From 1ecbdfa409135c311d00dd9f89f37cf6752f65a6 Mon Sep 17 00:00:00 2001
From: Gael <Gael@WL20-0067.corp.pasteur.fr>
Date: Fri, 16 Oct 2020 14:20:30 +0200
Subject: [PATCH] tempo
---
cute_little_R_functions.R | 23937 ++++++++++++++++-----------------
cute_little_R_functions.docx | Bin 460529 -> 461167 bytes
fun_gg_boxplot.docx | Bin 105393 -> 105410 bytes
fun_gg_scatter.docx | Bin 114104 -> 114123 bytes
other/cute_checks.docx | Bin 138628 -> 110509 bytes
5 files changed, 11968 insertions(+), 11969 deletions(-)
diff --git a/cute_little_R_functions.R b/cute_little_R_functions.R
index 40c7baf..bed9ef5 100644
--- a/cute_little_R_functions.R
+++ b/cute_little_R_functions.R
@@ -92,235 +92,234 @@
######## fun_check() #### check class, type, length, etc., of objects
-# Check r_debugging_tools-v1.2.R OK
+# Check r_debugging_tools-v1.4.R OK
# Check fun_test() (see cute_checks.docx) Ok
-# check manual: example to scan again
# clear to go Apollo
fun_check <- function(
- data,
- class = NULL,
- typeof = NULL,
- mode = NULL,
- length = NULL,
- prop = FALSE,
- double.as.integer.allowed = FALSE,
- options = NULL,
- all.options.in.data = FALSE,
- na.contain = FALSE,
- neg.values = TRUE,
- print = FALSE,
- data.name = NULL,
- fun.name = NULL
+data,
+class = NULL,
+typeof = NULL,
+mode = NULL,
+length = NULL,
+prop = FALSE,
+double.as.integer.allowed = FALSE,
+options = NULL,
+all.options.in.data = FALSE,
+na.contain = FALSE,
+neg.values = TRUE,
+print = FALSE,
+data.name = NULL,
+fun.name = NULL
){
- # AIM
- # check the class, type, mode and length of the data argument
- # mainly used to check the arguments of other functions
- # check also other kind of data parameters, is it a proportion? Is it type double but numbers without decimal part?
- # if options == NULL, then at least class or type or mode or length argument must be non null
- # if options is non null, then class, type and mode must be NULL, and length can be NULL or specified
- # WARNINGS
- # The function tests what is written in arguments, even if what is written is incoherent. For instance, fun_check(data = factor(1), class = "factor", mode = "character") will return a problem, and this, what ever the object tested in the data argument, because no object can be class "factor" and mode "character" (factors are class "factor" and mode "numeric")
- # Since R >= 4.0.0, class(matrix()) returns "matrix" "array", and not "matrix" alone as before. However, use argument class = "matrix" to check for matrix object (of class "matrix" "array" in R >= 4.0.0) and use argument class = "array" to check for array object (of class "array" in R >= 4.0.0)
- # ARGUMENTS
- # data: object to test
- # class: character string. Either one of the class() result (But see the warning section above) or "vector" or "ggplot2" (i.e., objects of class c("gg", "ggplot")) or NULL
- # typeof: character string. Either one of the typeof() result or NULL
- # mode: character string. Either one of the mode() result (for non vector object) or NULL
- # length: numeric value indicating the length of the object. Not considered if NULL
- # prop: logical. Are the numeric values between 0 and 1 (proportion)? If TRUE, can be used alone, without considering class, etc.
- # double.as.integer.allowed: logical. If TRUE, no error is reported if argument is set to typeof == "integer" or class == "integer", while the reality is typeof == "double" or class == "numeric" but the numbers strictly have zero as modulo (remainder of a division). This means that i <- 1 , which is typeof(i) -> "double" is considered as integer with double.as.integer.allowed = TRUE. WARNING: data%%1 == 0 but not isTRUE(all.equal(data%%1, 0)) is used here because the argument checks for integers stored as double (does not check for decimal numbers that are approximate integers)
- # options: a vector of character strings indicating all the possible option values for data
- # all.options.in.data: logical. If TRUE, all of the options must be present at least once in data, and nothing else. If FALSE, some or all of the options must be present in data, and nothing else. Ignored if options is NULL
- # na.contain: logical. Can data contain NA?
- # neg.values: logical. Are negative numeric values authorized? Warning: only considered if set to FALSE, to check for non negative values when class is set to "vector", "numeric", "matrix", "array", "data.frame", "table", or typeof is set to "double", "integer", or mode is set to "numeric". Ignored in other cases, notably with prop argument (which checks for values between 0 and 1 anyhow)
- # print: logical. Print the error message if $problem is TRUE? WARNING: set by default to FALSE, which facilitates the control of the error message output when using fun_check() inside functions. See the example section
- # data.name: character string indicating the name of the object to test. If NULL, use the name of the object assigned to the data argument
- # fun.name: character string indicating the name of the function checked (i.e., when fun_check() is used to check its argument). If non NULL, name will be added into the error message returned by fun_check()
- # RETURN
- # a list containing:
- # $problem: logical. Is there any problem detected?
- # $text: the problem detected
- # $fun.name: name of the checked parameter
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # none
- # EXAMPLES
- # test <- matrix(1:3) ; fun_check(data = test, print = TRUE, class = "vector", mode = "numeric")
- # DEBUGGING
- # data = expression(TEST) ; data.name = NULL ; class = "vector" ; typeof = NULL ; mode = NULL ; length = 1 ; prop = FALSE ; double.as.integer.allowed = FALSE ; options = NULL ; all.options.in.data = FALSE ; na.contain = FALSE ; neg.values = TRUE ; print = TRUE ; fun.name = NULL
- # function name: no used in this function for the error message, to avoid env colliding
- # argument checking
- # fun.name checked first because required next
- if( ! is.null(fun.name)){
- if( ! (all(class(fun.name) == "character") & length(fun.name) == 1)){
- tempo.cat <- paste0("ERROR IN fun_check(): THE fun.name ARGUMENT MUST BE A CHARACTER VECTOR OF LENGTH 1: ", paste(fun.name, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # end fun.name checked first because required next
- # arg with no default values
- if(missing(data)){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": ARGUMENT data HAS NO DEFAULT VALUE AND REQUIRES ONE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end arg with no default values
- # dealing with NA
- if(any(is.na(data.name)) | any(is.na(class)) | any(is.na(typeof)) | any(is.na(mode)) | any(is.na(length)) | any(is.na(prop)) | any(is.na(double.as.integer.allowed)) | any(is.na(all.options.in.data)) | any(is.na(na.contain)) | any(is.na(neg.values)) | any(is.na(print)) | any(is.na(fun.name))){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": NO ARGUMENT EXCEPT data AND options CAN HAVE NA VALUES\nPROBLEMATIC ARGUMENTS ARE: ", paste(c("data.name", "class", "typeof", "mode", "length", "prop", "double.as.integer.allowed", "all.options.in.data", "na.contain", "neg.values", "print", "fun.name")[c(any(is.na(data.name)), any(is.na(class)), any(is.na(typeof)), any(is.na(mode)), any(is.na(length)), any(is.na(prop)), any(is.na(double.as.integer.allowed)), any(is.na(all.options.in.data)), any(is.na(na.contain)), any(is.na(neg.values)), any(is.na(print)), any(is.na(fun.name)))], collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end dealing with NA
- # dealing with NULL
- if(is.null(prop) | is.null(double.as.integer.allowed) | is.null(all.options.in.data) | is.null(na.contain) | is.null(neg.values) | is.null(print)){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": THESE ARGUMENTS\nprop\ndouble.as.integer.allowed\nall.options.in.data\nna.contain\nneg.values\nprint\nCANNOT BE NULL\nPROBLEMATIC ARGUMENTS ARE: ", paste(c("prop", "double.as.integer.allowed", "all.options.in.data", "na.contain", "neg.values", "print")[c(is.null(prop), is.null(double.as.integer.allowed), is.null(all.options.in.data), is.null(na.contain), is.null(neg.values), is.null(print))], collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end dealing with NULL
- # dealing with logical
- # tested below
- # end dealing with logical
- if( ! is.null(data.name)){
- if( ! (length(data.name) == 1 & all(class(data.name) == "character"))){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": data.name ARGUMENT MUST BE A SINGLE CHARACTER ELEMENT AND NOT ", paste(data.name, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- if(is.null(options) & is.null(class) & is.null(typeof) & is.null(mode) & prop == FALSE & is.null(length)){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": AT LEAST ONE OF THE options, class, typeof, mode, prop, OR length ARGUMENT MUST BE SPECIFIED (I.E, TRUE FOR prop)")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! is.null(options) & ( ! is.null(class) | ! is.null(typeof) | ! is.null(mode) | prop == TRUE)){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": THE class, typeof, mode ARGUMENTS MUST BE NULL, AND prop FALSE, IF THE options ARGUMENT IS SPECIFIED\nTHE options ARGUMENT MUST BE NULL IF THE class AND/OR typeof AND/OR mode AND/OR prop ARGUMENT IS SPECIFIED")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! (all(class(neg.values) == "logical") & length(neg.values) == 1 & any(is.na(neg.values)) != TRUE)){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": THE neg.values ARGUMENT MUST BE TRUE OR FALSE ONLY")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(neg.values == FALSE & is.null(class) & is.null(typeof) & is.null(mode)){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": THE neg.values ARGUMENT CANNOT BE SWITCHED TO FALSE IF class, typeof AND mode ARGUMENTS ARE NULL")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! is.null(class)){
- if( ! all(class %in% c("vector", "logical", "integer", "numeric", "complex", "character", "matrix", "array", "data.frame", "list", "factor", "table", "expression", "name", "symbol", "function", "uneval", "environment", "ggplot2", "ggplot_built") & any(is.na(class)) != TRUE & length(class) == 1)){ # length == 1 here because of class(matrix()) since R4.0.0
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": class ARGUMENT MUST BE ONE OF THESE VALUE:\n\"vector\", \"logical\", \"integer\", \"numeric\", \"complex\", \"character\", \"matrix\", \"array\", \"data.frame\", \"list\", \"factor\", \"table\", \"expression\", \"name\", \"symbol\", \"function\", \"environment\", \"ggplot2\", \"ggplot_built\"")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(neg.values == FALSE & ! any(class %in% c("vector", "numeric", "integer", "matrix", "array", "data.frame", "table"))){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": class ARGUMENT CANNOT BE OTHER THAN \"vector\", \"numeric\", \"integer\", \"matrix\", \"array\", \"data.frame\", \"table\" IF neg.values ARGUMENT IS SWITCHED TO FALSE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- if( ! is.null(typeof)){
- if( ! (all(typeof %in% c("logical", "integer", "double", "complex", "character", "list", "expression", "name", "symbol", "closure", "special", "builtin", "environment", "S4")) & length(typeof) == 1 & any(is.na(typeof)) != TRUE)){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": typeof ARGUMENT MUST BE ONE OF THESE VALUE:\n\"logical\", \"integer\", \"double\", \"complex\", \"character\", \"list\", \"expression\", \"name\", \"symbol\", \"closure\", \"special\", \"builtin\", \"environment\", \"S4\"")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(neg.values == FALSE & ! typeof %in% c("double", "integer")){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": typeof ARGUMENT CANNOT BE OTHER THAN \"double\" OR \"integer\" IF neg.values ARGUMENT IS SWITCHED TO FALSE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- if( ! is.null(mode)){
- if( ! (all(mode %in% c("logical", "numeric", "complex", "character", "list", "expression", "name", "symbol", "function", "environment", "S4")) & length(mode) == 1 & any(is.na(mode)) != TRUE)){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": mode ARGUMENT MUST BE ONE OF THESE VALUE:\n\"logical\", \"numeric\", \"complex\", \"character\", \"list\", \"expression\", \"name\", \"symbol\", \"function\", \"environment\", \"S4\"")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(neg.values == FALSE & mode != "numeric"){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": mode ARGUMENT CANNOT BE OTHER THAN \"numeric\" IF neg.values ARGUMENT IS SWITCHED TO FALSE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- if( ! is.null(length)){
- if( ! (is.numeric(length) & length(length) == 1 & ! grepl(length, pattern = "\\.") & any(is.na(length)) != TRUE)){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": length ARGUMENT MUST BE A SINGLE INTEGER VALUE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- if( ! (is.logical(prop) | (length(prop) == 1 & any(is.na(prop)) != TRUE))){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": prop ARGUMENT MUST BE TRUE OR FALSE ONLY")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else if(prop == TRUE){
- if( ! is.null(class)){
- if( ! any(class %in% c("vector", "numeric", "matrix", "array", "data.frame", "table"))){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": class ARGUMENT CANNOT BE OTHER THAN NULL, \"vector\", \"numeric\", \"matrix\", \"array\", \"data.frame\", \"table\" IF prop ARGUMENT IS TRUE") # not integer because prop
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- if( ! is.null(mode)){
- if(mode != "numeric"){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": mode ARGUMENT CANNOT BE OTHER THAN NULL OR \"numeric\" IF prop ARGUMENT IS TRUE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- if( ! is.null(typeof)){
- if(typeof != "double"){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": typeof ARGUMENT CANNOT BE OTHER THAN NULL OR \"double\" IF prop ARGUMENT IS TRUE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- }
- if( ! (all(class(double.as.integer.allowed) == "logical") & length(double.as.integer.allowed) == 1 & any(is.na(double.as.integer.allowed)) != TRUE)){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": THE double.as.integer.allowed ARGUMENT MUST BE TRUE OR FALSE ONLY: ", paste(double.as.integer.allowed, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! (is.logical(all.options.in.data) & length(all.options.in.data) == 1 & any(is.na(all.options.in.data)) != TRUE)){
- tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": all.options.in.data ARGUMENT MUST BE A SINGLE LOGICAL VALUE (TRUE OR FALSE ONLY): ", paste(all.options.in.data, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! (all(class(na.contain) == "logical") & length(na.contain) == 1 & any(is.na(na.contain)) != TRUE)){
- tempo.cat <- paste0("ERROR IN fun_check(): THE na.contain ARGUMENT MUST BE TRUE OR FALSE ONLY: ", paste(na.contain, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! (all(class(print) == "logical") & length(print) == 1 & any(is.na(print)) != TRUE)){
- tempo.cat <- paste0("ERROR IN fun_check(): THE print ARGUMENT MUST BE TRUE OR FALSE ONLY: ", paste(print, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # data.name and fun.name tested at the beginning
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) # activate this line and use the function to check arguments status
- # end argument checking
- # main code
- if(is.null(data.name)){
- data.name <- deparse(substitute(data))
- }
- problem <- FALSE
- text <- paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " PARAMETER")
- if( ! is.null(options)){
- text <- ""
- if( ! all(data %in% options, na.rm = TRUE)){
- problem <- TRUE
- text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": THE ", data.name, " PARAMETER MUST BE SOME OF THESE OPTIONS: ", paste(options, collapse = " "), "\nTHE PROBLEMATIC ELEMENTS OF ", data.name, " ARE: ", paste(unique(data[ ! (data %in% options)]), collapse = " "))
- }
- if(all.options.in.data == TRUE){
- if( ! all(options %in% data)){ # no need of na.rm = TRUE for all because %in% does not output NA
- problem <- TRUE
- text <- paste0(ifelse(text == "", "", paste0(text, "\n")), ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": THE ", data.name, " PARAMETER MUST BE MADE OF ALL THESE OPTIONS: ", paste(options, collapse = " "), "\nTHE MISSING ELEMENTS OF THE options ARGUMENT ARE: ", paste(unique(options[ ! (options %in% data)]), collapse = " "))
- }
- }
- if( ! is.null(length)){
- if(length(data) != length){
- problem <- TRUE
- text <- paste0(ifelse(text == "", "", paste0(text, "\n")), ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": THE LENGTH OF ", data.name, " MUST BE ", length, " AND NOT ", length(data))
- }
- }
- if(text == ""){
- text <- paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " PARAMETER")
- }
- }
- arg.names <- c("class", "typeof", "mode", "length")
- if( ! is.null(class)){
- if(class == "matrix"){ # because of class(matric()) since R4.0.0
- class <- c("matrix", "array")
- }else if(class == "factor" & all(class(data) %in% c("factor", "ordered"))){ # to deal with ordered factors
- class <- c("factor", "ordered")
- }
- }
- if(is.null(options)){
- for(i2 in 1:length(arg.names)){
- if( ! is.null(get(arg.names[i2], env = sys.nframe(), inherit = FALSE))){
- # script to execute
- tempo.script <- '
+# AIM
+# check the class, type, mode and length of the data argument
+# mainly used to check the arguments of other functions
+# check also other kind of data parameters, is it a proportion? Is it type double but numbers without decimal part?
+# if options == NULL, then at least class or type or mode or length argument must be non null
+# if options is non null, then class, type and mode must be NULL, and length can be NULL or specified
+# WARNINGS
+# The function tests what is written in arguments, even if what is written is incoherent. For instance, fun_check(data = factor(1), class = "factor", mode = "character") will return a problem, and this, what ever the object tested in the data argument, because no object can be class "factor" and mode "character" (factors are class "factor" and mode "numeric")
+# Since R >= 4.0.0, class(matrix()) returns "matrix" "array", and not "matrix" alone as before. However, use argument class = "matrix" to check for matrix object (of class "matrix" "array" in R >= 4.0.0) and use argument class = "array" to check for array object (of class "array" in R >= 4.0.0)
+# ARGUMENTS
+# data: object to test
+# class: character string. Either one of the class() result (But see the warning section above) or "vector" or "ggplot2" (i.e., objects of class c("gg", "ggplot")) or NULL
+# typeof: character string. Either one of the typeof() result or NULL
+# mode: character string. Either one of the mode() result (for non vector object) or NULL
+# length: numeric value indicating the length of the object. Not considered if NULL
+# prop: logical. Are the numeric values between 0 and 1 (proportion)? If TRUE, can be used alone, without considering class, etc.
+# double.as.integer.allowed: logical. If TRUE, no error is reported if argument is set to typeof == "integer" or class == "integer", while the reality is typeof == "double" or class == "numeric" but the numbers strictly have zero as modulo (remainder of a division). This means that i <- 1 , which is typeof(i) -> "double" is considered as integer with double.as.integer.allowed = TRUE. WARNING: data%%1 == 0 but not isTRUE(all.equal(data%%1, 0)) is used here because the argument checks for integers stored as double (does not check for decimal numbers that are approximate integers)
+# options: a vector of character strings indicating all the possible option values for data
+# all.options.in.data: logical. If TRUE, all of the options must be present at least once in data, and nothing else. If FALSE, some or all of the options must be present in data, and nothing else. Ignored if options is NULL
+# na.contain: logical. Can data contain NA?
+# neg.values: logical. Are negative numeric values authorized? Warning: only considered if set to FALSE, to check for non negative values when class is set to "vector", "numeric", "matrix", "array", "data.frame", "table", or typeof is set to "double", "integer", or mode is set to "numeric". Ignored in other cases, notably with prop argument (which checks for values between 0 and 1 anyhow)
+# print: logical. Print the error message if $problem is TRUE? WARNING: set by default to FALSE, which facilitates the control of the error message output when using fun_check() inside functions. See the example section
+# data.name: character string indicating the name of the object to test. If NULL, use the name of the object assigned to the data argument
+# fun.name: character string indicating the name of the function checked (i.e., when fun_check() is used to check its argument). If non NULL, name will be added into the error message returned by fun_check()
+# RETURN
+# a list containing:
+# $problem: logical. Is there any problem detected?
+# $text: the problem detected
+# $fun.name: name of the checked parameter
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# none
+# EXAMPLE
+# test <- matrix(1:3) ; fun_check(data = test, print = TRUE, class = "vector", mode = "numeric")
+# DEBUGGING
+# data = expression(TEST) ; data.name = NULL ; class = "vector" ; typeof = NULL ; mode = NULL ; length = 1 ; prop = FALSE ; double.as.integer.allowed = FALSE ; options = NULL ; all.options.in.data = FALSE ; na.contain = FALSE ; neg.values = TRUE ; print = TRUE ; fun.name = NULL
+# function name: no used in this function for the error message, to avoid env colliding
+# argument checking
+# fun.name checked first because required next
+if( ! is.null(fun.name)){
+if( ! (all(class(fun.name) == "character") & length(fun.name) == 1)){
+tempo.cat <- paste0("ERROR IN fun_check(): THE fun.name ARGUMENT MUST BE A CHARACTER VECTOR OF LENGTH 1: ", paste(fun.name, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# end fun.name checked first because required next
+# arg with no default values
+if(missing(data)){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": ARGUMENT data HAS NO DEFAULT VALUE AND REQUIRES ONE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end arg with no default values
+# dealing with NA
+if(any(is.na(data.name)) | any(is.na(class)) | any(is.na(typeof)) | any(is.na(mode)) | any(is.na(length)) | any(is.na(prop)) | any(is.na(double.as.integer.allowed)) | any(is.na(all.options.in.data)) | any(is.na(na.contain)) | any(is.na(neg.values)) | any(is.na(print)) | any(is.na(fun.name))){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": NO ARGUMENT EXCEPT data AND options CAN HAVE NA VALUES\nPROBLEMATIC ARGUMENTS ARE: ", paste(c("data.name", "class", "typeof", "mode", "length", "prop", "double.as.integer.allowed", "all.options.in.data", "na.contain", "neg.values", "print", "fun.name")[c(any(is.na(data.name)), any(is.na(class)), any(is.na(typeof)), any(is.na(mode)), any(is.na(length)), any(is.na(prop)), any(is.na(double.as.integer.allowed)), any(is.na(all.options.in.data)), any(is.na(na.contain)), any(is.na(neg.values)), any(is.na(print)), any(is.na(fun.name)))], collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end dealing with NA
+# dealing with NULL
+if(is.null(prop) | is.null(double.as.integer.allowed) | is.null(all.options.in.data) | is.null(na.contain) | is.null(neg.values) | is.null(print)){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": THESE ARGUMENTS\nprop\ndouble.as.integer.allowed\nall.options.in.data\nna.contain\nneg.values\nprint\nCANNOT BE NULL\nPROBLEMATIC ARGUMENTS ARE: ", paste(c("prop", "double.as.integer.allowed", "all.options.in.data", "na.contain", "neg.values", "print")[c(is.null(prop), is.null(double.as.integer.allowed), is.null(all.options.in.data), is.null(na.contain), is.null(neg.values), is.null(print))], collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end dealing with NULL
+# dealing with logical
+# tested below
+# end dealing with logical
+if( ! is.null(data.name)){
+if( ! (length(data.name) == 1 & all(class(data.name) == "character"))){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": data.name ARGUMENT MUST BE A SINGLE CHARACTER ELEMENT AND NOT ", paste(data.name, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+if(is.null(options) & is.null(class) & is.null(typeof) & is.null(mode) & prop == FALSE & is.null(length)){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": AT LEAST ONE OF THE options, class, typeof, mode, prop, OR length ARGUMENT MUST BE SPECIFIED (I.E, TRUE FOR prop)")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! is.null(options) & ( ! is.null(class) | ! is.null(typeof) | ! is.null(mode) | prop == TRUE)){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": THE class, typeof, mode ARGUMENTS MUST BE NULL, AND prop FALSE, IF THE options ARGUMENT IS SPECIFIED\nTHE options ARGUMENT MUST BE NULL IF THE class AND/OR typeof AND/OR mode AND/OR prop ARGUMENT IS SPECIFIED")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! (all(class(neg.values) == "logical") & length(neg.values) == 1 & any(is.na(neg.values)) != TRUE)){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": THE neg.values ARGUMENT MUST BE TRUE OR FALSE ONLY")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(neg.values == FALSE & is.null(class) & is.null(typeof) & is.null(mode)){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": THE neg.values ARGUMENT CANNOT BE SWITCHED TO FALSE IF class, typeof AND mode ARGUMENTS ARE NULL")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! is.null(class)){
+if( ! all(class %in% c("vector", "logical", "integer", "numeric", "complex", "character", "matrix", "array", "data.frame", "list", "factor", "table", "expression", "name", "symbol", "function", "uneval", "environment", "ggplot2", "ggplot_built") & any(is.na(class)) != TRUE & length(class) == 1)){ # length == 1 here because of class(matrix()) since R4.0.0
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": class ARGUMENT MUST BE ONE OF THESE VALUE:\n\"vector\", \"logical\", \"integer\", \"numeric\", \"complex\", \"character\", \"matrix\", \"array\", \"data.frame\", \"list\", \"factor\", \"table\", \"expression\", \"name\", \"symbol\", \"function\", \"environment\", \"ggplot2\", \"ggplot_built\"")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(neg.values == FALSE & ! any(class %in% c("vector", "numeric", "integer", "matrix", "array", "data.frame", "table"))){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": class ARGUMENT CANNOT BE OTHER THAN \"vector\", \"numeric\", \"integer\", \"matrix\", \"array\", \"data.frame\", \"table\" IF neg.values ARGUMENT IS SWITCHED TO FALSE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+if( ! is.null(typeof)){
+if( ! (all(typeof %in% c("logical", "integer", "double", "complex", "character", "list", "expression", "name", "symbol", "closure", "special", "builtin", "environment", "S4")) & length(typeof) == 1 & any(is.na(typeof)) != TRUE)){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": typeof ARGUMENT MUST BE ONE OF THESE VALUE:\n\"logical\", \"integer\", \"double\", \"complex\", \"character\", \"list\", \"expression\", \"name\", \"symbol\", \"closure\", \"special\", \"builtin\", \"environment\", \"S4\"")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(neg.values == FALSE & ! typeof %in% c("double", "integer")){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": typeof ARGUMENT CANNOT BE OTHER THAN \"double\" OR \"integer\" IF neg.values ARGUMENT IS SWITCHED TO FALSE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+if( ! is.null(mode)){
+if( ! (all(mode %in% c("logical", "numeric", "complex", "character", "list", "expression", "name", "symbol", "function", "environment", "S4")) & length(mode) == 1 & any(is.na(mode)) != TRUE)){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": mode ARGUMENT MUST BE ONE OF THESE VALUE:\n\"logical\", \"numeric\", \"complex\", \"character\", \"list\", \"expression\", \"name\", \"symbol\", \"function\", \"environment\", \"S4\"")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(neg.values == FALSE & mode != "numeric"){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": mode ARGUMENT CANNOT BE OTHER THAN \"numeric\" IF neg.values ARGUMENT IS SWITCHED TO FALSE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+if( ! is.null(length)){
+if( ! (is.numeric(length) & length(length) == 1 & ! grepl(length, pattern = "\\.") & any(is.na(length)) != TRUE)){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": length ARGUMENT MUST BE A SINGLE INTEGER VALUE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+if( ! (is.logical(prop) | (length(prop) == 1 & any(is.na(prop)) != TRUE))){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": prop ARGUMENT MUST BE TRUE OR FALSE ONLY")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else if(prop == TRUE){
+if( ! is.null(class)){
+if( ! any(class %in% c("vector", "numeric", "matrix", "array", "data.frame", "table"))){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": class ARGUMENT CANNOT BE OTHER THAN NULL, \"vector\", \"numeric\", \"matrix\", \"array\", \"data.frame\", \"table\" IF prop ARGUMENT IS TRUE") # not integer because prop
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+if( ! is.null(mode)){
+if(mode != "numeric"){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": mode ARGUMENT CANNOT BE OTHER THAN NULL OR \"numeric\" IF prop ARGUMENT IS TRUE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+if( ! is.null(typeof)){
+if(typeof != "double"){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": typeof ARGUMENT CANNOT BE OTHER THAN NULL OR \"double\" IF prop ARGUMENT IS TRUE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+}
+if( ! (all(class(double.as.integer.allowed) == "logical") & length(double.as.integer.allowed) == 1 & any(is.na(double.as.integer.allowed)) != TRUE)){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": THE double.as.integer.allowed ARGUMENT MUST BE TRUE OR FALSE ONLY: ", paste(double.as.integer.allowed, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! (is.logical(all.options.in.data) & length(all.options.in.data) == 1 & any(is.na(all.options.in.data)) != TRUE)){
+tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": all.options.in.data ARGUMENT MUST BE A SINGLE LOGICAL VALUE (TRUE OR FALSE ONLY): ", paste(all.options.in.data, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! (all(class(na.contain) == "logical") & length(na.contain) == 1 & any(is.na(na.contain)) != TRUE)){
+tempo.cat <- paste0("ERROR IN fun_check(): THE na.contain ARGUMENT MUST BE TRUE OR FALSE ONLY: ", paste(na.contain, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! (all(class(print) == "logical") & length(print) == 1 & any(is.na(print)) != TRUE)){
+tempo.cat <- paste0("ERROR IN fun_check(): THE print ARGUMENT MUST BE TRUE OR FALSE ONLY: ", paste(print, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# data.name and fun.name tested at the beginning
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) # activate this line and use the function to check arguments status
+# end argument checking
+# main code
+if(is.null(data.name)){
+data.name <- deparse(substitute(data))
+}
+problem <- FALSE
+text <- paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " PARAMETER")
+if( ! is.null(options)){
+text <- ""
+if( ! all(data %in% options, na.rm = TRUE)){
+problem <- TRUE
+text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": THE ", data.name, " PARAMETER MUST BE SOME OF THESE OPTIONS: ", paste(options, collapse = " "), "\nTHE PROBLEMATIC ELEMENTS OF ", data.name, " ARE: ", paste(unique(data[ ! (data %in% options)]), collapse = " "))
+}
+if(all.options.in.data == TRUE){
+if( ! all(options %in% data)){ # no need of na.rm = TRUE for all because %in% does not output NA
+problem <- TRUE
+text <- paste0(ifelse(text == "", "", paste0(text, "\n")), ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": THE ", data.name, " PARAMETER MUST BE MADE OF ALL THESE OPTIONS: ", paste(options, collapse = " "), "\nTHE MISSING ELEMENTS OF THE options ARGUMENT ARE: ", paste(unique(options[ ! (options %in% data)]), collapse = " "))
+}
+}
+if( ! is.null(length)){
+if(length(data) != length){
+problem <- TRUE
+text <- paste0(ifelse(text == "", "", paste0(text, "\n")), ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": THE LENGTH OF ", data.name, " MUST BE ", length, " AND NOT ", length(data))
+}
+}
+if(text == ""){
+text <- paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " PARAMETER")
+}
+}
+arg.names <- c("class", "typeof", "mode", "length")
+if( ! is.null(class)){
+if(class == "matrix"){ # because of class(matric()) since R4.0.0
+class <- c("matrix", "array")
+}else if(class == "factor" & all(class(data) %in% c("factor", "ordered"))){ # to deal with ordered factors
+class <- c("factor", "ordered")
+}
+}
+if(is.null(options)){
+for(i2 in 1:length(arg.names)){
+if( ! is.null(get(arg.names[i2], env = sys.nframe(), inherit = FALSE))){
+# script to execute
+tempo.script <- '
problem <- TRUE ;
if(identical(text, paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " PARAMETER"))){
text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": THE ", data.name, " PARAMETER MUST BE ") ;
@@ -329,59 +328,59 @@ text <- paste0(text, " AND ") ;
}
text <- paste0(text, toupper(arg.names[i2]), " ", if(all(get(arg.names[i2], env = sys.nframe(), inherit = FALSE) %in% c("matrix", "array"))){"matrix"}else if(all(get(arg.names[i2], env = sys.nframe(), inherit = FALSE) %in% c("factor", "ordered"))){"factor"}else{get(arg.names[i2], env = sys.nframe(), inherit = FALSE)})
'
- # end script to execute
- if(typeof(data) == "double" & double.as.integer.allowed == TRUE & ((arg.names[i2] == "class" & all(get(arg.names[i2], env = sys.nframe(), inherit = FALSE) == "integer")) | (arg.names[i2] == "typeof" & all(get(arg.names[i2], env = sys.nframe(), inherit = FALSE) == "integer")))){
- if( ! all(data %% 1 == 0, na.rm = TRUE)){ # to check integers (use %%, meaning the remaining of a division): see the precedent line. isTRUE(all.equal(data%%1, rep(0, length(data)))) not used because we strictly need zero as a result
- eval(parse(text = tempo.script)) # execute tempo.script
- }
- }else if( ! any(all(get(arg.names[i2], env = sys.nframe(), inherit = FALSE) %in% c("vector", "ggplot2"))) & ! all(eval(parse(text = paste0(arg.names[i2], "(data)"))) %in% get(arg.names[i2], env = sys.nframe(), inherit = FALSE))){ # test the four c("class", "typeof", "mode", "length") arguments with their corresponding function. No need of na.rm = TRUE for all because %in% does not output NA
- eval(parse(text = tempo.script)) # execute tempo.script
- }else if(arg.names[i2] == "class" & all(get(arg.names[i2], env = sys.nframe(), inherit = FALSE) == "vector") & ! (all(class(data) %in% "numeric") | all(class(data) %in% "integer") | all(class(data) %in% "character") | all(class(data) %in% "logical"))){ # test class == "vector". No need of na.rm = TRUE for all because %in% does not output NA
- eval(parse(text = tempo.script)) # execute tempo.script
- }else if(arg.names[i2] == "class" & all(get(arg.names[i2], env = sys.nframe(), inherit = FALSE) == "ggplot2") & ! all(class(data) %in% c("gg", "ggplot"))){ # test ggplot object
- eval(parse(text = tempo.script)) # execute tempo.script
- }
- }
- }
- }
+# end script to execute
+if(typeof(data) == "double" & double.as.integer.allowed == TRUE & ((arg.names[i2] == "class" & all(get(arg.names[i2], env = sys.nframe(), inherit = FALSE) == "integer")) | (arg.names[i2] == "typeof" & all(get(arg.names[i2], env = sys.nframe(), inherit = FALSE) == "integer")))){
+if( ! all(data %% 1 == 0, na.rm = TRUE)){ # to check integers (use %%, meaning the remaining of a division): see the precedent line. isTRUE(all.equal(data%%1, rep(0, length(data)))) not used because we strictly need zero as a result
+eval(parse(text = tempo.script)) # execute tempo.script
+}
+}else if( ! any(all(get(arg.names[i2], env = sys.nframe(), inherit = FALSE) %in% c("vector", "ggplot2"))) & ! all(eval(parse(text = paste0(arg.names[i2], "(data)"))) %in% get(arg.names[i2], env = sys.nframe(), inherit = FALSE))){ # test the four c("class", "typeof", "mode", "length") arguments with their corresponding function. No need of na.rm = TRUE for all because %in% does not output NA
+eval(parse(text = tempo.script)) # execute tempo.script
+}else if(arg.names[i2] == "class" & all(get(arg.names[i2], env = sys.nframe(), inherit = FALSE) == "vector") & ! (all(class(data) %in% "numeric") | all(class(data) %in% "integer") | all(class(data) %in% "character") | all(class(data) %in% "logical"))){ # test class == "vector". No need of na.rm = TRUE for all because %in% does not output NA
+eval(parse(text = tempo.script)) # execute tempo.script
+}else if(arg.names[i2] == "class" & all(get(arg.names[i2], env = sys.nframe(), inherit = FALSE) == "ggplot2") & ! all(class(data) %in% c("gg", "ggplot"))){ # test ggplot object
+eval(parse(text = tempo.script)) # execute tempo.script
+}
+}
+}
+}
if(prop == TRUE){
- if(is.null(data) | any(data < 0 | data > 1, na.rm = TRUE)){
- problem <- TRUE
- if(identical(text, paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " PARAMETER"))){
- text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": ")
- }else{
- text <- paste0(text, " AND ")
- }
- text <- paste0(text, "THE ", data.name, " PARAMETER MUST BE DECIMAL VALUES BETWEEN 0 AND 1")
- }
+if(is.null(data) | any(data < 0 | data > 1, na.rm = TRUE)){
+problem <- TRUE
+if(identical(text, paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " PARAMETER"))){
+text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": ")
+}else{
+text <- paste0(text, " AND ")
+}
+text <- paste0(text, "THE ", data.name, " PARAMETER MUST BE DECIMAL VALUES BETWEEN 0 AND 1")
+}
}
if(all(class(data) %in% "expression")){ # no need of na.rm = TRUE for all because %in% does not output NA
- data <- as.character(data) # to evaluate the presence of NA
+data <- as.character(data) # to evaluate the presence of NA
}
if(na.contain == FALSE & (mode(data) %in% c("logical", "numeric", "complex", "character", "list", "expression", "name", "symbol"))){ # before it was ! (class(data) %in% c("function", "environment"))
- if(any(is.na(data)) == TRUE){ # not on the same line because when data is class envir or function , do not like that
- problem <- TRUE
- if(identical(text, paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " PARAMETER"))){
- text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": ")
- }else{
- text <- paste0(text, " AND ")
- }
- text <- paste0(text, "THE ", data.name, " PARAMETER CONTAINS NA WHILE NOT AUTHORIZED")
- }
+if(any(is.na(data)) == TRUE){ # not on the same line because when data is class envir or function , do not like that
+problem <- TRUE
+if(identical(text, paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " PARAMETER"))){
+text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": ")
+}else{
+text <- paste0(text, " AND ")
+}
+text <- paste0(text, "THE ", data.name, " PARAMETER CONTAINS NA WHILE NOT AUTHORIZED")
+}
}
if(neg.values == FALSE){
- if(any(data < 0, na.rm = TRUE)){
- problem <- TRUE
- if(identical(text, paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " PARAMETER"))){
- text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": ")
- }else{
- text <- paste0(text, " AND ")
- }
- text <- paste0(text, "THE ", data.name, " PARAMETER MUST BE NON NEGATIVE NUMERIC VALUES")
- }
+if(any(data < 0, na.rm = TRUE)){
+problem <- TRUE
+if(identical(text, paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " PARAMETER"))){
+text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": ")
+}else{
+text <- paste0(text, " AND ")
+}
+text <- paste0(text, "THE ", data.name, " PARAMETER MUST BE NON NEGATIVE NUMERIC VALUES")
+}
}
if(print == TRUE & problem == TRUE){
- cat(paste0("\n\n================\n\n", text, "\n\n================\n\n"))
+cat(paste0("\n\n================\n\n", text, "\n\n================\n\n"))
}
output <- list(problem = problem, text = text, fun.name = data.name)
return(output)
@@ -391,79 +390,79 @@ return(output)
fun_secu <- function(pos = 1, name = NULL){
- # AIM
- # verif that local variables are not present in other environments, in order to avoid scope preference usage. The fun_secu() function checks by default the parent environment. This means that when used inside a function, it checks the local environment of this function. When used in the Global environment, it would check this environment
- # ARGUMENTS
- # pos: single integer indicating the position of the environment checked (argument n of parent.frame()). VaLue 1 means one step above the fun_secu() local environment. Thus, if fun_secu() is used in the working environment, with pos ==1, variables of this env will be checked in the above envs. If fun_secu() is used in a function, with pos ==1, variables presents in the local env of the functions will be checked in the above envs (which includes the working environment (Global env)
- # name: single character string indicating the name of the function checked
- # RETURN
- # a character string of the local variables that match variables in the different environments of the R scope, or NULL if no match
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # fun_secu()
- # fun_secu(pos = 2)
- # mean <- 0 ; fun1 <- function(){sd <- 1 ; fun_secu(name = as.character(sys.calls()[[length(sys.calls())]]))} ; fun2 <- function(){cor <- 2 ; fun1()} ; fun1() ; fun2() ; rm(mean) # sys.calls() gives the the function name at top stack of the imbricated functions, sys.calls()[[length(sys.calls())]] the name of the just above function. This can also been used for the above function: as.character(sys.call(1))
- # test.pos <- 2 ; mean <- 0 ; fun1 <- function(){sd <- 1 ; fun_secu(pos = test.pos, name = if(length(sys.calls()) >= test.pos){as.character(sys.calls()[[length(sys.calls()) + 1 - test.pos]])}else{search()[ (1:length(search()))[test.pos - length(sys.calls())]]})} ; fun2 <- function(){cor <- 2 ; fun1()} ; fun1() ; fun2() ; rm(mean) # for argument name, here is a way to have the name of the tested environment according to test.pos value
- # DEBUGGING
- # pos = 1 ; name = NULL # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = pos, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(name)){
- tempo <- fun_check(data = name, class = "vector", typeof = "character", length = 1, fun.name = function.name) ; eval(ee)
- }
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- # match.list <- vector("list", length = (length(sys.calls()) - 1 + length(search()) + ifelse(length(sys.calls()) == 1, -1, 0))) # match.list is a list of all the environment tested (local of functions and R envs), length(sys.calls()) - 1 to remove the level of the fun_secu() function, sys.calls() giving all the names of the imbricated functions, including fun_secu, ifelse(length(sys.calls()) == 1, -1, 0) to remove Global env if this one is tested
- tempo.name <- rev(as.character(unlist(sys.calls()))) # get names of frames (i.e., enclosed env)
- tempo.frame <- rev(sys.frames()) # get frames (i.e., enclosed env)
- # dealing with source()
- # source() used in the Global env creates three frames above the Global env, which should be removed because not very interesting for variable duplications. Add a <<-(sys.frames()) in this code and source anova_contrasts code to see this. With ls(a[[4]]), we can see the content of each env, which are probably elements of source()
- if(any(sapply(tempo.frame, FUN = environmentName) %in% "R_GlobalEnv")){
- global.pos <- which(sapply(tempo.frame, FUN = environmentName) %in% "R_GlobalEnv")
- # remove the global env (because already in search(), and all the oabove env
- tempo.name <- tempo.name[-c(global.pos:length(tempo.frame))]
- tempo.frame <- tempo.frame[-c(global.pos:length(tempo.frame))]
- }
- # end dealing with source()
- # might have a problem if(length(tempo.name) == 0){
- match.list <- vector("list", length = length(tempo.name) + length(search())) # match.list is a list of all the environment tested (local of functions and R envs), length(sys.calls()) - 1 to remove the level of the fun_secu() function, sys.calls() giving all the names of the imbricated functions, including fun_secu, ifelse(length(sys.calls()) == 1, -1, 0) to remove Global env if this one is tested
- ls.names <- c(tempo.name, search()) # names of the functions + names of the search() environments
- ls.input <- c(tempo.frame, as.list(search())) # environements of the functions + names of the search() environments
- names(match.list) <- ls.names #
- match.list <- match.list[-c(1:(pos + 1))] # because we check only above pos
- ls.tested <- ls.input[[pos + 1]]
- ls.input <- ls.input[-c(1:(pos + 1))]
- for(i1 in 1:length(match.list)){
- if(any(ls(name = ls.input[[i1]], all.names = TRUE) %in% ls(name = ls.tested, all.names = TRUE))){
- match.list[i1] <- list(ls(name = ls.input[[i1]], all.names = TRUE)[ls(name = ls.input[[i1]], all.names = TRUE) %in% ls(name = ls.tested, all.names = TRUE)])
- }
- }
- if( ! all(sapply(match.list, FUN = is.null))){
- output <- paste0("SOME VARIABLES ", ifelse(is.null(name), "OF THE CHECKED ENVIRONMENT", paste0("OF ", name)), " ARE ALSO PRESENT IN :\n", paste0(names(match.list[ ! sapply(match.list, FUN = is.null)]), ": ", sapply(match.list[ ! sapply(match.list, FUN = is.null)], FUN = paste0, collapse = " "), collapse = "\n"))
- }else{
- output <- NULL
- }
- return(output)
+# AIM
+# verif that local variables are not present in other environments, in order to avoid scope preference usage. The fun_secu() function checks by default the parent environment. This means that when used inside a function, it checks the local environment of this function. When used in the Global environment, it would check this environment
+# ARGUMENTS
+# pos: single integer indicating the position of the environment checked (argument n of parent.frame()). VaLue 1 means one step above the fun_secu() local environment. Thus, if fun_secu() is used in the working environment, with pos ==1, variables of this env will be checked in the above envs. If fun_secu() is used in a function, with pos ==1, variables presents in the local env of the functions will be checked in the above envs (which includes the working environment (Global env)
+# name: single character string indicating the name of the function checked
+# RETURN
+# a character string of the local variables that match variables in the different environments of the R scope, or NULL if no match
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# fun_secu()
+# fun_secu(pos = 2)
+# mean <- 0 ; fun1 <- function(){sd <- 1 ; fun_secu(name = as.character(sys.calls()[[length(sys.calls())]]))} ; fun2 <- function(){cor <- 2 ; fun1()} ; fun1() ; fun2() ; rm(mean) # sys.calls() gives the the function name at top stack of the imbricated functions, sys.calls()[[length(sys.calls())]] the name of the just above function. This can also been used for the above function: as.character(sys.call(1))
+# test.pos <- 2 ; mean <- 0 ; fun1 <- function(){sd <- 1 ; fun_secu(pos = test.pos, name = if(length(sys.calls()) >= test.pos){as.character(sys.calls()[[length(sys.calls()) + 1 - test.pos]])}else{search()[ (1:length(search()))[test.pos - length(sys.calls())]]})} ; fun2 <- function(){cor <- 2 ; fun1()} ; fun1() ; fun2() ; rm(mean) # for argument name, here is a way to have the name of the tested environment according to test.pos value
+# DEBUGGING
+# pos = 1 ; name = NULL # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = pos, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(name)){
+tempo <- fun_check(data = name, class = "vector", typeof = "character", length = 1, fun.name = function.name) ; eval(ee)
+}
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+# match.list <- vector("list", length = (length(sys.calls()) - 1 + length(search()) + ifelse(length(sys.calls()) == 1, -1, 0))) # match.list is a list of all the environment tested (local of functions and R envs), length(sys.calls()) - 1 to remove the level of the fun_secu() function, sys.calls() giving all the names of the imbricated functions, including fun_secu, ifelse(length(sys.calls()) == 1, -1, 0) to remove Global env if this one is tested
+tempo.name <- rev(as.character(unlist(sys.calls()))) # get names of frames (i.e., enclosed env)
+tempo.frame <- rev(sys.frames()) # get frames (i.e., enclosed env)
+# dealing with source()
+# source() used in the Global env creates three frames above the Global env, which should be removed because not very interesting for variable duplications. Add a <<-(sys.frames()) in this code and source anova_contrasts code to see this. With ls(a[[4]]), we can see the content of each env, which are probably elements of source()
+if(any(sapply(tempo.frame, FUN = environmentName) %in% "R_GlobalEnv")){
+global.pos <- which(sapply(tempo.frame, FUN = environmentName) %in% "R_GlobalEnv")
+# remove the global env (because already in search(), and all the oabove env
+tempo.name <- tempo.name[-c(global.pos:length(tempo.frame))]
+tempo.frame <- tempo.frame[-c(global.pos:length(tempo.frame))]
+}
+# end dealing with source()
+# might have a problem if(length(tempo.name) == 0){
+match.list <- vector("list", length = length(tempo.name) + length(search())) # match.list is a list of all the environment tested (local of functions and R envs), length(sys.calls()) - 1 to remove the level of the fun_secu() function, sys.calls() giving all the names of the imbricated functions, including fun_secu, ifelse(length(sys.calls()) == 1, -1, 0) to remove Global env if this one is tested
+ls.names <- c(tempo.name, search()) # names of the functions + names of the search() environments
+ls.input <- c(tempo.frame, as.list(search())) # environements of the functions + names of the search() environments
+names(match.list) <- ls.names #
+match.list <- match.list[-c(1:(pos + 1))] # because we check only above pos
+ls.tested <- ls.input[[pos + 1]]
+ls.input <- ls.input[-c(1:(pos + 1))]
+for(i1 in 1:length(match.list)){
+if(any(ls(name = ls.input[[i1]], all.names = TRUE) %in% ls(name = ls.tested, all.names = TRUE))){
+match.list[i1] <- list(ls(name = ls.input[[i1]], all.names = TRUE)[ls(name = ls.input[[i1]], all.names = TRUE) %in% ls(name = ls.tested, all.names = TRUE)])
+}
+}
+if( ! all(sapply(match.list, FUN = is.null))){
+output <- paste0("SOME VARIABLES ", ifelse(is.null(name), "OF THE CHECKED ENVIRONMENT", paste0("OF ", name)), " ARE ALSO PRESENT IN :\n", paste0(names(match.list[ ! sapply(match.list, FUN = is.null)]), ": ", sapply(match.list[ ! sapply(match.list, FUN = is.null)], FUN = paste0, collapse = " "), collapse = "\n"))
+}else{
+output <- NULL
+}
+return(output)
}
@@ -472,147 +471,147 @@ fun_secu <- function(pos = 1, name = NULL){
# Check OK: clear to go Apollo
fun_info <- function(data, n = 20, full = FALSE, warn.print = TRUE){
- # AIM
- # provide a complete description of an object. Using complete = TRUE allows data recovering. Using complete = FALSE is a convenient display feature (in case of large dataset)
- # ARGUMENTS
- # data: object to test
- # n: positive integer value indicating the number of element to display per compartment of the output list (i.e., head(..., n)). Ignored if full argument is TRUE. Also ignored for the STRUCTURE compartment output, corresponding to ls.str() information (because head() removes almost everything)
- # full: logical. Return the full information?
- # warn.print: logical. Print potential warnings at the end of the execution? If FALSE the message or NULL (if no message) is added in the output as an additional compartment
- # RETURN
- # a list containing information, depending on the class and type of data
- # if data is made of numerics, provide range, sum, mean, number of NA and number of Inf
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # fun_info(data = 1:3)
- # fun_info(data = 1:3, n = 2)
- # fun_info(data = 1:3, n = 2, full = TRUE)
- # fun_info(data.frame(a = 1:2, b = ordered(factor(c("A", "B"))), stringsAsFactors = TRUE))
- # fun_info(list(a = 1:3, b = ordered(factor(c("A", "B")))))
- # DEBUGGING
- # data = NULL # for function debugging
- # data = 1:3 # for function debugging
- # data = matrix(1:3) # for function debugging
- # data = data.frame(a = 1:2, b = c("A", "B"), stringsAsFactors = TRUE) # for function debugging
- # data = factor(c("b", "a")) # for function debugging
- # data = ordered(factor(c("b", "a"))) # for function debugging
- # data = list(a = 1:3, b = factor(c("A", "B"))) # for function debugging
- # data = list(a = 1:3, b = ordered(factor(c("A", "B")))) # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = n, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & n < 1){
- tempo.cat <- paste0("ERROR IN ", function.name, ": n ARGUMENT MUST BE A POSITIVE AND NON NULL INTEGER")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- tempo <- fun_check(data = full, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = warn.print, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # end argument checking
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) # activate this line and use the function to check arguments status
- # main code
- warn <- NULL
- if(full == FALSE){
- warn <- paste0("FROM ", function.name, ":\n\nSOME COMPARTMENTS CAN BE TRUNCATED (n ARGUMENT IS ", n, ")\n\n")
- }
- data.name <- deparse(substitute(data))
- output <- list("NAME" = data.name)
- tempo <- list("CLASS" = class(data))
- output <- c(output, tempo)
- tempo <- list("TYPE" = typeof(data))
- output <- c(output, tempo)
- tempo <- list("LENGTH" = length(data))
- output <- c(output, tempo)
- if(all(typeof(data) %in% c("integer", "numeric", "double"))){
- tempo <- list("RANGE" = range(data[ ! is.infinite(data)], na.rm = TRUE))
- output <- c(output, tempo)
- tempo <- list("SUM" = sum(data[ ! is.infinite(data)], na.rm = TRUE))
- output <- c(output, tempo)
- tempo <- list("MEAN" = mean(data[ ! is.infinite(data)], na.rm = TRUE))
- output <- c(output, tempo)
- tempo <- list("NA.NB" = sum(is.na(data)))
- output <- c(output, tempo)
- tempo <- list("INF.NB" = sum(is.infinite(data)))
- output <- c(output, tempo)
- }
- tempo <- list("HEAD" = head(data))
- output <- c(output, tempo)
- if( ! is.null(data)){
- tempo <- list("TAIL" = tail(data))
- output <- c(output, tempo)
- if( ! is.null(dim(data))){
- tempo <- list("DIMENSION" = dim(data))
- names(tempo[[1]]) <- c("NROW", "NCOL")
- output <- c(output, tempo)
- }
- tempo <- list("SUMMARY" = summary(data))
- output <- c(output, tempo)
- }
- if(all(class(data) == "data.frame" | all(class(data) %in% c("matrix", "array")))){
- tempo <- list("ROW_NAMES" = dimnames(data)[[1]])
- output <- c(output, tempo)
- tempo <- list("COLUM_NAMES" = dimnames(data)[[2]])
- output <- c(output, tempo)
- }
- if(all(class(data) == "data.frame")){
- tempo <- list("STRUCTURE" = ls.str(data)) # str() print automatically, ls.str() not but does not give the order of the data.frame
- output <- c(output, tempo)
- tempo <- list("COLUMN_TYPE" = sapply(data, FUN = "typeof"))
- if(any(sapply(data, FUN = "class") %in% "factor")){ # if an ordered factor is present, then sapply(data, FUN = "class") return a list but works with any(sapply(data, FUN = "class") %in% "factor")
- tempo.class <- sapply(data, FUN = "class")
- if(any(unlist(tempo.class) %in% "ordered")){
- tempo2 <- sapply(tempo.class, paste, collapse = " ") # paste the "ordered" factor" in "ordered factor"
- }else{
- tempo2 <- unlist(tempo.class)
- }
- tempo[["COLUMN_TYPE"]][grepl(x = tempo2, pattern = "factor")] <- tempo2[grepl(x = tempo2, pattern = "factor")]
- }
- output <- c(output, tempo)
- }
- if(all(class(data) == "list")){
- tempo <- list("COMPARTMENT_NAMES" = names(data))
- output <- c(output, tempo)
- tempo <- list("COMPARTMENT_TYPE" = sapply(data, FUN = "typeof"))
- if(any(unlist(sapply(data, FUN = "class")) %in% "factor")){ # if an ordered factor is present, then sapply(data, FUN = "class") return a list but works with any(sapply(data, FUN = "class") %in% "factor")
- tempo.class <- sapply(data, FUN = "class")
- if(any(unlist(tempo.class) %in% "ordered")){
- tempo2 <- sapply(tempo.class, paste, collapse = " ") # paste the "ordered" factor" in "ordered factor"
- }else{
- tempo2 <- unlist(tempo.class)
- }
- tempo[["COMPARTMENT_TYPE"]][grepl(x = tempo2, pattern = "factor")] <- tempo2[grepl(x = tempo2, pattern = "factor")]
- }
- output <- c(output, tempo)
- tempo <- list("STRUCTURE" = ls.str(data)) # str() print automatically, ls.str() not but does not give the order of the data.frame
- output <- c(output, tempo)
- }
- if(full == FALSE){
- output[names(output) != "STRUCTURE"] <- lapply(X = output[names(output) != "STRUCTURE"], FUN = head, n = n, simplify = FALSE)
- }
- if(warn.print == FALSE){
- output <- c(output, WARNING = warn)
- }else if(warn.print == TRUE & ! is.null(warn)){
- on.exit(warning(warn, call. = FALSE))
- }
- return(output)
+# AIM
+# provide a complete description of an object. Using complete = TRUE allows data recovering. Using complete = FALSE is a convenient display feature (in case of large dataset)
+# ARGUMENTS
+# data: object to test
+# n: positive integer value indicating the number of element to display per compartment of the output list (i.e., head(..., n)). Ignored if full argument is TRUE. Also ignored for the STRUCTURE compartment output, corresponding to ls.str() information (because head() removes almost everything)
+# full: logical. Return the full information?
+# warn.print: logical. Print potential warnings at the end of the execution? If FALSE the message or NULL (if no message) is added in the output as an additional compartment
+# RETURN
+# a list containing information, depending on the class and type of data
+# if data is made of numerics, provide range, sum, mean, number of NA and number of Inf
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# fun_info(data = 1:3)
+# fun_info(data = 1:3, n = 2)
+# fun_info(data = 1:3, n = 2, full = TRUE)
+# fun_info(data.frame(a = 1:2, b = ordered(factor(c("A", "B"))), stringsAsFactors = TRUE))
+# fun_info(list(a = 1:3, b = ordered(factor(c("A", "B")))))
+# DEBUGGING
+# data = NULL # for function debugging
+# data = 1:3 # for function debugging
+# data = matrix(1:3) # for function debugging
+# data = data.frame(a = 1:2, b = c("A", "B"), stringsAsFactors = TRUE) # for function debugging
+# data = factor(c("b", "a")) # for function debugging
+# data = ordered(factor(c("b", "a"))) # for function debugging
+# data = list(a = 1:3, b = factor(c("A", "B"))) # for function debugging
+# data = list(a = 1:3, b = ordered(factor(c("A", "B")))) # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = n, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & n < 1){
+tempo.cat <- paste0("ERROR IN ", function.name, ": n ARGUMENT MUST BE A POSITIVE AND NON NULL INTEGER")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+tempo <- fun_check(data = full, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = warn.print, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# end argument checking
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) # activate this line and use the function to check arguments status
+# main code
+warn <- NULL
+if(full == FALSE){
+warn <- paste0("FROM ", function.name, ":\n\nSOME COMPARTMENTS CAN BE TRUNCATED (n ARGUMENT IS ", n, ")\n\n")
+}
+data.name <- deparse(substitute(data))
+output <- list("NAME" = data.name)
+tempo <- list("CLASS" = class(data))
+output <- c(output, tempo)
+tempo <- list("TYPE" = typeof(data))
+output <- c(output, tempo)
+tempo <- list("LENGTH" = length(data))
+output <- c(output, tempo)
+if(all(typeof(data) %in% c("integer", "numeric", "double"))){
+tempo <- list("RANGE" = range(data[ ! is.infinite(data)], na.rm = TRUE))
+output <- c(output, tempo)
+tempo <- list("SUM" = sum(data[ ! is.infinite(data)], na.rm = TRUE))
+output <- c(output, tempo)
+tempo <- list("MEAN" = mean(data[ ! is.infinite(data)], na.rm = TRUE))
+output <- c(output, tempo)
+tempo <- list("NA.NB" = sum(is.na(data)))
+output <- c(output, tempo)
+tempo <- list("INF.NB" = sum(is.infinite(data)))
+output <- c(output, tempo)
+}
+tempo <- list("HEAD" = head(data))
+output <- c(output, tempo)
+if( ! is.null(data)){
+tempo <- list("TAIL" = tail(data))
+output <- c(output, tempo)
+if( ! is.null(dim(data))){
+tempo <- list("DIMENSION" = dim(data))
+names(tempo[[1]]) <- c("NROW", "NCOL")
+output <- c(output, tempo)
+}
+tempo <- list("SUMMARY" = summary(data))
+output <- c(output, tempo)
+}
+if(all(class(data) == "data.frame" | all(class(data) %in% c("matrix", "array")))){
+tempo <- list("ROW_NAMES" = dimnames(data)[[1]])
+output <- c(output, tempo)
+tempo <- list("COLUM_NAMES" = dimnames(data)[[2]])
+output <- c(output, tempo)
+}
+if(all(class(data) == "data.frame")){
+tempo <- list("STRUCTURE" = ls.str(data)) # str() print automatically, ls.str() not but does not give the order of the data.frame
+output <- c(output, tempo)
+tempo <- list("COLUMN_TYPE" = sapply(data, FUN = "typeof"))
+if(any(sapply(data, FUN = "class") %in% "factor")){ # if an ordered factor is present, then sapply(data, FUN = "class") return a list but works with any(sapply(data, FUN = "class") %in% "factor")
+tempo.class <- sapply(data, FUN = "class")
+if(any(unlist(tempo.class) %in% "ordered")){
+tempo2 <- sapply(tempo.class, paste, collapse = " ") # paste the "ordered" factor" in "ordered factor"
+}else{
+tempo2 <- unlist(tempo.class)
+}
+tempo[["COLUMN_TYPE"]][grepl(x = tempo2, pattern = "factor")] <- tempo2[grepl(x = tempo2, pattern = "factor")]
+}
+output <- c(output, tempo)
+}
+if(all(class(data) == "list")){
+tempo <- list("COMPARTMENT_NAMES" = names(data))
+output <- c(output, tempo)
+tempo <- list("COMPARTMENT_TYPE" = sapply(data, FUN = "typeof"))
+if(any(unlist(sapply(data, FUN = "class")) %in% "factor")){ # if an ordered factor is present, then sapply(data, FUN = "class") return a list but works with any(sapply(data, FUN = "class") %in% "factor")
+tempo.class <- sapply(data, FUN = "class")
+if(any(unlist(tempo.class) %in% "ordered")){
+tempo2 <- sapply(tempo.class, paste, collapse = " ") # paste the "ordered" factor" in "ordered factor"
+}else{
+tempo2 <- unlist(tempo.class)
+}
+tempo[["COMPARTMENT_TYPE"]][grepl(x = tempo2, pattern = "factor")] <- tempo2[grepl(x = tempo2, pattern = "factor")]
+}
+output <- c(output, tempo)
+tempo <- list("STRUCTURE" = ls.str(data)) # str() print automatically, ls.str() not but does not give the order of the data.frame
+output <- c(output, tempo)
+}
+if(full == FALSE){
+output[names(output) != "STRUCTURE"] <- lapply(X = output[names(output) != "STRUCTURE"], FUN = head, n = n, simplify = FALSE)
+}
+if(warn.print == FALSE){
+output <- c(output, WARNING = warn)
+}else if(warn.print == TRUE & ! is.null(warn)){
+on.exit(warning(warn, call. = FALSE))
+}
+return(output)
}
@@ -621,64 +620,64 @@ fun_info <- function(data, n = 20, full = FALSE, warn.print = TRUE){
# Check OK: clear to go Apollo
fun_head <- function(
- data1,
- n = 6,
- side = "l"
+data1,
+n = 6,
+side = "l"
){
- # AIM
- # as head() but display the left or right head of big 2D objects
- # ARGUMENTS
- # data1: any object but more dedicated for matrix, data frame or table
- # n: as in head() but for for matrix, data frame or table, number of dimension to print (10 means 10 rows and columns)
- # side: either "l" or "r" for the left or right side of the 2D object (only for matrix, data frame or table)
- # BEWARE: other arguments of head() not used
- # RETURN
- # the head
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # obs1 = matrix(1:30, ncol = 5, dimnames = list(letters[1:6], LETTERS[1:5])) ; obs1 ; fun_head(obs1, 3)
- # obs1 = matrix(1:30, ncol = 5, dimnames = list(letters[1:6], LETTERS[1:5])) ; obs1 ; fun_head(obs1, 3, "right")
- # DEBUGGING
- # data1 = matrix(1:30, ncol = 5) # for function debugging
- # data1 = matrix(1:30, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = n, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = side, options = c("l", "r"), length = 1, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- if( ! (any(class(data1) %in% c("data.frame", "table")) | all(class(data1) %in% c("matrix", "array")))){ # before R4.0.0, it was ! any(class(data1) %in% c("matrix", "data.frame", "table"))
- return(head(data1, n))
- }else{
- obs.dim <- dim(data1)
- row <- 1:ifelse(obs.dim[1] < n, obs.dim[1], n)
- if(side == "l"){
- col <- 1:ifelse(obs.dim[2] < n, obs.dim[2], n)
- }
- if(side == "r"){
- col <- ifelse(obs.dim[2] < n, 1, obs.dim[2] - n + 1):obs.dim[2]
- }
- return(data1[row, col])
- }
+# AIM
+# as head() but display the left or right head of big 2D objects
+# ARGUMENTS
+# data1: any object but more dedicated for matrix, data frame or table
+# n: as in head() but for for matrix, data frame or table, number of dimension to print (10 means 10 rows and columns)
+# side: either "l" or "r" for the left or right side of the 2D object (only for matrix, data frame or table)
+# BEWARE: other arguments of head() not used
+# RETURN
+# the head
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# obs1 = matrix(1:30, ncol = 5, dimnames = list(letters[1:6], LETTERS[1:5])) ; obs1 ; fun_head(obs1, 3)
+# obs1 = matrix(1:30, ncol = 5, dimnames = list(letters[1:6], LETTERS[1:5])) ; obs1 ; fun_head(obs1, 3, "right")
+# DEBUGGING
+# data1 = matrix(1:30, ncol = 5) # for function debugging
+# data1 = matrix(1:30, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = n, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = side, options = c("l", "r"), length = 1, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+if( ! (any(class(data1) %in% c("data.frame", "table")) | all(class(data1) %in% c("matrix", "array")))){ # before R4.0.0, it was ! any(class(data1) %in% c("matrix", "data.frame", "table"))
+return(head(data1, n))
+}else{
+obs.dim <- dim(data1)
+row <- 1:ifelse(obs.dim[1] < n, obs.dim[1], n)
+if(side == "l"){
+col <- 1:ifelse(obs.dim[2] < n, obs.dim[2], n)
+}
+if(side == "r"){
+col <- ifelse(obs.dim[2] < n, 1, obs.dim[2] - n + 1):obs.dim[2]
+}
+return(data1[row, col])
+}
}
@@ -687,64 +686,64 @@ fun_head <- function(
# Check OK: clear to go Apollo
fun_tail <- function(
- data1,
- n = 10,
- side = "l"
+data1,
+n = 10,
+side = "l"
){
- # AIM
- # as tail() but display the left or right head of big 2D objects
- # ARGUMENTS
- # data1: any object but more dedicated for matrix, data frame or table
- # n: as in tail() but for for matrix, data frame or table, number of dimension to print (10 means 10 rows and columns)
- # side: either "l" or "r" for the left or right side of the 2D object (only for matrix, data frame or table)
- # BEWARE: other arguments of tail() not used
- # RETURN
- # the tail
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # obs1 = matrix(1:30, ncol = 5, dimnames = list(letters[1:6], LETTERS[1:5])) ; obs1 ; fun_tail(obs1, 3)
- # obs1 = matrix(1:30, ncol = 5, dimnames = list(letters[1:6], LETTERS[1:5])) ; obs1 ; fun_tail(obs1, 3, "r")
- # DEBUGGING
- # data1 = matrix(1:10, ncol = 5) # for function debugging
- # data1 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = n, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = side, options = c("l", "r"), length = 1, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- if( ! (any(class(data1) %in% c("data.frame", "table")) | all(class(data1) %in% c("matrix", "array")))){ # before R4.0.0, it was ! any(class(data1) %in% c("matrix", "data.frame", "table"))
- return(tail(data1, n))
- }else{
- obs.dim <- dim(data1)
- row <- ifelse(obs.dim[1] < n, 1, obs.dim[1] - n + 1):obs.dim[1]
- if(side == "l"){
- col <- 1:ifelse(obs.dim[2] < n, obs.dim[2], n)
- }
- if(side == "r"){
- col <- ifelse(obs.dim[2] < n, 1, obs.dim[2] - n + 1):obs.dim[2]
- }
- return(data1[row, col])
- }
+# AIM
+# as tail() but display the left or right head of big 2D objects
+# ARGUMENTS
+# data1: any object but more dedicated for matrix, data frame or table
+# n: as in tail() but for for matrix, data frame or table, number of dimension to print (10 means 10 rows and columns)
+# side: either "l" or "r" for the left or right side of the 2D object (only for matrix, data frame or table)
+# BEWARE: other arguments of tail() not used
+# RETURN
+# the tail
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# obs1 = matrix(1:30, ncol = 5, dimnames = list(letters[1:6], LETTERS[1:5])) ; obs1 ; fun_tail(obs1, 3)
+# obs1 = matrix(1:30, ncol = 5, dimnames = list(letters[1:6], LETTERS[1:5])) ; obs1 ; fun_tail(obs1, 3, "r")
+# DEBUGGING
+# data1 = matrix(1:10, ncol = 5) # for function debugging
+# data1 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = n, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = side, options = c("l", "r"), length = 1, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+if( ! (any(class(data1) %in% c("data.frame", "table")) | all(class(data1) %in% c("matrix", "array")))){ # before R4.0.0, it was ! any(class(data1) %in% c("matrix", "data.frame", "table"))
+return(tail(data1, n))
+}else{
+obs.dim <- dim(data1)
+row <- ifelse(obs.dim[1] < n, 1, obs.dim[1] - n + 1):obs.dim[1]
+if(side == "l"){
+col <- 1:ifelse(obs.dim[2] < n, obs.dim[2], n)
+}
+if(side == "r"){
+col <- ifelse(obs.dim[2] < n, 1, obs.dim[2] - n + 1):obs.dim[2]
+}
+return(data1[row, col])
+}
}
@@ -753,217 +752,217 @@ fun_tail <- function(
# Check OK: clear to go Apollo
fun_comp_1d <- function(data1, data2){
- # AIM
- # compare two 1D datasets (vector or factor or 1D table, or 1D matrix or 1D array) of the same class or not. Check and report in a list if the 2 datasets have:
- # same class
- # common elements
- # common element names (except factors)
- # common levels (factors only)
- # ARGUMENTS
- # data1: vector or factor or 1D table, or 1D matrix or 1D array
- # data2: vector or factor or 1D table, or 1D matrix or 1D array
- # RETURN
- # a list containing:
- # $same.class: logical. Are class identical?
- # $class: class of the 2 datasets (NULL otherwise)
- # $same.length: logical. Are number of elements identical?
- # $length: number of elements in the 2 datasets (NULL otherwise)
- # $same.levels: logical. Are levels identical? NULL if data1 and data2 are not factors
- # $levels: levels of the 2 datasets if identical (NULL otherwise or NULL if data1 and data2 are not factors)
- # $any.id.levels: logical. Is there any identical levels? (NULL if data1 and data2 are not factors)
- # $same.levels.pos1: position, in data1, of the levels identical in data2 (NULL if data1 and data2 are not factors)
- # $same.levels.pos2: position, in data2, of the levels identical in data1 (NULL if data1 and data2 are not factors)
- # $common.levels: common levels between data1 and data2 (can be a subset of $levels or not). NULL if no common levels or if data1 and data2 are not factors
- # $same.name: logical. Are element names identical? NULL if data1 and data2 have no names
- # $name: name of elements of the 2 datasets if identical (NULL otherwise)
- # $any.id.name: logical. Is there any element names identical ?
- # $same.name.pos1: position, in data1, of the element names identical in data2. NULL if no identical names
- # $same.name.pos2: position, in data2, of the elements names identical in data1. NULL if no identical names
- # $common.names: common element names between data1 and data2 (can be a subset of $name or not). NULL if no common element names
- # $any.id.element: logical. is there any identical elements ?
- # $same.element.pos1: position, in data1, of the elements identical in data2. NULL if no identical elements
- # $same.element.pos2: position, in data2, of the elements identical in data1. NULL if no identical elements
- # $common.elements: common elements between data1 and data2. NULL if no common elements
- # $same.order: logical. Are all elements in the same order? TRUE or FALSE if elements of data1 and data2 are identical but not necessary in the same order. NULL otherwise (different length for instance)
- # $order1: order of all elements of data1. NULL if $same.order is FALSE
- # $order2: order of all elements of data2. NULL if $same.order is FALSE
- # $identical.object: logical. Are objects identical (kind of object, element names, content, including content order)?
- # $identical.content: logical. Are content objects identical (identical elements, including order, excluding kind of object and element names)?
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # none
- # EXAMPLES
- # obs1 = 1:5 ; obs2 = 1:5 ; names(obs1) <- LETTERS[1:5] ; names(obs2) <- LETTERS[1:5] ; fun_comp_1d(obs1, obs2)
- # obs1 = 1:5 ; obs2 = 1:5 ; names(obs1) <- LETTERS[1:5] ; fun_comp_1d(obs1, obs2)
- # obs1 = 1:5 ; obs2 = 3:6 ; names(obs1) <- LETTERS[1:5] ; names(obs2) <- LETTERS[1:4] ; fun_comp_1d(obs1, obs2)
- # obs1 = factor(LETTERS[1:5]) ; obs2 = factor(LETTERS[1:5]) ; fun_comp_1d(obs1, obs2)
- # obs1 = factor(LETTERS[1:5]) ; obs2 = factor(LETTERS[10:11]) ; fun_comp_1d(obs1, obs2)
- # obs1 = factor(LETTERS[1:5]) ; obs2 = factor(LETTERS[4:7]) ; fun_comp_1d(obs1, obs2)
- # obs1 = factor(c(LETTERS[1:4], "E")) ; obs2 = factor(c(LETTERS[1:4], "F")) ; fun_comp_1d(obs1, obs2)
- # obs1 = 1:5 ; obs2 = factor(LETTERS[1:5]) ; fun_comp_1d(obs1, obs2)
- # obs1 = 1:5 ; obs2 = 1.1:6.1 ; fun_comp_1d(obs1, obs2)
- # obs1 = as.table(1:5); obs2 = as.table(1:5) ; fun_comp_1d(obs1, obs2)
- # obs1 = as.table(1:5); obs2 = 1:5 ; fun_comp_1d(obs1, obs2)
- # DEBUGGING
- # data1 = 1:5 ; data2 = 1:5 ; names(data1) <- LETTERS[1:5] ; names(data2) <- LETTERS[1:5] # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # argument checking
- if( ! any(class(data1) %in% c("logical", "integer", "numeric", "character", "factor", "table"))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 ARGUMENT MUST BE A NON NULL VECTOR, FACTOR OR 1D TABLE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else if(all(class(data1) %in% "table")){
- if(length(dim(data1)) > 1){
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 ARGUMENT MUST BE A 1D TABLE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- if( ! any(class(data2) %in% c("logical", "integer", "numeric", "character", "factor", "table"))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE data2 ARGUMENT MUST BE A NON NULL VECTOR, FACTOR OR 1D TABLE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else if(all(class(data2) %in% "table")){
- if(length(dim(data2)) > 1){
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE data2 ARGUMENT MUST BE A 1D TABLE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) # activate this line and use the function to check arguments status
- # end argument checking
- # main code
- same.class <- FALSE
- class <- NULL
- same.length <- FALSE
- length <- NULL
- same.levels <- NULL # not FALSE to deal with no factors
- levels <- NULL
- any.id.levels <- NULL
- same.levels.pos1 <- NULL
- same.levels.pos2 <- NULL
- common.levels <- NULL
- same.name <- NULL # not FALSE to deal with absence of name
- name <- NULL
- any.id.name <- FALSE
- same.name.pos1 <- NULL
- same.name.pos2 <- NULL
- common.names <- NULL
- any.id.element <- FALSE
- same.element.pos1 <- NULL
- same.element.pos2 <- NULL
- common.elements <- NULL
- same.order <- NULL
- order1 <- NULL
- order2 <- NULL
- identical.object <- FALSE
- identical.content <- FALSE
- if(identical(data1, data2)){
- same.class <- TRUE
- class <- class(data1)
- same.length <- TRUE
- length <- length(data1)
- if(any(class(data1) %in% "factor")){
- same.levels <- TRUE
- levels <- levels(data1)
- any.id.levels <- TRUE
- same.levels.pos1 <- 1:length(levels(data1))
- same.levels.pos2 <- 1:length(levels(data2))
- common.levels <- levels(data1)
- }
- if( ! is.null(names(data1))){
- same.name <- TRUE
- name <- names(data1)
- any.id.name <- TRUE
- same.name.pos1 <- 1:length(data1)
- same.name.pos2 <- 1:length(data2)
- common.names <- names(data1)
- }
- any.id.element <- TRUE
- same.element.pos1 <- 1:length(data1)
- same.element.pos2 <- 1:length(data2)
- common.elements <- data1
- same.order <- TRUE
- order1 <- order(data1)
- order2 <- order(data2)
- identical.object <- TRUE
- identical.content <- TRUE
- }else{
- if(identical(class(data1), class(data2))){
- same.class <- TRUE
- class <- class(data1)
- }
- if(identical(length(data1), length(data2))){
- same.length<- TRUE
- length <- length(data1)
- }
- if(any(class(data1) %in% "factor") & any(class(data2) %in% "factor")){
- if(identical(levels(data1), levels(data2))){
- same.levels <- TRUE
- levels <- levels(data1)
- }else{
- same.levels <- FALSE
- }
- if(any(levels(data1) %in% levels(data2))){
- any.id.levels <- TRUE
- same.levels.pos1 <- which(levels(data1) %in% levels(data2))
- }
- if(any(levels(data2) %in% levels(data1))){
- any.id.levels <- TRUE
- same.levels.pos2 <- which(levels(data2) %in% levels(data1))
- }
- if(any.id.levels == TRUE){
- common.levels <- unique(c(levels(data1)[same.levels.pos1], levels(data2)[same.levels.pos2]))
- }
- }
- if(any(class(data1) %in% "factor")){ # to compare content
- data1 <- as.character(data1)
- }
- if(any(class(data2) %in% "factor")){ # to compare content
- data2 <- as.character(data2)
- }
- if( ! (is.null(names(data1)) & is.null(names(data2)))){
- if(identical(names(data1), names(data2))){
- same.name <- TRUE
- name <- names(data1)
- }else{
- same.name <- FALSE
- }
- if(any(names(data1) %in% names(data2))){
- any.id.name <- TRUE
- same.name.pos1 <- which(names(data1) %in% names(data2))
- }
- if(any(names(data2) %in% names(data1))){
- any.id.name <- TRUE
- same.name.pos2 <- which(names(data2) %in% names(data1))
- }
- if(any.id.name == TRUE){
- common.names <- unique(c(names(data1)[same.name.pos1], names(data2)[same.name.pos2]))
- }
- }
- names(data1) <- NULL # names solved -> to do not be disturbed by names
- names(data2) <- NULL # names solved -> to do not be disturbed by names
- if(any(data1 %in% data2)){
- any.id.element <- TRUE
- same.element.pos1 <- which(data1 %in% data2)
- }
- if(any(data2 %in% data1)){
- any.id.element <- TRUE
- same.element.pos2 <- which(data2 %in% data1)
- }
- if(any.id.element == TRUE){
- common.elements <- unique(c(data1[same.element.pos1], data2[same.element.pos2]))
- }
- if(identical(data1, data2)){
- identical.content <- TRUE
- same.order <- TRUE
- }else if(identical(sort(data1), sort(data2))){
- same.order <- FALSE
- order1 <- order(data1)
- order2 <- order(data2)
- }
- }
- output <- list(same.class = same.class, class = class, same.length = same.length, length = length, same.levels = same.levels, levels = levels, any.id.levels = any.id.levels, same.levels.pos1 = same.levels.pos1, same.levels.pos2 = same.levels.pos2, common.levels = common.levels, same.name = same.name, name = name, any.id.name = any.id.name, same.name.pos1 = same.name.pos1, same.name.pos2 = same.name.pos2, common.names = common.names, any.id.element = any.id.element, same.element.pos1 = same.element.pos1, same.element.pos2 = same.element.pos2, common.elements = common.elements, same.order = same.order, order1 = order1, order2 = order2, identical.object = identical.object, identical.content = identical.content)
- return(output)
+# AIM
+# compare two 1D datasets (vector or factor or 1D table, or 1D matrix or 1D array) of the same class or not. Check and report in a list if the 2 datasets have:
+# same class
+# common elements
+# common element names (except factors)
+# common levels (factors only)
+# ARGUMENTS
+# data1: vector or factor or 1D table, or 1D matrix or 1D array
+# data2: vector or factor or 1D table, or 1D matrix or 1D array
+# RETURN
+# a list containing:
+# $same.class: logical. Are class identical?
+# $class: class of the 2 datasets (NULL otherwise)
+# $same.length: logical. Are number of elements identical?
+# $length: number of elements in the 2 datasets (NULL otherwise)
+# $same.levels: logical. Are levels identical? NULL if data1 and data2 are not factors
+# $levels: levels of the 2 datasets if identical (NULL otherwise or NULL if data1 and data2 are not factors)
+# $any.id.levels: logical. Is there any identical levels? (NULL if data1 and data2 are not factors)
+# $same.levels.pos1: position, in data1, of the levels identical in data2 (NULL if data1 and data2 are not factors)
+# $same.levels.pos2: position, in data2, of the levels identical in data1 (NULL if data1 and data2 are not factors)
+# $common.levels: common levels between data1 and data2 (can be a subset of $levels or not). NULL if no common levels or if data1 and data2 are not factors
+# $same.name: logical. Are element names identical? NULL if data1 and data2 have no names
+# $name: name of elements of the 2 datasets if identical (NULL otherwise)
+# $any.id.name: logical. Is there any element names identical ?
+# $same.name.pos1: position, in data1, of the element names identical in data2. NULL if no identical names
+# $same.name.pos2: position, in data2, of the elements names identical in data1. NULL if no identical names
+# $common.names: common element names between data1 and data2 (can be a subset of $name or not). NULL if no common element names
+# $any.id.element: logical. is there any identical elements ?
+# $same.element.pos1: position, in data1, of the elements identical in data2. NULL if no identical elements
+# $same.element.pos2: position, in data2, of the elements identical in data1. NULL if no identical elements
+# $common.elements: common elements between data1 and data2. NULL if no common elements
+# $same.order: logical. Are all elements in the same order? TRUE or FALSE if elements of data1 and data2 are identical but not necessary in the same order. NULL otherwise (different length for instance)
+# $order1: order of all elements of data1. NULL if $same.order is FALSE
+# $order2: order of all elements of data2. NULL if $same.order is FALSE
+# $identical.object: logical. Are objects identical (kind of object, element names, content, including content order)?
+# $identical.content: logical. Are content objects identical (identical elements, including order, excluding kind of object and element names)?
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# none
+# EXAMPLES
+# obs1 = 1:5 ; obs2 = 1:5 ; names(obs1) <- LETTERS[1:5] ; names(obs2) <- LETTERS[1:5] ; fun_comp_1d(obs1, obs2)
+# obs1 = 1:5 ; obs2 = 1:5 ; names(obs1) <- LETTERS[1:5] ; fun_comp_1d(obs1, obs2)
+# obs1 = 1:5 ; obs2 = 3:6 ; names(obs1) <- LETTERS[1:5] ; names(obs2) <- LETTERS[1:4] ; fun_comp_1d(obs1, obs2)
+# obs1 = factor(LETTERS[1:5]) ; obs2 = factor(LETTERS[1:5]) ; fun_comp_1d(obs1, obs2)
+# obs1 = factor(LETTERS[1:5]) ; obs2 = factor(LETTERS[10:11]) ; fun_comp_1d(obs1, obs2)
+# obs1 = factor(LETTERS[1:5]) ; obs2 = factor(LETTERS[4:7]) ; fun_comp_1d(obs1, obs2)
+# obs1 = factor(c(LETTERS[1:4], "E")) ; obs2 = factor(c(LETTERS[1:4], "F")) ; fun_comp_1d(obs1, obs2)
+# obs1 = 1:5 ; obs2 = factor(LETTERS[1:5]) ; fun_comp_1d(obs1, obs2)
+# obs1 = 1:5 ; obs2 = 1.1:6.1 ; fun_comp_1d(obs1, obs2)
+# obs1 = as.table(1:5); obs2 = as.table(1:5) ; fun_comp_1d(obs1, obs2)
+# obs1 = as.table(1:5); obs2 = 1:5 ; fun_comp_1d(obs1, obs2)
+# DEBUGGING
+# data1 = 1:5 ; data2 = 1:5 ; names(data1) <- LETTERS[1:5] ; names(data2) <- LETTERS[1:5] # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# argument checking
+if( ! any(class(data1) %in% c("logical", "integer", "numeric", "character", "factor", "table"))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 ARGUMENT MUST BE A NON NULL VECTOR, FACTOR OR 1D TABLE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else if(all(class(data1) %in% "table")){
+if(length(dim(data1)) > 1){
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 ARGUMENT MUST BE A 1D TABLE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+if( ! any(class(data2) %in% c("logical", "integer", "numeric", "character", "factor", "table"))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE data2 ARGUMENT MUST BE A NON NULL VECTOR, FACTOR OR 1D TABLE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else if(all(class(data2) %in% "table")){
+if(length(dim(data2)) > 1){
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE data2 ARGUMENT MUST BE A 1D TABLE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) # activate this line and use the function to check arguments status
+# end argument checking
+# main code
+same.class <- FALSE
+class <- NULL
+same.length <- FALSE
+length <- NULL
+same.levels <- NULL # not FALSE to deal with no factors
+levels <- NULL
+any.id.levels <- NULL
+same.levels.pos1 <- NULL
+same.levels.pos2 <- NULL
+common.levels <- NULL
+same.name <- NULL # not FALSE to deal with absence of name
+name <- NULL
+any.id.name <- FALSE
+same.name.pos1 <- NULL
+same.name.pos2 <- NULL
+common.names <- NULL
+any.id.element <- FALSE
+same.element.pos1 <- NULL
+same.element.pos2 <- NULL
+common.elements <- NULL
+same.order <- NULL
+order1 <- NULL
+order2 <- NULL
+identical.object <- FALSE
+identical.content <- FALSE
+if(identical(data1, data2)){
+same.class <- TRUE
+class <- class(data1)
+same.length <- TRUE
+length <- length(data1)
+if(any(class(data1) %in% "factor")){
+same.levels <- TRUE
+levels <- levels(data1)
+any.id.levels <- TRUE
+same.levels.pos1 <- 1:length(levels(data1))
+same.levels.pos2 <- 1:length(levels(data2))
+common.levels <- levels(data1)
+}
+if( ! is.null(names(data1))){
+same.name <- TRUE
+name <- names(data1)
+any.id.name <- TRUE
+same.name.pos1 <- 1:length(data1)
+same.name.pos2 <- 1:length(data2)
+common.names <- names(data1)
+}
+any.id.element <- TRUE
+same.element.pos1 <- 1:length(data1)
+same.element.pos2 <- 1:length(data2)
+common.elements <- data1
+same.order <- TRUE
+order1 <- order(data1)
+order2 <- order(data2)
+identical.object <- TRUE
+identical.content <- TRUE
+}else{
+if(identical(class(data1), class(data2))){
+same.class <- TRUE
+class <- class(data1)
+}
+if(identical(length(data1), length(data2))){
+same.length<- TRUE
+length <- length(data1)
+}
+if(any(class(data1) %in% "factor") & any(class(data2) %in% "factor")){
+if(identical(levels(data1), levels(data2))){
+same.levels <- TRUE
+levels <- levels(data1)
+}else{
+same.levels <- FALSE
+}
+if(any(levels(data1) %in% levels(data2))){
+any.id.levels <- TRUE
+same.levels.pos1 <- which(levels(data1) %in% levels(data2))
+}
+if(any(levels(data2) %in% levels(data1))){
+any.id.levels <- TRUE
+same.levels.pos2 <- which(levels(data2) %in% levels(data1))
+}
+if(any.id.levels == TRUE){
+common.levels <- unique(c(levels(data1)[same.levels.pos1], levels(data2)[same.levels.pos2]))
+}
+}
+if(any(class(data1) %in% "factor")){ # to compare content
+data1 <- as.character(data1)
+}
+if(any(class(data2) %in% "factor")){ # to compare content
+data2 <- as.character(data2)
+}
+if( ! (is.null(names(data1)) & is.null(names(data2)))){
+if(identical(names(data1), names(data2))){
+same.name <- TRUE
+name <- names(data1)
+}else{
+same.name <- FALSE
+}
+if(any(names(data1) %in% names(data2))){
+any.id.name <- TRUE
+same.name.pos1 <- which(names(data1) %in% names(data2))
+}
+if(any(names(data2) %in% names(data1))){
+any.id.name <- TRUE
+same.name.pos2 <- which(names(data2) %in% names(data1))
+}
+if(any.id.name == TRUE){
+common.names <- unique(c(names(data1)[same.name.pos1], names(data2)[same.name.pos2]))
+}
+}
+names(data1) <- NULL # names solved -> to do not be disturbed by names
+names(data2) <- NULL # names solved -> to do not be disturbed by names
+if(any(data1 %in% data2)){
+any.id.element <- TRUE
+same.element.pos1 <- which(data1 %in% data2)
+}
+if(any(data2 %in% data1)){
+any.id.element <- TRUE
+same.element.pos2 <- which(data2 %in% data1)
+}
+if(any.id.element == TRUE){
+common.elements <- unique(c(data1[same.element.pos1], data2[same.element.pos2]))
+}
+if(identical(data1, data2)){
+identical.content <- TRUE
+same.order <- TRUE
+}else if(identical(sort(data1), sort(data2))){
+same.order <- FALSE
+order1 <- order(data1)
+order2 <- order(data2)
+}
+}
+output <- list(same.class = same.class, class = class, same.length = same.length, length = length, same.levels = same.levels, levels = levels, any.id.levels = any.id.levels, same.levels.pos1 = same.levels.pos1, same.levels.pos2 = same.levels.pos2, common.levels = common.levels, same.name = same.name, name = name, any.id.name = any.id.name, same.name.pos1 = same.name.pos1, same.name.pos2 = same.name.pos2, common.names = common.names, any.id.element = any.id.element, same.element.pos1 = same.element.pos1, same.element.pos2 = same.element.pos2, common.elements = common.elements, same.order = same.order, order1 = order1, order2 = order2, identical.object = identical.object, identical.content = identical.content)
+return(output)
}
@@ -972,371 +971,371 @@ fun_comp_1d <- function(data1, data2){
# Check OK: clear to go Apollo
fun_comp_2d <- function(data1, data2){
- # AIM
- # compare two 2D datasets of the same class or not. Check and report in a list if the 2 datasets have:
- # same class
- # common row names
- # common column names
- # same row number
- # same column number
- # potential identical rows between the 2 datasets
- # potential identical columns between the 2 datasets
- # ARGUMENTS
- # data1: matrix, data frame or table
- # data2: matrix, data frame or table
- # RETURN
- # a list containing:
- # $same.class: logical. Are class identical ?
- # $class: classes of the 2 datasets (NULL otherwise)
- # $same.dim: logical. Are dimension identical ?
- # $dim: dimension of the 2 datasets (NULL otherwise)
- # $same.row.nb: logical. Are number of rows identical ?
- # $row.nb: nb of rows of the 2 datasets if identical (NULL otherwise)
- # $same.col.nb: logical. Are number of columns identical ?
- # $col.nb: nb of columns of the 2 datasets if identical (NULL otherwise)
- # $same.row.name: logical. Are row names identical ? NULL if no row names in the two 2D datasets
- # $row.name: name of rows of the 2 datasets if identical (NULL otherwise)
- # $any.id.row.name: logical. Is there any row names identical ? NULL if no row names in the two 2D datasets
- # $same.row.name.pos1: position, in data1, of the row names identical in data2
- # $same.row.name.pos2: position, in data2, of the row names identical in data1
- # $common.row.names: common row names between data1 and data2 (can be a subset of $name or not). NULL if no common row names
- # $same.col.name: logical. Are column names identical ? NULL if no col names in the two 2D datasets
- # $col.name: name of columns of the 2 datasets if identical (NULL otherwise)
- # $any.id.col.name: logical. Is there any column names identical ? NULL if no col names in the two 2D datasets
- # $same.col.name.pos1: position, in data1, of the column names identical in data2
- # $same.col.name.pos2: position, in data2, of the column names identical in data1
- # $common.col.names: common column names between data1 and data2 (can be a subset of $name or not). NULL if no common column names
- # $any.id.row: logical. is there identical rows (not considering row names)? NULL if nrow(data1) * nrow(data2) > 1e10
- # $same.row.pos1: position, in data1, of the rows identical in data2 (not considering row names). Return "TOO BIG FOR EVALUATION" if nrow(data1) * nrow(data2) > 1e10
- # $same.row.pos2: position, in data2, of the rows identical in data1 (not considering row names). Return "TOO BIG FOR EVALUATION" if nrow(data1) * nrow(data2) > 1e10
- # $any.id.col: logical. is there identical columns (not considering column names)? NULL if ncol(data1) * ncol(data2) > 1e10
- # $same.col.pos1: position in data1 of the cols identical in data2 (not considering column names). Return "TOO BIG FOR EVALUATION" if ncol(data1) * ncol(data2) > 1e10
- # $same.col.pos2: position in data2 of the cols identical in data1 (not considering column names). Return "TOO BIG FOR EVALUATION" if ncol(data1) * ncol(data2) > 1e10
- # $identical.object: logical. Are objects identical (including row & column names)?
- # $identical.content: logical. Are content objects identical (identical excluding row & column names)?
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # none
- # EXAMPLES
- # obs1 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; obs2 = as.data.frame(matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])), stringsAsFactors = TRUE) ; obs1 ; obs2 ; fun_comp_2d(obs1, obs2)
- # obs1 = matrix(101:110, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; obs2 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; obs1 ; obs2 ; fun_comp_2d(obs1, obs2)
- # large matrices
- # obs1 = matrix(1:1e6, ncol = 5, dimnames = list(NULL, LETTERS[1:5])) ; obs2 = matrix(as.integer((1:1e6)+1e6/5), ncol = 5, dimnames = list(NULL, LETTERS[1:5])) ; head(obs1) ; head(obs2) ; fun_comp_2d(obs1, obs2)
- # WARNING: when comparing content (rows, columns, or total), double and integer data are considered as different -> double(1) != integer(1)
- # obs1 = matrix(1:1e6, ncol = 5, dimnames = list(NULL, LETTERS[1:5])) ; obs2 = matrix((1:1e6)+1e6/5, ncol = 5, dimnames = list(NULL, LETTERS[1:5])) ; head(obs1) ; head(obs2) ; fun_comp_2d(obs1, obs2)
- # obs1 = matrix(1:10, byrow = TRUE, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; obs2 = matrix(c(1:5, 101:105, 6:10), byrow = TRUE, ncol = 5, dimnames = list(c("a", "z", "b"), c(LETTERS[1:2], "k", LETTERS[5:4]))) ; obs1 ; obs2 ; fun_comp_2d(obs1, obs2)
- # obs1 = t(matrix(1:10, byrow = TRUE, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5]))) ; obs2 = t(matrix(c(1:5, 101:105, 6:10), byrow = TRUE, ncol = 5, dimnames = list(c("a", "z", "b"), c(LETTERS[1:2], "k", LETTERS[5:4])))) ; obs1 ; obs2 ; fun_comp_2d(obs1, obs2)
- # DEBUGGING
- # data1 = matrix(1:10, ncol = 5) ; data2 = matrix(1:10, ncol = 5) # for function debugging
- # data1 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; data2 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
- # data1 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; data2 = matrix(1:10, ncol = 5) # for function debugging
- # data1 = matrix(1:15, byrow = TRUE, ncol = 5, dimnames = list(letters[1:3], LETTERS[1:5])) ; data2 = matrix(1:10, byrow = TRUE, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
- # data1 = matrix(1:15, ncol = 5, dimnames = list(letters[1:3], LETTERS[1:5])) ; data2 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
- # data1 = matrix(1:15, ncol = 5, dimnames = list(paste0("A", letters[1:3]), LETTERS[1:5])) ; data2 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
- # data1 = matrix(1:15, ncol = 5, dimnames = list(letters[1:3], LETTERS[1:5])) ; data2 = matrix(1:12, ncol = 4, dimnames = list(letters[1:3], LETTERS[1:4])) # for function debugging
- # data1 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; data2 = matrix(101:110, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
- # data1 = data.frame(a = 1:3, b= letters[1:3], row.names = LETTERS[1:3], stringsAsFactors = TRUE) ; data2 = data.frame(A = 1:3, B= letters[1:3], stringsAsFactors = TRUE) # for function debugging
- # data1 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; data2 = as.data.frame(matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])), stringsAsFactors = TRUE) # for function debugging
- # data1 = matrix(1:10, byrow = TRUE, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; data2 = matrix(c(1:5, 101:105, 6:10), byrow = TRUE, ncol = 5, dimnames = list(c("a", "z", "b"), c(LETTERS[1:2], "k", LETTERS[5:4]))) # for function debugging
- # data1 = table(Exp1 = c("A", "A", "A", "B", "B", "B"), Exp2 = c("A1", "B1", "A1", "C1", "C1", "B1")) ; data2 = data.frame(A = 1:3, B= letters[1:3], stringsAsFactors = TRUE) # for function debugging
- # data1 = matrix(1:1e6, ncol = 5, dimnames = list(NULL, LETTERS[1:5])) ; data2 = matrix((1:1e6)+1e6/5, ncol = 5, dimnames = list(NULL, LETTERS[1:5]))
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # argument checking
- if( ! (any(class(data1) %in% c("data.frame", "table")) | all(class(data1) %in% c("matrix", "array")))){ # before R4.0.0, it was ! any(class(data1) %in% c("matrix", "data.frame", "table"))
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 ARGUMENT MUST BE A MATRIX, DATA FRAME OR TABLE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! (any(class(data2) %in% c("data.frame", "table")) | all(class(data2) %in% c("matrix", "array")))){ # before R4.0.0, it was ! any(class(data2) %in% c("matrix", "data.frame", "table"))
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE data2 ARGUMENT MUST BE A MATRIX, DATA FRAME OR TABLE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) # activate this line and use the function to check arguments status
- # end argument checking
- # main code
- same.class <- NULL
- class <- NULL
- same.dim <- NULL
- dim <- NULL
- same.row.nb <- NULL
- row.nb <- NULL
- same.col.nb <- NULL
- col.nb <- NULL
- same.row.name <- NULL
- row.name <- NULL
- any.id.row.name <- NULL
- same.row.name.pos1 <- NULL
- same.row.name.pos2 <- NULL
- common.row.names <- NULL
- same.col.name <- NULL
- any.id.col.name <- NULL
- same.col.name.pos1 <- NULL
- same.col.name.pos2 <- NULL
- common.col.names <- NULL
- col.name <- NULL
- any.id.row <- NULL
- same.row.pos1 <- NULL
- same.row.pos2 <- NULL
- any.id.col <- NULL
- same.col.pos1 <- NULL
- same.col.pos2 <- NULL
- identical.object <- NULL
- identical.content <- NULL
- if(identical(data1, data2) & (any(class(data1) %in% c("data.frame", "table")) | all(class(data1) %in% c("matrix", "array")))){ # before R4.0.0, it was ! any(class(data1) %in% c("matrix", "data.frame", "table"))
- same.class <- TRUE
- class <- class(data1)
- same.dim <- TRUE
- dim <- dim(data1)
- same.row.nb <- TRUE
- row.nb <- nrow(data1)
- same.col.nb <- TRUE
- col.nb <- ncol(data1)
- same.row.name <- TRUE
- row.name <- dimnames(data1)[[1]]
- any.id.row.name <- TRUE
- same.row.name.pos1 <- 1:row.nb
- same.row.name.pos2 <- 1:row.nb
- common.row.names <- dimnames(data1)[[1]]
- same.col.name <- TRUE
- col.name <- dimnames(data1)[[2]]
- any.id.col.name <- TRUE
- same.col.name.pos1 <- 1:col.nb
- same.col.name.pos2 <- 1:col.nb
- common.col.names <- dimnames(data1)[[2]]
- any.id.row <- TRUE
- same.row.pos1 <- 1:row.nb
- same.row.pos2 <- 1:row.nb
- any.id.col <- TRUE
- same.col.pos1 <- 1:col.nb
- same.col.pos2 <- 1:col.nb
- identical.object <- TRUE
- identical.content <- TRUE
- }else{
- identical.object <- FALSE
- if(all(class(data1) == "table") & length(dim(data1)) == 1){
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 ARGUMENT IS A 1D TABLE. USE THE fun_comp_1d FUNCTION")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(all(class(data2) == "table") & length(dim(data2)) == 1){
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE data2 ARGUMENT IS A 1D TABLE. USE THE fun_comp_1d FUNCTION")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! identical(class(data1), class(data2))){
- same.class <- FALSE
- }else if( ! (any(class(data1) %in% c("data.frame", "table")) | all(class(data1) %in% c("matrix", "array")))){ # before R4.0.0, it was ! any(class(data1) %in% c("matrix", "data.frame", "table"))
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 AND data2 ARGUMENTS MUST BE EITHER MATRIX, DATA FRAME OR TABLE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- same.class <- TRUE
- class <- class(data1)
- }
- if( ! identical(dim(data1), dim(data2))){
- same.dim <- FALSE
- }else{
- same.dim <- TRUE
- dim <- dim(data1)
- }
- if( ! identical(nrow(data1), nrow(data2))){
- same.row.nb <- FALSE
- }else{
- same.row.nb <- TRUE
- row.nb <- nrow(data1)
- }
- if( ! identical(ncol(data1), ncol(data2))){
- same.col.nb <- FALSE
- }else{
- same.col.nb <- TRUE
- col.nb <- ncol(data1)
- }
- # row and col names
- if(is.null(dimnames(data1)) & is.null(dimnames(data2))){
- same.row.name <- NULL
- same.col.name <- NULL
- # row and col names remain NULL
- }else if((is.null(dimnames(data1)) & ! is.null(dimnames(data2))) | ( ! is.null(dimnames(data1)) & is.null(dimnames(data2)))){
- same.row.name <- FALSE
- same.col.name <- FALSE
- # row and col names remain NULL
- }else{
- if( ! identical(dimnames(data1)[[1]], dimnames(data2)[[1]])){
- same.row.name <- FALSE
- # row names remain NULL
- }else{
- same.row.name <- TRUE
- row.name <- dimnames(data1)[[1]]
- }
- # row names
- any.id.row.name <- FALSE
- if(any(dimnames(data1)[[1]] %in% dimnames(data2)[[1]])){
- any.id.row.name <- TRUE
- same.row.name.pos1 <- which(dimnames(data1)[[1]] %in% dimnames(data2)[[1]])
- }
- if(any(dimnames(data2)[[1]] %in% dimnames(data1)[[1]])){
- any.id.row.name <- TRUE
- same.row.name.pos2 <- which(dimnames(data2)[[1]] %in% dimnames(data1)[[1]])
- }
- if(any.id.row.name == TRUE){
- common.row.names <- unique(c(dimnames(data1)[[1]][same.row.name.pos1], dimnames(data2)[[1]][same.row.name.pos2]))
- }
- # col names
- any.id.col.name <- FALSE
- if(any(dimnames(data1)[[2]] %in% dimnames(data2)[[2]])){
- any.id.col.name <- TRUE
- same.col.name.pos1 <- which(dimnames(data1)[[2]] %in% dimnames(data2)[[2]])
- }
- if(any(dimnames(data2)[[2]] %in% dimnames(data1)[[2]])){
- any.id.col.name <- TRUE
- same.col.name.pos2 <- which(dimnames(data2)[[2]] %in% dimnames(data1)[[2]])
- }
- if(any.id.col.name == TRUE){
- common.col.names <- unique(c(dimnames(data1)[[2]][same.col.name.pos1], dimnames(data2)[[2]][same.col.name.pos2]))
- }
- if( ! identical(dimnames(data1)[[2]], dimnames(data2)[[2]])){
- same.col.name <- FALSE
- # col names remain NULL
- }else{
- same.col.name <- TRUE
- col.name <- dimnames(data1)[[2]]
- }
- }
- # identical row and col content
- if(all(class(data1) == "table")){
- as.data.frame(matrix(data1, ncol = ncol(data1)), stringsAsFactors = FALSE)
- }else if(all(class(data1) %in% c("matrix", "array"))){
- data1 <- as.data.frame(data1, stringsAsFactors = FALSE)
- }else if(all(class(data1) == "data.frame")){
- data1 <- data.frame(lapply(data1, as.character), stringsAsFactors = FALSE)
- }
- if(all(class(data2) == "table")){
- as.data.frame(matrix(data2, ncol = ncol(data2)), stringsAsFactors = FALSE)
- }else if(all(class(data2) %in% c("matrix", "array"))){
- data2 <- as.data.frame(data2, stringsAsFactors = FALSE)
- }else if(all(class(data2) == "data.frame")){
- data2 <- data.frame(lapply(data2, as.character), stringsAsFactors = FALSE)
- }
- row.names(data1) <- paste0("A", 1:nrow(data1))
- row.names(data2) <- paste0("A", 1:nrow(data2))
- if(same.col.nb == TRUE){ # because if not the same col nb, the row cannot be identical
- if(all(sapply(data1, FUN = typeof) == "integer") & all(sapply(data2, FUN = typeof) == "integer") & as.double(nrow(data1)) * nrow(data2) <= 1e10){ # as.double(nrow(data1)) to prevent integer overflow because R is 32 bits for integers
- same.row.pos1 <- which(c(as.data.frame(t(data1), stringsAsFactors = FALSE)) %in% c(as.data.frame(t(data2), stringsAsFactors = FALSE))) # this work fast with only integers (because 32 bits)
- same.row.pos2 <- which(c(as.data.frame(t(data2), stringsAsFactors = FALSE)) %in% c(as.data.frame(t(data1), stringsAsFactors = FALSE)))
- }else if(as.double(nrow(data1)) * nrow(data2) <= 1e6){ # as.double(nrow(data1)) to prevent integer overflow because R is 32 bits for integers
- same.row.pos1 <- logical(length = nrow(data1)) # FALSE by default
- same.row.pos1[] <- FALSE # security
- for(i3 in 1:nrow(data1)){
- for(i4 in 1:nrow(data2)){
- same.row.pos1[i3] <- identical(data1[i3, ], data2[i4, ])
- }
- }
- same.row.pos1 <- which(same.row.pos1)
-
- same.row.pos2 <- logical(length = nrow(data2)) # FALSE by default
- same.row.pos2[] <- FALSE # security
- for(i3 in 1:nrow(data2)){
- for(i4 in 1:nrow(data1)){
- same.row.pos2[i3] <- identical(data2[i3, ], data1[i4, ])
- }
- }
- same.row.pos2 <- which(same.row.pos2)
- }else{
- same.row.pos1 <- "TOO BIG FOR EVALUATION"
- same.row.pos2 <- "TOO BIG FOR EVALUATION"
- }
-
- names(same.row.pos1) <- NULL
- names(same.row.pos2) <- NULL
- if(all(is.na(same.row.pos1))){
- same.row.pos1 <- NULL
- }else{
- same.row.pos1 <- same.row.pos1[ ! is.na(same.row.pos1)]
- any.id.row <- TRUE
- }
- if(all(is.na(same.row.pos2))){
- same.row.pos2 <- NULL
- }else{
- same.row.pos2 <- same.row.pos2[ ! is.na(same.row.pos2)]
- any.id.row <- TRUE
- }
- if(is.null(same.row.pos1) & is.null(same.row.pos2)){
- any.id.row <- FALSE
- }else if(length(same.row.pos1) == 0 & length(same.row.pos2) == 0){
- any.id.row <- FALSE
- }else if(all(same.row.pos1 == "TOO BIG FOR EVALUATION") & all(same.row.pos2 == "TOO BIG FOR EVALUATION")){
- any.id.row <- NULL
- }
- }else{
- any.id.row <- FALSE
- # same.row.pos1 and 2 remain NULL
- }
- if(same.row.nb == TRUE){ # because if not the same row nb, the col cannot be identical
- if(all(sapply(data1, FUN = typeof) == "integer") & all(sapply(data2, FUN = typeof) == "integer") & as.double(ncol(data1)) * ncol(data2) <= 1e10){ # as.double(ncol(data1)) to prevent integer overflow because R is 32 bits for integers
- same.col.pos1 <- which(c(data1) %in% c(data2))
- same.col.pos2 <- which(c(data2) %in% c(data1))
- }else if(as.double(ncol(data1)) * ncol(data2) <= 1e6){ # as.double(ncol(data1)) to prevent integer overflow because R is 32 bits for integers
- same.col.pos1 <- logical(length = ncol(data1)) # FALSE by default
- same.col.pos1[] <- FALSE # security
- for(i3 in 1:ncol(data1)){
- for(i4 in 1:ncol(data2)){
- same.col.pos1[i3] <- identical(data1[ , i3], data2[ ,i4])
- }
- }
- same.col.pos1 <- which(same.col.pos1)
-
- same.col.pos2 <- logical(length = ncol(data2)) # FALSE by default
- same.col.pos2[] <- FALSE # security
- for(i3 in 1:ncol(data2)){
- for(i4 in 1:ncol(data1)){
- same.col.pos2[i3] <- identical(data2[ , i3], data1[ , i4])
- }
- }
- same.col.pos2 <- which(same.col.pos2)
- }else{
- same.col.pos1 <- "TOO BIG FOR EVALUATION"
- same.col.pos2 <- "TOO BIG FOR EVALUATION"
- }
- names(same.col.pos1) <- NULL
- names(same.col.pos2) <- NULL
- if(all(is.na(same.col.pos1))){
- same.col.pos1 <- NULL
- }else{
- same.col.pos1 <- same.col.pos1[ ! is.na(same.col.pos1)]
- any.id.col <- TRUE
- }
- if(all(is.na(same.col.pos2))){
- same.col.pos2 <- NULL
- }else{
- same.col.pos2 <- same.col.pos2[ ! is.na(same.col.pos2)]
- any.id.col <- TRUE
- }
- if(is.null(same.col.pos1) & is.null(same.col.pos2)){
- any.id.col <- FALSE
- }else if(length(same.col.pos1) == 0 & length(same.col.pos2) == 0){
- any.id.col <- FALSE
- }else if(all(same.col.pos1 == "TOO BIG FOR EVALUATION") & all(same.col.pos2 == "TOO BIG FOR EVALUATION")){
- any.id.col <- NULL
- }
- }else{
- any.id.col <- FALSE
- # same.col.pos1 and 2 remain NULL
- }
- if(same.dim == TRUE){
- names(data1) <- NULL
- row.names(data1) <- NULL
- names(data2) <- NULL
- row.names(data2) <- NULL
- if(identical(data1, data2)){
- identical.content <- TRUE
- }else{
- identical.content <- FALSE
- }
- }else{
- identical.content <- FALSE
- }
- }
- output <- list(same.class = same.class, class = class, same.dim = same.dim, dim = dim, same.row.nb = same.row.nb, row.nb = row.nb, same.col.nb = same.col.nb , col.nb = col.nb, same.row.name = same.row.name, row.name = row.name, any.id.row.name = any.id.row.name, same.row.name.pos1 = same.row.name.pos1, same.row.name.pos2 = same.row.name.pos2, common.row.names = common.row.names, same.col.name = same.col.name, col.name = col.name,any.id.col.name = any.id.col.name, same.col.name.pos1 = same.col.name.pos1, same.col.name.pos2 = same.col.name.pos2, common.col.names = common.col.names, any.id.row = any.id.row, same.row.pos1 = same.row.pos1, same.row.pos2 = same.row.pos2, any.id.col = any.id.col, same.col.pos1 = same.col.pos1, same.col.pos2 = same.col.pos2, identical.object = identical.object, identical.content = identical.content)
- return(output)
+# AIM
+# compare two 2D datasets of the same class or not. Check and report in a list if the 2 datasets have:
+# same class
+# common row names
+# common column names
+# same row number
+# same column number
+# potential identical rows between the 2 datasets
+# potential identical columns between the 2 datasets
+# ARGUMENTS
+# data1: matrix, data frame or table
+# data2: matrix, data frame or table
+# RETURN
+# a list containing:
+# $same.class: logical. Are class identical ?
+# $class: classes of the 2 datasets (NULL otherwise)
+# $same.dim: logical. Are dimension identical ?
+# $dim: dimension of the 2 datasets (NULL otherwise)
+# $same.row.nb: logical. Are number of rows identical ?
+# $row.nb: nb of rows of the 2 datasets if identical (NULL otherwise)
+# $same.col.nb: logical. Are number of columns identical ?
+# $col.nb: nb of columns of the 2 datasets if identical (NULL otherwise)
+# $same.row.name: logical. Are row names identical ? NULL if no row names in the two 2D datasets
+# $row.name: name of rows of the 2 datasets if identical (NULL otherwise)
+# $any.id.row.name: logical. Is there any row names identical ? NULL if no row names in the two 2D datasets
+# $same.row.name.pos1: position, in data1, of the row names identical in data2
+# $same.row.name.pos2: position, in data2, of the row names identical in data1
+# $common.row.names: common row names between data1 and data2 (can be a subset of $name or not). NULL if no common row names
+# $same.col.name: logical. Are column names identical ? NULL if no col names in the two 2D datasets
+# $col.name: name of columns of the 2 datasets if identical (NULL otherwise)
+# $any.id.col.name: logical. Is there any column names identical ? NULL if no col names in the two 2D datasets
+# $same.col.name.pos1: position, in data1, of the column names identical in data2
+# $same.col.name.pos2: position, in data2, of the column names identical in data1
+# $common.col.names: common column names between data1 and data2 (can be a subset of $name or not). NULL if no common column names
+# $any.id.row: logical. is there identical rows (not considering row names)? NULL if nrow(data1) * nrow(data2) > 1e10
+# $same.row.pos1: position, in data1, of the rows identical in data2 (not considering row names). Return "TOO BIG FOR EVALUATION" if nrow(data1) * nrow(data2) > 1e10
+# $same.row.pos2: position, in data2, of the rows identical in data1 (not considering row names). Return "TOO BIG FOR EVALUATION" if nrow(data1) * nrow(data2) > 1e10
+# $any.id.col: logical. is there identical columns (not considering column names)? NULL if ncol(data1) * ncol(data2) > 1e10
+# $same.col.pos1: position in data1 of the cols identical in data2 (not considering column names). Return "TOO BIG FOR EVALUATION" if ncol(data1) * ncol(data2) > 1e10
+# $same.col.pos2: position in data2 of the cols identical in data1 (not considering column names). Return "TOO BIG FOR EVALUATION" if ncol(data1) * ncol(data2) > 1e10
+# $identical.object: logical. Are objects identical (including row & column names)?
+# $identical.content: logical. Are content objects identical (identical excluding row & column names)?
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# none
+# EXAMPLES
+# obs1 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; obs2 = as.data.frame(matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])), stringsAsFactors = TRUE) ; obs1 ; obs2 ; fun_comp_2d(obs1, obs2)
+# obs1 = matrix(101:110, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; obs2 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; obs1 ; obs2 ; fun_comp_2d(obs1, obs2)
+# large matrices
+# obs1 = matrix(1:1e6, ncol = 5, dimnames = list(NULL, LETTERS[1:5])) ; obs2 = matrix(as.integer((1:1e6)+1e6/5), ncol = 5, dimnames = list(NULL, LETTERS[1:5])) ; head(obs1) ; head(obs2) ; fun_comp_2d(obs1, obs2)
+# WARNING: when comparing content (rows, columns, or total), double and integer data are considered as different -> double(1) != integer(1)
+# obs1 = matrix(1:1e6, ncol = 5, dimnames = list(NULL, LETTERS[1:5])) ; obs2 = matrix((1:1e6)+1e6/5, ncol = 5, dimnames = list(NULL, LETTERS[1:5])) ; head(obs1) ; head(obs2) ; fun_comp_2d(obs1, obs2)
+# obs1 = matrix(1:10, byrow = TRUE, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; obs2 = matrix(c(1:5, 101:105, 6:10), byrow = TRUE, ncol = 5, dimnames = list(c("a", "z", "b"), c(LETTERS[1:2], "k", LETTERS[5:4]))) ; obs1 ; obs2 ; fun_comp_2d(obs1, obs2)
+# obs1 = t(matrix(1:10, byrow = TRUE, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5]))) ; obs2 = t(matrix(c(1:5, 101:105, 6:10), byrow = TRUE, ncol = 5, dimnames = list(c("a", "z", "b"), c(LETTERS[1:2], "k", LETTERS[5:4])))) ; obs1 ; obs2 ; fun_comp_2d(obs1, obs2)
+# DEBUGGING
+# data1 = matrix(1:10, ncol = 5) ; data2 = matrix(1:10, ncol = 5) # for function debugging
+# data1 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; data2 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
+# data1 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; data2 = matrix(1:10, ncol = 5) # for function debugging
+# data1 = matrix(1:15, byrow = TRUE, ncol = 5, dimnames = list(letters[1:3], LETTERS[1:5])) ; data2 = matrix(1:10, byrow = TRUE, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
+# data1 = matrix(1:15, ncol = 5, dimnames = list(letters[1:3], LETTERS[1:5])) ; data2 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
+# data1 = matrix(1:15, ncol = 5, dimnames = list(paste0("A", letters[1:3]), LETTERS[1:5])) ; data2 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
+# data1 = matrix(1:15, ncol = 5, dimnames = list(letters[1:3], LETTERS[1:5])) ; data2 = matrix(1:12, ncol = 4, dimnames = list(letters[1:3], LETTERS[1:4])) # for function debugging
+# data1 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; data2 = matrix(101:110, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
+# data1 = data.frame(a = 1:3, b= letters[1:3], row.names = LETTERS[1:3], stringsAsFactors = TRUE) ; data2 = data.frame(A = 1:3, B= letters[1:3], stringsAsFactors = TRUE) # for function debugging
+# data1 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; data2 = as.data.frame(matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])), stringsAsFactors = TRUE) # for function debugging
+# data1 = matrix(1:10, byrow = TRUE, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; data2 = matrix(c(1:5, 101:105, 6:10), byrow = TRUE, ncol = 5, dimnames = list(c("a", "z", "b"), c(LETTERS[1:2], "k", LETTERS[5:4]))) # for function debugging
+# data1 = table(Exp1 = c("A", "A", "A", "B", "B", "B"), Exp2 = c("A1", "B1", "A1", "C1", "C1", "B1")) ; data2 = data.frame(A = 1:3, B= letters[1:3], stringsAsFactors = TRUE) # for function debugging
+# data1 = matrix(1:1e6, ncol = 5, dimnames = list(NULL, LETTERS[1:5])) ; data2 = matrix((1:1e6)+1e6/5, ncol = 5, dimnames = list(NULL, LETTERS[1:5]))
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# argument checking
+if( ! (any(class(data1) %in% c("data.frame", "table")) | all(class(data1) %in% c("matrix", "array")))){ # before R4.0.0, it was ! any(class(data1) %in% c("matrix", "data.frame", "table"))
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 ARGUMENT MUST BE A MATRIX, DATA FRAME OR TABLE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! (any(class(data2) %in% c("data.frame", "table")) | all(class(data2) %in% c("matrix", "array")))){ # before R4.0.0, it was ! any(class(data2) %in% c("matrix", "data.frame", "table"))
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE data2 ARGUMENT MUST BE A MATRIX, DATA FRAME OR TABLE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) # activate this line and use the function to check arguments status
+# end argument checking
+# main code
+same.class <- NULL
+class <- NULL
+same.dim <- NULL
+dim <- NULL
+same.row.nb <- NULL
+row.nb <- NULL
+same.col.nb <- NULL
+col.nb <- NULL
+same.row.name <- NULL
+row.name <- NULL
+any.id.row.name <- NULL
+same.row.name.pos1 <- NULL
+same.row.name.pos2 <- NULL
+common.row.names <- NULL
+same.col.name <- NULL
+any.id.col.name <- NULL
+same.col.name.pos1 <- NULL
+same.col.name.pos2 <- NULL
+common.col.names <- NULL
+col.name <- NULL
+any.id.row <- NULL
+same.row.pos1 <- NULL
+same.row.pos2 <- NULL
+any.id.col <- NULL
+same.col.pos1 <- NULL
+same.col.pos2 <- NULL
+identical.object <- NULL
+identical.content <- NULL
+if(identical(data1, data2) & (any(class(data1) %in% c("data.frame", "table")) | all(class(data1) %in% c("matrix", "array")))){ # before R4.0.0, it was ! any(class(data1) %in% c("matrix", "data.frame", "table"))
+same.class <- TRUE
+class <- class(data1)
+same.dim <- TRUE
+dim <- dim(data1)
+same.row.nb <- TRUE
+row.nb <- nrow(data1)
+same.col.nb <- TRUE
+col.nb <- ncol(data1)
+same.row.name <- TRUE
+row.name <- dimnames(data1)[[1]]
+any.id.row.name <- TRUE
+same.row.name.pos1 <- 1:row.nb
+same.row.name.pos2 <- 1:row.nb
+common.row.names <- dimnames(data1)[[1]]
+same.col.name <- TRUE
+col.name <- dimnames(data1)[[2]]
+any.id.col.name <- TRUE
+same.col.name.pos1 <- 1:col.nb
+same.col.name.pos2 <- 1:col.nb
+common.col.names <- dimnames(data1)[[2]]
+any.id.row <- TRUE
+same.row.pos1 <- 1:row.nb
+same.row.pos2 <- 1:row.nb
+any.id.col <- TRUE
+same.col.pos1 <- 1:col.nb
+same.col.pos2 <- 1:col.nb
+identical.object <- TRUE
+identical.content <- TRUE
+}else{
+identical.object <- FALSE
+if(all(class(data1) == "table") & length(dim(data1)) == 1){
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 ARGUMENT IS A 1D TABLE. USE THE fun_comp_1d FUNCTION")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(all(class(data2) == "table") & length(dim(data2)) == 1){
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE data2 ARGUMENT IS A 1D TABLE. USE THE fun_comp_1d FUNCTION")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! identical(class(data1), class(data2))){
+same.class <- FALSE
+}else if( ! (any(class(data1) %in% c("data.frame", "table")) | all(class(data1) %in% c("matrix", "array")))){ # before R4.0.0, it was ! any(class(data1) %in% c("matrix", "data.frame", "table"))
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 AND data2 ARGUMENTS MUST BE EITHER MATRIX, DATA FRAME OR TABLE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+same.class <- TRUE
+class <- class(data1)
+}
+if( ! identical(dim(data1), dim(data2))){
+same.dim <- FALSE
+}else{
+same.dim <- TRUE
+dim <- dim(data1)
+}
+if( ! identical(nrow(data1), nrow(data2))){
+same.row.nb <- FALSE
+}else{
+same.row.nb <- TRUE
+row.nb <- nrow(data1)
+}
+if( ! identical(ncol(data1), ncol(data2))){
+same.col.nb <- FALSE
+}else{
+same.col.nb <- TRUE
+col.nb <- ncol(data1)
+}
+# row and col names
+if(is.null(dimnames(data1)) & is.null(dimnames(data2))){
+same.row.name <- NULL
+same.col.name <- NULL
+# row and col names remain NULL
+}else if((is.null(dimnames(data1)) & ! is.null(dimnames(data2))) | ( ! is.null(dimnames(data1)) & is.null(dimnames(data2)))){
+same.row.name <- FALSE
+same.col.name <- FALSE
+# row and col names remain NULL
+}else{
+if( ! identical(dimnames(data1)[[1]], dimnames(data2)[[1]])){
+same.row.name <- FALSE
+# row names remain NULL
+}else{
+same.row.name <- TRUE
+row.name <- dimnames(data1)[[1]]
+}
+# row names
+any.id.row.name <- FALSE
+if(any(dimnames(data1)[[1]] %in% dimnames(data2)[[1]])){
+any.id.row.name <- TRUE
+same.row.name.pos1 <- which(dimnames(data1)[[1]] %in% dimnames(data2)[[1]])
+}
+if(any(dimnames(data2)[[1]] %in% dimnames(data1)[[1]])){
+any.id.row.name <- TRUE
+same.row.name.pos2 <- which(dimnames(data2)[[1]] %in% dimnames(data1)[[1]])
+}
+if(any.id.row.name == TRUE){
+common.row.names <- unique(c(dimnames(data1)[[1]][same.row.name.pos1], dimnames(data2)[[1]][same.row.name.pos2]))
+}
+# col names
+any.id.col.name <- FALSE
+if(any(dimnames(data1)[[2]] %in% dimnames(data2)[[2]])){
+any.id.col.name <- TRUE
+same.col.name.pos1 <- which(dimnames(data1)[[2]] %in% dimnames(data2)[[2]])
+}
+if(any(dimnames(data2)[[2]] %in% dimnames(data1)[[2]])){
+any.id.col.name <- TRUE
+same.col.name.pos2 <- which(dimnames(data2)[[2]] %in% dimnames(data1)[[2]])
+}
+if(any.id.col.name == TRUE){
+common.col.names <- unique(c(dimnames(data1)[[2]][same.col.name.pos1], dimnames(data2)[[2]][same.col.name.pos2]))
+}
+if( ! identical(dimnames(data1)[[2]], dimnames(data2)[[2]])){
+same.col.name <- FALSE
+# col names remain NULL
+}else{
+same.col.name <- TRUE
+col.name <- dimnames(data1)[[2]]
+}
+}
+# identical row and col content
+if(all(class(data1) == "table")){
+as.data.frame(matrix(data1, ncol = ncol(data1)), stringsAsFactors = FALSE)
+}else if(all(class(data1) %in% c("matrix", "array"))){
+data1 <- as.data.frame(data1, stringsAsFactors = FALSE)
+}else if(all(class(data1) == "data.frame")){
+data1 <- data.frame(lapply(data1, as.character), stringsAsFactors = FALSE)
+}
+if(all(class(data2) == "table")){
+as.data.frame(matrix(data2, ncol = ncol(data2)), stringsAsFactors = FALSE)
+}else if(all(class(data2) %in% c("matrix", "array"))){
+data2 <- as.data.frame(data2, stringsAsFactors = FALSE)
+}else if(all(class(data2) == "data.frame")){
+data2 <- data.frame(lapply(data2, as.character), stringsAsFactors = FALSE)
+}
+row.names(data1) <- paste0("A", 1:nrow(data1))
+row.names(data2) <- paste0("A", 1:nrow(data2))
+if(same.col.nb == TRUE){ # because if not the same col nb, the row cannot be identical
+if(all(sapply(data1, FUN = typeof) == "integer") & all(sapply(data2, FUN = typeof) == "integer") & as.double(nrow(data1)) * nrow(data2) <= 1e10){ # as.double(nrow(data1)) to prevent integer overflow because R is 32 bits for integers
+same.row.pos1 <- which(c(as.data.frame(t(data1), stringsAsFactors = FALSE)) %in% c(as.data.frame(t(data2), stringsAsFactors = FALSE))) # this work fast with only integers (because 32 bits)
+same.row.pos2 <- which(c(as.data.frame(t(data2), stringsAsFactors = FALSE)) %in% c(as.data.frame(t(data1), stringsAsFactors = FALSE)))
+}else if(as.double(nrow(data1)) * nrow(data2) <= 1e6){ # as.double(nrow(data1)) to prevent integer overflow because R is 32 bits for integers
+same.row.pos1 <- logical(length = nrow(data1)) # FALSE by default
+same.row.pos1[] <- FALSE # security
+for(i3 in 1:nrow(data1)){
+for(i4 in 1:nrow(data2)){
+same.row.pos1[i3] <- identical(data1[i3, ], data2[i4, ])
+}
+}
+same.row.pos1 <- which(same.row.pos1)
+
+same.row.pos2 <- logical(length = nrow(data2)) # FALSE by default
+same.row.pos2[] <- FALSE # security
+for(i3 in 1:nrow(data2)){
+for(i4 in 1:nrow(data1)){
+same.row.pos2[i3] <- identical(data2[i3, ], data1[i4, ])
+}
+}
+same.row.pos2 <- which(same.row.pos2)
+}else{
+same.row.pos1 <- "TOO BIG FOR EVALUATION"
+same.row.pos2 <- "TOO BIG FOR EVALUATION"
+}
+
+names(same.row.pos1) <- NULL
+names(same.row.pos2) <- NULL
+if(all(is.na(same.row.pos1))){
+same.row.pos1 <- NULL
+}else{
+same.row.pos1 <- same.row.pos1[ ! is.na(same.row.pos1)]
+any.id.row <- TRUE
+}
+if(all(is.na(same.row.pos2))){
+same.row.pos2 <- NULL
+}else{
+same.row.pos2 <- same.row.pos2[ ! is.na(same.row.pos2)]
+any.id.row <- TRUE
+}
+if(is.null(same.row.pos1) & is.null(same.row.pos2)){
+any.id.row <- FALSE
+}else if(length(same.row.pos1) == 0 & length(same.row.pos2) == 0){
+any.id.row <- FALSE
+}else if(all(same.row.pos1 == "TOO BIG FOR EVALUATION") & all(same.row.pos2 == "TOO BIG FOR EVALUATION")){
+any.id.row <- NULL
+}
+}else{
+any.id.row <- FALSE
+# same.row.pos1 and 2 remain NULL
+}
+if(same.row.nb == TRUE){ # because if not the same row nb, the col cannot be identical
+if(all(sapply(data1, FUN = typeof) == "integer") & all(sapply(data2, FUN = typeof) == "integer") & as.double(ncol(data1)) * ncol(data2) <= 1e10){ # as.double(ncol(data1)) to prevent integer overflow because R is 32 bits for integers
+same.col.pos1 <- which(c(data1) %in% c(data2))
+same.col.pos2 <- which(c(data2) %in% c(data1))
+}else if(as.double(ncol(data1)) * ncol(data2) <= 1e6){ # as.double(ncol(data1)) to prevent integer overflow because R is 32 bits for integers
+same.col.pos1 <- logical(length = ncol(data1)) # FALSE by default
+same.col.pos1[] <- FALSE # security
+for(i3 in 1:ncol(data1)){
+for(i4 in 1:ncol(data2)){
+same.col.pos1[i3] <- identical(data1[ , i3], data2[ ,i4])
+}
+}
+same.col.pos1 <- which(same.col.pos1)
+
+same.col.pos2 <- logical(length = ncol(data2)) # FALSE by default
+same.col.pos2[] <- FALSE # security
+for(i3 in 1:ncol(data2)){
+for(i4 in 1:ncol(data1)){
+same.col.pos2[i3] <- identical(data2[ , i3], data1[ , i4])
+}
+}
+same.col.pos2 <- which(same.col.pos2)
+}else{
+same.col.pos1 <- "TOO BIG FOR EVALUATION"
+same.col.pos2 <- "TOO BIG FOR EVALUATION"
+}
+names(same.col.pos1) <- NULL
+names(same.col.pos2) <- NULL
+if(all(is.na(same.col.pos1))){
+same.col.pos1 <- NULL
+}else{
+same.col.pos1 <- same.col.pos1[ ! is.na(same.col.pos1)]
+any.id.col <- TRUE
+}
+if(all(is.na(same.col.pos2))){
+same.col.pos2 <- NULL
+}else{
+same.col.pos2 <- same.col.pos2[ ! is.na(same.col.pos2)]
+any.id.col <- TRUE
+}
+if(is.null(same.col.pos1) & is.null(same.col.pos2)){
+any.id.col <- FALSE
+}else if(length(same.col.pos1) == 0 & length(same.col.pos2) == 0){
+any.id.col <- FALSE
+}else if(all(same.col.pos1 == "TOO BIG FOR EVALUATION") & all(same.col.pos2 == "TOO BIG FOR EVALUATION")){
+any.id.col <- NULL
+}
+}else{
+any.id.col <- FALSE
+# same.col.pos1 and 2 remain NULL
+}
+if(same.dim == TRUE){
+names(data1) <- NULL
+row.names(data1) <- NULL
+names(data2) <- NULL
+row.names(data2) <- NULL
+if(identical(data1, data2)){
+identical.content <- TRUE
+}else{
+identical.content <- FALSE
+}
+}else{
+identical.content <- FALSE
+}
+}
+output <- list(same.class = same.class, class = class, same.dim = same.dim, dim = dim, same.row.nb = same.row.nb, row.nb = row.nb, same.col.nb = same.col.nb , col.nb = col.nb, same.row.name = same.row.name, row.name = row.name, any.id.row.name = any.id.row.name, same.row.name.pos1 = same.row.name.pos1, same.row.name.pos2 = same.row.name.pos2, common.row.names = common.row.names, same.col.name = same.col.name, col.name = col.name,any.id.col.name = any.id.col.name, same.col.name.pos1 = same.col.name.pos1, same.col.name.pos2 = same.col.name.pos2, common.col.names = common.col.names, any.id.row = any.id.row, same.row.pos1 = same.row.pos1, same.row.pos2 = same.row.pos2, any.id.col = any.id.col, same.col.pos1 = same.col.pos1, same.col.pos2 = same.col.pos2, identical.object = identical.object, identical.content = identical.content)
+return(output)
}
@@ -1345,130 +1344,130 @@ fun_comp_2d <- function(data1, data2){
# Check OK: clear to go Apollo
fun_comp_list <- function(data1, data2){
- # AIM
- # compare two lists. Check and report in a list if the 2 datasets have:
- # same length
- # common names
- # common compartments
- # ARGUMENTS
- # data1: list
- # data2: list
- # RETURN
- # a list containing:
- # $same.length: logical. Are number of elements identical?
- # $length: number of elements in the 2 datasets (NULL otherwise)
- # $same.name: logical. Are element names identical ?
- # $name: name of elements of the 2 datasets if identical (NULL otherwise)
- # $any.id.name: logical. Is there any element names identical ?
- # $same.name.pos1: position, in data1, of the element names identical in data2
- # $same.name.pos2: position, in data2, of the compartment names identical in data1
- # $any.id.compartment: logical. is there any identical compartments ?
- # $same.compartment.pos1: position, in data1, of the compartments identical in data2
- # $same.compartment.pos2: position, in data2, of the compartments identical in data1
- # $identical.object: logical. Are objects identical (kind of object, compartment names and content)?
- # $identical.content: logical. Are content objects identical (identical compartments excluding compartment names)?
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # none
- # EXAMPLES
- # obs1 = list(a = 1:5, b = LETTERS[1:2], d = matrix(1:6)) ; obs2 = list(a = 1:5, b = LETTERS[1:2], d = matrix(1:6)) ; fun_comp_list(obs1, obs2)
- # obs1 = list(1:5, LETTERS[1:2]) ; obs2 = list(a = 1:5, b = LETTERS[1:2]) ; fun_comp_list(obs1, obs2)
- # obs1 = list(b = 1:5, c = LETTERS[1:2]) ; obs2 = list(a = 1:5, b = LETTERS[1:2], d = matrix(1:6)) ; fun_comp_list(obs1, obs2)
- # obs1 = list(b = 1:5, c = LETTERS[1:2]) ; obs2 = list(LETTERS[5:9], matrix(1:6), 1:5) ; fun_comp_list(obs1, obs2)
- # DEBUGGING
- # data1 = list(a = 1:5, b = LETTERS[1:2], d = matrix(1:6)) ; data2 = list(a = 1:5, b = LETTERS[1:2], d = matrix(1:6)) # for function debugging
- # data1 = list(a = 1:5, b = LETTERS[1:2]) ; data2 = list(a = 1:5, b = LETTERS[1:2], d = matrix(1:6)) # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # argument checking
- if( ! any(class(data1) %in% "list")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 ARGUMENT MUST BE A LIST")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! any(class(data2) %in% "list")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE data2 ARGUMENT MUST BE A LIST")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) # activate this line and use the function to check arguments status
- # end argument checking
- # main code
- same.length <- NULL
- length <- NULL
- same.name <- NULL
- name <- NULL
- any.id.name <- NULL
- same.name.pos1 <- NULL
- same.name.pos2 <- NULL
- any.id.compartment <- NULL
- same.compartment.pos1 <- NULL
- same.compartment.pos2 <- NULL
- identical.object <- NULL
- identical.content <- NULL
- if(identical(data1, data2)){
- same.length <- TRUE
- length <- length(data1)
- if( ! is.null(names(data1))){
- same.name <- TRUE
- name <- names(data1)
- any.id.name <- TRUE
- same.name.pos1 <- 1:length(data1)
- same.name.pos2 <- 1:length(data2)
- }
- any.id.compartment <- TRUE
- same.compartment.pos1 <- 1:length(data1)
- same.compartment.pos2 <- 1:length(data2)
- identical.object <- TRUE
- identical.content <- TRUE
- }else{
- identical.object <- FALSE
- if( ! identical(length(data1), length(data2))){
- same.length<- FALSE
- }else{
- same.length<- TRUE
- length <- length(data1)
- }
- if( ! (is.null(names(data1)) & is.null(names(data2)))){
- if( ! identical(names(data1), names(data2))){
- same.name <- FALSE
- }else{
- same.name <- TRUE
- name <- names(data1)
- }
- any.id.name <- FALSE
- if(any(names(data1) %in% names(data2))){
- any.id.name <- TRUE
- same.name.pos1 <- which(names(data1) %in% names(data2))
- }
- if(any(names(data2) %in% names(data1))){
- any.id.name <- TRUE
- same.name.pos2 <- which(names(data2) %in% names(data1))
- }
- }
- names(data1) <- NULL
- names(data2) <- NULL
- any.id.compartment <- FALSE
- if(any(data1 %in% data2)){
- any.id.compartment <- TRUE
- same.compartment.pos1 <- which(data1 %in% data2)
- }
- if(any(data2 %in% data1)){
- any.id.compartment <- TRUE
- same.compartment.pos2 <- which(data2 %in% data1)
- }
- if(same.length == TRUE & ! all(is.null(same.compartment.pos1), is.null(same.compartment.pos2))){
- if(identical(same.compartment.pos1, same.compartment.pos2)){
- identical.content <- TRUE
- }else{
- identical.content <- FALSE
- }
- }else{
- identical.content <- FALSE
- }
- }
- output <- list(same.length = same.length, length = length, same.name = same.name, name = name, any.id.name = any.id.name, same.name.pos1 = same.name.pos1, same.name.pos2 = same.name.pos2, any.id.compartment = any.id.compartment, same.compartment.pos1 = same.compartment.pos1, same.compartment.pos2 = same.compartment.pos2, identical.object = identical.object, identical.content = identical.content)
- return(output)
+# AIM
+# compare two lists. Check and report in a list if the 2 datasets have:
+# same length
+# common names
+# common compartments
+# ARGUMENTS
+# data1: list
+# data2: list
+# RETURN
+# a list containing:
+# $same.length: logical. Are number of elements identical?
+# $length: number of elements in the 2 datasets (NULL otherwise)
+# $same.name: logical. Are element names identical ?
+# $name: name of elements of the 2 datasets if identical (NULL otherwise)
+# $any.id.name: logical. Is there any element names identical ?
+# $same.name.pos1: position, in data1, of the element names identical in data2
+# $same.name.pos2: position, in data2, of the compartment names identical in data1
+# $any.id.compartment: logical. is there any identical compartments ?
+# $same.compartment.pos1: position, in data1, of the compartments identical in data2
+# $same.compartment.pos2: position, in data2, of the compartments identical in data1
+# $identical.object: logical. Are objects identical (kind of object, compartment names and content)?
+# $identical.content: logical. Are content objects identical (identical compartments excluding compartment names)?
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# none
+# EXAMPLES
+# obs1 = list(a = 1:5, b = LETTERS[1:2], d = matrix(1:6)) ; obs2 = list(a = 1:5, b = LETTERS[1:2], d = matrix(1:6)) ; fun_comp_list(obs1, obs2)
+# obs1 = list(1:5, LETTERS[1:2]) ; obs2 = list(a = 1:5, b = LETTERS[1:2]) ; fun_comp_list(obs1, obs2)
+# obs1 = list(b = 1:5, c = LETTERS[1:2]) ; obs2 = list(a = 1:5, b = LETTERS[1:2], d = matrix(1:6)) ; fun_comp_list(obs1, obs2)
+# obs1 = list(b = 1:5, c = LETTERS[1:2]) ; obs2 = list(LETTERS[5:9], matrix(1:6), 1:5) ; fun_comp_list(obs1, obs2)
+# DEBUGGING
+# data1 = list(a = 1:5, b = LETTERS[1:2], d = matrix(1:6)) ; data2 = list(a = 1:5, b = LETTERS[1:2], d = matrix(1:6)) # for function debugging
+# data1 = list(a = 1:5, b = LETTERS[1:2]) ; data2 = list(a = 1:5, b = LETTERS[1:2], d = matrix(1:6)) # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# argument checking
+if( ! any(class(data1) %in% "list")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 ARGUMENT MUST BE A LIST")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! any(class(data2) %in% "list")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE data2 ARGUMENT MUST BE A LIST")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) # activate this line and use the function to check arguments status
+# end argument checking
+# main code
+same.length <- NULL
+length <- NULL
+same.name <- NULL
+name <- NULL
+any.id.name <- NULL
+same.name.pos1 <- NULL
+same.name.pos2 <- NULL
+any.id.compartment <- NULL
+same.compartment.pos1 <- NULL
+same.compartment.pos2 <- NULL
+identical.object <- NULL
+identical.content <- NULL
+if(identical(data1, data2)){
+same.length <- TRUE
+length <- length(data1)
+if( ! is.null(names(data1))){
+same.name <- TRUE
+name <- names(data1)
+any.id.name <- TRUE
+same.name.pos1 <- 1:length(data1)
+same.name.pos2 <- 1:length(data2)
+}
+any.id.compartment <- TRUE
+same.compartment.pos1 <- 1:length(data1)
+same.compartment.pos2 <- 1:length(data2)
+identical.object <- TRUE
+identical.content <- TRUE
+}else{
+identical.object <- FALSE
+if( ! identical(length(data1), length(data2))){
+same.length<- FALSE
+}else{
+same.length<- TRUE
+length <- length(data1)
+}
+if( ! (is.null(names(data1)) & is.null(names(data2)))){
+if( ! identical(names(data1), names(data2))){
+same.name <- FALSE
+}else{
+same.name <- TRUE
+name <- names(data1)
+}
+any.id.name <- FALSE
+if(any(names(data1) %in% names(data2))){
+any.id.name <- TRUE
+same.name.pos1 <- which(names(data1) %in% names(data2))
+}
+if(any(names(data2) %in% names(data1))){
+any.id.name <- TRUE
+same.name.pos2 <- which(names(data2) %in% names(data1))
+}
+}
+names(data1) <- NULL
+names(data2) <- NULL
+any.id.compartment <- FALSE
+if(any(data1 %in% data2)){
+any.id.compartment <- TRUE
+same.compartment.pos1 <- which(data1 %in% data2)
+}
+if(any(data2 %in% data1)){
+any.id.compartment <- TRUE
+same.compartment.pos2 <- which(data2 %in% data1)
+}
+if(same.length == TRUE & ! all(is.null(same.compartment.pos1), is.null(same.compartment.pos2))){
+if(identical(same.compartment.pos1, same.compartment.pos2)){
+identical.content <- TRUE
+}else{
+identical.content <- FALSE
+}
+}else{
+identical.content <- FALSE
+}
+}
+output <- list(same.length = same.length, length = length, same.name = same.name, name = name, any.id.name = any.id.name, same.name.pos1 = same.name.pos1, same.name.pos2 = same.name.pos2, any.id.compartment = any.id.compartment, same.compartment.pos1 = same.compartment.pos1, same.compartment.pos2 = same.compartment.pos2, identical.object = identical.object, identical.content = identical.content)
+return(output)
}
@@ -1479,393 +1478,393 @@ fun_comp_list <- function(data1, data2){
# Check OK: clear to go Apollo
fun_test <- function(
- fun,
- arg,
- val,
- expect.error = NULL,
- thread.nb = NULL,
- print.count = 10,
- plot.fun = FALSE,
- export = FALSE,
- res.path = NULL,
- lib.path = NULL,
- cute.path = "C:\\Users\\Gael\\Documents\\Git_projects\\cute_little_R_functions\\cute_little_R_functions.R"
+fun,
+arg,
+val,
+expect.error = NULL,
+thread.nb = NULL,
+print.count = 10,
+plot.fun = FALSE,
+export = FALSE,
+res.path = NULL,
+lib.path = NULL,
+cute.path = "C:\\Users\\Gael\\Documents\\Git_projects\\cute_little_R_functions\\cute_little_R_functions.R"
){
- # AIM
- # test combinations of argument values of a function
- # WARNINGS
- # Limited to 43 arguments with at least 2 values each. The total number of arguments tested can be more if the additional arguments have a single value. The limit is due to nested "for" loops (https://stat.ethz.ch/pipermail/r-help/2008-March/157341.html), but it should not be a problem since the number of tests would be 2^43 > 8e12
- # ARGUMENTS
- # fun: character string indicating the name of the function tested (without brackets)
- # arg: vector of character strings of arguments of fun. At least arguments that do not have default values must be present in this vector
- # val: list with number of compartments equal to length of arg, each compartment containing values of the corresponding argument in arg. Each different value must be in a list or in a vector. For instance, argument 3 in arg is a logical argument (values accepted TRUE, FALSE, NA). Thus, compartment 3 of val can be either list(TRUE, FALSE, NA), or c(TRUE, FALSE, NA)
- # expect.error: list of exactly the same structure as val argument, but containing FALSE or TRUE, depending on whether error is expected (TRUE) or not (FALSE) for each corresponding value of val. A message is returned depending on discrepancies between the expected and observed errors. BEWARE: not always possible to write the expected errors for all the combination of argument values. Ignored if NULL
- # thread.nb: numeric value indicating the number of available threads. Write NULL if no parallelization wanted
- # print.count: interger value. Print a working progress message every print.count during loops. BEWARE: can increase substentially the time to complete the process using a small value, like 10 for instance. Use Inf is no loop message desired
- # plot.fun: logical. Plot the plotting function tested for each test?
- # export: logical. Export the results into a .RData file and into a .txt file? If FALSE, return a list into the console (see below). BEWARE: will be automatically set to TRUE if thread.nb is not NULL. This means that when using parallelization, the results are systematically exported, not returned into the console
- # res.path: character string indicating the absolute pathway of folder where the txt results and pdfs, containing all the plots, will be saved. Several txt and pdf, one per thread, if parallelization. Ignored if export is FALSE. Must be specified if thread.nb is not NULL or if export is TRUE
- # lib.path: character vector specifying the absolute pathways of the directories containing the required packages if not in the default directories. Ignored if NULL
- # cute.path: character string indicating the absolute path of the cute.R file. Will be remove when cute will be a package. Not considered if thread.nb is NULL
- # REQUIRED PACKAGES
- # lubridate
- # parallel if thread.nb argument is not NULL (included in the R installation packages but not automatically loaded)
- # if the tested function is in a package, this package must be imported first (no parallelization) or must be in the classical R package folder indicated by the lib.path argument (parallelization)
- # RETURN
- # if export is FALSE a list containing:
- # $fun: the tested function
- # $data: a data frame of all the combination tested, containing the following columns:
- # the different values tested, named by arguments
- # $kind: a vector of character strings indicating the kind of test result: either "ERROR", or "WARNING", or "OK"
- # $problem: a logical vector indicating if error or not
- # $expected.error: optional logical vector indicating the expected error specified in the expect.error argument
- # $message: either NULL if $kind is always "OK", or the messages
- # $instruction: the initial instruction
- # $sys.info: system and packages info
- # if export is TRUE 1) the same list object into a .RData file, 2) also the $data data frame into a .txt file, and 3) if expect.error is non NULL and if any discrepancy, the $data data frame into a .txt file but containing only the rows with discrepancies between expected and observed errors
- # one or several pdf if a plotting function is tested and if the plot.fun argument is TRUE
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # fun_get_message()
- # fun_pack()
- # EXAMPLES
- # fun_test(fun = "unique", arg = c("x", "incomparables"), val = list(x = list(1:10, c(1,1,2,8), NA), incomparable = c(TRUE, FALSE, NA)))
- # fun_test(fun = "fun_round", arg = c("data", "dec.nb", "after.lead.zero"), val = list(L1 = list(c(1, 1.0002256, 1.23568), "a", NA), L2 = list(2, c(1,3), NA), L3 = c(TRUE, FALSE, NA)))
- # fun_test(fun = "plot", arg = c("x", "y"), val = list(x = list(1:10, 12:13, NA, (1:10)^2), y = list(1:10, NA, NA)), expect.error = list(x = list(FALSE, TRUE, TRUE, FALSE), y = list(FALSE, TRUE, TRUE)), thread.nb = NULL, plot.fun = TRUE, res.path = "C:\\Users\\Gael\\Desktop\\", lib.path = NULL)
- # fun_test(fun = "plot", arg = c("x", "y"), val = list(x = list(1:10, 12:13, NA, (1:10)^2), y = list(1:10, NA, NA)), thread.nb = 4, plot.fun = TRUE, res.path = "C:\\Users\\Gael\\Desktop\\", lib.path = "C:\\Program Files\\R\\R-4.0.2\\library\\")
- # set.seed(1) ; obs1 <- data.frame(Time = c(rnorm(10), rnorm(10) + 2), Group1 = rep(c("G", "H"), each = 10), stringsAsFactors = TRUE) ; fun_test(fun = "fun_gg_boxplot", arg = c("data1", "y", "categ"), val = list(L1 = list(L1 = obs1), L2 = list(L1 = "Time"), L3 = list(L1 = "Group1")))
- # set.seed(1) ; obs1 <- data.frame(Time = c(rnorm(10), rnorm(10) + 2), Group1 = rep(c("G", "H"), each = 10), stringsAsFactors = TRUE) ; fun_test(fun = "fun_gg_boxplot", arg = c("data1", "y", "categ"), val = list(L1 = list(obs1), L2 = "Time", L3 = "Group1"), thread.nb = NULL, plot.fun = TRUE, res.path = "C:\\Users\\Gael\\Desktop\\", lib.path = "C:\\Program Files\\R\\R-4.0.2\\library\\")
- # library(ggplot2) ; fun_test(fun = "geom_histogram", arg = c("data", "mapping"), val = list(x = list(data.frame(X = "a", stringsAsFactors = TRUE)), y = list(ggplot2::aes(x = X))), thread.nb = NULL, plot.fun = TRUE, res.path = "C:\\Users\\Gael\\Desktop\\", lib.path = "C:\\Program Files\\R\\R-4.0.2\\library\\") # BEWARE: ggplot2::geom_histogram does not work
- # DEBUGGING
- # fun = "unique" ; arg = "x" ; val = list(x = list(1:10, c(1,1,2,8), NA)) ; expect.error = list(x = list(FALSE, FALSE, TRUE)) ; thread.nb = NULL ; plot.fun = FALSE ; export = FALSE ; res.path = "C:\\Users\\Gael\\Desktop\\" ; lib.path = NULL ; print.count = 1 ; cute.path = "C:\\Users\\Gael\\Documents\\Git_projects\\cute_little_R_functions\\cute_little_R_functions.R" # for function debugging
- # fun = "unique" ; arg = c("x", "incomparables") ; val = list(x = list(1:10, c(1,1,2,8), NA), incomparable = c(TRUE, FALSE, NA)) ; expect.error = NULL ; thread.nb = 2 ; plot.fun = FALSE ; export = TRUE ; res.path = "C:\\Users\\Gael\\Desktop\\" ; lib.path = NULL ; print.count = 10 ; cute.path = "C:\\Users\\Gael\\Documents\\Git_projects\\cute_little_R_functions\\cute_little_R_functions.R" # for function debugging
- # fun = "plot" ; arg = c("x", "y") ; val = list(x = list(1:10, 12:13, NA), y = list(1:10, NA, NA)) ; expect.error = list(x = list(FALSE, FALSE, TRUE, FALSE), y = list(FALSE, TRUE, TRUE)) ; print.count = 10 ; thread.nb = NULL ; plot.fun = TRUE ; export = TRUE ; res.path = "C:\\Users\\Gael\\Desktop\\" ; lib.path = NULL # for function debugging
- # set.seed(1) ; obs1 <- data.frame(Time = c(rnorm(10), rnorm(10) + 2), Group1 = rep(c("G", "H"), each = 10), stringsAsFactors = TRUE) ; fun = "fun_gg_boxplot" ; arg = c("data1", "y", "categ") ; val = list(L1 = list(L1 = obs1), L2 = list(L1 = "Time"), L3 = list(L1 = "Group1")) ; expect.error = NULL ; print.count = 10 ; thread.nb = NULL ; plot.fun = TRUE ; export = TRUE ; res.path = "C:\\Users\\Gael\\Desktop\\" ; lib.path = NULL # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- instruction <- match.call()
- # end function name
- # required function checking
- req.function <- c(
- "fun_check",
- "fun_get_message",
- "fun_pack"
- )
- for(i1 in req.function){
- if(length(find(i1, mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED ", i1, "() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # end required function checking
- # argument primary checking
- # arg with no default values
- if(any(missing(fun) | missing(arg) | missing(val))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ARGUMENTS fun, arg AND val HAVE NO DEFAULT VALUE AND REQUIRE ONE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end arg with no default values
- # using fun_check()
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = fun, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- if(grepl(x = fun, pattern = "()$")){ # remove ()
- fun <- sub(x = fun, pattern = "()$", replacement = "")
- }
- if( ! exists(fun)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CHARACTER STRING IN fun ARGUMENT DOES NOT EXIST IN THE R WORKING ENVIRONMENT: ", paste(fun, collapse = "\n"))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }else if( ! all(class(get(fun)) == "function")){ # here no env = sys.nframe(), inherit = FALSE for get() because fun is a function in the classical scope
- tempo.cat <- paste0("ERROR IN ", function.name, ": fun ARGUMENT IS NOT CLASS \"function\" BUT: ", paste(class(get(fun)), collapse = "\n"), "\nCHECK IF ANY CREATED OBJECT WOULD HAVE THE NAME OF THE TESTED FUNCTION")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- tempo <- fun_check(data = arg, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & length(arg) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": arg ARGUMENT CANNOT BE LENGTH 0")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- tempo <- fun_check(data = val, class = "list", fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- for(i2 in 1:length(val)){
- tempo1 <- fun_check(data = val[[i2]], class = "vector", na.contain = TRUE, fun.name = function.name)
- tempo2 <- fun_check(data = val[[i2]], class = "list", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": COMPARTMENT ", i2, " OF val ARGUMENT MUST BE A VECTOR OR A LIST")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }else if(tempo1$problem == FALSE){ # vector split into list compartments
- val[[i2]] <- split(x = val[[i2]], f = 1:length(val[[i2]]))
- }
- }
- }
- if( ! is.null(expect.error)){
- tempo <- fun_check(data = expect.error, class = "list", fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- for(i3 in 1:length(expect.error)){
- tempo1 <- fun_check(data = expect.error[[i3]], class = "vector", mode = "logical", fun.name = function.name)
- tempo2 <- fun_check(data = expect.error[[i3]], class = "list", fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": COMPARTMENT ", i3, " OF expect.error ARGUMENT MUST BE TRUE OR FALSE")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }else if(tempo1$problem == FALSE){ # vector split into list compartments
- expect.error[[i3]] <- split(x = expect.error[[i3]], f = 1:length(expect.error[[i3]]))
- }
- }
- }
- }
- if( ! is.null(thread.nb)){
- tempo <- fun_check(data = thread.nb, typeof = "integer", double.as.integer.allowed = TRUE, neg.values = FALSE, length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & thread.nb < 1){
- tempo.cat <- paste0("ERROR IN ", function.name, ": thread.nb PARAMETER MUST EQUAL OR GREATER THAN 1: ", thread.nb)
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- tempo <- fun_check(data = print.count, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = plot.fun, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = export, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(res.path)){
- tempo <- fun_check(data = res.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- if( ! all(dir.exists(res.path))){ # separation to avoid the problem of tempo$problem == FALSE and res.path == NA
- tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE res.path ARGUMENT DOES NOT EXISTS:\n", paste(res.path, collapse = "\n"))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- }
- if( ! is.null(lib.path)){
- tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
- tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- }
- if( ! is.null(thread.nb)){
- tempo <- fun_check(data = cute.path, class = "vector", typeof = "character", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- if( ! file.exists(cute.path)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": FILE PATH INDICATED IN THE cute.path PARAMETER DOES NOT EXISTS:\n", cute.path)
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- }
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # end using fun_check()
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument primary checking
- # second round of checking and data preparation
- # dealing with NA
- if(any(is.na(fun)) | any(is.na(arg)) | any(is.na(expect.error)) | any(is.na(thread.nb)) | any(is.na(print.count)) | any(is.na(plot.fun)) | any(is.na(export)) | any(is.na(res.path)) | any(is.na(lib.path))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": NO ARGUMENT EXCEPT val CAN HAVE NA VALUES")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end dealing with NA
- # dealing with NULL
- if(is.null(fun) | is.null(arg) | is.null(val) | is.null(print.count) | is.null(plot.fun) | is.null(export)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": THESE ARGUMENTS\nfun\narg\nval\nprint.count\nplot.fun\nexport\nCANNOT BE NULL") #problematic arguments are -> put everywhere
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end dealing with NULL
- if(length(arg) != length(val)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": LENGTH OF arg ARGUMENT MUST BE IDENTICAL TO LENGTH OF val ARGUMENT:\nHERE IT IS: ", length(arg), " VERSUS ", length(val))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
- }
- args <- names(formals(get(fun))) # here no env = sys.nframe(), inherit = FALSE for get() because fun is a function in the classical scope
- if( ! all(arg %in% args)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": SOME OF THE STRINGS IN arg ARE NOT ARGUMENTS OF fun\nfun ARGUMENTS: ", paste(args, collapse = " "),"\nPROBLEMATIC STRINGS IN arg: ", paste(arg[ ! arg %in% args], collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
- }
- if(sum(sapply(val, FUN = length) > 1) > 43){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CANNOT TEST MORE THAN 43 ARGUMENTS IF THEY ALL HAVE AT LEAST 2 VALUES EACH\nHERE THE NUMBER IS: ", sum(sapply(val, FUN = length) > 1))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
- }
- if( ! is.null(expect.error)){
- if(length(val) != length(expect.error)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": LENGTH OF val ARGUMENT MUST BE IDENTICAL TO LENGTH OF expect.error ARGUMENT:\nHERE IT IS: ", length(val), " VERSUS ", length(expect.error))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
- }
- }
- if( ! is.null(thread.nb) & is.null(res.path)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": res.path ARGUMENT MUST BE SPECIFIED IF thread.nb ARGUMENT IS NOT NULL")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
- }
- if(is.null(res.path) & export == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": res.path ARGUMENT MUST BE SPECIFIED IF export ARGUMENT TRUE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
- }
- if( ! is.null(thread.nb) & export == FALSE){
- export <- TRUE
- tempo.cat <- paste0("WARNING FROM ", function.name, ": export ARGUMENT CONVERTED TO TRUE BECAUSE thread.nb ARGUMENT IS NOT NULL")
- warning(paste0("\n", tempo.cat, "\n"), call. = FALSE)
- }
- # end second round of checking and data preparation
- # package checking
- fun_pack(req.package = c("lubridate"), lib.path = lib.path)
- if( ! is.null(thread.nb)){
- fun_pack(req.package = c("parallel"), lib.path = lib.path)
- }
- # end package checking
- # declaration of special plot functions
- sp.plot.fun <- c("fun_gg_scatter", "fun_gg_bar", "fun_gg_boxplot")
- # end declaration of special plot functions
- # main code
- ini.warning.length <- options()$warning.length
- warn <- NULL
- warn.count <- 0
- cat("\nfun_test JOB IGNITION\n")
- ini.date <- Sys.time()
- ini.time <- as.numeric(ini.date) # time of process begin, converted into seconds
- if(export == TRUE){
- res.path <- paste0(res.path, "/fun_test_res_", trunc(ini.time))
- if(dir.exists(res.path)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": FOLDER ALREADY EXISTS\n", res.path, "\nPLEASE RERUN ONCE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- dir.create(res.path)
- }
- }
- total.comp.nb <- prod(sapply(val, FUN = "length"))
- cat(paste0("\nTHE TOTAL NUMBER OF TESTS IS: ", total.comp.nb, "\n"))
- # creation of the txt instruction that includes several loops
- loop.string <- NULL
- end.loop.string <- NULL
- fun.args <- NULL
- fun.args2 <- NULL
- error.values <- NULL
- arg.values <- "list("
- for(i1 in 1:length(arg)){
- if(is.null(thread.nb)){
- if(length(val[[i1]]) > 1){ # loop only if more than one value in length(val[[i1]])
- loop.string <- paste0(loop.string, "for(i", i1, " in 1:", length(val[[i1]]), "){")
- end.loop.string <- paste0(end.loop.string, "}")
- }
- }else{
- loop.string <- "for(i in x){"
- end.loop.string <- "}"
- }
- fun.args <- paste0(
- fun.args,
- ifelse(i1 == 1, "", ", "),
- arg[i1],
- " = val[[",
- i1,
- "]][[",
- if(is.null(thread.nb)){
- if(length(val[[i1]]) > 1){
- paste0("i", i1)
- }else{
- "1" # a unique element in val[[i1]]
- }
- }else{
- paste0("i.list[[", i1, "]][i]")
- },
- "]]"
- )
- fun.args2 <- paste0(
- fun.args2,
- ifelse(i1 == 1, "", ", "),
- arg[i1],
- " = val[[",
- i1,
- "]][[', ",
- if(is.null(thread.nb)){
- if(length(val[[i1]]) > 1){
- paste0("i", i1)
- }else{
- "1" # a unique element in val[[i1]]
- }
- }else{
- paste0("i.list[[", i1, "]][i]")
- },
- ", ']]"
- )
- arg.values <- paste0(
- arg.values,
- "val[[", i1, "]][[",
- if(is.null(thread.nb)){
- if(length(val[[i1]]) > 1){
- paste0("i", i1)
- }else{
- "1" # a unique element in val[[i1]]
- }
- }else{
- paste0("i.list[[", i1, "]][i]")
- },
- "]]",
- ifelse(i1 == length(arg), "", ", ")
- )
- error.values <- paste0(
- error.values,
- ifelse(i1 == 1, "", " | "),
- "expect.error[[", i1, "]][[",
- if(is.null(thread.nb)){
- if(length(expect.error[[i1]]) > 1){
- paste0("i", i1)
- }else{
- "1" # a unique element in expect.error[[i1]]
- }
- }else{
- paste0("i.list[[", i1, "]][i]")
- },
- "]]"
- )
- }
- arg.values <- paste0(arg.values, ")")
- fun.test <- paste0(fun, "(", fun.args, ")")
- fun.test2 <- paste0("paste0('", fun, "(", fun.args2, ")')")
- # plot title for special plot functions
- if(plot.fun == TRUE){
- plot.kind <- "classic"
- if(fun %in% sp.plot.fun){
- plot.kind <- "special"
- if(any(arg %in% "title")){ # this is for the special functions
- tempo.match <- regmatches(x = fun.test, m = regexpr(text = fun.test, pattern = "title = .+[,)]"))
- tempo.match <- substring(tempo.match , 1, nchar(tempo.match) - 1)
- fun.test <- sub(x = fun.test, pattern = tempo.match, replacement = paste0(tempo.match, "\ntempo.title"))
- }else{
- fun.test <- sub(x = fun.test, pattern = ")$", replacement = ", title = tempo.title)")
- }
- }
- }
- # end plot title for special plot functions
- kind <- character()
- problem <- logical()
- expected.error <- logical()
- res <- character()
- count <- 0
- print.count.loop <- 0
- plot.count <- 0
- if(length(arg) == 1){
- data <- data.frame()
- }else{ # length(arg) == 0 already tested above
- data <- data.frame(t(vector("character", length(arg))), stringsAsFactors = FALSE)[-1, ] # -1 to remove the single row created and to have an empty data frame with length(arg) columns
- }
- code <- paste(
- loop.string, '
-count <- count + 1
-print.count.loop <- print.count.loop + 1
-data <- rbind(data, as.character(sapply(eval(parse(text = arg.values)), FUN = "paste", collapse = " ")), stringsAsFactors = FALSE) # each colum is a test
-tempo.try.error <- fun_get_message(data = eval(parse(text = fun.test2)), kind = "error", header = FALSE, env = get(env.name, env = sys.nframe(), inherit = FALSE)) # data argument needs a character string but eval(parse(text = fun.test2)) provides it (eval parse replace the i1, i2, etc., by the correct values, meaning that only val is required in the env.name environment)
-tempo.try.warning <- fun_get_message(data = eval(parse(text = fun.test2)), kind = "warning", header = FALSE, env = get(env.name, env = sys.nframe(), inherit = FALSE), print.no = TRUE) # data argument needs a character string but eval(parse(text = fun.test2)) provides it (eval parse replace the i1, i2, etc., by the correct values, meaning that only val is required in the env.name environment)
-if( ! is.null(expect.error)){
-expected.error <- c(expected.error, eval(parse(text = error.values)))
+# AIM
+# test combinations of argument values of a function
+# WARNINGS
+# Limited to 43 arguments with at least 2 values each. The total number of arguments tested can be more if the additional arguments have a single value. The limit is due to nested "for" loops (https://stat.ethz.ch/pipermail/r-help/2008-March/157341.html), but it should not be a problem since the number of tests would be 2^43 > 8e12
+# ARGUMENTS
+# fun: character string indicating the name of the function tested (without brackets)
+# arg: vector of character strings of arguments of fun. At least arguments that do not have default values must be present in this vector
+# val: list with number of compartments equal to length of arg, each compartment containing values of the corresponding argument in arg. Each different value must be in a list or in a vector. For instance, argument 3 in arg is a logical argument (values accepted TRUE, FALSE, NA). Thus, compartment 3 of val can be either list(TRUE, FALSE, NA), or c(TRUE, FALSE, NA)
+# expect.error: list of exactly the same structure as val argument, but containing FALSE or TRUE, depending on whether error is expected (TRUE) or not (FALSE) for each corresponding value of val. A message is returned depending on discrepancies between the expected and observed errors. BEWARE: not always possible to write the expected errors for all the combination of argument values. Ignored if NULL
+# thread.nb: numeric value indicating the number of available threads. Write NULL if no parallelization wanted
+# print.count: interger value. Print a working progress message every print.count during loops. BEWARE: can increase substentially the time to complete the process using a small value, like 10 for instance. Use Inf is no loop message desired
+# plot.fun: logical. Plot the plotting function tested for each test?
+# export: logical. Export the results into a .RData file and into a .txt file? If FALSE, return a list into the console (see below). BEWARE: will be automatically set to TRUE if thread.nb is not NULL. This means that when using parallelization, the results are systematically exported, not returned into the console
+# res.path: character string indicating the absolute pathway of folder where the txt results and pdfs, containing all the plots, will be saved. Several txt and pdf, one per thread, if parallelization. Ignored if export is FALSE. Must be specified if thread.nb is not NULL or if export is TRUE
+# lib.path: character vector specifying the absolute pathways of the directories containing the required packages if not in the default directories. Ignored if NULL
+# cute.path: character string indicating the absolute path of the cute.R file. Will be remove when cute will be a package. Not considered if thread.nb is NULL
+# REQUIRED PACKAGES
+# lubridate
+# parallel if thread.nb argument is not NULL (included in the R installation packages but not automatically loaded)
+# if the tested function is in a package, this package must be imported first (no parallelization) or must be in the classical R package folder indicated by the lib.path argument (parallelization)
+# RETURN
+# if export is FALSE a list containing:
+# $fun: the tested function
+# $data: a data frame of all the combination tested, containing the following columns:
+# the different values tested, named by arguments
+# $kind: a vector of character strings indicating the kind of test result: either "ERROR", or "WARNING", or "OK"
+# $problem: a logical vector indicating if error or not
+# $expected.error: optional logical vector indicating the expected error specified in the expect.error argument
+# $message: either NULL if $kind is always "OK", or the messages
+# $instruction: the initial instruction
+# $sys.info: system and packages info
+# if export is TRUE 1) the same list object into a .RData file, 2) also the $data data frame into a .txt file, and 3) if expect.error is non NULL and if any discrepancy, the $data data frame into a .txt file but containing only the rows with discrepancies between expected and observed errors
+# one or several pdf if a plotting function is tested and if the plot.fun argument is TRUE
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# fun_get_message()
+# fun_pack()
+# EXAMPLES
+# fun_test(fun = "unique", arg = c("x", "incomparables"), val = list(x = list(1:10, c(1,1,2,8), NA), incomparable = c(TRUE, FALSE, NA)))
+# fun_test(fun = "fun_round", arg = c("data", "dec.nb", "after.lead.zero"), val = list(L1 = list(c(1, 1.0002256, 1.23568), "a", NA), L2 = list(2, c(1,3), NA), L3 = c(TRUE, FALSE, NA)))
+# fun_test(fun = "plot", arg = c("x", "y"), val = list(x = list(1:10, 12:13, NA, (1:10)^2), y = list(1:10, NA, NA)), expect.error = list(x = list(FALSE, TRUE, TRUE, FALSE), y = list(FALSE, TRUE, TRUE)), thread.nb = NULL, plot.fun = TRUE, res.path = "C:\\Users\\Gael\\Desktop\\", lib.path = NULL)
+# fun_test(fun = "plot", arg = c("x", "y"), val = list(x = list(1:10, 12:13, NA, (1:10)^2), y = list(1:10, NA, NA)), thread.nb = 4, plot.fun = TRUE, res.path = "C:\\Users\\Gael\\Desktop\\", lib.path = "C:\\Program Files\\R\\R-4.0.2\\library\\")
+# set.seed(1) ; obs1 <- data.frame(Time = c(rnorm(10), rnorm(10) + 2), Group1 = rep(c("G", "H"), each = 10), stringsAsFactors = TRUE) ; fun_test(fun = "fun_gg_boxplot", arg = c("data1", "y", "categ"), val = list(L1 = list(L1 = obs1), L2 = list(L1 = "Time"), L3 = list(L1 = "Group1")))
+# set.seed(1) ; obs1 <- data.frame(Time = c(rnorm(10), rnorm(10) + 2), Group1 = rep(c("G", "H"), each = 10), stringsAsFactors = TRUE) ; fun_test(fun = "fun_gg_boxplot", arg = c("data1", "y", "categ"), val = list(L1 = list(obs1), L2 = "Time", L3 = "Group1"), thread.nb = NULL, plot.fun = TRUE, res.path = "C:\\Users\\Gael\\Desktop\\", lib.path = "C:\\Program Files\\R\\R-4.0.2\\library\\")
+# library(ggplot2) ; fun_test(fun = "geom_histogram", arg = c("data", "mapping"), val = list(x = list(data.frame(X = "a", stringsAsFactors = TRUE)), y = list(ggplot2::aes(x = X))), thread.nb = NULL, plot.fun = TRUE, res.path = "C:\\Users\\Gael\\Desktop\\", lib.path = "C:\\Program Files\\R\\R-4.0.2\\library\\") # BEWARE: ggplot2::geom_histogram does not work
+# DEBUGGING
+# fun = "unique" ; arg = "x" ; val = list(x = list(1:10, c(1,1,2,8), NA)) ; expect.error = list(x = list(FALSE, FALSE, TRUE)) ; thread.nb = NULL ; plot.fun = FALSE ; export = FALSE ; res.path = "C:\\Users\\Gael\\Desktop\\" ; lib.path = NULL ; print.count = 1 ; cute.path = "C:\\Users\\Gael\\Documents\\Git_projects\\cute_little_R_functions\\cute_little_R_functions.R" # for function debugging
+# fun = "unique" ; arg = c("x", "incomparables") ; val = list(x = list(1:10, c(1,1,2,8), NA), incomparable = c(TRUE, FALSE, NA)) ; expect.error = NULL ; thread.nb = 2 ; plot.fun = FALSE ; export = TRUE ; res.path = "C:\\Users\\Gael\\Desktop\\" ; lib.path = NULL ; print.count = 10 ; cute.path = "C:\\Users\\Gael\\Documents\\Git_projects\\cute_little_R_functions\\cute_little_R_functions.R" # for function debugging
+# fun = "plot" ; arg = c("x", "y") ; val = list(x = list(1:10, 12:13, NA), y = list(1:10, NA, NA)) ; expect.error = list(x = list(FALSE, FALSE, TRUE, FALSE), y = list(FALSE, TRUE, TRUE)) ; print.count = 10 ; thread.nb = NULL ; plot.fun = TRUE ; export = TRUE ; res.path = "C:\\Users\\Gael\\Desktop\\" ; lib.path = NULL # for function debugging
+# set.seed(1) ; obs1 <- data.frame(Time = c(rnorm(10), rnorm(10) + 2), Group1 = rep(c("G", "H"), each = 10), stringsAsFactors = TRUE) ; fun = "fun_gg_boxplot" ; arg = c("data1", "y", "categ") ; val = list(L1 = list(L1 = obs1), L2 = list(L1 = "Time"), L3 = list(L1 = "Group1")) ; expect.error = NULL ; print.count = 10 ; thread.nb = NULL ; plot.fun = TRUE ; export = TRUE ; res.path = "C:\\Users\\Gael\\Desktop\\" ; lib.path = NULL # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+instruction <- match.call()
+# end function name
+# required function checking
+req.function <- c(
+"fun_check",
+"fun_get_message",
+"fun_pack"
+)
+for(i1 in req.function){
+if(length(find(i1, mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED ", i1, "() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# end required function checking
+# argument primary checking
+# arg with no default values
+if(any(missing(fun) | missing(arg) | missing(val))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ARGUMENTS fun, arg AND val HAVE NO DEFAULT VALUE AND REQUIRE ONE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end arg with no default values
+# using fun_check()
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = fun, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+if(grepl(x = fun, pattern = "()$")){ # remove ()
+fun <- sub(x = fun, pattern = "()$", replacement = "")
+}
+if( ! exists(fun)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CHARACTER STRING IN fun ARGUMENT DOES NOT EXIST IN THE R WORKING ENVIRONMENT: ", paste(fun, collapse = "\n"))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}else if( ! all(class(get(fun)) == "function")){ # here no env = sys.nframe(), inherit = FALSE for get() because fun is a function in the classical scope
+tempo.cat <- paste0("ERROR IN ", function.name, ": fun ARGUMENT IS NOT CLASS \"function\" BUT: ", paste(class(get(fun)), collapse = "\n"), "\nCHECK IF ANY CREATED OBJECT WOULD HAVE THE NAME OF THE TESTED FUNCTION")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+tempo <- fun_check(data = arg, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & length(arg) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": arg ARGUMENT CANNOT BE LENGTH 0")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+tempo <- fun_check(data = val, class = "list", fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+for(i2 in 1:length(val)){
+tempo1 <- fun_check(data = val[[i2]], class = "vector", na.contain = TRUE, fun.name = function.name)
+tempo2 <- fun_check(data = val[[i2]], class = "list", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": COMPARTMENT ", i2, " OF val ARGUMENT MUST BE A VECTOR OR A LIST")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}else if(tempo1$problem == FALSE){ # vector split into list compartments
+val[[i2]] <- split(x = val[[i2]], f = 1:length(val[[i2]]))
+}
+}
+}
+if( ! is.null(expect.error)){
+tempo <- fun_check(data = expect.error, class = "list", fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+for(i3 in 1:length(expect.error)){
+tempo1 <- fun_check(data = expect.error[[i3]], class = "vector", mode = "logical", fun.name = function.name)
+tempo2 <- fun_check(data = expect.error[[i3]], class = "list", fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": COMPARTMENT ", i3, " OF expect.error ARGUMENT MUST BE TRUE OR FALSE")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}else if(tempo1$problem == FALSE){ # vector split into list compartments
+expect.error[[i3]] <- split(x = expect.error[[i3]], f = 1:length(expect.error[[i3]]))
+}
+}
+}
+}
+if( ! is.null(thread.nb)){
+tempo <- fun_check(data = thread.nb, typeof = "integer", double.as.integer.allowed = TRUE, neg.values = FALSE, length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & thread.nb < 1){
+tempo.cat <- paste0("ERROR IN ", function.name, ": thread.nb PARAMETER MUST EQUAL OR GREATER THAN 1: ", thread.nb)
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+tempo <- fun_check(data = print.count, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = plot.fun, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = export, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(res.path)){
+tempo <- fun_check(data = res.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+if( ! all(dir.exists(res.path))){ # separation to avoid the problem of tempo$problem == FALSE and res.path == NA
+tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE res.path ARGUMENT DOES NOT EXISTS:\n", paste(res.path, collapse = "\n"))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+}
+if( ! is.null(lib.path)){
+tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
+tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+}
+if( ! is.null(thread.nb)){
+tempo <- fun_check(data = cute.path, class = "vector", typeof = "character", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+if( ! file.exists(cute.path)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": FILE PATH INDICATED IN THE cute.path PARAMETER DOES NOT EXISTS:\n", cute.path)
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+}
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# end using fun_check()
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument primary checking
+# second round of checking and data preparation
+# dealing with NA
+if(any(is.na(fun)) | any(is.na(arg)) | any(is.na(expect.error)) | any(is.na(thread.nb)) | any(is.na(print.count)) | any(is.na(plot.fun)) | any(is.na(export)) | any(is.na(res.path)) | any(is.na(lib.path))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": NO ARGUMENT EXCEPT val CAN HAVE NA VALUES")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end dealing with NA
+# dealing with NULL
+if(is.null(fun) | is.null(arg) | is.null(val) | is.null(print.count) | is.null(plot.fun) | is.null(export)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": THESE ARGUMENTS\nfun\narg\nval\nprint.count\nplot.fun\nexport\nCANNOT BE NULL") #problematic arguments are -> put everywhere
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end dealing with NULL
+if(length(arg) != length(val)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": LENGTH OF arg ARGUMENT MUST BE IDENTICAL TO LENGTH OF val ARGUMENT:\nHERE IT IS: ", length(arg), " VERSUS ", length(val))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
+}
+args <- names(formals(get(fun))) # here no env = sys.nframe(), inherit = FALSE for get() because fun is a function in the classical scope
+if( ! all(arg %in% args)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": SOME OF THE STRINGS IN arg ARE NOT ARGUMENTS OF fun\nfun ARGUMENTS: ", paste(args, collapse = " "),"\nPROBLEMATIC STRINGS IN arg: ", paste(arg[ ! arg %in% args], collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
+}
+if(sum(sapply(val, FUN = length) > 1) > 43){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CANNOT TEST MORE THAN 43 ARGUMENTS IF THEY ALL HAVE AT LEAST 2 VALUES EACH\nHERE THE NUMBER IS: ", sum(sapply(val, FUN = length) > 1))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
+}
+if( ! is.null(expect.error)){
+if(length(val) != length(expect.error)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": LENGTH OF val ARGUMENT MUST BE IDENTICAL TO LENGTH OF expect.error ARGUMENT:\nHERE IT IS: ", length(val), " VERSUS ", length(expect.error))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
+}
+}
+if( ! is.null(thread.nb) & is.null(res.path)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": res.path ARGUMENT MUST BE SPECIFIED IF thread.nb ARGUMENT IS NOT NULL")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
+}
+if(is.null(res.path) & export == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": res.path ARGUMENT MUST BE SPECIFIED IF export ARGUMENT TRUE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
+}
+if( ! is.null(thread.nb) & export == FALSE){
+export <- TRUE
+tempo.cat <- paste0("WARNING FROM ", function.name, ": export ARGUMENT CONVERTED TO TRUE BECAUSE thread.nb ARGUMENT IS NOT NULL")
+warning(paste0("\n", tempo.cat, "\n"), call. = FALSE)
+}
+# end second round of checking and data preparation
+# package checking
+fun_pack(req.package = c("lubridate"), lib.path = lib.path)
+if( ! is.null(thread.nb)){
+fun_pack(req.package = c("parallel"), lib.path = lib.path)
+}
+# end package checking
+# declaration of special plot functions
+sp.plot.fun <- c("fun_gg_scatter", "fun_gg_bar", "fun_gg_boxplot")
+# end declaration of special plot functions
+# main code
+ini.warning.length <- options()$warning.length
+warn <- NULL
+warn.count <- 0
+cat("\nfun_test JOB IGNITION\n")
+ini.date <- Sys.time()
+ini.time <- as.numeric(ini.date) # time of process begin, converted into seconds
+if(export == TRUE){
+res.path <- paste0(res.path, "/fun_test_res_", trunc(ini.time))
+if(dir.exists(res.path)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": FOLDER ALREADY EXISTS\n", res.path, "\nPLEASE RERUN ONCE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+dir.create(res.path)
+}
+}
+total.comp.nb <- prod(sapply(val, FUN = "length"))
+cat(paste0("\nTHE TOTAL NUMBER OF TESTS IS: ", total.comp.nb, "\n"))
+# creation of the txt instruction that includes several loops
+loop.string <- NULL
+end.loop.string <- NULL
+fun.args <- NULL
+fun.args2 <- NULL
+error.values <- NULL
+arg.values <- "list("
+for(i1 in 1:length(arg)){
+if(is.null(thread.nb)){
+if(length(val[[i1]]) > 1){ # loop only if more than one value in length(val[[i1]])
+loop.string <- paste0(loop.string, "for(i", i1, " in 1:", length(val[[i1]]), "){")
+end.loop.string <- paste0(end.loop.string, "}")
+}
+}else{
+loop.string <- "for(i in x){"
+end.loop.string <- "}"
+}
+fun.args <- paste0(
+fun.args,
+ifelse(i1 == 1, "", ", "),
+arg[i1],
+" = val[[",
+i1,
+"]][[",
+if(is.null(thread.nb)){
+if(length(val[[i1]]) > 1){
+paste0("i", i1)
+}else{
+"1" # a unique element in val[[i1]]
+}
+}else{
+paste0("i.list[[", i1, "]][i]")
+},
+"]]"
+)
+fun.args2 <- paste0(
+fun.args2,
+ifelse(i1 == 1, "", ", "),
+arg[i1],
+" = val[[",
+i1,
+"]][[', ",
+if(is.null(thread.nb)){
+if(length(val[[i1]]) > 1){
+paste0("i", i1)
+}else{
+"1" # a unique element in val[[i1]]
+}
+}else{
+paste0("i.list[[", i1, "]][i]")
+},
+", ']]"
+)
+arg.values <- paste0(
+arg.values,
+"val[[", i1, "]][[",
+if(is.null(thread.nb)){
+if(length(val[[i1]]) > 1){
+paste0("i", i1)
+}else{
+"1" # a unique element in val[[i1]]
+}
+}else{
+paste0("i.list[[", i1, "]][i]")
+},
+"]]",
+ifelse(i1 == length(arg), "", ", ")
+)
+error.values <- paste0(
+error.values,
+ifelse(i1 == 1, "", " | "),
+"expect.error[[", i1, "]][[",
+if(is.null(thread.nb)){
+if(length(expect.error[[i1]]) > 1){
+paste0("i", i1)
+}else{
+"1" # a unique element in expect.error[[i1]]
+}
+}else{
+paste0("i.list[[", i1, "]][i]")
+},
+"]]"
+)
+}
+arg.values <- paste0(arg.values, ")")
+fun.test <- paste0(fun, "(", fun.args, ")")
+fun.test2 <- paste0("paste0('", fun, "(", fun.args2, ")')")
+# plot title for special plot functions
+if(plot.fun == TRUE){
+plot.kind <- "classic"
+if(fun %in% sp.plot.fun){
+plot.kind <- "special"
+if(any(arg %in% "title")){ # this is for the special functions
+tempo.match <- regmatches(x = fun.test, m = regexpr(text = fun.test, pattern = "title = .+[,)]"))
+tempo.match <- substring(tempo.match , 1, nchar(tempo.match) - 1)
+fun.test <- sub(x = fun.test, pattern = tempo.match, replacement = paste0(tempo.match, "\ntempo.title"))
+}else{
+fun.test <- sub(x = fun.test, pattern = ")$", replacement = ", title = tempo.title)")
+}
+}
+}
+# end plot title for special plot functions
+kind <- character()
+problem <- logical()
+expected.error <- logical()
+res <- character()
+count <- 0
+print.count.loop <- 0
+plot.count <- 0
+if(length(arg) == 1){
+data <- data.frame()
+}else{ # length(arg) == 0 already tested above
+data <- data.frame(t(vector("character", length(arg))), stringsAsFactors = FALSE)[-1, ] # -1 to remove the single row created and to have an empty data frame with length(arg) columns
+}
+code <- paste(
+loop.string, '
+count <- count + 1
+print.count.loop <- print.count.loop + 1
+data <- rbind(data, as.character(sapply(eval(parse(text = arg.values)), FUN = "paste", collapse = " ")), stringsAsFactors = FALSE) # each colum is a test
+tempo.try.error <- fun_get_message(data = eval(parse(text = fun.test2)), kind = "error", header = FALSE, env = get(env.name, env = sys.nframe(), inherit = FALSE)) # data argument needs a character string but eval(parse(text = fun.test2)) provides it (eval parse replace the i1, i2, etc., by the correct values, meaning that only val is required in the env.name environment)
+tempo.try.warning <- fun_get_message(data = eval(parse(text = fun.test2)), kind = "warning", header = FALSE, env = get(env.name, env = sys.nframe(), inherit = FALSE), print.no = TRUE) # data argument needs a character string but eval(parse(text = fun.test2)) provides it (eval parse replace the i1, i2, etc., by the correct values, meaning that only val is required in the env.name environment)
+if( ! is.null(expect.error)){
+expected.error <- c(expected.error, eval(parse(text = error.values)))
}
if( ! is.null(tempo.try.error)){
kind <- c(kind, "ERROR")
@@ -1911,239 +1910,239 @@ cat(paste0(ifelse(is.null(thread.nb), "\nLOOP PROCESS ENDED | ", paste0("\nPROCE
}
',
end.loop.string
- )
- # end creation of the txt instruction that includes several loops
- if( ! is.null(thread.nb)){
- # list of i numbers that will be split
- i.list <- vector("list", length(val)) # positions to split in parallel jobs
- for(i2 in 1:length(arg)){
- if(i2 == 1){
- tempo.divisor <- total.comp.nb / length(val[[i2]])
- i.list[[i2]] <- rep(1:length(val[[i2]]), each = as.integer(tempo.divisor))
- tempo.multi <- length(val[[i2]])
- }else{
- tempo.divisor <- tempo.divisor / length(val[[i2]])
- i.list[[i2]] <- rep(rep(1:length(val[[i2]]), each = as.integer(tempo.divisor)), time = as.integer(tempo.multi))
- tempo.multi <- tempo.multi * length(val[[i2]])
- }
- }
- # end list of i numbers that will be split
- tempo.cat <- paste0("PARALLELIZATION INITIATED AT: ", ini.date)
- cat(paste0("\n", tempo.cat, "\n"))
- tempo.thread.nb = parallel::detectCores(all.tests = FALSE, logical = TRUE) # detect the number of threads
- if(tempo.thread.nb < thread.nb){
- thread.nb <- tempo.thread.nb
- }
- tempo.cat <- paste0("NUMBER OF THREADS USED: ", thread.nb)
- cat(paste0("\n ", tempo.cat, "\n"))
- Clust <- parallel::makeCluster(thread.nb, outfile = paste0(res.path, "/fun_test_parall_log.txt")) # outfile to print or cat during parallelization (only possible in a file, outfile = "" do not work on windows)
- tempo.cat <- paste0("SPLIT OF TEST NUMBERS IN PARALLELISATION:")
- cat(paste0("\n ", tempo.cat, "\n"))
- cluster.list <- parallel::clusterSplit(Clust, 1:total.comp.nb) # split according to the number of cluster
- str(cluster.list) # using print(str()) add a NULL below the result
- cat("\n")
- paral.output.list <- parallel::clusterApply( # paral.output.list is a list made of thread.nb compartments, each made of n / thread.nb (mat theo column number) compartment. Each compartment receive the corresponding results of fun_permut(), i.e., data (permuted mat1.perm), warning message, cor (final correlation) and count (number of permutations)
- cl = Clust,
- x = cluster.list,
- function.name = function.name,
- instruction = instruction,
- thread.nb = thread.nb,
- print.count = print.count,
- total.comp.nb = total.comp.nb,
- sp.plot.fun = sp.plot.fun,
- i.list = i.list,
- fun.tested = fun,
- arg.values = arg.values,
- fun.test = fun.test,
- fun.test2 = fun.test2,
- kind = kind,
- problem = problem,
- res = res,
- count = count,
- plot.count = plot.count,
- data = data,
- code = code,
- plot.fun = plot.fun,
- res.path = res.path,
- lib.path = lib.path,
- cute.path = cute.path,
- fun = function(
- x,
- function.name,
- instruction,
- thread.nb,
- print.count,
- total.comp.nb,
- sp.plot.fun,
- i.list,
- fun.tested,
- arg.values,
- fun.test,
- fun.test2,
- kind,
- problem,
- res,
- count,
- plot.count,
- data,
- code,
- plot.fun,
- res.path,
- lib.path,
- cute.path
- ){
- # check again: very important because another R
- process.id <- Sys.getpid()
- cat(paste0("\nPROCESS ID ", process.id, " -> TESTS ", x[1], " TO ", x[length(x)], "\n"))
- source(cute.path, local = .GlobalEnv)
- fun_pack(req.package = "lubridate", lib.path = lib.path, load = TRUE) # load = TRUE to be sure that functions are present in the environment. And this prevent to use R.lib.path argument of fun_python_pack()
- # end check again: very important because another R
- # plot management
- if(plot.fun == TRUE){
- pdf(file = paste0(res.path, "/plots_from_fun_test_", x[1], ifelse(length(x) == 1, ".pdf", paste0("-", x[length(x)], ".pdf"))))
- }else{
- pdf(file = NULL) # send plots into a NULL file, no pdf file created
- }
- window.nb <- dev.cur()
- invisible(dev.set(window.nb))
- # end plot management
- # new environment
- env.name <- paste0("env", ini.time)
- if(exists(env.name, where = -1)){ # verify if still ok when fun_test() is inside a function
- tempo.cat <- paste0("ERROR IN ", function.name, ": ENVIRONMENT env.name ALREADY EXISTS. PLEASE RERUN ONCE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- assign(env.name, new.env())
- assign("val", val, envir = get(env.name, env = sys.nframe(), inherit = FALSE)) # var replaced by val
- }
- # end new environment
- ini.date <- Sys.time()
- ini.time <- as.numeric(ini.date) # time of process begin, converted into
- print.count.loop <- 0
- suppressMessages(suppressWarnings(eval(parse(text = code))))
- colnames(data) <- arg
- if( ! is.null(expect.error)){
- data <- data.frame(data, kind = kind, problem = problem, expected.error = expected.error, message = res, stringsAsFactors = FALSE)
- }else{
- data <- data.frame(data, kind = kind, problem = problem, message = res, stringsAsFactors = FALSE)
- }
- row.names(data) <- paste0("test_", sprintf(paste0("%0", nchar(total.comp.nb), "d"), x))
- sys.info <- sessionInfo()
- sys.info$loadedOnly <- sys.info$loadedOnly[order(names(sys.info$loadedOnly))] # sort the packages
- invisible(dev.off(window.nb))
- rm(env.name) # optional, because should disappear at the end of the function execution
- # output
- output <- list(fun = fun, data = data, instruction = instruction, sys.info = sys.info)
- save(output, file = paste0(res.path, "/fun_test_", x[1], ifelse(length(x) == 1, ".RData", paste0("-", x[length(x)], ".RData"))))
- if(plot.fun == TRUE & plot.count == 0){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") IN PROCESS ", process.id, ": NO PDF PLOT BECAUSE ONLY ERRORS REPORTED")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- file.remove(paste0(res.path, "/plots_from_fun_test_", x[1], ifelse(length(x) == 1, ".pdf", paste0("-", x[length(x)], ".pdf"))))
- }
- table.out <- as.matrix(output$data)
- # table.out[table.out == ""] <- " " # does not work # because otherwise read.table() converts "" into NA
- table.out <- gsub(table.out, pattern = "\n", replacement = " ")
- write.table(table.out, file = paste0(res.path, "/table_from_fun_test_", x[1], ifelse(length(x) == 1, ".txt", paste0("-", x[length(x)], ".txt"))), row.names = TRUE, col.names = NA, append = FALSE, quote = FALSE, sep = "\t", eol = "\n")
- }
- )
- parallel::stopCluster(Clust)
- # txt files assembly
- if(length(cluster.list) > 1){
- for(i2 in 1:length(cluster.list)){
- tempo.name <- paste0(res.path, "/table_from_fun_test_", min(cluster.list[[i2]], na.rm = TRUE), ifelse(length(cluster.list[[i2]]) == 1, ".txt", paste0("-", max(cluster.list[[i2]], na.rm = TRUE), ".txt")))
- tempo <- read.table(file = tempo.name, header = TRUE, stringsAsFactors = FALSE, sep = "\t", row.names = 1, comment.char = "", colClasses = "character") # row.names = 1 (1st column) because now read.table() adds a NA in the header if the header starts by a tabulation, comment.char = "" because colors with #, colClasses = "character" otherwise convert "" (from NULL) into NA
- file.remove(tempo.name)
- if(i2 == 1){
- final.file <- tempo
- }else{
- final.file <- rbind(final.file, tempo, stringsAsFactors = TRUE)
- }
- }
- write.table(final.file, file = paste0(res.path, "/table_from_fun_test_1-", total.comp.nb, ".txt"), row.names = TRUE, col.names = NA, append = FALSE, quote = FALSE, sep = "\t", eol = "\n")
- if( ! is.null(expect.error)){
- final.file <- final.file[ ! final.file$problem == final.file$expected.error, ]
- if(nrow(final.file) == 0){
- cat(paste0("NO DISCREPANCY BETWEEN EXPECTED AND OBSERVED ERRORS\n\n"))
- }else{
- cat(paste0("DISCREPANCIES BETWEEN EXPECTED AND OBSERVED ERRORS (SEE THE discrepancy_table_from_fun_test_1-", total.comp.nb, ".txt FILE)\n\n"))
- write.table(final.file, file = paste0(res.path, "/discrepancy_table_from_fun_test_1-", total.comp.nb, ".txt"), row.names = TRUE, col.names = NA, append = FALSE, quote = FALSE, sep = "\t", eol = "\n")
- }
- }
- }
- # end txt files assembly
- }else{
- # plot management
- if(plot.fun == TRUE){
- pdf(file = paste0(res.path, "/plots_from_fun_test_1", ifelse(total.comp.nb == 1, ".pdf", paste0("-", total.comp.nb, ".pdf"))))
- }else{
- pdf(file = NULL) # send plots into a NULL file, no pdf file created
- }
- window.nb <- dev.cur()
- invisible(dev.set(window.nb))
- # end plot management
- # new environment
- env.name <- paste0("env", ini.time)
- if(exists(env.name, where = -1)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ENVIRONMENT env.name ALREADY EXISTS. PLEASE RERUN ONCE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- assign(env.name, new.env())
- assign("val", val, envir = get(env.name, env = sys.nframe(), inherit = FALSE)) # var replaced by val
- }
- # end new environment
- suppressMessages(suppressWarnings(eval(parse(text = code))))
- colnames(data) <- arg
- expect.data <- data.frame()
- if( ! is.null(expect.error)){
- data <- data.frame(data, kind = kind, problem = problem, expected.error = expected.error, message = res, stringsAsFactors = FALSE)
- }else{
- data <- data.frame(data, kind = kind, problem = problem, message = res, stringsAsFactors = FALSE)
- }
- row.names(data) <- paste0("test_", sprintf(paste0("%0", nchar(total.comp.nb), "d"), 1:total.comp.nb))
- sys.info <- sessionInfo()
- sys.info$loadedOnly <- sys.info$loadedOnly[order(names(sys.info$loadedOnly))] # sort the packages
- invisible(dev.off(window.nb))
- rm(env.name) # optional, because should disappear at the end of the function execution
- # output
- output <- list(fun = fun, data = data, instruction = instruction, sys.info = sys.info)
- if(plot.fun == TRUE & plot.count == 0){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NO PDF PLOT BECAUSE ONLY ERRORS REPORTED")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- file.remove(paste0(res.path, "/plots_from_fun_test_1", ifelse(total.comp.nb == 1, ".pdf", paste0("-", total.comp.nb, ".pdf"))))
- }
- if( ! is.null(expect.error)){
- expect.data <- output$data[ ! output$data$problem == output$data$expected.error, ]
- if(nrow(expect.data) == 0){
- cat(paste0("NO DISCREPANCY BETWEEN EXPECTED AND OBSERVED ERRORS\n\n"))
- }else{
- cat(paste0("DISCREPANCIES BETWEEN EXPECTED AND OBSERVED ERRORS (SEE THE ", if(export == TRUE){paste0("discrepancy_table_from_fun_test_1", ifelse(total.comp.nb == 1, "", paste0("-", total.comp.nb)), ".txt FILE")}else{"$data RESULT"}, ")\n\n"))
- if(export == TRUE){
- expect.data <- as.matrix(expect.data)
- expect.data <- gsub(expect.data, pattern = "\n", replacement = " ")
- write.table(expect.data, file = paste0(res.path, "/discrepancy_table_from_fun_test_1", ifelse(total.comp.nb == 1, ".txt", paste0("-", total.comp.nb, ".txt"))), row.names = TRUE, col.names = NA, append = FALSE, quote = FALSE, sep = "\t", eol = "\n")
- }
- }
- }
- if( ! is.null(warn)){
- options(warning.length = 8170)
- on.exit(warning(paste0("FROM ", function.name, ":\n\n", warn), call. = FALSE))
- on.exit(exp = options(warning.length = ini.warning.length), add = TRUE)
- }
- if(export == TRUE){
- save(output, file = paste0(res.path, "/fun_test_1", ifelse(total.comp.nb == 1, ".RData", paste0("-", total.comp.nb, ".RData"))))
- table.out <- as.matrix(output$data)
- table.out <- gsub(table.out, pattern = "\n", replacement = " ")
- write.table(table.out, file = paste0(res.path, "/table_from_fun_test_1", ifelse(total.comp.nb == 1, ".txt", paste0("-", total.comp.nb, ".txt"))), row.names = TRUE, col.names = NA, append = FALSE, quote = FALSE, sep = "\t", eol = "\n")
- }else{
- return(output)
- }
- }
- end.date <- Sys.time()
- end.time <- as.numeric(end.date)
- total.lapse <- round(lubridate::seconds_to_period(end.time - ini.time))
- cat(paste0("fun_test JOB END\n\nTIME: ", end.date, "\n\nTOTAL TIME LAPSE: ", total.lapse, "\n\n\n"))
+)
+# end creation of the txt instruction that includes several loops
+if( ! is.null(thread.nb)){
+# list of i numbers that will be split
+i.list <- vector("list", length(val)) # positions to split in parallel jobs
+for(i2 in 1:length(arg)){
+if(i2 == 1){
+tempo.divisor <- total.comp.nb / length(val[[i2]])
+i.list[[i2]] <- rep(1:length(val[[i2]]), each = as.integer(tempo.divisor))
+tempo.multi <- length(val[[i2]])
+}else{
+tempo.divisor <- tempo.divisor / length(val[[i2]])
+i.list[[i2]] <- rep(rep(1:length(val[[i2]]), each = as.integer(tempo.divisor)), time = as.integer(tempo.multi))
+tempo.multi <- tempo.multi * length(val[[i2]])
+}
+}
+# end list of i numbers that will be split
+tempo.cat <- paste0("PARALLELIZATION INITIATED AT: ", ini.date)
+cat(paste0("\n", tempo.cat, "\n"))
+tempo.thread.nb = parallel::detectCores(all.tests = FALSE, logical = TRUE) # detect the number of threads
+if(tempo.thread.nb < thread.nb){
+thread.nb <- tempo.thread.nb
+}
+tempo.cat <- paste0("NUMBER OF THREADS USED: ", thread.nb)
+cat(paste0("\n ", tempo.cat, "\n"))
+Clust <- parallel::makeCluster(thread.nb, outfile = paste0(res.path, "/fun_test_parall_log.txt")) # outfile to print or cat during parallelization (only possible in a file, outfile = "" do not work on windows)
+tempo.cat <- paste0("SPLIT OF TEST NUMBERS IN PARALLELISATION:")
+cat(paste0("\n ", tempo.cat, "\n"))
+cluster.list <- parallel::clusterSplit(Clust, 1:total.comp.nb) # split according to the number of cluster
+str(cluster.list) # using print(str()) add a NULL below the result
+cat("\n")
+paral.output.list <- parallel::clusterApply( # paral.output.list is a list made of thread.nb compartments, each made of n / thread.nb (mat theo column number) compartment. Each compartment receive the corresponding results of fun_permut(), i.e., data (permuted mat1.perm), warning message, cor (final correlation) and count (number of permutations)
+cl = Clust,
+x = cluster.list,
+function.name = function.name,
+instruction = instruction,
+thread.nb = thread.nb,
+print.count = print.count,
+total.comp.nb = total.comp.nb,
+sp.plot.fun = sp.plot.fun,
+i.list = i.list,
+fun.tested = fun,
+arg.values = arg.values,
+fun.test = fun.test,
+fun.test2 = fun.test2,
+kind = kind,
+problem = problem,
+res = res,
+count = count,
+plot.count = plot.count,
+data = data,
+code = code,
+plot.fun = plot.fun,
+res.path = res.path,
+lib.path = lib.path,
+cute.path = cute.path,
+fun = function(
+x,
+function.name,
+instruction,
+thread.nb,
+print.count,
+total.comp.nb,
+sp.plot.fun,
+i.list,
+fun.tested,
+arg.values,
+fun.test,
+fun.test2,
+kind,
+problem,
+res,
+count,
+plot.count,
+data,
+code,
+plot.fun,
+res.path,
+lib.path,
+cute.path
+){
+# check again: very important because another R
+process.id <- Sys.getpid()
+cat(paste0("\nPROCESS ID ", process.id, " -> TESTS ", x[1], " TO ", x[length(x)], "\n"))
+source(cute.path, local = .GlobalEnv)
+fun_pack(req.package = "lubridate", lib.path = lib.path, load = TRUE) # load = TRUE to be sure that functions are present in the environment. And this prevent to use R.lib.path argument of fun_python_pack()
+# end check again: very important because another R
+# plot management
+if(plot.fun == TRUE){
+pdf(file = paste0(res.path, "/plots_from_fun_test_", x[1], ifelse(length(x) == 1, ".pdf", paste0("-", x[length(x)], ".pdf"))))
+}else{
+pdf(file = NULL) # send plots into a NULL file, no pdf file created
+}
+window.nb <- dev.cur()
+invisible(dev.set(window.nb))
+# end plot management
+# new environment
+env.name <- paste0("env", ini.time)
+if(exists(env.name, where = -1)){ # verify if still ok when fun_test() is inside a function
+tempo.cat <- paste0("ERROR IN ", function.name, ": ENVIRONMENT env.name ALREADY EXISTS. PLEASE RERUN ONCE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+assign(env.name, new.env())
+assign("val", val, envir = get(env.name, env = sys.nframe(), inherit = FALSE)) # var replaced by val
+}
+# end new environment
+ini.date <- Sys.time()
+ini.time <- as.numeric(ini.date) # time of process begin, converted into
+print.count.loop <- 0
+suppressMessages(suppressWarnings(eval(parse(text = code))))
+colnames(data) <- arg
+if( ! is.null(expect.error)){
+data <- data.frame(data, kind = kind, problem = problem, expected.error = expected.error, message = res, stringsAsFactors = FALSE)
+}else{
+data <- data.frame(data, kind = kind, problem = problem, message = res, stringsAsFactors = FALSE)
+}
+row.names(data) <- paste0("test_", sprintf(paste0("%0", nchar(total.comp.nb), "d"), x))
+sys.info <- sessionInfo()
+sys.info$loadedOnly <- sys.info$loadedOnly[order(names(sys.info$loadedOnly))] # sort the packages
+invisible(dev.off(window.nb))
+rm(env.name) # optional, because should disappear at the end of the function execution
+# output
+output <- list(fun = fun, data = data, instruction = instruction, sys.info = sys.info)
+save(output, file = paste0(res.path, "/fun_test_", x[1], ifelse(length(x) == 1, ".RData", paste0("-", x[length(x)], ".RData"))))
+if(plot.fun == TRUE & plot.count == 0){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") IN PROCESS ", process.id, ": NO PDF PLOT BECAUSE ONLY ERRORS REPORTED")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+file.remove(paste0(res.path, "/plots_from_fun_test_", x[1], ifelse(length(x) == 1, ".pdf", paste0("-", x[length(x)], ".pdf"))))
+}
+table.out <- as.matrix(output$data)
+# table.out[table.out == ""] <- " " # does not work # because otherwise read.table() converts "" into NA
+table.out <- gsub(table.out, pattern = "\n", replacement = " ")
+write.table(table.out, file = paste0(res.path, "/table_from_fun_test_", x[1], ifelse(length(x) == 1, ".txt", paste0("-", x[length(x)], ".txt"))), row.names = TRUE, col.names = NA, append = FALSE, quote = FALSE, sep = "\t", eol = "\n")
+}
+)
+parallel::stopCluster(Clust)
+# txt files assembly
+if(length(cluster.list) > 1){
+for(i2 in 1:length(cluster.list)){
+tempo.name <- paste0(res.path, "/table_from_fun_test_", min(cluster.list[[i2]], na.rm = TRUE), ifelse(length(cluster.list[[i2]]) == 1, ".txt", paste0("-", max(cluster.list[[i2]], na.rm = TRUE), ".txt")))
+tempo <- read.table(file = tempo.name, header = TRUE, stringsAsFactors = FALSE, sep = "\t", row.names = 1, comment.char = "", colClasses = "character") # row.names = 1 (1st column) because now read.table() adds a NA in the header if the header starts by a tabulation, comment.char = "" because colors with #, colClasses = "character" otherwise convert "" (from NULL) into NA
+file.remove(tempo.name)
+if(i2 == 1){
+final.file <- tempo
+}else{
+final.file <- rbind(final.file, tempo, stringsAsFactors = TRUE)
+}
+}
+write.table(final.file, file = paste0(res.path, "/table_from_fun_test_1-", total.comp.nb, ".txt"), row.names = TRUE, col.names = NA, append = FALSE, quote = FALSE, sep = "\t", eol = "\n")
+if( ! is.null(expect.error)){
+final.file <- final.file[ ! final.file$problem == final.file$expected.error, ]
+if(nrow(final.file) == 0){
+cat(paste0("NO DISCREPANCY BETWEEN EXPECTED AND OBSERVED ERRORS\n\n"))
+}else{
+cat(paste0("DISCREPANCIES BETWEEN EXPECTED AND OBSERVED ERRORS (SEE THE discrepancy_table_from_fun_test_1-", total.comp.nb, ".txt FILE)\n\n"))
+write.table(final.file, file = paste0(res.path, "/discrepancy_table_from_fun_test_1-", total.comp.nb, ".txt"), row.names = TRUE, col.names = NA, append = FALSE, quote = FALSE, sep = "\t", eol = "\n")
+}
+}
+}
+# end txt files assembly
+}else{
+# plot management
+if(plot.fun == TRUE){
+pdf(file = paste0(res.path, "/plots_from_fun_test_1", ifelse(total.comp.nb == 1, ".pdf", paste0("-", total.comp.nb, ".pdf"))))
+}else{
+pdf(file = NULL) # send plots into a NULL file, no pdf file created
+}
+window.nb <- dev.cur()
+invisible(dev.set(window.nb))
+# end plot management
+# new environment
+env.name <- paste0("env", ini.time)
+if(exists(env.name, where = -1)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ENVIRONMENT env.name ALREADY EXISTS. PLEASE RERUN ONCE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+assign(env.name, new.env())
+assign("val", val, envir = get(env.name, env = sys.nframe(), inherit = FALSE)) # var replaced by val
+}
+# end new environment
+suppressMessages(suppressWarnings(eval(parse(text = code))))
+colnames(data) <- arg
+expect.data <- data.frame()
+if( ! is.null(expect.error)){
+data <- data.frame(data, kind = kind, problem = problem, expected.error = expected.error, message = res, stringsAsFactors = FALSE)
+}else{
+data <- data.frame(data, kind = kind, problem = problem, message = res, stringsAsFactors = FALSE)
+}
+row.names(data) <- paste0("test_", sprintf(paste0("%0", nchar(total.comp.nb), "d"), 1:total.comp.nb))
+sys.info <- sessionInfo()
+sys.info$loadedOnly <- sys.info$loadedOnly[order(names(sys.info$loadedOnly))] # sort the packages
+invisible(dev.off(window.nb))
+rm(env.name) # optional, because should disappear at the end of the function execution
+# output
+output <- list(fun = fun, data = data, instruction = instruction, sys.info = sys.info)
+if(plot.fun == TRUE & plot.count == 0){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NO PDF PLOT BECAUSE ONLY ERRORS REPORTED")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+file.remove(paste0(res.path, "/plots_from_fun_test_1", ifelse(total.comp.nb == 1, ".pdf", paste0("-", total.comp.nb, ".pdf"))))
+}
+if( ! is.null(expect.error)){
+expect.data <- output$data[ ! output$data$problem == output$data$expected.error, ]
+if(nrow(expect.data) == 0){
+cat(paste0("NO DISCREPANCY BETWEEN EXPECTED AND OBSERVED ERRORS\n\n"))
+}else{
+cat(paste0("DISCREPANCIES BETWEEN EXPECTED AND OBSERVED ERRORS (SEE THE ", if(export == TRUE){paste0("discrepancy_table_from_fun_test_1", ifelse(total.comp.nb == 1, "", paste0("-", total.comp.nb)), ".txt FILE")}else{"$data RESULT"}, ")\n\n"))
+if(export == TRUE){
+expect.data <- as.matrix(expect.data)
+expect.data <- gsub(expect.data, pattern = "\n", replacement = " ")
+write.table(expect.data, file = paste0(res.path, "/discrepancy_table_from_fun_test_1", ifelse(total.comp.nb == 1, ".txt", paste0("-", total.comp.nb, ".txt"))), row.names = TRUE, col.names = NA, append = FALSE, quote = FALSE, sep = "\t", eol = "\n")
+}
+}
+}
+if( ! is.null(warn)){
+options(warning.length = 8170)
+on.exit(warning(paste0("FROM ", function.name, ":\n\n", warn), call. = FALSE))
+on.exit(exp = options(warning.length = ini.warning.length), add = TRUE)
+}
+if(export == TRUE){
+save(output, file = paste0(res.path, "/fun_test_1", ifelse(total.comp.nb == 1, ".RData", paste0("-", total.comp.nb, ".RData"))))
+table.out <- as.matrix(output$data)
+table.out <- gsub(table.out, pattern = "\n", replacement = " ")
+write.table(table.out, file = paste0(res.path, "/table_from_fun_test_1", ifelse(total.comp.nb == 1, ".txt", paste0("-", total.comp.nb, ".txt"))), row.names = TRUE, col.names = NA, append = FALSE, quote = FALSE, sep = "\t", eol = "\n")
+}else{
+return(output)
+}
+}
+end.date <- Sys.time()
+end.time <- as.numeric(end.date)
+total.lapse <- round(lubridate::seconds_to_period(end.time - ini.time))
+cat(paste0("fun_test JOB END\n\nTIME: ", end.date, "\n\nTOTAL TIME LAPSE: ", total.lapse, "\n\n\n"))
}
@@ -2155,74 +2154,74 @@ end.loop.string
# Check OK: clear to go Apollo
fun_name_change <- function(data1, data2, added.string = "_modif"){
- # AIM
- # this function allow to check if a vector of character strings, like column names of a data frame, has elements present in another vector (vector of reserved words or column names of another data frame before merging)
- # ARGUMENTS
- # data1: vector of character strings to check and modify
- # data2: reference vector of character strings
- # added.string: string added at the end of the modified string in data1 if present in data2
- # RETURN
- # a list containing
- # $data: the modified data1 (in the same order as in the initial data1)
- # $ini: the initial elements before modification. NULL if no modification
- # $post: the modified elements in the same order as in ini. NULL if no modification
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # obs1 <- c("A", "B", "C", "D") ; obs2 <- c("A", "C") ; fun_name_change(obs1, obs2)
- # obs1 <- c("A", "B", "C", "C_modif1", "D") ; obs2 <- c("A", "A_modif1", "C") ; fun_name_change(obs1, obs2) # the function checks that the new names are neither in obs1 nor in obs2 (increment the number after the added string)
- # DEBUGGING
- # data1 = c("A", "B", "C", "D") ; data2 <- c("A", "C") ; added.string = "_modif" # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = data1, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = data2, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = added.string, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- ini <- NULL
- post <- NULL
- if(any(data1 %in% data2)){
- tempo.names <- data1[data1 %in% data2]
- ini <- NULL
- post <- NULL
- for(i2 in 1:length(tempo.names)){
- count <- 0
- tempo <- tempo.names[i2]
- while(any(tempo %in% data2) | any(tempo %in% data1)){
- count <- count + 1
- tempo <- paste0(tempo.names[i2], "_modif", count)
- }
- data1[data1 %in% tempo.names[i2]] <- paste0(tempo.names[i2], "_modif", count)
- if(count != 0){
- ini <- c(ini, tempo.names[i2])
- post <- c(post, paste0(tempo.names[i2], "_modif", count))
- }
- }
- data <- data1
- }else{
- data <- data1
- }
- output <- list(data = data, ini = ini, post = post)
- return(output)
+# AIM
+# this function allow to check if a vector of character strings, like column names of a data frame, has elements present in another vector (vector of reserved words or column names of another data frame before merging)
+# ARGUMENTS
+# data1: vector of character strings to check and modify
+# data2: reference vector of character strings
+# added.string: string added at the end of the modified string in data1 if present in data2
+# RETURN
+# a list containing
+# $data: the modified data1 (in the same order as in the initial data1)
+# $ini: the initial elements before modification. NULL if no modification
+# $post: the modified elements in the same order as in ini. NULL if no modification
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# obs1 <- c("A", "B", "C", "D") ; obs2 <- c("A", "C") ; fun_name_change(obs1, obs2)
+# obs1 <- c("A", "B", "C", "C_modif1", "D") ; obs2 <- c("A", "A_modif1", "C") ; fun_name_change(obs1, obs2) # the function checks that the new names are neither in obs1 nor in obs2 (increment the number after the added string)
+# DEBUGGING
+# data1 = c("A", "B", "C", "D") ; data2 <- c("A", "C") ; added.string = "_modif" # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = data1, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = data2, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = added.string, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+ini <- NULL
+post <- NULL
+if(any(data1 %in% data2)){
+tempo.names <- data1[data1 %in% data2]
+ini <- NULL
+post <- NULL
+for(i2 in 1:length(tempo.names)){
+count <- 0
+tempo <- tempo.names[i2]
+while(any(tempo %in% data2) | any(tempo %in% data1)){
+count <- count + 1
+tempo <- paste0(tempo.names[i2], "_modif", count)
+}
+data1[data1 %in% tempo.names[i2]] <- paste0(tempo.names[i2], "_modif", count)
+if(count != 0){
+ini <- c(ini, tempo.names[i2])
+post <- c(post, paste0(tempo.names[i2], "_modif", count))
+}
+}
+data <- data1
+}else{
+data <- data1
+}
+output <- list(data = data, ini = ini, post = post)
+return(output)
}
@@ -2231,123 +2230,123 @@ fun_name_change <- function(data1, data2, added.string = "_modif"){
# Check OK: clear to go Apollo
fun_df_remod <- function(
- data,
- quanti.col.name = "quanti",
- quali.col.name = "quali"
+data,
+quanti.col.name = "quanti",
+quali.col.name = "quali"
){
- # AIM
- # if the data frame is made of numeric columns, a new data frame is created, with the 1st column gathering all the numeric values, and the 2nd column being the name of the columns of the initial data frame. If row names were present in the initial data frame, then a new ini_rowname column is added with the names of the rows
-
-
- # If the data frame is made of one numeric column and one character or factor column, a new data frame is created, with the new columns corresponding to the split numeric values (according to the character column). NA are added a the end of each column to have the same number of rows. BEWARE: in such data frame, rows are not individuals. This means that in the example below, values 10 and 20 are associated on the same row but that means nothing in term of association
-
-
-
- # ARGUMENTS
- # data: data frame to convert
- # quanti.col.name: optional name for the quanti column of the new data frame
- # quali.col.name: optional name for the quali column of the new data frame
- # RETURN
- # the modified data frame
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # obs <- data.frame(col1 = (1:4)*10, col2 = c("A", "B", "A", "A"), stringsAsFactors = TRUE) ; obs ; fun_df_remod(obs)
- # obs <- data.frame(col1 = (1:4)*10, col2 = 5:8, stringsAsFactors = TRUE) ; obs ; fun_df_remod(obs, quanti.col.name = "quanti", quali.col.name = "quali")
- # obs <- data.frame(col1 = (1:4)*10, col2 = 5:8, stringsAsFactors = TRUE) ; rownames(obs) <- paste0("row", 1:4) ; obs ; fun_df_remod(obs, quanti.col.name = "quanti", quali.col.name = "quali")
- # DEBUGGING
- # data = data.frame(a = 1:3, b = 4:6, stringsAsFactors = TRUE) ; quanti.col.name = "quanti" ; quali.col.name = "quali" # for function debugging
- # data = data.frame(a = 1:3, b = 4:6, c = 11:13, stringsAsFactors = TRUE) ; quanti.col.name = "quanti" ; quali.col.name = "quali" # for function debugging
- # data = data.frame(a = 1:3, b = letters[1:3], stringsAsFactors = TRUE) ; quanti.col.name = "quanti" ; quali.col.name = "quali" # for function debugging
- # data = data.frame(a = 1:3, b = letters[1:3], stringsAsFactors = TRUE) ; quanti.col.name = "TEST" ; quali.col.name = "quali" # for function debugging
- # data = data.frame(b = letters[1:3], a = 1:3, stringsAsFactors = TRUE) ; quanti.col.name = "quanti" ; quali.col.name = "quali" # for function debugging
- # data = data.frame(b = c("e", "e", "h"), a = 1:3, stringsAsFactors = TRUE) ; quanti.col.name = "quanti" ; quali.col.name = "quali" # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- # argument checking without fun_check()
- if( ! any(class(data) %in% "data.frame")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE data ARGUMENT MUST BE A DATA FRAME")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end argument checking without fun_check()
- # argument checking with fun_check()
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = quanti.col.name, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = quali.col.name, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # end argument checking with fun_check()
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- tempo.factor <- unlist(lapply(data, class))
- for(i in 1:length(tempo.factor)){ # convert factor columns as character
- if(all(tempo.factor[i] == "factor")){
- data[, i] <- as.character(data[, i])
- }
- }
- tempo.factor <- unlist(lapply(data, mode))
- if(length(data) == 2){
- if( ! ((mode(data[, 1]) == "character" & mode(data[, 2]) == "numeric") | mode(data[, 2]) == "character" & mode(data[, 1]) == "numeric" | mode(data[, 2]) == "numeric" & mode(data[, 1]) == "numeric") ){
- tempo.cat <- paste0("ERROR IN ", function.name, ": IF data ARGUMENT IS A DATA FRAME MADE OF 2 COLUMNS, EITHER A COLUMN MUST BE NUMERIC AND THE OTHER CHARACTER, OR THE TWO COLUMNS MUST BE NUMERIC")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if((mode(data[, 1]) == "character" | mode(data[, 2]) == "character") & (quanti.col.name != "quanti" | quali.col.name != "quali")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": IMPROPER quanti.col.name OR quali.col.name RESETTINGS. THESE ARGUMENTS ARE RESERVED FOR DATA FRAMES MADE OF n NUMERIC COLUMNS ONLY")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }else{
- if( ! all(tempo.factor %in% "numeric")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": IF data ARGUMENT IS A DATA FRAME MADE OF ONE COLUMN, OR MORE THAN 2 COLUMNS, THESE COLUMNS MUST BE NUMERIC")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- if(( ! any(tempo.factor %in% "character")) & is.null(names(data))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": NUMERIC DATA FRAME in the data ARGUMENT MUST HAVE COLUMN NAMES")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(all(tempo.factor %in% "numeric")){ # transfo 1
- quanti <- NULL
- for(i in 1:length(data)){
- quanti <-c(quanti, data[, i])
- }
- quali <- rep(names(data), each = nrow(data))
- output.data <- data.frame(quanti, quali, stringsAsFactors = TRUE)
- names(output.data) <- c(quanti.col.name, quali.col.name)
- # add the ini_rowname column
- ini.rownames <- rownames(data)
- tempo.data <- data
- rownames(tempo.data) <- NULL
- null.rownames <- (tempo.data)
- if( ! identical(ini.rownames, null.rownames)){
- ini_rowname <- rep(ini.rownames, times = ncol(data))
- output.data <- cbind(output.data, ini_rowname, stringsAsFactors = TRUE)
- }
- }else{ # transfo 2
- if(class(data[, 1]) == "character"){
- data <- cbind(data[2], data[1], stringsAsFactors = TRUE)
- }
- nc.max <- max(table(data[, 2])) # effectif maximum des classes
- nb.na <- nc.max - table(data[,2]) # nombre de NA à ajouter pour réaliser la data frame
- tempo<-split(data[, 1], data[, 2])
- for(i in 1:length(tempo)){tempo[[i]] <- append(tempo[[i]], rep(NA, nb.na[i]))} # des NA doivent être ajoutés lorsque les effectifs sont différents entre les classes. C'est uniquement pour que chaque colonne ait le même nombre de lignes
- output.data<-data.frame(tempo, stringsAsFactors = TRUE)
- }
- return(output.data)
+# AIM
+# if the data frame is made of numeric columns, a new data frame is created, with the 1st column gathering all the numeric values, and the 2nd column being the name of the columns of the initial data frame. If row names were present in the initial data frame, then a new ini_rowname column is added with the names of the rows
+
+
+# If the data frame is made of one numeric column and one character or factor column, a new data frame is created, with the new columns corresponding to the split numeric values (according to the character column). NA are added a the end of each column to have the same number of rows. BEWARE: in such data frame, rows are not individuals. This means that in the example below, values 10 and 20 are associated on the same row but that means nothing in term of association
+
+
+
+# ARGUMENTS
+# data: data frame to convert
+# quanti.col.name: optional name for the quanti column of the new data frame
+# quali.col.name: optional name for the quali column of the new data frame
+# RETURN
+# the modified data frame
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# obs <- data.frame(col1 = (1:4)*10, col2 = c("A", "B", "A", "A"), stringsAsFactors = TRUE) ; obs ; fun_df_remod(obs)
+# obs <- data.frame(col1 = (1:4)*10, col2 = 5:8, stringsAsFactors = TRUE) ; obs ; fun_df_remod(obs, quanti.col.name = "quanti", quali.col.name = "quali")
+# obs <- data.frame(col1 = (1:4)*10, col2 = 5:8, stringsAsFactors = TRUE) ; rownames(obs) <- paste0("row", 1:4) ; obs ; fun_df_remod(obs, quanti.col.name = "quanti", quali.col.name = "quali")
+# DEBUGGING
+# data = data.frame(a = 1:3, b = 4:6, stringsAsFactors = TRUE) ; quanti.col.name = "quanti" ; quali.col.name = "quali" # for function debugging
+# data = data.frame(a = 1:3, b = 4:6, c = 11:13, stringsAsFactors = TRUE) ; quanti.col.name = "quanti" ; quali.col.name = "quali" # for function debugging
+# data = data.frame(a = 1:3, b = letters[1:3], stringsAsFactors = TRUE) ; quanti.col.name = "quanti" ; quali.col.name = "quali" # for function debugging
+# data = data.frame(a = 1:3, b = letters[1:3], stringsAsFactors = TRUE) ; quanti.col.name = "TEST" ; quali.col.name = "quali" # for function debugging
+# data = data.frame(b = letters[1:3], a = 1:3, stringsAsFactors = TRUE) ; quanti.col.name = "quanti" ; quali.col.name = "quali" # for function debugging
+# data = data.frame(b = c("e", "e", "h"), a = 1:3, stringsAsFactors = TRUE) ; quanti.col.name = "quanti" ; quali.col.name = "quali" # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+# argument checking without fun_check()
+if( ! any(class(data) %in% "data.frame")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE data ARGUMENT MUST BE A DATA FRAME")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end argument checking without fun_check()
+# argument checking with fun_check()
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = quanti.col.name, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = quali.col.name, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# end argument checking with fun_check()
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+tempo.factor <- unlist(lapply(data, class))
+for(i in 1:length(tempo.factor)){ # convert factor columns as character
+if(all(tempo.factor[i] == "factor")){
+data[, i] <- as.character(data[, i])
+}
+}
+tempo.factor <- unlist(lapply(data, mode))
+if(length(data) == 2){
+if( ! ((mode(data[, 1]) == "character" & mode(data[, 2]) == "numeric") | mode(data[, 2]) == "character" & mode(data[, 1]) == "numeric" | mode(data[, 2]) == "numeric" & mode(data[, 1]) == "numeric") ){
+tempo.cat <- paste0("ERROR IN ", function.name, ": IF data ARGUMENT IS A DATA FRAME MADE OF 2 COLUMNS, EITHER A COLUMN MUST BE NUMERIC AND THE OTHER CHARACTER, OR THE TWO COLUMNS MUST BE NUMERIC")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if((mode(data[, 1]) == "character" | mode(data[, 2]) == "character") & (quanti.col.name != "quanti" | quali.col.name != "quali")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": IMPROPER quanti.col.name OR quali.col.name RESETTINGS. THESE ARGUMENTS ARE RESERVED FOR DATA FRAMES MADE OF n NUMERIC COLUMNS ONLY")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}else{
+if( ! all(tempo.factor %in% "numeric")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": IF data ARGUMENT IS A DATA FRAME MADE OF ONE COLUMN, OR MORE THAN 2 COLUMNS, THESE COLUMNS MUST BE NUMERIC")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+if(( ! any(tempo.factor %in% "character")) & is.null(names(data))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": NUMERIC DATA FRAME in the data ARGUMENT MUST HAVE COLUMN NAMES")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(all(tempo.factor %in% "numeric")){ # transfo 1
+quanti <- NULL
+for(i in 1:length(data)){
+quanti <-c(quanti, data[, i])
+}
+quali <- rep(names(data), each = nrow(data))
+output.data <- data.frame(quanti, quali, stringsAsFactors = TRUE)
+names(output.data) <- c(quanti.col.name, quali.col.name)
+# add the ini_rowname column
+ini.rownames <- rownames(data)
+tempo.data <- data
+rownames(tempo.data) <- NULL
+null.rownames <- (tempo.data)
+if( ! identical(ini.rownames, null.rownames)){
+ini_rowname <- rep(ini.rownames, times = ncol(data))
+output.data <- cbind(output.data, ini_rowname, stringsAsFactors = TRUE)
+}
+}else{ # transfo 2
+if(class(data[, 1]) == "character"){
+data <- cbind(data[2], data[1], stringsAsFactors = TRUE)
+}
+nc.max <- max(table(data[, 2])) # effectif maximum des classes
+nb.na <- nc.max - table(data[,2]) # nombre de NA à ajouter pour réaliser la data frame
+tempo<-split(data[, 1], data[, 2])
+for(i in 1:length(tempo)){tempo[[i]] <- append(tempo[[i]], rep(NA, nb.na[i]))} # des NA doivent être ajoutés lorsque les effectifs sont différents entre les classes. C'est uniquement pour que chaque colonne ait le même nombre de lignes
+output.data<-data.frame(tempo, stringsAsFactors = TRUE)
+}
+return(output.data)
}
@@ -2358,85 +2357,85 @@ fun_df_remod <- function(
# Check OK: clear to go Apollo
fun_round <- function(data, dec.nb = 2, after.lead.zero = TRUE){
- # AIM
- # round a vector of values, if decimal, with the desired number of decimal digits after the decimal leading zeros
- # WARNINGS
- # Work well with numbers as character strings, but not always with numerical numbers because of the floating point
- # Numeric values are really truncated from a part of their decimal digits, whatever options(digits) settings
- # See ?.Machine or https://stackoverflow.com/questions/5173692/how-to-return-number-of-decimal-places-in-r, with the interexting formula: abs(x - round(x)) > .Machine$double.eps^0.5
- # ARGUMENTS
- # data: a vector of numbers (numeric or character mode)
- # dec.nb: number of required decimal digits
- # after.lead.zero: logical. If FALSE, rounding is performed for all the decimal numbers, whatever the leading zeros (e.g., 0.123 -> 0.12 and 0.00128 -> 0.00). If TRUE, dec.nb are taken after the leading zeros (e.g., 0.123 -> 0.12 and 0.00128 -> 0.0013)
- # RETURN
- # the modified vector
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # ini.options <- options()$digits ; options(digits = 8) ; cat(fun_round(data = c(NA, 10, 100.001, 333.0001254, 12312.1235), dec.nb = 2, after.lead.zero = FALSE), "\n\n") ; options(digits = ini.options)
- # ini.options <- options()$digits ; options(digits = 8) ; cat(fun_round(data = c(NA, 10, 100.001, 333.0001254, 12312.1235), dec.nb = 2, after.lead.zero = TRUE), "\n\n") ; options(digits = ini.options)
- # ini.options <- options()$digits ; options(digits = 8) ; cat(fun_round(data = c(NA, "10", "100.001", "333.0001254", "12312.1235"), dec.nb = 2, after.lead.zero = FALSE), "\n\n") ; options(digits = ini.options)
- # ini.options <- options()$digits ; options(digits = 8) ; cat(fun_round(data = c(NA, "10", "100.001", "333.0001254", "12312.1235"), dec.nb = 2, after.lead.zero = TRUE), "\n\n") ; options(digits = ini.options)
- # DEBUGGING
- # data = data = c(10, 100.001, 333.0001254, 12312.1235) ; dec.nb = 2 ; after.lead.zero = FALSE # # for function debugging
- # data = data = c("10", "100.001", "333.0001254", "12312.1235") ; dec.nb = 2 ; after.lead.zero = TRUE # # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- # argument checking without fun_check()
- if( ! (all(typeof(data) == "character") | all(typeof(data) == "double") | all(typeof(data) == "integer"))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": data ARGUMENT MUST BE A VECTOR OF NUMBERS (IN NUMERIC OR CHARACTER MODE)")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end argument checking without fun_check()
- # argument checking with fun_check()
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = data, class = "vector", na.contain = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = dec.nb, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = after.lead.zero, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # end argument checking with fun_check()
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- tempo <- grepl(x = data, pattern = "\\.") # detection of decimal numbers
- ini.mode <- mode(data)
- data <- as.character(data) # to really truncate decimal digits
- for(i in 1:length(data)){ # scan all the numbers of the vector
- if(tempo[i] == TRUE){ # means decimal number
- if(after.lead.zero == TRUE){
- zero.pos <- unlist(gregexpr(text=data[i], pattern = 0)) # recover all the position of the zeros in the number. -1 if no zeros (do not record the leading and trailing zeros)
- }else{
- zero.pos <- -1 # -1 as if no zero
- }
- dot.pos <- unlist(gregexpr(text=data[i], pattern = "\\.")) # recover all the position of the zeros in the number
- digit.pos <- unlist(gregexpr(text=data[i], pattern = "[[:digit:]]")) # recover all the position of the digits in the number
- dec.pos <- digit.pos[digit.pos > dot.pos]
- count <- 0
- while((dot.pos + count + 1) %in% zero.pos & (dot.pos + count + 1) <= max(dec.pos) & (count + dec.nb) < length(dec.pos)){ # count the number of leading zeros in the decimal part
- count <- count + 1
- }
- data[i] <- formatC(as.numeric(data[i]), digits = (count + dec.nb), format = "f")
- }
- }
- if(ini.mode != "character"){
- data <- as.numeric(data)
- }
- return(data)
+# AIM
+# round a vector of values, if decimal, with the desired number of decimal digits after the decimal leading zeros
+# WARNINGS
+# Work well with numbers as character strings, but not always with numerical numbers because of the floating point
+# Numeric values are really truncated from a part of their decimal digits, whatever options(digits) settings
+# See ?.Machine or https://stackoverflow.com/questions/5173692/how-to-return-number-of-decimal-places-in-r, with the interexting formula: abs(x - round(x)) > .Machine$double.eps^0.5
+# ARGUMENTS
+# data: a vector of numbers (numeric or character mode)
+# dec.nb: number of required decimal digits
+# after.lead.zero: logical. If FALSE, rounding is performed for all the decimal numbers, whatever the leading zeros (e.g., 0.123 -> 0.12 and 0.00128 -> 0.00). If TRUE, dec.nb are taken after the leading zeros (e.g., 0.123 -> 0.12 and 0.00128 -> 0.0013)
+# RETURN
+# the modified vector
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# ini.options <- options()$digits ; options(digits = 8) ; cat(fun_round(data = c(NA, 10, 100.001, 333.0001254, 12312.1235), dec.nb = 2, after.lead.zero = FALSE), "\n\n") ; options(digits = ini.options)
+# ini.options <- options()$digits ; options(digits = 8) ; cat(fun_round(data = c(NA, 10, 100.001, 333.0001254, 12312.1235), dec.nb = 2, after.lead.zero = TRUE), "\n\n") ; options(digits = ini.options)
+# ini.options <- options()$digits ; options(digits = 8) ; cat(fun_round(data = c(NA, "10", "100.001", "333.0001254", "12312.1235"), dec.nb = 2, after.lead.zero = FALSE), "\n\n") ; options(digits = ini.options)
+# ini.options <- options()$digits ; options(digits = 8) ; cat(fun_round(data = c(NA, "10", "100.001", "333.0001254", "12312.1235"), dec.nb = 2, after.lead.zero = TRUE), "\n\n") ; options(digits = ini.options)
+# DEBUGGING
+# data = data = c(10, 100.001, 333.0001254, 12312.1235) ; dec.nb = 2 ; after.lead.zero = FALSE # # for function debugging
+# data = data = c("10", "100.001", "333.0001254", "12312.1235") ; dec.nb = 2 ; after.lead.zero = TRUE # # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+# argument checking without fun_check()
+if( ! (all(typeof(data) == "character") | all(typeof(data) == "double") | all(typeof(data) == "integer"))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": data ARGUMENT MUST BE A VECTOR OF NUMBERS (IN NUMERIC OR CHARACTER MODE)")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end argument checking without fun_check()
+# argument checking with fun_check()
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = data, class = "vector", na.contain = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = dec.nb, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = after.lead.zero, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# end argument checking with fun_check()
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+tempo <- grepl(x = data, pattern = "\\.") # detection of decimal numbers
+ini.mode <- mode(data)
+data <- as.character(data) # to really truncate decimal digits
+for(i in 1:length(data)){ # scan all the numbers of the vector
+if(tempo[i] == TRUE){ # means decimal number
+if(after.lead.zero == TRUE){
+zero.pos <- unlist(gregexpr(text=data[i], pattern = 0)) # recover all the position of the zeros in the number. -1 if no zeros (do not record the leading and trailing zeros)
+}else{
+zero.pos <- -1 # -1 as if no zero
+}
+dot.pos <- unlist(gregexpr(text=data[i], pattern = "\\.")) # recover all the position of the zeros in the number
+digit.pos <- unlist(gregexpr(text=data[i], pattern = "[[:digit:]]")) # recover all the position of the digits in the number
+dec.pos <- digit.pos[digit.pos > dot.pos]
+count <- 0
+while((dot.pos + count + 1) %in% zero.pos & (dot.pos + count + 1) <= max(dec.pos) & (count + dec.nb) < length(dec.pos)){ # count the number of leading zeros in the decimal part
+count <- count + 1
+}
+data[i] <- formatC(as.numeric(data[i]), digits = (count + dec.nb), format = "f")
+}
+}
+if(ini.mode != "character"){
+data <- as.numeric(data)
+}
+return(data)
}
@@ -2445,46 +2444,46 @@ fun_round <- function(data, dec.nb = 2, after.lead.zero = TRUE){
# Check OK: clear to go Apollo
fun_mat_rotate <- function(data){
- # AIM
- # 90° clockwise matrix rotation
- # applied twice, the function provide the mirror matrix, according to vertical and horizontal symmetry
- # ARGUMENTS
- # data: matrix (matrix class)
- # RETURN
- # the modified matrix
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # obs <- matrix(1:10, ncol = 1) ; obs ; fun_mat_rotate(obs)
- # obs <- matrix(LETTERS[1:10], ncol = 5) ; obs ; fun_mat_rotate(obs)
- # DEBUGGING
- # data = matrix(1:10, ncol = 1)
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = data, class = "matrix", fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- for (i in 1:ncol(data)){data[,i] <- rev(data[,i])}
- data <- t(data)
- return(data)
+# AIM
+# 90° clockwise matrix rotation
+# applied twice, the function provide the mirror matrix, according to vertical and horizontal symmetry
+# ARGUMENTS
+# data: matrix (matrix class)
+# RETURN
+# the modified matrix
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# obs <- matrix(1:10, ncol = 1) ; obs ; fun_mat_rotate(obs)
+# obs <- matrix(LETTERS[1:10], ncol = 5) ; obs ; fun_mat_rotate(obs)
+# DEBUGGING
+# data = matrix(1:10, ncol = 1)
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = data, class = "matrix", fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+for (i in 1:ncol(data)){data[,i] <- rev(data[,i])}
+data <- t(data)
+return(data)
}
@@ -2493,136 +2492,136 @@ fun_mat_rotate <- function(data){
# Check OK: clear to go Apollo
fun_mat_num2color <- function(
- mat1,
- mat.hsv.h = TRUE,
- notch = 1,
- s = 1,
- v = 1,
- forced.color = NULL
+mat1,
+mat.hsv.h = TRUE,
+notch = 1,
+s = 1,
+v = 1,
+forced.color = NULL
){
- # AIM
- # convert a matrix made of numbers into a hexadecimal matrix for rgb colorization
- # ARGUMENTS:
- # mat1: matrix 1 of non negative numerical values that has to be colored (matrix class). NA allowed
- # mat.hsv.h: logical. Is mat1 the h of hsv colors ? (if TRUE, mat1 must be between zero and 1)
- # notch: single value between 0 and 1 to shift the successive colors on the hsv circle by + notch
- # s: s argument of hsv(). Must be between 0 and 1
- # v: v argument of hsv(). Must be between 0 and 1
- # forced.color: Must be NULL or hexadecimal color code or name given by colors(). The first minimal values of mat1 will be these colors. All the color of mat1 can be forced using this argument
- # RETURN
- # a list containing:
- # $mat1.name: name of mat1
- # $colored.mat: colors of mat1 in hexa
- # $problem: logical. Is any colors of forced.color overlap the colors designed by the function. NULL if forced.color = NULL
- # $text.problem: text when overlapping colors. NULL if forced.color = NULL or problem == FALSE
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # mat1 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2) ; dimnames(mat1) <- list(LETTERS[1:4], letters[1:2]) ; fun_mat_num2color(mat1, mat.hsv.h = FALSE, notch = 1, s = 1, v = 1, forced.color = NULL)
- # DEBUGGING
- # mat1 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2) ; dimnames(mat1) <- list(LETTERS[1:4], letters[1:2]); mat.hsv.h = FALSE ; notch = 1 ; s = 1 ; v = 1 ; forced.color = c(hsv(1,1,1), hsv(0,0,0)) # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- # argument checking with fun_check()
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = mat1, mode = "numeric", class = "matrix", na.contain = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = mat.hsv.h, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = notch, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = s, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = v, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # end argument checking with fun_check()
- # argument checking without fun_check()
- if(mat.hsv.h == TRUE & fun_check(data = mat1, mode = "numeric", prop = TRUE)$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": mat1 ARGUMENT MUST BE A MATRIX OF PROPORTIONS SINCE THE mat.hsv.h ARGUMENT IS SET TO TRUE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! is.null(forced.color)){
- tempo <- fun_check(data = forced.color, class = "character")
- if(any(tempo$problem == TRUE)){
- paste0("\n\n================\n\n", paste(tempo$text[tempo$problem], collapse = "\n"))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! all(forced.color %in% colors() | grepl(pattern = "^#", forced.color))){ # check that all strings of forced.color start by #
- tempo.cat <- paste0("ERROR IN ", function.name, ": forced.color ARGUMENT MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # AND/OR COLOR NAMES GIVEN BY colors()")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # end argument checking without fun_check()
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- problem <- NULL
- text.problem <- NULL
- mat1.name <- deparse(substitute(mat1))
- # change the scale of the plotted matrix
- if(mat.hsv.h == TRUE){
- if(any(min(mat1, na.rm = TRUE) < 0 | max(mat1, na.rm = TRUE) > 1, na.rm = TRUE)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": mat1 MUST BE MADE OF VALUES BETWEEN 0 AND 1 BECAUSE mat.hsv.h ARGUMENT SET TO TRUE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }else{
- if(any(mat1 - floor(mat1) > 0, na.rm = TRUE) | any(mat1 == 0, na.rm = TRUE)){ # no need of isTRUE(all.equal()) because we do not require approx here but strictly 0, thus == is ok
- tempo.cat <- paste0("ERROR IN ", function.name, ": mat1 MUST BE MADE OF INTEGER VALUES WITHOUT 0 BECAUSE mat.hsv.h ARGUMENT SET TO FALSE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- mat1 <- mat1 / max(mat1, na.rm = TRUE)
- }
- }
- if(notch != 1){
- different.color <- unique(as.vector(mat1))
- different.color <- different.color[ ! is.na(different.color)]
- tempo.different.color <- different.color + c(0, cumsum(rep(notch, length(different.color) - 1)))
- tempo.different.color <- tempo.different.color - floor(tempo.different.color)
- if(any(duplicated(tempo.different.color) == TRUE)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": DUPLICATED VALUES AFTER USING notch (", paste(tempo.different.color[duplicated(tempo.different.color)], collapse = " "), "). TRY ANOTHER notch VALUE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else if(length(different.color) != length(tempo.different.color)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": LENGTH OF different.color (", paste(different.color, collapse = " "), ") DIFFERENT FROM LENGTH OF tempo.different.color (", paste(tempo.different.color, collapse = " "), ")")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- for(i in 1:length(different.color)){
- mat1[mat1 == different.color[i]] <- tempo.different.color[i] # no need of isTRUE(all.equal()) because different.color comes from mat1
- }
- }
- }
- if( ! is.null(forced.color)){
- hexa.values.to.change <- hsv(unique(sort(mat1))[1:length(forced.color)], s, v)
- }
- mat1[ ! is.na(mat1)] <- hsv(mat1[ ! is.na(mat1)], s, v)
- if( ! is.null(forced.color)){
- if(any(forced.color %in% mat1, na.rm = TRUE)){
- problem <- TRUE
- text.problem <- paste0("THE FOLLOWING COLORS WHERE INTRODUCED USING forced.color BUT WHERE ALREADY PRESENT IN THE COLORED MATRIX :", paste(forced.color[forced.color %in% mat1], collapse = " "))
- }else{
- problem <- FALSE
- }
- for(i in 1:length(hexa.values.to.change)){
- if( ! any(mat1 == hexa.values.to.change[i], na.rm = TRUE)){# no need of isTRUE(all.equal()) because character
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE ", hexa.values.to.change[i], " VALUE FROM hexa.values.to.change IS NOT REPRESENTED IN mat1 : ", paste(unique(as.vector(mat1)), collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- mat1[which(mat1 == hexa.values.to.change[i])] <- forced.color[i] # no need of isTRUE(all.equal()) because character
- }
- }
- }
- output <- list(mat1.name = mat1.name, colored.mat = mat1, problem = problem, text.problem = text.problem)
- return(output)
+# AIM
+# convert a matrix made of numbers into a hexadecimal matrix for rgb colorization
+# ARGUMENTS:
+# mat1: matrix 1 of non negative numerical values that has to be colored (matrix class). NA allowed
+# mat.hsv.h: logical. Is mat1 the h of hsv colors ? (if TRUE, mat1 must be between zero and 1)
+# notch: single value between 0 and 1 to shift the successive colors on the hsv circle by + notch
+# s: s argument of hsv(). Must be between 0 and 1
+# v: v argument of hsv(). Must be between 0 and 1
+# forced.color: Must be NULL or hexadecimal color code or name given by colors(). The first minimal values of mat1 will be these colors. All the color of mat1 can be forced using this argument
+# RETURN
+# a list containing:
+# $mat1.name: name of mat1
+# $colored.mat: colors of mat1 in hexa
+# $problem: logical. Is any colors of forced.color overlap the colors designed by the function. NULL if forced.color = NULL
+# $text.problem: text when overlapping colors. NULL if forced.color = NULL or problem == FALSE
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# mat1 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2) ; dimnames(mat1) <- list(LETTERS[1:4], letters[1:2]) ; fun_mat_num2color(mat1, mat.hsv.h = FALSE, notch = 1, s = 1, v = 1, forced.color = NULL)
+# DEBUGGING
+# mat1 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2) ; dimnames(mat1) <- list(LETTERS[1:4], letters[1:2]); mat.hsv.h = FALSE ; notch = 1 ; s = 1 ; v = 1 ; forced.color = c(hsv(1,1,1), hsv(0,0,0)) # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+# argument checking with fun_check()
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = mat1, mode = "numeric", class = "matrix", na.contain = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = mat.hsv.h, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = notch, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = s, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = v, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# end argument checking with fun_check()
+# argument checking without fun_check()
+if(mat.hsv.h == TRUE & fun_check(data = mat1, mode = "numeric", prop = TRUE)$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": mat1 ARGUMENT MUST BE A MATRIX OF PROPORTIONS SINCE THE mat.hsv.h ARGUMENT IS SET TO TRUE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! is.null(forced.color)){
+tempo <- fun_check(data = forced.color, class = "character")
+if(any(tempo$problem == TRUE)){
+paste0("\n\n================\n\n", paste(tempo$text[tempo$problem], collapse = "\n"))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! all(forced.color %in% colors() | grepl(pattern = "^#", forced.color))){ # check that all strings of forced.color start by #
+tempo.cat <- paste0("ERROR IN ", function.name, ": forced.color ARGUMENT MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # AND/OR COLOR NAMES GIVEN BY colors()")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# end argument checking without fun_check()
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+problem <- NULL
+text.problem <- NULL
+mat1.name <- deparse(substitute(mat1))
+# change the scale of the plotted matrix
+if(mat.hsv.h == TRUE){
+if(any(min(mat1, na.rm = TRUE) < 0 | max(mat1, na.rm = TRUE) > 1, na.rm = TRUE)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": mat1 MUST BE MADE OF VALUES BETWEEN 0 AND 1 BECAUSE mat.hsv.h ARGUMENT SET TO TRUE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}else{
+if(any(mat1 - floor(mat1) > 0, na.rm = TRUE) | any(mat1 == 0, na.rm = TRUE)){ # no need of isTRUE(all.equal()) because we do not require approx here but strictly 0, thus == is ok
+tempo.cat <- paste0("ERROR IN ", function.name, ": mat1 MUST BE MADE OF INTEGER VALUES WITHOUT 0 BECAUSE mat.hsv.h ARGUMENT SET TO FALSE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+mat1 <- mat1 / max(mat1, na.rm = TRUE)
+}
+}
+if(notch != 1){
+different.color <- unique(as.vector(mat1))
+different.color <- different.color[ ! is.na(different.color)]
+tempo.different.color <- different.color + c(0, cumsum(rep(notch, length(different.color) - 1)))
+tempo.different.color <- tempo.different.color - floor(tempo.different.color)
+if(any(duplicated(tempo.different.color) == TRUE)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": DUPLICATED VALUES AFTER USING notch (", paste(tempo.different.color[duplicated(tempo.different.color)], collapse = " "), "). TRY ANOTHER notch VALUE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else if(length(different.color) != length(tempo.different.color)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": LENGTH OF different.color (", paste(different.color, collapse = " "), ") DIFFERENT FROM LENGTH OF tempo.different.color (", paste(tempo.different.color, collapse = " "), ")")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+for(i in 1:length(different.color)){
+mat1[mat1 == different.color[i]] <- tempo.different.color[i] # no need of isTRUE(all.equal()) because different.color comes from mat1
+}
+}
+}
+if( ! is.null(forced.color)){
+hexa.values.to.change <- hsv(unique(sort(mat1))[1:length(forced.color)], s, v)
+}
+mat1[ ! is.na(mat1)] <- hsv(mat1[ ! is.na(mat1)], s, v)
+if( ! is.null(forced.color)){
+if(any(forced.color %in% mat1, na.rm = TRUE)){
+problem <- TRUE
+text.problem <- paste0("THE FOLLOWING COLORS WHERE INTRODUCED USING forced.color BUT WHERE ALREADY PRESENT IN THE COLORED MATRIX :", paste(forced.color[forced.color %in% mat1], collapse = " "))
+}else{
+problem <- FALSE
+}
+for(i in 1:length(hexa.values.to.change)){
+if( ! any(mat1 == hexa.values.to.change[i], na.rm = TRUE)){# no need of isTRUE(all.equal()) because character
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE ", hexa.values.to.change[i], " VALUE FROM hexa.values.to.change IS NOT REPRESENTED IN mat1 : ", paste(unique(as.vector(mat1)), collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+mat1[which(mat1 == hexa.values.to.change[i])] <- forced.color[i] # no need of isTRUE(all.equal()) because character
+}
+}
+}
+output <- list(mat1.name = mat1.name, colored.mat = mat1, problem = problem, text.problem = text.problem)
+return(output)
}
@@ -2631,104 +2630,104 @@ fun_mat_num2color <- function(
# Check OK: clear to go Apollo
fun_mat_op <- function(mat.list, kind.of.operation = "+"){
- # AIM
- # assemble several matrices of same dimensions by performing by case operation. For instance add the value of all the case 1 (row1 & column1) of the matrices and put it in the case 1 of a new matrix M, add the value of all the case 2 (row2 & column1) of the matrices and put it in the case 2 of a new matrix M, etc.
-
- # c: case
- # i: row number
- # j: column number
- # k: matrix number
- # z: number of matrices
- # ARGUMENTS:
- # mat.list: list of matrices
- # kind.of.operation: either "+" (by case addition), "-" (by case subtraction) or "*" (by case multiplication)
- # RETURN
- # the assembled matrix, with row and/or column names only if all the matrices have identical row/column names
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # fun_comp_2d()
- # EXAMPLES
- # mat1 = matrix(c(1,1,1,2,1,5,9,8), ncol = 2) ; mat2 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2) ; fun_mat_op(mat.list = list(mat1, mat2), kind.of.operation = "+")
- # mat1 = matrix(c(1,1,1,2,1,5,9,8), ncol = 2, dimnames = list(LETTERS[1:4], letters[1:2])) ; mat2 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2, dimnames = list(LETTERS[1:4], letters[1:2])) ; fun_mat_op(mat.list = list(mat1, mat2), kind.of.operation = "*")
- # mat1 = matrix(c(1,1,1,2,1,5,9,8), ncol = 2, dimnames = list(LETTERS[1:4], c(NA, NA))) ; mat2 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2, dimnames = list(LETTERS[1:4], letters[1:2])) ; fun_mat_op(mat.list = list(mat1, mat2), kind.of.operation = "-")
- # mat1 = matrix(c(1,1,1,2,1,5,9,8), ncol = 2, dimnames = list(c("A1", "A2", "A3", "A4"), letters[1:2])) ; mat2 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2, dimnames = list(LETTERS[1:4], letters[1:2])) ; mat3 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2, dimnames = list(LETTERS[1:4], letters[1:2])) ; fun_mat_op(mat.list = list(mat1, mat2, mat3), kind.of.operation = "+")
- # DEBUGGING
- # mat1 = matrix(c(1,1,1,2,1,5,9,8), ncol = 2) ; mat2 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2) ; mat.list = list(mat1, mat2) ; kind.of.operation = "+" # for function debugging
- # mat1 = matrix(c(1,1,1,2,1,5,9,8), ncol = 2, dimnames = list(LETTERS[1:4], c(NA, NA))) ; mat2 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2, dimnames = list(LETTERS[1:4], letters[1:2])) ; mat.list = list(mat1, mat2) ; kind.of.operation = "*" # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_comp_2d() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- # argument checking with fun_check()
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = mat.list, class = "list", fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = kind.of.operation, options = c("+", "-", "*"), length = 1, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # end argument checking with fun_check()
- # argument checking without fun_check()
- if(length(mat.list) < 2){
- tempo.cat <- paste0("ERROR IN ", function.name, ": mat.list ARGUMENT MUST BE A LIST CONTAINING AT LEAST 2 MATRICES")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- for(i1 in 1:length(mat.list)){
- tempo <- fun_check(data = mat.list[[i1]], class = "matrix", mode = "numeric", na.contain = TRUE)
- if(tempo$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ELEMENT ", i1, " OF mat.list ARGUMENT MUST BE A NUMERIC MATRIX")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- ident.row.names <- TRUE
- ident.col.names <- TRUE
- for(i1 in 2:length(mat.list)){
- tempo <- fun_comp_2d(data1 = mat.list[[1]], data2 = mat.list[[i1]])
- if(tempo$same.dim == FALSE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": MATRIX ", i1, " OF mat.list ARGUMENT MUST HAVE THE SAME DIMENSION (", paste(dim(mat.list[[i1]]), collapse = " "), ") THAN THE MATRIX 1 IN mat.list (", paste(dim(mat.list[[1]]), collapse = " "), ")")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! is.null(tempo$same.row.name)){
- if(tempo$same.row.name != TRUE){ # != TRUE to deal with NA
- ident.row.names <- FALSE
- }
- }
- if( ! is.null(tempo$same.col.name)){
- if(tempo$same.col.name != TRUE){ # != TRUE to deal with NA
- ident.col.names <- FALSE
- }
- }
- }
- # end argument checking without fun_check()
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- output <- mat.list[[1]]
- for(i1 in 2:length(mat.list)){
- output <- get(kind.of.operation)(output, mat.list[[i1]]) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
- }
- dimnames(output) <- NULL
- if(ident.row.names == TRUE){
- rownames(output) <- rownames(mat.list[[1]])
- }
- if(ident.col.names == TRUE){
- colnames(output) <- colnames(mat.list[[1]])
- }
- return(output)
+# AIM
+# assemble several matrices of same dimensions by performing by case operation. For instance add the value of all the case 1 (row1 & column1) of the matrices and put it in the case 1 of a new matrix M, add the value of all the case 2 (row2 & column1) of the matrices and put it in the case 2 of a new matrix M, etc.
+
+# c: case
+# i: row number
+# j: column number
+# k: matrix number
+# z: number of matrices
+# ARGUMENTS:
+# mat.list: list of matrices
+# kind.of.operation: either "+" (by case addition), "-" (by case subtraction) or "*" (by case multiplication)
+# RETURN
+# the assembled matrix, with row and/or column names only if all the matrices have identical row/column names
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# fun_comp_2d()
+# EXAMPLES
+# mat1 = matrix(c(1,1,1,2,1,5,9,8), ncol = 2) ; mat2 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2) ; fun_mat_op(mat.list = list(mat1, mat2), kind.of.operation = "+")
+# mat1 = matrix(c(1,1,1,2,1,5,9,8), ncol = 2, dimnames = list(LETTERS[1:4], letters[1:2])) ; mat2 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2, dimnames = list(LETTERS[1:4], letters[1:2])) ; fun_mat_op(mat.list = list(mat1, mat2), kind.of.operation = "*")
+# mat1 = matrix(c(1,1,1,2,1,5,9,8), ncol = 2, dimnames = list(LETTERS[1:4], c(NA, NA))) ; mat2 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2, dimnames = list(LETTERS[1:4], letters[1:2])) ; fun_mat_op(mat.list = list(mat1, mat2), kind.of.operation = "-")
+# mat1 = matrix(c(1,1,1,2,1,5,9,8), ncol = 2, dimnames = list(c("A1", "A2", "A3", "A4"), letters[1:2])) ; mat2 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2, dimnames = list(LETTERS[1:4], letters[1:2])) ; mat3 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2, dimnames = list(LETTERS[1:4], letters[1:2])) ; fun_mat_op(mat.list = list(mat1, mat2, mat3), kind.of.operation = "+")
+# DEBUGGING
+# mat1 = matrix(c(1,1,1,2,1,5,9,8), ncol = 2) ; mat2 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2) ; mat.list = list(mat1, mat2) ; kind.of.operation = "+" # for function debugging
+# mat1 = matrix(c(1,1,1,2,1,5,9,8), ncol = 2, dimnames = list(LETTERS[1:4], c(NA, NA))) ; mat2 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2, dimnames = list(LETTERS[1:4], letters[1:2])) ; mat.list = list(mat1, mat2) ; kind.of.operation = "*" # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_comp_2d() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+# argument checking with fun_check()
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = mat.list, class = "list", fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = kind.of.operation, options = c("+", "-", "*"), length = 1, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# end argument checking with fun_check()
+# argument checking without fun_check()
+if(length(mat.list) < 2){
+tempo.cat <- paste0("ERROR IN ", function.name, ": mat.list ARGUMENT MUST BE A LIST CONTAINING AT LEAST 2 MATRICES")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+for(i1 in 1:length(mat.list)){
+tempo <- fun_check(data = mat.list[[i1]], class = "matrix", mode = "numeric", na.contain = TRUE)
+if(tempo$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ELEMENT ", i1, " OF mat.list ARGUMENT MUST BE A NUMERIC MATRIX")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+ident.row.names <- TRUE
+ident.col.names <- TRUE
+for(i1 in 2:length(mat.list)){
+tempo <- fun_comp_2d(data1 = mat.list[[1]], data2 = mat.list[[i1]])
+if(tempo$same.dim == FALSE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": MATRIX ", i1, " OF mat.list ARGUMENT MUST HAVE THE SAME DIMENSION (", paste(dim(mat.list[[i1]]), collapse = " "), ") THAN THE MATRIX 1 IN mat.list (", paste(dim(mat.list[[1]]), collapse = " "), ")")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! is.null(tempo$same.row.name)){
+if(tempo$same.row.name != TRUE){ # != TRUE to deal with NA
+ident.row.names <- FALSE
+}
+}
+if( ! is.null(tempo$same.col.name)){
+if(tempo$same.col.name != TRUE){ # != TRUE to deal with NA
+ident.col.names <- FALSE
+}
+}
+}
+# end argument checking without fun_check()
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+output <- mat.list[[1]]
+for(i1 in 2:length(mat.list)){
+output <- get(kind.of.operation)(output, mat.list[[i1]]) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
+}
+dimnames(output) <- NULL
+if(ident.row.names == TRUE){
+rownames(output) <- rownames(mat.list[[1]])
+}
+if(ident.col.names == TRUE){
+colnames(output) <- colnames(mat.list[[1]])
+}
+return(output)
}
@@ -2737,74 +2736,74 @@ fun_mat_op <- function(mat.list, kind.of.operation = "+"){
# Check OK: clear to go Apollo
fun_mat_inv <- function(mat){
- # AIM
- # return the inverse of a square matrix when solve() cannot
- # ARGUMENTS:
- # mat: a square numeric matrix without NULL, NA, Inf or single case (dimension 1, 1) of 0
- # RETURN
- # the inversed matrix
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # mat1 = matrix(c(1,1,1,2,1,5,9,8,9), ncol = 3) ; fun_mat_inv(mat = mat1) # use solve()
- # mat1 = matrix(c(0,0,0,0,0,0,0,0,0), ncol = 3) ; fun_mat_inv(mat = mat1) # use the trick
- # mat1 = matrix(c(1,1,1,2,Inf,5,9,8,9), ncol = 3) ; fun_mat_inv(mat = mat1)
- # mat1 = matrix(c(1,1,1,2,NA,5,9,8,9), ncol = 3) ; fun_mat_inv(mat = mat1)
- # mat1 = matrix(c(1,2), ncol = 1) ; fun_mat_inv(mat = mat1)
- # mat1 = matrix(0, ncol = 1) ; fun_mat_inv(mat = mat1)
- # mat1 = matrix(2, ncol = 1) ; fun_mat_inv(mat = mat1)
- # DEBUGGING
- # mat = matrix(c(1,1,1,2,1,5,9,8,9), ncol = 3) # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- # argument checking with fun_check()
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = mat, class = "matrix", mode = "numeric", fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # end argument checking with fun_check()
- # argument checking without fun_check()
- if(ncol(mat) != nrow(mat)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": mat ARGUMENT MUST BE A SQUARE MATRIX")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(mat %in% c(Inf, -Inf, NA))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": mat ARGUMENT MUST BE A MATRIX WITHOUT Inf, -Inf OR NA")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(all(mat == 0) & ncol(mat) == 1){
- tempo.cat <- paste0("ERROR IN ", function.name, ": mat ARGUMENT CANNOT BE A SQUARE MATRIX MADE OF A SINGLE CASE OF 0")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end argument checking without fun_check()
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- if(any(grepl(x = try(solve(mat), silent = TRUE)[], pattern = "[Ee]rror"))){
- tempo <- svd(mat)
- val.critique <- which(tempo$d < 10^-8)
- Diag.mod <- diag(1 / tempo$d)
- for(i in val.critique){
- Diag.mod[i, i] <- 0
- }
- return(tempo$v %*% Diag.mod %*% t(tempo$u))
- }else{
- return(solve(mat))
- }
+# AIM
+# return the inverse of a square matrix when solve() cannot
+# ARGUMENTS:
+# mat: a square numeric matrix without NULL, NA, Inf or single case (dimension 1, 1) of 0
+# RETURN
+# the inversed matrix
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# mat1 = matrix(c(1,1,1,2,1,5,9,8,9), ncol = 3) ; fun_mat_inv(mat = mat1) # use solve()
+# mat1 = matrix(c(0,0,0,0,0,0,0,0,0), ncol = 3) ; fun_mat_inv(mat = mat1) # use the trick
+# mat1 = matrix(c(1,1,1,2,Inf,5,9,8,9), ncol = 3) ; fun_mat_inv(mat = mat1)
+# mat1 = matrix(c(1,1,1,2,NA,5,9,8,9), ncol = 3) ; fun_mat_inv(mat = mat1)
+# mat1 = matrix(c(1,2), ncol = 1) ; fun_mat_inv(mat = mat1)
+# mat1 = matrix(0, ncol = 1) ; fun_mat_inv(mat = mat1)
+# mat1 = matrix(2, ncol = 1) ; fun_mat_inv(mat = mat1)
+# DEBUGGING
+# mat = matrix(c(1,1,1,2,1,5,9,8,9), ncol = 3) # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+# argument checking with fun_check()
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = mat, class = "matrix", mode = "numeric", fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# end argument checking with fun_check()
+# argument checking without fun_check()
+if(ncol(mat) != nrow(mat)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": mat ARGUMENT MUST BE A SQUARE MATRIX")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(mat %in% c(Inf, -Inf, NA))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": mat ARGUMENT MUST BE A MATRIX WITHOUT Inf, -Inf OR NA")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(all(mat == 0) & ncol(mat) == 1){
+tempo.cat <- paste0("ERROR IN ", function.name, ": mat ARGUMENT CANNOT BE A SQUARE MATRIX MADE OF A SINGLE CASE OF 0")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end argument checking without fun_check()
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+if(any(grepl(x = try(solve(mat), silent = TRUE)[], pattern = "[Ee]rror"))){
+tempo <- svd(mat)
+val.critique <- which(tempo$d < 10^-8)
+Diag.mod <- diag(1 / tempo$d)
+for(i in val.critique){
+Diag.mod[i, i] <- 0
+}
+return(tempo$v %*% Diag.mod %*% t(tempo$u))
+}else{
+return(solve(mat))
+}
}
@@ -2813,931 +2812,931 @@ fun_mat_inv <- function(mat){
# Check OK: clear to go Apollo
fun_mat_fill <- function(mat, empty.cell.string = 0, warn.print = FALSE){
- # AIM
- # detect the empty half part of a symmetric square matrix (either topleft, topright, bottomleft or bottomright)
- # fill this empty half part using the other symmetric half part of the matrix
- # WARNINGS
- # a plot verification using fun_gg_heatmap() is recommanded
- # ARGUMENTS:
- # mat: a numeric or character square matrix with the half part (according to the grand diagonal) filled with NA (any kind of matrix), "0" (character matrix) or 0 (numeric matrix) exclusively (not a mix of 0 and NA in the empty part)
- # empty.cell.string: a numeric, character or NA (no quotes) indicating what empty cells are filled with
- # warn.print: logical. Print warnings at the end of the execution? No print if no warning messages
- # RETURN
- # a list containing:
- # $mat: the filled matrix
- # $warn: the warning messages. Use cat() for proper display. NULL if no warning
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # mat1 = matrix(c(1,NA,NA,NA, 0,2,NA,NA, NA,3,4,NA, 5,6,7,8), ncol = 4) ; mat1 ; fun_mat_fill(mat = mat1, empty.cell.string = NA, warn.print = TRUE) # bottomleft example
- # mat1 = matrix(c(1,1,1,2, 0,2,3,0, NA,3,0,0, 5,0,0,0), ncol = 4) ; mat1 ; fun_mat_fill(mat = mat1, empty.cell.string = NA, warn.print = TRUE) # error example
- # mat1 = matrix(c(1,1,1,2, 0,2,3,0, NA,3,0,0, 5,0,0,0), ncol = 4) ; mat1 ; fun_mat_fill(mat = mat1, empty.cell.string = 0, warn.print = TRUE) # bottomright example
- # mat1 = matrix(c(1,1,1,2, "a",2,3,NA, "a","a",0,0, "a","a","a",0), ncol = 4) ; mat1 ; fun_mat_fill(mat = mat1, empty.cell.string = "a", warn.print = TRUE) # topright example
- # mat1 = matrix(c(0,0,0,2, 0,0,3,0, 0,3,0,NA, 5,0,0,0), ncol = 4) ; mat1 ; fun_mat_fill(mat = mat1, empty.cell.string = 0, warn.print = TRUE) # topleft example
- # mat1 = matrix(c(0,0,0,2, 0,0,3,0, 0,3,0,0, 5,0,0,0), ncol = 4) ; mat1 ; fun_mat_fill(mat = mat1, empty.cell.string = 0, warn.print = TRUE) # error example
- # DEBUGGING
- # mat = matrix(c(1,NA,NA,NA, 0,2,NA,NA, NA,3,4,NA, 5,6,7,8), ncol = 4) ; empty.cell.string = NA ; warn.print = TRUE # for function debugging
- # mat = matrix(c(0,0,0,2, 0,0,3,0, 0,3,0,NA, 5,0,0,0), ncol = 4) ; empty.cell.string = 0 ; warn.print = TRUE # for function debugging # topleft example
- # mat = matrix(c(0,0,0,2, 0,0,3,0, 0,3,0,NA, 5,0,0,0), ncol = 4) ; empty.cell.string = NA ; warn.print = TRUE # for function debugging # topleft example
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- # argument checking with fun_check()
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = mat, class = "matrix", na.contain = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = empty.cell.string, class = "vector", na.contain = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = warn.print, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # end argument checking with fun_check()
- # argument checking without fun_check()
- if(ncol(mat) != nrow(mat)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": mat ARGUMENT MUST BE A SQUARE MATRIX")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! (mode(mat) %in% c("numeric", "character"))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": mat ARGUMENT MUST BE A NUMERIC OR CHARACTER MATRIX")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(nrow(mat) == 1 & ncol(mat) == 1){
- tempo.cat <- paste0("ERROR IN ", function.name, ": mat ARGUMENT CANNOT BE A SQUARE MATRIX MADE OF A SINGLE CASE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(ifelse(is.na(empty.cell.string), ! any(is.na(mat)), ! any(mat == empty.cell.string, na.rm = TRUE))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": mat ARGUMENT MATRIX MUST HAVE CELLS WITH THE EMPTY STRING SPECIFIED IN empty.cell.string ARGUMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end argument checking without fun_check()
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- list.diag <- vector("list", length = nrow(mat) - 1)
- for(i1 in 1:(nrow(mat) - 1)){
- list.diag[[i1]] <- numeric(length = nrow(mat) - i1)
- }
- sector <- c("topleft", "topright", "bottomright", "bottomleft")
- diag.scan <-c( # same order as sector. Recover each diag from center to corner
- "mat[as.matrix(as.data.frame(list(1:(nrow(mat) - i2), (ncol(mat) -i2):1), stringsAsFactors = TRUE))]", # topleft part
- "mat[as.matrix(as.data.frame(list(1:(nrow(mat) - i2), (1:ncol(mat))[-(1:i2)]), stringsAsFactors = TRUE))]", # topright part
- "mat[as.matrix(as.data.frame(list((1 + i2):nrow(mat), ncol(mat):(1 + i2)), stringsAsFactors = TRUE))]", # bottomright part
- "mat[as.matrix(as.data.frame(list((1 + i2):nrow(mat), 1:(ncol(mat) -i2)), stringsAsFactors = TRUE))]" # bottomleft part
- )
- # empty part detection
- tempo.list.diag <- list.diag
- empty.sector <- NULL
- full.sector <- NULL
- ini.warning.length <- options()$warning.length
- warn <- NULL
- warn.count <- 0
- for(i1 in 1:length(sector)){
- tempo.list.diag <- list.diag
- for(i2 in 1:(nrow(mat) - 1)){
- tempo.list.diag[[i2]] <- eval(parse(text = diag.scan[i1]))
- if(ifelse(is.na(empty.cell.string), ! all(is.na(tempo.list.diag[[i2]])), ! (all(tempo.list.diag[[i2]] == empty.cell.string, na.rm = TRUE) & ! (is.na(all(tempo.list.diag[[i2]] == empty.cell.string, na.rm = FALSE)))))){ # I had to add this ! (is.na(all(tempo.list.diag[[i2]] == empty.cell.string, na.rm = FALSE))) because all(tempo.list.diag[[i2]] == empty.cell.string, na.rm = FALSE) gives NA and not FALSE if one NA in tempo.list.diag[[i2]] -> not good for if()
- full.sector <- c(full.sector, sector[i1])
- break
- }
- }
- if(i1 == nrow(mat) - 1){
- if(all(unlist(lapply(tempo.list.diag, FUN = function(x){if(is.na(empty.cell.string)){is.na(x)}else{x == empty.cell.string}})), na.rm = TRUE)){
- empty.sector <- c(empty.sector, sector[i1])
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") EMPTY SECTOR DETECTED ON THE ", toupper(sector[i1]), " CORNER, FULL OF ", empty.cell.string)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE ", toupper(sector[i1]), " SECTOR, DETECTED AS EMPTY, IS NOT? DIFFERENT VALUES IN THIS SECTOR:\n", paste(names(table(unlist(tempo.list.diag), useNA = "ifany")), collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- }
- # end empty part detection
- if(length(empty.sector) == 0){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") ACCORDING TO empty.cell.string ARGUMENT (", empty.cell.string, "), mat ARGUMENT MATRIX HAS ZERO EMPTY HALF PART")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }else{
- if(length(empty.sector) > 1){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ACCORDING TO empty.cell.string ARGUMENT (", empty.cell.string, "), mat ARGUMENT MATRIX HAS MORE THAN ONE EMPTY HALF PART (ACCORDING TO THE GRAND DIAGONAL): ", paste(empty.sector, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else if(any(full.sector %in% empty.sector, na.rm = TRUE)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE FUNCTION HAS DETECTED EMPTY AND NON EMPTY HALF PART IN THE SAME SECTOR: ", paste(full.sector[full.sector %in% empty.sector], collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else if(length(empty.sector) + length(full.sector)!= 4){
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE FUNCTION HAS DETECTED MORE OR LESS SECTORS THAN 4:\nHALF SECTORS:", paste(empty.sector, collapse = " "), "\nFULL SECTORS:", paste(full.sector, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") ", toupper(empty.sector), " SECTOR HAS BEEN COMPLETED TO BECOME SYMMETRICAL")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- # matrix filling
- for(i1 in 1:(nrow(mat) - 1)){
- if(empty.sector == "topleft"){
- eval(parse(text = paste0(diag.scan[1], " <- ", diag.scan[3])))
- }else if(empty.sector == "topright"){
- eval(parse(text = paste0(diag.scan[2], " <- ", diag.scan[4])))
- }else if(empty.sector == "bottomright"){
- eval(parse(text = paste0(diag.scan[3], " <- ", diag.scan[1])))
- }else if(empty.sector == "bottomleft"){
- eval(parse(text = paste0(diag.scan[4], " <- ", diag.scan[2])))
- }
- }
- # end matrix filling
- }
- if(warn.print == TRUE & ! is.null(warn)){
- options(warning.length = 8170)
- on.exit(warning(paste0("FROM ", function.name, ":\n\n", warn), call. = FALSE))
- on.exit(exp = options(warning.length = ini.warning.length), add = TRUE)
- }
- return(list(mat = mat, warn = warn))
+# AIM
+# detect the empty half part of a symmetric square matrix (either topleft, topright, bottomleft or bottomright)
+# fill this empty half part using the other symmetric half part of the matrix
+# WARNINGS
+# a plot verification using fun_gg_heatmap() is recommanded
+# ARGUMENTS:
+# mat: a numeric or character square matrix with the half part (according to the grand diagonal) filled with NA (any kind of matrix), "0" (character matrix) or 0 (numeric matrix) exclusively (not a mix of 0 and NA in the empty part)
+# empty.cell.string: a numeric, character or NA (no quotes) indicating what empty cells are filled with
+# warn.print: logical. Print warnings at the end of the execution? No print if no warning messages
+# RETURN
+# a list containing:
+# $mat: the filled matrix
+# $warn: the warning messages. Use cat() for proper display. NULL if no warning
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# mat1 = matrix(c(1,NA,NA,NA, 0,2,NA,NA, NA,3,4,NA, 5,6,7,8), ncol = 4) ; mat1 ; fun_mat_fill(mat = mat1, empty.cell.string = NA, warn.print = TRUE) # bottomleft example
+# mat1 = matrix(c(1,1,1,2, 0,2,3,0, NA,3,0,0, 5,0,0,0), ncol = 4) ; mat1 ; fun_mat_fill(mat = mat1, empty.cell.string = NA, warn.print = TRUE) # error example
+# mat1 = matrix(c(1,1,1,2, 0,2,3,0, NA,3,0,0, 5,0,0,0), ncol = 4) ; mat1 ; fun_mat_fill(mat = mat1, empty.cell.string = 0, warn.print = TRUE) # bottomright example
+# mat1 = matrix(c(1,1,1,2, "a",2,3,NA, "a","a",0,0, "a","a","a",0), ncol = 4) ; mat1 ; fun_mat_fill(mat = mat1, empty.cell.string = "a", warn.print = TRUE) # topright example
+# mat1 = matrix(c(0,0,0,2, 0,0,3,0, 0,3,0,NA, 5,0,0,0), ncol = 4) ; mat1 ; fun_mat_fill(mat = mat1, empty.cell.string = 0, warn.print = TRUE) # topleft example
+# mat1 = matrix(c(0,0,0,2, 0,0,3,0, 0,3,0,0, 5,0,0,0), ncol = 4) ; mat1 ; fun_mat_fill(mat = mat1, empty.cell.string = 0, warn.print = TRUE) # error example
+# DEBUGGING
+# mat = matrix(c(1,NA,NA,NA, 0,2,NA,NA, NA,3,4,NA, 5,6,7,8), ncol = 4) ; empty.cell.string = NA ; warn.print = TRUE # for function debugging
+# mat = matrix(c(0,0,0,2, 0,0,3,0, 0,3,0,NA, 5,0,0,0), ncol = 4) ; empty.cell.string = 0 ; warn.print = TRUE # for function debugging # topleft example
+# mat = matrix(c(0,0,0,2, 0,0,3,0, 0,3,0,NA, 5,0,0,0), ncol = 4) ; empty.cell.string = NA ; warn.print = TRUE # for function debugging # topleft example
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
-
-
-######## fun_permut() #### progressively breaks a vector order
-
-
-fun_permut <- function(
- data1,
- data2 = NULL,
- n = NULL,
- seed = NULL,
- print.count = 10,
- text.print = "",
- cor.method = "spearman",
- cor.limit = 0.2,
- warn.print = FALSE,
- lib.path = NULL
-){
- # AIM
- # reorder the elements of the data1 vector by flipping 2 randomly selected consecutive positions either:
- # 1) n times (when n is precised) or
- # 2) until the correlation between data1 and data2 decreases down to the cor.limit (0.2 by default). See cor.limit below to deal with negative correlations
- # Example of consecutive position flipping: ABCD -> BACD -> BADC, etc.
- # designed for discrete values, but worls also for continuous values
- # WARNINGS
- # see # https://www.r-bloggers.com/strategies-to-speedup-r-code/ for code speedup
- # the random switch of non consecutive positions (ABCD -> DBCA for instance) does not work very well as the correlation is quickly obtained but the initial vector structure is mainly kept (no much order). Ths code would be: pos <- ini.pos[1:2] ; pos <- sample.int(n = n , size = 2, replace = FALSE) ; tempo.pos[pos] <- tempo.pos[rev(pos)]
- # ARGUMENTS
- # data1: a vector of at least 2 elements. Must be numeric if data2 is specified
- # data2: a numeric vector of same length as data1
- # n: number of times "flipping 2 randomly selected consecutive positions". Ignored if data2 is specified
- # seed: integer number used by set.seed(). Write NULL if random result is required, an integer otherwise. BEWARE: if not NULL, fun_permut() will systematically return the same result when the other parameters keep the same settings
- # print.count: interger value. Print a working progress message every print.count during loops. BEWARE: can increase substentially the time to complete the process using a small value, like 10 for instance. Use Inf is no loop message desired
- # text.print: optional message to add to the working progress message every print.count loop
- # cor.method: correlation method. Either "pearson", "kendall" or "spearman". Ignored if data2 is not specified
- # cor.limit: a correlation limit (between 0 and 1). Ignored if data2 is not specified. Compute the correlation between data1 and data2, permute the data1 values, and stop the permutation process when the correlation between data1 and data2 decreases down below the cor limit value (0.2 by default). If cor(data1, data2) is negative, then -cor.limit is used and the process stops until the correlation between data1 and data2 increases up over cor.limit (-0.2 by default). BEWARE: write a positive cor.limit even if cor(data1, data2) is known to be negative. The function will automatically uses -cor.limit. If the initial correlation is already below cor.limit (positive correlation) or over -cor.limit (negative correlation), then the data1 value positions are completely randomized (correlation between data1 and data2 is expected to be 0)
- # warn.print: logical. Print warnings at the end of the execution? No print if no warning messages
- # lib.path: character vector specifying the absolute pathways of the directories containing the required packages if not in the default directories. Ignored if NULL
- # RETURN
- # a list containing:
- # $data: the modified vector
- # $warn: potential warning messages (in case of negative correlation when data2 is specified). NULL if non warning message
- # $cor: a spearman correlation between the initial positions (1:length(data1) and the final positions if data2 is not specified and the final correlation between data1 and data2 otherwise, according to cor.method
- # $count: the number of loops used
- # REQUIRED PACKAGES
- # lubridate
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # fun_pack()
- # fun_round()
- # EXAMPLES
- # example (1) showing that for loop, used in fun_permut(), is faster than while loop
- # ini.time <- as.numeric(Sys.time()) ; count <- 0 ; for(i0 in 1:1e9){count <- count + 1} ; tempo.time <- as.numeric(Sys.time()) ; tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - ini.time)) ; tempo.lapse
- # example (2) showing that for loop, used in fun_permut(), is faster than while loop
- # ini.time <- as.numeric(Sys.time()) ; count <- 0 ; while(count < 1e9){count <- count + 1} ; tempo.time <- as.numeric(Sys.time()) ; tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - ini.time)) ; tempo.lapse
- # fun_permut(data1 = LETTERS[1:5], data2 = NULL, n = 100, seed = 1, print.count = 10, text.print = "CPU NB 4")
- # fun_permut(data1 = 101:110, data2 = 21:30, seed = 1, print.count = 1e4, text.print = "", cor.method = "spearman", cor.limit = 0.2)
- # a way to use the cor.limit argument just considering data1
- # obs1 <- 101:110 ; fun_permut(data1 = obs1, data2 = obs1, seed = 1, print.count = 10, cor.method = "spearman", cor.limit = 0.2)
- # fun_permut(data1 = 1:1e3, data2 = 1e3:1, seed = 1, print.count = 1e6, text.print = "", cor.method = "spearman", cor.limit = 0.7)
- # fun_permut(data1 = 1:1e2, data2 = 1e2:1, seed = 1, print.count = 1e3, cor.limit = 0.5)
- # fun_permut(data1 = c(0,0,0,0,0), n = 5, data2 = NULL, seed = 1, print.count = 1e3, cor.limit = 0.5)
- # DEBUGGING
- # data1 = LETTERS[1:5] ; data2 = NULL ; n = 1e6 ; seed = NULL ; print.count = 1e3 ; text.print = "" ; cor.method = "spearman" ; cor.limit = 0.2 ; warn.print = TRUE ; lib.path = NULL
- # data1 = LETTERS[1:5] ; data2 = NULL ; n = 10 ; seed = 22 ; print.count = 10 ; text.print = "" ; cor.method = "spearman" ; cor.limit = 0.2 ; warn.print = TRUE ; lib.path = NULL
- # data1 = 101:110 ; data2 = 21:30 ; n = 10 ; seed = 22 ; print.count = 10 ; text.print = "" ; cor.method = "spearman" ; cor.limit = 0.2 ; warn.print = TRUE ; lib.path = NULL
- # data1 = 1:1e3 ; data2 = 1e3:1 ; n = 20 ; seed = 22 ; print.count = 1e6 ; text.print = "" ; cor.method = "spearman" ; cor.limit = 0.5 ; warn.print = TRUE ; lib.path = NULL
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(length(utils::find("fun_pack", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_pack() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(length(utils::find("fun_round", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_pack() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = data1, class = "vector", fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & length(data1) < 2){
- tempo.cat <- paste0("ERROR IN ", function.name, ": data1 ARGUMENT MUST BE A VECTOR OF MINIMUM LENGTH 2. HERE IT IS: ", length(data1))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- if( ! is.null(data2)){
- tempo <- fun_check(data = data1, class = "vector", mode = "numeric", fun.name = function.name) ; eval(ee)
- if(tempo$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": data1 MUST BE A NUMERIC VECTOR IF data2 ARGUMENT IS SPECIFIED")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- tempo <- fun_check(data = data2, class = "vector", mode = "numeric", fun.name = function.name) ; eval(ee)
- if(length(data1) != length(data2)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": data1 AND data2 MUST BE VECTOR OF SAME LENGTH. HERE IT IS ", length(data1)," AND ", length(data2))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }else if(is.null(n)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": n ARGUMENT CANNOT BE NULL IF data2 ARGUMENT IS NULL")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- if( ! is.null(n)){
- tempo <- fun_check(data = n, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- }
- if( ! is.null(seed)){
- tempo <- fun_check(data = seed, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = TRUE, fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = print.count, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = text.print, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = cor.method, options = c("pearson", "kendall", "spearman"), length =1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = cor.limit, class = "vector", mode = "numeric", prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = warn.print, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(lib.path)){
- tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
- tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- }
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # package checking
- fun_pack(req.package = "lubridate", lib.path = lib.path)
- # end package checking
- # main code
- # code that protects set.seed() in the global environment
- # see also Protocol 100-rev0 Parallelization in R.docx
- if(exists(".Random.seed", envir = .GlobalEnv)){ # if .Random.seed does not exists, it means that no random operation has been performed yet in any R environment
- tempo.random.seed <- .Random.seed
- on.exit(assign(".Random.seed", tempo.random.seed, env = .GlobalEnv))
- }else{
- on.exit(set.seed(NULL)) # inactivate seeding -> return to complete randomness
- }
- set.seed(seed)
- # end code that protects set.seed() in the global environment
- ini.date <- Sys.time() # time of process begin, converted into seconds
- ini.time <- as.numeric(ini.date) # time of process begin, converted into seconds
- ini.pos <- 1:length(data1) # positions of data1 before permutation loops
- tempo.pos <- ini.pos # positions of data1 that will be modified during loops
- # pos.selec.seq <- ini.pos[-length(data1)] # selection of 1 position in initial position, without the last because always up permutation (pos -> pos+1 & pos+1 -> pos)
- pos.selec.seq.max <- length(ini.pos) - 1 # max position (used by sample.int() function). See below for - 1
- ini.warning.length <- options()$warning.length
- warn <- NULL
- warn.count <- 0
- count <- 0
- round <- 0
- BREAK <- FALSE
- tempo.cor <- 0
- if(is.null(data2)){
- if(length(table(data1)) == 1){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NO PERMUTATION PERFORMED BECAUSE data1 ARGUMENT SEEMS TO BE MADE OF IDENTICAL ELEMENTS: ", names(table(data1)))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn))) #
- }else{
- if(print.count > n){
- print.count <- n
- }
- cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "FOR LOOP OF ", n, " LOOPS INITIATED | LOOP COUNT: ", format(count, big.mark=",")))
- print.count.loop <- logical(length = print.count)
- print.count.loop[length(print.count.loop)] <- TRUE # not this to avoid long vector, but not forget to reset during printing: print.count.loop[(1:trunc(n / print.count) * print.count)] <- TRUE # counter to speedup
- count.loop <- 0
- pos <- sample.int(n = pos.selec.seq.max , size = print.count, replace = TRUE) # selection of random positions. BEWARE: n = pos.selec.seq.max because already - 1 (see above) but is connected to tempo.pos[c(pos2 + 1, pos2)] <- tempo.pos[c(pos2, pos2 + 1)]
- tempo.date.loop <- Sys.time()
- tempo.time.loop <- as.numeric(tempo.date.loop)
- for(i3 in 1:n){
- count.loop <- count.loop + 1
- pos2 <- pos[count.loop] # selection of 1 position
- tempo.pos[c(pos2 + 1, pos2)] <- tempo.pos[c(pos2, pos2 + 1)]
- if(print.count.loop[count.loop]){
- count.loop <- 0
- pos <- sample.int(n = pos.selec.seq.max , size = print.count, replace = TRUE) # BEWARE: never forget to resample here
- tempo.time <- as.numeric(Sys.time())
- tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - tempo.time.loop))
- final.loop <- (tempo.time - tempo.time.loop) / i3 * n # expected duration in seconds
- final.exp <- as.POSIXct(final.loop, origin = tempo.date.loop)
- cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "FOR LOOP ", i3, " / ", n, " | TIME SPENT: ", tempo.lapse, " | EXPECTED END: ", final.exp))
- }
- }
- count <- count + n # out of the loop to speedup
- cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "FOR LOOP ENDED | LOOP COUNT: ", format(count, big.mark=",")))
- cat("\n\n")
- }
- }else{
- if(length(table(data1)) == 1){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NO PERMUTATION PERFORMED BECAUSE data1 ARGUMENT SEEMS TO BE MADE OF IDENTICAL ELEMENTS: ", names(table(data1)))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn))) #
- tempo.cor <- 1
- }else if(length(table(data2)) == 1){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NO PERMUTATION PERFORMED BECAUSE data2 ARGUMENT SEEMS TO BE MADE OF IDENTICAL ELEMENTS: ", names(table(data2)))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn))) #
- tempo.cor <- 1
- }else{
- cor.ini <- cor(x = data1, y = data2, use = "pairwise.complete.obs", method = cor.method)
- tempo.cor <- cor.ini # correlation that will be modified during loops
- neg.cor <- FALSE
- if(tempo.cor < 0){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") INITIAL ", toupper(cor.method), " CORRELATION BETWEEN data1 AND data2 HAS BEEN DETECTED AS NEGATIVE: ", tempo.cor, ". THE LOOP STEPS WILL BE PERFORMED USING POSITIVE CORRELATIONS BUT THE FINAL CORRELATION WILL BE NEGATIVE")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn))) #
- neg.cor <- TRUE
- tempo.cor <- abs(tempo.cor)
- cor.ini <- abs(cor.ini)
- }
- if(tempo.cor < cor.limit){ # randomize directly all the position to be close to correlation zero
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") INITIAL ABSOLUTE VALUE OF THE ", toupper(cor.method), " CORRELATION ", fun_round(tempo.cor), " BETWEEN data1 AND data2 HAS BEEN DETECTED AS BELOW THE CORRELATION LIMIT PARAMETER ", cor.limit, "\nTHE data1 SEQUENCE HAS BEEN COMPLETELY RANDOMIZED TO CORRESPOND TO CORRELATION ZERO")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn))) #
- for(i4 in 1:5){ # done 5 times to be sure of the complete randomness
- tempo.pos <- sample(x = tempo.pos, size = length(tempo.pos), replace = FALSE)
- }
- count <- count + 5 # out of the loop to speedup
- }else{
- # smallest correlation decrease
- count <- count + 1 # 1 and not 0 because already 1 performed just below
- pos <- sample.int(n = pos.selec.seq.max , size = 1, replace = TRUE) # selection of 1 position # pos.selec.seq.max because selection of 1 position in initial position, without the last because always up permutation (pos -> pos+1 & pos+1 -> pos)
- tempo.pos[c(pos + 1, pos)] <- tempo.pos[c(pos, pos + 1)]
- tempo.cor <- abs(cor(x = data1[tempo.pos], y = data2, use = "pairwise.complete.obs", method = cor.method))
- smallest.cor.dec <- cor.ini - tempo.cor
- # end smallest correlation decrease
- # going out of tempo.cor == cor.ini
- cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "CORRELATION DECREASE AFTER A SINGLE PERMUTATION: ", fun_round(smallest.cor.dec, 4)))
- cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "FIRST WHILE LOOP STEP -> GOING OUT FROM EQUALITY | LOOP COUNT: ", format(count, big.mark=","), " | CORRELATION LIMIT: ", fun_round(cor.limit, 4), " | ABS TEMPO CORRELATION: ", fun_round(tempo.cor, 4)))
- print.count.loop <- logical(length = print.count)
- print.count.loop[length(print.count.loop)] <- TRUE # counter to speedup
- count.loop <- 0 #
- pos <- sample.int(n = pos.selec.seq.max , size = print.count, replace = TRUE) # selection of random positions. BEWARE: n = pos.selec.seq.max because already - 1 (see above) but is connected to tempo.pos[c(pos2 + 1, pos2)] <- tempo.pos[c(pos2, pos2 + 1)]
- tempo.date.loop <- Sys.time()
- tempo.time.loop <- as.numeric(tempo.date.loop)
- while(tempo.cor == cor.ini){ # to be out of equality between tempo.cor and cor.ini at the beginning (only valid for very long vector)
- count <- count + 1
- count.loop <- count.loop + 1
- pos2 <- pos[count.loop]
- tempo.pos[c(pos2 + 1, pos2)] <- tempo.pos[c(pos2, pos2 + 1)]
- tempo.cor <- abs(cor(x = data1[tempo.pos], y = data2, use = "pairwise.complete.obs", method = cor.method))
- if(print.count.loop[count.loop]){
- count.loop <- 0
- pos <- sample.int(n = pos.selec.seq.max , size = print.count, replace = TRUE) # BEWARE: never forget to resample here
- tempo.time <- as.numeric(Sys.time())
- tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - tempo.time.loop))
- cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "FIRST WHILE LOOP STEP", format(count.loop, big.mark=","), " / ? | COUNT: ", format(count, big.mark=","), " | CORRELATION LIMIT: ", fun_round(cor.limit, 4), " | ABS TEMPO CORRELATION: ", fun_round(tempo.cor, 4), " | TIME SPENT: ", tempo.lapse))
- }
- }
- tempo.time <- as.numeric(Sys.time())
- tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - ini.time))
- cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "FIRST WHILE LOOP STEP END | LOOP COUNT: ", format(count, big.mark=","), " | CORRELATION LIMIT: ", fun_round(cor.limit, 4), " | ABS TEMPO CORRELATION: ", fun_round(tempo.cor, 4), " | TOTAL SPENT TIME: ", tempo.lapse))
- if(tempo.cor < cor.limit){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE FIRST FOR & WHILE LOOP STEPS HAVE BEEN TOO FAR AND SUBSEQUENT LOOP STEPS WILL NOT RUN")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- # end going out of tempo.cor == cor.ini
- # estimation of the average correlation decrease per loop on x loops and for loop execution
- cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "WHILE/FOR LOOPS INITIATION | LOOP COUNT: ", format(count, big.mark=","), " | CORRELATION LIMIT: ", fun_round(cor.limit, 4), " | ABS TEMPO CORRELATION: ", fun_round(tempo.cor, 4)))
- count.est <- 1e5
- first.round <- TRUE
- GOBACK <- FALSE
- while(tempo.cor > cor.limit){
- round <- round + 1
- # estimation step
- if(first.round == TRUE){
- first.round <- FALSE
- cor.dec.per.loop <- numeric(length = 5)
- loop.nb.est <- Inf
- cor.est.ini <- tempo.cor
- cor.est <- numeric(length = 5)
- for(i6 in 1:5){ # connected to cor.dec.per.loop
- tempo.pos.est <- tempo.pos
- pos <- sample.int(n = pos.selec.seq.max , size = count.est, replace = TRUE) # selection of n position
- for(i7 in 1:count.est){
- pos2 <- pos[i7] # selection of 1 position
- tempo.pos.est[c(pos2 + 1, pos2)] <- tempo.pos.est[c(pos2, pos2 + 1)]
- }
- tempo.cor.est <- abs(cor(x = data1[tempo.pos.est], y = data2, use = "pairwise.complete.obs", method = cor.method))
- cor.est[i6] <- tempo.cor.est
- tempo.cor.dec.per.loop <- (cor.est.ini - tempo.cor.est) / count.est # correlation decrease per loop
- if(is.na(tempo.cor.dec.per.loop) | ! is.finite(tempo.cor.dec.per.loop)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 2\ncor.est.ini: ", cor.est.ini, "\ntempo.cor.est: ", tempo.cor.est)
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- cor.dec.per.loop[i6] <- tempo.cor.dec.per.loop
- }
- cor.est <- cor.est[which.max(cor.dec.per.loop)] # max to avoid to go to far with for loop (tempo.cor below tempo.limit)
- cor.dec.per.loop <- max(cor.dec.per.loop, na.rm = TRUE) # max to avoid to go to far with for loop (tempo.cor below tempo.limit)
- loop.nb.est <- round((tempo.cor - cor.limit) / cor.dec.per.loop)
- }else{
- if(GOBACK == TRUE){
- loop.nb.est <- round(loop.nb.est / 2)
- }else{
- cor.dec.per.loop <- (cor.ini - tempo.cor) / count
- loop.nb.est <- round((tempo.cor - cor.limit) / cor.dec.per.loop)
- }
- }
- # end estimation step
- # loop step
- if(is.na(loop.nb.est) | ! is.finite(loop.nb.est)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 1\nloop.nb.est: ", loop.nb.est, "\ncor.ini: ", cor.ini, "\ntempo.cor: ", tempo.cor, "\ncor.limit: ", cor.limit, "\ncor.dec.per.loop: ", cor.dec.per.loop)
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else if(loop.nb.est > 1e4){ # below -> leave the while loop
- tempo.pos.secu <- tempo.pos
- count.secu <- count
- tempo.cor.secu <- tempo.cor
- cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "INITIAL SETTINGS BEFORE ROUND: ", round, " | LOOP COUNT: ", format(count, big.mark=","), " | GO BACK: ", GOBACK, " | LOOP NUMBER ESTIMATION: ", format(loop.nb.est, big.mark=","), " | CORRELATION LIMIT: ", fun_round(cor.limit, 4), " | ABS TEMPO CORRELATION: ", fun_round(tempo.cor, 4)))
- print.count.loop <- logical(length = print.count)
- print.count.loop[length(print.count.loop)] <- TRUE # not this to avoid long vector, but not forget to reset during printing: print.count.loop[(1:trunc(n / print.count) * print.count)] <- TRUE # counter to speedup
- count.loop <- 0
- pos <- sample.int(n = pos.selec.seq.max , size = print.count, replace = TRUE) # selection of random positions. BEWARE: n = pos.selec.seq.max because already - 1 (see above) but is connected to tempo.pos[c(pos2 + 1, pos2)] <- tempo.pos[c(pos2, pos2 + 1)]
- tempo.date.loop <- Sys.time()
- tempo.time.loop <- as.numeric(tempo.date.loop)
- for(i6 in 1:loop.nb.est){
- count.loop <- count.loop + 1
- pos2 <- pos[count.loop] # selection of 1 position
- tempo.pos[c(pos2 + 1, pos2)] <- tempo.pos[c(pos2, pos2 + 1)]
- if(print.count.loop[count.loop]){
- count.loop <- 0
- pos <- sample.int(n = pos.selec.seq.max , size = print.count, replace = TRUE) # BEWARE: never forget to resample here
- tempo.time <- as.numeric(Sys.time())
- tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - tempo.time.loop))
- final.loop <- (tempo.time - tempo.time.loop) / i6 * loop.nb.est # expected duration in seconds # intra nb.compar loop lapse: time lapse / cycles done * cycles remaining
- final.exp <- as.POSIXct(final.loop, origin = tempo.date.loop)
- cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "FOR LOOP | ROUND ", round, " | LOOP: ", format(i6, big.mark=","), " / ", format(loop.nb.est, big.mark=","), " | TIME SPENT: ", tempo.lapse, " | EXPECTED END: ", final.exp))
- }
- }
- count <- count + loop.nb.est # out of the loop to speedup
- tempo.cor <- abs(cor(x = data1[tempo.pos], y = data2, use = "pairwise.complete.obs", method = cor.method))
- if(tempo.cor > tempo.cor.secu | ((tempo.cor - cor.limit) < 0 & abs(tempo.cor - cor.limit) > smallest.cor.dec * round(log10(max(ini.pos, na.rm = TRUE))))){
- GOBACK <- TRUE
- tempo.pos <- tempo.pos.secu
- count <- count.secu
- tempo.cor <- tempo.cor.secu
- }else{
- GOBACK <- FALSE
- }
- }else{
- cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "FINAL WHILE LOOP | LOOP COUNT: ", format(count, big.mark=","), " | CORRELATION LIMIT: ", fun_round(cor.limit, 4), " | ABS TEMPO CORRELATION: ", fun_round(tempo.cor, 4)))
- print.count.loop <- logical(length = print.count)
- print.count.loop[length(print.count.loop)] <- TRUE # counter to speedup
- count.loop <- 0 #
- pos <- sample.int(n = pos.selec.seq.max , size = print.count, replace = TRUE) # selection of random positions. BEWARE: n = pos.selec.seq.max because already - 1 (see above) but is connected to tempo.pos[c(pos2 + 1, pos2)] <- tempo.pos[c(pos2, pos2 + 1)]
- tempo.cor.loop <- tempo.cor
- tempo.date.loop <- Sys.time()
- tempo.time.loop <- as.numeric(tempo.date.loop)
- while(tempo.cor > cor.limit){
- count <- count + 1
- count.loop <- count.loop + 1
- pos2 <- pos[count.loop]
- tempo.pos[c(pos2 + 1, pos2)] <- tempo.pos[c(pos2, pos2 + 1)]
- tempo.cor <- abs(cor(x = data1[tempo.pos], y = data2, use = "pairwise.complete.obs", method = cor.method))
- if(print.count.loop[count.loop]){
- count.loop <- 0
- pos <- sample.int(n = pos.selec.seq.max , size = print.count, replace = TRUE) # BEWARE: never forget to resample here
- tempo.time <- as.numeric(Sys.time())
- tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - tempo.time.loop))
- final.loop <- (tempo.time - tempo.time.loop) / (tempo.cor.loop - tempo.cor) * (tempo.cor - cor.limit) # expected duration in seconds # tempo.cor.loop - tempo.cor always positive and tempo.cor decreases progressively starting from tempo.cor.loop
- final.exp <- as.POSIXct(final.loop, origin = tempo.date.loop)
- cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "WHILE LOOP | LOOP NB: ", format(count.loop, big.mark=","), " | COUNT: ", format(count, big.mark=","), " | CORRELATION LIMIT: ", fun_round(cor.limit, 4), " | ABS TEMPO CORRELATION: ", fun_round(tempo.cor, 4), " | TIME SPENT: ", tempo.lapse, " | EXPECTED END: ", final.exp))
- }
- }
- }
- }
- tempo.time <- as.numeric(Sys.time())
- tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - ini.time))
- cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "WHILE/FOR LOOPS END | LOOP COUNT: ", format(count, big.mark=","), " | NB OF ROUNDS: ", round, " | CORRELATION LIMIT: ", fun_round(cor.limit, 4), " | ABS TEMPO CORRELATION: ", fun_round(tempo.cor, 4), " | TOTAL SPENT TIME: ", tempo.lapse))
- }
- tempo.cor <- ifelse(neg.cor == TRUE, -tempo.cor, tempo.cor)
- }
- }
- cat("\n\n")
- if(warn.print == TRUE & ! is.null(warn)){
- options(warning.length = 8170)
- on.exit(warning(paste0("FROM ", function.name, ":\n\n", warn), call. = FALSE), add = TRUE)
- on.exit(exp = options(warning.length = ini.warning.length), add = TRUE)
- }
- output <- list(data = data1[tempo.pos], warn = warn, cor = if(is.null(data2)){cor(ini.pos, tempo.pos, method = "spearman")}else{tempo.cor}, count = count)
- return(output)
+# end required function checking
+# argument checking
+# argument checking with fun_check()
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = mat, class = "matrix", na.contain = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = empty.cell.string, class = "vector", na.contain = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = warn.print, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# end argument checking with fun_check()
+# argument checking without fun_check()
+if(ncol(mat) != nrow(mat)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": mat ARGUMENT MUST BE A SQUARE MATRIX")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! (mode(mat) %in% c("numeric", "character"))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": mat ARGUMENT MUST BE A NUMERIC OR CHARACTER MATRIX")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(nrow(mat) == 1 & ncol(mat) == 1){
+tempo.cat <- paste0("ERROR IN ", function.name, ": mat ARGUMENT CANNOT BE A SQUARE MATRIX MADE OF A SINGLE CASE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(ifelse(is.na(empty.cell.string), ! any(is.na(mat)), ! any(mat == empty.cell.string, na.rm = TRUE))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": mat ARGUMENT MATRIX MUST HAVE CELLS WITH THE EMPTY STRING SPECIFIED IN empty.cell.string ARGUMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end argument checking without fun_check()
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+list.diag <- vector("list", length = nrow(mat) - 1)
+for(i1 in 1:(nrow(mat) - 1)){
+list.diag[[i1]] <- numeric(length = nrow(mat) - i1)
+}
+sector <- c("topleft", "topright", "bottomright", "bottomleft")
+diag.scan <-c( # same order as sector. Recover each diag from center to corner
+"mat[as.matrix(as.data.frame(list(1:(nrow(mat) - i2), (ncol(mat) -i2):1), stringsAsFactors = TRUE))]", # topleft part
+"mat[as.matrix(as.data.frame(list(1:(nrow(mat) - i2), (1:ncol(mat))[-(1:i2)]), stringsAsFactors = TRUE))]", # topright part
+"mat[as.matrix(as.data.frame(list((1 + i2):nrow(mat), ncol(mat):(1 + i2)), stringsAsFactors = TRUE))]", # bottomright part
+"mat[as.matrix(as.data.frame(list((1 + i2):nrow(mat), 1:(ncol(mat) -i2)), stringsAsFactors = TRUE))]" # bottomleft part
+)
+# empty part detection
+tempo.list.diag <- list.diag
+empty.sector <- NULL
+full.sector <- NULL
+ini.warning.length <- options()$warning.length
+warn <- NULL
+warn.count <- 0
+for(i1 in 1:length(sector)){
+tempo.list.diag <- list.diag
+for(i2 in 1:(nrow(mat) - 1)){
+tempo.list.diag[[i2]] <- eval(parse(text = diag.scan[i1]))
+if(ifelse(is.na(empty.cell.string), ! all(is.na(tempo.list.diag[[i2]])), ! (all(tempo.list.diag[[i2]] == empty.cell.string, na.rm = TRUE) & ! (is.na(all(tempo.list.diag[[i2]] == empty.cell.string, na.rm = FALSE)))))){ # I had to add this ! (is.na(all(tempo.list.diag[[i2]] == empty.cell.string, na.rm = FALSE))) because all(tempo.list.diag[[i2]] == empty.cell.string, na.rm = FALSE) gives NA and not FALSE if one NA in tempo.list.diag[[i2]] -> not good for if()
+full.sector <- c(full.sector, sector[i1])
+break
+}
+}
+if(i1 == nrow(mat) - 1){
+if(all(unlist(lapply(tempo.list.diag, FUN = function(x){if(is.na(empty.cell.string)){is.na(x)}else{x == empty.cell.string}})), na.rm = TRUE)){
+empty.sector <- c(empty.sector, sector[i1])
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") EMPTY SECTOR DETECTED ON THE ", toupper(sector[i1]), " CORNER, FULL OF ", empty.cell.string)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE ", toupper(sector[i1]), " SECTOR, DETECTED AS EMPTY, IS NOT? DIFFERENT VALUES IN THIS SECTOR:\n", paste(names(table(unlist(tempo.list.diag), useNA = "ifany")), collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+}
+# end empty part detection
+if(length(empty.sector) == 0){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") ACCORDING TO empty.cell.string ARGUMENT (", empty.cell.string, "), mat ARGUMENT MATRIX HAS ZERO EMPTY HALF PART")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}else{
+if(length(empty.sector) > 1){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ACCORDING TO empty.cell.string ARGUMENT (", empty.cell.string, "), mat ARGUMENT MATRIX HAS MORE THAN ONE EMPTY HALF PART (ACCORDING TO THE GRAND DIAGONAL): ", paste(empty.sector, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else if(any(full.sector %in% empty.sector, na.rm = TRUE)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE FUNCTION HAS DETECTED EMPTY AND NON EMPTY HALF PART IN THE SAME SECTOR: ", paste(full.sector[full.sector %in% empty.sector], collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else if(length(empty.sector) + length(full.sector)!= 4){
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE FUNCTION HAS DETECTED MORE OR LESS SECTORS THAN 4:\nHALF SECTORS:", paste(empty.sector, collapse = " "), "\nFULL SECTORS:", paste(full.sector, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") ", toupper(empty.sector), " SECTOR HAS BEEN COMPLETED TO BECOME SYMMETRICAL")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+# matrix filling
+for(i1 in 1:(nrow(mat) - 1)){
+if(empty.sector == "topleft"){
+eval(parse(text = paste0(diag.scan[1], " <- ", diag.scan[3])))
+}else if(empty.sector == "topright"){
+eval(parse(text = paste0(diag.scan[2], " <- ", diag.scan[4])))
+}else if(empty.sector == "bottomright"){
+eval(parse(text = paste0(diag.scan[3], " <- ", diag.scan[1])))
+}else if(empty.sector == "bottomleft"){
+eval(parse(text = paste0(diag.scan[4], " <- ", diag.scan[2])))
+}
+}
+# end matrix filling
+}
+if(warn.print == TRUE & ! is.null(warn)){
+options(warning.length = 8170)
+on.exit(warning(paste0("FROM ", function.name, ":\n\n", warn), call. = FALSE))
+on.exit(exp = options(warning.length = ini.warning.length), add = TRUE)
+}
+return(list(mat = mat, warn = warn))
+}
+
+
+######## fun_permut() #### progressively breaks a vector order
+
+
+fun_permut <- function(
+data1,
+data2 = NULL,
+n = NULL,
+seed = NULL,
+print.count = 10,
+text.print = "",
+cor.method = "spearman",
+cor.limit = 0.2,
+warn.print = FALSE,
+lib.path = NULL
+){
+# AIM
+# reorder the elements of the data1 vector by flipping 2 randomly selected consecutive positions either:
+# 1) n times (when n is precised) or
+# 2) until the correlation between data1 and data2 decreases down to the cor.limit (0.2 by default). See cor.limit below to deal with negative correlations
+# Example of consecutive position flipping: ABCD -> BACD -> BADC, etc.
+# designed for discrete values, but worls also for continuous values
+# WARNINGS
+# see # https://www.r-bloggers.com/strategies-to-speedup-r-code/ for code speedup
+# the random switch of non consecutive positions (ABCD -> DBCA for instance) does not work very well as the correlation is quickly obtained but the initial vector structure is mainly kept (no much order). Ths code would be: pos <- ini.pos[1:2] ; pos <- sample.int(n = n , size = 2, replace = FALSE) ; tempo.pos[pos] <- tempo.pos[rev(pos)]
+# ARGUMENTS
+# data1: a vector of at least 2 elements. Must be numeric if data2 is specified
+# data2: a numeric vector of same length as data1
+# n: number of times "flipping 2 randomly selected consecutive positions". Ignored if data2 is specified
+# seed: integer number used by set.seed(). Write NULL if random result is required, an integer otherwise. BEWARE: if not NULL, fun_permut() will systematically return the same result when the other parameters keep the same settings
+# print.count: interger value. Print a working progress message every print.count during loops. BEWARE: can increase substentially the time to complete the process using a small value, like 10 for instance. Use Inf is no loop message desired
+# text.print: optional message to add to the working progress message every print.count loop
+# cor.method: correlation method. Either "pearson", "kendall" or "spearman". Ignored if data2 is not specified
+# cor.limit: a correlation limit (between 0 and 1). Ignored if data2 is not specified. Compute the correlation between data1 and data2, permute the data1 values, and stop the permutation process when the correlation between data1 and data2 decreases down below the cor limit value (0.2 by default). If cor(data1, data2) is negative, then -cor.limit is used and the process stops until the correlation between data1 and data2 increases up over cor.limit (-0.2 by default). BEWARE: write a positive cor.limit even if cor(data1, data2) is known to be negative. The function will automatically uses -cor.limit. If the initial correlation is already below cor.limit (positive correlation) or over -cor.limit (negative correlation), then the data1 value positions are completely randomized (correlation between data1 and data2 is expected to be 0)
+# warn.print: logical. Print warnings at the end of the execution? No print if no warning messages
+# lib.path: character vector specifying the absolute pathways of the directories containing the required packages if not in the default directories. Ignored if NULL
+# RETURN
+# a list containing:
+# $data: the modified vector
+# $warn: potential warning messages (in case of negative correlation when data2 is specified). NULL if non warning message
+# $cor: a spearman correlation between the initial positions (1:length(data1) and the final positions if data2 is not specified and the final correlation between data1 and data2 otherwise, according to cor.method
+# $count: the number of loops used
+# REQUIRED PACKAGES
+# lubridate
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# fun_pack()
+# fun_round()
+# EXAMPLES
+# example (1) showing that for loop, used in fun_permut(), is faster than while loop
+# ini.time <- as.numeric(Sys.time()) ; count <- 0 ; for(i0 in 1:1e9){count <- count + 1} ; tempo.time <- as.numeric(Sys.time()) ; tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - ini.time)) ; tempo.lapse
+# example (2) showing that for loop, used in fun_permut(), is faster than while loop
+# ini.time <- as.numeric(Sys.time()) ; count <- 0 ; while(count < 1e9){count <- count + 1} ; tempo.time <- as.numeric(Sys.time()) ; tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - ini.time)) ; tempo.lapse
+# fun_permut(data1 = LETTERS[1:5], data2 = NULL, n = 100, seed = 1, print.count = 10, text.print = "CPU NB 4")
+# fun_permut(data1 = 101:110, data2 = 21:30, seed = 1, print.count = 1e4, text.print = "", cor.method = "spearman", cor.limit = 0.2)
+# a way to use the cor.limit argument just considering data1
+# obs1 <- 101:110 ; fun_permut(data1 = obs1, data2 = obs1, seed = 1, print.count = 10, cor.method = "spearman", cor.limit = 0.2)
+# fun_permut(data1 = 1:1e3, data2 = 1e3:1, seed = 1, print.count = 1e6, text.print = "", cor.method = "spearman", cor.limit = 0.7)
+# fun_permut(data1 = 1:1e2, data2 = 1e2:1, seed = 1, print.count = 1e3, cor.limit = 0.5)
+# fun_permut(data1 = c(0,0,0,0,0), n = 5, data2 = NULL, seed = 1, print.count = 1e3, cor.limit = 0.5)
+# DEBUGGING
+# data1 = LETTERS[1:5] ; data2 = NULL ; n = 1e6 ; seed = NULL ; print.count = 1e3 ; text.print = "" ; cor.method = "spearman" ; cor.limit = 0.2 ; warn.print = TRUE ; lib.path = NULL
+# data1 = LETTERS[1:5] ; data2 = NULL ; n = 10 ; seed = 22 ; print.count = 10 ; text.print = "" ; cor.method = "spearman" ; cor.limit = 0.2 ; warn.print = TRUE ; lib.path = NULL
+# data1 = 101:110 ; data2 = 21:30 ; n = 10 ; seed = 22 ; print.count = 10 ; text.print = "" ; cor.method = "spearman" ; cor.limit = 0.2 ; warn.print = TRUE ; lib.path = NULL
+# data1 = 1:1e3 ; data2 = 1e3:1 ; n = 20 ; seed = 22 ; print.count = 1e6 ; text.print = "" ; cor.method = "spearman" ; cor.limit = 0.5 ; warn.print = TRUE ; lib.path = NULL
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(length(utils::find("fun_pack", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_pack() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(length(utils::find("fun_round", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_pack() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = data1, class = "vector", fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & length(data1) < 2){
+tempo.cat <- paste0("ERROR IN ", function.name, ": data1 ARGUMENT MUST BE A VECTOR OF MINIMUM LENGTH 2. HERE IT IS: ", length(data1))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+if( ! is.null(data2)){
+tempo <- fun_check(data = data1, class = "vector", mode = "numeric", fun.name = function.name) ; eval(ee)
+if(tempo$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": data1 MUST BE A NUMERIC VECTOR IF data2 ARGUMENT IS SPECIFIED")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+tempo <- fun_check(data = data2, class = "vector", mode = "numeric", fun.name = function.name) ; eval(ee)
+if(length(data1) != length(data2)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": data1 AND data2 MUST BE VECTOR OF SAME LENGTH. HERE IT IS ", length(data1)," AND ", length(data2))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}else if(is.null(n)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": n ARGUMENT CANNOT BE NULL IF data2 ARGUMENT IS NULL")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+if( ! is.null(n)){
+tempo <- fun_check(data = n, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+}
+if( ! is.null(seed)){
+tempo <- fun_check(data = seed, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = TRUE, fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = print.count, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = text.print, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = cor.method, options = c("pearson", "kendall", "spearman"), length =1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = cor.limit, class = "vector", mode = "numeric", prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = warn.print, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(lib.path)){
+tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
+tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+}
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# package checking
+fun_pack(req.package = "lubridate", lib.path = lib.path)
+# end package checking
+# main code
+# code that protects set.seed() in the global environment
+# see also Protocol 100-rev0 Parallelization in R.docx
+if(exists(".Random.seed", envir = .GlobalEnv)){ # if .Random.seed does not exists, it means that no random operation has been performed yet in any R environment
+tempo.random.seed <- .Random.seed
+on.exit(assign(".Random.seed", tempo.random.seed, env = .GlobalEnv))
+}else{
+on.exit(set.seed(NULL)) # inactivate seeding -> return to complete randomness
+}
+set.seed(seed)
+# end code that protects set.seed() in the global environment
+ini.date <- Sys.time() # time of process begin, converted into seconds
+ini.time <- as.numeric(ini.date) # time of process begin, converted into seconds
+ini.pos <- 1:length(data1) # positions of data1 before permutation loops
+tempo.pos <- ini.pos # positions of data1 that will be modified during loops
+# pos.selec.seq <- ini.pos[-length(data1)] # selection of 1 position in initial position, without the last because always up permutation (pos -> pos+1 & pos+1 -> pos)
+pos.selec.seq.max <- length(ini.pos) - 1 # max position (used by sample.int() function). See below for - 1
+ini.warning.length <- options()$warning.length
+warn <- NULL
+warn.count <- 0
+count <- 0
+round <- 0
+BREAK <- FALSE
+tempo.cor <- 0
+if(is.null(data2)){
+if(length(table(data1)) == 1){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NO PERMUTATION PERFORMED BECAUSE data1 ARGUMENT SEEMS TO BE MADE OF IDENTICAL ELEMENTS: ", names(table(data1)))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn))) #
+}else{
+if(print.count > n){
+print.count <- n
+}
+cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "FOR LOOP OF ", n, " LOOPS INITIATED | LOOP COUNT: ", format(count, big.mark=",")))
+print.count.loop <- logical(length = print.count)
+print.count.loop[length(print.count.loop)] <- TRUE # not this to avoid long vector, but not forget to reset during printing: print.count.loop[(1:trunc(n / print.count) * print.count)] <- TRUE # counter to speedup
+count.loop <- 0
+pos <- sample.int(n = pos.selec.seq.max , size = print.count, replace = TRUE) # selection of random positions. BEWARE: n = pos.selec.seq.max because already - 1 (see above) but is connected to tempo.pos[c(pos2 + 1, pos2)] <- tempo.pos[c(pos2, pos2 + 1)]
+tempo.date.loop <- Sys.time()
+tempo.time.loop <- as.numeric(tempo.date.loop)
+for(i3 in 1:n){
+count.loop <- count.loop + 1
+pos2 <- pos[count.loop] # selection of 1 position
+tempo.pos[c(pos2 + 1, pos2)] <- tempo.pos[c(pos2, pos2 + 1)]
+if(print.count.loop[count.loop]){
+count.loop <- 0
+pos <- sample.int(n = pos.selec.seq.max , size = print.count, replace = TRUE) # BEWARE: never forget to resample here
+tempo.time <- as.numeric(Sys.time())
+tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - tempo.time.loop))
+final.loop <- (tempo.time - tempo.time.loop) / i3 * n # expected duration in seconds
+final.exp <- as.POSIXct(final.loop, origin = tempo.date.loop)
+cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "FOR LOOP ", i3, " / ", n, " | TIME SPENT: ", tempo.lapse, " | EXPECTED END: ", final.exp))
+}
+}
+count <- count + n # out of the loop to speedup
+cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "FOR LOOP ENDED | LOOP COUNT: ", format(count, big.mark=",")))
+cat("\n\n")
+}
+}else{
+if(length(table(data1)) == 1){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NO PERMUTATION PERFORMED BECAUSE data1 ARGUMENT SEEMS TO BE MADE OF IDENTICAL ELEMENTS: ", names(table(data1)))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn))) #
+tempo.cor <- 1
+}else if(length(table(data2)) == 1){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NO PERMUTATION PERFORMED BECAUSE data2 ARGUMENT SEEMS TO BE MADE OF IDENTICAL ELEMENTS: ", names(table(data2)))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn))) #
+tempo.cor <- 1
+}else{
+cor.ini <- cor(x = data1, y = data2, use = "pairwise.complete.obs", method = cor.method)
+tempo.cor <- cor.ini # correlation that will be modified during loops
+neg.cor <- FALSE
+if(tempo.cor < 0){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") INITIAL ", toupper(cor.method), " CORRELATION BETWEEN data1 AND data2 HAS BEEN DETECTED AS NEGATIVE: ", tempo.cor, ". THE LOOP STEPS WILL BE PERFORMED USING POSITIVE CORRELATIONS BUT THE FINAL CORRELATION WILL BE NEGATIVE")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn))) #
+neg.cor <- TRUE
+tempo.cor <- abs(tempo.cor)
+cor.ini <- abs(cor.ini)
+}
+if(tempo.cor < cor.limit){ # randomize directly all the position to be close to correlation zero
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") INITIAL ABSOLUTE VALUE OF THE ", toupper(cor.method), " CORRELATION ", fun_round(tempo.cor), " BETWEEN data1 AND data2 HAS BEEN DETECTED AS BELOW THE CORRELATION LIMIT PARAMETER ", cor.limit, "\nTHE data1 SEQUENCE HAS BEEN COMPLETELY RANDOMIZED TO CORRESPOND TO CORRELATION ZERO")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn))) #
+for(i4 in 1:5){ # done 5 times to be sure of the complete randomness
+tempo.pos <- sample(x = tempo.pos, size = length(tempo.pos), replace = FALSE)
+}
+count <- count + 5 # out of the loop to speedup
+}else{
+# smallest correlation decrease
+count <- count + 1 # 1 and not 0 because already 1 performed just below
+pos <- sample.int(n = pos.selec.seq.max , size = 1, replace = TRUE) # selection of 1 position # pos.selec.seq.max because selection of 1 position in initial position, without the last because always up permutation (pos -> pos+1 & pos+1 -> pos)
+tempo.pos[c(pos + 1, pos)] <- tempo.pos[c(pos, pos + 1)]
+tempo.cor <- abs(cor(x = data1[tempo.pos], y = data2, use = "pairwise.complete.obs", method = cor.method))
+smallest.cor.dec <- cor.ini - tempo.cor
+# end smallest correlation decrease
+# going out of tempo.cor == cor.ini
+cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "CORRELATION DECREASE AFTER A SINGLE PERMUTATION: ", fun_round(smallest.cor.dec, 4)))
+cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "FIRST WHILE LOOP STEP -> GOING OUT FROM EQUALITY | LOOP COUNT: ", format(count, big.mark=","), " | CORRELATION LIMIT: ", fun_round(cor.limit, 4), " | ABS TEMPO CORRELATION: ", fun_round(tempo.cor, 4)))
+print.count.loop <- logical(length = print.count)
+print.count.loop[length(print.count.loop)] <- TRUE # counter to speedup
+count.loop <- 0 #
+pos <- sample.int(n = pos.selec.seq.max , size = print.count, replace = TRUE) # selection of random positions. BEWARE: n = pos.selec.seq.max because already - 1 (see above) but is connected to tempo.pos[c(pos2 + 1, pos2)] <- tempo.pos[c(pos2, pos2 + 1)]
+tempo.date.loop <- Sys.time()
+tempo.time.loop <- as.numeric(tempo.date.loop)
+while(tempo.cor == cor.ini){ # to be out of equality between tempo.cor and cor.ini at the beginning (only valid for very long vector)
+count <- count + 1
+count.loop <- count.loop + 1
+pos2 <- pos[count.loop]
+tempo.pos[c(pos2 + 1, pos2)] <- tempo.pos[c(pos2, pos2 + 1)]
+tempo.cor <- abs(cor(x = data1[tempo.pos], y = data2, use = "pairwise.complete.obs", method = cor.method))
+if(print.count.loop[count.loop]){
+count.loop <- 0
+pos <- sample.int(n = pos.selec.seq.max , size = print.count, replace = TRUE) # BEWARE: never forget to resample here
+tempo.time <- as.numeric(Sys.time())
+tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - tempo.time.loop))
+cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "FIRST WHILE LOOP STEP", format(count.loop, big.mark=","), " / ? | COUNT: ", format(count, big.mark=","), " | CORRELATION LIMIT: ", fun_round(cor.limit, 4), " | ABS TEMPO CORRELATION: ", fun_round(tempo.cor, 4), " | TIME SPENT: ", tempo.lapse))
+}
+}
+tempo.time <- as.numeric(Sys.time())
+tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - ini.time))
+cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "FIRST WHILE LOOP STEP END | LOOP COUNT: ", format(count, big.mark=","), " | CORRELATION LIMIT: ", fun_round(cor.limit, 4), " | ABS TEMPO CORRELATION: ", fun_round(tempo.cor, 4), " | TOTAL SPENT TIME: ", tempo.lapse))
+if(tempo.cor < cor.limit){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE FIRST FOR & WHILE LOOP STEPS HAVE BEEN TOO FAR AND SUBSEQUENT LOOP STEPS WILL NOT RUN")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+# end going out of tempo.cor == cor.ini
+# estimation of the average correlation decrease per loop on x loops and for loop execution
+cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "WHILE/FOR LOOPS INITIATION | LOOP COUNT: ", format(count, big.mark=","), " | CORRELATION LIMIT: ", fun_round(cor.limit, 4), " | ABS TEMPO CORRELATION: ", fun_round(tempo.cor, 4)))
+count.est <- 1e5
+first.round <- TRUE
+GOBACK <- FALSE
+while(tempo.cor > cor.limit){
+round <- round + 1
+# estimation step
+if(first.round == TRUE){
+first.round <- FALSE
+cor.dec.per.loop <- numeric(length = 5)
+loop.nb.est <- Inf
+cor.est.ini <- tempo.cor
+cor.est <- numeric(length = 5)
+for(i6 in 1:5){ # connected to cor.dec.per.loop
+tempo.pos.est <- tempo.pos
+pos <- sample.int(n = pos.selec.seq.max , size = count.est, replace = TRUE) # selection of n position
+for(i7 in 1:count.est){
+pos2 <- pos[i7] # selection of 1 position
+tempo.pos.est[c(pos2 + 1, pos2)] <- tempo.pos.est[c(pos2, pos2 + 1)]
+}
+tempo.cor.est <- abs(cor(x = data1[tempo.pos.est], y = data2, use = "pairwise.complete.obs", method = cor.method))
+cor.est[i6] <- tempo.cor.est
+tempo.cor.dec.per.loop <- (cor.est.ini - tempo.cor.est) / count.est # correlation decrease per loop
+if(is.na(tempo.cor.dec.per.loop) | ! is.finite(tempo.cor.dec.per.loop)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 2\ncor.est.ini: ", cor.est.ini, "\ntempo.cor.est: ", tempo.cor.est)
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+cor.dec.per.loop[i6] <- tempo.cor.dec.per.loop
+}
+cor.est <- cor.est[which.max(cor.dec.per.loop)] # max to avoid to go to far with for loop (tempo.cor below tempo.limit)
+cor.dec.per.loop <- max(cor.dec.per.loop, na.rm = TRUE) # max to avoid to go to far with for loop (tempo.cor below tempo.limit)
+loop.nb.est <- round((tempo.cor - cor.limit) / cor.dec.per.loop)
+}else{
+if(GOBACK == TRUE){
+loop.nb.est <- round(loop.nb.est / 2)
+}else{
+cor.dec.per.loop <- (cor.ini - tempo.cor) / count
+loop.nb.est <- round((tempo.cor - cor.limit) / cor.dec.per.loop)
+}
+}
+# end estimation step
+# loop step
+if(is.na(loop.nb.est) | ! is.finite(loop.nb.est)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 1\nloop.nb.est: ", loop.nb.est, "\ncor.ini: ", cor.ini, "\ntempo.cor: ", tempo.cor, "\ncor.limit: ", cor.limit, "\ncor.dec.per.loop: ", cor.dec.per.loop)
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else if(loop.nb.est > 1e4){ # below -> leave the while loop
+tempo.pos.secu <- tempo.pos
+count.secu <- count
+tempo.cor.secu <- tempo.cor
+cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "INITIAL SETTINGS BEFORE ROUND: ", round, " | LOOP COUNT: ", format(count, big.mark=","), " | GO BACK: ", GOBACK, " | LOOP NUMBER ESTIMATION: ", format(loop.nb.est, big.mark=","), " | CORRELATION LIMIT: ", fun_round(cor.limit, 4), " | ABS TEMPO CORRELATION: ", fun_round(tempo.cor, 4)))
+print.count.loop <- logical(length = print.count)
+print.count.loop[length(print.count.loop)] <- TRUE # not this to avoid long vector, but not forget to reset during printing: print.count.loop[(1:trunc(n / print.count) * print.count)] <- TRUE # counter to speedup
+count.loop <- 0
+pos <- sample.int(n = pos.selec.seq.max , size = print.count, replace = TRUE) # selection of random positions. BEWARE: n = pos.selec.seq.max because already - 1 (see above) but is connected to tempo.pos[c(pos2 + 1, pos2)] <- tempo.pos[c(pos2, pos2 + 1)]
+tempo.date.loop <- Sys.time()
+tempo.time.loop <- as.numeric(tempo.date.loop)
+for(i6 in 1:loop.nb.est){
+count.loop <- count.loop + 1
+pos2 <- pos[count.loop] # selection of 1 position
+tempo.pos[c(pos2 + 1, pos2)] <- tempo.pos[c(pos2, pos2 + 1)]
+if(print.count.loop[count.loop]){
+count.loop <- 0
+pos <- sample.int(n = pos.selec.seq.max , size = print.count, replace = TRUE) # BEWARE: never forget to resample here
+tempo.time <- as.numeric(Sys.time())
+tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - tempo.time.loop))
+final.loop <- (tempo.time - tempo.time.loop) / i6 * loop.nb.est # expected duration in seconds # intra nb.compar loop lapse: time lapse / cycles done * cycles remaining
+final.exp <- as.POSIXct(final.loop, origin = tempo.date.loop)
+cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "FOR LOOP | ROUND ", round, " | LOOP: ", format(i6, big.mark=","), " / ", format(loop.nb.est, big.mark=","), " | TIME SPENT: ", tempo.lapse, " | EXPECTED END: ", final.exp))
+}
+}
+count <- count + loop.nb.est # out of the loop to speedup
+tempo.cor <- abs(cor(x = data1[tempo.pos], y = data2, use = "pairwise.complete.obs", method = cor.method))
+if(tempo.cor > tempo.cor.secu | ((tempo.cor - cor.limit) < 0 & abs(tempo.cor - cor.limit) > smallest.cor.dec * round(log10(max(ini.pos, na.rm = TRUE))))){
+GOBACK <- TRUE
+tempo.pos <- tempo.pos.secu
+count <- count.secu
+tempo.cor <- tempo.cor.secu
+}else{
+GOBACK <- FALSE
+}
+}else{
+cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "FINAL WHILE LOOP | LOOP COUNT: ", format(count, big.mark=","), " | CORRELATION LIMIT: ", fun_round(cor.limit, 4), " | ABS TEMPO CORRELATION: ", fun_round(tempo.cor, 4)))
+print.count.loop <- logical(length = print.count)
+print.count.loop[length(print.count.loop)] <- TRUE # counter to speedup
+count.loop <- 0 #
+pos <- sample.int(n = pos.selec.seq.max , size = print.count, replace = TRUE) # selection of random positions. BEWARE: n = pos.selec.seq.max because already - 1 (see above) but is connected to tempo.pos[c(pos2 + 1, pos2)] <- tempo.pos[c(pos2, pos2 + 1)]
+tempo.cor.loop <- tempo.cor
+tempo.date.loop <- Sys.time()
+tempo.time.loop <- as.numeric(tempo.date.loop)
+while(tempo.cor > cor.limit){
+count <- count + 1
+count.loop <- count.loop + 1
+pos2 <- pos[count.loop]
+tempo.pos[c(pos2 + 1, pos2)] <- tempo.pos[c(pos2, pos2 + 1)]
+tempo.cor <- abs(cor(x = data1[tempo.pos], y = data2, use = "pairwise.complete.obs", method = cor.method))
+if(print.count.loop[count.loop]){
+count.loop <- 0
+pos <- sample.int(n = pos.selec.seq.max , size = print.count, replace = TRUE) # BEWARE: never forget to resample here
+tempo.time <- as.numeric(Sys.time())
+tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - tempo.time.loop))
+final.loop <- (tempo.time - tempo.time.loop) / (tempo.cor.loop - tempo.cor) * (tempo.cor - cor.limit) # expected duration in seconds # tempo.cor.loop - tempo.cor always positive and tempo.cor decreases progressively starting from tempo.cor.loop
+final.exp <- as.POSIXct(final.loop, origin = tempo.date.loop)
+cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "WHILE LOOP | LOOP NB: ", format(count.loop, big.mark=","), " | COUNT: ", format(count, big.mark=","), " | CORRELATION LIMIT: ", fun_round(cor.limit, 4), " | ABS TEMPO CORRELATION: ", fun_round(tempo.cor, 4), " | TIME SPENT: ", tempo.lapse, " | EXPECTED END: ", final.exp))
+}
+}
+}
+}
+tempo.time <- as.numeric(Sys.time())
+tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - ini.time))
+cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "WHILE/FOR LOOPS END | LOOP COUNT: ", format(count, big.mark=","), " | NB OF ROUNDS: ", round, " | CORRELATION LIMIT: ", fun_round(cor.limit, 4), " | ABS TEMPO CORRELATION: ", fun_round(tempo.cor, 4), " | TOTAL SPENT TIME: ", tempo.lapse))
+}
+tempo.cor <- ifelse(neg.cor == TRUE, -tempo.cor, tempo.cor)
+}
+}
+cat("\n\n")
+if(warn.print == TRUE & ! is.null(warn)){
+options(warning.length = 8170)
+on.exit(warning(paste0("FROM ", function.name, ":\n\n", warn), call. = FALSE), add = TRUE)
+on.exit(exp = options(warning.length = ini.warning.length), add = TRUE)
+}
+output <- list(data = data1[tempo.pos], warn = warn, cor = if(is.null(data2)){cor(ini.pos, tempo.pos, method = "spearman")}else{tempo.cor}, count = count)
+return(output)
}
-
+
######## fun_slide() #### return a computation made on a vector using a sliding window
fun_slide <- function(
- data,
- window.size,
- step,
- from = NULL,
- to = NULL,
- fun,
- args = NULL,
- boundary = "left",
- parall = FALSE,
- thread.nb = NULL,
- print.count = 100,
- res.path = NULL,
- lib.path = NULL,
- verbose = TRUE,
- cute.path = "C:\\Users\\Gael\\Documents\\Git_projects\\cute_little_R_functions\\cute_little_R_functions.R"
+data,
+window.size,
+step,
+from = NULL,
+to = NULL,
+fun,
+args = NULL,
+boundary = "left",
+parall = FALSE,
+thread.nb = NULL,
+print.count = 100,
+res.path = NULL,
+lib.path = NULL,
+verbose = TRUE,
+cute.path = "C:\\Users\\Gael\\Documents\\Git_projects\\cute_little_R_functions\\cute_little_R_functions.R"
+){
+# AIM
+# return a computation made on a vector using a sliding window
+# WARNINGS
+# The function uses two strategies, depending on the amout of memory required which depends on the data, window.size and step arguments. The first one uses lapply(), is generally fast but requires lots of memory. The second one uses a parallelized loop. The choice between the two strategies is automatic if parall argument is FALSE, and is forced toward parallelization if parall argument is TRUE
+# The parall argument forces the parallelization, which is convenient when the data argument is big, because the lapply() function is sometimes slower than the parallelization
+# ARGUMENTS
+# data: vector, matrix, table or array of numeric values (mode must be numeric). Inf not allowed. NA will be removed before computation
+# window.size: single numeric value indicating the width of the window sliding across data (in the same unit as data value)
+# step: single numeric value indicating the step between each window (in the same unit as data value). Cannot be larger than window.size
+# from: value of the left boundary of the first sliding window. If NULL, min(data) is used. The first window will strictly have from or min(data) as left boundary
+# to: value of the right boundary of the last sliding window. If NULL, max(data) is used. Warning: (1) the final last window will not necessary have to|max(data) as right boundary. In fact the last window will be the one that contains to|max(data) for the first time, i.e., min[from|min(data) + window.size + n * step >= to|max(data)]; (2) In fact, the >= in min[from|min(data) + window.size + n * step >= to|max(data)] depends on the boundary argument (>= for "right" and > for "left"); (3) to have the rule (1) but for the center of the last window, use to argument as to = to|max(data) + window.size / 2
+# fun: function or character string (without brackets) indicating the name of the function to apply in each window. Example: fun = "mean", or fun = mean
+# args: character string of additional arguments of fun (separated by a comma between the quotes). Example args = "na.rm = TRUE" for fun = mean. Ignored if NULL
+# boundary: either "left" or "right". Indicates if the sliding window includes values equal to left boundary and exclude values equal to right boundary ("left") or the opposite ("right")
+# parall: logical. Force parallelization ?
+# thread.nb: numeric value indicating the number of threads to use if ever parallelization is required. If NULL, all the available threads will be used
+# print.count: interger value. Print a working progress message every print.count during loops. BEWARE: can increase substentially the time to complete the process using a small value, like 10 for instance. Use Inf is no loop message desired
+# res.path: character string indicating the absolute pathway where the parallelization log file will be created if parallelization is used. If NULL, will be created in the R current directory
+# lib.path: character vector specifying the absolute pathways of the directories containing the required packages if not in the default directories. Ignored if NULL
+# verbose: logical. Display messages?
+# cute.path: character string indicating the absolute path of the cute.R file. Will be remove when cute will be a package. Not considered if thread.nb is NULL
+# RETURN
+# a data frame containing
+#$left : the left boundary of each window (in the unit of the data argument)
+#$right : the right boundary of each window (in the unit of data argument)
+#$center : the center of each window (in the unit of data argument)
+#$value : the computed value by the fun argument in each window)
+# REQUIRED PACKAGES
+# lubridate
+# parallel if parallelization is used (included in the R installation packages but not automatically loaded)
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# fun_get_message
+# fun_pack()
+# EXAMPLES
+# fun_slide(data = c(1:10, 100:110, 500), window.size = 5, step = 2, fun = length, boundary = "left")
+# fun_slide(data = c(1:10, 100:110, 500), window.size = 5, step = 2, fun = length, boundary = "right") # effect of boundary argument
+# fun_slide(data = c(1:10, 100:110, 500), window.size = 5, step = 2, fun = length, boundary = "left", parall = TRUE) # effect of parall argument
+# DEBUGGING
+# data = c(1:10, 100:110, 500) ; window.size = 5 ; step = 2 ; from = NULL ; to = NULL ; fun = length ; args = NULL ; boundary = "left" ; parall = FALSE ; thread.nb = NULL ; print.count = 100 ; res.path = NULL ; lib.path = NULL ; verbose = TRUE ; cute.path = "C:\\Users\\Gael\\Documents\\Git_projects\\cute_little_R_functions\\cute_little_R_functions.R"
+# data = lag.pos; window.size = window.size; step = step; fun = length; from = min(a$pos); to = max(a$pos)
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+instruction <- match.call()
+# end function name
+# required function checking
+req.function <- c(
+"fun_check",
+"fun_get_message",
+"fun_pack"
+)
+for(i1 in req.function){
+if(length(find(i1, mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED ", i1, "() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# end required function checking
+# argument primary checking
+# arg with no default values
+if(any(missing(data) | missing(window.size) | missing(step) | missing(fun))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ARGUMENTS fun, args AND val HAVE NO DEFAULT VALUE AND REQUIRE ONE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end arg with no default values
+# using fun_check()
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = data, mode = "numeric", na.contain = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = window.size, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = step, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(from)){
+tempo <- fun_check(data = from, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
+}
+if( ! is.null(to)){
+tempo <- fun_check(data = to, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
+}
+tempo1 <- fun_check(data = fun, class = "vector", mode = "character", length = 1, fun.name = function.name)
+tempo2 <- fun_check(data = fun, class = "function", length = 1, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": fun ARGUMENT MUST BE A FUNCTION OR A CHARACTER STRING OF THE NAME OF A FUNCTION")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+if( ! is.null(args)){
+tempo <- fun_check(data = args, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = boundary, options = c("left", "right"), length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(thread.nb)){
+tempo <- fun_check(data = thread.nb, typeof = "integer", double.as.integer.allowed = TRUE, neg.values = FALSE, length = 1, fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = parall, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = print.count, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+if( ! is.null(res.path)){
+tempo <- fun_check(data = res.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+}
+if( ! is.null(lib.path)){
+tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = cute.path, class = "vector", typeof = "character", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+if( ! file.exists(cute.path)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": FILE PATH INDICATED IN THE cute.path PARAMETER DOES NOT EXISTS:\n", cute.path)
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+tempo <- fun_check(data = verbose, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# end using fun_check()
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument primary checking
+# second round of checking and data preparation
+# dealing with NA
+if(any(is.na(window.size)) | any(is.na(step)) | any(is.na(from)) | any(is.na(to)) | suppressWarnings(any(is.na(fun))) | any(is.na(args)) | any(is.na(boundary)) | any(is.na(parall)) | any(is.na(thread.nb)) | any(is.na(print.count)) | any(is.na(res.path)) | any(is.na(lib.path)) | any(is.na(verbose))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": NO ARGUMENT EXCEPT data CAN HAVE NA VALUES\nPROBLEMATIC ARGUMENTS ARE: ", paste(c("window.size", "step", "from", "to", "fun", "args", "boundary", "parall", "thread.nb", "print.count", "res.path", "lib.path", "verbose")[c(any(is.na(window.size)), any(is.na(step)), any(is.na(from)), any(is.na(to)), suppressWarnings(any(is.na(fun))), any(is.na(args)), any(is.na(boundary)), any(is.na(parall)), any(is.na(thread.nb)), any(is.na(print.count)), any(is.na(res.path)), any(is.na(lib.path)), any(is.na(verbose)))], collapse = "\n"))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end dealing with NA
+# dealing with NULL
+if(is.null(data) | is.null(window.size) | is.null(step) | is.null(fun) | is.null(boundary) | is.null(parall) | is.null(print.count) | is.null(verbose)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": THESE ARGUMENTS\ndata\nwindow.size\nstep\nfun\nboundary\nparall\nprint.count\nverbose\nCANNOT BE NULL\nPROBLEMATIC ARGUMENTS ARE: ", paste(c("data", "window.size", "step", "fun", "boundary", "parall", "print.count", "verbose")[c(is.null(data), is.null(window.size), is.null(step), is.null(fun), is.null(boundary), is.null(parall), is.null(print.count), is.null(verbose))], collapse = "\n"))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end dealing with NULL
+if(any( ! is.finite(data))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": data ARGUMENT CANNOT CONTAIN Inf VALUES")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
+}
+if(step > window.size){
+tempo.cat <- paste0("ERROR IN ", function.name, ": step ARGUMENT MUST BE LOWER THAN window.size ARGUMENT\nstep: ", paste(step, collapse = " "), "\nwindow.size: ", paste(window.size, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
+}
+if( ! is.null(thread.nb)){
+if(thread.nb < 1){
+tempo.cat <- paste0("ERROR IN ", function.name, ": thread.nb PARAMETER MUST EQUAL OR GREATER THAN 1: ", thread.nb)
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
+}
+}
+if( ! is.null(res.path)){
+if( ! all(dir.exists(res.path))){ # separation to avoid the problem of tempo$problem == FALSE and res.path == NA
+tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE res.path ARGUMENT DOES NOT EXISTS:\n", paste(res.path, collapse = "\n"))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
+}
+}else{
+res.path <- getwd() # working directory
+}
+if( ! is.null(lib.path)){
+if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
+tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
+}
+}
+# end second round of checking and data preparation
+# package checking
+fun_pack(req.package = c("lubridate"), lib.path = lib.path)
+fun_pack(req.package = c("parallel"), lib.path = lib.path)
+# end package checking
+# main code
+if(verbose == TRUE){
+cat("\nfun_slide JOB IGNITION\n")
+}
+ini.date <- Sys.time()
+ini.time <- as.numeric(ini.date) # time of process begin, converted into seconds
+fun <- match.fun(fun) # make fun <- get(fun) is fun is a function name written as character string of length 1
+if(boundary == "left"){
+left <- ">="
+right <- "<"
+right.last.wind <- ">"
+}else if(boundary == "right"){
+left <- ">"
+right <- "<="
+right.last.wind <- ">="
+}else{
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 1")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+data <- as.vector(data)
+data <- sort(data, na.last = NA) # NA removed
+wind <- data.frame(left = seq(from = if(is.null(from)){min(data, na.rm = TRUE)}else{from}, to = if(is.null(to)){max(data, na.rm = TRUE)}else{to}, by = step), stringsAsFactors = TRUE)
+wind <- data.frame(wind, right = wind$left + window.size, stringsAsFactors = TRUE)
+wind <- data.frame(wind, center = (wind$left + wind$right) / 2, stringsAsFactors = TRUE)
+if(all(wind$right < if(is.null(to)){max(data, na.rm = TRUE)}else{to})){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 2")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# The 3 next lines is for the rule of to argument with center (see to argument description)
+# if(any(wind$center > max(data, na.rm = TRUE))){
+# wind <- wind[ ! wind$center > max(data, na.rm = TRUE),]
+# }
+if(sum(get(right.last.wind)(wind$right, if(is.null(to)){max(data, na.rm = TRUE)}else{to}), na.rm = TRUE) > 1){ # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
+tempo.log <- get(right.last.wind)(wind$right, if(is.null(to)){max(data, na.rm = TRUE)}else{to}) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
+tempo.log[min(which(tempo.log), na.rm = TRUE)] <- FALSE # convert the first left boundary that goes above max(data, na.rm = TRUE) to FALSE to keep it (the next ones will be removed)
+wind <- wind[ ! tempo.log,]
+}
+
+# test if lapply can be used
+if(parall == FALSE){
+# new environment
+env.name <- paste0("env", ini.time)
+if(exists(env.name, where = -1)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ENVIRONMENT env.name ALREADY EXISTS. PLEASE RERUN ONCE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+assign(env.name, new.env())
+assign("wind", wind, envir = get(env.name, env = sys.nframe(), inherit = FALSE))
+assign("data", data, envir = get(env.name, env = sys.nframe(), inherit = FALSE))
+}
+# end new environment
+tempo.message <- fun_get_message(data="lapply(X = wind$left, Y = data, FUN = function(X, Y){res <- get(left)(Y, X) ; return(res)})", kind = "error", header = FALSE, env = get(env.name, env = sys.nframe(), inherit = FALSE), print.no = FALSE) # no env = sys.nframe(), inherit = FALSE in get(left) because look for function in the classical scope
+rm(env.name) # optional, because should disappear at the end of the function execution
+}else{
+tempo.message <- "ERROR" # with this, force the parallelization by default
+}
+# end test if lapply can be used
+if( ! any(grepl(x = tempo.message, pattern = "ERROR.*"))){
+left.log <- lapply(X = wind$left, Y = data, FUN = function(X, Y){
+res <- get(left)(Y, X) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
+return(res)
+})
+right.log <- lapply(X = wind$right, Y = data, FUN = function(X, Y){
+res <- get(right)(Y, X) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
+return(res)
+})
+log <- mapply(FUN = "&", left.log, right.log, SIMPLIFY = FALSE)
+output <- eval(parse(text = paste0("sapply(lapply(log, FUN = function(X){(data[X])}), FUN = fun", if( ! is.null(args)){paste0(", ", args)}, ")"))) # take the values of the data vector according to log (list of logical, each compartment of length(data)) and apply fun with args of fun
+if(length(output) != nrow(wind)){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 3")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+output <- data.frame(wind, value = output, stringsAsFactors = TRUE)
+}
+}else{
+if(verbose == TRUE){
+tempo.cat <- paste0("PARALLELIZATION INITIATED AT: ", ini.date)
+cat(paste0("\n", tempo.cat, "\n"))
+}
+tempo.thread.nb = parallel::detectCores(all.tests = FALSE, logical = TRUE) # detect the number of threads
+if( ! is.null(thread.nb)){
+if(tempo.thread.nb < thread.nb){
+thread.nb <- tempo.thread.nb
+if(verbose == TRUE){
+tempo.cat <- paste0("ONLY: ", tempo.thread.nb, " THREADS AVAILABLE")
+cat(paste0("\n", tempo.cat, "\n"))
+}
+}
+}else{
+thread.nb <- tempo.thread.nb
+}
+if(verbose == TRUE){
+tempo.cat <- paste0("NUMBER OF THREADS USED: ", thread.nb)
+cat(paste0("\n ", tempo.cat, "\n"))
+}
+Clust <- parallel::makeCluster(thread.nb, outfile = paste0(res.path, "/fun_slide_parall_log.txt")) # outfile to print or cat during parallelization (only possible in a file, outfile = "" do not work on windows)
+cluster.list <- parallel::clusterSplit(Clust, 1:nrow(wind)) # split according to the number of cluster
+if(verbose == TRUE){
+tempo.cat <- paste0("SPLIT OF TEST NUMBERS IN PARALLELISATION:")
+cat(paste0("\n ", tempo.cat, "\n"))
+str(cluster.list) # using print(str()) add a NULL below the result
+cat("\n")
+}
+paral.output.list <- parallel::clusterApply( #
+cl = Clust,
+x = cluster.list,
+function.name = function.name,
+data = data,
+FUN = fun, # because fun argument of clusterApply
+args = args,
+thread.nb = thread.nb,
+print.count = print.count,
+wind = wind,
+left = left,
+right = right,
+res.path = res.path,
+lib.path = lib.path,
+verbose = verbose,
+cute.path = cute.path,
+fun = function(
+x,
+function.name,
+data,
+FUN,
+args,
+thread.nb,
+print.count,
+wind,
+left,
+right,
+res.path,
+lib.path,
+verbose,
+cute.path
){
- # AIM
- # return a computation made on a vector using a sliding window
- # WARNINGS
- # The function uses two strategies, depending on the amout of memory required which depends on the data, window.size and step arguments. The first one uses lapply(), is generally fast but requires lots of memory. The second one uses a parallelized loop. The choice between the two strategies is automatic if parall argument is FALSE, and is forced toward parallelization if parall argument is TRUE
- # The parall argument forces the parallelization, which is convenient when the data argument is big, because the lapply() function is sometimes slower than the parallelization
- # ARGUMENTS
- # data: vector, matrix, table or array of numeric values (mode must be numeric). Inf not allowed. NA will be removed before computation
- # window.size: single numeric value indicating the width of the window sliding across data (in the same unit as data value)
- # step: single numeric value indicating the step between each window (in the same unit as data value). Cannot be larger than window.size
- # from: value of the left boundary of the first sliding window. If NULL, min(data) is used. The first window will strictly have from or min(data) as left boundary
- # to: value of the right boundary of the last sliding window. If NULL, max(data) is used. Warning: (1) the final last window will not necessary have to|max(data) as right boundary. In fact the last window will be the one that contains to|max(data) for the first time, i.e., min[from|min(data) + window.size + n * step >= to|max(data)]; (2) In fact, the >= in min[from|min(data) + window.size + n * step >= to|max(data)] depends on the boundary argument (>= for "right" and > for "left"); (3) to have the rule (1) but for the center of the last window, use to argument as to = to|max(data) + window.size / 2
- # fun: function or character string (without brackets) indicating the name of the function to apply in each window. Example: fun = "mean", or fun = mean
- # args: character string of additional arguments of fun (separated by a comma between the quotes). Example args = "na.rm = TRUE" for fun = mean. Ignored if NULL
- # boundary: either "left" or "right". Indicates if the sliding window includes values equal to left boundary and exclude values equal to right boundary ("left") or the opposite ("right")
- # parall: logical. Force parallelization ?
- # thread.nb: numeric value indicating the number of threads to use if ever parallelization is required. If NULL, all the available threads will be used
- # print.count: interger value. Print a working progress message every print.count during loops. BEWARE: can increase substentially the time to complete the process using a small value, like 10 for instance. Use Inf is no loop message desired
- # res.path: character string indicating the absolute pathway where the parallelization log file will be created if parallelization is used. If NULL, will be created in the R current directory
- # lib.path: character vector specifying the absolute pathways of the directories containing the required packages if not in the default directories. Ignored if NULL
- # verbose: logical. Display messages?
- # cute.path: character string indicating the absolute path of the cute.R file. Will be remove when cute will be a package. Not considered if thread.nb is NULL
- # RETURN
- # a data frame containing
- #$left : the left boundary of each window (in the unit of the data argument)
- #$right : the right boundary of each window (in the unit of data argument)
- #$center : the center of each window (in the unit of data argument)
- #$value : the computed value by the fun argument in each window)
- # REQUIRED PACKAGES
- # lubridate
- # parallel if parallelization is used (included in the R installation packages but not automatically loaded)
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # fun_get_message
- # fun_pack()
- # EXAMPLES
- # fun_slide(data = c(1:10, 100:110, 500), window.size = 5, step = 2, fun = length, boundary = "left")
- # fun_slide(data = c(1:10, 100:110, 500), window.size = 5, step = 2, fun = length, boundary = "right") # effect of boundary argument
- # fun_slide(data = c(1:10, 100:110, 500), window.size = 5, step = 2, fun = length, boundary = "left", parall = TRUE) # effect of parall argument
- # DEBUGGING
- # data = c(1:10, 100:110, 500) ; window.size = 5 ; step = 2 ; from = NULL ; to = NULL ; fun = length ; args = NULL ; boundary = "left" ; parall = FALSE ; thread.nb = NULL ; print.count = 100 ; res.path = NULL ; lib.path = NULL ; verbose = TRUE ; cute.path = "C:\\Users\\Gael\\Documents\\Git_projects\\cute_little_R_functions\\cute_little_R_functions.R"
- # data = lag.pos; window.size = window.size; step = step; fun = length; from = min(a$pos); to = max(a$pos)
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- instruction <- match.call()
- # end function name
- # required function checking
- req.function <- c(
- "fun_check",
- "fun_get_message",
- "fun_pack"
- )
- for(i1 in req.function){
- if(length(find(i1, mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED ", i1, "() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # end required function checking
- # argument primary checking
- # arg with no default values
- if(any(missing(data) | missing(window.size) | missing(step) | missing(fun))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ARGUMENTS fun, args AND val HAVE NO DEFAULT VALUE AND REQUIRE ONE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end arg with no default values
- # using fun_check()
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = data, mode = "numeric", na.contain = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = window.size, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = step, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(from)){
- tempo <- fun_check(data = from, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
- }
- if( ! is.null(to)){
- tempo <- fun_check(data = to, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
- }
- tempo1 <- fun_check(data = fun, class = "vector", mode = "character", length = 1, fun.name = function.name)
- tempo2 <- fun_check(data = fun, class = "function", length = 1, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": fun ARGUMENT MUST BE A FUNCTION OR A CHARACTER STRING OF THE NAME OF A FUNCTION")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- if( ! is.null(args)){
- tempo <- fun_check(data = args, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = boundary, options = c("left", "right"), length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(thread.nb)){
- tempo <- fun_check(data = thread.nb, typeof = "integer", double.as.integer.allowed = TRUE, neg.values = FALSE, length = 1, fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = parall, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = print.count, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- if( ! is.null(res.path)){
- tempo <- fun_check(data = res.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- }
- if( ! is.null(lib.path)){
- tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = cute.path, class = "vector", typeof = "character", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- if( ! file.exists(cute.path)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": FILE PATH INDICATED IN THE cute.path PARAMETER DOES NOT EXISTS:\n", cute.path)
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- tempo <- fun_check(data = verbose, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # end using fun_check()
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument primary checking
- # second round of checking and data preparation
- # dealing with NA
- if(any(is.na(window.size)) | any(is.na(step)) | any(is.na(from)) | any(is.na(to)) | suppressWarnings(any(is.na(fun))) | any(is.na(args)) | any(is.na(boundary)) | any(is.na(parall)) | any(is.na(thread.nb)) | any(is.na(print.count)) | any(is.na(res.path)) | any(is.na(lib.path)) | any(is.na(verbose))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": NO ARGUMENT EXCEPT data CAN HAVE NA VALUES\nPROBLEMATIC ARGUMENTS ARE: ", paste(c("window.size", "step", "from", "to", "fun", "args", "boundary", "parall", "thread.nb", "print.count", "res.path", "lib.path", "verbose")[c(any(is.na(window.size)), any(is.na(step)), any(is.na(from)), any(is.na(to)), suppressWarnings(any(is.na(fun))), any(is.na(args)), any(is.na(boundary)), any(is.na(parall)), any(is.na(thread.nb)), any(is.na(print.count)), any(is.na(res.path)), any(is.na(lib.path)), any(is.na(verbose)))], collapse = "\n"))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end dealing with NA
- # dealing with NULL
- if(is.null(data) | is.null(window.size) | is.null(step) | is.null(fun) | is.null(boundary) | is.null(parall) | is.null(print.count) | is.null(verbose)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": THESE ARGUMENTS\ndata\nwindow.size\nstep\nfun\nboundary\nparall\nprint.count\nverbose\nCANNOT BE NULL\nPROBLEMATIC ARGUMENTS ARE: ", paste(c("data", "window.size", "step", "fun", "boundary", "parall", "print.count", "verbose")[c(is.null(data), is.null(window.size), is.null(step), is.null(fun), is.null(boundary), is.null(parall), is.null(print.count), is.null(verbose))], collapse = "\n"))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end dealing with NULL
- if(any( ! is.finite(data))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": data ARGUMENT CANNOT CONTAIN Inf VALUES")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
- }
- if(step > window.size){
- tempo.cat <- paste0("ERROR IN ", function.name, ": step ARGUMENT MUST BE LOWER THAN window.size ARGUMENT\nstep: ", paste(step, collapse = " "), "\nwindow.size: ", paste(window.size, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
- }
- if( ! is.null(thread.nb)){
- if(thread.nb < 1){
- tempo.cat <- paste0("ERROR IN ", function.name, ": thread.nb PARAMETER MUST EQUAL OR GREATER THAN 1: ", thread.nb)
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
- }
- }
- if( ! is.null(res.path)){
- if( ! all(dir.exists(res.path))){ # separation to avoid the problem of tempo$problem == FALSE and res.path == NA
- tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE res.path ARGUMENT DOES NOT EXISTS:\n", paste(res.path, collapse = "\n"))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
- }
- }else{
- res.path <- getwd() # working directory
- }
- if( ! is.null(lib.path)){
- if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
- tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
- }
- }
- # end second round of checking and data preparation
- # package checking
- fun_pack(req.package = c("lubridate"), lib.path = lib.path)
- fun_pack(req.package = c("parallel"), lib.path = lib.path)
- # end package checking
- # main code
- if(verbose == TRUE){
- cat("\nfun_slide JOB IGNITION\n")
- }
- ini.date <- Sys.time()
- ini.time <- as.numeric(ini.date) # time of process begin, converted into seconds
- fun <- match.fun(fun) # make fun <- get(fun) is fun is a function name written as character string of length 1
- if(boundary == "left"){
- left <- ">="
- right <- "<"
- right.last.wind <- ">"
- }else if(boundary == "right"){
- left <- ">"
- right <- "<="
- right.last.wind <- ">="
- }else{
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 1")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- data <- as.vector(data)
- data <- sort(data, na.last = NA) # NA removed
- wind <- data.frame(left = seq(from = if(is.null(from)){min(data, na.rm = TRUE)}else{from}, to = if(is.null(to)){max(data, na.rm = TRUE)}else{to}, by = step), stringsAsFactors = TRUE)
- wind <- data.frame(wind, right = wind$left + window.size, stringsAsFactors = TRUE)
- wind <- data.frame(wind, center = (wind$left + wind$right) / 2, stringsAsFactors = TRUE)
- if(all(wind$right < if(is.null(to)){max(data, na.rm = TRUE)}else{to})){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 2")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # The 3 next lines is for the rule of to argument with center (see to argument description)
- # if(any(wind$center > max(data, na.rm = TRUE))){
- # wind <- wind[ ! wind$center > max(data, na.rm = TRUE),]
- # }
- if(sum(get(right.last.wind)(wind$right, if(is.null(to)){max(data, na.rm = TRUE)}else{to}), na.rm = TRUE) > 1){ # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
- tempo.log <- get(right.last.wind)(wind$right, if(is.null(to)){max(data, na.rm = TRUE)}else{to}) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
- tempo.log[min(which(tempo.log), na.rm = TRUE)] <- FALSE # convert the first left boundary that goes above max(data, na.rm = TRUE) to FALSE to keep it (the next ones will be removed)
- wind <- wind[ ! tempo.log,]
- }
-
- # test if lapply can be used
- if(parall == FALSE){
- # new environment
- env.name <- paste0("env", ini.time)
- if(exists(env.name, where = -1)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ENVIRONMENT env.name ALREADY EXISTS. PLEASE RERUN ONCE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- assign(env.name, new.env())
- assign("wind", wind, envir = get(env.name, env = sys.nframe(), inherit = FALSE))
- assign("data", data, envir = get(env.name, env = sys.nframe(), inherit = FALSE))
- }
- # end new environment
- tempo.message <- fun_get_message(data="lapply(X = wind$left, Y = data, FUN = function(X, Y){res <- get(left)(Y, X) ; return(res)})", kind = "error", header = FALSE, env = get(env.name, env = sys.nframe(), inherit = FALSE), print.no = FALSE) # no env = sys.nframe(), inherit = FALSE in get(left) because look for function in the classical scope
- rm(env.name) # optional, because should disappear at the end of the function execution
- }else{
- tempo.message <- "ERROR" # with this, force the parallelization by default
- }
- # end test if lapply can be used
- if( ! any(grepl(x = tempo.message, pattern = "ERROR.*"))){
- left.log <- lapply(X = wind$left, Y = data, FUN = function(X, Y){
- res <- get(left)(Y, X) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
- return(res)
- })
- right.log <- lapply(X = wind$right, Y = data, FUN = function(X, Y){
- res <- get(right)(Y, X) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
- return(res)
- })
- log <- mapply(FUN = "&", left.log, right.log, SIMPLIFY = FALSE)
- output <- eval(parse(text = paste0("sapply(lapply(log, FUN = function(X){(data[X])}), FUN = fun", if( ! is.null(args)){paste0(", ", args)}, ")"))) # take the values of the data vector according to log (list of logical, each compartment of length(data)) and apply fun with args of fun
- if(length(output) != nrow(wind)){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 3")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- output <- data.frame(wind, value = output, stringsAsFactors = TRUE)
- }
- }else{
- if(verbose == TRUE){
- tempo.cat <- paste0("PARALLELIZATION INITIATED AT: ", ini.date)
- cat(paste0("\n", tempo.cat, "\n"))
- }
- tempo.thread.nb = parallel::detectCores(all.tests = FALSE, logical = TRUE) # detect the number of threads
- if( ! is.null(thread.nb)){
- if(tempo.thread.nb < thread.nb){
- thread.nb <- tempo.thread.nb
- if(verbose == TRUE){
- tempo.cat <- paste0("ONLY: ", tempo.thread.nb, " THREADS AVAILABLE")
- cat(paste0("\n", tempo.cat, "\n"))
- }
- }
- }else{
- thread.nb <- tempo.thread.nb
- }
- if(verbose == TRUE){
- tempo.cat <- paste0("NUMBER OF THREADS USED: ", thread.nb)
- cat(paste0("\n ", tempo.cat, "\n"))
- }
- Clust <- parallel::makeCluster(thread.nb, outfile = paste0(res.path, "/fun_slide_parall_log.txt")) # outfile to print or cat during parallelization (only possible in a file, outfile = "" do not work on windows)
- cluster.list <- parallel::clusterSplit(Clust, 1:nrow(wind)) # split according to the number of cluster
- if(verbose == TRUE){
- tempo.cat <- paste0("SPLIT OF TEST NUMBERS IN PARALLELISATION:")
- cat(paste0("\n ", tempo.cat, "\n"))
- str(cluster.list) # using print(str()) add a NULL below the result
- cat("\n")
- }
- paral.output.list <- parallel::clusterApply( #
- cl = Clust,
- x = cluster.list,
- function.name = function.name,
- data = data,
- FUN = fun, # because fun argument of clusterApply
- args = args,
- thread.nb = thread.nb,
- print.count = print.count,
- wind = wind,
- left = left,
- right = right,
- res.path = res.path,
- lib.path = lib.path,
- verbose = verbose,
- cute.path = cute.path,
- fun = function(
- x,
- function.name,
- data,
- FUN,
- args,
- thread.nb,
- print.count,
- wind,
- left,
- right,
- res.path,
- lib.path,
- verbose,
- cute.path
- ){
- # check again: very important because another R
- process.id <- Sys.getpid()
- cat(paste0("\nPROCESS ID ", process.id, " -> TESTS ", x[1], " TO ", x[length(x)], "\n"))
- source(cute.path, local = .GlobalEnv)
- fun_pack(req.package = "lubridate", lib.path = lib.path, load = TRUE) # load = TRUE to be sure that functions are present in the environment. And this prevent to use R.lib.path argument of fun_python_pack()
- # end check again: very important because another R
- ini.date <- Sys.time()
- ini.time <- as.numeric(ini.date) # time of process begin, converted into
- output <- NULL
- print.count.loop <- 0
- for(i4 in 1:length(x)){
- print.count.loop <- print.count.loop + 1
- log <- get(left)(data, wind$left[x[i4]]) & get(right)(data, wind$right[x[i4]]) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
- output <- c(output, eval(parse(text = paste0("FUN(data[log]", if( ! is.null(args)){paste0(", ", args)}, ")"))))
- if(verbose == TRUE){
- if(print.count.loop == print.count){
- print.count.loop <- 0
- tempo.time <- as.numeric(Sys.time())
- tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - ini.time))
- final.loop <- (tempo.time - ini.time) / i4 * length(x) # expected duration in seconds # intra nb.compar loop lapse: time lapse / cycles done * cycles remaining
- final.exp <- as.POSIXct(final.loop, origin = ini.date)
- cat(paste0("\nIN PROCESS ", process.id, " | LOOP ", format(i4, big.mark=","), " / ", format(length(x), big.mark=","), " | TIME SPENT: ", tempo.lapse, " | EXPECTED END: ", final.exp))
- }
- if(i4 == length(x)){
- tempo.time <- as.numeric(Sys.time())
- tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - ini.time))
- cat(paste0("\nPROCESS ", process.id, " ENDED | LOOP ", format(i4, big.mark=","), " / ", format(length(x), big.mark=","), " | TIME SPENT: ", tempo.lapse, "\n\n"))
- }
- }
- }
- wind <- wind[x, ]
- if(length(output) != nrow(wind)){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 4")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- output <- data.frame(wind, value = output, stringsAsFactors = TRUE)
- return(output)
- }
- }
- )
- parallel::stopCluster(Clust)
- # result assembly
- output <- data.frame()
- for(i2 in 1:length(paral.output.list)){ # compartment relatives to each parallelization
- output <- rbind(output, paral.output.list[[i2]], stringsAsFactors = TRUE)
- }
- # end result assembly
- if(nrow(output) != nrow(wind)){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 5\nlength(output): ", length(output), "\nnrow(wind): ", nrow(wind))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- output <- output[order(output$left), ]
- }
- }
- if(verbose == TRUE){
- end.date <- Sys.time()
- end.time <- as.numeric(end.date)
- total.lapse <- round(lubridate::seconds_to_period(end.time - ini.time))
- cat(paste0("fun_slide JOB END\n\nTIME: ", end.date, "\n\nTOTAL TIME LAPSE: ", total.lapse, "\n\n\n"))
- }
- return(output)
+# check again: very important because another R
+process.id <- Sys.getpid()
+cat(paste0("\nPROCESS ID ", process.id, " -> TESTS ", x[1], " TO ", x[length(x)], "\n"))
+source(cute.path, local = .GlobalEnv)
+fun_pack(req.package = "lubridate", lib.path = lib.path, load = TRUE) # load = TRUE to be sure that functions are present in the environment. And this prevent to use R.lib.path argument of fun_python_pack()
+# end check again: very important because another R
+ini.date <- Sys.time()
+ini.time <- as.numeric(ini.date) # time of process begin, converted into
+output <- NULL
+print.count.loop <- 0
+for(i4 in 1:length(x)){
+print.count.loop <- print.count.loop + 1
+log <- get(left)(data, wind$left[x[i4]]) & get(right)(data, wind$right[x[i4]]) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
+output <- c(output, eval(parse(text = paste0("FUN(data[log]", if( ! is.null(args)){paste0(", ", args)}, ")"))))
+if(verbose == TRUE){
+if(print.count.loop == print.count){
+print.count.loop <- 0
+tempo.time <- as.numeric(Sys.time())
+tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - ini.time))
+final.loop <- (tempo.time - ini.time) / i4 * length(x) # expected duration in seconds # intra nb.compar loop lapse: time lapse / cycles done * cycles remaining
+final.exp <- as.POSIXct(final.loop, origin = ini.date)
+cat(paste0("\nIN PROCESS ", process.id, " | LOOP ", format(i4, big.mark=","), " / ", format(length(x), big.mark=","), " | TIME SPENT: ", tempo.lapse, " | EXPECTED END: ", final.exp))
+}
+if(i4 == length(x)){
+tempo.time <- as.numeric(Sys.time())
+tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - ini.time))
+cat(paste0("\nPROCESS ", process.id, " ENDED | LOOP ", format(i4, big.mark=","), " / ", format(length(x), big.mark=","), " | TIME SPENT: ", tempo.lapse, "\n\n"))
+}
+}
+}
+wind <- wind[x, ]
+if(length(output) != nrow(wind)){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 4")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+output <- data.frame(wind, value = output, stringsAsFactors = TRUE)
+return(output)
+}
+}
+)
+parallel::stopCluster(Clust)
+# result assembly
+output <- data.frame()
+for(i2 in 1:length(paral.output.list)){ # compartment relatives to each parallelization
+output <- rbind(output, paral.output.list[[i2]], stringsAsFactors = TRUE)
+}
+# end result assembly
+if(nrow(output) != nrow(wind)){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 5\nlength(output): ", length(output), "\nnrow(wind): ", nrow(wind))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+output <- output[order(output$left), ]
+}
+}
+if(verbose == TRUE){
+end.date <- Sys.time()
+end.time <- as.numeric(end.date)
+total.lapse <- round(lubridate::seconds_to_period(end.time - ini.time))
+cat(paste0("fun_slide JOB END\n\nTIME: ", end.date, "\n\nTOTAL TIME LAPSE: ", total.lapse, "\n\n\n"))
+}
+return(output)
}
@@ -3758,70 +3757,70 @@ fun_slide <- function(
# Check OK: clear to go Apollo
fun_width <- function(
- class.nb,
- inches.per.class.nb = 1,
- ini.window.width = 7,
- inch.left.space,
- inch.right.space,
- boundarie.space = 0.5
+class.nb,
+inches.per.class.nb = 1,
+ini.window.width = 7,
+inch.left.space,
+inch.right.space,
+boundarie.space = 0.5
){
- # AIM
- # rescale the width of a window to open depending on the number of classes to plot
- # can be used for height, considering that it is as if it was a width
- # this order can be used:
- # fun_width()
- # fun_open()
- # fun_prior_plot() # not for ggplot2
- # plot() or any other plotting
- # fun_post_plot() if fun_prior_plot() has been used # not for ggplot2
- # fun_close()
- # ARGUMENTS
- # class.nb: number of class to plot
- # inches.per.class.nb: number of inches per unit of class.nb. 2 means 2 inches for each boxplot for instance
- # ini.window.width:initial window width in inches
- # inch.left.space: left horizontal margin of the figure region (in inches)
- # inch.right.space: right horizontal margin of the figure region (in inches)
- # boundarie.space: space between the right and left limits of the plotting region and the plot (0.5 means half a class width)
- # RETURN
- # the new window width in inches
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # fun_width(class.nb = 10, inches.per.class.nb = 0.2, ini.window.width = 7, inch.left.space = 1, inch.right.space = 1, boundarie.space = 0.5)
- # DEBUGGING
- # class.nb = 10 ; inches.per.class.nb = 0.2 ; ini.window.width = 7 ; inch.left.space = 1 ; inch.right.space = 1 ; boundarie.space = 0.5 # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = class.nb, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = inches.per.class.nb, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = ini.window.width, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = inch.left.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = inch.right.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = boundarie.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- range.max <- class.nb + boundarie.space # the max range of the future plot
- range.min <- boundarie.space # the min range of the future plot
- window.width <- inch.left.space + inch.right.space + inches.per.class.nb * (range.max - range.min)
- return(window.width)
+# AIM
+# rescale the width of a window to open depending on the number of classes to plot
+# can be used for height, considering that it is as if it was a width
+# this order can be used:
+# fun_width()
+# fun_open()
+# fun_prior_plot() # not for ggplot2
+# plot() or any other plotting
+# fun_post_plot() if fun_prior_plot() has been used # not for ggplot2
+# fun_close()
+# ARGUMENTS
+# class.nb: number of class to plot
+# inches.per.class.nb: number of inches per unit of class.nb. 2 means 2 inches for each boxplot for instance
+# ini.window.width:initial window width in inches
+# inch.left.space: left horizontal margin of the figure region (in inches)
+# inch.right.space: right horizontal margin of the figure region (in inches)
+# boundarie.space: space between the right and left limits of the plotting region and the plot (0.5 means half a class width)
+# RETURN
+# the new window width in inches
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# fun_width(class.nb = 10, inches.per.class.nb = 0.2, ini.window.width = 7, inch.left.space = 1, inch.right.space = 1, boundarie.space = 0.5)
+# DEBUGGING
+# class.nb = 10 ; inches.per.class.nb = 0.2 ; ini.window.width = 7 ; inch.left.space = 1 ; inch.right.space = 1 ; boundarie.space = 0.5 # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = class.nb, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = inches.per.class.nb, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = ini.window.width, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = inch.left.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = inch.right.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = boundarie.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+range.max <- class.nb + boundarie.space # the max range of the future plot
+range.min <- boundarie.space # the min range of the future plot
+window.width <- inch.left.space + inch.right.space + inches.per.class.nb * (range.max - range.min)
+return(window.width)
}
@@ -3830,168 +3829,168 @@ fun_width <- function(
# Check OK: clear to go Apollo
fun_open <- function(
- pdf = TRUE,
- pdf.path = "working.dir",
- pdf.name = "graph",
- width = 7,
- height = 7,
- paper = "special",
- pdf.overwrite = FALSE,
- rescale = "fixed",
- remove.read.only = TRUE,
- return.output = FALSE
+pdf = TRUE,
+pdf.path = "working.dir",
+pdf.name = "graph",
+width = 7,
+height = 7,
+paper = "special",
+pdf.overwrite = FALSE,
+rescale = "fixed",
+remove.read.only = TRUE,
+return.output = FALSE
){
- # AIM
- # open a pdf or screen (GUI) graphic window and return initial graphic parameters
- # this order can be used:
- # fun_width()
- # fun_open()
- # fun_prior_plot() # not for ggplot2
- # plot() or any other plotting
- # fun_post_plot() if fun_prior_plot() has been used # not for ggplot2
- # fun_close()
- # WARNINGS
- # On Linux, use pdf = TRUE, if (GUI) graphic window is not always available, meaning that X is not installed (clusters for instance). Use X11() in R to test if available
- # ARGUMENTS:
- # pdf: logical. Use pdf display? If FALSE, a GUI is opened
- # pdf.path: where the pdf is saved (do not terminate by / or \\). Write "working.dir" if working directory is required (default). Ignored if pdf == FALSE
- # pdf.name: name of the pdf file containing the graphs (the .pdf extension is added by the function, if not detected in the name end). Ignored if pdf == FALSE
- # width: width of the window (in inches)
- # height: height of the window (in inches)
- # paper: paper argument of the pdf function (paper format). Only used for pdf(). Either "a4", "letter", "legal", "us", "executive", "a4r", "USr" or "special". If "special", means that the paper dimension will be width and height. With another paper format, if width or height is over the size of the paper, width or height will be modified such that the plot is adjusted to the paper dimension (see $dim in the returned list below to see the modified dimensions). Ignored if pdf == FALSE
- # pdf.overwrite: logical. Existing pdf can be overwritten? . Ignored if pdf == FALSE
- # rescale: kind of GUI. Either "R", "fit", or "fixed". Ignored on Mac and Linux OS. See ?windows for details
- # remove.read.only: logical. remove the read only (R.O.) graphical parameters? If TRUE, the graphical parameters are returned without the R.O. parameters. The returned $ini.par list can be used to set the par() of a new graphical device. If FALSE, graphical parameters are returned with the R.O. parameters, which provides information like text dimension (see ?par() ). The returned $ini.par list can be used to set the par() of a new graphical device, but generate a warning message. Ignored if return.output == FALSE.
- # return.output: logical. Return output ? If TRUE the output list is displayed
- # RETURN
- # a list containing:
- # $pdf.loc: path of the pdf created
- # $ini.par: initial par() parameters
- # $zone.ini: initial window spliting
- # $dim: dimension of the graphical device (in inches)
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # fun_open(pdf = FALSE, pdf.path = "C:/Users/Gael/Desktop", pdf.name = "graph", width = 7, height = 7, paper = "special", pdf.overwrite = FALSE, return.output = TRUE)
- # DEBUGGING
- # pdf = TRUE ; pdf.path = "C:/Users/Gael/Desktop" ; pdf.name = "graphs" ; width = 7 ; height = 7 ; paper = "special" ; pdf.overwrite = FALSE ; rescale = "fixed" ; remove.read.only = TRUE ; return.output = TRUE # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = pdf, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = pdf.path, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = pdf.name, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = width, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = height, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = paper, options = c("a4", "letter", "legal", "us", "executive", "a4r", "USr", "special", "A4", "LETTER", "LEGAL", "US"), length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data =pdf.overwrite, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = rescale, options = c("R", "fit", "fixed"), length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = remove.read.only, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = return.output, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- if(pdf.path == "working.dir"){
- pdf.path <- getwd()
- }else{
- if(grepl(x = pdf.path, pattern = ".+/$")){
- pdf.path <- sub(x = pdf.path, pattern = "/$", replacement = "") # remove the last /
- }else if(grepl(x = pdf.path, pattern = ".+[\\]$")){ # or ".+\\\\$" # cannot be ".+\$" because \$ does not exist contrary to \n
- pdf.path <- sub(x = pdf.path, pattern = "[\\]$", replacement = "") # remove the last /
- }
- if(dir.exists(pdf.path) == FALSE){
- tempo.cat <- paste0("ERROR IN ", function.name, "\npdf.path ARGUMENT DOES NOT CORRESPOND TO EXISTING DIRECTORY\n", pdf.path)
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # par.ini recovery
- # cannot use pdf(file = NULL), because some small differences between pdf() and other devices. For instance, differences with windows() for par()$fin, par()$pin and par()$plt
- if(Sys.info()["sysname"] == "Windows"){ # Note that .Platform$OS.type() only says "unix" for macOS and Linux and "Windows" for Windows
- open.fail <- NULL
- grDevices::windows()
- ini.par <- par(no.readonly = remove.read.only) # to recover the initial graphical parameters if required (reset). BEWARE: this command alone opens a pdf of GUI window if no window already opened. But here, protected with the code because always a tempo window opened
- invisible(dev.off()) # close the new window
- }else if(Sys.info()["sysname"] == "Linux"){
- if(pdf == TRUE){
- tempo.code <- 0
- while(file.exists(paste0(pdf.path, "/recover_ini_par", tempo.code, ".pdf")) == TRUE){
- tempo.code <- tempo.code + 1
- }
- grDevices::pdf(width = width, height = height, file=paste0(pdf.path, "/recover_ini_par", tempo.code, ".pdf"), paper = paper)
- ini.par <- par(no.readonly = remove.read.only) # to recover the initial graphical parameters if required (reset). BEWARE: this command alone opens a pdf of GUI window if no window already opened. But here, protected with the code because always a tempo window opened
- invisible(dev.off()) # close the pdf window
- file.remove(paste0(pdf.path, "/recover_ini_par", tempo.code, ".pdf")) # remove the pdf file
- }else{
- # test if X11 can be opened
- if(file.exists(paste0(getwd(), "/Rplots.pdf"))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nTHIS FUNCTION CANNOT BE USED ON LINUX IF A Rplots.pdf FILE ALREADY EXISTS HERE\n", getwd())
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- open.fail <- suppressWarnings(try(grDevices::X11(), silent = TRUE))[] # try to open a X11 window. If open.fail == NULL, no problem, meaning that the X11 window is opened. If open.fail != NULL, a pdf can be opened here paste0(getwd(), "/Rplots.pdf")
- if(is.null(open.fail)){
- ini.par <- par(no.readonly = remove.read.only) # to recover the initial graphical parameters if required (reset). BEWARE: this command alone opens a pdf of GUI window if no window already opened. But here, protected with the code because always a tempo window opened
- invisible(dev.off()) # close the new window
- }else if(file.exists(paste0(getwd(), "/Rplots.pdf"))){
- file.remove(paste0(getwd(), "/Rplots.pdf")) # remove the pdf file
- tempo.cat <- ("ERROR IN fun_open()\nTHIS FUNCTION CANNOT OPEN GUI ON LINUX OR NON MACOS UNIX SYSTEM (X GRAPHIC INTERFACE HAS TO BE SET)\nTO OVERCOME THIS, PLEASE SET pdf ARGUMENT TO TRUE AND RERUN")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- }
- }else{
- open.fail <- NULL
- grDevices::quartz()
- ini.par <- par(no.readonly = remove.read.only) # to recover the initial graphical parameters if required (reset). BEWARE: this command alone opens a pdf of GUI window if no window already opened. But here, protected with the code because always a tempo window opened
- invisible(dev.off()) # close the new window
- }
- # end par.ini recovery
- zone.ini <- matrix(1, ncol=1) # to recover the initial parameters for next figure region when device region split into several figure regions
- if(pdf == TRUE){
- if(grepl(x = pdf.name, pattern = "\\.pdf$")){
- pdf.name <- sub(x = pdf.name, pattern = "\\.pdf$", replacement = "") # remove the last .pdf
- }
- pdf.loc <- paste0(pdf.path, "/", pdf.name, ".pdf")
- if(file.exists(pdf.loc) == TRUE & pdf.overwrite == FALSE){
- tempo.cat <- paste0("ERROR IN ", function.name, "\npdf.loc FILE ALREADY EXISTS AND CANNOT BE OVERWRITTEN DUE TO pdf.overwrite ARGUMENT SET TO TRUE\n", pdf.loc)
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- grDevices::pdf(width = width, height = height, file=pdf.loc, paper = paper)
- }
- }else if(pdf == FALSE){
- pdf.loc <- NULL
- if(Sys.info()["sysname"] == "Windows"){ # .Platform$OS.type() only says "unix" for macOS and Linux and "Windows" for Windows
- grDevices::windows(width = width, height = height, rescale = rescale)
- }else if(Sys.info()["sysname"] == "Linux"){
- if( ! is.null(open.fail)){
- tempo.cat <- "ERROR IN fun_open()\nTHIS FUNCTION CANNOT OPEN GUI ON LINUX OR NON MACOS UNIX SYSTEM (X GRAPHIC INTERFACE HAS TO BE SET)\nTO OVERCOME THIS, PLEASE SET pdf ARGUMENT TO TRUE AND RERUN"
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- grDevices::X11(width = width, height = height)
- }
- }else{
- grDevices::quartz(width = width, height = height)
- }
- }
- if(return.output == TRUE){
- output <- list(pdf.loc = pdf.loc, ini.par = ini.par, zone.ini = zone.ini, dim = dev.size())
- return(output)
- }
+# AIM
+# open a pdf or screen (GUI) graphic window and return initial graphic parameters
+# this order can be used:
+# fun_width()
+# fun_open()
+# fun_prior_plot() # not for ggplot2
+# plot() or any other plotting
+# fun_post_plot() if fun_prior_plot() has been used # not for ggplot2
+# fun_close()
+# WARNINGS
+# On Linux, use pdf = TRUE, if (GUI) graphic window is not always available, meaning that X is not installed (clusters for instance). Use X11() in R to test if available
+# ARGUMENTS:
+# pdf: logical. Use pdf display? If FALSE, a GUI is opened
+# pdf.path: where the pdf is saved (do not terminate by / or \\). Write "working.dir" if working directory is required (default). Ignored if pdf == FALSE
+# pdf.name: name of the pdf file containing the graphs (the .pdf extension is added by the function, if not detected in the name end). Ignored if pdf == FALSE
+# width: width of the window (in inches)
+# height: height of the window (in inches)
+# paper: paper argument of the pdf function (paper format). Only used for pdf(). Either "a4", "letter", "legal", "us", "executive", "a4r", "USr" or "special". If "special", means that the paper dimension will be width and height. With another paper format, if width or height is over the size of the paper, width or height will be modified such that the plot is adjusted to the paper dimension (see $dim in the returned list below to see the modified dimensions). Ignored if pdf == FALSE
+# pdf.overwrite: logical. Existing pdf can be overwritten? . Ignored if pdf == FALSE
+# rescale: kind of GUI. Either "R", "fit", or "fixed". Ignored on Mac and Linux OS. See ?windows for details
+# remove.read.only: logical. remove the read only (R.O.) graphical parameters? If TRUE, the graphical parameters are returned without the R.O. parameters. The returned $ini.par list can be used to set the par() of a new graphical device. If FALSE, graphical parameters are returned with the R.O. parameters, which provides information like text dimension (see ?par() ). The returned $ini.par list can be used to set the par() of a new graphical device, but generate a warning message. Ignored if return.output == FALSE.
+# return.output: logical. Return output ? If TRUE the output list is displayed
+# RETURN
+# a list containing:
+# $pdf.loc: path of the pdf created
+# $ini.par: initial par() parameters
+# $zone.ini: initial window spliting
+# $dim: dimension of the graphical device (in inches)
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# fun_open(pdf = FALSE, pdf.path = "C:/Users/Gael/Desktop", pdf.name = "graph", width = 7, height = 7, paper = "special", pdf.overwrite = FALSE, return.output = TRUE)
+# DEBUGGING
+# pdf = TRUE ; pdf.path = "C:/Users/Gael/Desktop" ; pdf.name = "graphs" ; width = 7 ; height = 7 ; paper = "special" ; pdf.overwrite = FALSE ; rescale = "fixed" ; remove.read.only = TRUE ; return.output = TRUE # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = pdf, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = pdf.path, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = pdf.name, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = width, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = height, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = paper, options = c("a4", "letter", "legal", "us", "executive", "a4r", "USr", "special", "A4", "LETTER", "LEGAL", "US"), length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data =pdf.overwrite, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = rescale, options = c("R", "fit", "fixed"), length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = remove.read.only, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = return.output, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+if(pdf.path == "working.dir"){
+pdf.path <- getwd()
+}else{
+if(grepl(x = pdf.path, pattern = ".+/$")){
+pdf.path <- sub(x = pdf.path, pattern = "/$", replacement = "") # remove the last /
+}else if(grepl(x = pdf.path, pattern = ".+[\\]$")){ # or ".+\\\\$" # cannot be ".+\$" because \$ does not exist contrary to \n
+pdf.path <- sub(x = pdf.path, pattern = "[\\]$", replacement = "") # remove the last /
+}
+if(dir.exists(pdf.path) == FALSE){
+tempo.cat <- paste0("ERROR IN ", function.name, "\npdf.path ARGUMENT DOES NOT CORRESPOND TO EXISTING DIRECTORY\n", pdf.path)
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# par.ini recovery
+# cannot use pdf(file = NULL), because some small differences between pdf() and other devices. For instance, differences with windows() for par()$fin, par()$pin and par()$plt
+if(Sys.info()["sysname"] == "Windows"){ # Note that .Platform$OS.type() only says "unix" for macOS and Linux and "Windows" for Windows
+open.fail <- NULL
+grDevices::windows()
+ini.par <- par(no.readonly = remove.read.only) # to recover the initial graphical parameters if required (reset). BEWARE: this command alone opens a pdf of GUI window if no window already opened. But here, protected with the code because always a tempo window opened
+invisible(dev.off()) # close the new window
+}else if(Sys.info()["sysname"] == "Linux"){
+if(pdf == TRUE){
+tempo.code <- 0
+while(file.exists(paste0(pdf.path, "/recover_ini_par", tempo.code, ".pdf")) == TRUE){
+tempo.code <- tempo.code + 1
+}
+grDevices::pdf(width = width, height = height, file=paste0(pdf.path, "/recover_ini_par", tempo.code, ".pdf"), paper = paper)
+ini.par <- par(no.readonly = remove.read.only) # to recover the initial graphical parameters if required (reset). BEWARE: this command alone opens a pdf of GUI window if no window already opened. But here, protected with the code because always a tempo window opened
+invisible(dev.off()) # close the pdf window
+file.remove(paste0(pdf.path, "/recover_ini_par", tempo.code, ".pdf")) # remove the pdf file
+}else{
+# test if X11 can be opened
+if(file.exists(paste0(getwd(), "/Rplots.pdf"))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nTHIS FUNCTION CANNOT BE USED ON LINUX IF A Rplots.pdf FILE ALREADY EXISTS HERE\n", getwd())
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+open.fail <- suppressWarnings(try(grDevices::X11(), silent = TRUE))[] # try to open a X11 window. If open.fail == NULL, no problem, meaning that the X11 window is opened. If open.fail != NULL, a pdf can be opened here paste0(getwd(), "/Rplots.pdf")
+if(is.null(open.fail)){
+ini.par <- par(no.readonly = remove.read.only) # to recover the initial graphical parameters if required (reset). BEWARE: this command alone opens a pdf of GUI window if no window already opened. But here, protected with the code because always a tempo window opened
+invisible(dev.off()) # close the new window
+}else if(file.exists(paste0(getwd(), "/Rplots.pdf"))){
+file.remove(paste0(getwd(), "/Rplots.pdf")) # remove the pdf file
+tempo.cat <- ("ERROR IN fun_open()\nTHIS FUNCTION CANNOT OPEN GUI ON LINUX OR NON MACOS UNIX SYSTEM (X GRAPHIC INTERFACE HAS TO BE SET)\nTO OVERCOME THIS, PLEASE SET pdf ARGUMENT TO TRUE AND RERUN")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+}
+}else{
+open.fail <- NULL
+grDevices::quartz()
+ini.par <- par(no.readonly = remove.read.only) # to recover the initial graphical parameters if required (reset). BEWARE: this command alone opens a pdf of GUI window if no window already opened. But here, protected with the code because always a tempo window opened
+invisible(dev.off()) # close the new window
+}
+# end par.ini recovery
+zone.ini <- matrix(1, ncol=1) # to recover the initial parameters for next figure region when device region split into several figure regions
+if(pdf == TRUE){
+if(grepl(x = pdf.name, pattern = "\\.pdf$")){
+pdf.name <- sub(x = pdf.name, pattern = "\\.pdf$", replacement = "") # remove the last .pdf
+}
+pdf.loc <- paste0(pdf.path, "/", pdf.name, ".pdf")
+if(file.exists(pdf.loc) == TRUE & pdf.overwrite == FALSE){
+tempo.cat <- paste0("ERROR IN ", function.name, "\npdf.loc FILE ALREADY EXISTS AND CANNOT BE OVERWRITTEN DUE TO pdf.overwrite ARGUMENT SET TO TRUE\n", pdf.loc)
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+grDevices::pdf(width = width, height = height, file=pdf.loc, paper = paper)
+}
+}else if(pdf == FALSE){
+pdf.loc <- NULL
+if(Sys.info()["sysname"] == "Windows"){ # .Platform$OS.type() only says "unix" for macOS and Linux and "Windows" for Windows
+grDevices::windows(width = width, height = height, rescale = rescale)
+}else if(Sys.info()["sysname"] == "Linux"){
+if( ! is.null(open.fail)){
+tempo.cat <- "ERROR IN fun_open()\nTHIS FUNCTION CANNOT OPEN GUI ON LINUX OR NON MACOS UNIX SYSTEM (X GRAPHIC INTERFACE HAS TO BE SET)\nTO OVERCOME THIS, PLEASE SET pdf ARGUMENT TO TRUE AND RERUN"
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+grDevices::X11(width = width, height = height)
+}
+}else{
+grDevices::quartz(width = width, height = height)
+}
+}
+if(return.output == TRUE){
+output <- list(pdf.loc = pdf.loc, ini.par = ini.par, zone.ini = zone.ini, dim = dev.size())
+return(output)
+}
}
@@ -4000,8222 +3999,8222 @@ fun_open <- function(
# Check OK: clear to go Apollo
fun_prior_plot <- function(
- param.reinitial = FALSE,
- xlog.scale = FALSE,
- ylog.scale = FALSE,
- remove.label = TRUE,
- remove.x.axis = TRUE,
- remove.y.axis = TRUE,
- std.x.range = TRUE,
- std.y.range = TRUE,
- down.space = 1,
- left.space = 1,
- up.space = 1,
- right.space = 1,
- orient = 1,
- dist.legend = 3.5,
- tick.length = 0.5,
- box.type = "n",
- amplif.label = 1,
- amplif.axis = 1,
- display.extend = FALSE,
- return.par = FALSE
+param.reinitial = FALSE,
+xlog.scale = FALSE,
+ylog.scale = FALSE,
+remove.label = TRUE,
+remove.x.axis = TRUE,
+remove.y.axis = TRUE,
+std.x.range = TRUE,
+std.y.range = TRUE,
+down.space = 1,
+left.space = 1,
+up.space = 1,
+right.space = 1,
+orient = 1,
+dist.legend = 3.5,
+tick.length = 0.5,
+box.type = "n",
+amplif.label = 1,
+amplif.axis = 1,
+display.extend = FALSE,
+return.par = FALSE
){
- # AIM
- # very convenient to erase the axes for post plot axis redrawing using fun_post_plot()
- # reinitialize and set the graphic parameters before plotting
- # CANNOT be used if no graphic device already opened
- # ARGUMENTS
- # param.reinitial: reinitialize graphic parameters before applying the new ones, as defined by the other arguments? Either TRUE or FALSE
- # xlog.scale: Log scale for the x-axis? Either TRUE or FALSE. If TRUE, erases the x-axis, except legend, for further drawing by fun_post_plot()(xlog argument of par())
- # ylog.scale: Log scale for the y-axis? Either TRUE or FALSE. If TRUE, erases the y-axis, except legend, for further drawing by fun_post_plot()(ylog argument of par())
- # remove.label: remove labels (axis legend) of the two axes? Either TRUE or FALSE (ann argument of par())
- # remove.x.axis: remove x-axis except legend? Either TRUE or FALSE (control the xaxt argument of par()). Automately set to TRUE if xlog.scale == TRUE
- # remove.y.axis: remove y-axis except legend? Either TRUE or FALSE (control the yaxt argument of par()). Automately set to TRUE if ylog.scale == TRUE
- # std.x.range: standard range on the x-axis? TRUE (no range extend) or FALSE (4% range extend). Controls xaxs argument of par() (TRUE is xaxs = "i", FALSE is xaxs = "r")
- # std.y.range: standard range on the y-axis? TRUE (no range extend) or FALSE (4% range extend). Controls yaxs argument of par() (TRUE is yaxs = "i", FALSE is yaxs = "r")
- # down.space: lower vertical margin (in inches, mai argument of par())
- # left.space: left horizontal margin (in inches, mai argument of par())
- # up.space: upper vertical margin between plot region and grapical window (in inches, mai argument of par())
- # right.space: right horizontal margin (in inches, mai argument of par())
- # orient: scale number orientation (las argument of par()). 0, always parallel to the axis; 1, always horizontal; 2, always perpendicular to the axis; 3, always vertical
- # dist.legend: numeric value that moves axis legends away in inches (first number of mgp argument of par() but in inches thus / 0.2)
- # tick.length: length of the ticks (1 means complete the distance between the plot region and the axis numbers, 0.5 means half the length, etc. 0 means no tick
- # box.type: bty argument of par(). Either "o", "l", "7", "c", "u", "]", the resulting box resembles the corresponding upper case letter. A value of "n" suppresses the box
- # amplif.label: increase or decrease the size of the text in legends
- # amplif.axis: increase or decrease the size of the scale numbers in axis
- # display.extend: extend display beyond plotting region? Either TRUE or FALSE (xpd argument of par() without NA)
- # return.par: return graphic parameter modification?
- # RETURN
- # return graphic parameter modification
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # fun_prior_plot(param.reinitial = FALSE, xlog.scale = FALSE, ylog.scale = FALSE, remove.label = TRUE, remove.x.axis = TRUE, remove.y.axis = TRUE, std.x.range = TRUE, std.y.range = TRUE, down.space = 1, left.space = 1, up.space = 1, right.space = 1, orient = 1, dist.legend = 4.5, tick.length = 0.5, box.type = "n", amplif.label = 1, amplif.axis = 1, display.extend = FALSE, return.par = FALSE)
- # DEBUGGING
- # param.reinitial = FALSE ; xlog.scale = FALSE ; ylog.scale = FALSE ; remove.label = TRUE ; remove.x.axis = TRUE ; remove.y.axis = TRUE ; std.x.range = TRUE ; std.y.range = TRUE ; down.space = 1 ; left.space = 1 ; up.space = 1 ; right.space = 1 ; orient = 1 ; dist.legend = 4.5 ; tick.length = 0.5 ; box.type = "n" ; amplif.label = 1 ; amplif.axis = 1 ; display.extend = FALSE ; return.par = FALSE # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = param.reinitial, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = xlog.scale, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = ylog.scale, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = remove.label, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = remove.x.axis, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = remove.y.axis, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = std.x.range, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = std.y.range, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = down.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = left.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = up.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = right.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = orient, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = dist.legend, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = tick.length, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = box.type, options = c("o", "l", "7", "c", "u", "]", "n"), length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = amplif.label, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = amplif.axis, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = display.extend, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = return.par, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- if(is.null(dev.list())){
- tempo.cat <- paste0("ERROR IN ", function.name, ": THIS FUNCTION CANNOT BE USED IF NO GRAPHIC DEVICE ALREADY OPENED (dev.list() IS CURRENTLY NULL)")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # par.ini recovery
- # cannot use pdf(file = NULL), because some small differences between pdf() and other devices. For instance, differences with windows() for par()$fin, par()$pin and par()$plt
- if(param.reinitial == TRUE){
- if( ! all(names(dev.cur()) == "null device")){
- active.wind.nb <- dev.cur()
- }else{
- active.wind.nb <- 0
- }
- if(Sys.info()["sysname"] == "Windows"){ # Note that .Platform$OS.type() only says "unix" for macOS and Linux and "Windows" for Windows
- grDevices::windows()
- ini.par <- par(no.readonly = FALSE) # to recover the initial graphical parameters if required (reset). BEWARE: this command alone opens a pdf of GUI window if no window already opened. But here, protected with the code because always a tempo window opened
- invisible(dev.off()) # close the new window
- }else if(Sys.info()["sysname"] == "Linux"){
- if(file.exists(paste0(getwd(), "/Rplots.pdf"))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": THIS FUNCTION CANNOT BE USED ON LINUX WITH param.reinitial SET TO TRUE IF A Rplots.pdf FILE ALREADY EXISTS HERE: ", getwd())
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- open.fail <- suppressWarnings(try(grDevices::X11(), silent = TRUE))[] # try to open a X11 window. If open.fail == NULL, no problem, meaning that the X11 window is opened. If open.fail != NULL, a pdf can be opened here paste0(getwd(), "/Rplots.pdf")
- if(is.null(open.fail)){
- ini.par <- par(no.readonly = FALSE) # to recover the initial graphical parameters if required (reset). BEWARE: this command alone opens a pdf of GUI window if no window already opened. But here, protected with the code because always a tempo window opened
- invisible(dev.off()) # close the new window
- }else if(file.exists(paste0(getwd(), "/Rplots.pdf"))){
- ini.par <- par(no.readonly = FALSE) # to recover the initial graphical parameters if required (reset). BEWARE: this command alone opens a pdf of GUI window if no window already opened. But here, protected with the code because always a tempo window opened
- invisible(dev.off()) # close the new window
- file.remove(paste0(getwd(), "/Rplots.pdf")) # remove the pdf file
- }else{
- tempo.cat <- ("ERROR IN fun_prior_plot()\nTHIS FUNCTION CANNOT OPEN GUI ON LINUX OR NON MACOS UNIX SYSTEM (X GRAPHIC INTERFACE HAS TO BE SET)\nTO OVERCOME THIS, PLEASE USE PDF GRAPHIC INTERFACES AND RERUN")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- }else{ # macOS
- grDevices::quartz()
- ini.par <- par(no.readonly = FALSE) # to recover the initial graphical parameters if required (reset). BEWARE: this command alone opens a pdf of GUI window if no window already opened. But here, protected with the code because always a tempo window opened)
- invisible(dev.off()) # close the new window
- }
- if( ! all(names(dev.cur()) == "null device")){
- invisible(dev.set(active.wind.nb)) # go back to the active window if exists
- par(ini.par) # apply the initial par to current window
- }
- }
- # end par.ini recovery
- if(remove.x.axis == TRUE){
- par(xaxt = "n") # suppress the y-axis label
- }else{
- par(xaxt = "s")
- }
- if(remove.y.axis == TRUE){
- par(yaxt = "n") # suppress the y-axis label
- }else{
- par(yaxt = "s")
- }
- if(std.x.range == TRUE){
- par(xaxs = "i")
- }else{
- par(xaxs = "r")
- }
- if(std.y.range == TRUE){
- par(yaxs = "i")
- }else{
- par(yaxs = "r")
- }
- par(mai = c(down.space, left.space, up.space, right.space), ann = ! remove.label, las = orient, mgp = c(dist.legend/0.2, 1, 0), xpd = display.extend, bty= box.type, cex.lab = amplif.label, cex.axis = amplif.axis)
- par(tcl = -par()$mgp[2] * tick.length) # tcl gives the length of the ticks as proportion of line text, knowing that mgp is in text lines. So the main ticks are a 0.5 of the distance of the axis numbers by default. The sign provides the side of the tick (negative for outside of the plot region)
- if(xlog.scale == TRUE){
- par(xaxt = "n", xlog = TRUE) # suppress the x-axis label
- }else{
- par(xlog = FALSE)
- }
- if(ylog.scale == TRUE){
- par(yaxt = "n", ylog = TRUE) # suppress the y-axis label
- }else{
- par(ylog = FALSE)
- }
- if(return.par == TRUE){
- tempo.par <- par()
- return(tempo.par)
- }
+# AIM
+# very convenient to erase the axes for post plot axis redrawing using fun_post_plot()
+# reinitialize and set the graphic parameters before plotting
+# CANNOT be used if no graphic device already opened
+# ARGUMENTS
+# param.reinitial: reinitialize graphic parameters before applying the new ones, as defined by the other arguments? Either TRUE or FALSE
+# xlog.scale: Log scale for the x-axis? Either TRUE or FALSE. If TRUE, erases the x-axis, except legend, for further drawing by fun_post_plot()(xlog argument of par())
+# ylog.scale: Log scale for the y-axis? Either TRUE or FALSE. If TRUE, erases the y-axis, except legend, for further drawing by fun_post_plot()(ylog argument of par())
+# remove.label: remove labels (axis legend) of the two axes? Either TRUE or FALSE (ann argument of par())
+# remove.x.axis: remove x-axis except legend? Either TRUE or FALSE (control the xaxt argument of par()). Automately set to TRUE if xlog.scale == TRUE
+# remove.y.axis: remove y-axis except legend? Either TRUE or FALSE (control the yaxt argument of par()). Automately set to TRUE if ylog.scale == TRUE
+# std.x.range: standard range on the x-axis? TRUE (no range extend) or FALSE (4% range extend). Controls xaxs argument of par() (TRUE is xaxs = "i", FALSE is xaxs = "r")
+# std.y.range: standard range on the y-axis? TRUE (no range extend) or FALSE (4% range extend). Controls yaxs argument of par() (TRUE is yaxs = "i", FALSE is yaxs = "r")
+# down.space: lower vertical margin (in inches, mai argument of par())
+# left.space: left horizontal margin (in inches, mai argument of par())
+# up.space: upper vertical margin between plot region and grapical window (in inches, mai argument of par())
+# right.space: right horizontal margin (in inches, mai argument of par())
+# orient: scale number orientation (las argument of par()). 0, always parallel to the axis; 1, always horizontal; 2, always perpendicular to the axis; 3, always vertical
+# dist.legend: numeric value that moves axis legends away in inches (first number of mgp argument of par() but in inches thus / 0.2)
+# tick.length: length of the ticks (1 means complete the distance between the plot region and the axis numbers, 0.5 means half the length, etc. 0 means no tick
+# box.type: bty argument of par(). Either "o", "l", "7", "c", "u", "]", the resulting box resembles the corresponding upper case letter. A value of "n" suppresses the box
+# amplif.label: increase or decrease the size of the text in legends
+# amplif.axis: increase or decrease the size of the scale numbers in axis
+# display.extend: extend display beyond plotting region? Either TRUE or FALSE (xpd argument of par() without NA)
+# return.par: return graphic parameter modification?
+# RETURN
+# return graphic parameter modification
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# fun_prior_plot(param.reinitial = FALSE, xlog.scale = FALSE, ylog.scale = FALSE, remove.label = TRUE, remove.x.axis = TRUE, remove.y.axis = TRUE, std.x.range = TRUE, std.y.range = TRUE, down.space = 1, left.space = 1, up.space = 1, right.space = 1, orient = 1, dist.legend = 4.5, tick.length = 0.5, box.type = "n", amplif.label = 1, amplif.axis = 1, display.extend = FALSE, return.par = FALSE)
+# DEBUGGING
+# param.reinitial = FALSE ; xlog.scale = FALSE ; ylog.scale = FALSE ; remove.label = TRUE ; remove.x.axis = TRUE ; remove.y.axis = TRUE ; std.x.range = TRUE ; std.y.range = TRUE ; down.space = 1 ; left.space = 1 ; up.space = 1 ; right.space = 1 ; orient = 1 ; dist.legend = 4.5 ; tick.length = 0.5 ; box.type = "n" ; amplif.label = 1 ; amplif.axis = 1 ; display.extend = FALSE ; return.par = FALSE # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = param.reinitial, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = xlog.scale, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = ylog.scale, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = remove.label, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = remove.x.axis, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = remove.y.axis, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = std.x.range, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = std.y.range, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = down.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = left.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = up.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = right.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = orient, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = dist.legend, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = tick.length, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = box.type, options = c("o", "l", "7", "c", "u", "]", "n"), length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = amplif.label, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = amplif.axis, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = display.extend, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = return.par, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+if(is.null(dev.list())){
+tempo.cat <- paste0("ERROR IN ", function.name, ": THIS FUNCTION CANNOT BE USED IF NO GRAPHIC DEVICE ALREADY OPENED (dev.list() IS CURRENTLY NULL)")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# par.ini recovery
+# cannot use pdf(file = NULL), because some small differences between pdf() and other devices. For instance, differences with windows() for par()$fin, par()$pin and par()$plt
+if(param.reinitial == TRUE){
+if( ! all(names(dev.cur()) == "null device")){
+active.wind.nb <- dev.cur()
+}else{
+active.wind.nb <- 0
+}
+if(Sys.info()["sysname"] == "Windows"){ # Note that .Platform$OS.type() only says "unix" for macOS and Linux and "Windows" for Windows
+grDevices::windows()
+ini.par <- par(no.readonly = FALSE) # to recover the initial graphical parameters if required (reset). BEWARE: this command alone opens a pdf of GUI window if no window already opened. But here, protected with the code because always a tempo window opened
+invisible(dev.off()) # close the new window
+}else if(Sys.info()["sysname"] == "Linux"){
+if(file.exists(paste0(getwd(), "/Rplots.pdf"))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": THIS FUNCTION CANNOT BE USED ON LINUX WITH param.reinitial SET TO TRUE IF A Rplots.pdf FILE ALREADY EXISTS HERE: ", getwd())
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+open.fail <- suppressWarnings(try(grDevices::X11(), silent = TRUE))[] # try to open a X11 window. If open.fail == NULL, no problem, meaning that the X11 window is opened. If open.fail != NULL, a pdf can be opened here paste0(getwd(), "/Rplots.pdf")
+if(is.null(open.fail)){
+ini.par <- par(no.readonly = FALSE) # to recover the initial graphical parameters if required (reset). BEWARE: this command alone opens a pdf of GUI window if no window already opened. But here, protected with the code because always a tempo window opened
+invisible(dev.off()) # close the new window
+}else if(file.exists(paste0(getwd(), "/Rplots.pdf"))){
+ini.par <- par(no.readonly = FALSE) # to recover the initial graphical parameters if required (reset). BEWARE: this command alone opens a pdf of GUI window if no window already opened. But here, protected with the code because always a tempo window opened
+invisible(dev.off()) # close the new window
+file.remove(paste0(getwd(), "/Rplots.pdf")) # remove the pdf file
+}else{
+tempo.cat <- ("ERROR IN fun_prior_plot()\nTHIS FUNCTION CANNOT OPEN GUI ON LINUX OR NON MACOS UNIX SYSTEM (X GRAPHIC INTERFACE HAS TO BE SET)\nTO OVERCOME THIS, PLEASE USE PDF GRAPHIC INTERFACES AND RERUN")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+}else{ # macOS
+grDevices::quartz()
+ini.par <- par(no.readonly = FALSE) # to recover the initial graphical parameters if required (reset). BEWARE: this command alone opens a pdf of GUI window if no window already opened. But here, protected with the code because always a tempo window opened)
+invisible(dev.off()) # close the new window
+}
+if( ! all(names(dev.cur()) == "null device")){
+invisible(dev.set(active.wind.nb)) # go back to the active window if exists
+par(ini.par) # apply the initial par to current window
+}
+}
+# end par.ini recovery
+if(remove.x.axis == TRUE){
+par(xaxt = "n") # suppress the y-axis label
+}else{
+par(xaxt = "s")
+}
+if(remove.y.axis == TRUE){
+par(yaxt = "n") # suppress the y-axis label
+}else{
+par(yaxt = "s")
+}
+if(std.x.range == TRUE){
+par(xaxs = "i")
+}else{
+par(xaxs = "r")
+}
+if(std.y.range == TRUE){
+par(yaxs = "i")
+}else{
+par(yaxs = "r")
+}
+par(mai = c(down.space, left.space, up.space, right.space), ann = ! remove.label, las = orient, mgp = c(dist.legend/0.2, 1, 0), xpd = display.extend, bty= box.type, cex.lab = amplif.label, cex.axis = amplif.axis)
+par(tcl = -par()$mgp[2] * tick.length) # tcl gives the length of the ticks as proportion of line text, knowing that mgp is in text lines. So the main ticks are a 0.5 of the distance of the axis numbers by default. The sign provides the side of the tick (negative for outside of the plot region)
+if(xlog.scale == TRUE){
+par(xaxt = "n", xlog = TRUE) # suppress the x-axis label
+}else{
+par(xlog = FALSE)
+}
+if(ylog.scale == TRUE){
+par(yaxt = "n", ylog = TRUE) # suppress the y-axis label
+}else{
+par(ylog = FALSE)
+}
+if(return.par == TRUE){
+tempo.par <- par()
+return(tempo.par)
+}
}
######## fun_scale() #### select nice label numbers when setting number of ticks on an axis
-
+
# Check OK: clear to go Apollo
fun_scale <- function(n, lim, kind = "approx", lib.path = NULL){
- # AIM
- # attempt to select nice scale numbers when setting n ticks on a lim axis range
- # ARGUMENTS
- # n: desired number of main ticks on the axis (integer above 0)
- # lim: vector of 2 numbers indicating the limit range of the axis. Order of the 2 values matters (for inverted axis). Can be log transformed values
- # kind: either "approx" (approximative), "strict" (strict) or "strict.cl" (strict clean). If "approx", use the scales::trans_breaks() function to provide an easy to read scale of approximately n ticks spanning the range of the lim argument. If "strict", cut the range of the lim argument into n + 1 equidistant part and return the n numbers at each boundary. This often generates numbers uneasy to read. If "strict.cl", provide an easy to read scale of exactly n ticks, but sometimes not completely spanning the range of the lim argument
- # lib.path: character vector specifying the absolute pathways of the directories containing the required packages if not in the default directories. Ignored if NULL
- # RETURN
- # a vector of numbers
- # REQUIRED PACKAGES
- # if kind = "approx":
- # ggplot2
- # scales
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # fun_round()
- # EXAMPLES
- # approximate number of main ticks
- # ymin = 2 ; ymax = 3.101 ; n = 5 ; scale <- fun_scale(n = n, lim = c(ymin, ymax), kind = "approx") ; scale ; par(yaxt = "n", yaxs = "i", las = 1) ; plot(ymin:ymax, ymin:ymax, xlim = range(scale, ymin, ymax)[order(c(ymin, ymax))], ylim = range(scale, ymin, ymax)[order(c(ymin, ymax))], xlab = "DEFAULT SCALE", ylab = "NEW SCALE") ; par(yaxt = "s") ; axis(side = 2, at = scale)
- # strict number of main ticks
- # ymin = 2 ; ymax = 3.101 ; n = 5 ; scale <- fun_scale(n = n, lim = c(ymin, ymax), kind = "strict") ; scale ; par(yaxt = "n", yaxs = "i", las = 1) ; plot(ymin:ymax, ymin:ymax, xlim = range(scale, ymin, ymax)[order(c(ymin, ymax))], ylim = range(scale, ymin, ymax)[order(c(ymin, ymax))], xlab = "DEFAULT SCALE", ylab = "NEW SCALE") ; par(yaxt = "s") ; axis(side = 2, at = scale)
- # strict "clean" number of main ticks
- # ymin = 2 ; ymax = 3.101 ; n = 5 ; scale <- fun_scale(n = n, lim = c(ymin, ymax), kind = "strict.cl") ; scale ; par(yaxt = "n", yaxs = "i", las = 1) ; plot(ymin:ymax, ymin:ymax, xlim = range(scale, ymin, ymax)[order(c(ymin, ymax))], ylim = range(scale, ymin, ymax)[order(c(ymin, ymax))], xlab = "DEFAULT SCALE", ylab = "NEW SCALE") ; par(yaxt = "s") ; axis(side = 2, at = scale)
- # approximate number of main ticks, scale inversion
- # ymin = 3.101 ; ymax = 2 ; n = 5 ; scale <- fun_scale(n = n, lim = c(ymin, ymax), kind = "approx") ; scale ; par(yaxt = "n", yaxs = "i", las = 1) ; plot(ymin:ymax, ymin:ymax, xlim = range(scale, ymin, ymax)[order(c(ymin, ymax))], ylim = range(scale, ymin, ymax)[order(c(ymin, ymax))], xlab = "DEFAULT SCALE", ylab = "NEW SCALE") ; par(yaxt = "s") ; axis(side = 2, at = scale)
- # DEBUGGING
- # n = 9 ; lim = c(2, 3.101) ; kind = "approx" ; lib.path = NULL # for function debugging
- # n = 10 ; lim = c(1e-4, 1e6) ; kind = "approx" ; lib.path = NULL # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # end initial argument checking
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(length(utils::find("fun_round", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_round() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = n, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & isTRUE(all.equal(n, 0))){ # isTRUE(all.equal(n, 0)) equivalent to n == 0 but deals with floats (approx ok)
- tempo.cat <- paste0("ERROR IN ", function.name, ": n ARGUMENT MUST BE A NON NULL AND POSITIVE INTEGER")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE) #
- }
- tempo <- fun_check(data = lim, class = "vector", mode = "numeric", length = 2, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & all(diff(lim) == 0)){ # isTRUE(all.equal(diff(lim), rep(0, length(diff(lim))))) not used because we strictly need zero as a result
- tempo.cat <- paste0("ERROR IN ", function.name, ": lim ARGUMENT HAS A NULL RANGE (2 IDENTICAL VALUES)")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }else if(tempo$problem == FALSE & any(lim %in% c(Inf, -Inf))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": lim ARGUMENT CANNOT CONTAIN -Inf OR Inf VALUES")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- tempo <- fun_check(data = kind, options = c("approx", "strict", "strict.cl"), length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(lib.path)){
- tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
- tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- }
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # end argument checking with fun_check()
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- lim.rank <- rank(lim) # to deal with inverted axis
- lim <- sort(lim)
- if(kind == "approx"){
- # package checking
- fun_pack(req.package = c("ggplot2"), lib.path = lib.path)
- fun_pack(req.package = c("scales"), lib.path = lib.path)
- # end package checking
- output <- ggplot2::ggplot_build(ggplot2::ggplot() + ggplot2::scale_y_continuous(
- breaks = scales::trans_breaks(
- trans = "identity",
- inv = "identity",
- n = n
- ),
- limits = lim
- ))$layout$panel_params[[1]]$y$breaks # pretty() alone is not appropriate: tempo.pret <- pretty(seq(lim[1] ,lim[2], length.out = n)) ; tempo.pret[tempo.pret > = lim[1] & tempo.pret < = lim[2]]. # in ggplot 3.3.0, tempo.coord$y.major_source replaced by tempo.coord$y$breaks
- if( ! is.null(attributes(output))){ # layout$panel_params[[1]]$y$breaks can be characters (labels of the axis). In that case, it has attributes that corresponds to positions
- output <- unlist(attributes(output))
- }
- output <- output[ ! is.na(output)]
- }else if(kind == "strict"){
- output <- fun_round(seq(lim[1] ,lim[2], length.out = n), 2)
- }else if(kind == "strict.cl"){
- tempo.range <- diff(sort(lim))
- tempo.max <- max(lim)
- tempo.min <- min(lim)
- mid <- tempo.min + (tempo.range/2) # middle of axis
- tempo.inter <- tempo.range / (n + 1) # current interval between two ticks, between 0 and Inf
- if(tempo.inter == 0){ # isTRUE(all.equal(tempo.inter, rep(0, length(tempo.inter)))) not used because we strictly need zero as a result
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE INTERVAL BETWEEN TWO TICKS OF THE SCALE IS NULL. MODIFY THE lim OR n ARGUMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- log10.abs.lim <- 200
- log10.range <- (-log10.abs.lim):log10.abs.lim
- log10.vec <- 10^log10.range
- round.vec <- c(5, 4, 3, 2.5, 2, 1.25, 1)
- dec.table <- outer(log10.vec, round.vec) # table containing the scale units (row: power of ten from -201 to +199, column: the 5, 2.5, 2, 1.25, 1 notches
-
-
-
- # recover the number of leading zeros in tempo.inter
- ini.scipen <- options()$scipen
- options(scipen = -1000) # force scientific format
- if(any(grepl(pattern = "\\+", x = tempo.inter))){ # tempo.inter > 1
- power10.exp <- as.integer(substring(text = tempo.inter, first = (regexpr(pattern = "\\+", text = tempo.inter) + 1))) # recover the power of 10. Example recover 08 from 1e+08
- mantisse <- as.numeric(substr(x = tempo.inter, start = 1, stop = (regexpr(pattern = "\\+", text = tempo.inter) - 2))) # recover the mantisse. Example recover 1.22 from 1.22e+08
- }else if(any(grepl(pattern = "\\-", x = tempo.inter))){ # tempo.inter < 1
- power10.exp <- as.integer(substring(text = tempo.inter, first = (regexpr(pattern = "\\-", text = tempo.inter)))) # recover the power of 10. Example recover 08 from 1e+08
- mantisse <- as.numeric(substr(x = tempo.inter, start = 1, stop = (regexpr(pattern = "\\-", text = tempo.inter) - 2))) # recover the mantisse. Example recover 1.22 from 1.22e+08
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 1")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- tempo.scale <- dec.table[log10.range == power10.exp, ]
- # new interval
- inter.select <- NULL
- for(i1 in 1:length(tempo.scale)){
- tempo.first.tick <- trunc((tempo.min + tempo.scale[i1]) / tempo.scale[i1]) * (tempo.scale[i1]) # this would be use to have a number not multiple of tempo.scale[i1]: ceiling(tempo.min) + tempo.scale[i1] * 10^power10.exp
- tempo.last.tick <- tempo.first.tick + tempo.scale[i1] * (n - 1)
- if((tempo.first.tick >= tempo.min) & (tempo.last.tick <= tempo.max)){
- inter.select <- tempo.scale[i1]
- break()
- }
- }
- if(is.null(inter.select)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 2")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- options(scipen = ini.scipen) # restore the initial scientific penalty
- # end new interval
- # centering the new scale
- tempo.mid <- trunc((mid + (-1:1) * inter.select) / inter.select) * inter.select # tempo middle tick closest to the middle axis
- mid.tick <- tempo.mid[which.min(abs(tempo.mid - mid))]
- if(isTRUE(all.equal(n, rep(1, length(n))))){ # isTRUE(all.equal(n, rep(1, length(n)))) is similar to n == 1 but deals with float
- output <- mid.tick
- }else if(isTRUE(all.equal(n, rep(2, length(n))))){ # isTRUE(all.equal(n, rep(0, length(n)))) is similar to n == 2 but deals with float
- output <- mid.tick
- tempo.min.dist <- mid.tick - inter.select - tempo.min
- tempo.max.dist <- tempo.max - mid.tick + inter.select
- if(tempo.min.dist <= tempo.max.dist){ # distance between lowest tick and bottom axis <= distance between highest tick and top axis. If yes, extra tick but at the top, otherwise at the bottom
- output <- c(mid.tick, mid.tick + inter.select)
- }else{
- output <- c(mid.tick - inter.select, mid.tick)
- }
- }else if((n / 2 - trunc(n / 2)) > 0.1){ # > 0.1 to avoid floating point. Because result can only be 0 or 0.5. Thus, > 0.1 means odd number
- output <- c(mid.tick - (trunc(n / 2):1) * inter.select, mid.tick, mid.tick + (1:trunc(n / 2)) * inter.select)
- }else if((n / 2 - trunc(n / 2)) < 0.1){ # < 0.1 to avoid floating point. Because result can only be 0 or 0.5. Thus, < 0.1 means even number
- tempo.min.dist <- mid.tick - trunc(n / 2) * inter.select - tempo.min
- tempo.max.dist <- tempo.max - mid.tick + trunc(n / 2) * inter.select
- if(tempo.min.dist <= tempo.max.dist){ # distance between lowest tick and bottom axis <= distance between highest tick and top axis. If yes, extra tick but at the bottom, otherwise at the top
- output <- c(mid.tick - ((trunc(n / 2) - 1):1) * inter.select, mid.tick, mid.tick + (1:trunc(n / 2)) * inter.select)
- }else{
- output <- c(mid.tick - (trunc(n / 2):1) * inter.select, mid.tick, mid.tick + (1:(trunc(n / 2) - 1)) * inter.select)
- }
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 3")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end centering the new scale
- # last check
- if(min(output) < tempo.min){
- output <- c(output[-1], max(output) + inter.select) # remove the lowest tick and add a tick at the top
- }else if( max(output) > tempo.max){
- output <- c(min(output) - inter.select, output[-length(output)])
- }
- if(min(output) < tempo.min | max(output) > tempo.max){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 4")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(is.na(output))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 5 (NA GENERATION)")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end last check
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 6")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(diff(lim.rank) < 0){
- output <- rev(output)
- }
- return(output)
+# AIM
+# attempt to select nice scale numbers when setting n ticks on a lim axis range
+# ARGUMENTS
+# n: desired number of main ticks on the axis (integer above 0)
+# lim: vector of 2 numbers indicating the limit range of the axis. Order of the 2 values matters (for inverted axis). Can be log transformed values
+# kind: either "approx" (approximative), "strict" (strict) or "strict.cl" (strict clean). If "approx", use the scales::trans_breaks() function to provide an easy to read scale of approximately n ticks spanning the range of the lim argument. If "strict", cut the range of the lim argument into n + 1 equidistant part and return the n numbers at each boundary. This often generates numbers uneasy to read. If "strict.cl", provide an easy to read scale of exactly n ticks, but sometimes not completely spanning the range of the lim argument
+# lib.path: character vector specifying the absolute pathways of the directories containing the required packages if not in the default directories. Ignored if NULL
+# RETURN
+# a vector of numbers
+# REQUIRED PACKAGES
+# if kind = "approx":
+# ggplot2
+# scales
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# fun_round()
+# EXAMPLES
+# approximate number of main ticks
+# ymin = 2 ; ymax = 3.101 ; n = 5 ; scale <- fun_scale(n = n, lim = c(ymin, ymax), kind = "approx") ; scale ; par(yaxt = "n", yaxs = "i", las = 1) ; plot(ymin:ymax, ymin:ymax, xlim = range(scale, ymin, ymax)[order(c(ymin, ymax))], ylim = range(scale, ymin, ymax)[order(c(ymin, ymax))], xlab = "DEFAULT SCALE", ylab = "NEW SCALE") ; par(yaxt = "s") ; axis(side = 2, at = scale)
+# strict number of main ticks
+# ymin = 2 ; ymax = 3.101 ; n = 5 ; scale <- fun_scale(n = n, lim = c(ymin, ymax), kind = "strict") ; scale ; par(yaxt = "n", yaxs = "i", las = 1) ; plot(ymin:ymax, ymin:ymax, xlim = range(scale, ymin, ymax)[order(c(ymin, ymax))], ylim = range(scale, ymin, ymax)[order(c(ymin, ymax))], xlab = "DEFAULT SCALE", ylab = "NEW SCALE") ; par(yaxt = "s") ; axis(side = 2, at = scale)
+# strict "clean" number of main ticks
+# ymin = 2 ; ymax = 3.101 ; n = 5 ; scale <- fun_scale(n = n, lim = c(ymin, ymax), kind = "strict.cl") ; scale ; par(yaxt = "n", yaxs = "i", las = 1) ; plot(ymin:ymax, ymin:ymax, xlim = range(scale, ymin, ymax)[order(c(ymin, ymax))], ylim = range(scale, ymin, ymax)[order(c(ymin, ymax))], xlab = "DEFAULT SCALE", ylab = "NEW SCALE") ; par(yaxt = "s") ; axis(side = 2, at = scale)
+# approximate number of main ticks, scale inversion
+# ymin = 3.101 ; ymax = 2 ; n = 5 ; scale <- fun_scale(n = n, lim = c(ymin, ymax), kind = "approx") ; scale ; par(yaxt = "n", yaxs = "i", las = 1) ; plot(ymin:ymax, ymin:ymax, xlim = range(scale, ymin, ymax)[order(c(ymin, ymax))], ylim = range(scale, ymin, ymax)[order(c(ymin, ymax))], xlab = "DEFAULT SCALE", ylab = "NEW SCALE") ; par(yaxt = "s") ; axis(side = 2, at = scale)
+# DEBUGGING
+# n = 9 ; lim = c(2, 3.101) ; kind = "approx" ; lib.path = NULL # for function debugging
+# n = 10 ; lim = c(1e-4, 1e6) ; kind = "approx" ; lib.path = NULL # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# end initial argument checking
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(length(utils::find("fun_round", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_round() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = n, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & isTRUE(all.equal(n, 0))){ # isTRUE(all.equal(n, 0)) equivalent to n == 0 but deals with floats (approx ok)
+tempo.cat <- paste0("ERROR IN ", function.name, ": n ARGUMENT MUST BE A NON NULL AND POSITIVE INTEGER")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE) #
+}
+tempo <- fun_check(data = lim, class = "vector", mode = "numeric", length = 2, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & all(diff(lim) == 0)){ # isTRUE(all.equal(diff(lim), rep(0, length(diff(lim))))) not used because we strictly need zero as a result
+tempo.cat <- paste0("ERROR IN ", function.name, ": lim ARGUMENT HAS A NULL RANGE (2 IDENTICAL VALUES)")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}else if(tempo$problem == FALSE & any(lim %in% c(Inf, -Inf))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": lim ARGUMENT CANNOT CONTAIN -Inf OR Inf VALUES")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+tempo <- fun_check(data = kind, options = c("approx", "strict", "strict.cl"), length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(lib.path)){
+tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
+tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+}
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# end argument checking with fun_check()
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+lim.rank <- rank(lim) # to deal with inverted axis
+lim <- sort(lim)
+if(kind == "approx"){
+# package checking
+fun_pack(req.package = c("ggplot2"), lib.path = lib.path)
+fun_pack(req.package = c("scales"), lib.path = lib.path)
+# end package checking
+output <- ggplot2::ggplot_build(ggplot2::ggplot() + ggplot2::scale_y_continuous(
+breaks = scales::trans_breaks(
+trans = "identity",
+inv = "identity",
+n = n
+),
+limits = lim
+))$layout$panel_params[[1]]$y$breaks # pretty() alone is not appropriate: tempo.pret <- pretty(seq(lim[1] ,lim[2], length.out = n)) ; tempo.pret[tempo.pret > = lim[1] & tempo.pret < = lim[2]]. # in ggplot 3.3.0, tempo.coord$y.major_source replaced by tempo.coord$y$breaks
+if( ! is.null(attributes(output))){ # layout$panel_params[[1]]$y$breaks can be characters (labels of the axis). In that case, it has attributes that corresponds to positions
+output <- unlist(attributes(output))
+}
+output <- output[ ! is.na(output)]
+}else if(kind == "strict"){
+output <- fun_round(seq(lim[1] ,lim[2], length.out = n), 2)
+}else if(kind == "strict.cl"){
+tempo.range <- diff(sort(lim))
+tempo.max <- max(lim)
+tempo.min <- min(lim)
+mid <- tempo.min + (tempo.range/2) # middle of axis
+tempo.inter <- tempo.range / (n + 1) # current interval between two ticks, between 0 and Inf
+if(tempo.inter == 0){ # isTRUE(all.equal(tempo.inter, rep(0, length(tempo.inter)))) not used because we strictly need zero as a result
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE INTERVAL BETWEEN TWO TICKS OF THE SCALE IS NULL. MODIFY THE lim OR n ARGUMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+log10.abs.lim <- 200
+log10.range <- (-log10.abs.lim):log10.abs.lim
+log10.vec <- 10^log10.range
+round.vec <- c(5, 4, 3, 2.5, 2, 1.25, 1)
+dec.table <- outer(log10.vec, round.vec) # table containing the scale units (row: power of ten from -201 to +199, column: the 5, 2.5, 2, 1.25, 1 notches
+
+
+
+# recover the number of leading zeros in tempo.inter
+ini.scipen <- options()$scipen
+options(scipen = -1000) # force scientific format
+if(any(grepl(pattern = "\\+", x = tempo.inter))){ # tempo.inter > 1
+power10.exp <- as.integer(substring(text = tempo.inter, first = (regexpr(pattern = "\\+", text = tempo.inter) + 1))) # recover the power of 10. Example recover 08 from 1e+08
+mantisse <- as.numeric(substr(x = tempo.inter, start = 1, stop = (regexpr(pattern = "\\+", text = tempo.inter) - 2))) # recover the mantisse. Example recover 1.22 from 1.22e+08
+}else if(any(grepl(pattern = "\\-", x = tempo.inter))){ # tempo.inter < 1
+power10.exp <- as.integer(substring(text = tempo.inter, first = (regexpr(pattern = "\\-", text = tempo.inter)))) # recover the power of 10. Example recover 08 from 1e+08
+mantisse <- as.numeric(substr(x = tempo.inter, start = 1, stop = (regexpr(pattern = "\\-", text = tempo.inter) - 2))) # recover the mantisse. Example recover 1.22 from 1.22e+08
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 1")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+tempo.scale <- dec.table[log10.range == power10.exp, ]
+# new interval
+inter.select <- NULL
+for(i1 in 1:length(tempo.scale)){
+tempo.first.tick <- trunc((tempo.min + tempo.scale[i1]) / tempo.scale[i1]) * (tempo.scale[i1]) # this would be use to have a number not multiple of tempo.scale[i1]: ceiling(tempo.min) + tempo.scale[i1] * 10^power10.exp
+tempo.last.tick <- tempo.first.tick + tempo.scale[i1] * (n - 1)
+if((tempo.first.tick >= tempo.min) & (tempo.last.tick <= tempo.max)){
+inter.select <- tempo.scale[i1]
+break()
+}
+}
+if(is.null(inter.select)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 2")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+options(scipen = ini.scipen) # restore the initial scientific penalty
+# end new interval
+# centering the new scale
+tempo.mid <- trunc((mid + (-1:1) * inter.select) / inter.select) * inter.select # tempo middle tick closest to the middle axis
+mid.tick <- tempo.mid[which.min(abs(tempo.mid - mid))]
+if(isTRUE(all.equal(n, rep(1, length(n))))){ # isTRUE(all.equal(n, rep(1, length(n)))) is similar to n == 1 but deals with float
+output <- mid.tick
+}else if(isTRUE(all.equal(n, rep(2, length(n))))){ # isTRUE(all.equal(n, rep(0, length(n)))) is similar to n == 2 but deals with float
+output <- mid.tick
+tempo.min.dist <- mid.tick - inter.select - tempo.min
+tempo.max.dist <- tempo.max - mid.tick + inter.select
+if(tempo.min.dist <= tempo.max.dist){ # distance between lowest tick and bottom axis <= distance between highest tick and top axis. If yes, extra tick but at the top, otherwise at the bottom
+output <- c(mid.tick, mid.tick + inter.select)
+}else{
+output <- c(mid.tick - inter.select, mid.tick)
+}
+}else if((n / 2 - trunc(n / 2)) > 0.1){ # > 0.1 to avoid floating point. Because result can only be 0 or 0.5. Thus, > 0.1 means odd number
+output <- c(mid.tick - (trunc(n / 2):1) * inter.select, mid.tick, mid.tick + (1:trunc(n / 2)) * inter.select)
+}else if((n / 2 - trunc(n / 2)) < 0.1){ # < 0.1 to avoid floating point. Because result can only be 0 or 0.5. Thus, < 0.1 means even number
+tempo.min.dist <- mid.tick - trunc(n / 2) * inter.select - tempo.min
+tempo.max.dist <- tempo.max - mid.tick + trunc(n / 2) * inter.select
+if(tempo.min.dist <= tempo.max.dist){ # distance between lowest tick and bottom axis <= distance between highest tick and top axis. If yes, extra tick but at the bottom, otherwise at the top
+output <- c(mid.tick - ((trunc(n / 2) - 1):1) * inter.select, mid.tick, mid.tick + (1:trunc(n / 2)) * inter.select)
+}else{
+output <- c(mid.tick - (trunc(n / 2):1) * inter.select, mid.tick, mid.tick + (1:(trunc(n / 2) - 1)) * inter.select)
+}
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 3")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end centering the new scale
+# last check
+if(min(output) < tempo.min){
+output <- c(output[-1], max(output) + inter.select) # remove the lowest tick and add a tick at the top
+}else if( max(output) > tempo.max){
+output <- c(min(output) - inter.select, output[-length(output)])
+}
+if(min(output) < tempo.min | max(output) > tempo.max){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 4")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(is.na(output))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 5 (NA GENERATION)")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end last check
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 6")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(diff(lim.rank) < 0){
+output <- rev(output)
+}
+return(output)
+}
+
+
+######## fun_inter_ticks() #### define coordinates of secondary ticks
+
+
+fun_inter_ticks <- function(
+lim,
+log = "log10",
+breaks = NULL,
+n = NULL,
+warn.print = TRUE
+){
+# AIM
+# define coordinates and values of secondary ticks
+# ARGUMENTS
+# lim: vector of 2 numbers indicating the limit range of the axis. Order of the 2 values matters (for inverted axis). If log argument is "log2" or "log10", values in lim must be already log transformed. Thus, negative or zero values are allowed
+# log: either "log2" (values in the lim argument are log2 transformed) or "log10" (values in the lim argument are log10 transformed), or "no"
+# breaks: mandatory vector of numbers indicating the main ticks values/positions when log argument is "no". Ignored when log argument is "log2" or "log10"
+# n: number of secondary ticks between each main tick when log argument is "no". Ignored when log argument is "log2" or "log10"
+# warn.print: logical. Print potential warning messages at the end of the execution? If FALSE, warning messages are never printed, but can still be recovered in the returned list
+# RETURN
+# a list containing
+# $log: value of the log argument used
+# $coordinates: the coordinates of the secondary ticks on the axis, between the lim values
+# $values: the corresponding values associated to each coordinate (with log scale, 2^$values or 10^$values is equivalent to the labels of the axis)
+# $warn: the potential warning messages. Use cat() for proper display. NULL if no warning
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# no log scale
+# fun_inter_ticks(lim = c(-4,4), log = "no", breaks = c(-2, 0, 2), n = 3)
+# fun_inter_ticks(lim = c(10, 0), log = "no", breaks = c(10, 8, 6, 4, 2, 0), n = 4)
+# log2
+# fun_inter_ticks(lim = c(-4,4), log = "log2")
+# log10
+# fun_inter_ticks(lim = c(-2,3), log = "log10")
+# DEBUGGING
+# lim = c(2, 3.101) ; log = "no" ; breaks = NULL ; n = NULL ; warn.print = TRUE # for function debugging
+# lim = c(0, 26.5) ; log = "no" ; breaks = c(0, 10, 20) ; n = 3 # for function debugging
+# lim = c(10, 0); log = "no"; breaks = c(10, 8, 6, 4, 2, 0); n = 4 # for function debugging
+# lim = c(-10, -20); log = "no"; breaks = c(-20, -15, -10); n = 4 # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+req.function <- c(
+"fun_check"
+)
+for(i1 in req.function){
+if(length(find(i1, mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nREQUIRED ", i1, "() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# end required function checking
+# argument primary checking
+# arg with no default values
+if(any(missing(lim))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nARGUMENT lim HAS NO DEFAULT VALUE AND REQUIRES ONE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end arg with no default values
+# using fun_check()
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = lim, class = "vector", mode = "numeric", length = 2, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = log, options = c("no", "log2", "log10"), length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(breaks)){
+tempo <- fun_check(data = breaks, class = "vector", mode = "numeric", fun.name = function.name) ; eval(ee)
+}
+if( ! is.null(n)){
+tempo <- fun_check(data = n, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = warn.print, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# end using fun_check()
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument primary checking
+# second round of checking and data preparation
+# dealing with NA
+if(any(is.na(lim)) | any(is.na(log)) | any(is.na(breaks)) | any(is.na(n)) | any(is.na(warn.print))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nNO ARGUMENT CAN HAVE NA VALUES")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end dealing with NA
+# dealing with NULL
+if(is.null(lim) | is.null(log)){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nTHESE ARGUMENTS\nlim\nlog\nCANNOT BE NULL")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end dealing with NULL
+if(all(diff(lim) == 0)){ # isTRUE(all.equal(diff(lim), rep(0, length(diff(lim))))) not used because we strictly need zero as a result
+tempo.cat <- paste0("ERROR IN ", function.name, "\nlim ARGUMENT HAS A NULL RANGE (2 IDENTICAL VALUES): ", paste(lim, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else if(any(lim %in% c(Inf, -Inf))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nlim ARGUMENT CANNOT CONTAIN -Inf OR Inf VALUES")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(log == "no" & is.null(breaks)){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nbreaks ARGUMENT CANNOT BE NULL IF log ARGUMENT IS \"no\"")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! is.null(breaks)){
+if(length(breaks) < 2){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nbreaks ARGUMENT MUST HAVE 2 VALUES AT LEAST (OTHERWISE, INTER TICK POSITIONS CANNOT BE COMPUTED): ", paste(breaks, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! isTRUE(all.equal(diff(sort(breaks)), rep(diff(sort(breaks))[1], length(diff(sort(breaks))))))){ # isTRUE(all.equal(n, 0)) equivalent to n == 0 but deals with floats (approx ok)
+tempo.cat <- paste0("ERROR IN ", function.name, "\nbreaks ARGUMENT MUST HAVE EQUIDISTANT VALUES (OTHERWISE, EQUAL NUMBER OF INTER TICK BETWEEN MAIN TICKS CANNOT BE COMPUTED): ", paste(breaks, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+if( ! is.null(n)){
+if(n <= 0){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nn ARGUMENT MUST BE A POSITIVE AND NON NULL INTEGER: ", paste(n, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# end second round of checking and data preparation
+# main code
+ini.warning.length <- options()$warning.length
+warn <- NULL
+warn.count <- 0
+lim.rank <- rank(lim) # to deal with inverse axis
+if(log != "no"){
+ini.scipen <- options()$scipen
+options(scipen = -1000) # force scientific format
+power10.exp <- as.integer(substring(text = 10^lim, first = (regexpr(pattern = "\\+|\\-", text = 10^lim)))) # recover the power of 10, i.e., integer part of lim. Example recover 08 from 1e+08. Works for log2
+# mantisse <- as.numeric(substr(x = 10^lim, start = 1, stop = (regexpr(pattern = "\\+|\\-", text = 10^lim) - 2))) # recover the mantisse. Example recover 1.22 from 1.22e+08
+options(scipen = ini.scipen) # restore the initial scientific penalty
+tick.pos <- unique(as.vector(outer(2:10, ifelse(log == "log2", 2, 10)^((power10.exp[1] - ifelse(diff(lim.rank) > 0, 1, -1)):(power10.exp[2] + ifelse(diff(lim.rank) > 0, 1, -1)))))) # use log10(2:10) even if log2: it is to get log values between 0 and 1
+tick.pos <- sort(tick.pos, decreasing = ifelse(diff(lim.rank) > 0, FALSE, TRUE))
+if(log == "log2"){
+tick.values <- tick.pos[tick.pos >= min(2^lim) & tick.pos <= max(2^lim)]
+tick.pos <- log2(tick.values)
+}else if(log == "log10"){
+tick.values <- tick.pos[tick.pos >= min(10^lim) & tick.pos <= max(10^lim)]
+tick.pos <- log10(tick.values)
+}
+}else{
+# if(length(breaks) > 1){ # not required because already checked above
+breaks.rank <- rank(c(breaks[1], breaks[length(breaks)]))
+if(diff(breaks.rank) != diff(lim.rank)){
+breaks <- sort(breaks, decreasing = ifelse(diff(lim.rank) < 0, TRUE, FALSE))
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") VALUES IN breaks ARGUMENT NOT IN THE SAME ORDER AS IN lim ARGUMENT -> VALUES REORDERED AS IN lim: ", paste(breaks, collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+breaks.rank <- rank(c(breaks[1], breaks[length(breaks)]))
+}
+# }
+main.tick.dist <- mean(diff(breaks), na.rm = TRUE)
+tick.dist <- main.tick.dist / (n + 1)
+tempo.extra.margin <- max(abs(diff(breaks)), na.rm = TRUE)
+tick.pos <- seq(
+if(diff(breaks.rank) > 0){breaks[1] - tempo.extra.margin}else{breaks[1] + tempo.extra.margin},
+if(diff(breaks.rank) > 0){breaks[length(breaks)] + tempo.extra.margin}else{breaks[length(breaks)] - tempo.extra.margin},
+by = tick.dist
+)
+tick.pos <- tick.pos[tick.pos >= min(lim) & tick.pos <= max(lim)]
+tick.values <- tick.pos
+}
+if(any(is.na(tick.pos) | ! is.finite(tick.pos))){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, ": NA or Inf GENERATED FOR THE INTER TICK POSITIONS: ", paste(tick.pos, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(length(tick.pos) == 0){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NO INTER TICKS COMPUTED BETWEEN THEN LIMITS INDICATED: ", paste(lim, collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+output <- list(log = log, coordinates = tick.pos, values = tick.values, warn = warn)
+if(warn.print == TRUE & ! is.null(warn)){
+options(warning.length = 8170)
+on.exit(warning(paste0("FROM ", function.name, ":\n\n", warn), call. = FALSE)) # to recover the warning messages, see $warn
+on.exit(exp = options(warning.length = ini.warning.length), add = TRUE)
+}
+return(output)
+}
+
+
+######## fun_post_plot() #### set graph param after plotting (axes redesign for instance)
+
+
+
+
+
+# Check OK: clear to go Apollo
+fun_post_plot <- function(
+x.side = 0,
+x.log.scale = FALSE,
+x.categ = NULL,
+x.categ.pos = NULL,
+x.lab = "",
+x.axis.size = 1.5,
+x.label.size = 1.5,
+x.dist.legend = 0.5,
+x.nb.inter.tick = 1,
+y.side = 0,
+y.log.scale = FALSE,
+y.categ = NULL,
+y.categ.pos = NULL,
+y.lab = "",
+y.axis.size = 1.5,
+y.label.size = 1.5,
+y.dist.legend = 0.5,
+y.nb.inter.tick = 1,
+text.angle = 90,
+tick.length = 0.5,
+sec.tick.length = 0.3,
+bg.color = NULL,
+grid.lwd = NULL,
+grid.col = "white",
+corner.text = "",
+corner.text.size = 1,
+just.label.add = FALSE,
+par.reset = FALSE,
+custom.par = NULL
+){
+# AIM
+# redesign axis. If x.side = 0, y.side = 0, the function just adds text at topright of the graph and reset par() for next graphics and provides outputs (see below)
+# provide also positions for legend or additional text on the graph
+# use fun_prior_plot() before this function for initial inactivation of the axis drawings
+# ARGUMENTS
+# x.side: axis at the bottom (1) or top (3) of the region figure. Write 0 for no change
+# x.log.scale: Log scale for the x-axis? Either TRUE or FALSE
+# x.categ: character vector representing the classes (levels()) to specify when the x-axis is qualititative(stripchart, boxplot)
+# x.categ.pos: position of the classes names (numeric vector of identical length than x.categ). If left NULL, this will be 1:length(levels())
+# x.lab: label of the x-axis. If x.side == 0 and x.lab != "", then x.lab is printed
+# x.axis.size: positive numeric. Increase or decrease the size of the x axis numbers. Value 1 does not change it, 0.5 decreases by half, 2 increases by 2. Also control the size of displayed categories
+# x.label.size: positive numeric. Increase or decrease the size of the x axis legend text. Value 1 does not change it, 0.5 decreases by half, 2 increases by 2
+# x.dist.legend: increase the number to move x-axis legends away in inches (first number of mgp argument of par() but in inches)
+# x.nb.inter.tick: number of secondary ticks between main ticks on x-axis (only if not log scale). 0 means no secondary ticks
+# y.side: axis at the left (2) or right (4) of the region figure. Write 0 for no change
+# y.log.scale: Log scale for the y-axis? Either TRUE or FALSE
+# y.categ: classes (levels()) to specify when the y-axis is qualititative(stripchart, boxplot)
+# y.categ.pos: position of the classes names (numeric vector of identical length than y.categ). If left NULL, this will be 1:length(levels())
+# y.lab: label of the y-axis. If y.side == 0 and y.lab != "", then y.lab is printed
+# y.axis.size: positive numeric. Increase or decrease the size of the y axis numbers. Value 1 does not change it, 0.5 decreases by half, 2 increases by 2. Also control the size of displayed categories
+# y.label.size: positive numeric. Increase or decrease the size of the y axis legend text. Value 1 does not change it, 0.5 decreases by half, 2 increases by 2
+# y.dist.legend: increase the number to move y-axis legends away in inches (first number of mgp argument of par() but in inches)
+# y.nb.inter.tick: number of secondary ticks between main ticks on y-axis (only if not log scale). 0 means non secondary ticks
+# text.angle: angle of the text when axis is qualitative
+# tick.length: length of the main ticks (1 means complete the distance between the plot region and the axis numbers, 0.5 means half the length, etc., 0 for no ticks)
+# sec.tick.length: length of the secondary ticks (1 means complete the distance between the plot region and the axis numbers, 0.5 means half the length, etc., 0 for no ticks)
+# bg.color: background color of the plot region. NULL for no color. BEWARE: cover/hide an existing plot !
+# grid.lwd: if non NULL, activate the grid line (specify the line width)
+# grid.col: grid line color (only if grid.lwd non NULL)
+# corner.text: text to add at the top right corner of the window
+# corner.text.size: positive numeric. Increase or decrease the size of the text. Value 1 does not change it, 0.5 decreases by half, 2 increases by 2
+# par.reset: to reset all the graphics parameters. BEWARE: TRUE can generate display problems, mainly in graphic devices with multiple figure regions
+# just.label.add: just add axis labels (legend)? Either TRUE or FALSE. If TRUE, at least (x.side == 0 & x.lab != "") or (y.side == 0 & y.lab != "") must be set to display the corresponding x.lab or y.lab
+# custom.par: list that provides the parameters that reset all the graphics parameters. BEWARE: if NULL and par.reset == TRUE, the default par() parameters are used
+# RETURN
+# a list containing:
+# $x.mid.left.dev.region: middle of the left margin of the device region, in coordinates of the x-axis
+# $x.left.dev.region: left side of the left margin (including the potential margin of the device region), in coordinates of the x-axis
+# $x.mid.right.dev.region: middle of the right margin of the device region, in coordinates of the x-axis
+# $x.right.dev.region: right side of the right margin (including the potential margin of the device region), in coordinates of the x-axis
+# $x.mid.left.fig.region: middle of the left margin of the figure region, in coordinates of the x-axis
+# $x.left.fig.region: left side of the left margin, in coordinates of the x-axis
+# $x.mid.right.fig.region: middle of the right margin of the figure region, in coordinates of the x-axis
+# $x.right.fig.region: right side of the right margin, in coordinates of the x-axis
+# $x.left.plot.region: left side of the plot region, in coordinates of the x-axis
+# $x.right.plot.region: right side of the plot region, in coordinates of the x-axis
+# $x.mid.plot.region: middle of the plot region, in coordinates of the x-axis
+# $y.mid.bottom.dev.region: middle of the bottom margin of the device region, in coordinates of the y-axis
+# $y.bottom.dev.region: bottom side of the bottom margin (including the potential margin of the device region), in coordinates of the y-axis
+# $y.mid.top.dev.region: middle of the top margin of the device region, in coordinates of the y-axis
+# $y.top.dev.region: top side of the top margin (including the potential margin of the device region), in coordinates of the y-axis
+# $y.mid.bottom.fig.region: middle of the bottom margin of the figure region, in coordinates of the y-axis
+# $y.bottom.fig.region: bottom of the bottom margin of the figure region, in coordinates of the y-axis
+# $y.mid.top.fig.region: middle of the top margin of the figure region, in coordinates of the y-axis
+# $y.top.fig.region: top of the top margin of the figure region, in coordinates of the y-axis
+# $y.top.plot.region: top of the plot region, in coordinates of the y-axis
+# $y.bottom.plot.region: bottom of the plot region, in coordinates of the y-axis
+# $y.mid.plot.region: middle of the plot region, in coordinates of the y-axis
+# $text: warning text
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# fun_open() to reinitialize graph parameters if par.reset = TRUE and custom.par = NULL
+# EXAMPLES
+# Example of log axis with log y-axis and unmodified x-axis:
+# prior.par <- fun_prior_plot(param.reinitial = TRUE, xlog.scale = FALSE, ylog.scale = TRUE, remove.label = TRUE, remove.x.axis = FALSE, remove.y.axis = TRUE, down.space = 1, left.space = 1, up.space = 1, right.space = 1, orient = 1, dist.legend = 0.5, tick.length = 0.5, box.type = "n", amplif.label = 1, amplif.axis = 1, display.extend = FALSE, return.par = TRUE) ; plot(1:100, log = "y") ; fun_post_plot(y.side = 2, y.log.scale = prior.par$ylog, x.lab = "Values", y.lab = "TEST", y.axis.size = 1.25, y.label.size = 1.5, y.dist.legend = 0.7, just.label.add = ! prior.par$ann)
+# Example of log axis with redrawn x-axis and y-axis:
+# prior.par <- fun_prior_plot(param.reinitial = TRUE) ; plot(1:100) ; fun_post_plot(x.side = 1, x.lab = "Values", y.side = 2, y.lab = "TEST", y.axis.size = 1, y.label.size = 2, y.dist.legend = 0.6)
+# Example of title easily added to a plot:
+# plot(1:100) ; para <- fun_post_plot(corner.text = "TITLE ADDED") # try also: par(xpd = TRUE) ; text(x = para$x.mid.left.fig.region, y = para$y.mid.top.fig.region, labels = "TITLE ADDED", cex = 0.5)
+# example with margins in the device region:
+# windows(5,5) ; fun_prior_plot(box.type = "o") ; par(mai=c(0.5,0.5,0.5,0.5), omi = c(0.25,0.25,1,0.25), xaxs = "i", yaxs = "i") ; plot(0:10) ; a <- fun_post_plot(x.side = 0, y.side = 0) ; x <- c(a$x.mid.left.dev.region, a$x.left.dev.region, a$x.mid.right.dev.region, a$x.right.dev.region, a$x.mid.left.fig.region, a$x.left.fig.region, a$x.mid.right.fig.region, a$x.right.fig.region, a$x.right.plot.region, a$x.left.plot.region, a$x.mid.plot.region) ; y <- c(a$y.mid.bottom.dev.region, a$y.bottom.dev.region, a$y.mid.top.dev.region, a$y.top.dev.region, a$y.mid.bottom.fig.region, a$y.bottom.fig.region, a$y.mid.top.fig.region, a$y.top.fig.region, a$y.top.plot.region, a$y.bottom.plot.region, a$y.mid.plot.region) ; par(xpd = NA) ; points(x = rep(5, length(y)), y = y, pch = 16, col = "red") ; text(x = rep(5, length(y)), y = y, c("y.mid.bottom.dev.region", "y.bottom.dev.region", "y.mid.top.dev.region", "y.top.dev.region", "y.mid.bottom.fig.region", "y.bottom.fig.region", "y.mid.top.fig.region", "y.top.fig.region", "y.top.plot.region", "y.bottom.plot.region", "y.mid.plot.region"), cex = 0.65, col = grey(0.25)) ; points(y = rep(5, length(x)), x = x, pch = 16, col = "blue") ; text(y = rep(5, length(x)), x = x, c("x.mid.left.dev.region", "x.left.dev.region", "x.mid.right.dev.region", "x.right.dev.region", "x.mid.left.fig.region", "x.left.fig.region", "x.mid.right.fig.region", "x.right.fig.region", "x.right.plot.region", "x.left.plot.region", "x.mid.plot.region"), cex = 0.65, srt = 90, col = grey(0.25))
+# DEBUGGING
+# x.side = 0 ; x.log.scale = FALSE ; x.categ = NULL ; x.categ.pos = NULL ; x.lab = "" ; x.axis.size = 1.5 ; x.label.size = 1.5 ; x.dist.legend = 1 ; x.nb.inter.tick = 1 ; y.side = 0 ; y.log.scale = FALSE ; y.categ = NULL ; y.categ.pos = NULL ; y.lab = "" ; y.axis.size = 1.5 ; y.label.size = 1.5 ; y.dist.legend = 0.7 ; y.nb.inter.tick = 1 ; text.angle = 90 ; tick.length = 0.5 ; sec.tick.length = 0.3 ; bg.color = NULL ; grid.lwd = NULL ; grid.col = "white" ; corner.text = "" ; corner.text.size = 1 ; just.label.add = FALSE ; par.reset = FALSE ; custom.par = NULL # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(length(utils::find("fun_open", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_open() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = x.side, options = c(0, 1, 3), length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = x.log.scale, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(x.categ)){
+tempo <- fun_check(data = x.categ, class = "character", na.contain = TRUE, fun.name = function.name) ; eval(ee)
+}
+if( ! is.null(x.categ.pos)){
+tempo <- fun_check(data = x.categ.pos, class = "vector", mode = "numeric", fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = x.lab, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = x.axis.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = x.label.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = x.dist.legend, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = x.nb.inter.tick, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = y.side, options = c(0, 2, 4), length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = y.log.scale, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(y.categ)){
+tempo <- fun_check(data = y.categ, class = "character", na.contain = TRUE, fun.name = function.name) ; eval(ee)
+}
+if( ! is.null(y.categ.pos)){
+tempo <- fun_check(data = y.categ.pos, class = "vector", mode = "numeric", fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = y.lab, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = y.axis.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = y.label.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = y.dist.legend, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = y.nb.inter.tick, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = text.angle, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = tick.length, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = sec.tick.length, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
+if( ! is.null(bg.color)){
+tempo <- fun_check(data = bg.color, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+if( ! (bg.color %in% colors() | grepl(pattern = "^#", bg.color))){ # check color
+tempo.cat <- paste0("ERROR IN ", function.name, ": bg.color ARGUMENT MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # OR A COLOR NAME GIVEN BY colors()")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+if( ! is.null(grid.lwd)){
+tempo <- fun_check(data = grid.lwd, class = "vector", mode = "numeric", neg.values = FALSE, fun.name = function.name) ; eval(ee)
+}
+if( ! is.null(grid.col)){
+tempo <- fun_check(data = grid.col, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+if( ! (grid.col %in% colors() | grepl(pattern = "^#", grid.col))){ # check color
+tempo.cat <- paste0("ERROR IN ", function.name, ": grid.col ARGUMENT MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # OR A COLOR NAME GIVEN BY colors()")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+tempo <- fun_check(data = corner.text, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = corner.text.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = just.label.add, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = par.reset, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(custom.par)){
+tempo <- fun_check(data = custom.par, typeof = "list", length = 1, fun.name = function.name) ; eval(ee)
+}
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+text <- NULL
+par(tcl = -par()$mgp[2] * tick.length)
+if(x.log.scale == TRUE){
+grid.coord.x <- c(10^par("usr")[1], 10^par("usr")[2])
+}else{
+grid.coord.x <- c(par("usr")[1], par("usr")[2])
+}
+if(y.log.scale == TRUE){
+grid.coord.y <- c(10^par("usr")[3], 10^par("usr")[4])
+}else{
+grid.coord.y <- c(par("usr")[3], par("usr")[4])
+}
+if( ! is.null(bg.color)){
+rect(grid.coord.x[1], grid.coord.y[1], grid.coord.x[2], grid.coord.y[2], col = bg.color, border = NA)
+}
+if( ! is.null(grid.lwd)){
+grid(nx = NA, ny = NULL, col = grid.col, lty = 1, lwd = grid.lwd)
+}
+if(x.log.scale == TRUE){
+x.mid.left.dev.region <- 10^(par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] - ((par("usr")[2] - par("usr")[1]) / ((par("omd")[2] - par("omd")[1]) * (par("plt")[2] - par("plt")[1]))) * par("omd")[1] / 2) # in x coordinates, to position axis labeling at the bottom of the graph (according to x scale)
+x.left.dev.region <- 10^(par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] - ((par("usr")[2] - par("usr")[1]) / ((par("omd")[2] - par("omd")[1]) * (par("plt")[2] - par("plt")[1]))) * par("omd")[1]) # in x coordinates
+x.mid.right.dev.region <- 10^(par("usr")[2] + ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * (1 - par("plt")[2]) + ((par("usr")[2] - par("usr")[1]) / ((par("omd")[2] - par("omd")[1]) * (par("plt")[2] - par("plt")[1]))) * (1 - par("omd")[2]) / 2) # in x coordinates, to position axis labeling at the top of the graph (according to x scale)
+x.right.dev.region <- 10^(par("usr")[2] + ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * (1 - par("plt")[2]) + ((par("usr")[2] - par("usr")[1]) / ((par("omd")[2] - par("omd")[1]) * (par("plt")[2] - par("plt")[1]))) * (1 - par("omd")[2])) # in x coordinates
+x.mid.left.fig.region <- 10^(par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] / 2) # in x coordinates, to position axis labeling at the bottom of the graph (according to x scale)
+x.left.fig.region <- 10^(par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1]) # in x coordinates
+x.mid.right.fig.region <- 10^(par("usr")[2] + ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * (1 - par("plt")[2]) / 2) # in x coordinates, to position axis labeling at the top of the graph (according to x scale)
+x.right.fig.region <- 10^(par("usr")[2] + ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * (1 - par("plt")[2])) # in x coordinates
+x.left.plot.region <- 10^par("usr")[1] # in x coordinates, left of the plot region (according to x scale)
+x.right.plot.region <- 10^par("usr")[2] # in x coordinates, right of the plot region (according to x scale)
+x.mid.plot.region <- 10^((par("usr")[2] + par("usr")[1]) / 2) # in x coordinates, right of the plot region (according to x scale)
+}else{
+x.mid.left.dev.region <- (par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] - ((par("usr")[2] - par("usr")[1]) / ((par("omd")[2] - par("omd")[1]) * (par("plt")[2] - par("plt")[1]))) * par("omd")[1] / 2) # in x coordinates, to position axis labeling at the bottom of the graph (according to x scale)
+x.left.dev.region <- (par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] - ((par("usr")[2] - par("usr")[1]) / ((par("omd")[2] - par("omd")[1]) * (par("plt")[2] - par("plt")[1]))) * par("omd")[1]) # in x coordinates
+x.mid.right.dev.region <- (par("usr")[2] + ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * (1 - par("plt")[2]) + ((par("usr")[2] - par("usr")[1]) / ((par("omd")[2] - par("omd")[1]) * (par("plt")[2] - par("plt")[1]))) * (1 - par("omd")[2]) / 2) # in x coordinates, to position axis labeling at the top of the graph (according to x scale)
+x.right.dev.region <- (par("usr")[2] + ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * (1 - par("plt")[2]) + ((par("usr")[2] - par("usr")[1]) / ((par("omd")[2] - par("omd")[1]) * (par("plt")[2] - par("plt")[1]))) * (1 - par("omd")[2])) # in x coordinates
+x.mid.left.fig.region <- (par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] / 2) # in x coordinates, to position axis labeling at the bottom of the graph (according to x scale)
+x.left.fig.region <- (par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1]) # in x coordinates
+x.mid.right.fig.region <- (par("usr")[2] + ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * (1 - par("plt")[2]) / 2) # in x coordinates, to position axis labeling at the top of the graph (according to x scale)
+x.right.fig.region <- (par("usr")[2] + ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * (1 - par("plt")[2])) # in x coordinates
+x.left.plot.region <- par("usr")[1] # in x coordinates, left of the plot region (according to x scale)
+x.right.plot.region <- par("usr")[2] # in x coordinates, right of the plot region (according to x scale)
+x.mid.plot.region <- (par("usr")[2] + par("usr")[1]) / 2 # in x coordinates, right of the plot region (according to x scale)
+}
+if(y.log.scale == TRUE){
+y.mid.bottom.dev.region <- 10^(par("usr")[3] - ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * par("plt")[3] - ((par("usr")[4] - par("usr")[3]) / ((par("omd")[4] - par("omd")[3]) * (par("plt")[4] - par("plt")[3]))) * (par("omd")[3] / 2)) # in y coordinates, to position axis labeling at the bottom of the graph (according to y scale). Ex mid.bottom.space
+y.bottom.dev.region <- 10^(par("usr")[3] - ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * par("plt")[3] - ((par("usr")[4] - par("usr")[3]) / ((par("omd")[4] - par("omd")[3]) * (par("plt")[4] - par("plt")[3]))) * par("omd")[3]) # in y coordinates
+y.mid.top.dev.region <- 10^(par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / ((par("omd")[4] - par("omd")[3]) * (par("plt")[4] - par("plt")[3]))) * (1 - par("omd")[4]) / 2) # in y coordinates, to position axis labeling at the top of the graph (according to y scale). Ex mid.top.space
+y.top.dev.region <- 10^(par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / ((par("omd")[4] - par("omd")[3]) * (par("plt")[4] - par("plt")[3]))) * (1 - par("omd")[4])) # in y coordinates
+y.mid.bottom.fig.region <- 10^(par("usr")[3] - ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * par("plt")[3] / 2) # in y coordinates, to position axis labeling at the bottom of the graph (according to y scale). Ex mid.bottom.space
+y.bottom.fig.region <- 10^(par("usr")[3] - ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * par("plt")[3]) # in y coordinates
+y.mid.top.fig.region <- 10^(par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) / 2) # in y coordinates, to position axis labeling at the top of the graph (according to y scale). Ex mid.top.space
+y.top.fig.region <- 10^(par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4])) # in y coordinates
+y.top.plot.region <- 10^par("usr")[4] # in y coordinates, top of the plot region (according to y scale)
+y.bottom.plot.region <- 10^par("usr")[3] # in y coordinates, bottom of the plot region (according to y scale)
+y.mid.plot.region <- (par("usr")[3] + par("usr")[4]) / 2 # in x coordinates, right of the plot region (according to x scale)
+}else{
+y.mid.bottom.dev.region <- (par("usr")[3] - ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * par("plt")[3] - ((par("usr")[4] - par("usr")[3]) / ((par("omd")[4] - par("omd")[3]) * (par("plt")[4] - par("plt")[3]))) * (par("omd")[3] / 2)) # in y coordinates, to position axis labeling at the bottom of the graph (according to y scale). Ex mid.bottom.space
+y.bottom.dev.region <- (par("usr")[3] - ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * par("plt")[3] - ((par("usr")[4] - par("usr")[3]) / ((par("omd")[4] - par("omd")[3]) * (par("plt")[4] - par("plt")[3]))) * par("omd")[3]) # in y coordinates
+y.mid.top.dev.region <- (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / ((par("omd")[4] - par("omd")[3]) * (par("plt")[4] - par("plt")[3]))) * (1 - par("omd")[4]) / 2) # in y coordinates, to position axis labeling at the top of the graph (according to y scale). Ex mid.top.space
+y.top.dev.region <- (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / ((par("omd")[4] - par("omd")[3]) * (par("plt")[4] - par("plt")[3]))) * (1 - par("omd")[4])) # in y coordinates
+y.mid.bottom.fig.region <- (par("usr")[3] - ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * par("plt")[3] / 2) # in y coordinates, to position axis labeling at the bottom of the graph (according to y scale). Ex mid.bottom.space
+y.bottom.fig.region <- (par("usr")[3] - ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * par("plt")[3]) # in y coordinates
+y.mid.top.fig.region <- (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) / 2) # in y coordinates, to position axis labeling at the top of the graph (according to y scale). Ex mid.top.space
+y.top.fig.region <- (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4])) # in y coordinates
+y.top.plot.region <- par("usr")[4] # in y coordinates, top of the plot region (according to y scale)
+y.bottom.plot.region <- par("usr")[3] # in y coordinates, bottom of the plot region (according to y scale)
+y.mid.plot.region <- ((par("usr")[3] + par("usr")[4]) / 2) # in x coordinates, right of the plot region (according to x scale)
+}
+if(any(sapply(FUN = all.equal, c(1, 3), x.side) == TRUE)){
+par(xpd=FALSE, xaxt="s")
+if(is.null(x.categ) & x.log.scale == TRUE){
+if(any(par()$xaxp[1:2] == 0)){ # any(sapply(FUN = all.equal, par()$xaxp[1:2], 0) == TRUE) not used because we strictly need zero as a result. Beware: write "== TRUE", because the result is otherwise character and a warning message appears using any()
+if(par()$xaxp[1] == 0){ # isTRUE(all.equal(par()$xaxp[1], 0)) not used because we strictly need zero as a result
+par(xaxp = c(10^-30, par()$xaxp[2:3])) # because log10(par()$xaxp[1] == 0) == -Inf
+}
+if(par()$xaxp[2] == 0){ # isTRUE(all.equal(par()$xaxp[1], 0)) not used because we strictly need zero as a result
+par(xaxp = c(par()$xaxp[1], 10^-30, par()$xaxp[3])) # because log10(par()$xaxp[2] == 0) == -Inf
+}
+}
+axis(side = x.side, at = c(10^par()$usr[1], 10^par()$usr[2]), labels=rep("", 2), lwd=1, lwd.ticks = 0) # draw the axis line
+mtext(side = x.side, text = x.lab, line = x.dist.legend / 0.2, las = 0, cex = x.label.size)
+par(tcl = -par()$mgp[2] * sec.tick.length) # length of the secondary ticks are reduced
+suppressWarnings(rug(10^outer(c((log10(par("xaxp")[1]) -1):log10(par("xaxp")[2])), log10(1:10), "+"), ticksize = NA, side = x.side)) # ticksize = NA to allow the use of par()$tcl value
+par(tcl = -par()$mgp[2] * tick.length) # back to main ticks
+axis(side = x.side, at = c(1e-15, 1e-14, 1e-13, 1e-12, 1e-11, 1e-10, 1e-9, 1e-8, 1e-7, 1e-6, 1e-5, 1e-4, 1e-3, 1e-2, 1e-1, 1e0, 1e1, 1e2, 1e3, 1e4, 1e5, 1e6, 1e7, 1e8, 1e9, 1e10), labels = expression(10^-15, 10^-14, 10^-13, 10^-12, 10^-11, 10^-10, 10^-9, 10^-8, 10^-7, 10^-6, 10^-5, 10^-4, 10^-3, 10^-2, 10^-1, 10^0, 10^1, 10^2, 10^3, 10^4, 10^5, 10^6, 10^7, 10^8, 10^9, 10^10), lwd = 0, lwd.ticks = 1, cex.axis = x.axis.size)
+x.text <- 10^par("usr")[2]
+}else if(is.null(x.categ) & x.log.scale == FALSE){
+axis(side=x.side, at=c(par()$usr[1], par()$usr[2]), labels=rep("", 2), lwd=1, lwd.ticks=0) # draw the axis line
+axis(side=x.side, at=round(seq(par()$xaxp[1], par()$xaxp[2], length.out=par()$xaxp[3]+1), 2), cex.axis = x.axis.size) # axis(side=x.side, at=round(seq(par()$xaxp[1], par()$xaxp[2], length.out=par()$xaxp[3]+1), 2), labels = format(round(seq(par()$xaxp[1], par()$xaxp[2], length.out=par()$xaxp[3]+1), 2), big.mark=','), cex.axis = x.axis.size) # to get the 1000 comma separator
+mtext(side = x.side, text = x.lab, line = x.dist.legend / 0.2, las = 0, cex = x.label.size)
+if(x.nb.inter.tick > 0){
+inter.tick.unit <- (par("xaxp")[2] - par("xaxp")[1]) / par("xaxp")[3]
+par(tcl = -par()$mgp[2] * sec.tick.length) # length of the ticks are reduced
+suppressWarnings(rug(seq(par("xaxp")[1] - 10 * inter.tick.unit, par("xaxp")[2] + 10 * inter.tick.unit, by = inter.tick.unit / (1 + x.nb.inter.tick)), ticksize = NA, x.side)) # ticksize = NA to allow the use of par()$tcl value
+par(tcl = -par()$mgp[2] * tick.length) # back to main ticks
+}
+x.text <- par("usr")[2]
+}else if(( ! is.null(x.categ)) & x.log.scale == FALSE){
+if(is.null(x.categ.pos)){
+x.categ.pos <- 1:length(x.categ)
+}else if(length(x.categ.pos) != length(x.categ)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": x.categ.pos MUST BE THE SAME LENGTH AS x.categ")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+par(xpd = TRUE)
+if(isTRUE(all.equal(x.side, 1))){ #isTRUE(all.equal(x.side, 1)) is similar to x.side == 1 but deals with float
+segments(x0 = x.left.plot.region, x1 = x.right.plot.region, y0 = y.bottom.plot.region, y1 = y.bottom.plot.region) # draw the line of the axis
+text(x = x.categ.pos, y = y.mid.bottom.fig.region, labels = x.categ, srt = text.angle, cex = x.axis.size)
+}else if(isTRUE(all.equal(x.side, 3))){ #isTRUE(all.equal(x.side, 1)) is similar to x.side == 3 but deals with float
+segments(x0 = x.left.plot.region, x1 = x.right.plot.region, y0 = y.top.plot.region, y1 = y.top.plot.region) # draw the line of the axis
+text(x = x.categ.pos, y = y.mid.top.fig.region, labels = x.categ, srt = text.angle, cex = x.axis.size)
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, ": ARGUMENT x.side CAN ONLY BE 1 OR 3")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+par(xpd = FALSE)
+x.text <- par("usr")[2]
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, ": PROBLEM WITH THE x.side (", x.side ,") OR x.log.scale (", x.log.scale,") ARGUMENTS")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}else{
+x.text <- par("usr")[2]
+}
+if(any(sapply(FUN = all.equal, c(2, 4), y.side) == TRUE)){
+par(xpd=FALSE, yaxt="s")
+if(is.null(y.categ) & y.log.scale == TRUE){
+if(any(par()$yaxp[1:2] == 0)){ # any(sapply(FUN = all.equal, par()$yaxp[1:2], 0) == TRUE) not used because we strictly need zero as a result. Beware: write "== TRUE", because the result is otherwise character and a warning message appears using any()
+if(par()$yaxp[1] == 0){ # strict zero needed
+par(yaxp = c(10^-30, par()$yaxp[2:3])) # because log10(par()$yaxp[1] == 0) == -Inf
+}
+if(par()$yaxp[2] == 0){ # strict zero needed
+par(yaxp = c(par()$yaxp[1], 10^-30, par()$yaxp[3])) # because log10(par()$yaxp[2] == 0) == -Inf
+}
+}
+axis(side=y.side, at=c(10^par()$usr[3], 10^par()$usr[4]), labels=rep("", 2), lwd=1, lwd.ticks=0) # draw the axis line
+par(tcl = -par()$mgp[2] * sec.tick.length) # length of the ticks are reduced
+suppressWarnings(rug(10^outer(c((log10(par("yaxp")[1])-1):log10(par("yaxp")[2])), log10(1:10), "+"), ticksize = NA, side = y.side)) # ticksize = NA to allow the use of par()$tcl value
+par(tcl = -par()$mgp[2] * tick.length) # back to main tick length
+axis(side = y.side, at = c(1e-15, 1e-14, 1e-13, 1e-12, 1e-11, 1e-10, 1e-9, 1e-8, 1e-7, 1e-6, 1e-5, 1e-4, 1e-3, 1e-2, 1e-1, 1e0, 1e1, 1e2, 1e3, 1e4, 1e5, 1e6, 1e7, 1e8, 1e9, 1e10), labels = expression(10^-15, 10^-14, 10^-13, 10^-12, 10^-11, 10^-10, 10^-9, 10^-8, 10^-7, 10^-6, 10^-5, 10^-4, 10^-3, 10^-2, 10^-1, 10^0, 10^1, 10^2, 10^3, 10^4, 10^5, 10^6, 10^7, 10^8, 10^9, 10^10), lwd = 0, lwd.ticks = 1, cex.axis = y.axis.size)
+y.text <- 10^(par("usr")[4] + (par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3]) * (1 - par("plt")[4]))
+mtext(side = y.side, text = y.lab, line = y.dist.legend / 0.2, las = 0, cex = y.label.size)
+}else if(is.null(y.categ) & y.log.scale == FALSE){
+axis(side=y.side, at=c(par()$usr[3], par()$usr[4]), labels=rep("", 2), lwd=1, lwd.ticks=0) # draw the axis line
+axis(side=y.side, at=round(seq(par()$yaxp[1], par()$yaxp[2], length.out=par()$yaxp[3]+1), 2), cex.axis = y.axis.size)
+mtext(side = y.side, text = y.lab, line = y.dist.legend / 0.2, las = 0, cex = y.label.size)
+if(y.nb.inter.tick > 0){
+inter.tick.unit <- (par("yaxp")[2] - par("yaxp")[1]) / par("yaxp")[3]
+par(tcl = -par()$mgp[2] * sec.tick.length) # length of the ticks are reduced
+suppressWarnings(rug(seq(par("yaxp")[1] - 10 * inter.tick.unit, par("yaxp")[2] + 10 * inter.tick.unit, by = inter.tick.unit / (1 + y.nb.inter.tick)), ticksize = NA, side=y.side)) # ticksize = NA to allow the use of par()$tcl value
+par(tcl = -par()$mgp[2] * tick.length) # back to main tick length
+}
+y.text <- (par("usr")[4] + (par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3]) * (1 - par("plt")[4]))
+}else if(( ! is.null(y.categ)) & y.log.scale == FALSE){
+if(is.null(y.categ.pos)){
+y.categ.pos <- 1:length(y.categ)
+}else if(length(y.categ.pos) != length(y.categ)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": y.categ.pos MUST BE THE SAME LENGTH AS y.categ")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+axis(side = y.side, at = y.categ.pos, labels = rep("", length(y.categ)), lwd=0, lwd.ticks=1) # draw the line of the axis
+par(xpd = TRUE)
+if(isTRUE(all.equal(y.side, 2))){ #isTRUE(all.equal(y.side, 2)) is similar to y.side == 2 but deals with float
+text(x = x.mid.left.fig.region, y = y.categ.pos, labels = y.categ, srt = text.angle, cex = y.axis.size)
+}else if(isTRUE(all.equal(y.side, 4))){ # idem
+text(x = x.mid.right.fig.region, y = y.categ.pos, labels = y.categ, srt = text.angle, cex = y.axis.size)
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, ": ARGUMENT y.side CAN ONLY BE 2 OR 4")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+par(xpd = FALSE)
+y.text <- (par("usr")[4] + (par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3]) * (1 - par("plt")[4]))
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, ": PROBLEM WITH THE y.side (", y.side ,") OR y.log.scale (", y.log.scale,") ARGUMENTS")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}else{
+y.text <- (par("usr")[4] + (par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3]) * (1 - par("plt")[4]))
+}
+par(xpd=NA)
+text(x = x.mid.right.fig.region, y = y.text, corner.text, adj=c(1, 1.1), cex = corner.text.size) # text at the topright corner. Replace x.right.fig.region by x.text if text at the right edge of the plot region
+if(just.label.add == TRUE & isTRUE(all.equal(x.side, 0)) & x.lab != ""){
+text(x = x.mid.plot.region, y = y.mid.bottom.fig.region, x.lab, adj=c(0.5, 0.5), cex = x.label.size) # x label
+}
+if(just.label.add == TRUE & isTRUE(all.equal(y.side, 0)) & y.lab != ""){
+text(x = y.mid.plot.region, y = x.mid.left.fig.region, y.lab, adj=c(0.5, 0.5), cex = y.label.size) # x label
+}
+par(xpd=FALSE)
+if(par.reset == TRUE){
+tempo.par <- fun_open(pdf = FALSE, return.output = TRUE)
+invisible(dev.off()) # close the new window
+if( ! is.null(custom.par)){
+if( ! names(custom.par) %in% names(tempo.par$ini.par)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": custom.par ARGUMENT SHOULD HAVE THE NAMES OF THE COMPARTMENT LIST COMING FROM THE par() LIST")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+par(custom.par)
+text <- c(text, "\nGRAPH PARAMETERS SET TO VALUES DEFINED BY custom.par ARGUMENT\n")
+}else{
+par(tempo.par$ini.par)
+text <- c(text, "\nGRAPH PARAMETERS RESET TO par() DEFAULT VALUES\n")
+}
+}
+output <- list(x.mid.left.dev.region = x.mid.left.dev.region, x.left.dev.region = x.left.dev.region, x.mid.right.dev.region = x.mid.right.dev.region, x.right.dev.region = x.right.dev.region, x.mid.left.fig.region = x.mid.left.fig.region, x.left.fig.region = x.left.fig.region, x.mid.right.fig.region = x.mid.right.fig.region, x.right.fig.region = x.right.fig.region, x.left.plot.region = x.left.plot.region, x.right.plot.region = x.right.plot.region, x.mid.plot.region = x.mid.plot.region, y.mid.bottom.dev.region = y.mid.bottom.dev.region, y.bottom.dev.region = y.bottom.dev.region, y.mid.top.dev.region = y.mid.top.dev.region, y.top.dev.region = y.top.dev.region, y.mid.bottom.fig.region = y.mid.bottom.fig.region, y.bottom.fig.region = y.bottom.fig.region, y.mid.top.fig.region = y.mid.top.fig.region, y.top.fig.region = y.top.fig.region, y.top.plot.region = y.top.plot.region, y.bottom.plot.region = y.bottom.plot.region, y.mid.plot.region = y.mid.plot.region, text = text)
+return(output)
+}
+
+
+######## fun_close() #### close specific graphic windows
+
+
+# Check OK: clear to go Apollo
+fun_close <- function(kind = "pdf", return.text = FALSE){
+# AIM
+# close only specific graphic windows (devices)
+# ARGUMENTS:
+# kind: vector, among c("windows", "quartz", "x11", "X11", "pdf", "bmp", "png", "tiff"), indicating the kind of graphic windows (devices) to close. BEWARE: either "windows", "quartz", "x11" or "X11" means that all the X11 GUI graphics devices will be closed, whatever the OS used
+# return.text: print text regarding the kind parameter and the devices that were finally closed?
+# RETURN
+# text regarding the kind parameter and the devices that were finally closed
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# windows() ; windows() ; pdf() ; dev.list() ; fun_close(kind = c("pdf", "x11"), return.text = TRUE) ; dev.list()
+# DEBUGGING
+# kind = c("windows", "pdf") ; return.text = FALSE # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = kind, options = c("windows", "quartz", "x11", "X11", "pdf", "bmp", "png", "tiff"), fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = return.text, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+text <- paste0("THE REQUIRED KIND OF GRAPHIC DEVICES TO CLOSE ARE ", paste(kind, collapse = " "))
+if(Sys.info()["sysname"] == "Windows"){ # Note that .Platform$OS.type() only says "unix" for macOS and Linux and "Windows" for Windows
+if(any(kind %in% c("windows", "quartz", "x11", "X11"))){
+tempo <- kind %in% c("windows", "quartz", "x11", "X11")
+kind[tempo] <- "windows" # term are replaced by what is displayed when using a <- dev.list() ; names(a)
+}
+}else if(Sys.info()["sysname"] == "Linux"){
+if(any(kind %in% c("windows", "quartz", "x11", "X11"))){
+tempo.device <- suppressWarnings(try(X11(), silent = TRUE))[] # open a X11 window to try to recover the X11 system used
+if( ! is.null(tempo.device)){
+text <- paste0(text, "\nCANNOT CLOSE GUI GRAPHIC DEVICES AS REQUIRED BECAUSE THIS LINUX SYSTEM DOES NOT HAVE IT")
+}else{
+tempo <- kind %in% c("windows", "quartz", "x11", "X11")
+kind[tempo] <- names(dev.list()[length(dev.list())]) # term are replaced by what is displayed when using a <- dev.list() ; names(a)
+invisible(dev.off()) # close the X11 opened by tempo
+}
+}
+}else{ # for macOS
+if(any(kind %in% c("windows", "quartz", "x11", "X11"))){
+tempo <- kind %in% c("windows", "quartz", "x11", "X11")
+kind[tempo] <- "quartz" # term are replaced by what is displayed when using a <- dev.list() ; names(a)
+}
+}
+kind <- unique(kind)
+if(length(dev.list()) != 0){
+for(i in length(names(dev.list())):1){
+if(names(dev.list())[i] %in% kind){
+text <- paste0(text, "\n", names(dev.list())[i], " DEVICE NUMBER ", dev.list()[i], " HAS BEEN CLOSED")
+invisible(dev.off(dev.list()[i]))
+}
+}
+}
+if(return.text == TRUE){
+return(text)
+}
+}
+
+
+################ Standard graphics
+
+
+######## fun_empty_graph() #### text to display for empty graphs
+
+
+
+
+
+# Check OK: clear to go Apollo
+fun_empty_graph <- function(
+text = NULL,
+text.size = 1,
+title = NULL,
+title.size = 1.5
+){
+# AIM
+# display an empty plot with a text in the middle of the window (for instance to specify that no plot can be drawn)
+# ARGUMENTS
+# text: character string of the message to display
+# text.size: numeric value of the text size
+# title: character string of the graph title
+# title.size: numeric value of the title size (in points)
+# RETURN
+# an empty plot
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# simple example
+# fun_empty_graph(text = "NO GRAPH")
+# white page
+# fun_empty_graph() # white page
+# all the arguments
+# fun_empty_graph(text = "NO GRAPH", text.size = 2, title = "GRAPH1", title.size = 1)
+# DEBUGGING
+# text = "NO GRAPH" ; title = "GRAPH1" ; text.size = 1
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+if( ! is.null(text)){
+tempo <- fun_check(data = text, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = text.size, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(title)){
+tempo <- fun_check(data = title, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = title.size, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+ini.par <- par(no.readonly = TRUE) # to recover the initial graphical parameters if required (reset). BEWARE: this command alone opens a pdf of GUI window if no window already opened. But here, protected with the code because always a tempo window opened
+par(ann=FALSE, xaxt="n", yaxt="n", mar = rep(1, 4), bty = "n", xpd = NA)
+plot(1, 1, type = "n") # no display with type = "n"
+x.left.dev.region <- (par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] - ((par("usr")[2] - par("usr")[1]) / ((par("omd")[2] - par("omd")[1]) * (par("plt")[2] - par("plt")[1]))) * par("omd")[1])
+y.top.dev.region <- (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / ((par("omd")[4] - par("omd")[3]) * (par("plt")[4] - par("plt")[3]))) * (1 - par("omd")[4]))
+if( ! is.null(text)){
+text(x = 1, y = 1, labels = text, cex = text.size)
+}
+if( ! is.null(title)){
+text(x = x.left.dev.region, y = y.top.dev.region, labels = title, adj=c(0, 1), cex = title.size)
+}
+par(ini.par)
+}
+
+
+################ gg graphics
+
+
+######## fun_gg_palette() #### ggplot2 default color palette
+
+
+
+
+
+# Check OK: clear to go Apollo
+fun_gg_palette <- function(n, kind = "std"){
+# AIM
+# provide colors used by ggplot2
+# the interest is to use another single color that is not the red one used by default
+# for ggplot2 specifications, see: https://ggplot2.tidyverse.org/articles/ggplot2-specs.html
+# ARGUMENTS
+# n: number of groups on the graph
+# kind: either "std" for standard gg colors, "dark" for darkened gg colors, or "light" for pastel gg colors
+# RETURN
+# the vector of hexadecimal colors
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# output of the function
+# fun_gg_palette(n = 2)
+# the ggplot2 palette when asking for 7 different colors
+# plot(1:7, pch = 16, cex = 5, col = fun_gg_palette(n = 7))
+# selection of the 5th color of the ggplot2 palette made of 7 different colors
+# plot(1:7, pch = 16, cex = 5, col = fun_gg_palette(n = 7)[5])
+# the ggplot2 palette made of 7 darkened colors
+# plot(1:7, pch = 16, cex = 5, col = fun_gg_palette(n = 7, kind = "dark"))
+# the ggplot2 palette made of 7 lighten colors
+# plot(1:7, pch = 16, cex = 5, col = fun_gg_palette(n = 7, kind = "light"))
+# DEBUGGING
+# n = 0
+# kind = "std"
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = n, class = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & isTRUE(all.equal(n, 0))){ # isTRUE(all.equal(n, 0))) is similar to n == 0 but deals with float
+tempo.cat <- paste0("ERROR IN ", function.name, ": n ARGUMENT MUST BE A NON ZERO INTEGER. HERE IT IS: ", paste(n, collapse = " "))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+tempo <- fun_check(data = kind, options = c("std", "dark", "light"), length = 1, fun.name = function.name) ; eval(ee)
+}
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+hues = seq(15, 375, length = n + 1)
+hcl(h = hues, l = if(kind == "std"){65}else if(kind == "dark"){35}else if(kind == "light"){85}, c = 100)[1:n]
+}
+
+
+######## fun_gg_just() #### ggplot2 justification of the axis labeling, depending on angle
+
+
+
+
+
+# Check OK: clear to go Apollo
+fun_gg_just <- function(angle, pos, kind = "axis"){
+# AIM
+# provide correct justification for text labeling, depending on the chosen angle
+# WARNINGS
+# justification behave differently on plot, depending whether it is used for annotayed text or for axis labelling. Indeed the latter has labelling constrained
+# Of note, a bug in ggplot2: vjust sometimes does not work, i.e., the same justification result is obtained whatever the value used. This is the case with angle = 90, pos = "top", kind = "axis". While everything is fine with angle = 90, pos = "bottom", kind = "axis". At least, everything seems fine for kind = "axis" and pos = c("left", "bottom")
+# ARGUMENTS
+# angle: integer value of the text angle, using the same rules as in ggplot2. Positive values for counterclockwise rotation: 0 for horizontal, 90 for vertical, 180 for upside down etc. Negative values for clockwise rotation: 0 for horizontal, -90 for vertical, -180 for upside down etc.
+# pos: where text is? Either "top", "right", "bottom" or "left" of the elements to justify from
+# kind: kind of text? Either "axis" or "text". In the first case, the pos argument refers to the axis position, and in the second to annotated text (using ggplot2::annotate() or ggplot2::geom_text())
+# RETURN
+# a list containing:
+# $angle: the submitted angle (value potentially reduced to fit the [-360 ; 360] interval, e.g., 460 -> 100, without impact on the final angle displayed)
+# $pos: the selected position (argument pos)
+# $kind: the selected kind of text (argument kind)
+# $hjust: the horizontal justification
+# $vjust: the vertical justification
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# fun_gg_just(angle = 45, pos = "bottom")
+# fun_gg_just(angle = (360*2 + 45), pos = "left")
+# output <- fun_gg_just(angle = 45, pos = "bottom") ; obs1 <- data.frame(time = 1:20, group = rep(c("CLASS_1", "CLASS_2"), times = 10), stringsAsFactors = TRUE) ; ggplot2::ggplot() + ggplot2::geom_bar(data = obs1, mapping = ggplot2::aes(x = group, y = time), stat = "identity") + ggplot2::theme(axis.text.x = ggplot2::element_text(angle = output$angle, hjust = output$hjust, vjust = output$vjust))
+# output <- fun_gg_just(angle = -45, pos = "left") ; obs1 <- data.frame(time = 1:20, group = rep(c("CLASS_1", "CLASS_2"), times = 10), stringsAsFactors = TRUE) ; ggplot2::ggplot() + ggplot2::geom_bar(data = obs1, mapping = ggplot2::aes(x = group, y = time), stat = "identity") + ggplot2::theme(axis.text.y = ggplot2::element_text(angle = output$angle, hjust = output$hjust, vjust = output$vjust)) + ggplot2::coord_flip()
+# output1 <- fun_gg_just(angle = 90, pos = "bottom") ; output2 <- fun_gg_just(angle = -45, pos = "left") ; obs1 <- data.frame(time = 1:20, group = rep(c("CLASS_1", "CLASS_2"), times = 10), stringsAsFactors = TRUE) ; ggplot2::ggplot() + ggplot2::geom_bar(data = obs1, mapping = ggplot2::aes(x = group, y = time), stat = "identity") + ggplot2::theme(axis.text.x = ggplot2::element_text(angle = output1$angle, hjust = output1$hjust, vjust = output1$vjust), axis.text.y = ggplot2::element_text(angle = output2$angle, hjust = output2$hjust, vjust = output2$vjust))
+# output <- fun_gg_just(angle = -45, pos = "left") ; obs1 <- data.frame(time = 1, km = 1, bird = "pigeon", stringsAsFactors = FALSE) ; ggplot2::ggplot(data = obs1, mapping = ggplot2::aes(x = time, y = km)) + ggplot2::geom_point() + ggplot2::geom_text(mapping = ggplot2::aes(label = bird), angle = output$angle, hjust = output$hjust, vjust = output$vjust)
+# obs1 <- data.frame(time = 1:10, km = 1:10, bird = c(NA, NA, NA, "pigeon", NA, "cat", NA, NA, NA, NA), stringsAsFactors = FALSE) ; fun_open(width = 4, height = 4) ; for(i0 in c("text", "axis")){for(i1 in c("top", "right", "bottom", "left")){for(i2 in c(0, 45, 90, 135, 180, 225, 270, 315, 360)){output <- fun_gg_just(angle = i2, pos = i1, kind = i0) ; title <- paste0("kind: ", i0, " | pos: ", i1, " | angle = ", i2, " | hjust: ", output$hjust, " | vjust: ", output$vjust) ; if(i0 == "text"){print(ggplot2::ggplot(data = obs1, mapping = ggplot2::aes(x = time, y = km)) + ggplot2::geom_point(color = fun_gg_palette(1), alpha = 0.5) + ggplot2::ggtitle(title) + ggplot2::geom_text(mapping = ggplot2::aes(label = bird), angle = output$angle, hjust = output$hjust, vjust = output$vjust) + ggplot2::theme(title = ggplot2::element_text(size = 5)))}else{print(ggplot2::ggplot(data = obs1, mapping = ggplot2::aes(x = time, y = km)) + ggplot2::geom_point(color = fun_gg_palette(1), alpha = 0.5) + ggplot2::ggtitle(title) + ggplot2::geom_text(mapping = ggplot2::aes(label = bird)) + ggplot2::scale_x_continuous(position = ifelse(i1 == "top", "top", "bottom")) + ggplot2::scale_y_continuous(position = ifelse(i1 == "right", "right", "left")) + ggplot2::theme(title = ggplot2::element_text(size = 5), axis.text.x = if(i1 %in% c("top", "bottom")){ggplot2::element_text(angle = output$angle, hjust = output$hjust, vjust = output$vjust)}, axis.text.y = if(i1 %in% c("right", "left")){ggplot2::element_text(angle = output$angle, hjust = output$hjust, vjust = output$vjust)}))}}}} ; dev.off()
+# DEBUGGING
+# angle = 45 ; pos = "left" ; kind = "axis"
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+arg.user.setting <- as.list(match.call(expand.dots = FALSE))[-1] # list of the argument settings (excluding default values not provided by the user)
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument primary checking
+# arg with no default values
+if(any(missing(angle) | missing(pos))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ARGUMENTS angle AND pos HAVE NO DEFAULT VALUE AND REQUIRE ONE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end arg with no default values
+# using fun_check()
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = angle, class = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = pos, options = c("left", "top", "right", "bottom"), length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = kind, options = c("axis", "text"), length = 1, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# end using fun_check()
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument primary checking
+# second round of checking and data preparation
+# dealing with NA arguments
+tempo.arg <- names(arg.user.setting) # values provided by the user
+tempo.log <- sapply(lapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = is.na), FUN = any) & lapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = length) == 1 # no argument provided by the user can be just NA
+if(any(tempo.log) == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ":\n", ifelse(sum(tempo.log, na.rm = TRUE) > 1, "THESE ARGUMENTS\n", "THIS ARGUMENT\n"), paste0(tempo.arg[tempo.log], collapse = "\n"),"\nCANNOT JUST BE NA")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end dealing with NA arguments
+# dealing with NULL arguments
+tempo.arg <- c(
+"angle",
+"pos",
+"kind"
+)
+tempo.log <- sapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = is.null)
+if(any(tempo.log) == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ":\n", ifelse(sum(tempo.log, na.rm = TRUE) > 1, "THESE ARGUMENTS\n", "THIS ARGUMENT\n"), paste0(tempo.arg[tempo.log], collapse = "\n"),"\nCANNOT BE NULL")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end dealing with NULL arguments
+# end second round of checking and data preparation
+# main code
+# to get angle between -360 and 360
+while(angle > 360){
+angle <- angle - 360
+}
+while(angle < -360){
+angle <- angle + 360
+}
+# end to get angle between -360 and 360
+# justifications
+if(pos %in% c("bottom", "top")){
+# code below is for if(pos == "bottom"){
+if(any(sapply(FUN = all.equal, c(-360, -180, 0, 180, 360), angle) == TRUE)){ # equivalent of angle == -360 | angle == -180 | angle == 0 | angle == 180 | angle == 360 but deals with floats
+hjust <- 0.5
+if(kind == "text"){
+if(any(sapply(FUN = all.equal, c(-360, 0, 360), angle) == TRUE)){
+vjust <- 1
+}else if(any(sapply(FUN = all.equal, c(-180, 180), angle) == TRUE)){
+vjust <- 0
+}
+}else{
+vjust <- 0.5
+}
+}else if(any(sapply(FUN = all.equal, c(-270, 90), angle) == TRUE)){
+hjust <- 1
+vjust <- 0.5
+}else if(any(sapply(FUN = all.equal, c(-90, 270), angle) == TRUE)){
+hjust <- 0
+vjust <- 0.5
+}else if((angle > -360 & angle < -270) | (angle > 0 & angle < 90)){
+hjust <- 1
+vjust <- 1
+}else if((angle > -270 & angle < -180) | (angle > 90 & angle < 180)){
+hjust <- 1
+vjust <- 0
+}else if((angle > -180 & angle < -90) | (angle > 180 & angle < 270)){
+hjust <- 0
+vjust <- 0
+if(kind == "text" & pos == "top"){
+hjust <- 1
+}
+}else if((angle > -90 & angle < 0) | (angle > 270 & angle < 360)){
+hjust <- 0
+vjust <- 1
+}
+if(pos == "top"){
+if( ! ((angle > -180 & angle < -90) | (angle > 180 & angle < 270))){
+hjust <- 1 - hjust
+}
+vjust <- 1 - vjust
+}
+}else if(pos %in% c("left", "right")){
+# code below is for if(pos == "left"){
+if(any(sapply(FUN = all.equal, c(-270, -90, 90, 270), angle) == TRUE)){ # equivalent of angle == -270 | angle == -90 | angle == 90 | angle == 270 but deals with floats
+hjust <- 0.5
+if(kind == "text"){
+if(any(sapply(FUN = all.equal, c(-90, 90), angle) == TRUE)){
+vjust <- 0
+}else if(any(sapply(FUN = all.equal, c(-270, 270), angle) == TRUE)){
+vjust <- 1
+}
+}else{
+vjust <- 0.5
+}
+}else if(any(sapply(FUN = all.equal, c(-360, 0, 360), angle) == TRUE)){
+hjust <- 1
+vjust <- 0.5
+}else if(any(sapply(FUN = all.equal, c(-180, 180), angle) == TRUE)){
+hjust <- 0
+vjust <- 0.5
+}else if((angle > -360 & angle < -270) | (angle > 0 & angle < 90)){
+hjust <- 1
+vjust <- 0
+}else if((angle > -270 & angle < -180) | (angle > 90 & angle < 180)){
+hjust <- 0
+vjust <- 0
+}else if((angle > -180 & angle < -90) | (angle > 180 & angle < 270)){
+hjust <- 0
+vjust <- 1
+}else if((angle > -90 & angle < 0) | (angle > 270 & angle < 360)){
+hjust <- 1
+vjust <- 1
+}
+if(pos == "right"){
+hjust <- 1 - hjust
+if( ! (((angle > -270 & angle < -180) | (angle > 90 & angle < 180)) | ((angle > -180 & angle < -90) | (angle > 180 & angle < 270)))){
+vjust <- 1 - vjust
+}
+}
+}
+# end justifications
+output <- list(angle = angle, pos = pos, kind = kind, hjust = hjust, vjust = vjust)
+return(output)
+}
+
+
+######## fun_gg_get_legend() #### get the legend of ggplot objects
+
+
+
+
+
+# Check OK: clear to go Apollo
+fun_gg_get_legend <- function(ggplot_built, fun.name = NULL, lib.path = NULL){
+# AIM
+# get legend of ggplot objects
+# # from https://stackoverflow.com/questions/12539348/ggplot-separate-legend-and-plot
+# ARGUMENTS
+# ggplot_built: a ggplot build object
+# fun.name: single character string indicating the name of the function using fun_gg_get_legend() for warning and error messages. Ignored if NULL
+# lib.path: character vector specifying the absolute pathways of the directories containing the required packages if not in the default directories. Ignored if NULL
+# RETURN
+# a list of class c("gtable", "gTree", "grob", "gDesc"), providing legend information of ggplot_built objet, or NULL if the ggplot_built object has no legend
+# REQUIRED PACKAGES
+# ggplot2
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# fun_pack()
+# EXAMPLES
+# Simple example
+# obs1 <- data.frame(time = 1:20, group = rep(c("CLASS_1", "CLASS_2"), times = 10), stringsAsFactors = TRUE) ; p <- ggplot2::ggplot() + ggplot2::geom_point(data = obs1, mapping = ggplot2::aes(x = group, y = time, fill = group)) ; fun_gg_get_legend(ggplot_built = ggplot2::ggplot_build(p))
+# Error message because no legend in the ggplot
+# obs1 <- data.frame(time = 1:20, group = rep(c("CLASS_1", "CLASS_2"), times = 10), stringsAsFactors = TRUE) ; p <- ggplot2::ggplot() + ggplot2::geom_point(data = obs1, mapping = ggplot2::aes(x = group, y = time)) ; fun_gg_get_legend(ggplot_built = ggplot2::ggplot_build(p))
+# DEBUGGING
+# obs1 <- data.frame(time = 1:20, group = rep(c("CLASS_1", "CLASS_2"), times = 10), stringsAsFactors = TRUE) ; p <- ggplot2::ggplot() + ggplot2::geom_point(data = obs1, mapping = ggplot2::aes(x = group, y = time)) ; ggplot_built = ggplot2::ggplot_build(p) ; fun.name = NULL ; lib.path = NULL
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+req.function <- c(
+"fun_check",
+"fun_pack"
+)
+for(i1 in req.function){
+if(length(find(i1, mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED ", i1, "() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# end required function checking
+# argument primary checking
+# arg with no default values
+if(any(missing(ggplot_built))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ARGUMENTS ggplot_built HAVE NO DEFAULT VALUE AND REQUIRE ONE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end arg with no default values
+# using fun_check()
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = ggplot_built, class = "ggplot_built", mode = "list", fun.name = function.name) ; eval(ee)
+if( ! is.null(fun.name)){
+tempo <- fun_check(data = fun.name, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+}
+if( ! is.null(lib.path)){
+tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+}
+if( ! is.null(arg.check)){
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+}
+# end using fun_check()
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument primary checking
+# second round of checking
+# dealing with NA
+if(any(is.na(ggplot_built)) | any(is.na(fun.name)) | any(is.na(lib.path))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": NO ARGUMENT CAN HAVE NA VALUES")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end dealing with NA
+# dealing with NULL
+if(is.null(ggplot_built)){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nggplot_built ARGUMENT CANNOT BE NULL")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end dealing with NULL
+if( ! is.null(lib.path)){
+if( ! all(dir.exists(lib.path))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# end second round of checking
+# package checking
+fun_pack(req.package = c("ggplot2"), lib.path = lib.path)
+# end package checking
+# main code
+win.nb <- dev.cur()
+pdf(file = NULL)
+tmp <- ggplot2::ggplot_gtable(ggplot_built)
+# BEWARE with ggplot_gtable : open a blanck device https://stackoverflow.com/questions/17012518/why-does-this-r-ggplot2-code-bring-up-a-blank-display-device
+invisible(dev.off())
+if(win.nb > 1){ # to go back to the previous active device, if == 1 means no opened device
+dev.set(win.nb)
+}
+leg <- which(sapply(tmp$grobs, function(x) x$name) == "guide-box")
+if(length(leg) == 0){
+legend <- NULL
+}else{
+legend <- tmp$grobs[[leg]]
+}
+return(legend)
+}
+
+
+######## fun_gg_point_rast() #### ggplot2 raster scatterplot layer
+
+
+
+
+
+# Check OK: clear to go Apollo
+fun_gg_point_rast <- function(
+data = NULL,
+mapping = NULL,
+stat = "identity",
+position = "identity",
+...,
+na.rm = FALSE,
+show.legend = NA,
+inherit.aes = TRUE,
+raster.width = NULL,
+raster.height = NULL,
+raster.dpi = 300,
+inactivate = TRUE,
+lib.path = NULL
+){
+# AIM
+# equivalent to ggplot2::geom_point() but in raster mode
+# use it like ggplot2::geom_point() with the main raster.dpi additional argument
+# WARNINGS
+# can be long to generate the plot
+# use a square plot region. Otherwise, the dots will have ellipsoid shape
+# solve the transparency problems with some GUI
+# this function is derived from the geom_point_rast() function, created by Viktor Petukhov , and present in the ggrastr package (https://rdrr.io/github/VPetukhov/ggrastr/src/R/geom-point-rast.R, MIT License, Copyright (c) 2017 Viktor Petukhov). Has been placed here to minimize package dependencies
+# ARGUMENTS
+# classical arguments of geom_point(), shown here https://rdrr.io/github/VPetukhov/ggrastr/man/geom_point_rast.html
+# raster.width : width of the result image (in inches). Default: deterined by the current device parameters
+# raster.height: height of the result image (in inches). Default: deterined by the current device parameters
+# raster.dpi: resolution of the result image
+# inactivate: logical. Inactivate the fun.name argument of the fun_check() function? If TRUE, the name of the fun_check() function in error messages coming from this function. Use TRUE if fun_gg_point_rast() is used like this: eval(parse(text = "fun_gg_point_rast"))
+# lib.path: character vector specifying the absolute pathways of the directories containing the required packages if not in the default directories. Ignored if NULL
+# RETURN
+# a raster scatter plot
+# REQUIRED PACKAGES
+# ggplot2
+# grid (included in the R installation packages but not automatically loaded)
+# Cairo
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# fun_pack()
+# EXAMPLES
+# Two pdf in the current directory
+# set.seed(1) ; data1 = data.frame(x = rnorm(100000), y = rnorm(10000), stringsAsFactors = TRUE) ; fun_open(pdf.name = "Raster") ; ggplot2::ggplot() + fun_gg_point_rast(data = data1, mapping = ggplot2::aes(x = x, y = y)) ; fun_open(pdf.name = "Vectorial") ; ggplot2::ggplot() + ggplot2::geom_point(data = data1, mapping = ggplot2::aes(x = x, y = y)) ; dev.off() ; dev.off()
+# DEBUGGING
+#
+# function name
+if(all(inactivate == FALSE)){ # inactivate has to be used here but will be fully checked below
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+}else if(all(inactivate == TRUE)){
+function.name <- NULL
+}else{
+tempo.cat <- paste0("ERROR IN fun_gg_point_rast(): CODE INCONSISTENCY 1")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(length(utils::find("fun_pack", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_pack() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+if( ! is.null(data)){
+tempo <- fun_check(data = data, class = "data.frame", na.contain = TRUE, fun.name = function.name) ; eval(ee)
+}
+if( ! is.null(mapping)){
+tempo <- fun_check(data = mapping, class = "uneval", typeof = "list", fun.name = function.name) ; eval(ee) # aes() is tested
+}
+# stat and position not tested because too complicate
+tempo <- fun_check(data = na.rm, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = show.legend, class = "vector", mode = "logical", length = 1, na.contain = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = inherit.aes, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(raster.width)){
+tempo <- fun_check(data = raster.width, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+}
+if( ! is.null(raster.height)){
+tempo <- fun_check(data = raster.height, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = raster.dpi, class = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = inactivate, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(lib.path)){
+tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
+tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+}
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# package checking
+fun_pack(req.package = c("ggplot2"), lib.path = lib.path)
+fun_pack(req.package = c("grid"), lib.path = lib.path)
+fun_pack(req.package = c("Cairo"), lib.path = lib.path)
+# end package checking
+# additional functions
+DrawGeomPointRast <- function(data, panel_params, coord, na.rm = FALSE, raster.width = NULL, raster.height= NULL, raster.dpi = 300){
+if (is.null(raster.width)){
+raster.width <- par('fin')[1]
+}
+if (is.null(raster.height)){
+ raster.height <- par('fin')[2]
+}
+prev_dev_id <- dev.cur()
+p <- ggplot2::GeomPoint$draw_panel(data, panel_params, coord)
+dev_id <- Cairo::Cairo(type='raster', width = raster.width*raster.dpi, height = raster.height*raster.dpi, dpi = raster.dpi, units = 'px', bg = "transparent")[1]
+grid::pushViewport(grid::viewport(width = 1, height = 1))
+grid::grid.points(x = p$x, y = p$y, pch = p$pch, size = p$size,
+name = p$name, gp = p$gp, vp = p$vp, draw = T)
+grid::popViewport()
+cap <- grid::grid.cap()
+invisible(dev.off(dev_id))
+invisible(dev.set(prev_dev_id))
+grid::rasterGrob(cap, x = 0, y = 0, width = 1, height = 1, default.units = "native", just = c("left","bottom"))
+}
+# end additional functions
+# main code
+GeomPointRast <- ggplot2::ggproto("GeomPointRast", ggplot2::GeomPoint, draw_panel = DrawGeomPointRast)
+ggplot2::layer(
+data = data,
+mapping = mapping,
+stat = stat,
+geom = GeomPointRast,
+position = position,
+show.legend = show.legend,
+inherit.aes = inherit.aes,
+params = list(
+na.rm = na.rm,
+raster.width = raster.width,
+raster.height = raster.height,
+raster.dpi = raster.dpi,
+...
+)
+)
+# end main code
+}
+
+
+######## fun_gg_boxplot() #### ggplot2 boxplot + background dots if required
+
+
+
+
+######## fun_gg_scatter() #### ggplot2 scatterplot + lines (up to 6 overlays totally)
+
+
+
+
+######## fun_gg_heatmap() #### ggplot2 heatmap + overlaid mask if required
+
+
+#test plot.margin = margin(up.space.mds, right.space.mds, down.space.mds, left.space.mds, "inches") to set the dim of the region plot ?
+# if matrix is full of zero (or same value I guess), heatmap is complicate. Test it and error message
+
+# Check OK: clear to go Apollo
+fun_gg_heatmap <- function(
+data1,
+legend.name1 = "",
+low.color1 = "blue",
+mid.color1 = "white",
+high.color1 = "red",
+limit1 = NULL,
+midpoint1 = NULL,
+data2 = NULL,
+color2 = "black",
+alpha2 = 0.5,
+invert2 = FALSE,
+text.size = 12,
+title = "",
+title.text.size = 12,
+show.scale = TRUE,
+rotate = FALSE,
+return = FALSE,
+plot = TRUE,
+add = NULL,
+warn.print = FALSE,
+lib.path = NULL
+){
+# AIM
+# ggplot2 heatmap with the possibility to overlay a mask
+# see also:
+# draw : http://www.sthda.com/english/wiki/ggplot2-quick-correlation-matrix-heatmap-r-software-and-data-visualization
+# same range scale : https://stackoverflow.com/questions/44655723/r-ggplot2-heatmap-fixed-scale-color-between-graphs
+# for ggplot2 specifications, see: https://ggplot2.tidyverse.org/articles/ggplot2-specs.html
+# ARGUMENTS
+# data1: numeric matrix or data frame resulting from the conversion of the numeric matrix by reshape2::melt()
+# legend.name1: character string of the data1 heatmap scale legend
+# low.color1: character string of the color (i.e., "blue" or "#0000FF") of the lowest scale value
+# mid.color1: same as low.color1 but for the middle scale value. If NULL, the middle color is the default color between low.color1 and high.color1. BEWARE: argument midpoint1 is not ignored, even if mid.color1 is NULL, meaning that the default mid color can still be controled
+# high.color1: same as low.color1 but for the highest scale value
+# limit1: 2 numeric values defining the lowest and higest color scale values. If NULL, take the range of data1 values
+# midpoint1: single numeric value defining the value corresponding to the mid.color1 argument. A warning message is returned if midpoint1 does not correspond to the mean of limit1 values, because the color scale is not linear anymore. If NULL, takes the mean of limit1 values. Mean of data1, instead of mean of limit1, can be used here if required
+# data2: binary mask matrix (made of 0 and 1) of same dimension as data1 or a data frame resulting from the conversion of the binary mask matrix by reshape2::melt(). Value 1 of data2 will correspond to color2 argument (value 0 will be NA color), and the opposite if invert2 argument is TRUE (inverted mask)
+# color2: color of the 1 values of the binary mask matrix. The 0 values will be color NA
+# alpha2: numeric value (from 0 to 1) of the mask transparency
+# invert2: logical. Invert the mask (1 -> 0 and 0 -> 1)?
+# text.size: numeric value of the size of the texts in scale
+# title: character string of the graph title
+# title.text.size: numeric value of the title size (in points)
+# show.scale: logical. Show color scale?
+# rotate: logical. Rotate the heatmap 90° clockwise?
+# return: logical. Return the graph parameters?
+# plot: logical. Plot the graphic? If FALSE and return argument is TRUE, graphical parameters and associated warnings are provided without plotting
+# add: character string allowing to add more ggplot2 features (dots, lines, themes, etc.). BEWARE: (1) must start with "+" just after the simple or double opening quote (no space, end of line, carriage return, etc., allowed), (2) must finish with ")" just before the simple or double closing quote (no space, end of line, carriage return, etc., allowed) and (3) each function must be preceded by "ggplot2::" (for instance: "ggplot2::coord_flip()). If the character string contains the "ggplot2::theme" string, then internal ggplot2 theme() and theme_classic() functions will be inactivated to be reused by add. BEWARE: handle this argument with caution since added functions can create conflicts with the preexisting internal ggplot2 functions
+# warn.print: logical. Print warnings at the end of the execution? No print if no warning messages
+# lib.path: absolute path of the required packages, if not in the default folders
+# RETURN
+# a heatmap if plot argument is TRUE
+# a list of the graph info if return argument is TRUE:
+# $data: a list of the graphic info
+# $axes: a list of the axes info
+# $scale: the scale info (lowest, mid and highest values)
+# $warn: the warning messages. Use cat() for proper display. NULL if no warning
+# REQUIRED PACKAGES
+# ggplot2
+# reshape2
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# fun_pack()
+# fun_round()
+# EXAMPLES
+# fun_gg_heatmap(data1 = matrix(1:16, ncol = 4), title = "GRAPH 1")
+# fun_gg_heatmap(data1 = matrix(1:16, ncol = 4), return = TRUE)
+# fun_gg_heatmap(data1 = matrix(1:16, ncol = 4), legend.name1 = "VALUE", title = "GRAPH 1", text.size = 5, data2 = matrix(rep(c(1,0,0,0), 4), ncol = 4), invert2 = FALSE, return = TRUE)
+# diagonal matrix
+# fun_gg_heatmap(data1 = matrix(c(1,0,0,0,0,1,0,0,0,0,1,0,0,0,0,1), ncol = 4))
+# fun_gg_heatmap(data1 = reshape2::melt(matrix(c(1,0,0,0,0,1,0,0,0,0,1,0,0,0,0,1), ncol = 4)))
+# error message
+# fun_gg_heatmap(data1 = matrix(1:16, ncol = 4), data2 = matrix(rep(c(1,0,0,0), 5), ncol = 5))
+# fun_gg_heatmap(data1 = matrix(1:16, ncol = 4), data2 = reshape2::melt(matrix(rep(c(1,0,0,0), 4), ncol = 4)))
+# fun_gg_heatmap(data1 = reshape2::melt(matrix(1:16, ncol = 4)), data2 = reshape2::melt(matrix(rep(c(1,0,0,0), 4), ncol = 4)))
+# DEBUGGING
+# data1 = matrix(1:16, ncol = 4) ; legend.name1 = "" ; low.color1 = "blue" ; mid.color1 = "white" ; high.color1 = "red" ; limit1 = NULL ; midpoint1 = NULL ; data2 = matrix(rep(c(1,0,0,0), 4), ncol = 4) ; color2 = "black" ; alpha2 = 0.5 ; invert2 = FALSE ; text.size = 12 ; title = "" ; title.text.size = 12 ; show.scale = TRUE ; rotate = FALSE ; return = FALSE ; plot = TRUE ; add = NULL ; warn.print = TRUE ; lib.path = NULL
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(length(utils::find("fun_pack", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_pack() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(length(utils::find("fun_round", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_round() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# no reserved words required for this function
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+if(all(is.matrix(data1))){
+tempo <- fun_check(data = data1, class = "matrix", mode = "numeric", na.contain = TRUE, fun.name = function.name) ; eval(ee)
+}else if(all(is.data.frame(data1))){
+tempo <- fun_check(data = data1, class = "data.frame", length = 3, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+# structure of reshape2::melt() data frame
+tempo <- fun_check(data = data1[, 1], data.name = "COLUMN 1 OF data1 (reshape2::melt() DATA FRAME)", typeof = "integer", fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = data1[, 2], data.name = "COLUMN 2 OF data1 (reshape2::melt() DATA FRAME)", typeof = "integer", fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = data1[, 3], data.name = "COLUMN 3 OF data1 (reshape2::melt() DATA FRAME)", mode = "numeric", na.contain = TRUE, fun.name = function.name) ; eval(ee)
+}
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 ARGUMENT MUST BE A NUMERIC MATRIX OR A DATA FRAME OUTPUT OF THE reshape::melt() FUNCTION")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+tempo <- fun_check(data = legend.name1, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = low.color1, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & ! (all(low.color1 %in% colors() | grepl(pattern = "^#", low.color1)))){ # check that all strings of low.color1 start by #
+tempo.cat <- paste0("ERROR IN ", function.name, ": low.color1 ARGUMENT MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # AND/OR COLOR NAMES GIVEN BY colors()")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+if( ! is.null(mid.color1)){
+tempo <- fun_check(data = mid.color1, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & ! (all(mid.color1 %in% colors() | grepl(pattern = "^#", mid.color1)))){ # check that all strings of mid.color1 start by #
+tempo.cat <- paste0("ERROR IN ", function.name, ": mid.color1 ARGUMENT MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # AND/OR COLOR NAMES GIVEN BY colors()")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+tempo <- fun_check(data = high.color1, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & ! (all(high.color1 %in% colors() | grepl(pattern = "^#", high.color1)))){ # check that all strings of high.color1 start by #
+tempo.cat <- paste0("ERROR IN ", function.name, ": high.color1 ARGUMENT MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # AND/OR COLOR NAMES GIVEN BY colors()")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+if( ! is.null(limit1)){
+tempo <- fun_check(data = limit1, class = "vector", mode = "numeric", length = 2, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & any(limit1 %in% c(Inf, -Inf))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": limit1 ARGUMENT CANNOT CONTAIN -Inf OR Inf VALUES")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+if( ! is.null(midpoint1)){
+tempo <- fun_check(data = midpoint1, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
+}
+if( ! is.null(data2)){
+if(all(is.matrix(data2))){
+tempo <- fun_check(data = data2, class = "matrix", mode = "numeric", fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & ! all(unique(data2) %in% c(0,1))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": MATRIX IN data2 MUST BE MADE OF 0 AND 1 ONLY (MASK MATRIX)")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}else if(tempo$problem == FALSE & all(is.matrix(data1)) & ! identical(dim(data1), dim(data2))){ # matrix and matrix
+tempo.cat <- paste0("ERROR IN ", function.name, ": MATRIX DIMENSION IN data2 MUST BE IDENTICAL AS MATRIX DIMENSION IN data1. HERE IT IS RESPECTIVELY:\n", paste(dim(data2), collapse = " "), "\n", paste(dim(data1), collapse = " "))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}else if(tempo$problem == FALSE & all(is.data.frame(data1)) & nrow(data1) != prod(dim(data2))){ # reshape2 and matrix
+tempo.cat <- paste0("ERROR IN ", function.name, ": DATA FRAME IN data2 MUST HAVE ROW NUMBER EQUAL TO PRODUCT OF DIMENSIONS OF data1 MATRIX. HERE IT IS RESPECTIVELY:\n", paste(nrow(data1), collapse = " "), "\n", paste(prod(dim(data2)), collapse = " "))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}else if(all(is.data.frame(data2))){
+tempo <- fun_check(data = data2, class = "data.frame", length = 3, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+# structure of reshape2::melt() data frame
+tempo <- fun_check(data = data2[, 1], data.name = "COLUMN 1 OF data2 (reshape2::melt() DATA FRAME)", typeof = "integer", fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = data2[, 2], data.name = "COLUMN 2 OF data2 (reshape2::melt() DATA FRAME)", typeof = "integer", fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = data2[, 3], data.name = "COLUMN 3 OF data2 (reshape2::melt() DATA FRAME)", mode = "numeric", fun.name = function.name) ; eval(ee)
+}
+if(tempo$problem == FALSE & ! all(unique(data2[, 3]) %in% c(0,1))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": THIRD COLUMN OF DATA FRAME IN data2 MUST BE MADE OF 0 AND 1 ONLY (MASK DATA FRAME)")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}else if(tempo$problem == FALSE & all(is.data.frame(data1)) & ! identical(dim(data1), dim(data2))){ # data frame and data frame
+tempo.cat <- paste0("ERROR IN ", function.name, ": DATA FRAME DIMENSION IN data2 MUST BE IDENTICAL TO DATA FRAME DIMENSION IN data1. HERE IT IS RESPECTIVELY:\n", paste(dim(data2), collapse = " "), "\n", paste(dim(data1), collapse = " "))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}else if(tempo$problem == FALSE & all(is.matrix(data1)) & nrow(data2) != prod(dim(data1))){ # reshape2 and matrix
+tempo.cat <- paste0("ERROR IN ", function.name, ": DATA FRAME IN data2 MUST HAVE ROW NUMBER EQUAL TO PRODUCT OF DIMENSION OF data1 MATRIX. HERE IT IS RESPECTIVELY:\n", paste(nrow(data2), collapse = " "), "\n", paste(prod(dim(data1)), collapse = " "))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE data2 ARGUMENT MUST BE A NUMERIC MATRIX OR A DATA FRAME OUTPUT OF THE reshape::melt() FUNCTION")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+tempo <- fun_check(data = color2, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & ! (all(color2 %in% colors() | grepl(pattern = "^#", color2)))){ # check that all strings of color2 start by #
+tempo.cat <- paste0("ERROR IN ", function.name, ": color2 ARGUMENT MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # AND/OR COLOR NAMES GIVEN BY colors()")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+tempo <- fun_check(data = alpha2, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = invert2, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = text.size, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = title, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = title.text.size, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = show.scale, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = return, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = plot, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(add)){
+tempo <- fun_check(data = add, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & ! grepl(pattern = "^\\+", add)){ # check that the add string start by +
+tempo.cat <- paste0("ERROR IN ", function.name, ": add ARGUMENT MUST START WITH \"+\": ", paste(unique(add), collapse = " "))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}else if(tempo$problem == FALSE & ! grepl(pattern = "ggplot2::", add)){ #
+tempo.cat <- paste0("ERROR IN ", function.name, ": add ARGUMENT MUST CONTAIN \"ggplot2::\" IN FRONT OF EACH GGPLOT2 FUNCTION: ", paste(unique(add), collapse = " "))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}else if(tempo$problem == FALSE & ! grepl(pattern = ")$", add)){ # check that the add string finished by )
+tempo.cat <- paste0("ERROR IN ", function.name, ": add ARGUMENT MUST FINISH BY \")\": ", paste(unique(add), collapse = " "))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+tempo <- fun_check(data = warn.print, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(lib.path)){
+tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
+tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+}
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# package checking
+fun_pack(req.package = c("reshape2", "ggplot2"), lib.path = lib.path)
+# end package checking
+# main code
+ini.warning.length <- options()$warning.length
+warn <- NULL
+warn.count <- 0
+if(all(is.matrix(data1))){
+data1 <- reshape2::melt(data1) # transform a matrix into a data frame with 2 coordinates columns and the third intensity column
+}
+if(rotate == TRUE){
+data1[, 1] <- rev(data1[, 1])
+}
+if(is.null(limit1)){
+if(any(data1[, 3] %in% c(Inf, -Inf))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE data1 ARGUMENT CONTAINS -Inf OR Inf VALUES IN THE THIRD COLUMN, THAT WILL NOT BE CONSIDERED IN THE PLOT RANGE")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+limit1 <- range(data1[, 3], na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE limit1 ARGUMENT IS NULL -> RANGE OF data1 ARGUMENT HAS BEEN TAKEN: ", paste(fun_round(limit1), collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+if(suppressWarnings(any(limit1 %in% c(Inf, -Inf)))){
+tempo.cat <- paste0("ERROR IN ", function.name, " COMPUTED LIMIT CONTAINS Inf VALUES, BECAUSE VALUES FROM data1 ARGUMENTS ARE NA OR Inf ONLY")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+if(is.null(midpoint1)){
+midpoint1 <- mean(limit1, na.rm = TRUE)
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE midpoint1 ARGUMENT IS NULL -> MEAN OF limit1 ARGUMENT HAS BEEN TAKEN: ", paste(fun_round(midpoint1), collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}else if(fun_round(midpoint1, 9) != fun_round(mean(limit1), 9)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE midpoint1 ARGUMENT (", fun_round(mean(midpoint1), 9), ") DOES NOT CORRESPOND TO THE MEAN OF THE limit1 ARGUMENT (", fun_round(mean(limit1), 9), "). COLOR SCALE IS NOT LINEAR")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+if( ! is.null(data2)){
+if(all(is.matrix(data2))){
+data2 <- reshape2::melt(data2) # transform a matrix into a data frame with 2 coordinates columns and the third intensity column
+}
+if(rotate == TRUE){
+data2[, 1] <- rev(data2[, 1])
+}
+data2[, 3] <- factor(data2[, 3]) # to converte continuous scale into discrete scale
+}
+tempo.gg.name <- "gg.indiv.plot."
+tempo.gg.count <- 0 # to facilitate debugging
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::ggplot())
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_raster(data = data1, mapping = ggplot2::aes_string(x = names(data1)[ifelse(rotate == FALSE, 2, 1)], y = names(data1)[ifelse(rotate == FALSE, 1, 2)], fill = names(data1)[3]), show.legend = show.scale)) # show.legend option do not remove the legend, only the aesthetic of the legend (dot, line, etc.)
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_fill_gradient2(low = low.color1, high = high.color1, mid = mid.color1, midpoint = midpoint1, limit = limit1, breaks = c(limit1[1], midpoint1, limit1[2]), labels = fun_round(c(limit1[1], midpoint1, limit1[2])), name = legend.name1))
+if( ! is.null(data2)){
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_raster(data = data2, mapping = ggplot2::aes_string(x = names(data2)[ifelse(rotate == FALSE, 2, 1)], y = names(data2)[ifelse(rotate == FALSE, 1, 2)], alpha = names(data2)[3]), fill = color2, show.legend = FALSE))
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "alpha", values = if(invert2 == FALSE){c(0, alpha2)}else{c(alpha2, 0)}, guide = FALSE))
+# assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_raster(data = data2, mapping = ggplot2::aes_string(x = names(data2)[ifelse(rotate == FALSE, 2, 1)], y = names(data2)[ifelse(rotate == FALSE, 1, 2)], group = names(data2)[3]), fill = data2[, 3], alpha = alpha2, show.legend = FALSE)) # BEWARE: this does not work if NA present, because geom_raster() has a tendency to complete empty spaces, and thus, behave differently than geom_tile(). See https://github.com/tidyverse/ggplot2/issues/3025
+}
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::coord_fixed()) # x = y
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_y_reverse())
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::ggtitle(title))
+add.check <- TRUE
+if( ! is.null(add)){ # if add is NULL, then = 0
+if(grepl(pattern = "ggplot2::theme", add) == TRUE){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") \"ggplot2::theme\" STRING DETECTED IN THE add ARGUMENT -> INTERNAL GGPLOT2 THEME FUNCTIONS theme() AND theme_classic() HAVE BEEN INACTIVATED, TO BE USED BY THE USER")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+add.check <- FALSE
+}
+}
+if(add.check == TRUE){
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::theme_classic(base_size = text.size))
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::theme(
+text = ggplot2::element_text(size = text.size),
+plot.title = ggplot2::element_text(size = title.text.size), # stronger than text
+line = ggplot2::element_blank(),
+axis.title = ggplot2::element_blank(),
+axis.text = ggplot2::element_blank(),
+axis.ticks = ggplot2::element_blank(),
+panel.background = ggplot2::element_blank()
+))
+}
+if(plot == TRUE){
+# suppressWarnings(
+print(eval(parse(text = paste(paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "), if(is.null(add)){NULL}else{add}))))
+# )
+}else{
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") PLOT NOT SHOWN AS REQUESTED")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+if(warn.print == TRUE & ! is.null(warn)){
+options(warning.length = 8170)
+on.exit(warning(paste0("FROM ", function.name, ":\n\n", warn), call. = FALSE))
+on.exit(exp = options(warning.length = ini.warning.length), add = TRUE)
+}
+if(return == TRUE){
+output <- ggplot2::ggplot_build(eval(parse(text = paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "))))
+output <- output$data
+names(output)[1] <- "heatmap"
+if( ! is.null(data2)){
+names(output)[2] <- "mask"
+}
+return(list(data = output, axes = output$layout$panel_params[[1]], scale = c(limit1[1], midpoint1, limit1[2]), warn = warn))
+}
+}
+
+
+######## fun_gg_empty_graph() #### text to display for empty graphs
+
+
+
+
+
+# Check OK: clear to go Apollo
+fun_gg_empty_graph <- function(
+text = NULL,
+text.size = 12,
+title = NULL,
+title.size = 8,
+lib.path = NULL
+){
+# AIM
+# display an empty ggplot2 plot with a text in the middle of the window (for instance to specify that no plot can be drawn)
+# ARGUMENTS
+# text: character string of the message to display
+# text.size: numeric value of the text size (in points)
+# title: character string of the graph title
+# title.size: numeric value of the title size (in points)
+# lib.path: character vector specifying the absolute pathways of the directories containing the required packages if not in the default directories. Ignored if NULL
+# RETURN
+# an empty plot
+# REQUIRED PACKAGES
+# ggplot2
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# fun_pack()
+# EXAMPLES
+### simple example
+# fun_gg_empty_graph(text = "NO GRAPH")
+### white page
+# fun_gg_empty_graph()
+### all the arguments
+# fun_gg_empty_graph(text = "NO GRAPH", text.size = 8, title = "GRAPH1", title.size = 10, lib.path = NULL)
+# DEBUGGING
+# text = "NO GRAPH" ; text.size = 12 ; title = "GRAPH1" ; title.size = 8 ; lib.path = NULL
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(length(utils::find("fun_pack", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_pack() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+if( ! is.null(text)){
+tempo <- fun_check(data = text, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = text.size, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(title)){
+tempo <- fun_check(data = title, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = title.size, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# package checking
+fun_pack(req.package = c("ggplot2"), lib.path = lib.path)
+# end package checking
+# main code
+tempo.gg.name <- "gg.indiv.plot."
+tempo.gg.count <- 0
+# no need loop part
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::ggplot())
+if( ! is.null(text)){
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_text(data = data.frame(x = 1, y = 1, stringsAsFactors = TRUE), ggplot2::aes(x = x, y = y, label = text), size = text.size))
+}
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::ggtitle(title))
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::theme_void())
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), m.gg <- ggplot2::theme(
+plot.title = ggplot2::element_text(size = title.size) # stronger than text
+))
+suppressWarnings(print(eval(parse(text = paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + ")))))
+}
+
+
+################ Graphic extraction
+
+
+######## fun_trim() #### display values from a quantitative variable and trim according to defined cut-offs
+
+
+# Check OK: clear to go Apollo
+fun_trim <- function(
+data,
+displayed.nb = NULL,
+single.value.display = FALSE,
+trim.method = "",
+trim.cutoffs = c(0.05,
+0.975),
+interval.scale.disp = TRUE,
+down.space = 0.75,
+left.space = 0.75,
+up.space = 0.3,
+right.space = 0.25,
+orient = 1,
+dist.legend = 0.37,
+box.type = "l",
+amplif.label = 1.25,
+amplif.axis = 1.25,
+std.x.range = TRUE,
+std.y.range = TRUE,
+cex.pt = 0.2,
+col.box = hsv(0.55,
+0.8,
+0.8),
+x.nb.inter.tick = 4,
+y.nb.inter.tick = 0,
+tick.length = 1,
+sec.tick.length = 0.75,
+corner.text = "",
+amplif.legend = 1,
+corner.text.size = 0.75,
+trim.return = FALSE
+){
+# AIM
+# trim and display values from a numeric vector or matrix
+# plot 4 graphs: stripchart of values, stripchart of rank of values, histogram and normal QQPlot
+# different kinds of intervals are displayed on the top of graphes to facilitate the analysis of the variable and a trimming setting
+# the trimming interval chosen is displayed on top of graphs
+# both trimmed and not trimmed values are returned in a list
+# ARGUMENTS
+# data: values to plot (either a numeric vector or a numeric matrix)
+# displayed.nb: number of values displayed. If NULL, all the values are displayed. Otherwise, if the number of values is over displayed.nb, then displayed.nb values are displayed after random selection
+# single.value.display: provide the 4 graphs if data is made of a single (potentially repeated value)? If FALSE, an empty graph is displayed if data is made of a single (potentially repeated value). And the return list is made of NULL compartments
+# trim.method: Write "" if not required. write "mean.sd" if mean +/- sd has to be displayed as a trimming interval (only recommanded for normal distribution). Write "quantile" to display a trimming interval based on quantile cut-offs. No other possibility allowed. See trim.cutoffs below
+# trim.cutoffs: 2 values cutoff for the trimming interval displayed, each value between 0 and 1. Not used if trim.method == "".The couple of values c(lower, upper) represents the lower and upper boundaries of the trimming interval (in proportion), which represent the interval of distribution kept (between 0 and 1). Example: trim.cutoffs = c(0.05, 0.975). What is strictly kept for the display is ]lower , upper[, boundaries excluded. Using the "mean.sd" method, 0.025 and 0.975 represent 95% CI which is mean +/- 1.96 * sd
+# interval.scale.disp: display sd and quantiles intervals on top of graphs ?
+# down.space: lower vertical margin (in inches, mai argument of par())
+# left.space: left horizontal margin (in inches, mai argument of par())
+# up.space: upper vertical margin between plot region and grapical window (in inches, mai argument of par())
+# right.space: right horizontal margin (in inches, mai argument of par())
+# orient: scale number orientation (las argument of par()). 0, always parallel to the axis; 1, always horizontal; 2, always perpendicular to the axis; 3, always vertical
+# dist.legend: numeric value that moves axis legends away in inches (first number of mgp argument of par() but in inches thus / 0.2)
+# box.type: bty argument of par(). Either "o", "l", "7", "c", "u", "]", the resulting box resembles the corresponding upper case letter. A value of "n" suppresses the box
+# amplif.label: increase or decrease the size of the text in legends
+# amplif.axis: increase or decrease the size of the scale numbers in axis
+# std.x.range: standard range on the x-axis? TRUE (no range extend) or FALSE (4% range extend). Controls xaxs argument of par() (TRUE is xaxs = "i", FALSE is xaxs = "r")
+# std.y.range: standard range on the y-axis? TRUE (no range extend) or FALSE (4% range extend). Controls yaxs argument of par() (TRUE is yaxs = "i", FALSE is yaxs = "r")
+# cex.pt: size of points in stripcharts (in inches, thus cex.pt will be thereafter / 0.2)
+# col.box: color of boxplot
+# x.nb.inter.tick: number of secondary ticks between main ticks on x-axis (only if not log scale). Zero means non secondary ticks
+# y.nb.inter.tick: number of secondary ticks between main ticks on y-axis (only if not log scale). Zero means non secondary ticks
+# tick.length: length of the ticks (1 means complete the distance between the plot region and the axis numbers, 0.5 means half the length, etc. 0 means no tick
+# sec.tick.length: length of the secondary ticks (1 means complete the distance between the plot region and the axis numbers, 0.5 means half the length, etc., 0 for no ticks)
+# corner.text: text to add at the top right corner of the window
+# amplif.legend: increase or decrease the size of the text of legend
+# corner.text.size: positive numeric. Increase or decrease the size of the text. Value 1 does not change it, 0.5 decreases by half, 2 increases by 2
+# trim.return: return the trimmed and non trimmed values? NULL returned for trimmed and non trimmed values if trim.method == ""
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# RETURN
+# a list containing:
+# $trim.method: correspond to trim.method above
+# $trim.cutoffs: correspond to trim.cutoffs above
+# $real.trim.cutoffs: the two boundary values (in the unit of the numeric vector or numeric matrix analyzed). NULL
+# $trimmed.values: the values outside of the trimming interval as defined in trim.cutoffs above
+# $kept.values: the values inside the trimming interval as defined in trim.cutoffs above
+# EXAMPLES
+# fun_trim(data = c(1:100, 1:10), displayed.nb = NULL, single.value.display = FALSE, trim.method = "mean.sd", trim.cutoffs = c(0.05, 0.975), interval.scale.disp = TRUE, down.space = 0.75, left.space = 0.75, up.space = 0.3, right.space = 0.25, orient = 1, dist.legend = 0.37, box.type = "l", amplif.label = 1.25, amplif.axis = 1.25, std.x.range = TRUE, std.y.range = TRUE, cex.pt = 0.2, col.box = hsv(0.55, 0.8, 0.8), x.nb.inter.tick = 4, y.nb.inter.tick = 0, tick.length = 0.5, sec.tick.length = 0.3, corner.text = "", amplif.legend = 1, corner.text.size = 0.75, trim.return = TRUE)
+# DEBUGGING
+# data = c(1:100, 1:10) ; displayed.nb = NULL ; single.value.display = FALSE ; trim.method = "quantile" ; trim.cutoffs = c(0.05, 0.975) ; interval.scale.disp = TRUE ; down.space = 1 ; left.space = 1 ; up.space = 0.5 ; right.space = 0.25 ; orient = 1 ; dist.legend = 0.5 ; box.type = "l" ; amplif.label = 1 ; amplif.axis = 1 ; std.x.range = TRUE ; std.y.range = TRUE ; cex.pt = 0.1 ; col.box = hsv(0.55, 0.8, 0.8) ; x.nb.inter.tick = 4 ; y.nb.inter.tick = 0 ; tick.length = 0.5 ; sec.tick.length = 0.3 ; corner.text = "" ; amplif.legend = 1 ; corner.text.size = 0.75 ; trim.return = TRUE # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+# argument checking without fun_check()
+if( ! (all(class(data) == "numeric") | all(class(data) == "integer") | (all(class(data) %in% c("matrix", "array")) & mode(data) == "numeric"))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": data ARGUMENT MUST BE A NUMERIC VECTOR OR NUMERIC MATRIX")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end argument checking without fun_check()
+# argument checking with fun_check()
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+if( ! is.null(displayed.nb)){
+tempo <- fun_check(data = displayed.nb, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
+if(displayed.nb < 2){
+tempo.cat <- paste0("ERROR IN ", function.name, ": displayed.nb ARGUMENT MUST BE A SINGLE INTEGER VALUE GREATER THAN 1 AND NOT: ", paste(displayed.nb, collapse = " "))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+tempo <- fun_check(data = single.value.display, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = trim.method, options = c("", "mean.sd", "quantile"), length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = trim.cutoffs, class = "vector", mode = "numeric", length = 2, prop = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = interval.scale.disp, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = down.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = left.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = up.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = right.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = orient, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = dist.legend, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = box.type, options = c("o", "l", "7", "c", "u", "]", "n"), length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = amplif.label, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = amplif.axis, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = std.x.range, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = std.y.range, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = cex.pt, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = col.box, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = x.nb.inter.tick, class = "integer", length = 1, neg.values = FALSE, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = y.nb.inter.tick, class = "integer", length = 1, neg.values = FALSE, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = tick.length, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = sec.tick.length, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = corner.text, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = amplif.legend, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = corner.text.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = trim.return, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# end argument checking with fun_check()
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+if(all(is.na(data) | ! is.finite(data))){
+tempo.cat <- paste0("ERROR IN fun_trim FUNCTION\ndata ARGUMENT CONTAINS ONLY NA OR Inf")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end argument checking
+# main code
+if(all(class(data)%in% c("matrix", "array"))){
+data <- as.vector(data)
+}
+na.nb <- NULL
+if(any(is.na(data))){
+na.nb <- sum(c(is.na(data)))
+data <- data[ ! is.na(data)]
+}
+color.cut <- hsv(0.75, 1, 1) # color of interval selected
+col.mean <- hsv(0.25, 1, 0.8) # color of interval using mean+/-sd
+col.quantile <- "orange" # color of interval using quantiles
+quantiles.selection <- c(0.01, 0.025, 0.05, 0.1, 0.25, 0.5, 0.75, 0.9, 0.95, 0.975, 0.99) # quantiles used in axis to help for choosing trimming cutoffs
+if(single.value.display == FALSE & length(unique(data)) == 1){
+par(bty = "n", xaxt = "n", yaxt = "n", xpd = TRUE)
+plot(1, pch = 16, col = "white", xlab = "", ylab = "")
+text(x = 1, y = 1, paste0("No graphic displayed\nBecause data made of a single different value (", formatC(as.double(table(data))), ")"), cex = 2)
+output <- list(trim.method = NULL, trim.cutoffs = NULL, real.trim.cutoffs = NULL, trimmed.values = NULL, kept.values = NULL)
+}else{
+output <- list(trim.method = trim.method, trim.cutoffs = trim.cutoffs, real.trim.cutoffs = NULL, trimmed.values = NULL, kept.values = NULL)
+fun.rug <- function(sec.tick.length.f = sec.tick.length, x.nb.inter.tick.f = x.nb.inter.tick, y.nb.inter.tick.f = y.nb.inter.tick){
+if(x.nb.inter.tick.f > 0){
+inter.tick.unit <- (par("xaxp")[2] - par("xaxp")[1]) / par("xaxp")[3]
+par.ini <- par()[c("xpd", "tcl")]
+par(xpd = FALSE)
+par(tcl = -par()$mgp[2] * sec.tick.length.f) # tcl gives the length of the ticks as proportion of line text, knowing that mgp is in text lines. So the main ticks are a 0.5 of the distance of the axis numbers by default. The sign provides the side of the tick (negative for outside of the plot region)
+suppressWarnings(rug(seq(par("xaxp")[1] - 10 * inter.tick.unit, par("xaxp")[2] + 10 * inter.tick.unit, by = inter.tick.unit / (1 + x.nb.inter.tick.f)), ticksize = NA, side = 1)) # ticksize = NA to allow the use of par()$tcl value
+par(par.ini)
+rm(par.ini)
+}
+if(y.nb.inter.tick.f > 0){
+inter.tick.unit <- (par("yaxp")[2] - par("yaxp")[1]) / par("yaxp")[3]
+par.ini <- par()[c("xpd", "tcl")]
+par(xpd = FALSE)
+par(tcl = -par()$mgp[2] * sec.tick.length.f) # tcl gives the length of the ticks as proportion of line text, knowing that mgp is in text lines. So the main ticks are a 0.5 of the distance of the axis numbers by default. The sign provides the side of the tick (negative for outside of the plot region)
+suppressWarnings(rug(seq(par("yaxp")[1] - 10 * inter.tick.unit, par("yaxp")[2] + 10 * inter.tick.unit, by = inter.tick.unit / (1 + y.nb.inter.tick.f)), ticksize = NA, side = 2)) # ticksize = NA to allow the use of par()$tcl value
+par(par.ini)
+rm(par.ini)
+}
+}
+fun.add.cut <- function(data.f, trim.method.f = trim.method, trim.cutoffs.f = trim.cutoffs, color.cut.f = color.cut, return.f = FALSE){
+# DEBUGGING
+# data.f = data ; trim.method.f = "mean.sd"; trim.cutoffs.f = trim.cutoffs ; color.cut.f = color.cut ; return.f = TRUE
+real.trim.cutoffs.f <- NULL
+if(trim.method.f != ""){
+data.f <- sort(data.f)
+par.ini <- par()$xpd
+par(xpd = FALSE)
+if(trim.method.f == "mean.sd"){
+real.trim.cutoffs.f <- qnorm(trim.cutoffs.f, mean(data.f, na.rm = TRUE), sd(data.f, na.rm = TRUE))
+abline(v = qnorm(trim.cutoffs.f, mean(data.f, na.rm = TRUE), sd(data.f, na.rm = TRUE)), col = color.cut.f)
+segments(qnorm(trim.cutoffs.f[1], mean(data.f, na.rm = TRUE), sd(data.f, na.rm = TRUE)), par()$usr[4] * 0.75, qnorm(trim.cutoffs.f[2], mean(data.f, na.rm = TRUE), sd(data.f, na.rm = TRUE)), par()$usr[4] * 0.75, col = color.cut.f)
+}
+if(trim.method.f == "quantile"){
+real.trim.cutoffs.f <- quantile(data.f, probs = trim.cutoffs.f, type = 7, na.rm = TRUE)
+abline(v = quantile(data.f, probs = trim.cutoffs.f, type = 7, na.rm = TRUE), col = color.cut.f)
+segments(quantile(data.f, probs = trim.cutoffs.f[1], type = 7, na.rm = TRUE), par()$usr[4] * 0.75, quantile(data.f, probs = trim.cutoffs.f[2], type = 7, na.rm = TRUE), par()$usr[4] * 0.75, col = color.cut.f)
+}
+par(par.ini)
+if(return.f == TRUE){
+trimmed.values.f <- data.f[data.f <= real.trim.cutoffs.f[1] | data.f >= real.trim.cutoffs.f[2]]
+kept.values.f <- data.f[data.f > real.trim.cutoffs.f[1] & data.f < real.trim.cutoffs.f[2]]
+}
+}else{
+real.trim.cutoffs.f <- NULL
+trimmed.values.f <- NULL
+kept.values.f <- NULL
+}
+if(return.f == TRUE){
+output <- list(trim.method = trim.method.f, trim.cutoffs = trim.cutoffs.f, real.trim.cutoffs = real.trim.cutoffs.f, trimmed.values = trimmed.values.f, kept.values = kept.values.f)
+return(output)
+}
+}
+fun.interval.scale.display <- function(data.f, col.quantile.f = col.quantile, quantiles.selection.f = quantiles.selection, col.mean.f = col.mean){ # intervals on top of graphs
+par.ini <- par()[c("mgp", "xpd")]
+par(mgp = c(0.25, 0.25, 0), xpd = NA)
+axis(side = 3, at = c(par()$usr[1], par()$usr[2]), labels = rep("", 2), col = col.quantile.f, lwd.ticks = 0)
+par(xpd = FALSE)
+axis(side = 3, at = quantile(as.vector(data.f), probs = quantiles.selection.f, type = 7, na.rm = TRUE), labels = quantiles.selection.f, col.axis = col.quantile.f, col = col.quantile.f)
+par(mgp = c(1.75, 1.75, 1.5), xpd = NA)
+axis(side = 3, at = c(par()$usr[1], par()$usr[2]), labels = rep("", 2), col = col.mean.f, lwd.ticks = 0)
+par(xpd = FALSE)
+axis(side = 3, at = m + s * qnorm(quantiles.selection.f), labels = formatC(round(qnorm(quantiles.selection.f), 2)), col.axis = col.mean.f, col = col.mean.f, lwd.ticks = 1)
+par(par.ini)
+}
+zone<-matrix(1:4, ncol=2)
+layout(zone)
+par(omi = c(0, 0, 1.5, 0), mai = c(down.space, left.space, up.space, right.space), las = orient, mgp = c(dist.legend / 0.2, 0.5, 0), xpd = FALSE, bty= box.type, cex.lab = amplif.label, cex.axis = amplif.axis, xaxs = ifelse(std.x.range, "i", "r"), yaxs = ifelse(std.y.range, "i", "r"))
+par(tcl = -par()$mgp[2] * tick.length) # tcl gives the length of the ticks as proportion of line text, knowing that mgp is in text lines. So the main ticks are a 0.5 of the distance of the axis numbers by default. The sign provides the side of the tick (negative for outside of the plot region)
+if(is.null(displayed.nb)){
+sampled.data <- as.vector(data)
+if(corner.text == ""){
+corner.text <- paste0("ALL VALUES OF THE DATASET DISPLAYED")
+}else{
+corner.text <- paste0(corner.text, "\nALL VALUES OF THE DATASET DISPLAYED")
+}
+}else{
+if(length(as.vector(data)) > displayed.nb){
+sampled.data <- sample(as.vector(data), displayed.nb, replace = FALSE)
+if(corner.text == ""){
+corner.text <- paste0("WARNING: ONLY ", displayed.nb, " VALUES ARE DISPLAYED AMONG THE ", length(as.vector(data)), " VALUES OF THE DATASET ANALYZED")
+}else{
+corner.text <- paste0(corner.text, "\nWARNING: ONLY ", displayed.nb, " VALUES ARE DISPLAYED AMONG THE ", length(as.vector(data)), " VALUES OF THE DATASET ANALYZED")
+}
+}else{
+sampled.data <- as.vector(data)
+if(corner.text == ""){
+corner.text <- paste0("WARNING: THE DISPLAYED NUMBER OF VALUES PARAMETER ", deparse(substitute(displayed.nb)), " HAS BEEN SET TO ", displayed.nb, " WHICH IS ABOVE THE NUMBER OF VALUES OF THE DATASET ANALYZED -> ALL VALUES DISPLAYED")
+}else{
+corner.text <- paste0(corner.text, "\nWARNING: THE DISPLAYED NUMBER OF VALUES PARAMETER ", deparse(substitute(displayed.nb)), " HAS BEEN SET TO ", displayed.nb, " WHICH IS ABOVE THE NUMBER OF VALUES OF THE DATASET ANALYZED -> ALL VALUES DISPLAYED")
+}
+}
+}
+if( ! is.null(na.nb)){
+if(corner.text == ""){
+corner.text <- paste0("WARNING: NUMBER OF NA REMOVED IS ", na.nb)
+}else{
+corner.text <- paste0("WARNING: NUMBER OF NA REMOVED IS ", na.nb)
+}
+}
+stripchart(sampled.data, method="jitter", jitter=0.4, vertical=FALSE, ylim=c(0.5, 1.5), group.names = "", xlab = "Value", ylab="", pch=1, cex = cex.pt / 0.2)
+fun.rug(y.nb.inter.tick.f = 0)
+boxplot(as.vector(data), horizontal=TRUE, add=TRUE, boxwex = 0.4, staplecol = col.box, whiskcol = col.box, medcol = col.box, boxcol = col.box, range = 0, whisklty = 1)
+m <- mean(as.vector(data), na.rm = TRUE)
+s <- sd(as.vector(data), na.rm = TRUE)
+segments(m, 0.8, m, 1, lwd=2, col="red") # mean
+segments(m -1.96 * s, 0.9, m + 1.96 * s, 0.9, lwd=1, col="red") # mean
+graph.xlim <- par()$usr[1:2] # for hist() and qqnorm() below
+if(interval.scale.disp == TRUE){
+fun.interval.scale.display(data.f = data)
+if(corner.text == ""){
+corner.text <- paste0("MULTIPLYING FACTOR DISPLAYED (MEAN +/- SD) ON SCALES: ", paste(formatC(round(qnorm(quantiles.selection), 2))[-(1:(length(quantiles.selection) - 1) / 2)], collapse = ", "), "\nQUANTILES DISPLAYED ON SCALES: ", paste(quantiles.selection, collapse = ", "))
+}else{
+corner.text <- paste0(corner.text, "\nMULTIPLYING FACTOR DISPLAYED (MEAN +/- SD) ON SCALES: ", paste(formatC(round(qnorm(quantiles.selection), 2))[-(1:(length(quantiles.selection) - 1) / 2)], collapse = ", "), "\nQUANTILES DISPLAYED ON SCALES: ", paste(quantiles.selection, collapse = ", "))
+}
+}
+output.tempo <- fun.add.cut(data.f = data, return.f = TRUE) # to recover real.trim.cutoffs
+if(trim.return == TRUE){
+output <- output.tempo
+}
+par(xpd = NA)
+if(trim.method != ""){
+if(corner.text == ""){
+corner.text <- paste0("SELECTED CUT-OFFS (PROPORTION): ", paste(trim.cutoffs, collapse = ", "), "\nSELECTED CUT-OFFS: ", paste(output.tempo$real.trim.cutoffs, collapse = ", "))
+}else{
+corner.text <- paste0(corner.text, "\nSELECTED CUT-OFFS (PROPORTION): ", paste(trim.cutoffs, collapse = ", "), "\nSELECTED CUT-OFFS: ", paste(output.tempo$real.trim.cutoffs, collapse = ", "))
+}
+if(interval.scale.disp == TRUE){
+legend(x = (par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] - ((par("usr")[2] - par("usr")[1]) / (par("omd")[2] - par("omd")[1])) * par("omd")[1]), y = (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / (par("omd")[4] - par("omd")[3])) * (1 - par("omd")[4]) / 2), legend = c(c("min, Q1, Median, Q3, max"), "mean +/- 1.96sd", paste0("Trimming interval: ", paste0(trim.cutoffs, collapse = " , ")), "Mean +/- sd multiplying factor", "Quantile"), yjust = 0, lty=1, col=c(col.box, "red", color.cut, col.mean, col.quantile), bty="n", cex = amplif.legend)
+}else{
+legend(x = (par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] - ((par("usr")[2] - par("usr")[1]) / (par("omd")[2] - par("omd")[1])) * par("omd")[1]), y = (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / (par("omd")[4] - par("omd")[3])) * (1 - par("omd")[4]) / 2), legend = c(c("min, Q1, Median, Q3, max"), "mean +/- 1.96sd", paste0("Trimming interval: ", paste0(trim.cutoffs, collapse = " , "))), yjust = 0, lty=1, col=c(col.box, "red", color.cut), bty="n", cex = amplif.legend, y.intersp=1.25)
+}
+}else{
+if(interval.scale.disp == TRUE){
+legend(x = (par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] - ((par("usr")[2] - par("usr")[1]) / (par("omd")[2] - par("omd")[1])) * par("omd")[1]), y = (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / (par("omd")[4] - par("omd")[3])) * (1 - par("omd")[4]) / 2), legend = c(c("min, Q1, Median, Q3, max"), "mean +/- sd", "Mean +/- sd multiplying factor", "Quantile"), yjust = 0, lty=1, col=c(col.box, "red", col.mean, col.quantile), bty="n", cex = amplif.legend)
+}else{
+legend(x = (par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] - ((par("usr")[2] - par("usr")[1]) / (par("omd")[2] - par("omd")[1])) * par("omd")[1]), y = (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / (par("omd")[4] - par("omd")[3])) * (1 - par("omd")[4]) / 2), legend = c(c("min, Q1, Median, Q3, max"), "mean +/- sd"), yjust = 0, lty=1, col=c(col.box, "red"), bty="n", cex = amplif.legend, y.intersp=1.25)
+}
+}
+par(xpd = FALSE, xaxs = ifelse(std.x.range, "i", "r"), yaxs = ifelse(std.y.range, "i", "r"))
+hist(as.vector(data), main = "", xlim = graph.xlim, xlab = "Value", ylab="Density", col = grey(0.25)) # removed: breaks = seq(min(as.vector(data), na.rm = TRUE), max(as.vector(data), na.rm = TRUE), length.out = length(as.vector(data)) / 10)
+abline(h = par()$usr[3])
+fun.rug()
+if(interval.scale.disp == TRUE){
+fun.interval.scale.display(data.f = data)
+}
+fun.add.cut(data.f = data)
+par(xaxs = ifelse(std.x.range, "i", "r"))
+stripchart(rank(sampled.data), method="stack", vertical=FALSE, ylim=c(0.99, 1.3), group.names = "", xlab = "Rank of values", ylab="", pch=1, cex = cex.pt / 0.2)
+fun.rug(y.nb.inter.tick.f = 0)
+x.text <- par("usr")[2] + (par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1]) * (1 - par("plt")[2]) / 2
+y.text <- (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / ((par()$omd[4] / 2) * ((par("plt")[4] - par("plt")[3])))) * (1 - par("omd")[4])) # BEWARE. Here in "(par()$omd[4] / 2", division by two because there are 2 graphs staked on the y axis, and not one
+par(xpd=NA)
+text(x = x.text, y = y.text, paste0(corner.text), adj=c(1, 1.1), cex = corner.text.size) # text at the topright corner
+par(xpd=FALSE)
+par(xaxs = ifelse(std.x.range, "i", "r"), yaxs = ifelse(std.y.range, "i", "r"))
+qqnorm(as.vector(sampled.data), main = "", datax = TRUE, ylab = "Value", pch = 1, col = "red", cex = cex.pt / 0.2)
+fun.rug()
+if(diff(quantile(as.vector(data), probs = c(0.25, 0.75), na.rm = TRUE)) != 0){ # otherwise, error generated
+qqline(as.vector(data), datax = TRUE)
+}
+if(interval.scale.disp == TRUE){
+fun.interval.scale.display(data.f = data)
+}
+fun.add.cut(data.f = data)
+}
+if(trim.return == TRUE){
+return(output)
+}
+}
+
+
+######## fun_segmentation() #### segment a dot cloud on a scatterplot and define the dots from another cloud outside the segmentation
+
+
+# Check OK: clear to go Apollo
+fun_segmentation <- function(
+data1,
+x1,
+y1,
+x.range.split = NULL,
+x.step.factor = 10,
+y.range.split = NULL,
+y.step.factor = 10,
+error = 0,
+data2 = NULL,
+x2,
+y2,
+data2.pb.dot = "unknown",
+xy.cross.kind = "&",
+plot = FALSE,
+graph.in.file = FALSE,
+raster = TRUE,
+warn.print = FALSE,
+lib.path = NULL
+){
+# AIM
+# if data1 is a data frame corresponding to the data set of a scatterplot (with a x column for x-axis values and a y column for the y-axis column), then fun_segmentation() delimits a frame around the dots cloud using a sliding window set by x.range.split and x.step.factor to frame the top and bottom part of the cloud, and set by y.range.split and y.step.factor to frame the left and right part of the cloud
+# if a second data frame is provided, corresponding to the data set of a scatterplot (with a x column for x-axis values and a y column for the y-axis column), then fun_segmentation() defines the dots of this data frame, outside of the frame of the first data frame
+# WARNINGS
+# if dots from data2 look significant on the graph (outside the frame) but are not (not black on the last figure), this is probably because the frame is flat on the zero coordinate (no volume inside the frame at this position). Thus, no way to conclude that data2 dots here are significant. These dots are refered to as "unknown". The pb.dot argument deals with such dots
+# dots that are sometimes inside and outside the frame, depending on the sliding window, are treated differently: they are removed. Such dots are neither classified as "signif", "non signif" or "unknown", but as "inconsistent"
+# unknown dots are treated as finally significant, not significant, or unknown (data2.pb.dot argument) for each x-axis and y-axis separately. Then, the union or intersection of significant dots is performed (argument xy.cross.kind). See the example section
+# ARGUMENTS
+# data1: a data frame containing a column of x-axis values and a column of y-axis values
+# x1: character string of the data1 column name for x-axis (first column of data1 by default)
+# y1: character string of the data1 column name for y-axis (second column of data1 by default)
+# x.range.split: positive non null numeric value giving the number of interval on the x value range. if x.range is the range of the dots on the x-axis, then abs(diff(x.range) / x.range.split) gives the window size. Window size decreases when range.split increases. In unit of x-axis. Write NULL if not required. At least one of the x.range.split and y.range.split must be non NULL
+# x.step.factor: positive non null numeric value giving the shift step of the window. If x.step.factor = 1, no overlap during the sliding (when the window slides from position n to position n+1, no overlap between the two positions). If x.step.factor = 2, 50% of overlap (when the window slides from position n to position n+1, the window on position n+1 overlap 50% of the window when it was on position n)
+# y.range.split: same as x.range.split for the y-axis. At least one of the x.range.split and y.range.split must be non NULL
+# y.step.factor: same as x.step.factor for the y-axis
+# error: proportion (from 0 to 1) of false positives (i.e., proportion of dots from data1 outside of the frame). 0.05 means 5% of the dots from data1 outside of the frame
+# data2: a data frame containing a column of x-axis values and a column of y-axis values, for which outside dots of the data1 cloud has to be determined. Write NULL if not required
+# x2: character string of the data1 column name for x-axis (first column of data1 by default)
+# y2: character string of the data1 column name for y-axis (second column of data1 by default)
+# data2.pb.dot: unknown dots are explain in the warning section above. If "signif", then the unknown dots are finally considered as significant (outside the frame). If "not.signif", then the unknown dots are finally considered as non significant (inside the frame). If "unknown", no conclusion are drawn from these dots. See the examples below
+# xy.cross.kind: if data2 is non null and if both x.range.split and y.range.split are non null, which dots are finally significants? Write "&" for intersection of outside dots on x and on y. Write "|" for union of outside dots on x and on y. See the examples below
+# plot: logical. Print graphs that check the frame?
+# graph.in.file: logical. Graphs sent into a graphic device already opened? If FALSE, GUI are opened for each graph. If TRUE, no GUI are opended. The graphs are displayed on the current active graphic device. Ignored if plot is FALSE
+# raster: logical. Dots in raster mode? If FALSE, dots from each geom_point from geom argument are in vectorial mode (bigger pdf and long to display if millions of dots). If TRUE, dots from each geom_point from geom argument are in matricial mode (smaller pdf and easy display if millions of dots, but long to generate the layer). If TRUE, the region plot will be square to avoid a bug in fun_gg_point_rast(). If TRUE, solve the transparency problem with some GUI. Not considered if plot is FALSE
+# warn.print: logical. Print warnings at the end of the execution? No print if no warning messages
+# lib.path: character vector specifying the absolute pathways of the directories containing the required packages if not in the default directories. Ignored if NULL. Ignored if plot is FALSE
+# RETURN
+# several graphs if plot is TRUE
+# a list containing:
+# $data1.removed.row.nb: which rows have been removed due to NA; NaN, -Inf or Inf detection in x1 or y1 columns (NULL if no row removed)
+# $data1.removed.rows: removed rows (NULL if no row removed)
+# $data2.removed.row.nb: which rows have been removed due to NA; NaN, -Inf or Inf detection in x2 or y2 columns (NULL if no row removed)
+# $data2.removed.rows: removed rows (NULL if no row removed)
+# $hframe: x and y coordinates of the bottom and top frames for frame plotting (frame1 for the left step and frame2 for the right step)
+# $vframe: x and y coordinates of the left and right frames for frame plotting (frame1 for the down step and frame2 for the top step)
+# $data1.signif.dot: the significant dots of data1 (i.e., dots outside the frame). A good segmentation should not have any data1.signif.dot
+# $data1.non.signif.dot: the non significant dots of data1 (i.e., dots inside the frame)
+# $data1.inconsistent.dot: see the warning section above
+# $data2.signif.dot: the significant dots of data2 if non NULL (i.e., dots outside the frame)
+# $data2.non.signif.dot: the non significant dots of data2 (i.e., dots inside the frame)
+# $data2.unknown.dot: the problematic dots of data2 (i.e., data2 dots outside of the range of data1, or data2 dots in a sliding window without data1 dots). Is systematically NULL except if argument data2.pb.dot = "unknown" and some data2 dots are in such situation. Modifying the segmentation x.range.split, x.step.factor, y.range.split, y.step.factor arguments can solve this problem
+# $data2.inconsistent.dot: see the warning section above
+# $axes: the x-axis and y-axis info
+# $warn: the warning messages. Use cat() for proper display. NULL if no warning
+# REQUIRED PACKAGES
+# ggplot2 if plot is TRUE
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# if plot is TRUE:
+# fun_pack()
+# fun_open()
+# fun_gg_palette()
+# fun_gg_scatter()
+# fun_gg_empty_graph()
+# fun_close()
+# EXAMPLES
+# example explaining the unknown and inconsistent dots, and the cross
+
+# set.seed(1) ; data1 = data.frame(x = rnorm(500), y = rnorm(500), stringsAsFactors = TRUE) ; data1[5:7, 2] <- NA ; data2 = data.frame(x = rnorm(500, 0, 2), y = rnorm(500, 0, 2), stringsAsFactors = TRUE) ; data2[11:13, 1] <- Inf ; set.seed(NULL) ; fun_segmentation(data1 = data1, x1 = names(data1)[1], y1 = names(data1)[2], x.range.split = 20, x.step.factor = 10, y.range.split = 23, y.step.factor = 10, error = 0, data2 = data2, x2 = names(data2)[1], y2 = names(data2)[2], data2.pb.dot = "not.signif", xy.cross.kind = "|", plot = TRUE, graph.in.file = FALSE, raster = FALSE, lib.path = NULL)
+# set.seed(1) ; data1 = data.frame(x = rnorm(500), y = rnorm(500), stringsAsFactors = TRUE) ; data2 = data.frame(x = rnorm(500, 0, 2), y = rnorm(500, 0, 2), stringsAsFactors = TRUE) ; set.seed(NULL) ; fun_segmentation(data1 = data1, x1 = names(data1)[1], y1 = names(data1)[2], x.range.split = NULL, x.step.factor = 10, y.range.split = 23, y.step.factor = 10, error = 0, data2 = data2, x2 = names(data2)[1], y2 = names(data2)[2], data2.pb.dot = "unknown", xy.cross.kind = "|", plot = TRUE, graph.in.file = FALSE, raster = FALSE, lib.path = NULL)
+# set.seed(1) ; data1 = data.frame(x = rnorm(500), y = rnorm(500), stringsAsFactors = TRUE) ; data2 = data.frame(x = rnorm(500, 0, 2), y = rnorm(500, 0, 2), stringsAsFactors = TRUE) ; set.seed(NULL) ; fun_segmentation(data1 = data1, x1 = names(data1)[1], y1 = names(data1)[2], x.range.split = 20, x.step.factor = 10, y.range.split = NULL, y.step.factor = 10, error = 0, data2 = data2, x2 = names(data2)[1], y2 = names(data2)[2], data2.pb.dot = "unknown", xy.cross.kind = "&", plot = TRUE, graph.in.file = FALSE, raster = FALSE, lib.path = NULL)
+# DEBUGGING
+# set.seed(1) ; data1 = data.frame(x = rnorm(50), y = rnorm(50), stringsAsFactors = TRUE) ; data1[5:7, 2] <- NA ; x1 = names(data1)[1] ; y1 = names(data1)[2] ; x.range.split = 5 ; x.step.factor = 10 ; y.range.split = 5 ; y.step.factor = 10 ; error = 0 ; data2 = data.frame(x = rnorm(50, 0, 2), y = rnorm(50, 0, 2), stringsAsFactors = TRUE) ; set.seed(NULL) ; x2 = names(data2)[1] ; y2 = names(data2)[2] ; data2.pb.dot = "unknown" ; xy.cross.kind = "|" ; plot = TRUE ; graph.in.file = FALSE ; raster = FALSE ; warn.print = TRUE ; lib.path = NULL
+# set.seed(1) ; data1 = data.frame(x = rnorm(500), y = rnorm(500), stringsAsFactors = TRUE) ; data2 = data.frame(x = rnorm(500, 0, 2), y = rnorm(500, 0, 2), stringsAsFactors = TRUE) ; set.seed(NULL) ; x1 = names(data1)[1] ; y1 = names(data1)[2] ; x.range.split = 20 ; x.step.factor = 10 ; y.range.split = 23 ; y.step.factor = 10 ; error = 0 ; x2 = names(data2)[1] ; y2 = names(data2)[2] ; data2.pb.dot = "not.signif" ; xy.cross.kind = "|" ; plot = TRUE ; graph.in.file = FALSE ; raster = FALSE ; warn.print = TRUE ; lib.path = NULL
+# set.seed(1) ; data1 = data.frame(x = rnorm(500), y = rnorm(500), stringsAsFactors = TRUE) ; data2 = data.frame(x = rnorm(500, 0, 2), y = rnorm(500, 0, 2), stringsAsFactors = TRUE) ; set.seed(NULL) ; x1 = names(data1)[1] ; y1 = names(data1)[2] ; x.range.split = 20 ; x.step.factor = 10 ; y.range.split = NULL ; y.step.factor = 10 ; error = 0 ; x2 = names(data2)[1] ; y2 = names(data2)[2] ; data2.pb.dot = "unknown" ; xy.cross.kind = "&" ; plot = TRUE ; graph.in.file = FALSE ; raster = FALSE ; warn.print = TRUE ; lib.path = NULL
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+ini.warning.length <- options()$warning.length
+warn <- NULL
+warn.count <- 0
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = data1, class = "data.frame", na.contain = TRUE, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & length(data1) < 2){
+tempo.cat <- paste0("ERROR IN ", function.name, ": data1 ARGUMENT MUST BE A DATA FRAME OF AT LEAST 2 COLUMNS")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+tempo <- fun_check(data = x1, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & ! (x1 %in% names(data1))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": x1 ARGUMENT MUST BE A COLUMN NAME OF data1")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}else if(tempo$problem == FALSE & x1 %in% names(data1)){
+tempo <- fun_check(data = data1[, x1], data.name = "x1 COLUMN OF data1", class = "vector", mode = "numeric", na.contain = TRUE, fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = y1, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & ! (y1 %in% names(data1))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": y1 ARGUMENT MUST BE A COLUMN NAME OF data1")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}else if(tempo$problem == FALSE & y1 %in% names(data1)){
+tempo <- fun_check(data = data1[, y1], data.name = "y1 COLUMN OF data1", class = "vector", mode = "numeric", na.contain = TRUE, fun.name = function.name) ; eval(ee)
+}
+if(is.null(x.range.split) & is.null(y.range.split)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": AT LEAST ONE OF THE x.range.split AND y.range.split ARGUMENTS MUST BE NON NULL")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+if( ! is.null(x.range.split)){
+tempo <- fun_check(data = x.range.split, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & x.range.split < 1){
+tempo.cat <- paste0("ERROR IN ", function.name, ": x.range.split ARGUMENT CANNOT BE LOWER THAN 1")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+if( ! is.null(y.range.split)){
+tempo <- fun_check(data = y.range.split, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & y.range.split < 1){
+tempo.cat <- paste0("ERROR IN ", function.name, ": y.range.split ARGUMENT CANNOT BE LOWER THAN 1")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+tempo <- fun_check(data = x.step.factor, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & x.step.factor < 1){
+tempo.cat <- paste0("ERROR IN ", function.name, ": x.step.factor ARGUMENT CANNOT BE LOWER THAN 1")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+tempo <- fun_check(data = y.step.factor, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & y.step.factor < 1){
+tempo.cat <- paste0("ERROR IN ", function.name, ": y.step.factor ARGUMENT CANNOT BE LOWER THAN 1")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+tempo <- fun_check(data = error, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(data2)){
+if(is.null(x2) | is.null(y2)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": x2 AND y2 ARGUMENTS CANNOT BE NULL IF data2 ARGUMENT IS NON NULL")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+tempo <- fun_check(data = data2, class = "data.frame", na.contain = TRUE, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & length(data2) < 2){
+tempo.cat <- paste0("ERROR IN ", function.name, ": data2 ARGUMENT MUST BE A DATA FRAME OF AT LEAST 2 COLUMNS")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+if( ! is.null(x2)){
+tempo <- fun_check(data = x2, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & ! (x2 %in% names(data2))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": x2 ARGUMENT MUST BE A COLUMN NAME OF data2")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}else if(tempo$problem == FALSE & x2 %in% names(data2)){
+tempo <- fun_check(data = data2[, x2], data.name = "x2 COLUMN OF data2", class = "vector", mode = "numeric", na.contain = TRUE, fun.name = function.name) ; eval(ee)
+}
+}
+if( ! is.null(y2)){
+tempo <- fun_check(data = y2, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & ! (y2 %in% names(data2))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": y2 ARGUMENT MUST BE A COLUMN NAME OF data2")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}else if(tempo$problem == FALSE & y2 %in% names(data2)){
+tempo <- fun_check(data = data2[, y2], data.name = "y2 COLUMN OF data2", class = "vector", mode = "numeric", na.contain = TRUE, fun.name = function.name) ; eval(ee)
+}
+}
+}
+if( ! is.null(data2)){
+tempo <- fun_check(data = data2.pb.dot, options = c("signif", "not.signif", "unknown"), length = 1, fun.name = function.name) ; eval(ee)
+}
+if( ! (is.null(x.range.split)) & ! (is.null(y.range.split))){
+tempo <- fun_check(data = xy.cross.kind, options = c("&", "|"), length = 1, fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = plot, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = warn.print, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & plot == TRUE){
+tempo <- fun_check(data = raster, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = graph.in.file, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & graph.in.file == TRUE & is.null(dev.list())){
+tempo.cat <- paste0("ERROR IN ", function.name, ": \ngraph.in.file PARAMETER SET TO TRUE BUT NO ACTIVE GRAPHIC DEVICE DETECTED")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}else if(tempo$problem == FALSE & graph.in.file == TRUE & ! is.null(dev.list())){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") GRAPHS PRINTED IN THE CURRENT DEVICE (TYPE ", toupper(names(dev.cur())), ")")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+if( ! is.null(lib.path)){
+tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
+tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+}
+}
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# other required function checking
+if(plot == TRUE){
+if(length(utils::find("fun_pack", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_pack() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(length(utils::find("fun_open", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_open() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(length(utils::find("fun_gg_palette", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_gg_palette() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(length(utils::find("fun_gg_empty_graph", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_gg_empty_graph() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(length(utils::find("fun_gg_scatter", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_gg_scatter() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(length(utils::find("fun_close", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_close() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# end other required function checking
+# package checking
+if(plot == TRUE){
+fun_pack(req.package = c("ggplot2"), lib.path = lib.path)
+}
+# end package checking
+# main code
+# na and Inf detection and removal (done now to be sure of the correct length of categ)
+data1.removed.row.nb <- NULL
+data1.removed.rows <- NULL
+data2.removed.row.nb <- NULL
+data2.removed.rows <- NULL
+if(any(is.na(data1[, c(x1, y1)])) | any(is.infinite(data1[, x1])) | any(is.infinite(data1[, y1]))){
+tempo.na <- unlist(lapply(lapply(c(data1[c(x1, y1)]), FUN = is.na), FUN = which))
+tempo.inf <- unlist(lapply(lapply(c(data1[c(x1, y1)]), FUN = is.infinite), FUN = which))
+data1.removed.row.nb <- sort(unique(c(tempo.na, tempo.inf)))
+if(length(data1.removed.row.nb) > 0){
+data1.removed.rows <- data1[data1.removed.row.nb, ]
+}
+if(length(data1.removed.row.nb) == nrow(data1)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": AT LEAST ONE NA, NaN, -Inf OR Inf DETECTED IN EACH ROW OF data1. FUNCTION CANNOT BE USED ON EMPTY DATA FRAME")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(length(data1.removed.row.nb) > 0){
+data1 <- data1[-data1.removed.row.nb, ]
+}
+if(nrow(data1) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 1")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NA, NaN, -Inf OR Inf DETECTED IN COLUMN ", paste(c(x1, y1), collapse = " "), " OF data1 AND CORRESPONDING ROWS REMOVED (SEE $data1.removed.row.nb AND $data1.removed.rows)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}else{
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NO NA, NaN, -Inf OR Inf DETECTED IN COLUMN ", paste(c(x1, y1), collapse = " "), " OF data1. NO ROW REMOVED")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+if( ! is.null(data2)){
+if(any(is.na(data2[, c(x2, y2)])) | any(is.infinite(data2[, x2])) | any(is.infinite(data2[, y2]))){
+tempo.na <- unlist(lapply(lapply(c(data2[c(x2, y2)]), FUN = is.na), FUN = which))
+tempo.inf <- unlist(lapply(lapply(c(data2[c(x2, y2)]), FUN = is.infinite), FUN = which))
+data2.removed.row.nb <- sort(unique(c(tempo.na, tempo.inf)))
+if(length(data2.removed.row.nb) > 0){
+data2.removed.rows <- data2[data2.removed.row.nb, ]
+}
+if(length(data2.removed.row.nb) == nrow(data2)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": AT LEAST ONE NA, NaN, -Inf OR Inf DETECTED IN EACH ROW OF data2. FUNCTION CANNOT BE USED ON EMPTY DATA FRAME")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(length(data2.removed.row.nb) > 0){
+data2 <- data2[-data2.removed.row.nb, ]
+}
+if(nrow(data2) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 2")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NA, NaN, -Inf OR Inf DETECTED IN COLUMN ", paste(c(x2, y2), collapse = " "), " OF data2 AND CORRESPONDING ROWS REMOVED (SEE $data2.removed.row.nb AND $data2.removed.rows)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}else{
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NO NA, NaN, -Inf OR Inf DETECTED IN COLUMN ", paste(c(x2, y2), collapse = " "), " OF data2. NO ROW REMOVED")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+# end na and Inf detection and removal (done now to be sure of the correct length of categ)
+# row annotation (dot number)
+# data1 <- data1[ ! duplicated(data1[, c(x1, y1)]), ] # do not remove the dots that have same x and y values, because they will have different dot number -> not the same position on the matrices (so true for symmetric matrices)
+data1 <- cbind(data1, DOT_NB = 1:nrow(data1), stringsAsFactors = TRUE)
+if( ! is.null(data2)){
+# data2 <- data2[ ! duplicated(data2[, c(x2, y2)]), ] # do not remove the dots that have same x and y values, because they will have different dot number -> not the same position on the matrices (so true for symmetric matrices)
+data2 <- cbind(data2, DOT_NB = 1:nrow(data2), stringsAsFactors = TRUE)
+}
+# end row annotation (dot number)
+
+
+
+
+# Method using x unit interval
+# may be create vector of each column to increase speed
+x.data1.l <- NULL # x coord of the y upper and lower limits defined on the data1 cloud for left step line
+x.data1.r <- NULL # x coord of the y upper and lower limits defined on the data1 cloud for right step line
+y.data1.down.limit.l <- NULL # lower limit of the data1 cloud for left step line
+y.data1.top.limit.l <- NULL # upper limit of the data1 cloud for left step line
+y.data1.down.limit.r <- NULL # lower limit of the data1 cloud for right step line
+y.data1.top.limit.r <- NULL # upper limit of the data1 cloud for left step line
+if(any(data1[, x1] %in% c(Inf, -Inf))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE data1 ARGUMENT CONTAINS -Inf OR Inf VALUES IN THE x1 COLUMN, THAT WILL NOT BE CONSIDERED IN THE PLOT RANGE")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+x.range <- range(data1[, x1], na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
+if(suppressWarnings(any(x.range %in% c(Inf, -Inf)))){
+tempo.cat <- paste0("ERROR IN ", function.name, " COMPUTED x.range CONTAINS Inf VALUES, BECAUSE VALUES FROM data1 ARGUMENTS ARE NA OR Inf ONLY")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(data1[, y1] %in% c(Inf, -Inf))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE data1 ARGUMENT CONTAINS -Inf OR Inf VALUES IN THE y1 COLUMN, THAT WILL NOT BE CONSIDERED IN THE PLOT RANGE")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+y.range <- range(data1[, y1], na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
+if(suppressWarnings(any(x.range %in% c(Inf, -Inf)))){
+tempo.cat <- paste0("ERROR IN ", function.name, " COMPUTED y.range CONTAINS Inf VALUES, BECAUSE VALUES FROM data1 ARGUMENTS ARE NA OR Inf ONLY")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+x.range.plot <- range(data1[, x1], na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
+y.range.plot <- range(data1[, y1], na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
+if( ! is.null(data2)){
+if(any(data2[, x2] %in% c(Inf, -Inf))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE data2 ARGUMENT CONTAINS -Inf OR Inf VALUES IN THE x2 COLUMN, THAT WILL NOT BE CONSIDERED IN THE PLOT RANGE")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+x.range.plot <- range(data1[, x1], data2[, x2], na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
+if(any(data2[, y2] %in% c(Inf, -Inf))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE data2 ARGUMENT CONTAINS -Inf OR Inf VALUES IN THE y2 COLUMN, THAT WILL NOT BE CONSIDERED IN THE PLOT RANGE")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+y.range.plot <- range(data1[, y1], data2[, y2], na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
+}
+if(suppressWarnings(any(x.range.plot %in% c(Inf, -Inf)))){
+tempo.cat <- paste0("ERROR IN ", function.name, " COMPUTED x.range.plot CONTAINS Inf VALUES, BECAUSE VALUES FROM data1 (AND data2?) ARGUMENTS ARE NA OR Inf ONLY")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(suppressWarnings(any(y.range.plot %in% c(Inf, -Inf)))){
+tempo.cat <- paste0("ERROR IN ", function.name, " COMPUTED y.range.plot CONTAINS Inf VALUES, BECAUSE VALUES FROM data1 (AND data2?) ARGUMENTS ARE NA OR Inf ONLY")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! is.null(x.range.split)){
+# data.frame ordering to slide the window from small to big values + sliding window definition
+data1 <- data1[order(data1[, x1], na.last = TRUE), ]
+if( ! is.null(data2)){
+data2 <- data2[order(data2[, x2], na.last = TRUE), ]
+}
+x.win.size <- abs(diff(x.range) / x.range.split) # in unit of x-axis
+step <- x.win.size / x.step.factor
+# end data.frame ordering to slide the window from small to big values + sliding window definition
+# x-axis sliding and y-axis limits of the data1 cloud -> y significant data2
+loop.nb <- ceiling((diff(x.range) - x.win.size) / step) # x.win.size + n * step covers the x range if x.win.size + n * step >= diff(x.range), thus if n >= (diff(x.range) - x.win.size) / step
+y.outside.data1.dot.nb <- integer() # vector that will contain the selected rows numbers of data1 that are upper or lower than the frame
+y.inside.data1.dot.nb <- integer() # vector that will contain the selected rows numbers of data1 that are not upper or lower than the frame
+y.data1.median <- median(data1[, y1], na.rm = TRUE) # will be used for sliding windows without data1 in it
+if( ! is.null(data2)){
+y.outside.data2.dot.nb <- integer() # vector that will contain the selected 1D coordinates (i.e., dots) of data2 that are upper or lower than the data1 frame
+y.inside.data2.dot.nb <- integer() # vector that will contain the 1D coordinates (i.e., dots) of data2 that are not upper or lower than the data1 frame
+y.unknown.data2.dot.nb <- integer() # vector that will contain the 1D coordinates (i.e., dots) of data2 that are problematic: data2 dots outside of the range of data1, or data2 dots in a sliding window without data1 dots
+# recover data2 dots outside the range of data1
+if(any(data2[, x2] < x.range[1])){
+y.unknown.data2.dot.nb <- c(y.unknown.data2.dot.nb, data2$DOT_NB[data2[, x2] < x.range[1]])
+#tempo.warn & indicate the interval
+}
+if(any(data2[, x2] > x.range[2])){
+y.unknown.data2.dot.nb <- c(y.unknown.data2.dot.nb, data2$DOT_NB[data2[, x2] > x.range[2]])
+#tempo.warn & indicate the interval
+}
+# end recover data2 dots outside the range of data1
+}
+# loop.ini.time <- as.numeric(Sys.time())
+for(i1 in 0:(loop.nb + 1)){
+min.pos <- x.range[1] + step * i1 # lower position of the sliding window in data1
+max.pos <- min.pos + x.win.size # upper position of the sliding window in data1
+x.data1.l <- c(x.data1.l, min.pos, min.pos + step) # min.pos + step to make the steps
+x.data1.r <- c(x.data1.r, max.pos, max.pos + step) # max.pos + step to make the steps
+x.data1.dot.here <- data1[, x1] >= min.pos & data1[, x1] < max.pos # is there data1 dot present in the sliding window, considering the x axis?
+if( ! is.null(data2)){
+x.data2.dot.here <- data2[, x2] >= min.pos & data2[, x2] < max.pos # is there data2 dot present in the sliding window, considering the x axis?
+}
+# recover the data1 dots outside the frame
+if(any(x.data1.dot.here == TRUE)){
+tempo.y.data1.top.limit <- quantile(data1[x.data1.dot.here, y1], probs = 1 - error, na.rm = TRUE)
+tempo.y.data1.down.limit <- quantile(data1[x.data1.dot.here, y1], probs = 0 + error, na.rm = TRUE)
+y.data1.top.limit.l <- c(y.data1.top.limit.l, tempo.y.data1.top.limit, tempo.y.data1.top.limit)
+y.data1.down.limit.l <- c(y.data1.down.limit.l, tempo.y.data1.down.limit, tempo.y.data1.down.limit)
+y.data1.top.limit.r <- c(y.data1.top.limit.r, tempo.y.data1.top.limit, tempo.y.data1.top.limit)
+y.data1.down.limit.r <- c(y.data1.down.limit.r, tempo.y.data1.down.limit, tempo.y.data1.down.limit)
+y.data1.dot.signif <- ( ! ((data1[, y1] <= tempo.y.data1.top.limit) & (data1[, y1] >= tempo.y.data1.down.limit))) & x.data1.dot.here # is there data1 dot present in the sliding window, above or below the data1 limits, considering the y axis?
+y.data1.dot.not.signif <- x.data1.dot.here & ! y.data1.dot.signif
+y.outside.data1.dot.nb <- c(y.outside.data1.dot.nb, data1$DOT_NB[y.data1.dot.signif]) # recover the row number of data1
+y.outside.data1.dot.nb <- unique(y.outside.data1.dot.nb)
+y.inside.data1.dot.nb <- c(y.inside.data1.dot.nb, data1$DOT_NB[y.data1.dot.not.signif])
+y.inside.data1.dot.nb <- unique(y.inside.data1.dot.nb)
+}else{
+y.data1.top.limit.l <- c(y.data1.top.limit.l, y.data1.median, y.data1.median)
+y.data1.down.limit.l <- c(y.data1.down.limit.l, y.data1.median, y.data1.median)
+y.data1.top.limit.r <- c(y.data1.top.limit.r, y.data1.median, y.data1.median)
+y.data1.down.limit.r <- c(y.data1.down.limit.r, y.data1.median, y.data1.median)
+}
+# end recover the data1 dots outside the frame
+# recover the data2 dots outside the frame
+if( ! is.null(data2)){
+if(any(x.data1.dot.here == TRUE) & any(x.data2.dot.here == TRUE)){
+y.data2.dot.signif <- ( ! ((data2[, y2] <= tempo.y.data1.top.limit) & (data2[, y2] >= tempo.y.data1.down.limit))) & x.data2.dot.here # is there data2 dot present in the sliding window, above or below the data1 limits, considering the y axis?
+y.data2.dot.not.signif <- x.data2.dot.here & ! y.data2.dot.signif
+y.outside.data2.dot.nb <- c(y.outside.data2.dot.nb, data2$DOT_NB[y.data2.dot.signif])
+y.outside.data2.dot.nb <- unique(y.outside.data2.dot.nb)
+y.inside.data2.dot.nb <- c(y.inside.data2.dot.nb, data2$DOT_NB[y.data2.dot.not.signif])
+y.inside.data2.dot.nb <- unique(y.inside.data2.dot.nb)
+}else if(any(x.data1.dot.here == FALSE) & any(x.data2.dot.here == TRUE)){ # problem: data2 dots in the the window but no data1 dots to generates the quantiles
+y.unknown.data2.dot.nb <- c(y.unknown.data2.dot.nb, data2$DOT_NB[x.data2.dot.here])
+y.unknown.data2.dot.nb <- unique(y.unknown.data2.dot.nb)
+#tempo.warn & indicate the interval
+
+
+
+
+# tempo.warn <- paste0("FROM FUNCTION ", function.name, ": THE [", round(min.pos, 3), " ; ", round(max.pos, 3), "] INTERVAL DOES NOT CONTAIN data1 X VALUES BUT CONTAINS data2 X VALUES WHICH CANNOT BE EVALUATED.\nTHE CONCERNED data2 ROW NUMBERS ARE:\n", paste(which(x.data1.dot.here == FALSE & x.data2.dot.here == TRUE), collapse = "\n"))
+# warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+# end recover the data2 dots outside the frame
+# if(any(i1 == seq(1, loop.nb, 500))){
+# loop.fin.time <- as.numeric(Sys.time()) # time of process end
+# cat(paste0("COMPUTATION TIME OF LOOP ", i1, " / ", loop.nb, ": ", as.character(lubridate::seconds_to_period(round(loop.fin.time - loop.ini.time))), "\n"))
+# }
+}
+if(max.pos < x.range[2]){
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE SLIDING WINDOW HAS NOT REACHED THE MAX VALUE OF data1 ON THE X-AXIS: ", max.pos, " VERSUS ", x.range[2])
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+y.incon.data1.dot.nb.final <- unique(c(y.outside.data1.dot.nb[y.outside.data1.dot.nb %in% y.inside.data1.dot.nb], y.inside.data1.dot.nb[y.inside.data1.dot.nb %in% y.outside.data1.dot.nb])) # inconsistent dots: if a row number of y.inside.data1.dot.nb is present in y.outside.data1.dot.nb (and vice versa), it means that during the sliding, a dot has been sometime inside, sometime outside -> removed from the outside list
+y.outside.data1.dot.nb.final <- y.outside.data1.dot.nb[ ! (y.outside.data1.dot.nb %in% y.incon.data1.dot.nb.final)] # inconsistent dots removed from the outside list
+y.inside.data1.dot.nb.final <- y.inside.data1.dot.nb[ ! (y.inside.data1.dot.nb %in% y.incon.data1.dot.nb.final)] # inconsistent dots removed from the inside list
+if( ! is.null(data2)){
+# if some unknown dots are also inside, and/or outside, they are put in the inside and/or outside. Ok, because then the intersection between inside and outside is treated -> inconsistent dots
+tempo.unknown.out <- y.unknown.data2.dot.nb[y.unknown.data2.dot.nb %in% y.outside.data2.dot.nb]
+y.outside.data2.dot.nb <- unique(c(y.outside.data2.dot.nb, tempo.unknown.out)) # if a row number of y.unknown.data2.dot.nb is present in y.outside.data2.dot.nb, it is put into outside
+tempo.unknown.in <- y.unknown.data2.dot.nb[y.unknown.data2.dot.nb %in% y.inside.data2.dot.nb]
+y.inside.data2.dot.nb <- unique(c(y.inside.data2.dot.nb, tempo.unknown.in)) # if a row number of y.unknown.data2.dot.nb is present in y.inside.data2.dot.nb, it is put into inside
+y.unknown.data2.dot.nb.final <- y.unknown.data2.dot.nb[ ! (y.unknown.data2.dot.nb %in% c(y.outside.data2.dot.nb, y.inside.data2.dot.nb))] # then dots also in inside and outside are remove from unknown
+y.incon.data2.dot.nb.final <- unique(c(y.outside.data2.dot.nb[y.outside.data2.dot.nb %in% y.inside.data2.dot.nb], y.inside.data2.dot.nb[y.inside.data2.dot.nb %in% y.outside.data2.dot.nb])) # inconsistent dots: if a row number of y.inside.data2.dot.nb is present in y.outside.data2.dot.nb (and vice versa), it means that during the sliding, a dot has been sometime inside, sometime outside -> removed from the outside list
+y.outside.data2.dot.nb.final <- y.outside.data2.dot.nb[ ! (y.outside.data2.dot.nb %in% y.incon.data2.dot.nb.final)] # inconsistent dots removed from the outside list
+y.inside.data2.dot.nb.final <- y.inside.data2.dot.nb[ ! (y.inside.data2.dot.nb %in% y.incon.data2.dot.nb.final)] # inconsistent dots removed from the inside list
+}
+# end x-axis sliding and y-axis limits of the data1 cloud -> y significant data2
+}
+# end Method using x unit interval
+
+
+
+
+# Method using y unit interval
+y.data1.d <- NULL # y coord of the x upper and lower limits defined on the data1 cloud for down step line
+y.data1.t <- NULL # y coord of the x upper and lower limits defined on the data1 cloud for top step line
+x.data1.left.limit.d <- NULL # left limit of the data1 cloud for down step line
+x.data1.right.limit.d <- NULL # right limit of the data1 cloud for down step line
+x.data1.left.limit.t <- NULL # left limit of the data1 cloud for top step line
+x.data1.right.limit.t <- NULL # right limit of the data1 cloud for top step line
+if( ! is.null(y.range.split)){
+# data.frame ordering to slide the window from small to big values + sliding window definition
+data1 <- data1[order(data1[, y1], na.last = TRUE), ]
+if( ! is.null(data2)){
+data2 <- data2[order(data2[, y2], na.last = TRUE), ]
+}
+y.win.size <- abs(diff(y.range) / y.range.split) # in unit of y-axis
+step <- y.win.size / y.step.factor
+# end data.frame ordering to slide the window from small to big values + sliding window definition
+# y-axis sliding and x-axis limits of the data1 cloud -> x significant data2
+loop.nb <- ceiling((diff(y.range) - y.win.size) / step) # y.win.size + n * step covers the y range if y.win.size + n * step >= diff(y.range), thus if n >= (diff(y.range) - y.win.size) / step
+x.outside.data1.dot.nb <- integer() # vector that will contain the selected rows numbers of data1 that are upper or lower than the frame
+x.inside.data1.dot.nb <- integer() # vector that will contain the selected rows numbers of data1 that are not upper or lower than the frame
+x.data1.median <- median(data1[, x1], na.rm = TRUE) # will be used for sliding window without data1 in it
+if( ! is.null(data2)){
+x.outside.data2.dot.nb <- integer() # vector that will contain the selected 1D coordinates (i.e., dots) of data2 that are upper or lower than the data1 frame
+x.inside.data2.dot.nb <- integer() # vector that will contain the 1D coordinates (i.e., dots) of data2 that are not upper or lower than the data1 frame
+x.unknown.data2.dot.nb <- integer() # vector that will contain the 1D coordinates (i.e., dots) of data2 that are problematic: data2 dots outside of the range of data1, or data2 dots in a sliding window without data1 dots
+# recover data2 dots outside the range of data1
+if(any(data2[, y2] < y.range[1])){
+x.unknown.data2.dot.nb <- c(x.unknown.data2.dot.nb, data2$DOT_NB[data2[, y2] < y.range[1]])
+}
+if(any(data2[, y2] > y.range[2])){
+x.unknown.data2.dot.nb <- c(x.unknown.data2.dot.nb, data2$DOT_NB[data2[, y2] > y.range[2]])
+}
+# end recover data2 dots outside the range of data1
+}
+# loop.ini.time <- as.numeric(Sys.time())
+for(i1 in 0:(loop.nb + 1)){
+min.pos <- y.range[1] + step * i1 # lower position of the sliding window in data1
+max.pos <- min.pos + y.win.size # upper position of the sliding window in data1
+y.data1.d <- c(y.data1.d, min.pos, min.pos + step) # min.pos + step to make the steps
+y.data1.t <- c(y.data1.t, max.pos, max.pos + step) # max.pos + step to make the steps
+y.data1.dot.here <- data1[, y1] >= min.pos & data1[, y1] < max.pos # is there data1 dot present in the sliding window, considering the y axis?
+if( ! is.null(data2)){
+y.data2.dot.here <- data2[, y2] >= min.pos & data2[, y2] < max.pos # is there data2 dot present in the sliding window, considering the y axis?
+}
+# recover the data1 dots outside the frame
+if(any(y.data1.dot.here == TRUE)){
+tempo.x.data1.right.limit <- quantile(data1[y.data1.dot.here, x1], probs = 1 - error, na.rm = TRUE)
+tempo.x.data1.left.limit <- quantile(data1[y.data1.dot.here, x1], probs = 0 + error, na.rm = TRUE)
+x.data1.right.limit.d <- c(x.data1.right.limit.d, tempo.x.data1.right.limit, tempo.x.data1.right.limit)
+x.data1.left.limit.d <- c(x.data1.left.limit.d, tempo.x.data1.left.limit, tempo.x.data1.left.limit)
+x.data1.right.limit.t <- c(x.data1.right.limit.t, tempo.x.data1.right.limit, tempo.x.data1.right.limit)
+x.data1.left.limit.t <- c(x.data1.left.limit.t, tempo.x.data1.left.limit, tempo.x.data1.left.limit)
+x.data1.dot.signif <- ( ! ((data1[, x1] <= tempo.x.data1.right.limit) & (data1[, x1] >= tempo.x.data1.left.limit))) & y.data1.dot.here # is there data2 dot present in the sliding window, above or below the data1 limits, considering the x axis?
+x.data1.dot.not.signif <- y.data1.dot.here & ! x.data1.dot.signif
+x.outside.data1.dot.nb <- c(x.outside.data1.dot.nb, data1$DOT_NB[x.data1.dot.signif]) # recover the row number of data1
+x.outside.data1.dot.nb <- unique(x.outside.data1.dot.nb)
+x.inside.data1.dot.nb <- c(x.inside.data1.dot.nb, data1$DOT_NB[x.data1.dot.not.signif])
+x.inside.data1.dot.nb <- unique(x.inside.data1.dot.nb)
+}else{
+x.data1.right.limit.d <- c(x.data1.right.limit.d, x.data1.median, x.data1.median)
+x.data1.left.limit.d <- c(x.data1.left.limit.d, x.data1.median, x.data1.median)
+x.data1.right.limit.t <- c(x.data1.right.limit.t, x.data1.median, x.data1.median)
+x.data1.left.limit.t <- c(x.data1.left.limit.t, x.data1.median, x.data1.median)
+}
+# end recover the data1 dots outside the frame
+# recover the data2 dots outside the frame
+if( ! is.null(data2)){
+if(any(y.data1.dot.here == TRUE) & any(y.data2.dot.here == TRUE)){
+x.data2.dot.signif <- ( ! ((data2[, x2] <= tempo.x.data1.right.limit) & (data2[, x2] >= tempo.x.data1.left.limit))) & y.data2.dot.here # is there data2 dot present in the sliding window, above or below the data1 limits, considering the x axis?
+x.data2.dot.not.signif <- y.data2.dot.here & ! x.data2.dot.signif
+x.outside.data2.dot.nb <- c(x.outside.data2.dot.nb, data2$DOT_NB[x.data2.dot.signif])
+x.outside.data2.dot.nb <- unique(x.outside.data2.dot.nb)
+x.inside.data2.dot.nb <- c(x.inside.data2.dot.nb, data2$DOT_NB[x.data2.dot.not.signif])
+x.inside.data2.dot.nb <- unique(x.inside.data2.dot.nb)
+}else if(any(y.data1.dot.here == FALSE) & any(y.data2.dot.here == TRUE)){ # recover the data2 dots outside the range of the data1 cloud
+x.unknown.data2.dot.nb <- c(x.unknown.data2.dot.nb, data2$DOT_NB[y.data2.dot.here])
+x.unknown.data2.dot.nb <- unique(x.unknown.data2.dot.nb)
+
+
+
+# tempo.warn <- paste0("FROM FUNCTION ", function.name, ": THE [", round(min.pos, 3), " ; ", round(max.pos, 3), "] INTERVAL DOES NOT CONTAIN data1 Y VALUES BUT CONTAINS data2 Y VALUES WHICH CANNOT BE EVALUATED.\nTHE CONCERNED data2 ROW NUMBERS ARE:\n", paste(which(y.data1.dot.here == FALSE & y.data2.dot.here == TRUE), collapse = "\n"))
+# warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+# end recover the data2 dots outside the frame
+# if(any(i1 == seq(1, loop.nb, 500))){
+# loop.fin.time <- as.numeric(Sys.time()) # time of process end
+# cat(paste0("COMPUTATION TIME OF LOOP ", i1, " / ", loop.nb, ": ", as.character(lubridate::seconds_to_period(round(loop.fin.time - loop.ini.time))), "\n"))
+# }
+}
+if(max.pos < y.range[2]){
+tempo.cat <- paste0("ERROR IN ", function.name, ": THE SLIDING WINDOW HAS NOT REACHED THE MAX VALUE OF data1 ON THE Y-AXIS: ", max.pos, " VERSUS ", y.range[2])
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+x.incon.data1.dot.nb.final <- unique(c(x.outside.data1.dot.nb[x.outside.data1.dot.nb %in% x.inside.data1.dot.nb], x.inside.data1.dot.nb[x.inside.data1.dot.nb %in% x.outside.data1.dot.nb])) # inconsistent dots: if a row number of x.inside.data1.dot.nb is present in x.outside.data1.dot.nb (and vice versa), it means that during the sliding, a dot has been sometime inside, sometime outside -> removed from the outside list
+x.outside.data1.dot.nb.final <- x.outside.data1.dot.nb[ ! (x.outside.data1.dot.nb %in% x.incon.data1.dot.nb.final)] # inconsistent dots removed from the outside list
+x.inside.data1.dot.nb.final <- x.inside.data1.dot.nb[ ! (x.inside.data1.dot.nb %in% x.incon.data1.dot.nb.final)] # inconsistent dots removed from the inside list
+if( ! is.null(data2)){
+# if some unknown dots are also inside, and/or outside, they are put in the inside and/or outside. Ok, because then the intersection between inside and outside is treated -> inconsistent dots
+tempo.unknown.out <- x.unknown.data2.dot.nb[x.unknown.data2.dot.nb %in% x.outside.data2.dot.nb]
+x.outside.data2.dot.nb <- unique(c(x.outside.data2.dot.nb, tempo.unknown.out)) # if a row number of x.unknown.data2.dot.nb is present in x.outside.data2.dot.nb, it is put into outside
+tempo.unknown.in <- x.unknown.data2.dot.nb[x.unknown.data2.dot.nb %in% x.inside.data2.dot.nb]
+x.inside.data2.dot.nb <- unique(c(x.inside.data2.dot.nb, tempo.unknown.in)) # if a row number of x.unknown.data2.dot.nb is present in x.inside.data2.dot.nb, it is put into inside
+x.unknown.data2.dot.nb.final <- x.unknown.data2.dot.nb[ ! (x.unknown.data2.dot.nb %in% c(x.outside.data2.dot.nb, x.inside.data2.dot.nb))] # then dots also in inside and outside are remove from unknown
+x.incon.data2.dot.nb.final <- unique(c(x.outside.data2.dot.nb[x.outside.data2.dot.nb %in% x.inside.data2.dot.nb], x.inside.data2.dot.nb[x.inside.data2.dot.nb %in% x.outside.data2.dot.nb])) # inconsistent dots: if a row number of x.inside.data2.dot.nb is present in x.outside.data2.dot.nb (and vice versa), it means that during the sliding, a dot has been sometime inside, sometime outside -> removed from the outside list
+x.outside.data2.dot.nb.final <- x.outside.data2.dot.nb[ ! (x.outside.data2.dot.nb %in% x.incon.data2.dot.nb.final)] # inconsistent dots removed from the outside list
+x.inside.data2.dot.nb.final <- x.inside.data2.dot.nb[ ! (x.inside.data2.dot.nb %in% x.incon.data2.dot.nb.final)] # inconsistent dots removed from the inside list
+}
+# end y-axis sliding and x-axis limits of the data1 cloud -> x significant data2
+}
+# end Method using y unit interval
+
+
+
+# recovering the frame coordinates
+hframe = rbind(
+data.frame(
+x = if(is.null(x.data1.l)){NULL}else{x.data1.l},
+y = if(is.null(x.data1.l)){NULL}else{y.data1.down.limit.l},
+kind = if(is.null(x.data1.l)){NULL}else{"down.frame1"},
+stringsAsFactors = TRUE
+),
+data.frame(
+x = if(is.null(x.data1.r)){NULL}else{x.data1.r},
+y = if(is.null(x.data1.r)){NULL}else{y.data1.down.limit.r},
+kind = if(is.null(x.data1.r)){NULL}else{"down.frame2"},
+stringsAsFactors = TRUE
+),
+data.frame(
+x = if(is.null(x.data1.l)){NULL}else{x.data1.l},
+y = if(is.null(x.data1.l)){NULL}else{y.data1.top.limit.l},
+kind = if(is.null(x.data1.l)){NULL}else{"top.frame1"},
+stringsAsFactors = TRUE
+),
+data.frame(
+x = if(is.null(x.data1.r)){NULL}else{x.data1.r},
+y = if(is.null(x.data1.r)){NULL}else{y.data1.top.limit.r},
+kind = if(is.null(x.data1.r)){NULL}else{"top.frame2"},
+stringsAsFactors = TRUE
+),
+stringsAsFactors = TRUE
+)
+vframe = rbind(
+data.frame(
+x = if(is.null(y.data1.d)){NULL}else{x.data1.left.limit.d},
+y = if(is.null(y.data1.d)){NULL}else{y.data1.d},
+kind = if(is.null(y.data1.d)){NULL}else{"left.frame1"},
+stringsAsFactors = TRUE
+),
+data.frame(
+x = if(is.null(y.data1.t)){NULL}else{x.data1.left.limit.t},
+y = if(is.null(y.data1.t)){NULL}else{y.data1.t},
+kind = if(is.null(y.data1.t)){NULL}else{"left.frame2"},
+stringsAsFactors = TRUE
+),
+data.frame(
+x = if(is.null(y.data1.d)){NULL}else{x.data1.right.limit.d},
+y = if(is.null(y.data1.d)){NULL}else{y.data1.d},
+kind = if(is.null(y.data1.d)){NULL}else{"right.frame1"},
+stringsAsFactors = TRUE
+),
+data.frame(
+x = if(is.null(y.data1.t)){NULL}else{x.data1.right.limit.t},
+y = if(is.null(y.data1.t)){NULL}else{y.data1.t},
+kind = if(is.null(y.data1.t)){NULL}else{"right.frame2"},
+stringsAsFactors = TRUE
+),
+stringsAsFactors = TRUE
+)
+# end recovering the frame coordinates
+# recovering the dot coordinates
+data1.signif.dot <- NULL
+data1.non.signif.dot <- NULL
+data1.incon.dot <- NULL
+data2.signif.dot <- NULL
+data2.non.signif.dot <- NULL
+data2.unknown.dot <- NULL
+data2.incon.dot <- NULL
+if(( ! is.null(x.range.split)) & ( ! is.null(y.range.split))){
+# inconsistent dots recovery
+if(length(unique(c(x.incon.data1.dot.nb.final, y.incon.data1.dot.nb.final))) > 0){
+data1.incon.dot <- data1[data1$DOT_NB %in% unique(c(x.incon.data1.dot.nb.final, y.incon.data1.dot.nb.final)), ] # if a dot in inconsistent in x or y -> classified as inconsistent (so unique() used)
+# removal of the inconsistent dot in the other classifications
+x.inside.data1.dot.nb.final <- x.inside.data1.dot.nb.final[ ! x.inside.data1.dot.nb.final %in% data1.incon.dot$DOT_NB]
+y.inside.data1.dot.nb.final <- y.inside.data1.dot.nb.final[ ! y.inside.data1.dot.nb.final %in% data1.incon.dot$DOT_NB]
+x.outside.data1.dot.nb.final <- x.outside.data1.dot.nb.final[ ! x.outside.data1.dot.nb.final %in% data1.incon.dot$DOT_NB]
+y.outside.data1.dot.nb.final <- y.outside.data1.dot.nb.final[ ! y.outside.data1.dot.nb.final %in% data1.incon.dot$DOT_NB]
+x.unknown.data1.dot.nb.final <- x.unknown.data1.dot.nb.final[ ! x.unknown.data1.dot.nb.final %in% data1.incon.dot$DOT_NB]
+y.unknown.data1.dot.nb.final <- y.unknown.data1.dot.nb.final[ ! y.unknown.data1.dot.nb.final %in% data1.incon.dot$DOT_NB]
+# end removal of the inconsistent dot in the other classifications
+}
+if( ! is.null(data2)){
+if(length(unique(c(x.incon.data2.dot.nb.final, y.incon.data2.dot.nb.final))) > 0){
+data2.incon.dot <- data2[data2$DOT_NB %in% unique(c(x.incon.data2.dot.nb.final, y.incon.data2.dot.nb.final)), ]
+# removal of the inconsistent dot in the other classifications
+x.inside.data2.dot.nb.final <- x.inside.data2.dot.nb.final[ ! x.inside.data2.dot.nb.final %in% data2.incon.dot$DOT_NB]
+y.inside.data2.dot.nb.final <- y.inside.data2.dot.nb.final[ ! y.inside.data2.dot.nb.final %in% data2.incon.dot$DOT_NB]
+x.outside.data2.dot.nb.final <- x.outside.data2.dot.nb.final[ ! x.outside.data2.dot.nb.final %in% data2.incon.dot$DOT_NB]
+y.outside.data2.dot.nb.final <- y.outside.data2.dot.nb.final[ ! y.outside.data2.dot.nb.final %in% data2.incon.dot$DOT_NB]
+x.unknown.data2.dot.nb.final <- x.unknown.data2.dot.nb.final[ ! x.unknown.data2.dot.nb.final %in% data2.incon.dot$DOT_NB]
+y.unknown.data2.dot.nb.final <- y.unknown.data2.dot.nb.final[ ! y.unknown.data2.dot.nb.final %in% data2.incon.dot$DOT_NB]
+# end removal of the inconsistent dot in the other classifications
+}
+}
+# end inconsistent dots recovery
+# unknown dots recovery
+if( ! is.null(data2)){
+if(data2.pb.dot == "signif"){
+x.outside.data2.dot.nb.final <- unique(c(x.outside.data2.dot.nb.final, x.unknown.data2.dot.nb.final))
+x.inside.data2.dot.nb.final <- x.inside.data2.dot.nb.final[ ! x.inside.data2.dot.nb.final %in% x.unknown.data2.dot.nb.final] # remove x.unknown.data2.dot.nb.final from x.inside.data2.dot.nb.final
+y.outside.data2.dot.nb.final <- unique(c(y.outside.data2.dot.nb.final, y.unknown.data2.dot.nb.final))
+y.inside.data2.dot.nb.final <- y.inside.data2.dot.nb.final[ ! y.inside.data2.dot.nb.final %in% y.unknown.data2.dot.nb.final] # remove y.unknown.data2.dot.nb.final from y.inside.data2.dot.nb.final
+x.unknown.data2.dot.nb.final <- NULL
+y.unknown.data2.dot.nb.final <- NULL
+data2.unknown.dot <- NULL
+}else if(data2.pb.dot == "not.signif"){
+x.inside.data2.dot.nb.final <- unique(c(x.inside.data2.dot.nb.final, x.unknown.data2.dot.nb.final))
+x.outside.data2.dot.nb.final <- x.outside.data2.dot.nb.final[ ! x.outside.data2.dot.nb.final %in% x.unknown.data2.dot.nb.final] # remove x.unknown.data2.dot.nb.final from x.outside.data2.dot.nb.final
+y.inside.data2.dot.nb.final <- unique(c(y.inside.data2.dot.nb.final, y.unknown.data2.dot.nb.final))
+y.outside.data2.dot.nb.final <- y.outside.data2.dot.nb.final[ ! y.outside.data2.dot.nb.final %in% y.unknown.data2.dot.nb.final] # remove y.unknown.data2.dot.nb.final from y.outside.data2.dot.nb.final
+x.unknown.data2.dot.nb.final <- NULL
+y.unknown.data2.dot.nb.final <- NULL
+data2.unknown.dot <- NULL
+}else if(data2.pb.dot == "unknown"){
+if(length(unique(c(x.unknown.data2.dot.nb.final, y.unknown.data2.dot.nb.final))) > 0){
+data2.unknown.dot <- data2[data2$DOT_NB %in% unique(c(x.unknown.data2.dot.nb.final, y.unknown.data2.dot.nb.final)), ] # if a dot in unknown in x or y -> classified as unknown (so unique() used)
+x.outside.data2.dot.nb.final <- x.outside.data2.dot.nb.final[ ! x.outside.data2.dot.nb.final %in% data2.unknown.dot$DOT_NB] # remove x.unknown.data2.dot.nb.final from x.outside.data2.dot.nb.final
+x.inside.data2.dot.nb.final <- x.inside.data2.dot.nb.final[ ! x.inside.data2.dot.nb.final %in% data2.unknown.dot$DOT_NB] # remove x.unknown.data2.dot.nb.final from x.inside.data2.dot.nb.final
+y.outside.data2.dot.nb.final <- y.outside.data2.dot.nb.final[ ! y.outside.data2.dot.nb.final %in% data2.unknown.dot$DOT_NB] # remove y.unknown.data2.dot.nb.final from y.outside.data2.dot.nb.final
+y.inside.data2.dot.nb.final <- y.inside.data2.dot.nb.final[ ! y.inside.data2.dot.nb.final %in% data2.unknown.dot$DOT_NB] # remove y.unknown.data2.dot.nb.final from y.inside.data2.dot.nb.final
+}
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 3")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# end unknown dots recovery
+# sign and non sign dot recovery
+if(xy.cross.kind == "|"){ # here the problem is to deal with significant dots depending on x and y. Thus I start with that, recover dots finally non significant in outside and put them in inside (when &), and remove from inside the dots in outside
+if(length(unique(c(x.outside.data1.dot.nb.final, y.outside.data1.dot.nb.final))) > 0){
+tempo.outside <- unique(c(x.outside.data1.dot.nb.final, y.outside.data1.dot.nb.final)) # union so unique() used
+tempo.inside <- unique(c(x.inside.data1.dot.nb.final, y.inside.data1.dot.nb.final))
+tempo.inside <- tempo.inside[ ! tempo.inside %in% tempo.outside]
+data1.signif.dot <- data1[data1$DOT_NB %in% tempo.outside, ]
+data1.non.signif.dot <- data1[data1$DOT_NB %in% tempo.inside, ]
+}else{
+data1.non.signif.dot <- data1[unique(c(x.inside.data1.dot.nb.final, y.inside.data1.dot.nb.final)), ] # if no outside dots, I recover all the inside dots and that's it
+}
+}else if(xy.cross.kind == "&"){
+if(sum(x.outside.data1.dot.nb.final %in% y.outside.data1.dot.nb.final) > 0){ # that is intersection
+tempo.outside <- unique(x.outside.data1.dot.nb.final[x.outside.data1.dot.nb.final %in% y.outside.data1.dot.nb.final]) # intersection
+tempo.outside.removed <- unique(c(x.outside.data1.dot.nb.final, y.outside.data1.dot.nb.final))[ ! unique(c(x.outside.data1.dot.nb.final, y.outside.data1.dot.nb.final)) %in% tempo.outside]
+tempo.inside <- unique(c(x.inside.data1.dot.nb.final, y.inside.data1.dot.nb.final))
+data1.signif.dot <- data1[data1$DOT_NB %in% tempo.outside, ]
+data1.non.signif.dot <- data1[data1$DOT_NB %in% tempo.inside, ]
+}else{
+data1.non.signif.dot <- data1[unique(c(x.inside.data1.dot.nb.final, y.inside.data1.dot.nb.final)), ] # if no outside dots, I recover all the inside dots and that's it
+}
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 4")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! is.null(data2)){
+if(xy.cross.kind == "|"){ # here the problem is to deal with significant dots depending on x and y. Thus I start with that, recover dots finally non significant in outside and put them in inside (when &), and remove from inside the dots in outside
+if(length(unique(c(x.outside.data2.dot.nb.final, y.outside.data2.dot.nb.final))) > 0){
+tempo.outside <- unique(c(x.outside.data2.dot.nb.final, y.outside.data2.dot.nb.final)) # union so unique() used
+tempo.inside <- unique(c(x.inside.data2.dot.nb.final, y.inside.data2.dot.nb.final))
+tempo.inside <- tempo.inside[ ! tempo.inside %in% tempo.outside]
+data2.signif.dot <- data2[data2$DOT_NB %in% tempo.outside, ]
+data2.non.signif.dot <- data2[data2$DOT_NB %in% tempo.inside, ]
+}else{
+data2.non.signif.dot <- data2[unique(c(x.inside.data2.dot.nb.final, y.inside.data2.dot.nb.final)), ] # if no outside dots, I recover all the inside dots and that's it
+}
+}else if(xy.cross.kind == "&"){
+if(sum(x.outside.data2.dot.nb.final %in% y.outside.data2.dot.nb.final) > 0){ # that is intersection
+tempo.outside <- unique(x.outside.data2.dot.nb.final[x.outside.data2.dot.nb.final %in% y.outside.data2.dot.nb.final]) # intersection
+tempo.outside.removed <- unique(c(x.outside.data2.dot.nb.final, y.outside.data2.dot.nb.final))[ ! unique(c(x.outside.data2.dot.nb.final, y.outside.data2.dot.nb.final)) %in% tempo.outside]
+tempo.inside <- unique(c(x.inside.data2.dot.nb.final, y.inside.data2.dot.nb.final))
+data2.signif.dot <- data2[data2$DOT_NB %in% tempo.outside, ]
+data2.non.signif.dot <- data2[data2$DOT_NB %in% tempo.inside, ]
+}else{
+data2.non.signif.dot <- data2[unique(c(x.inside.data2.dot.nb.final, y.inside.data2.dot.nb.final)), ] # if no outside dots, I recover all the inside dots and that's it
+}
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 5")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# end sign and non sign dot recovery
+}else if(( ! is.null(x.range.split)) & is.null(y.range.split)){
+# inconsistent dots recovery
+if(length(y.incon.data1.dot.nb.final) > 0){
+data1.incon.dot <- data1[data1$DOT_NB %in% y.incon.data1.dot.nb.final, ]
+}
+if( ! is.null(data2)){
+if(length(y.incon.data2.dot.nb.final) > 0){
+data2.incon.dot <- data2[data2$DOT_NB %in% y.incon.data2.dot.nb.final, ]
+}
+}# end inconsistent dots recovery
+# unknown dots recovery
+if( ! is.null(data2)){
+if(data2.pb.dot == "signif"){
+y.outside.data2.dot.nb.final <- unique(c(y.outside.data2.dot.nb.final, y.unknown.data2.dot.nb.final))
+}else if(data2.pb.dot == "not.signif"){
+y.inside.data2.dot.nb.final <- unique(c(y.inside.data2.dot.nb.final, y.unknown.data2.dot.nb.final))
+}else if(data2.pb.dot == "unknown"){
+if(length(y.unknown.data2.dot.nb.final) > 0){
+data2.unknown.dot <- data2[data2$DOT_NB %in% y.unknown.data2.dot.nb.final, ]
+}
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 6")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# end unknown dots recovery
+# sign and non sign dot recovery
+if(length(y.outside.data1.dot.nb.final) > 0){
+data1.signif.dot <- data1[data1$DOT_NB %in% y.outside.data1.dot.nb.final, ]
+}
+if(length(y.inside.data1.dot.nb.final) > 0){
+data1.non.signif.dot <- data1[data1$DOT_NB %in% y.inside.data1.dot.nb.final, ]
+}
+if( ! is.null(data2)){
+if(length(y.outside.data2.dot.nb.final) > 0){
+data2.signif.dot <- data2[data2$DOT_NB %in% y.outside.data2.dot.nb.final, ]
+}
+if(length(y.inside.data2.dot.nb.final) > 0){
+data2.non.signif.dot <- data2[data2$DOT_NB %in% y.inside.data2.dot.nb.final, ]
+}
+}
+# end sign and non sign dot recovery
+}else if(is.null(x.range.split) & ( ! is.null(y.range.split))){
+# inconsistent dots recovery
+if(length(x.incon.data1.dot.nb.final) > 0){
+data1.incon.dot <- data1[data1$DOT_NB %in% x.incon.data1.dot.nb.final, ]
+}
+if( ! is.null(data2)){
+if(length(x.incon.data2.dot.nb.final) > 0){
+data2.incon.dot <- data2[data2$DOT_NB %in% x.incon.data2.dot.nb.final, ]
+}
+}# end inconsistent dots recovery
+# unknown dots recovery
+if( ! is.null(data2)){
+if(data2.pb.dot == "signif"){
+x.outside.data2.dot.nb.final <- unique(c(x.outside.data2.dot.nb.final, x.unknown.data2.dot.nb.final))
+}else if(data2.pb.dot == "not.signif"){
+x.inside.data2.dot.nb.final <- unique(c(x.inside.data2.dot.nb.final, x.unknown.data2.dot.nb.final))
+}else if(data2.pb.dot == "unknown"){
+if(length(x.unknown.data2.dot.nb.final) > 0){
+data2.unknown.dot <- data2[data2$DOT_NB %in% x.unknown.data2.dot.nb.final, ]
+}
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 7")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# end unknown dots recovery
+# sign and non sign dot recovery
+if(length(x.outside.data1.dot.nb.final) > 0){
+data1.signif.dot <- data1[data1$DOT_NB %in% x.outside.data1.dot.nb.final, ]
+}
+if(length(x.inside.data1.dot.nb.final) > 0){
+data1.non.signif.dot <- data1[data1$DOT_NB %in% x.inside.data1.dot.nb.final, ]
+}
+if( ! is.null(data2)){
+if(length(x.outside.data2.dot.nb.final) > 0){
+data2.signif.dot <- data2[data2$DOT_NB %in% x.outside.data2.dot.nb.final, ]
+}
+if(length(x.inside.data2.dot.nb.final) > 0){
+data2.non.signif.dot <- data2[data2$DOT_NB %in% x.inside.data2.dot.nb.final, ]
+}
+}
+# end sign and non sign dot recovery
+}
+# end recovering the dot coordinates
+# verif
+if(any(data1.signif.dot$DOT_NB %in% data1.non.signif.dot$DOT_NB)){
+tempo.cat <- paste0("ERROR IN ", FUNCTION.NAME, ": CODE INCONSISTENCY 8")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(data1.non.signif.dot$DOT_NB %in% data1.signif.dot$DOT_NB)){
+tempo.cat <- paste0("ERROR IN ", FUNCTION.NAME, ": CODE INCONSISTENCY 9")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(data1.signif.dot$DOT_NB %in% data1.incon.dot$DOT_NB)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 10")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(data1.incon.dot$DOT_NB %in% data1.signif.dot$DOT_NB)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 11")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(data1.non.signif.dot$DOT_NB %in% data1.incon.dot$DOT_NB)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 12")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(data1.incon.dot$DOT_NB %in% data1.non.signif.dot$DOT_NB)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 13")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! is.null(data2)){
+if(any(data2.signif.dot$DOT_NB %in% data2.non.signif.dot$DOT_NB)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 14")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(data2.non.signif.dot$DOT_NB %in% data2.signif.dot$DOT_NB)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 15")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(data2.signif.dot$DOT_NB %in% data2.unknown.dot$DOT_NB)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 16")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(data2.unknown.dot$DOT_NB %in% data2.signif.dot$DOT_NB)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 17")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(data2.signif.dot$DOT_NB %in% data2.incon.dot$DOT_NB)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 18")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(data2.incon.dot$DOT_NB %in% data2.signif.dot$DOT_NB)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 19")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(data2.non.signif.dot$DOT_NB %in% data2.unknown.dot$DOT_NB)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 20")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(data2.unknown.dot$DOT_NB %in% data2.non.signif.dot$DOT_NB)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 21")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(data2.non.signif.dot$DOT_NB %in% data2.incon.dot$DOT_NB)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 22")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(data2.incon.dot$DOT_NB %in% data2.non.signif.dot$DOT_NB)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 23")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(data2.unknown.dot$DOT_NB %in% data2.incon.dot$DOT_NB)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 24")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(data2.incon.dot$DOT_NB %in% data2.unknown.dot$DOT_NB)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 25")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# end verif
+# plot
+# recovering the axes data whatever plot or not
+if(is.null(data2)){
+axes <- fun_gg_scatter(data1 = list(data1), x = list(x1), y = list(y1), categ = list(NULL), color = list(fun_gg_palette(2)[2]), geom = list("geom_point"), alpha = list(0.5), x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, plot = FALSE, return = TRUE)$axes
+}else{
+axes <- fun_gg_scatter(data1 = list(data1, data2), x = list(x1, x2), y = list(y1, y2), categ = list(NULL, NULL), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1]), geom = list("geom_point", "geom_point"), alpha = list(0.5, 0.5), x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, plot = FALSE, return = TRUE)$axes
+}
+# end recovering the axes data whatever plot or not
+if(plot == TRUE){
+# add a categ for plot legend
+tempo.df.name <- c("data1", "data1.signif.dot", "data1.incon.dot", "data2", "data2.signif.dot", "data2.unknown.dot", "data2.incon.dot")
+tempo.class.name <- c("data1", "data1", "data1", "data2", "data2", "data2", "data2")
+for(i2 in 1:length(tempo.df.name)){
+if( ! is.null(get(tempo.df.name[i2], env = sys.nframe(), inherit = FALSE))){
+assign(tempo.df.name[i2], data.frame(get(tempo.df.name[i2], env = sys.nframe(), inherit = FALSE), kind = tempo.class.name[i2]),
+stringsAsFactors = TRUE)
+}
+}
+# end add a categ for plot legend
+if(( ! is.null(x.range.split)) & ( ! is.null(y.range.split))){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, hframe, vframe), x = list(x1, "x", "x"), y = list(y1, "y", "y"), categ = list("kind", "kind", "kind"), legend.name = list("DATASET", "HORIZ FRAME" , "VERT FRAME"), color = list(fun_gg_palette(2)[2], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2), rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_path", "geom_path"), alpha = list(0.5, 0.5, 0.5), title = "DATA1", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+if( ! is.null(data1.signif.dot)){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, hframe, vframe, data1.signif.dot), x = list(x1, "x", "x", x1), y = list(y1, "y", "y", y1), categ = list("kind", "kind", "kind", "kind"), legend.name = list("DATASET", "HORIZ FRAME" , "VERT FRAME", "SIGNIF DOTS"), color = list(fun_gg_palette(2)[2], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2), rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2), "black"), geom = list("geom_point", "geom_path", "geom_path", "geom_point"), alpha = list(0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA1 SIGNIFICANT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else{
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA1 DOTS\nOUTSIDE THE FRAMES", text.size = 8, title = "DATA1 + DATA1 SIGNIFICANT DOTS")
+}
+if( ! is.null(data1.incon.dot)){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, hframe, vframe, data1.incon.dot), x = list(x1, "x", "x", x1), y = list(y1, "y", "y", y1), categ = list("kind", "kind", "kind", "kind"), legend.name = list("DATASET", "HORIZ FRAME" , "VERT FRAME", "INCONSISTENT DOTS"), color = list(fun_gg_palette(2)[2], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2), rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2), fun_gg_palette(7)[6]), geom = list("geom_point", "geom_path", "geom_path", "geom_point"), alpha = list(0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA1 INCONSISTENT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else{
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA1\nINCONSISTENT DOTS", text.size = 8, title = "DATA1 + DATA1 INCONSISTENT DOTS")
+}
+if( ! is.null(data2)){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, hframe , vframe), x = list(x1, x2, "x", "x"), y = list(y1, y2, "y", "y"), categ = list("kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "HORIZ FRAME" , "VERT FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2), rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_point", "geom_path", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+if( ! is.null(data2.signif.dot)){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, data2.signif.dot, hframe , vframe), x = list(x1, x2, x2, "x", "x"), y = list(y1, y2, y2, "y", "y"), categ = list("kind", "kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "SIGNIF DOTS", "HORIZ FRAME" , "VERT FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], "black", rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2), rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_point", "geom_point", "geom_path", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2 + DATA2 SIGNIFICANT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else{
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA2 DOTS\nOUTSIDE THE FRAMES", text.size = 8, title = "DATA1 + DATA2 + DATA2 SIGNIFICANT DOTS")
+}
+if( ! is.null(data2.incon.dot)){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, data2.incon.dot, hframe , vframe), x = list(x1, x2, x2, "x", "x"), y = list(y1, y2, y2, "y", "y"), categ = list("kind", "kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "INCONSISTENT DOTS", "HORIZ FRAME" , "VERT FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], fun_gg_palette(7)[6], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2), rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_point", "geom_point", "geom_path", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2 + DATA2 INCONSISTENT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else{
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA2\nINCONSISTENT DOTS", text.size = 8, title = "DATA2 + DATA2 INCONSISTENT DOTS")
+}
+if( ! is.null(data2.unknown.dot)){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, data2.unknown.dot, hframe , vframe), x = list(x1, x2, x2, "x", "x"), y = list(y1, y2, y2, "y", "y"), categ = list("kind", "kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "UNKNOWN DOTS", "HORIZ FRAME" , "VERT FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], fun_gg_palette(7)[5], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2), rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_point", "geom_point", "geom_path", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2 + DATA2 UNKNOWN DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else{
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA2\nUNKNOWN DOTS", text.size = 12, title = "DATA2 + DATA2 UNKNOWN DOTS")
+}
+}
+}else if(( ! is.null(x.range.split)) & is.null(y.range.split)){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, hframe), x = list(x1, "x"), y = list(y1, "y"), categ = list("kind", "kind"), legend.name = list("DATASET", "HORIZ FRAME"), color = list(fun_gg_palette(2)[2], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2)), geom = list("geom_point", "geom_path"), alpha = list(0.5, 0.5), title = "DATA1", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+if( ! is.null(data1.signif.dot)){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, hframe, data1.signif.dot), x = list(x1, "x", x1), y = list(y1, "y", y1), categ = list("kind", "kind", "kind"), legend.name = list("DATASET", "HORIZ FRAME", "SIGNIF DOTS"), color = list(fun_gg_palette(2)[2], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2), "black"), geom = list("geom_point", "geom_path", "geom_point"), alpha = list(0.5, 0.5, 0.5), title = "DATA1 + DATA1 SIGNIFICANT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else{
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA1 DOTS\nOUTSIDE THE FRAMES", text.size = 8, title = "DATA1 + DATA1 SIGNIFICANT DOTS")
+}
+if( ! is.null(data1.incon.dot)){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, hframe, data1.incon.dot), x = list(x1, "x", x1), y = list(y1, "y", y1), categ = list("kind", "kind", "kind"), legend.name = list("DATASET", "HORIZ FRAME", "INCONSISTENT DOTS"), color = list(fun_gg_palette(2)[2], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2), fun_gg_palette(7)[6]), geom = list("geom_point", "geom_path", "geom_point"), alpha = list(0.5, 0.5, 0.5), title = "DATA1 + DATA1 INCONSISTENT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else{
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA1\nINCONSISTENT DOTS", text.size = 8, title = "DATA1 + DATA1 INCONSISTENT DOTS")
+}
+if( ! is.null(data2)){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, hframe), x = list(x1, x2, "x"), y = list(y1, y2, "y"), categ = list("kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "HORIZ FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2)), geom = list("geom_point", "geom_point", "geom_path"), alpha = list(0.5, 0.5, 0.5), title = "DATA1 + DATA2", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+if( ! is.null(data2.signif.dot)){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, data2.signif.dot, hframe), x = list(x1, x2, x2, "x"), y = list(y1, y2, y2, "y"), categ = list("kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "SIGNIF DOTS", "HORIZ FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], "black", rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2)), geom = list("geom_point", "geom_point", "geom_point", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2 + DATA2 SIGNIFICANT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else{
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA2 DOTS\nOUTSIDE THE FRAMES", text.size = 8, title = "DATA1 + DATA2 + DATA2 SIGNIFICANT DOTS")
+}
+if( ! is.null(data2.incon.dot)){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, data2.incon.dot, hframe), x = list(x1, x2, x2, "x"), y = list(y1, y2, y2, "y"), categ = list("kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "INCONSISTENT DOTS", "HORIZ FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], fun_gg_palette(7)[6], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2)), geom = list("geom_point", "geom_point", "geom_point", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2 + DATA2 INCONSISTENT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else{
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA2\nINCONSISTENT DOTS", text.size = 8, title = "DATA2 + DATA2 INCONSISTENT DOTS")
+}
+if( ! is.null(data2.unknown.dot)){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, data2.unknown.dot, hframe), x = list(x1, x2, x2, "x"), y = list(y1, y2, y2, "y"), categ = list("kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "UNKNOWN DOTS", "HORIZ FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], fun_gg_palette(7)[5], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2)), geom = list("geom_point", "geom_point", "geom_point", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2 + DATA2 UNKNOWN DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else{
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA2\nUNKNOWN DOTS", text.size = 8, title = "DATA2 + DATA2 UNKNOWN DOTS")
+}
+}
+}else if(is.null(x.range.split) & ( ! is.null(y.range.split))){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, vframe), x = list(x1, "x"), y = list(y1, "y"), categ = list("kind", "kind"), legend.name = list("DATASET", "VERT FRAME"), color = list(fun_gg_palette(2)[2], rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_path"), alpha = list(0.5, 0.5), title = "DATA1", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+if( ! is.null(data1.signif.dot)){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, vframe, data1.signif.dot), x = list(x1, "x", x1), y = list(y1, "y", y1), categ = list("kind", "kind", "kind"), legend.name = list("DATASET", "VERT FRAME", "SIGNIF DOTS"), color = list(fun_gg_palette(2)[2], rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2), "black"), geom = list("geom_point", "geom_path", "geom_point"), alpha = list(0.5, 0.5, 0.5), title = "DATA1 + DATA1 SIGNIFICANT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else{
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA1 DOTS\nOUTSIDE THE FRAMES", text.size = 8, title = "DATA1 + DATA1 SIGNIFICANT DOTS")
+}
+if( ! is.null(data1.incon.dot)){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, vframe, data1.incon.dot), x = list(x1, "x", x1), y = list(y1, "y", y1), categ = list("kind", "kind", "kind"), legend.name = list("DATASET", "VERT FRAME", "INCONSISTENT DOTS"), color = list(fun_gg_palette(2)[2], rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2), fun_gg_palette(7)[6]), geom = list("geom_point", "geom_path", "geom_point"), alpha = list(0.5, 0.5, 0.5), title = "DATA1 + DATA1 INCONSISTENT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else{
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA1\nINCONSISTENT DOTS", text.size = 8, title = "DATA1 + DATA1 INCONSISTENT DOTS")
+}
+if( ! is.null(data2)){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, vframe), x = list(x1, x2, "x"), y = list(y1, y2, "y"), categ = list("kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "VERT FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_point", "geom_path"), alpha = list(0.5, 0.5, 0.5), title = "DATA1 + DATA2", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+if( ! is.null(data2.signif.dot)){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, data2.signif.dot, vframe), x = list(x1, x2, x2, "x"), y = list(y1, y2, y2, "y"), categ = list("kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "SIGNIF DOTS", "VERT FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], "black", rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_point", "geom_point", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2 + DATA2 SIGNIFICANT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else{
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA2 DOTS\nOUTSIDE THE FRAMES", text.size = 8, title = "DATA1 + DATA2 + DATA2 SIGNIFICANT DOTS")
+}
+if( ! is.null(data2.incon.dot)){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, data2.incon.dot, vframe), x = list(x1, x2, x2, "x"), y = list(y1, y2, y2, "y"), categ = list("kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "INCONSISTENT DOTS", "VERT FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], fun_gg_palette(7)[6], rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_point", "geom_point", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2 + DATA2 INCONSISTENT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else{
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA2\nINCONSISTENT DOTS", text.size = 8, title = "DATA2 + DATA2 INCONSISTENT DOTS")
+}
+if( ! is.null(data2.unknown.dot)){
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, data2.unknown.dot, vframe), x = list(x1, x2, x2, "x"), y = list(y1, y2, y2, "y"), categ = list("kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "UNKNOWN DOTS", "VERT FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], fun_gg_palette(7)[5], rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_point", "geom_point", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2 + DATA2 UNKNOWN DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
+if( ! is.null(tempo.graph$warn)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else{
+if(graph.in.file == FALSE){
+fun_open(pdf = FALSE)
+}
+fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA2\nUNKNOWN DOTS", text.size = 8, title = "DATA2 + DATA2 UNKNOWN DOTS")
+}
+}
+}
+}
+# end plot
+if(warn.print == TRUE & ! is.null(warn)){
+options(warning.length = 8170)
+on.exit(warning(paste0("FROM ", function.name, ":\n\n", warn), call. = FALSE))
+on.exit(exp = options(warning.length = ini.warning.length), add = TRUE)
+}
+tempo.list <- list(data1.removed.row.nb = data1.removed.row.nb, data1.removed.rows = data1.removed.rows, data2.removed.row.nb = data2.removed.row.nb, data2.removed.rows = data2.removed.rows, hframe = hframe, vframe = vframe, data1.signif.dot = data1.signif.dot, data1.non.signif.dot = data1.non.signif.dot, data1.inconsistent.dot = data1.incon.dot, data2.signif.dot = data2.signif.dot, data2.non.signif.dot = data2.non.signif.dot, data2.unknown.dot = data2.unknown.dot, data2.inconsistent.dot = data2.incon.dot, axes = axes, warn = warn)
+return(tempo.list)
+}
+
+
+################ Import
+
+
+######## fun_pack() #### check if R packages are present and import into the working environment
+
+
+# Check OK: clear to go Apollo
+fun_pack <- function(
+req.package,
+load = FALSE,
+lib.path = NULL
+){
+# AIM
+# check if the specified R packages are present in the computer and import them into the working environment
+# ARGUMENTS
+# req.package: character vector of package names to import
+# load: logical. Load the package into the environement (using library())? Interesting if packages are not in default folders or for checking the functions names of packages using search()
+# lib.path: optional character vector specifying the absolute pathways of the directories containing some of the listed packages in the req.package argument, if not in the default directories. Ignored if NULL
+# RETURN
+# nothing
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# fun_pack(req.package = "nopackage")
+# fun_pack(req.package = "ggplot2")
+# fun_pack(req.package = "ggplot2", lib.path = "blablabla")
+# DEBUGGING
+# req.package = "ggplot2" ; lib.path = "C:/Program Files/R/R-3.5.1/library"
+# req.package = "serpentine" ; lib.path = "C:/users/gael/appdata/roaming/python/python36/site-packages"
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = req.package, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = load, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(lib.path)){
+tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
+tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+}
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+if(is.null(lib.path)){
+lib.path <- .libPaths() # .libPaths(new = lib.path) # or .libPaths(new = c(.libPaths(), lib.path))
+}else{
+.libPaths(new = sub(x = lib.path, pattern = "/$|\\\\$", replacement = "")) # .libPaths(new = ) add path to default path. BEWARE: .libPaths() does not support / at the end of a submitted path. Thus check and replace last / or \\ in path
+}
+tempo <- NULL
+for(i1 in 1:length(req.package)){
+if( ! req.package[i1] %in% rownames(utils::installed.packages(lib.loc = lib.path))){
+tempo <- c(tempo, req.package[i1])
+}
+}
+if( ! is.null(tempo)){
+tempo.cat <- paste0(
+"ERROR IN ",
+function.name,
+": PACKAGE",
+ifelse(length(tempo) == 1, paste0("\n\n", tempo, "\n\n"), paste0("S\n", paste(tempo, collapse = "\n"), "\n")),
+"MUST BE INSTALLED IN",
+ifelse(length(lib.path) == 1, "", " ONE OF THESE FOLDERS"),
+":\n",
+paste(lib.path, collapse = "\n")
+)
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else if(load == TRUE){
+suppressMessages(suppressWarnings(suppressPackageStartupMessages(library(req.package[i1], lib.loc = lib.path, quietly = TRUE, character.only = TRUE))))
+}
+}
+
+
+######## fun_python_pack() #### check if python packages are present
+
+
+# Check OK: clear to go Apollo
+fun_python_pack <- function(
+req.package,
+python.exec.path = NULL,
+lib.path = NULL,
+R.lib.path = NULL
+){
+# AIM
+# check if the specified python packages are present in the computer (no import)
+# WARNINGS
+# for python 3.7. Previous versions return an error "Error in sys$stdout$flush() : attempt to apply non-function"
+# ARGUMENTS
+# req.package: character vector of package names to import
+# python.exec.path: optional character vector specifying the absolute pathways of the executable python file to use (associated to the packages to use). If NULL, the reticulate::import_from_path() function used in fun_python_pack() seeks for an available version of python.exe, and then uses python_config(python_version, required_module, python_versions). But might not be the correct one for the lib.path parameter specified. Thus, it is recommanded to do not leave NULL, notably when using computing clusters
+# lib.path: optional character vector specifying the absolute pathways of the directories containing some of the listed packages in the req.package argument, if not in the default directories
+# R.lib.path: absolute path of the reticulate packages, if not in the default folders
+# RETURN
+# nothing
+# REQUIRED PACKAGES
+# reticulate
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# fun_pack()
+# EXAMPLES
+# example of error message
+# fun_python_pack(req.package = "nopackage")
+# example without error message (require the installation of the python serpentine package from https://github.com/koszullab/serpentine
+# fun_python_pack(req.package = "serpentine", python.exec.path = "C:/ProgramData/Anaconda3/python.exe", lib.path = "c:/programdata/anaconda3/lib/site-packages/")
+# another example of error message
+# fun_python_pack(req.package = "serpentine", lib.path = "blablabla")
+# DEBUGGING
+# req.package = "serpentine" ; python.exec.path = "C:/ProgramData/Anaconda3/python.exe" ; lib.path = "c:/programdata/anaconda3/lib/site-packages/" ; R.lib.path = NULL
+# req.package = "bad" ; lib.path = NULL ; R.lib.path = NULL
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(length(utils::find("fun_pack", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_pack() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = req.package, class = "character", fun.name = function.name) ; eval(ee)
+if( ! is.null(python.exec.path)){
+tempo <- fun_check(data = python.exec.path, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+if( ! all(file.exists(python.exec.path))){ # separation to avoid the problem of tempo$problem == FALSE and python.exec.path == NA
+tempo.cat <- paste0("ERROR IN ", function.name, ": FILE PATH INDICATED IN THE python.exec.path ARGUMENT DOES NOT EXISTS:\n", paste(python.exec.path, collapse = "\n"))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+}
+if( ! is.null(lib.path)){
+tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
+tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+}
+if( ! is.null(R.lib.path)){
+tempo <- fun_check(data = R.lib.path, class = "character", fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+if( ! all(dir.exists(R.lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and R.lib.path == NA
+tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE R.lib.path ARGUMENT DOES NOT EXISTS:\n", paste(R.lib.path, collapse = "\n"))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+}
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# package checking
+fun_pack(req.package = "reticulate", lib.path = R.lib.path)
+# end package checking
+# main code
+if(is.null(python.exec.path)){
+python.exec.path <- reticulate::py_run_string("
+import sys ;
+path_lib = sys.path
+") # python string
+python.exec.path <- python.exec.path$path_lib
+}
+if(is.null(lib.path)){
+lib.path <- reticulate::py_run_string("
+import sys ;
+path_lib = sys.path
+") # python string
+lib.path <- lib.path$path_lib
+}
+reticulate::use_python(Sys.which(python.exec.path), required = TRUE) # required to avoid the use of erratic python exec by reticulate::import_from_path()
+for(i1 in 1:length(req.package)){
+tempo.try <- vector("list", length = length(lib.path))
+for(i2 in 1:length(lib.path)){
+tempo.try[[i2]] <- suppressWarnings(try(reticulate::import_from_path(req.package[i1], path = lib.path[i2]), silent = TRUE))
+tempo.try[[i2]] <- suppressWarnings(try(reticulate::import_from_path(req.package[i1], path = lib.path[i2]), silent = TRUE)) # done twice to avoid the error message about flushing present the first time but not the second time. see https://stackoverflow.com/questions/57357001/reticulate-1-13-error-in-sysstdoutflush-attempt-to-apply-non-function
+}
+if(all(sapply(tempo.try, FUN = grepl, pattern = "[Ee]rror"))){
+print(tempo.try)
+tempo.cat <- paste0("ERROR IN ", function.name, ": PACKAGE ", req.package[i1], " MUST BE INSTALLED IN THE MENTIONNED DIRECTORY:\n", paste(lib.path, collapse = "\n"))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+} # else{
+# suppressMessages(suppressWarnings(suppressPackageStartupMessages(assign(req.package[i1], reticulate::import(req.package[i1]))))) # not required because try() already evaluates
+# }
+}
+}
+
+
+################ Print / Exporting results (text & tables)
+
+
+######## fun_report() #### print string or data object into output file
+
+
+# Check OK: clear to go Apollo
+fun_report <- function(
+data,
+output = "results.txt",
+path = "C:/Users/Gael/Desktop/",
+no.overwrite = TRUE,
+rownames.kept = FALSE,
+vector.cat = FALSE,
+noquote = TRUE,
+sep = 2
+){
+# AIM
+# log file function: print a character string or a data object into a same output file
+# ARGUMENTS
+# data: object to print in the output file. If NULL, nothing is done, with no warning
+# output: name of the output file
+# path: location of the output file
+# no.overwrite: (logical) if output file already exists, defines if the printing is appended (default TRUE) or if the output file content is erased before printing (FALSE)
+# rownames.kept: (logical) defines whether row names have to be removed or not in small tables (less than length.rows rows)
+# vector.cat (logical). If TRUE print a vector of length > 1 using cat() instead of capture.output(). Otherwise (default FALSE) the opposite
+# noquote: (logical). If TRUE no quote are present for the characters
+# sep: number of separating lines after printed data (must be integer)
+# RETURN
+# nothing
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# fun_report()
+# fun_report(data = 1:3, output = "results.txt", path = "C:/Users/Gael/Desktop", no.overwrite = TRUE, rownames.kept = FALSE, vector.cat = FALSE, noquote = FALSE, sep = 2)
+# DEBUGGING
+# data = 1:3 ; output = "results.txt" ; path = "C:/Users/Gael/Desktop" ; no.overwrite = TRUE ; rownames.kept = FALSE ; vector.cat = FALSE ; noquote = FALSE ; sep = 2 # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = output, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & output == ""){
+tempo.cat <- paste0("ERROR IN ", function.name, ": output ARGUMENT AS \"\" DOES NOT CORRESPOND TO A VALID FILE NAME")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+tempo <- fun_check(data = path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+if( ! all(dir.exists(path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
+tempo.cat <- paste0("ERROR IN ", function.name, ": path ARGUMENT DOES NOT CORRESPOND TO EXISTING DIRECTORY\n", paste(path, collapse = "\n"))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+tempo <- fun_check(data = no.overwrite, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = rownames.kept, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = vector.cat, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = noquote, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = sep, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# end argument checking
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# the 4 next lines are inactivated but kept because at a time, I might have a problem with data (solved with data = NULL). These 4 lines are just to know how to detect a missing argument. Important here because if data is not provided, print the code of the data function
+# arg.user.list <- as.list(match.call(expand.dots = FALSE))[-1] # recover all the arguments provided by the function user (excluding the argument with defaults values not provided by the user. Thus, it is really the list indicated by the user)
+# default.arg.list <- formals(fun = sys.function(sys.parent())) # list of all the arguments of the function with their default values (not the values of the user !). It seems that ls() as first line of the function provide the names of the arguments (empty, called, etc., or not)
+# arg.without.default.value <- sapply(default.arg.list, is.symbol) & sapply(sapply(default.arg.list, as.character), identical, "") # logical to detect argument without default values (these are typeof "symbol" and class "name" and empty character
+# if( ! all(names(default.arg.list)[arg.without.default.value] %in% names(arg.user.list))){ # test that the arguments with no null values are provided by the user
+# tempo.cat <- paste0("ERROR IN ", function.name, ": VALUE REQUIRED FOR THESE ARGUMENTS WITH NO DEFAULTS VALUES: ", paste(names(default.arg.list)[arg.without.default.value][ ! names(default.arg.list)[arg.without.default.value] %in% names(arg.user.list)], collapse = " "))
+# stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+# }
+# end argument checking
+# main code
+if( ! is.null(data)){
+if(all(class(data) == "data.frame") | all(class(data) == "table") | all(class(data) %in% c("matrix", "array"))){ # before R4.0.0, it was all(class(data) %in% c("matrix", "data.frame", "table"))
+if(rownames.kept == FALSE & all(class(data) == "data.frame") & nrow(data) != 0 & nrow(data) <= 4){ # for data frames with nrows <= 4
+rownames.output.tables <- ""
+length.rows <- nrow(data)
+for(i in 1:length.rows){ # replace the rownames of the first 4 rows by increasing number of spaces (because identical row names not allowed in data frames). This method cannot be extended to more rows as the printed data frame is shifted on the right because of "big empty rownames"
+rownames.output.tables <- c(rownames.output.tables, paste0(rownames.output.tables[i]," ", collapse=""))
+}
+row.names(data) <- rownames.output.tables[1:length.rows]
+}else if(rownames.kept == FALSE & (all(class(data) == "table") | all(class(data) %in% c("matrix", "array")))){ # before R4.0.0, it was & all(class(data) %in% c("matrix", "table"))
+rownames(data) <- rep("", nrow(data)) # identical row names allowed in matrices and tables
+}
+if(noquote == TRUE){
+utils::capture.output(noquote(data), file=paste0(path, "/", output), append = no.overwrite)
+}else{
+utils::capture.output(data, file=paste0(path, "/", output), append = no.overwrite)
+}
+}else if(is.vector(data) & all(class(data) != "list") & (length(data) == 1 | vector.cat == TRUE)){
+if(noquote == TRUE){
+cat(noquote(data), file= paste0(path, "/", output), append = no.overwrite)
+}else{
+cat(data, file= paste0(path, "/", output), append = no.overwrite)
+}
+}else if(all(mode(data) == "character")){ # characters (array, list, factor or vector with vector.cat = FALSE)
+if(noquote == TRUE){
+utils::capture.output(noquote(data), file=paste0(path, "/", output), append = no.overwrite)
+}else{
+utils::capture.output(data, file=paste0(path, "/", output), append = no.overwrite)
+}
+}else{ # other object (S4 for instance, which do not like noquote()
+utils::capture.output(data, file=paste0(path, "/", output), append = no.overwrite)
+}
+sep.final <- paste0(rep("\n", sep), collapse = "")
+write(sep.final, file= paste0(path, "/", output), append = TRUE) # add a sep
+}
+}
+
+
+######## fun_get_message() #### return error/warning/other messages of an expression (that can be exported)
+
+
+# Check OK: clear to go Apollo
+fun_get_message <- function(
+data,
+kind = "error",
+header = TRUE,
+print.no = FALSE,
+text = NULL,
+env = NULL
+){
+# AIM
+# evaluate an instruction written between "" and return the first of the error, or warning or standard (non error non warning) messages if ever exist
+# using argument print.no = FALSE, return NULL if no message, which is convenient in some cases
+# WARNINGS
+# Only the first message is returned
+# Always use the env argument when fun_get_message() is used inside functions
+# ARGUMENTS
+# data: character string to evaluate
+# kind: character string. Either "error" to get error messages, or "warning" to get warning messages, or "message" to get non error and non warning messages
+# header: logical. Add a header in the returned message?
+# print.no: logical. Print a message saying that no message reported?
+# text: character string added to the output message (even if no message exists and print.no is TRUE). Inactivated if header is FALSE
+# env: the name of an existing environment. NULL if not required
+# RETURN
+# the message or NULL if no message and print.no is FALSE
+# REQUIRED PACKAGES
+# none
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_check()
+# EXAMPLES
+# fun_get_message(data = "wilcox.test(c(1,1,3), c(1, 2, 4), paired = TRUE)", kind = "error", print.no = TRUE, text = "IN A")
+# fun_get_message(data = "wilcox.test(c(1,1,3), c(1, 2, 4), paired = TRUE)", kind = "warning", print.no = TRUE, text = "IN A")
+# fun_get_message(data = "wilcox.test(c(1,1,3), c(1, 2, 4), paired = TRUE)", kind = "message", print.no = TRUE, text = "IN A")
+# fun_get_message(data = "wilcox.test()", kind = "error", print.no = TRUE, text = "IN A")
+# fun_get_message(data = "sum(1)", kind = "error", print.no = TRUE, text = "IN A")
+# fun_get_message(data = "message('ahah')", kind = "error", print.no = TRUE, text = "IN A")
+# fun_get_message(data = "message('ahah')", kind = "message", print.no = TRUE, text = "IN A")
+# fun_get_message(data = "ggplot2::ggplot(data = data.frame(X = 1:10, stringsAsFactors = TRUE), mapping = ggplot2::aes(x = X)) + ggplot2::geom_histogram()", kind = "message", print.no = TRUE, text = "IN FUNCTION 1")
+# set.seed(1) ; obs1 <- data.frame(Time = c(rnorm(10), rnorm(10) + 2), Group1 = rep(c("G", "H"), each = 10), stringsAsFactors = TRUE) ; fun_get_message(data = 'fun_gg_boxplot(data = obs1, y = "Time", categ = "Group1")', kind = "message", print.no = TRUE, text = "IN FUNCTION 1")
+# DEBUGGING
+# data = "wilcox.test(c(1,1,3), c(1, 2, 4), paired = TRUE)" ; kind = "warning" ; header = TRUE ; print.no = FALSE ; text = NULL ; env = NULL # for function debugging
+# data = "sum(1)" ; kind = "warning" ; header = TRUE ; print.no = FALSE ; text = NULL ; env = NULL # for function debugging
+# set.seed(1) ; obs1 <- data.frame(Time = c(rnorm(10), rnorm(10) + 2), Group1 = rep(c("G", "H"), each = 10), stringsAsFactors = TRUE) ; data = 'fun_gg_boxplot(data1 = obs1, y = "Time", categ = "Group1")' ; kind = "warning" ; header = TRUE ; print.no = FALSE ; text = NULL ; env = NULL # for function debugging
+# data = "message('ahah')" ; kind = "error" ; header = TRUE ; print.no = TRUE ; text = "IN A" ; env = NULL
+# data = 'ggplot2::ggplot(data = data.frame(X = "a", stringsAsFactors = TRUE), mapping = ggplot2::aes(x = X)) + ggplot2::geom_histogram()' ; kind = "message" ; header = TRUE ; print.no = FALSE ; text = NULL # for function debugging
+# data = 'ggplot2::ggplot(data = data.frame(X = "a", stringsAsFactors = TRUE), mapping = ggplot2::aes(x = X)) + ggplot2::geom_histogram()' ; kind = "warning" ; header = TRUE ; print.no = FALSE ; text = NULL # for function debugging
+# data = "emmeans::emmeans(object = emm.rg, specs = contrast.var)" ; kind = "message" ; header = TRUE ; print.no = FALSE ; text = NULL ; env = NULL # for function debugging
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+# end function name
+# required function checking
+if(length(utils::find("fun_check", mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end required function checking
+# no need to use reserved words to avoid bugs, because it is local, and exists("tempo.warning", inherit = FALSE), never use the scope
+# argument checking
+# argument checking with fun_check()
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = data, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = kind, options = c("error", "warning", "message"), length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = print.no, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = header, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(text)){
+tempo <- fun_check(data = text, class = "character", length = 1, fun.name = function.name) ; eval(ee)
+}
+if( ! is.null(env)){
+tempo <- fun_check(data = env, class = "environment", fun.name = function.name) ; eval(ee) #
+}
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# end argument checking with fun_check()
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument checking
+# main code
+pdf(file = NULL) # send plots into a NULL file, no pdf file created
+window.nb <- dev.cur()
+invisible(dev.set(window.nb))
+# last warning cannot be used because suppressWarnings() does not modify last.warning present in the base evironment (created at first warning in a new R session), or warnings() # to reset the warning history : unlockBinding("last.warning", baseenv()) ; assign("last.warning", NULL, envir = baseenv())
+output <- NULL
+tempo.error <- try(suppressMessages(suppressWarnings(eval(parse(text = data), envir = if(is.null(env)){parent.frame()}else{env}))), silent = TRUE) # get error message, not warning or messages
+if(any(class(tempo.error) %in% c("gg", "ggplot"))){
+tempo.error <- try(suppressMessages(suppressWarnings(ggplot2::ggplot_build(tempo.error))), silent = TRUE)[1]
+}
+if(exists("tempo.error", inherit = FALSE) == TRUE){ # inherit = FALSE avoid the portee lexical and thus the declared word
+if( ! all(class(tempo.error) == "try-error")){ # deal with NULL and S4 objects. Old code: ! (all(class(tempo.error) == "try-error") & any(grepl(x = tempo.error, pattern = "^Error|^error|^ERROR"))) but problem with S4 objects. Old code : if((length(tempo.error) > 0 & ! any(grepl(x = tempo.error, pattern = "^Error|^error|^ERROR"))) | (length(tempo.error) == 0) ){ but problem when tempo.error is a list but added this did not work: | ! all(class(tempo.error) == "character")
+tempo.error <- NULL
+}
+}else{
+tempo.error <- NULL
+}
+if(kind == "error" & ! is.null(tempo.error)){ #
+if(header == TRUE){
+tempo.error[1] <- gsub(x = tempo.error[1], pattern = "^Error i|^error i|^ERROR I", replacement = "I")
+output <- paste0("ERROR MESSAGE REPORTED", ifelse(is.null(text), "", " "), text, ":\n", tempo.error[1]) #
+}else{
+output <- tempo.error[1] #
+}
+}else if(kind == "error" & is.null(tempo.error) & print.no == TRUE){
+output <- paste0("NO ERROR MESSAGE REPORTED", ifelse(is.null(text), "", " "), text)
+}else if(kind != "error" & ( ! is.null(tempo.error)) & print.no == TRUE){
+output <- paste0("NO ", ifelse(kind == "warning", "WARNING", "STANDARD (NON ERROR AND NON WARNING)"), " MESSAGE BECAUSE OF ERROR MESSAGE REPORTED", ifelse(is.null(text), "", " "), text)
+}else if(is.null(tempo.error)){
+fun.warning.capture <- function(expr){
+# from demo(error.catching) typed in the R console, coming from ?tryCatch
+# see also http://mazamascience.com/WorkingWithData/?p=912
+# return a character string or NULL
+# expr <- wilcox.test.default(c(1, 1, 3), c(1, 2, 4), paired = TRUE)
+W <- NULL
+w.handler <- function(w){ # warning handler
+W <<- w # send to the above env, i.e., the inside of the fun.warning.capture function
+invokeRestart("muffleWarning") # here w.handler() muffles all the warnings. See http://romainfrancois.blog.free.fr/index.php?post/2009/05/20/Disable-specific-warnings to muffle specific warnings and print others
+}
+output <- list(
+value = suppressMessages(withCallingHandlers(tryCatch(expr, error = function(e){e}), warning = w.handler)), # BEWARE: w.handler is a function written without (), like in other functions with FUN argument
+warning = W # processed by w.handler()
+)
+return(if(is.null(output$warning)){NULL}else{as.character(output$warning)})
+}
+tempo.warn <- fun.warning.capture(eval(parse(text = data), envir = if(is.null(env)){parent.frame()}else{env}))
+# warn.options.ini <- options()$warn ; options(warn = 1) ; tempo.warn <- utils::capture.output({tempo <- suppressMessages(eval(parse(text = data), envir = if(is.null(env)){parent.frame()}else{env}))}, type = "message") ; options(warn = warn.options.ini) # this recover warnings not messages and not errors but does not work in all enviroments
+tempo.message <- utils::capture.output({
+tempo <- suppressMessages(suppressWarnings(eval(parse(text = data), envir = if(is.null(env)){parent.frame()}else{env})))
+if(any(class(tempo) %in% c("gg", "ggplot"))){
+tempo <- ggplot2::ggplot_build(tempo)
+}else{
+tempo <- suppressWarnings(eval(parse(text = data), envir = if(is.null(env)){parent.frame()}else{env}))
+}
+}, type = "message") # recover messages not warnings and not errors
+if(kind == "warning" & ! is.null(tempo.warn)){
+if(length(tempo.warn) > 0){ # to avoid character(0)
+if( ! any(sapply(tempo.warn, FUN = "grepl", pattern = "() FUNCTION:$"))){
+tempo.warn <- paste(unique(tempo.warn), collapse = "\n") # if FALSE, means that the tested data is a special function. If TRUE, means that the data is a standard function. In that case, the output of capture.output() is two strings per warning messages: if several warning messages -> identical first string, which is removed in next messages by unique()
+}else{
+tempo.warn <- paste(tempo.warn, collapse = "\n")
+}
+if(header == TRUE){
+if(any(grepl(x = tempo.warn[[1]], pattern = "^simpleWarning i"))){
+tempo.warn[[1]] <- gsub(x = tempo.warn[[1]], pattern = "^Warning i", replacement = "I")
+}
+if(any(grepl(x = tempo.warn[[1]], pattern = "^Warning i"))){
+tempo.warn[[1]] <- gsub(x = tempo.warn[[1]], pattern = "^Warning i", replacement = "I")
+}
+output <- paste0("WARNING MESSAGE REPORTED", ifelse(is.null(text), "", " "), text, ":\n", tempo.warn) #
+}else{
+output <- tempo.warn #
+}
+}else{
+if(print.no == TRUE){
+output <- paste0("NO WARNING MESSAGE REPORTED", ifelse(is.null(text), "", " "), text)
+} # no need else{} here because output is already NULL at first
+}
+}else if(kind == "warning" & is.null(tempo.warn) & print.no == TRUE){
+output <- paste0("NO WARNING MESSAGE REPORTED", ifelse(is.null(text), "", " "), text)
+}else if(kind == "message" & exists("tempo.message", inherit = FALSE) == TRUE){ # inherit = FALSE avoid the portee lexical and thus the declared word
+if(length(tempo.message) > 0){ # if something is returned by capture.ouptput() (only in this env) with a length more than 1
+if(header == TRUE){
+output <- paste0("STANDARD (NON ERROR AND NON WARNING) MESSAGE REPORTED", ifelse(is.null(text), "", " "), text, ":\n", tempo.message) #
+}else{
+output <- tempo.message #
+}
+}else{
+if(print.no == TRUE){
+output <- paste0("NO STANDARD (NON ERROR AND NON WARNING) MESSAGE REPORTED", ifelse(is.null(text), "", " "), text)
+} # no need else{} here because output is already NULL at first
+}
+}else if(kind == "message" & exists("tempo.message", inherit = FALSE) == FALSE & print.no == TRUE){
+output <- paste0("NO STANDARD (NON ERROR AND NON WARNING) MESSAGE REPORTED", ifelse(is.null(text), "", " "), text)
+} # no need else{} here because output is already NULL at first
+} # no need else{} here because output is already NULL at first
+invisible(dev.off(window.nb)) # end send plots into a NULL file
+return(output) # do not use cat() because the idea is to reuse the message
+}
+
+
+
+
+# add the new NA and NULL: remain all (slitherine, cute, anova, comat)
+# complete the fun_check(): slitherine anova, comat
+
+fun_gg_boxplot <- function(
+data1,
+y,
+categ,
+categ.class.order = NULL,
+categ.color = NULL,
+box.legend.name = NULL,
+box.fill = FALSE,
+box.width = 0.5,
+box.space = 0.1,
+box.line.size = 0.75,
+box.notch = FALSE,
+box.alpha = 1,
+box.mean = TRUE,
+box.whisker.kind = "std",
+box.whisker.width = 0,
+dot.color = grey(0.25),
+dot.categ = NULL,
+dot.categ.class.order = NULL,
+dot.legend.name = NULL,
+dot.tidy = FALSE,
+dot.tidy.bin.nb = 50,
+dot.jitter = 0.5,
+dot.seed = 2,
+dot.size = 3,
+dot.alpha = 0.5,
+dot.border.size = 0.5,
+dot.border.color = NULL,
+x.lab = NULL,
+x.angle = 0,
+y.lab = NULL,
+y.lim = NULL,
+y.log = "no",
+y.tick.nb = NULL,
+y.second.tick.nb = 1,
+y.include.zero = FALSE,
+y.top.extra.margin = 0.05,
+y.bottom.extra.margin = 0.05,
+stat.disp = "top",
+stat.disp.mean = FALSE,
+stat.size = 4,
+stat.dist = 5,
+stat.angle = 0,
+vertical = TRUE,
+text.size = 12,
+title = "",
+title.text.size = 8,
+legend.show = TRUE,
+legend.width = 0.5,
+article = TRUE,
+grid = FALSE,
+add = NULL,
+return = FALSE,
+return.ggplot = FALSE,
+return.gtable = TRUE,
+plot = TRUE,
+warn.print = FALSE,
+lib.path = NULL
+){
+# AIM
+# plot ggplot2 boxplots + dots + means
+# for ggplot2 specifications, see: https://ggplot2.tidyverse.org/articles/ggplot2-specs.html
+# WARNINGS
+# Rows containing NA in data1[, c(y, categ)] will be removed before processing, with a warning (see below)
+# Hinges are not computed like in the classical boxplot() function of R. See https://ggplot2.tidyverse.org/reference/geom_boxplot.html
+# To have a single box, please create a factor column with a single class and specify the name of this column in the categ argument. For a single set of grouped boxes, create a factor column with a single class and specify this column in categ argument as first element (i.e., as categ1, knowing that categ2 must also be specified in this situation). See categ argument below
+# The dot.alpha argument can alter the display of the color boxes when using pdf output
+# Size arguments (box.line.size, dot.size, dot.border.size, stat.size, text.size and title.text.size) are in mm. See Hadley comment in https://stackoverflow.com/questions/17311917/ggplot2-the-unit-of-size. See also http://sape.inf.usi.ch/quick-reference/ggplot2/size). Unit object are not accepted, but conversion can be used (e.g., grid::convertUnit(grid::unit(0.2, "inches"), "mm", valueOnly = TRUE))
+# Display seems to be done twice on Windows devices (like a blink). However, no double plots on pdf devices. Thus, the blink remains mysterious
+# To remove boxes and have only dots, use box.alpha = 0
+# ARGUMENTS
+# data1: data frame containing one column of quantitative values (see the y argument below) and one or two columns of categories (see the categ argument below). Duplicated column names are not allowed
+# y: character string of the data1 column name for y-axis (column containing numeric values). Numeric values will be split according to the classes of the column names indicated in the categ argument to generate the boxes and will also be used to plot the dots
+# categ: vector of character strings of the data1 column name for categories (column of characters or factors). Must be either one or two column names. If a single column name (further referred to as categ1), then one box per class of categ1. If two column names (further referred to as categ1 and categ2), then one box per class of categ2, which form a group of boxes in each class of categ1. WARNING: no empty classes allowed. To have a single box, create a factor column with a single class and specify the name of this column in the categ argument (here, no categ2 in categ argument). For a single set of grouped boxes, create a factor column with a single class and specify this column in categ argument as first element (i.e., as categ1), in addition to the already used category (as categ2 in this situation)
+# categ.class.order: list indicating the order of the classes of categ1 and categ2 represented on the boxplot (the first compartment for categ1 and and the second for categ2). If categ.class.order == NULL, classes are represented according to the alphabetical order. Some compartments can be NULL and others not. See the categ argument for categ1 and categ2 description
+# categ.color: vector of color character string for box frames (see the categ argument for categ1 and categ2 description)
+# If categ.color == NULL, default colors of ggplot2, whatever categ1 and categ2
+# If categ.color is non-null and only categ1 in categ argument, categ.color can be either:
+# (1) a single color string. All the boxes will have this color, whatever the number of classes of categ1
+# (2) a vector of string colors, one for each class of categ1. Each color will be associated according to categ.class.order of categ1
+# (3) a vector or factor of string colors, like if it was one of the column of data1 data frame. WARNING: a single color per class of categ1 and a single class of categ1 per color must be respected
+# Integers are also accepted instead of character strings, as long as above rules about length are respected. Integers will be processed by fun_gg_palette() using the maximal integer value among all the integers in categ.color (see fun_gg_palette())
+# If categ.color is non-null and categ1 and categ2 are specified, all the rules described above will apply to categ2 instead of categ1 (colors will be determined for boxes inside a group of boxes)
+# box.legend.name: character string of the legend title. If box.legend.name is NULL, then box.legend.name <- categ1 if only categ1 is present, and box.legend.name <- categ2 if categ1 and categ2 are present in the categ argument. Write "" if no legend required. See the categ argument for categ1 and categ2 description
+# box.fill: logical. Fill the box? If TRUE, the categ.color argument will be used to generate filled boxplots (the box frames being black) as well as filled outlier dots (the dot border being controlled by the dot.border.color argument). If all the dots are plotted (argument dot.color other than NULL), they will be over the boxes. If FALSE, the categ.color argument will be used to color the box frames and the outlier dot borders. If all the dots are plotted, they will be beneath the boxes
+# box.width: single numeric value (from 0 to 1) of width of either boxes or group of boxes
+# When categ argument has a single categ1 element (i.e., separate boxes. See the categ argument for categ1 and categ2 description), then each class of categ1 is represented by a single box. In that case, box.width argument defines each box width, from 0 (no box width) to 1 (max box width), but also the space between boxes (the code uses 1 - box.width for the box spaces). Of note, xmin and xmax of the fun_gg_boxplot() output report the box boundaries (around x-axis unit 1, 2, 3, etc., for each box)
+# When categ argument has a two categ1 and categ2 elements (i.e., grouped boxes), box.width argument defines the width allocated for each set of grouped boxes, from 0 (no group width) to 1 (max group width), but also the space between grouped boxes (the code uses 1 - box.width for the spaces). Of note, xmin and xmax of the fun_gg_boxplot() output report the box boundaries (around x-axis unit 1, 2, 3, etc., for each set of grouped box)
+# box.space: single numeric value (from 0 to 1) indicating the box separation inside grouped boxes, when categ argument has a two categ1 and categ2 elements. 0 means no space and 1 means boxes shrunk to a vertical line. Ignored if categ argument has a single categ1 element
+# box.line.size: single numeric value of line width of boxes and whiskers in mm
+# box.notch: logical. Notched boxplot? It TRUE, display notched boxplot, notches corresponding approximately to the 95% confidence interval of the median (the notch interval is exactly 1.58 x Inter Quartile Range (IQR) / sqrt(n), with n the number of values that made the box). If notch intervals between two boxes do not overlap, it can be interpreted as significant median differences
+# box.alpha: single numeric value (from 0 to 1) of box transparency (full transparent to full opaque, respectively). To remove boxplots, use box.alpha = 0
+# box.mean: logical. Add mean value? If TRUE, a diamond-shaped dot, with the horizontal diagonal corresponding to the mean value, is displayed over each boxplot
+# box.whisker.kind: range of the whiskers. Either "no" (no whiskers), or "std" (length of each whisker equal to 1.5 x Inter Quartile Range (IQR)), or "max" (length of the whiskers up or down to the most distant dot)
+# box.whisker.width: single numeric value (from 0 to 1) of the whisker width, with 0 meaning no whiskers and 1 meaning a width equal to the box width
+# dot.color: vector of color character string ruling the dot colors and the dot display. See the example section below for easier understanding of the rules described here
+# If NULL, no dots plotted
+# If "same", the dots will have the same colors as the respective boxplots
+# Otherwise, as in the rule (1), (2) or (3) described in the categ.color argument, except that in the possibility (3), the rule "a single color per class of categ and a single class of categ per color", does not have to be respected (for instance, each dot can have a different color). Colors will also depend on the dot.categ argument. If dot.categ is NULL, then colors will be applied to each class of the last column name specified in categ. If dot.categ is non-NULL, colors will be applied to each class of the column name specified in dot.categ. See examples
+# dot.categ: optional single character string of a column name (further referred to as categ3) of the data1 argument. This column of data1 will be used to generate a legend for dots, in addition to the legend for boxes. See the dot.color argument for details about the way the legend is built using the two dot.categ and dot.color arguments. If NULL, no legend created and the colors of dots will depend on dot.color and categ arguments (as explained in the dot.color argument)
+# dot.categ.class.order: optional vector of character strings indicating the order of the classes of categ3 (see the dot.categ argument). If dot.categ is non-NULL and dot.categ.class.order is NULL, classes are displayed in the legend according to the alphabetical order. Ignored if dot.categ is NULL
+# dot.legend.name: optional character string of the legend title for categ3 (see the dot.categ argument). If dot.legend.name == NULL, dot.categ value is used (name of the column in data1). Write "" if no legend required. Ignored if dot.categ is NULL
+# dot.tidy: logical. Nice dot spreading? If TRUE, use the geom_dotplot() function for a nice representation. WARNING: change the true quantitative coordinates of dots (i.e., y-axis values for vertical display) because of binning. Thus, the gain in aestheticism is associated with a loss in precision that can be very important. If FALSE, dots are randomly spread on the qualitative axis, using the dot.jitter argument (see below) keeping the true quantitative coordinates
+# dot.tidy.bin.nb: positive integer indicating the number of bins (i.e., nb of separations) of the y.lim range. Each dot will then be put in one of the bin, with a diameter of the width of the bin. In other words, increase the number of bins to have smaller dots. Not considered if dot.tidy is FALSE
+# dot.jitter: numeric value (from 0 to 1) of random dot horizontal dispersion (for vertical display), with 0 meaning no dispersion and 1 meaning dispersion in the corresponding box width interval. Not considered if dot.tidy is TRUE
+# dot.seed: integer value that set the random seed. Using the same number will generate the same dot jittering. Write NULL to have different jittering each time the same instruction is run. Ignored if dot.tidy is TRUE
+# dot.size: numeric value of dot diameter in mm. Not considered if dot.tidy is TRUE
+# dot.alpha: numeric value (from 0 to 1) of dot transparency (full transparent to full opaque, respectively)
+# dot.border.size: numeric value of border dot width in mm. Write zero for no dot border. If dot.tidy is TRUE, value 0 remove the border and other values leave the border without size control (geom_doplot() feature)
+# dot.border.color: single character color string defining the color of the dot border (same color for all the dots, whatever their categories). If dot.border.color == NULL, the border color will be the same as the dot color. A single integer is also accepted instead of a character string, that will be processed by fun_gg_palette()
+# x.lab: a character string or expression for x-axis legend. If NULL, character string of categ1 (see the categ argument for categ1 and categ2 description)
+# x.angle: integer value of the text angle for the x-axis numbers, using the same rules as in ggplot2. Positive values for counterclockwise rotation: 0 for horizontal, 90 for vertical, 180 for upside down etc. Negative values for clockwise rotation: 0 for horizontal, -90 for vertical, -180 for upside down etc.
+# y.lab: a character string or expression for y-axis legend. If NULL, character string of the y argument
+# y.lim: 2 numeric values indicating the range of the y-axis. Order matters (for inverted axis). If NULL, the range of the x column name of data1 will be used.
+# y.log: either "no", "log2" (values in the y argument column of the data1 data frame will be log2 transformed and y-axis will be log2 scaled) or "log10" (values in the y argument column of the data1 data frame will be log10 transformed and y-axis will be log10 scaled). WARNING: not possible to have horizontal boxes with a log axis, due to a bug in ggplot2 (see https://github.com/tidyverse/ggplot2/issues/881)
+# y.tick.nb: approximate number of desired values labeling the y-axis (i.e., main ticks, see the n argument of the the cute::fun_scale() function). If NULL and if y.log is "no", then the number of labeling values is set by ggplot2. If NULL and if y.log is "log2" or "log10", then the number of labeling values corresponds to all the exposant integers in the y.lim range (e.g., 10^1, 10^2 and 10^3, meaning 3 main ticks for y.lim = c(9, 1200)). WARNING: if non-NULL and if y.log is "log2" or "log10", labeling can be difficult to read (e.g., ..., 10^2, 10^2.5, 10^3, ...)
+ # y.second.tick.nb: number of desired secondary ticks between main ticks. Ignored if y.log is other than "no" (log scale plotted). Use argument return = TRUE and see $plot$y.second.tick.values to have the values associated to secondary ticks. IF NULL, no secondary ticks
+# y.include.zero: logical. Does y.lim range include 0? Ignored if y.log is "log2" or "log10"
+# y.top.extra.margin: single proportion (between 0 and 1) indicating if extra margins must be added to y.lim. If different from 0, add the range of the axis multiplied by y.top.extra.margin (e.g., abs(y.lim[2] - y.lim[1]) * y.top.extra.margin) to the top of y-axis
+# y.bottom.extra.margin: idem as y.top.extra.margin but to the bottom of y-axis
+# stat.disp: add the median number above the corresponding box. Either NULL (no number shown), "top" (at the top of the plot region) or "above" (above each box)
+# stat.disp.mean: logical. Display mean numbers instead of median numbers? Ignored if stat.disp is NULL
+# stat.size: numeric value of the stat font size in mm. Ignored if stat.disp is NULL
+# stat.dist: numeric value of the stat distance in percentage of the y-axis range (stat.dist = 5 means move the number displayed at 5% of the y-axis range). Ignored if stat.disp is NULL or "top"
+# stat.angle: integer value of the angle of stat, using the same rules as in ggplot2. Positive values for counterclockwise rotation: 0 for horizontal, 90 for vertical, 180 for upside down etc. Negative values for clockwise rotation: 0 for horizontal, -90 for vertical, -180 for upside down etc.
+# vertical: logical. Vertical boxes? WARNING: will be automatically set to TRUE if y.log argument is other than "no". Indeed, not possible to have horizontal boxes with a log axis, due to a bug in ggplot2 (see https://github.com/tidyverse/ggplot2/issues/881)
+# text.size: numeric value of the font size of the (1) axis numbers, (2) axis labels and (3) texts in the graphic legend (in mm)
+# title: character string of the graph title
+# title.text.size: numeric value of the title font size in mm
+# legend.show: logical. Show legend? Not considered if categ argument is NULL, because this already generate no legend, excepted if legend.width argument is non-NULL. In that specific case (categ is NULL, legend.show is TRUE and legend.width is non-NULL), an empty legend space is created. This can be useful when desiring graphs of exactly the same width, whatever they have legends or not
+# legend.width: single proportion (between 0 and 1) indicating the relative width of the legend sector (on the right of the plot) relative to the width of the plot. Value 1 means that the window device width is split in 2, half for the plot and half for the legend. Value 0 means no room for the legend, which will overlay the plot region. Write NULL to inactivate the legend sector. In such case, ggplot2 will manage the room required for the legend display, meaning that the width of the plotting region can vary between graphs, depending on the text in the legend
+# article: logical. If TRUE, use an article theme (article like). If FALSE, use a classic related ggplot theme. Use the add argument (e.g., add = "+ggplot2::theme_classic()" for the exact classic ggplot theme
+# grid: logical. Draw lines in the background to better read the box values? Not considered if article == FALSE (grid systematically present)
+# add: character string allowing to add more ggplot2 features (dots, lines, themes, facet, etc.). Ignored if NULL
+# WARNING: (1) the string must start with "+", (2) the string must finish with ")" and (3) each function must be preceded by "ggplot2::". Example: "+ ggplot2::coord_flip() + ggplot2::theme_bw()"
+# If the character string contains the "ggplot2::theme" string, then the article argument of fun_gg_boxplot() (see above) is ignored with a warning
+# Handle the add argument with caution since added functions can create conflicts with the preexisting internal ggplot2 functions
+# WARNING: the call of objects inside the quotes of add can lead to an error if the name of these objects are some of the fun_gg_scatter() arguments. Indeed, the function will use the internal argument instead of the global environment object. Example article <- "a" in the working environment and add = '+ ggplot2::ggtitle(article)'. The risk here is to have TRUE as title. To solve this, use add = '+ ggplot2::ggtitle(get("article", envir = .GlobalEnv))'
+# return: logical. Return the graph parameters?
+# return.ggplot: logical. Return the ggplot object in the output list? Ignored if return argument is FALSE. WARNING: always assign the fun_gg_boxplot() function (e.g., a <- fun_gg_boxplot()) if return.ggplot argument is TRUE, otherwise, double plotting is performed. See $ggplot in the RETURN section below for more details
+# return.gtable: logical. Return the ggplot object as gtable of grobs in the output list? Ignored if plot argument is FALSE. Indeed, the graph must be plotted to get the grobs dispositions. See $gtable in the RETURN section below for more details
+# plot: logical. Plot the graphic? If FALSE and return argument is TRUE, graphical parameters and associated warnings are provided without plotting
+# warn.print: logical. Print warnings at the end of the execution? ? If FALSE, warning messages are never printed, but can still be recovered in the returned list. Some of the warning messages (those delivered by the internal ggplot2 functions) are not apparent when using the argument plot = FALSE
+# lib.path: character string indicating the absolute path of the required packages (see below). if NULL, the function will use the R library default folders
+# REQUIRED PACKAGES
+# ggplot2
+# gridExtra
+# lemon (in case of use in the add argument)
+# scales
+# REQUIRED FUNCTIONS FROM THE cute PACKAGE
+# fun_check()
+# fun_comp_1d()
+# fun_comp_2d()
+# fun_gg_just()
+# fun_gg_palette()
+# fun_inter_ticks()
+# fun_name_change()
+# fun_pack()
+# fun_round()
+# fun_scale()
+# RETURN
+# a boxplot if plot argument is TRUE
+# a list of the graph info if return argument is TRUE:
+# $data: the initial data with graphic information added
+# $stat: the graphic statistics (mostly equivalent to ggplot_build()$data[[2]])
+# $removed.row.nb: which rows have been removed due to NA/Inf detection in y and categ columns (NULL if no row removed)
+# $removed.rows: removed rows (NULL if no row removed)
+# $plot: the graphic box and dot coordinates
+# $dots: dot coordinates
+# $main.box: coordinates of boxes
+# $median: median coordinates
+# $sup.whisker: coordinates of top whiskers (y for base and y.end for extremities)
+# $inf.whisker: coordinates of bottom whiskers (y for base and y.end for extremities)
+# $sup.whisker.edge: coordinates of top whisker edges (x and xend)
+# $inf.whisker.edge: coordinates of bottom whisker edges(x and xend)
+# $mean: diamond mean coordinates (only if box.mean argument is TRUE)
+# $stat.display: coordinates of stat numbers (only if stat.disp argument is not NULL)
+# y.second.tick.positions: coordinates of secondary ticks (only if y.second.tick.nb argument is non-NULL or if y.log argument is different from "no")
+# y.second.tick.values: values of secondary ticks. NULL except if y.second.tick.nb argument is non-NULL or if y.log argument is different from "no")
+# $panel: the variable names used for the panels (NULL if no panels). WARNING: NA can be present according to ggplot2 upgrade to v3.3.0
+# $axes: the x-axis and y-axis info
+# $warn: the warning messages. Use cat() for proper display. NULL if no warning. WARNING: warning messages delivered by the internal ggplot2 functions are not apparent when using the argument plot = FALSE
+# $ggplot: ggplot object that can be used for reprint (use print(...$ggplot) or update (use ...$ggplot + ggplot2::...). NULL if return.ggplot argument is FALSE. Of note, a non-NULL $ggplot in the output list is sometimes annoying as the manipulation of this list prints the plot
+# $gtable: gtable object that can be used for reprint (use gridExtra::grid.arrange(...$ggplot) or with additionnal grobs (see the grob decomposition in the examples). NULL if return.ggplot argument is FALSE. Contrary to $ggplot, a non-NULL $gtable in the output list is not annoying as the manipulation of this list does not print the plot
+# EXAMPLE
+# DEBUGGING
+# set.seed(1) ; obs1 <- data.frame(Time = c(rnorm(10), rnorm(10) + 2), Categ1 = rep(c("G", "H"), each = 10), stringsAsFactors = TRUE) ; set.seed(NULL) ; obs1$Time[1:10] <- NA ; data1 = obs1 ; y = "Time" ; categ = c("Categ1") ; categ.class.order = NULL ; box.legend.name = NULL ; categ.color = c("green") ; box.fill = FALSE ; box.width = 0.5 ; box.space = 0.1 ; box.notch = FALSE ; box.line.size = 0.5 ; box.alpha = 0.5 ; box.mean = TRUE ; box.whisker.kind = "std" ; box.whisker.width = 0.5 ; dot.color = "black" ; dot.categ = "Categ1"; dot.categ.class.order = c("G", "H") ; dot.legend.name = NULL ; dot.tidy = TRUE ; dot.tidy.bin.nb = 50 ; dot.jitter = 0.25 ; dot.size = 3 ; dot.alpha = 0.5 ; dot.border.size = 0.5 ; dot.border.color = NULL ; dot.seed = 2 ; y.lim = NULL ; y.log = "no" ; y.tick.nb = NULL ; y.second.tick.nb = NULL ; y.include.zero = FALSE ; y.top.extra.margin = 0.05 ; y.bottom.extra.margin = 0.05 ; stat.disp = NULL ; stat.disp.mean = FALSE ; stat.size = 4 ; stat.dist = 2 ; stat.angle = 0 ; x.lab = NULL ; y.lab = NULL ; vertical = TRUE ; text.size = 12 ; title = "" ; title.text.size = 8 ; legend.show = TRUE ; legend.width = 0.5 ; x.angle = 0 ; article = FALSE ; grid = FALSE ; return = TRUE ; return.ggplot = FALSE ; return.gtable = FALSE ; plot = TRUE ; add = NULL ; warn.print = FALSE ; lib.path = NULL
+# set.seed(1) ; obs1 <- data.frame(Time = c(rnorm(10), rnorm(10) + 2), Categ1 = rep(c("G", "H"), each = 10), Categ2 = rep(c("A", "B"), time = 10), Categ3 = rep(c("I", "J"), time = 10), stringsAsFactors = TRUE) ; set.seed(NULL) ; obs1$Time[1:10] <- NA ; data1 = obs1 ; y = "Time" ; categ = c("Categ1", "Categ2") ; categ.class.order = list(c("G", "H"), c("A", "B")); box.legend.name = NULL ; categ.color = c("green", "blue") ; box.fill = FALSE ; box.width = 0.5 ; box.space = 0.1 ; box.notch = FALSE ; box.line.size = 0.5 ; box.alpha = 0.5 ; box.mean = TRUE ; box.whisker.kind = "std" ; box.whisker.width = 0.5 ; dot.color = "black" ; dot.categ = "Categ1" ; dot.categ.class.order = NULL ; dot.legend.name = NULL ; dot.tidy = TRUE ; dot.tidy.bin.nb = 30 ; dot.jitter = 0.25 ; dot.size = 3 ; dot.alpha = 0.5 ; dot.border.size = 0.5 ; dot.border.color = NULL ; dot.seed = 2 ; y.lim = NULL ; y.log = "no" ; y.tick.nb = NULL ; y.second.tick.nb = NULL ; y.include.zero = FALSE ; y.top.extra.margin = 0.05 ; y.bottom.extra.margin = 0.05 ; stat.disp = NULL ; stat.disp.mean = FALSE ; stat.size = 4 ; stat.dist = 2 ; stat.angle = 0 ; x.lab = NULL ; y.lab = NULL ; vertical = TRUE ; text.size = 12 ; title = "" ; title.text.size = 8 ; legend.show = TRUE ; legend.width = 0.5 ; x.angle = 0 ; article = FALSE ; grid = FALSE ; return = FALSE ; return.ggplot = FALSE ; return.gtable = FALSE ; plot = TRUE ; add = NULL ; warn.print = FALSE ; lib.path = NULL
+# set.seed(1) ; obs1 <- data.frame(Time = c(rnorm(10), rnorm(10) + 2), Categ1 = rep(c("G", "H"), each = 10), Categ2 = rep(c("A", "B"), time = 10), stringsAsFactors = TRUE) ; set.seed(NULL) ; data1 = obs1 ; y = "Time" ; categ = c("Categ1") ; categ.class.order = list(c("H", "G")); box.legend.name = NULL ; categ.color = c("blue") ; box.fill = FALSE ; box.width = 0.5 ; box.space = 0.1 ; box.notch = TRUE ; box.line.size = 1 ; box.alpha = 1 ; box.mean = FALSE ; box.whisker.kind = "max" ; box.whisker.width = 0 ; dot.color = "black" ; dot.categ = "Categ1" ; dot.categ.class.order = NULL ; dot.legend.name = NULL ; dot.tidy = TRUE ; dot.tidy.bin.nb = 30 ; dot.jitter = 0.25 ; dot.size = 3 ; dot.alpha = 0.5 ; dot.border.size = 0.5 ; dot.border.color = NULL ; dot.seed = 2 ; y.lim = NULL ; y.log = "log10" ; y.tick.nb = NULL ; y.second.tick.nb = NULL ; y.include.zero = FALSE ; y.top.extra.margin = 0.05 ; y.bottom.extra.margin = 0.05 ; stat.disp = "above" ; stat.disp.mean = FALSE ; stat.size = 4 ; stat.dist = 2 ; stat.angle = 0 ; x.lab = NULL ; y.lab = NULL ; vertical = TRUE ; text.size = 12 ; title = "" ; title.text.size = 8 ; legend.width = 0.5 ; legend.show = TRUE ; x.angle = 0 ; article = FALSE ; grid = FALSE ; return = FALSE ; return.ggplot = FALSE ; return.gtable = FALSE ; plot = TRUE ; add = NULL ; warn.print = FALSE ; lib.path = NULL
+# function name
+function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
+arg.names <- names(formals(fun = sys.function(sys.parent(n = 2)))) # names of all the arguments
+arg.user.setting <- as.list(match.call(expand.dots = FALSE))[-1] # list of the argument settings (excluding default values not provided by the user)
+# end function name
+# required function checking
+req.function <- c(
+"fun_comp_2d",
+"fun_gg_just",
+"fun_gg_palette",
+"fun_name_change",
+"fun_pack",
+"fun_check",
+"fun_round",
+"fun_scale",
+"fun_inter_ticks"
+)
+for(i1 in req.function){
+if(length(find(i1, mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nREQUIRED ", i1, "() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# end required function checking
+# reserved words to avoid bugs (names of dataframe columns used in this function)
+reserved.words <- c("categ.check", "categ.color", "dot.color", "dot.categ", "dot.max", "dot.min", "group", "PANEL", "group.check", "MEAN", "tempo.categ1", "tempo.categ2", "text.max.pos", "text.min.pos", "x", "x.y", "y", "y.check", "y_from.dot.max", "ymax", "tidy_group")
+# end reserved words to avoid bugs (used in this function)
+# argument primary checking
+# arg with no default values
+if(any(missing(data1) | missing(y) | missing(categ))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ARGUMENTS angle AND pos HAVE NO DEFAULT VALUE AND REQUIRE ONE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end arg with no default values
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo <- fun_check(data = data1, class = "data.frame", na.contain = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = y, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = categ, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+if( ! is.null(categ.class.order)){
+tempo <- fun_check(data = categ.class.order, class = "list", fun.name = function.name) ; eval(ee)
+}
+if( ! is.null(box.legend.name)){
+tempo <- fun_check(data = box.legend.name, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+}
+if( ! is.null(categ.color)){
+tempo1 <- fun_check(data = categ.color, class = "vector", mode = "character", na.contain = TRUE, fun.name = function.name)
+tempo2 <- fun_check(data = categ.color, class = "factor", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.check.color <- fun_check(data = categ.color, class = "integer", double.as.integer.allowed = TRUE, na.contain = TRUE, fun.name = function.name)$problem
+if(tempo.check.color == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": categ.color ARGUMENT MUST BE A FACTOR OR CHARACTER VECTOR OR INTEGER VECTOR") # integer possible because dealt above
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+}
+tempo <- fun_check(data = box.fill, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = box.width, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = box.space, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = box.line.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = box.notch, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = box.alpha, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = box.mean, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = box.whisker.kind, options = c("no", "std", "max"), length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = box.whisker.width, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(dot.color)){
+tempo1 <- fun_check(data = dot.color, class = "vector", mode = "character", na.contain = TRUE, fun.name = function.name)
+tempo2 <- fun_check(data = dot.color, class = "factor", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.check.color <- fun_check(data = dot.color, class = "integer", double.as.integer.allowed = TRUE, na.contain = TRUE, fun.name = function.name)$problem
+if(tempo.check.color == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": dot.color MUST BE A FACTOR OR CHARACTER VECTOR OR INTEGER VECTOR") # integer possible because dealt above
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+}
+if( ! is.null(dot.categ)){
+tempo <- fun_check(data = dot.categ, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+}
+if( ! is.null(dot.categ.class.order)){
+tempo <- fun_check(data = dot.categ.class.order, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+}
+if( ! is.null(dot.legend.name)){
+tempo <- fun_check(data = dot.legend.name, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = dot.tidy, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = dot.tidy.bin.nb, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = dot.jitter, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(dot.seed)){
+tempo <- fun_check(data = dot.seed, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = TRUE, fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = dot.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = dot.alpha, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = dot.border.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+if( ! is.null(dot.border.color)){
+tempo1 <- fun_check(data = dot.border.color, class = "vector", mode = "character", length = 1, fun.name = function.name)
+tempo2 <- fun_check(data = dot.border.color, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, fun.name = function.name)
+if((tempo1$problem == TRUE & tempo2$problem == TRUE) | (tempo1$problem == FALSE & tempo2$problem == TRUE & ! (all(dot.border.color %in% colors() | grepl(pattern = "^#", dot.border.color))))){ # check that all strings of low.color start by #
+tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.border.color ARGUMENT MUST BE (1) A HEXADECIMAL COLOR STRING STARTING BY #, OR (2) A COLOR NAME GIVEN BY colors(), OR (3) AN INTEGER VALUE\nHERE IT IS: ", paste(unique(dot.border.color), collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+if( ! is.null(x.lab)){
+if(all(class(x.lab) %in% "expression")){ # to deal with math symbols
+tempo <- fun_check(data = x.lab, class = "expression", length = 1, fun.name = function.name) ; eval(ee)
+}else{
+tempo <- fun_check(data = x.lab, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+}
+}
+tempo <- fun_check(data = x.angle, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, neg.values = TRUE, fun.name = function.name) ; eval(ee)
+if( ! is.null(y.lab)){
+if(all(class(y.lab) %in% "expression")){ # to deal with math symbols
+tempo <- fun_check(data = y.lab, class = "expression", length = 1, fun.name = function.name) ; eval(ee)
+}else{
+tempo <- fun_check(data = y.lab, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+}
+}
+if( ! is.null(y.lim)){
+tempo <- fun_check(data = y.lim, class = "vector", mode = "numeric", length = 2, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & any(y.lim %in% c(Inf, -Inf))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": y.lim ARGUMENT CANNOT CONTAIN -Inf OR Inf VALUES")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+tempo <- fun_check(data = y.log, options = c("no", "log2", "log10"), length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(y.tick.nb)){
+tempo <- fun_check(data = y.tick.nb, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & y.tick.nb < 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": y.tick.nb ARGUMENT MUST BE A NON NULL POSITIVE INTEGER")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+if( ! is.null(y.second.tick.nb)){
+tempo <- fun_check(data = y.second.tick.nb, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & y.second.tick.nb <= 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": y.second.tick.nb ARGUMENT MUST BE A NON NULL POSITIVE INTEGER")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+tempo <- fun_check(data = y.include.zero, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = y.top.extra.margin, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = y.bottom.extra.margin, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(stat.disp)){
+tempo <- fun_check(data = stat.disp, options = c("top", "above"), length = 1, fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = stat.disp.mean, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = stat.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = stat.dist, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = stat.angle, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, neg.values = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = vertical, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = text.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = title, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = title.text.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = legend.show, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(legend.width)){
+tempo <- fun_check(data = legend.width, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = article, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = grid, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(add)){
+tempo <- fun_check(data = add, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = return, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = return.ggplot, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = return.gtable, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = plot, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = warn.print, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(lib.path)){
+tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
+tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+}
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between == #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end argument primary checking
+# second round of checking and data preparation
+# dealing with NA arguments
+tempo.arg <- names(arg.user.setting) # values provided by the user
+tempo.log <- sapply(lapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = is.na), FUN = any) & lapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = length) == 1 # no argument provided by the user can be just NA
+if(any(tempo.log) == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ":\n", ifelse(sum(tempo.log, na.rm = TRUE) > 1, "THESE ARGUMENTS\n", "THIS ARGUMENT\n"), paste0(tempo.arg[tempo.log], collapse = "\n"),"\nCANNOT JUST BE NA")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end dealing with NA arguments
+# dealing with NULL arguments
+tempo.arg <-c(
+"data1",
+"y",
+"categ",
+"box.fill",
+"box.width",
+"box.space",
+"box.line.size",
+"box.notch",
+"box.alpha",
+"box.mean",
+"box.whisker.kind",
+"box.whisker.width",
+# "dot.color", # inactivated because can be null
+"dot.tidy",
+"dot.tidy.bin.nb",
+"dot.jitter",
+"dot.size",
+"dot.alpha",
+"dot.border.size",
+"x.angle",
+"y.log",
+"y.include.zero",
+"y.top.extra.margin",
+"y.bottom.extra.margin",
+"stat.disp.mean",
+"stat.size",
+"stat.dist",
+"stat.angle",
+"vertical",
+"text.size",
+"title",
+"title.text.size",
+"legend.show",
+# "legend.width", # inactivated because can be null
+"article",
+"grid",
+"return",
+"return.ggplot",
+"return.gtable",
+"plot",
+"warn.print"
+)
+tempo.log <- sapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = is.null)
+if(any(tempo.log) == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ":\n", ifelse(sum(tempo.log, na.rm = TRUE) > 1, "THESE ARGUMENTS\n", "THIS ARGUMENT\n"), paste0(tempo.arg[tempo.log], collapse = "\n"),"\nCANNOT BE NULL")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end dealing with NULL arguments
+# code that protects set.seed() in the global environment
+# see also Protocol 100-rev0 Parallelization in R.docx
+if(exists(".Random.seed", envir = .GlobalEnv)){ # if .Random.seed does not exists, it means that no random operation has been performed yet in any R environment
+tempo.random.seed <- .Random.seed
+on.exit(assign(".Random.seed", tempo.random.seed, env = .GlobalEnv))
+}else{
+on.exit(set.seed(NULL)) # inactivate seeding -> return to complete randomness
+}
+set.seed(dot.seed)
+# end code that protects set.seed() in the global environment
+ini.warning.length <- options()$warning.length
+warn <- NULL
+warn.count <- 0
+if(any(duplicated(names(data1)))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nDUPLICATED COLUMN NAMES OF data1 ARGUMENT NOT ALLOWED:\n", paste(names(data1)[duplicated(names(data1))], collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! (y %in% names(data1))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\ny ARGUMENT MUST BE A COLUMN NAME OF data1")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+tempo <- fun_check(data = data1[, y], data.name = "y COLUMN OF data1", class = "vector", mode = "numeric", na.contain = TRUE, fun.name = function.name)
+if(tempo$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, "\ny ARGUMENT MUST BE NUMERIC COLUMN IN data1")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+if(length(categ) > 2){
+tempo.cat <- paste0("ERROR IN ", function.name, "\ncateg ARGUMENT CANNOT HAVE MORE THAN 2 COLUMN NAMES OF data1")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else if( ! all(categ %in% names(data1))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\ncateg ARGUMENT MUST BE COLUMN NAMES OF data1. HERE IT IS:\n", paste(categ, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# reserved word checking
+if(any(names(data1) %in% reserved.words)){
+if(any(duplicated(names(data1)))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nDUPLICATED COLUMN NAMES OF data1 ARGUMENT NOT ALLOWED:\n", paste(names(data1)[duplicated(names(data1))], collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! is.null(dot.categ)){
+if(dot.categ %in% categ){
+reserved.words <- c(reserved.words, paste0(dot.categ, "_DOT")) # paste0(dot.categ, "_DOT") is added to the reserved words because in such situation, a new column will be added to data1 that is named paste0(dot.categ, "_DOT")
+}
+}
+tempo.output <- fun_name_change(names(data1), reserved.words)
+for(i2 in 1:length(tempo.output$ini)){ # a loop to be sure to take the good ones
+names(data1)[names(data1) == tempo.output$ini[i2]] <- tempo.output$post[i2]
+if(any(y == tempo.output$ini[i2])){
+y[y == tempo.output$ini[i2]] <- tempo.output$post[i2]
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") IN y ARGUMENT (COLUMN NAMES OF data1 ARGUMENT),\n", tempo.output$ini[i2], " HAS BEEN REPLACED BY ", tempo.output$post[i2], "\nBECAUSE RISK OF BUG AS SOME NAMES IN y ARGUMENT ARE RESERVED WORD USED BY THE ", function.name, " FUNCTION")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+# WARNING: names of y argument potentially replaced
+if(any(categ == tempo.output$ini[i2])){
+categ[categ == tempo.output$ini[i2]] <- tempo.output$post[i2]
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") IN categ ARGUMENT (COLUMN NAMES OF data1 ARGUMENT),\n", tempo.output$ini[i2], " HAS BEEN REPLACED BY ", tempo.output$post[i2], "\nBECAUSE RISK OF BUG AS SOME NAMES IN categ ARGUMENT ARE RESERVED WORD USED BY THE ", function.name, " FUNCTION")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+# WARNING: names of categ argument potentially replaced
+if( ! is.null(dot.categ)){
+if(any(dot.categ == tempo.output$ini[i2])){
+dot.categ[dot.categ == tempo.output$ini[i2]] <- tempo.output$post[i2]
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") IN dot.categ ARGUMENT (COLUMN NAMES OF data1 ARGUMENT),\n", tempo.output$ini[i2], " HAS BEEN REPLACED BY ", tempo.output$post[i2], "\nBECAUSE RISK OF BUG AS SOME NAMES IN dot.categ ARGUMENT ARE RESERVED WORD USED BY THE ", function.name, " FUNCTION")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+# WARNING: names of dot.categ argument potentially replaced
+}
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") REGARDING COLUMN NAMES REPLACEMENT, THE NAMES\n", paste(tempo.output$ini, collapse = " "), "\nHAVE BEEN REPLACED BY\n", paste(tempo.output$post, collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+if( ! (is.null(add) | is.null(tempo.output$ini))){
+if(grepl(x = add, pattern = paste(tempo.output$ini, collapse = "|"))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nDETECTION OF COLUMN NAMES OF data1 IN THE add ARGUMENT STRING, THAT CORRESPOND TO RESERVED STRINGS FOR ", function.name, "\nCOLUMN NAMES HAVE TO BE CHANGED\nTHE PROBLEMATIC COLUMN NAMES ARE SOME OF THESE NAMES:\n", paste(tempo.output$ini, collapse = " "), "\nIN THE DATA FRAME OF data1 AND IN THE STRING OF add ARGUMENT, TRY TO REPLACE NAMES BY:\n", paste(tempo.output$post, collapse = " "), "\n\nFOR INFORMATION, THE RESERVED WORDS ARE:\n", paste(reserved.words, collapse = "\n"))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+}
+if( ! (is.null(add))){
+if(any(sapply(X = arg.names, FUN = grepl, x = add))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NAMES OF ", function.name, " ARGUMENTS DETECTED IN THE add STRING:\n", paste(arg.names[sapply(X = arg.names, FUN = grepl, x = add)], collapse = "\n"), "\nRISK OF WRONG OBJECT USAGE INSIDE ", function.name)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+# end reserved word checking
+# verif of add
+if( ! is.null(add)){
+if( ! grepl(pattern = "^\\s*\\+", add)){ # check that the add string start by +
+tempo.cat <- paste0("ERROR IN ", function.name, ": add ARGUMENT MUST START WITH \"+\": ", paste(unique(add), collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else if( ! grepl(pattern = "(ggplot2|lemon)\\s*::", add)){ #
+tempo.cat <- paste0("ERROR IN ", function.name, ": FOR EASIER FUNCTION DETECTION, add ARGUMENT MUST CONTAIN \"ggplot2::\" OR \"lemon::\" IN FRONT OF EACH GGPLOT2 FUNCTION: ", paste(unique(add), collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else if( ! grepl(pattern = ")\\s*$", add)){ # check that the add string finished by )
+tempo.cat <- paste0("ERROR IN ", function.name, ": add ARGUMENT MUST FINISH BY \")\": ", paste(unique(add), collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# end verif of add
+# management of add containing facet
+facet.categ <- NULL
+if( ! is.null(add)){
+facet.check <- TRUE
+tempo <- unlist(strsplit(x = add, split = "\\s*\\+\\s*(ggplot2|lemon)\\s*::\\s*")) #
+tempo <- sub(x = tempo, pattern = "^facet_wrap", replacement = "ggplot2::facet_wrap")
+tempo <- sub(x = tempo, pattern = "^facet_grid", replacement = "ggplot2::facet_grid")
+tempo <- sub(x = tempo, pattern = "^facet_rep", replacement = "lemon::facet_rep")
+if(any(grepl(x = tempo, pattern = "ggplot2::facet_wrap|lemon::facet_rep_wrap"))){
+tempo1 <- suppressWarnings(eval(parse(text = tempo[grepl(x = tempo, pattern = "ggplot2::facet_wrap|lemon::facet_rep_wrap")])))
+facet.categ <- names(tempo1$params$facets)
+tempo.text <- "facet_wrap OR facet_rep_wrap"
+facet.check <- FALSE
+}else if(grepl(x = add, pattern = "ggplot2::facet_grid|lemon::facet_rep_grid")){
+tempo1 <- suppressWarnings(eval(parse(text = tempo[grepl(x = tempo, pattern = "ggplot2::facet_grid|lemon::facet_rep_grid")])))
+facet.categ <- c(names(tempo1$params$rows), names(tempo1$params$cols))
+tempo.text <- "facet_grid OR facet_rep_grid"
+facet.check <- FALSE
+}
+if(facet.check == FALSE & ! all(facet.categ %in% names(data1))){ # WARNING: all(facet.categ %in% names(data1)) is TRUE when facet.categ is NULL
+tempo.cat <- paste0("ERROR IN ", function.name, "\nDETECTION OF \"", tempo.text, "\" STRING IN THE add ARGUMENT BUT PROBLEM OF VARIABLE DETECTION (COLUMN NAMES OF data1)\nTHE DETECTED VARIABLES ARE:\n", paste(facet.categ, collapse = " "), "\nTHE data1 COLUMN NAMES ARE:\n", paste(names(data1), collapse = " "), "\nPLEASE REWRITE THE add STRING AND RERUN")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# end management of add containing facet
+# conversion of categ columns in data1 into factors
+for(i1 in 1:length(categ)){
+tempo1 <- fun_check(data = data1[, categ[i1]], data.name = paste0("categ NUMBER ", i1, " OF data1"), class = "vector", mode = "character", na.contain = TRUE, fun.name = function.name)
+tempo2 <- fun_check(data = data1[, categ[i1]], data.name = paste0("categ NUMBER ", i1, " OF data1"), class = "factor", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, "\n", paste0("categ NUMBER ", i1, " OF data1"), " MUST BE A FACTOR OR CHARACTER VECTOR")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else if(tempo1$problem == FALSE){ # character vector
+if(box.alpha != 0){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") IN categ NUMBER ", i1, " IN data1, THE CHARACTER COLUMN HAS BEEN CONVERTED TO FACTOR, WITH LEVELS ACCORDING TO THE ALPHABETICAL ORDER")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+data1[, categ[i1]] <- factor(data1[, categ[i1]]) # if already a factor, change nothing, if characters, levels according to alphabetical order
+}
+# OK: all the categ columns of data1 are factors from here
+# end conversion of categ columns in data1 into factors
+
+
+
+# management of log scale and Inf removal
+if(any(( ! is.finite(data1[, y])) & ( ! is.na(data1[, y])))){ # is.finite also detects NA: ( ! is.finite(data1[, y])) & ( ! is.na(data1[, y])) detects only Inf
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") PRESENCE OF -Inf OR Inf VALUES IN THE ", y, " COLUMN OF THE data1 ARGUMENT AND CORRESPONDING ROWS REMOVED (SEE $removed.row.nb AND $removed.rows)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+data1.ini <- data1 # strictly identical to data1 except that in data1 y is log converted if and only if y.log != "no"
+if(y.log != "no"){
+tempo1 <- ! is.finite(data1[, y]) # where are initial NA and Inf
+data1[, y] <- suppressWarnings(get(y.log)(data1[, y]))# no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
+if(any( ! (tempo1 | is.finite(data1[, y])))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") LOG CONVERSION INTRODUCED -Inf OR Inf OR NaN VALUES IN THE ", y, " COLUMN OF THE data1 ARGUMENT AND CORRESPONDING ROWS REMOVED (SEE $removed.row.nb AND $removed.rows)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+# Inf removal
+if(any(( ! is.finite(data1[, y])) & ( ! is.na(data1[, y])))){ # is.finite also detects NA: ( ! is.finite(data1[, y])) & ( ! is.na(data1[, y])) detects only Inf
+removed.row.nb <- which(( ! is.finite(data1[, y])) & ( ! is.na(data1[, y])))
+removed.rows <- data1.ini[removed.row.nb, ] # here data1.ini used to have the y = O rows that will be removed because of Inf creation after log transformation
+data1 <- data1[-removed.row.nb, ] #
+data1.ini <- data1.ini[-removed.row.nb, ] #
+}else{
+removed.row.nb <- NULL
+removed.rows <- data.frame(stringsAsFactors = FALSE)
+}
+# From here, data1 and data.ini have no more Inf
+# end Inf removal
+if(y.log != "no" & ! is.null(y.lim)){
+if(any(y.lim <= 0)){
+tempo.cat <- paste0("ERROR IN ", function.name, "\ny.lim ARGUMENT CANNOT HAVE ZERO OR NEGATIVE VALUES WITH THE y.log ARGUMENT SET TO ", y.log, ":\n", paste(y.lim, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else if(any( ! is.finite(if(y.log == "log10"){log10(y.lim)}else{log2(y.lim)}))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\ny.lim ARGUMENT RETURNS INF/NA WITH THE y.log ARGUMENT SET TO ", y.log, "\nAS SCALE COMPUTATION IS ", ifelse(y.log == "log10", "log10", "log2"), ":\n", paste(if(y.log == "log10"){log10(y.lim)}else{log2(y.lim)}, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+if(y.log != "no" & y.include.zero == TRUE){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") y.log ARGUMENT SET TO ", y.log, " AND y.include.zero ARGUMENT SET TO TRUE -> y.include.zero ARGUMENT RESET TO FALSE BECAUSE 0 VALUE CANNOT BE REPRESENTED IN LOG SCALE")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+y.include.zero <- FALSE
+}
+if(y.log != "no" & vertical == FALSE){
+vertical <- TRUE
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") BECAUSE OF A BUG IN ggplot2, CANNOT FLIP BOXES HORIZONTALLY WITH A Y.LOG SCALE -> vertical ARGUMENT RESET TO TRUE")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+# end management of log scale and Inf removal
+# na detection and removal (done now to be sure of the correct length of categ)
+column.check <- unique(c(y, categ, if( ! is.null(dot.color) & ! is.null(dot.categ)){dot.categ}, if( ! is.null(facet.categ)){facet.categ})) # dot.categ because can be a 3rd column of data1, categ.color and dot.color will be tested later
+if(any(is.na(data1[, column.check]))){ # data1 used here instead of data1.ini in case of new NaN created by log conversion (neg values)
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NA DETECTED IN COLUMNS OF data1 AND CORRESPONDING ROWS REMOVED (SEE $removed.row.nb AND $removed.rows)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+for(i2 in 1:length(column.check)){
+if(any(is.na(data1[, column.check[i2]]))){
+tempo.warn <- paste0("NA REMOVAL DUE TO COLUMN ", column.check[i2], " OF data1")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n", tempo.warn)))
+}
+}
+tempo <- unique(unlist(lapply(lapply(c(data1[column.check]), FUN = is.na), FUN = which)))
+removed.row.nb <- c(removed.row.nb, tempo) # removed.row.nb created to remove Inf
+removed.rows <- rbind(removed.rows, data1.ini[tempo, ], stringsAsFactors = FALSE) # here data1.ini used to have the non NA rows that will be removed because of NAN creation after log transformation (neg values for instance)
+column.check <- column.check[ ! column.check == y] # remove y to keep quali columns
+if(length(tempo) != 0){
+data1 <- data1[-tempo, ] # WARNING tempo here and not removed.row.nb because the latter contain more numbers thant the former
+data1.ini <- data1.ini[-tempo, ] # WARNING tempo here and not removed.row.nb because the latter contain more numbers than the former
+for(i3 in 1:length(column.check)){
+if(any( ! unique(removed.rows[, column.check[i3]]) %in% unique(data1[, column.check[i3]]))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") IN COLUMN ", column.check[i3], " OF data1, THE FOLLOWING CLASSES HAVE DISAPPEARED AFTER NA/Inf REMOVAL (IF COLUMN USED IN THE PLOT, THIS CLASS WILL NOT BE DISPLAYED):\n", paste(unique(removed.rows[, column.check[i3]])[ ! unique(removed.rows[, column.check[i3]]) %in% unique(data1[, column.check[i3]])], collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+}
+count.categ <- 0
+for(i2 in 1:length(column.check)){
+if(column.check[i2] %in% categ){
+count.categ <- count.categ + 1
+}
+if(column.check[i2] == categ[count.categ]){
+categ.class.order[count.categ] <- list(levels(data1[, column.check[i2]])[levels(data1[, column.check[i2]]) %in% unique(data1[, column.check[i2]])]) # remove the absent color in the character vector
+data1[, column.check[i2]] <- factor(as.character(data1[, column.check[i2]]), levels = unique(categ.class.order[[count.categ]]))
+}
+if( ! is.null(dot.color) & ! is.null(dot.categ)){ # reminder : dot.categ cannot be a column name of categ anymore (because in that case dot.categ name is changed into "..._DOT"
+if(column.check[i2] == dot.categ){
+dot.categ.class.order <- levels(data1[, column.check[i2]])[levels(data1[, column.check[i2]]) %in% unique(data1[, column.check[i2]])] # remove the absent color in the character vector
+data1[, column.check[i2]] <- factor(as.character(data1[, column.check[i2]]), levels = unique(dot.categ.class.order))
+}
+}
+if(column.check[i2] %in% facet.categ){ # works if facet.categ == NULL this method should keep the order of levels when removing some levels
+tempo.levels <- levels(data1[, column.check[i2]])[levels(data1[, column.check[i2]]) %in% unique(as.character(data1[, column.check[i2]]))]
+data1[, column.check[i2]] <- factor(as.character(data1[, column.check[i2]]), levels = tempo.levels)
+}
+}
+}
+# end na detection and removal (done now to be sure of the correct length of categ)
+# From here, data1 and data.ini have no more NA or NaN in y, categ, dot.categ (if dot.color != NULL) and facet.categ
+
+
+
+if( ! is.null(categ.class.order)){
+if(length(categ.class.order) != length(categ)){
+tempo.cat <- paste0("ERROR IN ", function.name, "\ncateg.class.order ARGUMENT MUST BE A LIST OF LENGTH EQUAL TO LENGTH OF categ\nHERE IT IS LENGTH: ", length(categ.class.order), " VERSUS ", length(categ))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+for(i3 in 1:length(categ.class.order)){
+if(is.null(categ.class.order[[i3]])){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE categ.class.order COMPARTMENT ", i3, " IS NULL. ALPHABETICAL ORDER WILL BE APPLIED")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+data1[, categ[i3]] <- factor(as.character(data1[, categ[i3]])) # if already a factor, change nothing, if characters, levels according to alphabetical order
+categ.class.order[[i3]] <- levels(data1[, categ[i3]]) # character vector that will be used later
+}else{
+tempo <- fun_check(data = categ.class.order[[i3]], data.name = paste0("COMPARTMENT ", i3 , " OF categ.class.order ARGUMENT"), class = "vector", mode = "character", length = length(levels(data1[, categ[i3]])), fun.name = function.name) # length(data1[, categ[i1]) -> if data1[, categ[i1] was initially character vector, then conversion as factor after the NA removal, thus class number ok. If data1[, categ[i1] was initially factor, no modification after the NA removal, thus class number ok
+if(tempo$problem == TRUE){
+stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+if(any(duplicated(categ.class.order[[i3]]))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nCOMPARTMENT ", i3, " OF categ.class.order ARGUMENT CANNOT HAVE DUPLICATED CLASSES: ", paste(categ.class.order[[i3]], collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else if( ! (all(categ.class.order[[i3]] %in% unique(data1[, categ[i3]])) & all(unique(data1[, categ[i3]]) %in% categ.class.order[[i3]]))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nCOMPARTMENT ", i3, " OF categ.class.order ARGUMENT MUST BE CLASSES OF ELEMENT ", i3, " OF categ ARGUMENT\nHERE IT IS:\n", paste(categ.class.order[[i3]], collapse = " "), "\nFOR COMPARTMENT ", i3, " OF categ.class.order AND IT IS:\n", paste(unique(data1[, categ[i3]]), collapse = " "), "\nFOR COLUMN ", categ[i3], " OF data1")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+data1[, categ[i3]] <- factor(data1[, categ[i3]], levels = categ.class.order[[i3]]) # reorder the factor
+
+}
+names(categ.class.order)[i3] <- categ[i3]
+}
+}
+}else{
+categ.class.order <- vector("list", length = length(categ))
+tempo.categ.class.order <- NULL
+for(i2 in 1:length(categ.class.order)){
+categ.class.order[[i2]] <- levels(data1[, categ[i2]])
+names(categ.class.order)[i2] <- categ[i2]
+tempo.categ.class.order <- c(tempo.categ.class.order, ifelse(i2 != 1, "\n", ""), categ.class.order[[i2]])
+}
+if(box.alpha != 0){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE categ.class.order SETTING IS NULL. ALPHABETICAL ORDER WILL BE APPLIED FOR BOX ORDERING:\n", paste(tempo.categ.class.order, collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+# categ.class.order not NULL anymore (list)
+if(is.null(box.legend.name) & box.alpha != 0){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE box.legend.name SETTING IS NULL. NAMES OF categ WILL BE USED: ", paste(categ, collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+box.legend.name <- categ[length(categ)] # if only categ1, then legend name of categ1, if length(categ) == 2, then legend name of categ2
+}
+# box.legend.name not NULL anymore (character string)
+# management of categ.color
+if( ! is.null(categ.color)){
+# check the nature of color
+# integer colors into gg_palette
+tempo.check.color <- fun_check(data = categ.color, class = "integer", double.as.integer.allowed = TRUE, na.contain = TRUE, fun.name = function.name)$problem
+if(tempo.check.color == FALSE){
+# convert integers into colors
+categ.color <- fun_gg_palette(max(categ.color, na.rm = TRUE))[categ.color]
+}
+# end integer colors into gg_palette
+if( ! (all(categ.color %in% colors() | grepl(pattern = "^#", categ.color)))){ # check that all strings of low.color start by #
+tempo.cat <- paste0("ERROR IN ", function.name, "\ncateg.color ARGUMENT MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # AND/OR COLOR NAMES GIVEN BY colors(): ", paste(unique(categ.color), collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(is.na(categ.color)) & box.alpha != 0){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") categ.color ARGUMENT CONTAINS NA")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+# end check the nature of color
+# check the length of color
+categ.len <- length(categ) # if only categ1, then colors for classes of categ1, if length(categ) == 2, then colors for classes of categ2
+if(length(categ.color) == length(levels(data1[, categ[categ.len]]))){ # here length(categ.color) is equal to the different number of categ
+# data1[, categ[categ.len]] <- factor(data1[, categ[categ.len]]) # not required because sure that is is a factor
+data1 <- data.frame(data1, categ.color = data1[, categ[categ.len]], stringsAsFactors = TRUE) # no need stringsAsFactors here for stat.nolog as factors remain factors
+data1$categ.color <- factor(data1$categ.color, labels = categ.color)
+if(box.alpha != 0){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") IN ", categ[categ.len], " OF categ ARGUMENT, THE FOLLOWING COLORS:\n", paste(categ.color, collapse = " "), "\nHAVE BEEN ATTRIBUTED TO THESE CLASSES:\n", paste(levels(factor(data1[, categ[categ.len]])), collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else if(length(categ.color) == length(data1[, categ[categ.len]])){# here length(categ.color) is equal to nrow(data1) -> Modif to have length(categ.color) equal to the different number of categ (length(categ.color) == length(levels(data1[, categ[categ.len]])))
+data1 <- data.frame(data1, categ.color = categ.color, stringsAsFactors = TRUE)
+tempo.check <- unique(data1[ , c(categ[categ.len], "categ.color")])
+if( ! (nrow(tempo.check) == length(unique(categ.color)) & nrow(tempo.check) == length(unique(data1[ , categ[categ.len]])))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\ncateg.color ARGUMENT HAS THE LENGTH OF data1 ROW NUMBER\nBUT IS INCORRECTLY ASSOCIATED TO EACH CLASS OF categ ", categ[categ.len], ":\n", paste(unique(mapply(FUN = "paste", data1[ ,categ[categ.len]], data1[ ,"categ.color"])), collapse = "\n"))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+# data1[, categ[categ.len]] <- factor(data1[, categ[categ.len]]) # not required because sure that is is a factor
+categ.color <- unique(data1$categ.color[order(data1[, categ[categ.len]])]) # Modif to have length(categ.color) equal to the different number of categ (length(categ.color) == length(levels(data1[, categ[categ.len]])))
+if(box.alpha != 0){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") categ.color ARGUMENT HAS THE LENGTH OF data1 ROW NUMBER\nCOLORS HAVE BEEN RESPECTIVELY ASSOCIATED TO EACH CLASS OF categ ", categ[categ.len], " AS:\n", paste(levels(factor(data1[, categ[categ.len]])), collapse = " "), "\n", paste(categ.color, collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+}else if(length(categ.color) == 1){
+# data1[, categ[categ.len]] <- factor(data1[, categ[categ.len]]) # not required because sure that is is a factor
+data1 <- data.frame(data1, categ.color = categ.color, stringsAsFactors = TRUE)
+categ.color <- rep(categ.color, length(levels(data1[, categ[categ.len]])))
+if(box.alpha != 0){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") categ.color ARGUMENT HAS LENGTH 1, MEANING THAT ALL THE DIFFERENT CLASSES OF ", categ[categ.len], "\n", paste(levels(factor(data1[, categ[categ.len]])), collapse = " "), "\nWILL HAVE THE SAME COLOR\n", paste(categ.color, collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, "\ncateg.color ARGUMENT MUST BE (1) LENGTH 1, OR (2) THE LENGTH OF data1 NROWS AFTER NA/Inf REMOVAL, OR (3) THE LENGTH OF THE CLASSES IN THE categ ", categ[categ.len], " COLUMN. HERE IT IS COLOR LENGTH ", length(categ.color), " VERSUS CATEG LENGTH ", length(data1[, categ[categ.len]]), " AND CATEG CLASS LENGTH ", length(unique(data1[, categ[categ.len]])), "\nPRESENCE OF NA/Inf COULD BE THE PROBLEM")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}else{
+categ.len <- length(categ) # if only categ1, then colors for classes of categ1, if length(categ) == 2, then colors for classes of categ2
+# data1[, categ[categ.len]] <- factor(data1[, categ[categ.len]]) # not required because sure that is is a factor
+categ.color <- fun_gg_palette(length(levels(data1[, categ[categ.len]])))
+data1 <- data.frame(data1, categ.color = data1[, categ[categ.len]], stringsAsFactors = TRUE)
+data1$categ.color <- factor(data1$categ.color, labels = categ.color)
+if(box.alpha != 0){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NULL categ.color ARGUMENT -> COLORS RESPECTIVELY ATTRIBUTED TO EACH CLASS OF ", categ[categ.len], " IN data1:\n", paste(categ.color, collapse = " "), "\n", paste(levels(data1[, categ[categ.len]]), collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+# categ.color not NULL anymore
+categ.color <- as.character(categ.color)
+# categ.color is a character string representing the diff classes
+data1$categ.color <- factor(data1$categ.color, levels = unique(categ.color)) # ok because if categ.color is a character string, the order make class 1, class 2, etc. unique() because no duplicates allowed
+# data1$categ.color is a factor with order of levels -> categ.color
+# end management of categ.color
+# management of dot.color
+if( ! is.null(dot.color)){
+# optional legend of dot colors
+if( ! is.null(dot.categ)){
+ini.dot.categ <- dot.categ
+if( ! dot.categ %in% names(data1)){ # no need to use all() because length(dot.categ) = 1
+tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.categ ARGUMENT MUST BE A COLUMN NAME OF data1. HERE IT IS:\n", dot.categ)
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else if(dot.categ %in% categ){ # no need to use all() because length(dot.categ) = 1. Do not use dot.categ %in% categ[length(categ)] -> error
+# management of dot legend if dot.categ %in% categ (because legends with the same name are joined in ggplot2)
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE COLUMN NAME OF data1 INDICATED IN THE dot.categ ARGUMENT (", dot.categ, ") HAS BEEN REPLACED BY ", paste0(dot.categ, "_DOT"), " TO AVOID MERGED LEGEND BY GGPLOT2")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+data1 <- data.frame(data1, dot.categ = data1[, dot.categ], stringsAsFactors = TRUE) # dot.categ is not a column name of data1 (checked above with reserved words)
+dot.categ <- paste0(dot.categ, "_DOT")
+names(data1)[names(data1) == "dot.categ"] <- dot.categ # paste0(dot.categ, "_DOT") is not a column name of data1 (checked above with reserved words)
+# tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.categ ARGUMENT CANNOT BE A COLUMN NAME OF data1 ALREADY SPECIFIED IN THE categ ARGUMENT:\n", dot.categ, "\nINDEED, dot.categ ARGUMENT IS MADE TO HAVE MULTIPLE DOT COLORS NOT RELATED TO THE BOXPLOT CATEGORIES")
+# stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+tempo1 <- fun_check(data = data1[, dot.categ], data.name = paste0(dot.categ, " COLUMN OF data1"), class = "vector", mode = "character", na.contain = TRUE, fun.name = function.name)
+tempo2 <- fun_check(data = data1[, dot.categ], data.name = paste0(dot.categ, " COLUMN OF data1"), class = "factor", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.categ COLUMN MUST BE A FACTOR OR CHARACTER VECTOR") #
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+data1[, dot.categ] <- factor(data1[, dot.categ]) # if already a factor, change nothing, if characters, levels according to alphabetical order
+# dot.categ column of data1 is factor from here
+if( ! is.null(dot.categ.class.order)){
+if(any(duplicated(dot.categ.class.order))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.categ.class.order ARGUMENT CANNOT HAVE DUPLICATED CLASSES: ", paste(dot.categ.class.order, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else if( ! (all(dot.categ.class.order %in% levels(data1[, dot.categ])) & all(levels(data1[, dot.categ]) %in% dot.categ.class.order))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.categ.class.order ARGUMENT MUST BE CLASSES OF dot.categ ARGUMENT\nHERE IT IS:\n", paste(dot.categ.class.order, collapse = " "), "\nFOR dot.categ.class.order AND IT IS:\n", paste(levels(data1[, dot.categ]), collapse = " "), "\nFOR dot.categ COLUMN (", ini.dot.categ, ") OF data1")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+data1[, dot.categ] <- factor(data1[, dot.categ], levels = dot.categ.class.order) # reorder the factor
+}
+}else{
+if(all(dot.color == "same") & length(dot.color) == 1){
+dot.categ.class.order <- unlist(categ.class.order[length(categ)])
+data1[, dot.categ] <- factor(data1[, dot.categ], levels = dot.categ.class.order) # reorder the factor
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE dot.categ.class.order SETTING IS NULL AND dot.color IS \"same\". ORDER OF categ.class.order WILL BE APPLIED FOR LEGEND DISPLAY: ", paste(dot.categ.class.order, collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}else{
+dot.categ.class.order <- sort(levels(data1[, dot.categ]))
+data1[, dot.categ] <- factor(data1[, dot.categ], levels = dot.categ.class.order) # reorder the factor
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE dot.categ.class.order SETTING IS NULL. ALPHABETICAL ORDER WILL BE APPLIED FOR LEGEND DISPLAY: ", paste(dot.categ.class.order, collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+# dot.categ.class.order not NULL anymore (character string) if dot.categ is not NULL
+if(all(dot.color == "same") & length(dot.color) == 1){
+if( ! identical(ini.dot.categ, categ[length(categ)])){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nWHEN dot.color ARGUMENT IS \"same\", THE COLUMN NAME IN dot.categ ARGUMENT MUST BE IDENTICAL TO THE LAST COLUMN NAME IN categ ARGUMENT. HERE IT IS:\ndot.categ: ", paste(ini.dot.categ, collapse = " "), "\ncateg: ", paste(categ, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else if( ! fun_comp_1d(unlist(categ.class.order[length(categ)]), dot.categ.class.order)$identical.content){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nWHEN dot.color ARGUMENT IS \"same\",\nLAST COMPARTMENT OF categ.class.order ARGUMENT AND dot.categ.class.order ARGUMENT CANNOT BE DIFFERENT:\nLAST COMPARTMENT OF categ.class.order: ", paste(unlist(categ.class.order[length(categ)]), collapse = " "), "\ndot.categ.class.order: ", paste(dot.categ.class.order, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+for(i3 in 1:length(categ)){
+if(identical(categ[i3], ini.dot.categ) & ! identical(unlist(categ.class.order[i3]), dot.categ.class.order) & identical(sort(unlist(categ.class.order[i3])), sort(dot.categ.class.order))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE dot.categ ARGUMENT SETTING IS PRESENT IN THE categ ARGUMENT SETTING, BUT ORDER OF THE CLASSES IS NOT THE SAME:\ncateg.class.order: ", paste(unlist(categ.class.order[i3]), collapse = " "), "\ndot.categ.class.order: ", paste(dot.categ.class.order, collapse = " "), "\nNOTE THAT ORDER OF categ.class.order IS THE ONE USED FOR THE AXIS REPRESENTATION")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+if(is.null(dot.legend.name)){
+dot.legend.name <- if(ini.dot.categ %in% categ[length(categ)]){dot.categ}else{ini.dot.categ} #
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE dot.legend.name SETTING IS NULL -> ", dot.legend.name, " WILL BE USED AS LEGEND TITLE OF DOTS")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+# dot.legend.name not NULL anymore (character string)
+}else{
+if( ! is.null(dot.categ.class.order)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE dot.categ.class.order ARGUMENT IS NOT NULL, BUT IS THE dot.categ ARGUMENT\n-> dot.categ.class.order NOT CONSIDERED AS NO LEGEND WILL BE DRAWN")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+# But dot.categ.class.order will be converted to NULL below (not now)
+}
+# end optional legend of dot colors
+# check the nature of color
+# integer colors into gg_palette
+tempo.check.color <- fun_check(data = dot.color, class = "integer", double.as.integer.allowed = TRUE, na.contain = TRUE, fun.name = function.name)$problem
+if(tempo.check.color == FALSE){
+# convert integers into colors
+dot.color <- fun_gg_palette(max(dot.color, na.rm = TRUE))[dot.color]
+}
+# end integer colors into gg_palette
+if(all(dot.color == "same") & length(dot.color) == 1){
+dot.color <- categ.color # same color of the dots as the corresponding box color
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") dot.color ARGUMENT HAS BEEN SET TO \"same\"\nTHUS, DOTS WILL HAVE THE SAME COLORS AS THE CORRESPONDING BOXPLOT")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}else if( ! (all(dot.color %in% colors() | grepl(pattern = "^#", dot.color)))){ # check that all strings of low.color start by #
+tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.color ARGUMENT MUST BE (1) A HEXADECIMAL COLOR VECTOR STARTING BY #, OR (2) COLOR NAMES GIVEN BY colors(), OR (3) INTEGERS, OR THE STRING\"same\"\nHERE IT IS: ", paste(unique(dot.color), collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(any(is.na(dot.color))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") dot.color ARGUMENT CONTAINS NA")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+# end check the nature of color
+# check the length of color
+if( ! is.null(dot.categ)){
+# optional legend of dot colors
+if(length(dot.color) > 1 & length(unique(data1[, dot.categ])) != length(dot.color)){
+tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.color ARGUMENT IS NOT THE SAME LENGTH AS LEVELS OF dot.categ COLUMN (", dot.categ, "):\ndot.color: ", paste(dot.color, collapse = " "), "\ndot.categ LEVELS: ", paste(levels(data1[, dot.categ]), collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else if(length(dot.color) == 1 & length(dot.categ.class.order) > 1){ # to deal with single color
+dot.color <- rep(dot.color, length(dot.categ.class.order))
+}
+data1 <- data.frame(data1, dot.color = data1[, dot.categ], stringsAsFactors = TRUE)
+data1$dot.color <- factor(data1$dot.color, labels = dot.color) # do not use labels = unique(dot.color). Otherwise, we can have green1 green2 when dot.color is c("green", "green")
+dot.color <- as.character(unique(data1$dot.color[order(data1[, dot.categ])])) # reorder the dot.color character vector
+if(length(dot.color) == 1 & length(dot.categ.class.order) > 1){ # to deal with single color
+dot.color <- rep(dot.color, length(dot.categ.class.order))
+}
+tempo.check <- unique(data1[ , c(dot.categ, "dot.color")])
+if(length(unique(data1[ , "dot.color"])) > 1 & ( ! (nrow(tempo.check) == length(unique(data1[ , "dot.color"])) & nrow(tempo.check) == length(unique(data1[ , dot.categ]))))){ # length(unique(data1[ , "dot.color"])) > 1 because if only one color, can be attributed to each class of dot.categ
+tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.color ARGUMENT IS INCORRECTLY ASSOCIATED TO EACH CLASS OF dot.categ (", dot.categ, ") COLUMN:\n", paste(unique(mapply(FUN = "paste", data1[ , dot.categ], data1[ ,"dot.color"])), collapse = "\n"))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") IN dot.categ ARGUMENT (", ini.dot.categ, "), THE FOLLOWING COLORS OF DOTS:\n", paste(dot.color, collapse = " "), "\nHAVE BEEN ATTRIBUTED TO THESE CLASSES:\n", paste(levels(data1[, dot.categ]), collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+# dot.color is a character string representing the diff classes of dot.categ
+# data1$dot.color is a factor with order of levels -> dot.categ
+# end optional legend of dot colors
+}else{
+categ.len <- length(categ) # if only categ1, then colors for classes of categ1, if length(categ) == 2, then colors for classes of categ2
+if(length(dot.color) == length(levels(data1[, categ[categ.len]]))){ # here length(dot.color) is equal to the different number of categ
+# data1[, categ[categ.len]] <- factor(data1[, categ[categ.len]]) # not required because sure that is is a factor
+data1 <- data.frame(data1, dot.color = data1[, categ[categ.len]], stringsAsFactors = TRUE)
+data1$dot.color <- factor(data1$dot.color, labels = dot.color)
+if(box.alpha != 0){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") IN ", categ[categ.len], " OF categ ARGUMENT, THE FOLLOWING COLORS:\n", paste(dot.color, collapse = " "), "\nHAVE BEEN ATTRIBUTED TO THESE CLASSES:\n", paste(levels(factor(data1[, categ[categ.len]])), collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else if(length(dot.color) == length(data1[, categ[categ.len]])){# here length(dot.color) is equal to nrow(data1) -> Modif to have length(dot.color) equal to the different number of categ (length(dot.color) == length(levels(data1[, categ[categ.len]])))
+data1 <- data.frame(data1, dot.color = dot.color, stringsAsFactors = TRUE)
+}else if(length(dot.color) == 1 & ! all(dot.color == "same")){
+# data1[, categ[categ.len]] <- factor(data1[, categ[categ.len]]) # not required because sure that is is a factor
+data1 <- data.frame(data1, dot.color = dot.color, stringsAsFactors = TRUE)
+dot.color <- rep(dot.color, length(levels(data1[, categ[categ.len]])))
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") dot.color ARGUMENT HAS LENGTH 1, MEANING THAT ALL THE DIFFERENT CLASSES OF ", categ[categ.len], "\n", paste(levels(factor(data1[, categ[categ.len]])), collapse = " "), "\nWILL HAVE THE SAME COLOR\n", paste(dot.color, collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.color ARGUMENT MUST BE (1) LENGTH 1, OR (2) THE LENGTH OF data1 NROWS AFTER NA/Inf REMOVAL, OR (3) THE LENGTH OF THE CLASSES IN THE categ ", categ[categ.len], " COLUMN. HERE IT IS COLOR LENGTH ", length(dot.color), " VERSUS CATEG LENGTH ", length(data1[, categ[categ.len]]), " AND CATEG CLASS LENGTH ", length(unique(data1[, categ[categ.len]])), "\nPRESENCE OF NA/Inf COULD BE THE PROBLEM")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end check the length of color
+dot.color <- as.character(dot.color)
+# dot.color is a character string representing the diff classes
+data1$dot.color <- factor(data1$dot.color, levels = unique(dot.color)) # ok because if dot.color is a character string, the order make class 1, class 2, etc. If dot.color is a column of data1, then levels will be created, without incidence, except if dot.categ specified (see below). unique() because no duplicates allowed
+# data1$dot.color is a factor with order of levels -> dot.color
+}
+# end optional legend of dot colors
+}else if(is.null(dot.color) & ! (is.null(dot.categ) & is.null(dot.categ.class.order) & is.null(dot.legend.name))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") dot.categ OR dot.categ.class.order OR dot.legend.name ARGUMENT HAS BEEN SPECIFIED BUT dot.color ARGUMENT IS NULL (NO DOT PLOTTED)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+# dot.color either NULL (no dot plotted) or character string (potentially representing the diff classes of dot.categ)
+# data1$dot.color is either NA or a factor (with order of levels -> depending on dot.categ or categ[length(categ)], or other
+if(is.null(dot.categ)){
+dot.categ.class.order <- NULL # because not used anyway
+}
+# dot.categ.class.order either NULL if dot.categ is NULL (no legend displayed) or character string (potentially representing the diff classes of dot.categ)
+# end management of dot.color
+if(is.null(dot.color) & box.fill == FALSE & dot.alpha <= 0.025){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE FOLLOWING ARGUMENTS WERE SET AS:\ndot.color = NULL (NOT ALL DOTS BUT ONLY POTENTIAL OUTLIER DOTS DISPLAYED)\nbox.fill = FALSE (NO FILLING COLOR FOR BOTH BOXES AND POTENTIAL OUTLIER DOTS)\ndot.alpha = ", fun_round(dot.alpha, 4), "\n-> POTENTIAL OUTLIER DOTS MIGHT NOT BE VISIBLE BECAUSE ALMOST TRANSPARENT")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+if(is.null(dot.color) & box.fill == FALSE & dot.border.size == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nTHE FOLLOWING ARGUMENTS WERE SET AS:\ndot.color = NULL (NOT ALL DOTS BUT ONLY POTENTIAL OUTLIER DOTS DISPLAYED)\nbox.fill = FALSE (NO FILLING COLOR FOR BOTH BOXES AND POTENTIAL OUTLIER DOTS)\ndot.border.size = 0 (NO BORDER FOR POTENTIAL OUTLIER DOTS)\n-> THESE SETTINGS ARE NOT ALLOWED BECAUSE THE POTENTIAL OUTLIER DOTS WILL NOT BE VISIBLE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# integer dot.border.color into gg_palette
+if( ! is.null(dot.border.color)){
+tempo <- fun_check(data = dot.border.color, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, fun.name = function.name)
+if(tempo$problem == FALSE){ # convert integers into colors
+dot.border.color <- fun_gg_palette(max(dot.border.color, na.rm = TRUE))[dot.border.color]
+}
+}
+# end integer dot.border.color into gg_palette
+# na detection and removal (done now to be sure of the correct length of categ)
+column.check <- c("categ.color", if( ! is.null(dot.color)){"dot.color"}) #
+if(any(is.na(data1[, column.check]))){ # data1 used here instead of data1.ini in case of new NaN created by log conversion (neg values)
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NA DETECTED IN COLUMNS ", paste(column.check, collapse = " "), " OF data1 AND CORRESPONDING ROWS REMOVED (SEE $removed.row.nb AND $removed.rows)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+for(i2 in 1:length(column.check)){
+if(any(is.na(data1[, column.check[i2]]))){
+tempo.warn <- paste0("NA REMOVAL DUE TO COLUMN ", column.check[i2], " OF data1")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n", tempo.warn)))
+}
+}
+tempo <- unique(unlist(lapply(lapply(c(data1[column.check]), FUN = is.na), FUN = which)))
+removed.row.nb <- c(removed.row.nb, tempo)
+removed.rows <- rbind(removed.rows, data1[tempo, ], stringsAsFactors = FALSE) # here data1 used because categorical columns tested
+if(length(tempo) != 0){
+data1 <- data1[-tempo, ] # WARNING tempo here and not removed.row.nb because the latter contain more numbers thant the former
+data1.ini <- data1.ini[-tempo, ] # WARNING tempo here and not removed.row.nb because the latter contain more numbers thant the former
+for(i3 in 1:length(column.check)){
+if(any( ! unique(removed.rows[, column.check[i3]]) %in% unique(data1[, column.check[i3]]))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") IN COLUMN ", column.check[i3], " OF data1, THE FOLLOWING CLASSES HAVE DISAPPEARED AFTER NA/Inf REMOVAL (IF COLUMN USED IN THE PLOT, THIS CLASS WILL NOT BE DISPLAYED):\n", paste(unique(removed.rows[, column.check[i3]])[ ! unique(removed.rows[, column.check[i3]]) %in% unique(data1[, column.check[i3]])], collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+}
+for(i2 in 1:length(column.check)){
+if(column.check[i2] == "categ.color"){
+categ.color <- levels(data1[, column.check[i2]])[levels(data1[, column.check[i2]]) %in% unique(data1[, column.check[i2]])] # remove the absent color in the character vector
+if(length(categ.color) == 1 & length(unlist(categ.class.order[length(categ)])) > 1){ # to deal with single color
+categ.color <- rep(categ.color, length(unlist(categ.class.order[length(categ)])))
+}
+data1[, column.check[i2]] <- factor(as.character(data1[, column.check[i2]]), levels = unique(categ.color))
+}
+if(column.check[i2] == "dot.color"){
+dot.color <- levels(data1[, column.check[i2]])[levels(data1[, column.check[i2]]) %in% unique(data1[, column.check[i2]])] # remove the absent color in the character vector
+if(length(dot.color) == 1 & length(dot.categ.class.order) > 1){ # to deal with single color. If dot.categ.class.order == NULL (which is systematically the case if dot.categ == NULL), no rep(dot.color, length(dot.categ.class.order)
+dot.color <- rep(dot.color, length(dot.categ.class.order))
+}
+data1[, column.check[i2]] <- factor(as.character(data1[, column.check[i2]]), levels = unique(dot.color))
+}
+}
+}
+# end na detection and removal (done now to be sure of the correct length of categ)
+# From here, data1 and data.ini have no more NA or NaN
+
+
+
+# end second round of checking and data preparation
+
+
+# package checking
+fun_pack(req.package = c(
+"ggplot2",
+"gridExtra",
+"lemon",
+"scales"
+), lib.path = lib.path)
+# end package checking
+
+
+
+
+
+# main code
+# y coordinates recovery (create ini.box.coord, dot.coord and modify data1)
+if(length(categ) == 1){
+# width commputations
+box.width2 <- box.width
+box.space <- 0 # to inactivate the shrink that add space between grouped boxes, because no grouped boxes here
+# end width commputations
+# data1 check categ order for dots coordinates recovery
+data1 <- data.frame(data1, categ.check = data1[, categ[1]], stringsAsFactors = TRUE)
+data1$categ.check <- as.integer(data1$categ.check) # to check that data1[, categ[1]] and dot.coord$group are similar, during merging
+# end data1 check categ order for dots coordinates recovery
+# per box dots coordinates recovery
+tempo.gg.name <- "gg.indiv.plot."
+tempo.gg.count <- 0
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = paste0("ggplot2::ggplot()", if(is.null(add)){""}else{add})))) # add added here to have the facets
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_point(data = data1, mapping = ggplot2::aes_string(x = categ[1], y = y, color = categ[1]), stroke = dot.border.size, size = dot.size, alpha = dot.alpha, shape = 21))
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "color", name = box.legend.name, values = if(is.null(dot.color)){rep(NA, length(unique(data1[, categ[1]])))}else if(length(dot.color) == 1){rep(dot.color, length(unique(data1[, categ[1]])))}else{dot.color}))
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_boxplot(data = data1, mapping = ggplot2::aes_string(x = categ[1], y = y, fill = categ[1]), coef = if(box.whisker.kind == "no"){0}else if(box.whisker.kind == "std"){1.5}else if(box.whisker.kind == "max"){Inf})) # fill because this is what is used with geom_box # to easily have the equivalent of the grouped boxes
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "fill", name = box.legend.name, values = if(length(categ.color) == 1){rep(categ.color, length(unique(data1[, categ[1]])))}else{categ.color}))
+# end per box dots coordinates recovery
+}else if(length(categ) == 2){
+# width commputations
+box.width2 <- box.width / length(unique(data1[, categ[length(categ)]])) # real width of each box in x-axis unit, among the set of grouped box. Not relevant if no grouped boxes length(categ) == 1
+# end width commputations
+# data1 check categ order for dots coordinates recovery
+tempo.factor <- paste0(data1[order(data1[, categ[2]], data1[, categ[1]]), categ[2]], "_", data1[order(data1[, categ[2]], data1[, categ[1]]), categ[1]])
+data1 <- data.frame(data1[order(data1[, categ[2]], data1[, categ[1]]), ], categ.check = factor(tempo.factor, levels = unique(tempo.factor)), stringsAsFactors = TRUE)
+data1$categ.check <- as.integer(data1$categ.check)
+# end data1 check categ order for dots coordinates recovery
+# per box dots coordinates recovery
+tempo.gg.name <- "gg.indiv.plot."
+tempo.gg.count <- 0
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = paste0("ggplot2::ggplot()", if(is.null(add)){""}else{add})))) # add added here to have the facets
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_point(data = data1, mapping = ggplot2::aes_string(x = categ[1], y = y, color = categ[2]), stroke = dot.border.size, size = dot.size, alpha = dot.alpha, shape = 21))
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "color", name = box.legend.name, values = if(is.null(dot.color)){rep(NA, length(unique(data1[, categ[2]])))}else if(length(dot.color) == 1){rep(dot.color, length(unique(data1[, categ[2]])))}else{dot.color}))
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_boxplot(data = data1, mapping = ggplot2::aes_string(x = categ[1], y = y, fill = categ[2]), coef = if(box.whisker.kind == "no"){0}else if(box.whisker.kind == "std"){1.5}else if(box.whisker.kind == "max"){Inf})) # fill because this is what is used with geom_box # to easily have the equivalent of the grouped boxes
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "fill", name = box.legend.name, values = if(length(categ.color) == 1){rep(categ.color, length(unique(data1[, categ[2]])))}else{categ.color}))
+# end per box dots coordinates recovery
+}else{
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 1")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if( ! is.null(stat.disp)){
+stat.just <- fun_gg_just(
+angle = stat.angle,
+pos = ifelse(
+vertical == TRUE,
+ifelse(stat.disp == "top", "bottom", "top"), # "bottom" because we want justification for text that are below the ref point which is the top of the graph. The opposite for "above"
+ifelse(stat.disp == "top", "left", "right") # "left" because we want justification for text that are on the left of the ref point which is the right border of the graph. The opposite for "above"
+),
+kind = "text"
+)
+}
+# has in fact no interest because ggplot2 does not create room for geom_text()
+tempo.data.max <- data1[which.max(data1[, y]), ]
+tempo.data.max <- data.frame(tempo.data.max, label = formatC(tempo.data.max[, y], digit = 2, drop0trailing = TRUE, format = "f"), stringsAsFactors = TRUE)
+# end has in fact no interest because ggplot2 does not create room for geom_text()
+tempo.graph.info.ini <- ggplot2::ggplot_build(eval(parse(text = paste(paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "), if( ! is.null(stat.disp)){' + ggplot2::geom_text(data = tempo.data.max, mapping = ggplot2::aes_string(x = 1, y = y, label = "label"), size = stat.size, color = "black", angle = stat.angle, hjust = stat.just$hjust, vjust = stat.just$vjust)'})))) # added here to have room for annotation
+dot.coord <- tempo.graph.info.ini$data[[1]]
+dot.coord$x <- as.numeric(dot.coord$x) # because weird class
+dot.coord$PANEL <- as.numeric(dot.coord$PANEL) # because numbers as levels. But may be a problem is facet are reordered ?
+tempo.mean <- aggregate(x = dot.coord$y, by = list(dot.coord$group, dot.coord$PANEL), FUN = mean, na.rm = TRUE)
+names(tempo.mean)[names(tempo.mean) == "x"] <- "MEAN"
+names(tempo.mean)[names(tempo.mean) == "Group.1"] <- "BOX"
+names(tempo.mean)[names(tempo.mean) == "Group.2"] <- "PANEL"
+dot.coord <- data.frame(
+dot.coord[order(dot.coord$group, dot.coord$y), ], # dot.coord$PANEL deals below
+y.check = as.double(data1[order(data1$categ.check, data1[, y]), y]),
+categ.check = data1[order(data1$categ.check, data1[, y]), "categ.check"],
+dot.color = if(is.null(dot.color)){NA}else{data1[order(data1$categ.check, data1[, y]), "dot.color"]},
+data1[order(data1$categ.check, data1[, y]), ][categ], # avoid the renaming below
+stringsAsFactors = TRUE
+) # y.check to be sure that the order is the same between the y of data1 and the y of dot.coord
+# names(dot.coord)[names(dot.coord) == "tempo.categ1"] <- categ[1]
+if( ! is.null(dot.categ)){
+dot.coord <- data.frame(dot.coord, data1[order(data1$categ.check, data1[, y]), ][dot.categ], stringsAsFactors = TRUE) # avoid the renaming
+}
+if( ! is.null(facet.categ)){
+dot.coord <- data.frame(dot.coord, data1[order(data1$categ.check, data1[, y]), ][facet.categ], stringsAsFactors = TRUE) # for facet panels
+tempo.test <- NULL
+for(i2 in 1:length(facet.categ)){
+tempo.test <- paste0(tempo.test, ".", formatC(as.numeric(dot.coord[, facet.categ[i2]]), width = nchar(max(as.numeric(dot.coord[, facet.categ[i2]]), na.rm = TRUE)), flag = "0")) # convert factor into numeric with leading zero for proper ranking # merge the formatC() to create a new factor. The convertion to integer should recreate the correct group number. Here as.numeric is used and not as.integer in case of numeric in facet.categ (because comes from add and not checked by fun_check, contrary to categ)
+}
+tempo.test <- as.integer(factor(tempo.test))
+if( ! identical(as.integer(dot.coord$PANEL), tempo.test)){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nas.integer(dot.coord$PANEL) AND tempo.test MUST BE IDENTICAL. CODE HAS TO BE MODIFIED")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+if(dot.tidy == TRUE){
+if( ! is.null(dot.categ)){
+dot.coord <- data.frame(dot.coord, tidy_group = data1[order(data1$categ.check, data1[, y]), ][, dot.categ], stringsAsFactors = TRUE) # avoid the renaming
+# tidy_group_coord is to be able to fuse table when creating the table for dot coordinates
+if(dot.categ %in% categ){
+dot.coord <- data.frame(dot.coord, tidy_group_coord = dot.coord$group, stringsAsFactors = TRUE)
+}else{
+dot.coord <- data.frame(dot.coord, tidy_group_coord = as.integer(factor(paste0(
+formatC(as.integer(dot.coord[, categ[1]]), width = nchar(max(as.integer(dot.coord[, categ[1]]), na.rm = TRUE)), flag = "0"), # convert factor into numeric with leading zero for proper ranking
+".",
+if(length(categ) == 2){formatC(as.integer(dot.coord[, categ[2]]), width = nchar(max(as.integer(dot.coord[, categ[2]]), na.rm = TRUE)), flag = "0")}, # convert factor into numeric with leading zero for proper ranking
+if(length(categ) == 2){"."},
+formatC(as.integer(dot.coord[, dot.categ]), width = nchar(max(as.integer(dot.coord[, dot.categ]), na.rm = TRUE)), flag = "0") # convert factor into numeric with leading zero for proper ranking
+)), stringsAsFactors = TRUE) # merge the 2 or 3 formatC() to create a new factor. The convertion to integer should recreate the correct group number
+) # for tidy dot plots
+}
+}else{
+dot.coord <- data.frame(dot.coord, tidy_group = if(length(categ) == 1){
+dot.coord[, categ]}else{as.integer(factor(paste0(
+formatC(as.integer(dot.coord[, categ[1]]), width = nchar(max(as.integer(dot.coord[, categ[1]]), na.rm = TRUE)), flag = "0"), # convert factor into numeric with leading zero for proper ranking
+".",
+formatC(as.integer(dot.coord[, categ[2]]), width = nchar(max(as.integer(dot.coord[, categ[2]]), na.rm = TRUE)), flag = "0")# convert factor into numeric with leading zero for proper ranking
+)), stringsAsFactors = TRUE) # merge the 2 formatC() to create a new factor. The convertion to integer should recreate the correct group number
+}) # for tidy dot plots
+# tidy_group_coord is to be able to fuse table when creating the table for dot coordinates
+dot.coord <- data.frame(dot.coord, tidy_group_coord = dot.coord$group, stringsAsFactors = TRUE)
+}
+}
+if( ! (identical(dot.coord$y, dot.coord$y.check) & identical(dot.coord$group, dot.coord$categ.check))){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\n(dot.coord$y AND dot.coord$y.check) AS WELL AS (dot.coord$group AND dot.coord$categ.check) MUST BE IDENTICAL. CODE HAS TO BE MODIFIED")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+if( ! identical(tempo.mean[order(tempo.mean$BOX, tempo.mean$PANEL), ]$BOX, unique(dot.coord[order(dot.coord$group, dot.coord$PANEL), c("group", "PANEL")])$group)){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\n(tempo.mean$BOX, tempo.mean$PANEL) AND (dot.coord$group, dot.coord$PANEL) MUST BE IDENTICAL. CODE HAS TO BE MODIFIED")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+tempo <- unique(dot.coord[order(dot.coord$group, dot.coord$PANEL), c(categ, if( ! is.null(dot.color) & ! is.null(dot.categ)){if(dot.categ != ini.dot.categ){dot.categ}}, if( ! is.null(facet.categ)){facet.categ}), drop = FALSE])
+# names(tempo) <- paste0(names(tempo), ".mean")
+tempo.mean <- data.frame(tempo.mean[order(tempo.mean$BOX, tempo.mean$PANEL), ], tempo, stringsAsFactors = TRUE)
+}
+}
+# at that stage, categ color and dot color are correctly attributed in data1, box.coord and dot.coord
+# end y dot coordinates recovery (create ini.box.coord, dot.coord and modify data1)
+# ylim range
+if(is.null(y.lim)){
+y.lim <- tempo.graph.info.ini$layout$panel_params[[1]]$y.range # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
+if(any(( ! is.finite(y.lim)) | is.na(y.lim)) | length(y.lim) != 2){ # kept but normally no more Inf in data1
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\ntempo.graph.info.ini$layout$panel_params[[1]]$y.range[1] CONTAINS NA OR Inf OR HAS LENGTH 1")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}else if(y.log != "no"){
+y.lim <- get(y.log)(y.lim) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
+}
+if(y.log != "no"){
+# normally this control is not necessary anymore
+if(any( ! is.finite(y.lim))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\ny.lim ARGUMENT CANNOT HAVE ZERO OR NEGATIVE VALUES WITH THE y.log ARGUMENT SET TO ", y.log, ":\n", paste(y.lim, collapse = " "), "\nPLEASE, CHECK DATA VALUES (PRESENCE OF ZERO OR INF VALUES)")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+if(suppressWarnings(all(y.lim %in% c(Inf, -Inf)))){
+# normally this control is not necessary anymore
+tempo.cat <- paste0("ERROR IN ", function.name, " y.lim CONTAINS Inf VALUES, MAYBE BECAUSE VALUES FROM data1 ARGUMENTS ARE NA OR Inf ONLY OR BECAUSE OF LOG SCALE REQUIREMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+if(suppressWarnings(any(is.na(y.lim)))){
+# normally this control is not necessary anymore
+tempo.cat <- paste0("ERROR IN ", function.name, " y.lim CONTAINS NA OR NaN VALUES, MAYBE BECAUSE VALUES FROM data1 ARGUMENTS ARE NA OR Inf ONLY OR BECAUSE OF LOG SCALE REQUIREMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+y.lim.order <- order(y.lim) # to deal with inverse axis
+y.lim <- sort(y.lim)
+y.lim[1] <- y.lim[1] - abs(y.lim[2] - y.lim[1]) * ifelse(diff(y.lim.order) > 0, y.bottom.extra.margin, y.top.extra.margin) # diff(y.lim.order) > 0 medians not inversed axis
+y.lim[2] <- y.lim[2] + abs(y.lim[2] - y.lim[1]) * ifelse(diff(y.lim.order) > 0, y.top.extra.margin, y.bottom.extra.margin) # diff(y.lim.order) > 0 medians not inversed axis
+if(y.include.zero == TRUE){ # no need to check y.log != "no" because done before
+y.lim <- range(c(y.lim, 0), na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
+}
+y.lim <- y.lim[y.lim.order]
+if(any(is.na(y.lim))){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 2")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end ylim range
+
+
+
+
+
+
+# drawing
+# constant part
+tempo.gg.name <- "gg.indiv.plot."
+tempo.gg.count <- 0
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = paste0("ggplot2::ggplot()", if(is.null(add)){""}else{add})))) # add is directly put here to deal with additional variable of data, like when using facet_grid. No problem if add is a theme, will be dealt below
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::xlab(if(is.null(x.lab)){categ[1]}else{x.lab}))
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::ylab(if(is.null(y.lab)){y}else{y.lab}))
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::ggtitle(title))
+# text angle management
+axis.just <- fun_gg_just(angle = x.angle, pos = ifelse(vertical == TRUE, "bottom", "left"), kind = "axis")
+# end text angle management
+add.check <- TRUE
+if( ! is.null(add)){ # if add is NULL, then = 0
+if(grepl(pattern = "ggplot2\\s*::\\s*theme", add) == TRUE){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") \"ggplot2::theme\" STRING DETECTED IN THE add ARGUMENT\n-> INTERNAL GGPLOT2 THEME FUNCTIONS theme() AND theme_classic() HAVE BEEN INACTIVATED, TO BE USED BY THE USER\n-> article ARGUMENT WILL BE IGNORED")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+add.check <- FALSE
+}
+}
+if(add.check == TRUE & article == TRUE){
+# WARNING: not possible to add theme()several times. NO message but the last one overwrites the others
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::theme_classic(base_size = text.size))
+if(grid == TRUE){
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), m.gg <- ggplot2::theme(
+text = ggplot2::element_text(size = text.size),
+plot.title = ggplot2::element_text(size = title.text.size), # stronger than text
+line = ggplot2::element_line(size = 0.5),
+legend.key = ggplot2::element_rect(color = "white", size = 1.5), # size of the frame of the legend
+axis.line.y.left = ggplot2::element_line(colour = "black"), # draw lines for the y axis
+axis.line.x.bottom = ggplot2::element_line(colour = "black"), # draw lines for the x axis
+panel.grid.major.x = if(vertical == TRUE){NULL}else{ggplot2::element_line(colour = "grey85", size = 0.75)},
+panel.grid.major.y = if(vertical == TRUE){ggplot2::element_line(colour = "grey85", size = 0.75)}else{NULL},
+panel.grid.minor.y = if(vertical == TRUE){ggplot2::element_line(colour = "grey90", size = 0.25)}else{NULL},
+axis.text.x = if(vertical == TRUE){ggplot2::element_text(angle = axis.just$angle, hjust = axis.just$hjust, vjust = axis.just$vjust)}else{NULL},
+axis.text.y = if(vertical == TRUE){NULL}else{ggplot2::element_text(angle = axis.just$angle, hjust = axis.just$hjust, vjust = axis.just$vjust)},
+strip.background = ggplot2::element_rect(fill = NA, colour = NA) # for facet background
+))
+}else{
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), m.gg <- ggplot2::theme(
+text = ggplot2::element_text(size = text.size),
+plot.title = ggplot2::element_text(size = title.text.size), # stronger than text
+line = ggplot2::element_line(size = 0.5),
+legend.key = ggplot2::element_rect(color = "white", size = 1.5), # size of the frame of the legend
+axis.line.y.left = ggplot2::element_line(colour = "black"),
+axis.line.x.bottom = ggplot2::element_line(colour = "black"),
+axis.text.x = if(vertical == TRUE){ggplot2::element_text(angle = axis.just$angle, hjust = axis.just$hjust, vjust = axis.just$vjust)}else{NULL},
+axis.text.y = if(vertical == TRUE){NULL}else{ggplot2::element_text(angle = axis.just$angle, hjust = axis.just$hjust, vjust = axis.just$vjust)},
+strip.background = ggplot2::element_rect(fill = NA, colour = NA)
+))
+}
+}else if(add.check == TRUE & article == FALSE){
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), m.gg <- ggplot2::theme(
+text = ggplot2::element_text(size = text.size),
+plot.title = ggplot2::element_text(size = title.text.size), # stronger than text
+line = ggplot2::element_line(size = 0.5),
+legend.key = ggplot2::element_rect(color = "white", size = 1.5), # size of the frame of the legend
+panel.background = ggplot2::element_rect(fill = "grey95"),
+axis.line.y.left = ggplot2::element_line(colour = "black"),
+axis.line.x.bottom = ggplot2::element_line(colour = "black"),
+panel.grid.major.x = ggplot2::element_line(colour = "grey85", size = 0.75),
+panel.grid.major.y = ggplot2::element_line(colour = "grey85", size = 0.75),
+panel.grid.minor.x = ggplot2::element_blank(),
+panel.grid.minor.y = ggplot2::element_line(colour = "grey90", size = 0.25),
+strip.background = ggplot2::element_rect(fill = NA, colour = NA),
+axis.text.x = if(vertical == TRUE){ggplot2::element_text(angle = axis.just$angle, hjust = axis.just$hjust, vjust = axis.just$vjust)}else{NULL},
+axis.text.y = if(vertical == TRUE){NULL}else{ggplot2::element_text(angle = axis.just$angle, hjust = axis.just$hjust, vjust = axis.just$vjust)}
+))
+}
+# Contrary to fun_gg_bar(), cannot plot the boxplot right now, because I need the dots plotted first
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_boxplot(data = data1, mapping = ggplot2::aes_string(x = categ[1], y = y, group = categ[length(categ)]), position = ggplot2::position_dodge(width = NULL), color = NA, width = box.width, fill = NA)) # this is to set the graph (i.e., a blanck boxplot to be able to use x coordinates to plot dots before boxes)
+# end constant part
+
+
+
+
+# graphic info recovery (including means)
+tempo.graph.info <- ggplot2::ggplot_build(eval(parse(text = paste0(paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "), ' + ggplot2::geom_boxplot(data = data1, mapping = ggplot2::aes_string(x = categ[1], y = y, fill = categ[length(categ)]), position = ggplot2::position_dodge(width = NULL), width = box.width, notch = box.notch, coef = if(box.whisker.kind == "no"){0}else if(box.whisker.kind == "std"){1.5}else if(box.whisker.kind == "max"){Inf}) + ggplot2::scale_discrete_manual(aesthetics = "fill", name = box.legend.name, values = if(length(categ.color) == 1){rep(categ.color, length(unique(data1[, categ[length(categ)]])))}else{categ.color})')))) # will be recovered later again, when ylim will be considered
+tempo.yx.ratio <- (tempo.graph.info$layout$panel_params[[1]]$y.range[2] - tempo.graph.info$layout$panel_params[[1]]$y.range[1]) / (tempo.graph.info$layout$panel_params[[1]]$x.range[2] - tempo.graph.info$layout$panel_params[[1]]$x.range[1])
+box.coord <- tempo.graph.info$data[[2]] # to have the summary statistics of the plot. Contrary to ini.box.plot, now integrates ylim Here because can be required for stat.disp when just box are plotted
+box.coord$x <- as.numeric(box.coord$x) # because x is of special class that block comparison of values using identical
+box.coord$PANEL <- as.numeric(box.coord$PANEL) # because numbers as levels. But may be a problem is facet are reordered ?
+box.coord <- box.coord[order(box.coord$group, box.coord$PANEL), ]
+if( ! (identical(tempo.mean$BOX, box.coord$group) & identical(tempo.mean$PANEL, box.coord$PANEL))){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nidentical(tempo.mean$BOX, box.coord$group) & identical(tempo.mean$PANEL, box.coord$PANEL) DO NOT HAVE THE SAME VALUE ORDER")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+# tempo <- c(categ, if( ! is.null(dot.color) & ! is.null(dot.categ)){if(dot.categ != ini.dot.categ){dot.categ}}, if( ! is.null(facet.categ)){facet.categ})
+if(any(names(tempo.mean) %in% names(box.coord))){
+names(tempo.mean)[names(tempo.mean) %in% names(box.coord)] <- paste0(names(tempo.mean)[names(tempo.mean) %in% names(box.coord)], ".mean")
+}
+box.coord <- data.frame(box.coord, tempo.mean, stringsAsFactors = TRUE)
+}
+# end graphic info recovery (including means)
+
+
+
+# stat output (will also serve for boxplot and mean display)
+# x not added know to do not have them in stat.nolog
+stat <- data.frame(
+MIN = box.coord$ymin_final,
+QUART1 = box.coord$lower,
+MEDIAN = box.coord$middle,
+MEAN = box.coord$MEAN,
+QUART3 = box.coord$upper,
+MAX = box.coord$ymax_final,
+WHISK_INF = box.coord$ymin,
+BOX_INF = box.coord$lower,
+NOTCH_INF = box.coord$notchlower,
+NOTCH_SUP = box.coord$notchupper,
+BOX_SUP = box.coord$upper,
+WHISK_SUP = box.coord$ymax,
+OUTLIERS = box.coord["outliers"],
+tempo.mean[colnames(tempo.mean) != "MEAN"],
+COLOR = box.coord$fill,
+stringsAsFactors = TRUE
+) # box.coord["outliers"] written like this because it is a list. X coordinates not put now because several features to set
+names(stat)[names(stat) == "outliers"] <- "OUTLIERS"
+stat.nolog <- stat # stat.nolog ini will serve for outputs
+if(y.log != "no"){
+stat.nolog[c("MIN", "QUART1", "MEDIAN", "MEAN", "QUART3", "MAX", "WHISK_INF", "BOX_INF", "NOTCH_INF", "NOTCH_SUP", "BOX_SUP", "WHISK_SUP")] <- ifelse(y.log == "log2", 2, 10)^(stat.nolog[c("MIN", "QUART1", "MEDIAN", "MEAN", "QUART3", "MAX", "WHISK_INF", "BOX_INF", "NOTCH_INF", "NOTCH_SUP", "BOX_SUP", "WHISK_SUP")])
+stat.nolog$OUTLIERS <- lapply(stat.nolog$OUTLIERS, FUN = function(X){ifelse(y.log == "log2", 2, 10)^X})
+}
+# end stat output (will also serve for boxplot and mean display)
+
+
+
+
+
+
+# x coordinates management (for random plotting and for stat display)
+# width commputations
+width.ini <- c(box.coord$xmax - box.coord$xmin)[1] # all the box widths are equal here. Only the first one taken
+width.correct <- width.ini * box.space / 2
+if( ! (identical(stat$BOX, box.coord$group) & identical(stat$PANEL, box.coord$PANEL))){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nidentical(stat$BOX, box.coord$group) & identical(stat$PANEL, box.coord$PANEL) MUST BE IDENTICAL. CODE HAS TO BE MODIFIED")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+stat <- data.frame(
+stat,
+X = box.coord$x,
+X_BOX_INF = box.coord$xmin + width.correct,
+X_BOX_SUP = box.coord$xmax - width.correct,
+X_NOTCH_INF = box.coord$x - (box.coord$x - (box.coord$xmin + width.correct)) / 2,
+X_NOTCH_SUP = box.coord$x + (box.coord$x - (box.coord$xmin + width.correct)) / 2,
+X_WHISK_INF = box.coord$x - (box.coord$x - (box.coord$xmin + width.correct)) * box.whisker.width,
+X_WHISK_SUP = box.coord$x + (box.coord$x - (box.coord$xmin + width.correct)) * box.whisker.width,
+# tempo.mean[colnames(tempo.mean) != "MEAN"], # already added above
+stringsAsFactors = TRUE
+)
+stat$COLOR <- factor(stat$COLOR, levels = unique(categ.color))
+if( ! all(stat$NOTCH_SUP < stat$BOX_SUP & stat$NOTCH_INF > stat$BOX_INF) & box.notch == TRUE){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") SOME NOTCHES ARE BEYOND BOX HINGES. TRY ARGUMENT box.notch = FALSE")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+dot.jitter <- c((box.coord$xmax - width.correct) - (box.coord$xmin + width.correct))[1] * dot.jitter # real dot.jitter. (box.coord$xmin + width.correct) - (box.coord$xmax - width.correct))[1] is the width of the box. Is equivalent to (box.coord$x - (box.coord$xmin + width.correct))[1] * 2
+# end width commputations
+if( ! is.null(dot.color)){
+# random dots
+if(dot.tidy == FALSE){
+dot.coord.rd1 <- merge(dot.coord, box.coord[c("fill", "PANEL", "group", "x")], by = c("PANEL", "group"), sort = FALSE) # rd for random. Send the coord of the boxes into the coord data.frame of the dots (in the column x.y). WARNING: by = c("PANEL", "group") without fill column because PANEL & group columns are enough as only one value of x column per group number in box.coord. Thus, no need to consider fill column
+if(nrow(dot.coord.rd1) != nrow(dot.coord)){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nTHE merge() FUNCTION DID NOT RETURN A CORRECT dot.coord.rd1 DATA FRAME. CODE HAS TO BE MODIFIED")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+sampled.dot.jitter <- if(nrow(dot.coord.rd1) == 1){runif(n = nrow(dot.coord.rd1), min = - dot.jitter / 2, max = dot.jitter / 2)}else{sample(x = runif(n = nrow(dot.coord.rd1), min = - dot.jitter / 2, max = dot.jitter / 2), size = nrow(dot.coord.rd1), replace = FALSE)}
+dot.coord.rd2 <- data.frame(dot.coord.rd1, dot.x = dot.coord.rd1$x.y + sampled.dot.jitter, stringsAsFactors = TRUE) # set the dot.jitter thanks to runif and dot.jitter range. Then, send the coord of the boxes into the coord data.frame of the dots (in the column x.y)
+if(length(categ) == 1){
+tempo.data1 <- unique(data.frame(data1[categ[1]], group = as.integer(data1[, categ[1]]), stringsAsFactors = TRUE)) # categ[1] is factor
+names(tempo.data1)[names(tempo.data1) == categ[1]] <- paste0(categ[1], ".check")
+verif <- paste0(categ[1], ".check")
+}else if(length(categ) == 2){
+tempo.data1 <- unique(
+data.frame(
+data1[c(categ[1], categ[2])],
+group = as.integer(factor(paste0(
+formatC(as.integer(data1[, categ[2]]), width = nchar(max(as.integer(data1[, categ[2]]), na.rm = TRUE)), flag = "0"), # convert factor into numeric with leading zero for proper ranking
+".",
+formatC(as.integer(data1[, categ[1]]), width = nchar(max(as.integer(data1[, categ[1]]), na.rm = TRUE)), flag = "0")# convert factor into numeric with leading zero for proper ranking
+)), stringsAsFactors = TRUE) # merge the 2 formatC() to create a new factor. The convertion to integer should recreate the correct group number
+)
+) # categ[2] first if categ[2] is used to make the categories in ggplot and categ[1] is used to make the x-axis
+names(tempo.data1)[names(tempo.data1) == categ[1]] <- paste0(categ[1], ".check")
+names(tempo.data1)[names(tempo.data1) == categ[2]] <- paste0(categ[2], ".check")
+verif <- c(paste0(categ[1], ".check"), paste0(categ[2], ".check"))
+}else{
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 3")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+dot.coord.rd3 <- merge(dot.coord.rd2, tempo.data1, by = intersect("group", "group"), sort = FALSE) # send the factors of data1 into coord. WARNING: I have replaced by = "group" by intersect("group", "group") because of an error due to wrong group group merging in dot.coord.rd3
+if(nrow(dot.coord.rd3) != nrow(dot.coord) | ( ! fun_comp_2d(dot.coord.rd3[categ], dot.coord.rd3[verif])$identical.content)){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nTHE merge() FUNCTION DID NOT RETURN A CORRECT dot.coord.rd3 DATA FRAME. CODE HAS TO BE MODIFIED")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end random dots
+}
+# tidy dots
+# coordinates are recovered during plotting (see dot.coord.tidy1 below)
+# end tidy dots
+}
+# end x coordinates management (for random plotting and for stat display)
+
+
+
+
+
+# boxplot display before dot display if box.fill = TRUE
+coord.names <- NULL
+# creation of the data frame for (main box + legend) and data frame for means
+if(box.notch == FALSE){
+for(i3 in 1:length(categ)){
+if(i3 == 1){
+tempo.polygon <- data.frame(GROUPX = c(t(stat[, rep(categ[i3], 5)])), stringsAsFactors = TRUE)
+}else{
+tempo.polygon <- cbind(tempo.polygon, c(t(stat[, rep(categ[i3], 5)])), stringsAsFactors = TRUE)
+}
+}
+names(tempo.polygon) <- categ
+tempo.polygon <- data.frame(X = c(t(stat[, c("X_BOX_INF", "X_BOX_SUP", "X_BOX_SUP", "X_BOX_INF", "X_BOX_INF")])), Y = c(t(stat[, c("BOX_INF", "BOX_INF", "BOX_SUP", "BOX_SUP", "BOX_INF")])), COLOR = c(t(stat[, c("COLOR", "COLOR", "COLOR", "COLOR", "COLOR")])), BOX = as.character(c(t(stat[, c("BOX", "BOX", "BOX", "BOX", "BOX")]))), tempo.polygon, stringsAsFactors = TRUE)
+if( ! is.null(facet.categ)){
+for(i4 in 1:length(facet.categ)){
+tempo.polygon <- data.frame(tempo.polygon, c(t(stat[, c(facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4])])), stringsAsFactors = TRUE)
+names(tempo.polygon)[length(names(tempo.polygon))] <- facet.categ[i4]
+}
+}
+}else{
+for(i3 in 1:length(categ)){
+if(i3 == 1){
+tempo.polygon <- data.frame(GROUPX = c(t(stat[, rep(categ[i3], 11)])), stringsAsFactors = TRUE)
+}else{
+tempo.polygon <- cbind(tempo.polygon, c(t(stat[, rep(categ[i3], 11)])), stringsAsFactors = TRUE)
+}
+}
+names(tempo.polygon) <- categ
+tempo.polygon <- data.frame(X = c(t(stat[, c("X_BOX_INF", "X_BOX_SUP", "X_BOX_SUP", "X_NOTCH_SUP", "X_BOX_SUP", "X_BOX_SUP", "X_BOX_INF", "X_BOX_INF", "X_NOTCH_INF", "X_BOX_INF", "X_BOX_INF")])), Y = c(t(stat[, c("BOX_INF", "BOX_INF", "NOTCH_INF", "MEDIAN", "NOTCH_SUP", "BOX_SUP", "BOX_SUP", "NOTCH_SUP", "MEDIAN", "NOTCH_INF", "BOX_INF")])), COLOR = c(t(stat[, c("COLOR", "COLOR", "COLOR", "COLOR", "COLOR", "COLOR", "COLOR", "COLOR", "COLOR", "COLOR", "COLOR")])), BOX = as.character(c(t(stat[, c("BOX", "BOX", "BOX", "BOX", "BOX", "BOX", "BOX", "BOX", "BOX", "BOX", "BOX")]))), tempo.polygon, stringsAsFactors = TRUE)
+if( ! is.null(facet.categ)){
+for(i4 in 1:length(facet.categ)){
+tempo.polygon <- data.frame(tempo.polygon, c(t(stat[, c(facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4])])), stringsAsFactors = TRUE)
+names(tempo.polygon)[length(names(tempo.polygon))] <- facet.categ[i4]
+}
+}
+}
+tempo.polygon$COLOR <- factor(tempo.polygon$COLOR, levels = unique(categ.color))
+if( ! is.null(categ.class.order)){
+for(i3 in 1:length(categ)){
+tempo.polygon[, categ[i3]] <- factor(tempo.polygon[, categ[i3]], levels = categ.class.order[[i3]])
+}
+}
+# modified name of dot.categ column (e.g., "Categ1_DOT") must be included for boxplot using ridy dots
+if( ! is.null(dot.color) & ! is.null(dot.categ)){
+if(dot.categ != ini.dot.categ){
+tempo.polygon <- data.frame(tempo.polygon, GROUPX = tempo.polygon[, ini.dot.categ], stringsAsFactors = TRUE)
+names(tempo.polygon)[names(tempo.polygon) == "GROUPX"] <- dot.categ
+
+}
+}
+tempo.diamon.mean <- data.frame(X = c(t(stat[, c("X", "X_NOTCH_INF", "X", "X_NOTCH_SUP", "X")])), Y = c(t(cbind(stat["MEAN"] - (stat[, "X"] - stat[, "X_NOTCH_INF"]) * tempo.yx.ratio, stat["MEAN"], stat["MEAN"] + (stat[, "X"] - stat[, "X_NOTCH_INF"]) * tempo.yx.ratio, stat["MEAN"], stat["MEAN"] - (stat[, "X"] - stat[, "X_NOTCH_INF"]) * tempo.yx.ratio, stringsAsFactors = TRUE))), COLOR = c(t(stat[, c("COLOR", "COLOR", "COLOR", "COLOR", "COLOR")])), GROUP = c(t(stat[, c("BOX", "BOX", "BOX", "BOX", "BOX")])), stringsAsFactors = TRUE) # stringsAsFactors = TRUE for cbind() because stat["MEAN"] is a data frame. Otherwise, stringsAsFactors is not an argument for cbind() on vectors
+if( ! is.null(facet.categ)){
+for(i3 in 1:length(facet.categ)){
+tempo.diamon.mean <- data.frame(tempo.diamon.mean, c(t(stat[, c(facet.categ[i3], facet.categ[i3], facet.categ[i3], facet.categ[i3], facet.categ[i3])])), stringsAsFactors = TRUE)
+names(tempo.diamon.mean)[length(names(tempo.diamon.mean))] <- facet.categ[i3]
+}
+}
+tempo.diamon.mean$COLOR <- factor(tempo.diamon.mean$COLOR, levels = unique(categ.color))
+# end creation of the data frame for (main box + legend) and data frame for means
+if(box.fill == TRUE){
+# assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_boxplot(data = data1, mapping = ggplot2::aes_string(x = categ[1], y = y, color = categ[length(categ)], fill = categ[length(categ)]), position = ggplot2::position_dodge(width = NULL), width = box.width, size = box.line.size, notch = box.notch, coef = if(box.whisker.kind == "no"){0}else if(box.whisker.kind == "std"){1.5}else if(box.whisker.kind == "max"){Inf}, alpha = box.alpha, outlier.shape = if( ! is.null(dot.color)){NA}else{21}, outlier.color = if( ! is.null(dot.color)){NA}else{dot.border.color}, outlier.fill = if( ! is.null(dot.color)){NA}else{NULL}, outlier.size = if( ! is.null(dot.color)){NA}else{dot.size}, outlier.stroke = if( ! is.null(dot.color)){NA}else{dot.border.size}, outlier.alpha = if( ! is.null(dot.color)){NA}else{dot.alpha})) # the color, size, etc. of the outliers are dealt here. outlier.color = NA to do not plot outliers when dots are already plotted. Finally, boxplot redrawn (see below)
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_polygon(
+data = tempo.polygon,
+mapping = ggplot2::aes_string(x = "X", y = "Y", group = "BOX", fill = categ[length(categ)], color = categ[length(categ)]),
+size = box.line.size,
+alpha = box.alpha # works only for fill, not for color
+))
+coord.names <- c(coord.names, "main.box")
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = X, xend = X, y = BOX_SUP, yend = WHISK_SUP, group = categ[length(categ)]), color = "black", size = box.line.size, alpha = box.alpha)) #
+coord.names <- c(coord.names, "sup.whisker")
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = X, xend = X, y = BOX_INF, yend = WHISK_INF, group = categ[length(categ)]), color = "black", size = box.line.size, alpha = box.alpha)) #
+coord.names <- c(coord.names, "inf.whisker")
+if(box.whisker.width > 0){
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = X_WHISK_INF, xend = X_WHISK_SUP, y = WHISK_SUP, yend = WHISK_SUP, group = categ[length(categ)]), color = "black", size = box.line.size, alpha = box.alpha, lineend = "round")) #
+coord.names <- c(coord.names, "sup.whisker.edge")
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = X_WHISK_INF, xend = X_WHISK_SUP, y = WHISK_INF, yend = WHISK_INF, group = categ[length(categ)]), color = "black", size = box.line.size, alpha = box.alpha, lineend = "round")) #
+coord.names <- c(coord.names, "inf.whisker.edge")
+}
+if(box.mean == TRUE){
+# assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_point(data = stat, mapping = ggplot2::aes_string(x = "X", y = "MEAN", group = categ[length(categ)]), shape = 23, stroke = box.line.size * 2, fill = stat$COLOR, size = box.mean.size, color = "black", alpha = box.alpha)) # group used in aesthetic to do not have it in the legend
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_polygon(
+data = tempo.diamon.mean,
+mapping = ggplot2::aes(x = X, y = Y, group = GROUP),
+fill = tempo.diamon.mean[, "COLOR"],
+color = hsv(0, 0, 0, alpha = box.alpha), # outline of the polygon in black but with alpha
+size = box.line.size,
+alpha = box.alpha
+))
+coord.names <- c(coord.names, "mean")
+}
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = if(box.notch == FALSE){X_BOX_INF}else{X_NOTCH_INF}, xend = if(box.notch == FALSE){X_BOX_SUP}else{X_NOTCH_SUP}, y = MEDIAN, yend = MEDIAN, group = categ[length(categ)]), color = "black", size = box.line.size * 2, alpha = box.alpha)) #
+coord.names <- c(coord.names, "median")
+}
+# end boxplot display before dot display if box.fill = TRUE
+
+
+
+
+
+
+# dot display
+if( ! is.null(dot.color)){
+if(dot.tidy == FALSE){
+if(is.null(dot.categ)){
+if(dot.border.size == 0){
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_point(
+data = dot.coord.rd3,
+mapping = ggplot2::aes_string(x = "dot.x", y = "y", group = categ[length(categ)]),
+size = dot.size,
+shape = 19,
+color = dot.coord.rd3$dot.color,
+alpha = dot.alpha
+)) # group used in aesthetic to do not have it in the legend. Here ggplot2::scale_discrete_manual() cannot be used because of the group easthetic
+}else{
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_point(
+data = dot.coord.rd3,
+mapping = ggplot2::aes_string(x = "dot.x", y = "y", group = categ[length(categ)]),
+shape = 21,
+stroke = dot.border.size,
+color = if(is.null(dot.border.color)){dot.coord.rd3$dot.color}else{rep(dot.border.color, nrow(dot.coord.rd3))},
+size = dot.size,
+fill = dot.coord.rd3$dot.color,
+alpha = dot.alpha
+)) # group used in aesthetic to do not have it in the legend. Here ggplot2::scale_discrete_manual() cannot be used because of the group easthetic
+}
+}else{
+if(dot.border.size == 0){
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_point(
+data = dot.coord.rd3,
+mapping = ggplot2::aes_string(x = "dot.x", y = "y", alpha = dot.categ),
+size = dot.size,
+shape = 19,
+color = dot.coord.rd3$dot.color
+)) # group used in aesthetic to do not have it in the legend. Here ggplot2::scale_discrete_manual() cannot be used because of the group easthetic
+}else{
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_point(
+data = dot.coord.rd3,
+mapping = ggplot2::aes_string(x = "dot.x", y = "y", alpha = dot.categ),
+size = dot.size,
+shape = 21,
+stroke = dot.border.size,
+color = if(is.null(dot.border.color)){dot.coord.rd3$dot.color}else{rep(dot.border.color, nrow(dot.coord.rd3))},
+fill = dot.coord.rd3$dot.color
+)) # group used in aesthetic to do not have it in the legend. Here ggplot2::scale_discrete_manual() cannot be used because of the group easthetic
+}
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "alpha", name = dot.legend.name, values = rep(dot.alpha, length(dot.categ.class.order)), guide = ggplot2::guide_legend(override.aes = list(fill = dot.color, color = if(is.null(dot.border.color)){dot.color}else{dot.border.color}, stroke = dot.border.size, alpha = dot.alpha)))) # values are the values of color (which is the border color in geom_box. WARNING: values = categ.color takes the numbers to make the colors if categ.color is a factor
+}
+coord.names <- c(coord.names, "dots")
+}else if(dot.tidy == TRUE){
+# here plot using group -> no scale
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_dotplot(
+data = dot.coord,
+mapping = ggplot2::aes_string(x = categ[1], y = "y", group = "group"), # not dot.categ here because the classes of dot.categ create new separations
+position = ggplot2::position_dodge(width = box.width),
+binpositions = "all",
+binaxis = "y",
+stackdir = "center",
+alpha = dot.alpha,
+fill = dot.coord$dot.color,
+stroke = dot.border.size,
+color = if(is.null(dot.border.color)){dot.coord$dot.color}else{rep(dot.border.color, nrow(dot.coord))},
+show.legend = FALSE, # WARNING: do not use show.legend = TRUE because it uses the arguments outside aes() as aesthetics (here color and fill). Thus I must find a way using ggplot2::scale_discrete_manual()
+binwidth = (y.lim[2] - y.lim[1]) / dot.tidy.bin.nb
+)) # geom_dotplot ggplot2 v3.3.0: I had to remove rev() in fill and color # very weird behavior of geom_dotplot ggplot2 v3.2.1, (1) because with aes group = (to avoid legend), the dot plotting is not good in term of coordinates, and (2) because data1 seems reorderer according to x = categ[1] before plotting. Thus, I have to use fill = dot.coord[rev(order(dot.coord[, categ[1]], decreasing = TRUE)), "dot.color"] to have the good corresponding colors # show.legend option do not remove the legend, only the aesthetic of the legend (dot, line, etc.)
+coord.names <- c(coord.names, "dots")
+if( ! is.null(dot.categ)){
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_dotplot(
+data = dot.coord,
+mapping = ggplot2::aes_string(x = categ[1], y = "y", alpha = dot.categ), # not dot.categ here because the classes of dot.categ create new separations
+position = ggplot2::position_dodge(width = box.width),
+binpositions = "all",
+binaxis = "y",
+stackdir = "center",
+fill = NA,
+stroke = NA,
+color = NA,
+# WARNING: do not use show.legend = TRUE because it uses the arguments outside aes() as aesthetics (here color and fill). Thus I must find a way using ggplot2::scale_discrete_manual()
+binwidth = (y.lim[2] - y.lim[1]) / dot.tidy.bin.nb
+)) # geom_dotplot ggplot2 v3.3.0: I had to remove rev() in fill and color # very weird behavior of geom_dotplot ggplot2 v3.2.1, (1) because with aes group = (to avoid legend), the dot plotting is not good in term of coordinates, and (2) because data1 seems reorderer according to x = categ[1] before plotting. Thus, I have to use fill = dot.coord[rev(order(dot.coord[, categ[1]], decreasing = TRUE)), "dot.color"] to have the good corresponding colors # show.legend option do not remove the legend, only the aesthetic of the legend (dot, line, etc.)
+# assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "linetype", name = dot.legend.name, values = rep(1, length(categ.color)))) # values = rep("black", length(categ.color)) are the values of color (which is the border color of dots), and this modify the border color on the plot. WARNING: values = categ.color takes the numbers to make the colors if categ.color is a factor
+coord.names <- c(coord.names, "bad_remove")
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "alpha", name = dot.legend.name, values = rep(dot.alpha, length(dot.categ.class.order)), labels = dot.categ.class.order, guide = ggplot2::guide_legend(title = if(ini.dot.categ == categ[length(categ)]){dot.categ}else{ini.dot.categ}, override.aes = list(fill = levels(dot.coord$dot.color), color = if(is.null(dot.border.color)){levels(dot.coord$dot.color)}else{dot.border.color}, stroke = dot.border.size, alpha = dot.alpha)))) # values are the values of color (which is the border color in geom_box. WARNING: values = categ.color takes the numbers to make the colors if categ.color is a factor
+}
+# coordinates of tidy dots
+tempo.coord <- ggplot2::ggplot_build(eval(parse(text = paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "))))$data # to have the tidy dot coordinates
+if(length(which(sapply(tempo.coord, FUN = nrow) == nrow(data1))) > if(is.null(dot.categ)){1}else{2}){ # if(is.null(dot.categ)){1}else{2} because 1 dotplot if dot.categ is NULL and 2 dotplots is not, with the second being a blank dotplot with wrong coordinates. Thus take the first in that situation
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nMORE THAN ", if(is.null(dot.categ)){1}else{2}, " COMPARTMENT WITH NROW EQUAL TO nrow(data1) IN THE tempo.coord LIST (FOR TIDY DOT COORDINATES). CODE HAS TO BE MODIFIED")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+dot.coord.tidy1 <- tempo.coord[[which(sapply(tempo.coord, FUN = nrow) == nrow(data1))[1]]] # the second being a blank dotplot with wrong coordinates. Thus take the first whatever situation
+dot.coord.tidy1$x <- as.numeric(dot.coord.tidy1$x) # because weird class
+dot.coord.tidy1$PANEL <- as.numeric(dot.coord.tidy1$PANEL) # because numbers as levels. But may be a problem is facet are reordered ?
+}
+# tempo.box.coord <- merge(box.coord, unique(dot.coord[, c("PANEL", "group", categ)]), by = c("PANEL", "group"), sort = FALSE) # not required anymore because box.coord already contains categ do not add dot.categ and tidy_group_coord here because the coordinates are for stats. Add the categ in box.coord. WARNING: by = c("PANEL", "group") without fill column because PANEL & group columns are enough as only one value of x column per group number in box.coord. Thus, no need to consider fill column
+# below inactivated because not true when dealing with dot.categ different from categ
+# if(nrow(tempo.box.coord) != nrow(box.coord)){
+# tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nTHE merge() FUNCTION DID NOT RETURN A CORRECT tempo.box.coord DATA FRAME. CODE HAS TO BE MODIFIED")
+# stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+# }
+dot.coord.tidy2 <- merge(dot.coord.tidy1, box.coord[c("fill", "PANEL", "group", "x", categ)], by = c("PANEL", "group"), sort = FALSE) # send the coord of the boxes into the coord data.frame of the dots (in the column x.y).WARNING: by = c("PANEL", "group") without fill column because PANEL & group columns are enough as only one value of x column per group number in tempo.box.coord. Thus, no need to consider fill colum # DANGER: from here the fill.y and x.y (from tempo.box.coord) are not good in dot.coord.tidy2. It is ok because Categ1 Categ2 from tempo.box.coord are ok with the group column from dot.coord.tidy1. This is due to the fact that dot.coord.tidy resulting from geom_dotplot does not make the same groups as the other functions
+if(nrow(dot.coord.tidy2) != nrow(dot.coord)){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nTHE merge() FUNCTION DID NOT RETURN A CORRECT dot.coord.tidy2 DATA FRAME. CODE HAS TO BE MODIFIED")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# From here, check for dot.coord.tidy3 which wil be important for stat over the plot. WARNING: dot.categ has nothing to do here for stat coordinates. Thus, not in tempo.data1
+if(length(categ) == 1){
+tempo.data1 <- unique(data.frame(data1[categ[1]], group = as.integer(data1[, categ[1]]), stringsAsFactors = TRUE)) # categ[1] is factor
+names(tempo.data1)[names(tempo.data1) == categ[1]] <- paste0(categ[1], ".check")
+verif <- paste0(categ[1], ".check")
+}else if(length(categ) == 2){
+tempo.data1 <- unique(
+data.frame(
+data1[c(categ[1], categ[2])],
+group = as.integer(factor(paste0(
+formatC(as.integer(data1[, categ[2]]), width = nchar(max(as.integer(data1[, categ[2]]), na.rm = TRUE)), flag = "0"), # convert factor into numeric with leading zero for proper ranking
+".",
+formatC(as.integer(data1[, categ[1]]), width = nchar(max(as.integer(data1[, categ[1]]), na.rm = TRUE)), flag = "0")# convert factor into numeric with leading zero for proper ranking
+)), stringsAsFactors = TRUE) # merge the 2 formatC() to create a new factor. The convertion to integer should recreate the correct group number
+)
+) # categ[2] first if categ[2] is used to make the categories in ggplot and categ[1] is used to make the x-axis
+names(tempo.data1)[names(tempo.data1) == categ[1]] <- paste0(categ[1], ".check")
+names(tempo.data1)[names(tempo.data1) == categ[2]] <- paste0(categ[2], ".check")
+verif <- c(paste0(categ[1], ".check"), paste0(categ[2], ".check"))
+}else{
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 4")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+dot.coord.tidy3 <- merge(dot.coord.tidy2, tempo.data1, by = intersect("group", "group"), sort = FALSE) # send the factors of data1 into coord. WARNING: I have tested intersect("group", "group") instead of by = "group". May be come back to by = "group" in case of error. But I did this because of an error in dot.coord.rd3 above
+if(nrow(dot.coord.tidy3) != nrow(dot.coord) | ( ! fun_comp_2d(dot.coord.tidy3[categ], dot.coord.tidy3[verif])$identical.content)){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nTHE merge() FUNCTION DID NOT RETURN A CORRECT dot.coord.tidy3 DATA FRAME. CODE HAS TO BE MODIFIED")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end coordinates of tidy dots
+}
+}
+# end dot display
+
+
+
+# boxplot display (if box.fill = FALSE, otherwise, already plotted above)
+if(box.fill == TRUE){
+# overcome "work only for the filling of boxes, not for the frame. See https://github.com/tidyverse/ggplot2/issues/252"
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "fill", name = box.legend.name, values = if(length(categ.color) == 1){rep(categ.color, length(unique(data1[, categ[length(categ)]])))}else{categ.color}, guide = ggplot2::guide_legend(order = 1))) #, guide = ggplot2::guide_legend(override.aes = list(fill = levels(tempo.polygon$COLOR), color = "black")))) # values are the values of color (which is the border color in geom_box. WARNING: values = categ.color takes the numbers to make the colors if categ.color is a factor
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "color", name = box.legend.name, values = rep(hsv(0, 0, 0, alpha = box.alpha), length(unique(data1[, categ[length(categ)]]))), guide = ggplot2::guide_legend(order = 1))) # , guide = ggplot2::guide_legend(override.aes = list(color = "black", alpha = box.alpha)))) # values are the values of color (which is the border color in geom_box. WARNING: values = categ.color takes the numbers to make the colors if categ.color is a factor # outline of the polygon in black but with alpha
+}else{
+# assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_boxplot(data = data1, mapping = ggplot2::aes_string(x = categ[1], y = y, color = categ[length(categ)], fill = categ[length(categ)]), position = ggplot2::position_dodge(width = NULL), width = box.width, size = box.line.size, notch = box.notch, alpha = box.alpha, coef = if(box.whisker.kind == "no"){0}else if(box.whisker.kind == "std"){1.5}else if(box.whisker.kind == "max"){Inf}, outlier.shape = if( ! is.null(dot.color)){NA}else{21}, outlier.color = if( ! is.null(dot.color)){NA}else{if(dot.border.size == 0){NA}else{dot.border.color}}, outlier.fill = if( ! is.null(dot.color)){NA}else{NULL}, outlier.size = if( ! is.null(dot.color)){NA}else{dot.size}, outlier.stroke = if( ! is.null(dot.color)){NA}else{dot.border.size}, outlier.alpha = if( ! is.null(dot.color)){NA}else{dot.alpha})) # the color, size, etc. of the outliers are dealt here. outlier.color = NA to do not plot outliers when dots are already plotted
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_path(
+data = tempo.polygon,
+mapping = ggplot2::aes_string(x = "X", y = "Y", group = "BOX", color = categ[length(categ)]),
+size = box.line.size,
+alpha = box.alpha,
+lineend = "round",
+linejoin = "round"
+))
+coord.names <- c(coord.names, "main.box")
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = if(box.notch == FALSE){X_BOX_INF}else{X_NOTCH_INF}, xend = if(box.notch == FALSE){X_BOX_SUP}else{X_NOTCH_SUP}, y = MEDIAN, yend = MEDIAN, group = categ[length(categ)]), color = stat$COLOR, size = box.line.size * 2, alpha = box.alpha)) #
+coord.names <- c(coord.names, "median")
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = X, xend = X, y = BOX_SUP, yend = WHISK_SUP, group = categ[length(categ)]), color = stat$COLOR, size = box.line.size, alpha = box.alpha)) #
+coord.names <- c(coord.names, "sup.whisker")
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = X, xend = X, y = BOX_INF, yend = WHISK_INF, group = categ[length(categ)]), color = stat$COLOR, size = box.line.size, alpha = box.alpha)) #
+coord.names <- c(coord.names, "inf.whisker")
+if(box.whisker.width > 0){
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = X_WHISK_INF, xend = X_WHISK_SUP, y = WHISK_SUP, yend = WHISK_SUP, group = categ[length(categ)]), color = stat$COLOR, size = box.line.size, alpha = box.alpha, lineend = "round")) #
+coord.names <- c(coord.names, "sup.whisker.edge")
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = X_WHISK_INF, xend = X_WHISK_SUP, y = WHISK_INF, yend = WHISK_INF, group = categ[length(categ)]), color = stat$COLOR, size = box.line.size, alpha = box.alpha, lineend = "round")) #
+coord.names <- c(coord.names, "inf.whisker.edge")
+}
+if(box.mean == TRUE){
+# assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_point(data = stat, mapping = ggplot2::aes_string(x = "X", y = "MEAN", group = categ[length(categ)]), shape = 23, stroke = box.line.size * 2, color = stat$COLOR, size = box.mean.size, fill = NA, alpha = box.alpha)) # group used in aesthetic to do not have it in the legend. Here ggplot2::scale_discrete_manual() cannot be used because of the group easthetic
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_path(
+data = tempo.diamon.mean,
+mapping = ggplot2::aes(x = X, y = Y, group = GROUP),
+color = tempo.diamon.mean[, "COLOR"],
+size = box.line.size,
+alpha = box.alpha,
+lineend = "round",
+linejoin = "round"
+))
+coord.names <- c(coord.names, "mean")
+}
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "fill", name = box.legend.name, values = rep(NA, length(unique(data1[, categ[length(categ)]]))))) #, guide = ggplot2::guide_legend(override.aes = list(color = categ.color)))) # values are the values of color (which is the border color in geom_box. WARNING: values = categ.color takes the numbers to make the colors if categ.color is a factor
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "color", name = box.legend.name, values = if(length(categ.color) == 1){rep(categ.color, length(unique(data1[, categ[length(categ)]])))}else{categ.color}, guide = ggplot2::guide_legend(override.aes = list(alpha = if(plot == TRUE & ((length(dev.list()) > 0 & names(dev.cur()) == "windows") | (length(dev.list()) == 0 & Sys.info()["sysname"] == "Windows"))){1}else{box.alpha})))) # , guide = ggplot2::guide_legend(override.aes = list(color = as.character(categ.color))))) # values are the values of color (which is the border color in geom_box. WARNING: values = categ.color takes the numbers to make the colors if categ.color is a factor
+if(plot == TRUE & ((length(dev.list()) > 0 & names(dev.cur()) == "windows") | (length(dev.list()) == 0 & Sys.info()["sysname"] == "Windows"))){ # if any Graph device already open and this device is "windows", or if no Graph device opened yet and we are on windows system -> prevention of alpha legend bug on windows using value 1
+# to avoid a bug on windows: if alpha argument is different from 1 for lines (transparency), then lines are not correctly displayed in the legend when using the R GUI (bug https://github.com/tidyverse/ggplot2/issues/2452). No bug when using a pdf
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") GRAPHIC DEVICE USED ON A WINDOWS SYSTEM ->\nTRANSPARENCY OF THE LINES IS INACTIVATED IN THE LEGEND TO PREVENT A WINDOWS DEPENDENT BUG (SEE https://github.com/tidyverse/ggplot2/issues/2452)\nTO OVERCOME THIS ON WINDOWS, USE ANOTHER DEVICE (pdf() FOR INSTANCE)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+if(box.alpha == 0){ # remove box legend because no boxes drawn
+# add this after the scale_xxx_manual() for boxplots
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::guides(fill = FALSE, color = FALSE)) # inactivate the legend
+}
+# end boxplot display (if box.fill = FALSE, otherwise, already plotted above)
+
+
+
+
+# stat display
+# layer after dots but ok, behind dots on the plot
+if( ! is.null(stat.disp)){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NUMBERS DISPLAYED ARE ", ifelse(stat.disp.mean == FALSE, "MEDIANS", "MEANS"))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+if(stat.disp == "top"){
+tempo.stat <- data.frame(stat, Y = y.lim[2], stringsAsFactors = TRUE) # I had to create a data frame for geom_tex() so that facet is taken into account, (ggplot2::annotate() does not deal with facet because no data and mapping arguments). Of note, facet.categ is in tempo.stat, via tempo.mean, via dot.coord
+if(stat.disp.mean == FALSE){tempo.stat$MEDIAN <- formatC(stat.nolog$MEDIAN, digit = 2, drop0trailing = TRUE, format = "f")}else{tempo.stat$MEAN <- formatC(stat.nolog$MEAN, digit = 2, drop0trailing = TRUE, format = "f")}
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_text(
+data = tempo.stat,
+mapping = ggplot2::aes_string(x = "X", y = "Y", label = ifelse(stat.disp.mean == FALSE, "MEDIAN", "MEAN")),
+size = stat.size,
+color = "black",
+angle = stat.angle,
+hjust = stat.just$hjust,
+vjust = stat.just$vjust
+)) # stat$X used here because identical to stat.nolog but has the X. WARNING: no need of order() for labels because box.coord$x set the order. For justification, see https://stackoverflow.com/questions/7263849/what-do-hjust-and-vjust-do-when-making-a-plot-using-ggplot
+coord.names <- c(coord.names, "stat.display")
+}else if(stat.disp == "above"){
+# stat coordinates
+if( ! is.null(dot.color)){ # for text just above max dot
+if(dot.tidy == FALSE){
+tempo.stat.ini <- dot.coord.rd3
+}else if(dot.tidy == TRUE){
+tempo.stat.ini <- dot.coord.tidy3
+tempo.stat.ini$x.y <- tempo.stat.ini$x.x # this is just to be able to use tempo.stat.ini$x.y for untidy or tidy dots (remember that dot.coord.tidy3$x.y is not good, see above)
+}
+stat.coord1 <- aggregate(x = tempo.stat.ini["y"], by = {x.env <- if(length(categ) == 1){list(tempo.stat.ini$group, tempo.stat.ini$PANEL, tempo.stat.ini$x.y, tempo.stat.ini[, categ[1]])}else if(length(categ) == 2){list(tempo.stat.ini$group, tempo.stat.ini$PANEL, tempo.stat.ini$x.y, tempo.stat.ini[, categ[1]], tempo.stat.ini[, categ[2]])} ; names(x.env) <- if(length(categ) == 1){c("group", "PANEL", "x.y", categ[1])}else if(length(categ) == 2){c("group", "PANEL", "x.y", categ[1], categ[2])} ; x.env}, FUN = min, na.rm = TRUE)
+names(stat.coord1)[names(stat.coord1) == "y"] <- "dot.min"
+stat.coord2 <- aggregate(x = tempo.stat.ini["y"], by = {x.env <- if(length(categ) == 1){list(tempo.stat.ini$group, tempo.stat.ini$PANEL, tempo.stat.ini$x.y, tempo.stat.ini[, categ[1]])}else if(length(categ) == 2){list(tempo.stat.ini$group, tempo.stat.ini$PANEL, tempo.stat.ini$x.y, tempo.stat.ini[, categ[1]], tempo.stat.ini[, categ[2]])} ; names(x.env) <- if(length(categ) == 1){c("group", "PANEL", "x.y", categ[1])}else if(length(categ) == 2){c("group", "PANEL", "x.y", categ[1], categ[2])} ; x.env}, FUN = max, na.rm = TRUE)
+names(stat.coord2) <- paste0(names(stat.coord2), "_from.dot.max")
+names(stat.coord2)[names(stat.coord2) == "y_from.dot.max"] <- "dot.max"
+stat.coord3 <- cbind(box.coord[order(box.coord$group, box.coord$PANEL), ], stat.coord1[order(stat.coord1$group, stat.coord1$x.y), ], stat.coord2[order(stat.coord2$group, stat.coord2$x.y), ], stringsAsFactors = TRUE) #
+if( ! all(identical(round(stat.coord3$x, 9), round(as.numeric(stat.coord3$x.y), 9)))){ # as.numeric() because stat.coord3$x is class "mapped_discrete" "numeric"
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nFUSION OF box.coord, stat.coord1 AND stat.coord2 ACCORDING TO box.coord$x, stat.coord1$x.y AND stat.coord2$x.y IS NOT CORRECT. CODE HAS TO BE MODIFIED")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# text.coord <- stat.coord3[, c("x", "group", "dot.min", "dot.max")]
+# names(text.coord)[names(text.coord) == "dot.min"] <- "text.min.pos"
+#names(text.coord)[names(text.coord) == "dot.max"] <- "text.max.pos"
+box.coord <- box.coord[order(box.coord$x, box.coord$group, box.coord$PANEL), ]
+# text.coord <- text.coord[order(text.coord$x), ] # to be sure to have the two objects in the same order for x. WARNING: cannot add identical(as.integer(text.coord$group), as.integer(box.coord$group)) because with error, the correspondence between x and group is not the same
+stat.coord3 <- stat.coord3[order(stat.coord3$x, stat.coord3$group, stat.coord3$PANEL), ] # to be sure to have the two objects in the same order for x. WARNING: cannot add identical(as.integer(text.coord$group), as.integer(box.coord$group)) because with error, the correspondence between x and group is not the same
+if( ! (identical(box.coord$x, stat.coord3$x) & identical(box.coord$group, stat.coord3$group) & identical(box.coord$PANEL, stat.coord3$PANEL))){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\ntext.coord AND box.coord DO NOT HAVE THE SAME x, group AND PANEL COLUMN CONTENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}else{
+stat.coord3 <- box.coord
+}
+stat.coord3 <- data.frame(
+stat.coord3,
+Y = stat.coord3[, ifelse(
+is.null(dot.color),
+ifelse(diff(y.lim) > 0, "ymax", "ymin"),
+ifelse(diff(y.lim) > 0, "ymax_final", "ymin_final")
+)],
+stringsAsFactors = TRUE
+) # ymax is top whisker, ymax_final is top dot
+# stat.coord3 <- data.frame(stat.coord3, Y = vector("numeric", length = nrow(stat.coord3)), stringsAsFactors = TRUE)
+# check.Y <- as.logical(stat.coord3$Y) # convert everything in Y into FALSE (because Y is full of zero)
+# end stat coordinates
+# stat display
+# performed twice: first for y values >=0, then y values < 0, because only a single value allowed for hjust anf vjust
+if(stat.disp.mean == FALSE){
+tempo.center.ref <- "middle"
+}else{
+tempo.center.ref <- "MEAN"
+}
+# if(is.null(dot.color)){
+# tempo.low.ref <- "ymin"
+# tempo.high.ref <- "ymax"
+# }else{
+# tempo.low.ref <- "ymin_final"
+# tempo.high.ref <- "ymax_final"
+# }
+# tempo.log.high <- if(diff(y.lim) > 0){stat.coord3[, tempo.center.ref] >= 0}else{stat.coord3[, tempo.center.ref] < 0}
+# tempo.log.low <- if(diff(y.lim) > 0){stat.coord3[, tempo.center.ref] < 0}else{stat.coord3[, tempo.center.ref] >= 0}
+# stat.coord3$Y[tempo.log.high] <- stat.coord3[tempo.log.high, tempo.high.ref]
+# stat.coord3$Y[tempo.log.low] <- stat.coord3[tempo.log.low, tempo.low.ref]
+# add distance
+stat.coord3$Y <- stat.coord3$Y + diff(y.lim) * stat.dist / 100
+# end add distance
+# correct median or mean text format
+if(y.log != "no"){
+stat.coord3[, tempo.center.ref] <- ifelse(y.log == "log2", 2, 10)^(stat.coord3[, tempo.center.ref])
+}
+stat.coord3[, tempo.center.ref] <- formatC(stat.coord3[, tempo.center.ref], digit = 2, drop0trailing = TRUE, format = "f")
+# end correct median or mean text format
+# if(any(tempo.log.high) == TRUE){
+# tempo.stat <- stat.coord3[tempo.log.high,]
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_text(
+data = stat.coord3,
+mapping = ggplot2::aes_string(x = "x", y = "Y", label = tempo.center.ref),
+size = stat.size,
+color = "black",
+angle = stat.angle,
+hjust = stat.just$hjust,
+vjust = stat.just$vjust
+)) # WARNING: no need of order() for labels because box.coord$x set the order
+coord.names <- c(coord.names, "stat.display")
+# }
+# if(any(tempo.log.low) == TRUE){
+# tempo.stat <- stat.coord3[tempo.log.low,]
+# assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_text(
+# data = tempo.stat,
+# mapping = ggplot2::aes_string(x = "x", y = "Y", label = tempo.center.ref),
+# size = stat.size,
+# color = "black",
+# hjust = ifelse(vertical == TRUE, 0.5, 0.5 + stat.dist),
+# vjust = ifelse(vertical == TRUE, 0.5 + stat.dist, 0.5)
+# )) # WARNING: no need of order() for labels because box.coord$x set the order
+# coord.names <- c(coord.names, "stat.display.negative")
+# }
+# end stat display
+}else{
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 5")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+}
+# end stat display
+# legend management
+if(legend.show == FALSE){ # must be here because must be before bef.final.plot <-
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::guides(fill = FALSE, color = FALSE, alpha = FALSE)) # inactivate the initial legend
+}
+# end legend management
+
+
+
+# y scale management (cannot be before dot plot management)
+# the rescaling aspect is complicated and not intuitive. See:
+# explaination: https://github.com/tidyverse/ggplot2/issues/3948
+# the oob argument of scale_y_continuous() https://ggplot2.tidyverse.org/reference/scale_continuous.html
+# see also https://github.com/rstudio/cheatsheets/blob/master/data-visualization-2.1.pdf
+# secondary ticks
+bef.final.plot <- ggplot2::ggplot_build(eval(parse(text = paste(paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "), ' + if(vertical == TRUE){ggplot2::scale_y_continuous(expand = c(0, 0), limits = sort(y.lim), oob = scales::rescale_none)}else{ggplot2::coord_flip(ylim = y.lim)}')))) # here I do not need the x-axis and y-axis orientation, I just need the number of main ticks and the legend
+tempo.coord <- bef.final.plot$layout$panel_params[[1]]
+# y.second.tick.positions: coordinates of secondary ticks (only if y.second.tick.nb argument is non NULL or if y.log argument is different from "no")
+if(y.log != "no"){ # integer main ticks for log2 and log10
+tempo.scale <- (as.integer(min(y.lim, na.rm = TRUE)) - 1):(as.integer(max(y.lim, na.rm = TRUE)) + 1)
+}else{
+tempo <- if(is.null(attributes(tempo.coord$y$breaks))){tempo.coord$y$breaks}else{unlist(attributes(tempo.coord$y$breaks))}
+if(all(is.na(tempo))){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nONLY NA IN tempo.coord$y$breaks")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+tempo.scale <- fun_scale(lim = y.lim, n = ifelse(is.null(y.tick.nb), length(tempo[ ! is.na(tempo)]), y.tick.nb)) # in ggplot 3.3.0, tempo.coord$y.major_source replaced by tempo.coord$y$breaks. If fact: n = ifelse(is.null(y.tick.nb), length(tempo[ ! is.na(tempo)]), y.tick.nb)) replaced by n = ifelse(is.null(y.tick.nb), 4, y.tick.nb))
+}
+y.second.tick.values <- NULL
+y.second.tick.pos <- NULL
+if(y.log != "no"){
+tempo <- fun_inter_ticks(lim = y.lim, log = y.log)
+y.second.tick.values <- tempo$values
+y.second.tick.pos <- tempo$coordinates
+# if(vertical == TRUE){ # do not remove in case the bug is fixed
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::annotate(geom = "segment", y = y.second.tick.pos, yend = y.second.tick.pos, x = tempo.coord$x.range[1], xend = tempo.coord$x.range[1] + diff(tempo.coord$x.range) / 80))
+# }else{ # not working because of the ggplot2 bug
+# assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::annotate(geom = "segment", x = y.second.tick.pos, xend = y.second.tick.pos, y = tempo.coord$y.range[1], yend = tempo.coord$y.range[1] + diff(tempo.coord$y.range) / 80))
+# }
+coord.names <- c(coord.names, "y.second.tick.positions")
+}else if(( ! is.null(y.second.tick.nb)) & y.log == "no"){
+# if(y.second.tick.nb > 0){ #inactivated because already checked before
+tempo <- fun_inter_ticks(lim = y.lim, log = y.log, breaks = tempo.scale, n = y.second.tick.nb)
+y.second.tick.values <- tempo$values
+y.second.tick.pos <- tempo$coordinates
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::annotate(
+geom = "segment",
+y = y.second.tick.pos,
+yend = y.second.tick.pos,
+x = if(vertical == TRUE){tempo.coord$x.range[1]}else{tempo.coord$y.range[1]},
+xend = if(vertical == TRUE){tempo.coord$x.range[1] + diff(tempo.coord$x.range) / 80}else{tempo.coord$y.range[1] + diff(tempo.coord$y.range) / 80}
+))
+coord.names <- c(coord.names, "y.second.tick.positions")
+}
+# end y.second.tick.positions
+# for the ggplot2 bug with y.log, this does not work: eval(parse(text = ifelse(vertical == FALSE & y.log == "log10", "ggplot2::scale_x_continuous", "ggplot2::scale_y_continuous")))
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_y_continuous(
+breaks = tempo.scale,
+minor_breaks = y.second.tick.pos,
+labels = if(y.log == "log10"){scales::trans_format("identity", scales::math_format(10^.x))}else if(y.log == "log2"){scales::trans_format("identity", scales::math_format(2^.x))}else if(y.log == "no"){ggplot2::waiver()}else{tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 6") ; stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)}, # == in stop() to be able to add several messages between ==
+expand = c(0, 0), # remove space after after axis limits
+limits = sort(y.lim), # NA indicate that limits must correspond to data limits but ylim() already used
+oob = scales::rescale_none,
+trans = ifelse(diff(y.lim) < 0, "reverse", "identity") # equivalent to ggplot2::scale_y_reverse() but create the problem of y-axis label disappearance with y.lim decreasing. Thus, do not use. Use ylim() below and after this
+))
+if(vertical == TRUE){
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::coord_cartesian(ylim = y.lim)) #problem of ggplot2::ylim() is taht it redraw new breaks # coord_cartesian(ylim = y.lim)) not used because bug -> y-axis label disappearance with y.lim decreasing # clip = "off" to have secondary ticks outside plot region does not work
+}else{
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::coord_flip(ylim = y.lim)) # clip = "off" to have secondary ticks outside plot region does not work # create the problem of y-axis label disappearance with y.lim decreasing
+
+}
+# end y scale management (cannot be before dot plot management)
+
+
+# legend management
+if( ! is.null(legend.width)){
+legend.final <- fun_gg_get_legend(ggplot_built = bef.final.plot, fun.name = function.name, lib.path = lib.path) # get legend
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::guides(fill = FALSE, color = FALSE, alpha = FALSE)) # inactivate the initial legend
+if(is.null(legend.final) & plot == TRUE){ # even if any(unlist(legend.disp)) is TRUE
+legend.final <- ggplot2::ggplot()+ggplot2::theme_void() # empty graph instead of legend
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") LEGEND REQUESTED (NON NULL categ ARGUMENT OR legend.show ARGUMENT SET TO TRUE)\nBUT IT SEEMS THAT THE PLOT HAS NO LEGEND -> EMPTY LEGEND SPACE CREATED BECAUSE OF THE NON NULL legend.width ARGUMENT\n")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+# end legend management
+
+
+# drawing
+fin.plot <- suppressMessages(suppressWarnings(eval(parse(text = paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + ")))))
+grob.save <- NULL
+if(plot == TRUE){
+# following lines inactivated because of problem in warn.recov and message.recov
+# assign("env_fun_get_message", new.env())
+# assign("tempo.gg.name", tempo.gg.name, envir = env_fun_get_message)
+# assign("tempo.gg.count", tempo.gg.count, envir = env_fun_get_message)
+# assign("add", add, envir = env_fun_get_message)
+# two next line: for the moment, I cannot prevent the warning printing
+# warn.recov <- fun_get_message(paste(paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "), if(is.null(add)){NULL}else{add}), kind = "warning", header = FALSE, print.no = FALSE, env = env_fun_get_message) # for recovering warnings printed by ggplot() functions
+# message.recov <- fun_get_message('print(eval(parse(text = paste(paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "), if(is.null(add)){NULL}else{add}))))', kind = "message", header = FALSE, print.no = FALSE, env = env_fun_get_message) # for recovering messages printed by ggplot() functions
+# if( ! (return == TRUE & return.ggplot == TRUE)){ # because return() plots when return.ggplot is TRUE # finally not used -> see return.ggplot description
+if(is.null(legend.width)){
+grob.save <- suppressMessages(suppressWarnings(gridExtra::grid.arrange(fin.plot)))
+}else{
+grob.save <-suppressMessages(suppressWarnings(gridExtra::grid.arrange(fin.plot, legend.final, ncol=2, widths=c(1, legend.width))))
+}
+# }
+# suppressMessages(suppressWarnings(print(eval(parse(text = paste(paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "), if(is.null(add)){NULL}else{add}))))))
+}else{
+# following lines inactivated because of problem in warn.recov and message.recov
+# message.recov <- NULL
+# warn.recov <- NULL
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") PLOT NOT SHOWN AS REQUESTED")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+# end drawing
+
+
+
+# outputs
+# following lines inactivated because of problem in warn.recov and message.recov
+# if( ! (is.null(warn) & is.null(warn.recov) & is.null(message.recov))){
+# warn <- paste0(warn, "\n\n", if(length(warn.recov) > 0 | length(message.recov) > 0){paste0(paste0("MESSAGES FROM ggplot2 FUNCTIONS: ", ifelse( ! is.null(warn.recov), unique(message.recov), ""), ifelse( ! is.null(message.recov), unique(message.recov), ""), collapse = "\n\n"), "\n\n")})
+# }else if( ! (is.null(warn) & is.null(warn.recov)) & is.null(message.recov)){
+# warn <- paste0(warn, "\n\n", if(length(warn.recov) > 0){paste0(paste0("MESSAGES FROM ggplot2 FUNCTIONS: ", unique(warn.recov), collapse = "\n\n"), "\n\n")})
+# }else if( ! (is.null(warn) & is.null(message.recov)) & is.null(warn.recov)){
+# warn <- paste0(warn, "\n\n", if(length(message.recov) > 0){paste0(paste0("MESSAGES FROM ggplot2 FUNCTIONS: ", unique(message.recov), collapse = "\n\n"), "\n\n")})
+# }
+if(warn.print == TRUE & ! is.null(warn)){
+options(warning.length = 8170)
+on.exit(warning(paste0("FROM ", function.name, ":\n\n", warn), call. = FALSE))
+on.exit(exp = options(warning.length = ini.warning.length), add = TRUE)
+}
+if(return == TRUE){
+tempo.output <- ggplot2::ggplot_build(fin.plot)
+tempo.output$data <- tempo.output$data[-1] # remove the first data because corresponds to the initial empty boxplot
+if(length(tempo.output$data) != length(coord.names)){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nlength(tempo.output$data) AND length(coord.names) MUST BE IDENTICAL. CODE HAS TO BE MODIFIED")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
+}else{
+names(tempo.output$data) <- coord.names
+tempo.output$data <- tempo.output$data[coord.names != "bad_remove"]
+}
+tempo <- tempo.output$layout$panel_params[[1]]
+output <- list(
+data = data1.ini,
+stat = stat.nolog,
+removed.row.nb = removed.row.nb,
+removed.rows = removed.rows,
+plot = c(tempo.output$data, y.second.tick.values = list(y.second.tick.values)),
+panel = facet.categ,
+axes = list(
+x.range = tempo$x.range,
+x.labels = if(is.null(attributes(tempo$x$breaks))){tempo$x$breaks}else{tempo$x$scale$get_labels()}, # is.null(attributes(tempo$x$breaks)) test if it is number (TRUE) or character (FALSE)
+x.positions = if(is.null(attributes(tempo$x$breaks))){tempo$x$breaks}else{unlist(attributes(tempo$x$breaks))},
+y.range = tempo$y.range,
+y.labels = if(is.null(attributes(tempo$y$breaks))){tempo$y$breaks}else{tempo$y$scale$get_labels()},
+y.positions = if(is.null(attributes(tempo$y$breaks))){tempo$y$breaks}else{unlist(attributes(tempo$y$breaks))}
+),
+warn = paste0("\n", warn, "\n\n"),
+ggplot = if(return.ggplot == TRUE){fin.plot}else{NULL}, # fin.plot plots the graph if return == TRUE
+gtable = if(return.gtable == TRUE){grob.save}else{NULL}
+)
+return(output) # this plots the graph if return.ggplot is TRUE and if no assignment
+}
+# end outputs
+# end main code
+}
+
+
+
+
+
+# add density
+# add legend order
+
+
+fun_gg_scatter <- function(
+data1,
+x,
+y,
+categ = NULL,
+categ.class.order = NULL,
+color = NULL,
+geom = "geom_point",
+geom.step.dir = "hv",
+alpha = 0.5,
+dot.size = 2,
+dot.shape = 21,
+dot.border.size = 0.5,
+dot.border.color = NULL,
+line.size = 0.5,
+line.type = "solid",
+x.lim = NULL,
+x.lab = NULL,
+x.log = "no",
+x.tick.nb = NULL,
+x.second.tick.nb = NULL,
+x.include.zero = FALSE,
+x.left.extra.margin = 0.05,
+x.right.extra.margin = 0.05,
+x.text.angle = 0,
+y.lim = NULL,
+y.lab = NULL,
+y.log = "no",
+y.tick.nb = NULL,
+y.second.tick.nb = NULL,
+y.include.zero = FALSE,
+y.top.extra.margin = 0.05,
+y.bottom.extra.margin = 0.05,
+y.text.angle = 0,
+raster = FALSE,
+raster.ratio = 1,
+raster.threshold = NULL,
+text.size = 12,
+title = "",
+title.text.size = 12,
+legend.show = TRUE,
+legend.width = 0.5,
+legend.name = NULL,
+article = TRUE,
+grid = FALSE,
+add = NULL,
+return = FALSE,
+return.ggplot = FALSE,
+return.gtable = TRUE,
+plot = TRUE,
+warn.print = FALSE,
+lib.path = NULL
+){
+# AIM
+# plot ggplot2 scatterplot with the possibility to overlay dots from up to 3 different data frames (-> three different legends) and lines from up to 3 different data frames (-> three different legends) -> up to 6 overlays totally
+# for ggplot2 specifications, see: https://ggplot2.tidyverse.org/articles/ggplot2-specs.html
+# WARNINGS
+# Rows containing NA in data1[, c(x, y, categ)] will be removed before processing, with a warning (see below)
+# Size arguments (dot.size, dot.border.size, line.size, text.size and title.text.size) are in mm. See Hadley comment in https://stackoverflow.com/questions/17311917/ggplot2-the-unit-of-size. See also http://sape.inf.usi.ch/quick-reference/ggplot2/size). Unit object are not accepted, but conversion can be used (e.g., grid::convertUnit(grid::unit(0.2, "inches"), "mm", valueOnly = TRUE))
+# ARGUMENTS
+# data1: a dataframe compatible with ggplot2, or a list of data frames. Order matters for the order of the legend and for the layer staking (starting from below to top)
+# x: single character string of the data1 column name for x-axis coordinates. If data1 is a list, then x must be a list of single character strings, of same size as data1, with compartment 1 related to compartment 1 of data1, etc. Write NULL for each "geom_hline" in geom argument
+# y: single character string of the data1 column name for y-axis coordinates. If data1 is a list, then y must be a list of single character strings, of same size as data1, with compartment 1 related to compartment 1 of data1, etc. Write NULL for each "geom_vline" in geom argument
+# categ: either NULL or a single character string or a list of single character strings, indicating the data1 column names to use for categories which creates legend display
+# If categ == NULL, no categories -> no legend displayed
+# If data1 is a data frame, categ must be a single character string of the data1 column name for categories
+# If data1 is a list, then categ must be a list of single character strings, of same size as data1, with compartment 1 related to compartment 1 of data1, etc. Some of the list compartments can be NULL (no legend display for these compartments), and other not
+# categ.class.order: either (1) NULL or (2) a vector of character strings or (3) a list of these vectors, setting the order of the classes of categ in the legend display
+# If categ.class.order is NULL, classes are represented according to the alphabetical order
+# If data1 is a data frame, categ.class.order must be a vector of character strings specifying the different classes in the categ column name of data1
+# If data1 is a list, then categ.class.order must be a list of vector of character strings, of same size as data1, with compartment 1 related to compartment 1 of data1, etc. Some of the list compartments can be NULL (alphabetical order for these compartments), and other not
+# color: either (1) NULL, or (2) a vector of character strings or integers, or (3) a list of vectors of character strings or integers
+# If color is NULL, default colors of ggplot2
+# If data1 is a data frame, color argument can be either:
+# (1) a single color string. All the dots of the corresponding data1 will have this color, whatever the categ value (NULL or not)
+# (2) if categ is non-null, a vector of string colors, one for each class of categ. Each color will be associated according to the categ.class.order argument if specified, or to the alphabetical order of categ classes otherwise
+# (3) if categ is non-null, a vector or factor of string colors, like if it was one of the column of data1 data frame. WARNING: a single color per class of categ and a single class of categ per color must be respected
+# Integers are also accepted instead of character strings, as long as above rules about length are respected. Integers will be processed by fun_gg_palette() using the max integer value among all the integers in color (see fun_gg_palette())
+# If data1 is a list, then color argument must be either:
+# (1) a list of character strings or integers, of same size as data1, with compartment 1 related to compartment 1 of data1, etc.
+# (2) a single character string or a single integer
+# With a list (first possibility), the rules described for when data1 is a data frame apply to each compartment of the list. Some of the compartments can be NULL. In that case, a different grey color will be used for each NULL compartment. With a single value (second possibility), the same color will be used for all the dots and lines, whatever the data1 list
+# geom: single character string of the kind of plot, or a list of single character strings
+# Either:
+# "geom_point" (scatterplot)
+# "geom_line" (coordinates plotted then line connection, from the lowest to highest x coordinates first and from the lowest to highest y coordinates thenafter)
+# "geom_path" (coordinates plotted then line connection respecting the row order in data1)
+# "geom_step" coordinates plotted then line connection respecting the row order in data1 but drawn in steps). See the geom.step.dir argument
+# "geom_hline" (horizontal line, no x value provided)
+# "geom_vline" (vertical line, no y value provided)
+# "geom_stick" (dots as vertical bars)
+# If data1 is a list, then geom must be either:
+# (1) a list of single character strings, of same size as data1, with compartment 1 related to compartment 1 of data1, etc.
+# (2) a single character string. In that case the same kind of plot will apply for the different compartments of the data1 list
+# WARNING concerning "geom_hline" or "geom_vline":
+# (1) x or y argument must be NULL, respectively
+# (2) x.lim or y.lim argument must NOT be NULL, respectively, if only these kind of lines are drawn (if other geom present, then x.lim = NULL and y.lim = NULL will generate x.lim and y.lim defined by these other geom, which is not possible with "geom_hline" or "geom_vline" alone)
+# (3) the function will draw n lines for n values in the x argument column name of the data1 data frame. If several colors required, the categ argument must be specified and the corresponding categ column name must exist in the data1 data frame with a different class name for each row
+# geom.step.dir: single character string indicating the direction when using "geom_step" of the geom argument, or a list of single character strings
+# Either:
+# "vh" (vertical then horizontal)
+# "hv" (horizontal then vertical)
+# "mid" (step half-way between adjacent x-values)
+# See https://ggplot2.tidyverse.org/reference/geom_path.html
+# If data1 is a list, then geom.step.dir must be either:
+# (1) a list of single character strings, of same size as data1, with compartment 1 related to compartment 1 of data1, etc. The value in compartments related to other geom values than "geom_step" will be ignored
+# (2) a single character string, which will be used for all the "geom_step" values of the geom argument, whatever the data1 list
+# alpha: single numeric value (from 0 to 1) of transparency. If data1 is a list, then alpha must be either (1) a list of single numeric values, of same size as data1, with compartment 1 related to compartment 1 of data1, etc., or (2) a single numeric value. In that case the same transparency will apply for the different compartments of the data1 list
+# dot.size: single numeric value of dot shape radius? in mm. If data1 is a list, then dot.size must be either (1) a list of single numeric values, of same size as data1, with compartment 1 related to compartment 1 of data1, etc., or (2) a single numeric value. With a list (former possibility), the value in compartments related to lines will be ignored. With a single value (latter possibility), the same dot.size will be used for all the dots, whatever the data1 list
+# dot.shape: value indicating the shape of the dots (see https://ggplot2.tidyverse.org/articles/ggplot2-specs.html) If data1 is a list, then dot.shape must be either (1) a list of single shape values, of same size as data1, with compartment 1 related to compartment 1 of data1, etc., or (2) a single shape value. With a list (former possibility), the value in compartments related to lines will be ignored. With a single value (latter possibility), the same dot.shape will be used for all the dots, whatever the data1 list
+# dot.border.size: single numeric value of border dot width in mm. Write zero for no dot border. If data1 is a list, then dot.border.size must be either (1) a list of single numeric values, of same size as data1, with compartment 1 related to compartment 1 of data1, etc., or (2) a single numeric value. With a list (former possibility), the value in compartments related to lines will be ignored. With a single value (latter possibility), the same dot.border.size will be used for all the dots, whatever the data1 list
+# dot.border.color: single character color string defining the color of the dot border (same border color for all the dots, whatever their categories). If dot.border.color == NULL, the border color will be the same as the dot color. A single integer is also accepted instead of a character string, that will be processed by fun_gg_palette()
+# line.size: single numeric value of line width in mm. If data1 is a list, then line.size must be either (1) a list of single numeric values, of same size as data1, with compartment 1 related to compartment 1 of data1, etc., or (2) a single numeric value. With a list (former possibility), the value in compartments related to dots will be ignored. With a single value (latter possibility), the same line.size will be used for all the lines, whatever the data1 list
+# line.type: value indicating the kind of lines (see https://ggplot2.tidyverse.org/articles/ggplot2-specs.html) If data1 is a list, then line.type must be either (1) a list of single line kind values, of same size as data1, with compartment 1 related to compartment 1 of data1, etc., or (2) a single line kind value. With a list (former possibility), the value in compartments related to dots will be ignored. With a single value (latter possibility), the same line.type will be used for all the lines, whatever the data1 list
+# x.lim: 2 numeric values setting the x-axis range. Order of the 2 values matters (for inverted axis). If NULL, the range of the x column name of data1 will be used
+# x.lab: a character string or expression for x-axis label. If NULL, will use the first value of x (x column name of the first data frame in data1). Warning message if the elements in x are different between data frames in data1
+# x.log: either "no", "log2" (values in the x column name of the data1 data frame will be log2 transformed and x-axis will be log2 scaled) or "log10" (values in the x column name of the data1 data frame will be log10 transformed and x-axis will be log10 scaled)
+# x.tick.nb: approximate number of desired values labeling the x-axis (i.e., main ticks, see the n argument of the the cute::fun_scale() function). If NULL and if x.log is "no", then the number of labeling values is set by ggplot2. If NULL and if x.log is "log2" or "log10", then the number of labeling values corresponds to all the exposant integers in the x.lim range (e.g., 10^1, 10^2 and 10^3, meaning 3 main ticks for x.lim = c(9, 1200)). WARNING: if non-NULL and if x.log is "log2" or "log10", labeling can be difficult to read (e.g., ..., 10^2, 10^2.5, 10^3, ...)
+# x.second.tick.nb: number of desired secondary ticks between main ticks. Ignored if x.log is other than "no" (log scale plotted). Use argument return = TRUE and see $plot$x.second.tick.values to have the values associated to secondary ticks. IF NULL, no secondary ticks
+# x.include.zero: logical. Does x.lim range include 0? Ignored if x.log is "log2" or "log10"
+# x.left.extra.margin: single proportion (between 0 and 1) indicating if extra margins must be added to x.lim. If different from 0, add the range of the axis multiplied by x.left.extra.margin (e.g., abs(x.lim[2] - x.lim[1]) * x.left.extra.margin) to the left of x-axis
+# x.right.extra.margin: idem as x.left.extra.margin but to the right of x-axis
+# x.text.angle: integer value of the text angle for the x-axis labeling values, using the same rules as in ggplot2. Use positive value for clockwise rotation: 0 for horizontal, 90 for vertical, 180 for upside down etc. Use negative values for counterclockwise rotation: 0 for horizontal, -90 for vertical, -180 for upside down etc.
+# y.lim: 2 numeric values setting the y-axis range. Order of the 2 values matters (for inverted axis). If NULL, the range of the y column name of data1 will be used
+# y.lab: a character string or expression for y-axis label. If NULL, will use the first value of y (y column name of the first data frame in data1). Warning message if the elements in y are different between data frames in data1
+# y.log: either "no", "log2" (values in the y column name of the data1 data frame will be log2 transformed and y-axis will be log2 scaled) or "log10" (values in the y column name of the data1 data frame will be log10 transformed and y-axis will be log10 scaled)
+# y.tick.nb: approximate number of desired values labeling the y-axis (i.e., main ticks, see the n argument of the the cute::fun_scale() function). If NULL and if y.log is "no", then the number of labeling values is set by ggplot2. If NULL and if y.log is "log2" or "log10", then the number of labeling values corresponds to all the exposant integers in the y.lim range (e.g., 10^1, 10^2 and 10^3, meaning 3 main ticks for y.lim = c(9, 1200)). WARNING: if non-NULL and if y.log is "log2" or "log10", labeling can be difficult to read (e.g., ..., 10^2, 10^2.5, 10^3, ...)
+# y.second.tick.nb: number of desired secondary ticks between main ticks. Ignored if y.log is other than "no" (log scale plotted). Use argument return = TRUE and see $plot$y.second.tick.values to have the values associated to secondary ticks. IF NULL, no secondary ticks
+# y.include.zero: logical. Does y.lim range include 0? Ignored if y.log is "log2" or "log10"
+# y.top.extra.margin: single proportion (between 0 and 1) indicating if extra margins must be added to y.lim. If different from 0, add the range of the axis multiplied by y.top.extra.margin (e.g., abs(y.lim[2] - y.lim[1]) * y.top.extra.margin) to the top of y-axis
+# y.bottom.extra.margin: idem as y.top.extra.margin but to the bottom of y-axis
+# y.text.angle: integer value of the text angle for the y-axis labeling values, using the same rules as in ggplot2. Use positive value for clockwise rotation: 0 for horizontal, 90 for vertical, 180 for upside down etc. Use negative values for counterclockwise rotation: 0 for horizontal, -90 for vertical, -180 for upside down etc.
+# raster: logical. Dots in raster mode? If FALSE, dots from each "geom_point" from geom argument are plotted in vectorial mode (bigger pdf and long to display if lots of dots). If TRUE, dots from each "geom_point" from geom argument are plotted in matricial mode (smaller pdf and easy display if lots of dots, but it takes time to generate the layer). If TRUE, the raster.ratio argument is used to avoid an ellipsoid representation of the dots. If TRUE, solve the transparency problem with some GUI. Overriden by the non-NULL raster.threshold argument
+# raster.ratio: single numeric value indicating the height / width ratio of the graphic device used (for instance provided by the $dim compartment in the output of the fun_open() function). The default value is 1 because by default R opens a square graphic device. But this argument has to be set when using other device dimensions. Ignored if raster == FALSE
+# raster.threshold: positive integer value indicating the limit of the dot number above which "geom_point" layers from the geom argument switch from vectorial mode to matricial mode (see the raster argument). If any layer is matricial, then the raster.ratio argument is used to avoid an ellipsoid representation of the dots. If non-NULL, it overrides the raster argument
+# text.size: numeric value of the font size of the (1) axis numbers and axis legends and (2) texts in the graphic legend (in mm)
+# title: character string of the graph title
+# title.text.size: numeric value of the title font size in mm
+# legend.show: logical. Show legend? Not considered if categ argument is NULL, because this already generate no legend, excepted if legend.width argument is non-NULL. In that specific case (categ is NULL, legend.show is TRUE and legend.width is non-NULL), an empty legend space is created. This can be useful when desiring graphs of exactly the same width, whatever they have legends or not
+# legend.width: single proportion (between 0 and 1) indicating the relative width of the legend sector (on the right of the plot) relative to the width of the plot. Value 1 means that the window device width is split in 2, half for the plot and half for the legend. Value 0 means no room for the legend, which will overlay the plot region. Write NULL to inactivate the legend sector. In such case, ggplot2 will manage the room required for the legend display, meaning that the width of the plotting region can vary between graphs, depending on the text in the legend
+# legend.name: character string of the legend title. If legend.name is NULL and categ argument is not NULL, then legend.name <- categ. If data1 is a list, then legend.name must be a list of character strings, of same size as data1, with compartment 1 related to compartment 1 of data1, etc. Some of the list compartments can be NULL, and other not
+# article: logical. If TRUE, use an article theme (article like). If FALSE, use a classic related ggplot theme. Use the add argument (e.g., add = "+ggplot2::theme_classic()" for the exact classic ggplot theme
+# grid: logical. Draw lines in the background to better read the box values? Not considered if article == FALSE (grid systematically present)
+# add: character string allowing to add more ggplot2 features (dots, lines, themes, facet, etc.). Ignored if NULL
+# WARNING: (1) the string must start with "+", (2) the string must finish with ")" and (3) each function must be preceded by "ggplot2::". Example: "+ ggplot2::coord_flip() + ggplot2::theme_bw()"
+# If the character string contains the "ggplot2::theme" string, then the article argument of fun_gg_scatter() (see above) is ignored with a warning
+# Handle the add argument with caution since added functions can create conflicts with the preexisting internal ggplot2 functions
+# WARNING: the call of objects inside the quotes of add can lead to an error if the name of these objects are some of the fun_gg_scatter() arguments. Indeed, the function will use the internal argument instead of the global environment object. Example article <- "a" in the working environment and add = '+ ggplot2::ggtitle(article)'. The risk here is to have TRUE as title. To solve this, use add = '+ ggplot2::ggtitle(get("article", envir = .GlobalEnv))'
+# return: logical. Return the graph parameters?
+# return.ggplot: logical. Return the ggplot object in the output list? Ignored if return argument is FALSE. WARNING: always assign the fun_gg_scatter() function (e.g., a <- fun_gg_scatter()) if return.ggplot argument is TRUE, otherwise, double plotting is performed. See $ggplot in the RETURN section below for more details
+# return.gtable: logical. Return the ggplot object as gtable of grobs in the output list? Ignored if plot argument is FALSE. Indeed, the graph must be plotted to get the grobs dispositions. See $gtable in the RETURN section below for more details
+# plot: logical. Plot the graphic? If FALSE and return argument is TRUE, graphical parameters and associated warnings are provided without plotting
+# warn.print: logical. Print warnings at the end of the execution? ? If FALSE, warning messages are never printed, but can still be recovered in the returned list. Some of the warning messages (those delivered by the internal ggplot2 functions) are not apparent when using the argument plot = FALSE
+# lib.path: character string indicating the absolute path of the required packages (see below). if NULL, the function will use the R library default folders
+# REQUIRED PACKAGES
+# ggplot2
+# gridExtra
+# lemon (in case of use in the add argument)
+# scales
+# if raster plots are drawn (see the raster and raster.threshold arguments):
+# Cairo
+# grid
+# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
+# fun_gg_empty_graph()
+# fun_gg_palette()
+# fun_gg_point_rast()
+# fun_pack()
+# fun_check()
+# fun_round()
+# fun_scale()
+# fun_inter_ticks()
+# RETURN
+# a scatter plot if plot argument is TRUE
+# a list of the graph info if return argument is TRUE:
+# $data: the initial data with graphic information added. WARNING: if the x.log or y.log argument is not "no", x or y argument column of the data1 data frame are log2 or log10 converted in $data, respectively. Use 2^values or 10^$values to recover the initial values
+# $removed.row.nb: a list of the removed rows numbers in data frames (because of NA). NULL if no row removed
+# $removed.rows: a list of the removed rows in data frames (because of NA). NULL if no row removed
+# $plot: the graphic box and dot coordinates
+# $dots: dot coordinates
+# y.second.tick.positions: coordinates of secondary ticks (only if y.second.tick.nb argument is non-null or if y.log argument is different from "no")
+# y.second.tick.values: values of secondary ticks. NULL except if y.second.tick.nb argument is non-null or if y.log argument is different from "no")
+# $panel: the variable names used for the panels (NULL if no panels). WARNING: NA can be present according to ggplot2 upgrade to v3.3.0
+# $axes: the x-axis and y-axis info
+# $warn: the warning messages. Use cat() for proper display. NULL if no warning. WARNING: warning messages delivered by the internal ggplot2 functions are not apparent when using the argument plot = FALSE
+# $ggplot: ggplot object that can be used for reprint (use print($ggplot) or update (use $ggplot + ggplot2::...). NULL if return.ggplot argument is FALSE. Of note, a non-null $ggplot in the output list is sometimes annoying as the manipulation of this list prints the plot
+# $gtable: gtable object that can be used for reprint (use gridExtra::grid.arrange(...$ggplot) or with additionnal grobs (see the grob decomposition in the examples). NULL if return.ggplot argument is FALSE. Contrary to $ggplot, a non-NULL $gtable in the output list is not annoying as the manipulation of this list does not print the plot
+# EXAMPLES
+# DEBUGGING
+# set.seed(1) ; obs1 <- data.frame(km = rnorm(1000, 10, 3), time = rnorm(1000, 10, 3), group1 = rep(c("A1", "A2"), 500), stringsAsFactors = TRUE) ; obs2 <-data.frame(km = rnorm(1000, 15, 3), time = rnorm(1000, 15, 3), group2 = rep(c("G1", "G2"), 500), stringsAsFactors = TRUE) ; set.seed(NULL) ; obs1$km[2:3] <- NA ; data1 = list(L1 = obs1, L2 = obs2) ; x = list(L1 = "km", L2 = "km") ; y = list(L1 = "time", L2 = "time") ; categ = list(L1 = "group1", L2 = "group2") ; categ.class.order = NULL ; legend.name = NULL ; color = list(L1 = 4:5, L2 = 7:8) ; geom = list(L1 = "geom_point", L2 = "geom_point") ; geom.step.dir = "hv" ; alpha = list(L1 = 0.5, L2 = 0.5) ; dot.size = 3 ; dot.shape = 21 ; dot.border.size = 0.5 ; dot.border.color = NULL ; line.size = 0.5 ; line.type = "solid" ; x.lim = NULL ; x.lab = "KM" ; x.log = "no" ; x.tick.nb = 10 ; x.second.tick.nb = 1 ; x.left.extra.margin = 0 ; x.right.extra.margin = 0 ; y.lim = c(1, 25) ; y.lab = "TIME (s)" ; y.log = "no" ; y.tick.nb = 5 ; y.second.tick.nb = 2 ; y.top.extra.margin = 0 ; y.bottom.extra.margin = 0 ; x.include.zero = TRUE ; y.include.zero = TRUE ; x.text.angle = 0 ; y.text.angle = 0 ; text.size = 12 ; title = "" ; title.text.size = 8 ; legend.show = TRUE ; legend.width = 0.5 ; article = FALSE ; grid = FALSE ; raster = TRUE ; raster.ratio = 1 ; raster.threshold = NULL ; return = FALSE ; return.ggplot = FALSE ; return.gtable = FALSE ; plot = TRUE ; add = NULL ; warn.print = TRUE ; lib.path = NULL
+# data1 <- list(L1 = data.frame(a = 1:6, b = (1:6)^2, group = c("A", "A", "A", "B", "B", "B"), stringsAsFactors = TRUE), L2 = data.frame(a = (1:6)*2, b = ((1:6)^2)*2, group = c("A1", "A1", "A1", "B1", "B1", "B1"), stringsAsFactors = TRUE), L3 = data.frame(a = (1:6)*3, b = ((1:6)^2)*3, group3 = c("A4", "A5", "A6", "A7", "B4", "B5"), stringsAsFactors = TRUE)) ; data1$L1$a[3] <- NA ; data1$L1$group[5] <- NA ; data1$L3$group3[4] <- NA ; data1 ; x = list(L1 = names(data1$L1)[1], L2 = names(data1$L2)[1], L3 = NULL) ; y = list(L1 = names(data1$L1)[2], L2 = names(data1$L2)[2], L3 = "a") ; categ = list(L1 = "group", L2 = NULL, L3 = NULL) ; categ.class.order = NULL ; legend.name = NULL ; color = NULL ; geom = list(L1 = "geom_point", L2 = "geom_point", L3 = "geom_hline") ; geom.step.dir = "hv" ; alpha = list(L1 = 0.5, L2 = 0.5, L3 = 0.5) ; dot.size = 1 ; dot.shape = 21 ; dot.border.size = 0.5 ; dot.border.color = NULL ; line.size = 0.5 ; line.type = "solid" ; x.lim = c(14, 4) ; x.lab = NULL ; x.log = "log10" ; x.tick.nb = 10 ; x.second.tick.nb = 4 ; x.left.extra.margin = 0 ; x.right.extra.margin = 0 ; y.lim = c(60, 5) ; y.lab = NULL ; y.log = "log10" ; y.tick.nb = 10 ; y.second.tick.nb = 2 ; y.top.extra.margin = 0 ; y.bottom.extra.margin = 0 ; x.include.zero = TRUE ; y.include.zero = TRUE ; x.text.angle = 0 ; y.text.angle = 0 ; text.size = 12 ; title = "" ; title.text.size = 8 ; legend.show = TRUE ; legend.width = 0.5 ; article = TRUE ; grid = FALSE ; raster = FALSE ; raster.ratio = 1 ; raster.threshold = NULL ; return = TRUE ; return.ggplot = FALSE ; return.gtable = FALSE ; plot = TRUE ; add = NULL ; warn.print = TRUE ; lib.path = NULL
+# data1 <- data.frame(km = 2:7, time = (2:7)^2, group = c("A", "A", "A", "B", "B", "B"), stringsAsFactors = TRUE) ; data1 ; x = "km"; y = "time"; categ = "group" ; categ.class.order = NULL ; legend.name = NULL ; color = NULL ; geom = "geom_point" ; geom.step.dir = "hv" ; alpha = 0.1 ; dot.size = 3 ; dot.shape = 21 ; dot.border.size = 0.5 ; dot.border.color = NULL ; line.size = 0.5 ; line.type = "solid" ; x.lim = c(1,10) ; x.lab = NULL ; x.log = "log10" ; x.tick.nb = 10 ; x.second.tick.nb = 4 ; x.left.extra.margin = 0 ; x.right.extra.margin = 0 ; y.lim = NULL ; y.lab = expression(paste("TIME (", 10^-20, " s)")) ; y.log = "log10" ; y.tick.nb = 10 ; y.second.tick.nb = 2 ; y.top.extra.margin = 0 ; y.bottom.extra.margin = 0 ; x.include.zero = TRUE ; y.include.zero = TRUE ; x.text.angle = 0 ; y.text.angle = 0 ; text.size = 12 ; title = "" ; title.text.size = 8 ; legend.show = TRUE ; legend.width = 0.5 ; article = FALSE ; grid = FALSE ; raster = FALSE ; raster.ratio = 1 ; raster.threshold = NULL ; return = FALSE ; return.ggplot = FALSE ; return.gtable = FALSE ; plot = TRUE ; add = NULL ; warn.print = TRUE ; lib.path = NULL
+# function name
+function.name <- paste0(as.list(match.call(expand.dots=FALSE))[[1]], "()")
+arg.names <- names(formals(fun = sys.function(sys.parent(n = 2)))) # names of all the arguments
+arg.user.setting <- as.list(match.call(expand.dots=FALSE))[-1] # list of the argument settings (excluding default values not provided by the user)
+# end function name
+# required function checking
+req.function <- c(
+"fun_check",
+"fun_gg_just",
+"fun_gg_empty_graph",
+"fun_gg_palette",
+"fun_gg_point_rast",
+"fun_round",
+"fun_pack",
+"fun_scale",
+"fun_inter_ticks"
+)
+for(i1 in req.function){
+if(length(find(i1, mode = "function")) == 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED ", i1, "() FUNCTION IS MISSING IN THE R ENVIRONMENT")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
+}
+}
+# end required function checking
+# reserved words to avoid bugs (used in this function)
+reserved.words <- c("fake_x", "fake_y", "fake_categ")
+# end reserved words to avoid bugs (used in this function)
+# arg with no default values
+if(any(missing(data1) | missing(x) | missing(y))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ARGUMENTS angle AND pos HAVE NO DEFAULT VALUE AND REQUIRE ONE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end arg with no default values
+# primary argument checking
+arg.check <- NULL #
+text.check <- NULL #
+checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
+ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
+tempo1 <- fun_check(data = data1, class = "data.frame", na.contain = TRUE, fun.name = function.name)
+tempo2 <- fun_check(data = data1, class = "list", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": data1 ARGUMENT MUST BE A DATA FRAME OR A LIST OF DATA FRAMES")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+if( ! is.null(x)){
+tempo1 <- fun_check(data = x, class = "vector", mode = "character", na.contain = TRUE, length = 1, fun.name = function.name)
+tempo2 <- fun_check(data = x, class = "list", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": x ARGUMENT MUST BE A SINGLE CHARACTER STRING OR A LIST OF CHARACTER STRINGS")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+if( ! is.null(y)){
+tempo1 <- fun_check(data = y, class = "vector", mode = "character", na.contain = TRUE, length = 1, fun.name = function.name)
+tempo2 <- fun_check(data = y, class = "list", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": y ARGUMENT MUST BE A SINGLE CHARACTER STRING OR A LIST OF CHARACTER STRINGS")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+if( ! is.null(categ)){
+tempo1 <- fun_check(data = categ, class = "vector", mode = "character", length = 1, fun.name = function.name)
+tempo2 <- fun_check(data = categ, class = "list", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": categ ARGUMENT MUST BE A SINGLE CHARACTER STRING OR A LIST OF CHARACTER STRINGS")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+if( ! is.null(categ.class.order)){
+if(is.null(categ)){
+tempo.cat <- paste0("ERROR IN ", function.name, ": categ.class.order ARGUMENT IS NOT NULL, BUT categ IS")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+tempo1 <- fun_check(data = categ.class.order, class = "vector", mode = "character", fun.name = function.name)
+tempo2 <- fun_check(data = categ.class.order, class = "list", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": categ.class.order ARGUMENT MUST BE A VECTOR OF CHARACTER STRINGS OR A LIST OF VECTOR OF CHARACTER STRINGS")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+if( ! is.null(legend.name)){
+tempo1 <- fun_check(data = legend.name, class = "vector", mode = "character", na.contain = TRUE, length = 1, fun.name = function.name)
+tempo2 <- fun_check(data = legend.name, class = "list", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": legend.name ARGUMENT MUST BE A SINGLE CHARACTER STRING OR A LIST OF CHARACTER STRINGS")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+if( ! is.null(color)){
+tempo1 <- fun_check(data = color, class = "vector", mode = "character", na.contain = TRUE, fun.name = function.name)
+tempo2 <- fun_check(data = color, class = "factor", na.contain = TRUE, fun.name = function.name)
+tempo3 <- fun_check(data = color, class = "integer", double.as.integer.allowed = TRUE, na.contain = TRUE, fun.name = function.name)
+tempo4 <- fun_check(data = color, class = "list", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE & tempo3$problem == TRUE & tempo4$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": color ARGUMENT MUST BE A VECTOR (OF CHARACTER STRINGS OR INTEGERS) OR A FACTOR OR A LIST OF THESE POSSIBILITIES")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+tempo1 <- fun_check(data = geom, class = "vector", mode = "character", na.contain = FALSE, length = 1, fun.name = function.name)
+tempo2 <- fun_check(data = geom, class = "list", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": geom ARGUMENT MUST BE A SINGLE CHARACTER STRING OR A LIST OF CHARACTER STRINGS")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+tempo1 <- fun_check(data = geom.step.dir, options = c("vh", "hv", "mid"), na.contain = FALSE, length = 1, fun.name = function.name)
+tempo2 <- fun_check(data = geom.step.dir, class = "list", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": geom.step.dir ARGUMENT MUST BE A SINGLE CHARACTER STRING (\"vh\" OR \"hv\" OR \"mid\") OR A LIST OF THESE CHARACTER STRINGS")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+tempo1 <- fun_check(data = alpha, prop = TRUE, length = 1, fun.name = function.name)
+tempo2 <- fun_check(data = alpha, class = "list", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": alpha ARGUMENT MUST BE A SINGLE NUMERIC VALUE BETWEEN 0 AND 1 OR A LIST OF SUCH VALUES")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+tempo1 <- fun_check(data = dot.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name)
+tempo2 <- fun_check(data = dot.size, class = "list", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": dot.size ARGUMENT MUST BE A SINGLE NUMERIC VALUE OR A LIST OF SINGLE NUMERIC VALUES")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+tempo1 <- fun_check(data = dot.shape, class = "vector", length = 1, fun.name = function.name)
+tempo2 <- fun_check(data = dot.shape, class = "list", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": dot.shape ARGUMENT MUST BE A SINGLE SHAPE VALUE OR A LIST OF SINGLE SHAPE VALUES (SEE https://ggplot2.tidyverse.org/articles/ggplot2-specs.html)")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+tempo1 <- fun_check(data = dot.border.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name)
+tempo2 <- fun_check(data = dot.border.size, class = "list", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": dot.border.size ARGUMENT MUST BE A SINGLE NUMERIC VALUE OR A LIST OF SINGLE NUMERIC VALUES")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+if( ! is.null(dot.border.color)){
+tempo1 <- fun_check(data = dot.border.color, class = "vector", mode = "character", length = 1, fun.name = function.name)
+tempo2 <- fun_check(data = dot.border.color, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+# integer colors -> gg_palette
+tempo.cat <- paste0("ERROR IN ", function.name, ": dot.border.color MUST BE A SINGLE CHARACTER STRING OF COLOR OR A SINGLE INTEGER VALUE")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+tempo1 <- fun_check(data = line.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name)
+tempo2 <- fun_check(data = line.size, class = "list", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": line.size ARGUMENT MUST BE A SINGLE NUMERIC VALUE OR A LIST OF SINGLE NUMERIC VALUES")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+tempo1 <- fun_check(data = line.type, class = "vector", typeof = "integer", double.as.integer.allowed = FALSE, length = 1, fun.name = function.name)
+tempo2 <- fun_check(data = line.type, class = "vector", mode = "character", length = 1, fun.name = function.name)
+tempo3 <- fun_check(data = line.type, class = "list", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE & tempo3$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": line.type ARGUMENT MUST BE A SINGLE LINE KIND VALUE OR A LIST OF SINGLE LINE KIND VALUES (SEE https://ggplot2.tidyverse.org/articles/ggplot2-specs.html)")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+if( ! is.null(x.lim)){
+tempo <- fun_check(data = x.lim, class = "vector", mode = "numeric", length = 2, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & any(x.lim %in% c(Inf, -Inf))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": x.lim ARGUMENT CANNOT CONTAIN -Inf OR Inf VALUES")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+if( ! is.null(x.lab)){
+if(all(class(x.lab) %in% "expression")){ # to deal with math symbols
+tempo <- fun_check(data = x.lab, class = "expression", length = 1, fun.name = function.name) ; eval(ee)
+}else{
+tempo <- fun_check(data = x.lab, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+}
+}
+tempo <- fun_check(data = x.log, options = c("no", "log2", "log10"), length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(x.tick.nb)){
+tempo <- fun_check(data = x.tick.nb, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & x.tick.nb < 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": x.tick.nb ARGUMENT MUST BE A NON-NULL POSITIVE INTEGER")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+if( ! is.null(x.second.tick.nb)){
+tempo <- fun_check(data = x.second.tick.nb, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & x.second.tick.nb <= 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": x.second.tick.nb ARGUMENT MUST BE A NON-NULL POSITIVE INTEGER")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+tempo <- fun_check(data = x.include.zero, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = x.left.extra.margin, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = x.right.extra.margin, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = x.text.angle, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, neg.values = TRUE, fun.name = function.name) ; eval(ee)
+if( ! is.null(y.lim)){
+tempo <- fun_check(data = y.lim, class = "vector", mode = "numeric", length = 2, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & any(y.lim %in% c(Inf, -Inf))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": y.lim ARGUMENT CANNOT CONTAIN -Inf OR Inf VALUES")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+if( ! is.null(y.lab)){
+if(all(class(y.lab) %in% "expression")){ # to deal with math symbols
+tempo <- fun_check(data = y.lab, class = "expression", length = 1, fun.name = function.name) ; eval(ee)
+}else{
+tempo <- fun_check(data = y.lab, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+}
+}
+tempo <- fun_check(data = y.log, options = c("no", "log2", "log10"), length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(y.tick.nb)){
+tempo <- fun_check(data = y.tick.nb, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & y.tick.nb < 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": y.tick.nb ARGUMENT MUST BE A NON-NULL POSITIVE INTEGER")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+if( ! is.null(y.second.tick.nb)){
+tempo <- fun_check(data = y.second.tick.nb, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE & y.second.tick.nb <= 0){
+tempo.cat <- paste0("ERROR IN ", function.name, ": y.second.tick.nb ARGUMENT MUST BE A NON-NULL POSITIVE INTEGER")
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+tempo <- fun_check(data = y.include.zero, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = y.top.extra.margin, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = y.bottom.extra.margin, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = y.text.angle, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, neg.values = TRUE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = raster, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = raster.ratio, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+if( ! is.null(raster.threshold)){
+tempo <- fun_check(data = raster.threshold, class = "vector", typeof = "integer", neg.values = FALSE, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = text.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = title, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = title.text.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = legend.show, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(legend.width)){
+tempo <- fun_check(data = legend.width, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = article, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = grid, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(add)){
+tempo <- fun_check(data = add, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
+}
+tempo <- fun_check(data = return, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = return.ggplot, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = return.gtable, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = plot, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+tempo <- fun_check(data = warn.print, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
+if( ! is.null(lib.path)){
+tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
+if(tempo$problem == FALSE){
+if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
+tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
+text.check <- c(text.check, tempo.cat)
+arg.check <- c(arg.check, TRUE)
+}
+}
+}
+if(any(arg.check) == TRUE){
+stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
+}
+# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
+# end primary argument checking
+
+
+# second round of checking and data preparation
+# dealing with NA arguments
+tempo.arg <- names(arg.user.setting) # values provided by the user
+tempo.log <- sapply(lapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = is.na), FUN = any) & lapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = length) == 1 # no argument provided by the user can be just NA
+if(any(tempo.log) == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ":\n", ifelse(sum(tempo.log, na.rm = TRUE) > 1, "THESE ARGUMENTS\n", "THIS ARGUMENT\n"), paste0(tempo.arg[tempo.log], collapse = "\n"),"\nCANNOT JUST BE NA")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end dealing with NA arguments
+# dealing with NULL arguments
+tempo.arg <-c(
+"data1",
+# "x", # inactivated because of hline or vline
+# "y", # inactivated because of hline or vline
+"geom",
+"geom.step.dir",
+"alpha",
+"dot.size",
+"dot.shape",
+"dot.border.size",
+"line.size",
+"line.type",
+"x.log",
+"x.include.zero",
+"x.left.extra.margin",
+"x.right.extra.margin",
+"x.text.angle",
+"y.log",
+"y.include.zero",
+"y.top.extra.margin",
+"y.bottom.extra.margin",
+"y.text.angle",
+"raster",
+"raster.ratio",
+"text.size",
+"title",
+"title.text.size",
+"legend.show",
+# "legend.width", # inactivated because can be null
+"article",
+"grid",
+"return",
+"return.ggplot",
+"return.gtable",
+"plot",
+"warn.print"
+)
+tempo.log <- sapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = is.null)
+if(any(tempo.log) == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ":\n", ifelse(sum(tempo.log, na.rm = TRUE) > 1, "THESE ARGUMENTS\n", "THIS ARGUMENT\n"), paste0(tempo.arg[tempo.log], collapse = "\n"),"\nCANNOT BE NULL")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
+}
+# end dealing with NULL arguments
+# check list lengths (and names of data1 compartments if present)
+ini.warning.length <- options()$warning.length
+warn <- NULL
+warn.count <- 0
+list.color <- NULL
+list.geom <- NULL
+list.geom.step.dir <- NULL
+list.alpha <- NULL
+list.dot.size <- NULL
+list.dot.shape <- NULL
+list.dot.border.size <- NULL
+list.dot.border.color <- NULL
+list.line.size <- NULL
+list.line.type <- NULL
+if(all(class(data1) == "list")){
+if(length(data1) > 6){
+tempo.cat <- paste0("ERROR IN ", function.name, ": data1 ARGUMENT MUST BE A LIST OF 6 DATA FRAMES MAXIMUM (6 OVERLAYS MAX)")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+if(is.null(names(data1))){
+names(data1) <- paste0("L", 1:length(data1))
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NULL NAME COMPARTMENT OF data1 LIST -> NAMES RESPECTIVELY ATTRIBUTED TO EACH COMPARTMENT:\n", paste(names(data1), collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+if( ! is.null(x)){
+if( ! (all(class(x) == "list") & length(data1) == length(x))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": x ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+}else{
+x <- vector("list", length(data1))
+}
+if( ! is.null(y)){
+if( ! (all(class(y) == "list") & length(data1) == length(y))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": y ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+}else{
+y <- vector("list", length(data1))
+}
+if( ! is.null(categ)){
+if( ! (all(class(categ) == "list") & length(data1) == length(categ))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": categ ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+}
+if( ! is.null(categ.class.order)){
+if( ! (all(class(categ.class.order) == "list") & length(data1) == length(categ.class.order))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": categ.class.order ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+}
+if( ! is.null(color)){
+if( ! ((all(class(color) == "list") & length(data1) == length(color)) | ((all(mode(color) == "character") | all(mode(color) == "numeric")) & length(color) == 1))){ # list of same length as data1 or single value
+tempo.cat <- paste0("ERROR IN ", function.name, ": color ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE CHARACTER STRING OR INTEGER")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if((all(mode(color) == "character") | all(mode(color) == "numeric")) & length(color) == 1){ # convert the single value into a list of single value
+list.color <- vector(mode = "list", length = length(data1))
+list.color[] <- color
+}
+}
+if( ! ((all(class(geom) == "list") & length(data1) == length(geom)) | (all(mode(geom) == "character") & length(geom) == 1))){ # list of same length as data1 or single value
+tempo.cat <- paste0("ERROR IN ", function.name, ": geom ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE CHARACTER VALUE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if(all(mode(geom) == "character") & length(geom) == 1){ # convert the single value into a list of single value
+list.geom <- vector(mode = "list", length = length(data1))
+list.geom[] <- geom
+}
+if( ! ((all(class(geom.step.dir) == "list") & length(data1) == length(geom.step.dir)) | (all(mode(geom.step.dir) == "character") & length(geom.step.dir) == 1))){ # list of same length as data1 or single value
+tempo.cat <- paste0("ERROR IN ", function.name, ": geom.step.dir ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE CHARACTER VALUE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if(all(mode(geom.step.dir) == "character") & length(geom.step.dir) == 1){ # convert the single value into a list of single value
+list.geom.step.dir <- vector(mode = "list", length = length(data1))
+list.geom.step.dir[] <- geom.step.dir
+}
+if( ! ((all(class(alpha) == "list") & length(data1) == length(alpha)) | (all(mode(alpha) == "numeric") & length(alpha) == 1))){ # list of same length as data1 or single value
+tempo.cat <- paste0("ERROR IN ", function.name, ": alpha ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE NUMERIC VALUE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if(all(mode(alpha) == "numeric") & length(alpha) == 1){ # convert the single value into a list of single value
+list.alpha <- vector(mode = "list", length = length(data1))
+list.alpha[] <- alpha
+}
+if( ! ((all(class(dot.size) == "list") & length(data1) == length(dot.size)) | (all(mode(dot.size) == "numeric") & length(dot.size) == 1))){ # list of same length as data1 or single value
+tempo.cat <- paste0("ERROR IN ", function.name, ": dot.size ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE NUMERIC VALUE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if(all(mode(dot.size) == "numeric") & length(dot.size) == 1){ # convert the single value into a list of single value
+list.dot.size <- vector(mode = "list", length = length(data1))
+list.dot.size[] <- dot.size
+}
+if( ! ((all(class(dot.shape) == "list") & length(data1) == length(dot.shape)) | (all(mode(dot.shape) != "list") & length(dot.shape) == 1))){ # list of same length as data1 or single value
+tempo.cat <- paste0("ERROR IN ", function.name, ": dot.shape ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE SHAPE VALUE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if(all(mode(dot.shape) != "list") & length(dot.shape) == 1){ # convert the single value into a list of single value
+list.dot.shape <- vector(mode = "list", length = length(data1))
+list.dot.shape[] <- dot.shape
+}
+if( ! ((all(class(dot.border.size) == "list") & length(data1) == length(dot.border.size)) | (all(mode(dot.border.size) == "numeric") & length(dot.border.size) == 1))){ # list of same length as data1 or single value
+tempo.cat <- paste0("ERROR IN ", function.name, ": dot.border.size ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE NUMERIC VALUE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if(all(mode(dot.border.size) == "numeric") & length(dot.border.size) == 1){ # convert the single value into a list of single value
+list.dot.border.size <- vector(mode = "list", length = length(data1))
+list.dot.border.size[] <- dot.border.size
+}
+if( ! is.null(dot.border.color)){
+if( ! ((all(class(dot.border.color) == "list") & length(data1) == length(dot.border.color)) | ((all(mode(dot.border.color) == "character") | all(mode(dot.border.color) == "numeric")) & length(dot.border.color) == 1))){ # list of same length as data1 or single value
+tempo.cat <- paste0("ERROR IN ", function.name, ": dot.border.color ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE CHARACTER STRING OR INTEGER")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if((all(mode(dot.border.color) == "character") | all(mode(dot.border.color) == "numeric")) & length(dot.border.color) == 1){ # convert the single value into a list of single value
+list.dot.border.color <- vector(mode = "list", length = length(data1))
+list.dot.border.color[] <- dot.border.color
+}
+}
+if( ! ((all(class(line.size) == "list") & length(data1) == length(line.size)) | (all(mode(line.size) == "numeric") & length(line.size) == 1))){ # list of same length as data1 or single value
+tempo.cat <- paste0("ERROR IN ", function.name, ": line.size ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE NUMERIC VALUE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if(all(mode(line.size) == "numeric") & length(line.size) == 1){ # convert the single value into a list of single value
+list.line.size <- vector(mode = "list", length = length(data1))
+list.line.size[] <- line.size
+}
+if( ! ((all(class(line.type) == "list") & length(data1) == length(line.type)) | (all(mode(line.type) != "list") & length(line.type) == 1))){ # list of same length as data1 or single value
+tempo.cat <- paste0("ERROR IN ", function.name, ": line.type ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE LINE KIND VALUE")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if(all(mode(line.type) != "list") & length(line.type) == 1){ # convert the single value into a list of single value
+list.line.type <- vector(mode = "list", length = length(data1))
+list.line.type[] <- line.type
+}
+if( ! is.null(legend.name)){
+if( ! (all(class(legend.name) == "list") & length(data1) == length(legend.name))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": legend.name ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+}
+}
+# end check list lengths (and names of data1 compartments if present)
+# conversion into lists
+if(all(is.data.frame(data1))){
+data1 <- list(L1 = data1)
+if(all(class(x) == "list")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": x ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+x <- list(L1 = x)
+}
+if(all(class(y) == "list")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": y ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+y <- list(L1 = y)
+}
+if( ! is.null(categ)){
+if(all(class(categ) == "list")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": categ ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+categ <- list(L1 = categ)
+}
+}
+if( ! is.null(categ.class.order)){
+if(all(class(categ.class.order) == "list")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": categ.class.order ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+categ.class.order <- list(L1 = categ.class.order)
+}
+}
+if( ! is.null(color)){
+if(all(class(color) == "list")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": color ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+color <- list(L1 = color)
+}
+}
+if(all(class(geom) == "list")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": geom ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+geom <- list(L1 = geom)
+}
+if(all(class(geom.step.dir) == "list")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": geom.step.dir ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+geom.step.dir <- list(L1 = geom.step.dir)
+}
+if(all(class(alpha) == "list")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": alpha ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+alpha <- list(L1 = alpha)
+}
+if(all(class(dot.size) == "list")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": dot.size ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+dot.size <- list(L1 = dot.size)
+}
+if(all(class(dot.shape) == "list")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": dot.shape ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+dot.shape <- list(L1 = dot.shape)
+}
+if(all(class(dot.border.size) == "list")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": dot.border.size ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+dot.border.size <- list(L1 = dot.border.size)
+}
+if( ! is.null(dot.border.color)){
+if(all(class(dot.border.color) == "list")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": dot.border.color ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+dot.border.color <- list(L1 = dot.border.color)
+}
+}
+if(all(class(line.size) == "list")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": line.size ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+line.size <- list(L1 = line.size)
+}
+if(all(class(line.type) == "list")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": line.type ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+line.type <- list(L1 = line.type)
+}
+if( ! is.null(legend.name)){
+if(all(class(legend.name) == "list")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": legend.name ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+legend.name <- list(L1 = legend.name)
+}
+}
+}else if( ! all(sapply(data1, FUN = "class") == "data.frame")){ # if not a data frame, data1 can only be a list, as tested above
+tempo.cat <- paste0("ERROR IN ", function.name, ": data1 ARGUMENT MUST BE A DATA FRAME OR A LIST OF DATA FRAMES")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+# single value converted into list now reattributed to the argument name
+if( ! is.null(color)){
+if( ! is.null(list.color)){
+color <- list.color
+}
+}
+if( ! is.null(list.geom)){
+geom <- list.geom
+}
+if( ! is.null(list.geom.step.dir)){
+geom.step.dir <- list.geom.step.dir
+}
+if( ! is.null(list.alpha)){
+alpha <- list.alpha
+}
+if( ! is.null(list.dot.size)){
+dot.size <- list.dot.size
+}
+if( ! is.null(list.dot.shape)){
+dot.shape <- list.dot.shape
+}
+if( ! is.null(list.dot.border.size)){
+dot.border.size <- list.dot.border.size
+}
+if( ! is.null(dot.border.color)){
+if( ! is.null(list.dot.border.color)){
+dot.border.color <- list.dot.border.color
+}
+}
+if( ! is.null(list.line.size)){
+line.size <- list.line.size
+}
+if( ! is.null(list.line.type)){
+line.type <- list.line.type
+}
+# end single value converted into list now reattributed to the argument name
+# data, x, y, geom, alpha, dot.size, shape, dot.border.size, line.size, line.type, legend.name are list now
+# if non-null, categ, categ.class.order, legend.name, color, dot.border.color are list now
+# end conversion into lists
+# verif of add
+if( ! is.null(add)){
+if( ! grepl(pattern = "^\\s*\\+", add)){ # check that the add string start by +
+tempo.cat <- paste0("ERROR IN ", function.name, ": add ARGUMENT MUST START WITH \"+\": ", paste(unique(add), collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+
+}else if( ! grepl(pattern = "(ggplot2|lemon)\\s*::", add)){ #
+tempo.cat <- paste0("ERROR IN ", function.name, ": FOR EASIER FUNCTION DETECTION, add ARGUMENT MUST CONTAIN \"ggplot2::\" OR \"lemon::\" IN FRONT OF EACH GGPLOT2 FUNCTION: ", paste(unique(add), collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if( ! grepl(pattern = ")\\s*$", add)){ # check that the add string finished by )
+tempo.cat <- paste0("ERROR IN ", function.name, ": add ARGUMENT MUST FINISH BY \")\": ", paste(unique(add), collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+}
+# end verif of add
+# management of add containing facet
+facet.categ <- NULL
+if( ! is.null(add)){
+facet.check <- TRUE
+tempo <- unlist(strsplit(x = add, split = "\\s*\\+\\s*(ggplot2|lemon)\\s*::\\s*")) #
+tempo <- sub(x = tempo, pattern = "^facet_wrap", replacement = "ggplot2::facet_wrap")
+tempo <- sub(x = tempo, pattern = "^facet_grid", replacement = "ggplot2::facet_grid")
+tempo <- sub(x = tempo, pattern = "^facet_rep", replacement = "lemon::facet_rep")
+if(length(data1) > 1 & (any(grepl(x = tempo, pattern = "ggplot2::facet_wrap|lemon::facet_rep_wrap")) | grepl(x = add, pattern = "ggplot2::facet_grid|lemon::facet_rep_grid"))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nfacet PANELS CANNOT BE USED IF MORE THAN ONE DATA FRAME IN THE data1 ARGUMENT\nPLEASE REWRITE THE add STRING AND RERUN")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+if(any(grepl(x = tempo, pattern = "ggplot2::facet_wrap|lemon::facet_rep_wrap"))){
+tempo1 <- suppressWarnings(eval(parse(text = tempo[grepl(x = tempo, pattern = "ggplot2::facet_wrap|lemon::facet_rep_wrap")])))
+facet.categ <- list(names(tempo1$params$facets)) # list of length 1
+tempo.text <- "facet_wrap OR facet_rep_wrap"
+facet.check <- FALSE
+}else if(grepl(x = add, pattern = "ggplot2::facet_grid|lemon::facet_rep_grid")){
+tempo1 <- suppressWarnings(eval(parse(text = tempo[grepl(x = tempo, pattern = "ggplot2::facet_grid|lemon::facet_rep_grid")])))
+facet.categ <- list(c(names(tempo1$params$rows), names(tempo1$params$cols))) # list of length 1
+tempo.text <- "facet_grid OR facet_rep_grid"
+facet.check <- FALSE
+}
+if(facet.check == FALSE & ! all(facet.categ %in% names(data1[[1]]))){ # WARNING: all(facet.categ %in% names(data1)) is TRUE when facet.categ is NULL
+tempo.cat <- paste0("ERROR IN ", function.name, "\nDETECTION OF \"", tempo.text, "\" STRING IN THE add ARGUMENT BUT PROBLEM OF VARIABLE DETECTION (COLUMN NAMES OF data1)\nTHE DETECTED VARIABLES ARE:\n", paste(facet.categ, collapse = " "), "\nTHE data1 COLUMN NAMES ARE:\n", paste(names(data1[[1]]), collapse = " "), "\nPLEASE REWRITE THE add STRING AND RERUN")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+}
+}
+# if facet.categ is not NULL, it is a list of length 1 now
+# end management of add containing facet
+# legend name filling
+if(is.null(legend.name) & ! is.null(categ)){
+legend.name <- categ
+}else if(is.null(legend.name) & is.null(categ)){
+legend.name <- vector("list", length(data1)) # null list
+}
+# legend.name not NULL anymore (list)
+# end legend name filling
+# ini categ for legend display
+fin.lg.disp <- vector("list", 6) # will be used at the end to display or not legends
+fin.lg.disp[] <- FALSE
+legend.disp <- vector("list", length(data1))
+if(is.null(categ) | legend.show == FALSE){
+legend.disp[] <- FALSE
+}else{
+for(i2 in 1:length(data1)){
+if(is.null(categ[[i2]])){
+legend.disp[[i2]] <- FALSE
+}else{
+legend.disp[[i2]] <- TRUE
+}
+}
+}
+# end ini categ for legend display
+# integer colors into gg_palette
+tempo.check.color <- NULL
+for(i1 in 1:length(data1)){
+if(any(is.na(color[[i1]]))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), ": color ARGUMENT CANNOT CONTAIN NA")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+tempo.check.color <- c(tempo.check.color, fun_check(data = color[[i1]], data.name = ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), class = "integer", double.as.integer.allowed = TRUE, na.contain = TRUE, fun.name = function.name)$problem)
+}
+tempo.check.color <- ! tempo.check.color # invert TRUE and FALSE because if integer, then problem = FALSE
+if(any(tempo.check.color == TRUE)){ # convert integers into colors
+tempo.integer <- unlist(color[tempo.check.color])
+tempo.color <- fun_gg_palette(max(tempo.integer, na.rm = TRUE))
+for(i1 in 1:length(data1)){
+if(tempo.check.color[i1] == TRUE){
+color[[i1]] <-tempo.color[color[[i1]]]
+}
+}
+}
+# end integer colors into gg_palette
+# loop (checking inside list compartment)
+compart.null.color <- 0 # will be used to attribute a color when color is non-null but a compartment of color is NULL
+data1.ini <- data1 # to report NA removal
+removed.row.nb <- vector("list", length = length(data1)) # to report NA removal. Contains NULL
+removed.rows <- vector("list", length = length(data1)) # to report NA removal. Contains NULL
+for(i1 in 1:length(data1)){
+tempo <- fun_check(data = data1[[i1]], data.name = ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), class = "data.frame", na.contain = TRUE, fun.name = function.name)
+if(tempo$problem == TRUE){
+stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+# reserved word checking
+if(any(names(data1[[i1]]) %in% reserved.words)){ # I do not use fun_name_change() because cannot control y before creating "fake_y". But ok because reserved are not that common
+tempo.cat <- paste0("ERROR IN ", function.name, ": COLUMN NAMES OF ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " ARGUMENT CANNOT BE ONE OF THESE WORDS\n", paste(reserved.words, collapse = " "), "\nTHESE ARE RESERVED FOR THE ", function.name, " FUNCTION")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+if( ! (is.null(add))){
+if(any(sapply(X = reserved.words, FUN = grepl, x = add))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nDETECTION OF COLUMN NAMES OF data1 IN THE add ARGUMENT STRING, THAT CORRESPOND TO RESERVED STRINGS FOR ", function.name, "\nFOLLOWING COLUMN NAMES HAVE TO BE CHANGED:\n", paste(arg.names[sapply(X = reserved.words, FUN = grepl, x = add)], collapse = "\n"), "\nFOR INFORMATION, THE RESERVED WORDS ARE:\n", paste(reserved.words, collapse = "\n"))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if(any(sapply(X = arg.names, FUN = grepl, x = add))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NAMES OF ", function.name, " ARGUMENTS DETECTED IN THE add STRING:\n", paste(arg.names[sapply(X = arg.names, FUN = grepl, x = add)], collapse = "\n"), "\nRISK OF WRONG OBJECT USAGE INSIDE ", function.name)
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+# end reserved word checking
+# check of geom now because required for y argument
+tempo <- fun_check(data = geom[[i1]], data.name = ifelse(length(geom) == 1, "geom", paste0("geom NUMBER ", i1)), options = c("geom_point", "geom_line", "geom_path", "geom_step", "geom_hline", "geom_vline", "geom_stick"), length = 1, fun.name = function.name)
+if(tempo$problem == TRUE){
+stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+if(geom[[i1]] == "geom_step" & is.null(geom.step.dir[[i1]])){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(geom.step.dir) == 1, "geom.step.dir", paste0("ELEMENT ", i1, " OF geom.step.dir ARGUMENT")), ": geom.step.dir ARGUMENT CANNOT BE NULL IF ", ifelse(length(geom) == 1, "geom", paste0("ELEMENT ", i1, " OF geom")), " ARGUMENT IS \"geom_step\"\nHERE geom.step.dir ARGUMENT IS: ", paste(geom.step.dir[[i1]], collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if(geom[[i1]] == "geom_step" & ! is.null(geom.step.dir[[i1]])){
+tempo <- fun_check(data = geom.step.dir[[i1]], data.name = ifelse(length(geom.step.dir) == 1, "geom.step.dir", paste0("geom.step.dir NUMBER ", i1)), options = c("vh", "hv", "mid"), length = 1, fun.name = function.name)
+if(tempo$problem == TRUE){
+stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+}
+# end check of geom now because required for y argument
+if(is.null(x[[i1]])){
+if(all(geom[[i1]] != "geom_hline")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(x) == 1, "x", paste0("ELEMENT ", i1, " OF x ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ": x ARGUMENT CANNOT BE NULL EXCEPT IF ", ifelse(length(geom) == 1, "x", paste0("geom NUMBER ", i1)), " ARGUMENT IS \"geom_hline\"\nHERE geom ARGUMENT IS: ", paste(geom[[i1]], collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+x[[i1]] <- "fake_x"
+data1[[i1]] <- cbind(data1[[i1]], fake_x = NA, stringsAsFactors = TRUE)
+data1[[i1]][, "fake_x"] <- as.numeric(data1[[i1]][, "fake_x"])
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NULL ", ifelse(length(x) == 1, "x", paste0("ELEMENT ", i1, " OF x")), " ARGUMENT ASSOCIATED TO ", ifelse(length(geom) == 1, "geom", paste0("geom NUMBER ", i1)), " ARGUMENT ", geom[[i1]], " -> FAKE COLUMN ADDED TO DATA FRAME ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ", NAMED \"fake_x\" FOR FINAL DRAWING")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else{
+if(all(geom[[i1]] == "geom_hline")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(x) == 1, "x", paste0("ELEMENT ", i1, " OF x ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ": x ARGUMENT MUST BE NULL IF ", ifelse(length(geom) == 1, "geom", paste0("geom NUMBER ", i1)), " ARGUMENT IS \"geom_hline\"")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+tempo <- fun_check(data = x[[i1]], data.name = ifelse(length(x) == 1, "x", paste0("ELEMENT ", i1, " OF x ARGUMENT")), class = "vector", mode = "character", length = 1, fun.name = function.name)
+if(tempo$problem == TRUE){
+stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+}
+if(is.null(y[[i1]])){
+if(all(geom[[i1]] != "geom_vline")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(y) == 1, "y", paste0("ELEMENT ", i1, " OF y ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ": y ARGUMENT CANNOT BE NULL EXCEPT IF ", ifelse(length(geom) == 1, "y", paste0("geom NUMBER ", i1)), " ARGUMENT IS \"geom_vline\"\nHERE geom ARGUMENT IS: ", paste(geom[[i1]], collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+y[[i1]] <- "fake_y"
+data1[[i1]] <- cbind(data1[[i1]], fake_y = NA, stringsAsFactors = TRUE)
+data1[[i1]][, "fake_y"] <- as.numeric(data1[[i1]][, "fake_y"])
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NULL ", ifelse(length(y) == 1, "y", paste0("ELEMENT ", i1, " OF y")), " ARGUMENT ASSOCIATED TO ", ifelse(length(geom) == 1, "geom", paste0("geom NUMBER ", i1)), " ARGUMENT ", geom[[i1]], " -> FAKE COLUMN ADDED TO DATA FRAME ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ", NAMED \"fake_y\" FOR FINAL DRAWING")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else{
+if(all(geom[[i1]] == "geom_vline")){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(y) == 1, "y", paste0("ELEMENT ", i1, " OF y ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ": y ARGUMENT MUST BE NULL IF ", ifelse(length(geom) == 1, "geom", paste0("geom NUMBER ", i1)), " ARGUMENT IS \"geom_vline\"")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+tempo <- fun_check(data = y[[i1]], data.name = ifelse(length(y) == 1, "y", paste0("ELEMENT ", i1, " OF y ARGUMENT")), class = "vector", mode = "character", length = 1, fun.name = function.name)
+if(tempo$problem == TRUE){
+stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+}
+# x[[i1]] and y[[i1]] not NULL anymore
+if( ! (x[[i1]] %in% names(data1[[i1]]))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(x) == 1, "x", paste0("ELEMENT ", i1, " OF x")), " ARGUMENT MUST BE A COLUMN NAME OF ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT\nHERE IT IS: ", paste(x[[i1]], collapse = " "))))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+if( ! (y[[i1]] %in% names(data1[[i1]]))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(y) == 1, "y", paste0("ELEMENT ", i1, " OF y")), " ARGUMENT MUST BE A COLUMN NAME OF ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT\nHERE IT IS: ", paste(y[[i1]], collapse = " "))))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+tempo <- fun_check(data = data1[[i1]][, x[[i1]]], data.name = ifelse(length(x) == 1, "x ARGUMENT (AS COLUMN NAME OF data1 DATA FRAME)", paste0("ELEMENT ", i1, " OF x ARGUMENT", " (AS COLUMN NAME OF data1 DATA FRAME NUMBER ", i1, ")")), class = "vector", mode = "numeric", na.contain = TRUE, fun.name = function.name)
+if(tempo$problem == TRUE){
+stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+tempo <- fun_check(data = data1[[i1]][, y[[i1]]], data.name = ifelse(length(y) == 1, "y ARGUMENT (AS COLUMN NAME OF data1 DATA FRAME)", paste0("ELEMENT ", i1, " OF y ARGUMENT", " (AS COLUMN NAME OF data1 DATA FRAME NUMBER ", i1, ")")), class = "vector", mode = "numeric", na.contain = TRUE, fun.name = function.name)
+if(tempo$problem == TRUE){
+stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+if(x[[i1]] == "fake_x" & y[[i1]] == "fake_y"){ # because the code cannot accept to be both "fake_x" and "fake_y" at the same time
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 2\nTHE CODE CANNOT ACCEPT x AND y TO BE \"fake_x\" AND \"fake_y\" IN THE SAME DATA FRAME ", i1, " ")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+
+if(( ! is.null(categ)) & ( ! is.null(categ[[i1]]))){ # is.null(categ[[i1]]) works even if categ is NULL # is.null(categ[[i1]]) works even if categ is NULL # if categ[[i1]] = NULL, fake_categ will be created later on
+tempo <- fun_check(data = categ[[i1]], data.name = ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")),, class = "vector", mode = "character", length = 1, fun.name = function.name)
+if(tempo$problem == TRUE){
+stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+if( ! (categ[[i1]] %in% names(data1[[i1]]))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ")), " ARGUMENT MUST BE A COLUMN NAME OF ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT\nHERE IT IS: ", paste(categ[[i1]], collapse = " "))))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+tempo1 <- fun_check(data = data1[[i1]][, categ[[i1]]], data.name = ifelse(length(categ) == 1, "categ OF data1 ARGUMENT", paste0("ELEMENT ", i1, " OF categ ARGUMENT IN DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), class = "vector", mode = "character", na.contain = TRUE, fun.name = function.name)
+tempo2 <- fun_check(data = data1[[i1]][, categ[[i1]]], data.name = ifelse(length(categ) == 1, "categ OF data1 ARGUMENT", paste0("ELEMENT ", i1, " OF categ ARGUMENT IN DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), class = "factor", na.contain = TRUE, fun.name = function.name)
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(categ) == 1, "categ OF data1 ARGUMENT", paste0("ELEMENT ", i1, " OF categ ARGUMENT IN DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " MUST BE A FACTOR OR CHARACTER VECTOR")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if(tempo1$problem == FALSE){
+data1[[i1]][, categ[[i1]]] <- factor(data1[[i1]][, categ[[i1]]]) # if already a factor, change nothing, if characters, levels according to alphabetical order
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") IN ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ", THE CHARACTER COLUMN HAS BEEN CONVERTED TO FACTOR")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+
+}
+if(geom[[i1]] == "geom_vline" | geom[[i1]] == "geom_hline"){
+if(length(unique(data1[[i1]][, categ[[i1]]])) != nrow(data1[[i1]])){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(geom) == 1, "geom OF data1 ARGUMENT", paste0("geom NUMBER ", i1, " OF DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " ARGUMENT IS ", geom[[i1]], ", MEANING THAT ", ifelse(length(categ) == 1, "categ OF data1 ARGUMENT", paste0("ELEMENT ", i1, " OF categ ARGUMENT IN DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " MUST HAVE A DIFFERENT CLASS PER LINE OF data1 (ONE x VALUE PER CLASS)")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+}
+}else if(( ! is.null(categ)) & is.null(categ[[i1]])){ # is.null(categ[[i1]]) works even if categ is NULL # if categ[[i1]] = NULL, fake_categ will be created. WARNING: is.null(categ[[i1]]) means no legend display (see above), because categ has not been precised. This also means a single color for data1[[i1]]
+if(length(color[[i1]]) > 1){ # 0 means is.null(color[[i1]]) or is.null(color) and 1 is ok -> single color for data1[[i1]]
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NULL ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ")), " ARGUMENT BUT CORRESPONDING COLORS IN ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " HAS LENGTH OVER 1\n", paste(color[[i1]], collapse = " "), "\nWHICH IS NOT COMPATIBLE WITH NULL CATEG -> COLOR RESET TO A SINGLE COLOR")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+color[i1] <- list(NULL) # will provide a single color below # Warning color[[i1]] <- NULL removes the compartment
+}
+categ[[i1]] <- "fake_categ"
+data1[[i1]] <- cbind(data1[[i1]], fake_categ = "", stringsAsFactors = TRUE)
+# inactivated because give a different color to different "Line_" categ while a single color for all the data1[[i1]] required. Thus, put back after the color management
+# if(geom[[i1]] == "geom_hline" | geom[[i1]] == "geom_vline"){
+# data1[[i1]][, "fake_categ"] <- paste0("Line_", 1:nrow(data1[[i1]]))
+# }else{
+data1[[i1]][, "fake_categ"] <- data1[[i1]][, "fake_categ"] # as.numeric("") create a vector of NA but class numeric
+# }
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NULL ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ")), " ARGUMENT -> FOR DATA FRAME ", ifelse(length(data1) == 1, "data1 ARGUMENT:", paste0("NUMBER ", i1, " OF data1 ARGUMENT:")), "\n- FAKE \"fake_categ\" COLUMN ADDED FILLED WITH \"\"(OR WITH \"Line_...\" FOR LINES)\n- SINGLE COLOR USED FOR PLOTTING\n- NO LEGEND DISPLAYED")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+# OK: if categ is not NULL, all the non-null categ columns of data1 are factors from here
+
+# management of log scale and Inf removal
+if(x[[i1]] != "fake_x"){
+if(any(( ! is.finite(data1[[i1]][, x[[i1]]])) & ( ! is.na(data1[[i1]][, x[[i1]]])))){ # is.finite also detects NA: ( ! is.finite(data1[, y])) & ( ! is.na(data1[, y])) detects only Inf
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") PRESENCE OF -Inf OR Inf VALUES IN ", ifelse(length(categ) == 1, "x", paste0("ELEMENT ", i1, " OF x ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " AND CORRESPONDING ROWS REMOVED (SEE $removed.row.nb AND $removed.rows)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+if(y[[i1]] != "fake_y"){
+if(any(( ! is.finite(data1[[i1]][, y[[i1]]])) & ( ! is.na(data1[[i1]][, y[[i1]]])))){ # is.finite also detects NA: ( ! is.finite(data1[, y])) & ( ! is.na(data1[, y])) detects only Inf
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") PRESENCE OF -Inf OR Inf VALUES IN ", ifelse(length(categ) == 1, "y", paste0("ELEMENT ", i1, " OF y ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " AND CORRESPONDING ROWS REMOVED (SEE $removed.row.nb AND $removed.rows)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+# log conversion
+if(x.log != "no"){
+tempo1 <- ! is.finite(data1[[i1]][, x[[i1]]]) # where are initial NA and Inf
+data1[[i1]][, x[[i1]]] <- suppressWarnings(get(x.log)(data1[[i1]][, x[[i1]]]))# no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
+if(any( ! (tempo1 | is.finite(data1[[i1]][, x[[i1]]])))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") LOG CONVERSION INTRODUCED -Inf OR Inf OR NaN VALUES IN ", ifelse(length(categ) == 1, "x", paste0("ELEMENT ", i1, " OF x ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " AND CORRESPONDING ROWS REMOVED (SEE $removed.row.nb AND $removed.rows)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+if(y.log != "no"){
+tempo1 <- ! is.finite(data1[[i1]][, y[[i1]]]) # where are initial NA and Inf
+data1[[i1]][, y[[i1]]] <- suppressWarnings(get(y.log)(data1[[i1]][, y[[i1]]]))# no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
+if(any( ! (tempo1 | is.finite(data1[[i1]][, y[[i1]]])))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") LOG CONVERSION INTRODUCED -Inf OR Inf OR NaN VALUES IN ", ifelse(length(categ) == 1, "y", paste0("ELEMENT ", i1, " OF y ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " AND CORRESPONDING ROWS REMOVED (SEE $removed.row.nb AND $removed.rows)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+# Inf removal
+# removed.row.nb[[i1]] <- NULL # already NULL and Warning this removes the compartment
+removed.rows[[i1]] <- data.frame(stringsAsFactors = FALSE)
+if(any(( ! is.finite(data1[[i1]][, x[[i1]]])) & ( ! is.na(data1[[i1]][, x[[i1]]])))){ # is.finite also detects NA: ( ! is.finite(data1[[i1]][, x[[i1]]])) & ( ! is.na(data1[[i1]][, x[[i1]]])) detects only Inf
+removed.row.nb[[i1]] <- c(removed.row.nb[[i1]], which(( ! is.finite(data1[[i1]][, x[[i1]]])) & ( ! is.na(data1[[i1]][, x[[i1]]]))))
+}
+if(any(( ! is.finite(data1[[i1]][, y[[i1]]])) & ( ! is.na(data1[[i1]][, y[[i1]]])))){ # is.finite also detects NA: ( ! is.finite(data1[[i1]][, y[[i1]]])) & ( ! is.na(data1[[i1]][, y[[i1]]])) detects only Inf
+removed.row.nb[[i1]] <- c(removed.row.nb[[i1]], which(( ! is.finite(data1[[i1]][, y[[i1]]])) & ( ! is.na(data1[[i1]][, y[[i1]]]))))
+}
+if( ! is.null(removed.row.nb[[i1]])){
+removed.row.nb[[i1]] <- unique(removed.row.nb[[i1]]) # to remove the duplicated positions (NA in both x and y)
+removed.rows[[i1]] <- rbind(removed.rows[[i1]], data1.ini[[i1]][removed.row.nb[[i1]], ]) # here data1.ini used to have the y = O rows that will be removed because of Inf creation after log transformation
+data1[[i1]] <- data1[[i1]][-removed.row.nb[[i1]], ]
+data1.ini[[i1]] <- data1.ini[[i1]][-removed.row.nb[[i1]], ] #
+}
+# From here, data1 and data.ini have no more Inf
+# end Inf removal
+# x.lim and y.lim dealt later on, after the end f the loop
+# end management of log scale and Inf removal
+# management of log scale
+if(x.log != "no"){
+data1[[i1]][, x[[i1]]] <- suppressWarnings(get(x.log)(data1[[i1]][, x[[i1]]])) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
+}
+if(y.log != "no"){
+data1[[i1]][, y[[i1]]] <- suppressWarnings(get(y.log)(data1[[i1]][, y[[i1]]])) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
+}
+# end management of log scale
+# na detection and removal
+column.check <- unique(unlist(c( # unlist because creates a list
+if(x[[i1]] == "fake_x"){NULL}else{x[[i1]]},
+if(y[[i1]] == "fake_y"){NULL}else{y[[i1]]},
+if( ! is.null(categ)){if(is.null(categ[[i1]])){NULL}else{categ[[i1]]}},
+if( ! is.null(facet.categ)){if(is.null(facet.categ[[i1]])){NULL}else{facet.categ[[i1]]}}
+))) # dot.categ because can be a 3rd column of data1
+if(any(is.na(data1[[i1]][, column.check]))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NA DETECTED IN COLUMNS ", paste(column.check, collapse = " "), " OF ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " AND CORRESPONDING ROWS REMOVED (SEE $removed.row.nb AND $removed.rows)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+for(i3 in 1:length(column.check)){
+if(any(is.na(data1[[i1]][, column.check[i3]]))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("NA REMOVAL DUE TO COLUMN ", column.check[i3], " OF ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+tempo <- unique(unlist(lapply(lapply(c(data1[[i1]][column.check]), FUN = is.na), FUN = which)))
+removed.row.nb[[i1]] <- c(removed.row.nb[[i1]], tempo)
+removed.rows[[i1]] <- rbind(removed.rows[[i1]], data1.ini[[i1]][tempo, ]) # # tempo used because removed.row.nb is not empty. Here data1.ini used to have the non NA rows that will be removed because of NAN creation after log transformation (neg values for instance)
+column.check <- column.check[ ! (column.check == x[[i1]] | column.check == y[[i1]])] # remove x and y to keep quali columns
+if(length(tempo) != 0){
+data1[[i1]] <- data1[[i1]][-tempo, ] # WARNING tempo here and not removed.row.nb because the latter contain more numbers thant the former
+data1.ini[[i1]] <- data1.ini[[i1]][-tempo, ] # WARNING tempo here and not removed.row.nb because the latter contain more numbers than the former
+for(i4 in 1:length(column.check)){
+if(any( ! unique(removed.rows[[i1]][, column.check[i4]]) %in% unique(data1[[i1]][, column.check[i4]]))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") IN COLUMN ", column.check[i4], " OF ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ", THE FOLLOWING CLASSES HAVE DISAPPEARED AFTER NA REMOVAL\n(IF COLUMN USED IN THE PLOT, THIS CLASS WILL NOT BE DISPLAYED):\n", paste(unique(removed.rows[[i1]][, column.check[i4]])[ ! unique(removed.rows[[i1]][, column.check[i4]]) %in% unique(data1[[i1]][, column.check[i4]])], collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+tempo.levels <- levels(data1[[i1]][, column.check[i4]])[levels(data1[[i1]][, column.check[i4]]) %in% unique(as.character(data1[[i1]][, column.check[i4]]))]
+data1[[i1]][, column.check[i4]] <- factor(as.character(data1[[i1]][, column.check[i4]]), levels = tempo.levels)
+if(column.check[i4] %in% categ[[i1]] & ! is.null(categ.class.order)){
+categ.class.order[[i1]] <- levels(data1[[i1]][, column.check[i4]])[levels(data1[[i1]][, column.check[i4]]) %in% unique(data1[[i1]][, column.check[i4]])] # remove the absent class in the categ.class.order vector
+data1[[i1]][, column.check[i4]] <- factor(as.character(data1[[i1]][, column.check[i4]]), levels = unique(categ.class.order[[i1]]))
+}
+}
+}
+}
+}
+# end na detection and removal
+# From here, data1 and data.ini have no more NA or NaN in x, y, categ (if categ != NULL) and facet.categ (if categ != NULL)
+if( ! is.null(categ.class.order)){
+# the following check will be done several times but I prefer to keep it here, after the creation of categ
+if(is.null(categ[[i1]]) & ! is.null(categ.class.order[[i1]])){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nCOMPARTMENT ", i1, " OF categ ARGUMENT CANNOT BE NULL IF COMPARTMENT ", i1, " OF categ.class.order ARGUMENT IS NOT NULL: ", paste(categ.class.order, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+if(is.null(categ.class.order[[i1]])){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE categ.class.order COMPARTMENT ", i1, " IS NULL. ALPHABETICAL ORDER WILL BE APPLIED")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+data1[[i1]][, categ[[i1]]] <- factor(as.character(data1[[i1]][, categ[[i1]]])) # if already a factor, change nothing, if characters, levels according to alphabetical order
+categ.class.order[[i1]] <- levels(data1[[i1]][, categ[[i1]]]) # character vector that will be used later
+}else{
+tempo <- fun_check(data = categ.class.order[[i1]], data.name = paste0("COMPARTMENT ", i1 , " OF categ.class.order ARGUMENT"), class = "vector", mode = "character", length = length(levels(data1[[i1]][, categ[[i1]]])), fun.name = function.name) # length(data1[, categ[i1]) -> if data1[, categ[i1] was initially character vector, then conversion as factor after the NA removal, thus class number ok. If data1[, categ[i1] was initially factor, no modification after the NA removal, thus class number ok
+if(tempo$problem == TRUE){
+stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+}
+if(any(duplicated(categ.class.order[[i1]]))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nCOMPARTMENT ", i1, " OF categ.class.order ARGUMENT CANNOT HAVE DUPLICATED CLASSES: ", paste(categ.class.order[[i1]], collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if( ! (all(categ.class.order[[i1]] %in% unique(data1[[i1]][, categ[[i1]]])) & all(unique(data1[[i1]][, categ[[i1]]]) %in% categ.class.order[[i1]]))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nCOMPARTMENT ", i1, " OF categ.class.order ARGUMENT MUST BE CLASSES OF COMPARTMENT ", i1, " OF categ ARGUMENT\nHERE IT IS:\n", paste(categ.class.order[[i1]], collapse = " "), "\nFOR COMPARTMENT ", i1, " OF categ.class.order AND IT IS:\n", paste(unique(data1[[i1]][, categ[[i1]]]), collapse = " "), "\nFOR COLUMN ", categ[[i1]], " OF data1 NUMBER ", i1)
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+data1[[i1]][, categ[[i1]]] <- factor(data1[[i1]][, categ[[i1]]], levels = categ.class.order[[i1]]) # reorder the factor
+}
+names(categ.class.order)[i1] <- categ[[i1]]
+}
+}
+# OK: if categ.class.order is not NULL, all the NULL categ.class.order columns of data1 are character from here
+
+if( ! is.null(legend.name[[i1]])){
+tempo <- fun_check(data = legend.name[[i1]], data.name = ifelse(length(legend.name) == 1, "legend.name", paste0("legend.name NUMBER ", i1)),, class = "vector", mode = "character", length = 1, fun.name = function.name)
+if(tempo$problem == TRUE){
+stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+}
+if( ! is.null(color)){ # if color is NULL, will be filled later on
+# check the nature of color
+if(is.null(color[[i1]])){
+compart.null.color <- compart.null.color + 1
+color[[i1]] <- grey(compart.null.color / 8) # cannot be more than 7 overlays. Thus 7 different greys. 8/8 is excluded because white dots
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NULL COLOR IN ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " ASSOCIATED TO ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ", SINGLE COLOR ", paste(color[[i1]], collapse = " "), " HAS BEEN ATTRIBUTED")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+tempo1 <- fun_check(data = color[[i1]], data.name = ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), class = "vector", mode = "character", na.contain = TRUE, fun.name = function.name) # na.contain = TRUE in case of colum of data1
+tempo2 <- fun_check(data = color[[i1]], data.name = ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), class = "factor", na.contain = TRUE, fun.name = function.name) # idem
+if(tempo1$problem == TRUE & tempo2$problem == TRUE){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " MUST BE A FACTOR OR CHARACTER VECTOR OR INTEGER VECTOR") # integer possible because dealt above
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if( ! (all(color[[i1]] %in% colors() | grepl(pattern = "^#", color[[i1]])))){ # check that all strings of low.color start by #
+tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # AND/OR COLOR NAMES GIVEN BY colors(): ", paste(unique(color[[i1]]), collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+if(any(is.na(color[[i1]]))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") IN ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), ", THE COLORS:\n", paste(unique(color[[i1]]), collapse = " "), "\nCONTAINS NA")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+# end check the nature of color
+# check the length of color
+if(is.null(categ) & length(color[[i1]]) != 1){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " MUST BE A SINGLE COLOR IF categ IS NULL")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if( ! is.null(categ)){
+# No problem of NA management by ggplot2 because already removed
+if(categ[[i1]] == "fake_categ" & length(color[[i1]]) != 1){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " MUST BE A SINGLE COLOR IF ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IS NULL")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if(length(color[[i1]]) == length(unique(data1[[i1]][, categ[[i1]]]))){ # here length(color) is equal to the different number of categ
+data1[[i1]][, categ[[i1]]] <- factor(data1[[i1]][, categ[[i1]]]) # if already a factor, change nothing, if characters, levels according to alphabetical order
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") IN ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ", THE FOLLOWING COLORS:\n", paste(color[[i1]], collapse = " "), "\nHAVE BEEN ATTRIBUTED TO THESE CLASSES:\n", paste(levels(factor(data1[[i1]][, categ[[i1]]])), collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}else if(length(color[[i1]]) == length(data1[[i1]][, categ[[i1]]])){# here length(color) is equal to nrow(data1[[i1]]) -> Modif to have length(color) equal to the different number of categ (length(color) == length(levels(data1[[i1]][, categ[[i1]]])))
+data1[[i1]] <- cbind(data1[[i1]], color = color[[i1]], stringsAsFactors = TRUE)
+tempo.check <- unique(data1[[i1]][ , c(categ[[i1]], "color")])
+if( ! (nrow(data1[[i1]]) == length(color[[i1]]) & nrow(tempo.check) == length(unique(data1[[i1]][ , categ[[i1]]])))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color")), " ARGUMENT HAS THE LENGTH OF ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), "\nBUT IS INCORRECTLY ASSOCIATED TO EACH CLASS OF THIS categ:\n", paste(unique(mapply(FUN = "paste", data1[[i1]][ ,categ[[i1]]], data1[[i1]][ ,"color"])), collapse = "\n"))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+data1[[i1]][, categ[[i1]]] <- factor(data1[[i1]][, categ[[i1]]]) # if already a factor, change nothing, if characters, levels according to alphabetical order
+color[[i1]] <- unique(color[[i1]][order(data1[[i1]][, categ[[i1]]])]) # Modif to have length(color) equal to the different number of categ (length(color) == length(levels(data1[[i1]][, categ[[i1]]])))
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count, ") FROM FUNCTION ", function.name, ": ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " HAS THE LENGTH OF ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " COLUMN VALUES\nCOLORS HAVE BEEN RESPECTIVELY ASSOCIATED TO EACH CLASS OF categ AS:\n", paste(levels(factor(data1[[i1]][, categ[[i1]]])), collapse = " "), "\n", paste(color[[i1]], collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else if(length(color[[i1]]) == 1){
+data1[[i1]][, categ[[i1]]] <- factor(data1[[i1]][, categ[[i1]]]) # if already a factor, change nothing, if characters, levels according to alphabetical order
+color[[i1]] <- rep(color[[i1]], length(levels(data1[[i1]][, categ[[i1]]])))
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") IN ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ", COLOR HAS LENGTH 1 MEANING THAT ALL THE DIFFERENT CLASSES OF ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), "\n", paste(levels(factor(data1[[i1]][, categ[[i1]]])), collapse = " "), "\nWILL HAVE THE SAME COLOR\n", paste(color[[i1]], collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " MUST BE\n(1) LENGTH 1\nOR (2) THE LENGTH OF ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " COLUMN VALUES\nOR (3) THE LENGTH OF THE CLASSES IN THIS COLUMN\nHERE IT IS COLOR LENGTH ", length(color[[i1]]), " VERSUS CATEG LENGTH ", length(data1[[i1]][, categ[[i1]]]), " AND CATEG CLASS LENGTH ", length(unique(data1[[i1]][, categ[[i1]]])), "\nPRESENCE OF NA IN THE COLUMN x, y OR categ OF data1 COULD BE THE PROBLEME")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+}
+}
+if((geom[[i1]] == "geom_hline" | geom[[i1]] == "geom_vline") & ! is.null(categ[[i1]])){ # add here after the color management, to deal with the different lines to plot inside any data[[i1]]
+if(categ[[i1]] == "fake_categ"){
+data1[[i1]][, "fake_categ"] <- factor(paste0("Line_", formatC(1:nrow(data1[[i2]]), width = nchar(nrow(data1[[i2]])), flag = "0")))
+}
+}
+tempo <- fun_check(data = alpha[[i1]], data.name = ifelse(length(alpha) == 1, "alpha", paste0("alpha NUMBER ", i1)), prop = TRUE, length = 1, fun.name = function.name)
+if(tempo$problem == TRUE){
+stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+}
+# end loop (checking inside list compartment)
+if(length(data1) > 1){
+if(length(unique(unlist(x)[ ! x == "fake_x"])) > 1){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE x ARGUMENT DOES NOT CONTAIN IDENTICAL COLUMN NAMES:\n", paste(unlist(x), collapse = " "), "\nX-AXIS OVERLAYING DIFFERENT VARIABLES?")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+if(length(data1) > 1){
+if(length(unique(unlist(y)[ ! y == "fake_y"])) > 1){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE y ARGUMENT DOES NOT CONTAIN IDENTICAL COLUMN NAMES:\n", paste(unlist(y), collapse = " "), "\nY-AXIS OVERLAYING DIFFERENT VARIABLES?")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+if(sum(geom %in% "geom_point") > 3){
+tempo.cat <- paste0("ERROR IN ", function.name, ": geom ARGUMENT CANNOT HAVE MORE THAN THREE \"geom_point\" ELEMENTS")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if(length(geom) - sum(geom %in% "geom_point") > 3){
+tempo.cat <- paste0("ERROR IN ", function.name, ": geom ARGUMENT CANNOT HAVE MORE THAN THREE LINE ELEMENTS")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+# x.lim management before transfo by x.log
+if(x.log != "no" & ! is.null(x.lim)){
+if(any(x.lim <= 0)){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nx.lim ARGUMENT CANNOT HAVE ZERO OR NEGATIVE VALUES WITH THE x.log ARGUMENT SET TO ", x.log, ":\n", paste(x.lim, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if(any( ! is.finite(if(x.log == "log10"){log10(x.lim)}else{log2(x.lim)}))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nx.lim ARGUMENT RETURNS INF/NA WITH THE x.log ARGUMENT SET TO ", x.log, "\nAS SCALE COMPUTATION IS ", ifelse(x.log == "log10", "log10", "log2"), ":\n", paste(if(x.log == "log10"){log10(x.lim)}else{log2(x.lim)}, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
}
-
-
-######## fun_inter_ticks() #### define coordinates of secondary ticks
-
-
-fun_inter_ticks <- function(
- lim,
- log = "log10",
- breaks = NULL,
- n = NULL,
- warn.print = TRUE
-){
- # AIM
- # define coordinates and values of secondary ticks
- # ARGUMENTS
- # lim: vector of 2 numbers indicating the limit range of the axis. Order of the 2 values matters (for inverted axis). If log argument is "log2" or "log10", values in lim must be already log transformed. Thus, negative or zero values are allowed
- # log: either "log2" (values in the lim argument are log2 transformed) or "log10" (values in the lim argument are log10 transformed), or "no"
- # breaks: mandatory vector of numbers indicating the main ticks values/positions when log argument is "no". Ignored when log argument is "log2" or "log10"
- # n: number of secondary ticks between each main tick when log argument is "no". Ignored when log argument is "log2" or "log10"
- # warn.print: logical. Print potential warning messages at the end of the execution? If FALSE, warning messages are never printed, but can still be recovered in the returned list
- # RETURN
- # a list containing
- # $log: value of the log argument used
- # $coordinates: the coordinates of the secondary ticks on the axis, between the lim values
- # $values: the corresponding values associated to each coordinate (with log scale, 2^$values or 10^$values is equivalent to the labels of the axis)
- # $warn: the potential warning messages. Use cat() for proper display. NULL if no warning
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # no log scale
- # fun_inter_ticks(lim = c(-4,4), log = "no", breaks = c(-2, 0, 2), n = 3)
- # fun_inter_ticks(lim = c(10, 0), log = "no", breaks = c(10, 8, 6, 4, 2, 0), n = 4)
- # log2
- # fun_inter_ticks(lim = c(-4,4), log = "log2")
- # log10
- # fun_inter_ticks(lim = c(-2,3), log = "log10")
- # DEBUGGING
- # lim = c(2, 3.101) ; log = "no" ; breaks = NULL ; n = NULL ; warn.print = TRUE # for function debugging
- # lim = c(0, 26.5) ; log = "no" ; breaks = c(0, 10, 20) ; n = 3 # for function debugging
- # lim = c(10, 0); log = "no"; breaks = c(10, 8, 6, 4, 2, 0); n = 4 # for function debugging
- # lim = c(-10, -20); log = "no"; breaks = c(-20, -15, -10); n = 4 # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- req.function <- c(
- "fun_check"
- )
- for(i1 in req.function){
- if(length(find(i1, mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nREQUIRED ", i1, "() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # end required function checking
- # argument primary checking
- # arg with no default values
- if(any(missing(lim))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nARGUMENT lim HAS NO DEFAULT VALUE AND REQUIRES ONE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end arg with no default values
- # using fun_check()
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = lim, class = "vector", mode = "numeric", length = 2, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = log, options = c("no", "log2", "log10"), length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(breaks)){
- tempo <- fun_check(data = breaks, class = "vector", mode = "numeric", fun.name = function.name) ; eval(ee)
- }
- if( ! is.null(n)){
- tempo <- fun_check(data = n, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = warn.print, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # end using fun_check()
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument primary checking
- # second round of checking and data preparation
- # dealing with NA
- if(any(is.na(lim)) | any(is.na(log)) | any(is.na(breaks)) | any(is.na(n)) | any(is.na(warn.print))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nNO ARGUMENT CAN HAVE NA VALUES")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end dealing with NA
- # dealing with NULL
- if(is.null(lim) | is.null(log)){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nTHESE ARGUMENTS\nlim\nlog\nCANNOT BE NULL")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end dealing with NULL
- if(all(diff(lim) == 0)){ # isTRUE(all.equal(diff(lim), rep(0, length(diff(lim))))) not used because we strictly need zero as a result
- tempo.cat <- paste0("ERROR IN ", function.name, "\nlim ARGUMENT HAS A NULL RANGE (2 IDENTICAL VALUES): ", paste(lim, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else if(any(lim %in% c(Inf, -Inf))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nlim ARGUMENT CANNOT CONTAIN -Inf OR Inf VALUES")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(log == "no" & is.null(breaks)){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nbreaks ARGUMENT CANNOT BE NULL IF log ARGUMENT IS \"no\"")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! is.null(breaks)){
- if(length(breaks) < 2){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nbreaks ARGUMENT MUST HAVE 2 VALUES AT LEAST (OTHERWISE, INTER TICK POSITIONS CANNOT BE COMPUTED): ", paste(breaks, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! isTRUE(all.equal(diff(sort(breaks)), rep(diff(sort(breaks))[1], length(diff(sort(breaks))))))){ # isTRUE(all.equal(n, 0)) equivalent to n == 0 but deals with floats (approx ok)
- tempo.cat <- paste0("ERROR IN ", function.name, "\nbreaks ARGUMENT MUST HAVE EQUIDISTANT VALUES (OTHERWISE, EQUAL NUMBER OF INTER TICK BETWEEN MAIN TICKS CANNOT BE COMPUTED): ", paste(breaks, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- if( ! is.null(n)){
- if(n <= 0){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nn ARGUMENT MUST BE A POSITIVE AND NON NULL INTEGER: ", paste(n, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # end second round of checking and data preparation
- # main code
- ini.warning.length <- options()$warning.length
- warn <- NULL
- warn.count <- 0
- lim.rank <- rank(lim) # to deal with inverse axis
- if(log != "no"){
- ini.scipen <- options()$scipen
- options(scipen = -1000) # force scientific format
- power10.exp <- as.integer(substring(text = 10^lim, first = (regexpr(pattern = "\\+|\\-", text = 10^lim)))) # recover the power of 10, i.e., integer part of lim. Example recover 08 from 1e+08. Works for log2
- # mantisse <- as.numeric(substr(x = 10^lim, start = 1, stop = (regexpr(pattern = "\\+|\\-", text = 10^lim) - 2))) # recover the mantisse. Example recover 1.22 from 1.22e+08
- options(scipen = ini.scipen) # restore the initial scientific penalty
- tick.pos <- unique(as.vector(outer(2:10, ifelse(log == "log2", 2, 10)^((power10.exp[1] - ifelse(diff(lim.rank) > 0, 1, -1)):(power10.exp[2] + ifelse(diff(lim.rank) > 0, 1, -1)))))) # use log10(2:10) even if log2: it is to get log values between 0 and 1
- tick.pos <- sort(tick.pos, decreasing = ifelse(diff(lim.rank) > 0, FALSE, TRUE))
- if(log == "log2"){
- tick.values <- tick.pos[tick.pos >= min(2^lim) & tick.pos <= max(2^lim)]
- tick.pos <- log2(tick.values)
- }else if(log == "log10"){
- tick.values <- tick.pos[tick.pos >= min(10^lim) & tick.pos <= max(10^lim)]
- tick.pos <- log10(tick.values)
- }
- }else{
- # if(length(breaks) > 1){ # not required because already checked above
- breaks.rank <- rank(c(breaks[1], breaks[length(breaks)]))
- if(diff(breaks.rank) != diff(lim.rank)){
- breaks <- sort(breaks, decreasing = ifelse(diff(lim.rank) < 0, TRUE, FALSE))
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") VALUES IN breaks ARGUMENT NOT IN THE SAME ORDER AS IN lim ARGUMENT -> VALUES REORDERED AS IN lim: ", paste(breaks, collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- breaks.rank <- rank(c(breaks[1], breaks[length(breaks)]))
- }
- # }
- main.tick.dist <- mean(diff(breaks), na.rm = TRUE)
- tick.dist <- main.tick.dist / (n + 1)
- tempo.extra.margin <- max(abs(diff(breaks)), na.rm = TRUE)
- tick.pos <- seq(
- if(diff(breaks.rank) > 0){breaks[1] - tempo.extra.margin}else{breaks[1] + tempo.extra.margin},
- if(diff(breaks.rank) > 0){breaks[length(breaks)] + tempo.extra.margin}else{breaks[length(breaks)] - tempo.extra.margin},
- by = tick.dist
- )
- tick.pos <- tick.pos[tick.pos >= min(lim) & tick.pos <= max(lim)]
- tick.values <- tick.pos
- }
- if(any(is.na(tick.pos) | ! is.finite(tick.pos))){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, ": NA or Inf GENERATED FOR THE INTER TICK POSITIONS: ", paste(tick.pos, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(length(tick.pos) == 0){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NO INTER TICKS COMPUTED BETWEEN THEN LIMITS INDICATED: ", paste(lim, collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- output <- list(log = log, coordinates = tick.pos, values = tick.values, warn = warn)
- if(warn.print == TRUE & ! is.null(warn)){
- options(warning.length = 8170)
- on.exit(warning(paste0("FROM ", function.name, ":\n\n", warn), call. = FALSE)) # to recover the warning messages, see $warn
- on.exit(exp = options(warning.length = ini.warning.length), add = TRUE)
- }
- return(output)
}
-
-
-######## fun_post_plot() #### set graph param after plotting (axes redesign for instance)
-
-
-
-
-
-# Check OK: clear to go Apollo
-fun_post_plot <- function(
- x.side = 0,
- x.log.scale = FALSE,
- x.categ = NULL,
- x.categ.pos = NULL,
- x.lab = "",
- x.axis.size = 1.5,
- x.label.size = 1.5,
- x.dist.legend = 0.5,
- x.nb.inter.tick = 1,
- y.side = 0,
- y.log.scale = FALSE,
- y.categ = NULL,
- y.categ.pos = NULL,
- y.lab = "",
- y.axis.size = 1.5,
- y.label.size = 1.5,
- y.dist.legend = 0.5,
- y.nb.inter.tick = 1,
- text.angle = 90,
- tick.length = 0.5,
- sec.tick.length = 0.3,
- bg.color = NULL,
- grid.lwd = NULL,
- grid.col = "white",
- corner.text = "",
- corner.text.size = 1,
- just.label.add = FALSE,
- par.reset = FALSE,
- custom.par = NULL
-){
- # AIM
- # redesign axis. If x.side = 0, y.side = 0, the function just adds text at topright of the graph and reset par() for next graphics and provides outputs (see below)
- # provide also positions for legend or additional text on the graph
- # use fun_prior_plot() before this function for initial inactivation of the axis drawings
- # ARGUMENTS
- # x.side: axis at the bottom (1) or top (3) of the region figure. Write 0 for no change
- # x.log.scale: Log scale for the x-axis? Either TRUE or FALSE
- # x.categ: character vector representing the classes (levels()) to specify when the x-axis is qualititative(stripchart, boxplot)
- # x.categ.pos: position of the classes names (numeric vector of identical length than x.categ). If left NULL, this will be 1:length(levels())
- # x.lab: label of the x-axis. If x.side == 0 and x.lab != "", then x.lab is printed
- # x.axis.size: positive numeric. Increase or decrease the size of the x axis numbers. Value 1 does not change it, 0.5 decreases by half, 2 increases by 2. Also control the size of displayed categories
- # x.label.size: positive numeric. Increase or decrease the size of the x axis legend text. Value 1 does not change it, 0.5 decreases by half, 2 increases by 2
- # x.dist.legend: increase the number to move x-axis legends away in inches (first number of mgp argument of par() but in inches)
- # x.nb.inter.tick: number of secondary ticks between main ticks on x-axis (only if not log scale). 0 means no secondary ticks
- # y.side: axis at the left (2) or right (4) of the region figure. Write 0 for no change
- # y.log.scale: Log scale for the y-axis? Either TRUE or FALSE
- # y.categ: classes (levels()) to specify when the y-axis is qualititative(stripchart, boxplot)
- # y.categ.pos: position of the classes names (numeric vector of identical length than y.categ). If left NULL, this will be 1:length(levels())
- # y.lab: label of the y-axis. If y.side == 0 and y.lab != "", then y.lab is printed
- # y.axis.size: positive numeric. Increase or decrease the size of the y axis numbers. Value 1 does not change it, 0.5 decreases by half, 2 increases by 2. Also control the size of displayed categories
- # y.label.size: positive numeric. Increase or decrease the size of the y axis legend text. Value 1 does not change it, 0.5 decreases by half, 2 increases by 2
- # y.dist.legend: increase the number to move y-axis legends away in inches (first number of mgp argument of par() but in inches)
- # y.nb.inter.tick: number of secondary ticks between main ticks on y-axis (only if not log scale). 0 means non secondary ticks
- # text.angle: angle of the text when axis is qualitative
- # tick.length: length of the main ticks (1 means complete the distance between the plot region and the axis numbers, 0.5 means half the length, etc., 0 for no ticks)
- # sec.tick.length: length of the secondary ticks (1 means complete the distance between the plot region and the axis numbers, 0.5 means half the length, etc., 0 for no ticks)
- # bg.color: background color of the plot region. NULL for no color. BEWARE: cover/hide an existing plot !
- # grid.lwd: if non NULL, activate the grid line (specify the line width)
- # grid.col: grid line color (only if grid.lwd non NULL)
- # corner.text: text to add at the top right corner of the window
- # corner.text.size: positive numeric. Increase or decrease the size of the text. Value 1 does not change it, 0.5 decreases by half, 2 increases by 2
- # par.reset: to reset all the graphics parameters. BEWARE: TRUE can generate display problems, mainly in graphic devices with multiple figure regions
- # just.label.add: just add axis labels (legend)? Either TRUE or FALSE. If TRUE, at least (x.side == 0 & x.lab != "") or (y.side == 0 & y.lab != "") must be set to display the corresponding x.lab or y.lab
- # custom.par: list that provides the parameters that reset all the graphics parameters. BEWARE: if NULL and par.reset == TRUE, the default par() parameters are used
- # RETURN
- # a list containing:
- # $x.mid.left.dev.region: middle of the left margin of the device region, in coordinates of the x-axis
- # $x.left.dev.region: left side of the left margin (including the potential margin of the device region), in coordinates of the x-axis
- # $x.mid.right.dev.region: middle of the right margin of the device region, in coordinates of the x-axis
- # $x.right.dev.region: right side of the right margin (including the potential margin of the device region), in coordinates of the x-axis
- # $x.mid.left.fig.region: middle of the left margin of the figure region, in coordinates of the x-axis
- # $x.left.fig.region: left side of the left margin, in coordinates of the x-axis
- # $x.mid.right.fig.region: middle of the right margin of the figure region, in coordinates of the x-axis
- # $x.right.fig.region: right side of the right margin, in coordinates of the x-axis
- # $x.left.plot.region: left side of the plot region, in coordinates of the x-axis
- # $x.right.plot.region: right side of the plot region, in coordinates of the x-axis
- # $x.mid.plot.region: middle of the plot region, in coordinates of the x-axis
- # $y.mid.bottom.dev.region: middle of the bottom margin of the device region, in coordinates of the y-axis
- # $y.bottom.dev.region: bottom side of the bottom margin (including the potential margin of the device region), in coordinates of the y-axis
- # $y.mid.top.dev.region: middle of the top margin of the device region, in coordinates of the y-axis
- # $y.top.dev.region: top side of the top margin (including the potential margin of the device region), in coordinates of the y-axis
- # $y.mid.bottom.fig.region: middle of the bottom margin of the figure region, in coordinates of the y-axis
- # $y.bottom.fig.region: bottom of the bottom margin of the figure region, in coordinates of the y-axis
- # $y.mid.top.fig.region: middle of the top margin of the figure region, in coordinates of the y-axis
- # $y.top.fig.region: top of the top margin of the figure region, in coordinates of the y-axis
- # $y.top.plot.region: top of the plot region, in coordinates of the y-axis
- # $y.bottom.plot.region: bottom of the plot region, in coordinates of the y-axis
- # $y.mid.plot.region: middle of the plot region, in coordinates of the y-axis
- # $text: warning text
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # fun_open() to reinitialize graph parameters if par.reset = TRUE and custom.par = NULL
- # EXAMPLES
- # Example of log axis with log y-axis and unmodified x-axis:
- # prior.par <- fun_prior_plot(param.reinitial = TRUE, xlog.scale = FALSE, ylog.scale = TRUE, remove.label = TRUE, remove.x.axis = FALSE, remove.y.axis = TRUE, down.space = 1, left.space = 1, up.space = 1, right.space = 1, orient = 1, dist.legend = 0.5, tick.length = 0.5, box.type = "n", amplif.label = 1, amplif.axis = 1, display.extend = FALSE, return.par = TRUE) ; plot(1:100, log = "y") ; fun_post_plot(y.side = 2, y.log.scale = prior.par$ylog, x.lab = "Values", y.lab = "TEST", y.axis.size = 1.25, y.label.size = 1.5, y.dist.legend = 0.7, just.label.add = ! prior.par$ann)
- # Example of log axis with redrawn x-axis and y-axis:
- # prior.par <- fun_prior_plot(param.reinitial = TRUE) ; plot(1:100) ; fun_post_plot(x.side = 1, x.lab = "Values", y.side = 2, y.lab = "TEST", y.axis.size = 1, y.label.size = 2, y.dist.legend = 0.6)
- # Example of title easily added to a plot:
- # plot(1:100) ; para <- fun_post_plot(corner.text = "TITLE ADDED") # try also: par(xpd = TRUE) ; text(x = para$x.mid.left.fig.region, y = para$y.mid.top.fig.region, labels = "TITLE ADDED", cex = 0.5)
- # example with margins in the device region:
- # windows(5,5) ; fun_prior_plot(box.type = "o") ; par(mai=c(0.5,0.5,0.5,0.5), omi = c(0.25,0.25,1,0.25), xaxs = "i", yaxs = "i") ; plot(0:10) ; a <- fun_post_plot(x.side = 0, y.side = 0) ; x <- c(a$x.mid.left.dev.region, a$x.left.dev.region, a$x.mid.right.dev.region, a$x.right.dev.region, a$x.mid.left.fig.region, a$x.left.fig.region, a$x.mid.right.fig.region, a$x.right.fig.region, a$x.right.plot.region, a$x.left.plot.region, a$x.mid.plot.region) ; y <- c(a$y.mid.bottom.dev.region, a$y.bottom.dev.region, a$y.mid.top.dev.region, a$y.top.dev.region, a$y.mid.bottom.fig.region, a$y.bottom.fig.region, a$y.mid.top.fig.region, a$y.top.fig.region, a$y.top.plot.region, a$y.bottom.plot.region, a$y.mid.plot.region) ; par(xpd = NA) ; points(x = rep(5, length(y)), y = y, pch = 16, col = "red") ; text(x = rep(5, length(y)), y = y, c("y.mid.bottom.dev.region", "y.bottom.dev.region", "y.mid.top.dev.region", "y.top.dev.region", "y.mid.bottom.fig.region", "y.bottom.fig.region", "y.mid.top.fig.region", "y.top.fig.region", "y.top.plot.region", "y.bottom.plot.region", "y.mid.plot.region"), cex = 0.65, col = grey(0.25)) ; points(y = rep(5, length(x)), x = x, pch = 16, col = "blue") ; text(y = rep(5, length(x)), x = x, c("x.mid.left.dev.region", "x.left.dev.region", "x.mid.right.dev.region", "x.right.dev.region", "x.mid.left.fig.region", "x.left.fig.region", "x.mid.right.fig.region", "x.right.fig.region", "x.right.plot.region", "x.left.plot.region", "x.mid.plot.region"), cex = 0.65, srt = 90, col = grey(0.25))
- # DEBUGGING
- # x.side = 0 ; x.log.scale = FALSE ; x.categ = NULL ; x.categ.pos = NULL ; x.lab = "" ; x.axis.size = 1.5 ; x.label.size = 1.5 ; x.dist.legend = 1 ; x.nb.inter.tick = 1 ; y.side = 0 ; y.log.scale = FALSE ; y.categ = NULL ; y.categ.pos = NULL ; y.lab = "" ; y.axis.size = 1.5 ; y.label.size = 1.5 ; y.dist.legend = 0.7 ; y.nb.inter.tick = 1 ; text.angle = 90 ; tick.length = 0.5 ; sec.tick.length = 0.3 ; bg.color = NULL ; grid.lwd = NULL ; grid.col = "white" ; corner.text = "" ; corner.text.size = 1 ; just.label.add = FALSE ; par.reset = FALSE ; custom.par = NULL # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(length(utils::find("fun_open", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_open() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = x.side, options = c(0, 1, 3), length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = x.log.scale, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(x.categ)){
- tempo <- fun_check(data = x.categ, class = "character", na.contain = TRUE, fun.name = function.name) ; eval(ee)
- }
- if( ! is.null(x.categ.pos)){
- tempo <- fun_check(data = x.categ.pos, class = "vector", mode = "numeric", fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = x.lab, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = x.axis.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = x.label.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = x.dist.legend, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = x.nb.inter.tick, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = y.side, options = c(0, 2, 4), length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = y.log.scale, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(y.categ)){
- tempo <- fun_check(data = y.categ, class = "character", na.contain = TRUE, fun.name = function.name) ; eval(ee)
- }
- if( ! is.null(y.categ.pos)){
- tempo <- fun_check(data = y.categ.pos, class = "vector", mode = "numeric", fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = y.lab, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = y.axis.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = y.label.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = y.dist.legend, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = y.nb.inter.tick, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = text.angle, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = tick.length, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = sec.tick.length, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
- if( ! is.null(bg.color)){
- tempo <- fun_check(data = bg.color, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- if( ! (bg.color %in% colors() | grepl(pattern = "^#", bg.color))){ # check color
- tempo.cat <- paste0("ERROR IN ", function.name, ": bg.color ARGUMENT MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # OR A COLOR NAME GIVEN BY colors()")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- if( ! is.null(grid.lwd)){
- tempo <- fun_check(data = grid.lwd, class = "vector", mode = "numeric", neg.values = FALSE, fun.name = function.name) ; eval(ee)
- }
- if( ! is.null(grid.col)){
- tempo <- fun_check(data = grid.col, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- if( ! (grid.col %in% colors() | grepl(pattern = "^#", grid.col))){ # check color
- tempo.cat <- paste0("ERROR IN ", function.name, ": grid.col ARGUMENT MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # OR A COLOR NAME GIVEN BY colors()")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- tempo <- fun_check(data = corner.text, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = corner.text.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = just.label.add, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = par.reset, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(custom.par)){
- tempo <- fun_check(data = custom.par, typeof = "list", length = 1, fun.name = function.name) ; eval(ee)
- }
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- text <- NULL
- par(tcl = -par()$mgp[2] * tick.length)
- if(x.log.scale == TRUE){
- grid.coord.x <- c(10^par("usr")[1], 10^par("usr")[2])
- }else{
- grid.coord.x <- c(par("usr")[1], par("usr")[2])
- }
- if(y.log.scale == TRUE){
- grid.coord.y <- c(10^par("usr")[3], 10^par("usr")[4])
- }else{
- grid.coord.y <- c(par("usr")[3], par("usr")[4])
- }
- if( ! is.null(bg.color)){
- rect(grid.coord.x[1], grid.coord.y[1], grid.coord.x[2], grid.coord.y[2], col = bg.color, border = NA)
- }
- if( ! is.null(grid.lwd)){
- grid(nx = NA, ny = NULL, col = grid.col, lty = 1, lwd = grid.lwd)
- }
- if(x.log.scale == TRUE){
- x.mid.left.dev.region <- 10^(par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] - ((par("usr")[2] - par("usr")[1]) / ((par("omd")[2] - par("omd")[1]) * (par("plt")[2] - par("plt")[1]))) * par("omd")[1] / 2) # in x coordinates, to position axis labeling at the bottom of the graph (according to x scale)
- x.left.dev.region <- 10^(par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] - ((par("usr")[2] - par("usr")[1]) / ((par("omd")[2] - par("omd")[1]) * (par("plt")[2] - par("plt")[1]))) * par("omd")[1]) # in x coordinates
- x.mid.right.dev.region <- 10^(par("usr")[2] + ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * (1 - par("plt")[2]) + ((par("usr")[2] - par("usr")[1]) / ((par("omd")[2] - par("omd")[1]) * (par("plt")[2] - par("plt")[1]))) * (1 - par("omd")[2]) / 2) # in x coordinates, to position axis labeling at the top of the graph (according to x scale)
- x.right.dev.region <- 10^(par("usr")[2] + ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * (1 - par("plt")[2]) + ((par("usr")[2] - par("usr")[1]) / ((par("omd")[2] - par("omd")[1]) * (par("plt")[2] - par("plt")[1]))) * (1 - par("omd")[2])) # in x coordinates
- x.mid.left.fig.region <- 10^(par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] / 2) # in x coordinates, to position axis labeling at the bottom of the graph (according to x scale)
- x.left.fig.region <- 10^(par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1]) # in x coordinates
- x.mid.right.fig.region <- 10^(par("usr")[2] + ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * (1 - par("plt")[2]) / 2) # in x coordinates, to position axis labeling at the top of the graph (according to x scale)
- x.right.fig.region <- 10^(par("usr")[2] + ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * (1 - par("plt")[2])) # in x coordinates
- x.left.plot.region <- 10^par("usr")[1] # in x coordinates, left of the plot region (according to x scale)
- x.right.plot.region <- 10^par("usr")[2] # in x coordinates, right of the plot region (according to x scale)
- x.mid.plot.region <- 10^((par("usr")[2] + par("usr")[1]) / 2) # in x coordinates, right of the plot region (according to x scale)
- }else{
- x.mid.left.dev.region <- (par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] - ((par("usr")[2] - par("usr")[1]) / ((par("omd")[2] - par("omd")[1]) * (par("plt")[2] - par("plt")[1]))) * par("omd")[1] / 2) # in x coordinates, to position axis labeling at the bottom of the graph (according to x scale)
- x.left.dev.region <- (par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] - ((par("usr")[2] - par("usr")[1]) / ((par("omd")[2] - par("omd")[1]) * (par("plt")[2] - par("plt")[1]))) * par("omd")[1]) # in x coordinates
- x.mid.right.dev.region <- (par("usr")[2] + ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * (1 - par("plt")[2]) + ((par("usr")[2] - par("usr")[1]) / ((par("omd")[2] - par("omd")[1]) * (par("plt")[2] - par("plt")[1]))) * (1 - par("omd")[2]) / 2) # in x coordinates, to position axis labeling at the top of the graph (according to x scale)
- x.right.dev.region <- (par("usr")[2] + ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * (1 - par("plt")[2]) + ((par("usr")[2] - par("usr")[1]) / ((par("omd")[2] - par("omd")[1]) * (par("plt")[2] - par("plt")[1]))) * (1 - par("omd")[2])) # in x coordinates
- x.mid.left.fig.region <- (par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] / 2) # in x coordinates, to position axis labeling at the bottom of the graph (according to x scale)
- x.left.fig.region <- (par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1]) # in x coordinates
- x.mid.right.fig.region <- (par("usr")[2] + ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * (1 - par("plt")[2]) / 2) # in x coordinates, to position axis labeling at the top of the graph (according to x scale)
- x.right.fig.region <- (par("usr")[2] + ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * (1 - par("plt")[2])) # in x coordinates
- x.left.plot.region <- par("usr")[1] # in x coordinates, left of the plot region (according to x scale)
- x.right.plot.region <- par("usr")[2] # in x coordinates, right of the plot region (according to x scale)
- x.mid.plot.region <- (par("usr")[2] + par("usr")[1]) / 2 # in x coordinates, right of the plot region (according to x scale)
- }
- if(y.log.scale == TRUE){
- y.mid.bottom.dev.region <- 10^(par("usr")[3] - ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * par("plt")[3] - ((par("usr")[4] - par("usr")[3]) / ((par("omd")[4] - par("omd")[3]) * (par("plt")[4] - par("plt")[3]))) * (par("omd")[3] / 2)) # in y coordinates, to position axis labeling at the bottom of the graph (according to y scale). Ex mid.bottom.space
- y.bottom.dev.region <- 10^(par("usr")[3] - ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * par("plt")[3] - ((par("usr")[4] - par("usr")[3]) / ((par("omd")[4] - par("omd")[3]) * (par("plt")[4] - par("plt")[3]))) * par("omd")[3]) # in y coordinates
- y.mid.top.dev.region <- 10^(par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / ((par("omd")[4] - par("omd")[3]) * (par("plt")[4] - par("plt")[3]))) * (1 - par("omd")[4]) / 2) # in y coordinates, to position axis labeling at the top of the graph (according to y scale). Ex mid.top.space
- y.top.dev.region <- 10^(par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / ((par("omd")[4] - par("omd")[3]) * (par("plt")[4] - par("plt")[3]))) * (1 - par("omd")[4])) # in y coordinates
- y.mid.bottom.fig.region <- 10^(par("usr")[3] - ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * par("plt")[3] / 2) # in y coordinates, to position axis labeling at the bottom of the graph (according to y scale). Ex mid.bottom.space
- y.bottom.fig.region <- 10^(par("usr")[3] - ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * par("plt")[3]) # in y coordinates
- y.mid.top.fig.region <- 10^(par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) / 2) # in y coordinates, to position axis labeling at the top of the graph (according to y scale). Ex mid.top.space
- y.top.fig.region <- 10^(par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4])) # in y coordinates
- y.top.plot.region <- 10^par("usr")[4] # in y coordinates, top of the plot region (according to y scale)
- y.bottom.plot.region <- 10^par("usr")[3] # in y coordinates, bottom of the plot region (according to y scale)
- y.mid.plot.region <- (par("usr")[3] + par("usr")[4]) / 2 # in x coordinates, right of the plot region (according to x scale)
- }else{
- y.mid.bottom.dev.region <- (par("usr")[3] - ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * par("plt")[3] - ((par("usr")[4] - par("usr")[3]) / ((par("omd")[4] - par("omd")[3]) * (par("plt")[4] - par("plt")[3]))) * (par("omd")[3] / 2)) # in y coordinates, to position axis labeling at the bottom of the graph (according to y scale). Ex mid.bottom.space
- y.bottom.dev.region <- (par("usr")[3] - ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * par("plt")[3] - ((par("usr")[4] - par("usr")[3]) / ((par("omd")[4] - par("omd")[3]) * (par("plt")[4] - par("plt")[3]))) * par("omd")[3]) # in y coordinates
- y.mid.top.dev.region <- (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / ((par("omd")[4] - par("omd")[3]) * (par("plt")[4] - par("plt")[3]))) * (1 - par("omd")[4]) / 2) # in y coordinates, to position axis labeling at the top of the graph (according to y scale). Ex mid.top.space
- y.top.dev.region <- (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / ((par("omd")[4] - par("omd")[3]) * (par("plt")[4] - par("plt")[3]))) * (1 - par("omd")[4])) # in y coordinates
- y.mid.bottom.fig.region <- (par("usr")[3] - ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * par("plt")[3] / 2) # in y coordinates, to position axis labeling at the bottom of the graph (according to y scale). Ex mid.bottom.space
- y.bottom.fig.region <- (par("usr")[3] - ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * par("plt")[3]) # in y coordinates
- y.mid.top.fig.region <- (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) / 2) # in y coordinates, to position axis labeling at the top of the graph (according to y scale). Ex mid.top.space
- y.top.fig.region <- (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4])) # in y coordinates
- y.top.plot.region <- par("usr")[4] # in y coordinates, top of the plot region (according to y scale)
- y.bottom.plot.region <- par("usr")[3] # in y coordinates, bottom of the plot region (according to y scale)
- y.mid.plot.region <- ((par("usr")[3] + par("usr")[4]) / 2) # in x coordinates, right of the plot region (according to x scale)
- }
- if(any(sapply(FUN = all.equal, c(1, 3), x.side) == TRUE)){
- par(xpd=FALSE, xaxt="s")
- if(is.null(x.categ) & x.log.scale == TRUE){
- if(any(par()$xaxp[1:2] == 0)){ # any(sapply(FUN = all.equal, par()$xaxp[1:2], 0) == TRUE) not used because we strictly need zero as a result. Beware: write "== TRUE", because the result is otherwise character and a warning message appears using any()
- if(par()$xaxp[1] == 0){ # isTRUE(all.equal(par()$xaxp[1], 0)) not used because we strictly need zero as a result
- par(xaxp = c(10^-30, par()$xaxp[2:3])) # because log10(par()$xaxp[1] == 0) == -Inf
- }
- if(par()$xaxp[2] == 0){ # isTRUE(all.equal(par()$xaxp[1], 0)) not used because we strictly need zero as a result
- par(xaxp = c(par()$xaxp[1], 10^-30, par()$xaxp[3])) # because log10(par()$xaxp[2] == 0) == -Inf
- }
- }
- axis(side = x.side, at = c(10^par()$usr[1], 10^par()$usr[2]), labels=rep("", 2), lwd=1, lwd.ticks = 0) # draw the axis line
- mtext(side = x.side, text = x.lab, line = x.dist.legend / 0.2, las = 0, cex = x.label.size)
- par(tcl = -par()$mgp[2] * sec.tick.length) # length of the secondary ticks are reduced
- suppressWarnings(rug(10^outer(c((log10(par("xaxp")[1]) -1):log10(par("xaxp")[2])), log10(1:10), "+"), ticksize = NA, side = x.side)) # ticksize = NA to allow the use of par()$tcl value
- par(tcl = -par()$mgp[2] * tick.length) # back to main ticks
- axis(side = x.side, at = c(1e-15, 1e-14, 1e-13, 1e-12, 1e-11, 1e-10, 1e-9, 1e-8, 1e-7, 1e-6, 1e-5, 1e-4, 1e-3, 1e-2, 1e-1, 1e0, 1e1, 1e2, 1e3, 1e4, 1e5, 1e6, 1e7, 1e8, 1e9, 1e10), labels = expression(10^-15, 10^-14, 10^-13, 10^-12, 10^-11, 10^-10, 10^-9, 10^-8, 10^-7, 10^-6, 10^-5, 10^-4, 10^-3, 10^-2, 10^-1, 10^0, 10^1, 10^2, 10^3, 10^4, 10^5, 10^6, 10^7, 10^8, 10^9, 10^10), lwd = 0, lwd.ticks = 1, cex.axis = x.axis.size)
- x.text <- 10^par("usr")[2]
- }else if(is.null(x.categ) & x.log.scale == FALSE){
- axis(side=x.side, at=c(par()$usr[1], par()$usr[2]), labels=rep("", 2), lwd=1, lwd.ticks=0) # draw the axis line
- axis(side=x.side, at=round(seq(par()$xaxp[1], par()$xaxp[2], length.out=par()$xaxp[3]+1), 2), cex.axis = x.axis.size) # axis(side=x.side, at=round(seq(par()$xaxp[1], par()$xaxp[2], length.out=par()$xaxp[3]+1), 2), labels = format(round(seq(par()$xaxp[1], par()$xaxp[2], length.out=par()$xaxp[3]+1), 2), big.mark=','), cex.axis = x.axis.size) # to get the 1000 comma separator
- mtext(side = x.side, text = x.lab, line = x.dist.legend / 0.2, las = 0, cex = x.label.size)
- if(x.nb.inter.tick > 0){
- inter.tick.unit <- (par("xaxp")[2] - par("xaxp")[1]) / par("xaxp")[3]
- par(tcl = -par()$mgp[2] * sec.tick.length) # length of the ticks are reduced
- suppressWarnings(rug(seq(par("xaxp")[1] - 10 * inter.tick.unit, par("xaxp")[2] + 10 * inter.tick.unit, by = inter.tick.unit / (1 + x.nb.inter.tick)), ticksize = NA, x.side)) # ticksize = NA to allow the use of par()$tcl value
- par(tcl = -par()$mgp[2] * tick.length) # back to main ticks
- }
- x.text <- par("usr")[2]
- }else if(( ! is.null(x.categ)) & x.log.scale == FALSE){
- if(is.null(x.categ.pos)){
- x.categ.pos <- 1:length(x.categ)
- }else if(length(x.categ.pos) != length(x.categ)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": x.categ.pos MUST BE THE SAME LENGTH AS x.categ")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- par(xpd = TRUE)
- if(isTRUE(all.equal(x.side, 1))){ #isTRUE(all.equal(x.side, 1)) is similar to x.side == 1 but deals with float
- segments(x0 = x.left.plot.region, x1 = x.right.plot.region, y0 = y.bottom.plot.region, y1 = y.bottom.plot.region) # draw the line of the axis
- text(x = x.categ.pos, y = y.mid.bottom.fig.region, labels = x.categ, srt = text.angle, cex = x.axis.size)
- }else if(isTRUE(all.equal(x.side, 3))){ #isTRUE(all.equal(x.side, 1)) is similar to x.side == 3 but deals with float
- segments(x0 = x.left.plot.region, x1 = x.right.plot.region, y0 = y.top.plot.region, y1 = y.top.plot.region) # draw the line of the axis
- text(x = x.categ.pos, y = y.mid.top.fig.region, labels = x.categ, srt = text.angle, cex = x.axis.size)
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, ": ARGUMENT x.side CAN ONLY BE 1 OR 3")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- par(xpd = FALSE)
- x.text <- par("usr")[2]
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, ": PROBLEM WITH THE x.side (", x.side ,") OR x.log.scale (", x.log.scale,") ARGUMENTS")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }else{
- x.text <- par("usr")[2]
- }
- if(any(sapply(FUN = all.equal, c(2, 4), y.side) == TRUE)){
- par(xpd=FALSE, yaxt="s")
- if(is.null(y.categ) & y.log.scale == TRUE){
- if(any(par()$yaxp[1:2] == 0)){ # any(sapply(FUN = all.equal, par()$yaxp[1:2], 0) == TRUE) not used because we strictly need zero as a result. Beware: write "== TRUE", because the result is otherwise character and a warning message appears using any()
- if(par()$yaxp[1] == 0){ # strict zero needed
- par(yaxp = c(10^-30, par()$yaxp[2:3])) # because log10(par()$yaxp[1] == 0) == -Inf
- }
- if(par()$yaxp[2] == 0){ # strict zero needed
- par(yaxp = c(par()$yaxp[1], 10^-30, par()$yaxp[3])) # because log10(par()$yaxp[2] == 0) == -Inf
- }
- }
- axis(side=y.side, at=c(10^par()$usr[3], 10^par()$usr[4]), labels=rep("", 2), lwd=1, lwd.ticks=0) # draw the axis line
- par(tcl = -par()$mgp[2] * sec.tick.length) # length of the ticks are reduced
- suppressWarnings(rug(10^outer(c((log10(par("yaxp")[1])-1):log10(par("yaxp")[2])), log10(1:10), "+"), ticksize = NA, side = y.side)) # ticksize = NA to allow the use of par()$tcl value
- par(tcl = -par()$mgp[2] * tick.length) # back to main tick length
- axis(side = y.side, at = c(1e-15, 1e-14, 1e-13, 1e-12, 1e-11, 1e-10, 1e-9, 1e-8, 1e-7, 1e-6, 1e-5, 1e-4, 1e-3, 1e-2, 1e-1, 1e0, 1e1, 1e2, 1e3, 1e4, 1e5, 1e6, 1e7, 1e8, 1e9, 1e10), labels = expression(10^-15, 10^-14, 10^-13, 10^-12, 10^-11, 10^-10, 10^-9, 10^-8, 10^-7, 10^-6, 10^-5, 10^-4, 10^-3, 10^-2, 10^-1, 10^0, 10^1, 10^2, 10^3, 10^4, 10^5, 10^6, 10^7, 10^8, 10^9, 10^10), lwd = 0, lwd.ticks = 1, cex.axis = y.axis.size)
- y.text <- 10^(par("usr")[4] + (par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3]) * (1 - par("plt")[4]))
- mtext(side = y.side, text = y.lab, line = y.dist.legend / 0.2, las = 0, cex = y.label.size)
- }else if(is.null(y.categ) & y.log.scale == FALSE){
- axis(side=y.side, at=c(par()$usr[3], par()$usr[4]), labels=rep("", 2), lwd=1, lwd.ticks=0) # draw the axis line
- axis(side=y.side, at=round(seq(par()$yaxp[1], par()$yaxp[2], length.out=par()$yaxp[3]+1), 2), cex.axis = y.axis.size)
- mtext(side = y.side, text = y.lab, line = y.dist.legend / 0.2, las = 0, cex = y.label.size)
- if(y.nb.inter.tick > 0){
- inter.tick.unit <- (par("yaxp")[2] - par("yaxp")[1]) / par("yaxp")[3]
- par(tcl = -par()$mgp[2] * sec.tick.length) # length of the ticks are reduced
- suppressWarnings(rug(seq(par("yaxp")[1] - 10 * inter.tick.unit, par("yaxp")[2] + 10 * inter.tick.unit, by = inter.tick.unit / (1 + y.nb.inter.tick)), ticksize = NA, side=y.side)) # ticksize = NA to allow the use of par()$tcl value
- par(tcl = -par()$mgp[2] * tick.length) # back to main tick length
- }
- y.text <- (par("usr")[4] + (par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3]) * (1 - par("plt")[4]))
- }else if(( ! is.null(y.categ)) & y.log.scale == FALSE){
- if(is.null(y.categ.pos)){
- y.categ.pos <- 1:length(y.categ)
- }else if(length(y.categ.pos) != length(y.categ)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": y.categ.pos MUST BE THE SAME LENGTH AS y.categ")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- axis(side = y.side, at = y.categ.pos, labels = rep("", length(y.categ)), lwd=0, lwd.ticks=1) # draw the line of the axis
- par(xpd = TRUE)
- if(isTRUE(all.equal(y.side, 2))){ #isTRUE(all.equal(y.side, 2)) is similar to y.side == 2 but deals with float
- text(x = x.mid.left.fig.region, y = y.categ.pos, labels = y.categ, srt = text.angle, cex = y.axis.size)
- }else if(isTRUE(all.equal(y.side, 4))){ # idem
- text(x = x.mid.right.fig.region, y = y.categ.pos, labels = y.categ, srt = text.angle, cex = y.axis.size)
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, ": ARGUMENT y.side CAN ONLY BE 2 OR 4")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- par(xpd = FALSE)
- y.text <- (par("usr")[4] + (par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3]) * (1 - par("plt")[4]))
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, ": PROBLEM WITH THE y.side (", y.side ,") OR y.log.scale (", y.log.scale,") ARGUMENTS")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }else{
- y.text <- (par("usr")[4] + (par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3]) * (1 - par("plt")[4]))
- }
- par(xpd=NA)
- text(x = x.mid.right.fig.region, y = y.text, corner.text, adj=c(1, 1.1), cex = corner.text.size) # text at the topright corner. Replace x.right.fig.region by x.text if text at the right edge of the plot region
- if(just.label.add == TRUE & isTRUE(all.equal(x.side, 0)) & x.lab != ""){
- text(x = x.mid.plot.region, y = y.mid.bottom.fig.region, x.lab, adj=c(0.5, 0.5), cex = x.label.size) # x label
- }
- if(just.label.add == TRUE & isTRUE(all.equal(y.side, 0)) & y.lab != ""){
- text(x = y.mid.plot.region, y = x.mid.left.fig.region, y.lab, adj=c(0.5, 0.5), cex = y.label.size) # x label
- }
- par(xpd=FALSE)
- if(par.reset == TRUE){
- tempo.par <- fun_open(pdf = FALSE, return.output = TRUE)
- invisible(dev.off()) # close the new window
- if( ! is.null(custom.par)){
- if( ! names(custom.par) %in% names(tempo.par$ini.par)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": custom.par ARGUMENT SHOULD HAVE THE NAMES OF THE COMPARTMENT LIST COMING FROM THE par() LIST")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- par(custom.par)
- text <- c(text, "\nGRAPH PARAMETERS SET TO VALUES DEFINED BY custom.par ARGUMENT\n")
- }else{
- par(tempo.par$ini.par)
- text <- c(text, "\nGRAPH PARAMETERS RESET TO par() DEFAULT VALUES\n")
- }
- }
- output <- list(x.mid.left.dev.region = x.mid.left.dev.region, x.left.dev.region = x.left.dev.region, x.mid.right.dev.region = x.mid.right.dev.region, x.right.dev.region = x.right.dev.region, x.mid.left.fig.region = x.mid.left.fig.region, x.left.fig.region = x.left.fig.region, x.mid.right.fig.region = x.mid.right.fig.region, x.right.fig.region = x.right.fig.region, x.left.plot.region = x.left.plot.region, x.right.plot.region = x.right.plot.region, x.mid.plot.region = x.mid.plot.region, y.mid.bottom.dev.region = y.mid.bottom.dev.region, y.bottom.dev.region = y.bottom.dev.region, y.mid.top.dev.region = y.mid.top.dev.region, y.top.dev.region = y.top.dev.region, y.mid.bottom.fig.region = y.mid.bottom.fig.region, y.bottom.fig.region = y.bottom.fig.region, y.mid.top.fig.region = y.mid.top.fig.region, y.top.fig.region = y.top.fig.region, y.top.plot.region = y.top.plot.region, y.bottom.plot.region = y.bottom.plot.region, y.mid.plot.region = y.mid.plot.region, text = text)
- return(output)
+if(x.log != "no" & x.include.zero == TRUE){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") x.log ARGUMENT SET TO ", x.log, " AND x.include.zero ARGUMENT SET TO TRUE -> x.include.zero ARGUMENT RESET TO FALSE BECAUSE 0 VALUE CANNOT BE REPRESENTED IN LOG SCALE")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+x.include.zero <- FALSE
}
-
-
-######## fun_close() #### close specific graphic windows
-
-
-# Check OK: clear to go Apollo
-fun_close <- function(kind = "pdf", return.text = FALSE){
- # AIM
- # close only specific graphic windows (devices)
- # ARGUMENTS:
- # kind: vector, among c("windows", "quartz", "x11", "X11", "pdf", "bmp", "png", "tiff"), indicating the kind of graphic windows (devices) to close. BEWARE: either "windows", "quartz", "x11" or "X11" means that all the X11 GUI graphics devices will be closed, whatever the OS used
- # return.text: print text regarding the kind parameter and the devices that were finally closed?
- # RETURN
- # text regarding the kind parameter and the devices that were finally closed
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # windows() ; windows() ; pdf() ; dev.list() ; fun_close(kind = c("pdf", "x11"), return.text = TRUE) ; dev.list()
- # DEBUGGING
- # kind = c("windows", "pdf") ; return.text = FALSE # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = kind, options = c("windows", "quartz", "x11", "X11", "pdf", "bmp", "png", "tiff"), fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = return.text, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- text <- paste0("THE REQUIRED KIND OF GRAPHIC DEVICES TO CLOSE ARE ", paste(kind, collapse = " "))
- if(Sys.info()["sysname"] == "Windows"){ # Note that .Platform$OS.type() only says "unix" for macOS and Linux and "Windows" for Windows
- if(any(kind %in% c("windows", "quartz", "x11", "X11"))){
- tempo <- kind %in% c("windows", "quartz", "x11", "X11")
- kind[tempo] <- "windows" # term are replaced by what is displayed when using a <- dev.list() ; names(a)
- }
- }else if(Sys.info()["sysname"] == "Linux"){
- if(any(kind %in% c("windows", "quartz", "x11", "X11"))){
- tempo.device <- suppressWarnings(try(X11(), silent = TRUE))[] # open a X11 window to try to recover the X11 system used
- if( ! is.null(tempo.device)){
- text <- paste0(text, "\nCANNOT CLOSE GUI GRAPHIC DEVICES AS REQUIRED BECAUSE THIS LINUX SYSTEM DOES NOT HAVE IT")
- }else{
- tempo <- kind %in% c("windows", "quartz", "x11", "X11")
- kind[tempo] <- names(dev.list()[length(dev.list())]) # term are replaced by what is displayed when using a <- dev.list() ; names(a)
- invisible(dev.off()) # close the X11 opened by tempo
- }
- }
- }else{ # for macOS
- if(any(kind %in% c("windows", "quartz", "x11", "X11"))){
- tempo <- kind %in% c("windows", "quartz", "x11", "X11")
- kind[tempo] <- "quartz" # term are replaced by what is displayed when using a <- dev.list() ; names(a)
- }
- }
- kind <- unique(kind)
- if(length(dev.list()) != 0){
- for(i in length(names(dev.list())):1){
- if(names(dev.list())[i] %in% kind){
- text <- paste0(text, "\n", names(dev.list())[i], " DEVICE NUMBER ", dev.list()[i], " HAS BEEN CLOSED")
- invisible(dev.off(dev.list()[i]))
- }
- }
- }
- if(return.text == TRUE){
- return(text)
- }
+# end x.lim management before transfo by x.log
+# y.lim management before transfo by y.log
+if(y.log != "no" & ! is.null(y.lim)){
+if(any(y.lim <= 0)){
+tempo.cat <- paste0("ERROR IN ", function.name, "\ny.lim ARGUMENT CANNOT HAVE ZERO OR NEGATIVE VALUES WITH THE y.log ARGUMENT SET TO ", y.log, ":\n", paste(y.lim, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if(any( ! is.finite(if(y.log == "log10"){log10(y.lim)}else{log2(y.lim)}))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\ny.lim ARGUMENT RETURNS INF/NA WITH THE y.log ARGUMENT SET TO ", y.log, "\nAS SCALE COMPUTATION IS ", ifelse(y.log == "log10", "log10", "log2"), ":\n", paste(if(y.log == "log10"){log10(y.lim)}else{log2(y.lim)}, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
}
-
-
-################ Standard graphics
-
-
-######## fun_empty_graph() #### text to display for empty graphs
-
-
-
-
-
-# Check OK: clear to go Apollo
-fun_empty_graph <- function(
- text = NULL,
- text.size = 1,
- title = NULL,
- title.size = 1.5
-){
- # AIM
- # display an empty plot with a text in the middle of the window (for instance to specify that no plot can be drawn)
- # ARGUMENTS
- # text: character string of the message to display
- # text.size: numeric value of the text size
- # title: character string of the graph title
- # title.size: numeric value of the title size (in points)
- # RETURN
- # an empty plot
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # simple example
- # fun_empty_graph(text = "NO GRAPH")
- # white page
- # fun_empty_graph() # white page
- # all the arguments
- # fun_empty_graph(text = "NO GRAPH", text.size = 2, title = "GRAPH1", title.size = 1)
- # DEBUGGING
- # text = "NO GRAPH" ; title = "GRAPH1" ; text.size = 1
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- if( ! is.null(text)){
- tempo <- fun_check(data = text, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = text.size, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(title)){
- tempo <- fun_check(data = title, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = title.size, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- ini.par <- par(no.readonly = TRUE) # to recover the initial graphical parameters if required (reset). BEWARE: this command alone opens a pdf of GUI window if no window already opened. But here, protected with the code because always a tempo window opened
- par(ann=FALSE, xaxt="n", yaxt="n", mar = rep(1, 4), bty = "n", xpd = NA)
- plot(1, 1, type = "n") # no display with type = "n"
- x.left.dev.region <- (par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] - ((par("usr")[2] - par("usr")[1]) / ((par("omd")[2] - par("omd")[1]) * (par("plt")[2] - par("plt")[1]))) * par("omd")[1])
- y.top.dev.region <- (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / ((par("omd")[4] - par("omd")[3]) * (par("plt")[4] - par("plt")[3]))) * (1 - par("omd")[4]))
- if( ! is.null(text)){
- text(x = 1, y = 1, labels = text, cex = text.size)
- }
- if( ! is.null(title)){
- text(x = x.left.dev.region, y = y.top.dev.region, labels = title, adj=c(0, 1), cex = title.size)
- }
- par(ini.par)
}
+if(y.log != "no" & y.include.zero == TRUE){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") y.log ARGUMENT SET TO ", y.log, " AND y.include.zero ARGUMENT SET TO TRUE -> y.include.zero ARGUMENT RESET TO FALSE BECAUSE 0 VALUE CANNOT BE REPRESENTED IN LOG SCALE")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+y.include.zero <- FALSE
+}
+# end y.lim management before transfo by y.log
+# end second round of checking and data preparation
-################ gg graphics
-######## fun_gg_palette() #### ggplot2 default color palette
+# package checking
+fun_pack(req.package = c(
+"gridExtra",
+"ggplot2",
+"lemon",
+"scales"
+), lib.path = lib.path)
+# packages Cairo and grid tested by fun_gg_point_rast()
+# end package checking
-# Check OK: clear to go Apollo
-fun_gg_palette <- function(n, kind = "std"){
- # AIM
- # provide colors used by ggplot2
- # the interest is to use another single color that is not the red one used by default
- # for ggplot2 specifications, see: https://ggplot2.tidyverse.org/articles/ggplot2-specs.html
- # ARGUMENTS
- # n: number of groups on the graph
- # kind: either "std" for standard gg colors, "dark" for darkened gg colors, or "light" for pastel gg colors
- # RETURN
- # the vector of hexadecimal colors
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # output of the function
- # fun_gg_palette(n = 2)
- # the ggplot2 palette when asking for 7 different colors
- # plot(1:7, pch = 16, cex = 5, col = fun_gg_palette(n = 7))
- # selection of the 5th color of the ggplot2 palette made of 7 different colors
- # plot(1:7, pch = 16, cex = 5, col = fun_gg_palette(n = 7)[5])
- # the ggplot2 palette made of 7 darkened colors
- # plot(1:7, pch = 16, cex = 5, col = fun_gg_palette(n = 7, kind = "dark"))
- # the ggplot2 palette made of 7 lighten colors
- # plot(1:7, pch = 16, cex = 5, col = fun_gg_palette(n = 7, kind = "light"))
- # DEBUGGING
- # n = 0
- # kind = "std"
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = n, class = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & isTRUE(all.equal(n, 0))){ # isTRUE(all.equal(n, 0))) is similar to n == 0 but deals with float
- tempo.cat <- paste0("ERROR IN ", function.name, ": n ARGUMENT MUST BE A NON ZERO INTEGER. HERE IT IS: ", paste(n, collapse = " "))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- tempo <- fun_check(data = kind, options = c("std", "dark", "light"), length = 1, fun.name = function.name) ; eval(ee)
- }
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- hues = seq(15, 375, length = n + 1)
- hcl(h = hues, l = if(kind == "std"){65}else if(kind == "dark"){35}else if(kind == "light"){85}, c = 100)[1:n]
+# main code
+# axes management
+if(is.null(x.lim)){
+if(any(unlist(mapply(FUN = "[[", data1, x, SIMPLIFY = FALSE)) %in% c(Inf, -Inf))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE x COLUMN IN data1 CONTAINS -Inf OR Inf VALUES THAT WILL NOT BE CONSIDERED IN THE PLOT RANGE")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
}
-
-
-######## fun_gg_just() #### ggplot2 justification of the axis labeling, depending on angle
-
-
-
-
-
-# Check OK: clear to go Apollo
-fun_gg_just <- function(angle, pos, kind = "axis"){
- # AIM
- # provide correct justification for text labeling, depending on the chosen angle
- # WARNINGS
- # justification behave differently on plot, depending whether it is used for annotayed text or for axis labelling. Indeed the latter has labelling constrained
- # Of note, a bug in ggplot2: vjust sometimes does not work, i.e., the same justification result is obtained whatever the value used. This is the case with angle = 90, pos = "top", kind = "axis". While everything is fine with angle = 90, pos = "bottom", kind = "axis". At least, everything seems fine for kind = "axis" and pos = c("left", "bottom")
- # ARGUMENTS
- # angle: integer value of the text angle, using the same rules as in ggplot2. Positive values for counterclockwise rotation: 0 for horizontal, 90 for vertical, 180 for upside down etc. Negative values for clockwise rotation: 0 for horizontal, -90 for vertical, -180 for upside down etc.
- # pos: where text is? Either "top", "right", "bottom" or "left" of the elements to justify from
- # kind: kind of text? Either "axis" or "text". In the first case, the pos argument refers to the axis position, and in the second to annotated text (using ggplot2::annotate() or ggplot2::geom_text())
- # RETURN
- # a list containing:
- # $angle: the submitted angle (value potentially reduced to fit the [-360 ; 360] interval, e.g., 460 -> 100, without impact on the final angle displayed)
- # $pos: the selected position (argument pos)
- # $kind: the selected kind of text (argument kind)
- # $hjust: the horizontal justification
- # $vjust: the vertical justification
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # fun_gg_just(angle = 45, pos = "bottom")
- # fun_gg_just(angle = (360*2 + 45), pos = "left")
- # output <- fun_gg_just(angle = 45, pos = "bottom") ; obs1 <- data.frame(time = 1:20, group = rep(c("CLASS_1", "CLASS_2"), times = 10), stringsAsFactors = TRUE) ; ggplot2::ggplot() + ggplot2::geom_bar(data = obs1, mapping = ggplot2::aes(x = group, y = time), stat = "identity") + ggplot2::theme(axis.text.x = ggplot2::element_text(angle = output$angle, hjust = output$hjust, vjust = output$vjust))
- # output <- fun_gg_just(angle = -45, pos = "left") ; obs1 <- data.frame(time = 1:20, group = rep(c("CLASS_1", "CLASS_2"), times = 10), stringsAsFactors = TRUE) ; ggplot2::ggplot() + ggplot2::geom_bar(data = obs1, mapping = ggplot2::aes(x = group, y = time), stat = "identity") + ggplot2::theme(axis.text.y = ggplot2::element_text(angle = output$angle, hjust = output$hjust, vjust = output$vjust)) + ggplot2::coord_flip()
- # output1 <- fun_gg_just(angle = 90, pos = "bottom") ; output2 <- fun_gg_just(angle = -45, pos = "left") ; obs1 <- data.frame(time = 1:20, group = rep(c("CLASS_1", "CLASS_2"), times = 10), stringsAsFactors = TRUE) ; ggplot2::ggplot() + ggplot2::geom_bar(data = obs1, mapping = ggplot2::aes(x = group, y = time), stat = "identity") + ggplot2::theme(axis.text.x = ggplot2::element_text(angle = output1$angle, hjust = output1$hjust, vjust = output1$vjust), axis.text.y = ggplot2::element_text(angle = output2$angle, hjust = output2$hjust, vjust = output2$vjust))
- # output <- fun_gg_just(angle = -45, pos = "left") ; obs1 <- data.frame(time = 1, km = 1, bird = "pigeon", stringsAsFactors = FALSE) ; ggplot2::ggplot(data = obs1, mapping = ggplot2::aes(x = time, y = km)) + ggplot2::geom_point() + ggplot2::geom_text(mapping = ggplot2::aes(label = bird), angle = output$angle, hjust = output$hjust, vjust = output$vjust)
- # obs1 <- data.frame(time = 1:10, km = 1:10, bird = c(NA, NA, NA, "pigeon", NA, "cat", NA, NA, NA, NA), stringsAsFactors = FALSE) ; fun_open(width = 4, height = 4) ; for(i0 in c("text", "axis")){for(i1 in c("top", "right", "bottom", "left")){for(i2 in c(0, 45, 90, 135, 180, 225, 270, 315, 360)){output <- fun_gg_just(angle = i2, pos = i1, kind = i0) ; title <- paste0("kind: ", i0, " | pos: ", i1, " | angle = ", i2, " | hjust: ", output$hjust, " | vjust: ", output$vjust) ; if(i0 == "text"){print(ggplot2::ggplot(data = obs1, mapping = ggplot2::aes(x = time, y = km)) + ggplot2::geom_point(color = fun_gg_palette(1), alpha = 0.5) + ggplot2::ggtitle(title) + ggplot2::geom_text(mapping = ggplot2::aes(label = bird), angle = output$angle, hjust = output$hjust, vjust = output$vjust) + ggplot2::theme(title = ggplot2::element_text(size = 5)))}else{print(ggplot2::ggplot(data = obs1, mapping = ggplot2::aes(x = time, y = km)) + ggplot2::geom_point(color = fun_gg_palette(1), alpha = 0.5) + ggplot2::ggtitle(title) + ggplot2::geom_text(mapping = ggplot2::aes(label = bird)) + ggplot2::scale_x_continuous(position = ifelse(i1 == "top", "top", "bottom")) + ggplot2::scale_y_continuous(position = ifelse(i1 == "right", "right", "left")) + ggplot2::theme(title = ggplot2::element_text(size = 5), axis.text.x = if(i1 %in% c("top", "bottom")){ggplot2::element_text(angle = output$angle, hjust = output$hjust, vjust = output$vjust)}, axis.text.y = if(i1 %in% c("right", "left")){ggplot2::element_text(angle = output$angle, hjust = output$hjust, vjust = output$vjust)}))}}}} ; dev.off()
- # DEBUGGING
- # angle = 45 ; pos = "left" ; kind = "axis"
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- arg.user.setting <- as.list(match.call(expand.dots = FALSE))[-1] # list of the argument settings (excluding default values not provided by the user)
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument primary checking
- # arg with no default values
- if(any(missing(angle) | missing(pos))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ARGUMENTS angle AND pos HAVE NO DEFAULT VALUE AND REQUIRE ONE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end arg with no default values
- # using fun_check()
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = angle, class = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = pos, options = c("left", "top", "right", "bottom"), length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = kind, options = c("axis", "text"), length = 1, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # end using fun_check()
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument primary checking
- # second round of checking and data preparation
- # dealing with NA arguments
- tempo.arg <- names(arg.user.setting) # values provided by the user
- tempo.log <- sapply(lapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = is.na), FUN = any) & lapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = length) == 1 # no argument provided by the user can be just NA
- if(any(tempo.log) == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ":\n", ifelse(sum(tempo.log, na.rm = TRUE) > 1, "THESE ARGUMENTS\n", "THIS ARGUMENT\n"), paste0(tempo.arg[tempo.log], collapse = "\n"),"\nCANNOT JUST BE NA")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end dealing with NA arguments
- # dealing with NULL arguments
- tempo.arg <- c(
- "angle",
- "pos",
- "kind"
- )
- tempo.log <- sapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = is.null)
- if(any(tempo.log) == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ":\n", ifelse(sum(tempo.log, na.rm = TRUE) > 1, "THESE ARGUMENTS\n", "THIS ARGUMENT\n"), paste0(tempo.arg[tempo.log], collapse = "\n"),"\nCANNOT BE NULL")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end dealing with NULL arguments
- # end second round of checking and data preparation
- # main code
- # to get angle between -360 and 360
- while(angle > 360){
- angle <- angle - 360
- }
- while(angle < -360){
- angle <- angle + 360
- }
- # end to get angle between -360 and 360
- # justifications
- if(pos %in% c("bottom", "top")){
- # code below is for if(pos == "bottom"){
- if(any(sapply(FUN = all.equal, c(-360, -180, 0, 180, 360), angle) == TRUE)){ # equivalent of angle == -360 | angle == -180 | angle == 0 | angle == 180 | angle == 360 but deals with floats
- hjust <- 0.5
- if(kind == "text"){
- if(any(sapply(FUN = all.equal, c(-360, 0, 360), angle) == TRUE)){
- vjust <- 1
- }else if(any(sapply(FUN = all.equal, c(-180, 180), angle) == TRUE)){
- vjust <- 0
- }
- }else{
- vjust <- 0.5
- }
- }else if(any(sapply(FUN = all.equal, c(-270, 90), angle) == TRUE)){
- hjust <- 1
- vjust <- 0.5
- }else if(any(sapply(FUN = all.equal, c(-90, 270), angle) == TRUE)){
- hjust <- 0
- vjust <- 0.5
- }else if((angle > -360 & angle < -270) | (angle > 0 & angle < 90)){
- hjust <- 1
- vjust <- 1
- }else if((angle > -270 & angle < -180) | (angle > 90 & angle < 180)){
- hjust <- 1
- vjust <- 0
- }else if((angle > -180 & angle < -90) | (angle > 180 & angle < 270)){
- hjust <- 0
- vjust <- 0
- if(kind == "text" & pos == "top"){
- hjust <- 1
- }
- }else if((angle > -90 & angle < 0) | (angle > 270 & angle < 360)){
- hjust <- 0
- vjust <- 1
- }
- if(pos == "top"){
- if( ! ((angle > -180 & angle < -90) | (angle > 180 & angle < 270))){
- hjust <- 1 - hjust
- }
- vjust <- 1 - vjust
- }
- }else if(pos %in% c("left", "right")){
- # code below is for if(pos == "left"){
- if(any(sapply(FUN = all.equal, c(-270, -90, 90, 270), angle) == TRUE)){ # equivalent of angle == -270 | angle == -90 | angle == 90 | angle == 270 but deals with floats
- hjust <- 0.5
- if(kind == "text"){
- if(any(sapply(FUN = all.equal, c(-90, 90), angle) == TRUE)){
- vjust <- 0
- }else if(any(sapply(FUN = all.equal, c(-270, 270), angle) == TRUE)){
- vjust <- 1
- }
- }else{
- vjust <- 0.5
- }
- }else if(any(sapply(FUN = all.equal, c(-360, 0, 360), angle) == TRUE)){
- hjust <- 1
- vjust <- 0.5
- }else if(any(sapply(FUN = all.equal, c(-180, 180), angle) == TRUE)){
- hjust <- 0
- vjust <- 0.5
- }else if((angle > -360 & angle < -270) | (angle > 0 & angle < 90)){
- hjust <- 1
- vjust <- 0
- }else if((angle > -270 & angle < -180) | (angle > 90 & angle < 180)){
- hjust <- 0
- vjust <- 0
- }else if((angle > -180 & angle < -90) | (angle > 180 & angle < 270)){
- hjust <- 0
- vjust <- 1
- }else if((angle > -90 & angle < 0) | (angle > 270 & angle < 360)){
- hjust <- 1
- vjust <- 1
- }
- if(pos == "right"){
- hjust <- 1 - hjust
- if( ! (((angle > -270 & angle < -180) | (angle > 90 & angle < 180)) | ((angle > -180 & angle < -90) | (angle > 180 & angle < 270)))){
- vjust <- 1 - vjust
- }
- }
- }
- # end justifications
- output <- list(angle = angle, pos = pos, kind = kind, hjust = hjust, vjust = vjust)
- return(output)
+x.lim <- suppressWarnings(range(unlist(mapply(FUN = "[[", data1, x, SIMPLIFY = FALSE)), na.rm = TRUE, finite = TRUE)) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only. y.lim added here. If NULL, ok if y argument has values
+}else if(x.log != "no"){
+x.lim <- get(x.log)(x.lim) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
}
-
-
-######## fun_gg_get_legend() #### get the legend of ggplot objects
-
-
-
-
-
-# Check OK: clear to go Apollo
-fun_gg_get_legend <- function(ggplot_built, fun.name = NULL, lib.path = NULL){
- # AIM
- # get legend of ggplot objects
- # # from https://stackoverflow.com/questions/12539348/ggplot-separate-legend-and-plot
- # ARGUMENTS
- # ggplot_built: a ggplot build object
- # fun.name: single character string indicating the name of the function using fun_gg_get_legend() for warning and error messages. Ignored if NULL
- # lib.path: character vector specifying the absolute pathways of the directories containing the required packages if not in the default directories. Ignored if NULL
- # RETURN
- # a list of class c("gtable", "gTree", "grob", "gDesc"), providing legend information of ggplot_built objet, or NULL if the ggplot_built object has no legend
- # REQUIRED PACKAGES
- # ggplot2
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # fun_pack()
- # EXAMPLES
- # Simple example
- # obs1 <- data.frame(time = 1:20, group = rep(c("CLASS_1", "CLASS_2"), times = 10), stringsAsFactors = TRUE) ; p <- ggplot2::ggplot() + ggplot2::geom_point(data = obs1, mapping = ggplot2::aes(x = group, y = time, fill = group)) ; fun_gg_get_legend(ggplot_built = ggplot2::ggplot_build(p))
- # Error message because no legend in the ggplot
- # obs1 <- data.frame(time = 1:20, group = rep(c("CLASS_1", "CLASS_2"), times = 10), stringsAsFactors = TRUE) ; p <- ggplot2::ggplot() + ggplot2::geom_point(data = obs1, mapping = ggplot2::aes(x = group, y = time)) ; fun_gg_get_legend(ggplot_built = ggplot2::ggplot_build(p))
- # DEBUGGING
- # obs1 <- data.frame(time = 1:20, group = rep(c("CLASS_1", "CLASS_2"), times = 10), stringsAsFactors = TRUE) ; p <- ggplot2::ggplot() + ggplot2::geom_point(data = obs1, mapping = ggplot2::aes(x = group, y = time)) ; ggplot_built = ggplot2::ggplot_build(p) ; fun.name = NULL ; lib.path = NULL
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- req.function <- c(
- "fun_check",
- "fun_pack"
- )
- for(i1 in req.function){
- if(length(find(i1, mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED ", i1, "() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # end required function checking
- # argument primary checking
- # arg with no default values
- if(any(missing(ggplot_built))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ARGUMENTS ggplot_built HAVE NO DEFAULT VALUE AND REQUIRE ONE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end arg with no default values
- # using fun_check()
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = ggplot_built, class = "ggplot_built", mode = "list", fun.name = function.name) ; eval(ee)
- if( ! is.null(fun.name)){
- tempo <- fun_check(data = fun.name, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- }
- if( ! is.null(lib.path)){
- tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- }
- if( ! is.null(arg.check)){
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- }
- # end using fun_check()
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument primary checking
- # second round of checking
- # dealing with NA
- if(any(is.na(ggplot_built)) | any(is.na(fun.name)) | any(is.na(lib.path))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": NO ARGUMENT CAN HAVE NA VALUES")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end dealing with NA
- # dealing with NULL
- if(is.null(ggplot_built)){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nggplot_built ARGUMENT CANNOT BE NULL")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end dealing with NULL
- if( ! is.null(lib.path)){
- if( ! all(dir.exists(lib.path))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # end second round of checking
- # package checking
- fun_pack(req.package = c("ggplot2"), lib.path = lib.path)
- # end package checking
- # main code
- win.nb <- dev.cur()
- pdf(file = NULL)
- tmp <- ggplot2::ggplot_gtable(ggplot_built)
- # BEWARE with ggplot_gtable : open a blanck device https://stackoverflow.com/questions/17012518/why-does-this-r-ggplot2-code-bring-up-a-blank-display-device
- invisible(dev.off())
- if(win.nb > 1){ # to go back to the previous active device, if == 1 means no opened device
- dev.set(win.nb)
- }
- leg <- which(sapply(tmp$grobs, function(x) x$name) == "guide-box")
- if(length(leg) == 0){
- legend <- NULL
- }else{
- legend <- tmp$grobs[[leg]]
- }
- return(legend)
+if(x.log != "no"){
+if(any( ! is.finite(x.lim))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nx.lim ARGUMENT CANNOT HAVE ZERO OR NEGATIVE VALUES WITH THE x.log ARGUMENT SET TO ", x.log, ":\n", paste(x.lim, collapse = " "), "\nPLEASE, CHECK DATA VALUES (PRESENCE OF ZERO OR INF VALUES)")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
}
-
-
-######## fun_gg_point_rast() #### ggplot2 raster scatterplot layer
-
-
-
-
-
-# Check OK: clear to go Apollo
-fun_gg_point_rast <- function(
- data = NULL,
- mapping = NULL,
- stat = "identity",
- position = "identity",
- ...,
- na.rm = FALSE,
- show.legend = NA,
- inherit.aes = TRUE,
- raster.width = NULL,
- raster.height = NULL,
- raster.dpi = 300,
- inactivate = TRUE,
- lib.path = NULL
-){
- # AIM
- # equivalent to ggplot2::geom_point() but in raster mode
- # use it like ggplot2::geom_point() with the main raster.dpi additional argument
- # WARNINGS
- # can be long to generate the plot
- # use a square plot region. Otherwise, the dots will have ellipsoid shape
- # solve the transparency problems with some GUI
- # this function is derived from the geom_point_rast() function, created by Viktor Petukhov , and present in the ggrastr package (https://rdrr.io/github/VPetukhov/ggrastr/src/R/geom-point-rast.R, MIT License, Copyright (c) 2017 Viktor Petukhov). Has been placed here to minimize package dependencies
- # ARGUMENTS
- # classical arguments of geom_point(), shown here https://rdrr.io/github/VPetukhov/ggrastr/man/geom_point_rast.html
- # raster.width : width of the result image (in inches). Default: deterined by the current device parameters
- # raster.height: height of the result image (in inches). Default: deterined by the current device parameters
- # raster.dpi: resolution of the result image
- # inactivate: logical. Inactivate the fun.name argument of the fun_check() function? If TRUE, the name of the fun_check() function in error messages coming from this function. Use TRUE if fun_gg_point_rast() is used like this: eval(parse(text = "fun_gg_point_rast"))
- # lib.path: character vector specifying the absolute pathways of the directories containing the required packages if not in the default directories. Ignored if NULL
- # RETURN
- # a raster scatter plot
- # REQUIRED PACKAGES
- # ggplot2
- # grid (included in the R installation packages but not automatically loaded)
- # Cairo
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # fun_pack()
- # EXAMPLES
- # Two pdf in the current directory
- # set.seed(1) ; data1 = data.frame(x = rnorm(100000), y = rnorm(10000), stringsAsFactors = TRUE) ; fun_open(pdf.name = "Raster") ; ggplot2::ggplot() + fun_gg_point_rast(data = data1, mapping = ggplot2::aes(x = x, y = y)) ; fun_open(pdf.name = "Vectorial") ; ggplot2::ggplot() + ggplot2::geom_point(data = data1, mapping = ggplot2::aes(x = x, y = y)) ; dev.off() ; dev.off()
- # DEBUGGING
- #
- # function name
- if(all(inactivate == FALSE)){ # inactivate has to be used here but will be fully checked below
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- }else if(all(inactivate == TRUE)){
- function.name <- NULL
- }else{
- tempo.cat <- paste0("ERROR IN fun_gg_point_rast(): CODE INCONSISTENCY 1")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(length(utils::find("fun_pack", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_pack() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- if( ! is.null(data)){
- tempo <- fun_check(data = data, class = "data.frame", na.contain = TRUE, fun.name = function.name) ; eval(ee)
- }
- if( ! is.null(mapping)){
- tempo <- fun_check(data = mapping, class = "uneval", typeof = "list", fun.name = function.name) ; eval(ee) # aes() is tested
- }
- # stat and position not tested because too complicate
- tempo <- fun_check(data = na.rm, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = show.legend, class = "vector", mode = "logical", length = 1, na.contain = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = inherit.aes, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(raster.width)){
- tempo <- fun_check(data = raster.width, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- }
- if( ! is.null(raster.height)){
- tempo <- fun_check(data = raster.height, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = raster.dpi, class = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = inactivate, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(lib.path)){
- tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
- tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- }
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # package checking
- fun_pack(req.package = c("ggplot2"), lib.path = lib.path)
- fun_pack(req.package = c("grid"), lib.path = lib.path)
- fun_pack(req.package = c("Cairo"), lib.path = lib.path)
- # end package checking
- # additional functions
- DrawGeomPointRast <- function(data, panel_params, coord, na.rm = FALSE, raster.width = NULL, raster.height= NULL, raster.dpi = 300){
- if (is.null(raster.width)){
- raster.width <- par('fin')[1]
- }
- if (is.null(raster.height)){
- raster.height <- par('fin')[2]
- }
- prev_dev_id <- dev.cur()
- p <- ggplot2::GeomPoint$draw_panel(data, panel_params, coord)
- dev_id <- Cairo::Cairo(type='raster', width = raster.width*raster.dpi, height = raster.height*raster.dpi, dpi = raster.dpi, units = 'px', bg = "transparent")[1]
- grid::pushViewport(grid::viewport(width = 1, height = 1))
- grid::grid.points(x = p$x, y = p$y, pch = p$pch, size = p$size,
- name = p$name, gp = p$gp, vp = p$vp, draw = T)
- grid::popViewport()
- cap <- grid::grid.cap()
- invisible(dev.off(dev_id))
- invisible(dev.set(prev_dev_id))
- grid::rasterGrob(cap, x = 0, y = 0, width = 1, height = 1, default.units = "native", just = c("left","bottom"))
- }
- # end additional functions
- # main code
- GeomPointRast <- ggplot2::ggproto("GeomPointRast", ggplot2::GeomPoint, draw_panel = DrawGeomPointRast)
- ggplot2::layer(
- data = data,
- mapping = mapping,
- stat = stat,
- geom = GeomPointRast,
- position = position,
- show.legend = show.legend,
- inherit.aes = inherit.aes,
- params = list(
- na.rm = na.rm,
- raster.width = raster.width,
- raster.height = raster.height,
- raster.dpi = raster.dpi,
- ...
- )
- )
- # end main code
}
-
-
-######## fun_gg_boxplot() #### ggplot2 boxplot + background dots if required
-
-
-
-
-######## fun_gg_scatter() #### ggplot2 scatterplot + lines (up to 6 overlays totally)
-
-
-
-
-######## fun_gg_heatmap() #### ggplot2 heatmap + overlaid mask if required
-
-
-#test plot.margin = margin(up.space.mds, right.space.mds, down.space.mds, left.space.mds, "inches") to set the dim of the region plot ?
-# if matrix is full of zero (or same value I guess), heatmap is complicate. Test it and error message
-
-# Check OK: clear to go Apollo
-fun_gg_heatmap <- function(
- data1,
- legend.name1 = "",
- low.color1 = "blue",
- mid.color1 = "white",
- high.color1 = "red",
- limit1 = NULL,
- midpoint1 = NULL,
- data2 = NULL,
- color2 = "black",
- alpha2 = 0.5,
- invert2 = FALSE,
- text.size = 12,
- title = "",
- title.text.size = 12,
- show.scale = TRUE,
- rotate = FALSE,
- return = FALSE,
- plot = TRUE,
- add = NULL,
- warn.print = FALSE,
- lib.path = NULL
-){
- # AIM
- # ggplot2 heatmap with the possibility to overlay a mask
- # see also:
- # draw : http://www.sthda.com/english/wiki/ggplot2-quick-correlation-matrix-heatmap-r-software-and-data-visualization
- # same range scale : https://stackoverflow.com/questions/44655723/r-ggplot2-heatmap-fixed-scale-color-between-graphs
- # for ggplot2 specifications, see: https://ggplot2.tidyverse.org/articles/ggplot2-specs.html
- # ARGUMENTS
- # data1: numeric matrix or data frame resulting from the conversion of the numeric matrix by reshape2::melt()
- # legend.name1: character string of the data1 heatmap scale legend
- # low.color1: character string of the color (i.e., "blue" or "#0000FF") of the lowest scale value
- # mid.color1: same as low.color1 but for the middle scale value. If NULL, the middle color is the default color between low.color1 and high.color1. BEWARE: argument midpoint1 is not ignored, even if mid.color1 is NULL, meaning that the default mid color can still be controled
- # high.color1: same as low.color1 but for the highest scale value
- # limit1: 2 numeric values defining the lowest and higest color scale values. If NULL, take the range of data1 values
- # midpoint1: single numeric value defining the value corresponding to the mid.color1 argument. A warning message is returned if midpoint1 does not correspond to the mean of limit1 values, because the color scale is not linear anymore. If NULL, takes the mean of limit1 values. Mean of data1, instead of mean of limit1, can be used here if required
- # data2: binary mask matrix (made of 0 and 1) of same dimension as data1 or a data frame resulting from the conversion of the binary mask matrix by reshape2::melt(). Value 1 of data2 will correspond to color2 argument (value 0 will be NA color), and the opposite if invert2 argument is TRUE (inverted mask)
- # color2: color of the 1 values of the binary mask matrix. The 0 values will be color NA
- # alpha2: numeric value (from 0 to 1) of the mask transparency
- # invert2: logical. Invert the mask (1 -> 0 and 0 -> 1)?
- # text.size: numeric value of the size of the texts in scale
- # title: character string of the graph title
- # title.text.size: numeric value of the title size (in points)
- # show.scale: logical. Show color scale?
- # rotate: logical. Rotate the heatmap 90° clockwise?
- # return: logical. Return the graph parameters?
- # plot: logical. Plot the graphic? If FALSE and return argument is TRUE, graphical parameters and associated warnings are provided without plotting
- # add: character string allowing to add more ggplot2 features (dots, lines, themes, etc.). BEWARE: (1) must start with "+" just after the simple or double opening quote (no space, end of line, carriage return, etc., allowed), (2) must finish with ")" just before the simple or double closing quote (no space, end of line, carriage return, etc., allowed) and (3) each function must be preceded by "ggplot2::" (for instance: "ggplot2::coord_flip()). If the character string contains the "ggplot2::theme" string, then internal ggplot2 theme() and theme_classic() functions will be inactivated to be reused by add. BEWARE: handle this argument with caution since added functions can create conflicts with the preexisting internal ggplot2 functions
- # warn.print: logical. Print warnings at the end of the execution? No print if no warning messages
- # lib.path: absolute path of the required packages, if not in the default folders
- # RETURN
- # a heatmap if plot argument is TRUE
- # a list of the graph info if return argument is TRUE:
- # $data: a list of the graphic info
- # $axes: a list of the axes info
- # $scale: the scale info (lowest, mid and highest values)
- # $warn: the warning messages. Use cat() for proper display. NULL if no warning
- # REQUIRED PACKAGES
- # ggplot2
- # reshape2
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # fun_pack()
- # fun_round()
- # EXAMPLES
- # fun_gg_heatmap(data1 = matrix(1:16, ncol = 4), title = "GRAPH 1")
- # fun_gg_heatmap(data1 = matrix(1:16, ncol = 4), return = TRUE)
- # fun_gg_heatmap(data1 = matrix(1:16, ncol = 4), legend.name1 = "VALUE", title = "GRAPH 1", text.size = 5, data2 = matrix(rep(c(1,0,0,0), 4), ncol = 4), invert2 = FALSE, return = TRUE)
- # diagonal matrix
- # fun_gg_heatmap(data1 = matrix(c(1,0,0,0,0,1,0,0,0,0,1,0,0,0,0,1), ncol = 4))
- # fun_gg_heatmap(data1 = reshape2::melt(matrix(c(1,0,0,0,0,1,0,0,0,0,1,0,0,0,0,1), ncol = 4)))
- # error message
- # fun_gg_heatmap(data1 = matrix(1:16, ncol = 4), data2 = matrix(rep(c(1,0,0,0), 5), ncol = 5))
- # fun_gg_heatmap(data1 = matrix(1:16, ncol = 4), data2 = reshape2::melt(matrix(rep(c(1,0,0,0), 4), ncol = 4)))
- # fun_gg_heatmap(data1 = reshape2::melt(matrix(1:16, ncol = 4)), data2 = reshape2::melt(matrix(rep(c(1,0,0,0), 4), ncol = 4)))
- # DEBUGGING
- # data1 = matrix(1:16, ncol = 4) ; legend.name1 = "" ; low.color1 = "blue" ; mid.color1 = "white" ; high.color1 = "red" ; limit1 = NULL ; midpoint1 = NULL ; data2 = matrix(rep(c(1,0,0,0), 4), ncol = 4) ; color2 = "black" ; alpha2 = 0.5 ; invert2 = FALSE ; text.size = 12 ; title = "" ; title.text.size = 12 ; show.scale = TRUE ; rotate = FALSE ; return = FALSE ; plot = TRUE ; add = NULL ; warn.print = TRUE ; lib.path = NULL
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(length(utils::find("fun_pack", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_pack() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(length(utils::find("fun_round", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_round() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # no reserved words required for this function
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- if(all(is.matrix(data1))){
- tempo <- fun_check(data = data1, class = "matrix", mode = "numeric", na.contain = TRUE, fun.name = function.name) ; eval(ee)
- }else if(all(is.data.frame(data1))){
- tempo <- fun_check(data = data1, class = "data.frame", length = 3, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- # structure of reshape2::melt() data frame
- tempo <- fun_check(data = data1[, 1], data.name = "COLUMN 1 OF data1 (reshape2::melt() DATA FRAME)", typeof = "integer", fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = data1[, 2], data.name = "COLUMN 2 OF data1 (reshape2::melt() DATA FRAME)", typeof = "integer", fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = data1[, 3], data.name = "COLUMN 3 OF data1 (reshape2::melt() DATA FRAME)", mode = "numeric", na.contain = TRUE, fun.name = function.name) ; eval(ee)
- }
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 ARGUMENT MUST BE A NUMERIC MATRIX OR A DATA FRAME OUTPUT OF THE reshape::melt() FUNCTION")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- tempo <- fun_check(data = legend.name1, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = low.color1, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & ! (all(low.color1 %in% colors() | grepl(pattern = "^#", low.color1)))){ # check that all strings of low.color1 start by #
- tempo.cat <- paste0("ERROR IN ", function.name, ": low.color1 ARGUMENT MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # AND/OR COLOR NAMES GIVEN BY colors()")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- if( ! is.null(mid.color1)){
- tempo <- fun_check(data = mid.color1, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & ! (all(mid.color1 %in% colors() | grepl(pattern = "^#", mid.color1)))){ # check that all strings of mid.color1 start by #
- tempo.cat <- paste0("ERROR IN ", function.name, ": mid.color1 ARGUMENT MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # AND/OR COLOR NAMES GIVEN BY colors()")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- tempo <- fun_check(data = high.color1, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & ! (all(high.color1 %in% colors() | grepl(pattern = "^#", high.color1)))){ # check that all strings of high.color1 start by #
- tempo.cat <- paste0("ERROR IN ", function.name, ": high.color1 ARGUMENT MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # AND/OR COLOR NAMES GIVEN BY colors()")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- if( ! is.null(limit1)){
- tempo <- fun_check(data = limit1, class = "vector", mode = "numeric", length = 2, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & any(limit1 %in% c(Inf, -Inf))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": limit1 ARGUMENT CANNOT CONTAIN -Inf OR Inf VALUES")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- if( ! is.null(midpoint1)){
- tempo <- fun_check(data = midpoint1, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
- }
- if( ! is.null(data2)){
- if(all(is.matrix(data2))){
- tempo <- fun_check(data = data2, class = "matrix", mode = "numeric", fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & ! all(unique(data2) %in% c(0,1))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": MATRIX IN data2 MUST BE MADE OF 0 AND 1 ONLY (MASK MATRIX)")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }else if(tempo$problem == FALSE & all(is.matrix(data1)) & ! identical(dim(data1), dim(data2))){ # matrix and matrix
- tempo.cat <- paste0("ERROR IN ", function.name, ": MATRIX DIMENSION IN data2 MUST BE IDENTICAL AS MATRIX DIMENSION IN data1. HERE IT IS RESPECTIVELY:\n", paste(dim(data2), collapse = " "), "\n", paste(dim(data1), collapse = " "))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }else if(tempo$problem == FALSE & all(is.data.frame(data1)) & nrow(data1) != prod(dim(data2))){ # reshape2 and matrix
- tempo.cat <- paste0("ERROR IN ", function.name, ": DATA FRAME IN data2 MUST HAVE ROW NUMBER EQUAL TO PRODUCT OF DIMENSIONS OF data1 MATRIX. HERE IT IS RESPECTIVELY:\n", paste(nrow(data1), collapse = " "), "\n", paste(prod(dim(data2)), collapse = " "))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }else if(all(is.data.frame(data2))){
- tempo <- fun_check(data = data2, class = "data.frame", length = 3, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- # structure of reshape2::melt() data frame
- tempo <- fun_check(data = data2[, 1], data.name = "COLUMN 1 OF data2 (reshape2::melt() DATA FRAME)", typeof = "integer", fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = data2[, 2], data.name = "COLUMN 2 OF data2 (reshape2::melt() DATA FRAME)", typeof = "integer", fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = data2[, 3], data.name = "COLUMN 3 OF data2 (reshape2::melt() DATA FRAME)", mode = "numeric", fun.name = function.name) ; eval(ee)
- }
- if(tempo$problem == FALSE & ! all(unique(data2[, 3]) %in% c(0,1))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": THIRD COLUMN OF DATA FRAME IN data2 MUST BE MADE OF 0 AND 1 ONLY (MASK DATA FRAME)")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }else if(tempo$problem == FALSE & all(is.data.frame(data1)) & ! identical(dim(data1), dim(data2))){ # data frame and data frame
- tempo.cat <- paste0("ERROR IN ", function.name, ": DATA FRAME DIMENSION IN data2 MUST BE IDENTICAL TO DATA FRAME DIMENSION IN data1. HERE IT IS RESPECTIVELY:\n", paste(dim(data2), collapse = " "), "\n", paste(dim(data1), collapse = " "))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }else if(tempo$problem == FALSE & all(is.matrix(data1)) & nrow(data2) != prod(dim(data1))){ # reshape2 and matrix
- tempo.cat <- paste0("ERROR IN ", function.name, ": DATA FRAME IN data2 MUST HAVE ROW NUMBER EQUAL TO PRODUCT OF DIMENSION OF data1 MATRIX. HERE IT IS RESPECTIVELY:\n", paste(nrow(data2), collapse = " "), "\n", paste(prod(dim(data1)), collapse = " "))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE data2 ARGUMENT MUST BE A NUMERIC MATRIX OR A DATA FRAME OUTPUT OF THE reshape::melt() FUNCTION")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- tempo <- fun_check(data = color2, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & ! (all(color2 %in% colors() | grepl(pattern = "^#", color2)))){ # check that all strings of color2 start by #
- tempo.cat <- paste0("ERROR IN ", function.name, ": color2 ARGUMENT MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # AND/OR COLOR NAMES GIVEN BY colors()")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- tempo <- fun_check(data = alpha2, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = invert2, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = text.size, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = title, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = title.text.size, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = show.scale, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = return, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = plot, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(add)){
- tempo <- fun_check(data = add, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & ! grepl(pattern = "^\\+", add)){ # check that the add string start by +
- tempo.cat <- paste0("ERROR IN ", function.name, ": add ARGUMENT MUST START WITH \"+\": ", paste(unique(add), collapse = " "))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }else if(tempo$problem == FALSE & ! grepl(pattern = "ggplot2::", add)){ #
- tempo.cat <- paste0("ERROR IN ", function.name, ": add ARGUMENT MUST CONTAIN \"ggplot2::\" IN FRONT OF EACH GGPLOT2 FUNCTION: ", paste(unique(add), collapse = " "))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }else if(tempo$problem == FALSE & ! grepl(pattern = ")$", add)){ # check that the add string finished by )
- tempo.cat <- paste0("ERROR IN ", function.name, ": add ARGUMENT MUST FINISH BY \")\": ", paste(unique(add), collapse = " "))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- tempo <- fun_check(data = warn.print, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(lib.path)){
- tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
- tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- }
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # package checking
- fun_pack(req.package = c("reshape2", "ggplot2"), lib.path = lib.path)
- # end package checking
- # main code
- ini.warning.length <- options()$warning.length
- warn <- NULL
- warn.count <- 0
- if(all(is.matrix(data1))){
- data1 <- reshape2::melt(data1) # transform a matrix into a data frame with 2 coordinates columns and the third intensity column
- }
- if(rotate == TRUE){
- data1[, 1] <- rev(data1[, 1])
- }
- if(is.null(limit1)){
- if(any(data1[, 3] %in% c(Inf, -Inf))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE data1 ARGUMENT CONTAINS -Inf OR Inf VALUES IN THE THIRD COLUMN, THAT WILL NOT BE CONSIDERED IN THE PLOT RANGE")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- limit1 <- range(data1[, 3], na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE limit1 ARGUMENT IS NULL -> RANGE OF data1 ARGUMENT HAS BEEN TAKEN: ", paste(fun_round(limit1), collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- if(suppressWarnings(any(limit1 %in% c(Inf, -Inf)))){
- tempo.cat <- paste0("ERROR IN ", function.name, " COMPUTED LIMIT CONTAINS Inf VALUES, BECAUSE VALUES FROM data1 ARGUMENTS ARE NA OR Inf ONLY")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- if(is.null(midpoint1)){
- midpoint1 <- mean(limit1, na.rm = TRUE)
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE midpoint1 ARGUMENT IS NULL -> MEAN OF limit1 ARGUMENT HAS BEEN TAKEN: ", paste(fun_round(midpoint1), collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }else if(fun_round(midpoint1, 9) != fun_round(mean(limit1), 9)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE midpoint1 ARGUMENT (", fun_round(mean(midpoint1), 9), ") DOES NOT CORRESPOND TO THE MEAN OF THE limit1 ARGUMENT (", fun_round(mean(limit1), 9), "). COLOR SCALE IS NOT LINEAR")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- if( ! is.null(data2)){
- if(all(is.matrix(data2))){
- data2 <- reshape2::melt(data2) # transform a matrix into a data frame with 2 coordinates columns and the third intensity column
- }
- if(rotate == TRUE){
- data2[, 1] <- rev(data2[, 1])
- }
- data2[, 3] <- factor(data2[, 3]) # to converte continuous scale into discrete scale
- }
- tempo.gg.name <- "gg.indiv.plot."
- tempo.gg.count <- 0 # to facilitate debugging
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::ggplot())
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_raster(data = data1, mapping = ggplot2::aes_string(x = names(data1)[ifelse(rotate == FALSE, 2, 1)], y = names(data1)[ifelse(rotate == FALSE, 1, 2)], fill = names(data1)[3]), show.legend = show.scale)) # show.legend option do not remove the legend, only the aesthetic of the legend (dot, line, etc.)
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_fill_gradient2(low = low.color1, high = high.color1, mid = mid.color1, midpoint = midpoint1, limit = limit1, breaks = c(limit1[1], midpoint1, limit1[2]), labels = fun_round(c(limit1[1], midpoint1, limit1[2])), name = legend.name1))
- if( ! is.null(data2)){
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_raster(data = data2, mapping = ggplot2::aes_string(x = names(data2)[ifelse(rotate == FALSE, 2, 1)], y = names(data2)[ifelse(rotate == FALSE, 1, 2)], alpha = names(data2)[3]), fill = color2, show.legend = FALSE))
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "alpha", values = if(invert2 == FALSE){c(0, alpha2)}else{c(alpha2, 0)}, guide = FALSE))
- # assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_raster(data = data2, mapping = ggplot2::aes_string(x = names(data2)[ifelse(rotate == FALSE, 2, 1)], y = names(data2)[ifelse(rotate == FALSE, 1, 2)], group = names(data2)[3]), fill = data2[, 3], alpha = alpha2, show.legend = FALSE)) # BEWARE: this does not work if NA present, because geom_raster() has a tendency to complete empty spaces, and thus, behave differently than geom_tile(). See https://github.com/tidyverse/ggplot2/issues/3025
- }
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::coord_fixed()) # x = y
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_y_reverse())
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::ggtitle(title))
- add.check <- TRUE
- if( ! is.null(add)){ # if add is NULL, then = 0
- if(grepl(pattern = "ggplot2::theme", add) == TRUE){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") \"ggplot2::theme\" STRING DETECTED IN THE add ARGUMENT -> INTERNAL GGPLOT2 THEME FUNCTIONS theme() AND theme_classic() HAVE BEEN INACTIVATED, TO BE USED BY THE USER")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- add.check <- FALSE
- }
- }
- if(add.check == TRUE){
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::theme_classic(base_size = text.size))
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::theme(
- text = ggplot2::element_text(size = text.size),
- plot.title = ggplot2::element_text(size = title.text.size), # stronger than text
- line = ggplot2::element_blank(),
- axis.title = ggplot2::element_blank(),
- axis.text = ggplot2::element_blank(),
- axis.ticks = ggplot2::element_blank(),
- panel.background = ggplot2::element_blank()
- ))
- }
- if(plot == TRUE){
- # suppressWarnings(
- print(eval(parse(text = paste(paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "), if(is.null(add)){NULL}else{add}))))
- # )
- }else{
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") PLOT NOT SHOWN AS REQUESTED")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- if(warn.print == TRUE & ! is.null(warn)){
- options(warning.length = 8170)
- on.exit(warning(paste0("FROM ", function.name, ":\n\n", warn), call. = FALSE))
- on.exit(exp = options(warning.length = ini.warning.length), add = TRUE)
- }
- if(return == TRUE){
- output <- ggplot2::ggplot_build(eval(parse(text = paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "))))
- output <- output$data
- names(output)[1] <- "heatmap"
- if( ! is.null(data2)){
- names(output)[2] <- "mask"
- }
- return(list(data = output, axes = output$layout$panel_params[[1]], scale = c(limit1[1], midpoint1, limit1[2]), warn = warn))
- }
+if(suppressWarnings(all(x.lim %in% c(Inf, -Inf)))){ # happen when x is only NULL
+if(all(unlist(geom) %in% c("geom_vline", "geom_stick"))){
+tempo.cat <- paste0("ERROR IN ", function.name, " NOT POSSIBLE TO DRAW geom_vline OR geom_stick KIND OF LINES ALONE IF x.lim ARGUMENT IS SET TO NULL, SINCE NO X-AXIS DEFINED (", ifelse(length(x) == 1, "x", paste0("ELEMENT ", i1, " OF x")), " ARGUMENT MUST BE NULL FOR THESE KIND OF LINES)")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, " x.lim ARGUMENT MADE OF NA, -Inf OR Inf ONLY: ", paste(x.lim, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
}
-
-
-######## fun_gg_empty_graph() #### text to display for empty graphs
-
-
-
-
-
-# Check OK: clear to go Apollo
-fun_gg_empty_graph <- function(
- text = NULL,
- text.size = 12,
- title = NULL,
- title.size = 8,
- lib.path = NULL
-){
- # AIM
- # display an empty ggplot2 plot with a text in the middle of the window (for instance to specify that no plot can be drawn)
- # ARGUMENTS
- # text: character string of the message to display
- # text.size: numeric value of the text size (in points)
- # title: character string of the graph title
- # title.size: numeric value of the title size (in points)
- # lib.path: character vector specifying the absolute pathways of the directories containing the required packages if not in the default directories. Ignored if NULL
- # RETURN
- # an empty plot
- # REQUIRED PACKAGES
- # ggplot2
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # fun_pack()
- # EXAMPLES
- ### simple example
- # fun_gg_empty_graph(text = "NO GRAPH")
- ### white page
- # fun_gg_empty_graph()
- ### all the arguments
- # fun_gg_empty_graph(text = "NO GRAPH", text.size = 8, title = "GRAPH1", title.size = 10, lib.path = NULL)
- # DEBUGGING
- # text = "NO GRAPH" ; text.size = 12 ; title = "GRAPH1" ; title.size = 8 ; lib.path = NULL
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(length(utils::find("fun_pack", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_pack() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- if( ! is.null(text)){
- tempo <- fun_check(data = text, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = text.size, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(title)){
- tempo <- fun_check(data = title, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = title.size, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # package checking
- fun_pack(req.package = c("ggplot2"), lib.path = lib.path)
- # end package checking
- # main code
- tempo.gg.name <- "gg.indiv.plot."
- tempo.gg.count <- 0
- # no need loop part
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::ggplot())
- if( ! is.null(text)){
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_text(data = data.frame(x = 1, y = 1, stringsAsFactors = TRUE), ggplot2::aes(x = x, y = y, label = text), size = text.size))
- }
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::ggtitle(title))
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::theme_void())
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), m.gg <- ggplot2::theme(
- plot.title = ggplot2::element_text(size = title.size) # stronger than text
- ))
- suppressWarnings(print(eval(parse(text = paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + ")))))
}
-
-
-################ Graphic extraction
-
-
-######## fun_trim() #### display values from a quantitative variable and trim according to defined cut-offs
-
-
-# Check OK: clear to go Apollo
-fun_trim <- function(
- data,
- displayed.nb = NULL,
- single.value.display = FALSE,
- trim.method = "",
- trim.cutoffs = c(0.05,
- 0.975),
- interval.scale.disp = TRUE,
- down.space = 0.75,
- left.space = 0.75,
- up.space = 0.3,
- right.space = 0.25,
- orient = 1,
- dist.legend = 0.37,
- box.type = "l",
- amplif.label = 1.25,
- amplif.axis = 1.25,
- std.x.range = TRUE,
- std.y.range = TRUE,
- cex.pt = 0.2,
- col.box = hsv(0.55,
- 0.8,
- 0.8),
- x.nb.inter.tick = 4,
- y.nb.inter.tick = 0,
- tick.length = 1,
- sec.tick.length = 0.75,
- corner.text = "",
- amplif.legend = 1,
- corner.text.size = 0.75,
- trim.return = FALSE
-){
- # AIM
- # trim and display values from a numeric vector or matrix
- # plot 4 graphs: stripchart of values, stripchart of rank of values, histogram and normal QQPlot
- # different kinds of intervals are displayed on the top of graphes to facilitate the analysis of the variable and a trimming setting
- # the trimming interval chosen is displayed on top of graphs
- # both trimmed and not trimmed values are returned in a list
- # ARGUMENTS
- # data: values to plot (either a numeric vector or a numeric matrix)
- # displayed.nb: number of values displayed. If NULL, all the values are displayed. Otherwise, if the number of values is over displayed.nb, then displayed.nb values are displayed after random selection
- # single.value.display: provide the 4 graphs if data is made of a single (potentially repeated value)? If FALSE, an empty graph is displayed if data is made of a single (potentially repeated value). And the return list is made of NULL compartments
- # trim.method: Write "" if not required. write "mean.sd" if mean +/- sd has to be displayed as a trimming interval (only recommanded for normal distribution). Write "quantile" to display a trimming interval based on quantile cut-offs. No other possibility allowed. See trim.cutoffs below
- # trim.cutoffs: 2 values cutoff for the trimming interval displayed, each value between 0 and 1. Not used if trim.method == "".The couple of values c(lower, upper) represents the lower and upper boundaries of the trimming interval (in proportion), which represent the interval of distribution kept (between 0 and 1). Example: trim.cutoffs = c(0.05, 0.975). What is strictly kept for the display is ]lower , upper[, boundaries excluded. Using the "mean.sd" method, 0.025 and 0.975 represent 95% CI which is mean +/- 1.96 * sd
- # interval.scale.disp: display sd and quantiles intervals on top of graphs ?
- # down.space: lower vertical margin (in inches, mai argument of par())
- # left.space: left horizontal margin (in inches, mai argument of par())
- # up.space: upper vertical margin between plot region and grapical window (in inches, mai argument of par())
- # right.space: right horizontal margin (in inches, mai argument of par())
- # orient: scale number orientation (las argument of par()). 0, always parallel to the axis; 1, always horizontal; 2, always perpendicular to the axis; 3, always vertical
- # dist.legend: numeric value that moves axis legends away in inches (first number of mgp argument of par() but in inches thus / 0.2)
- # box.type: bty argument of par(). Either "o", "l", "7", "c", "u", "]", the resulting box resembles the corresponding upper case letter. A value of "n" suppresses the box
- # amplif.label: increase or decrease the size of the text in legends
- # amplif.axis: increase or decrease the size of the scale numbers in axis
- # std.x.range: standard range on the x-axis? TRUE (no range extend) or FALSE (4% range extend). Controls xaxs argument of par() (TRUE is xaxs = "i", FALSE is xaxs = "r")
- # std.y.range: standard range on the y-axis? TRUE (no range extend) or FALSE (4% range extend). Controls yaxs argument of par() (TRUE is yaxs = "i", FALSE is yaxs = "r")
- # cex.pt: size of points in stripcharts (in inches, thus cex.pt will be thereafter / 0.2)
- # col.box: color of boxplot
- # x.nb.inter.tick: number of secondary ticks between main ticks on x-axis (only if not log scale). Zero means non secondary ticks
- # y.nb.inter.tick: number of secondary ticks between main ticks on y-axis (only if not log scale). Zero means non secondary ticks
- # tick.length: length of the ticks (1 means complete the distance between the plot region and the axis numbers, 0.5 means half the length, etc. 0 means no tick
- # sec.tick.length: length of the secondary ticks (1 means complete the distance between the plot region and the axis numbers, 0.5 means half the length, etc., 0 for no ticks)
- # corner.text: text to add at the top right corner of the window
- # amplif.legend: increase or decrease the size of the text of legend
- # corner.text.size: positive numeric. Increase or decrease the size of the text. Value 1 does not change it, 0.5 decreases by half, 2 increases by 2
- # trim.return: return the trimmed and non trimmed values? NULL returned for trimmed and non trimmed values if trim.method == ""
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # RETURN
- # a list containing:
- # $trim.method: correspond to trim.method above
- # $trim.cutoffs: correspond to trim.cutoffs above
- # $real.trim.cutoffs: the two boundary values (in the unit of the numeric vector or numeric matrix analyzed). NULL
- # $trimmed.values: the values outside of the trimming interval as defined in trim.cutoffs above
- # $kept.values: the values inside the trimming interval as defined in trim.cutoffs above
- # EXAMPLES
- # fun_trim(data = c(1:100, 1:10), displayed.nb = NULL, single.value.display = FALSE, trim.method = "mean.sd", trim.cutoffs = c(0.05, 0.975), interval.scale.disp = TRUE, down.space = 0.75, left.space = 0.75, up.space = 0.3, right.space = 0.25, orient = 1, dist.legend = 0.37, box.type = "l", amplif.label = 1.25, amplif.axis = 1.25, std.x.range = TRUE, std.y.range = TRUE, cex.pt = 0.2, col.box = hsv(0.55, 0.8, 0.8), x.nb.inter.tick = 4, y.nb.inter.tick = 0, tick.length = 0.5, sec.tick.length = 0.3, corner.text = "", amplif.legend = 1, corner.text.size = 0.75, trim.return = TRUE)
- # DEBUGGING
- # data = c(1:100, 1:10) ; displayed.nb = NULL ; single.value.display = FALSE ; trim.method = "quantile" ; trim.cutoffs = c(0.05, 0.975) ; interval.scale.disp = TRUE ; down.space = 1 ; left.space = 1 ; up.space = 0.5 ; right.space = 0.25 ; orient = 1 ; dist.legend = 0.5 ; box.type = "l" ; amplif.label = 1 ; amplif.axis = 1 ; std.x.range = TRUE ; std.y.range = TRUE ; cex.pt = 0.1 ; col.box = hsv(0.55, 0.8, 0.8) ; x.nb.inter.tick = 4 ; y.nb.inter.tick = 0 ; tick.length = 0.5 ; sec.tick.length = 0.3 ; corner.text = "" ; amplif.legend = 1 ; corner.text.size = 0.75 ; trim.return = TRUE # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- # argument checking without fun_check()
- if( ! (all(class(data) == "numeric") | all(class(data) == "integer") | (all(class(data) %in% c("matrix", "array")) & mode(data) == "numeric"))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": data ARGUMENT MUST BE A NUMERIC VECTOR OR NUMERIC MATRIX")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end argument checking without fun_check()
- # argument checking with fun_check()
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- if( ! is.null(displayed.nb)){
- tempo <- fun_check(data = displayed.nb, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
- if(displayed.nb < 2){
- tempo.cat <- paste0("ERROR IN ", function.name, ": displayed.nb ARGUMENT MUST BE A SINGLE INTEGER VALUE GREATER THAN 1 AND NOT: ", paste(displayed.nb, collapse = " "))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- tempo <- fun_check(data = single.value.display, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = trim.method, options = c("", "mean.sd", "quantile"), length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = trim.cutoffs, class = "vector", mode = "numeric", length = 2, prop = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = interval.scale.disp, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = down.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = left.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = up.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = right.space, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = orient, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = dist.legend, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = box.type, options = c("o", "l", "7", "c", "u", "]", "n"), length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = amplif.label, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = amplif.axis, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = std.x.range, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = std.y.range, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = cex.pt, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = col.box, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = x.nb.inter.tick, class = "integer", length = 1, neg.values = FALSE, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = y.nb.inter.tick, class = "integer", length = 1, neg.values = FALSE, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = tick.length, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = sec.tick.length, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = corner.text, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = amplif.legend, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = corner.text.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = trim.return, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # end argument checking with fun_check()
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- if(all(is.na(data) | ! is.finite(data))){
- tempo.cat <- paste0("ERROR IN fun_trim FUNCTION\ndata ARGUMENT CONTAINS ONLY NA OR Inf")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end argument checking
- # main code
- if(all(class(data)%in% c("matrix", "array"))){
- data <- as.vector(data)
- }
- na.nb <- NULL
- if(any(is.na(data))){
- na.nb <- sum(c(is.na(data)))
- data <- data[ ! is.na(data)]
- }
- color.cut <- hsv(0.75, 1, 1) # color of interval selected
- col.mean <- hsv(0.25, 1, 0.8) # color of interval using mean+/-sd
- col.quantile <- "orange" # color of interval using quantiles
- quantiles.selection <- c(0.01, 0.025, 0.05, 0.1, 0.25, 0.5, 0.75, 0.9, 0.95, 0.975, 0.99) # quantiles used in axis to help for choosing trimming cutoffs
- if(single.value.display == FALSE & length(unique(data)) == 1){
- par(bty = "n", xaxt = "n", yaxt = "n", xpd = TRUE)
- plot(1, pch = 16, col = "white", xlab = "", ylab = "")
- text(x = 1, y = 1, paste0("No graphic displayed\nBecause data made of a single different value (", formatC(as.double(table(data))), ")"), cex = 2)
- output <- list(trim.method = NULL, trim.cutoffs = NULL, real.trim.cutoffs = NULL, trimmed.values = NULL, kept.values = NULL)
- }else{
- output <- list(trim.method = trim.method, trim.cutoffs = trim.cutoffs, real.trim.cutoffs = NULL, trimmed.values = NULL, kept.values = NULL)
- fun.rug <- function(sec.tick.length.f = sec.tick.length, x.nb.inter.tick.f = x.nb.inter.tick, y.nb.inter.tick.f = y.nb.inter.tick){
- if(x.nb.inter.tick.f > 0){
- inter.tick.unit <- (par("xaxp")[2] - par("xaxp")[1]) / par("xaxp")[3]
- par.ini <- par()[c("xpd", "tcl")]
- par(xpd = FALSE)
- par(tcl = -par()$mgp[2] * sec.tick.length.f) # tcl gives the length of the ticks as proportion of line text, knowing that mgp is in text lines. So the main ticks are a 0.5 of the distance of the axis numbers by default. The sign provides the side of the tick (negative for outside of the plot region)
- suppressWarnings(rug(seq(par("xaxp")[1] - 10 * inter.tick.unit, par("xaxp")[2] + 10 * inter.tick.unit, by = inter.tick.unit / (1 + x.nb.inter.tick.f)), ticksize = NA, side = 1)) # ticksize = NA to allow the use of par()$tcl value
- par(par.ini)
- rm(par.ini)
- }
- if(y.nb.inter.tick.f > 0){
- inter.tick.unit <- (par("yaxp")[2] - par("yaxp")[1]) / par("yaxp")[3]
- par.ini <- par()[c("xpd", "tcl")]
- par(xpd = FALSE)
- par(tcl = -par()$mgp[2] * sec.tick.length.f) # tcl gives the length of the ticks as proportion of line text, knowing that mgp is in text lines. So the main ticks are a 0.5 of the distance of the axis numbers by default. The sign provides the side of the tick (negative for outside of the plot region)
- suppressWarnings(rug(seq(par("yaxp")[1] - 10 * inter.tick.unit, par("yaxp")[2] + 10 * inter.tick.unit, by = inter.tick.unit / (1 + y.nb.inter.tick.f)), ticksize = NA, side = 2)) # ticksize = NA to allow the use of par()$tcl value
- par(par.ini)
- rm(par.ini)
- }
- }
- fun.add.cut <- function(data.f, trim.method.f = trim.method, trim.cutoffs.f = trim.cutoffs, color.cut.f = color.cut, return.f = FALSE){
- # DEBUGGING
- # data.f = data ; trim.method.f = "mean.sd"; trim.cutoffs.f = trim.cutoffs ; color.cut.f = color.cut ; return.f = TRUE
- real.trim.cutoffs.f <- NULL
- if(trim.method.f != ""){
- data.f <- sort(data.f)
- par.ini <- par()$xpd
- par(xpd = FALSE)
- if(trim.method.f == "mean.sd"){
- real.trim.cutoffs.f <- qnorm(trim.cutoffs.f, mean(data.f, na.rm = TRUE), sd(data.f, na.rm = TRUE))
- abline(v = qnorm(trim.cutoffs.f, mean(data.f, na.rm = TRUE), sd(data.f, na.rm = TRUE)), col = color.cut.f)
- segments(qnorm(trim.cutoffs.f[1], mean(data.f, na.rm = TRUE), sd(data.f, na.rm = TRUE)), par()$usr[4] * 0.75, qnorm(trim.cutoffs.f[2], mean(data.f, na.rm = TRUE), sd(data.f, na.rm = TRUE)), par()$usr[4] * 0.75, col = color.cut.f)
- }
- if(trim.method.f == "quantile"){
- real.trim.cutoffs.f <- quantile(data.f, probs = trim.cutoffs.f, type = 7, na.rm = TRUE)
- abline(v = quantile(data.f, probs = trim.cutoffs.f, type = 7, na.rm = TRUE), col = color.cut.f)
- segments(quantile(data.f, probs = trim.cutoffs.f[1], type = 7, na.rm = TRUE), par()$usr[4] * 0.75, quantile(data.f, probs = trim.cutoffs.f[2], type = 7, na.rm = TRUE), par()$usr[4] * 0.75, col = color.cut.f)
- }
- par(par.ini)
- if(return.f == TRUE){
- trimmed.values.f <- data.f[data.f <= real.trim.cutoffs.f[1] | data.f >= real.trim.cutoffs.f[2]]
- kept.values.f <- data.f[data.f > real.trim.cutoffs.f[1] & data.f < real.trim.cutoffs.f[2]]
- }
- }else{
- real.trim.cutoffs.f <- NULL
- trimmed.values.f <- NULL
- kept.values.f <- NULL
- }
- if(return.f == TRUE){
- output <- list(trim.method = trim.method.f, trim.cutoffs = trim.cutoffs.f, real.trim.cutoffs = real.trim.cutoffs.f, trimmed.values = trimmed.values.f, kept.values = kept.values.f)
- return(output)
- }
- }
- fun.interval.scale.display <- function(data.f, col.quantile.f = col.quantile, quantiles.selection.f = quantiles.selection, col.mean.f = col.mean){ # intervals on top of graphs
- par.ini <- par()[c("mgp", "xpd")]
- par(mgp = c(0.25, 0.25, 0), xpd = NA)
- axis(side = 3, at = c(par()$usr[1], par()$usr[2]), labels = rep("", 2), col = col.quantile.f, lwd.ticks = 0)
- par(xpd = FALSE)
- axis(side = 3, at = quantile(as.vector(data.f), probs = quantiles.selection.f, type = 7, na.rm = TRUE), labels = quantiles.selection.f, col.axis = col.quantile.f, col = col.quantile.f)
- par(mgp = c(1.75, 1.75, 1.5), xpd = NA)
- axis(side = 3, at = c(par()$usr[1], par()$usr[2]), labels = rep("", 2), col = col.mean.f, lwd.ticks = 0)
- par(xpd = FALSE)
- axis(side = 3, at = m + s * qnorm(quantiles.selection.f), labels = formatC(round(qnorm(quantiles.selection.f), 2)), col.axis = col.mean.f, col = col.mean.f, lwd.ticks = 1)
- par(par.ini)
- }
- zone<-matrix(1:4, ncol=2)
- layout(zone)
- par(omi = c(0, 0, 1.5, 0), mai = c(down.space, left.space, up.space, right.space), las = orient, mgp = c(dist.legend / 0.2, 0.5, 0), xpd = FALSE, bty= box.type, cex.lab = amplif.label, cex.axis = amplif.axis, xaxs = ifelse(std.x.range, "i", "r"), yaxs = ifelse(std.y.range, "i", "r"))
- par(tcl = -par()$mgp[2] * tick.length) # tcl gives the length of the ticks as proportion of line text, knowing that mgp is in text lines. So the main ticks are a 0.5 of the distance of the axis numbers by default. The sign provides the side of the tick (negative for outside of the plot region)
- if(is.null(displayed.nb)){
- sampled.data <- as.vector(data)
- if(corner.text == ""){
- corner.text <- paste0("ALL VALUES OF THE DATASET DISPLAYED")
- }else{
- corner.text <- paste0(corner.text, "\nALL VALUES OF THE DATASET DISPLAYED")
- }
- }else{
- if(length(as.vector(data)) > displayed.nb){
- sampled.data <- sample(as.vector(data), displayed.nb, replace = FALSE)
- if(corner.text == ""){
- corner.text <- paste0("WARNING: ONLY ", displayed.nb, " VALUES ARE DISPLAYED AMONG THE ", length(as.vector(data)), " VALUES OF THE DATASET ANALYZED")
- }else{
- corner.text <- paste0(corner.text, "\nWARNING: ONLY ", displayed.nb, " VALUES ARE DISPLAYED AMONG THE ", length(as.vector(data)), " VALUES OF THE DATASET ANALYZED")
- }
- }else{
- sampled.data <- as.vector(data)
- if(corner.text == ""){
- corner.text <- paste0("WARNING: THE DISPLAYED NUMBER OF VALUES PARAMETER ", deparse(substitute(displayed.nb)), " HAS BEEN SET TO ", displayed.nb, " WHICH IS ABOVE THE NUMBER OF VALUES OF THE DATASET ANALYZED -> ALL VALUES DISPLAYED")
- }else{
- corner.text <- paste0(corner.text, "\nWARNING: THE DISPLAYED NUMBER OF VALUES PARAMETER ", deparse(substitute(displayed.nb)), " HAS BEEN SET TO ", displayed.nb, " WHICH IS ABOVE THE NUMBER OF VALUES OF THE DATASET ANALYZED -> ALL VALUES DISPLAYED")
- }
- }
- }
- if( ! is.null(na.nb)){
- if(corner.text == ""){
- corner.text <- paste0("WARNING: NUMBER OF NA REMOVED IS ", na.nb)
- }else{
- corner.text <- paste0("WARNING: NUMBER OF NA REMOVED IS ", na.nb)
- }
- }
- stripchart(sampled.data, method="jitter", jitter=0.4, vertical=FALSE, ylim=c(0.5, 1.5), group.names = "", xlab = "Value", ylab="", pch=1, cex = cex.pt / 0.2)
- fun.rug(y.nb.inter.tick.f = 0)
- boxplot(as.vector(data), horizontal=TRUE, add=TRUE, boxwex = 0.4, staplecol = col.box, whiskcol = col.box, medcol = col.box, boxcol = col.box, range = 0, whisklty = 1)
- m <- mean(as.vector(data), na.rm = TRUE)
- s <- sd(as.vector(data), na.rm = TRUE)
- segments(m, 0.8, m, 1, lwd=2, col="red") # mean
- segments(m -1.96 * s, 0.9, m + 1.96 * s, 0.9, lwd=1, col="red") # mean
- graph.xlim <- par()$usr[1:2] # for hist() and qqnorm() below
- if(interval.scale.disp == TRUE){
- fun.interval.scale.display(data.f = data)
- if(corner.text == ""){
- corner.text <- paste0("MULTIPLYING FACTOR DISPLAYED (MEAN +/- SD) ON SCALES: ", paste(formatC(round(qnorm(quantiles.selection), 2))[-(1:(length(quantiles.selection) - 1) / 2)], collapse = ", "), "\nQUANTILES DISPLAYED ON SCALES: ", paste(quantiles.selection, collapse = ", "))
- }else{
- corner.text <- paste0(corner.text, "\nMULTIPLYING FACTOR DISPLAYED (MEAN +/- SD) ON SCALES: ", paste(formatC(round(qnorm(quantiles.selection), 2))[-(1:(length(quantiles.selection) - 1) / 2)], collapse = ", "), "\nQUANTILES DISPLAYED ON SCALES: ", paste(quantiles.selection, collapse = ", "))
- }
- }
- output.tempo <- fun.add.cut(data.f = data, return.f = TRUE) # to recover real.trim.cutoffs
- if(trim.return == TRUE){
- output <- output.tempo
- }
- par(xpd = NA)
- if(trim.method != ""){
- if(corner.text == ""){
- corner.text <- paste0("SELECTED CUT-OFFS (PROPORTION): ", paste(trim.cutoffs, collapse = ", "), "\nSELECTED CUT-OFFS: ", paste(output.tempo$real.trim.cutoffs, collapse = ", "))
- }else{
- corner.text <- paste0(corner.text, "\nSELECTED CUT-OFFS (PROPORTION): ", paste(trim.cutoffs, collapse = ", "), "\nSELECTED CUT-OFFS: ", paste(output.tempo$real.trim.cutoffs, collapse = ", "))
- }
- if(interval.scale.disp == TRUE){
- legend(x = (par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] - ((par("usr")[2] - par("usr")[1]) / (par("omd")[2] - par("omd")[1])) * par("omd")[1]), y = (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / (par("omd")[4] - par("omd")[3])) * (1 - par("omd")[4]) / 2), legend = c(c("min, Q1, Median, Q3, max"), "mean +/- 1.96sd", paste0("Trimming interval: ", paste0(trim.cutoffs, collapse = " , ")), "Mean +/- sd multiplying factor", "Quantile"), yjust = 0, lty=1, col=c(col.box, "red", color.cut, col.mean, col.quantile), bty="n", cex = amplif.legend)
- }else{
- legend(x = (par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] - ((par("usr")[2] - par("usr")[1]) / (par("omd")[2] - par("omd")[1])) * par("omd")[1]), y = (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / (par("omd")[4] - par("omd")[3])) * (1 - par("omd")[4]) / 2), legend = c(c("min, Q1, Median, Q3, max"), "mean +/- 1.96sd", paste0("Trimming interval: ", paste0(trim.cutoffs, collapse = " , "))), yjust = 0, lty=1, col=c(col.box, "red", color.cut), bty="n", cex = amplif.legend, y.intersp=1.25)
- }
- }else{
- if(interval.scale.disp == TRUE){
- legend(x = (par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] - ((par("usr")[2] - par("usr")[1]) / (par("omd")[2] - par("omd")[1])) * par("omd")[1]), y = (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / (par("omd")[4] - par("omd")[3])) * (1 - par("omd")[4]) / 2), legend = c(c("min, Q1, Median, Q3, max"), "mean +/- sd", "Mean +/- sd multiplying factor", "Quantile"), yjust = 0, lty=1, col=c(col.box, "red", col.mean, col.quantile), bty="n", cex = amplif.legend)
- }else{
- legend(x = (par("usr")[1] - ((par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1])) * par("plt")[1] - ((par("usr")[2] - par("usr")[1]) / (par("omd")[2] - par("omd")[1])) * par("omd")[1]), y = (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / (par("omd")[4] - par("omd")[3])) * (1 - par("omd")[4]) / 2), legend = c(c("min, Q1, Median, Q3, max"), "mean +/- sd"), yjust = 0, lty=1, col=c(col.box, "red"), bty="n", cex = amplif.legend, y.intersp=1.25)
- }
- }
- par(xpd = FALSE, xaxs = ifelse(std.x.range, "i", "r"), yaxs = ifelse(std.y.range, "i", "r"))
- hist(as.vector(data), main = "", xlim = graph.xlim, xlab = "Value", ylab="Density", col = grey(0.25)) # removed: breaks = seq(min(as.vector(data), na.rm = TRUE), max(as.vector(data), na.rm = TRUE), length.out = length(as.vector(data)) / 10)
- abline(h = par()$usr[3])
- fun.rug()
- if(interval.scale.disp == TRUE){
- fun.interval.scale.display(data.f = data)
- }
- fun.add.cut(data.f = data)
- par(xaxs = ifelse(std.x.range, "i", "r"))
- stripchart(rank(sampled.data), method="stack", vertical=FALSE, ylim=c(0.99, 1.3), group.names = "", xlab = "Rank of values", ylab="", pch=1, cex = cex.pt / 0.2)
- fun.rug(y.nb.inter.tick.f = 0)
- x.text <- par("usr")[2] + (par("usr")[2] - par("usr")[1]) / (par("plt")[2] - par("plt")[1]) * (1 - par("plt")[2]) / 2
- y.text <- (par("usr")[4] + ((par("usr")[4] - par("usr")[3]) / (par("plt")[4] - par("plt")[3])) * (1 - par("plt")[4]) + ((par("usr")[4] - par("usr")[3]) / ((par()$omd[4] / 2) * ((par("plt")[4] - par("plt")[3])))) * (1 - par("omd")[4])) # BEWARE. Here in "(par()$omd[4] / 2", division by two because there are 2 graphs staked on the y axis, and not one
- par(xpd=NA)
- text(x = x.text, y = y.text, paste0(corner.text), adj=c(1, 1.1), cex = corner.text.size) # text at the topright corner
- par(xpd=FALSE)
- par(xaxs = ifelse(std.x.range, "i", "r"), yaxs = ifelse(std.y.range, "i", "r"))
- qqnorm(as.vector(sampled.data), main = "", datax = TRUE, ylab = "Value", pch = 1, col = "red", cex = cex.pt / 0.2)
- fun.rug()
- if(diff(quantile(as.vector(data), probs = c(0.25, 0.75), na.rm = TRUE)) != 0){ # otherwise, error generated
- qqline(as.vector(data), datax = TRUE)
- }
- if(interval.scale.disp == TRUE){
- fun.interval.scale.display(data.f = data)
- }
- fun.add.cut(data.f = data)
- }
- if(trim.return == TRUE){
- return(output)
- }
+x.lim.order <- order(x.lim) # to deal with inverse axis
+# print(x.lim.order)
+x.lim <- sort(x.lim)
+x.lim[1] <- x.lim[1] - abs(x.lim[2] - x.lim[1]) * ifelse(diff(x.lim.order) > 0, x.right.extra.margin, x.left.extra.margin) # diff(x.lim.order) > 0 means not inversed axis
+x.lim[2] <- x.lim[2] + abs(x.lim[2] - x.lim[1]) * ifelse(diff(x.lim.order) > 0, x.left.extra.margin, x.right.extra.margin) # diff(x.lim.order) > 0 means not inversed axis
+if(x.include.zero == TRUE){ # no need to check x.log != "no" because done before
+x.lim <- range(c(x.lim, 0), na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
+}
+x.lim <- x.lim[x.lim.order]
+if(any(is.na(x.lim))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 3")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+if(is.null(y.lim)){
+if(any(unlist(mapply(FUN = "[[", data1, y, SIMPLIFY = FALSE)) %in% c(Inf, -Inf))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE y COLUMN IN data1 CONTAINS -Inf OR Inf VALUES THAT WILL NOT BE CONSIDERED IN THE PLOT RANGE")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+y.lim <- suppressWarnings(range(unlist(mapply(FUN = "[[", data1, y, SIMPLIFY = FALSE)), na.rm = TRUE, finite = TRUE)) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only. y.lim added here. If NULL, ok if y argument has values
+}else if(y.log != "no"){
+y.lim <- get(y.log)(y.lim) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
+}
+if(y.log != "no"){
+if(any( ! is.finite(y.lim))){
+tempo.cat <- paste0("ERROR IN ", function.name, "\ny.lim ARGUMENT CANNOT HAVE ZERO OR NEGATIVE VALUES WITH THE y.log ARGUMENT SET TO ", y.log, ":\n", paste(y.lim, collapse = " "), "\nPLEASE, CHECK DATA VALUES (PRESENCE OF ZERO OR INF VALUES)")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+}
+if(suppressWarnings(all(y.lim %in% c(Inf, -Inf)))){ # happen when y is only NULL
+if(all(unlist(geom) == "geom_vline")){
+tempo.cat <- paste0("ERROR IN ", function.name, " NOT POSSIBLE TO DRAW geom_vline KIND OF LINES ALONE IF y.lim ARGUMENT IS SET TO NULL, SINCE NO Y-AXIS DEFINED (", ifelse(length(y) == 1, "y", paste0("ELEMENT ", i1, " OF y")), " ARGUMENT MUST BE NULL FOR THESE KIND OF LINES)")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, " y.lim ARGUMENT MADE OF NA, -Inf OR Inf ONLY: ", paste(y.lim, collapse = " "))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
}
+}
+y.lim.order <- order(y.lim) # to deal with inverse axis
+y.lim <- sort(y.lim)
+y.lim[1] <- y.lim[1] - abs(y.lim[2] - y.lim[1]) * ifelse(diff(y.lim.order) > 0, y.bottom.extra.margin, y.top.extra.margin) # diff(y.lim.order) > 0 means not inversed axis
+y.lim[2] <- y.lim[2] + abs(y.lim[2] - y.lim[1]) * ifelse(diff(y.lim.order) > 0, y.top.extra.margin, y.bottom.extra.margin) # diff(y.lim.order) > 0 means not inversed axis
+if(y.include.zero == TRUE){ # no need to check y.log != "no" because done before
+y.lim <- range(c(y.lim, 0), na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
+}
+y.lim <- y.lim[y.lim.order]
+if(any(is.na(y.lim))){
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 4")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+# end axes management
-######## fun_segmentation() #### segment a dot cloud on a scatterplot and define the dots from another cloud outside the segmentation
-# Check OK: clear to go Apollo
-fun_segmentation <- function(
- data1,
- x1,
- y1,
- x.range.split = NULL,
- x.step.factor = 10,
- y.range.split = NULL,
- y.step.factor = 10,
- error = 0,
- data2 = NULL,
- x2,
- y2,
- data2.pb.dot = "unknown",
- xy.cross.kind = "&",
- plot = FALSE,
- graph.in.file = FALSE,
- raster = TRUE,
- warn.print = FALSE,
- lib.path = NULL
-){
- # AIM
- # if data1 is a data frame corresponding to the data set of a scatterplot (with a x column for x-axis values and a y column for the y-axis column), then fun_segmentation() delimits a frame around the dots cloud using a sliding window set by x.range.split and x.step.factor to frame the top and bottom part of the cloud, and set by y.range.split and y.step.factor to frame the left and right part of the cloud
- # if a second data frame is provided, corresponding to the data set of a scatterplot (with a x column for x-axis values and a y column for the y-axis column), then fun_segmentation() defines the dots of this data frame, outside of the frame of the first data frame
- # WARNINGS
- # if dots from data2 look significant on the graph (outside the frame) but are not (not black on the last figure), this is probably because the frame is flat on the zero coordinate (no volume inside the frame at this position). Thus, no way to conclude that data2 dots here are significant. These dots are refered to as "unknown". The pb.dot argument deals with such dots
- # dots that are sometimes inside and outside the frame, depending on the sliding window, are treated differently: they are removed. Such dots are neither classified as "signif", "non signif" or "unknown", but as "inconsistent"
- # unknown dots are treated as finally significant, not significant, or unknown (data2.pb.dot argument) for each x-axis and y-axis separately. Then, the union or intersection of significant dots is performed (argument xy.cross.kind). See the example section
- # ARGUMENTS
- # data1: a data frame containing a column of x-axis values and a column of y-axis values
- # x1: character string of the data1 column name for x-axis (first column of data1 by default)
- # y1: character string of the data1 column name for y-axis (second column of data1 by default)
- # x.range.split: positive non null numeric value giving the number of interval on the x value range. if x.range is the range of the dots on the x-axis, then abs(diff(x.range) / x.range.split) gives the window size. Window size decreases when range.split increases. In unit of x-axis. Write NULL if not required. At least one of the x.range.split and y.range.split must be non NULL
- # x.step.factor: positive non null numeric value giving the shift step of the window. If x.step.factor = 1, no overlap during the sliding (when the window slides from position n to position n+1, no overlap between the two positions). If x.step.factor = 2, 50% of overlap (when the window slides from position n to position n+1, the window on position n+1 overlap 50% of the window when it was on position n)
- # y.range.split: same as x.range.split for the y-axis. At least one of the x.range.split and y.range.split must be non NULL
- # y.step.factor: same as x.step.factor for the y-axis
- # error: proportion (from 0 to 1) of false positives (i.e., proportion of dots from data1 outside of the frame). 0.05 means 5% of the dots from data1 outside of the frame
- # data2: a data frame containing a column of x-axis values and a column of y-axis values, for which outside dots of the data1 cloud has to be determined. Write NULL if not required
- # x2: character string of the data1 column name for x-axis (first column of data1 by default)
- # y2: character string of the data1 column name for y-axis (second column of data1 by default)
- # data2.pb.dot: unknown dots are explain in the warning section above. If "signif", then the unknown dots are finally considered as significant (outside the frame). If "not.signif", then the unknown dots are finally considered as non significant (inside the frame). If "unknown", no conclusion are drawn from these dots. See the examples below
- # xy.cross.kind: if data2 is non null and if both x.range.split and y.range.split are non null, which dots are finally significants? Write "&" for intersection of outside dots on x and on y. Write "|" for union of outside dots on x and on y. See the examples below
- # plot: logical. Print graphs that check the frame?
- # graph.in.file: logical. Graphs sent into a graphic device already opened? If FALSE, GUI are opened for each graph. If TRUE, no GUI are opended. The graphs are displayed on the current active graphic device. Ignored if plot is FALSE
- # raster: logical. Dots in raster mode? If FALSE, dots from each geom_point from geom argument are in vectorial mode (bigger pdf and long to display if millions of dots). If TRUE, dots from each geom_point from geom argument are in matricial mode (smaller pdf and easy display if millions of dots, but long to generate the layer). If TRUE, the region plot will be square to avoid a bug in fun_gg_point_rast(). If TRUE, solve the transparency problem with some GUI. Not considered if plot is FALSE
- # warn.print: logical. Print warnings at the end of the execution? No print if no warning messages
- # lib.path: character vector specifying the absolute pathways of the directories containing the required packages if not in the default directories. Ignored if NULL. Ignored if plot is FALSE
- # RETURN
- # several graphs if plot is TRUE
- # a list containing:
- # $data1.removed.row.nb: which rows have been removed due to NA; NaN, -Inf or Inf detection in x1 or y1 columns (NULL if no row removed)
- # $data1.removed.rows: removed rows (NULL if no row removed)
- # $data2.removed.row.nb: which rows have been removed due to NA; NaN, -Inf or Inf detection in x2 or y2 columns (NULL if no row removed)
- # $data2.removed.rows: removed rows (NULL if no row removed)
- # $hframe: x and y coordinates of the bottom and top frames for frame plotting (frame1 for the left step and frame2 for the right step)
- # $vframe: x and y coordinates of the left and right frames for frame plotting (frame1 for the down step and frame2 for the top step)
- # $data1.signif.dot: the significant dots of data1 (i.e., dots outside the frame). A good segmentation should not have any data1.signif.dot
- # $data1.non.signif.dot: the non significant dots of data1 (i.e., dots inside the frame)
- # $data1.inconsistent.dot: see the warning section above
- # $data2.signif.dot: the significant dots of data2 if non NULL (i.e., dots outside the frame)
- # $data2.non.signif.dot: the non significant dots of data2 (i.e., dots inside the frame)
- # $data2.unknown.dot: the problematic dots of data2 (i.e., data2 dots outside of the range of data1, or data2 dots in a sliding window without data1 dots). Is systematically NULL except if argument data2.pb.dot = "unknown" and some data2 dots are in such situation. Modifying the segmentation x.range.split, x.step.factor, y.range.split, y.step.factor arguments can solve this problem
- # $data2.inconsistent.dot: see the warning section above
- # $axes: the x-axis and y-axis info
- # $warn: the warning messages. Use cat() for proper display. NULL if no warning
- # REQUIRED PACKAGES
- # ggplot2 if plot is TRUE
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # if plot is TRUE:
- # fun_pack()
- # fun_open()
- # fun_gg_palette()
- # fun_gg_scatter()
- # fun_gg_empty_graph()
- # fun_close()
- # EXAMPLES
- # example explaining the unknown and inconsistent dots, and the cross
-
- # set.seed(1) ; data1 = data.frame(x = rnorm(500), y = rnorm(500), stringsAsFactors = TRUE) ; data1[5:7, 2] <- NA ; data2 = data.frame(x = rnorm(500, 0, 2), y = rnorm(500, 0, 2), stringsAsFactors = TRUE) ; data2[11:13, 1] <- Inf ; set.seed(NULL) ; fun_segmentation(data1 = data1, x1 = names(data1)[1], y1 = names(data1)[2], x.range.split = 20, x.step.factor = 10, y.range.split = 23, y.step.factor = 10, error = 0, data2 = data2, x2 = names(data2)[1], y2 = names(data2)[2], data2.pb.dot = "not.signif", xy.cross.kind = "|", plot = TRUE, graph.in.file = FALSE, raster = FALSE, lib.path = NULL)
- # set.seed(1) ; data1 = data.frame(x = rnorm(500), y = rnorm(500), stringsAsFactors = TRUE) ; data2 = data.frame(x = rnorm(500, 0, 2), y = rnorm(500, 0, 2), stringsAsFactors = TRUE) ; set.seed(NULL) ; fun_segmentation(data1 = data1, x1 = names(data1)[1], y1 = names(data1)[2], x.range.split = NULL, x.step.factor = 10, y.range.split = 23, y.step.factor = 10, error = 0, data2 = data2, x2 = names(data2)[1], y2 = names(data2)[2], data2.pb.dot = "unknown", xy.cross.kind = "|", plot = TRUE, graph.in.file = FALSE, raster = FALSE, lib.path = NULL)
- # set.seed(1) ; data1 = data.frame(x = rnorm(500), y = rnorm(500), stringsAsFactors = TRUE) ; data2 = data.frame(x = rnorm(500, 0, 2), y = rnorm(500, 0, 2), stringsAsFactors = TRUE) ; set.seed(NULL) ; fun_segmentation(data1 = data1, x1 = names(data1)[1], y1 = names(data1)[2], x.range.split = 20, x.step.factor = 10, y.range.split = NULL, y.step.factor = 10, error = 0, data2 = data2, x2 = names(data2)[1], y2 = names(data2)[2], data2.pb.dot = "unknown", xy.cross.kind = "&", plot = TRUE, graph.in.file = FALSE, raster = FALSE, lib.path = NULL)
- # DEBUGGING
- # set.seed(1) ; data1 = data.frame(x = rnorm(50), y = rnorm(50), stringsAsFactors = TRUE) ; data1[5:7, 2] <- NA ; x1 = names(data1)[1] ; y1 = names(data1)[2] ; x.range.split = 5 ; x.step.factor = 10 ; y.range.split = 5 ; y.step.factor = 10 ; error = 0 ; data2 = data.frame(x = rnorm(50, 0, 2), y = rnorm(50, 0, 2), stringsAsFactors = TRUE) ; set.seed(NULL) ; x2 = names(data2)[1] ; y2 = names(data2)[2] ; data2.pb.dot = "unknown" ; xy.cross.kind = "|" ; plot = TRUE ; graph.in.file = FALSE ; raster = FALSE ; warn.print = TRUE ; lib.path = NULL
- # set.seed(1) ; data1 = data.frame(x = rnorm(500), y = rnorm(500), stringsAsFactors = TRUE) ; data2 = data.frame(x = rnorm(500, 0, 2), y = rnorm(500, 0, 2), stringsAsFactors = TRUE) ; set.seed(NULL) ; x1 = names(data1)[1] ; y1 = names(data1)[2] ; x.range.split = 20 ; x.step.factor = 10 ; y.range.split = 23 ; y.step.factor = 10 ; error = 0 ; x2 = names(data2)[1] ; y2 = names(data2)[2] ; data2.pb.dot = "not.signif" ; xy.cross.kind = "|" ; plot = TRUE ; graph.in.file = FALSE ; raster = FALSE ; warn.print = TRUE ; lib.path = NULL
- # set.seed(1) ; data1 = data.frame(x = rnorm(500), y = rnorm(500), stringsAsFactors = TRUE) ; data2 = data.frame(x = rnorm(500, 0, 2), y = rnorm(500, 0, 2), stringsAsFactors = TRUE) ; set.seed(NULL) ; x1 = names(data1)[1] ; y1 = names(data1)[2] ; x.range.split = 20 ; x.step.factor = 10 ; y.range.split = NULL ; y.step.factor = 10 ; error = 0 ; x2 = names(data2)[1] ; y2 = names(data2)[2] ; data2.pb.dot = "unknown" ; xy.cross.kind = "&" ; plot = TRUE ; graph.in.file = FALSE ; raster = FALSE ; warn.print = TRUE ; lib.path = NULL
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- ini.warning.length <- options()$warning.length
- warn <- NULL
- warn.count <- 0
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = data1, class = "data.frame", na.contain = TRUE, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & length(data1) < 2){
- tempo.cat <- paste0("ERROR IN ", function.name, ": data1 ARGUMENT MUST BE A DATA FRAME OF AT LEAST 2 COLUMNS")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- tempo <- fun_check(data = x1, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & ! (x1 %in% names(data1))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": x1 ARGUMENT MUST BE A COLUMN NAME OF data1")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }else if(tempo$problem == FALSE & x1 %in% names(data1)){
- tempo <- fun_check(data = data1[, x1], data.name = "x1 COLUMN OF data1", class = "vector", mode = "numeric", na.contain = TRUE, fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = y1, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & ! (y1 %in% names(data1))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": y1 ARGUMENT MUST BE A COLUMN NAME OF data1")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }else if(tempo$problem == FALSE & y1 %in% names(data1)){
- tempo <- fun_check(data = data1[, y1], data.name = "y1 COLUMN OF data1", class = "vector", mode = "numeric", na.contain = TRUE, fun.name = function.name) ; eval(ee)
- }
- if(is.null(x.range.split) & is.null(y.range.split)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": AT LEAST ONE OF THE x.range.split AND y.range.split ARGUMENTS MUST BE NON NULL")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- if( ! is.null(x.range.split)){
- tempo <- fun_check(data = x.range.split, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & x.range.split < 1){
- tempo.cat <- paste0("ERROR IN ", function.name, ": x.range.split ARGUMENT CANNOT BE LOWER THAN 1")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- if( ! is.null(y.range.split)){
- tempo <- fun_check(data = y.range.split, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & y.range.split < 1){
- tempo.cat <- paste0("ERROR IN ", function.name, ": y.range.split ARGUMENT CANNOT BE LOWER THAN 1")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- tempo <- fun_check(data = x.step.factor, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & x.step.factor < 1){
- tempo.cat <- paste0("ERROR IN ", function.name, ": x.step.factor ARGUMENT CANNOT BE LOWER THAN 1")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- tempo <- fun_check(data = y.step.factor, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & y.step.factor < 1){
- tempo.cat <- paste0("ERROR IN ", function.name, ": y.step.factor ARGUMENT CANNOT BE LOWER THAN 1")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- tempo <- fun_check(data = error, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(data2)){
- if(is.null(x2) | is.null(y2)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": x2 AND y2 ARGUMENTS CANNOT BE NULL IF data2 ARGUMENT IS NON NULL")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- tempo <- fun_check(data = data2, class = "data.frame", na.contain = TRUE, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & length(data2) < 2){
- tempo.cat <- paste0("ERROR IN ", function.name, ": data2 ARGUMENT MUST BE A DATA FRAME OF AT LEAST 2 COLUMNS")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- if( ! is.null(x2)){
- tempo <- fun_check(data = x2, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & ! (x2 %in% names(data2))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": x2 ARGUMENT MUST BE A COLUMN NAME OF data2")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }else if(tempo$problem == FALSE & x2 %in% names(data2)){
- tempo <- fun_check(data = data2[, x2], data.name = "x2 COLUMN OF data2", class = "vector", mode = "numeric", na.contain = TRUE, fun.name = function.name) ; eval(ee)
- }
- }
- if( ! is.null(y2)){
- tempo <- fun_check(data = y2, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & ! (y2 %in% names(data2))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": y2 ARGUMENT MUST BE A COLUMN NAME OF data2")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }else if(tempo$problem == FALSE & y2 %in% names(data2)){
- tempo <- fun_check(data = data2[, y2], data.name = "y2 COLUMN OF data2", class = "vector", mode = "numeric", na.contain = TRUE, fun.name = function.name) ; eval(ee)
- }
- }
- }
- if( ! is.null(data2)){
- tempo <- fun_check(data = data2.pb.dot, options = c("signif", "not.signif", "unknown"), length = 1, fun.name = function.name) ; eval(ee)
- }
- if( ! (is.null(x.range.split)) & ! (is.null(y.range.split))){
- tempo <- fun_check(data = xy.cross.kind, options = c("&", "|"), length = 1, fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = plot, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = warn.print, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & plot == TRUE){
- tempo <- fun_check(data = raster, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = graph.in.file, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & graph.in.file == TRUE & is.null(dev.list())){
- tempo.cat <- paste0("ERROR IN ", function.name, ": \ngraph.in.file PARAMETER SET TO TRUE BUT NO ACTIVE GRAPHIC DEVICE DETECTED")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }else if(tempo$problem == FALSE & graph.in.file == TRUE & ! is.null(dev.list())){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") GRAPHS PRINTED IN THE CURRENT DEVICE (TYPE ", toupper(names(dev.cur())), ")")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- if( ! is.null(lib.path)){
- tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
- tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- }
- }
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # other required function checking
- if(plot == TRUE){
- if(length(utils::find("fun_pack", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_pack() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(length(utils::find("fun_open", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_open() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(length(utils::find("fun_gg_palette", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_gg_palette() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(length(utils::find("fun_gg_empty_graph", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_gg_empty_graph() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(length(utils::find("fun_gg_scatter", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_gg_scatter() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(length(utils::find("fun_close", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_close() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # end other required function checking
- # package checking
- if(plot == TRUE){
- fun_pack(req.package = c("ggplot2"), lib.path = lib.path)
- }
- # end package checking
- # main code
- # na and Inf detection and removal (done now to be sure of the correct length of categ)
- data1.removed.row.nb <- NULL
- data1.removed.rows <- NULL
- data2.removed.row.nb <- NULL
- data2.removed.rows <- NULL
- if(any(is.na(data1[, c(x1, y1)])) | any(is.infinite(data1[, x1])) | any(is.infinite(data1[, y1]))){
- tempo.na <- unlist(lapply(lapply(c(data1[c(x1, y1)]), FUN = is.na), FUN = which))
- tempo.inf <- unlist(lapply(lapply(c(data1[c(x1, y1)]), FUN = is.infinite), FUN = which))
- data1.removed.row.nb <- sort(unique(c(tempo.na, tempo.inf)))
- if(length(data1.removed.row.nb) > 0){
- data1.removed.rows <- data1[data1.removed.row.nb, ]
- }
- if(length(data1.removed.row.nb) == nrow(data1)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": AT LEAST ONE NA, NaN, -Inf OR Inf DETECTED IN EACH ROW OF data1. FUNCTION CANNOT BE USED ON EMPTY DATA FRAME")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(length(data1.removed.row.nb) > 0){
- data1 <- data1[-data1.removed.row.nb, ]
- }
- if(nrow(data1) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 1")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NA, NaN, -Inf OR Inf DETECTED IN COLUMN ", paste(c(x1, y1), collapse = " "), " OF data1 AND CORRESPONDING ROWS REMOVED (SEE $data1.removed.row.nb AND $data1.removed.rows)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }else{
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NO NA, NaN, -Inf OR Inf DETECTED IN COLUMN ", paste(c(x1, y1), collapse = " "), " OF data1. NO ROW REMOVED")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- if( ! is.null(data2)){
- if(any(is.na(data2[, c(x2, y2)])) | any(is.infinite(data2[, x2])) | any(is.infinite(data2[, y2]))){
- tempo.na <- unlist(lapply(lapply(c(data2[c(x2, y2)]), FUN = is.na), FUN = which))
- tempo.inf <- unlist(lapply(lapply(c(data2[c(x2, y2)]), FUN = is.infinite), FUN = which))
- data2.removed.row.nb <- sort(unique(c(tempo.na, tempo.inf)))
- if(length(data2.removed.row.nb) > 0){
- data2.removed.rows <- data2[data2.removed.row.nb, ]
- }
- if(length(data2.removed.row.nb) == nrow(data2)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": AT LEAST ONE NA, NaN, -Inf OR Inf DETECTED IN EACH ROW OF data2. FUNCTION CANNOT BE USED ON EMPTY DATA FRAME")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(length(data2.removed.row.nb) > 0){
- data2 <- data2[-data2.removed.row.nb, ]
- }
- if(nrow(data2) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 2")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NA, NaN, -Inf OR Inf DETECTED IN COLUMN ", paste(c(x2, y2), collapse = " "), " OF data2 AND CORRESPONDING ROWS REMOVED (SEE $data2.removed.row.nb AND $data2.removed.rows)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }else{
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NO NA, NaN, -Inf OR Inf DETECTED IN COLUMN ", paste(c(x2, y2), collapse = " "), " OF data2. NO ROW REMOVED")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- # end na and Inf detection and removal (done now to be sure of the correct length of categ)
- # row annotation (dot number)
- # data1 <- data1[ ! duplicated(data1[, c(x1, y1)]), ] # do not remove the dots that have same x and y values, because they will have different dot number -> not the same position on the matrices (so true for symmetric matrices)
- data1 <- cbind(data1, DOT_NB = 1:nrow(data1), stringsAsFactors = TRUE)
- if( ! is.null(data2)){
- # data2 <- data2[ ! duplicated(data2[, c(x2, y2)]), ] # do not remove the dots that have same x and y values, because they will have different dot number -> not the same position on the matrices (so true for symmetric matrices)
- data2 <- cbind(data2, DOT_NB = 1:nrow(data2), stringsAsFactors = TRUE)
- }
- # end row annotation (dot number)
-
-
-
-
- # Method using x unit interval
- # may be create vector of each column to increase speed
- x.data1.l <- NULL # x coord of the y upper and lower limits defined on the data1 cloud for left step line
- x.data1.r <- NULL # x coord of the y upper and lower limits defined on the data1 cloud for right step line
- y.data1.down.limit.l <- NULL # lower limit of the data1 cloud for left step line
- y.data1.top.limit.l <- NULL # upper limit of the data1 cloud for left step line
- y.data1.down.limit.r <- NULL # lower limit of the data1 cloud for right step line
- y.data1.top.limit.r <- NULL # upper limit of the data1 cloud for left step line
- if(any(data1[, x1] %in% c(Inf, -Inf))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE data1 ARGUMENT CONTAINS -Inf OR Inf VALUES IN THE x1 COLUMN, THAT WILL NOT BE CONSIDERED IN THE PLOT RANGE")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- x.range <- range(data1[, x1], na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
- if(suppressWarnings(any(x.range %in% c(Inf, -Inf)))){
- tempo.cat <- paste0("ERROR IN ", function.name, " COMPUTED x.range CONTAINS Inf VALUES, BECAUSE VALUES FROM data1 ARGUMENTS ARE NA OR Inf ONLY")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(data1[, y1] %in% c(Inf, -Inf))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE data1 ARGUMENT CONTAINS -Inf OR Inf VALUES IN THE y1 COLUMN, THAT WILL NOT BE CONSIDERED IN THE PLOT RANGE")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- y.range <- range(data1[, y1], na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
- if(suppressWarnings(any(x.range %in% c(Inf, -Inf)))){
- tempo.cat <- paste0("ERROR IN ", function.name, " COMPUTED y.range CONTAINS Inf VALUES, BECAUSE VALUES FROM data1 ARGUMENTS ARE NA OR Inf ONLY")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- x.range.plot <- range(data1[, x1], na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
- y.range.plot <- range(data1[, y1], na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
- if( ! is.null(data2)){
- if(any(data2[, x2] %in% c(Inf, -Inf))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE data2 ARGUMENT CONTAINS -Inf OR Inf VALUES IN THE x2 COLUMN, THAT WILL NOT BE CONSIDERED IN THE PLOT RANGE")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- x.range.plot <- range(data1[, x1], data2[, x2], na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
- if(any(data2[, y2] %in% c(Inf, -Inf))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE data2 ARGUMENT CONTAINS -Inf OR Inf VALUES IN THE y2 COLUMN, THAT WILL NOT BE CONSIDERED IN THE PLOT RANGE")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- y.range.plot <- range(data1[, y1], data2[, y2], na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
- }
- if(suppressWarnings(any(x.range.plot %in% c(Inf, -Inf)))){
- tempo.cat <- paste0("ERROR IN ", function.name, " COMPUTED x.range.plot CONTAINS Inf VALUES, BECAUSE VALUES FROM data1 (AND data2?) ARGUMENTS ARE NA OR Inf ONLY")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(suppressWarnings(any(y.range.plot %in% c(Inf, -Inf)))){
- tempo.cat <- paste0("ERROR IN ", function.name, " COMPUTED y.range.plot CONTAINS Inf VALUES, BECAUSE VALUES FROM data1 (AND data2?) ARGUMENTS ARE NA OR Inf ONLY")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! is.null(x.range.split)){
- # data.frame ordering to slide the window from small to big values + sliding window definition
- data1 <- data1[order(data1[, x1], na.last = TRUE), ]
- if( ! is.null(data2)){
- data2 <- data2[order(data2[, x2], na.last = TRUE), ]
- }
- x.win.size <- abs(diff(x.range) / x.range.split) # in unit of x-axis
- step <- x.win.size / x.step.factor
- # end data.frame ordering to slide the window from small to big values + sliding window definition
- # x-axis sliding and y-axis limits of the data1 cloud -> y significant data2
- loop.nb <- ceiling((diff(x.range) - x.win.size) / step) # x.win.size + n * step covers the x range if x.win.size + n * step >= diff(x.range), thus if n >= (diff(x.range) - x.win.size) / step
- y.outside.data1.dot.nb <- integer() # vector that will contain the selected rows numbers of data1 that are upper or lower than the frame
- y.inside.data1.dot.nb <- integer() # vector that will contain the selected rows numbers of data1 that are not upper or lower than the frame
- y.data1.median <- median(data1[, y1], na.rm = TRUE) # will be used for sliding windows without data1 in it
- if( ! is.null(data2)){
- y.outside.data2.dot.nb <- integer() # vector that will contain the selected 1D coordinates (i.e., dots) of data2 that are upper or lower than the data1 frame
- y.inside.data2.dot.nb <- integer() # vector that will contain the 1D coordinates (i.e., dots) of data2 that are not upper or lower than the data1 frame
- y.unknown.data2.dot.nb <- integer() # vector that will contain the 1D coordinates (i.e., dots) of data2 that are problematic: data2 dots outside of the range of data1, or data2 dots in a sliding window without data1 dots
- # recover data2 dots outside the range of data1
- if(any(data2[, x2] < x.range[1])){
- y.unknown.data2.dot.nb <- c(y.unknown.data2.dot.nb, data2$DOT_NB[data2[, x2] < x.range[1]])
- #tempo.warn & indicate the interval
- }
- if(any(data2[, x2] > x.range[2])){
- y.unknown.data2.dot.nb <- c(y.unknown.data2.dot.nb, data2$DOT_NB[data2[, x2] > x.range[2]])
- #tempo.warn & indicate the interval
- }
- # end recover data2 dots outside the range of data1
- }
- # loop.ini.time <- as.numeric(Sys.time())
- for(i1 in 0:(loop.nb + 1)){
- min.pos <- x.range[1] + step * i1 # lower position of the sliding window in data1
- max.pos <- min.pos + x.win.size # upper position of the sliding window in data1
- x.data1.l <- c(x.data1.l, min.pos, min.pos + step) # min.pos + step to make the steps
- x.data1.r <- c(x.data1.r, max.pos, max.pos + step) # max.pos + step to make the steps
- x.data1.dot.here <- data1[, x1] >= min.pos & data1[, x1] < max.pos # is there data1 dot present in the sliding window, considering the x axis?
- if( ! is.null(data2)){
- x.data2.dot.here <- data2[, x2] >= min.pos & data2[, x2] < max.pos # is there data2 dot present in the sliding window, considering the x axis?
- }
- # recover the data1 dots outside the frame
- if(any(x.data1.dot.here == TRUE)){
- tempo.y.data1.top.limit <- quantile(data1[x.data1.dot.here, y1], probs = 1 - error, na.rm = TRUE)
- tempo.y.data1.down.limit <- quantile(data1[x.data1.dot.here, y1], probs = 0 + error, na.rm = TRUE)
- y.data1.top.limit.l <- c(y.data1.top.limit.l, tempo.y.data1.top.limit, tempo.y.data1.top.limit)
- y.data1.down.limit.l <- c(y.data1.down.limit.l, tempo.y.data1.down.limit, tempo.y.data1.down.limit)
- y.data1.top.limit.r <- c(y.data1.top.limit.r, tempo.y.data1.top.limit, tempo.y.data1.top.limit)
- y.data1.down.limit.r <- c(y.data1.down.limit.r, tempo.y.data1.down.limit, tempo.y.data1.down.limit)
- y.data1.dot.signif <- ( ! ((data1[, y1] <= tempo.y.data1.top.limit) & (data1[, y1] >= tempo.y.data1.down.limit))) & x.data1.dot.here # is there data1 dot present in the sliding window, above or below the data1 limits, considering the y axis?
- y.data1.dot.not.signif <- x.data1.dot.here & ! y.data1.dot.signif
- y.outside.data1.dot.nb <- c(y.outside.data1.dot.nb, data1$DOT_NB[y.data1.dot.signif]) # recover the row number of data1
- y.outside.data1.dot.nb <- unique(y.outside.data1.dot.nb)
- y.inside.data1.dot.nb <- c(y.inside.data1.dot.nb, data1$DOT_NB[y.data1.dot.not.signif])
- y.inside.data1.dot.nb <- unique(y.inside.data1.dot.nb)
- }else{
- y.data1.top.limit.l <- c(y.data1.top.limit.l, y.data1.median, y.data1.median)
- y.data1.down.limit.l <- c(y.data1.down.limit.l, y.data1.median, y.data1.median)
- y.data1.top.limit.r <- c(y.data1.top.limit.r, y.data1.median, y.data1.median)
- y.data1.down.limit.r <- c(y.data1.down.limit.r, y.data1.median, y.data1.median)
- }
- # end recover the data1 dots outside the frame
- # recover the data2 dots outside the frame
- if( ! is.null(data2)){
- if(any(x.data1.dot.here == TRUE) & any(x.data2.dot.here == TRUE)){
- y.data2.dot.signif <- ( ! ((data2[, y2] <= tempo.y.data1.top.limit) & (data2[, y2] >= tempo.y.data1.down.limit))) & x.data2.dot.here # is there data2 dot present in the sliding window, above or below the data1 limits, considering the y axis?
- y.data2.dot.not.signif <- x.data2.dot.here & ! y.data2.dot.signif
- y.outside.data2.dot.nb <- c(y.outside.data2.dot.nb, data2$DOT_NB[y.data2.dot.signif])
- y.outside.data2.dot.nb <- unique(y.outside.data2.dot.nb)
- y.inside.data2.dot.nb <- c(y.inside.data2.dot.nb, data2$DOT_NB[y.data2.dot.not.signif])
- y.inside.data2.dot.nb <- unique(y.inside.data2.dot.nb)
- }else if(any(x.data1.dot.here == FALSE) & any(x.data2.dot.here == TRUE)){ # problem: data2 dots in the the window but no data1 dots to generates the quantiles
- y.unknown.data2.dot.nb <- c(y.unknown.data2.dot.nb, data2$DOT_NB[x.data2.dot.here])
- y.unknown.data2.dot.nb <- unique(y.unknown.data2.dot.nb)
- #tempo.warn & indicate the interval
-
-
-
-
- # tempo.warn <- paste0("FROM FUNCTION ", function.name, ": THE [", round(min.pos, 3), " ; ", round(max.pos, 3), "] INTERVAL DOES NOT CONTAIN data1 X VALUES BUT CONTAINS data2 X VALUES WHICH CANNOT BE EVALUATED.\nTHE CONCERNED data2 ROW NUMBERS ARE:\n", paste(which(x.data1.dot.here == FALSE & x.data2.dot.here == TRUE), collapse = "\n"))
- # warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- # end recover the data2 dots outside the frame
- # if(any(i1 == seq(1, loop.nb, 500))){
- # loop.fin.time <- as.numeric(Sys.time()) # time of process end
- # cat(paste0("COMPUTATION TIME OF LOOP ", i1, " / ", loop.nb, ": ", as.character(lubridate::seconds_to_period(round(loop.fin.time - loop.ini.time))), "\n"))
- # }
- }
- if(max.pos < x.range[2]){
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE SLIDING WINDOW HAS NOT REACHED THE MAX VALUE OF data1 ON THE X-AXIS: ", max.pos, " VERSUS ", x.range[2])
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- y.incon.data1.dot.nb.final <- unique(c(y.outside.data1.dot.nb[y.outside.data1.dot.nb %in% y.inside.data1.dot.nb], y.inside.data1.dot.nb[y.inside.data1.dot.nb %in% y.outside.data1.dot.nb])) # inconsistent dots: if a row number of y.inside.data1.dot.nb is present in y.outside.data1.dot.nb (and vice versa), it means that during the sliding, a dot has been sometime inside, sometime outside -> removed from the outside list
- y.outside.data1.dot.nb.final <- y.outside.data1.dot.nb[ ! (y.outside.data1.dot.nb %in% y.incon.data1.dot.nb.final)] # inconsistent dots removed from the outside list
- y.inside.data1.dot.nb.final <- y.inside.data1.dot.nb[ ! (y.inside.data1.dot.nb %in% y.incon.data1.dot.nb.final)] # inconsistent dots removed from the inside list
- if( ! is.null(data2)){
- # if some unknown dots are also inside, and/or outside, they are put in the inside and/or outside. Ok, because then the intersection between inside and outside is treated -> inconsistent dots
- tempo.unknown.out <- y.unknown.data2.dot.nb[y.unknown.data2.dot.nb %in% y.outside.data2.dot.nb]
- y.outside.data2.dot.nb <- unique(c(y.outside.data2.dot.nb, tempo.unknown.out)) # if a row number of y.unknown.data2.dot.nb is present in y.outside.data2.dot.nb, it is put into outside
- tempo.unknown.in <- y.unknown.data2.dot.nb[y.unknown.data2.dot.nb %in% y.inside.data2.dot.nb]
- y.inside.data2.dot.nb <- unique(c(y.inside.data2.dot.nb, tempo.unknown.in)) # if a row number of y.unknown.data2.dot.nb is present in y.inside.data2.dot.nb, it is put into inside
- y.unknown.data2.dot.nb.final <- y.unknown.data2.dot.nb[ ! (y.unknown.data2.dot.nb %in% c(y.outside.data2.dot.nb, y.inside.data2.dot.nb))] # then dots also in inside and outside are remove from unknown
- y.incon.data2.dot.nb.final <- unique(c(y.outside.data2.dot.nb[y.outside.data2.dot.nb %in% y.inside.data2.dot.nb], y.inside.data2.dot.nb[y.inside.data2.dot.nb %in% y.outside.data2.dot.nb])) # inconsistent dots: if a row number of y.inside.data2.dot.nb is present in y.outside.data2.dot.nb (and vice versa), it means that during the sliding, a dot has been sometime inside, sometime outside -> removed from the outside list
- y.outside.data2.dot.nb.final <- y.outside.data2.dot.nb[ ! (y.outside.data2.dot.nb %in% y.incon.data2.dot.nb.final)] # inconsistent dots removed from the outside list
- y.inside.data2.dot.nb.final <- y.inside.data2.dot.nb[ ! (y.inside.data2.dot.nb %in% y.incon.data2.dot.nb.final)] # inconsistent dots removed from the inside list
- }
- # end x-axis sliding and y-axis limits of the data1 cloud -> y significant data2
- }
- # end Method using x unit interval
-
-
-
-
- # Method using y unit interval
- y.data1.d <- NULL # y coord of the x upper and lower limits defined on the data1 cloud for down step line
- y.data1.t <- NULL # y coord of the x upper and lower limits defined on the data1 cloud for top step line
- x.data1.left.limit.d <- NULL # left limit of the data1 cloud for down step line
- x.data1.right.limit.d <- NULL # right limit of the data1 cloud for down step line
- x.data1.left.limit.t <- NULL # left limit of the data1 cloud for top step line
- x.data1.right.limit.t <- NULL # right limit of the data1 cloud for top step line
- if( ! is.null(y.range.split)){
- # data.frame ordering to slide the window from small to big values + sliding window definition
- data1 <- data1[order(data1[, y1], na.last = TRUE), ]
- if( ! is.null(data2)){
- data2 <- data2[order(data2[, y2], na.last = TRUE), ]
- }
- y.win.size <- abs(diff(y.range) / y.range.split) # in unit of y-axis
- step <- y.win.size / y.step.factor
- # end data.frame ordering to slide the window from small to big values + sliding window definition
- # y-axis sliding and x-axis limits of the data1 cloud -> x significant data2
- loop.nb <- ceiling((diff(y.range) - y.win.size) / step) # y.win.size + n * step covers the y range if y.win.size + n * step >= diff(y.range), thus if n >= (diff(y.range) - y.win.size) / step
- x.outside.data1.dot.nb <- integer() # vector that will contain the selected rows numbers of data1 that are upper or lower than the frame
- x.inside.data1.dot.nb <- integer() # vector that will contain the selected rows numbers of data1 that are not upper or lower than the frame
- x.data1.median <- median(data1[, x1], na.rm = TRUE) # will be used for sliding window without data1 in it
- if( ! is.null(data2)){
- x.outside.data2.dot.nb <- integer() # vector that will contain the selected 1D coordinates (i.e., dots) of data2 that are upper or lower than the data1 frame
- x.inside.data2.dot.nb <- integer() # vector that will contain the 1D coordinates (i.e., dots) of data2 that are not upper or lower than the data1 frame
- x.unknown.data2.dot.nb <- integer() # vector that will contain the 1D coordinates (i.e., dots) of data2 that are problematic: data2 dots outside of the range of data1, or data2 dots in a sliding window without data1 dots
- # recover data2 dots outside the range of data1
- if(any(data2[, y2] < y.range[1])){
- x.unknown.data2.dot.nb <- c(x.unknown.data2.dot.nb, data2$DOT_NB[data2[, y2] < y.range[1]])
- }
- if(any(data2[, y2] > y.range[2])){
- x.unknown.data2.dot.nb <- c(x.unknown.data2.dot.nb, data2$DOT_NB[data2[, y2] > y.range[2]])
- }
- # end recover data2 dots outside the range of data1
- }
- # loop.ini.time <- as.numeric(Sys.time())
- for(i1 in 0:(loop.nb + 1)){
- min.pos <- y.range[1] + step * i1 # lower position of the sliding window in data1
- max.pos <- min.pos + y.win.size # upper position of the sliding window in data1
- y.data1.d <- c(y.data1.d, min.pos, min.pos + step) # min.pos + step to make the steps
- y.data1.t <- c(y.data1.t, max.pos, max.pos + step) # max.pos + step to make the steps
- y.data1.dot.here <- data1[, y1] >= min.pos & data1[, y1] < max.pos # is there data1 dot present in the sliding window, considering the y axis?
- if( ! is.null(data2)){
- y.data2.dot.here <- data2[, y2] >= min.pos & data2[, y2] < max.pos # is there data2 dot present in the sliding window, considering the y axis?
- }
- # recover the data1 dots outside the frame
- if(any(y.data1.dot.here == TRUE)){
- tempo.x.data1.right.limit <- quantile(data1[y.data1.dot.here, x1], probs = 1 - error, na.rm = TRUE)
- tempo.x.data1.left.limit <- quantile(data1[y.data1.dot.here, x1], probs = 0 + error, na.rm = TRUE)
- x.data1.right.limit.d <- c(x.data1.right.limit.d, tempo.x.data1.right.limit, tempo.x.data1.right.limit)
- x.data1.left.limit.d <- c(x.data1.left.limit.d, tempo.x.data1.left.limit, tempo.x.data1.left.limit)
- x.data1.right.limit.t <- c(x.data1.right.limit.t, tempo.x.data1.right.limit, tempo.x.data1.right.limit)
- x.data1.left.limit.t <- c(x.data1.left.limit.t, tempo.x.data1.left.limit, tempo.x.data1.left.limit)
- x.data1.dot.signif <- ( ! ((data1[, x1] <= tempo.x.data1.right.limit) & (data1[, x1] >= tempo.x.data1.left.limit))) & y.data1.dot.here # is there data2 dot present in the sliding window, above or below the data1 limits, considering the x axis?
- x.data1.dot.not.signif <- y.data1.dot.here & ! x.data1.dot.signif
- x.outside.data1.dot.nb <- c(x.outside.data1.dot.nb, data1$DOT_NB[x.data1.dot.signif]) # recover the row number of data1
- x.outside.data1.dot.nb <- unique(x.outside.data1.dot.nb)
- x.inside.data1.dot.nb <- c(x.inside.data1.dot.nb, data1$DOT_NB[x.data1.dot.not.signif])
- x.inside.data1.dot.nb <- unique(x.inside.data1.dot.nb)
- }else{
- x.data1.right.limit.d <- c(x.data1.right.limit.d, x.data1.median, x.data1.median)
- x.data1.left.limit.d <- c(x.data1.left.limit.d, x.data1.median, x.data1.median)
- x.data1.right.limit.t <- c(x.data1.right.limit.t, x.data1.median, x.data1.median)
- x.data1.left.limit.t <- c(x.data1.left.limit.t, x.data1.median, x.data1.median)
- }
- # end recover the data1 dots outside the frame
- # recover the data2 dots outside the frame
- if( ! is.null(data2)){
- if(any(y.data1.dot.here == TRUE) & any(y.data2.dot.here == TRUE)){
- x.data2.dot.signif <- ( ! ((data2[, x2] <= tempo.x.data1.right.limit) & (data2[, x2] >= tempo.x.data1.left.limit))) & y.data2.dot.here # is there data2 dot present in the sliding window, above or below the data1 limits, considering the x axis?
- x.data2.dot.not.signif <- y.data2.dot.here & ! x.data2.dot.signif
- x.outside.data2.dot.nb <- c(x.outside.data2.dot.nb, data2$DOT_NB[x.data2.dot.signif])
- x.outside.data2.dot.nb <- unique(x.outside.data2.dot.nb)
- x.inside.data2.dot.nb <- c(x.inside.data2.dot.nb, data2$DOT_NB[x.data2.dot.not.signif])
- x.inside.data2.dot.nb <- unique(x.inside.data2.dot.nb)
- }else if(any(y.data1.dot.here == FALSE) & any(y.data2.dot.here == TRUE)){ # recover the data2 dots outside the range of the data1 cloud
- x.unknown.data2.dot.nb <- c(x.unknown.data2.dot.nb, data2$DOT_NB[y.data2.dot.here])
- x.unknown.data2.dot.nb <- unique(x.unknown.data2.dot.nb)
-
-
-
- # tempo.warn <- paste0("FROM FUNCTION ", function.name, ": THE [", round(min.pos, 3), " ; ", round(max.pos, 3), "] INTERVAL DOES NOT CONTAIN data1 Y VALUES BUT CONTAINS data2 Y VALUES WHICH CANNOT BE EVALUATED.\nTHE CONCERNED data2 ROW NUMBERS ARE:\n", paste(which(y.data1.dot.here == FALSE & y.data2.dot.here == TRUE), collapse = "\n"))
- # warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- # end recover the data2 dots outside the frame
- # if(any(i1 == seq(1, loop.nb, 500))){
- # loop.fin.time <- as.numeric(Sys.time()) # time of process end
- # cat(paste0("COMPUTATION TIME OF LOOP ", i1, " / ", loop.nb, ": ", as.character(lubridate::seconds_to_period(round(loop.fin.time - loop.ini.time))), "\n"))
- # }
- }
- if(max.pos < y.range[2]){
- tempo.cat <- paste0("ERROR IN ", function.name, ": THE SLIDING WINDOW HAS NOT REACHED THE MAX VALUE OF data1 ON THE Y-AXIS: ", max.pos, " VERSUS ", y.range[2])
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- x.incon.data1.dot.nb.final <- unique(c(x.outside.data1.dot.nb[x.outside.data1.dot.nb %in% x.inside.data1.dot.nb], x.inside.data1.dot.nb[x.inside.data1.dot.nb %in% x.outside.data1.dot.nb])) # inconsistent dots: if a row number of x.inside.data1.dot.nb is present in x.outside.data1.dot.nb (and vice versa), it means that during the sliding, a dot has been sometime inside, sometime outside -> removed from the outside list
- x.outside.data1.dot.nb.final <- x.outside.data1.dot.nb[ ! (x.outside.data1.dot.nb %in% x.incon.data1.dot.nb.final)] # inconsistent dots removed from the outside list
- x.inside.data1.dot.nb.final <- x.inside.data1.dot.nb[ ! (x.inside.data1.dot.nb %in% x.incon.data1.dot.nb.final)] # inconsistent dots removed from the inside list
- if( ! is.null(data2)){
- # if some unknown dots are also inside, and/or outside, they are put in the inside and/or outside. Ok, because then the intersection between inside and outside is treated -> inconsistent dots
- tempo.unknown.out <- x.unknown.data2.dot.nb[x.unknown.data2.dot.nb %in% x.outside.data2.dot.nb]
- x.outside.data2.dot.nb <- unique(c(x.outside.data2.dot.nb, tempo.unknown.out)) # if a row number of x.unknown.data2.dot.nb is present in x.outside.data2.dot.nb, it is put into outside
- tempo.unknown.in <- x.unknown.data2.dot.nb[x.unknown.data2.dot.nb %in% x.inside.data2.dot.nb]
- x.inside.data2.dot.nb <- unique(c(x.inside.data2.dot.nb, tempo.unknown.in)) # if a row number of x.unknown.data2.dot.nb is present in x.inside.data2.dot.nb, it is put into inside
- x.unknown.data2.dot.nb.final <- x.unknown.data2.dot.nb[ ! (x.unknown.data2.dot.nb %in% c(x.outside.data2.dot.nb, x.inside.data2.dot.nb))] # then dots also in inside and outside are remove from unknown
- x.incon.data2.dot.nb.final <- unique(c(x.outside.data2.dot.nb[x.outside.data2.dot.nb %in% x.inside.data2.dot.nb], x.inside.data2.dot.nb[x.inside.data2.dot.nb %in% x.outside.data2.dot.nb])) # inconsistent dots: if a row number of x.inside.data2.dot.nb is present in x.outside.data2.dot.nb (and vice versa), it means that during the sliding, a dot has been sometime inside, sometime outside -> removed from the outside list
- x.outside.data2.dot.nb.final <- x.outside.data2.dot.nb[ ! (x.outside.data2.dot.nb %in% x.incon.data2.dot.nb.final)] # inconsistent dots removed from the outside list
- x.inside.data2.dot.nb.final <- x.inside.data2.dot.nb[ ! (x.inside.data2.dot.nb %in% x.incon.data2.dot.nb.final)] # inconsistent dots removed from the inside list
- }
- # end y-axis sliding and x-axis limits of the data1 cloud -> x significant data2
- }
- # end Method using y unit interval
-
-
-
- # recovering the frame coordinates
- hframe = rbind(
- data.frame(
- x = if(is.null(x.data1.l)){NULL}else{x.data1.l},
- y = if(is.null(x.data1.l)){NULL}else{y.data1.down.limit.l},
- kind = if(is.null(x.data1.l)){NULL}else{"down.frame1"},
- stringsAsFactors = TRUE
- ),
- data.frame(
- x = if(is.null(x.data1.r)){NULL}else{x.data1.r},
- y = if(is.null(x.data1.r)){NULL}else{y.data1.down.limit.r},
- kind = if(is.null(x.data1.r)){NULL}else{"down.frame2"},
- stringsAsFactors = TRUE
- ),
- data.frame(
- x = if(is.null(x.data1.l)){NULL}else{x.data1.l},
- y = if(is.null(x.data1.l)){NULL}else{y.data1.top.limit.l},
- kind = if(is.null(x.data1.l)){NULL}else{"top.frame1"},
- stringsAsFactors = TRUE
- ),
- data.frame(
- x = if(is.null(x.data1.r)){NULL}else{x.data1.r},
- y = if(is.null(x.data1.r)){NULL}else{y.data1.top.limit.r},
- kind = if(is.null(x.data1.r)){NULL}else{"top.frame2"},
- stringsAsFactors = TRUE
- ),
- stringsAsFactors = TRUE
- )
- vframe = rbind(
- data.frame(
- x = if(is.null(y.data1.d)){NULL}else{x.data1.left.limit.d},
- y = if(is.null(y.data1.d)){NULL}else{y.data1.d},
- kind = if(is.null(y.data1.d)){NULL}else{"left.frame1"},
- stringsAsFactors = TRUE
- ),
- data.frame(
- x = if(is.null(y.data1.t)){NULL}else{x.data1.left.limit.t},
- y = if(is.null(y.data1.t)){NULL}else{y.data1.t},
- kind = if(is.null(y.data1.t)){NULL}else{"left.frame2"},
- stringsAsFactors = TRUE
- ),
- data.frame(
- x = if(is.null(y.data1.d)){NULL}else{x.data1.right.limit.d},
- y = if(is.null(y.data1.d)){NULL}else{y.data1.d},
- kind = if(is.null(y.data1.d)){NULL}else{"right.frame1"},
- stringsAsFactors = TRUE
- ),
- data.frame(
- x = if(is.null(y.data1.t)){NULL}else{x.data1.right.limit.t},
- y = if(is.null(y.data1.t)){NULL}else{y.data1.t},
- kind = if(is.null(y.data1.t)){NULL}else{"right.frame2"},
- stringsAsFactors = TRUE
- ),
- stringsAsFactors = TRUE
- )
- # end recovering the frame coordinates
- # recovering the dot coordinates
- data1.signif.dot <- NULL
- data1.non.signif.dot <- NULL
- data1.incon.dot <- NULL
- data2.signif.dot <- NULL
- data2.non.signif.dot <- NULL
- data2.unknown.dot <- NULL
- data2.incon.dot <- NULL
- if(( ! is.null(x.range.split)) & ( ! is.null(y.range.split))){
- # inconsistent dots recovery
- if(length(unique(c(x.incon.data1.dot.nb.final, y.incon.data1.dot.nb.final))) > 0){
- data1.incon.dot <- data1[data1$DOT_NB %in% unique(c(x.incon.data1.dot.nb.final, y.incon.data1.dot.nb.final)), ] # if a dot in inconsistent in x or y -> classified as inconsistent (so unique() used)
- # removal of the inconsistent dot in the other classifications
- x.inside.data1.dot.nb.final <- x.inside.data1.dot.nb.final[ ! x.inside.data1.dot.nb.final %in% data1.incon.dot$DOT_NB]
- y.inside.data1.dot.nb.final <- y.inside.data1.dot.nb.final[ ! y.inside.data1.dot.nb.final %in% data1.incon.dot$DOT_NB]
- x.outside.data1.dot.nb.final <- x.outside.data1.dot.nb.final[ ! x.outside.data1.dot.nb.final %in% data1.incon.dot$DOT_NB]
- y.outside.data1.dot.nb.final <- y.outside.data1.dot.nb.final[ ! y.outside.data1.dot.nb.final %in% data1.incon.dot$DOT_NB]
- x.unknown.data1.dot.nb.final <- x.unknown.data1.dot.nb.final[ ! x.unknown.data1.dot.nb.final %in% data1.incon.dot$DOT_NB]
- y.unknown.data1.dot.nb.final <- y.unknown.data1.dot.nb.final[ ! y.unknown.data1.dot.nb.final %in% data1.incon.dot$DOT_NB]
- # end removal of the inconsistent dot in the other classifications
- }
- if( ! is.null(data2)){
- if(length(unique(c(x.incon.data2.dot.nb.final, y.incon.data2.dot.nb.final))) > 0){
- data2.incon.dot <- data2[data2$DOT_NB %in% unique(c(x.incon.data2.dot.nb.final, y.incon.data2.dot.nb.final)), ]
- # removal of the inconsistent dot in the other classifications
- x.inside.data2.dot.nb.final <- x.inside.data2.dot.nb.final[ ! x.inside.data2.dot.nb.final %in% data2.incon.dot$DOT_NB]
- y.inside.data2.dot.nb.final <- y.inside.data2.dot.nb.final[ ! y.inside.data2.dot.nb.final %in% data2.incon.dot$DOT_NB]
- x.outside.data2.dot.nb.final <- x.outside.data2.dot.nb.final[ ! x.outside.data2.dot.nb.final %in% data2.incon.dot$DOT_NB]
- y.outside.data2.dot.nb.final <- y.outside.data2.dot.nb.final[ ! y.outside.data2.dot.nb.final %in% data2.incon.dot$DOT_NB]
- x.unknown.data2.dot.nb.final <- x.unknown.data2.dot.nb.final[ ! x.unknown.data2.dot.nb.final %in% data2.incon.dot$DOT_NB]
- y.unknown.data2.dot.nb.final <- y.unknown.data2.dot.nb.final[ ! y.unknown.data2.dot.nb.final %in% data2.incon.dot$DOT_NB]
- # end removal of the inconsistent dot in the other classifications
- }
- }
- # end inconsistent dots recovery
- # unknown dots recovery
- if( ! is.null(data2)){
- if(data2.pb.dot == "signif"){
- x.outside.data2.dot.nb.final <- unique(c(x.outside.data2.dot.nb.final, x.unknown.data2.dot.nb.final))
- x.inside.data2.dot.nb.final <- x.inside.data2.dot.nb.final[ ! x.inside.data2.dot.nb.final %in% x.unknown.data2.dot.nb.final] # remove x.unknown.data2.dot.nb.final from x.inside.data2.dot.nb.final
- y.outside.data2.dot.nb.final <- unique(c(y.outside.data2.dot.nb.final, y.unknown.data2.dot.nb.final))
- y.inside.data2.dot.nb.final <- y.inside.data2.dot.nb.final[ ! y.inside.data2.dot.nb.final %in% y.unknown.data2.dot.nb.final] # remove y.unknown.data2.dot.nb.final from y.inside.data2.dot.nb.final
- x.unknown.data2.dot.nb.final <- NULL
- y.unknown.data2.dot.nb.final <- NULL
- data2.unknown.dot <- NULL
- }else if(data2.pb.dot == "not.signif"){
- x.inside.data2.dot.nb.final <- unique(c(x.inside.data2.dot.nb.final, x.unknown.data2.dot.nb.final))
- x.outside.data2.dot.nb.final <- x.outside.data2.dot.nb.final[ ! x.outside.data2.dot.nb.final %in% x.unknown.data2.dot.nb.final] # remove x.unknown.data2.dot.nb.final from x.outside.data2.dot.nb.final
- y.inside.data2.dot.nb.final <- unique(c(y.inside.data2.dot.nb.final, y.unknown.data2.dot.nb.final))
- y.outside.data2.dot.nb.final <- y.outside.data2.dot.nb.final[ ! y.outside.data2.dot.nb.final %in% y.unknown.data2.dot.nb.final] # remove y.unknown.data2.dot.nb.final from y.outside.data2.dot.nb.final
- x.unknown.data2.dot.nb.final <- NULL
- y.unknown.data2.dot.nb.final <- NULL
- data2.unknown.dot <- NULL
- }else if(data2.pb.dot == "unknown"){
- if(length(unique(c(x.unknown.data2.dot.nb.final, y.unknown.data2.dot.nb.final))) > 0){
- data2.unknown.dot <- data2[data2$DOT_NB %in% unique(c(x.unknown.data2.dot.nb.final, y.unknown.data2.dot.nb.final)), ] # if a dot in unknown in x or y -> classified as unknown (so unique() used)
- x.outside.data2.dot.nb.final <- x.outside.data2.dot.nb.final[ ! x.outside.data2.dot.nb.final %in% data2.unknown.dot$DOT_NB] # remove x.unknown.data2.dot.nb.final from x.outside.data2.dot.nb.final
- x.inside.data2.dot.nb.final <- x.inside.data2.dot.nb.final[ ! x.inside.data2.dot.nb.final %in% data2.unknown.dot$DOT_NB] # remove x.unknown.data2.dot.nb.final from x.inside.data2.dot.nb.final
- y.outside.data2.dot.nb.final <- y.outside.data2.dot.nb.final[ ! y.outside.data2.dot.nb.final %in% data2.unknown.dot$DOT_NB] # remove y.unknown.data2.dot.nb.final from y.outside.data2.dot.nb.final
- y.inside.data2.dot.nb.final <- y.inside.data2.dot.nb.final[ ! y.inside.data2.dot.nb.final %in% data2.unknown.dot$DOT_NB] # remove y.unknown.data2.dot.nb.final from y.inside.data2.dot.nb.final
- }
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 3")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # end unknown dots recovery
- # sign and non sign dot recovery
- if(xy.cross.kind == "|"){ # here the problem is to deal with significant dots depending on x and y. Thus I start with that, recover dots finally non significant in outside and put them in inside (when &), and remove from inside the dots in outside
- if(length(unique(c(x.outside.data1.dot.nb.final, y.outside.data1.dot.nb.final))) > 0){
- tempo.outside <- unique(c(x.outside.data1.dot.nb.final, y.outside.data1.dot.nb.final)) # union so unique() used
- tempo.inside <- unique(c(x.inside.data1.dot.nb.final, y.inside.data1.dot.nb.final))
- tempo.inside <- tempo.inside[ ! tempo.inside %in% tempo.outside]
- data1.signif.dot <- data1[data1$DOT_NB %in% tempo.outside, ]
- data1.non.signif.dot <- data1[data1$DOT_NB %in% tempo.inside, ]
- }else{
- data1.non.signif.dot <- data1[unique(c(x.inside.data1.dot.nb.final, y.inside.data1.dot.nb.final)), ] # if no outside dots, I recover all the inside dots and that's it
- }
- }else if(xy.cross.kind == "&"){
- if(sum(x.outside.data1.dot.nb.final %in% y.outside.data1.dot.nb.final) > 0){ # that is intersection
- tempo.outside <- unique(x.outside.data1.dot.nb.final[x.outside.data1.dot.nb.final %in% y.outside.data1.dot.nb.final]) # intersection
- tempo.outside.removed <- unique(c(x.outside.data1.dot.nb.final, y.outside.data1.dot.nb.final))[ ! unique(c(x.outside.data1.dot.nb.final, y.outside.data1.dot.nb.final)) %in% tempo.outside]
- tempo.inside <- unique(c(x.inside.data1.dot.nb.final, y.inside.data1.dot.nb.final))
- data1.signif.dot <- data1[data1$DOT_NB %in% tempo.outside, ]
- data1.non.signif.dot <- data1[data1$DOT_NB %in% tempo.inside, ]
- }else{
- data1.non.signif.dot <- data1[unique(c(x.inside.data1.dot.nb.final, y.inside.data1.dot.nb.final)), ] # if no outside dots, I recover all the inside dots and that's it
- }
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 4")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! is.null(data2)){
- if(xy.cross.kind == "|"){ # here the problem is to deal with significant dots depending on x and y. Thus I start with that, recover dots finally non significant in outside and put them in inside (when &), and remove from inside the dots in outside
- if(length(unique(c(x.outside.data2.dot.nb.final, y.outside.data2.dot.nb.final))) > 0){
- tempo.outside <- unique(c(x.outside.data2.dot.nb.final, y.outside.data2.dot.nb.final)) # union so unique() used
- tempo.inside <- unique(c(x.inside.data2.dot.nb.final, y.inside.data2.dot.nb.final))
- tempo.inside <- tempo.inside[ ! tempo.inside %in% tempo.outside]
- data2.signif.dot <- data2[data2$DOT_NB %in% tempo.outside, ]
- data2.non.signif.dot <- data2[data2$DOT_NB %in% tempo.inside, ]
- }else{
- data2.non.signif.dot <- data2[unique(c(x.inside.data2.dot.nb.final, y.inside.data2.dot.nb.final)), ] # if no outside dots, I recover all the inside dots and that's it
- }
- }else if(xy.cross.kind == "&"){
- if(sum(x.outside.data2.dot.nb.final %in% y.outside.data2.dot.nb.final) > 0){ # that is intersection
- tempo.outside <- unique(x.outside.data2.dot.nb.final[x.outside.data2.dot.nb.final %in% y.outside.data2.dot.nb.final]) # intersection
- tempo.outside.removed <- unique(c(x.outside.data2.dot.nb.final, y.outside.data2.dot.nb.final))[ ! unique(c(x.outside.data2.dot.nb.final, y.outside.data2.dot.nb.final)) %in% tempo.outside]
- tempo.inside <- unique(c(x.inside.data2.dot.nb.final, y.inside.data2.dot.nb.final))
- data2.signif.dot <- data2[data2$DOT_NB %in% tempo.outside, ]
- data2.non.signif.dot <- data2[data2$DOT_NB %in% tempo.inside, ]
- }else{
- data2.non.signif.dot <- data2[unique(c(x.inside.data2.dot.nb.final, y.inside.data2.dot.nb.final)), ] # if no outside dots, I recover all the inside dots and that's it
- }
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 5")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # end sign and non sign dot recovery
- }else if(( ! is.null(x.range.split)) & is.null(y.range.split)){
- # inconsistent dots recovery
- if(length(y.incon.data1.dot.nb.final) > 0){
- data1.incon.dot <- data1[data1$DOT_NB %in% y.incon.data1.dot.nb.final, ]
- }
- if( ! is.null(data2)){
- if(length(y.incon.data2.dot.nb.final) > 0){
- data2.incon.dot <- data2[data2$DOT_NB %in% y.incon.data2.dot.nb.final, ]
- }
- }# end inconsistent dots recovery
- # unknown dots recovery
- if( ! is.null(data2)){
- if(data2.pb.dot == "signif"){
- y.outside.data2.dot.nb.final <- unique(c(y.outside.data2.dot.nb.final, y.unknown.data2.dot.nb.final))
- }else if(data2.pb.dot == "not.signif"){
- y.inside.data2.dot.nb.final <- unique(c(y.inside.data2.dot.nb.final, y.unknown.data2.dot.nb.final))
- }else if(data2.pb.dot == "unknown"){
- if(length(y.unknown.data2.dot.nb.final) > 0){
- data2.unknown.dot <- data2[data2$DOT_NB %in% y.unknown.data2.dot.nb.final, ]
- }
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 6")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # end unknown dots recovery
- # sign and non sign dot recovery
- if(length(y.outside.data1.dot.nb.final) > 0){
- data1.signif.dot <- data1[data1$DOT_NB %in% y.outside.data1.dot.nb.final, ]
- }
- if(length(y.inside.data1.dot.nb.final) > 0){
- data1.non.signif.dot <- data1[data1$DOT_NB %in% y.inside.data1.dot.nb.final, ]
- }
- if( ! is.null(data2)){
- if(length(y.outside.data2.dot.nb.final) > 0){
- data2.signif.dot <- data2[data2$DOT_NB %in% y.outside.data2.dot.nb.final, ]
- }
- if(length(y.inside.data2.dot.nb.final) > 0){
- data2.non.signif.dot <- data2[data2$DOT_NB %in% y.inside.data2.dot.nb.final, ]
- }
- }
- # end sign and non sign dot recovery
- }else if(is.null(x.range.split) & ( ! is.null(y.range.split))){
- # inconsistent dots recovery
- if(length(x.incon.data1.dot.nb.final) > 0){
- data1.incon.dot <- data1[data1$DOT_NB %in% x.incon.data1.dot.nb.final, ]
- }
- if( ! is.null(data2)){
- if(length(x.incon.data2.dot.nb.final) > 0){
- data2.incon.dot <- data2[data2$DOT_NB %in% x.incon.data2.dot.nb.final, ]
- }
- }# end inconsistent dots recovery
- # unknown dots recovery
- if( ! is.null(data2)){
- if(data2.pb.dot == "signif"){
- x.outside.data2.dot.nb.final <- unique(c(x.outside.data2.dot.nb.final, x.unknown.data2.dot.nb.final))
- }else if(data2.pb.dot == "not.signif"){
- x.inside.data2.dot.nb.final <- unique(c(x.inside.data2.dot.nb.final, x.unknown.data2.dot.nb.final))
- }else if(data2.pb.dot == "unknown"){
- if(length(x.unknown.data2.dot.nb.final) > 0){
- data2.unknown.dot <- data2[data2$DOT_NB %in% x.unknown.data2.dot.nb.final, ]
- }
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 7")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # end unknown dots recovery
- # sign and non sign dot recovery
- if(length(x.outside.data1.dot.nb.final) > 0){
- data1.signif.dot <- data1[data1$DOT_NB %in% x.outside.data1.dot.nb.final, ]
- }
- if(length(x.inside.data1.dot.nb.final) > 0){
- data1.non.signif.dot <- data1[data1$DOT_NB %in% x.inside.data1.dot.nb.final, ]
- }
- if( ! is.null(data2)){
- if(length(x.outside.data2.dot.nb.final) > 0){
- data2.signif.dot <- data2[data2$DOT_NB %in% x.outside.data2.dot.nb.final, ]
- }
- if(length(x.inside.data2.dot.nb.final) > 0){
- data2.non.signif.dot <- data2[data2$DOT_NB %in% x.inside.data2.dot.nb.final, ]
- }
- }
- # end sign and non sign dot recovery
- }
- # end recovering the dot coordinates
- # verif
- if(any(data1.signif.dot$DOT_NB %in% data1.non.signif.dot$DOT_NB)){
- tempo.cat <- paste0("ERROR IN ", FUNCTION.NAME, ": CODE INCONSISTENCY 8")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(data1.non.signif.dot$DOT_NB %in% data1.signif.dot$DOT_NB)){
- tempo.cat <- paste0("ERROR IN ", FUNCTION.NAME, ": CODE INCONSISTENCY 9")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(data1.signif.dot$DOT_NB %in% data1.incon.dot$DOT_NB)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 10")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(data1.incon.dot$DOT_NB %in% data1.signif.dot$DOT_NB)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 11")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(data1.non.signif.dot$DOT_NB %in% data1.incon.dot$DOT_NB)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 12")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(data1.incon.dot$DOT_NB %in% data1.non.signif.dot$DOT_NB)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 13")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! is.null(data2)){
- if(any(data2.signif.dot$DOT_NB %in% data2.non.signif.dot$DOT_NB)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 14")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(data2.non.signif.dot$DOT_NB %in% data2.signif.dot$DOT_NB)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 15")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(data2.signif.dot$DOT_NB %in% data2.unknown.dot$DOT_NB)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 16")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(data2.unknown.dot$DOT_NB %in% data2.signif.dot$DOT_NB)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 17")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(data2.signif.dot$DOT_NB %in% data2.incon.dot$DOT_NB)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 18")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(data2.incon.dot$DOT_NB %in% data2.signif.dot$DOT_NB)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 19")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(data2.non.signif.dot$DOT_NB %in% data2.unknown.dot$DOT_NB)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 20")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(data2.unknown.dot$DOT_NB %in% data2.non.signif.dot$DOT_NB)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 21")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(data2.non.signif.dot$DOT_NB %in% data2.incon.dot$DOT_NB)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 22")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(data2.incon.dot$DOT_NB %in% data2.non.signif.dot$DOT_NB)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 23")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(data2.unknown.dot$DOT_NB %in% data2.incon.dot$DOT_NB)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 24")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(data2.incon.dot$DOT_NB %in% data2.unknown.dot$DOT_NB)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 25")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # end verif
- # plot
- # recovering the axes data whatever plot or not
- if(is.null(data2)){
- axes <- fun_gg_scatter(data1 = list(data1), x = list(x1), y = list(y1), categ = list(NULL), color = list(fun_gg_palette(2)[2]), geom = list("geom_point"), alpha = list(0.5), x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, plot = FALSE, return = TRUE)$axes
- }else{
- axes <- fun_gg_scatter(data1 = list(data1, data2), x = list(x1, x2), y = list(y1, y2), categ = list(NULL, NULL), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1]), geom = list("geom_point", "geom_point"), alpha = list(0.5, 0.5), x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, plot = FALSE, return = TRUE)$axes
- }
- # end recovering the axes data whatever plot or not
- if(plot == TRUE){
- # add a categ for plot legend
- tempo.df.name <- c("data1", "data1.signif.dot", "data1.incon.dot", "data2", "data2.signif.dot", "data2.unknown.dot", "data2.incon.dot")
- tempo.class.name <- c("data1", "data1", "data1", "data2", "data2", "data2", "data2")
- for(i2 in 1:length(tempo.df.name)){
- if( ! is.null(get(tempo.df.name[i2], env = sys.nframe(), inherit = FALSE))){
- assign(tempo.df.name[i2], data.frame(get(tempo.df.name[i2], env = sys.nframe(), inherit = FALSE), kind = tempo.class.name[i2]),
- stringsAsFactors = TRUE)
- }
- }
- # end add a categ for plot legend
- if(( ! is.null(x.range.split)) & ( ! is.null(y.range.split))){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, hframe, vframe), x = list(x1, "x", "x"), y = list(y1, "y", "y"), categ = list("kind", "kind", "kind"), legend.name = list("DATASET", "HORIZ FRAME" , "VERT FRAME"), color = list(fun_gg_palette(2)[2], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2), rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_path", "geom_path"), alpha = list(0.5, 0.5, 0.5), title = "DATA1", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- if( ! is.null(data1.signif.dot)){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, hframe, vframe, data1.signif.dot), x = list(x1, "x", "x", x1), y = list(y1, "y", "y", y1), categ = list("kind", "kind", "kind", "kind"), legend.name = list("DATASET", "HORIZ FRAME" , "VERT FRAME", "SIGNIF DOTS"), color = list(fun_gg_palette(2)[2], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2), rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2), "black"), geom = list("geom_point", "geom_path", "geom_path", "geom_point"), alpha = list(0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA1 SIGNIFICANT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else{
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA1 DOTS\nOUTSIDE THE FRAMES", text.size = 8, title = "DATA1 + DATA1 SIGNIFICANT DOTS")
- }
- if( ! is.null(data1.incon.dot)){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, hframe, vframe, data1.incon.dot), x = list(x1, "x", "x", x1), y = list(y1, "y", "y", y1), categ = list("kind", "kind", "kind", "kind"), legend.name = list("DATASET", "HORIZ FRAME" , "VERT FRAME", "INCONSISTENT DOTS"), color = list(fun_gg_palette(2)[2], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2), rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2), fun_gg_palette(7)[6]), geom = list("geom_point", "geom_path", "geom_path", "geom_point"), alpha = list(0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA1 INCONSISTENT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else{
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA1\nINCONSISTENT DOTS", text.size = 8, title = "DATA1 + DATA1 INCONSISTENT DOTS")
- }
- if( ! is.null(data2)){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, hframe , vframe), x = list(x1, x2, "x", "x"), y = list(y1, y2, "y", "y"), categ = list("kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "HORIZ FRAME" , "VERT FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2), rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_point", "geom_path", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- if( ! is.null(data2.signif.dot)){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, data2.signif.dot, hframe , vframe), x = list(x1, x2, x2, "x", "x"), y = list(y1, y2, y2, "y", "y"), categ = list("kind", "kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "SIGNIF DOTS", "HORIZ FRAME" , "VERT FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], "black", rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2), rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_point", "geom_point", "geom_path", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2 + DATA2 SIGNIFICANT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else{
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA2 DOTS\nOUTSIDE THE FRAMES", text.size = 8, title = "DATA1 + DATA2 + DATA2 SIGNIFICANT DOTS")
- }
- if( ! is.null(data2.incon.dot)){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, data2.incon.dot, hframe , vframe), x = list(x1, x2, x2, "x", "x"), y = list(y1, y2, y2, "y", "y"), categ = list("kind", "kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "INCONSISTENT DOTS", "HORIZ FRAME" , "VERT FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], fun_gg_palette(7)[6], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2), rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_point", "geom_point", "geom_path", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2 + DATA2 INCONSISTENT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else{
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA2\nINCONSISTENT DOTS", text.size = 8, title = "DATA2 + DATA2 INCONSISTENT DOTS")
- }
- if( ! is.null(data2.unknown.dot)){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, data2.unknown.dot, hframe , vframe), x = list(x1, x2, x2, "x", "x"), y = list(y1, y2, y2, "y", "y"), categ = list("kind", "kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "UNKNOWN DOTS", "HORIZ FRAME" , "VERT FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], fun_gg_palette(7)[5], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2), rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_point", "geom_point", "geom_path", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2 + DATA2 UNKNOWN DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
-
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else{
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA2\nUNKNOWN DOTS", text.size = 12, title = "DATA2 + DATA2 UNKNOWN DOTS")
- }
- }
- }else if(( ! is.null(x.range.split)) & is.null(y.range.split)){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, hframe), x = list(x1, "x"), y = list(y1, "y"), categ = list("kind", "kind"), legend.name = list("DATASET", "HORIZ FRAME"), color = list(fun_gg_palette(2)[2], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2)), geom = list("geom_point", "geom_path"), alpha = list(0.5, 0.5), title = "DATA1", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- if( ! is.null(data1.signif.dot)){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, hframe, data1.signif.dot), x = list(x1, "x", x1), y = list(y1, "y", y1), categ = list("kind", "kind", "kind"), legend.name = list("DATASET", "HORIZ FRAME", "SIGNIF DOTS"), color = list(fun_gg_palette(2)[2], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2), "black"), geom = list("geom_point", "geom_path", "geom_point"), alpha = list(0.5, 0.5, 0.5), title = "DATA1 + DATA1 SIGNIFICANT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else{
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA1 DOTS\nOUTSIDE THE FRAMES", text.size = 8, title = "DATA1 + DATA1 SIGNIFICANT DOTS")
- }
- if( ! is.null(data1.incon.dot)){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, hframe, data1.incon.dot), x = list(x1, "x", x1), y = list(y1, "y", y1), categ = list("kind", "kind", "kind"), legend.name = list("DATASET", "HORIZ FRAME", "INCONSISTENT DOTS"), color = list(fun_gg_palette(2)[2], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2), fun_gg_palette(7)[6]), geom = list("geom_point", "geom_path", "geom_point"), alpha = list(0.5, 0.5, 0.5), title = "DATA1 + DATA1 INCONSISTENT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else{
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA1\nINCONSISTENT DOTS", text.size = 8, title = "DATA1 + DATA1 INCONSISTENT DOTS")
- }
- if( ! is.null(data2)){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, hframe), x = list(x1, x2, "x"), y = list(y1, y2, "y"), categ = list("kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "HORIZ FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2)), geom = list("geom_point", "geom_point", "geom_path"), alpha = list(0.5, 0.5, 0.5), title = "DATA1 + DATA2", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- if( ! is.null(data2.signif.dot)){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, data2.signif.dot, hframe), x = list(x1, x2, x2, "x"), y = list(y1, y2, y2, "y"), categ = list("kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "SIGNIF DOTS", "HORIZ FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], "black", rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2)), geom = list("geom_point", "geom_point", "geom_point", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2 + DATA2 SIGNIFICANT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else{
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA2 DOTS\nOUTSIDE THE FRAMES", text.size = 8, title = "DATA1 + DATA2 + DATA2 SIGNIFICANT DOTS")
- }
- if( ! is.null(data2.incon.dot)){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, data2.incon.dot, hframe), x = list(x1, x2, x2, "x"), y = list(y1, y2, y2, "y"), categ = list("kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "INCONSISTENT DOTS", "HORIZ FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], fun_gg_palette(7)[6], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2)), geom = list("geom_point", "geom_point", "geom_point", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2 + DATA2 INCONSISTENT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else{
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA2\nINCONSISTENT DOTS", text.size = 8, title = "DATA2 + DATA2 INCONSISTENT DOTS")
- }
- if( ! is.null(data2.unknown.dot)){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, data2.unknown.dot, hframe), x = list(x1, x2, x2, "x"), y = list(y1, y2, y2, "y"), categ = list("kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "UNKNOWN DOTS", "HORIZ FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], fun_gg_palette(7)[5], rep(hsv(h = c(0.1, 0.15), v = c(0.75, 1)), 2)), geom = list("geom_point", "geom_point", "geom_point", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2 + DATA2 UNKNOWN DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else{
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA2\nUNKNOWN DOTS", text.size = 8, title = "DATA2 + DATA2 UNKNOWN DOTS")
- }
- }
- }else if(is.null(x.range.split) & ( ! is.null(y.range.split))){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, vframe), x = list(x1, "x"), y = list(y1, "y"), categ = list("kind", "kind"), legend.name = list("DATASET", "VERT FRAME"), color = list(fun_gg_palette(2)[2], rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_path"), alpha = list(0.5, 0.5), title = "DATA1", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- if( ! is.null(data1.signif.dot)){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, vframe, data1.signif.dot), x = list(x1, "x", x1), y = list(y1, "y", y1), categ = list("kind", "kind", "kind"), legend.name = list("DATASET", "VERT FRAME", "SIGNIF DOTS"), color = list(fun_gg_palette(2)[2], rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2), "black"), geom = list("geom_point", "geom_path", "geom_point"), alpha = list(0.5, 0.5, 0.5), title = "DATA1 + DATA1 SIGNIFICANT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else{
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA1 DOTS\nOUTSIDE THE FRAMES", text.size = 8, title = "DATA1 + DATA1 SIGNIFICANT DOTS")
- }
- if( ! is.null(data1.incon.dot)){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, vframe, data1.incon.dot), x = list(x1, "x", x1), y = list(y1, "y", y1), categ = list("kind", "kind", "kind"), legend.name = list("DATASET", "VERT FRAME", "INCONSISTENT DOTS"), color = list(fun_gg_palette(2)[2], rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2), fun_gg_palette(7)[6]), geom = list("geom_point", "geom_path", "geom_point"), alpha = list(0.5, 0.5, 0.5), title = "DATA1 + DATA1 INCONSISTENT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else{
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA1\nINCONSISTENT DOTS", text.size = 8, title = "DATA1 + DATA1 INCONSISTENT DOTS")
- }
- if( ! is.null(data2)){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, vframe), x = list(x1, x2, "x"), y = list(y1, y2, "y"), categ = list("kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "VERT FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_point", "geom_path"), alpha = list(0.5, 0.5, 0.5), title = "DATA1 + DATA2", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- if( ! is.null(data2.signif.dot)){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, data2.signif.dot, vframe), x = list(x1, x2, x2, "x"), y = list(y1, y2, y2, "y"), categ = list("kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "SIGNIF DOTS", "VERT FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], "black", rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_point", "geom_point", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2 + DATA2 SIGNIFICANT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else{
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA2 DOTS\nOUTSIDE THE FRAMES", text.size = 8, title = "DATA1 + DATA2 + DATA2 SIGNIFICANT DOTS")
- }
- if( ! is.null(data2.incon.dot)){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, data2.incon.dot, vframe), x = list(x1, x2, x2, "x"), y = list(y1, y2, y2, "y"), categ = list("kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "INCONSISTENT DOTS", "VERT FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], fun_gg_palette(7)[6], rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_point", "geom_point", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2 + DATA2 INCONSISTENT DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else{
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA2\nINCONSISTENT DOTS", text.size = 8, title = "DATA2 + DATA2 INCONSISTENT DOTS")
- }
- if( ! is.null(data2.unknown.dot)){
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- tempo.graph <- fun_gg_scatter(data1 = list(data1, data2, data2.unknown.dot, vframe), x = list(x1, x2, x2, "x"), y = list(y1, y2, y2, "y"), categ = list("kind", "kind", "kind", "kind"), legend.name = list("DATASET", "DATASET", "UNKNOWN DOTS", "VERT FRAME"), color = list(fun_gg_palette(2)[2], fun_gg_palette(2)[1], fun_gg_palette(7)[5], rep(hsv(h = c(0.5, 0.6), v = c(0.9, 1)), 2)), geom = list("geom_point", "geom_point", "geom_point", "geom_path"), alpha = list(0.5, 0.5, 0.5, 0.5), title = "DATA1 + DATA2 + DATA2 UNKNOWN DOTS", x.lim = x.range.plot, y.lim = y.range.plot, raster = raster, return = TRUE)
- if( ! is.null(tempo.graph$warn)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") FROM fun_gg_scatter():\n", tempo.graph$warn)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else{
- if(graph.in.file == FALSE){
- fun_open(pdf = FALSE)
- }
- fun_gg_empty_graph(text = "NO PLOT\nBECAUSE\nNO DATA2\nUNKNOWN DOTS", text.size = 8, title = "DATA2 + DATA2 UNKNOWN DOTS")
- }
- }
- }
- }
- # end plot
- if(warn.print == TRUE & ! is.null(warn)){
- options(warning.length = 8170)
- on.exit(warning(paste0("FROM ", function.name, ":\n\n", warn), call. = FALSE))
- on.exit(exp = options(warning.length = ini.warning.length), add = TRUE)
- }
- tempo.list <- list(data1.removed.row.nb = data1.removed.row.nb, data1.removed.rows = data1.removed.rows, data2.removed.row.nb = data2.removed.row.nb, data2.removed.rows = data2.removed.rows, hframe = hframe, vframe = vframe, data1.signif.dot = data1.signif.dot, data1.non.signif.dot = data1.non.signif.dot, data1.inconsistent.dot = data1.incon.dot, data2.signif.dot = data2.signif.dot, data2.non.signif.dot = data2.non.signif.dot, data2.unknown.dot = data2.unknown.dot, data2.inconsistent.dot = data2.incon.dot, axes = axes, warn = warn)
- return(tempo.list)
+# create a fake categ if NULL to deal with legend display
+if(is.null(categ)){
+categ <- vector("list", length(data1))
+categ[] <- "fake_categ"
+for(i2 in 1:length(data1)){
+data1[[i2]] <- cbind(data1[[i2]], fake_categ = "", stringsAsFactors = TRUE)
+if(geom[[i2]] == "geom_hline" | geom[[i2]] == "geom_vline"){
+data1[[i2]][, "fake_categ"] <- factor(paste0("Line_", 1:nrow(data1[[i2]])))
}
+}
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NULL categ ARGUMENT -> FAKE \"fake_categ\" COLUMN ADDED TO EACH DATA FRAME OF data1, AND FILLED WITH \"\"")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+# categ is not NULL anymore
+if(is.null(categ.class.order)){
+categ.class.order <- vector("list", length = length(data1))
+tempo.categ.class.order <- NULL
+for(i2 in 1:length(categ.class.order)){
+categ.class.order[[i2]] <- levels(data1[[i2]][, categ[[i2]]])
+names(categ.class.order)[i2] <- categ[[i2]]
+tempo.categ.class.order <- c(tempo.categ.class.order, ifelse(i2 != 1, "\n", ""), categ.class.order[[i2]])
+}
+if(any(unlist(legend.disp))){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") THE categ.class.order SETTING IS NULL. ALPHABETICAL ORDER WILL BE APPLIED FOR CLASS ORDERING:\n", paste(tempo.categ.class.order, collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+# end create a fake categ if NULL to deal with legend display
+# categ.class.order is not NULL anymore
+
+
+# vector of color with length as in levels(categ) of data1
+if(is.null(color)){
+color <- vector("list", length(data1))
+length.categ.list <- lapply(lapply(mapply(FUN = "[[", data1, categ, SIMPLIFY = FALSE), FUN = unique), FUN = function(x){length(x[ ! is.na(x)])})
+length.categ.list[sapply(categ, FUN = "==", "fake_categ")] <- 1 # when is.null(color), a single color for all the dots or lines of data[[i1]] that contain "fake_categ" category
+total.categ.length <- sum(unlist(length.categ.list), na.rm = TRUE)
+tempo.color <- fun_gg_palette(total.categ.length)
+tempo.count <- 0
+for(i2 in 1:length(data1)){
+color[[i2]] <- tempo.color[(1:length.categ.list[[i2]]) + tempo.count]
+tempo.count <- tempo.count + length.categ.list[[i2]]
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") NULL color ARGUMENT -> COLORS RESPECTIVELY ATTRIBUTED TO EACH CLASS OF ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i2, " OF categ ARGUMENT")), " (", categ[[i2]], ") IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i2, " OF data1 ARGUMENT")), ":\n", paste(color[[i2]], collapse = " "), "\n", paste(if(all(levels(data1[[i2]][, categ[[i2]]]) == "")){'\"\"'}else{levels(data1[[i2]][, categ[[i2]]])}, collapse = " "))
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+# end vector of color with length as in levels(categ) of data1
+# color is not NULL anymore
-################ Import
-
-######## fun_pack() #### check if R packages are present and import into the working environment
-# Check OK: clear to go Apollo
-fun_pack <- function(
- req.package,
- load = FALSE,
- lib.path = NULL
-){
- # AIM
- # check if the specified R packages are present in the computer and import them into the working environment
- # ARGUMENTS
- # req.package: character vector of package names to import
- # load: logical. Load the package into the environement (using library())? Interesting if packages are not in default folders or for checking the functions names of packages using search()
- # lib.path: optional character vector specifying the absolute pathways of the directories containing some of the listed packages in the req.package argument, if not in the default directories. Ignored if NULL
- # RETURN
- # nothing
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # fun_pack(req.package = "nopackage")
- # fun_pack(req.package = "ggplot2")
- # fun_pack(req.package = "ggplot2", lib.path = "blablabla")
- # DEBUGGING
- # req.package = "ggplot2" ; lib.path = "C:/Program Files/R/R-3.5.1/library"
- # req.package = "serpentine" ; lib.path = "C:/users/gael/appdata/roaming/python/python36/site-packages"
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = req.package, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = load, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(lib.path)){
- tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
- tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- }
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- if(is.null(lib.path)){
- lib.path <- .libPaths() # .libPaths(new = lib.path) # or .libPaths(new = c(.libPaths(), lib.path))
- }else{
- .libPaths(new = sub(x = lib.path, pattern = "/$|\\\\$", replacement = "")) # .libPaths(new = ) add path to default path. BEWARE: .libPaths() does not support / at the end of a submitted path. Thus check and replace last / or \\ in path
- }
- tempo <- NULL
- for(i1 in 1:length(req.package)){
- if( ! req.package[i1] %in% rownames(utils::installed.packages(lib.loc = lib.path))){
- tempo <- c(tempo, req.package[i1])
- }
- }
- if( ! is.null(tempo)){
- tempo.cat <- paste0(
- "ERROR IN ",
- function.name,
- ": PACKAGE",
- ifelse(length(tempo) == 1, paste0("\n\n", tempo, "\n\n"), paste0("S\n", paste(tempo, collapse = "\n"), "\n")),
- "MUST BE INSTALLED IN",
- ifelse(length(lib.path) == 1, "", " ONE OF THESE FOLDERS"),
- ":\n",
- paste(lib.path, collapse = "\n")
- )
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else if(load == TRUE){
- suppressMessages(suppressWarnings(suppressPackageStartupMessages(library(req.package[i1], lib.loc = lib.path, quietly = TRUE, character.only = TRUE))))
- }
+# last check
+for(i1 in 1:length(data1)){
+if(categ[[i1]] != "fake_categ" & length(color[[i1]]) != length(unique(data1[[i1]][, categ[[i1]]]))){
+tempo.cat <- paste0("ERROR IN ", function.name, " LAST CHECK: ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " MUST HAVE THE LENGTH OF LEVELS OF ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), "\nHERE IT IS COLOR LENGTH ", length(color[[i1]]), " VERSUS CATEG LEVELS LENGTH ", length(unique(data1[[i1]][, categ[[i1]]])))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else if(categ[[i1]] == "fake_categ" & length(color[[i1]]) != 1){
+tempo.cat <- paste0("ERROR IN ", function.name, " LAST CHECK: ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " MUST HAVE LENGTH 1 WHEN ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IS NULL\nHERE IT IS COLOR LENGTH ", length(color[[i1]]))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
}
-
-
-######## fun_python_pack() #### check if python packages are present
-
-
-# Check OK: clear to go Apollo
-fun_python_pack <- function(
- req.package,
- python.exec.path = NULL,
- lib.path = NULL,
- R.lib.path = NULL
-){
- # AIM
- # check if the specified python packages are present in the computer (no import)
- # WARNINGS
- # for python 3.7. Previous versions return an error "Error in sys$stdout$flush() : attempt to apply non-function"
- # ARGUMENTS
- # req.package: character vector of package names to import
- # python.exec.path: optional character vector specifying the absolute pathways of the executable python file to use (associated to the packages to use). If NULL, the reticulate::import_from_path() function used in fun_python_pack() seeks for an available version of python.exe, and then uses python_config(python_version, required_module, python_versions). But might not be the correct one for the lib.path parameter specified. Thus, it is recommanded to do not leave NULL, notably when using computing clusters
- # lib.path: optional character vector specifying the absolute pathways of the directories containing some of the listed packages in the req.package argument, if not in the default directories
- # R.lib.path: absolute path of the reticulate packages, if not in the default folders
- # RETURN
- # nothing
- # REQUIRED PACKAGES
- # reticulate
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # fun_pack()
- # EXAMPLES
- # example of error message
- # fun_python_pack(req.package = "nopackage")
- # example without error message (require the installation of the python serpentine package from https://github.com/koszullab/serpentine
- # fun_python_pack(req.package = "serpentine", python.exec.path = "C:/ProgramData/Anaconda3/python.exe", lib.path = "c:/programdata/anaconda3/lib/site-packages/")
- # another example of error message
- # fun_python_pack(req.package = "serpentine", lib.path = "blablabla")
- # DEBUGGING
- # req.package = "serpentine" ; python.exec.path = "C:/ProgramData/Anaconda3/python.exe" ; lib.path = "c:/programdata/anaconda3/lib/site-packages/" ; R.lib.path = NULL
- # req.package = "bad" ; lib.path = NULL ; R.lib.path = NULL
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(length(utils::find("fun_pack", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_pack() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = req.package, class = "character", fun.name = function.name) ; eval(ee)
- if( ! is.null(python.exec.path)){
- tempo <- fun_check(data = python.exec.path, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- if( ! all(file.exists(python.exec.path))){ # separation to avoid the problem of tempo$problem == FALSE and python.exec.path == NA
- tempo.cat <- paste0("ERROR IN ", function.name, ": FILE PATH INDICATED IN THE python.exec.path ARGUMENT DOES NOT EXISTS:\n", paste(python.exec.path, collapse = "\n"))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- }
- if( ! is.null(lib.path)){
- tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
- tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- }
- if( ! is.null(R.lib.path)){
- tempo <- fun_check(data = R.lib.path, class = "character", fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- if( ! all(dir.exists(R.lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and R.lib.path == NA
- tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE R.lib.path ARGUMENT DOES NOT EXISTS:\n", paste(R.lib.path, collapse = "\n"))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- }
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # package checking
- fun_pack(req.package = "reticulate", lib.path = R.lib.path)
- # end package checking
- # main code
- if(is.null(python.exec.path)){
- python.exec.path <- reticulate::py_run_string("
-import sys ;
-path_lib = sys.path
-") # python string
- python.exec.path <- python.exec.path$path_lib
- }
- if(is.null(lib.path)){
- lib.path <- reticulate::py_run_string("
-import sys ;
-path_lib = sys.path
-") # python string
- lib.path <- lib.path$path_lib
- }
- reticulate::use_python(Sys.which(python.exec.path), required = TRUE) # required to avoid the use of erratic python exec by reticulate::import_from_path()
- for(i1 in 1:length(req.package)){
- tempo.try <- vector("list", length = length(lib.path))
- for(i2 in 1:length(lib.path)){
- tempo.try[[i2]] <- suppressWarnings(try(reticulate::import_from_path(req.package[i1], path = lib.path[i2]), silent = TRUE))
- tempo.try[[i2]] <- suppressWarnings(try(reticulate::import_from_path(req.package[i1], path = lib.path[i2]), silent = TRUE)) # done twice to avoid the error message about flushing present the first time but not the second time. see https://stackoverflow.com/questions/57357001/reticulate-1-13-error-in-sysstdoutflush-attempt-to-apply-non-function
- }
- if(all(sapply(tempo.try, FUN = grepl, pattern = "[Ee]rror"))){
- print(tempo.try)
- tempo.cat <- paste0("ERROR IN ", function.name, ": PACKAGE ", req.package[i1], " MUST BE INSTALLED IN THE MENTIONNED DIRECTORY:\n", paste(lib.path, collapse = "\n"))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- } # else{
- # suppressMessages(suppressWarnings(suppressPackageStartupMessages(assign(req.package[i1], reticulate::import(req.package[i1]))))) # not required because try() already evaluates
- # }
- }
}
+# end last check
-################ Print / Exporting results (text & tables)
-
-######## fun_report() #### print string or data object into output file
-# Check OK: clear to go Apollo
-fun_report <- function(
- data,
- output = "results.txt",
- path = "C:/Users/Gael/Desktop/",
- no.overwrite = TRUE,
- rownames.kept = FALSE,
- vector.cat = FALSE,
- noquote = TRUE,
- sep = 2
-){
- # AIM
- # log file function: print a character string or a data object into a same output file
- # ARGUMENTS
- # data: object to print in the output file. If NULL, nothing is done, with no warning
- # output: name of the output file
- # path: location of the output file
- # no.overwrite: (logical) if output file already exists, defines if the printing is appended (default TRUE) or if the output file content is erased before printing (FALSE)
- # rownames.kept: (logical) defines whether row names have to be removed or not in small tables (less than length.rows rows)
- # vector.cat (logical). If TRUE print a vector of length > 1 using cat() instead of capture.output(). Otherwise (default FALSE) the opposite
- # noquote: (logical). If TRUE no quote are present for the characters
- # sep: number of separating lines after printed data (must be integer)
- # RETURN
- # nothing
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # fun_report()
- # fun_report(data = 1:3, output = "results.txt", path = "C:/Users/Gael/Desktop", no.overwrite = TRUE, rownames.kept = FALSE, vector.cat = FALSE, noquote = FALSE, sep = 2)
- # DEBUGGING
- # data = 1:3 ; output = "results.txt" ; path = "C:/Users/Gael/Desktop" ; no.overwrite = TRUE ; rownames.kept = FALSE ; vector.cat = FALSE ; noquote = FALSE ; sep = 2 # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = output, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & output == ""){
- tempo.cat <- paste0("ERROR IN ", function.name, ": output ARGUMENT AS \"\" DOES NOT CORRESPOND TO A VALID FILE NAME")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- tempo <- fun_check(data = path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- if( ! all(dir.exists(path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
- tempo.cat <- paste0("ERROR IN ", function.name, ": path ARGUMENT DOES NOT CORRESPOND TO EXISTING DIRECTORY\n", paste(path, collapse = "\n"))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- tempo <- fun_check(data = no.overwrite, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = rownames.kept, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = vector.cat, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = noquote, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = sep, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # end argument checking
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # the 4 next lines are inactivated but kept because at a time, I might have a problem with data (solved with data = NULL). These 4 lines are just to know how to detect a missing argument. Important here because if data is not provided, print the code of the data function
- # arg.user.list <- as.list(match.call(expand.dots = FALSE))[-1] # recover all the arguments provided by the function user (excluding the argument with defaults values not provided by the user. Thus, it is really the list indicated by the user)
- # default.arg.list <- formals(fun = sys.function(sys.parent())) # list of all the arguments of the function with their default values (not the values of the user !). It seems that ls() as first line of the function provide the names of the arguments (empty, called, etc., or not)
- # arg.without.default.value <- sapply(default.arg.list, is.symbol) & sapply(sapply(default.arg.list, as.character), identical, "") # logical to detect argument without default values (these are typeof "symbol" and class "name" and empty character
- # if( ! all(names(default.arg.list)[arg.without.default.value] %in% names(arg.user.list))){ # test that the arguments with no null values are provided by the user
- # tempo.cat <- paste0("ERROR IN ", function.name, ": VALUE REQUIRED FOR THESE ARGUMENTS WITH NO DEFAULTS VALUES: ", paste(names(default.arg.list)[arg.without.default.value][ ! names(default.arg.list)[arg.without.default.value] %in% names(arg.user.list)], collapse = " "))
- # stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- # }
- # end argument checking
- # main code
- if( ! is.null(data)){
- if(all(class(data) == "data.frame") | all(class(data) == "table") | all(class(data) %in% c("matrix", "array"))){ # before R4.0.0, it was all(class(data) %in% c("matrix", "data.frame", "table"))
- if(rownames.kept == FALSE & all(class(data) == "data.frame") & nrow(data) != 0 & nrow(data) <= 4){ # for data frames with nrows <= 4
- rownames.output.tables <- ""
- length.rows <- nrow(data)
- for(i in 1:length.rows){ # replace the rownames of the first 4 rows by increasing number of spaces (because identical row names not allowed in data frames). This method cannot be extended to more rows as the printed data frame is shifted on the right because of "big empty rownames"
- rownames.output.tables <- c(rownames.output.tables, paste0(rownames.output.tables[i]," ", collapse=""))
- }
- row.names(data) <- rownames.output.tables[1:length.rows]
- }else if(rownames.kept == FALSE & (all(class(data) == "table") | all(class(data) %in% c("matrix", "array")))){ # before R4.0.0, it was & all(class(data) %in% c("matrix", "table"))
- rownames(data) <- rep("", nrow(data)) # identical row names allowed in matrices and tables
- }
- if(noquote == TRUE){
- utils::capture.output(noquote(data), file=paste0(path, "/", output), append = no.overwrite)
- }else{
- utils::capture.output(data, file=paste0(path, "/", output), append = no.overwrite)
- }
- }else if(is.vector(data) & all(class(data) != "list") & (length(data) == 1 | vector.cat == TRUE)){
- if(noquote == TRUE){
- cat(noquote(data), file= paste0(path, "/", output), append = no.overwrite)
- }else{
- cat(data, file= paste0(path, "/", output), append = no.overwrite)
- }
- }else if(all(mode(data) == "character")){ # characters (array, list, factor or vector with vector.cat = FALSE)
- if(noquote == TRUE){
- utils::capture.output(noquote(data), file=paste0(path, "/", output), append = no.overwrite)
- }else{
- utils::capture.output(data, file=paste0(path, "/", output), append = no.overwrite)
- }
- }else{ # other object (S4 for instance, which do not like noquote()
- utils::capture.output(data, file=paste0(path, "/", output), append = no.overwrite)
- }
- sep.final <- paste0(rep("\n", sep), collapse = "")
- write(sep.final, file= paste0(path, "/", output), append = TRUE) # add a sep
- }
+# conversion of geom_hline and geom_vline
+for(i1 in 1:length(data1)){
+if(geom[[i1]] == "geom_hline" | geom[[i1]] == "geom_vline"){
+final.data.frame <- data.frame()
+for(i3 in 1:nrow(data1[[i1]])){
+tempo.data.frame <- rbind(data1[[i1]][i3, ], data1[[i1]][i3, ], stringsAsFactors = TRUE)
+if(geom[[i1]] == "geom_hline"){
+tempo.data.frame[, x[[i1]]] <- x.lim
+}else if(geom[[i1]] == "geom_vline"){
+tempo.data.frame[, y[[i1]]] <- y.lim
+}else{
+tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 5")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
}
-
-
-######## fun_get_message() #### return error/warning/other messages of an expression (that can be exported)
-
-
-# Check OK: clear to go Apollo
-fun_get_message <- function(
- data,
- kind = "error",
- header = TRUE,
- print.no = FALSE,
- text = NULL,
- env = NULL
-){
- # AIM
- # evaluate an instruction written between "" and return the first of the error, or warning or standard (non error non warning) messages if ever exist
- # using argument print.no = FALSE, return NULL if no message, which is convenient in some cases
- # WARNINGS
- # Only the first message is returned
- # Always use the env argument when fun_get_message() is used inside functions
- # ARGUMENTS
- # data: character string to evaluate
- # kind: character string. Either "error" to get error messages, or "warning" to get warning messages, or "message" to get non error and non warning messages
- # header: logical. Add a header in the returned message?
- # print.no: logical. Print a message saying that no message reported?
- # text: character string added to the output message (even if no message exists and print.no is TRUE). Inactivated if header is FALSE
- # env: the name of an existing environment. NULL if not required
- # RETURN
- # the message or NULL if no message and print.no is FALSE
- # REQUIRED PACKAGES
- # none
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_check()
- # EXAMPLES
- # fun_get_message(data = "wilcox.test(c(1,1,3), c(1, 2, 4), paired = TRUE)", kind = "error", print.no = TRUE, text = "IN A")
- # fun_get_message(data = "wilcox.test(c(1,1,3), c(1, 2, 4), paired = TRUE)", kind = "warning", print.no = TRUE, text = "IN A")
- # fun_get_message(data = "wilcox.test(c(1,1,3), c(1, 2, 4), paired = TRUE)", kind = "message", print.no = TRUE, text = "IN A")
- # fun_get_message(data = "wilcox.test()", kind = "error", print.no = TRUE, text = "IN A")
- # fun_get_message(data = "sum(1)", kind = "error", print.no = TRUE, text = "IN A")
- # fun_get_message(data = "message('ahah')", kind = "error", print.no = TRUE, text = "IN A")
- # fun_get_message(data = "message('ahah')", kind = "message", print.no = TRUE, text = "IN A")
- # fun_get_message(data = "ggplot2::ggplot(data = data.frame(X = 1:10, stringsAsFactors = TRUE), mapping = ggplot2::aes(x = X)) + ggplot2::geom_histogram()", kind = "message", print.no = TRUE, text = "IN FUNCTION 1")
- # set.seed(1) ; obs1 <- data.frame(Time = c(rnorm(10), rnorm(10) + 2), Group1 = rep(c("G", "H"), each = 10), stringsAsFactors = TRUE) ; fun_get_message(data = 'fun_gg_boxplot(data = obs1, y = "Time", categ = "Group1")', kind = "message", print.no = TRUE, text = "IN FUNCTION 1")
- # DEBUGGING
- # data = "wilcox.test(c(1,1,3), c(1, 2, 4), paired = TRUE)" ; kind = "warning" ; header = TRUE ; print.no = FALSE ; text = NULL ; env = NULL # for function debugging
- # data = "sum(1)" ; kind = "warning" ; header = TRUE ; print.no = FALSE ; text = NULL ; env = NULL # for function debugging
- # set.seed(1) ; obs1 <- data.frame(Time = c(rnorm(10), rnorm(10) + 2), Group1 = rep(c("G", "H"), each = 10), stringsAsFactors = TRUE) ; data = 'fun_gg_boxplot(data1 = obs1, y = "Time", categ = "Group1")' ; kind = "warning" ; header = TRUE ; print.no = FALSE ; text = NULL ; env = NULL # for function debugging
- # data = "message('ahah')" ; kind = "error" ; header = TRUE ; print.no = TRUE ; text = "IN A" ; env = NULL
- # data = 'ggplot2::ggplot(data = data.frame(X = "a", stringsAsFactors = TRUE), mapping = ggplot2::aes(x = X)) + ggplot2::geom_histogram()' ; kind = "message" ; header = TRUE ; print.no = FALSE ; text = NULL # for function debugging
- # data = 'ggplot2::ggplot(data = data.frame(X = "a", stringsAsFactors = TRUE), mapping = ggplot2::aes(x = X)) + ggplot2::geom_histogram()' ; kind = "warning" ; header = TRUE ; print.no = FALSE ; text = NULL # for function debugging
- # data = "emmeans::emmeans(object = emm.rg, specs = contrast.var)" ; kind = "message" ; header = TRUE ; print.no = FALSE ; text = NULL ; env = NULL # for function debugging
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- # end function name
- # required function checking
- if(length(utils::find("fun_check", mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end required function checking
- # no need to use reserved words to avoid bugs, because it is local, and exists("tempo.warning", inherit = FALSE), never use the scope
- # argument checking
- # argument checking with fun_check()
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = data, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = kind, options = c("error", "warning", "message"), length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = print.no, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = header, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(text)){
- tempo <- fun_check(data = text, class = "character", length = 1, fun.name = function.name) ; eval(ee)
- }
- if( ! is.null(env)){
- tempo <- fun_check(data = env, class = "environment", fun.name = function.name) ; eval(ee) #
- }
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # end argument checking with fun_check()
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument checking
- # main code
- pdf(file = NULL) # send plots into a NULL file, no pdf file created
- window.nb <- dev.cur()
- invisible(dev.set(window.nb))
- # last warning cannot be used because suppressWarnings() does not modify last.warning present in the base evironment (created at first warning in a new R session), or warnings() # to reset the warning history : unlockBinding("last.warning", baseenv()) ; assign("last.warning", NULL, envir = baseenv())
- output <- NULL
- tempo.error <- try(suppressMessages(suppressWarnings(eval(parse(text = data), envir = if(is.null(env)){parent.frame()}else{env}))), silent = TRUE) # get error message, not warning or messages
- if(any(class(tempo.error) %in% c("gg", "ggplot"))){
- tempo.error <- try(suppressMessages(suppressWarnings(ggplot2::ggplot_build(tempo.error))), silent = TRUE)[1]
- }
- if(exists("tempo.error", inherit = FALSE) == TRUE){ # inherit = FALSE avoid the portee lexical and thus the declared word
- if( ! all(class(tempo.error) == "try-error")){ # deal with NULL and S4 objects. Old code: ! (all(class(tempo.error) == "try-error") & any(grepl(x = tempo.error, pattern = "^Error|^error|^ERROR"))) but problem with S4 objects. Old code : if((length(tempo.error) > 0 & ! any(grepl(x = tempo.error, pattern = "^Error|^error|^ERROR"))) | (length(tempo.error) == 0) ){ but problem when tempo.error is a list but added this did not work: | ! all(class(tempo.error) == "character")
- tempo.error <- NULL
- }
- }else{
- tempo.error <- NULL
- }
- if(kind == "error" & ! is.null(tempo.error)){ #
- if(header == TRUE){
- tempo.error[1] <- gsub(x = tempo.error[1], pattern = "^Error i|^error i|^ERROR I", replacement = "I")
- output <- paste0("ERROR MESSAGE REPORTED", ifelse(is.null(text), "", " "), text, ":\n", tempo.error[1]) #
- }else{
- output <- tempo.error[1] #
- }
- }else if(kind == "error" & is.null(tempo.error) & print.no == TRUE){
- output <- paste0("NO ERROR MESSAGE REPORTED", ifelse(is.null(text), "", " "), text)
- }else if(kind != "error" & ( ! is.null(tempo.error)) & print.no == TRUE){
- output <- paste0("NO ", ifelse(kind == "warning", "WARNING", "STANDARD (NON ERROR AND NON WARNING)"), " MESSAGE BECAUSE OF ERROR MESSAGE REPORTED", ifelse(is.null(text), "", " "), text)
- }else if(is.null(tempo.error)){
- fun.warning.capture <- function(expr){
- # from demo(error.catching) typed in the R console, coming from ?tryCatch
- # see also http://mazamascience.com/WorkingWithData/?p=912
- # return a character string or NULL
- # expr <- wilcox.test.default(c(1, 1, 3), c(1, 2, 4), paired = TRUE)
- W <- NULL
- w.handler <- function(w){ # warning handler
- W <<- w # send to the above env, i.e., the inside of the fun.warning.capture function
- invokeRestart("muffleWarning") # here w.handler() muffles all the warnings. See http://romainfrancois.blog.free.fr/index.php?post/2009/05/20/Disable-specific-warnings to muffle specific warnings and print others
- }
- output <- list(
- value = suppressMessages(withCallingHandlers(tryCatch(expr, error = function(e){e}), warning = w.handler)), # BEWARE: w.handler is a function written without (), like in other functions with FUN argument
- warning = W # processed by w.handler()
- )
- return(if(is.null(output$warning)){NULL}else{as.character(output$warning)})
- }
- tempo.warn <- fun.warning.capture(eval(parse(text = data), envir = if(is.null(env)){parent.frame()}else{env}))
- # warn.options.ini <- options()$warn ; options(warn = 1) ; tempo.warn <- utils::capture.output({tempo <- suppressMessages(eval(parse(text = data), envir = if(is.null(env)){parent.frame()}else{env}))}, type = "message") ; options(warn = warn.options.ini) # this recover warnings not messages and not errors but does not work in all enviroments
- tempo.message <- utils::capture.output({
- tempo <- suppressMessages(suppressWarnings(eval(parse(text = data), envir = if(is.null(env)){parent.frame()}else{env})))
- if(any(class(tempo) %in% c("gg", "ggplot"))){
- tempo <- ggplot2::ggplot_build(tempo)
- }else{
- tempo <- suppressWarnings(eval(parse(text = data), envir = if(is.null(env)){parent.frame()}else{env}))
- }
- }, type = "message") # recover messages not warnings and not errors
- if(kind == "warning" & ! is.null(tempo.warn)){
- if(length(tempo.warn) > 0){ # to avoid character(0)
- if( ! any(sapply(tempo.warn, FUN = "grepl", pattern = "() FUNCTION:$"))){
- tempo.warn <- paste(unique(tempo.warn), collapse = "\n") # if FALSE, means that the tested data is a special function. If TRUE, means that the data is a standard function. In that case, the output of capture.output() is two strings per warning messages: if several warning messages -> identical first string, which is removed in next messages by unique()
- }else{
- tempo.warn <- paste(tempo.warn, collapse = "\n")
- }
- if(header == TRUE){
- if(any(grepl(x = tempo.warn[[1]], pattern = "^simpleWarning i"))){
- tempo.warn[[1]] <- gsub(x = tempo.warn[[1]], pattern = "^Warning i", replacement = "I")
- }
- if(any(grepl(x = tempo.warn[[1]], pattern = "^Warning i"))){
- tempo.warn[[1]] <- gsub(x = tempo.warn[[1]], pattern = "^Warning i", replacement = "I")
- }
- output <- paste0("WARNING MESSAGE REPORTED", ifelse(is.null(text), "", " "), text, ":\n", tempo.warn) #
- }else{
- output <- tempo.warn #
- }
- }else{
- if(print.no == TRUE){
- output <- paste0("NO WARNING MESSAGE REPORTED", ifelse(is.null(text), "", " "), text)
- } # no need else{} here because output is already NULL at first
- }
- }else if(kind == "warning" & is.null(tempo.warn) & print.no == TRUE){
- output <- paste0("NO WARNING MESSAGE REPORTED", ifelse(is.null(text), "", " "), text)
- }else if(kind == "message" & exists("tempo.message", inherit = FALSE) == TRUE){ # inherit = FALSE avoid the portee lexical and thus the declared word
- if(length(tempo.message) > 0){ # if something is returned by capture.ouptput() (only in this env) with a length more than 1
- if(header == TRUE){
- output <- paste0("STANDARD (NON ERROR AND NON WARNING) MESSAGE REPORTED", ifelse(is.null(text), "", " "), text, ":\n", tempo.message) #
- }else{
- output <- tempo.message #
- }
- }else{
- if(print.no == TRUE){
- output <- paste0("NO STANDARD (NON ERROR AND NON WARNING) MESSAGE REPORTED", ifelse(is.null(text), "", " "), text)
- } # no need else{} here because output is already NULL at first
- }
- }else if(kind == "message" & exists("tempo.message", inherit = FALSE) == FALSE & print.no == TRUE){
- output <- paste0("NO STANDARD (NON ERROR AND NON WARNING) MESSAGE REPORTED", ifelse(is.null(text), "", " "), text)
- } # no need else{} here because output is already NULL at first
- } # no need else{} here because output is already NULL at first
- invisible(dev.off(window.nb)) # end send plots into a NULL file
- return(output) # do not use cat() because the idea is to reuse the message
+# 3 lines below inactivated because I put that above
+# if(is.null(categ[[i1]])){
+# data1[, "fake_categ"] <- paste0("Line_", i3)
+# }
+final.data.frame <- rbind(final.data.frame, tempo.data.frame, stringsAsFactors = TRUE)
}
+data1[[i1]] <- final.data.frame
+geom[[i1]] <- "geom_line"
+if(length(color[[i1]]) == 1){
+color[[i1]] <- rep(color[[i1]], length(unique(data1[[i1]][ , categ[[i1]]])))
+}else if(length(color[[i1]]) != length(unique(data1[[i1]][ , categ[[i1]]]))){
+tempo.cat <- paste0("ERROR IN ", function.name, " geom_hline AND geom_vline CONVERSION TO FIT THE XLIM AND YLIM LIMITS OF THE DATA: ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " MUST HAVE THE LENGTH OF LEVELS OF ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), "\nHERE IT IS COLOR LENGTH ", length(color[[i1]]), " VERSUS CATEG LEVELS LENGTH ", length(unique(data1[[i1]][, categ[[i1]]])))
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+}
+}
+# end conversion of geom_hline and geom_vline
-# add the new NA and NULL: remain all (slitherine, cute, anova, comat)
-# complete the fun_check(): slitherine anova, comat
-
-fun_gg_boxplot <- function(
- data1,
- y,
- categ,
- categ.class.order = NULL,
- categ.color = NULL,
- box.legend.name = NULL,
- box.fill = FALSE,
- box.width = 0.5,
- box.space = 0.1,
- box.line.size = 0.75,
- box.notch = FALSE,
- box.alpha = 1,
- box.mean = TRUE,
- box.whisker.kind = "std",
- box.whisker.width = 0,
- dot.color = grey(0.25),
- dot.categ = NULL,
- dot.categ.class.order = NULL,
- dot.legend.name = NULL,
- dot.tidy = FALSE,
- dot.tidy.bin.nb = 50,
- dot.jitter = 0.5,
- dot.seed = 2,
- dot.size = 3,
- dot.alpha = 0.5,
- dot.border.size = 0.5,
- dot.border.color = NULL,
- x.lab = NULL,
- x.angle = 0,
- y.lab = NULL,
- y.lim = NULL,
- y.log = "no",
- y.tick.nb = NULL,
- y.second.tick.nb = 1,
- y.include.zero = FALSE,
- y.top.extra.margin = 0.05,
- y.bottom.extra.margin = 0.05,
- stat.disp = "top",
- stat.disp.mean = FALSE,
- stat.size = 4,
- stat.dist = 5,
- stat.angle = 0,
- vertical = TRUE,
- text.size = 12,
- title = "",
- title.text.size = 8,
- legend.show = TRUE,
- legend.width = 0.5,
- article = TRUE,
- grid = FALSE,
- add = NULL,
- return = FALSE,
- return.ggplot = FALSE,
- return.gtable = TRUE,
- plot = TRUE,
- warn.print = FALSE,
- lib.path = NULL
-){
- # AIM
- # plot ggplot2 boxplots + dots + means
- # for ggplot2 specifications, see: https://ggplot2.tidyverse.org/articles/ggplot2-specs.html
- # WARNINGS
- # Rows containing NA in data1[, c(y, categ)] will be removed before processing, with a warning (see below)
- # Hinges are not computed like in the classical boxplot() function of R. See https://ggplot2.tidyverse.org/reference/geom_boxplot.html
- # To have a single box, please create a factor column with a single class and specify the name of this column in the categ argument. For a single set of grouped boxes, create a factor column with a single class and specify this column in categ argument as first element (i.e., as categ1, knowing that categ2 must also be specified in this situation). See categ argument below
- # The dot.alpha argument can alter the display of the color boxes when using pdf output
- # Size arguments (box.line.size, dot.size, dot.border.size, stat.size, text.size and title.text.size) are in mm. See Hadley comment in https://stackoverflow.com/questions/17311917/ggplot2-the-unit-of-size. See also http://sape.inf.usi.ch/quick-reference/ggplot2/size). Unit object are not accepted, but conversion can be used (e.g., grid::convertUnit(grid::unit(0.2, "inches"), "mm", valueOnly = TRUE))
- # Display seems to be done twice on Windows devices (like a blink). However, no double plots on pdf devices. Thus, the blink remains mysterious
- # To remove boxes and have only dots, use box.alpha = 0
- # ARGUMENTS
- # data1: data frame containing one column of quantitative values (see the y argument below) and one or two columns of categories (see the categ argument below). Duplicated column names are not allowed
- # y: character string of the data1 column name for y-axis (column containing numeric values). Numeric values will be split according to the classes of the column names indicated in the categ argument to generate the boxes and will also be used to plot the dots
- # categ: vector of character strings of the data1 column name for categories (column of characters or factors). Must be either one or two column names. If a single column name (further referred to as categ1), then one box per class of categ1. If two column names (further referred to as categ1 and categ2), then one box per class of categ2, which form a group of boxes in each class of categ1. WARNING: no empty classes allowed. To have a single box, create a factor column with a single class and specify the name of this column in the categ argument (here, no categ2 in categ argument). For a single set of grouped boxes, create a factor column with a single class and specify this column in categ argument as first element (i.e., as categ1), in addition to the already used category (as categ2 in this situation)
- # categ.class.order: list indicating the order of the classes of categ1 and categ2 represented on the boxplot (the first compartment for categ1 and and the second for categ2). If categ.class.order == NULL, classes are represented according to the alphabetical order. Some compartments can be NULL and others not. See the categ argument for categ1 and categ2 description
- # categ.color: vector of color character string for box frames (see the categ argument for categ1 and categ2 description)
- # If categ.color == NULL, default colors of ggplot2, whatever categ1 and categ2
- # If categ.color is non-null and only categ1 in categ argument, categ.color can be either:
- # (1) a single color string. All the boxes will have this color, whatever the number of classes of categ1
- # (2) a vector of string colors, one for each class of categ1. Each color will be associated according to categ.class.order of categ1
- # (3) a vector or factor of string colors, like if it was one of the column of data1 data frame. WARNING: a single color per class of categ1 and a single class of categ1 per color must be respected
- # Integers are also accepted instead of character strings, as long as above rules about length are respected. Integers will be processed by fun_gg_palette() using the maximal integer value among all the integers in categ.color (see fun_gg_palette())
- # If categ.color is non-null and categ1 and categ2 are specified, all the rules described above will apply to categ2 instead of categ1 (colors will be determined for boxes inside a group of boxes)
- # box.legend.name: character string of the legend title. If box.legend.name is NULL, then box.legend.name <- categ1 if only categ1 is present, and box.legend.name <- categ2 if categ1 and categ2 are present in the categ argument. Write "" if no legend required. See the categ argument for categ1 and categ2 description
- # box.fill: logical. Fill the box? If TRUE, the categ.color argument will be used to generate filled boxplots (the box frames being black) as well as filled outlier dots (the dot border being controlled by the dot.border.color argument). If all the dots are plotted (argument dot.color other than NULL), they will be over the boxes. If FALSE, the categ.color argument will be used to color the box frames and the outlier dot borders. If all the dots are plotted, they will be beneath the boxes
- # box.width: single numeric value (from 0 to 1) of width of either boxes or group of boxes
- # When categ argument has a single categ1 element (i.e., separate boxes. See the categ argument for categ1 and categ2 description), then each class of categ1 is represented by a single box. In that case, box.width argument defines each box width, from 0 (no box width) to 1 (max box width), but also the space between boxes (the code uses 1 - box.width for the box spaces). Of note, xmin and xmax of the fun_gg_boxplot() output report the box boundaries (around x-axis unit 1, 2, 3, etc., for each box)
- # When categ argument has a two categ1 and categ2 elements (i.e., grouped boxes), box.width argument defines the width allocated for each set of grouped boxes, from 0 (no group width) to 1 (max group width), but also the space between grouped boxes (the code uses 1 - box.width for the spaces). Of note, xmin and xmax of the fun_gg_boxplot() output report the box boundaries (around x-axis unit 1, 2, 3, etc., for each set of grouped box)
- # box.space: single numeric value (from 0 to 1) indicating the box separation inside grouped boxes, when categ argument has a two categ1 and categ2 elements. 0 means no space and 1 means boxes shrunk to a vertical line. Ignored if categ argument has a single categ1 element
- # box.line.size: single numeric value of line width of boxes and whiskers in mm
- # box.notch: logical. Notched boxplot? It TRUE, display notched boxplot, notches corresponding approximately to the 95% confidence interval of the median (the notch interval is exactly 1.58 x Inter Quartile Range (IQR) / sqrt(n), with n the number of values that made the box). If notch intervals between two boxes do not overlap, it can be interpreted as significant median differences
- # box.alpha: single numeric value (from 0 to 1) of box transparency (full transparent to full opaque, respectively). To remove boxplots, use box.alpha = 0
- # box.mean: logical. Add mean value? If TRUE, a diamond-shaped dot, with the horizontal diagonal corresponding to the mean value, is displayed over each boxplot
- # box.whisker.kind: range of the whiskers. Either "no" (no whiskers), or "std" (length of each whisker equal to 1.5 x Inter Quartile Range (IQR)), or "max" (length of the whiskers up or down to the most distant dot)
- # box.whisker.width: single numeric value (from 0 to 1) of the whisker width, with 0 meaning no whiskers and 1 meaning a width equal to the box width
- # dot.color: vector of color character string ruling the dot colors and the dot display. See the example section below for easier understanding of the rules described here
- # If NULL, no dots plotted
- # If "same", the dots will have the same colors as the respective boxplots
- # Otherwise, as in the rule (1), (2) or (3) described in the categ.color argument, except that in the possibility (3), the rule "a single color per class of categ and a single class of categ per color", does not have to be respected (for instance, each dot can have a different color). Colors will also depend on the dot.categ argument. If dot.categ is NULL, then colors will be applied to each class of the last column name specified in categ. If dot.categ is non-NULL, colors will be applied to each class of the column name specified in dot.categ. See examples
- # dot.categ: optional single character string of a column name (further referred to as categ3) of the data1 argument. This column of data1 will be used to generate a legend for dots, in addition to the legend for boxes. See the dot.color argument for details about the way the legend is built using the two dot.categ and dot.color arguments. If NULL, no legend created and the colors of dots will depend on dot.color and categ arguments (as explained in the dot.color argument)
- # dot.categ.class.order: optional vector of character strings indicating the order of the classes of categ3 (see the dot.categ argument). If dot.categ is non-NULL and dot.categ.class.order is NULL, classes are displayed in the legend according to the alphabetical order. Ignored if dot.categ is NULL
- # dot.legend.name: optional character string of the legend title for categ3 (see the dot.categ argument). If dot.legend.name == NULL, dot.categ value is used (name of the column in data1). Write "" if no legend required. Ignored if dot.categ is NULL
- # dot.tidy: logical. Nice dot spreading? If TRUE, use the geom_dotplot() function for a nice representation. WARNING: change the true quantitative coordinates of dots (i.e., y-axis values for vertical display) because of binning. Thus, the gain in aestheticism is associated with a loss in precision that can be very important. If FALSE, dots are randomly spread on the qualitative axis, using the dot.jitter argument (see below) keeping the true quantitative coordinates
- # dot.tidy.bin.nb: positive integer indicating the number of bins (i.e., nb of separations) of the y.lim range. Each dot will then be put in one of the bin, with a diameter of the width of the bin. In other words, increase the number of bins to have smaller dots. Not considered if dot.tidy is FALSE
- # dot.jitter: numeric value (from 0 to 1) of random dot horizontal dispersion (for vertical display), with 0 meaning no dispersion and 1 meaning dispersion in the corresponding box width interval. Not considered if dot.tidy is TRUE
- # dot.seed: integer value that set the random seed. Using the same number will generate the same dot jittering. Write NULL to have different jittering each time the same instruction is run. Ignored if dot.tidy is TRUE
- # dot.size: numeric value of dot diameter in mm. Not considered if dot.tidy is TRUE
- # dot.alpha: numeric value (from 0 to 1) of dot transparency (full transparent to full opaque, respectively)
- # dot.border.size: numeric value of border dot width in mm. Write zero for no dot border. If dot.tidy is TRUE, value 0 remove the border and other values leave the border without size control (geom_doplot() feature)
- # dot.border.color: single character color string defining the color of the dot border (same color for all the dots, whatever their categories). If dot.border.color == NULL, the border color will be the same as the dot color. A single integer is also accepted instead of a character string, that will be processed by fun_gg_palette()
- # x.lab: a character string or expression for x-axis legend. If NULL, character string of categ1 (see the categ argument for categ1 and categ2 description)
- # x.angle: integer value of the text angle for the x-axis numbers, using the same rules as in ggplot2. Positive values for counterclockwise rotation: 0 for horizontal, 90 for vertical, 180 for upside down etc. Negative values for clockwise rotation: 0 for horizontal, -90 for vertical, -180 for upside down etc.
- # y.lab: a character string or expression for y-axis legend. If NULL, character string of the y argument
- # y.lim: 2 numeric values indicating the range of the y-axis. Order matters (for inverted axis). If NULL, the range of the x column name of data1 will be used.
- # y.log: either "no", "log2" (values in the y argument column of the data1 data frame will be log2 transformed and y-axis will be log2 scaled) or "log10" (values in the y argument column of the data1 data frame will be log10 transformed and y-axis will be log10 scaled). WARNING: not possible to have horizontal boxes with a log axis, due to a bug in ggplot2 (see https://github.com/tidyverse/ggplot2/issues/881)
- # y.tick.nb: approximate number of desired values labeling the y-axis (i.e., main ticks, see the n argument of the the cute::fun_scale() function). If NULL and if y.log is "no", then the number of labeling values is set by ggplot2. If NULL and if y.log is "log2" or "log10", then the number of labeling values corresponds to all the exposant integers in the y.lim range (e.g., 10^1, 10^2 and 10^3, meaning 3 main ticks for y.lim = c(9, 1200)). WARNING: if non-NULL and if y.log is "log2" or "log10", labeling can be difficult to read (e.g., ..., 10^2, 10^2.5, 10^3, ...)
- # y.second.tick.nb: number of desired secondary ticks between main ticks. Ignored if y.log is other than "no" (log scale plotted). Use argument return = TRUE and see $plot$y.second.tick.values to have the values associated to secondary ticks. IF NULL, no secondary ticks
- # y.include.zero: logical. Does y.lim range include 0? Ignored if y.log is "log2" or "log10"
- # y.top.extra.margin: single proportion (between 0 and 1) indicating if extra margins must be added to y.lim. If different from 0, add the range of the axis multiplied by y.top.extra.margin (e.g., abs(y.lim[2] - y.lim[1]) * y.top.extra.margin) to the top of y-axis
- # y.bottom.extra.margin: idem as y.top.extra.margin but to the bottom of y-axis
- # stat.disp: add the median number above the corresponding box. Either NULL (no number shown), "top" (at the top of the plot region) or "above" (above each box)
- # stat.disp.mean: logical. Display mean numbers instead of median numbers? Ignored if stat.disp is NULL
- # stat.size: numeric value of the stat font size in mm. Ignored if stat.disp is NULL
- # stat.dist: numeric value of the stat distance in percentage of the y-axis range (stat.dist = 5 means move the number displayed at 5% of the y-axis range). Ignored if stat.disp is NULL or "top"
- # stat.angle: integer value of the angle of stat, using the same rules as in ggplot2. Positive values for counterclockwise rotation: 0 for horizontal, 90 for vertical, 180 for upside down etc. Negative values for clockwise rotation: 0 for horizontal, -90 for vertical, -180 for upside down etc.
- # vertical: logical. Vertical boxes? WARNING: will be automatically set to TRUE if y.log argument is other than "no". Indeed, not possible to have horizontal boxes with a log axis, due to a bug in ggplot2 (see https://github.com/tidyverse/ggplot2/issues/881)
- # text.size: numeric value of the font size of the (1) axis numbers, (2) axis labels and (3) texts in the graphic legend (in mm)
- # title: character string of the graph title
- # title.text.size: numeric value of the title font size in mm
- # legend.show: logical. Show legend? Not considered if categ argument is NULL, because this already generate no legend, excepted if legend.width argument is non-NULL. In that specific case (categ is NULL, legend.show is TRUE and legend.width is non-NULL), an empty legend space is created. This can be useful when desiring graphs of exactly the same width, whatever they have legends or not
- # legend.width: single proportion (between 0 and 1) indicating the relative width of the legend sector (on the right of the plot) relative to the width of the plot. Value 1 means that the window device width is split in 2, half for the plot and half for the legend. Value 0 means no room for the legend, which will overlay the plot region. Write NULL to inactivate the legend sector. In such case, ggplot2 will manage the room required for the legend display, meaning that the width of the plotting region can vary between graphs, depending on the text in the legend
- # article: logical. If TRUE, use an article theme (article like). If FALSE, use a classic related ggplot theme. Use the add argument (e.g., add = "+ggplot2::theme_classic()" for the exact classic ggplot theme
- # grid: logical. Draw lines in the background to better read the box values? Not considered if article == FALSE (grid systematically present)
- # add: character string allowing to add more ggplot2 features (dots, lines, themes, facet, etc.). Ignored if NULL
- # WARNING: (1) the string must start with "+", (2) the string must finish with ")" and (3) each function must be preceded by "ggplot2::". Example: "+ ggplot2::coord_flip() + ggplot2::theme_bw()"
- # If the character string contains the "ggplot2::theme" string, then the article argument of fun_gg_boxplot() (see above) is ignored with a warning
- # Handle the add argument with caution since added functions can create conflicts with the preexisting internal ggplot2 functions
- # WARNING: the call of objects inside the quotes of add can lead to an error if the name of these objects are some of the fun_gg_scatter() arguments. Indeed, the function will use the internal argument instead of the global environment object. Example article <- "a" in the working environment and add = '+ ggplot2::ggtitle(article)'. The risk here is to have TRUE as title. To solve this, use add = '+ ggplot2::ggtitle(get("article", envir = .GlobalEnv))'
- # return: logical. Return the graph parameters?
- # return.ggplot: logical. Return the ggplot object in the output list? Ignored if return argument is FALSE. WARNING: always assign the fun_gg_boxplot() function (e.g., a <- fun_gg_boxplot()) if return.ggplot argument is TRUE, otherwise, double plotting is performed. See $ggplot in the RETURN section below for more details
- # return.gtable: logical. Return the ggplot object as gtable of grobs in the output list? Ignored if plot argument is FALSE. Indeed, the graph must be plotted to get the grobs dispositions. See $gtable in the RETURN section below for more details
- # plot: logical. Plot the graphic? If FALSE and return argument is TRUE, graphical parameters and associated warnings are provided without plotting
- # warn.print: logical. Print warnings at the end of the execution? ? If FALSE, warning messages are never printed, but can still be recovered in the returned list. Some of the warning messages (those delivered by the internal ggplot2 functions) are not apparent when using the argument plot = FALSE
- # lib.path: character string indicating the absolute path of the required packages (see below). if NULL, the function will use the R library default folders
- # REQUIRED PACKAGES
- # ggplot2
- # gridExtra
- # lemon (in case of use in the add argument)
- # scales
- # REQUIRED FUNCTIONS FROM THE cute PACKAGE
- # fun_check()
- # fun_comp_1d()
- # fun_comp_2d()
- # fun_gg_just()
- # fun_gg_palette()
- # fun_inter_ticks()
- # fun_name_change()
- # fun_pack()
- # fun_round()
- # fun_scale()
- # RETURN
- # a boxplot if plot argument is TRUE
- # a list of the graph info if return argument is TRUE:
- # $data: the initial data with graphic information added
- # $stat: the graphic statistics (mostly equivalent to ggplot_build()$data[[2]])
- # $removed.row.nb: which rows have been removed due to NA/Inf detection in y and categ columns (NULL if no row removed)
- # $removed.rows: removed rows (NULL if no row removed)
- # $plot: the graphic box and dot coordinates
- # $dots: dot coordinates
- # $main.box: coordinates of boxes
- # $median: median coordinates
- # $sup.whisker: coordinates of top whiskers (y for base and y.end for extremities)
- # $inf.whisker: coordinates of bottom whiskers (y for base and y.end for extremities)
- # $sup.whisker.edge: coordinates of top whisker edges (x and xend)
- # $inf.whisker.edge: coordinates of bottom whisker edges(x and xend)
- # $mean: diamond mean coordinates (only if box.mean argument is TRUE)
- # $stat.display: coordinates of stat numbers (only if stat.disp argument is not NULL)
- # y.second.tick.positions: coordinates of secondary ticks (only if y.second.tick.nb argument is non-NULL or if y.log argument is different from "no")
- # y.second.tick.values: values of secondary ticks. NULL except if y.second.tick.nb argument is non-NULL or if y.log argument is different from "no")
- # $panel: the variable names used for the panels (NULL if no panels). WARNING: NA can be present according to ggplot2 upgrade to v3.3.0
- # $axes: the x-axis and y-axis info
- # $warn: the warning messages. Use cat() for proper display. NULL if no warning. WARNING: warning messages delivered by the internal ggplot2 functions are not apparent when using the argument plot = FALSE
- # $ggplot: ggplot object that can be used for reprint (use print(...$ggplot) or update (use ...$ggplot + ggplot2::...). NULL if return.ggplot argument is FALSE. Of note, a non-NULL $ggplot in the output list is sometimes annoying as the manipulation of this list prints the plot
- # $gtable: gtable object that can be used for reprint (use gridExtra::grid.arrange(...$ggplot) or with additionnal grobs (see the grob decomposition in the examples). NULL if return.ggplot argument is FALSE. Contrary to $ggplot, a non-NULL $gtable in the output list is not annoying as the manipulation of this list does not print the plot
- # EXAMPLE
- # DEBUGGING
- # set.seed(1) ; obs1 <- data.frame(Time = c(rnorm(10), rnorm(10) + 2), Categ1 = rep(c("G", "H"), each = 10), stringsAsFactors = TRUE) ; set.seed(NULL) ; obs1$Time[1:10] <- NA ; data1 = obs1 ; y = "Time" ; categ = c("Categ1") ; categ.class.order = NULL ; box.legend.name = NULL ; categ.color = c("green") ; box.fill = FALSE ; box.width = 0.5 ; box.space = 0.1 ; box.notch = FALSE ; box.line.size = 0.5 ; box.alpha = 0.5 ; box.mean = TRUE ; box.whisker.kind = "std" ; box.whisker.width = 0.5 ; dot.color = "black" ; dot.categ = "Categ1"; dot.categ.class.order = c("G", "H") ; dot.legend.name = NULL ; dot.tidy = TRUE ; dot.tidy.bin.nb = 50 ; dot.jitter = 0.25 ; dot.size = 3 ; dot.alpha = 0.5 ; dot.border.size = 0.5 ; dot.border.color = NULL ; dot.seed = 2 ; y.lim = NULL ; y.log = "no" ; y.tick.nb = NULL ; y.second.tick.nb = NULL ; y.include.zero = FALSE ; y.top.extra.margin = 0.05 ; y.bottom.extra.margin = 0.05 ; stat.disp = NULL ; stat.disp.mean = FALSE ; stat.size = 4 ; stat.dist = 2 ; stat.angle = 0 ; x.lab = NULL ; y.lab = NULL ; vertical = TRUE ; text.size = 12 ; title = "" ; title.text.size = 8 ; legend.show = TRUE ; legend.width = 0.5 ; x.angle = 0 ; article = FALSE ; grid = FALSE ; return = TRUE ; return.ggplot = FALSE ; return.gtable = FALSE ; plot = TRUE ; add = NULL ; warn.print = FALSE ; lib.path = NULL
- # set.seed(1) ; obs1 <- data.frame(Time = c(rnorm(10), rnorm(10) + 2), Categ1 = rep(c("G", "H"), each = 10), Categ2 = rep(c("A", "B"), time = 10), Categ3 = rep(c("I", "J"), time = 10), stringsAsFactors = TRUE) ; set.seed(NULL) ; obs1$Time[1:10] <- NA ; data1 = obs1 ; y = "Time" ; categ = c("Categ1", "Categ2") ; categ.class.order = list(c("G", "H"), c("A", "B")); box.legend.name = NULL ; categ.color = c("green", "blue") ; box.fill = FALSE ; box.width = 0.5 ; box.space = 0.1 ; box.notch = FALSE ; box.line.size = 0.5 ; box.alpha = 0.5 ; box.mean = TRUE ; box.whisker.kind = "std" ; box.whisker.width = 0.5 ; dot.color = "black" ; dot.categ = "Categ1" ; dot.categ.class.order = NULL ; dot.legend.name = NULL ; dot.tidy = TRUE ; dot.tidy.bin.nb = 30 ; dot.jitter = 0.25 ; dot.size = 3 ; dot.alpha = 0.5 ; dot.border.size = 0.5 ; dot.border.color = NULL ; dot.seed = 2 ; y.lim = NULL ; y.log = "no" ; y.tick.nb = NULL ; y.second.tick.nb = NULL ; y.include.zero = FALSE ; y.top.extra.margin = 0.05 ; y.bottom.extra.margin = 0.05 ; stat.disp = NULL ; stat.disp.mean = FALSE ; stat.size = 4 ; stat.dist = 2 ; stat.angle = 0 ; x.lab = NULL ; y.lab = NULL ; vertical = TRUE ; text.size = 12 ; title = "" ; title.text.size = 8 ; legend.show = TRUE ; legend.width = 0.5 ; x.angle = 0 ; article = FALSE ; grid = FALSE ; return = FALSE ; return.ggplot = FALSE ; return.gtable = FALSE ; plot = TRUE ; add = NULL ; warn.print = FALSE ; lib.path = NULL
- # set.seed(1) ; obs1 <- data.frame(Time = c(rnorm(10), rnorm(10) + 2), Categ1 = rep(c("G", "H"), each = 10), Categ2 = rep(c("A", "B"), time = 10), stringsAsFactors = TRUE) ; set.seed(NULL) ; data1 = obs1 ; y = "Time" ; categ = c("Categ1") ; categ.class.order = list(c("H", "G")); box.legend.name = NULL ; categ.color = c("blue") ; box.fill = FALSE ; box.width = 0.5 ; box.space = 0.1 ; box.notch = TRUE ; box.line.size = 1 ; box.alpha = 1 ; box.mean = FALSE ; box.whisker.kind = "max" ; box.whisker.width = 0 ; dot.color = "black" ; dot.categ = "Categ1" ; dot.categ.class.order = NULL ; dot.legend.name = NULL ; dot.tidy = TRUE ; dot.tidy.bin.nb = 30 ; dot.jitter = 0.25 ; dot.size = 3 ; dot.alpha = 0.5 ; dot.border.size = 0.5 ; dot.border.color = NULL ; dot.seed = 2 ; y.lim = NULL ; y.log = "log10" ; y.tick.nb = NULL ; y.second.tick.nb = NULL ; y.include.zero = FALSE ; y.top.extra.margin = 0.05 ; y.bottom.extra.margin = 0.05 ; stat.disp = "above" ; stat.disp.mean = FALSE ; stat.size = 4 ; stat.dist = 2 ; stat.angle = 0 ; x.lab = NULL ; y.lab = NULL ; vertical = TRUE ; text.size = 12 ; title = "" ; title.text.size = 8 ; legend.width = 0.5 ; legend.show = TRUE ; x.angle = 0 ; article = FALSE ; grid = FALSE ; return = FALSE ; return.ggplot = FALSE ; return.gtable = FALSE ; plot = TRUE ; add = NULL ; warn.print = FALSE ; lib.path = NULL
- # function name
- function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
- arg.names <- names(formals(fun = sys.function(sys.parent(n = 2)))) # names of all the arguments
- arg.user.setting <- as.list(match.call(expand.dots = FALSE))[-1] # list of the argument settings (excluding default values not provided by the user)
- # end function name
- # required function checking
- req.function <- c(
- "fun_comp_2d",
- "fun_gg_just",
- "fun_gg_palette",
- "fun_name_change",
- "fun_pack",
- "fun_check",
- "fun_round",
- "fun_scale",
- "fun_inter_ticks"
- )
- for(i1 in req.function){
- if(length(find(i1, mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nREQUIRED ", i1, "() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # end required function checking
- # reserved words to avoid bugs (names of dataframe columns used in this function)
- reserved.words <- c("categ.check", "categ.color", "dot.color", "dot.categ", "dot.max", "dot.min", "group", "PANEL", "group.check", "MEAN", "tempo.categ1", "tempo.categ2", "text.max.pos", "text.min.pos", "x", "x.y", "y", "y.check", "y_from.dot.max", "ymax", "tidy_group")
- # end reserved words to avoid bugs (used in this function)
- # argument primary checking
- # arg with no default values
- if(any(missing(data1) | missing(y) | missing(categ))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ARGUMENTS angle AND pos HAVE NO DEFAULT VALUE AND REQUIRE ONE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end arg with no default values
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo <- fun_check(data = data1, class = "data.frame", na.contain = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = y, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = categ, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- if( ! is.null(categ.class.order)){
- tempo <- fun_check(data = categ.class.order, class = "list", fun.name = function.name) ; eval(ee)
- }
- if( ! is.null(box.legend.name)){
- tempo <- fun_check(data = box.legend.name, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- }
- if( ! is.null(categ.color)){
- tempo1 <- fun_check(data = categ.color, class = "vector", mode = "character", na.contain = TRUE, fun.name = function.name)
- tempo2 <- fun_check(data = categ.color, class = "factor", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.check.color <- fun_check(data = categ.color, class = "integer", double.as.integer.allowed = TRUE, na.contain = TRUE, fun.name = function.name)$problem
- if(tempo.check.color == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": categ.color ARGUMENT MUST BE A FACTOR OR CHARACTER VECTOR OR INTEGER VECTOR") # integer possible because dealt above
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- }
- tempo <- fun_check(data = box.fill, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = box.width, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = box.space, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = box.line.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = box.notch, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = box.alpha, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = box.mean, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = box.whisker.kind, options = c("no", "std", "max"), length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = box.whisker.width, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(dot.color)){
- tempo1 <- fun_check(data = dot.color, class = "vector", mode = "character", na.contain = TRUE, fun.name = function.name)
- tempo2 <- fun_check(data = dot.color, class = "factor", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.check.color <- fun_check(data = dot.color, class = "integer", double.as.integer.allowed = TRUE, na.contain = TRUE, fun.name = function.name)$problem
- if(tempo.check.color == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": dot.color MUST BE A FACTOR OR CHARACTER VECTOR OR INTEGER VECTOR") # integer possible because dealt above
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- }
- if( ! is.null(dot.categ)){
- tempo <- fun_check(data = dot.categ, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- }
- if( ! is.null(dot.categ.class.order)){
- tempo <- fun_check(data = dot.categ.class.order, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- }
- if( ! is.null(dot.legend.name)){
- tempo <- fun_check(data = dot.legend.name, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = dot.tidy, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = dot.tidy.bin.nb, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = dot.jitter, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(dot.seed)){
- tempo <- fun_check(data = dot.seed, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = TRUE, fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = dot.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = dot.alpha, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = dot.border.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- if( ! is.null(dot.border.color)){
- tempo1 <- fun_check(data = dot.border.color, class = "vector", mode = "character", length = 1, fun.name = function.name)
- tempo2 <- fun_check(data = dot.border.color, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, fun.name = function.name)
- if((tempo1$problem == TRUE & tempo2$problem == TRUE) | (tempo1$problem == FALSE & tempo2$problem == TRUE & ! (all(dot.border.color %in% colors() | grepl(pattern = "^#", dot.border.color))))){ # check that all strings of low.color start by #
- tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.border.color ARGUMENT MUST BE (1) A HEXADECIMAL COLOR STRING STARTING BY #, OR (2) A COLOR NAME GIVEN BY colors(), OR (3) AN INTEGER VALUE\nHERE IT IS: ", paste(unique(dot.border.color), collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- if( ! is.null(x.lab)){
- if(all(class(x.lab) %in% "expression")){ # to deal with math symbols
- tempo <- fun_check(data = x.lab, class = "expression", length = 1, fun.name = function.name) ; eval(ee)
- }else{
- tempo <- fun_check(data = x.lab, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- }
- }
- tempo <- fun_check(data = x.angle, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, neg.values = TRUE, fun.name = function.name) ; eval(ee)
- if( ! is.null(y.lab)){
- if(all(class(y.lab) %in% "expression")){ # to deal with math symbols
- tempo <- fun_check(data = y.lab, class = "expression", length = 1, fun.name = function.name) ; eval(ee)
- }else{
- tempo <- fun_check(data = y.lab, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- }
- }
- if( ! is.null(y.lim)){
- tempo <- fun_check(data = y.lim, class = "vector", mode = "numeric", length = 2, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & any(y.lim %in% c(Inf, -Inf))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": y.lim ARGUMENT CANNOT CONTAIN -Inf OR Inf VALUES")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- tempo <- fun_check(data = y.log, options = c("no", "log2", "log10"), length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(y.tick.nb)){
- tempo <- fun_check(data = y.tick.nb, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & y.tick.nb < 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": y.tick.nb ARGUMENT MUST BE A NON NULL POSITIVE INTEGER")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- if( ! is.null(y.second.tick.nb)){
- tempo <- fun_check(data = y.second.tick.nb, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & y.second.tick.nb <= 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": y.second.tick.nb ARGUMENT MUST BE A NON NULL POSITIVE INTEGER")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- tempo <- fun_check(data = y.include.zero, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = y.top.extra.margin, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = y.bottom.extra.margin, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(stat.disp)){
- tempo <- fun_check(data = stat.disp, options = c("top", "above"), length = 1, fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = stat.disp.mean, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = stat.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = stat.dist, class = "vector", mode = "numeric", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = stat.angle, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, neg.values = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = vertical, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = text.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = title, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = title.text.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = legend.show, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(legend.width)){
- tempo <- fun_check(data = legend.width, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = article, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = grid, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(add)){
- tempo <- fun_check(data = add, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = return, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = return.ggplot, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = return.gtable, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = plot, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = warn.print, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(lib.path)){
- tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
- tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- }
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between == #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end argument primary checking
- # second round of checking and data preparation
- # dealing with NA arguments
- tempo.arg <- names(arg.user.setting) # values provided by the user
- tempo.log <- sapply(lapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = is.na), FUN = any) & lapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = length) == 1 # no argument provided by the user can be just NA
- if(any(tempo.log) == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ":\n", ifelse(sum(tempo.log, na.rm = TRUE) > 1, "THESE ARGUMENTS\n", "THIS ARGUMENT\n"), paste0(tempo.arg[tempo.log], collapse = "\n"),"\nCANNOT JUST BE NA")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end dealing with NA arguments
- # dealing with NULL arguments
- tempo.arg <-c(
- "data1",
- "y",
- "categ",
- "box.fill",
- "box.width",
- "box.space",
- "box.line.size",
- "box.notch",
- "box.alpha",
- "box.mean",
- "box.whisker.kind",
- "box.whisker.width",
- # "dot.color", # inactivated because can be null
- "dot.tidy",
- "dot.tidy.bin.nb",
- "dot.jitter",
- "dot.size",
- "dot.alpha",
- "dot.border.size",
- "x.angle",
- "y.log",
- "y.include.zero",
- "y.top.extra.margin",
- "y.bottom.extra.margin",
- "stat.disp.mean",
- "stat.size",
- "stat.dist",
- "stat.angle",
- "vertical",
- "text.size",
- "title",
- "title.text.size",
- "legend.show",
- # "legend.width", # inactivated because can be null
- "article",
- "grid",
- "return",
- "return.ggplot",
- "return.gtable",
- "plot",
- "warn.print"
- )
- tempo.log <- sapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = is.null)
- if(any(tempo.log) == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ":\n", ifelse(sum(tempo.log, na.rm = TRUE) > 1, "THESE ARGUMENTS\n", "THIS ARGUMENT\n"), paste0(tempo.arg[tempo.log], collapse = "\n"),"\nCANNOT BE NULL")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end dealing with NULL arguments
- # code that protects set.seed() in the global environment
- # see also Protocol 100-rev0 Parallelization in R.docx
- if(exists(".Random.seed", envir = .GlobalEnv)){ # if .Random.seed does not exists, it means that no random operation has been performed yet in any R environment
- tempo.random.seed <- .Random.seed
- on.exit(assign(".Random.seed", tempo.random.seed, env = .GlobalEnv))
- }else{
- on.exit(set.seed(NULL)) # inactivate seeding -> return to complete randomness
- }
- set.seed(dot.seed)
- # end code that protects set.seed() in the global environment
- ini.warning.length <- options()$warning.length
- warn <- NULL
- warn.count <- 0
- if(any(duplicated(names(data1)))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nDUPLICATED COLUMN NAMES OF data1 ARGUMENT NOT ALLOWED:\n", paste(names(data1)[duplicated(names(data1))], collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! (y %in% names(data1))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\ny ARGUMENT MUST BE A COLUMN NAME OF data1")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- tempo <- fun_check(data = data1[, y], data.name = "y COLUMN OF data1", class = "vector", mode = "numeric", na.contain = TRUE, fun.name = function.name)
- if(tempo$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, "\ny ARGUMENT MUST BE NUMERIC COLUMN IN data1")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- if(length(categ) > 2){
- tempo.cat <- paste0("ERROR IN ", function.name, "\ncateg ARGUMENT CANNOT HAVE MORE THAN 2 COLUMN NAMES OF data1")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else if( ! all(categ %in% names(data1))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\ncateg ARGUMENT MUST BE COLUMN NAMES OF data1. HERE IT IS:\n", paste(categ, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # reserved word checking
- if(any(names(data1) %in% reserved.words)){
- if(any(duplicated(names(data1)))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nDUPLICATED COLUMN NAMES OF data1 ARGUMENT NOT ALLOWED:\n", paste(names(data1)[duplicated(names(data1))], collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! is.null(dot.categ)){
- if(dot.categ %in% categ){
- reserved.words <- c(reserved.words, paste0(dot.categ, "_DOT")) # paste0(dot.categ, "_DOT") is added to the reserved words because in such situation, a new column will be added to data1 that is named paste0(dot.categ, "_DOT")
- }
- }
- tempo.output <- fun_name_change(names(data1), reserved.words)
- for(i2 in 1:length(tempo.output$ini)){ # a loop to be sure to take the good ones
- names(data1)[names(data1) == tempo.output$ini[i2]] <- tempo.output$post[i2]
- if(any(y == tempo.output$ini[i2])){
- y[y == tempo.output$ini[i2]] <- tempo.output$post[i2]
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") IN y ARGUMENT (COLUMN NAMES OF data1 ARGUMENT),\n", tempo.output$ini[i2], " HAS BEEN REPLACED BY ", tempo.output$post[i2], "\nBECAUSE RISK OF BUG AS SOME NAMES IN y ARGUMENT ARE RESERVED WORD USED BY THE ", function.name, " FUNCTION")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- # WARNING: names of y argument potentially replaced
- if(any(categ == tempo.output$ini[i2])){
- categ[categ == tempo.output$ini[i2]] <- tempo.output$post[i2]
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") IN categ ARGUMENT (COLUMN NAMES OF data1 ARGUMENT),\n", tempo.output$ini[i2], " HAS BEEN REPLACED BY ", tempo.output$post[i2], "\nBECAUSE RISK OF BUG AS SOME NAMES IN categ ARGUMENT ARE RESERVED WORD USED BY THE ", function.name, " FUNCTION")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- # WARNING: names of categ argument potentially replaced
- if( ! is.null(dot.categ)){
- if(any(dot.categ == tempo.output$ini[i2])){
- dot.categ[dot.categ == tempo.output$ini[i2]] <- tempo.output$post[i2]
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") IN dot.categ ARGUMENT (COLUMN NAMES OF data1 ARGUMENT),\n", tempo.output$ini[i2], " HAS BEEN REPLACED BY ", tempo.output$post[i2], "\nBECAUSE RISK OF BUG AS SOME NAMES IN dot.categ ARGUMENT ARE RESERVED WORD USED BY THE ", function.name, " FUNCTION")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- # WARNING: names of dot.categ argument potentially replaced
- }
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") REGARDING COLUMN NAMES REPLACEMENT, THE NAMES\n", paste(tempo.output$ini, collapse = " "), "\nHAVE BEEN REPLACED BY\n", paste(tempo.output$post, collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- if( ! (is.null(add) | is.null(tempo.output$ini))){
- if(grepl(x = add, pattern = paste(tempo.output$ini, collapse = "|"))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nDETECTION OF COLUMN NAMES OF data1 IN THE add ARGUMENT STRING, THAT CORRESPOND TO RESERVED STRINGS FOR ", function.name, "\nCOLUMN NAMES HAVE TO BE CHANGED\nTHE PROBLEMATIC COLUMN NAMES ARE SOME OF THESE NAMES:\n", paste(tempo.output$ini, collapse = " "), "\nIN THE DATA FRAME OF data1 AND IN THE STRING OF add ARGUMENT, TRY TO REPLACE NAMES BY:\n", paste(tempo.output$post, collapse = " "), "\n\nFOR INFORMATION, THE RESERVED WORDS ARE:\n", paste(reserved.words, collapse = "\n"))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- }
- if( ! (is.null(add))){
- if(any(sapply(X = arg.names, FUN = grepl, x = add))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NAMES OF ", function.name, " ARGUMENTS DETECTED IN THE add STRING:\n", paste(arg.names[sapply(X = arg.names, FUN = grepl, x = add)], collapse = "\n"), "\nRISK OF WRONG OBJECT USAGE INSIDE ", function.name)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- # end reserved word checking
- # verif of add
- if( ! is.null(add)){
- if( ! grepl(pattern = "^\\s*\\+", add)){ # check that the add string start by +
- tempo.cat <- paste0("ERROR IN ", function.name, ": add ARGUMENT MUST START WITH \"+\": ", paste(unique(add), collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else if( ! grepl(pattern = "(ggplot2|lemon)\\s*::", add)){ #
- tempo.cat <- paste0("ERROR IN ", function.name, ": FOR EASIER FUNCTION DETECTION, add ARGUMENT MUST CONTAIN \"ggplot2::\" OR \"lemon::\" IN FRONT OF EACH GGPLOT2 FUNCTION: ", paste(unique(add), collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else if( ! grepl(pattern = ")\\s*$", add)){ # check that the add string finished by )
- tempo.cat <- paste0("ERROR IN ", function.name, ": add ARGUMENT MUST FINISH BY \")\": ", paste(unique(add), collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # end verif of add
- # management of add containing facet
- facet.categ <- NULL
- if( ! is.null(add)){
- facet.check <- TRUE
- tempo <- unlist(strsplit(x = add, split = "\\s*\\+\\s*(ggplot2|lemon)\\s*::\\s*")) #
- tempo <- sub(x = tempo, pattern = "^facet_wrap", replacement = "ggplot2::facet_wrap")
- tempo <- sub(x = tempo, pattern = "^facet_grid", replacement = "ggplot2::facet_grid")
- tempo <- sub(x = tempo, pattern = "^facet_rep", replacement = "lemon::facet_rep")
- if(any(grepl(x = tempo, pattern = "ggplot2::facet_wrap|lemon::facet_rep_wrap"))){
- tempo1 <- suppressWarnings(eval(parse(text = tempo[grepl(x = tempo, pattern = "ggplot2::facet_wrap|lemon::facet_rep_wrap")])))
- facet.categ <- names(tempo1$params$facets)
- tempo.text <- "facet_wrap OR facet_rep_wrap"
- facet.check <- FALSE
- }else if(grepl(x = add, pattern = "ggplot2::facet_grid|lemon::facet_rep_grid")){
- tempo1 <- suppressWarnings(eval(parse(text = tempo[grepl(x = tempo, pattern = "ggplot2::facet_grid|lemon::facet_rep_grid")])))
- facet.categ <- c(names(tempo1$params$rows), names(tempo1$params$cols))
- tempo.text <- "facet_grid OR facet_rep_grid"
- facet.check <- FALSE
- }
- if(facet.check == FALSE & ! all(facet.categ %in% names(data1))){ # WARNING: all(facet.categ %in% names(data1)) is TRUE when facet.categ is NULL
- tempo.cat <- paste0("ERROR IN ", function.name, "\nDETECTION OF \"", tempo.text, "\" STRING IN THE add ARGUMENT BUT PROBLEM OF VARIABLE DETECTION (COLUMN NAMES OF data1)\nTHE DETECTED VARIABLES ARE:\n", paste(facet.categ, collapse = " "), "\nTHE data1 COLUMN NAMES ARE:\n", paste(names(data1), collapse = " "), "\nPLEASE REWRITE THE add STRING AND RERUN")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # end management of add containing facet
- # conversion of categ columns in data1 into factors
- for(i1 in 1:length(categ)){
- tempo1 <- fun_check(data = data1[, categ[i1]], data.name = paste0("categ NUMBER ", i1, " OF data1"), class = "vector", mode = "character", na.contain = TRUE, fun.name = function.name)
- tempo2 <- fun_check(data = data1[, categ[i1]], data.name = paste0("categ NUMBER ", i1, " OF data1"), class = "factor", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, "\n", paste0("categ NUMBER ", i1, " OF data1"), " MUST BE A FACTOR OR CHARACTER VECTOR")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else if(tempo1$problem == FALSE){ # character vector
- if(box.alpha != 0){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") IN categ NUMBER ", i1, " IN data1, THE CHARACTER COLUMN HAS BEEN CONVERTED TO FACTOR, WITH LEVELS ACCORDING TO THE ALPHABETICAL ORDER")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- data1[, categ[i1]] <- factor(data1[, categ[i1]]) # if already a factor, change nothing, if characters, levels according to alphabetical order
- }
- # OK: all the categ columns of data1 are factors from here
- # end conversion of categ columns in data1 into factors
-
-
-
- # management of log scale and Inf removal
- if(any(( ! is.finite(data1[, y])) & ( ! is.na(data1[, y])))){ # is.finite also detects NA: ( ! is.finite(data1[, y])) & ( ! is.na(data1[, y])) detects only Inf
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") PRESENCE OF -Inf OR Inf VALUES IN THE ", y, " COLUMN OF THE data1 ARGUMENT AND CORRESPONDING ROWS REMOVED (SEE $removed.row.nb AND $removed.rows)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- data1.ini <- data1 # strictly identical to data1 except that in data1 y is log converted if and only if y.log != "no"
- if(y.log != "no"){
- tempo1 <- ! is.finite(data1[, y]) # where are initial NA and Inf
- data1[, y] <- suppressWarnings(get(y.log)(data1[, y]))# no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
- if(any( ! (tempo1 | is.finite(data1[, y])))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") LOG CONVERSION INTRODUCED -Inf OR Inf OR NaN VALUES IN THE ", y, " COLUMN OF THE data1 ARGUMENT AND CORRESPONDING ROWS REMOVED (SEE $removed.row.nb AND $removed.rows)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- # Inf removal
- if(any(( ! is.finite(data1[, y])) & ( ! is.na(data1[, y])))){ # is.finite also detects NA: ( ! is.finite(data1[, y])) & ( ! is.na(data1[, y])) detects only Inf
- removed.row.nb <- which(( ! is.finite(data1[, y])) & ( ! is.na(data1[, y])))
- removed.rows <- data1.ini[removed.row.nb, ] # here data1.ini used to have the y = O rows that will be removed because of Inf creation after log transformation
- data1 <- data1[-removed.row.nb, ] #
- data1.ini <- data1.ini[-removed.row.nb, ] #
- }else{
- removed.row.nb <- NULL
- removed.rows <- data.frame(stringsAsFactors = FALSE)
- }
- # From here, data1 and data.ini have no more Inf
- # end Inf removal
- if(y.log != "no" & ! is.null(y.lim)){
- if(any(y.lim <= 0)){
- tempo.cat <- paste0("ERROR IN ", function.name, "\ny.lim ARGUMENT CANNOT HAVE ZERO OR NEGATIVE VALUES WITH THE y.log ARGUMENT SET TO ", y.log, ":\n", paste(y.lim, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else if(any( ! is.finite(if(y.log == "log10"){log10(y.lim)}else{log2(y.lim)}))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\ny.lim ARGUMENT RETURNS INF/NA WITH THE y.log ARGUMENT SET TO ", y.log, "\nAS SCALE COMPUTATION IS ", ifelse(y.log == "log10", "log10", "log2"), ":\n", paste(if(y.log == "log10"){log10(y.lim)}else{log2(y.lim)}, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- if(y.log != "no" & y.include.zero == TRUE){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") y.log ARGUMENT SET TO ", y.log, " AND y.include.zero ARGUMENT SET TO TRUE -> y.include.zero ARGUMENT RESET TO FALSE BECAUSE 0 VALUE CANNOT BE REPRESENTED IN LOG SCALE")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- y.include.zero <- FALSE
- }
- if(y.log != "no" & vertical == FALSE){
- vertical <- TRUE
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") BECAUSE OF A BUG IN ggplot2, CANNOT FLIP BOXES HORIZONTALLY WITH A Y.LOG SCALE -> vertical ARGUMENT RESET TO TRUE")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- # end management of log scale and Inf removal
- # na detection and removal (done now to be sure of the correct length of categ)
- column.check <- unique(c(y, categ, if( ! is.null(dot.color) & ! is.null(dot.categ)){dot.categ}, if( ! is.null(facet.categ)){facet.categ})) # dot.categ because can be a 3rd column of data1, categ.color and dot.color will be tested later
- if(any(is.na(data1[, column.check]))){ # data1 used here instead of data1.ini in case of new NaN created by log conversion (neg values)
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NA DETECTED IN COLUMNS OF data1 AND CORRESPONDING ROWS REMOVED (SEE $removed.row.nb AND $removed.rows)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- for(i2 in 1:length(column.check)){
- if(any(is.na(data1[, column.check[i2]]))){
- tempo.warn <- paste0("NA REMOVAL DUE TO COLUMN ", column.check[i2], " OF data1")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n", tempo.warn)))
- }
- }
- tempo <- unique(unlist(lapply(lapply(c(data1[column.check]), FUN = is.na), FUN = which)))
- removed.row.nb <- c(removed.row.nb, tempo) # removed.row.nb created to remove Inf
- removed.rows <- rbind(removed.rows, data1.ini[tempo, ], stringsAsFactors = FALSE) # here data1.ini used to have the non NA rows that will be removed because of NAN creation after log transformation (neg values for instance)
- column.check <- column.check[ ! column.check == y] # remove y to keep quali columns
- if(length(tempo) != 0){
- data1 <- data1[-tempo, ] # WARNING tempo here and not removed.row.nb because the latter contain more numbers thant the former
- data1.ini <- data1.ini[-tempo, ] # WARNING tempo here and not removed.row.nb because the latter contain more numbers than the former
- for(i3 in 1:length(column.check)){
- if(any( ! unique(removed.rows[, column.check[i3]]) %in% unique(data1[, column.check[i3]]))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") IN COLUMN ", column.check[i3], " OF data1, THE FOLLOWING CLASSES HAVE DISAPPEARED AFTER NA/Inf REMOVAL (IF COLUMN USED IN THE PLOT, THIS CLASS WILL NOT BE DISPLAYED):\n", paste(unique(removed.rows[, column.check[i3]])[ ! unique(removed.rows[, column.check[i3]]) %in% unique(data1[, column.check[i3]])], collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- }
- count.categ <- 0
- for(i2 in 1:length(column.check)){
- if(column.check[i2] %in% categ){
- count.categ <- count.categ + 1
- }
- if(column.check[i2] == categ[count.categ]){
- categ.class.order[count.categ] <- list(levels(data1[, column.check[i2]])[levels(data1[, column.check[i2]]) %in% unique(data1[, column.check[i2]])]) # remove the absent color in the character vector
- data1[, column.check[i2]] <- factor(as.character(data1[, column.check[i2]]), levels = unique(categ.class.order[[count.categ]]))
- }
- if( ! is.null(dot.color) & ! is.null(dot.categ)){ # reminder : dot.categ cannot be a column name of categ anymore (because in that case dot.categ name is changed into "..._DOT"
- if(column.check[i2] == dot.categ){
- dot.categ.class.order <- levels(data1[, column.check[i2]])[levels(data1[, column.check[i2]]) %in% unique(data1[, column.check[i2]])] # remove the absent color in the character vector
- data1[, column.check[i2]] <- factor(as.character(data1[, column.check[i2]]), levels = unique(dot.categ.class.order))
- }
- }
- if(column.check[i2] %in% facet.categ){ # works if facet.categ == NULL this method should keep the order of levels when removing some levels
- tempo.levels <- levels(data1[, column.check[i2]])[levels(data1[, column.check[i2]]) %in% unique(as.character(data1[, column.check[i2]]))]
- data1[, column.check[i2]] <- factor(as.character(data1[, column.check[i2]]), levels = tempo.levels)
- }
- }
- }
- # end na detection and removal (done now to be sure of the correct length of categ)
- # From here, data1 and data.ini have no more NA or NaN in y, categ, dot.categ (if dot.color != NULL) and facet.categ
-
-
-
- if( ! is.null(categ.class.order)){
- if(length(categ.class.order) != length(categ)){
- tempo.cat <- paste0("ERROR IN ", function.name, "\ncateg.class.order ARGUMENT MUST BE A LIST OF LENGTH EQUAL TO LENGTH OF categ\nHERE IT IS LENGTH: ", length(categ.class.order), " VERSUS ", length(categ))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- for(i3 in 1:length(categ.class.order)){
- if(is.null(categ.class.order[[i3]])){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE categ.class.order COMPARTMENT ", i3, " IS NULL. ALPHABETICAL ORDER WILL BE APPLIED")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- data1[, categ[i3]] <- factor(as.character(data1[, categ[i3]])) # if already a factor, change nothing, if characters, levels according to alphabetical order
- categ.class.order[[i3]] <- levels(data1[, categ[i3]]) # character vector that will be used later
- }else{
- tempo <- fun_check(data = categ.class.order[[i3]], data.name = paste0("COMPARTMENT ", i3 , " OF categ.class.order ARGUMENT"), class = "vector", mode = "character", length = length(levels(data1[, categ[i3]])), fun.name = function.name) # length(data1[, categ[i1]) -> if data1[, categ[i1] was initially character vector, then conversion as factor after the NA removal, thus class number ok. If data1[, categ[i1] was initially factor, no modification after the NA removal, thus class number ok
- if(tempo$problem == TRUE){
- stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- if(any(duplicated(categ.class.order[[i3]]))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nCOMPARTMENT ", i3, " OF categ.class.order ARGUMENT CANNOT HAVE DUPLICATED CLASSES: ", paste(categ.class.order[[i3]], collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else if( ! (all(categ.class.order[[i3]] %in% unique(data1[, categ[i3]])) & all(unique(data1[, categ[i3]]) %in% categ.class.order[[i3]]))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nCOMPARTMENT ", i3, " OF categ.class.order ARGUMENT MUST BE CLASSES OF ELEMENT ", i3, " OF categ ARGUMENT\nHERE IT IS:\n", paste(categ.class.order[[i3]], collapse = " "), "\nFOR COMPARTMENT ", i3, " OF categ.class.order AND IT IS:\n", paste(unique(data1[, categ[i3]]), collapse = " "), "\nFOR COLUMN ", categ[i3], " OF data1")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- data1[, categ[i3]] <- factor(data1[, categ[i3]], levels = categ.class.order[[i3]]) # reorder the factor
-
- }
- names(categ.class.order)[i3] <- categ[i3]
- }
- }
- }else{
- categ.class.order <- vector("list", length = length(categ))
- tempo.categ.class.order <- NULL
- for(i2 in 1:length(categ.class.order)){
- categ.class.order[[i2]] <- levels(data1[, categ[i2]])
- names(categ.class.order)[i2] <- categ[i2]
- tempo.categ.class.order <- c(tempo.categ.class.order, ifelse(i2 != 1, "\n", ""), categ.class.order[[i2]])
- }
- if(box.alpha != 0){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE categ.class.order SETTING IS NULL. ALPHABETICAL ORDER WILL BE APPLIED FOR BOX ORDERING:\n", paste(tempo.categ.class.order, collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- # categ.class.order not NULL anymore (list)
- if(is.null(box.legend.name) & box.alpha != 0){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE box.legend.name SETTING IS NULL. NAMES OF categ WILL BE USED: ", paste(categ, collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- box.legend.name <- categ[length(categ)] # if only categ1, then legend name of categ1, if length(categ) == 2, then legend name of categ2
- }
- # box.legend.name not NULL anymore (character string)
- # management of categ.color
- if( ! is.null(categ.color)){
- # check the nature of color
- # integer colors into gg_palette
- tempo.check.color <- fun_check(data = categ.color, class = "integer", double.as.integer.allowed = TRUE, na.contain = TRUE, fun.name = function.name)$problem
- if(tempo.check.color == FALSE){
- # convert integers into colors
- categ.color <- fun_gg_palette(max(categ.color, na.rm = TRUE))[categ.color]
- }
- # end integer colors into gg_palette
- if( ! (all(categ.color %in% colors() | grepl(pattern = "^#", categ.color)))){ # check that all strings of low.color start by #
- tempo.cat <- paste0("ERROR IN ", function.name, "\ncateg.color ARGUMENT MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # AND/OR COLOR NAMES GIVEN BY colors(): ", paste(unique(categ.color), collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(is.na(categ.color)) & box.alpha != 0){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") categ.color ARGUMENT CONTAINS NA")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- # end check the nature of color
- # check the length of color
- categ.len <- length(categ) # if only categ1, then colors for classes of categ1, if length(categ) == 2, then colors for classes of categ2
- if(length(categ.color) == length(levels(data1[, categ[categ.len]]))){ # here length(categ.color) is equal to the different number of categ
- # data1[, categ[categ.len]] <- factor(data1[, categ[categ.len]]) # not required because sure that is is a factor
- data1 <- data.frame(data1, categ.color = data1[, categ[categ.len]], stringsAsFactors = TRUE) # no need stringsAsFactors here for stat.nolog as factors remain factors
- data1$categ.color <- factor(data1$categ.color, labels = categ.color)
- if(box.alpha != 0){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") IN ", categ[categ.len], " OF categ ARGUMENT, THE FOLLOWING COLORS:\n", paste(categ.color, collapse = " "), "\nHAVE BEEN ATTRIBUTED TO THESE CLASSES:\n", paste(levels(factor(data1[, categ[categ.len]])), collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else if(length(categ.color) == length(data1[, categ[categ.len]])){# here length(categ.color) is equal to nrow(data1) -> Modif to have length(categ.color) equal to the different number of categ (length(categ.color) == length(levels(data1[, categ[categ.len]])))
- data1 <- data.frame(data1, categ.color = categ.color, stringsAsFactors = TRUE)
- tempo.check <- unique(data1[ , c(categ[categ.len], "categ.color")])
- if( ! (nrow(tempo.check) == length(unique(categ.color)) & nrow(tempo.check) == length(unique(data1[ , categ[categ.len]])))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\ncateg.color ARGUMENT HAS THE LENGTH OF data1 ROW NUMBER\nBUT IS INCORRECTLY ASSOCIATED TO EACH CLASS OF categ ", categ[categ.len], ":\n", paste(unique(mapply(FUN = "paste", data1[ ,categ[categ.len]], data1[ ,"categ.color"])), collapse = "\n"))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- # data1[, categ[categ.len]] <- factor(data1[, categ[categ.len]]) # not required because sure that is is a factor
- categ.color <- unique(data1$categ.color[order(data1[, categ[categ.len]])]) # Modif to have length(categ.color) equal to the different number of categ (length(categ.color) == length(levels(data1[, categ[categ.len]])))
- if(box.alpha != 0){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") categ.color ARGUMENT HAS THE LENGTH OF data1 ROW NUMBER\nCOLORS HAVE BEEN RESPECTIVELY ASSOCIATED TO EACH CLASS OF categ ", categ[categ.len], " AS:\n", paste(levels(factor(data1[, categ[categ.len]])), collapse = " "), "\n", paste(categ.color, collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- }else if(length(categ.color) == 1){
- # data1[, categ[categ.len]] <- factor(data1[, categ[categ.len]]) # not required because sure that is is a factor
- data1 <- data.frame(data1, categ.color = categ.color, stringsAsFactors = TRUE)
- categ.color <- rep(categ.color, length(levels(data1[, categ[categ.len]])))
- if(box.alpha != 0){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") categ.color ARGUMENT HAS LENGTH 1, MEANING THAT ALL THE DIFFERENT CLASSES OF ", categ[categ.len], "\n", paste(levels(factor(data1[, categ[categ.len]])), collapse = " "), "\nWILL HAVE THE SAME COLOR\n", paste(categ.color, collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, "\ncateg.color ARGUMENT MUST BE (1) LENGTH 1, OR (2) THE LENGTH OF data1 NROWS AFTER NA/Inf REMOVAL, OR (3) THE LENGTH OF THE CLASSES IN THE categ ", categ[categ.len], " COLUMN. HERE IT IS COLOR LENGTH ", length(categ.color), " VERSUS CATEG LENGTH ", length(data1[, categ[categ.len]]), " AND CATEG CLASS LENGTH ", length(unique(data1[, categ[categ.len]])), "\nPRESENCE OF NA/Inf COULD BE THE PROBLEM")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }else{
- categ.len <- length(categ) # if only categ1, then colors for classes of categ1, if length(categ) == 2, then colors for classes of categ2
- # data1[, categ[categ.len]] <- factor(data1[, categ[categ.len]]) # not required because sure that is is a factor
- categ.color <- fun_gg_palette(length(levels(data1[, categ[categ.len]])))
- data1 <- data.frame(data1, categ.color = data1[, categ[categ.len]], stringsAsFactors = TRUE)
- data1$categ.color <- factor(data1$categ.color, labels = categ.color)
- if(box.alpha != 0){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NULL categ.color ARGUMENT -> COLORS RESPECTIVELY ATTRIBUTED TO EACH CLASS OF ", categ[categ.len], " IN data1:\n", paste(categ.color, collapse = " "), "\n", paste(levels(data1[, categ[categ.len]]), collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- # categ.color not NULL anymore
- categ.color <- as.character(categ.color)
- # categ.color is a character string representing the diff classes
- data1$categ.color <- factor(data1$categ.color, levels = unique(categ.color)) # ok because if categ.color is a character string, the order make class 1, class 2, etc. unique() because no duplicates allowed
- # data1$categ.color is a factor with order of levels -> categ.color
- # end management of categ.color
- # management of dot.color
- if( ! is.null(dot.color)){
- # optional legend of dot colors
- if( ! is.null(dot.categ)){
- ini.dot.categ <- dot.categ
- if( ! dot.categ %in% names(data1)){ # no need to use all() because length(dot.categ) = 1
- tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.categ ARGUMENT MUST BE A COLUMN NAME OF data1. HERE IT IS:\n", dot.categ)
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else if(dot.categ %in% categ){ # no need to use all() because length(dot.categ) = 1. Do not use dot.categ %in% categ[length(categ)] -> error
- # management of dot legend if dot.categ %in% categ (because legends with the same name are joined in ggplot2)
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE COLUMN NAME OF data1 INDICATED IN THE dot.categ ARGUMENT (", dot.categ, ") HAS BEEN REPLACED BY ", paste0(dot.categ, "_DOT"), " TO AVOID MERGED LEGEND BY GGPLOT2")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- data1 <- data.frame(data1, dot.categ = data1[, dot.categ], stringsAsFactors = TRUE) # dot.categ is not a column name of data1 (checked above with reserved words)
- dot.categ <- paste0(dot.categ, "_DOT")
- names(data1)[names(data1) == "dot.categ"] <- dot.categ # paste0(dot.categ, "_DOT") is not a column name of data1 (checked above with reserved words)
- # tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.categ ARGUMENT CANNOT BE A COLUMN NAME OF data1 ALREADY SPECIFIED IN THE categ ARGUMENT:\n", dot.categ, "\nINDEED, dot.categ ARGUMENT IS MADE TO HAVE MULTIPLE DOT COLORS NOT RELATED TO THE BOXPLOT CATEGORIES")
- # stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- tempo1 <- fun_check(data = data1[, dot.categ], data.name = paste0(dot.categ, " COLUMN OF data1"), class = "vector", mode = "character", na.contain = TRUE, fun.name = function.name)
- tempo2 <- fun_check(data = data1[, dot.categ], data.name = paste0(dot.categ, " COLUMN OF data1"), class = "factor", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.categ COLUMN MUST BE A FACTOR OR CHARACTER VECTOR") #
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- data1[, dot.categ] <- factor(data1[, dot.categ]) # if already a factor, change nothing, if characters, levels according to alphabetical order
- # dot.categ column of data1 is factor from here
- if( ! is.null(dot.categ.class.order)){
- if(any(duplicated(dot.categ.class.order))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.categ.class.order ARGUMENT CANNOT HAVE DUPLICATED CLASSES: ", paste(dot.categ.class.order, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else if( ! (all(dot.categ.class.order %in% levels(data1[, dot.categ])) & all(levels(data1[, dot.categ]) %in% dot.categ.class.order))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.categ.class.order ARGUMENT MUST BE CLASSES OF dot.categ ARGUMENT\nHERE IT IS:\n", paste(dot.categ.class.order, collapse = " "), "\nFOR dot.categ.class.order AND IT IS:\n", paste(levels(data1[, dot.categ]), collapse = " "), "\nFOR dot.categ COLUMN (", ini.dot.categ, ") OF data1")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- data1[, dot.categ] <- factor(data1[, dot.categ], levels = dot.categ.class.order) # reorder the factor
- }
- }else{
- if(all(dot.color == "same") & length(dot.color) == 1){
- dot.categ.class.order <- unlist(categ.class.order[length(categ)])
- data1[, dot.categ] <- factor(data1[, dot.categ], levels = dot.categ.class.order) # reorder the factor
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE dot.categ.class.order SETTING IS NULL AND dot.color IS \"same\". ORDER OF categ.class.order WILL BE APPLIED FOR LEGEND DISPLAY: ", paste(dot.categ.class.order, collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }else{
- dot.categ.class.order <- sort(levels(data1[, dot.categ]))
- data1[, dot.categ] <- factor(data1[, dot.categ], levels = dot.categ.class.order) # reorder the factor
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE dot.categ.class.order SETTING IS NULL. ALPHABETICAL ORDER WILL BE APPLIED FOR LEGEND DISPLAY: ", paste(dot.categ.class.order, collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- # dot.categ.class.order not NULL anymore (character string) if dot.categ is not NULL
- if(all(dot.color == "same") & length(dot.color) == 1){
- if( ! identical(ini.dot.categ, categ[length(categ)])){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nWHEN dot.color ARGUMENT IS \"same\", THE COLUMN NAME IN dot.categ ARGUMENT MUST BE IDENTICAL TO THE LAST COLUMN NAME IN categ ARGUMENT. HERE IT IS:\ndot.categ: ", paste(ini.dot.categ, collapse = " "), "\ncateg: ", paste(categ, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else if( ! fun_comp_1d(unlist(categ.class.order[length(categ)]), dot.categ.class.order)$identical.content){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nWHEN dot.color ARGUMENT IS \"same\",\nLAST COMPARTMENT OF categ.class.order ARGUMENT AND dot.categ.class.order ARGUMENT CANNOT BE DIFFERENT:\nLAST COMPARTMENT OF categ.class.order: ", paste(unlist(categ.class.order[length(categ)]), collapse = " "), "\ndot.categ.class.order: ", paste(dot.categ.class.order, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- for(i3 in 1:length(categ)){
- if(identical(categ[i3], ini.dot.categ) & ! identical(unlist(categ.class.order[i3]), dot.categ.class.order) & identical(sort(unlist(categ.class.order[i3])), sort(dot.categ.class.order))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE dot.categ ARGUMENT SETTING IS PRESENT IN THE categ ARGUMENT SETTING, BUT ORDER OF THE CLASSES IS NOT THE SAME:\ncateg.class.order: ", paste(unlist(categ.class.order[i3]), collapse = " "), "\ndot.categ.class.order: ", paste(dot.categ.class.order, collapse = " "), "\nNOTE THAT ORDER OF categ.class.order IS THE ONE USED FOR THE AXIS REPRESENTATION")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- if(is.null(dot.legend.name)){
- dot.legend.name <- if(ini.dot.categ %in% categ[length(categ)]){dot.categ}else{ini.dot.categ} #
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE dot.legend.name SETTING IS NULL -> ", dot.legend.name, " WILL BE USED AS LEGEND TITLE OF DOTS")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- # dot.legend.name not NULL anymore (character string)
- }else{
- if( ! is.null(dot.categ.class.order)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE dot.categ.class.order ARGUMENT IS NOT NULL, BUT IS THE dot.categ ARGUMENT\n-> dot.categ.class.order NOT CONSIDERED AS NO LEGEND WILL BE DRAWN")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- # But dot.categ.class.order will be converted to NULL below (not now)
- }
- # end optional legend of dot colors
- # check the nature of color
- # integer colors into gg_palette
- tempo.check.color <- fun_check(data = dot.color, class = "integer", double.as.integer.allowed = TRUE, na.contain = TRUE, fun.name = function.name)$problem
- if(tempo.check.color == FALSE){
- # convert integers into colors
- dot.color <- fun_gg_palette(max(dot.color, na.rm = TRUE))[dot.color]
- }
- # end integer colors into gg_palette
- if(all(dot.color == "same") & length(dot.color) == 1){
- dot.color <- categ.color # same color of the dots as the corresponding box color
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") dot.color ARGUMENT HAS BEEN SET TO \"same\"\nTHUS, DOTS WILL HAVE THE SAME COLORS AS THE CORRESPONDING BOXPLOT")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }else if( ! (all(dot.color %in% colors() | grepl(pattern = "^#", dot.color)))){ # check that all strings of low.color start by #
- tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.color ARGUMENT MUST BE (1) A HEXADECIMAL COLOR VECTOR STARTING BY #, OR (2) COLOR NAMES GIVEN BY colors(), OR (3) INTEGERS, OR THE STRING\"same\"\nHERE IT IS: ", paste(unique(dot.color), collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(any(is.na(dot.color))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") dot.color ARGUMENT CONTAINS NA")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- # end check the nature of color
- # check the length of color
- if( ! is.null(dot.categ)){
- # optional legend of dot colors
- if(length(dot.color) > 1 & length(unique(data1[, dot.categ])) != length(dot.color)){
- tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.color ARGUMENT IS NOT THE SAME LENGTH AS LEVELS OF dot.categ COLUMN (", dot.categ, "):\ndot.color: ", paste(dot.color, collapse = " "), "\ndot.categ LEVELS: ", paste(levels(data1[, dot.categ]), collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else if(length(dot.color) == 1 & length(dot.categ.class.order) > 1){ # to deal with single color
- dot.color <- rep(dot.color, length(dot.categ.class.order))
- }
- data1 <- data.frame(data1, dot.color = data1[, dot.categ], stringsAsFactors = TRUE)
- data1$dot.color <- factor(data1$dot.color, labels = dot.color) # do not use labels = unique(dot.color). Otherwise, we can have green1 green2 when dot.color is c("green", "green")
- dot.color <- as.character(unique(data1$dot.color[order(data1[, dot.categ])])) # reorder the dot.color character vector
- if(length(dot.color) == 1 & length(dot.categ.class.order) > 1){ # to deal with single color
- dot.color <- rep(dot.color, length(dot.categ.class.order))
- }
- tempo.check <- unique(data1[ , c(dot.categ, "dot.color")])
- if(length(unique(data1[ , "dot.color"])) > 1 & ( ! (nrow(tempo.check) == length(unique(data1[ , "dot.color"])) & nrow(tempo.check) == length(unique(data1[ , dot.categ]))))){ # length(unique(data1[ , "dot.color"])) > 1 because if only one color, can be attributed to each class of dot.categ
- tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.color ARGUMENT IS INCORRECTLY ASSOCIATED TO EACH CLASS OF dot.categ (", dot.categ, ") COLUMN:\n", paste(unique(mapply(FUN = "paste", data1[ , dot.categ], data1[ ,"dot.color"])), collapse = "\n"))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") IN dot.categ ARGUMENT (", ini.dot.categ, "), THE FOLLOWING COLORS OF DOTS:\n", paste(dot.color, collapse = " "), "\nHAVE BEEN ATTRIBUTED TO THESE CLASSES:\n", paste(levels(data1[, dot.categ]), collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- # dot.color is a character string representing the diff classes of dot.categ
- # data1$dot.color is a factor with order of levels -> dot.categ
- # end optional legend of dot colors
- }else{
- categ.len <- length(categ) # if only categ1, then colors for classes of categ1, if length(categ) == 2, then colors for classes of categ2
- if(length(dot.color) == length(levels(data1[, categ[categ.len]]))){ # here length(dot.color) is equal to the different number of categ
- # data1[, categ[categ.len]] <- factor(data1[, categ[categ.len]]) # not required because sure that is is a factor
- data1 <- data.frame(data1, dot.color = data1[, categ[categ.len]], stringsAsFactors = TRUE)
- data1$dot.color <- factor(data1$dot.color, labels = dot.color)
- if(box.alpha != 0){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") IN ", categ[categ.len], " OF categ ARGUMENT, THE FOLLOWING COLORS:\n", paste(dot.color, collapse = " "), "\nHAVE BEEN ATTRIBUTED TO THESE CLASSES:\n", paste(levels(factor(data1[, categ[categ.len]])), collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else if(length(dot.color) == length(data1[, categ[categ.len]])){# here length(dot.color) is equal to nrow(data1) -> Modif to have length(dot.color) equal to the different number of categ (length(dot.color) == length(levels(data1[, categ[categ.len]])))
- data1 <- data.frame(data1, dot.color = dot.color, stringsAsFactors = TRUE)
- }else if(length(dot.color) == 1 & ! all(dot.color == "same")){
- # data1[, categ[categ.len]] <- factor(data1[, categ[categ.len]]) # not required because sure that is is a factor
- data1 <- data.frame(data1, dot.color = dot.color, stringsAsFactors = TRUE)
- dot.color <- rep(dot.color, length(levels(data1[, categ[categ.len]])))
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") dot.color ARGUMENT HAS LENGTH 1, MEANING THAT ALL THE DIFFERENT CLASSES OF ", categ[categ.len], "\n", paste(levels(factor(data1[, categ[categ.len]])), collapse = " "), "\nWILL HAVE THE SAME COLOR\n", paste(dot.color, collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, "\ndot.color ARGUMENT MUST BE (1) LENGTH 1, OR (2) THE LENGTH OF data1 NROWS AFTER NA/Inf REMOVAL, OR (3) THE LENGTH OF THE CLASSES IN THE categ ", categ[categ.len], " COLUMN. HERE IT IS COLOR LENGTH ", length(dot.color), " VERSUS CATEG LENGTH ", length(data1[, categ[categ.len]]), " AND CATEG CLASS LENGTH ", length(unique(data1[, categ[categ.len]])), "\nPRESENCE OF NA/Inf COULD BE THE PROBLEM")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end check the length of color
- dot.color <- as.character(dot.color)
- # dot.color is a character string representing the diff classes
- data1$dot.color <- factor(data1$dot.color, levels = unique(dot.color)) # ok because if dot.color is a character string, the order make class 1, class 2, etc. If dot.color is a column of data1, then levels will be created, without incidence, except if dot.categ specified (see below). unique() because no duplicates allowed
- # data1$dot.color is a factor with order of levels -> dot.color
- }
- # end optional legend of dot colors
- }else if(is.null(dot.color) & ! (is.null(dot.categ) & is.null(dot.categ.class.order) & is.null(dot.legend.name))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") dot.categ OR dot.categ.class.order OR dot.legend.name ARGUMENT HAS BEEN SPECIFIED BUT dot.color ARGUMENT IS NULL (NO DOT PLOTTED)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- # dot.color either NULL (no dot plotted) or character string (potentially representing the diff classes of dot.categ)
- # data1$dot.color is either NA or a factor (with order of levels -> depending on dot.categ or categ[length(categ)], or other
- if(is.null(dot.categ)){
- dot.categ.class.order <- NULL # because not used anyway
- }
- # dot.categ.class.order either NULL if dot.categ is NULL (no legend displayed) or character string (potentially representing the diff classes of dot.categ)
- # end management of dot.color
- if(is.null(dot.color) & box.fill == FALSE & dot.alpha <= 0.025){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE FOLLOWING ARGUMENTS WERE SET AS:\ndot.color = NULL (NOT ALL DOTS BUT ONLY POTENTIAL OUTLIER DOTS DISPLAYED)\nbox.fill = FALSE (NO FILLING COLOR FOR BOTH BOXES AND POTENTIAL OUTLIER DOTS)\ndot.alpha = ", fun_round(dot.alpha, 4), "\n-> POTENTIAL OUTLIER DOTS MIGHT NOT BE VISIBLE BECAUSE ALMOST TRANSPARENT")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- if(is.null(dot.color) & box.fill == FALSE & dot.border.size == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nTHE FOLLOWING ARGUMENTS WERE SET AS:\ndot.color = NULL (NOT ALL DOTS BUT ONLY POTENTIAL OUTLIER DOTS DISPLAYED)\nbox.fill = FALSE (NO FILLING COLOR FOR BOTH BOXES AND POTENTIAL OUTLIER DOTS)\ndot.border.size = 0 (NO BORDER FOR POTENTIAL OUTLIER DOTS)\n-> THESE SETTINGS ARE NOT ALLOWED BECAUSE THE POTENTIAL OUTLIER DOTS WILL NOT BE VISIBLE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # integer dot.border.color into gg_palette
- if( ! is.null(dot.border.color)){
- tempo <- fun_check(data = dot.border.color, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, fun.name = function.name)
- if(tempo$problem == FALSE){ # convert integers into colors
- dot.border.color <- fun_gg_palette(max(dot.border.color, na.rm = TRUE))[dot.border.color]
- }
- }
- # end integer dot.border.color into gg_palette
- # na detection and removal (done now to be sure of the correct length of categ)
- column.check <- c("categ.color", if( ! is.null(dot.color)){"dot.color"}) #
- if(any(is.na(data1[, column.check]))){ # data1 used here instead of data1.ini in case of new NaN created by log conversion (neg values)
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NA DETECTED IN COLUMNS ", paste(column.check, collapse = " "), " OF data1 AND CORRESPONDING ROWS REMOVED (SEE $removed.row.nb AND $removed.rows)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- for(i2 in 1:length(column.check)){
- if(any(is.na(data1[, column.check[i2]]))){
- tempo.warn <- paste0("NA REMOVAL DUE TO COLUMN ", column.check[i2], " OF data1")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n", tempo.warn)))
- }
- }
- tempo <- unique(unlist(lapply(lapply(c(data1[column.check]), FUN = is.na), FUN = which)))
- removed.row.nb <- c(removed.row.nb, tempo)
- removed.rows <- rbind(removed.rows, data1[tempo, ], stringsAsFactors = FALSE) # here data1 used because categorical columns tested
- if(length(tempo) != 0){
- data1 <- data1[-tempo, ] # WARNING tempo here and not removed.row.nb because the latter contain more numbers thant the former
- data1.ini <- data1.ini[-tempo, ] # WARNING tempo here and not removed.row.nb because the latter contain more numbers thant the former
- for(i3 in 1:length(column.check)){
- if(any( ! unique(removed.rows[, column.check[i3]]) %in% unique(data1[, column.check[i3]]))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") IN COLUMN ", column.check[i3], " OF data1, THE FOLLOWING CLASSES HAVE DISAPPEARED AFTER NA/Inf REMOVAL (IF COLUMN USED IN THE PLOT, THIS CLASS WILL NOT BE DISPLAYED):\n", paste(unique(removed.rows[, column.check[i3]])[ ! unique(removed.rows[, column.check[i3]]) %in% unique(data1[, column.check[i3]])], collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- }
- for(i2 in 1:length(column.check)){
- if(column.check[i2] == "categ.color"){
- categ.color <- levels(data1[, column.check[i2]])[levels(data1[, column.check[i2]]) %in% unique(data1[, column.check[i2]])] # remove the absent color in the character vector
- if(length(categ.color) == 1 & length(unlist(categ.class.order[length(categ)])) > 1){ # to deal with single color
- categ.color <- rep(categ.color, length(unlist(categ.class.order[length(categ)])))
- }
- data1[, column.check[i2]] <- factor(as.character(data1[, column.check[i2]]), levels = unique(categ.color))
- }
- if(column.check[i2] == "dot.color"){
- dot.color <- levels(data1[, column.check[i2]])[levels(data1[, column.check[i2]]) %in% unique(data1[, column.check[i2]])] # remove the absent color in the character vector
- if(length(dot.color) == 1 & length(dot.categ.class.order) > 1){ # to deal with single color. If dot.categ.class.order == NULL (which is systematically the case if dot.categ == NULL), no rep(dot.color, length(dot.categ.class.order)
- dot.color <- rep(dot.color, length(dot.categ.class.order))
- }
- data1[, column.check[i2]] <- factor(as.character(data1[, column.check[i2]]), levels = unique(dot.color))
- }
- }
- }
- # end na detection and removal (done now to be sure of the correct length of categ)
- # From here, data1 and data.ini have no more NA or NaN
-
-
-
- # end second round of checking and data preparation
-
-
- # package checking
- fun_pack(req.package = c(
- "ggplot2",
- "gridExtra",
- "lemon",
- "scales"
- ), lib.path = lib.path)
- # end package checking
-
-
-
-
-
- # main code
- # y coordinates recovery (create ini.box.coord, dot.coord and modify data1)
- if(length(categ) == 1){
- # width commputations
- box.width2 <- box.width
- box.space <- 0 # to inactivate the shrink that add space between grouped boxes, because no grouped boxes here
- # end width commputations
- # data1 check categ order for dots coordinates recovery
- data1 <- data.frame(data1, categ.check = data1[, categ[1]], stringsAsFactors = TRUE)
- data1$categ.check <- as.integer(data1$categ.check) # to check that data1[, categ[1]] and dot.coord$group are similar, during merging
- # end data1 check categ order for dots coordinates recovery
- # per box dots coordinates recovery
- tempo.gg.name <- "gg.indiv.plot."
- tempo.gg.count <- 0
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = paste0("ggplot2::ggplot()", if(is.null(add)){""}else{add})))) # add added here to have the facets
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_point(data = data1, mapping = ggplot2::aes_string(x = categ[1], y = y, color = categ[1]), stroke = dot.border.size, size = dot.size, alpha = dot.alpha, shape = 21))
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "color", name = box.legend.name, values = if(is.null(dot.color)){rep(NA, length(unique(data1[, categ[1]])))}else if(length(dot.color) == 1){rep(dot.color, length(unique(data1[, categ[1]])))}else{dot.color}))
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_boxplot(data = data1, mapping = ggplot2::aes_string(x = categ[1], y = y, fill = categ[1]), coef = if(box.whisker.kind == "no"){0}else if(box.whisker.kind == "std"){1.5}else if(box.whisker.kind == "max"){Inf})) # fill because this is what is used with geom_box # to easily have the equivalent of the grouped boxes
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "fill", name = box.legend.name, values = if(length(categ.color) == 1){rep(categ.color, length(unique(data1[, categ[1]])))}else{categ.color}))
- # end per box dots coordinates recovery
- }else if(length(categ) == 2){
- # width commputations
- box.width2 <- box.width / length(unique(data1[, categ[length(categ)]])) # real width of each box in x-axis unit, among the set of grouped box. Not relevant if no grouped boxes length(categ) == 1
- # end width commputations
- # data1 check categ order for dots coordinates recovery
- tempo.factor <- paste0(data1[order(data1[, categ[2]], data1[, categ[1]]), categ[2]], "_", data1[order(data1[, categ[2]], data1[, categ[1]]), categ[1]])
- data1 <- data.frame(data1[order(data1[, categ[2]], data1[, categ[1]]), ], categ.check = factor(tempo.factor, levels = unique(tempo.factor)), stringsAsFactors = TRUE)
- data1$categ.check <- as.integer(data1$categ.check)
- # end data1 check categ order for dots coordinates recovery
- # per box dots coordinates recovery
- tempo.gg.name <- "gg.indiv.plot."
- tempo.gg.count <- 0
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = paste0("ggplot2::ggplot()", if(is.null(add)){""}else{add})))) # add added here to have the facets
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_point(data = data1, mapping = ggplot2::aes_string(x = categ[1], y = y, color = categ[2]), stroke = dot.border.size, size = dot.size, alpha = dot.alpha, shape = 21))
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "color", name = box.legend.name, values = if(is.null(dot.color)){rep(NA, length(unique(data1[, categ[2]])))}else if(length(dot.color) == 1){rep(dot.color, length(unique(data1[, categ[2]])))}else{dot.color}))
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_boxplot(data = data1, mapping = ggplot2::aes_string(x = categ[1], y = y, fill = categ[2]), coef = if(box.whisker.kind == "no"){0}else if(box.whisker.kind == "std"){1.5}else if(box.whisker.kind == "max"){Inf})) # fill because this is what is used with geom_box # to easily have the equivalent of the grouped boxes
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "fill", name = box.legend.name, values = if(length(categ.color) == 1){rep(categ.color, length(unique(data1[, categ[2]])))}else{categ.color}))
- # end per box dots coordinates recovery
- }else{
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 1")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if( ! is.null(stat.disp)){
- stat.just <- fun_gg_just(
- angle = stat.angle,
- pos = ifelse(
- vertical == TRUE,
- ifelse(stat.disp == "top", "bottom", "top"), # "bottom" because we want justification for text that are below the ref point which is the top of the graph. The opposite for "above"
- ifelse(stat.disp == "top", "left", "right") # "left" because we want justification for text that are on the left of the ref point which is the right border of the graph. The opposite for "above"
- ),
- kind = "text"
- )
- }
- # has in fact no interest because ggplot2 does not create room for geom_text()
- tempo.data.max <- data1[which.max(data1[, y]), ]
- tempo.data.max <- data.frame(tempo.data.max, label = formatC(tempo.data.max[, y], digit = 2, drop0trailing = TRUE, format = "f"), stringsAsFactors = TRUE)
- # end has in fact no interest because ggplot2 does not create room for geom_text()
- tempo.graph.info.ini <- ggplot2::ggplot_build(eval(parse(text = paste(paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "), if( ! is.null(stat.disp)){' + ggplot2::geom_text(data = tempo.data.max, mapping = ggplot2::aes_string(x = 1, y = y, label = "label"), size = stat.size, color = "black", angle = stat.angle, hjust = stat.just$hjust, vjust = stat.just$vjust)'})))) # added here to have room for annotation
- dot.coord <- tempo.graph.info.ini$data[[1]]
- dot.coord$x <- as.numeric(dot.coord$x) # because weird class
- dot.coord$PANEL <- as.numeric(dot.coord$PANEL) # because numbers as levels. But may be a problem is facet are reordered ?
- tempo.mean <- aggregate(x = dot.coord$y, by = list(dot.coord$group, dot.coord$PANEL), FUN = mean, na.rm = TRUE)
- names(tempo.mean)[names(tempo.mean) == "x"] <- "MEAN"
- names(tempo.mean)[names(tempo.mean) == "Group.1"] <- "BOX"
- names(tempo.mean)[names(tempo.mean) == "Group.2"] <- "PANEL"
- dot.coord <- data.frame(
- dot.coord[order(dot.coord$group, dot.coord$y), ], # dot.coord$PANEL deals below
- y.check = as.double(data1[order(data1$categ.check, data1[, y]), y]),
- categ.check = data1[order(data1$categ.check, data1[, y]), "categ.check"],
- dot.color = if(is.null(dot.color)){NA}else{data1[order(data1$categ.check, data1[, y]), "dot.color"]},
- data1[order(data1$categ.check, data1[, y]), ][categ], # avoid the renaming below
- stringsAsFactors = TRUE
- ) # y.check to be sure that the order is the same between the y of data1 and the y of dot.coord
- # names(dot.coord)[names(dot.coord) == "tempo.categ1"] <- categ[1]
- if( ! is.null(dot.categ)){
- dot.coord <- data.frame(dot.coord, data1[order(data1$categ.check, data1[, y]), ][dot.categ], stringsAsFactors = TRUE) # avoid the renaming
- }
- if( ! is.null(facet.categ)){
- dot.coord <- data.frame(dot.coord, data1[order(data1$categ.check, data1[, y]), ][facet.categ], stringsAsFactors = TRUE) # for facet panels
- tempo.test <- NULL
- for(i2 in 1:length(facet.categ)){
- tempo.test <- paste0(tempo.test, ".", formatC(as.numeric(dot.coord[, facet.categ[i2]]), width = nchar(max(as.numeric(dot.coord[, facet.categ[i2]]), na.rm = TRUE)), flag = "0")) # convert factor into numeric with leading zero for proper ranking # merge the formatC() to create a new factor. The convertion to integer should recreate the correct group number. Here as.numeric is used and not as.integer in case of numeric in facet.categ (because comes from add and not checked by fun_check, contrary to categ)
- }
- tempo.test <- as.integer(factor(tempo.test))
- if( ! identical(as.integer(dot.coord$PANEL), tempo.test)){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nas.integer(dot.coord$PANEL) AND tempo.test MUST BE IDENTICAL. CODE HAS TO BE MODIFIED")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- if(dot.tidy == TRUE){
- if( ! is.null(dot.categ)){
- dot.coord <- data.frame(dot.coord, tidy_group = data1[order(data1$categ.check, data1[, y]), ][, dot.categ], stringsAsFactors = TRUE) # avoid the renaming
- # tidy_group_coord is to be able to fuse table when creating the table for dot coordinates
- if(dot.categ %in% categ){
- dot.coord <- data.frame(dot.coord, tidy_group_coord = dot.coord$group, stringsAsFactors = TRUE)
- }else{
- dot.coord <- data.frame(dot.coord, tidy_group_coord = as.integer(factor(paste0(
- formatC(as.integer(dot.coord[, categ[1]]), width = nchar(max(as.integer(dot.coord[, categ[1]]), na.rm = TRUE)), flag = "0"), # convert factor into numeric with leading zero for proper ranking
- ".",
- if(length(categ) == 2){formatC(as.integer(dot.coord[, categ[2]]), width = nchar(max(as.integer(dot.coord[, categ[2]]), na.rm = TRUE)), flag = "0")}, # convert factor into numeric with leading zero for proper ranking
- if(length(categ) == 2){"."},
- formatC(as.integer(dot.coord[, dot.categ]), width = nchar(max(as.integer(dot.coord[, dot.categ]), na.rm = TRUE)), flag = "0") # convert factor into numeric with leading zero for proper ranking
- )), stringsAsFactors = TRUE) # merge the 2 or 3 formatC() to create a new factor. The convertion to integer should recreate the correct group number
- ) # for tidy dot plots
- }
- }else{
- dot.coord <- data.frame(dot.coord, tidy_group = if(length(categ) == 1){
- dot.coord[, categ]}else{as.integer(factor(paste0(
- formatC(as.integer(dot.coord[, categ[1]]), width = nchar(max(as.integer(dot.coord[, categ[1]]), na.rm = TRUE)), flag = "0"), # convert factor into numeric with leading zero for proper ranking
- ".",
- formatC(as.integer(dot.coord[, categ[2]]), width = nchar(max(as.integer(dot.coord[, categ[2]]), na.rm = TRUE)), flag = "0")# convert factor into numeric with leading zero for proper ranking
- )), stringsAsFactors = TRUE) # merge the 2 formatC() to create a new factor. The convertion to integer should recreate the correct group number
- }) # for tidy dot plots
- # tidy_group_coord is to be able to fuse table when creating the table for dot coordinates
- dot.coord <- data.frame(dot.coord, tidy_group_coord = dot.coord$group, stringsAsFactors = TRUE)
- }
- }
- if( ! (identical(dot.coord$y, dot.coord$y.check) & identical(dot.coord$group, dot.coord$categ.check))){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\n(dot.coord$y AND dot.coord$y.check) AS WELL AS (dot.coord$group AND dot.coord$categ.check) MUST BE IDENTICAL. CODE HAS TO BE MODIFIED")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- if( ! identical(tempo.mean[order(tempo.mean$BOX, tempo.mean$PANEL), ]$BOX, unique(dot.coord[order(dot.coord$group, dot.coord$PANEL), c("group", "PANEL")])$group)){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\n(tempo.mean$BOX, tempo.mean$PANEL) AND (dot.coord$group, dot.coord$PANEL) MUST BE IDENTICAL. CODE HAS TO BE MODIFIED")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- tempo <- unique(dot.coord[order(dot.coord$group, dot.coord$PANEL), c(categ, if( ! is.null(dot.color) & ! is.null(dot.categ)){if(dot.categ != ini.dot.categ){dot.categ}}, if( ! is.null(facet.categ)){facet.categ}), drop = FALSE])
- # names(tempo) <- paste0(names(tempo), ".mean")
- tempo.mean <- data.frame(tempo.mean[order(tempo.mean$BOX, tempo.mean$PANEL), ], tempo, stringsAsFactors = TRUE)
- }
- }
- # at that stage, categ color and dot color are correctly attributed in data1, box.coord and dot.coord
- # end y dot coordinates recovery (create ini.box.coord, dot.coord and modify data1)
- # ylim range
- if(is.null(y.lim)){
- y.lim <- tempo.graph.info.ini$layout$panel_params[[1]]$y.range # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
- if(any(( ! is.finite(y.lim)) | is.na(y.lim)) | length(y.lim) != 2){ # kept but normally no more Inf in data1
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\ntempo.graph.info.ini$layout$panel_params[[1]]$y.range[1] CONTAINS NA OR Inf OR HAS LENGTH 1")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }else if(y.log != "no"){
- y.lim <- get(y.log)(y.lim) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
- }
- if(y.log != "no"){
- # normally this control is not necessary anymore
- if(any( ! is.finite(y.lim))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\ny.lim ARGUMENT CANNOT HAVE ZERO OR NEGATIVE VALUES WITH THE y.log ARGUMENT SET TO ", y.log, ":\n", paste(y.lim, collapse = " "), "\nPLEASE, CHECK DATA VALUES (PRESENCE OF ZERO OR INF VALUES)")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- if(suppressWarnings(all(y.lim %in% c(Inf, -Inf)))){
- # normally this control is not necessary anymore
- tempo.cat <- paste0("ERROR IN ", function.name, " y.lim CONTAINS Inf VALUES, MAYBE BECAUSE VALUES FROM data1 ARGUMENTS ARE NA OR Inf ONLY OR BECAUSE OF LOG SCALE REQUIREMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- if(suppressWarnings(any(is.na(y.lim)))){
- # normally this control is not necessary anymore
- tempo.cat <- paste0("ERROR IN ", function.name, " y.lim CONTAINS NA OR NaN VALUES, MAYBE BECAUSE VALUES FROM data1 ARGUMENTS ARE NA OR Inf ONLY OR BECAUSE OF LOG SCALE REQUIREMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- y.lim.order <- order(y.lim) # to deal with inverse axis
- y.lim <- sort(y.lim)
- y.lim[1] <- y.lim[1] - abs(y.lim[2] - y.lim[1]) * ifelse(diff(y.lim.order) > 0, y.bottom.extra.margin, y.top.extra.margin) # diff(y.lim.order) > 0 medians not inversed axis
- y.lim[2] <- y.lim[2] + abs(y.lim[2] - y.lim[1]) * ifelse(diff(y.lim.order) > 0, y.top.extra.margin, y.bottom.extra.margin) # diff(y.lim.order) > 0 medians not inversed axis
- if(y.include.zero == TRUE){ # no need to check y.log != "no" because done before
- y.lim <- range(c(y.lim, 0), na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
- }
- y.lim <- y.lim[y.lim.order]
- if(any(is.na(y.lim))){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 2")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end ylim range
-
-
-
-
-
-
- # drawing
- # constant part
- tempo.gg.name <- "gg.indiv.plot."
- tempo.gg.count <- 0
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = paste0("ggplot2::ggplot()", if(is.null(add)){""}else{add})))) # add is directly put here to deal with additional variable of data, like when using facet_grid. No problem if add is a theme, will be dealt below
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::xlab(if(is.null(x.lab)){categ[1]}else{x.lab}))
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::ylab(if(is.null(y.lab)){y}else{y.lab}))
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::ggtitle(title))
- # text angle management
- axis.just <- fun_gg_just(angle = x.angle, pos = ifelse(vertical == TRUE, "bottom", "left"), kind = "axis")
- # end text angle management
- add.check <- TRUE
- if( ! is.null(add)){ # if add is NULL, then = 0
- if(grepl(pattern = "ggplot2\\s*::\\s*theme", add) == TRUE){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") \"ggplot2::theme\" STRING DETECTED IN THE add ARGUMENT\n-> INTERNAL GGPLOT2 THEME FUNCTIONS theme() AND theme_classic() HAVE BEEN INACTIVATED, TO BE USED BY THE USER\n-> article ARGUMENT WILL BE IGNORED")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- add.check <- FALSE
- }
- }
- if(add.check == TRUE & article == TRUE){
- # WARNING: not possible to add theme()several times. NO message but the last one overwrites the others
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::theme_classic(base_size = text.size))
- if(grid == TRUE){
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), m.gg <- ggplot2::theme(
- text = ggplot2::element_text(size = text.size),
- plot.title = ggplot2::element_text(size = title.text.size), # stronger than text
- line = ggplot2::element_line(size = 0.5),
- legend.key = ggplot2::element_rect(color = "white", size = 1.5), # size of the frame of the legend
- axis.line.y.left = ggplot2::element_line(colour = "black"), # draw lines for the y axis
- axis.line.x.bottom = ggplot2::element_line(colour = "black"), # draw lines for the x axis
- panel.grid.major.x = if(vertical == TRUE){NULL}else{ggplot2::element_line(colour = "grey85", size = 0.75)},
- panel.grid.major.y = if(vertical == TRUE){ggplot2::element_line(colour = "grey85", size = 0.75)}else{NULL},
- panel.grid.minor.y = if(vertical == TRUE){ggplot2::element_line(colour = "grey90", size = 0.25)}else{NULL},
- axis.text.x = if(vertical == TRUE){ggplot2::element_text(angle = axis.just$angle, hjust = axis.just$hjust, vjust = axis.just$vjust)}else{NULL},
- axis.text.y = if(vertical == TRUE){NULL}else{ggplot2::element_text(angle = axis.just$angle, hjust = axis.just$hjust, vjust = axis.just$vjust)},
- strip.background = ggplot2::element_rect(fill = NA, colour = NA) # for facet background
- ))
- }else{
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), m.gg <- ggplot2::theme(
- text = ggplot2::element_text(size = text.size),
- plot.title = ggplot2::element_text(size = title.text.size), # stronger than text
- line = ggplot2::element_line(size = 0.5),
- legend.key = ggplot2::element_rect(color = "white", size = 1.5), # size of the frame of the legend
- axis.line.y.left = ggplot2::element_line(colour = "black"),
- axis.line.x.bottom = ggplot2::element_line(colour = "black"),
- axis.text.x = if(vertical == TRUE){ggplot2::element_text(angle = axis.just$angle, hjust = axis.just$hjust, vjust = axis.just$vjust)}else{NULL},
- axis.text.y = if(vertical == TRUE){NULL}else{ggplot2::element_text(angle = axis.just$angle, hjust = axis.just$hjust, vjust = axis.just$vjust)},
- strip.background = ggplot2::element_rect(fill = NA, colour = NA)
- ))
- }
- }else if(add.check == TRUE & article == FALSE){
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), m.gg <- ggplot2::theme(
- text = ggplot2::element_text(size = text.size),
- plot.title = ggplot2::element_text(size = title.text.size), # stronger than text
- line = ggplot2::element_line(size = 0.5),
- legend.key = ggplot2::element_rect(color = "white", size = 1.5), # size of the frame of the legend
- panel.background = ggplot2::element_rect(fill = "grey95"),
- axis.line.y.left = ggplot2::element_line(colour = "black"),
- axis.line.x.bottom = ggplot2::element_line(colour = "black"),
- panel.grid.major.x = ggplot2::element_line(colour = "grey85", size = 0.75),
- panel.grid.major.y = ggplot2::element_line(colour = "grey85", size = 0.75),
- panel.grid.minor.x = ggplot2::element_blank(),
- panel.grid.minor.y = ggplot2::element_line(colour = "grey90", size = 0.25),
- strip.background = ggplot2::element_rect(fill = NA, colour = NA),
- axis.text.x = if(vertical == TRUE){ggplot2::element_text(angle = axis.just$angle, hjust = axis.just$hjust, vjust = axis.just$vjust)}else{NULL},
- axis.text.y = if(vertical == TRUE){NULL}else{ggplot2::element_text(angle = axis.just$angle, hjust = axis.just$hjust, vjust = axis.just$vjust)}
- ))
- }
- # Contrary to fun_gg_bar(), cannot plot the boxplot right now, because I need the dots plotted first
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_boxplot(data = data1, mapping = ggplot2::aes_string(x = categ[1], y = y, group = categ[length(categ)]), position = ggplot2::position_dodge(width = NULL), color = NA, width = box.width, fill = NA)) # this is to set the graph (i.e., a blanck boxplot to be able to use x coordinates to plot dots before boxes)
- # end constant part
-
-
-
-
- # graphic info recovery (including means)
- tempo.graph.info <- ggplot2::ggplot_build(eval(parse(text = paste0(paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "), ' + ggplot2::geom_boxplot(data = data1, mapping = ggplot2::aes_string(x = categ[1], y = y, fill = categ[length(categ)]), position = ggplot2::position_dodge(width = NULL), width = box.width, notch = box.notch, coef = if(box.whisker.kind == "no"){0}else if(box.whisker.kind == "std"){1.5}else if(box.whisker.kind == "max"){Inf}) + ggplot2::scale_discrete_manual(aesthetics = "fill", name = box.legend.name, values = if(length(categ.color) == 1){rep(categ.color, length(unique(data1[, categ[length(categ)]])))}else{categ.color})')))) # will be recovered later again, when ylim will be considered
- tempo.yx.ratio <- (tempo.graph.info$layout$panel_params[[1]]$y.range[2] - tempo.graph.info$layout$panel_params[[1]]$y.range[1]) / (tempo.graph.info$layout$panel_params[[1]]$x.range[2] - tempo.graph.info$layout$panel_params[[1]]$x.range[1])
- box.coord <- tempo.graph.info$data[[2]] # to have the summary statistics of the plot. Contrary to ini.box.plot, now integrates ylim Here because can be required for stat.disp when just box are plotted
- box.coord$x <- as.numeric(box.coord$x) # because x is of special class that block comparison of values using identical
- box.coord$PANEL <- as.numeric(box.coord$PANEL) # because numbers as levels. But may be a problem is facet are reordered ?
- box.coord <- box.coord[order(box.coord$group, box.coord$PANEL), ]
- if( ! (identical(tempo.mean$BOX, box.coord$group) & identical(tempo.mean$PANEL, box.coord$PANEL))){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nidentical(tempo.mean$BOX, box.coord$group) & identical(tempo.mean$PANEL, box.coord$PANEL) DO NOT HAVE THE SAME VALUE ORDER")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- # tempo <- c(categ, if( ! is.null(dot.color) & ! is.null(dot.categ)){if(dot.categ != ini.dot.categ){dot.categ}}, if( ! is.null(facet.categ)){facet.categ})
- if(any(names(tempo.mean) %in% names(box.coord))){
- names(tempo.mean)[names(tempo.mean) %in% names(box.coord)] <- paste0(names(tempo.mean)[names(tempo.mean) %in% names(box.coord)], ".mean")
- }
- box.coord <- data.frame(box.coord, tempo.mean, stringsAsFactors = TRUE)
- }
- # end graphic info recovery (including means)
-
-
-
- # stat output (will also serve for boxplot and mean display)
- # x not added know to do not have them in stat.nolog
- stat <- data.frame(
- MIN = box.coord$ymin_final,
- QUART1 = box.coord$lower,
- MEDIAN = box.coord$middle,
- MEAN = box.coord$MEAN,
- QUART3 = box.coord$upper,
- MAX = box.coord$ymax_final,
- WHISK_INF = box.coord$ymin,
- BOX_INF = box.coord$lower,
- NOTCH_INF = box.coord$notchlower,
- NOTCH_SUP = box.coord$notchupper,
- BOX_SUP = box.coord$upper,
- WHISK_SUP = box.coord$ymax,
- OUTLIERS = box.coord["outliers"],
- tempo.mean[colnames(tempo.mean) != "MEAN"],
- COLOR = box.coord$fill,
- stringsAsFactors = TRUE
- ) # box.coord["outliers"] written like this because it is a list. X coordinates not put now because several features to set
- names(stat)[names(stat) == "outliers"] <- "OUTLIERS"
- stat.nolog <- stat # stat.nolog ini will serve for outputs
- if(y.log != "no"){
- stat.nolog[c("MIN", "QUART1", "MEDIAN", "MEAN", "QUART3", "MAX", "WHISK_INF", "BOX_INF", "NOTCH_INF", "NOTCH_SUP", "BOX_SUP", "WHISK_SUP")] <- ifelse(y.log == "log2", 2, 10)^(stat.nolog[c("MIN", "QUART1", "MEDIAN", "MEAN", "QUART3", "MAX", "WHISK_INF", "BOX_INF", "NOTCH_INF", "NOTCH_SUP", "BOX_SUP", "WHISK_SUP")])
- stat.nolog$OUTLIERS <- lapply(stat.nolog$OUTLIERS, FUN = function(X){ifelse(y.log == "log2", 2, 10)^X})
- }
- # end stat output (will also serve for boxplot and mean display)
-
-
-
-
-
-
- # x coordinates management (for random plotting and for stat display)
- # width commputations
- width.ini <- c(box.coord$xmax - box.coord$xmin)[1] # all the box widths are equal here. Only the first one taken
- width.correct <- width.ini * box.space / 2
- if( ! (identical(stat$BOX, box.coord$group) & identical(stat$PANEL, box.coord$PANEL))){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nidentical(stat$BOX, box.coord$group) & identical(stat$PANEL, box.coord$PANEL) MUST BE IDENTICAL. CODE HAS TO BE MODIFIED")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- stat <- data.frame(
- stat,
- X = box.coord$x,
- X_BOX_INF = box.coord$xmin + width.correct,
- X_BOX_SUP = box.coord$xmax - width.correct,
- X_NOTCH_INF = box.coord$x - (box.coord$x - (box.coord$xmin + width.correct)) / 2,
- X_NOTCH_SUP = box.coord$x + (box.coord$x - (box.coord$xmin + width.correct)) / 2,
- X_WHISK_INF = box.coord$x - (box.coord$x - (box.coord$xmin + width.correct)) * box.whisker.width,
- X_WHISK_SUP = box.coord$x + (box.coord$x - (box.coord$xmin + width.correct)) * box.whisker.width,
- # tempo.mean[colnames(tempo.mean) != "MEAN"], # already added above
- stringsAsFactors = TRUE
- )
- stat$COLOR <- factor(stat$COLOR, levels = unique(categ.color))
- if( ! all(stat$NOTCH_SUP < stat$BOX_SUP & stat$NOTCH_INF > stat$BOX_INF) & box.notch == TRUE){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") SOME NOTCHES ARE BEYOND BOX HINGES. TRY ARGUMENT box.notch = FALSE")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- dot.jitter <- c((box.coord$xmax - width.correct) - (box.coord$xmin + width.correct))[1] * dot.jitter # real dot.jitter. (box.coord$xmin + width.correct) - (box.coord$xmax - width.correct))[1] is the width of the box. Is equivalent to (box.coord$x - (box.coord$xmin + width.correct))[1] * 2
- # end width commputations
- if( ! is.null(dot.color)){
- # random dots
- if(dot.tidy == FALSE){
- dot.coord.rd1 <- merge(dot.coord, box.coord[c("fill", "PANEL", "group", "x")], by = c("PANEL", "group"), sort = FALSE) # rd for random. Send the coord of the boxes into the coord data.frame of the dots (in the column x.y). WARNING: by = c("PANEL", "group") without fill column because PANEL & group columns are enough as only one value of x column per group number in box.coord. Thus, no need to consider fill column
- if(nrow(dot.coord.rd1) != nrow(dot.coord)){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nTHE merge() FUNCTION DID NOT RETURN A CORRECT dot.coord.rd1 DATA FRAME. CODE HAS TO BE MODIFIED")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- sampled.dot.jitter <- if(nrow(dot.coord.rd1) == 1){runif(n = nrow(dot.coord.rd1), min = - dot.jitter / 2, max = dot.jitter / 2)}else{sample(x = runif(n = nrow(dot.coord.rd1), min = - dot.jitter / 2, max = dot.jitter / 2), size = nrow(dot.coord.rd1), replace = FALSE)}
- dot.coord.rd2 <- data.frame(dot.coord.rd1, dot.x = dot.coord.rd1$x.y + sampled.dot.jitter, stringsAsFactors = TRUE) # set the dot.jitter thanks to runif and dot.jitter range. Then, send the coord of the boxes into the coord data.frame of the dots (in the column x.y)
- if(length(categ) == 1){
- tempo.data1 <- unique(data.frame(data1[categ[1]], group = as.integer(data1[, categ[1]]), stringsAsFactors = TRUE)) # categ[1] is factor
- names(tempo.data1)[names(tempo.data1) == categ[1]] <- paste0(categ[1], ".check")
- verif <- paste0(categ[1], ".check")
- }else if(length(categ) == 2){
- tempo.data1 <- unique(
- data.frame(
- data1[c(categ[1], categ[2])],
- group = as.integer(factor(paste0(
- formatC(as.integer(data1[, categ[2]]), width = nchar(max(as.integer(data1[, categ[2]]), na.rm = TRUE)), flag = "0"), # convert factor into numeric with leading zero for proper ranking
- ".",
- formatC(as.integer(data1[, categ[1]]), width = nchar(max(as.integer(data1[, categ[1]]), na.rm = TRUE)), flag = "0")# convert factor into numeric with leading zero for proper ranking
- )), stringsAsFactors = TRUE) # merge the 2 formatC() to create a new factor. The convertion to integer should recreate the correct group number
- )
- ) # categ[2] first if categ[2] is used to make the categories in ggplot and categ[1] is used to make the x-axis
- names(tempo.data1)[names(tempo.data1) == categ[1]] <- paste0(categ[1], ".check")
- names(tempo.data1)[names(tempo.data1) == categ[2]] <- paste0(categ[2], ".check")
- verif <- c(paste0(categ[1], ".check"), paste0(categ[2], ".check"))
- }else{
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 3")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- dot.coord.rd3 <- merge(dot.coord.rd2, tempo.data1, by = intersect("group", "group"), sort = FALSE) # send the factors of data1 into coord. WARNING: I have replaced by = "group" by intersect("group", "group") because of an error due to wrong group group merging in dot.coord.rd3
- if(nrow(dot.coord.rd3) != nrow(dot.coord) | ( ! fun_comp_2d(dot.coord.rd3[categ], dot.coord.rd3[verif])$identical.content)){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nTHE merge() FUNCTION DID NOT RETURN A CORRECT dot.coord.rd3 DATA FRAME. CODE HAS TO BE MODIFIED")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end random dots
- }
- # tidy dots
- # coordinates are recovered during plotting (see dot.coord.tidy1 below)
- # end tidy dots
- }
- # end x coordinates management (for random plotting and for stat display)
-
-
-
-
-
- # boxplot display before dot display if box.fill = TRUE
- coord.names <- NULL
- # creation of the data frame for (main box + legend) and data frame for means
- if(box.notch == FALSE){
- for(i3 in 1:length(categ)){
- if(i3 == 1){
- tempo.polygon <- data.frame(GROUPX = c(t(stat[, rep(categ[i3], 5)])), stringsAsFactors = TRUE)
- }else{
- tempo.polygon <- cbind(tempo.polygon, c(t(stat[, rep(categ[i3], 5)])), stringsAsFactors = TRUE)
- }
- }
- names(tempo.polygon) <- categ
- tempo.polygon <- data.frame(X = c(t(stat[, c("X_BOX_INF", "X_BOX_SUP", "X_BOX_SUP", "X_BOX_INF", "X_BOX_INF")])), Y = c(t(stat[, c("BOX_INF", "BOX_INF", "BOX_SUP", "BOX_SUP", "BOX_INF")])), COLOR = c(t(stat[, c("COLOR", "COLOR", "COLOR", "COLOR", "COLOR")])), BOX = as.character(c(t(stat[, c("BOX", "BOX", "BOX", "BOX", "BOX")]))), tempo.polygon, stringsAsFactors = TRUE)
- if( ! is.null(facet.categ)){
- for(i4 in 1:length(facet.categ)){
- tempo.polygon <- data.frame(tempo.polygon, c(t(stat[, c(facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4])])), stringsAsFactors = TRUE)
- names(tempo.polygon)[length(names(tempo.polygon))] <- facet.categ[i4]
- }
- }
- }else{
- for(i3 in 1:length(categ)){
- if(i3 == 1){
- tempo.polygon <- data.frame(GROUPX = c(t(stat[, rep(categ[i3], 11)])), stringsAsFactors = TRUE)
- }else{
- tempo.polygon <- cbind(tempo.polygon, c(t(stat[, rep(categ[i3], 11)])), stringsAsFactors = TRUE)
- }
- }
- names(tempo.polygon) <- categ
- tempo.polygon <- data.frame(X = c(t(stat[, c("X_BOX_INF", "X_BOX_SUP", "X_BOX_SUP", "X_NOTCH_SUP", "X_BOX_SUP", "X_BOX_SUP", "X_BOX_INF", "X_BOX_INF", "X_NOTCH_INF", "X_BOX_INF", "X_BOX_INF")])), Y = c(t(stat[, c("BOX_INF", "BOX_INF", "NOTCH_INF", "MEDIAN", "NOTCH_SUP", "BOX_SUP", "BOX_SUP", "NOTCH_SUP", "MEDIAN", "NOTCH_INF", "BOX_INF")])), COLOR = c(t(stat[, c("COLOR", "COLOR", "COLOR", "COLOR", "COLOR", "COLOR", "COLOR", "COLOR", "COLOR", "COLOR", "COLOR")])), BOX = as.character(c(t(stat[, c("BOX", "BOX", "BOX", "BOX", "BOX", "BOX", "BOX", "BOX", "BOX", "BOX", "BOX")]))), tempo.polygon, stringsAsFactors = TRUE)
- if( ! is.null(facet.categ)){
- for(i4 in 1:length(facet.categ)){
- tempo.polygon <- data.frame(tempo.polygon, c(t(stat[, c(facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4], facet.categ[i4])])), stringsAsFactors = TRUE)
- names(tempo.polygon)[length(names(tempo.polygon))] <- facet.categ[i4]
- }
- }
- }
- tempo.polygon$COLOR <- factor(tempo.polygon$COLOR, levels = unique(categ.color))
- if( ! is.null(categ.class.order)){
- for(i3 in 1:length(categ)){
- tempo.polygon[, categ[i3]] <- factor(tempo.polygon[, categ[i3]], levels = categ.class.order[[i3]])
- }
- }
- # modified name of dot.categ column (e.g., "Categ1_DOT") must be included for boxplot using ridy dots
- if( ! is.null(dot.color) & ! is.null(dot.categ)){
- if(dot.categ != ini.dot.categ){
- tempo.polygon <- data.frame(tempo.polygon, GROUPX = tempo.polygon[, ini.dot.categ], stringsAsFactors = TRUE)
- names(tempo.polygon)[names(tempo.polygon) == "GROUPX"] <- dot.categ
-
- }
- }
- tempo.diamon.mean <- data.frame(X = c(t(stat[, c("X", "X_NOTCH_INF", "X", "X_NOTCH_SUP", "X")])), Y = c(t(cbind(stat["MEAN"] - (stat[, "X"] - stat[, "X_NOTCH_INF"]) * tempo.yx.ratio, stat["MEAN"], stat["MEAN"] + (stat[, "X"] - stat[, "X_NOTCH_INF"]) * tempo.yx.ratio, stat["MEAN"], stat["MEAN"] - (stat[, "X"] - stat[, "X_NOTCH_INF"]) * tempo.yx.ratio, stringsAsFactors = TRUE))), COLOR = c(t(stat[, c("COLOR", "COLOR", "COLOR", "COLOR", "COLOR")])), GROUP = c(t(stat[, c("BOX", "BOX", "BOX", "BOX", "BOX")])), stringsAsFactors = TRUE) # stringsAsFactors = TRUE for cbind() because stat["MEAN"] is a data frame. Otherwise, stringsAsFactors is not an argument for cbind() on vectors
- if( ! is.null(facet.categ)){
- for(i3 in 1:length(facet.categ)){
- tempo.diamon.mean <- data.frame(tempo.diamon.mean, c(t(stat[, c(facet.categ[i3], facet.categ[i3], facet.categ[i3], facet.categ[i3], facet.categ[i3])])), stringsAsFactors = TRUE)
- names(tempo.diamon.mean)[length(names(tempo.diamon.mean))] <- facet.categ[i3]
- }
- }
- tempo.diamon.mean$COLOR <- factor(tempo.diamon.mean$COLOR, levels = unique(categ.color))
- # end creation of the data frame for (main box + legend) and data frame for means
- if(box.fill == TRUE){
- # assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_boxplot(data = data1, mapping = ggplot2::aes_string(x = categ[1], y = y, color = categ[length(categ)], fill = categ[length(categ)]), position = ggplot2::position_dodge(width = NULL), width = box.width, size = box.line.size, notch = box.notch, coef = if(box.whisker.kind == "no"){0}else if(box.whisker.kind == "std"){1.5}else if(box.whisker.kind == "max"){Inf}, alpha = box.alpha, outlier.shape = if( ! is.null(dot.color)){NA}else{21}, outlier.color = if( ! is.null(dot.color)){NA}else{dot.border.color}, outlier.fill = if( ! is.null(dot.color)){NA}else{NULL}, outlier.size = if( ! is.null(dot.color)){NA}else{dot.size}, outlier.stroke = if( ! is.null(dot.color)){NA}else{dot.border.size}, outlier.alpha = if( ! is.null(dot.color)){NA}else{dot.alpha})) # the color, size, etc. of the outliers are dealt here. outlier.color = NA to do not plot outliers when dots are already plotted. Finally, boxplot redrawn (see below)
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_polygon(
- data = tempo.polygon,
- mapping = ggplot2::aes_string(x = "X", y = "Y", group = "BOX", fill = categ[length(categ)], color = categ[length(categ)]),
- size = box.line.size,
- alpha = box.alpha # works only for fill, not for color
- ))
- coord.names <- c(coord.names, "main.box")
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = X, xend = X, y = BOX_SUP, yend = WHISK_SUP, group = categ[length(categ)]), color = "black", size = box.line.size, alpha = box.alpha)) #
- coord.names <- c(coord.names, "sup.whisker")
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = X, xend = X, y = BOX_INF, yend = WHISK_INF, group = categ[length(categ)]), color = "black", size = box.line.size, alpha = box.alpha)) #
- coord.names <- c(coord.names, "inf.whisker")
- if(box.whisker.width > 0){
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = X_WHISK_INF, xend = X_WHISK_SUP, y = WHISK_SUP, yend = WHISK_SUP, group = categ[length(categ)]), color = "black", size = box.line.size, alpha = box.alpha, lineend = "round")) #
- coord.names <- c(coord.names, "sup.whisker.edge")
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = X_WHISK_INF, xend = X_WHISK_SUP, y = WHISK_INF, yend = WHISK_INF, group = categ[length(categ)]), color = "black", size = box.line.size, alpha = box.alpha, lineend = "round")) #
- coord.names <- c(coord.names, "inf.whisker.edge")
- }
- if(box.mean == TRUE){
- # assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_point(data = stat, mapping = ggplot2::aes_string(x = "X", y = "MEAN", group = categ[length(categ)]), shape = 23, stroke = box.line.size * 2, fill = stat$COLOR, size = box.mean.size, color = "black", alpha = box.alpha)) # group used in aesthetic to do not have it in the legend
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_polygon(
- data = tempo.diamon.mean,
- mapping = ggplot2::aes(x = X, y = Y, group = GROUP),
- fill = tempo.diamon.mean[, "COLOR"],
- color = hsv(0, 0, 0, alpha = box.alpha), # outline of the polygon in black but with alpha
- size = box.line.size,
- alpha = box.alpha
- ))
- coord.names <- c(coord.names, "mean")
- }
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = if(box.notch == FALSE){X_BOX_INF}else{X_NOTCH_INF}, xend = if(box.notch == FALSE){X_BOX_SUP}else{X_NOTCH_SUP}, y = MEDIAN, yend = MEDIAN, group = categ[length(categ)]), color = "black", size = box.line.size * 2, alpha = box.alpha)) #
- coord.names <- c(coord.names, "median")
- }
- # end boxplot display before dot display if box.fill = TRUE
-
-
-
-
-
-
- # dot display
- if( ! is.null(dot.color)){
- if(dot.tidy == FALSE){
- if(is.null(dot.categ)){
- if(dot.border.size == 0){
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_point(
- data = dot.coord.rd3,
- mapping = ggplot2::aes_string(x = "dot.x", y = "y", group = categ[length(categ)]),
- size = dot.size,
- shape = 19,
- color = dot.coord.rd3$dot.color,
- alpha = dot.alpha
- )) # group used in aesthetic to do not have it in the legend. Here ggplot2::scale_discrete_manual() cannot be used because of the group easthetic
- }else{
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_point(
- data = dot.coord.rd3,
- mapping = ggplot2::aes_string(x = "dot.x", y = "y", group = categ[length(categ)]),
- shape = 21,
- stroke = dot.border.size,
- color = if(is.null(dot.border.color)){dot.coord.rd3$dot.color}else{rep(dot.border.color, nrow(dot.coord.rd3))},
- size = dot.size,
- fill = dot.coord.rd3$dot.color,
- alpha = dot.alpha
- )) # group used in aesthetic to do not have it in the legend. Here ggplot2::scale_discrete_manual() cannot be used because of the group easthetic
- }
- }else{
- if(dot.border.size == 0){
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_point(
- data = dot.coord.rd3,
- mapping = ggplot2::aes_string(x = "dot.x", y = "y", alpha = dot.categ),
- size = dot.size,
- shape = 19,
- color = dot.coord.rd3$dot.color
- )) # group used in aesthetic to do not have it in the legend. Here ggplot2::scale_discrete_manual() cannot be used because of the group easthetic
- }else{
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_point(
- data = dot.coord.rd3,
- mapping = ggplot2::aes_string(x = "dot.x", y = "y", alpha = dot.categ),
- size = dot.size,
- shape = 21,
- stroke = dot.border.size,
- color = if(is.null(dot.border.color)){dot.coord.rd3$dot.color}else{rep(dot.border.color, nrow(dot.coord.rd3))},
- fill = dot.coord.rd3$dot.color
- )) # group used in aesthetic to do not have it in the legend. Here ggplot2::scale_discrete_manual() cannot be used because of the group easthetic
- }
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "alpha", name = dot.legend.name, values = rep(dot.alpha, length(dot.categ.class.order)), guide = ggplot2::guide_legend(override.aes = list(fill = dot.color, color = if(is.null(dot.border.color)){dot.color}else{dot.border.color}, stroke = dot.border.size, alpha = dot.alpha)))) # values are the values of color (which is the border color in geom_box. WARNING: values = categ.color takes the numbers to make the colors if categ.color is a factor
- }
- coord.names <- c(coord.names, "dots")
- }else if(dot.tidy == TRUE){
- # here plot using group -> no scale
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_dotplot(
- data = dot.coord,
- mapping = ggplot2::aes_string(x = categ[1], y = "y", group = "group"), # not dot.categ here because the classes of dot.categ create new separations
- position = ggplot2::position_dodge(width = box.width),
- binpositions = "all",
- binaxis = "y",
- stackdir = "center",
- alpha = dot.alpha,
- fill = dot.coord$dot.color,
- stroke = dot.border.size,
- color = if(is.null(dot.border.color)){dot.coord$dot.color}else{rep(dot.border.color, nrow(dot.coord))},
- show.legend = FALSE, # WARNING: do not use show.legend = TRUE because it uses the arguments outside aes() as aesthetics (here color and fill). Thus I must find a way using ggplot2::scale_discrete_manual()
- binwidth = (y.lim[2] - y.lim[1]) / dot.tidy.bin.nb
- )) # geom_dotplot ggplot2 v3.3.0: I had to remove rev() in fill and color # very weird behavior of geom_dotplot ggplot2 v3.2.1, (1) because with aes group = (to avoid legend), the dot plotting is not good in term of coordinates, and (2) because data1 seems reorderer according to x = categ[1] before plotting. Thus, I have to use fill = dot.coord[rev(order(dot.coord[, categ[1]], decreasing = TRUE)), "dot.color"] to have the good corresponding colors # show.legend option do not remove the legend, only the aesthetic of the legend (dot, line, etc.)
- coord.names <- c(coord.names, "dots")
- if( ! is.null(dot.categ)){
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_dotplot(
- data = dot.coord,
- mapping = ggplot2::aes_string(x = categ[1], y = "y", alpha = dot.categ), # not dot.categ here because the classes of dot.categ create new separations
- position = ggplot2::position_dodge(width = box.width),
- binpositions = "all",
- binaxis = "y",
- stackdir = "center",
- fill = NA,
- stroke = NA,
- color = NA,
- # WARNING: do not use show.legend = TRUE because it uses the arguments outside aes() as aesthetics (here color and fill). Thus I must find a way using ggplot2::scale_discrete_manual()
- binwidth = (y.lim[2] - y.lim[1]) / dot.tidy.bin.nb
- )) # geom_dotplot ggplot2 v3.3.0: I had to remove rev() in fill and color # very weird behavior of geom_dotplot ggplot2 v3.2.1, (1) because with aes group = (to avoid legend), the dot plotting is not good in term of coordinates, and (2) because data1 seems reorderer according to x = categ[1] before plotting. Thus, I have to use fill = dot.coord[rev(order(dot.coord[, categ[1]], decreasing = TRUE)), "dot.color"] to have the good corresponding colors # show.legend option do not remove the legend, only the aesthetic of the legend (dot, line, etc.)
- # assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "linetype", name = dot.legend.name, values = rep(1, length(categ.color)))) # values = rep("black", length(categ.color)) are the values of color (which is the border color of dots), and this modify the border color on the plot. WARNING: values = categ.color takes the numbers to make the colors if categ.color is a factor
- coord.names <- c(coord.names, "bad_remove")
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "alpha", name = dot.legend.name, values = rep(dot.alpha, length(dot.categ.class.order)), labels = dot.categ.class.order, guide = ggplot2::guide_legend(title = if(ini.dot.categ == categ[length(categ)]){dot.categ}else{ini.dot.categ}, override.aes = list(fill = levels(dot.coord$dot.color), color = if(is.null(dot.border.color)){levels(dot.coord$dot.color)}else{dot.border.color}, stroke = dot.border.size, alpha = dot.alpha)))) # values are the values of color (which is the border color in geom_box. WARNING: values = categ.color takes the numbers to make the colors if categ.color is a factor
- }
- # coordinates of tidy dots
- tempo.coord <- ggplot2::ggplot_build(eval(parse(text = paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "))))$data # to have the tidy dot coordinates
- if(length(which(sapply(tempo.coord, FUN = nrow) == nrow(data1))) > if(is.null(dot.categ)){1}else{2}){ # if(is.null(dot.categ)){1}else{2} because 1 dotplot if dot.categ is NULL and 2 dotplots is not, with the second being a blank dotplot with wrong coordinates. Thus take the first in that situation
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nMORE THAN ", if(is.null(dot.categ)){1}else{2}, " COMPARTMENT WITH NROW EQUAL TO nrow(data1) IN THE tempo.coord LIST (FOR TIDY DOT COORDINATES). CODE HAS TO BE MODIFIED")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- dot.coord.tidy1 <- tempo.coord[[which(sapply(tempo.coord, FUN = nrow) == nrow(data1))[1]]] # the second being a blank dotplot with wrong coordinates. Thus take the first whatever situation
- dot.coord.tidy1$x <- as.numeric(dot.coord.tidy1$x) # because weird class
- dot.coord.tidy1$PANEL <- as.numeric(dot.coord.tidy1$PANEL) # because numbers as levels. But may be a problem is facet are reordered ?
- }
- # tempo.box.coord <- merge(box.coord, unique(dot.coord[, c("PANEL", "group", categ)]), by = c("PANEL", "group"), sort = FALSE) # not required anymore because box.coord already contains categ do not add dot.categ and tidy_group_coord here because the coordinates are for stats. Add the categ in box.coord. WARNING: by = c("PANEL", "group") without fill column because PANEL & group columns are enough as only one value of x column per group number in box.coord. Thus, no need to consider fill column
- # below inactivated because not true when dealing with dot.categ different from categ
- # if(nrow(tempo.box.coord) != nrow(box.coord)){
- # tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nTHE merge() FUNCTION DID NOT RETURN A CORRECT tempo.box.coord DATA FRAME. CODE HAS TO BE MODIFIED")
- # stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- # }
- dot.coord.tidy2 <- merge(dot.coord.tidy1, box.coord[c("fill", "PANEL", "group", "x", categ)], by = c("PANEL", "group"), sort = FALSE) # send the coord of the boxes into the coord data.frame of the dots (in the column x.y).WARNING: by = c("PANEL", "group") without fill column because PANEL & group columns are enough as only one value of x column per group number in tempo.box.coord. Thus, no need to consider fill colum # DANGER: from here the fill.y and x.y (from tempo.box.coord) are not good in dot.coord.tidy2. It is ok because Categ1 Categ2 from tempo.box.coord are ok with the group column from dot.coord.tidy1. This is due to the fact that dot.coord.tidy resulting from geom_dotplot does not make the same groups as the other functions
- if(nrow(dot.coord.tidy2) != nrow(dot.coord)){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nTHE merge() FUNCTION DID NOT RETURN A CORRECT dot.coord.tidy2 DATA FRAME. CODE HAS TO BE MODIFIED")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # From here, check for dot.coord.tidy3 which wil be important for stat over the plot. WARNING: dot.categ has nothing to do here for stat coordinates. Thus, not in tempo.data1
- if(length(categ) == 1){
- tempo.data1 <- unique(data.frame(data1[categ[1]], group = as.integer(data1[, categ[1]]), stringsAsFactors = TRUE)) # categ[1] is factor
- names(tempo.data1)[names(tempo.data1) == categ[1]] <- paste0(categ[1], ".check")
- verif <- paste0(categ[1], ".check")
- }else if(length(categ) == 2){
- tempo.data1 <- unique(
- data.frame(
- data1[c(categ[1], categ[2])],
- group = as.integer(factor(paste0(
- formatC(as.integer(data1[, categ[2]]), width = nchar(max(as.integer(data1[, categ[2]]), na.rm = TRUE)), flag = "0"), # convert factor into numeric with leading zero for proper ranking
- ".",
- formatC(as.integer(data1[, categ[1]]), width = nchar(max(as.integer(data1[, categ[1]]), na.rm = TRUE)), flag = "0")# convert factor into numeric with leading zero for proper ranking
- )), stringsAsFactors = TRUE) # merge the 2 formatC() to create a new factor. The convertion to integer should recreate the correct group number
- )
- ) # categ[2] first if categ[2] is used to make the categories in ggplot and categ[1] is used to make the x-axis
- names(tempo.data1)[names(tempo.data1) == categ[1]] <- paste0(categ[1], ".check")
- names(tempo.data1)[names(tempo.data1) == categ[2]] <- paste0(categ[2], ".check")
- verif <- c(paste0(categ[1], ".check"), paste0(categ[2], ".check"))
- }else{
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 4")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- dot.coord.tidy3 <- merge(dot.coord.tidy2, tempo.data1, by = intersect("group", "group"), sort = FALSE) # send the factors of data1 into coord. WARNING: I have tested intersect("group", "group") instead of by = "group". May be come back to by = "group" in case of error. But I did this because of an error in dot.coord.rd3 above
- if(nrow(dot.coord.tidy3) != nrow(dot.coord) | ( ! fun_comp_2d(dot.coord.tidy3[categ], dot.coord.tidy3[verif])$identical.content)){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nTHE merge() FUNCTION DID NOT RETURN A CORRECT dot.coord.tidy3 DATA FRAME. CODE HAS TO BE MODIFIED")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end coordinates of tidy dots
- }
- }
- # end dot display
-
-
-
- # boxplot display (if box.fill = FALSE, otherwise, already plotted above)
- if(box.fill == TRUE){
- # overcome "work only for the filling of boxes, not for the frame. See https://github.com/tidyverse/ggplot2/issues/252"
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "fill", name = box.legend.name, values = if(length(categ.color) == 1){rep(categ.color, length(unique(data1[, categ[length(categ)]])))}else{categ.color}, guide = ggplot2::guide_legend(order = 1))) #, guide = ggplot2::guide_legend(override.aes = list(fill = levels(tempo.polygon$COLOR), color = "black")))) # values are the values of color (which is the border color in geom_box. WARNING: values = categ.color takes the numbers to make the colors if categ.color is a factor
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "color", name = box.legend.name, values = rep(hsv(0, 0, 0, alpha = box.alpha), length(unique(data1[, categ[length(categ)]]))), guide = ggplot2::guide_legend(order = 1))) # , guide = ggplot2::guide_legend(override.aes = list(color = "black", alpha = box.alpha)))) # values are the values of color (which is the border color in geom_box. WARNING: values = categ.color takes the numbers to make the colors if categ.color is a factor # outline of the polygon in black but with alpha
- }else{
- # assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_boxplot(data = data1, mapping = ggplot2::aes_string(x = categ[1], y = y, color = categ[length(categ)], fill = categ[length(categ)]), position = ggplot2::position_dodge(width = NULL), width = box.width, size = box.line.size, notch = box.notch, alpha = box.alpha, coef = if(box.whisker.kind == "no"){0}else if(box.whisker.kind == "std"){1.5}else if(box.whisker.kind == "max"){Inf}, outlier.shape = if( ! is.null(dot.color)){NA}else{21}, outlier.color = if( ! is.null(dot.color)){NA}else{if(dot.border.size == 0){NA}else{dot.border.color}}, outlier.fill = if( ! is.null(dot.color)){NA}else{NULL}, outlier.size = if( ! is.null(dot.color)){NA}else{dot.size}, outlier.stroke = if( ! is.null(dot.color)){NA}else{dot.border.size}, outlier.alpha = if( ! is.null(dot.color)){NA}else{dot.alpha})) # the color, size, etc. of the outliers are dealt here. outlier.color = NA to do not plot outliers when dots are already plotted
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_path(
- data = tempo.polygon,
- mapping = ggplot2::aes_string(x = "X", y = "Y", group = "BOX", color = categ[length(categ)]),
- size = box.line.size,
- alpha = box.alpha,
- lineend = "round",
- linejoin = "round"
- ))
- coord.names <- c(coord.names, "main.box")
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = if(box.notch == FALSE){X_BOX_INF}else{X_NOTCH_INF}, xend = if(box.notch == FALSE){X_BOX_SUP}else{X_NOTCH_SUP}, y = MEDIAN, yend = MEDIAN, group = categ[length(categ)]), color = stat$COLOR, size = box.line.size * 2, alpha = box.alpha)) #
- coord.names <- c(coord.names, "median")
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = X, xend = X, y = BOX_SUP, yend = WHISK_SUP, group = categ[length(categ)]), color = stat$COLOR, size = box.line.size, alpha = box.alpha)) #
- coord.names <- c(coord.names, "sup.whisker")
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = X, xend = X, y = BOX_INF, yend = WHISK_INF, group = categ[length(categ)]), color = stat$COLOR, size = box.line.size, alpha = box.alpha)) #
- coord.names <- c(coord.names, "inf.whisker")
- if(box.whisker.width > 0){
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = X_WHISK_INF, xend = X_WHISK_SUP, y = WHISK_SUP, yend = WHISK_SUP, group = categ[length(categ)]), color = stat$COLOR, size = box.line.size, alpha = box.alpha, lineend = "round")) #
- coord.names <- c(coord.names, "sup.whisker.edge")
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_segment(data = stat, mapping = ggplot2::aes(x = X_WHISK_INF, xend = X_WHISK_SUP, y = WHISK_INF, yend = WHISK_INF, group = categ[length(categ)]), color = stat$COLOR, size = box.line.size, alpha = box.alpha, lineend = "round")) #
- coord.names <- c(coord.names, "inf.whisker.edge")
- }
- if(box.mean == TRUE){
- # assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_point(data = stat, mapping = ggplot2::aes_string(x = "X", y = "MEAN", group = categ[length(categ)]), shape = 23, stroke = box.line.size * 2, color = stat$COLOR, size = box.mean.size, fill = NA, alpha = box.alpha)) # group used in aesthetic to do not have it in the legend. Here ggplot2::scale_discrete_manual() cannot be used because of the group easthetic
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_path(
- data = tempo.diamon.mean,
- mapping = ggplot2::aes(x = X, y = Y, group = GROUP),
- color = tempo.diamon.mean[, "COLOR"],
- size = box.line.size,
- alpha = box.alpha,
- lineend = "round",
- linejoin = "round"
- ))
- coord.names <- c(coord.names, "mean")
- }
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "fill", name = box.legend.name, values = rep(NA, length(unique(data1[, categ[length(categ)]]))))) #, guide = ggplot2::guide_legend(override.aes = list(color = categ.color)))) # values are the values of color (which is the border color in geom_box. WARNING: values = categ.color takes the numbers to make the colors if categ.color is a factor
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "color", name = box.legend.name, values = if(length(categ.color) == 1){rep(categ.color, length(unique(data1[, categ[length(categ)]])))}else{categ.color}, guide = ggplot2::guide_legend(override.aes = list(alpha = if(plot == TRUE & ((length(dev.list()) > 0 & names(dev.cur()) == "windows") | (length(dev.list()) == 0 & Sys.info()["sysname"] == "Windows"))){1}else{box.alpha})))) # , guide = ggplot2::guide_legend(override.aes = list(color = as.character(categ.color))))) # values are the values of color (which is the border color in geom_box. WARNING: values = categ.color takes the numbers to make the colors if categ.color is a factor
- if(plot == TRUE & ((length(dev.list()) > 0 & names(dev.cur()) == "windows") | (length(dev.list()) == 0 & Sys.info()["sysname"] == "Windows"))){ # if any Graph device already open and this device is "windows", or if no Graph device opened yet and we are on windows system -> prevention of alpha legend bug on windows using value 1
- # to avoid a bug on windows: if alpha argument is different from 1 for lines (transparency), then lines are not correctly displayed in the legend when using the R GUI (bug https://github.com/tidyverse/ggplot2/issues/2452). No bug when using a pdf
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") GRAPHIC DEVICE USED ON A WINDOWS SYSTEM ->\nTRANSPARENCY OF THE LINES IS INACTIVATED IN THE LEGEND TO PREVENT A WINDOWS DEPENDENT BUG (SEE https://github.com/tidyverse/ggplot2/issues/2452)\nTO OVERCOME THIS ON WINDOWS, USE ANOTHER DEVICE (pdf() FOR INSTANCE)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- if(box.alpha == 0){ # remove box legend because no boxes drawn
- # add this after the scale_xxx_manual() for boxplots
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::guides(fill = FALSE, color = FALSE)) # inactivate the legend
- }
- # end boxplot display (if box.fill = FALSE, otherwise, already plotted above)
-
-
-
-
- # stat display
- # layer after dots but ok, behind dots on the plot
- if( ! is.null(stat.disp)){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NUMBERS DISPLAYED ARE ", ifelse(stat.disp.mean == FALSE, "MEDIANS", "MEANS"))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- if(stat.disp == "top"){
- tempo.stat <- data.frame(stat, Y = y.lim[2], stringsAsFactors = TRUE) # I had to create a data frame for geom_tex() so that facet is taken into account, (ggplot2::annotate() does not deal with facet because no data and mapping arguments). Of note, facet.categ is in tempo.stat, via tempo.mean, via dot.coord
- if(stat.disp.mean == FALSE){tempo.stat$MEDIAN <- formatC(stat.nolog$MEDIAN, digit = 2, drop0trailing = TRUE, format = "f")}else{tempo.stat$MEAN <- formatC(stat.nolog$MEAN, digit = 2, drop0trailing = TRUE, format = "f")}
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_text(
- data = tempo.stat,
- mapping = ggplot2::aes_string(x = "X", y = "Y", label = ifelse(stat.disp.mean == FALSE, "MEDIAN", "MEAN")),
- size = stat.size,
- color = "black",
- angle = stat.angle,
- hjust = stat.just$hjust,
- vjust = stat.just$vjust
- )) # stat$X used here because identical to stat.nolog but has the X. WARNING: no need of order() for labels because box.coord$x set the order. For justification, see https://stackoverflow.com/questions/7263849/what-do-hjust-and-vjust-do-when-making-a-plot-using-ggplot
- coord.names <- c(coord.names, "stat.display")
- }else if(stat.disp == "above"){
- # stat coordinates
- if( ! is.null(dot.color)){ # for text just above max dot
- if(dot.tidy == FALSE){
- tempo.stat.ini <- dot.coord.rd3
- }else if(dot.tidy == TRUE){
- tempo.stat.ini <- dot.coord.tidy3
- tempo.stat.ini$x.y <- tempo.stat.ini$x.x # this is just to be able to use tempo.stat.ini$x.y for untidy or tidy dots (remember that dot.coord.tidy3$x.y is not good, see above)
- }
- stat.coord1 <- aggregate(x = tempo.stat.ini["y"], by = {x.env <- if(length(categ) == 1){list(tempo.stat.ini$group, tempo.stat.ini$PANEL, tempo.stat.ini$x.y, tempo.stat.ini[, categ[1]])}else if(length(categ) == 2){list(tempo.stat.ini$group, tempo.stat.ini$PANEL, tempo.stat.ini$x.y, tempo.stat.ini[, categ[1]], tempo.stat.ini[, categ[2]])} ; names(x.env) <- if(length(categ) == 1){c("group", "PANEL", "x.y", categ[1])}else if(length(categ) == 2){c("group", "PANEL", "x.y", categ[1], categ[2])} ; x.env}, FUN = min, na.rm = TRUE)
- names(stat.coord1)[names(stat.coord1) == "y"] <- "dot.min"
- stat.coord2 <- aggregate(x = tempo.stat.ini["y"], by = {x.env <- if(length(categ) == 1){list(tempo.stat.ini$group, tempo.stat.ini$PANEL, tempo.stat.ini$x.y, tempo.stat.ini[, categ[1]])}else if(length(categ) == 2){list(tempo.stat.ini$group, tempo.stat.ini$PANEL, tempo.stat.ini$x.y, tempo.stat.ini[, categ[1]], tempo.stat.ini[, categ[2]])} ; names(x.env) <- if(length(categ) == 1){c("group", "PANEL", "x.y", categ[1])}else if(length(categ) == 2){c("group", "PANEL", "x.y", categ[1], categ[2])} ; x.env}, FUN = max, na.rm = TRUE)
- names(stat.coord2) <- paste0(names(stat.coord2), "_from.dot.max")
- names(stat.coord2)[names(stat.coord2) == "y_from.dot.max"] <- "dot.max"
- stat.coord3 <- cbind(box.coord[order(box.coord$group, box.coord$PANEL), ], stat.coord1[order(stat.coord1$group, stat.coord1$x.y), ], stat.coord2[order(stat.coord2$group, stat.coord2$x.y), ], stringsAsFactors = TRUE) #
- if( ! all(identical(round(stat.coord3$x, 9), round(as.numeric(stat.coord3$x.y), 9)))){ # as.numeric() because stat.coord3$x is class "mapped_discrete" "numeric"
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nFUSION OF box.coord, stat.coord1 AND stat.coord2 ACCORDING TO box.coord$x, stat.coord1$x.y AND stat.coord2$x.y IS NOT CORRECT. CODE HAS TO BE MODIFIED")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # text.coord <- stat.coord3[, c("x", "group", "dot.min", "dot.max")]
- # names(text.coord)[names(text.coord) == "dot.min"] <- "text.min.pos"
- #names(text.coord)[names(text.coord) == "dot.max"] <- "text.max.pos"
- box.coord <- box.coord[order(box.coord$x, box.coord$group, box.coord$PANEL), ]
- # text.coord <- text.coord[order(text.coord$x), ] # to be sure to have the two objects in the same order for x. WARNING: cannot add identical(as.integer(text.coord$group), as.integer(box.coord$group)) because with error, the correspondence between x and group is not the same
- stat.coord3 <- stat.coord3[order(stat.coord3$x, stat.coord3$group, stat.coord3$PANEL), ] # to be sure to have the two objects in the same order for x. WARNING: cannot add identical(as.integer(text.coord$group), as.integer(box.coord$group)) because with error, the correspondence between x and group is not the same
- if( ! (identical(box.coord$x, stat.coord3$x) & identical(box.coord$group, stat.coord3$group) & identical(box.coord$PANEL, stat.coord3$PANEL))){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\ntext.coord AND box.coord DO NOT HAVE THE SAME x, group AND PANEL COLUMN CONTENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }else{
- stat.coord3 <- box.coord
- }
- stat.coord3 <- data.frame(
- stat.coord3,
- Y = stat.coord3[, ifelse(
- is.null(dot.color),
- ifelse(diff(y.lim) > 0, "ymax", "ymin"),
- ifelse(diff(y.lim) > 0, "ymax_final", "ymin_final")
- )],
- stringsAsFactors = TRUE
- ) # ymax is top whisker, ymax_final is top dot
- # stat.coord3 <- data.frame(stat.coord3, Y = vector("numeric", length = nrow(stat.coord3)), stringsAsFactors = TRUE)
- # check.Y <- as.logical(stat.coord3$Y) # convert everything in Y into FALSE (because Y is full of zero)
- # end stat coordinates
- # stat display
- # performed twice: first for y values >=0, then y values < 0, because only a single value allowed for hjust anf vjust
- if(stat.disp.mean == FALSE){
- tempo.center.ref <- "middle"
- }else{
- tempo.center.ref <- "MEAN"
- }
- # if(is.null(dot.color)){
- # tempo.low.ref <- "ymin"
- # tempo.high.ref <- "ymax"
- # }else{
- # tempo.low.ref <- "ymin_final"
- # tempo.high.ref <- "ymax_final"
- # }
- # tempo.log.high <- if(diff(y.lim) > 0){stat.coord3[, tempo.center.ref] >= 0}else{stat.coord3[, tempo.center.ref] < 0}
- # tempo.log.low <- if(diff(y.lim) > 0){stat.coord3[, tempo.center.ref] < 0}else{stat.coord3[, tempo.center.ref] >= 0}
- # stat.coord3$Y[tempo.log.high] <- stat.coord3[tempo.log.high, tempo.high.ref]
- # stat.coord3$Y[tempo.log.low] <- stat.coord3[tempo.log.low, tempo.low.ref]
- # add distance
- stat.coord3$Y <- stat.coord3$Y + diff(y.lim) * stat.dist / 100
- # end add distance
- # correct median or mean text format
- if(y.log != "no"){
- stat.coord3[, tempo.center.ref] <- ifelse(y.log == "log2", 2, 10)^(stat.coord3[, tempo.center.ref])
- }
- stat.coord3[, tempo.center.ref] <- formatC(stat.coord3[, tempo.center.ref], digit = 2, drop0trailing = TRUE, format = "f")
- # end correct median or mean text format
- # if(any(tempo.log.high) == TRUE){
- # tempo.stat <- stat.coord3[tempo.log.high,]
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_text(
- data = stat.coord3,
- mapping = ggplot2::aes_string(x = "x", y = "Y", label = tempo.center.ref),
- size = stat.size,
- color = "black",
- angle = stat.angle,
- hjust = stat.just$hjust,
- vjust = stat.just$vjust
- )) # WARNING: no need of order() for labels because box.coord$x set the order
- coord.names <- c(coord.names, "stat.display")
- # }
- # if(any(tempo.log.low) == TRUE){
- # tempo.stat <- stat.coord3[tempo.log.low,]
- # assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::geom_text(
- # data = tempo.stat,
- # mapping = ggplot2::aes_string(x = "x", y = "Y", label = tempo.center.ref),
- # size = stat.size,
- # color = "black",
- # hjust = ifelse(vertical == TRUE, 0.5, 0.5 + stat.dist),
- # vjust = ifelse(vertical == TRUE, 0.5 + stat.dist, 0.5)
- # )) # WARNING: no need of order() for labels because box.coord$x set the order
- # coord.names <- c(coord.names, "stat.display.negative")
- # }
- # end stat display
- }else{
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 5")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- }
- # end stat display
- # legend management
- if(legend.show == FALSE){ # must be here because must be before bef.final.plot <-
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::guides(fill = FALSE, color = FALSE, alpha = FALSE)) # inactivate the initial legend
- }
- # end legend management
-
-
-
- # y scale management (cannot be before dot plot management)
- # the rescaling aspect is complicated and not intuitive. See:
- # explaination: https://github.com/tidyverse/ggplot2/issues/3948
- # the oob argument of scale_y_continuous() https://ggplot2.tidyverse.org/reference/scale_continuous.html
- # see also https://github.com/rstudio/cheatsheets/blob/master/data-visualization-2.1.pdf
- # secondary ticks
- bef.final.plot <- ggplot2::ggplot_build(eval(parse(text = paste(paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "), ' + if(vertical == TRUE){ggplot2::scale_y_continuous(expand = c(0, 0), limits = sort(y.lim), oob = scales::rescale_none)}else{ggplot2::coord_flip(ylim = y.lim)}')))) # here I do not need the x-axis and y-axis orientation, I just need the number of main ticks and the legend
- tempo.coord <- bef.final.plot$layout$panel_params[[1]]
- # y.second.tick.positions: coordinates of secondary ticks (only if y.second.tick.nb argument is non NULL or if y.log argument is different from "no")
- if(y.log != "no"){ # integer main ticks for log2 and log10
- tempo.scale <- (as.integer(min(y.lim, na.rm = TRUE)) - 1):(as.integer(max(y.lim, na.rm = TRUE)) + 1)
- }else{
- tempo <- if(is.null(attributes(tempo.coord$y$breaks))){tempo.coord$y$breaks}else{unlist(attributes(tempo.coord$y$breaks))}
- if(all(is.na(tempo))){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nONLY NA IN tempo.coord$y$breaks")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- tempo.scale <- fun_scale(lim = y.lim, n = ifelse(is.null(y.tick.nb), length(tempo[ ! is.na(tempo)]), y.tick.nb)) # in ggplot 3.3.0, tempo.coord$y.major_source replaced by tempo.coord$y$breaks. If fact: n = ifelse(is.null(y.tick.nb), length(tempo[ ! is.na(tempo)]), y.tick.nb)) replaced by n = ifelse(is.null(y.tick.nb), 4, y.tick.nb))
- }
- y.second.tick.values <- NULL
- y.second.tick.pos <- NULL
- if(y.log != "no"){
- tempo <- fun_inter_ticks(lim = y.lim, log = y.log)
- y.second.tick.values <- tempo$values
- y.second.tick.pos <- tempo$coordinates
- # if(vertical == TRUE){ # do not remove in case the bug is fixed
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::annotate(geom = "segment", y = y.second.tick.pos, yend = y.second.tick.pos, x = tempo.coord$x.range[1], xend = tempo.coord$x.range[1] + diff(tempo.coord$x.range) / 80))
- # }else{ # not working because of the ggplot2 bug
- # assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::annotate(geom = "segment", x = y.second.tick.pos, xend = y.second.tick.pos, y = tempo.coord$y.range[1], yend = tempo.coord$y.range[1] + diff(tempo.coord$y.range) / 80))
- # }
- coord.names <- c(coord.names, "y.second.tick.positions")
- }else if(( ! is.null(y.second.tick.nb)) & y.log == "no"){
- # if(y.second.tick.nb > 0){ #inactivated because already checked before
- tempo <- fun_inter_ticks(lim = y.lim, log = y.log, breaks = tempo.scale, n = y.second.tick.nb)
- y.second.tick.values <- tempo$values
- y.second.tick.pos <- tempo$coordinates
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::annotate(
- geom = "segment",
- y = y.second.tick.pos,
- yend = y.second.tick.pos,
- x = if(vertical == TRUE){tempo.coord$x.range[1]}else{tempo.coord$y.range[1]},
- xend = if(vertical == TRUE){tempo.coord$x.range[1] + diff(tempo.coord$x.range) / 80}else{tempo.coord$y.range[1] + diff(tempo.coord$y.range) / 80}
- ))
- coord.names <- c(coord.names, "y.second.tick.positions")
- }
- # end y.second.tick.positions
- # for the ggplot2 bug with y.log, this does not work: eval(parse(text = ifelse(vertical == FALSE & y.log == "log10", "ggplot2::scale_x_continuous", "ggplot2::scale_y_continuous")))
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_y_continuous(
- breaks = tempo.scale,
- minor_breaks = y.second.tick.pos,
- labels = if(y.log == "log10"){scales::trans_format("identity", scales::math_format(10^.x))}else if(y.log == "log2"){scales::trans_format("identity", scales::math_format(2^.x))}else if(y.log == "no"){ggplot2::waiver()}else{tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 6") ; stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)}, # == in stop() to be able to add several messages between ==
- expand = c(0, 0), # remove space after after axis limits
- limits = sort(y.lim), # NA indicate that limits must correspond to data limits but ylim() already used
- oob = scales::rescale_none,
- trans = ifelse(diff(y.lim) < 0, "reverse", "identity") # equivalent to ggplot2::scale_y_reverse() but create the problem of y-axis label disappearance with y.lim decreasing. Thus, do not use. Use ylim() below and after this
- ))
- if(vertical == TRUE){
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::coord_cartesian(ylim = y.lim)) #problem of ggplot2::ylim() is taht it redraw new breaks # coord_cartesian(ylim = y.lim)) not used because bug -> y-axis label disappearance with y.lim decreasing # clip = "off" to have secondary ticks outside plot region does not work
- }else{
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::coord_flip(ylim = y.lim)) # clip = "off" to have secondary ticks outside plot region does not work # create the problem of y-axis label disappearance with y.lim decreasing
-
- }
- # end y scale management (cannot be before dot plot management)
-
-
- # legend management
- if( ! is.null(legend.width)){
- legend.final <- fun_gg_get_legend(ggplot_built = bef.final.plot, fun.name = function.name, lib.path = lib.path) # get legend
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::guides(fill = FALSE, color = FALSE, alpha = FALSE)) # inactivate the initial legend
- if(is.null(legend.final) & plot == TRUE){ # even if any(unlist(legend.disp)) is TRUE
- legend.final <- ggplot2::ggplot()+ggplot2::theme_void() # empty graph instead of legend
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") LEGEND REQUESTED (NON NULL categ ARGUMENT OR legend.show ARGUMENT SET TO TRUE)\nBUT IT SEEMS THAT THE PLOT HAS NO LEGEND -> EMPTY LEGEND SPACE CREATED BECAUSE OF THE NON NULL legend.width ARGUMENT\n")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- # end legend management
-
-
- # drawing
- fin.plot <- suppressMessages(suppressWarnings(eval(parse(text = paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + ")))))
- grob.save <- NULL
- if(plot == TRUE){
- # following lines inactivated because of problem in warn.recov and message.recov
- # assign("env_fun_get_message", new.env())
- # assign("tempo.gg.name", tempo.gg.name, envir = env_fun_get_message)
- # assign("tempo.gg.count", tempo.gg.count, envir = env_fun_get_message)
- # assign("add", add, envir = env_fun_get_message)
- # two next line: for the moment, I cannot prevent the warning printing
- # warn.recov <- fun_get_message(paste(paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "), if(is.null(add)){NULL}else{add}), kind = "warning", header = FALSE, print.no = FALSE, env = env_fun_get_message) # for recovering warnings printed by ggplot() functions
- # message.recov <- fun_get_message('print(eval(parse(text = paste(paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "), if(is.null(add)){NULL}else{add}))))', kind = "message", header = FALSE, print.no = FALSE, env = env_fun_get_message) # for recovering messages printed by ggplot() functions
- # if( ! (return == TRUE & return.ggplot == TRUE)){ # because return() plots when return.ggplot is TRUE # finally not used -> see return.ggplot description
- if(is.null(legend.width)){
- grob.save <- suppressMessages(suppressWarnings(gridExtra::grid.arrange(fin.plot)))
- }else{
- grob.save <-suppressMessages(suppressWarnings(gridExtra::grid.arrange(fin.plot, legend.final, ncol=2, widths=c(1, legend.width))))
- }
- # }
- # suppressMessages(suppressWarnings(print(eval(parse(text = paste(paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "), if(is.null(add)){NULL}else{add}))))))
- }else{
- # following lines inactivated because of problem in warn.recov and message.recov
- # message.recov <- NULL
- # warn.recov <- NULL
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") PLOT NOT SHOWN AS REQUESTED")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- # end drawing
-
-
-
- # outputs
- # following lines inactivated because of problem in warn.recov and message.recov
- # if( ! (is.null(warn) & is.null(warn.recov) & is.null(message.recov))){
- # warn <- paste0(warn, "\n\n", if(length(warn.recov) > 0 | length(message.recov) > 0){paste0(paste0("MESSAGES FROM ggplot2 FUNCTIONS: ", ifelse( ! is.null(warn.recov), unique(message.recov), ""), ifelse( ! is.null(message.recov), unique(message.recov), ""), collapse = "\n\n"), "\n\n")})
- # }else if( ! (is.null(warn) & is.null(warn.recov)) & is.null(message.recov)){
- # warn <- paste0(warn, "\n\n", if(length(warn.recov) > 0){paste0(paste0("MESSAGES FROM ggplot2 FUNCTIONS: ", unique(warn.recov), collapse = "\n\n"), "\n\n")})
- # }else if( ! (is.null(warn) & is.null(message.recov)) & is.null(warn.recov)){
- # warn <- paste0(warn, "\n\n", if(length(message.recov) > 0){paste0(paste0("MESSAGES FROM ggplot2 FUNCTIONS: ", unique(message.recov), collapse = "\n\n"), "\n\n")})
- # }
- if(warn.print == TRUE & ! is.null(warn)){
- options(warning.length = 8170)
- on.exit(warning(paste0("FROM ", function.name, ":\n\n", warn), call. = FALSE))
- on.exit(exp = options(warning.length = ini.warning.length), add = TRUE)
- }
- if(return == TRUE){
- tempo.output <- ggplot2::ggplot_build(fin.plot)
- tempo.output$data <- tempo.output$data[-1] # remove the first data because corresponds to the initial empty boxplot
- if(length(tempo.output$data) != length(coord.names)){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nlength(tempo.output$data) AND length(coord.names) MUST BE IDENTICAL. CODE HAS TO BE MODIFIED")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE) # == in stop() to be able to add several messages between ==
- }else{
- names(tempo.output$data) <- coord.names
- tempo.output$data <- tempo.output$data[coord.names != "bad_remove"]
- }
- tempo <- tempo.output$layout$panel_params[[1]]
- output <- list(
- data = data1.ini,
- stat = stat.nolog,
- removed.row.nb = removed.row.nb,
- removed.rows = removed.rows,
- plot = c(tempo.output$data, y.second.tick.values = list(y.second.tick.values)),
- panel = facet.categ,
- axes = list(
- x.range = tempo$x.range,
- x.labels = if(is.null(attributes(tempo$x$breaks))){tempo$x$breaks}else{tempo$x$scale$get_labels()}, # is.null(attributes(tempo$x$breaks)) test if it is number (TRUE) or character (FALSE)
- x.positions = if(is.null(attributes(tempo$x$breaks))){tempo$x$breaks}else{unlist(attributes(tempo$x$breaks))},
- y.range = tempo$y.range,
- y.labels = if(is.null(attributes(tempo$y$breaks))){tempo$y$breaks}else{tempo$y$scale$get_labels()},
- y.positions = if(is.null(attributes(tempo$y$breaks))){tempo$y$breaks}else{unlist(attributes(tempo$y$breaks))}
- ),
- warn = paste0("\n", warn, "\n\n"),
- ggplot = if(return.ggplot == TRUE){fin.plot}else{NULL}, # fin.plot plots the graph if return == TRUE
- gtable = if(return.gtable == TRUE){grob.save}else{NULL}
- )
- return(output) # this plots the graph if return.ggplot is TRUE and if no assignment
- }
- # end outputs
- # end main code
+# kind of geom_point (vectorial or raster)
+scatter.kind <- vector("list", length = length(data1)) # list of same length as data1, that will be used to use either ggplot2::geom_point() (vectorial dot layer) or fun_gg_point_rast() (raster dot layer)
+fix.ratio <- FALSE
+if(is.null(raster.threshold)){
+if(raster == TRUE){
+scatter.kind[] <- "fun_gg_point_rast" # not important to fill everything: will be only used when geom == "geom_point"
+fix.ratio <- TRUE
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") RASTER PLOT GENERATED -> ASPECT RATIO OF THE PLOT REGION SET BY THE raster.ratio ARGUMENT (", fun_round(raster.ratio, 2), ") TO AVOID A BUG OF ELLIPSOID DOT DRAWING")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}else{
+scatter.kind[] <- "ggplot2::geom_point"
}
+}else{
+for(i2 in 1:length(data1)){
+if(geom[[i2]] == "geom_point"){
+if(nrow(data1[[i2]]) <= raster.threshold){
+scatter.kind[[i2]] <- "ggplot2::geom_point"
+}else{
+scatter.kind[[i2]] <- "fun_gg_point_rast"
+fix.ratio <- TRUE
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i2, " OF data1 ARGUMENT")), " LAYER AS RASTER (NOT VECTORIAL)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+}
+if(any(unlist(scatter.kind) == "fun_gg_point_rast")){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") RASTER PLOT GENERATED -> ASPECT RATIO OF THE PLOT REGION SET BY THE raster.ratio ARGUMENT (", fun_round(raster.ratio, 2), ") TO AVOID A BUG OF ELLIPSOID DOT DRAWING")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}
+# end kind of geom_point (vectorial or raster)
+
+
+
+
+# no need loop part
+coord.names <- NULL
+tempo.gg.name <- "gg.indiv.plot."
+tempo.gg.count <- 0
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = paste0("ggplot2::ggplot()", if(is.null(add)){""}else{add})))) # add added here to have the facets
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::xlab(if(is.null(x.lab)){x[[1]]}else{x.lab}))
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::ylab(if(is.null(y.lab)){y[[1]]}else{y.lab}))
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::ggtitle(title))
+# text angle management
+x.tempo.just <- fun_gg_just(angle = x.text.angle, pos = "bottom", kind = "axis")
+y.tempo.just <- fun_gg_just(angle = y.text.angle, pos = "left", kind = "axis")
+# end text angle management
+add.check <- TRUE
+if( ! is.null(add)){ # if add is NULL, then = 0
+if(grepl(pattern = "ggplot2::theme", add) == TRUE){
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") \"ggplot2::theme\" STRING DETECTED IN THE add ARGUMENT\n-> INTERNAL GGPLOT2 THEME FUNCTIONS theme() AND theme_classic() HAVE BEEN INACTIVATED, TO BE USED BY THE USER\n-> article ARGUMENT WILL BE IGNORED\nIT IS RECOMMENDED TO USE \"+ theme(aspect.ratio = raster.ratio)\" IF RASTER MODE IS ACTIVATED")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+add.check <- FALSE
+}
+}
+if(add.check == TRUE & article == TRUE){
+# WARNING: not possible to add several times theme(). NO message but the last one overwrites the others
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::theme_classic(base_size = text.size))
+if(grid == TRUE){
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), m.gg <- ggplot2::theme(
+text = ggplot2::element_text(size = text.size),
+plot.title = ggplot2::element_text(size = title.text.size), # stronger than text
+legend.key = ggplot2::element_rect(color = "white", size = 1.5), # size of the frame of the legend
+line = ggplot2::element_line(size = 0.5),
+axis.line.y.left = ggplot2::element_line(colour = "black"), # draw lines for the y axis
+axis.line.x.bottom = ggplot2::element_line(colour = "black"), # draw lines for the x axis
+panel.grid.major.x = ggplot2::element_line(colour = "grey85", size = 0.75),
+panel.grid.minor.x = ggplot2::element_line(colour = "grey90", size = 0.25),
+panel.grid.major.y = ggplot2::element_line(colour = "grey85", size = 0.75),
+panel.grid.minor.y = ggplot2::element_line(colour = "grey90", size = 0.25),
+axis.text.x = ggplot2::element_text(angle = x.tempo.just$angle, hjust = x.tempo.just$hjust, vjust = x.tempo.just$vjust),
+axis.text.y = ggplot2::element_text(angle = y.tempo.just$angle, hjust = y.tempo.just$hjust, vjust = y.tempo.just$vjust),
+aspect.ratio = if(fix.ratio == TRUE){raster.ratio}else{NULL} # for raster
+))
+}else{
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), m.gg <- ggplot2::theme(
+text = ggplot2::element_text(size = text.size),
+plot.title = ggplot2::element_text(size = title.text.size), # stronger than text
+line = ggplot2::element_line(size = 0.5),
+legend.key = ggplot2::element_rect(color = "white", size = 1.5), # size of the frame of the legend
+axis.line.y.left = ggplot2::element_line(colour = "black"),
+axis.line.x.bottom = ggplot2::element_line(colour = "black"),
+axis.text.x = ggplot2::element_text(angle = x.tempo.just$angle, hjust = x.tempo.just$hjust, vjust = x.tempo.just$vjust),
+axis.text.y = ggplot2::element_text(angle = y.tempo.just$angle, hjust = y.tempo.just$hjust, vjust = y.tempo.just$vjust),
+aspect.ratio = if(fix.ratio == TRUE){raster.ratio}else{NULL} # for raster
+))
+}
+}else if(add.check == TRUE & article == FALSE){
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), m.gg <- ggplot2::theme(
+text = ggplot2::element_text(size = text.size),
+plot.title = ggplot2::element_text(size = title.text.size), # stronger than text
+line = ggplot2::element_line(size = 0.5),
+legend.key = ggplot2::element_rect(color = "white", size = 1.5), # size of the frame of the legend
+panel.background = ggplot2::element_rect(fill = "grey95"),
+axis.line.y.left = ggplot2::element_line(colour = "black"),
+axis.line.x.bottom = ggplot2::element_line(colour = "black"),
+panel.grid.major.x = ggplot2::element_line(colour = "grey85", size = 0.75),
+panel.grid.minor.x = ggplot2::element_line(colour = "grey90", size = 0.25),
+panel.grid.major.y = ggplot2::element_line(colour = "grey85", size = 0.75),
+panel.grid.minor.y = ggplot2::element_line(colour = "grey90", size = 0.25),
+strip.background = ggplot2::element_rect(fill = "white", colour = "black"),
+axis.text.x = ggplot2::element_text(angle = x.tempo.just$angle, hjust = x.tempo.just$hjust, vjust = x.tempo.just$vjust),
+axis.text.y = ggplot2::element_text(angle = y.tempo.just$angle, hjust = y.tempo.just$hjust, vjust = y.tempo.just$vjust),
+aspect.ratio = if(fix.ratio == TRUE){raster.ratio}else{NULL} # for raster
+# do not work -> legend.position = "none" # to remove the legend completely: https://www.datanovia.com/en/blog/how-to-remove-legend-from-a-ggplot/
+))
+}
+# end no need loop part
+
+
+# loop part
+point.count <- 0
+line.count <- 0
+lg.order <- vector(mode = "list", length = 6) # order of the legend
+lg.order <- lapply(lg.order, as.numeric) # order of the legend
+lg.color <- vector(mode = "list", length = 6) # color of the legend
+lg.dot.shape <- vector(mode = "list", length = 6) # etc.
+lg.dot.size <- vector(mode = "list", length = 6) # etc.
+lg.dot.size <- lapply(lg.dot.size, as.numeric) # etc.
+lg.dot.border.size <- vector(mode = "list", length = 6) # etc.
+lg.dot.border.size <- lapply(lg.dot.border.size, as.numeric) # etc.
+lg.dot.border.color <- vector(mode = "list", length = 6) # etc.
+lg.line.size <- vector(mode = "list", length = 6) # etc.
+lg.line.size <- lapply(lg.line.size, as.numeric) # etc.
+lg.line.type <- vector(mode = "list", length = 6) # etc.
+lg.alpha <- vector(mode = "list", length = 6) # etc.
+lg.alpha <- lapply(lg.alpha, as.numeric) # etc.
+for(i1 in 1:length(data1)){
+if(geom[[i1]] == "geom_point"){
+point.count <- point.count + 1
+if(point.count == 1){
+fin.lg.disp[[1]] <- legend.disp[[point.count + line.count]]
+lg.order[[1]] <- point.count + line.count
+lg.color[[1]] <- color[[i1]] # if color == NULL -> NULL
+lg.dot.shape[[1]] <- dot.shape[[i1]]
+lg.dot.size[[1]] <- dot.size[[i1]]
+lg.dot.border.size[[1]] <- dot.border.size[[i1]]
+lg.dot.border.color[[1]] <- dot.border.color[[i1]] # if dot.border.color == NULL -> NULL
+if(plot == TRUE & fin.lg.disp[[1]] == TRUE & dot.shape[[1]] %in% 0:14 & ((length(dev.list()) > 0 & names(dev.cur()) == "windows") | (length(dev.list()) == 0 & Sys.info()["sysname"] == "Windows"))){ # if any Graph device already open and this device is "windows", or if no Graph device opened yet and we are on windows system -> prevention of alpha legend bug on windows using value 1
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") GRAPHIC DEVICE USED ON A WINDOWS SYSTEM ->\nTRANSPARENCY OF THE DOTS (DOT LAYER NUMBER ", point.count, ") IS INACTIVATED IN THE LEGEND TO PREVENT A WINDOWS DEPENDENT BUG (SEE https://github.com/tidyverse/ggplot2/issues/2452)\nTO OVERCOME THIS ON WINDOWS, USE ANOTHER DEVICE (pdf() FOR INSTANCE)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+lg.alpha[[1]] <- 1 # to avoid a bug on windows: if alpha argument is different from 1 for lines (transparency), then lines are not correctly displayed in the legend when using the R GUI (bug https://github.com/tidyverse/ggplot2/issues/2452). No bug when using a pdf
+}else{
+lg.alpha[[1]] <- alpha[[i1]]
+}
+class.categ <- levels(factor(data1[[i1]][, categ[[i1]]]))
+for(i5 in 1:length(color[[i1]])){ # or length(class.categ). It is the same because already checked that lengths are the same
+tempo.data.frame <- data1[[i1]][data1[[i1]][, categ[[i1]]] == class.categ[i5], ]
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = scatter.kind[[i1]]))(data = tempo.data.frame, mapping = ggplot2::aes_string(x = x[[i1]], y = y[[i1]], fill = categ[[i1]]), shape = dot.shape[[i1]], size = dot.size[[i1]], stroke = dot.border.size[[i1]], color = if(dot.shape[[i1]] %in% 21:24 & ! is.null(dot.border.color)){dot.border.color[[i1]]}else{color[[i1]][i5]}, alpha = alpha[[i1]], show.legend = if(i5 == 1){TRUE}else{FALSE})) # WARNING: a single color allowed for color argument outside aesthetic, but here a single color for border --> loop could be inactivated but kept for commodity # legend.show option do not remove the legend, only the aesthetic of the legend (dot, line, etc.). Used here to avoid multiple layers of legend which corrupt transparency
+coord.names <- c(coord.names, paste0(geom[[i1]], ".", class.categ[i5]))
+}
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_fill_manual(name = if(is.null(legend.name)){NULL}else{legend.name[[i1]]}, values = as.character(color[[i1]]), breaks = class.categ)) # values are the values of fill, breaks reorder the classes according to class.categ in the legend, order argument of guide_legend determines the order of the different aesthetics in the legend (not order of classes). See guide_legend settings of scale_..._manual below
+}
+if(point.count == 2){
+fin.lg.disp[[2]] <- legend.disp[[point.count + line.count]]
+lg.order[[2]] <- point.count + line.count
+lg.color[[2]] <- color[[i1]] # if color == NULL -> NULL
+lg.dot.shape[[2]] <- dot.shape[[i1]]
+lg.dot.size[[2]] <- dot.size[[i1]]
+lg.dot.border.size[[2]] <- dot.border.size[[i1]]
+lg.dot.border.color[[2]] <- dot.border.color[[i1]] # if dot.border.color == NULL -> NULL
+if(plot == TRUE & fin.lg.disp[[2]] == TRUE & dot.shape[[2]] %in% 0:14 & ((length(dev.list()) > 0 & names(dev.cur()) == "windows") | (length(dev.list()) == 0 & Sys.info()["sysname"] == "Windows"))){ # if any Graph device already open and this device is "windows", or if no Graph device opened yet and we are on windows system -> prevention of alpha legend bug on windows using value 1
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") GRAPHIC DEVICE USED ON A WINDOWS SYSTEM ->\nTRANSPARENCY OF THE DOTS (DOT LAYER NUMBER ", point.count, ") IS INACTIVATED IN THE LEGEND TO PREVENT A WINDOWS DEPENDENT BUG (SEE https://github.com/tidyverse/ggplot2/issues/2452)\nTO OVERCOME THIS ON WINDOWS, USE ANOTHER DEVICE (pdf() FOR INSTANCE)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+lg.alpha[[2]] <- 1 # to avoid a bug on windows: if alpha argument is different from 1 for lines (transparency), then lines are not correctly displayed in the legend when using the R GUI (bug https://github.com/tidyverse/ggplot2/issues/2452). No bug when using a pdf
+}else{
+lg.alpha[[2]] <- alpha[[i1]]
+}
+class.categ <- levels(factor(data1[[i1]][, categ[[i1]]]))
+for(i5 in 1:length(color[[i1]])){ # or length(class.categ). It is the same because already checked that lengths are the same
+tempo.data.frame <- data1[[i1]][data1[[i1]][, categ[[i1]]] == class.categ[i5], ]
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = scatter.kind[[i1]]))(data = tempo.data.frame, mapping = ggplot2::aes_string(x = x[[i1]], y = y[[i1]], shape = categ[[i1]]), size = dot.size[[i1]], stroke = dot.border.size[[i1]], fill = color[[i1]][i5], color = if(dot.shape[[i1]] %in% 21:24 & ! is.null(dot.border.color)){dot.border.color[[i1]]}else{color[[i1]][i5]}, alpha = alpha[[i1]], show.legend = FALSE)) # WARNING: a single color allowed for fill argument outside aesthetic, hence the loop # legend.show option do not remove the legend, only the aesthetic of the legend (dot, line, etc.). Used here to avoid multiple layers of legend which corrupt transparency
+coord.names <- c(coord.names, paste0(geom[[i1]], ".", class.categ[i5]))
+}
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_shape_manual(name = if(is.null(legend.name)){NULL}else{legend.name[[i1]]}, values = rep(dot.shape[[i1]], length(color[[i1]])), breaks = class.categ)) # values are the values of shape, breaks reorder the classes according to class.categ in the legend. See guide_legend settings of scale_..._manual below
+}
+if(point.count == 3){
+fin.lg.disp[[3]] <- legend.disp[[point.count + line.count]]
+lg.order[[3]] <- point.count + line.count
+lg.color[[3]] <- color[[i1]] # if color == NULL -> NULL
+lg.dot.shape[[3]] <- dot.shape[[i1]]
+lg.dot.size[[3]] <- dot.size[[i1]]
+lg.dot.border.size[[3]] <- dot.border.size[[i1]]
+lg.dot.border.color[[3]] <- dot.border.color[[i1]] # if dot.border.color == NULL -> NULL
+if(plot == TRUE & fin.lg.disp[[3]] == TRUE & dot.shape[[3]] %in% 0:14 & ((length(dev.list()) > 0 & names(dev.cur()) == "windows") | (length(dev.list()) == 0 & Sys.info()["sysname"] == "Windows"))){ # if any Graph device already open and this device is "windows", or if no Graph device opened yet and we are on windows system -> prevention of alpha legend bug on windows using value 1
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") GRAPHIC DEVICE USED ON A WINDOWS SYSTEM ->\nTRANSPARENCY OF THE DOTS (DOT LAYER NUMBER ", point.count, ") IS INACTIVATED IN THE LEGEND TO PREVENT A WINDOWS DEPENDENT BUG (SEE https://github.com/tidyverse/ggplot2/issues/2452)\nTO OVERCOME THIS ON WINDOWS, USE ANOTHER DEVICE (pdf() FOR INSTANCE)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+lg.alpha[[3]] <- 1 # to avoid a bug on windows: if alpha argument is different from 1 for lines (transparency), then lines are not correctly displayed in the legend when using the R GUI (bug https://github.com/tidyverse/ggplot2/issues/2452). No bug when using a pdf
+}else{
+lg.alpha[[3]] <- alpha[[i1]]
+}
+class.categ <- levels(factor(data1[[i1]][, categ[[i1]]]))
+for(i5 in 1:length(color[[i1]])){ # or length(class.categ). It is the same because already checked that lengths are the same
+tempo.data.frame <- data1[[i1]][data1[[i1]][, categ[[i1]]] == class.categ[i5], ]
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = scatter.kind[[i1]]))(data = tempo.data.frame, mapping = ggplot2::aes_string(x = x[[i1]], y = y[[i1]], stroke = categ[[i1]]), shape = dot.shape[[i1]], size = dot.size[[i1]], fill = color[[i1]][i5], stroke = dot.border.size[[i1]], color = if(dot.shape[[i1]] %in% 21:24 & ! is.null(dot.border.color)){dot.border.color[[i1]]}else{color[[i1]][i5]}, alpha = alpha[[i1]], show.legend = FALSE)) # WARNING: a single color allowed for color argument outside aesthetic, hence the loop # legend.show option do not remove the legend, only the aesthetic of the legend (dot, line, etc.). Used here to avoid multiple layers of legend which corrupt transparency
+coord.names <- c(coord.names, paste0(geom[[i1]], ".", class.categ[i5]))
+}
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "stroke", name = if(is.null(legend.name)){NULL}else{legend.name[[i1]]}, values = rep(dot.border.size[[i1]], length(color[[i1]])), breaks = class.categ)) # values are the values of stroke, breaks reorder the classes according to class.categ in the legend. See guide_legend settings of scale_..._manual below
+}
+}else{
+line.count <- line.count + 1
+if(line.count == 1){
+fin.lg.disp[[4]] <- legend.disp[[point.count + line.count]]
+lg.order[[4]] <- point.count + line.count
+lg.color[[4]] <- color[[i1]] # if color == NULL -> NULL
+lg.line.size[[4]] <- line.size[[i1]]
+lg.line.type[[4]] <- line.type[[i1]]
+if(plot == TRUE & fin.lg.disp[[4]] == TRUE & ((length(dev.list()) > 0 & names(dev.cur()) == "windows") | (length(dev.list()) == 0 & Sys.info()["sysname"] == "Windows"))){ # if any Graph device already open and this device is "windows", or if no Graph device opened yet and we are on windows system -> prevention of alpha legend bug on windows using value 1
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") GRAPHIC DEVICE USED ON A WINDOWS SYSTEM ->\nTRANSPARENCY OF THE LINES (LINE LAYER NUMBER ", line.count, ") IS INACTIVATED IN THE LEGEND TO PREVENT A WINDOWS DEPENDENT BUG (SEE https://github.com/tidyverse/ggplot2/issues/2452)\nTO OVERCOME THIS ON WINDOWS, USE ANOTHER DEVICE (pdf() FOR INSTANCE)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+lg.alpha[[4]] <- 1 # to avoid a bug on windows: if alpha argument is different from 1 for lines (transparency), then lines are not correctly displayed in the legend when using the R GUI (bug https://github.com/tidyverse/ggplot2/issues/2452). No bug when using a pdf
+}else{
+lg.alpha[[4]] <- alpha[[i1]]
+}
+class.categ <- levels(factor(data1[[i1]][, categ[[i1]]]))
+for(i5 in 1:length(color[[i1]])){ # or length(class.categ). It is the same because already checked that lengths are the same
+tempo.data.frame <- data1[[i1]][data1[[i1]][, categ[[i1]]] == class.categ[i5], ]
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = paste0("ggplot2::", # no CR here te0("ggpl
+ifelse(geom[[i1]] == 'geom_stick', 'geom_segment', geom[[i1]]), # geom_segment because geom_stick converted to geom_segment for plotting
+"(data = tempo.data.frame, mapping = ggplot2::aes(x = ",
+x[[i1]],
+ifelse(geom[[i1]] == 'geom_stick', ", yend = ", ", y = "),
+y[[i1]],
+if(geom[[i1]] == 'geom_stick'){paste0(', xend = ', x[[i1]], ', y = ', y.lim[1])},
+", linetype = ",
+categ[[i1]],
+"), color = \"",
+color[[i1]][i5],
+"\", size = ",
+line.size[[i1]],
+ifelse(geom[[i1]] == 'geom_path', ', lineend = \"round\"', ''),
+ifelse(geom[[i1]] == 'geom_step', paste0(', direction = \"', geom.step.dir[[i1]], '\"'), ''),
+", alpha = ",
+alpha[[i1]],
+", show.legend = ",
+ifelse(i5 == 1, TRUE, FALSE),
+")"
+)))) # WARNING: a single color allowed for color argument outside aesthetic, hence the loop # legend.show option do not remove the legend, only the aesthetic of the legend (dot, line, etc.). Used here to avoid multiple layers of legend which corrupt transparency
+coord.names <- c(coord.names, paste0(geom[[i1]], ".", class.categ[i5]))
+}
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "linetype", name = if(is.null(legend.name)){NULL}else{legend.name[[i1]]}, values = rep(line.type[[i1]], length(color[[i1]])), breaks = class.categ)) # values are the values of linetype. 1 means solid. Regarding the alpha bug, I have tried different things without success: alpha in guide alone, in geom alone, in both, with different values, breaks reorder the classes according to class.categ in the legend
+}
+if(line.count == 2){
+fin.lg.disp[[5]] <- legend.disp[[point.count + line.count]]
+lg.order[[5]] <- point.count + line.count
+lg.color[[5]] <- color[[i1]] # if color == NULL -> NULL
+lg.line.size[[5]] <- line.size[[i1]]
+lg.line.type[[5]] <- line.type[[i1]]
+if(plot == TRUE & fin.lg.disp[[5]] == TRUE & ((length(dev.list()) > 0 & names(dev.cur()) == "windows") | (length(dev.list()) == 0 & Sys.info()["sysname"] == "Windows"))){ # if any Graph device already open and this device is "windows", or if no Graph device opened yet and we are on windows system -> prevention of alpha legend bug on windows using value 1
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") GRAPHIC DEVICE USED ON A WINDOWS SYSTEM ->\nTRANSPARENCY OF THE LINES (LINE LAYER NUMBER ", line.count, ") IS INACTIVATED IN THE LEGEND TO PREVENT A WINDOWS DEPENDENT BUG (SEE https://github.com/tidyverse/ggplot2/issues/2452)\nTO OVERCOME THIS ON WINDOWS, USE ANOTHER DEVICE (pdf() FOR INSTANCE)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+lg.alpha[[5]] <- 1 # to avoid a bug on windows: if alpha argument is different from 1 for lines (transparency), then lines are not correctly displayed in the legend when using the R GUI (bug https://github.com/tidyverse/ggplot2/issues/2452). No bug when using a pdf
+}else{
+lg.alpha[[5]] <- alpha[[i1]]
+}
+class.categ <- levels(factor(data1[[i1]][, categ[[i1]]]))
+for(i5 in 1:length(color[[i1]])){ # or length(class.categ). It is the same because already checked that lengths are the same
+tempo.data.frame <- data1[[i1]][data1[[i1]][, categ[[i1]]] == class.categ[i5], ]
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = paste0("ggplot2::", # no CR here te0("ggpl
+ifelse(geom[[i1]] == 'geom_stick', 'geom_segment', geom[[i1]]), # geom_segment because geom_stick converted to geom_segment for plotting
+"(data = tempo.data.frame, mapping = ggplot2::aes(x = ",
+x[[i1]],
+ifelse(geom[[i1]] == 'geom_stick', ", yend = ", ", y = "),
+y[[i1]],
+if(geom[[i1]] == 'geom_stick'){paste0(', xend = ', x[[i1]], ', y = ', y.lim[1])},
+", alpha = ",
+categ[[i1]],
+"), color = \"",
+color[[i1]][i5],
+"\", size = ",
+line.size[[i1]],
+", linetype = ",
+ifelse(is.numeric(line.type[[i1]]), "", "\""),
+line.type[[i1]],
+ifelse(is.numeric(line.type[[i1]]), "", "\""),
+ifelse(geom[[i1]] == 'geom_path', ', lineend = \"round\"', ''),
+ifelse(geom[[i1]] == 'geom_step', paste0(', direction = \"', geom.step.dir[[i1]], '\"'), ''),
+", show.legend = FALSE)"
+)))) # WARNING: a single color allowed for color argument outside aesthetic, hence the loop # legend.show option do not remove the legend, only the aesthetic of the legend (dot, line, etc.). Used here to avoid multiple layers of legend which corrupt transparency
+coord.names <- c(coord.names, paste0(geom[[i1]], ".", class.categ[i5]))
+}
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "alpha", name = if(is.null(legend.name)){NULL}else{legend.name[[i1]]}, values = rep(alpha[[i1]], length(color[[i1]])), breaks = class.categ)) # values are the values of linetype. 1 means solid. Regarding the alpha bug, I have tried different things without success: alpha in guide alone, in geom alone, in both, with different values, breaks reorder the classes according to class.categ in the legend
+}
+if(line.count == 3){
+fin.lg.disp[[6]] <- legend.disp[[point.count + line.count]]
+lg.order[[6]] <- point.count + line.count
+lg.color[[6]] <- color[[i1]] # if color == NULL -> NULL
+lg.line.size[[6]] <- line.size[[i1]]
+lg.line.type[[6]] <- line.type[[i1]]
+if(plot == TRUE & fin.lg.disp[[6]] == TRUE & ((length(dev.list()) > 0 & names(dev.cur()) == "windows") | (length(dev.list()) == 0 & Sys.info()["sysname"] == "Windows"))){ # if any Graph device already open and this device is "windows", or if no Graph device opened yet and we are on windows system -> prevention of alpha legend bug on windows using value 1
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") GRAPHIC DEVICE USED ON A WINDOWS SYSTEM ->\nTRANSPARENCY OF THE LINES (LINE LAYER NUMBER ", line.count, ") IS INACTIVATED IN THE LEGEND TO PREVENT A WINDOWS DEPENDENT BUG (SEE https://github.com/tidyverse/ggplot2/issues/2452)\nTO OVERCOME THIS ON WINDOWS, USE ANOTHER DEVICE (pdf() FOR INSTANCE)")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+lg.alpha[[6]] <- 1 # to avoid a bug on windows: if alpha argument is different from 1 for lines (transparency), then lines are not correctly displayed in the legend when using the R GUI (bug https://github.com/tidyverse/ggplot2/issues/2452). No bug when using a pdf
+}else{
+lg.alpha[[6]] <- alpha[[i1]]
+}
+class.categ <- levels(factor(data1[[i1]][, categ[[i1]]]))
+for(i5 in 1:length(color[[i1]])){ # or length(class.categ). It is the same because already checked that lengths are the same
+tempo.data.frame <- data1[[i1]][data1[[i1]][, categ[[i1]]] == class.categ[i5], ]
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = paste0("ggplot2::", # no CR here te0("ggpl
+ifelse(geom[[i1]] == 'geom_stick', 'geom_segment', geom[[i1]]), # geom_segment because geom_stick converted to geom_segment for plotting
+"(data = tempo.data.frame, mapping = ggplot2::aes(x = ",
+x[[i1]],
+ifelse(geom[[i1]] == 'geom_stick', ", yend = ", ", y = "),
+y[[i1]],
+if(geom[[i1]] == 'geom_stick'){paste0(', xend = ', x[[i1]], ', y = ', y.lim[1])},
+", size = ",
+categ[[i1]],
+"), color = \"",
+color[[i1]][i5],
+"\", linetype = ",
+ifelse(is.numeric(line.type[[i1]]), "", "\""),
+line.type[[i1]],
+ifelse(is.numeric(line.type[[i1]]), "", "\""),
+ifelse(geom[[i1]] == 'geom_path', ', lineend = \"round\"', ''),
+ifelse(geom[[i1]] == 'geom_step', paste0(', direction = \"', geom.step.dir[[i1]], '\"'), ''),
+", alpha = ",
+alpha[[i1]],
+", show.legend = FALSE)"
+)))) # WARNING: a single color allowed for color argument outside aesthetic, hence the loop # legend.show option do not remove the legend, only the aesthetic of the legend (dot, line, etc.). Used here to avoid multiple layers of legend which corrupt transparency
+coord.names <- c(coord.names, paste0(geom[[i1]], ".", class.categ[i5]))
+}
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "size", name = if(is.null(legend.name)){NULL}else{legend.name[[i1]]}, values = rep(line.size[[i1]], length(color[[i1]])), breaks = class.categ)) # values are the values of linetype. 1 means solid. Regarding the alpha bug, I have tried different things without success: alpha in guide alone, in geom alone, in both, breaks reorder the classes according to class.categ in the legend
+}
+}
+}
+# end loop part
-# add density
-# add legend order
-
-fun_gg_scatter <- function(
- data1,
- x,
- y,
- categ = NULL,
- categ.class.order = NULL,
- color = NULL,
- geom = "geom_point",
- geom.step.dir = "hv",
- alpha = 0.5,
- dot.size = 2,
- dot.shape = 21,
- dot.border.size = 0.5,
- dot.border.color = NULL,
- line.size = 0.5,
- line.type = "solid",
- x.lim = NULL,
- x.lab = NULL,
- x.log = "no",
- x.tick.nb = NULL,
- x.second.tick.nb = NULL,
- x.include.zero = FALSE,
- x.left.extra.margin = 0.05,
- x.right.extra.margin = 0.05,
- x.text.angle = 0,
- y.lim = NULL,
- y.lab = NULL,
- y.log = "no",
- y.tick.nb = NULL,
- y.second.tick.nb = NULL,
- y.include.zero = FALSE,
- y.top.extra.margin = 0.05,
- y.bottom.extra.margin = 0.05,
- y.text.angle = 0,
- raster = FALSE,
- raster.ratio = 1,
- raster.threshold = NULL,
- text.size = 12,
- title = "",
- title.text.size = 12,
- legend.show = TRUE,
- legend.width = 0.5,
- legend.name = NULL,
- article = TRUE,
- grid = FALSE,
- add = NULL,
- return = FALSE,
- return.ggplot = FALSE,
- return.gtable = TRUE,
- plot = TRUE,
- warn.print = FALSE,
- lib.path = NULL
-){
- # AIM
- # plot ggplot2 scatterplot with the possibility to overlay dots from up to 3 different data frames (-> three different legends) and lines from up to 3 different data frames (-> three different legends) -> up to 6 overlays totally
- # for ggplot2 specifications, see: https://ggplot2.tidyverse.org/articles/ggplot2-specs.html
- # WARNINGS
- # Rows containing NA in data1[, c(x, y, categ)] will be removed before processing, with a warning (see below)
- # Size arguments (dot.size, dot.border.size, line.size, text.size and title.text.size) are in mm. See Hadley comment in https://stackoverflow.com/questions/17311917/ggplot2-the-unit-of-size. See also http://sape.inf.usi.ch/quick-reference/ggplot2/size). Unit object are not accepted, but conversion can be used (e.g., grid::convertUnit(grid::unit(0.2, "inches"), "mm", valueOnly = TRUE))
- # ARGUMENTS
- # data1: a dataframe compatible with ggplot2, or a list of data frames. Order matters for the order of the legend and for the layer staking (starting from below to top)
- # x: single character string of the data1 column name for x-axis coordinates. If data1 is a list, then x must be a list of single character strings, of same size as data1, with compartment 1 related to compartment 1 of data1, etc. Write NULL for each "geom_hline" in geom argument
- # y: single character string of the data1 column name for y-axis coordinates. If data1 is a list, then y must be a list of single character strings, of same size as data1, with compartment 1 related to compartment 1 of data1, etc. Write NULL for each "geom_vline" in geom argument
- # categ: either NULL or a single character string or a list of single character strings, indicating the data1 column names to use for categories which creates legend display
- # If categ == NULL, no categories -> no legend displayed
- # If data1 is a data frame, categ must be a single character string of the data1 column name for categories
- # If data1 is a list, then categ must be a list of single character strings, of same size as data1, with compartment 1 related to compartment 1 of data1, etc. Some of the list compartments can be NULL (no legend display for these compartments), and other not
- # categ.class.order: either (1) NULL or (2) a vector of character strings or (3) a list of these vectors, setting the order of the classes of categ in the legend display
- # If categ.class.order is NULL, classes are represented according to the alphabetical order
- # If data1 is a data frame, categ.class.order must be a vector of character strings specifying the different classes in the categ column name of data1
- # If data1 is a list, then categ.class.order must be a list of vector of character strings, of same size as data1, with compartment 1 related to compartment 1 of data1, etc. Some of the list compartments can be NULL (alphabetical order for these compartments), and other not
- # color: either (1) NULL, or (2) a vector of character strings or integers, or (3) a list of vectors of character strings or integers
- # If color is NULL, default colors of ggplot2
- # If data1 is a data frame, color argument can be either:
- # (1) a single color string. All the dots of the corresponding data1 will have this color, whatever the categ value (NULL or not)
- # (2) if categ is non-null, a vector of string colors, one for each class of categ. Each color will be associated according to the categ.class.order argument if specified, or to the alphabetical order of categ classes otherwise
- # (3) if categ is non-null, a vector or factor of string colors, like if it was one of the column of data1 data frame. WARNING: a single color per class of categ and a single class of categ per color must be respected
- # Integers are also accepted instead of character strings, as long as above rules about length are respected. Integers will be processed by fun_gg_palette() using the max integer value among all the integers in color (see fun_gg_palette())
- # If data1 is a list, then color argument must be either:
- # (1) a list of character strings or integers, of same size as data1, with compartment 1 related to compartment 1 of data1, etc.
- # (2) a single character string or a single integer
- # With a list (first possibility), the rules described for when data1 is a data frame apply to each compartment of the list. Some of the compartments can be NULL. In that case, a different grey color will be used for each NULL compartment. With a single value (second possibility), the same color will be used for all the dots and lines, whatever the data1 list
- # geom: single character string of the kind of plot, or a list of single character strings
- # Either:
- # "geom_point" (scatterplot)
- # "geom_line" (coordinates plotted then line connection, from the lowest to highest x coordinates first and from the lowest to highest y coordinates thenafter)
- # "geom_path" (coordinates plotted then line connection respecting the row order in data1)
- # "geom_step" coordinates plotted then line connection respecting the row order in data1 but drawn in steps). See the geom.step.dir argument
- # "geom_hline" (horizontal line, no x value provided)
- # "geom_vline" (vertical line, no y value provided)
- # "geom_stick" (dots as vertical bars)
- # If data1 is a list, then geom must be either:
- # (1) a list of single character strings, of same size as data1, with compartment 1 related to compartment 1 of data1, etc.
- # (2) a single character string. In that case the same kind of plot will apply for the different compartments of the data1 list
- # WARNING concerning "geom_hline" or "geom_vline":
- # (1) x or y argument must be NULL, respectively
- # (2) x.lim or y.lim argument must NOT be NULL, respectively, if only these kind of lines are drawn (if other geom present, then x.lim = NULL and y.lim = NULL will generate x.lim and y.lim defined by these other geom, which is not possible with "geom_hline" or "geom_vline" alone)
- # (3) the function will draw n lines for n values in the x argument column name of the data1 data frame. If several colors required, the categ argument must be specified and the corresponding categ column name must exist in the data1 data frame with a different class name for each row
- # geom.step.dir: single character string indicating the direction when using "geom_step" of the geom argument, or a list of single character strings
- # Either:
- # "vh" (vertical then horizontal)
- # "hv" (horizontal then vertical)
- # "mid" (step half-way between adjacent x-values)
- # See https://ggplot2.tidyverse.org/reference/geom_path.html
- # If data1 is a list, then geom.step.dir must be either:
- # (1) a list of single character strings, of same size as data1, with compartment 1 related to compartment 1 of data1, etc. The value in compartments related to other geom values than "geom_step" will be ignored
- # (2) a single character string, which will be used for all the "geom_step" values of the geom argument, whatever the data1 list
- # alpha: single numeric value (from 0 to 1) of transparency. If data1 is a list, then alpha must be either (1) a list of single numeric values, of same size as data1, with compartment 1 related to compartment 1 of data1, etc., or (2) a single numeric value. In that case the same transparency will apply for the different compartments of the data1 list
- # dot.size: single numeric value of dot shape radius? in mm. If data1 is a list, then dot.size must be either (1) a list of single numeric values, of same size as data1, with compartment 1 related to compartment 1 of data1, etc., or (2) a single numeric value. With a list (former possibility), the value in compartments related to lines will be ignored. With a single value (latter possibility), the same dot.size will be used for all the dots, whatever the data1 list
- # dot.shape: value indicating the shape of the dots (see https://ggplot2.tidyverse.org/articles/ggplot2-specs.html) If data1 is a list, then dot.shape must be either (1) a list of single shape values, of same size as data1, with compartment 1 related to compartment 1 of data1, etc., or (2) a single shape value. With a list (former possibility), the value in compartments related to lines will be ignored. With a single value (latter possibility), the same dot.shape will be used for all the dots, whatever the data1 list
- # dot.border.size: single numeric value of border dot width in mm. Write zero for no dot border. If data1 is a list, then dot.border.size must be either (1) a list of single numeric values, of same size as data1, with compartment 1 related to compartment 1 of data1, etc., or (2) a single numeric value. With a list (former possibility), the value in compartments related to lines will be ignored. With a single value (latter possibility), the same dot.border.size will be used for all the dots, whatever the data1 list
- # dot.border.color: single character color string defining the color of the dot border (same border color for all the dots, whatever their categories). If dot.border.color == NULL, the border color will be the same as the dot color. A single integer is also accepted instead of a character string, that will be processed by fun_gg_palette()
- # line.size: single numeric value of line width in mm. If data1 is a list, then line.size must be either (1) a list of single numeric values, of same size as data1, with compartment 1 related to compartment 1 of data1, etc., or (2) a single numeric value. With a list (former possibility), the value in compartments related to dots will be ignored. With a single value (latter possibility), the same line.size will be used for all the lines, whatever the data1 list
- # line.type: value indicating the kind of lines (see https://ggplot2.tidyverse.org/articles/ggplot2-specs.html) If data1 is a list, then line.type must be either (1) a list of single line kind values, of same size as data1, with compartment 1 related to compartment 1 of data1, etc., or (2) a single line kind value. With a list (former possibility), the value in compartments related to dots will be ignored. With a single value (latter possibility), the same line.type will be used for all the lines, whatever the data1 list
- # x.lim: 2 numeric values setting the x-axis range. Order of the 2 values matters (for inverted axis). If NULL, the range of the x column name of data1 will be used
- # x.lab: a character string or expression for x-axis label. If NULL, will use the first value of x (x column name of the first data frame in data1). Warning message if the elements in x are different between data frames in data1
- # x.log: either "no", "log2" (values in the x column name of the data1 data frame will be log2 transformed and x-axis will be log2 scaled) or "log10" (values in the x column name of the data1 data frame will be log10 transformed and x-axis will be log10 scaled)
- # x.tick.nb: approximate number of desired values labeling the x-axis (i.e., main ticks, see the n argument of the the cute::fun_scale() function). If NULL and if x.log is "no", then the number of labeling values is set by ggplot2. If NULL and if x.log is "log2" or "log10", then the number of labeling values corresponds to all the exposant integers in the x.lim range (e.g., 10^1, 10^2 and 10^3, meaning 3 main ticks for x.lim = c(9, 1200)). WARNING: if non-NULL and if x.log is "log2" or "log10", labeling can be difficult to read (e.g., ..., 10^2, 10^2.5, 10^3, ...)
- # x.second.tick.nb: number of desired secondary ticks between main ticks. Ignored if x.log is other than "no" (log scale plotted). Use argument return = TRUE and see $plot$x.second.tick.values to have the values associated to secondary ticks. IF NULL, no secondary ticks
- # x.include.zero: logical. Does x.lim range include 0? Ignored if x.log is "log2" or "log10"
- # x.left.extra.margin: single proportion (between 0 and 1) indicating if extra margins must be added to x.lim. If different from 0, add the range of the axis multiplied by x.left.extra.margin (e.g., abs(x.lim[2] - x.lim[1]) * x.left.extra.margin) to the left of x-axis
- # x.right.extra.margin: idem as x.left.extra.margin but to the right of x-axis
- # x.text.angle: integer value of the text angle for the x-axis labeling values, using the same rules as in ggplot2. Use positive value for clockwise rotation: 0 for horizontal, 90 for vertical, 180 for upside down etc. Use negative values for counterclockwise rotation: 0 for horizontal, -90 for vertical, -180 for upside down etc.
- # y.lim: 2 numeric values setting the y-axis range. Order of the 2 values matters (for inverted axis). If NULL, the range of the y column name of data1 will be used
- # y.lab: a character string or expression for y-axis label. If NULL, will use the first value of y (y column name of the first data frame in data1). Warning message if the elements in y are different between data frames in data1
- # y.log: either "no", "log2" (values in the y column name of the data1 data frame will be log2 transformed and y-axis will be log2 scaled) or "log10" (values in the y column name of the data1 data frame will be log10 transformed and y-axis will be log10 scaled)
- # y.tick.nb: approximate number of desired values labeling the y-axis (i.e., main ticks, see the n argument of the the cute::fun_scale() function). If NULL and if y.log is "no", then the number of labeling values is set by ggplot2. If NULL and if y.log is "log2" or "log10", then the number of labeling values corresponds to all the exposant integers in the y.lim range (e.g., 10^1, 10^2 and 10^3, meaning 3 main ticks for y.lim = c(9, 1200)). WARNING: if non-NULL and if y.log is "log2" or "log10", labeling can be difficult to read (e.g., ..., 10^2, 10^2.5, 10^3, ...)
- # y.second.tick.nb: number of desired secondary ticks between main ticks. Ignored if y.log is other than "no" (log scale plotted). Use argument return = TRUE and see $plot$y.second.tick.values to have the values associated to secondary ticks. IF NULL, no secondary ticks
- # y.include.zero: logical. Does y.lim range include 0? Ignored if y.log is "log2" or "log10"
- # y.top.extra.margin: single proportion (between 0 and 1) indicating if extra margins must be added to y.lim. If different from 0, add the range of the axis multiplied by y.top.extra.margin (e.g., abs(y.lim[2] - y.lim[1]) * y.top.extra.margin) to the top of y-axis
- # y.bottom.extra.margin: idem as y.top.extra.margin but to the bottom of y-axis
- # y.text.angle: integer value of the text angle for the y-axis labeling values, using the same rules as in ggplot2. Use positive value for clockwise rotation: 0 for horizontal, 90 for vertical, 180 for upside down etc. Use negative values for counterclockwise rotation: 0 for horizontal, -90 for vertical, -180 for upside down etc.
- # raster: logical. Dots in raster mode? If FALSE, dots from each "geom_point" from geom argument are plotted in vectorial mode (bigger pdf and long to display if lots of dots). If TRUE, dots from each "geom_point" from geom argument are plotted in matricial mode (smaller pdf and easy display if lots of dots, but it takes time to generate the layer). If TRUE, the raster.ratio argument is used to avoid an ellipsoid representation of the dots. If TRUE, solve the transparency problem with some GUI. Overriden by the non-NULL raster.threshold argument
- # raster.ratio: single numeric value indicating the height / width ratio of the graphic device used (for instance provided by the $dim compartment in the output of the fun_open() function). The default value is 1 because by default R opens a square graphic device. But this argument has to be set when using other device dimensions. Ignored if raster == FALSE
- # raster.threshold: positive integer value indicating the limit of the dot number above which "geom_point" layers from the geom argument switch from vectorial mode to matricial mode (see the raster argument). If any layer is matricial, then the raster.ratio argument is used to avoid an ellipsoid representation of the dots. If non-NULL, it overrides the raster argument
- # text.size: numeric value of the font size of the (1) axis numbers and axis legends and (2) texts in the graphic legend (in mm)
- # title: character string of the graph title
- # title.text.size: numeric value of the title font size in mm
- # legend.show: logical. Show legend? Not considered if categ argument is NULL, because this already generate no legend, excepted if legend.width argument is non-NULL. In that specific case (categ is NULL, legend.show is TRUE and legend.width is non-NULL), an empty legend space is created. This can be useful when desiring graphs of exactly the same width, whatever they have legends or not
- # legend.width: single proportion (between 0 and 1) indicating the relative width of the legend sector (on the right of the plot) relative to the width of the plot. Value 1 means that the window device width is split in 2, half for the plot and half for the legend. Value 0 means no room for the legend, which will overlay the plot region. Write NULL to inactivate the legend sector. In such case, ggplot2 will manage the room required for the legend display, meaning that the width of the plotting region can vary between graphs, depending on the text in the legend
- # legend.name: character string of the legend title. If legend.name is NULL and categ argument is not NULL, then legend.name <- categ. If data1 is a list, then legend.name must be a list of character strings, of same size as data1, with compartment 1 related to compartment 1 of data1, etc. Some of the list compartments can be NULL, and other not
- # article: logical. If TRUE, use an article theme (article like). If FALSE, use a classic related ggplot theme. Use the add argument (e.g., add = "+ggplot2::theme_classic()" for the exact classic ggplot theme
- # grid: logical. Draw lines in the background to better read the box values? Not considered if article == FALSE (grid systematically present)
- # add: character string allowing to add more ggplot2 features (dots, lines, themes, facet, etc.). Ignored if NULL
- # WARNING: (1) the string must start with "+", (2) the string must finish with ")" and (3) each function must be preceded by "ggplot2::". Example: "+ ggplot2::coord_flip() + ggplot2::theme_bw()"
- # If the character string contains the "ggplot2::theme" string, then the article argument of fun_gg_scatter() (see above) is ignored with a warning
- # Handle the add argument with caution since added functions can create conflicts with the preexisting internal ggplot2 functions
- # WARNING: the call of objects inside the quotes of add can lead to an error if the name of these objects are some of the fun_gg_scatter() arguments. Indeed, the function will use the internal argument instead of the global environment object. Example article <- "a" in the working environment and add = '+ ggplot2::ggtitle(article)'. The risk here is to have TRUE as title. To solve this, use add = '+ ggplot2::ggtitle(get("article", envir = .GlobalEnv))'
- # return: logical. Return the graph parameters?
- # return.ggplot: logical. Return the ggplot object in the output list? Ignored if return argument is FALSE. WARNING: always assign the fun_gg_scatter() function (e.g., a <- fun_gg_scatter()) if return.ggplot argument is TRUE, otherwise, double plotting is performed. See $ggplot in the RETURN section below for more details
- # return.gtable: logical. Return the ggplot object as gtable of grobs in the output list? Ignored if plot argument is FALSE. Indeed, the graph must be plotted to get the grobs dispositions. See $gtable in the RETURN section below for more details
- # plot: logical. Plot the graphic? If FALSE and return argument is TRUE, graphical parameters and associated warnings are provided without plotting
- # warn.print: logical. Print warnings at the end of the execution? ? If FALSE, warning messages are never printed, but can still be recovered in the returned list. Some of the warning messages (those delivered by the internal ggplot2 functions) are not apparent when using the argument plot = FALSE
- # lib.path: character string indicating the absolute path of the required packages (see below). if NULL, the function will use the R library default folders
- # REQUIRED PACKAGES
- # ggplot2
- # gridExtra
- # lemon (in case of use in the add argument)
- # scales
- # if raster plots are drawn (see the raster and raster.threshold arguments):
- # Cairo
- # grid
- # REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
- # fun_gg_empty_graph()
- # fun_gg_palette()
- # fun_gg_point_rast()
- # fun_pack()
- # fun_check()
- # fun_round()
- # fun_scale()
- # fun_inter_ticks()
- # RETURN
- # a scatter plot if plot argument is TRUE
- # a list of the graph info if return argument is TRUE:
- # $data: the initial data with graphic information added. WARNING: if the x.log or y.log argument is not "no", x or y argument column of the data1 data frame are log2 or log10 converted in $data, respectively. Use 2^values or 10^$values to recover the initial values
- # $removed.row.nb: a list of the removed rows numbers in data frames (because of NA). NULL if no row removed
- # $removed.rows: a list of the removed rows in data frames (because of NA). NULL if no row removed
- # $plot: the graphic box and dot coordinates
- # $dots: dot coordinates
- # y.second.tick.positions: coordinates of secondary ticks (only if y.second.tick.nb argument is non-null or if y.log argument is different from "no")
- # y.second.tick.values: values of secondary ticks. NULL except if y.second.tick.nb argument is non-null or if y.log argument is different from "no")
- # $panel: the variable names used for the panels (NULL if no panels). WARNING: NA can be present according to ggplot2 upgrade to v3.3.0
- # $axes: the x-axis and y-axis info
- # $warn: the warning messages. Use cat() for proper display. NULL if no warning. WARNING: warning messages delivered by the internal ggplot2 functions are not apparent when using the argument plot = FALSE
- # $ggplot: ggplot object that can be used for reprint (use print($ggplot) or update (use $ggplot + ggplot2::...). NULL if return.ggplot argument is FALSE. Of note, a non-null $ggplot in the output list is sometimes annoying as the manipulation of this list prints the plot
- # $gtable: gtable object that can be used for reprint (use gridExtra::grid.arrange(...$ggplot) or with additionnal grobs (see the grob decomposition in the examples). NULL if return.ggplot argument is FALSE. Contrary to $ggplot, a non-NULL $gtable in the output list is not annoying as the manipulation of this list does not print the plot
- # EXAMPLES
- # DEBUGGING
- # set.seed(1) ; obs1 <- data.frame(km = rnorm(1000, 10, 3), time = rnorm(1000, 10, 3), group1 = rep(c("A1", "A2"), 500), stringsAsFactors = TRUE) ; obs2 <-data.frame(km = rnorm(1000, 15, 3), time = rnorm(1000, 15, 3), group2 = rep(c("G1", "G2"), 500), stringsAsFactors = TRUE) ; set.seed(NULL) ; obs1$km[2:3] <- NA ; data1 = list(L1 = obs1, L2 = obs2) ; x = list(L1 = "km", L2 = "km") ; y = list(L1 = "time", L2 = "time") ; categ = list(L1 = "group1", L2 = "group2") ; categ.class.order = NULL ; legend.name = NULL ; color = list(L1 = 4:5, L2 = 7:8) ; geom = list(L1 = "geom_point", L2 = "geom_point") ; geom.step.dir = "hv" ; alpha = list(L1 = 0.5, L2 = 0.5) ; dot.size = 3 ; dot.shape = 21 ; dot.border.size = 0.5 ; dot.border.color = NULL ; line.size = 0.5 ; line.type = "solid" ; x.lim = NULL ; x.lab = "KM" ; x.log = "no" ; x.tick.nb = 10 ; x.second.tick.nb = 1 ; x.left.extra.margin = 0 ; x.right.extra.margin = 0 ; y.lim = c(1, 25) ; y.lab = "TIME (s)" ; y.log = "no" ; y.tick.nb = 5 ; y.second.tick.nb = 2 ; y.top.extra.margin = 0 ; y.bottom.extra.margin = 0 ; x.include.zero = TRUE ; y.include.zero = TRUE ; x.text.angle = 0 ; y.text.angle = 0 ; text.size = 12 ; title = "" ; title.text.size = 8 ; legend.show = TRUE ; legend.width = 0.5 ; article = FALSE ; grid = FALSE ; raster = TRUE ; raster.ratio = 1 ; raster.threshold = NULL ; return = FALSE ; return.ggplot = FALSE ; return.gtable = FALSE ; plot = TRUE ; add = NULL ; warn.print = TRUE ; lib.path = NULL
- # data1 <- list(L1 = data.frame(a = 1:6, b = (1:6)^2, group = c("A", "A", "A", "B", "B", "B"), stringsAsFactors = TRUE), L2 = data.frame(a = (1:6)*2, b = ((1:6)^2)*2, group = c("A1", "A1", "A1", "B1", "B1", "B1"), stringsAsFactors = TRUE), L3 = data.frame(a = (1:6)*3, b = ((1:6)^2)*3, group3 = c("A4", "A5", "A6", "A7", "B4", "B5"), stringsAsFactors = TRUE)) ; data1$L1$a[3] <- NA ; data1$L1$group[5] <- NA ; data1$L3$group3[4] <- NA ; data1 ; x = list(L1 = names(data1$L1)[1], L2 = names(data1$L2)[1], L3 = NULL) ; y = list(L1 = names(data1$L1)[2], L2 = names(data1$L2)[2], L3 = "a") ; categ = list(L1 = "group", L2 = NULL, L3 = NULL) ; categ.class.order = NULL ; legend.name = NULL ; color = NULL ; geom = list(L1 = "geom_point", L2 = "geom_point", L3 = "geom_hline") ; geom.step.dir = "hv" ; alpha = list(L1 = 0.5, L2 = 0.5, L3 = 0.5) ; dot.size = 1 ; dot.shape = 21 ; dot.border.size = 0.5 ; dot.border.color = NULL ; line.size = 0.5 ; line.type = "solid" ; x.lim = c(14, 4) ; x.lab = NULL ; x.log = "log10" ; x.tick.nb = 10 ; x.second.tick.nb = 4 ; x.left.extra.margin = 0 ; x.right.extra.margin = 0 ; y.lim = c(60, 5) ; y.lab = NULL ; y.log = "log10" ; y.tick.nb = 10 ; y.second.tick.nb = 2 ; y.top.extra.margin = 0 ; y.bottom.extra.margin = 0 ; x.include.zero = TRUE ; y.include.zero = TRUE ; x.text.angle = 0 ; y.text.angle = 0 ; text.size = 12 ; title = "" ; title.text.size = 8 ; legend.show = TRUE ; legend.width = 0.5 ; article = TRUE ; grid = FALSE ; raster = FALSE ; raster.ratio = 1 ; raster.threshold = NULL ; return = TRUE ; return.ggplot = FALSE ; return.gtable = FALSE ; plot = TRUE ; add = NULL ; warn.print = TRUE ; lib.path = NULL
- # data1 <- data.frame(km = 2:7, time = (2:7)^2, group = c("A", "A", "A", "B", "B", "B"), stringsAsFactors = TRUE) ; data1 ; x = "km"; y = "time"; categ = "group" ; categ.class.order = NULL ; legend.name = NULL ; color = NULL ; geom = "geom_point" ; geom.step.dir = "hv" ; alpha = 0.1 ; dot.size = 3 ; dot.shape = 21 ; dot.border.size = 0.5 ; dot.border.color = NULL ; line.size = 0.5 ; line.type = "solid" ; x.lim = c(1,10) ; x.lab = NULL ; x.log = "log10" ; x.tick.nb = 10 ; x.second.tick.nb = 4 ; x.left.extra.margin = 0 ; x.right.extra.margin = 0 ; y.lim = NULL ; y.lab = expression(paste("TIME (", 10^-20, " s)")) ; y.log = "log10" ; y.tick.nb = 10 ; y.second.tick.nb = 2 ; y.top.extra.margin = 0 ; y.bottom.extra.margin = 0 ; x.include.zero = TRUE ; y.include.zero = TRUE ; x.text.angle = 0 ; y.text.angle = 0 ; text.size = 12 ; title = "" ; title.text.size = 8 ; legend.show = TRUE ; legend.width = 0.5 ; article = FALSE ; grid = FALSE ; raster = FALSE ; raster.ratio = 1 ; raster.threshold = NULL ; return = FALSE ; return.ggplot = FALSE ; return.gtable = FALSE ; plot = TRUE ; add = NULL ; warn.print = TRUE ; lib.path = NULL
- # function name
- function.name <- paste0(as.list(match.call(expand.dots=FALSE))[[1]], "()")
- arg.names <- names(formals(fun = sys.function(sys.parent(n = 2)))) # names of all the arguments
- arg.user.setting <- as.list(match.call(expand.dots=FALSE))[-1] # list of the argument settings (excluding default values not provided by the user)
- # end function name
- # required function checking
- req.function <- c(
- "fun_check",
- "fun_gg_just",
- "fun_gg_empty_graph",
- "fun_gg_palette",
- "fun_gg_point_rast",
- "fun_round",
- "fun_pack",
- "fun_scale",
- "fun_inter_ticks"
- )
- for(i1 in req.function){
- if(length(find(i1, mode = "function")) == 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED ", i1, "() FUNCTION IS MISSING IN THE R ENVIRONMENT")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
- }
- }
- # end required function checking
- # reserved words to avoid bugs (used in this function)
- reserved.words <- c("fake_x", "fake_y", "fake_categ")
- # end reserved words to avoid bugs (used in this function)
- # arg with no default values
- if(any(missing(data1) | missing(x) | missing(y))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ARGUMENTS angle AND pos HAVE NO DEFAULT VALUE AND REQUIRE ONE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end arg with no default values
- # primary argument checking
- arg.check <- NULL #
- text.check <- NULL #
- checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
- ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
- tempo1 <- fun_check(data = data1, class = "data.frame", na.contain = TRUE, fun.name = function.name)
- tempo2 <- fun_check(data = data1, class = "list", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": data1 ARGUMENT MUST BE A DATA FRAME OR A LIST OF DATA FRAMES")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- if( ! is.null(x)){
- tempo1 <- fun_check(data = x, class = "vector", mode = "character", na.contain = TRUE, length = 1, fun.name = function.name)
- tempo2 <- fun_check(data = x, class = "list", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": x ARGUMENT MUST BE A SINGLE CHARACTER STRING OR A LIST OF CHARACTER STRINGS")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- if( ! is.null(y)){
- tempo1 <- fun_check(data = y, class = "vector", mode = "character", na.contain = TRUE, length = 1, fun.name = function.name)
- tempo2 <- fun_check(data = y, class = "list", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": y ARGUMENT MUST BE A SINGLE CHARACTER STRING OR A LIST OF CHARACTER STRINGS")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- if( ! is.null(categ)){
- tempo1 <- fun_check(data = categ, class = "vector", mode = "character", length = 1, fun.name = function.name)
- tempo2 <- fun_check(data = categ, class = "list", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": categ ARGUMENT MUST BE A SINGLE CHARACTER STRING OR A LIST OF CHARACTER STRINGS")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- if( ! is.null(categ.class.order)){
- if(is.null(categ)){
- tempo.cat <- paste0("ERROR IN ", function.name, ": categ.class.order ARGUMENT IS NOT NULL, BUT categ IS")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- tempo1 <- fun_check(data = categ.class.order, class = "vector", mode = "character", fun.name = function.name)
- tempo2 <- fun_check(data = categ.class.order, class = "list", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": categ.class.order ARGUMENT MUST BE A VECTOR OF CHARACTER STRINGS OR A LIST OF VECTOR OF CHARACTER STRINGS")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- if( ! is.null(legend.name)){
- tempo1 <- fun_check(data = legend.name, class = "vector", mode = "character", na.contain = TRUE, length = 1, fun.name = function.name)
- tempo2 <- fun_check(data = legend.name, class = "list", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": legend.name ARGUMENT MUST BE A SINGLE CHARACTER STRING OR A LIST OF CHARACTER STRINGS")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- if( ! is.null(color)){
- tempo1 <- fun_check(data = color, class = "vector", mode = "character", na.contain = TRUE, fun.name = function.name)
- tempo2 <- fun_check(data = color, class = "factor", na.contain = TRUE, fun.name = function.name)
- tempo3 <- fun_check(data = color, class = "integer", double.as.integer.allowed = TRUE, na.contain = TRUE, fun.name = function.name)
- tempo4 <- fun_check(data = color, class = "list", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE & tempo3$problem == TRUE & tempo4$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": color ARGUMENT MUST BE A VECTOR (OF CHARACTER STRINGS OR INTEGERS) OR A FACTOR OR A LIST OF THESE POSSIBILITIES")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- tempo1 <- fun_check(data = geom, class = "vector", mode = "character", na.contain = FALSE, length = 1, fun.name = function.name)
- tempo2 <- fun_check(data = geom, class = "list", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": geom ARGUMENT MUST BE A SINGLE CHARACTER STRING OR A LIST OF CHARACTER STRINGS")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- tempo1 <- fun_check(data = geom.step.dir, options = c("vh", "hv", "mid"), na.contain = FALSE, length = 1, fun.name = function.name)
- tempo2 <- fun_check(data = geom.step.dir, class = "list", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": geom.step.dir ARGUMENT MUST BE A SINGLE CHARACTER STRING (\"vh\" OR \"hv\" OR \"mid\") OR A LIST OF THESE CHARACTER STRINGS")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- tempo1 <- fun_check(data = alpha, prop = TRUE, length = 1, fun.name = function.name)
- tempo2 <- fun_check(data = alpha, class = "list", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": alpha ARGUMENT MUST BE A SINGLE NUMERIC VALUE BETWEEN 0 AND 1 OR A LIST OF SUCH VALUES")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- tempo1 <- fun_check(data = dot.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name)
- tempo2 <- fun_check(data = dot.size, class = "list", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": dot.size ARGUMENT MUST BE A SINGLE NUMERIC VALUE OR A LIST OF SINGLE NUMERIC VALUES")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- tempo1 <- fun_check(data = dot.shape, class = "vector", length = 1, fun.name = function.name)
- tempo2 <- fun_check(data = dot.shape, class = "list", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": dot.shape ARGUMENT MUST BE A SINGLE SHAPE VALUE OR A LIST OF SINGLE SHAPE VALUES (SEE https://ggplot2.tidyverse.org/articles/ggplot2-specs.html)")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- tempo1 <- fun_check(data = dot.border.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name)
- tempo2 <- fun_check(data = dot.border.size, class = "list", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": dot.border.size ARGUMENT MUST BE A SINGLE NUMERIC VALUE OR A LIST OF SINGLE NUMERIC VALUES")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- if( ! is.null(dot.border.color)){
- tempo1 <- fun_check(data = dot.border.color, class = "vector", mode = "character", length = 1, fun.name = function.name)
- tempo2 <- fun_check(data = dot.border.color, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- # integer colors -> gg_palette
- tempo.cat <- paste0("ERROR IN ", function.name, ": dot.border.color MUST BE A SINGLE CHARACTER STRING OF COLOR OR A SINGLE INTEGER VALUE")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- tempo1 <- fun_check(data = line.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name)
- tempo2 <- fun_check(data = line.size, class = "list", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": line.size ARGUMENT MUST BE A SINGLE NUMERIC VALUE OR A LIST OF SINGLE NUMERIC VALUES")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- tempo1 <- fun_check(data = line.type, class = "vector", typeof = "integer", double.as.integer.allowed = FALSE, length = 1, fun.name = function.name)
- tempo2 <- fun_check(data = line.type, class = "vector", mode = "character", length = 1, fun.name = function.name)
- tempo3 <- fun_check(data = line.type, class = "list", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE & tempo3$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": line.type ARGUMENT MUST BE A SINGLE LINE KIND VALUE OR A LIST OF SINGLE LINE KIND VALUES (SEE https://ggplot2.tidyverse.org/articles/ggplot2-specs.html)")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- if( ! is.null(x.lim)){
- tempo <- fun_check(data = x.lim, class = "vector", mode = "numeric", length = 2, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & any(x.lim %in% c(Inf, -Inf))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": x.lim ARGUMENT CANNOT CONTAIN -Inf OR Inf VALUES")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- if( ! is.null(x.lab)){
- if(all(class(x.lab) %in% "expression")){ # to deal with math symbols
- tempo <- fun_check(data = x.lab, class = "expression", length = 1, fun.name = function.name) ; eval(ee)
- }else{
- tempo <- fun_check(data = x.lab, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- }
- }
- tempo <- fun_check(data = x.log, options = c("no", "log2", "log10"), length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(x.tick.nb)){
- tempo <- fun_check(data = x.tick.nb, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & x.tick.nb < 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": x.tick.nb ARGUMENT MUST BE A NON-NULL POSITIVE INTEGER")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- if( ! is.null(x.second.tick.nb)){
- tempo <- fun_check(data = x.second.tick.nb, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & x.second.tick.nb <= 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": x.second.tick.nb ARGUMENT MUST BE A NON-NULL POSITIVE INTEGER")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- tempo <- fun_check(data = x.include.zero, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = x.left.extra.margin, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = x.right.extra.margin, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = x.text.angle, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, neg.values = TRUE, fun.name = function.name) ; eval(ee)
- if( ! is.null(y.lim)){
- tempo <- fun_check(data = y.lim, class = "vector", mode = "numeric", length = 2, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & any(y.lim %in% c(Inf, -Inf))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": y.lim ARGUMENT CANNOT CONTAIN -Inf OR Inf VALUES")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- if( ! is.null(y.lab)){
- if(all(class(y.lab) %in% "expression")){ # to deal with math symbols
- tempo <- fun_check(data = y.lab, class = "expression", length = 1, fun.name = function.name) ; eval(ee)
- }else{
- tempo <- fun_check(data = y.lab, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- }
- }
- tempo <- fun_check(data = y.log, options = c("no", "log2", "log10"), length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(y.tick.nb)){
- tempo <- fun_check(data = y.tick.nb, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & y.tick.nb < 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": y.tick.nb ARGUMENT MUST BE A NON-NULL POSITIVE INTEGER")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- if( ! is.null(y.second.tick.nb)){
- tempo <- fun_check(data = y.second.tick.nb, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE & y.second.tick.nb <= 0){
- tempo.cat <- paste0("ERROR IN ", function.name, ": y.second.tick.nb ARGUMENT MUST BE A NON-NULL POSITIVE INTEGER")
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- tempo <- fun_check(data = y.include.zero, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = y.top.extra.margin, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = y.bottom.extra.margin, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = y.text.angle, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, neg.values = TRUE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = raster, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = raster.ratio, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- if( ! is.null(raster.threshold)){
- tempo <- fun_check(data = raster.threshold, class = "vector", typeof = "integer", neg.values = FALSE, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = text.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = title, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = title.text.size, class = "vector", mode = "numeric", length = 1, neg.values = FALSE, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = legend.show, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(legend.width)){
- tempo <- fun_check(data = legend.width, prop = TRUE, length = 1, fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = article, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = grid, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(add)){
- tempo <- fun_check(data = add, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
- }
- tempo <- fun_check(data = return, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = return.ggplot, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = return.gtable, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = plot, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- tempo <- fun_check(data = warn.print, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
- if( ! is.null(lib.path)){
- tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
- if(tempo$problem == FALSE){
- if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
- tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
- text.check <- c(text.check, tempo.cat)
- arg.check <- c(arg.check, TRUE)
- }
- }
- }
- if(any(arg.check) == TRUE){
- stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
- }
- # source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
- # end primary argument checking
-
-
- # second round of checking and data preparation
- # dealing with NA arguments
- tempo.arg <- names(arg.user.setting) # values provided by the user
- tempo.log <- sapply(lapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = is.na), FUN = any) & lapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = length) == 1 # no argument provided by the user can be just NA
- if(any(tempo.log) == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ":\n", ifelse(sum(tempo.log, na.rm = TRUE) > 1, "THESE ARGUMENTS\n", "THIS ARGUMENT\n"), paste0(tempo.arg[tempo.log], collapse = "\n"),"\nCANNOT JUST BE NA")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end dealing with NA arguments
- # dealing with NULL arguments
- tempo.arg <-c(
- "data1",
- # "x", # inactivated because of hline or vline
- # "y", # inactivated because of hline or vline
- "geom",
- "geom.step.dir",
- "alpha",
- "dot.size",
- "dot.shape",
- "dot.border.size",
- "line.size",
- "line.type",
- "x.log",
- "x.include.zero",
- "x.left.extra.margin",
- "x.right.extra.margin",
- "x.text.angle",
- "y.log",
- "y.include.zero",
- "y.top.extra.margin",
- "y.bottom.extra.margin",
- "y.text.angle",
- "raster",
- "raster.ratio",
- "text.size",
- "title",
- "title.text.size",
- "legend.show",
- # "legend.width", # inactivated because can be null
- "article",
- "grid",
- "return",
- "return.ggplot",
- "return.gtable",
- "plot",
- "warn.print"
- )
- tempo.log <- sapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = is.null)
- if(any(tempo.log) == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ":\n", ifelse(sum(tempo.log, na.rm = TRUE) > 1, "THESE ARGUMENTS\n", "THIS ARGUMENT\n"), paste0(tempo.arg[tempo.log], collapse = "\n"),"\nCANNOT BE NULL")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
- }
- # end dealing with NULL arguments
- # check list lengths (and names of data1 compartments if present)
- ini.warning.length <- options()$warning.length
- warn <- NULL
- warn.count <- 0
- list.color <- NULL
- list.geom <- NULL
- list.geom.step.dir <- NULL
- list.alpha <- NULL
- list.dot.size <- NULL
- list.dot.shape <- NULL
- list.dot.border.size <- NULL
- list.dot.border.color <- NULL
- list.line.size <- NULL
- list.line.type <- NULL
- if(all(class(data1) == "list")){
- if(length(data1) > 6){
- tempo.cat <- paste0("ERROR IN ", function.name, ": data1 ARGUMENT MUST BE A LIST OF 6 DATA FRAMES MAXIMUM (6 OVERLAYS MAX)")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- if(is.null(names(data1))){
- names(data1) <- paste0("L", 1:length(data1))
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NULL NAME COMPARTMENT OF data1 LIST -> NAMES RESPECTIVELY ATTRIBUTED TO EACH COMPARTMENT:\n", paste(names(data1), collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- if( ! is.null(x)){
- if( ! (all(class(x) == "list") & length(data1) == length(x))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": x ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }else{
- x <- vector("list", length(data1))
- }
- if( ! is.null(y)){
- if( ! (all(class(y) == "list") & length(data1) == length(y))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": y ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }else{
- y <- vector("list", length(data1))
- }
- if( ! is.null(categ)){
- if( ! (all(class(categ) == "list") & length(data1) == length(categ))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": categ ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- if( ! is.null(categ.class.order)){
- if( ! (all(class(categ.class.order) == "list") & length(data1) == length(categ.class.order))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": categ.class.order ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- if( ! is.null(color)){
- if( ! ((all(class(color) == "list") & length(data1) == length(color)) | ((all(mode(color) == "character") | all(mode(color) == "numeric")) & length(color) == 1))){ # list of same length as data1 or single value
- tempo.cat <- paste0("ERROR IN ", function.name, ": color ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE CHARACTER STRING OR INTEGER")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if((all(mode(color) == "character") | all(mode(color) == "numeric")) & length(color) == 1){ # convert the single value into a list of single value
- list.color <- vector(mode = "list", length = length(data1))
- list.color[] <- color
- }
- }
- if( ! ((all(class(geom) == "list") & length(data1) == length(geom)) | (all(mode(geom) == "character") & length(geom) == 1))){ # list of same length as data1 or single value
- tempo.cat <- paste0("ERROR IN ", function.name, ": geom ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE CHARACTER VALUE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if(all(mode(geom) == "character") & length(geom) == 1){ # convert the single value into a list of single value
- list.geom <- vector(mode = "list", length = length(data1))
- list.geom[] <- geom
- }
- if( ! ((all(class(geom.step.dir) == "list") & length(data1) == length(geom.step.dir)) | (all(mode(geom.step.dir) == "character") & length(geom.step.dir) == 1))){ # list of same length as data1 or single value
- tempo.cat <- paste0("ERROR IN ", function.name, ": geom.step.dir ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE CHARACTER VALUE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if(all(mode(geom.step.dir) == "character") & length(geom.step.dir) == 1){ # convert the single value into a list of single value
- list.geom.step.dir <- vector(mode = "list", length = length(data1))
- list.geom.step.dir[] <- geom.step.dir
- }
- if( ! ((all(class(alpha) == "list") & length(data1) == length(alpha)) | (all(mode(alpha) == "numeric") & length(alpha) == 1))){ # list of same length as data1 or single value
- tempo.cat <- paste0("ERROR IN ", function.name, ": alpha ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE NUMERIC VALUE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if(all(mode(alpha) == "numeric") & length(alpha) == 1){ # convert the single value into a list of single value
- list.alpha <- vector(mode = "list", length = length(data1))
- list.alpha[] <- alpha
- }
- if( ! ((all(class(dot.size) == "list") & length(data1) == length(dot.size)) | (all(mode(dot.size) == "numeric") & length(dot.size) == 1))){ # list of same length as data1 or single value
- tempo.cat <- paste0("ERROR IN ", function.name, ": dot.size ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE NUMERIC VALUE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if(all(mode(dot.size) == "numeric") & length(dot.size) == 1){ # convert the single value into a list of single value
- list.dot.size <- vector(mode = "list", length = length(data1))
- list.dot.size[] <- dot.size
- }
- if( ! ((all(class(dot.shape) == "list") & length(data1) == length(dot.shape)) | (all(mode(dot.shape) != "list") & length(dot.shape) == 1))){ # list of same length as data1 or single value
- tempo.cat <- paste0("ERROR IN ", function.name, ": dot.shape ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE SHAPE VALUE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if(all(mode(dot.shape) != "list") & length(dot.shape) == 1){ # convert the single value into a list of single value
- list.dot.shape <- vector(mode = "list", length = length(data1))
- list.dot.shape[] <- dot.shape
- }
- if( ! ((all(class(dot.border.size) == "list") & length(data1) == length(dot.border.size)) | (all(mode(dot.border.size) == "numeric") & length(dot.border.size) == 1))){ # list of same length as data1 or single value
- tempo.cat <- paste0("ERROR IN ", function.name, ": dot.border.size ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE NUMERIC VALUE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if(all(mode(dot.border.size) == "numeric") & length(dot.border.size) == 1){ # convert the single value into a list of single value
- list.dot.border.size <- vector(mode = "list", length = length(data1))
- list.dot.border.size[] <- dot.border.size
- }
- if( ! is.null(dot.border.color)){
- if( ! ((all(class(dot.border.color) == "list") & length(data1) == length(dot.border.color)) | ((all(mode(dot.border.color) == "character") | all(mode(dot.border.color) == "numeric")) & length(dot.border.color) == 1))){ # list of same length as data1 or single value
- tempo.cat <- paste0("ERROR IN ", function.name, ": dot.border.color ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE CHARACTER STRING OR INTEGER")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if((all(mode(dot.border.color) == "character") | all(mode(dot.border.color) == "numeric")) & length(dot.border.color) == 1){ # convert the single value into a list of single value
- list.dot.border.color <- vector(mode = "list", length = length(data1))
- list.dot.border.color[] <- dot.border.color
- }
- }
- if( ! ((all(class(line.size) == "list") & length(data1) == length(line.size)) | (all(mode(line.size) == "numeric") & length(line.size) == 1))){ # list of same length as data1 or single value
- tempo.cat <- paste0("ERROR IN ", function.name, ": line.size ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE NUMERIC VALUE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if(all(mode(line.size) == "numeric") & length(line.size) == 1){ # convert the single value into a list of single value
- list.line.size <- vector(mode = "list", length = length(data1))
- list.line.size[] <- line.size
- }
- if( ! ((all(class(line.type) == "list") & length(data1) == length(line.type)) | (all(mode(line.type) != "list") & length(line.type) == 1))){ # list of same length as data1 or single value
- tempo.cat <- paste0("ERROR IN ", function.name, ": line.type ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST, OR A SINGLE LINE KIND VALUE")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if(all(mode(line.type) != "list") & length(line.type) == 1){ # convert the single value into a list of single value
- list.line.type <- vector(mode = "list", length = length(data1))
- list.line.type[] <- line.type
- }
- if( ! is.null(legend.name)){
- if( ! (all(class(legend.name) == "list") & length(data1) == length(legend.name))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": legend.name ARGUMENT MUST BE A LIST OF SAME LENGTH AS data1 IF data1 IS A LIST")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- }
- # end check list lengths (and names of data1 compartments if present)
- # conversion into lists
- if(all(is.data.frame(data1))){
- data1 <- list(L1 = data1)
- if(all(class(x) == "list")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": x ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- x <- list(L1 = x)
- }
- if(all(class(y) == "list")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": y ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- y <- list(L1 = y)
- }
- if( ! is.null(categ)){
- if(all(class(categ) == "list")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": categ ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- categ <- list(L1 = categ)
- }
- }
- if( ! is.null(categ.class.order)){
- if(all(class(categ.class.order) == "list")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": categ.class.order ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- categ.class.order <- list(L1 = categ.class.order)
- }
- }
- if( ! is.null(color)){
- if(all(class(color) == "list")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": color ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- color <- list(L1 = color)
- }
- }
- if(all(class(geom) == "list")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": geom ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- geom <- list(L1 = geom)
- }
- if(all(class(geom.step.dir) == "list")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": geom.step.dir ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- geom.step.dir <- list(L1 = geom.step.dir)
- }
- if(all(class(alpha) == "list")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": alpha ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- alpha <- list(L1 = alpha)
- }
- if(all(class(dot.size) == "list")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": dot.size ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- dot.size <- list(L1 = dot.size)
- }
- if(all(class(dot.shape) == "list")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": dot.shape ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- dot.shape <- list(L1 = dot.shape)
- }
- if(all(class(dot.border.size) == "list")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": dot.border.size ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- dot.border.size <- list(L1 = dot.border.size)
- }
- if( ! is.null(dot.border.color)){
- if(all(class(dot.border.color) == "list")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": dot.border.color ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- dot.border.color <- list(L1 = dot.border.color)
- }
- }
- if(all(class(line.size) == "list")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": line.size ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- line.size <- list(L1 = line.size)
- }
- if(all(class(line.type) == "list")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": line.type ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- line.type <- list(L1 = line.type)
- }
- if( ! is.null(legend.name)){
- if(all(class(legend.name) == "list")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": legend.name ARGUMENT CANNOT BE A LIST IF data1 IS A DATA FRAME")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- legend.name <- list(L1 = legend.name)
- }
- }
- }else if( ! all(sapply(data1, FUN = "class") == "data.frame")){ # if not a data frame, data1 can only be a list, as tested above
- tempo.cat <- paste0("ERROR IN ", function.name, ": data1 ARGUMENT MUST BE A DATA FRAME OR A LIST OF DATA FRAMES")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- # single value converted into list now reattributed to the argument name
- if( ! is.null(color)){
- if( ! is.null(list.color)){
- color <- list.color
- }
- }
- if( ! is.null(list.geom)){
- geom <- list.geom
- }
- if( ! is.null(list.geom.step.dir)){
- geom.step.dir <- list.geom.step.dir
- }
- if( ! is.null(list.alpha)){
- alpha <- list.alpha
- }
- if( ! is.null(list.dot.size)){
- dot.size <- list.dot.size
- }
- if( ! is.null(list.dot.shape)){
- dot.shape <- list.dot.shape
- }
- if( ! is.null(list.dot.border.size)){
- dot.border.size <- list.dot.border.size
- }
- if( ! is.null(dot.border.color)){
- if( ! is.null(list.dot.border.color)){
- dot.border.color <- list.dot.border.color
- }
- }
- if( ! is.null(list.line.size)){
- line.size <- list.line.size
- }
- if( ! is.null(list.line.type)){
- line.type <- list.line.type
- }
- # end single value converted into list now reattributed to the argument name
- # data, x, y, geom, alpha, dot.size, shape, dot.border.size, line.size, line.type, legend.name are list now
- # if non-null, categ, categ.class.order, legend.name, color, dot.border.color are list now
- # end conversion into lists
- # verif of add
- if( ! is.null(add)){
- if( ! grepl(pattern = "^\\s*\\+", add)){ # check that the add string start by +
- tempo.cat <- paste0("ERROR IN ", function.name, ": add ARGUMENT MUST START WITH \"+\": ", paste(unique(add), collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
-
- }else if( ! grepl(pattern = "(ggplot2|lemon)\\s*::", add)){ #
- tempo.cat <- paste0("ERROR IN ", function.name, ": FOR EASIER FUNCTION DETECTION, add ARGUMENT MUST CONTAIN \"ggplot2::\" OR \"lemon::\" IN FRONT OF EACH GGPLOT2 FUNCTION: ", paste(unique(add), collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if( ! grepl(pattern = ")\\s*$", add)){ # check that the add string finished by )
- tempo.cat <- paste0("ERROR IN ", function.name, ": add ARGUMENT MUST FINISH BY \")\": ", paste(unique(add), collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- # end verif of add
- # management of add containing facet
- facet.categ <- NULL
- if( ! is.null(add)){
- facet.check <- TRUE
- tempo <- unlist(strsplit(x = add, split = "\\s*\\+\\s*(ggplot2|lemon)\\s*::\\s*")) #
- tempo <- sub(x = tempo, pattern = "^facet_wrap", replacement = "ggplot2::facet_wrap")
- tempo <- sub(x = tempo, pattern = "^facet_grid", replacement = "ggplot2::facet_grid")
- tempo <- sub(x = tempo, pattern = "^facet_rep", replacement = "lemon::facet_rep")
- if(length(data1) > 1 & (any(grepl(x = tempo, pattern = "ggplot2::facet_wrap|lemon::facet_rep_wrap")) | grepl(x = add, pattern = "ggplot2::facet_grid|lemon::facet_rep_grid"))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nfacet PANELS CANNOT BE USED IF MORE THAN ONE DATA FRAME IN THE data1 ARGUMENT\nPLEASE REWRITE THE add STRING AND RERUN")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- if(any(grepl(x = tempo, pattern = "ggplot2::facet_wrap|lemon::facet_rep_wrap"))){
- tempo1 <- suppressWarnings(eval(parse(text = tempo[grepl(x = tempo, pattern = "ggplot2::facet_wrap|lemon::facet_rep_wrap")])))
- facet.categ <- list(names(tempo1$params$facets)) # list of length 1
- tempo.text <- "facet_wrap OR facet_rep_wrap"
- facet.check <- FALSE
- }else if(grepl(x = add, pattern = "ggplot2::facet_grid|lemon::facet_rep_grid")){
- tempo1 <- suppressWarnings(eval(parse(text = tempo[grepl(x = tempo, pattern = "ggplot2::facet_grid|lemon::facet_rep_grid")])))
- facet.categ <- list(c(names(tempo1$params$rows), names(tempo1$params$cols))) # list of length 1
- tempo.text <- "facet_grid OR facet_rep_grid"
- facet.check <- FALSE
- }
- if(facet.check == FALSE & ! all(facet.categ %in% names(data1[[1]]))){ # WARNING: all(facet.categ %in% names(data1)) is TRUE when facet.categ is NULL
- tempo.cat <- paste0("ERROR IN ", function.name, "\nDETECTION OF \"", tempo.text, "\" STRING IN THE add ARGUMENT BUT PROBLEM OF VARIABLE DETECTION (COLUMN NAMES OF data1)\nTHE DETECTED VARIABLES ARE:\n", paste(facet.categ, collapse = " "), "\nTHE data1 COLUMN NAMES ARE:\n", paste(names(data1[[1]]), collapse = " "), "\nPLEASE REWRITE THE add STRING AND RERUN")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- }
- # if facet.categ is not NULL, it is a list of length 1 now
- # end management of add containing facet
- # legend name filling
- if(is.null(legend.name) & ! is.null(categ)){
- legend.name <- categ
- }else if(is.null(legend.name) & is.null(categ)){
- legend.name <- vector("list", length(data1)) # null list
- }
- # legend.name not NULL anymore (list)
- # end legend name filling
- # ini categ for legend display
- fin.lg.disp <- vector("list", 6) # will be used at the end to display or not legends
- fin.lg.disp[] <- FALSE
- legend.disp <- vector("list", length(data1))
- if(is.null(categ) | legend.show == FALSE){
- legend.disp[] <- FALSE
- }else{
- for(i2 in 1:length(data1)){
- if(is.null(categ[[i2]])){
- legend.disp[[i2]] <- FALSE
- }else{
- legend.disp[[i2]] <- TRUE
- }
- }
- }
- # end ini categ for legend display
- # integer colors into gg_palette
- tempo.check.color <- NULL
- for(i1 in 1:length(data1)){
- if(any(is.na(color[[i1]]))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), ": color ARGUMENT CANNOT CONTAIN NA")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- tempo.check.color <- c(tempo.check.color, fun_check(data = color[[i1]], data.name = ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), class = "integer", double.as.integer.allowed = TRUE, na.contain = TRUE, fun.name = function.name)$problem)
- }
- tempo.check.color <- ! tempo.check.color # invert TRUE and FALSE because if integer, then problem = FALSE
- if(any(tempo.check.color == TRUE)){ # convert integers into colors
- tempo.integer <- unlist(color[tempo.check.color])
- tempo.color <- fun_gg_palette(max(tempo.integer, na.rm = TRUE))
- for(i1 in 1:length(data1)){
- if(tempo.check.color[i1] == TRUE){
- color[[i1]] <-tempo.color[color[[i1]]]
- }
- }
- }
- # end integer colors into gg_palette
- # loop (checking inside list compartment)
- compart.null.color <- 0 # will be used to attribute a color when color is non-null but a compartment of color is NULL
- data1.ini <- data1 # to report NA removal
- removed.row.nb <- vector("list", length = length(data1)) # to report NA removal. Contains NULL
- removed.rows <- vector("list", length = length(data1)) # to report NA removal. Contains NULL
- for(i1 in 1:length(data1)){
- tempo <- fun_check(data = data1[[i1]], data.name = ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), class = "data.frame", na.contain = TRUE, fun.name = function.name)
- if(tempo$problem == TRUE){
- stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- # reserved word checking
- if(any(names(data1[[i1]]) %in% reserved.words)){ # I do not use fun_name_change() because cannot control y before creating "fake_y". But ok because reserved are not that common
- tempo.cat <- paste0("ERROR IN ", function.name, ": COLUMN NAMES OF ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " ARGUMENT CANNOT BE ONE OF THESE WORDS\n", paste(reserved.words, collapse = " "), "\nTHESE ARE RESERVED FOR THE ", function.name, " FUNCTION")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- if( ! (is.null(add))){
- if(any(sapply(X = reserved.words, FUN = grepl, x = add))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nDETECTION OF COLUMN NAMES OF data1 IN THE add ARGUMENT STRING, THAT CORRESPOND TO RESERVED STRINGS FOR ", function.name, "\nFOLLOWING COLUMN NAMES HAVE TO BE CHANGED:\n", paste(arg.names[sapply(X = reserved.words, FUN = grepl, x = add)], collapse = "\n"), "\nFOR INFORMATION, THE RESERVED WORDS ARE:\n", paste(reserved.words, collapse = "\n"))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if(any(sapply(X = arg.names, FUN = grepl, x = add))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NAMES OF ", function.name, " ARGUMENTS DETECTED IN THE add STRING:\n", paste(arg.names[sapply(X = arg.names, FUN = grepl, x = add)], collapse = "\n"), "\nRISK OF WRONG OBJECT USAGE INSIDE ", function.name)
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- # end reserved word checking
- # check of geom now because required for y argument
- tempo <- fun_check(data = geom[[i1]], data.name = ifelse(length(geom) == 1, "geom", paste0("geom NUMBER ", i1)), options = c("geom_point", "geom_line", "geom_path", "geom_step", "geom_hline", "geom_vline", "geom_stick"), length = 1, fun.name = function.name)
- if(tempo$problem == TRUE){
- stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- if(geom[[i1]] == "geom_step" & is.null(geom.step.dir[[i1]])){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(geom.step.dir) == 1, "geom.step.dir", paste0("ELEMENT ", i1, " OF geom.step.dir ARGUMENT")), ": geom.step.dir ARGUMENT CANNOT BE NULL IF ", ifelse(length(geom) == 1, "geom", paste0("ELEMENT ", i1, " OF geom")), " ARGUMENT IS \"geom_step\"\nHERE geom.step.dir ARGUMENT IS: ", paste(geom.step.dir[[i1]], collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if(geom[[i1]] == "geom_step" & ! is.null(geom.step.dir[[i1]])){
- tempo <- fun_check(data = geom.step.dir[[i1]], data.name = ifelse(length(geom.step.dir) == 1, "geom.step.dir", paste0("geom.step.dir NUMBER ", i1)), options = c("vh", "hv", "mid"), length = 1, fun.name = function.name)
- if(tempo$problem == TRUE){
- stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- # end check of geom now because required for y argument
- if(is.null(x[[i1]])){
- if(all(geom[[i1]] != "geom_hline")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(x) == 1, "x", paste0("ELEMENT ", i1, " OF x ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ": x ARGUMENT CANNOT BE NULL EXCEPT IF ", ifelse(length(geom) == 1, "x", paste0("geom NUMBER ", i1)), " ARGUMENT IS \"geom_hline\"\nHERE geom ARGUMENT IS: ", paste(geom[[i1]], collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- x[[i1]] <- "fake_x"
- data1[[i1]] <- cbind(data1[[i1]], fake_x = NA, stringsAsFactors = TRUE)
- data1[[i1]][, "fake_x"] <- as.numeric(data1[[i1]][, "fake_x"])
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NULL ", ifelse(length(x) == 1, "x", paste0("ELEMENT ", i1, " OF x")), " ARGUMENT ASSOCIATED TO ", ifelse(length(geom) == 1, "geom", paste0("geom NUMBER ", i1)), " ARGUMENT ", geom[[i1]], " -> FAKE COLUMN ADDED TO DATA FRAME ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ", NAMED \"fake_x\" FOR FINAL DRAWING")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else{
- if(all(geom[[i1]] == "geom_hline")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(x) == 1, "x", paste0("ELEMENT ", i1, " OF x ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ": x ARGUMENT MUST BE NULL IF ", ifelse(length(geom) == 1, "geom", paste0("geom NUMBER ", i1)), " ARGUMENT IS \"geom_hline\"")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- tempo <- fun_check(data = x[[i1]], data.name = ifelse(length(x) == 1, "x", paste0("ELEMENT ", i1, " OF x ARGUMENT")), class = "vector", mode = "character", length = 1, fun.name = function.name)
- if(tempo$problem == TRUE){
- stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- if(is.null(y[[i1]])){
- if(all(geom[[i1]] != "geom_vline")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(y) == 1, "y", paste0("ELEMENT ", i1, " OF y ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ": y ARGUMENT CANNOT BE NULL EXCEPT IF ", ifelse(length(geom) == 1, "y", paste0("geom NUMBER ", i1)), " ARGUMENT IS \"geom_vline\"\nHERE geom ARGUMENT IS: ", paste(geom[[i1]], collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- y[[i1]] <- "fake_y"
- data1[[i1]] <- cbind(data1[[i1]], fake_y = NA, stringsAsFactors = TRUE)
- data1[[i1]][, "fake_y"] <- as.numeric(data1[[i1]][, "fake_y"])
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NULL ", ifelse(length(y) == 1, "y", paste0("ELEMENT ", i1, " OF y")), " ARGUMENT ASSOCIATED TO ", ifelse(length(geom) == 1, "geom", paste0("geom NUMBER ", i1)), " ARGUMENT ", geom[[i1]], " -> FAKE COLUMN ADDED TO DATA FRAME ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ", NAMED \"fake_y\" FOR FINAL DRAWING")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else{
- if(all(geom[[i1]] == "geom_vline")){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(y) == 1, "y", paste0("ELEMENT ", i1, " OF y ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ": y ARGUMENT MUST BE NULL IF ", ifelse(length(geom) == 1, "geom", paste0("geom NUMBER ", i1)), " ARGUMENT IS \"geom_vline\"")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- tempo <- fun_check(data = y[[i1]], data.name = ifelse(length(y) == 1, "y", paste0("ELEMENT ", i1, " OF y ARGUMENT")), class = "vector", mode = "character", length = 1, fun.name = function.name)
- if(tempo$problem == TRUE){
- stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- # x[[i1]] and y[[i1]] not NULL anymore
- if( ! (x[[i1]] %in% names(data1[[i1]]))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(x) == 1, "x", paste0("ELEMENT ", i1, " OF x")), " ARGUMENT MUST BE A COLUMN NAME OF ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT\nHERE IT IS: ", paste(x[[i1]], collapse = " "))))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- if( ! (y[[i1]] %in% names(data1[[i1]]))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(y) == 1, "y", paste0("ELEMENT ", i1, " OF y")), " ARGUMENT MUST BE A COLUMN NAME OF ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT\nHERE IT IS: ", paste(y[[i1]], collapse = " "))))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- tempo <- fun_check(data = data1[[i1]][, x[[i1]]], data.name = ifelse(length(x) == 1, "x ARGUMENT (AS COLUMN NAME OF data1 DATA FRAME)", paste0("ELEMENT ", i1, " OF x ARGUMENT", " (AS COLUMN NAME OF data1 DATA FRAME NUMBER ", i1, ")")), class = "vector", mode = "numeric", na.contain = TRUE, fun.name = function.name)
- if(tempo$problem == TRUE){
- stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- tempo <- fun_check(data = data1[[i1]][, y[[i1]]], data.name = ifelse(length(y) == 1, "y ARGUMENT (AS COLUMN NAME OF data1 DATA FRAME)", paste0("ELEMENT ", i1, " OF y ARGUMENT", " (AS COLUMN NAME OF data1 DATA FRAME NUMBER ", i1, ")")), class = "vector", mode = "numeric", na.contain = TRUE, fun.name = function.name)
- if(tempo$problem == TRUE){
- stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- if(x[[i1]] == "fake_x" & y[[i1]] == "fake_y"){ # because the code cannot accept to be both "fake_x" and "fake_y" at the same time
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 2\nTHE CODE CANNOT ACCEPT x AND y TO BE \"fake_x\" AND \"fake_y\" IN THE SAME DATA FRAME ", i1, " ")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
-
- if(( ! is.null(categ)) & ( ! is.null(categ[[i1]]))){ # is.null(categ[[i1]]) works even if categ is NULL # is.null(categ[[i1]]) works even if categ is NULL # if categ[[i1]] = NULL, fake_categ will be created later on
- tempo <- fun_check(data = categ[[i1]], data.name = ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")),, class = "vector", mode = "character", length = 1, fun.name = function.name)
- if(tempo$problem == TRUE){
- stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- if( ! (categ[[i1]] %in% names(data1[[i1]]))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ")), " ARGUMENT MUST BE A COLUMN NAME OF ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT\nHERE IT IS: ", paste(categ[[i1]], collapse = " "))))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- tempo1 <- fun_check(data = data1[[i1]][, categ[[i1]]], data.name = ifelse(length(categ) == 1, "categ OF data1 ARGUMENT", paste0("ELEMENT ", i1, " OF categ ARGUMENT IN DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), class = "vector", mode = "character", na.contain = TRUE, fun.name = function.name)
- tempo2 <- fun_check(data = data1[[i1]][, categ[[i1]]], data.name = ifelse(length(categ) == 1, "categ OF data1 ARGUMENT", paste0("ELEMENT ", i1, " OF categ ARGUMENT IN DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), class = "factor", na.contain = TRUE, fun.name = function.name)
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(categ) == 1, "categ OF data1 ARGUMENT", paste0("ELEMENT ", i1, " OF categ ARGUMENT IN DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " MUST BE A FACTOR OR CHARACTER VECTOR")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if(tempo1$problem == FALSE){
- data1[[i1]][, categ[[i1]]] <- factor(data1[[i1]][, categ[[i1]]]) # if already a factor, change nothing, if characters, levels according to alphabetical order
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") IN ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ", THE CHARACTER COLUMN HAS BEEN CONVERTED TO FACTOR")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
-
- }
- if(geom[[i1]] == "geom_vline" | geom[[i1]] == "geom_hline"){
- if(length(unique(data1[[i1]][, categ[[i1]]])) != nrow(data1[[i1]])){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(geom) == 1, "geom OF data1 ARGUMENT", paste0("geom NUMBER ", i1, " OF DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " ARGUMENT IS ", geom[[i1]], ", MEANING THAT ", ifelse(length(categ) == 1, "categ OF data1 ARGUMENT", paste0("ELEMENT ", i1, " OF categ ARGUMENT IN DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " MUST HAVE A DIFFERENT CLASS PER LINE OF data1 (ONE x VALUE PER CLASS)")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- }else if(( ! is.null(categ)) & is.null(categ[[i1]])){ # is.null(categ[[i1]]) works even if categ is NULL # if categ[[i1]] = NULL, fake_categ will be created. WARNING: is.null(categ[[i1]]) means no legend display (see above), because categ has not been precised. This also means a single color for data1[[i1]]
- if(length(color[[i1]]) > 1){ # 0 means is.null(color[[i1]]) or is.null(color) and 1 is ok -> single color for data1[[i1]]
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NULL ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ")), " ARGUMENT BUT CORRESPONDING COLORS IN ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " HAS LENGTH OVER 1\n", paste(color[[i1]], collapse = " "), "\nWHICH IS NOT COMPATIBLE WITH NULL CATEG -> COLOR RESET TO A SINGLE COLOR")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- color[i1] <- list(NULL) # will provide a single color below # Warning color[[i1]] <- NULL removes the compartment
- }
- categ[[i1]] <- "fake_categ"
- data1[[i1]] <- cbind(data1[[i1]], fake_categ = "", stringsAsFactors = TRUE)
- # inactivated because give a different color to different "Line_" categ while a single color for all the data1[[i1]] required. Thus, put back after the color management
- # if(geom[[i1]] == "geom_hline" | geom[[i1]] == "geom_vline"){
- # data1[[i1]][, "fake_categ"] <- paste0("Line_", 1:nrow(data1[[i1]]))
- # }else{
- data1[[i1]][, "fake_categ"] <- data1[[i1]][, "fake_categ"] # as.numeric("") create a vector of NA but class numeric
- # }
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NULL ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ")), " ARGUMENT -> FOR DATA FRAME ", ifelse(length(data1) == 1, "data1 ARGUMENT:", paste0("NUMBER ", i1, " OF data1 ARGUMENT:")), "\n- FAKE \"fake_categ\" COLUMN ADDED FILLED WITH \"\"(OR WITH \"Line_...\" FOR LINES)\n- SINGLE COLOR USED FOR PLOTTING\n- NO LEGEND DISPLAYED")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- # OK: if categ is not NULL, all the non-null categ columns of data1 are factors from here
-
- # management of log scale and Inf removal
- if(x[[i1]] != "fake_x"){
- if(any(( ! is.finite(data1[[i1]][, x[[i1]]])) & ( ! is.na(data1[[i1]][, x[[i1]]])))){ # is.finite also detects NA: ( ! is.finite(data1[, y])) & ( ! is.na(data1[, y])) detects only Inf
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") PRESENCE OF -Inf OR Inf VALUES IN ", ifelse(length(categ) == 1, "x", paste0("ELEMENT ", i1, " OF x ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " AND CORRESPONDING ROWS REMOVED (SEE $removed.row.nb AND $removed.rows)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- if(y[[i1]] != "fake_y"){
- if(any(( ! is.finite(data1[[i1]][, y[[i1]]])) & ( ! is.na(data1[[i1]][, y[[i1]]])))){ # is.finite also detects NA: ( ! is.finite(data1[, y])) & ( ! is.na(data1[, y])) detects only Inf
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") PRESENCE OF -Inf OR Inf VALUES IN ", ifelse(length(categ) == 1, "y", paste0("ELEMENT ", i1, " OF y ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " AND CORRESPONDING ROWS REMOVED (SEE $removed.row.nb AND $removed.rows)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- # log conversion
- if(x.log != "no"){
- tempo1 <- ! is.finite(data1[[i1]][, x[[i1]]]) # where are initial NA and Inf
- data1[[i1]][, x[[i1]]] <- suppressWarnings(get(x.log)(data1[[i1]][, x[[i1]]]))# no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
- if(any( ! (tempo1 | is.finite(data1[[i1]][, x[[i1]]])))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") LOG CONVERSION INTRODUCED -Inf OR Inf OR NaN VALUES IN ", ifelse(length(categ) == 1, "x", paste0("ELEMENT ", i1, " OF x ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " AND CORRESPONDING ROWS REMOVED (SEE $removed.row.nb AND $removed.rows)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- if(y.log != "no"){
- tempo1 <- ! is.finite(data1[[i1]][, y[[i1]]]) # where are initial NA and Inf
- data1[[i1]][, y[[i1]]] <- suppressWarnings(get(y.log)(data1[[i1]][, y[[i1]]]))# no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
- if(any( ! (tempo1 | is.finite(data1[[i1]][, y[[i1]]])))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") LOG CONVERSION INTRODUCED -Inf OR Inf OR NaN VALUES IN ", ifelse(length(categ) == 1, "y", paste0("ELEMENT ", i1, " OF y ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " AND CORRESPONDING ROWS REMOVED (SEE $removed.row.nb AND $removed.rows)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- # Inf removal
- # removed.row.nb[[i1]] <- NULL # already NULL and Warning this removes the compartment
- removed.rows[[i1]] <- data.frame(stringsAsFactors = FALSE)
- if(any(( ! is.finite(data1[[i1]][, x[[i1]]])) & ( ! is.na(data1[[i1]][, x[[i1]]])))){ # is.finite also detects NA: ( ! is.finite(data1[[i1]][, x[[i1]]])) & ( ! is.na(data1[[i1]][, x[[i1]]])) detects only Inf
- removed.row.nb[[i1]] <- c(removed.row.nb[[i1]], which(( ! is.finite(data1[[i1]][, x[[i1]]])) & ( ! is.na(data1[[i1]][, x[[i1]]]))))
- }
- if(any(( ! is.finite(data1[[i1]][, y[[i1]]])) & ( ! is.na(data1[[i1]][, y[[i1]]])))){ # is.finite also detects NA: ( ! is.finite(data1[[i1]][, y[[i1]]])) & ( ! is.na(data1[[i1]][, y[[i1]]])) detects only Inf
- removed.row.nb[[i1]] <- c(removed.row.nb[[i1]], which(( ! is.finite(data1[[i1]][, y[[i1]]])) & ( ! is.na(data1[[i1]][, y[[i1]]]))))
- }
- if( ! is.null(removed.row.nb[[i1]])){
- removed.row.nb[[i1]] <- unique(removed.row.nb[[i1]]) # to remove the duplicated positions (NA in both x and y)
- removed.rows[[i1]] <- rbind(removed.rows[[i1]], data1.ini[[i1]][removed.row.nb[[i1]], ]) # here data1.ini used to have the y = O rows that will be removed because of Inf creation after log transformation
- data1[[i1]] <- data1[[i1]][-removed.row.nb[[i1]], ]
- data1.ini[[i1]] <- data1.ini[[i1]][-removed.row.nb[[i1]], ] #
- }
- # From here, data1 and data.ini have no more Inf
- # end Inf removal
- # x.lim and y.lim dealt later on, after the end f the loop
- # end management of log scale and Inf removal
- # management of log scale
- if(x.log != "no"){
- data1[[i1]][, x[[i1]]] <- suppressWarnings(get(x.log)(data1[[i1]][, x[[i1]]])) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
- }
- if(y.log != "no"){
- data1[[i1]][, y[[i1]]] <- suppressWarnings(get(y.log)(data1[[i1]][, y[[i1]]])) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
- }
- # end management of log scale
- # na detection and removal
- column.check <- unique(unlist(c( # unlist because creates a list
- if(x[[i1]] == "fake_x"){NULL}else{x[[i1]]},
- if(y[[i1]] == "fake_y"){NULL}else{y[[i1]]},
- if( ! is.null(categ)){if(is.null(categ[[i1]])){NULL}else{categ[[i1]]}},
- if( ! is.null(facet.categ)){if(is.null(facet.categ[[i1]])){NULL}else{facet.categ[[i1]]}}
- ))) # dot.categ because can be a 3rd column of data1
- if(any(is.na(data1[[i1]][, column.check]))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NA DETECTED IN COLUMNS ", paste(column.check, collapse = " "), " OF ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " AND CORRESPONDING ROWS REMOVED (SEE $removed.row.nb AND $removed.rows)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- for(i3 in 1:length(column.check)){
- if(any(is.na(data1[[i1]][, column.check[i3]]))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("NA REMOVAL DUE TO COLUMN ", column.check[i3], " OF ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- tempo <- unique(unlist(lapply(lapply(c(data1[[i1]][column.check]), FUN = is.na), FUN = which)))
- removed.row.nb[[i1]] <- c(removed.row.nb[[i1]], tempo)
- removed.rows[[i1]] <- rbind(removed.rows[[i1]], data1.ini[[i1]][tempo, ]) # # tempo used because removed.row.nb is not empty. Here data1.ini used to have the non NA rows that will be removed because of NAN creation after log transformation (neg values for instance)
- column.check <- column.check[ ! (column.check == x[[i1]] | column.check == y[[i1]])] # remove x and y to keep quali columns
- if(length(tempo) != 0){
- data1[[i1]] <- data1[[i1]][-tempo, ] # WARNING tempo here and not removed.row.nb because the latter contain more numbers thant the former
- data1.ini[[i1]] <- data1.ini[[i1]][-tempo, ] # WARNING tempo here and not removed.row.nb because the latter contain more numbers than the former
- for(i4 in 1:length(column.check)){
- if(any( ! unique(removed.rows[[i1]][, column.check[i4]]) %in% unique(data1[[i1]][, column.check[i4]]))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") IN COLUMN ", column.check[i4], " OF ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ", THE FOLLOWING CLASSES HAVE DISAPPEARED AFTER NA REMOVAL\n(IF COLUMN USED IN THE PLOT, THIS CLASS WILL NOT BE DISPLAYED):\n", paste(unique(removed.rows[[i1]][, column.check[i4]])[ ! unique(removed.rows[[i1]][, column.check[i4]]) %in% unique(data1[[i1]][, column.check[i4]])], collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- tempo.levels <- levels(data1[[i1]][, column.check[i4]])[levels(data1[[i1]][, column.check[i4]]) %in% unique(as.character(data1[[i1]][, column.check[i4]]))]
- data1[[i1]][, column.check[i4]] <- factor(as.character(data1[[i1]][, column.check[i4]]), levels = tempo.levels)
- if(column.check[i4] %in% categ[[i1]] & ! is.null(categ.class.order)){
- categ.class.order[[i1]] <- levels(data1[[i1]][, column.check[i4]])[levels(data1[[i1]][, column.check[i4]]) %in% unique(data1[[i1]][, column.check[i4]])] # remove the absent class in the categ.class.order vector
- data1[[i1]][, column.check[i4]] <- factor(as.character(data1[[i1]][, column.check[i4]]), levels = unique(categ.class.order[[i1]]))
- }
- }
- }
- }
- }
- # end na detection and removal
- # From here, data1 and data.ini have no more NA or NaN in x, y, categ (if categ != NULL) and facet.categ (if categ != NULL)
- if( ! is.null(categ.class.order)){
- # the following check will be done several times but I prefer to keep it here, after the creation of categ
- if(is.null(categ[[i1]]) & ! is.null(categ.class.order[[i1]])){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nCOMPARTMENT ", i1, " OF categ ARGUMENT CANNOT BE NULL IF COMPARTMENT ", i1, " OF categ.class.order ARGUMENT IS NOT NULL: ", paste(categ.class.order, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- if(is.null(categ.class.order[[i1]])){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE categ.class.order COMPARTMENT ", i1, " IS NULL. ALPHABETICAL ORDER WILL BE APPLIED")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- data1[[i1]][, categ[[i1]]] <- factor(as.character(data1[[i1]][, categ[[i1]]])) # if already a factor, change nothing, if characters, levels according to alphabetical order
- categ.class.order[[i1]] <- levels(data1[[i1]][, categ[[i1]]]) # character vector that will be used later
- }else{
- tempo <- fun_check(data = categ.class.order[[i1]], data.name = paste0("COMPARTMENT ", i1 , " OF categ.class.order ARGUMENT"), class = "vector", mode = "character", length = length(levels(data1[[i1]][, categ[[i1]]])), fun.name = function.name) # length(data1[, categ[i1]) -> if data1[, categ[i1] was initially character vector, then conversion as factor after the NA removal, thus class number ok. If data1[, categ[i1] was initially factor, no modification after the NA removal, thus class number ok
- if(tempo$problem == TRUE){
- stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- if(any(duplicated(categ.class.order[[i1]]))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nCOMPARTMENT ", i1, " OF categ.class.order ARGUMENT CANNOT HAVE DUPLICATED CLASSES: ", paste(categ.class.order[[i1]], collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if( ! (all(categ.class.order[[i1]] %in% unique(data1[[i1]][, categ[[i1]]])) & all(unique(data1[[i1]][, categ[[i1]]]) %in% categ.class.order[[i1]]))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nCOMPARTMENT ", i1, " OF categ.class.order ARGUMENT MUST BE CLASSES OF COMPARTMENT ", i1, " OF categ ARGUMENT\nHERE IT IS:\n", paste(categ.class.order[[i1]], collapse = " "), "\nFOR COMPARTMENT ", i1, " OF categ.class.order AND IT IS:\n", paste(unique(data1[[i1]][, categ[[i1]]]), collapse = " "), "\nFOR COLUMN ", categ[[i1]], " OF data1 NUMBER ", i1)
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- data1[[i1]][, categ[[i1]]] <- factor(data1[[i1]][, categ[[i1]]], levels = categ.class.order[[i1]]) # reorder the factor
- }
- names(categ.class.order)[i1] <- categ[[i1]]
- }
- }
- # OK: if categ.class.order is not NULL, all the NULL categ.class.order columns of data1 are character from here
-
- if( ! is.null(legend.name[[i1]])){
- tempo <- fun_check(data = legend.name[[i1]], data.name = ifelse(length(legend.name) == 1, "legend.name", paste0("legend.name NUMBER ", i1)),, class = "vector", mode = "character", length = 1, fun.name = function.name)
- if(tempo$problem == TRUE){
- stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- if( ! is.null(color)){ # if color is NULL, will be filled later on
- # check the nature of color
- if(is.null(color[[i1]])){
- compart.null.color <- compart.null.color + 1
- color[[i1]] <- grey(compart.null.color / 8) # cannot be more than 7 overlays. Thus 7 different greys. 8/8 is excluded because white dots
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NULL COLOR IN ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " ASSOCIATED TO ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ", SINGLE COLOR ", paste(color[[i1]], collapse = " "), " HAS BEEN ATTRIBUTED")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- tempo1 <- fun_check(data = color[[i1]], data.name = ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), class = "vector", mode = "character", na.contain = TRUE, fun.name = function.name) # na.contain = TRUE in case of colum of data1
- tempo2 <- fun_check(data = color[[i1]], data.name = ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), class = "factor", na.contain = TRUE, fun.name = function.name) # idem
- if(tempo1$problem == TRUE & tempo2$problem == TRUE){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " MUST BE A FACTOR OR CHARACTER VECTOR OR INTEGER VECTOR") # integer possible because dealt above
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if( ! (all(color[[i1]] %in% colors() | grepl(pattern = "^#", color[[i1]])))){ # check that all strings of low.color start by #
- tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # AND/OR COLOR NAMES GIVEN BY colors(): ", paste(unique(color[[i1]]), collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- if(any(is.na(color[[i1]]))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") IN ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), ", THE COLORS:\n", paste(unique(color[[i1]]), collapse = " "), "\nCONTAINS NA")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- # end check the nature of color
- # check the length of color
- if(is.null(categ) & length(color[[i1]]) != 1){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " MUST BE A SINGLE COLOR IF categ IS NULL")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if( ! is.null(categ)){
- # No problem of NA management by ggplot2 because already removed
- if(categ[[i1]] == "fake_categ" & length(color[[i1]]) != 1){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " MUST BE A SINGLE COLOR IF ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IS NULL")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if(length(color[[i1]]) == length(unique(data1[[i1]][, categ[[i1]]]))){ # here length(color) is equal to the different number of categ
- data1[[i1]][, categ[[i1]]] <- factor(data1[[i1]][, categ[[i1]]]) # if already a factor, change nothing, if characters, levels according to alphabetical order
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") IN ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ", THE FOLLOWING COLORS:\n", paste(color[[i1]], collapse = " "), "\nHAVE BEEN ATTRIBUTED TO THESE CLASSES:\n", paste(levels(factor(data1[[i1]][, categ[[i1]]])), collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }else if(length(color[[i1]]) == length(data1[[i1]][, categ[[i1]]])){# here length(color) is equal to nrow(data1[[i1]]) -> Modif to have length(color) equal to the different number of categ (length(color) == length(levels(data1[[i1]][, categ[[i1]]])))
- data1[[i1]] <- cbind(data1[[i1]], color = color[[i1]], stringsAsFactors = TRUE)
- tempo.check <- unique(data1[[i1]][ , c(categ[[i1]], "color")])
- if( ! (nrow(data1[[i1]]) == length(color[[i1]]) & nrow(tempo.check) == length(unique(data1[[i1]][ , categ[[i1]]])))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color")), " ARGUMENT HAS THE LENGTH OF ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), "\nBUT IS INCORRECTLY ASSOCIATED TO EACH CLASS OF THIS categ:\n", paste(unique(mapply(FUN = "paste", data1[[i1]][ ,categ[[i1]]], data1[[i1]][ ,"color"])), collapse = "\n"))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- data1[[i1]][, categ[[i1]]] <- factor(data1[[i1]][, categ[[i1]]]) # if already a factor, change nothing, if characters, levels according to alphabetical order
- color[[i1]] <- unique(color[[i1]][order(data1[[i1]][, categ[[i1]]])]) # Modif to have length(color) equal to the different number of categ (length(color) == length(levels(data1[[i1]][, categ[[i1]]])))
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count, ") FROM FUNCTION ", function.name, ": ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " HAS THE LENGTH OF ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " COLUMN VALUES\nCOLORS HAVE BEEN RESPECTIVELY ASSOCIATED TO EACH CLASS OF categ AS:\n", paste(levels(factor(data1[[i1]][, categ[[i1]]])), collapse = " "), "\n", paste(color[[i1]], collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else if(length(color[[i1]]) == 1){
- data1[[i1]][, categ[[i1]]] <- factor(data1[[i1]][, categ[[i1]]]) # if already a factor, change nothing, if characters, levels according to alphabetical order
- color[[i1]] <- rep(color[[i1]], length(levels(data1[[i1]][, categ[[i1]]])))
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") IN ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), ", COLOR HAS LENGTH 1 MEANING THAT ALL THE DIFFERENT CLASSES OF ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), "\n", paste(levels(factor(data1[[i1]][, categ[[i1]]])), collapse = " "), "\nWILL HAVE THE SAME COLOR\n", paste(color[[i1]], collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, ": ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " MUST BE\n(1) LENGTH 1\nOR (2) THE LENGTH OF ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), " COLUMN VALUES\nOR (3) THE LENGTH OF THE CLASSES IN THIS COLUMN\nHERE IT IS COLOR LENGTH ", length(color[[i1]]), " VERSUS CATEG LENGTH ", length(data1[[i1]][, categ[[i1]]]), " AND CATEG CLASS LENGTH ", length(unique(data1[[i1]][, categ[[i1]]])), "\nPRESENCE OF NA IN THE COLUMN x, y OR categ OF data1 COULD BE THE PROBLEME")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- }
- if((geom[[i1]] == "geom_hline" | geom[[i1]] == "geom_vline") & ! is.null(categ[[i1]])){ # add here after the color management, to deal with the different lines to plot inside any data[[i1]]
- if(categ[[i1]] == "fake_categ"){
- data1[[i1]][, "fake_categ"] <- factor(paste0("Line_", formatC(1:nrow(data1[[i2]]), width = nchar(nrow(data1[[i2]])), flag = "0")))
- }
- }
- tempo <- fun_check(data = alpha[[i1]], data.name = ifelse(length(alpha) == 1, "alpha", paste0("alpha NUMBER ", i1)), prop = TRUE, length = 1, fun.name = function.name)
- if(tempo$problem == TRUE){
- stop(paste0("\n\n================\n\n", tempo$text, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- # end loop (checking inside list compartment)
- if(length(data1) > 1){
- if(length(unique(unlist(x)[ ! x == "fake_x"])) > 1){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE x ARGUMENT DOES NOT CONTAIN IDENTICAL COLUMN NAMES:\n", paste(unlist(x), collapse = " "), "\nX-AXIS OVERLAYING DIFFERENT VARIABLES?")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- if(length(data1) > 1){
- if(length(unique(unlist(y)[ ! y == "fake_y"])) > 1){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE y ARGUMENT DOES NOT CONTAIN IDENTICAL COLUMN NAMES:\n", paste(unlist(y), collapse = " "), "\nY-AXIS OVERLAYING DIFFERENT VARIABLES?")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- if(sum(geom %in% "geom_point") > 3){
- tempo.cat <- paste0("ERROR IN ", function.name, ": geom ARGUMENT CANNOT HAVE MORE THAN THREE \"geom_point\" ELEMENTS")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if(length(geom) - sum(geom %in% "geom_point") > 3){
- tempo.cat <- paste0("ERROR IN ", function.name, ": geom ARGUMENT CANNOT HAVE MORE THAN THREE LINE ELEMENTS")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- # x.lim management before transfo by x.log
- if(x.log != "no" & ! is.null(x.lim)){
- if(any(x.lim <= 0)){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nx.lim ARGUMENT CANNOT HAVE ZERO OR NEGATIVE VALUES WITH THE x.log ARGUMENT SET TO ", x.log, ":\n", paste(x.lim, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if(any( ! is.finite(if(x.log == "log10"){log10(x.lim)}else{log2(x.lim)}))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nx.lim ARGUMENT RETURNS INF/NA WITH THE x.log ARGUMENT SET TO ", x.log, "\nAS SCALE COMPUTATION IS ", ifelse(x.log == "log10", "log10", "log2"), ":\n", paste(if(x.log == "log10"){log10(x.lim)}else{log2(x.lim)}, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- if(x.log != "no" & x.include.zero == TRUE){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") x.log ARGUMENT SET TO ", x.log, " AND x.include.zero ARGUMENT SET TO TRUE -> x.include.zero ARGUMENT RESET TO FALSE BECAUSE 0 VALUE CANNOT BE REPRESENTED IN LOG SCALE")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- x.include.zero <- FALSE
- }
- # end x.lim management before transfo by x.log
- # y.lim management before transfo by y.log
- if(y.log != "no" & ! is.null(y.lim)){
- if(any(y.lim <= 0)){
- tempo.cat <- paste0("ERROR IN ", function.name, "\ny.lim ARGUMENT CANNOT HAVE ZERO OR NEGATIVE VALUES WITH THE y.log ARGUMENT SET TO ", y.log, ":\n", paste(y.lim, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if(any( ! is.finite(if(y.log == "log10"){log10(y.lim)}else{log2(y.lim)}))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\ny.lim ARGUMENT RETURNS INF/NA WITH THE y.log ARGUMENT SET TO ", y.log, "\nAS SCALE COMPUTATION IS ", ifelse(y.log == "log10", "log10", "log2"), ":\n", paste(if(y.log == "log10"){log10(y.lim)}else{log2(y.lim)}, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- if(y.log != "no" & y.include.zero == TRUE){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") y.log ARGUMENT SET TO ", y.log, " AND y.include.zero ARGUMENT SET TO TRUE -> y.include.zero ARGUMENT RESET TO FALSE BECAUSE 0 VALUE CANNOT BE REPRESENTED IN LOG SCALE")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- y.include.zero <- FALSE
- }
- # end y.lim management before transfo by y.log
- # end second round of checking and data preparation
-
-
-
-
-
- # package checking
- fun_pack(req.package = c(
- "gridExtra",
- "ggplot2",
- "lemon",
- "scales"
- ), lib.path = lib.path)
- # packages Cairo and grid tested by fun_gg_point_rast()
- # end package checking
-
-
-
-
- # main code
- # axes management
- if(is.null(x.lim)){
- if(any(unlist(mapply(FUN = "[[", data1, x, SIMPLIFY = FALSE)) %in% c(Inf, -Inf))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE x COLUMN IN data1 CONTAINS -Inf OR Inf VALUES THAT WILL NOT BE CONSIDERED IN THE PLOT RANGE")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- x.lim <- suppressWarnings(range(unlist(mapply(FUN = "[[", data1, x, SIMPLIFY = FALSE)), na.rm = TRUE, finite = TRUE)) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only. y.lim added here. If NULL, ok if y argument has values
- }else if(x.log != "no"){
- x.lim <- get(x.log)(x.lim) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
- }
- if(x.log != "no"){
- if(any( ! is.finite(x.lim))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nx.lim ARGUMENT CANNOT HAVE ZERO OR NEGATIVE VALUES WITH THE x.log ARGUMENT SET TO ", x.log, ":\n", paste(x.lim, collapse = " "), "\nPLEASE, CHECK DATA VALUES (PRESENCE OF ZERO OR INF VALUES)")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- if(suppressWarnings(all(x.lim %in% c(Inf, -Inf)))){ # happen when x is only NULL
- if(all(unlist(geom) %in% c("geom_vline", "geom_stick"))){
- tempo.cat <- paste0("ERROR IN ", function.name, " NOT POSSIBLE TO DRAW geom_vline OR geom_stick KIND OF LINES ALONE IF x.lim ARGUMENT IS SET TO NULL, SINCE NO X-AXIS DEFINED (", ifelse(length(x) == 1, "x", paste0("ELEMENT ", i1, " OF x")), " ARGUMENT MUST BE NULL FOR THESE KIND OF LINES)")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, " x.lim ARGUMENT MADE OF NA, -Inf OR Inf ONLY: ", paste(x.lim, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- x.lim.order <- order(x.lim) # to deal with inverse axis
- # print(x.lim.order)
- x.lim <- sort(x.lim)
- x.lim[1] <- x.lim[1] - abs(x.lim[2] - x.lim[1]) * ifelse(diff(x.lim.order) > 0, x.right.extra.margin, x.left.extra.margin) # diff(x.lim.order) > 0 means not inversed axis
- x.lim[2] <- x.lim[2] + abs(x.lim[2] - x.lim[1]) * ifelse(diff(x.lim.order) > 0, x.left.extra.margin, x.right.extra.margin) # diff(x.lim.order) > 0 means not inversed axis
- if(x.include.zero == TRUE){ # no need to check x.log != "no" because done before
- x.lim <- range(c(x.lim, 0), na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
- }
- x.lim <- x.lim[x.lim.order]
- if(any(is.na(x.lim))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 3")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- if(is.null(y.lim)){
- if(any(unlist(mapply(FUN = "[[", data1, y, SIMPLIFY = FALSE)) %in% c(Inf, -Inf))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE y COLUMN IN data1 CONTAINS -Inf OR Inf VALUES THAT WILL NOT BE CONSIDERED IN THE PLOT RANGE")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- y.lim <- suppressWarnings(range(unlist(mapply(FUN = "[[", data1, y, SIMPLIFY = FALSE)), na.rm = TRUE, finite = TRUE)) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only. y.lim added here. If NULL, ok if y argument has values
- }else if(y.log != "no"){
- y.lim <- get(y.log)(y.lim) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
- }
- if(y.log != "no"){
- if(any( ! is.finite(y.lim))){
- tempo.cat <- paste0("ERROR IN ", function.name, "\ny.lim ARGUMENT CANNOT HAVE ZERO OR NEGATIVE VALUES WITH THE y.log ARGUMENT SET TO ", y.log, ":\n", paste(y.lim, collapse = " "), "\nPLEASE, CHECK DATA VALUES (PRESENCE OF ZERO OR INF VALUES)")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- if(suppressWarnings(all(y.lim %in% c(Inf, -Inf)))){ # happen when y is only NULL
- if(all(unlist(geom) == "geom_vline")){
- tempo.cat <- paste0("ERROR IN ", function.name, " NOT POSSIBLE TO DRAW geom_vline KIND OF LINES ALONE IF y.lim ARGUMENT IS SET TO NULL, SINCE NO Y-AXIS DEFINED (", ifelse(length(y) == 1, "y", paste0("ELEMENT ", i1, " OF y")), " ARGUMENT MUST BE NULL FOR THESE KIND OF LINES)")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, " y.lim ARGUMENT MADE OF NA, -Inf OR Inf ONLY: ", paste(y.lim, collapse = " "))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- y.lim.order <- order(y.lim) # to deal with inverse axis
- y.lim <- sort(y.lim)
- y.lim[1] <- y.lim[1] - abs(y.lim[2] - y.lim[1]) * ifelse(diff(y.lim.order) > 0, y.bottom.extra.margin, y.top.extra.margin) # diff(y.lim.order) > 0 means not inversed axis
- y.lim[2] <- y.lim[2] + abs(y.lim[2] - y.lim[1]) * ifelse(diff(y.lim.order) > 0, y.top.extra.margin, y.bottom.extra.margin) # diff(y.lim.order) > 0 means not inversed axis
- if(y.include.zero == TRUE){ # no need to check y.log != "no" because done before
- y.lim <- range(c(y.lim, 0), na.rm = TRUE, finite = TRUE) # finite = TRUE removes all the -Inf and Inf except if only this. In that case, whatever the -Inf and/or Inf present, output -Inf;Inf range. Idem with NA only
- }
- y.lim <- y.lim[y.lim.order]
- if(any(is.na(y.lim))){
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 4")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- # end axes management
-
-
-
-
- # create a fake categ if NULL to deal with legend display
- if(is.null(categ)){
- categ <- vector("list", length(data1))
- categ[] <- "fake_categ"
- for(i2 in 1:length(data1)){
- data1[[i2]] <- cbind(data1[[i2]], fake_categ = "", stringsAsFactors = TRUE)
- if(geom[[i2]] == "geom_hline" | geom[[i2]] == "geom_vline"){
- data1[[i2]][, "fake_categ"] <- factor(paste0("Line_", 1:nrow(data1[[i2]])))
- }
- }
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NULL categ ARGUMENT -> FAKE \"fake_categ\" COLUMN ADDED TO EACH DATA FRAME OF data1, AND FILLED WITH \"\"")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- # categ is not NULL anymore
- if(is.null(categ.class.order)){
- categ.class.order <- vector("list", length = length(data1))
- tempo.categ.class.order <- NULL
- for(i2 in 1:length(categ.class.order)){
- categ.class.order[[i2]] <- levels(data1[[i2]][, categ[[i2]]])
- names(categ.class.order)[i2] <- categ[[i2]]
- tempo.categ.class.order <- c(tempo.categ.class.order, ifelse(i2 != 1, "\n", ""), categ.class.order[[i2]])
- }
- if(any(unlist(legend.disp))){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") THE categ.class.order SETTING IS NULL. ALPHABETICAL ORDER WILL BE APPLIED FOR CLASS ORDERING:\n", paste(tempo.categ.class.order, collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- # end create a fake categ if NULL to deal with legend display
- # categ.class.order is not NULL anymore
-
-
- # vector of color with length as in levels(categ) of data1
- if(is.null(color)){
- color <- vector("list", length(data1))
- length.categ.list <- lapply(lapply(mapply(FUN = "[[", data1, categ, SIMPLIFY = FALSE), FUN = unique), FUN = function(x){length(x[ ! is.na(x)])})
- length.categ.list[sapply(categ, FUN = "==", "fake_categ")] <- 1 # when is.null(color), a single color for all the dots or lines of data[[i1]] that contain "fake_categ" category
- total.categ.length <- sum(unlist(length.categ.list), na.rm = TRUE)
- tempo.color <- fun_gg_palette(total.categ.length)
- tempo.count <- 0
- for(i2 in 1:length(data1)){
- color[[i2]] <- tempo.color[(1:length.categ.list[[i2]]) + tempo.count]
- tempo.count <- tempo.count + length.categ.list[[i2]]
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") NULL color ARGUMENT -> COLORS RESPECTIVELY ATTRIBUTED TO EACH CLASS OF ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i2, " OF categ ARGUMENT")), " (", categ[[i2]], ") IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i2, " OF data1 ARGUMENT")), ":\n", paste(color[[i2]], collapse = " "), "\n", paste(if(all(levels(data1[[i2]][, categ[[i2]]]) == "")){'\"\"'}else{levels(data1[[i2]][, categ[[i2]]])}, collapse = " "))
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- # end vector of color with length as in levels(categ) of data1
- # color is not NULL anymore
-
-
-
-
-
- # last check
- for(i1 in 1:length(data1)){
- if(categ[[i1]] != "fake_categ" & length(color[[i1]]) != length(unique(data1[[i1]][, categ[[i1]]]))){
- tempo.cat <- paste0("ERROR IN ", function.name, " LAST CHECK: ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " MUST HAVE THE LENGTH OF LEVELS OF ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), "\nHERE IT IS COLOR LENGTH ", length(color[[i1]]), " VERSUS CATEG LEVELS LENGTH ", length(unique(data1[[i1]][, categ[[i1]]])))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else if(categ[[i1]] == "fake_categ" & length(color[[i1]]) != 1){
- tempo.cat <- paste0("ERROR IN ", function.name, " LAST CHECK: ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " MUST HAVE LENGTH 1 WHEN ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IS NULL\nHERE IT IS COLOR LENGTH ", length(color[[i1]]))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- # end last check
-
-
-
-
-
- # conversion of geom_hline and geom_vline
- for(i1 in 1:length(data1)){
- if(geom[[i1]] == "geom_hline" | geom[[i1]] == "geom_vline"){
- final.data.frame <- data.frame()
- for(i3 in 1:nrow(data1[[i1]])){
- tempo.data.frame <- rbind(data1[[i1]][i3, ], data1[[i1]][i3, ], stringsAsFactors = TRUE)
- if(geom[[i1]] == "geom_hline"){
- tempo.data.frame[, x[[i1]]] <- x.lim
- }else if(geom[[i1]] == "geom_vline"){
- tempo.data.frame[, y[[i1]]] <- y.lim
- }else{
- tempo.cat <- paste0("ERROR IN ", function.name, ": CODE INCONSISTENCY 5")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- # 3 lines below inactivated because I put that above
- # if(is.null(categ[[i1]])){
- # data1[, "fake_categ"] <- paste0("Line_", i3)
- # }
- final.data.frame <- rbind(final.data.frame, tempo.data.frame, stringsAsFactors = TRUE)
- }
- data1[[i1]] <- final.data.frame
- geom[[i1]] <- "geom_line"
- if(length(color[[i1]]) == 1){
- color[[i1]] <- rep(color[[i1]], length(unique(data1[[i1]][ , categ[[i1]]])))
- }else if(length(color[[i1]]) != length(unique(data1[[i1]][ , categ[[i1]]]))){
- tempo.cat <- paste0("ERROR IN ", function.name, " geom_hline AND geom_vline CONVERSION TO FIT THE XLIM AND YLIM LIMITS OF THE DATA: ", ifelse(length(color) == 1, "color", paste0("ELEMENT NUMBER ", i1, " OF color ARGUMENT")), " MUST HAVE THE LENGTH OF LEVELS OF ", ifelse(length(categ) == 1, "categ", paste0("ELEMENT ", i1, " OF categ ARGUMENT")), " IN ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i1, " OF data1 ARGUMENT")), "\nHERE IT IS COLOR LENGTH ", length(color[[i1]]), " VERSUS CATEG LEVELS LENGTH ", length(unique(data1[[i1]][, categ[[i1]]])))
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- }
- }
- # end conversion of geom_hline and geom_vline
-
-
-
-
- # kind of geom_point (vectorial or raster)
- scatter.kind <- vector("list", length = length(data1)) # list of same length as data1, that will be used to use either ggplot2::geom_point() (vectorial dot layer) or fun_gg_point_rast() (raster dot layer)
- fix.ratio <- FALSE
- if(is.null(raster.threshold)){
- if(raster == TRUE){
- scatter.kind[] <- "fun_gg_point_rast" # not important to fill everything: will be only used when geom == "geom_point"
- fix.ratio <- TRUE
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") RASTER PLOT GENERATED -> ASPECT RATIO OF THE PLOT REGION SET BY THE raster.ratio ARGUMENT (", fun_round(raster.ratio, 2), ") TO AVOID A BUG OF ELLIPSOID DOT DRAWING")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }else{
- scatter.kind[] <- "ggplot2::geom_point"
- }
- }else{
- for(i2 in 1:length(data1)){
- if(geom[[i2]] == "geom_point"){
- if(nrow(data1[[i2]]) <= raster.threshold){
- scatter.kind[[i2]] <- "ggplot2::geom_point"
- }else{
- scatter.kind[[i2]] <- "fun_gg_point_rast"
- fix.ratio <- TRUE
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") ", ifelse(length(data1) == 1, "data1 ARGUMENT", paste0("DATA FRAME NUMBER ", i2, " OF data1 ARGUMENT")), " LAYER AS RASTER (NOT VECTORIAL)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- }
- if(any(unlist(scatter.kind) == "fun_gg_point_rast")){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") RASTER PLOT GENERATED -> ASPECT RATIO OF THE PLOT REGION SET BY THE raster.ratio ARGUMENT (", fun_round(raster.ratio, 2), ") TO AVOID A BUG OF ELLIPSOID DOT DRAWING")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }
- # end kind of geom_point (vectorial or raster)
-
-
-
-
- # no need loop part
- coord.names <- NULL
- tempo.gg.name <- "gg.indiv.plot."
- tempo.gg.count <- 0
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = paste0("ggplot2::ggplot()", if(is.null(add)){""}else{add})))) # add added here to have the facets
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::xlab(if(is.null(x.lab)){x[[1]]}else{x.lab}))
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::ylab(if(is.null(y.lab)){y[[1]]}else{y.lab}))
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::ggtitle(title))
- # text angle management
- x.tempo.just <- fun_gg_just(angle = x.text.angle, pos = "bottom", kind = "axis")
- y.tempo.just <- fun_gg_just(angle = y.text.angle, pos = "left", kind = "axis")
- # end text angle management
- add.check <- TRUE
- if( ! is.null(add)){ # if add is NULL, then = 0
- if(grepl(pattern = "ggplot2::theme", add) == TRUE){
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") \"ggplot2::theme\" STRING DETECTED IN THE add ARGUMENT\n-> INTERNAL GGPLOT2 THEME FUNCTIONS theme() AND theme_classic() HAVE BEEN INACTIVATED, TO BE USED BY THE USER\n-> article ARGUMENT WILL BE IGNORED\nIT IS RECOMMENDED TO USE \"+ theme(aspect.ratio = raster.ratio)\" IF RASTER MODE IS ACTIVATED")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- add.check <- FALSE
- }
- }
- if(add.check == TRUE & article == TRUE){
- # WARNING: not possible to add several times theme(). NO message but the last one overwrites the others
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::theme_classic(base_size = text.size))
- if(grid == TRUE){
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), m.gg <- ggplot2::theme(
- text = ggplot2::element_text(size = text.size),
- plot.title = ggplot2::element_text(size = title.text.size), # stronger than text
- legend.key = ggplot2::element_rect(color = "white", size = 1.5), # size of the frame of the legend
- line = ggplot2::element_line(size = 0.5),
- axis.line.y.left = ggplot2::element_line(colour = "black"), # draw lines for the y axis
- axis.line.x.bottom = ggplot2::element_line(colour = "black"), # draw lines for the x axis
- panel.grid.major.x = ggplot2::element_line(colour = "grey85", size = 0.75),
- panel.grid.minor.x = ggplot2::element_line(colour = "grey90", size = 0.25),
- panel.grid.major.y = ggplot2::element_line(colour = "grey85", size = 0.75),
- panel.grid.minor.y = ggplot2::element_line(colour = "grey90", size = 0.25),
- axis.text.x = ggplot2::element_text(angle = x.tempo.just$angle, hjust = x.tempo.just$hjust, vjust = x.tempo.just$vjust),
- axis.text.y = ggplot2::element_text(angle = y.tempo.just$angle, hjust = y.tempo.just$hjust, vjust = y.tempo.just$vjust),
- aspect.ratio = if(fix.ratio == TRUE){raster.ratio}else{NULL} # for raster
- ))
- }else{
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), m.gg <- ggplot2::theme(
- text = ggplot2::element_text(size = text.size),
- plot.title = ggplot2::element_text(size = title.text.size), # stronger than text
- line = ggplot2::element_line(size = 0.5),
- legend.key = ggplot2::element_rect(color = "white", size = 1.5), # size of the frame of the legend
- axis.line.y.left = ggplot2::element_line(colour = "black"),
- axis.line.x.bottom = ggplot2::element_line(colour = "black"),
- axis.text.x = ggplot2::element_text(angle = x.tempo.just$angle, hjust = x.tempo.just$hjust, vjust = x.tempo.just$vjust),
- axis.text.y = ggplot2::element_text(angle = y.tempo.just$angle, hjust = y.tempo.just$hjust, vjust = y.tempo.just$vjust),
- aspect.ratio = if(fix.ratio == TRUE){raster.ratio}else{NULL} # for raster
- ))
- }
- }else if(add.check == TRUE & article == FALSE){
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), m.gg <- ggplot2::theme(
- text = ggplot2::element_text(size = text.size),
- plot.title = ggplot2::element_text(size = title.text.size), # stronger than text
- line = ggplot2::element_line(size = 0.5),
- legend.key = ggplot2::element_rect(color = "white", size = 1.5), # size of the frame of the legend
- panel.background = ggplot2::element_rect(fill = "grey95"),
- axis.line.y.left = ggplot2::element_line(colour = "black"),
- axis.line.x.bottom = ggplot2::element_line(colour = "black"),
- panel.grid.major.x = ggplot2::element_line(colour = "grey85", size = 0.75),
- panel.grid.minor.x = ggplot2::element_line(colour = "grey90", size = 0.25),
- panel.grid.major.y = ggplot2::element_line(colour = "grey85", size = 0.75),
- panel.grid.minor.y = ggplot2::element_line(colour = "grey90", size = 0.25),
- strip.background = ggplot2::element_rect(fill = "white", colour = "black"),
- axis.text.x = ggplot2::element_text(angle = x.tempo.just$angle, hjust = x.tempo.just$hjust, vjust = x.tempo.just$vjust),
- axis.text.y = ggplot2::element_text(angle = y.tempo.just$angle, hjust = y.tempo.just$hjust, vjust = y.tempo.just$vjust),
- aspect.ratio = if(fix.ratio == TRUE){raster.ratio}else{NULL} # for raster
- # do not work -> legend.position = "none" # to remove the legend completely: https://www.datanovia.com/en/blog/how-to-remove-legend-from-a-ggplot/
- ))
- }
- # end no need loop part
-
-
- # loop part
- point.count <- 0
- line.count <- 0
- lg.order <- vector(mode = "list", length = 6) # order of the legend
- lg.order <- lapply(lg.order, as.numeric) # order of the legend
- lg.color <- vector(mode = "list", length = 6) # color of the legend
- lg.dot.shape <- vector(mode = "list", length = 6) # etc.
- lg.dot.size <- vector(mode = "list", length = 6) # etc.
- lg.dot.size <- lapply(lg.dot.size, as.numeric) # etc.
- lg.dot.border.size <- vector(mode = "list", length = 6) # etc.
- lg.dot.border.size <- lapply(lg.dot.border.size, as.numeric) # etc.
- lg.dot.border.color <- vector(mode = "list", length = 6) # etc.
- lg.line.size <- vector(mode = "list", length = 6) # etc.
- lg.line.size <- lapply(lg.line.size, as.numeric) # etc.
- lg.line.type <- vector(mode = "list", length = 6) # etc.
- lg.alpha <- vector(mode = "list", length = 6) # etc.
- lg.alpha <- lapply(lg.alpha, as.numeric) # etc.
- for(i1 in 1:length(data1)){
- if(geom[[i1]] == "geom_point"){
- point.count <- point.count + 1
- if(point.count == 1){
- fin.lg.disp[[1]] <- legend.disp[[point.count + line.count]]
- lg.order[[1]] <- point.count + line.count
- lg.color[[1]] <- color[[i1]] # if color == NULL -> NULL
- lg.dot.shape[[1]] <- dot.shape[[i1]]
- lg.dot.size[[1]] <- dot.size[[i1]]
- lg.dot.border.size[[1]] <- dot.border.size[[i1]]
- lg.dot.border.color[[1]] <- dot.border.color[[i1]] # if dot.border.color == NULL -> NULL
- if(plot == TRUE & fin.lg.disp[[1]] == TRUE & dot.shape[[1]] %in% 0:14 & ((length(dev.list()) > 0 & names(dev.cur()) == "windows") | (length(dev.list()) == 0 & Sys.info()["sysname"] == "Windows"))){ # if any Graph device already open and this device is "windows", or if no Graph device opened yet and we are on windows system -> prevention of alpha legend bug on windows using value 1
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") GRAPHIC DEVICE USED ON A WINDOWS SYSTEM ->\nTRANSPARENCY OF THE DOTS (DOT LAYER NUMBER ", point.count, ") IS INACTIVATED IN THE LEGEND TO PREVENT A WINDOWS DEPENDENT BUG (SEE https://github.com/tidyverse/ggplot2/issues/2452)\nTO OVERCOME THIS ON WINDOWS, USE ANOTHER DEVICE (pdf() FOR INSTANCE)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- lg.alpha[[1]] <- 1 # to avoid a bug on windows: if alpha argument is different from 1 for lines (transparency), then lines are not correctly displayed in the legend when using the R GUI (bug https://github.com/tidyverse/ggplot2/issues/2452). No bug when using a pdf
- }else{
- lg.alpha[[1]] <- alpha[[i1]]
- }
- class.categ <- levels(factor(data1[[i1]][, categ[[i1]]]))
- for(i5 in 1:length(color[[i1]])){ # or length(class.categ). It is the same because already checked that lengths are the same
- tempo.data.frame <- data1[[i1]][data1[[i1]][, categ[[i1]]] == class.categ[i5], ]
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = scatter.kind[[i1]]))(data = tempo.data.frame, mapping = ggplot2::aes_string(x = x[[i1]], y = y[[i1]], fill = categ[[i1]]), shape = dot.shape[[i1]], size = dot.size[[i1]], stroke = dot.border.size[[i1]], color = if(dot.shape[[i1]] %in% 21:24 & ! is.null(dot.border.color)){dot.border.color[[i1]]}else{color[[i1]][i5]}, alpha = alpha[[i1]], show.legend = if(i5 == 1){TRUE}else{FALSE})) # WARNING: a single color allowed for color argument outside aesthetic, but here a single color for border --> loop could be inactivated but kept for commodity # legend.show option do not remove the legend, only the aesthetic of the legend (dot, line, etc.). Used here to avoid multiple layers of legend which corrupt transparency
- coord.names <- c(coord.names, paste0(geom[[i1]], ".", class.categ[i5]))
- }
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_fill_manual(name = if(is.null(legend.name)){NULL}else{legend.name[[i1]]}, values = as.character(color[[i1]]), breaks = class.categ)) # values are the values of fill, breaks reorder the classes according to class.categ in the legend, order argument of guide_legend determines the order of the different aesthetics in the legend (not order of classes). See guide_legend settings of scale_..._manual below
- }
- if(point.count == 2){
- fin.lg.disp[[2]] <- legend.disp[[point.count + line.count]]
- lg.order[[2]] <- point.count + line.count
- lg.color[[2]] <- color[[i1]] # if color == NULL -> NULL
- lg.dot.shape[[2]] <- dot.shape[[i1]]
- lg.dot.size[[2]] <- dot.size[[i1]]
- lg.dot.border.size[[2]] <- dot.border.size[[i1]]
- lg.dot.border.color[[2]] <- dot.border.color[[i1]] # if dot.border.color == NULL -> NULL
- if(plot == TRUE & fin.lg.disp[[2]] == TRUE & dot.shape[[2]] %in% 0:14 & ((length(dev.list()) > 0 & names(dev.cur()) == "windows") | (length(dev.list()) == 0 & Sys.info()["sysname"] == "Windows"))){ # if any Graph device already open and this device is "windows", or if no Graph device opened yet and we are on windows system -> prevention of alpha legend bug on windows using value 1
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") GRAPHIC DEVICE USED ON A WINDOWS SYSTEM ->\nTRANSPARENCY OF THE DOTS (DOT LAYER NUMBER ", point.count, ") IS INACTIVATED IN THE LEGEND TO PREVENT A WINDOWS DEPENDENT BUG (SEE https://github.com/tidyverse/ggplot2/issues/2452)\nTO OVERCOME THIS ON WINDOWS, USE ANOTHER DEVICE (pdf() FOR INSTANCE)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- lg.alpha[[2]] <- 1 # to avoid a bug on windows: if alpha argument is different from 1 for lines (transparency), then lines are not correctly displayed in the legend when using the R GUI (bug https://github.com/tidyverse/ggplot2/issues/2452). No bug when using a pdf
- }else{
- lg.alpha[[2]] <- alpha[[i1]]
- }
- class.categ <- levels(factor(data1[[i1]][, categ[[i1]]]))
- for(i5 in 1:length(color[[i1]])){ # or length(class.categ). It is the same because already checked that lengths are the same
- tempo.data.frame <- data1[[i1]][data1[[i1]][, categ[[i1]]] == class.categ[i5], ]
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = scatter.kind[[i1]]))(data = tempo.data.frame, mapping = ggplot2::aes_string(x = x[[i1]], y = y[[i1]], shape = categ[[i1]]), size = dot.size[[i1]], stroke = dot.border.size[[i1]], fill = color[[i1]][i5], color = if(dot.shape[[i1]] %in% 21:24 & ! is.null(dot.border.color)){dot.border.color[[i1]]}else{color[[i1]][i5]}, alpha = alpha[[i1]], show.legend = FALSE)) # WARNING: a single color allowed for fill argument outside aesthetic, hence the loop # legend.show option do not remove the legend, only the aesthetic of the legend (dot, line, etc.). Used here to avoid multiple layers of legend which corrupt transparency
- coord.names <- c(coord.names, paste0(geom[[i1]], ".", class.categ[i5]))
- }
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_shape_manual(name = if(is.null(legend.name)){NULL}else{legend.name[[i1]]}, values = rep(dot.shape[[i1]], length(color[[i1]])), breaks = class.categ)) # values are the values of shape, breaks reorder the classes according to class.categ in the legend. See guide_legend settings of scale_..._manual below
-
- }
- if(point.count == 3){
- fin.lg.disp[[3]] <- legend.disp[[point.count + line.count]]
- lg.order[[3]] <- point.count + line.count
- lg.color[[3]] <- color[[i1]] # if color == NULL -> NULL
- lg.dot.shape[[3]] <- dot.shape[[i1]]
- lg.dot.size[[3]] <- dot.size[[i1]]
- lg.dot.border.size[[3]] <- dot.border.size[[i1]]
- lg.dot.border.color[[3]] <- dot.border.color[[i1]] # if dot.border.color == NULL -> NULL
- if(plot == TRUE & fin.lg.disp[[3]] == TRUE & dot.shape[[3]] %in% 0:14 & ((length(dev.list()) > 0 & names(dev.cur()) == "windows") | (length(dev.list()) == 0 & Sys.info()["sysname"] == "Windows"))){ # if any Graph device already open and this device is "windows", or if no Graph device opened yet and we are on windows system -> prevention of alpha legend bug on windows using value 1
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") GRAPHIC DEVICE USED ON A WINDOWS SYSTEM ->\nTRANSPARENCY OF THE DOTS (DOT LAYER NUMBER ", point.count, ") IS INACTIVATED IN THE LEGEND TO PREVENT A WINDOWS DEPENDENT BUG (SEE https://github.com/tidyverse/ggplot2/issues/2452)\nTO OVERCOME THIS ON WINDOWS, USE ANOTHER DEVICE (pdf() FOR INSTANCE)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- lg.alpha[[3]] <- 1 # to avoid a bug on windows: if alpha argument is different from 1 for lines (transparency), then lines are not correctly displayed in the legend when using the R GUI (bug https://github.com/tidyverse/ggplot2/issues/2452). No bug when using a pdf
- }else{
- lg.alpha[[3]] <- alpha[[i1]]
- }
- class.categ <- levels(factor(data1[[i1]][, categ[[i1]]]))
- for(i5 in 1:length(color[[i1]])){ # or length(class.categ). It is the same because already checked that lengths are the same
- tempo.data.frame <- data1[[i1]][data1[[i1]][, categ[[i1]]] == class.categ[i5], ]
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = scatter.kind[[i1]]))(data = tempo.data.frame, mapping = ggplot2::aes_string(x = x[[i1]], y = y[[i1]], stroke = categ[[i1]]), shape = dot.shape[[i1]], size = dot.size[[i1]], fill = color[[i1]][i5], stroke = dot.border.size[[i1]], color = if(dot.shape[[i1]] %in% 21:24 & ! is.null(dot.border.color)){dot.border.color[[i1]]}else{color[[i1]][i5]}, alpha = alpha[[i1]], show.legend = FALSE)) # WARNING: a single color allowed for color argument outside aesthetic, hence the loop # legend.show option do not remove the legend, only the aesthetic of the legend (dot, line, etc.). Used here to avoid multiple layers of legend which corrupt transparency
- coord.names <- c(coord.names, paste0(geom[[i1]], ".", class.categ[i5]))
- }
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "stroke", name = if(is.null(legend.name)){NULL}else{legend.name[[i1]]}, values = rep(dot.border.size[[i1]], length(color[[i1]])), breaks = class.categ)) # values are the values of stroke, breaks reorder the classes according to class.categ in the legend. See guide_legend settings of scale_..._manual below
-
- }
- }else{
- line.count <- line.count + 1
- if(line.count == 1){
- fin.lg.disp[[4]] <- legend.disp[[point.count + line.count]]
- lg.order[[4]] <- point.count + line.count
- lg.color[[4]] <- color[[i1]] # if color == NULL -> NULL
- lg.line.size[[4]] <- line.size[[i1]]
- lg.line.type[[4]] <- line.type[[i1]]
- if(plot == TRUE & fin.lg.disp[[4]] == TRUE & ((length(dev.list()) > 0 & names(dev.cur()) == "windows") | (length(dev.list()) == 0 & Sys.info()["sysname"] == "Windows"))){ # if any Graph device already open and this device is "windows", or if no Graph device opened yet and we are on windows system -> prevention of alpha legend bug on windows using value 1
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") GRAPHIC DEVICE USED ON A WINDOWS SYSTEM ->\nTRANSPARENCY OF THE LINES (LINE LAYER NUMBER ", line.count, ") IS INACTIVATED IN THE LEGEND TO PREVENT A WINDOWS DEPENDENT BUG (SEE https://github.com/tidyverse/ggplot2/issues/2452)\nTO OVERCOME THIS ON WINDOWS, USE ANOTHER DEVICE (pdf() FOR INSTANCE)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- lg.alpha[[4]] <- 1 # to avoid a bug on windows: if alpha argument is different from 1 for lines (transparency), then lines are not correctly displayed in the legend when using the R GUI (bug https://github.com/tidyverse/ggplot2/issues/2452). No bug when using a pdf
- }else{
- lg.alpha[[4]] <- alpha[[i1]]
- }
- class.categ <- levels(factor(data1[[i1]][, categ[[i1]]]))
- for(i5 in 1:length(color[[i1]])){ # or length(class.categ). It is the same because already checked that lengths are the same
- tempo.data.frame <- data1[[i1]][data1[[i1]][, categ[[i1]]] == class.categ[i5], ]
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = paste0("ggplot2::", # no CR here te0("ggpl
- ifelse(geom[[i1]] == 'geom_stick', 'geom_segment', geom[[i1]]), # geom_segment because geom_stick converted to geom_segment for plotting
- "(data = tempo.data.frame, mapping = ggplot2::aes(x = ",
- x[[i1]],
- ifelse(geom[[i1]] == 'geom_stick', ", yend = ", ", y = "),
- y[[i1]],
- if(geom[[i1]] == 'geom_stick'){paste0(', xend = ', x[[i1]], ', y = ', y.lim[1])},
- ", linetype = ",
- categ[[i1]],
- "), color = \"",
- color[[i1]][i5],
- "\", size = ",
- line.size[[i1]],
- ifelse(geom[[i1]] == 'geom_path', ', lineend = \"round\"', ''),
- ifelse(geom[[i1]] == 'geom_step', paste0(', direction = \"', geom.step.dir[[i1]], '\"'), ''),
- ", alpha = ",
- alpha[[i1]],
- ", show.legend = ",
- ifelse(i5 == 1, TRUE, FALSE),
- ")"
- )))) # WARNING: a single color allowed for color argument outside aesthetic, hence the loop # legend.show option do not remove the legend, only the aesthetic of the legend (dot, line, etc.). Used here to avoid multiple layers of legend which corrupt transparency
- coord.names <- c(coord.names, paste0(geom[[i1]], ".", class.categ[i5]))
- }
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "linetype", name = if(is.null(legend.name)){NULL}else{legend.name[[i1]]}, values = rep(line.type[[i1]], length(color[[i1]])), breaks = class.categ)) # values are the values of linetype. 1 means solid. Regarding the alpha bug, I have tried different things without success: alpha in guide alone, in geom alone, in both, with different values, breaks reorder the classes according to class.categ in the legend
- }
- if(line.count == 2){
- fin.lg.disp[[5]] <- legend.disp[[point.count + line.count]]
- lg.order[[5]] <- point.count + line.count
- lg.color[[5]] <- color[[i1]] # if color == NULL -> NULL
- lg.line.size[[5]] <- line.size[[i1]]
- lg.line.type[[5]] <- line.type[[i1]]
- if(plot == TRUE & fin.lg.disp[[5]] == TRUE & ((length(dev.list()) > 0 & names(dev.cur()) == "windows") | (length(dev.list()) == 0 & Sys.info()["sysname"] == "Windows"))){ # if any Graph device already open and this device is "windows", or if no Graph device opened yet and we are on windows system -> prevention of alpha legend bug on windows using value 1
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") GRAPHIC DEVICE USED ON A WINDOWS SYSTEM ->\nTRANSPARENCY OF THE LINES (LINE LAYER NUMBER ", line.count, ") IS INACTIVATED IN THE LEGEND TO PREVENT A WINDOWS DEPENDENT BUG (SEE https://github.com/tidyverse/ggplot2/issues/2452)\nTO OVERCOME THIS ON WINDOWS, USE ANOTHER DEVICE (pdf() FOR INSTANCE)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- lg.alpha[[5]] <- 1 # to avoid a bug on windows: if alpha argument is different from 1 for lines (transparency), then lines are not correctly displayed in the legend when using the R GUI (bug https://github.com/tidyverse/ggplot2/issues/2452). No bug when using a pdf
- }else{
- lg.alpha[[5]] <- alpha[[i1]]
- }
- class.categ <- levels(factor(data1[[i1]][, categ[[i1]]]))
- for(i5 in 1:length(color[[i1]])){ # or length(class.categ). It is the same because already checked that lengths are the same
- tempo.data.frame <- data1[[i1]][data1[[i1]][, categ[[i1]]] == class.categ[i5], ]
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = paste0("ggplot2::", # no CR here te0("ggpl
- ifelse(geom[[i1]] == 'geom_stick', 'geom_segment', geom[[i1]]), # geom_segment because geom_stick converted to geom_segment for plotting
- "(data = tempo.data.frame, mapping = ggplot2::aes(x = ",
- x[[i1]],
- ifelse(geom[[i1]] == 'geom_stick', ", yend = ", ", y = "),
- y[[i1]],
- if(geom[[i1]] == 'geom_stick'){paste0(', xend = ', x[[i1]], ', y = ', y.lim[1])},
- ", alpha = ",
- categ[[i1]],
- "), color = \"",
- color[[i1]][i5],
- "\", size = ",
- line.size[[i1]],
- ", linetype = ",
- ifelse(is.numeric(line.type[[i1]]), "", "\""),
- line.type[[i1]],
- ifelse(is.numeric(line.type[[i1]]), "", "\""),
- ifelse(geom[[i1]] == 'geom_path', ', lineend = \"round\"', ''),
- ifelse(geom[[i1]] == 'geom_step', paste0(', direction = \"', geom.step.dir[[i1]], '\"'), ''),
- ", show.legend = FALSE)"
- )))) # WARNING: a single color allowed for color argument outside aesthetic, hence the loop # legend.show option do not remove the legend, only the aesthetic of the legend (dot, line, etc.). Used here to avoid multiple layers of legend which corrupt transparency
- coord.names <- c(coord.names, paste0(geom[[i1]], ".", class.categ[i5]))
- }
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "alpha", name = if(is.null(legend.name)){NULL}else{legend.name[[i1]]}, values = rep(alpha[[i1]], length(color[[i1]])), breaks = class.categ)) # values are the values of linetype. 1 means solid. Regarding the alpha bug, I have tried different things without success: alpha in guide alone, in geom alone, in both, with different values, breaks reorder the classes according to class.categ in the legend
- }
- if(line.count == 3){
- fin.lg.disp[[6]] <- legend.disp[[point.count + line.count]]
- lg.order[[6]] <- point.count + line.count
- lg.color[[6]] <- color[[i1]] # if color == NULL -> NULL
- lg.line.size[[6]] <- line.size[[i1]]
- lg.line.type[[6]] <- line.type[[i1]]
- if(plot == TRUE & fin.lg.disp[[6]] == TRUE & ((length(dev.list()) > 0 & names(dev.cur()) == "windows") | (length(dev.list()) == 0 & Sys.info()["sysname"] == "Windows"))){ # if any Graph device already open and this device is "windows", or if no Graph device opened yet and we are on windows system -> prevention of alpha legend bug on windows using value 1
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") GRAPHIC DEVICE USED ON A WINDOWS SYSTEM ->\nTRANSPARENCY OF THE LINES (LINE LAYER NUMBER ", line.count, ") IS INACTIVATED IN THE LEGEND TO PREVENT A WINDOWS DEPENDENT BUG (SEE https://github.com/tidyverse/ggplot2/issues/2452)\nTO OVERCOME THIS ON WINDOWS, USE ANOTHER DEVICE (pdf() FOR INSTANCE)")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- lg.alpha[[6]] <- 1 # to avoid a bug on windows: if alpha argument is different from 1 for lines (transparency), then lines are not correctly displayed in the legend when using the R GUI (bug https://github.com/tidyverse/ggplot2/issues/2452). No bug when using a pdf
- }else{
- lg.alpha[[6]] <- alpha[[i1]]
- }
- class.categ <- levels(factor(data1[[i1]][, categ[[i1]]]))
- for(i5 in 1:length(color[[i1]])){ # or length(class.categ). It is the same because already checked that lengths are the same
- tempo.data.frame <- data1[[i1]][data1[[i1]][, categ[[i1]]] == class.categ[i5], ]
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = paste0("ggplot2::", # no CR here te0("ggpl
- ifelse(geom[[i1]] == 'geom_stick', 'geom_segment', geom[[i1]]), # geom_segment because geom_stick converted to geom_segment for plotting
- "(data = tempo.data.frame, mapping = ggplot2::aes(x = ",
- x[[i1]],
- ifelse(geom[[i1]] == 'geom_stick', ", yend = ", ", y = "),
- y[[i1]],
- if(geom[[i1]] == 'geom_stick'){paste0(', xend = ', x[[i1]], ', y = ', y.lim[1])},
- ", size = ",
- categ[[i1]],
- "), color = \"",
- color[[i1]][i5],
- "\", linetype = ",
- ifelse(is.numeric(line.type[[i1]]), "", "\""),
- line.type[[i1]],
- ifelse(is.numeric(line.type[[i1]]), "", "\""),
- ifelse(geom[[i1]] == 'geom_path', ', lineend = \"round\"', ''),
- ifelse(geom[[i1]] == 'geom_step', paste0(', direction = \"', geom.step.dir[[i1]], '\"'), ''),
- ", alpha = ",
- alpha[[i1]],
- ", show.legend = FALSE)"
- )))) # WARNING: a single color allowed for color argument outside aesthetic, hence the loop # legend.show option do not remove the legend, only the aesthetic of the legend (dot, line, etc.). Used here to avoid multiple layers of legend which corrupt transparency
- coord.names <- c(coord.names, paste0(geom[[i1]], ".", class.categ[i5]))
- }
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_discrete_manual(aesthetics = "size", name = if(is.null(legend.name)){NULL}else{legend.name[[i1]]}, values = rep(line.size[[i1]], length(color[[i1]])), breaks = class.categ)) # values are the values of linetype. 1 means solid. Regarding the alpha bug, I have tried different things without success: alpha in guide alone, in geom alone, in both, breaks reorder the classes according to class.categ in the legend
- }
- }
- }
- # end loop part
-
-
-
-
- # legend display
- tempo.legend.final <- 'ggplot2::guides(
+# legend display
+tempo.legend.final <- 'ggplot2::guides(
fill = if(fin.lg.disp[[1]] == TRUE){
ggplot2::guide_legend(
order = lg.order[[1]],
@@ -12308,25 +12307,25 @@ FALSE
}
)' # clip = "off" to have secondary ticks outside plot region does not work
if( ! is.null(legend.width)){
- if(any(unlist(legend.disp))){ # means some TRUE
- tempo.graph.info <- suppressMessages(ggplot2::ggplot_build(eval(parse(text = paste0(paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "), ' + ', tempo.legend.final))))) # will be recovered later again, when ylim will be considered
- legend.final <- fun_gg_get_legend(ggplot_built = tempo.graph.info, fun.name = function.name) # get legend
- fin.lg.disp[] <- FALSE # remove all the legends. Must be done even if fin.lg.disp is not appearing in the code thenafter. Otherwise twice the legend
- if(is.null(legend.final) & plot == TRUE){ # even if any(unlist(legend.disp)) is TRUE
- legend.final <- fun_gg_empty_graph() # empty graph instead of legend
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") LEGEND REQUESTED (NON-NULL categ ARGUMENT OR legend.show ARGUMENT SET TO TRUE)\nBUT IT SEEMS THAT THE PLOT HAS NO LEGEND -> EMPTY LEGEND SPACE CREATED BECAUSE OF THE NON-NULL legend.width ARGUMENT\n")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
- }else if(plot == TRUE){ # means all FALSE
- legend.final <- ggplot2::ggplot()+ggplot2::theme_void() # empty graph instead of legend
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") LEGEND REQUESTED (NON-NULL categ ARGUMENT OR legend.show ARGUMENT SET TO TRUE)\nBUT IT SEEMS THAT THE PLOT HAS NO LEGEND -> EMPTY LEGEND SPACE CREATED BECAUSE OF THE NON-NULL legend.width ARGUMENT\n")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
- }
+if(any(unlist(legend.disp))){ # means some TRUE
+tempo.graph.info <- suppressMessages(ggplot2::ggplot_build(eval(parse(text = paste0(paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + "), ' + ', tempo.legend.final))))) # will be recovered later again, when ylim will be considered
+legend.final <- fun_gg_get_legend(ggplot_built = tempo.graph.info, fun.name = function.name) # get legend
+fin.lg.disp[] <- FALSE # remove all the legends. Must be done even if fin.lg.disp is not appearing in the code thenafter. Otherwise twice the legend
+if(is.null(legend.final) & plot == TRUE){ # even if any(unlist(legend.disp)) is TRUE
+legend.final <- fun_gg_empty_graph() # empty graph instead of legend
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") LEGEND REQUESTED (NON-NULL categ ARGUMENT OR legend.show ARGUMENT SET TO TRUE)\nBUT IT SEEMS THAT THE PLOT HAS NO LEGEND -> EMPTY LEGEND SPACE CREATED BECAUSE OF THE NON-NULL legend.width ARGUMENT\n")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
+}else if(plot == TRUE){ # means all FALSE
+legend.final <- ggplot2::ggplot()+ggplot2::theme_void() # empty graph instead of legend
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") LEGEND REQUESTED (NON-NULL categ ARGUMENT OR legend.show ARGUMENT SET TO TRUE)\nBUT IT SEEMS THAT THE PLOT HAS NO LEGEND -> EMPTY LEGEND SPACE CREATED BECAUSE OF THE NON-NULL legend.width ARGUMENT\n")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+}
}
if( ! any(unlist(legend.disp))){
- fin.lg.disp[] <- FALSE # remove all the legends. Must be done even if fin.lg.disp is not appearing in the code thenafter. Otherwise twice the legend
+fin.lg.disp[] <- FALSE # remove all the legends. Must be done even if fin.lg.disp is not appearing in the code thenafter. Otherwise twice the legend
}
assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(text = tempo.legend.final)))
# end legend display
@@ -12339,119 +12338,119 @@ assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), eval(parse(t
tempo.coord <- suppressMessages(ggplot2::ggplot_build(eval(parse(text = paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + ", ' + ggplot2::scale_x_continuous(expand = c(0, 0), limits = sort(x.lim), oob = scales::rescale_none) + ggplot2::scale_y_continuous(expand = c(0, 0), limits = sort(y.lim), oob = scales::rescale_none)'))))$layout$panel_params[[1]]) # here I do not need the x-axis and y-axis orientation, I just need the number of main ticks
# x.second.tick.positions # coordinates of secondary ticks (only if x.second.tick.nb argument is non-null or if x.log argument is different from "no")
if(x.log != "no"){ # integer main ticks for log2 and log10
- tempo.scale <- (as.integer(min(x.lim, na.rm = TRUE)) - 1):(as.integer(max(x.lim, na.rm = TRUE)) + 1)
-}else{
- tempo <- if(is.null(attributes(tempo.coord$x$breaks))){tempo.coord$x$breaks}else{unlist(attributes(tempo.coord$x$breaks))}
- if(all(is.na(tempo))){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nONLY NA IN tempo.coord$x$breaks")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- if(length(unique(x.lim)) <= 1){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nIT SEEMS THAT X-AXIS VALUES HAVE A NULL RANGE: ", paste(x.lim, collapse = " "), "\nPLEASE, USE THE x.lim ARGUMENT WITH 2 DIFFERENT VALUES TO SOLVE THIS")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- tempo.scale <- fun_scale(lim = x.lim, n = ifelse(is.null(x.tick.nb), length(tempo[ ! is.na(tempo)]), x.tick.nb)) # in ggplot 3.3.0, tempo.coord$x.major_source replaced by tempo.coord$x$breaks. If fact: n = ifelse(is.null(x.tick.nb), length(tempo[ ! is.na(tempo)]), x.tick.nb)) replaced by n = ifelse(is.null(x.tick.nb), 4, x.tick.nb))
- }
+tempo.scale <- (as.integer(min(x.lim, na.rm = TRUE)) - 1):(as.integer(max(x.lim, na.rm = TRUE)) + 1)
+}else{
+tempo <- if(is.null(attributes(tempo.coord$x$breaks))){tempo.coord$x$breaks}else{unlist(attributes(tempo.coord$x$breaks))}
+if(all(is.na(tempo))){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nONLY NA IN tempo.coord$x$breaks")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+if(length(unique(x.lim)) <= 1){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nIT SEEMS THAT X-AXIS VALUES HAVE A NULL RANGE: ", paste(x.lim, collapse = " "), "\nPLEASE, USE THE x.lim ARGUMENT WITH 2 DIFFERENT VALUES TO SOLVE THIS")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+tempo.scale <- fun_scale(lim = x.lim, n = ifelse(is.null(x.tick.nb), length(tempo[ ! is.na(tempo)]), x.tick.nb)) # in ggplot 3.3.0, tempo.coord$x.major_source replaced by tempo.coord$x$breaks. If fact: n = ifelse(is.null(x.tick.nb), length(tempo[ ! is.na(tempo)]), x.tick.nb)) replaced by n = ifelse(is.null(x.tick.nb), 4, x.tick.nb))
+}
}
x.second.tick.values <- NULL
x.second.tick.pos <- NULL
if(x.log != "no"){
- tempo <- fun_inter_ticks(lim = x.lim, log = x.log)
- x.second.tick.values <- tempo$values
- x.second.tick.pos <- tempo$coordinates
- # if(vertical == TRUE){ # do not remove in case the bug is fixed
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::annotate(
- geom = "segment", x = x.second.tick.pos,
- xend = x.second.tick.pos,
- y = if(diff(y.lim) > 0){tempo.coord$y.range[1]}else{tempo.coord$y.range[2]},
- yend = if(diff(y.lim) > 0){tempo.coord$y.range[1] + abs(diff(tempo.coord$y.range)) / 80}else{tempo.coord$y.range[2] - abs(diff(tempo.coord$y.range)) / 80}
- ))
- # }else{ # not working because of the ggplot2 bug
- # assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::annotate(geom = "segment", y = x.second.tick.pos, yend = x.second.tick.pos, x = tempo.coord$x.range[1], xend = tempo.coord$x.range[1] + diff(tempo.coord$x.range) / 80))
- # }
- coord.names <- c(coord.names, "x.second.tick.positions")
+tempo <- fun_inter_ticks(lim = x.lim, log = x.log)
+x.second.tick.values <- tempo$values
+x.second.tick.pos <- tempo$coordinates
+# if(vertical == TRUE){ # do not remove in case the bug is fixed
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::annotate(
+geom = "segment", x = x.second.tick.pos,
+xend = x.second.tick.pos,
+y = if(diff(y.lim) > 0){tempo.coord$y.range[1]}else{tempo.coord$y.range[2]},
+yend = if(diff(y.lim) > 0){tempo.coord$y.range[1] + abs(diff(tempo.coord$y.range)) / 80}else{tempo.coord$y.range[2] - abs(diff(tempo.coord$y.range)) / 80}
+))
+# }else{ # not working because of the ggplot2 bug
+# assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::annotate(geom = "segment", y = x.second.tick.pos, yend = x.second.tick.pos, x = tempo.coord$x.range[1], xend = tempo.coord$x.range[1] + diff(tempo.coord$x.range) / 80))
+# }
+coord.names <- c(coord.names, "x.second.tick.positions")
}else if(( ! is.null(x.second.tick.nb)) & x.log == "no"){
- # if(x.second.tick.nb > 0){ #inactivated because already checked before
- tempo <- fun_inter_ticks(lim = x.lim, log = x.log, breaks = tempo.scale, n = x.second.tick.nb)
- x.second.tick.values <- tempo$values
- x.second.tick.pos <- tempo$coordinates
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::annotate(
- geom = "segment",
- x = x.second.tick.pos,
- xend = x.second.tick.pos,
- y = if(diff(y.lim) > 0){tempo.coord$y.range[1]}else{tempo.coord$y.range[2]},
- yend = if(diff(y.lim) > 0){tempo.coord$y.range[1] + abs(diff(tempo.coord$y.range)) / 80}else{tempo.coord$y.range[2] - abs(diff(tempo.coord$y.range)) / 80}
- ))
- coord.names <- c(coord.names, "x.second.tick.positions")
+# if(x.second.tick.nb > 0){ #inactivated because already checked before
+tempo <- fun_inter_ticks(lim = x.lim, log = x.log, breaks = tempo.scale, n = x.second.tick.nb)
+x.second.tick.values <- tempo$values
+x.second.tick.pos <- tempo$coordinates
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::annotate(
+geom = "segment",
+x = x.second.tick.pos,
+xend = x.second.tick.pos,
+y = if(diff(y.lim) > 0){tempo.coord$y.range[1]}else{tempo.coord$y.range[2]},
+yend = if(diff(y.lim) > 0){tempo.coord$y.range[1] + abs(diff(tempo.coord$y.range)) / 80}else{tempo.coord$y.range[2] - abs(diff(tempo.coord$y.range)) / 80}
+))
+coord.names <- c(coord.names, "x.second.tick.positions")
}
# for the ggplot2 bug with x.log, this does not work: eval(parse(text = ifelse(vertical == FALSE & x.log == "log10", "ggplot2::scale_x_continuous", "ggplot2::scale_x_continuous")))
assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_x_continuous(
- breaks = tempo.scale,
- minor_breaks = x.second.tick.pos,
- labels = if(x.log == "log10"){scales::trans_format("identity", scales::math_format(10^.x))}else if(x.log == "log2"){scales::trans_format("identity", scales::math_format(2^.x))}else if(x.log == "no"){ggplot2::waiver()}else{tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 10") ; stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)},
- expand = c(0, 0), # remove space after after axis limits
- limits = sort(x.lim), # NA indicate that limits must correspond to data limits but xlim() already used
- oob = scales::rescale_none,
- trans = ifelse(diff(x.lim) < 0, "reverse", "identity") # equivalent to ggplot2::scale_x_reverse() but create the problem of x-axis label disappearance with x.lim decreasing. Thus, do not use. Use xlim() below and after this
+breaks = tempo.scale,
+minor_breaks = x.second.tick.pos,
+labels = if(x.log == "log10"){scales::trans_format("identity", scales::math_format(10^.x))}else if(x.log == "log2"){scales::trans_format("identity", scales::math_format(2^.x))}else if(x.log == "no"){ggplot2::waiver()}else{tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 10") ; stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)},
+expand = c(0, 0), # remove space after after axis limits
+limits = sort(x.lim), # NA indicate that limits must correspond to data limits but xlim() already used
+oob = scales::rescale_none,
+trans = ifelse(diff(x.lim) < 0, "reverse", "identity") # equivalent to ggplot2::scale_x_reverse() but create the problem of x-axis label disappearance with x.lim decreasing. Thus, do not use. Use xlim() below and after this
))
# end x.second.tick.positions
# y.second.tick.positions # coordinates of secondary ticks (only if y.second.tick.nb argument is non-null or if y.log argument is different from "no")
if(y.log != "no"){ # integer main ticks for log2 and log10
- tempo.scale <- (as.integer(min(y.lim, na.rm = TRUE)) - 1):(as.integer(max(y.lim, na.rm = TRUE)) + 1)
-}else{
- tempo <- if(is.null(attributes(tempo.coord$y$breaks))){tempo.coord$y$breaks}else{unlist(attributes(tempo.coord$y$breaks))}
- if(all(is.na(tempo))){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nONLY NA IN tempo.coord$y$breaks")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }
- if(length(unique(y.lim)) <= 1){
- tempo.cat <- paste0("ERROR IN ", function.name, "\nIT SEEMS THAT Y-AXIS VALUES HAVE A NULL RANGE: ", paste(y.lim, collapse = " "), "\nPLEASE, USE THE y.lim ARGUMENT WITH 2 DIFFERENT VALUES TO SOLVE THIS")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- tempo.scale <- fun_scale(lim = y.lim, n = ifelse(is.null(y.tick.nb), length(tempo[ ! is.na(tempo)]), y.tick.nb)) # in ggplot 3.3.0, tempo.coord$y.major_source replaced by tempo.coord$y$breaks. If fact: n = ifelse(is.null(y.tick.nb), length(tempo[ ! is.na(tempo)]), y.tick.nb)) replaced by n = ifelse(is.null(y.tick.nb), 4, y.tick.nb))
- }
+tempo.scale <- (as.integer(min(y.lim, na.rm = TRUE)) - 1):(as.integer(max(y.lim, na.rm = TRUE)) + 1)
+}else{
+tempo <- if(is.null(attributes(tempo.coord$y$breaks))){tempo.coord$y$breaks}else{unlist(attributes(tempo.coord$y$breaks))}
+if(all(is.na(tempo))){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nONLY NA IN tempo.coord$y$breaks")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}
+if(length(unique(y.lim)) <= 1){
+tempo.cat <- paste0("ERROR IN ", function.name, "\nIT SEEMS THAT Y-AXIS VALUES HAVE A NULL RANGE: ", paste(y.lim, collapse = " "), "\nPLEASE, USE THE y.lim ARGUMENT WITH 2 DIFFERENT VALUES TO SOLVE THIS")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+tempo.scale <- fun_scale(lim = y.lim, n = ifelse(is.null(y.tick.nb), length(tempo[ ! is.na(tempo)]), y.tick.nb)) # in ggplot 3.3.0, tempo.coord$y.major_source replaced by tempo.coord$y$breaks. If fact: n = ifelse(is.null(y.tick.nb), length(tempo[ ! is.na(tempo)]), y.tick.nb)) replaced by n = ifelse(is.null(y.tick.nb), 4, y.tick.nb))
+}
}
y.second.tick.values <- NULL
y.second.tick.pos <- NULL
if(y.log != "no"){
- tempo <- fun_inter_ticks(lim = y.lim, log = y.log)
- y.second.tick.values <- tempo$values
- y.second.tick.pos <- tempo$coordinates
- # if(vertical == TRUE){ # do not remove in case the bug is fixed
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::annotate(
- geom = "segment",
- y = y.second.tick.pos,
- yend = y.second.tick.pos,
- x = if(diff(x.lim) > 0){tempo.coord$x.range[1]}else{tempo.coord$x.range[2]},
- xend = if(diff(x.lim) > 0){tempo.coord$x.range[1] + abs(diff(tempo.coord$x.range)) / 80}else{tempo.coord$x.range[2] - abs(diff(tempo.coord$x.range)) / 80}
- ))
- # }else{ # not working because of the ggplot2 bug
- # assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::annotate(geom = "segment", x = y.second.tick.pos, xend = y.second.tick.pos, y = tempo.coord$y.range[1], yend = tempo.coord$y.range[1] + diff(tempo.coord$y.range) / 80))
- # }
- coord.names <- c(coord.names, "y.second.tick.positions")
+tempo <- fun_inter_ticks(lim = y.lim, log = y.log)
+y.second.tick.values <- tempo$values
+y.second.tick.pos <- tempo$coordinates
+# if(vertical == TRUE){ # do not remove in case the bug is fixed
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::annotate(
+geom = "segment",
+y = y.second.tick.pos,
+yend = y.second.tick.pos,
+x = if(diff(x.lim) > 0){tempo.coord$x.range[1]}else{tempo.coord$x.range[2]},
+xend = if(diff(x.lim) > 0){tempo.coord$x.range[1] + abs(diff(tempo.coord$x.range)) / 80}else{tempo.coord$x.range[2] - abs(diff(tempo.coord$x.range)) / 80}
+))
+# }else{ # not working because of the ggplot2 bug
+# assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::annotate(geom = "segment", x = y.second.tick.pos, xend = y.second.tick.pos, y = tempo.coord$y.range[1], yend = tempo.coord$y.range[1] + diff(tempo.coord$y.range) / 80))
+# }
+coord.names <- c(coord.names, "y.second.tick.positions")
}else if(( ! is.null(y.second.tick.nb)) & y.log == "no"){
- # if(y.second.tick.nb > 0){ #inactivated because already checked before
- tempo <- fun_inter_ticks(lim = y.lim, log = y.log, breaks = tempo.scale, n = y.second.tick.nb)
- y.second.tick.values <- tempo$values
- y.second.tick.pos <- tempo$coordinates
- assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::annotate(
- geom = "segment",
- y = y.second.tick.pos,
- yend = y.second.tick.pos,
- x = if(diff(x.lim) > 0){tempo.coord$x.range[1]}else{tempo.coord$x.range[2]},
- xend = if(diff(x.lim) > 0){tempo.coord$x.range[1] + abs(diff(tempo.coord$x.range)) / 80}else{tempo.coord$x.range[2] - abs(diff(tempo.coord$x.range)) / 80}
- ))
- coord.names <- c(coord.names, "y.second.tick.positions")
+# if(y.second.tick.nb > 0){ #inactivated because already checked before
+tempo <- fun_inter_ticks(lim = y.lim, log = y.log, breaks = tempo.scale, n = y.second.tick.nb)
+y.second.tick.values <- tempo$values
+y.second.tick.pos <- tempo$coordinates
+assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::annotate(
+geom = "segment",
+y = y.second.tick.pos,
+yend = y.second.tick.pos,
+x = if(diff(x.lim) > 0){tempo.coord$x.range[1]}else{tempo.coord$x.range[2]},
+xend = if(diff(x.lim) > 0){tempo.coord$x.range[1] + abs(diff(tempo.coord$x.range)) / 80}else{tempo.coord$x.range[2] - abs(diff(tempo.coord$x.range)) / 80}
+))
+coord.names <- c(coord.names, "y.second.tick.positions")
}
# for the ggplot2 bug with y.log, this does not work: eval(parse(text = ifelse(vertical == FALSE & y.log == "log10", "ggplot2::scale_x_continuous", "ggplot2::scale_y_continuous")))
assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::scale_y_continuous(
- breaks = tempo.scale,
- minor_breaks = y.second.tick.pos,
- labels = if(y.log == "log10"){scales::trans_format("identity", scales::math_format(10^.x))}else if(y.log == "log2"){scales::trans_format("identity", scales::math_format(2^.x))}else if(y.log == "no"){ggplot2::waiver()}else{tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 10") ; stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)},
- expand = c(0, 0), # remove space after axis limits
- limits = sort(y.lim), # NA indicate that limits must correspond to data limits but ylim() already used
- oob = scales::rescale_none,
- trans = ifelse(diff(y.lim) < 0, "reverse", "identity") # equivalent to ggplot2::scale_y_reverse() but create the problem of y-axis label disappearance with y.lim decreasing. Thus, do not use. Use ylim() below and after this
+breaks = tempo.scale,
+minor_breaks = y.second.tick.pos,
+labels = if(y.log == "log10"){scales::trans_format("identity", scales::math_format(10^.x))}else if(y.log == "log2"){scales::trans_format("identity", scales::math_format(2^.x))}else if(y.log == "no"){ggplot2::waiver()}else{tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, "\nCODE INCONSISTENCY 10") ; stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)},
+expand = c(0, 0), # remove space after axis limits
+limits = sort(y.lim), # NA indicate that limits must correspond to data limits but ylim() already used
+oob = scales::rescale_none,
+trans = ifelse(diff(y.lim) < 0, "reverse", "identity") # equivalent to ggplot2::scale_y_reverse() but create the problem of y-axis label disappearance with y.lim decreasing. Thus, do not use. Use ylim() below and after this
))
# end y.second.tick.positions
assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::coord_cartesian(xlim = x.lim, ylim = y.lim)) # clip = "off" to have secondary ticks outside plot region. The problem is that points out of bounds are also drawn outside the plot region. Thus, I cannot use it # at that stage, x.lim and y.lim not NULL anymore
@@ -12464,15 +12463,15 @@ assign(paste0(tempo.gg.name, tempo.gg.count <- tempo.gg.count + 1), ggplot2::coo
fin.plot <- eval(parse(text = paste(paste0(tempo.gg.name, 1:tempo.gg.count), collapse = " + ")))
grob.save <- NULL
if(plot == TRUE){
- if( ! is.null(legend.width)){ # any(unlist(legend.disp)) == TRUE removed to have empty legend space # not & any(unlist(fin.lg.disp)) == TRUE here because converted to FALSE
- grob.save <- suppressMessages(suppressWarnings(gridExtra::grid.arrange(fin.plot, legend.final, ncol=2, widths=c(1, legend.width))))
- }else{
- grob.save <- suppressMessages(suppressWarnings(print(fin.plot)))
- }
+if( ! is.null(legend.width)){ # any(unlist(legend.disp)) == TRUE removed to have empty legend space # not & any(unlist(fin.lg.disp)) == TRUE here because converted to FALSE
+grob.save <- suppressMessages(suppressWarnings(gridExtra::grid.arrange(fin.plot, legend.final, ncol=2, widths=c(1, legend.width))))
+}else{
+grob.save <- suppressMessages(suppressWarnings(print(fin.plot)))
+}
}else{
- warn.count <- warn.count + 1
- tempo.warn <- paste0("(", warn.count,") PLOT NOT SHOWN AS REQUESTED")
- warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
+warn.count <- warn.count + 1
+tempo.warn <- paste0("(", warn.count,") PLOT NOT SHOWN AS REQUESTED")
+warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
}
# end drawing
@@ -12480,50 +12479,50 @@ if(plot == TRUE){
# outputs
if(warn.print == TRUE & ! is.null(warn)){
- options(warning.length = 8170)
- on.exit(warning(paste0("FROM ", function.name, ":\n\n", warn), call. = FALSE))
- on.exit(exp = options(warning.length = ini.warning.length), add = TRUE)
+options(warning.length = 8170)
+on.exit(warning(paste0("FROM ", function.name, ":\n\n", warn), call. = FALSE))
+on.exit(exp = options(warning.length = ini.warning.length), add = TRUE)
}
if(return == TRUE){
- output <- suppressMessages(ggplot2::ggplot_build(fin.plot))
- # output$data <- output$data[-1] # yes for boxplot but not for scatter # remove the first data because corresponds to the initial empty boxplot
- if(length(output$data) != length(coord.names)){
- tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, ": length(output$data) AND length(coord.names) MUST BE IDENTICAL. CODE HAS TO BE MODIFIED")
- stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
- }else{
- names(output$data) <- coord.names
- }
- if(is.null(unlist(removed.row.nb))){
- removed.row.nb <- NULL
- removed.rows <- NULL
- }else{
- for(i3 in 1:length(data1)){
- if( ! is.null(removed.row.nb[[i3]])){
- removed.row.nb[[i3]] <- sort(removed.row.nb[[i3]])
- removed.rows[[i3]] <- data1.ini[[i3]][removed.row.nb[[i3]], ]
- }
- }
- }
- tempo <- output$layout$panel_params[[1]]
- output <- list(
- data = data1,
- removed.row.nb = removed.row.nb,
- removed.rows = removed.rows,
- plot = c(output$data, x.second.tick.values = list(x.second.tick.values), y.second.tick.values = list(y.second.tick.values)),
- panel = facet.categ,
- axes = list(
- x.range = tempo$x.range,
- x.labels = if(is.null(attributes(tempo$x$breaks))){tempo$x$breaks}else{tempo$x$scale$get_labels()}, # is.null(attributes(tempo$x$breaks)) test if it is number (TRUE) or character (FALSE)
- x.positions = if(is.null(attributes(tempo$x$breaks))){tempo$x$breaks}else{unlist(attributes(tempo$x$breaks))},
- y.range = tempo$y.range,
- y.labels = if(is.null(attributes(tempo$y$breaks))){tempo$y$breaks}else{tempo$y$scale$get_labels()},
- y.positions = if(is.null(attributes(tempo$y$breaks))){tempo$y$breaks}else{unlist(attributes(tempo$y$breaks))}
- ),
- warn = paste0("\n", warn, "\n\n"),
- ggplot = if(return.ggplot == TRUE){fin.plot}else{NULL}, # fin.plot plots the graph if return == TRUE
- gtable = if(return.gtable == TRUE){grob.save}else{NULL} #
- )
- return(output) # this plots the graph if return.ggplot is TRUE and if no assignment
+output <- suppressMessages(ggplot2::ggplot_build(fin.plot))
+# output$data <- output$data[-1] # yes for boxplot but not for scatter # remove the first data because corresponds to the initial empty boxplot
+if(length(output$data) != length(coord.names)){
+tempo.cat <- paste0("INTERNAL CODE ERROR IN ", function.name, ": length(output$data) AND length(coord.names) MUST BE IDENTICAL. CODE HAS TO BE MODIFIED")
+stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n", ifelse(is.null(warn), "", paste0("IN ADDITION\nWARNING", ifelse(warn.count > 1, "S", ""), ":\n\n", warn))), call. = FALSE)
+}else{
+names(output$data) <- coord.names
+}
+if(is.null(unlist(removed.row.nb))){
+removed.row.nb <- NULL
+removed.rows <- NULL
+}else{
+for(i3 in 1:length(data1)){
+if( ! is.null(removed.row.nb[[i3]])){
+removed.row.nb[[i3]] <- sort(removed.row.nb[[i3]])
+removed.rows[[i3]] <- data1.ini[[i3]][removed.row.nb[[i3]], ]
+}
+}
+}
+tempo <- output$layout$panel_params[[1]]
+output <- list(
+data = data1,
+removed.row.nb = removed.row.nb,
+removed.rows = removed.rows,
+plot = c(output$data, x.second.tick.values = list(x.second.tick.values), y.second.tick.values = list(y.second.tick.values)),
+panel = facet.categ,
+axes = list(
+x.range = tempo$x.range,
+x.labels = if(is.null(attributes(tempo$x$breaks))){tempo$x$breaks}else{tempo$x$scale$get_labels()}, # is.null(attributes(tempo$x$breaks)) test if it is number (TRUE) or character (FALSE)
+x.positions = if(is.null(attributes(tempo$x$breaks))){tempo$x$breaks}else{unlist(attributes(tempo$x$breaks))},
+y.range = tempo$y.range,
+y.labels = if(is.null(attributes(tempo$y$breaks))){tempo$y$breaks}else{tempo$y$scale$get_labels()},
+y.positions = if(is.null(attributes(tempo$y$breaks))){tempo$y$breaks}else{unlist(attributes(tempo$y$breaks))}
+),
+warn = paste0("\n", warn, "\n\n"),
+ggplot = if(return.ggplot == TRUE){fin.plot}else{NULL}, # fin.plot plots the graph if return == TRUE
+gtable = if(return.gtable == TRUE){grob.save}else{NULL} #
+)
+return(output) # this plots the graph if return.ggplot is TRUE and if no assignment
}
# end outputs
# end main code
diff --git a/cute_little_R_functions.docx b/cute_little_R_functions.docx
index db3bbcf49ae32fdab9130caec81ce630b6637220..1cd62281682a2335ac3446b3990c81fc74353138 100644
GIT binary patch
delta 376284
zcmYhhQ*b8G(zYGjd}7;}*tTuk_7h`bb7I?eGO=yj&cy%T`+E=mlU1v$y1M${uI^ra
zt@~K~+a&z@1z0#i$#R@ul8_*4flXVmd9YB5m40~!Sjj=?7QiusA*gj*mF7r(wpkll
zCYsG(<?^`|mO)g%70XA({)P7Mz!D=zO}?%Z8qcsX(oHhaQRr!mgf?gmyPJnx{*J64
zc0WGc1m%*O>30A**yM_jz!ePA5bjx{IVVjhIkh-vZG7UAO;0T099tOFB?o^oLRH%|
zceB_jKtOecf&t=4z{@cwP=U?>qApRL9iXs@YLIC33%yT<5cGw^2`$fPS)ZVAI<qLk
zk;+pM;I)Ce7#3Cl2D7uU04vz=T2j$?HaA2ZpAZBFzKEv5x$w=)AL>F}{vR4DBIXP2
z-b<Lu0B7m@qzv|afkUuXDz>#Z=wkUH?9bN7!lz{JVxR`|q(@}4t<=V0v42c)kMKg;
zPP;bgHCoaOv{V-cmHn(gN#C5ok7jwV6#j;3PMdtyXghsJ9MhL(LT`H6t!eQsOB8*<
z>ICKa2f6E$y{@UX_3AEh4GjJW@-*?J>I1CG0~bXeDTQM~k~k%n+rNzTtT4w>FP1Ag
zc7*hXHz3336movdM$vI-3*4Ix2Yp;W5r@&k0&_Bdklzs@JXEe|#%TxPZf$=_oaBc-
zJB_~)DBoX5Cpn%;o*QXqD=&HBp{Dd#7Y#DW!xEE3{&e$MR*CqyCQC@fhw~*4VI87v
zffN=$H%MBRE1LeTuc1IOGI}=QV)kkXs7_`_9iSFL40`Obn_mGMJTCB%X?Ji^DTfIQ
zLyQ0ii##txtdD)(w3`O)OCyR9TtQRp$>Jkol*`eAlq<3=WL;=eEoUgHbu^s>t-V+_
z4+nn;$tR90D$L|D<&!1971OEY&o06=c|9DTE)=hyMXqAve!7Jh!dH&bKXkKyh^&0!
z0l?bGeQ}Bx`Z?Za@Ud;2C&CW28FklW9F6#m#NTmlJ?#3X2xnaSnk*{7WVfDUfRg_i
z?(ab###NgpBQI`-&sQMokVkA8_N1jUzs@{ga|*Ix1B9We*x7%Msc*_RpGnjLpZi~{
zR<Z~Ezt9yDXV`uK1^Je2v#g{mSnjc%z^haVoi;wwG|$9|^imfDf<O5Wa4K(56@HTi
zMLx<Mo3kJhs=JmwlHc-q&WxvO1ehMAN(Ra8pE%`#z?c>m54Yf!`Q9Gs6)b`YnLAYG
zykkDP!R1V@DvX)cgp|Kv{+pHyqMJXaAmdib72m}Crgmps{Ht(Mvv|d)a(~0Yfc$L?
zcuf}5{C><j!GE;th`JLQTQ(ssO2X;CRxVlQ$`k7}hgShyJOitM$v)}ck^Xs16wc~j
zpP4<vD9T>DuDqYEj{IyN_SvD|XDsg!3cy6<c1zZ$J%ikOmO1C(dKQXnDA=yXV<jPf
zn9Vi@HuzrB0_ji@YF{8RzgW8o(4G|x^b*w(+3$Q8u_j!TDb2K#N$O}4%^T8`aWv{v
zYjmmw(9oM2`8<(ECPPKGA}AV%JbB06!qYrNED6(ix^NX06mxjx9c<&o(;}a+60vNA
zV5XOd`)QIX%n5E&uVxixg~XzrH=d7O3m#26HWlGD(tAy6%QU+I$cd~5_NcFn><wc!
zv`)UUfEOc96Q}qEw1+?N_%|`Ao@+yGPzg*d12VY=mNRJ0lJ#N}qv#GUdM!J4kd^ew
zMqae*<yWxq5giDcB?+}RpEH!I!ROjuGbH^b&geAT-w432hQdOOU6zh3@I$ahD?Two
zrZ}aK0nP2Cw6BOZmitq{IZKXWLwEWR)NBD(UOA+}<C@7|F1@`ltjKKPLW$6mUJrkR
zNPljZ2X&Aqo47|96zaKgl{v;w0B!B09Bny)-hvRnECLthS6cDMR!)5=)z}mM!uwgY
z$4U2aO7D+t%06;PFvdN4Gz2!VWRpA07gP_AJn0LfNWhB2@S^JgJ&CV33$8<3NfKYi
z+(=_63E46o-SL}qt0UW3ik*|%l=pYY9UZw#m5IVO>-7cE-BI%ihHu%xD_3LwXcxxy
z@M)TI9YSc@YU>d8RJI+e_Z(YIOV?fa*6w62!OK$XKqZ|%97Pp`B_Bm~q%LII2R8{+
zo@ubOI*I?T9m~B3iaME4!X3Nd!s(gl1nibN4ziLw`i>2@jjd0*cwiR6c!o47?{EJe
z)ZLZ<>^I-Uz`lhNZk^{teL8deEgyvCVAx01=UJAQ+d{VO*F~a}Bb;~V5qw{=7$`56
zE(0%<r!bkAM^c@SAQCyy&hlbrMvAiR8g4osYll5}`{*YEhSP~vgRk!Oi<Opx)-5oK
z!D;nA3Ag0T2(!ARS1EHWEIc%avKS>dhn|ygO3MNuIBZS1j8QR01ZQ&b9JIP239vrv
zwIgC}(jVx8D~{Fo?4{3!{Xcoxz_#ic73To!zoh2?C0)|NBXhN3Fg#ylP#l}J#}I;i
zJPUqp<A(*n1;lA%!uyVu=^xB)`$f*F{SVNU5tv%(6#?;;9b#KknAHodk9q&tQ;jhY
z^9Tmg9}&We`=$FWP|Ys+bojhwO2Q9QBkw7(Z-cBpmL%KYzDX@epX5&lp3u)8v~Q$e
zn+W%wdaK{qgC&7)W{z{v*p;}>c^yV`oKliFN}B<|p}E@QD16bbgt0c+Bfv%r59uu{
z?voYPg^;8eVQXY^yP5qYIsb)V0d6na%ZL$^+3j^BCsv&<fv@S<VB8!>c&7ZEhxn*l
z3ZSt*H6d&xE#ypvjtKQzJRWziE8?ny-gkw0&o+r)`CPuN6?Wmhd_i~9#r&6l%1(ci
zkM<bw15nxiX$vmZdDFo!GEM}30fD6k?r-s#W;rkU?R1d=ktJVl9Lb(YP(U6%l}O0~
zHBhkWP~_6h*JBESyYS(0P5{4I6S?ER5=5Fuszrf5PA2Zk)$+KF#bW1aGyyhyzbLWy
z%8bnHWe1;!+Cj|4*taLXiwDsi^}u%TKza{klK%vNatVK`6NuW+(KZrPNogG5MbNFg
z*WLPP$+>xt4DSuYjts-Ovz^%mmkU(WnmQf)Rx=N{rh&Ki?wUfG!SRI95#ybr!uY)?
zY%1)2ve7_cm93?)2lxE?_c^y8VKB#X(v?6dEO9aY#4W~BOktO?%#`o2lIx4%taBY$
z<gs2|@fp|M!|rG*c1Z4c37C&uiN4liYQ?YPw(j|N;*6MqpLjx7K9mpQ8n9<_l+3@+
z^9q0c45R?X$9)GBH1ai=l<z|{<eEU0x~*jIH6hat7mAjn0QE>-07~XiZu`A^4hmGH
zvbmCSxO2?d$<4OyA^CzZQk#QE-bBEB2Lwj3iycOHNA7kf`Z6l`fTxVxNBuU^;y<93
zFIvoZ_=9}2&v6Fjr%n?BNx6W1t>2{6DZnF4+#6AorJj6QaFe^}c!b(V^@z;!P^Q@<
zfECyHk#-G;kGo3spBTllLG=!;_AH6p@pxdgKAK{j*a*4NniaIrUy>RZnSEd~OLsyq
znBT;&tsu*LSl=Oo@GC**<-6qcVQ}N%`ybBFhU0v|x6u{^Nu4|+>H~1ydVNc%5B^$G
zsXBZ!3r5T<4-;XR<_A@nx9^ElK?c)Xqi3nT##%~)3xnk)k)JU)tt!<g2gthi?#bC^
z4Mc~NKKbwtcy?N}(%;?LDg)Nw^6W_jXdq{JiZRTSkRLA<=;fr)X{MUogN@qHJ9p8h
zDAr!c#&52jY{T(sV3Y|JwH(d6=lLx8b|RN!-zn-)0x|ej3hGUXuHj|&@8_GDB8fZZ
zv1!XjsY+~=zn7jIBip0z9a9wONK#ZKg{Su4uo5exC7c52rURA=UVzz-v62qr*LHB$
z$>6AQm}mnLkhl?6Gu(!DX|!dNboHty-8r32r`>meyo|?v*!CO&ctVBYJMC|PuvcSo
zX6dk}s3rDLX)jO+thbAYn8TGv_%*P;e~c(_+{))VY#i2b1h&t_y7t~YpEwkLRNio2
z=3gQP9JbPy&s&4+18^v^?I5MlL~(&Tg&iU)7F|zI#zd0M26>?4b;v9}1WD>FzTe&c
z=dL58?&#Kno>-?q-r<K~UY#70JPfCpILVS(nZkV5l0KFE$Vb;vjghH2AJ!Cwx#H)|
zd-0dSb5X>jzspB6GiGR4=)3c?h<y*-o3R1+Tc)EU;F<VC4me8j+-tje^Wie|LyKrU
zxz4cco1Iio+8?32yKZJko5F3EP5jYvA=<Y#P!~}@6XoXZOP6+aH$;=9PoHLPR5jQ&
zfe`(Y6Cu5q;+?Q$hw@J!#318(jLPNl?wJm*9|;ZwV7JX7(|;o)H2VwZps(xS<PdI3
z;C5<Y@97KoH&FAr?o}QmkN^W@k$`W5Uh%8@7@~Tpo3F+GH7t`Flb0UTfoKY}op;R`
zb&ov(`EQ`AsKKxUDaBPS9~Zmhordb9Fw2@bi_xFYCb_+zsOB3&B{&adI5V>!Fg_=}
zI?P%z%r0kGsU1`_6ax|1v3iI=fduF>^(9nKVVFHFb{M4B52LDcam?gbVj`}S5{}_{
zrO@wyvM?Ecec@u;e*WoI77|z)*v1&apsTeg0ETllBzN=BBVrpv^i&{hh|iw^u^YQ&
z1rjXg9CFqzu^Yv4VhRkn>rfFH&=!AAv2R$AeHY?#P_(rtt$scQ@-#(yYGcS#(*)0@
z<P;KK(AnfU5&=;2WMonbFc8aR4pL$$y+g%t{MckYQXC+c-4tKlNO6V~XCcoVzSz+5
zJKZR=Vl7vfXs^@-z-|hT2%J$4ikzyIXRbuzkUDP9;7oD8RASG^_pkqULN<YGK7H0H
zca)X`_8c68wak#Y+gt6<UiSNzz;59t2P_*wbPfOR&};Dmz8ga??r-%3rl{O{jIhiL
zV;pnJ>TDpo#4vcRF`nDrP%Xo|SGlLGbbPV+XE<O=D`YIUo6c6WBw%1)@K{@dF;mOC
zQlc&6Bo6lv816Q|;RQ|yS)W_5p<tp+Zvqp(l=k}Ne1DU{>&vTX=E(vX)3H>X52630
z^t#8Ekz2$@?ND31K>B>^Wdq;aL2ux2S)Cv;T`X|Yq{y#?CvQOP&+=sSp-BsxMV)3_
zx@dILcc8?@uu|u<q^DO}0ea`DJ#!2kLXo-Kti7b{0ifbMp5+V|c4QUqG82-N9aO1^
z@qrt##WHF}$ksQH)O82&M>9m25$*~UcW*yJfYnx5o*$=^b8@zy_(T%qWqf)(>pBZ9
zt_1eND@b|f5d}EzG@mf5EAn4d8r<>Ax&Au4Jj=s~i)STGY=sPFc+DK-fEo9)mJL?W
zqCr42tmvN$Q43XL5Ba^6>R4cV4F+A&t%6`bWIVsbXRR}+h-l|gfL+!66*P@JVN^!K
zW3i|1Ex+Ccn7XnkFzBwN1+It?_tEd+VF3+~ZjWgM{N3^_u^QD7Po%*EFXrxh>~Av1
zU9+{N)D>iwH#*TE9ILzLHWkVoyM$>ZU#j^+c#s>%z@1XImS%Co4;MxD)e4NcIA6RX
zT%ncQMima4A-oNbztY<DB1tPB8iGPWMb!neb=4EwSYktz(r3pCSt1yj67gvt^MO*t
zt|YG3lX`n|q1`1$5ho-~>_#OMhBazc7yUfyD$CYiR^sa+A)1k7Ngs<&MXh~LG9l4$
zjI=oD9yqt*lQON!a6K90Y)5fqIau@|P)4VrC1jap!Y*pLTkW7MdOLRezF^8S<@jXr
z){~|QHXz#chqkes``zZJ(%Z>pzkskh;AQW7VJs0CFzq9l<2%&A;cv&IP6CT51XD4n
zgd8!hBO-)FNwHAqB><G4D2=%pzZpM+Hq);kK9z!C0^EnggbVz1=GhPw^SoQ>Aovnm
zPB)yBRe!9&GAFh8GxYJW#)J@mn=@7XF1R3{GzQo;)1}s6PA|Eu50%Q1ULYalLR2$9
ziFfo8zjYy0u*U}d^2K4ZP6ftq?cae!V%UB&Q{B6d^8C!lfS|9^OoSQmvIG6Ag;QzY
zY*v4spJeZG+)7hob9<i}DWql4$afT9zIy$}Q<kyfQB0M!KdL4=vuVg1cDc=+Ij+7j
zJUQOtFA4P-R*5hH9GzqeJHV!wagtDvn2gNb=<Dk@<nSX#jD6<5ji*ZX`r0!l{x|Bx
z+2T0nlBW3*J4VG@x*+y^;F~O}#(8FgLl#Pi6y90z1$bi)eJAhAp-ai!f1c5q`z=bN
zS|>bzQTb+@1fiC4qY^b-m441Sc^88Sjc!1qZ*D`7NaU;%U~My^D1f<#o|FqM$%iV^
zCH>Z@(wEUn8%lHsx~Kh!&t=RM*L-p%To0v_#GDo=@C>=@=Zhvy`P~)NGIaRs^q~`o
z-XTw;*)Jct9}~VrQ#Um(=sa0}qa{0w+yyyOn2n6d%HrY=4(b}+1?4PqIBe(B@(?r5
z=J>1SNA$n?S=roH_JCcy1xC;MyKO?-2_TZAdG#1W0$^_lZ+4&pvM-H#j99W%7f&ua
zC3@uDcA^~K3b0dHZ+T%j|B9C03n%^9xU0|iL}Vl;QPnu*8rC2K%vxuMMTZ76dN<Mk
z5}tUySD(7$H%!WEKnq!Nk?YuFjw-iU7psVuHw#mQ40PmDqXPp}4ZFWl9~K!ORU6a8
zERG8c66z$XIe3!Nf(c}}UR~(l(q6p>htiRPYcI8dtdDJ(BV#rtE7hDBXghu5-E3LL
zJ|n5_ANKSy<WHOk?oVQID#|W&!2+S5=30JqC)FA7=)K*VZADsi?5V6W$lK(|;hE{k
z5^qPYg*U!<a6pQ|uE1cn_|Az*VYSS7#~G8ga0MA?=k&#LH^VE_d8xl1Rtj2L3;6<%
z10R?YXYxG1gz5XA-kVIfoONhdX(5+ap&O}s`b|dD_Cs;wlk#VDSN#W6QPi*&OQf&I
z80q{6i)6}!W&R-&W4!R<OwXD2D)J&UNBy8<p#`Uc+5w}Qo(CsK?(dJoE`*T%RgG!<
z{K!hlw=*;-I{M@hCtJaAneViW2HE#IM0KrP=(mOEh#*w9afQgiQ&>}$cVAu#h6z^p
zBw9Kj;m`;DGcFT*77TVRlxs&{vHR<v4%BWU8Aq`Ck4FE30qAxj;f*Bzik9+eYIZ*&
z?Go<!qyy~}MrPkeU0FfG$72vkCF0+Lxb~8sOUWW8X^2iX&8rUY_)-0AyPbAE<RU5s
zD&J|gl(`u-71bNzUH?GRD1+u;6$~4v+85-xWWlqMz+e9=XbV@k5M?XE6dF(*5BUg|
z9XO+YQ~M02R(&6X`6X|x0td@pK|h%k|0%HAYYG%>V{R~+CJ%#nvUFLs<KNo0!!jq(
zoOrh`P|@tQ3Y?H&k+GY^esk#84uoG>Pm<X;@A1;9t`l35E!luY+mL`|Xw3bSn5dXr
zkLm3P>b6rzyFF|Mx~tCY>YR>-0n-9OYE(Cn`NT>zL<nnweN3{!|A*q~R_gjjl`Hy$
z=L7gL=<Kb`Xs9Qb9CAcJ0=-{?!7ViH37>WQ<KNzEYmnslJkqx;$az|h%LCPJ7E<L>
zpqw`W6)hV?nAISo(%&{A%?3B)=5P7FJS1P_?(gfWSbO6Dq>7EucpP25)VQ>L9NmsT
zJ4s?saFE$i$Iajom>7!<CqZ#l-j118=m4&@ms1`AroC~BI#2pmfj=+5GH2|RGd!1a
z%MhkyB?fYr6qr(8LBkFYm>Ke)D0b*_Q(wpH;1(aw>cyEr#KD*!q;u|wQ1%-El)8M8
zE@vY-J7IKg^ks~MeExceHdvFi;&5kW&iWiogaW|c-nW>`d&w}I*;Q7N(<{tQa3GEE
ztIqTXR&@_)sN|Z8AZ#1>!{t@#RlA4Zz?+6MeUZ4FwwP)sX3e;23T=xq!sS-DNF|=#
zMggWxXk=>THbQj0{{lmJSv%Rks5_PGIrM$8_lDzYFzCX`u>u@P=854j*<3UydFh1+
zRe97aiSd3ER&`>$QW2SowU9`21Ykq^U!sADZxXew;PkU?RKnikID#1YH7WQ0Z*L4c
ztoMOV*5a7D2v7`Tyq$1-Ui1PJ(jiQL4~Ass(@a3Ez3z!uX`Oed#7wG5Q)p63LUF}m
zQIZ$p<Z+~{Jq991taeq!M1;wZI~yF4m2R%o5QD9x_f>olr1uv7OQ%;H7*PF)pQW3Y
zzOu@JdC#Whbxd*i;c=#4f$5g&^!i-A0;Th(a=xQE6#U}J2>3&k{RfEWHmGE~Qv;!t
z{XRet!a6whIpV$mA3slZ|JRCgrdoHkn;vuWveL9wrlCZ&o&2K)D;ISvNIS@s=z{n(
zd|Zlh;gXh{EeWZRb}zw17Wf|a^4^E$trh1=sp?wFMPimcVyfPgDf-4uL%3?2zL=7C
z^e0BVrOqb9NDf0eAowxx`Q2eBUzuJkuS&GKo-=vfG=WAcdAh4>lh?%`eOCTgCazu6
zmp+5G!ma0}tZS};L^``J)zeV&(yf5t=T5hE;x5sR?opxiMe%(}08rDhIm`2?ZdxKb
z`t2I?ZKL~b>s{oxCAlYY+3OS{;%{LKUUy;zDXU%VdwDbZM(b+NX`lPz##-ug_oo|C
zXPtFT`jjS{S{MyMrw`vmjr;Al8$aBS5Zj#0@ZV#g(m!hgd0=~tnB1FFY~Y2mN#iTr
z3><rmcWic)^bwKD+rWrJ)v*!S<G6kDW%j7q+)T;qvru=n`6E@<ZnJ^6RR5~u+C1FP
zQ;Yq-q6O7M(a^h#EO|TvHczDP`buwm+0Lj#L$}Wp-8>7#dI4t2Jfj3yx8Aw*`aFV=
zxhu5kzk)~^eJdG!HGJzV-;|uKkq<~uczg{D-f(7jR@aoar-8GgvXA(Gq8Oz|I@>FA
z*qdy--uzfr7E`7xk%$#;b*|2gBr><BpInb)ChhXBdrkw=XuFOlH~u=S7a-lvL=Aj{
zj*DA@<e2Uw*a@7Ahc>NNd}X#Oww$L<Rv1t<o(?tfpl*MZ1SmQ<jGI4Je&4l-f9w3b
z&yhc8Pg3{Y@dG=0)|HAkX?foo)O^=9XYOI3<DCSi?BNLZ^BYW?qC<NnxvVxYO68f|
z>$bYR)vK16LYcsSi@88b$HdFHn6TRD-q9o^XTK)}2YEQUr-M~}#eoB_+tnCTSICzy
z&rscWLY1T7M0?lbZ*P-<N%QWgmuw#LlpFY6FmL#yjKJ~yh=%9{4@PHeKPgPyo}Q&<
zyE)MwGYWBMLQq}`(-pu|QMcS7XZSjhGBhmya;;^|iQU>(S<>Zh_-fmDV+t;a%0gD*
z(;5`71Dw@Ei6y5>HFm2gvpUcI{=@?VZtn9{;!)Si93sRd1+G!KZ(dVqEdAXuTTd}9
ze97!t9Qb$O&(8ksOk3AxblvpsWzFRX%BSylB~2mU2_Wc9rCqX<UULPns>>wPkTvf3
zU!#j(wTfVDM<oBc^l>ZH5Mt1`*TZ-Rc;N3p_`wy>87}Z~Ru@5(YYHf`{PigkOk%3Y
z>;pPcTC{B+3N4eQX+BevG$R(F8pu|XWe}0&l)y-&cuy2%{Uy@gTG9-kz>Yww-V3jr
zA3;;+k7hyxf{mhVtd@FNuZSI<$Y&%Y0&ioIHC~X`r&`f+Pmqf-TeN{4(E1WidXXCb
zCD$jl0)=T8KMCU<(4DM{tJtwykOya~00=*9ka$Q-;yvDi<lp&t!#>dU3dp4kNH>pP
zKR`iAqUCc1H7_IMy_9IW{sp)Pw3Fk+A6>-|O^WJ_bW8VWmXqUo%Yd2;f%NH~_)P7Q
z|I>de7sgDFV2@}*33cq}GQ^GFFIaacf!h&31*z!9<d4sj{Yy<XIX4Zt>KEp(Fs^Ni
zAM*K+C)K94h*E-@VTaMH+xsdqGNApHjn(*&8QVHaGbSv6GJC*`O*+cT-+G5rl19?v
z`MOw-kz)ex`MRqJQ|%Tqg;ZekdPXI&2xcQ~wh1d5O1!;w^UoV8A8H9h5xeFRWnr_u
z|G^(AIS8w&?abZ+>m^4AhyNCve!|d9aPyP?w}%Uwr7H8JQR)Z+%6MUVAlqhv`cafH
zMpKjV7=abC;cu!Ki%7;iG>P$Z>ZK>|MaA;PmSN%5xEZ-^8N@zcjVP|m(8uOO3o$LK
zP*Xy1m}-swxEui#t)3;CyUlq*)tE6AOs<%*q;zId*>6XFM1$0N6R{PzYb8Kp$@HfV
zi~oPDO6LDpLQ=VyfuyqY0XHsth-gcUs40|54qXJ%m*ui>zF&oU<vE4*Ww5?n0vC}(
zvwo&SuL6{-L-wzbG$=GE?(9qR%{m=(a`lu6n%hSG%Qm-(c@iY^MQ{|Zp)`v6*$TF)
zlUe<Dhl7C}>#=Y{ZXC&IXq1ET(lA<t(MQ-Dmp?p@TXo*rKw4V&9FaG%H3ZwbqUXZ1
zDS#Yzku1Lr#FX!8jMT4kN}nR7Mb2#KLbb{$MDUX!Em2WW$tmz4Z{mZH5o@mYLterY
zmPfqn`v2E~n)OA+)^!QPFsCD^o#H;fotR86FOI(7Paan(t^F!?m+@*&2&@C_f5$I5
zGLBE%!vi5iMslsmp$A~r#>glhR_DM@8F>Fxmtd=6A7b)k+xf9bSfaaeyxsZKB{ad}
zB5`yLGu+q(ONfMG?@vQ@H{fR)U(E+ttwu9W%8Zy|4<)skYKq%0-QfmFiD|J9`*6Pu
zls8hI3{8bA6%asN6WbP~`$YTMuiWY$y_8Bj)B&X~Ws^hshJM)?Vbuo0GjbF66G8`b
zAi)r+jiU~c^NH#XezjQ~zF%~DOTW}UPJ=IW`dTH?eTP*P%&zCLn_9d<YHA^P^nX#s
z2<z!M{X`8Q;p7^&3k!Q;Z{!c>sExc(-|OxR9OpI2&NEU3$loY`b8qDaz6atZR>L4W
zhyk6*8nJMLVn<WHB<M#ioBIOqQox<*X{bfYb)k=V8a#2XWomCF@se(5mg0wr@cAX#
zAp?#gizx9hdXlG;>crtozgM`p*7K6vXXle`o%2FI@N*0=4F{a^irBi@zUl_(wBKJw
zs-lBj9RX#D(Hvq0WiM-^JFd%VQ`+U2mB82wpA1(9wqPDdk){&3F1x<G0Ci=jm)CJc
z7tb0-E-P#=U=D|9s};6TSHDgvm)(dYK3|75c3HF^L<#oL_4Mk`8u6d~8Q(?H0*vha
z89dJ&uD+E%iM<(JTdy}t-crzFFM7rc$SB^37#R2)e5ccTtPQD6?igCQouBK@W#GSs
zZt|MV1l8KQq=<xr4CgES#sI2#xKU`j4_Am+93K~=8#brBphw#;I=CCQ=5WsQacWKs
z*|LpJdhB?*uw|IRla+b*VxXD?hy}qtlA_nRA^j=SE&fujo(5z5<)t>rD(PEt3l=6c
z!|PN`|FeY}geFEsfhefv{-90BBG6{7a*@Px&NSCthzg-?wQar1i+d`n6g%K?<wQk_
z*q8l|9u)VP?;VOsOyy6jq%XqGdKyuGpFOU^xE{y`d$&;?GfeOC;0-x+ILt8brPqgJ
zkEG_EqVFDHPo@U%ZzVhuwE14WWd|Sg>H33S*=UYYDQ2^mPA9z~4Ft-qF0gE|Z9AX6
z$*}BLrK!-YFMtn;o|k}K{4}-mO-L0%{H!T+`5M7)iG7G-Tc7O$;8x|QQjubN;bwa>
z1?OKfl&tTRtf5IhiYSc^hO-g9vRG8lgvXNW2vrpOOg-f?SmKC%N0-F-2=F$8BKX&o
zyj`1Wvog2u+Qu;J@p=^J4vc53bmWF2^Gx+4F_e0Gwwk^n&barDMTEqkCxF-%#!`Y?
zjsPyO1*FVh_DcV6bm05Av4T8*<@Q0>U-%0sVK;hEJK?p=(hqddiDf)4!#CoLSE5BS
z@M<k3@z@W__CqJ^2y-kH<$Tkq9Diux+DK%QjNpCX$G;uZd<X<e_F_r$Jsh7UP}%k0
zRXtFXe7bmL)PoQHx-CdJ7{uZQW*}AuL!;~#p+|c9cqr58b1Yq~<9|sj2p;ofa+%M=
z_3DE-c1PRVy;o+%K-!%UFaomc6W2#Ew{JAdzl}ig{7+VY(*T{@e7Q+*V_k}rRnZ@p
zP59mQ8n=BKU>$*%C+Rtaqt&@OkLaL=sM8=KL_qWWh1F3*?R<%yPaol=!p6#IRX&d+
zedvJm+P=I_GG)1{{)3xaB=nf^_cmC%K~+o5_+MV`Q!Mm03Rm3HI>Ia&%C!LwZCEwT
zf08DTTc1n3C3OrhIC=MV>-8%P(sd(W`Wa@GZDtJJ=2d|9jB#6{POJxJ!{hDCJE2`q
z++PjsSb9B11UzHb(Q9rtU%4R=(LN06w`tJx0mvTybwh-_TzO+qJUdlap2o0>MXzxa
z^E;Wp66UY1QrTOT+luB+8`T(A#cW3J$HN0?WXg*&pxJeYJKce7p_(LK$<_u#XsEJB
z-rL4I8c@J1&nk~_>+ff`e=^coHgTS6e*`*0m|e+F;M9?OFgSae1<Ng(9CaL2zi+91
zOO3m}`;)Ohrs|K_F<GqBiad}eyiTOQUwVTp5SsX*M=Mn&&7103-P1mZl@|~I)iSYI
zt8%~8rr6~u!^elQ_6(lycgO8acoSBnW>eHQ4qJg_I3Z7YK%Y{~p|=RG%6UmuKmsa9
zPC(=#);l0~gV=%V2W8ZLs_(2EBXI~PlFOkW-f1m1fB#4nzin<bQ5H%hLKVe%ON3}&
z2t{~kJmS9&3vVdKL+Agy7K}0-gg+uXZ;d=qZC{~s<^cZ}c%=2efmtI>3xRqBW>%eb
za=~3ANHyhTT6Q)!`@s#cc$xlsESHmT<M7e)=WKl%-6%#474iQk8zH@86IHegnJ=*e
z8b-=!5Q=#f!QCnz`Sjkg`z?GeV{t?Rw^cXA>iS?z4WsI(n)*f|u{w=8W0KTsxLX~y
z-xPL9V?sFBxQ1EHkU;cHW$F<onstmKQp|Umi>EkK25|iq<ZAPT9!4XEkmC^~C|Wzc
z7vT}r8ubOh3hUZ|nQR>V3Ts9j3uV`?UI?ne^}hTH>sbfoVsFwwoancMI5<i$LAfRK
z4!8R|s>YgpjY<D*D9wG@al$IR+O-XS&M50=8+07%6b#2IEU=LV5+7p_0UnE*(!mrG
zKJ*~u7>;!tx?hS-t6mE<OO0(=sUI~O)2Jc%EmPC~sX<Lx#<NiNcV)F!rO+io&jDAs
z1U?Q8ZkF>PXCjGYFQjUm-!ImavDf;oe4#T^l!Ys@#8?Eu$XdoT6Q_1GU;g8naE(y9
z&iQH=(V&6-4d7H=jjyValNZ`$#bMI*3itsnbcCX1$5W*9zR=ww)iqor9sJ&$sSyV$
zfDj3JJP_+Q6%v3jZhzk=EMnF>yb#j~fZwY>uOE-q;5)w$D61eKGtRNF=nQnunv5kt
zNMWTFz|GK<Hr0(>q&Hl$X_*&q7NTt>HX$|~iFCiD0>Z=8iQG{hvYRh~N!VN<zE4TX
zlfan>5=x7*7E*KqqOIS~|Db0ss_Zv!Xq~FG9!SsMYwc=q6^fe_<4_xQBY~Yb>nB&;
z?LH<;N2Q+2{(gx?HG!L@ib36!A-WrW>2cnO2@j@fm2DIliS!TkT^<Fp%2X@d)Hw<O
z?OmL7fzz|oyf`$i!bv;1Uj+W0nEjaYH&r>`XNxt{pCXfVgVA@S-3P5i|3@~QS9<?P
zHvCrKal_MZpZVMR6u+fKqHX^Xs=}IgAO|K|)Snqsitx^Nfw$nKiSwdvL~$teY-mR{
zi`*2_q+t*K)s^_z-tqg2M`Yx%y4e7-U&t{0vbxcsh_5ic8AN=vNDumf5D&j(Gbw^s
z0gDd4$+hZ|D3$gU{SN@`i)q`RNY<q8HO}T?l#vQ?@iCa_d_f2OScdX#f{kZq5&Xrl
zUVx+4VQB%uxO6s$K*N7g@np59U6=#+zZUu*I_|d_vOx2x55nW*Am}cb4>YDst+0K!
z4smo=D62V-zD-b^SVpWlr`zQEyYh4u5_7K3+yfVrI&-cb3fF8qHa$~U2wuC$8}J|Y
zc&UdIye#&C&vr=d!A_d`e%k?l9i$t8IM5O4>WtolgQ?L!UJjbGJH+2HmBr+7{BUe#
z$haH*r(=XKC4}AVV2UDl7zl@CNJ(b5d+pKgZz$Ww>B4i~4%(0t=>CsMkTnF!j}nTC
z1x=EUtB@LNa`{2sjY{_<CKVQdptjb2H5+=3|8!XKUC0JM`=+LA%7K3o81WB=YLJ^V
zM>-EdxdB(7RUqbL$yr$hP{@j>Of#@m8bYNsug^5%<q&b|!K^_u0qoBB>GFesNQBu4
zCS*d*wt2(3^?&hAjWXM9cdEfQL-?yjF<Ufm!S+Z?!*10)pf4~O<rIK2Q#?;;f36ba
z65J->x_S6#JfLgd;N_mCtbFXjo4Q0pYf8>DKle#wFNhK%IBmi+Lb^g7i6PahBD7(F
zd(fzJp@`Sb%5Vka7WjCZ59#~|Y<jW@({G7mt{XoJDJy-xyZfqJ1`aVRy~#L!_-X|&
z3_IDKq#NeeNArnUoI4iH^IDo;Q1QzTz><9Wx`l1=oBg<hInmppI`HVq(b(m2>V*W8
zYr1}L&@1{@{=VBY77<fJM9wP?b`~9N{(XVGB85*|-GTM*3lJwJ0Z{++>n-p<%e2Mb
z+58K}Oajzyg^<5KS`Yqla;qjyvxylEc-LM;rN&chHM5^0WFUjQg5Aj!;|t}A4|*MX
z^l1v;5j{-C3i#tIUv%96R`epBRqYITudwMGvjKP?CR_fe%W7N;32RwsXk@UOZ_`si
zV22cO5P3hO2rx4z?#zK6;_T>cCW&gck$}Gq6D(7wZKT#AbLy6+afIm+V?uOFF9lL7
zRokTS`YJ%8Zz-bWs5-<505Qb+oR5Ma%LF(vR59e81c0h>(|C~hX-gts&pgM5JKWT6
zPQN3*_XIqa7krv%xu%weJs5GJ6hj83{>em<kFe@=1I}QHEpl4_A<3$Z@e^hzSo=u6
zFd`|L8U8Ck0e)#=g0#J5pTM53jcbvk62pxG2<<m2B~0nLZr3+Kij;?XUGlV9g8&{9
zg(eih;Lv7(ajMqao^($Oz-+p#rOB3FV6`5*Ciu5jJCfM}eVrPv7jEaFvAJ(<o4^C3
zw-#HLf&E{OJEq>RHaW2e3Ih#Sbzov-0V85U>>>m$VDm6Ww~Cb^VOGOyEApHyC@fvz
zCMcODL2S}2y+Z#$2az*<lweq?@y+&d-KPJ6{UC&55jPY6!3{wNu9=4T*&F)T_KjkJ
zIqUIq*V>m_du8#xVmX?UV()n$y^jftS08Xm0o{tquj%#et>-*^*dMen>B(c6A<gl7
zTwI<PX4}d<lCl;+3pc@DQ<bh2D5i@CUzE=sQxs>j5*G{^IPtp0tyW6%c>ZFUG8uL!
ztF)*wy}^G4mGn6j+rX(5g$^7NOT*f^Go?~yveYJv5>+Zbbez1MO8QK{`Nc-vcbGp@
z0m($hKmBUD6qh_*v<Ckr+w#lfKm6xYkV1TAY*8iHr8_tAY%(@wQ;OZBQ*L<Lmzf~q
z*d~+Xv57qsvug)cwS^rbOLkNHntssucS*v20#LfJMZ7jJKs+=c(m~%o;z<8ts3d!)
zCB4F1Bu@(#)LHN29EaEE2X>Q*lxI|J82Al#+F0LpP>+zIItRGCs$XJBX}`U2Tr0s#
z)Z7*hcldpc^_=!EJoHBOz4|2XZ&7=akeWUWGCVh<2bF?VrT`sAxC{fi@(L|Y5Y2W%
z=n+I>i8Ay$!1H{v7UeiT0*7x<xNtFs(hde3+G_Zr;A16_x~fyOs;!O+=G|%I189mX
z^EJk-0;nNDU9=Q-O*CgEHW|8<%O%kCq^izi>M?hy18?kmEsN+kh+5-qt+5P=;lnwt
zK<{@-#hE!wE$ANTxG^Ak0p{(NtLcu$qGz&teI3qwHx?HVm&RMRzh+tWKvhoyAiKd0
z;yed0w+cRP^baMv#2ISJe);Oa1NFBXgqv|qx0f~x+X_E?+5%)9F9M7Rsi{q)?nq>e
z{iZLU$v?(sWTgJ7H?8m;bdj$)h+7Le`X3EQ_G_IEo_~S-?tu(b!maj;!$EwK<`g7R
zbRXshiJ2+&a8XMf|0HTGg@1t&Up17T8J44dTarWz;bYnkW-=m306hVA0yXnigCFL=
zl%b@H5V~JjQEW0+Oho#Hu^1^m57Je+C6coM@|jpm0$bZQXE(q98fXb$``7uedc!O&
z`RqK2s53ALnF;QTyIk~mpj$tMt(^1MGsWndFd`lo713o@&u;PDfcq)+R5|01*;*!N
z|C7Cdk(c|*LwDZ!4112#2Cn_X>SZk6nh@QdSpI0daTk5^3m;qVy|EO~i9!%kLV$Ts
zX!*$^8XV*S)ZdLph?*Kix}tMQvw<AcT)!+1Q6RbdDxpdk9{mYZMkD3`gCp@98~E!*
z^$e1U#R`i<Yzz3ESX5x%v@9}P9Iw3{ro|A=^@PsYNO|LRP^jSl3bbi=(K((1{nfqL
zBZ5UX8yJH?EIKgen$wbyDOzoQAz~y!h%3Tv2lo&$3V&>ltf&{_ipSLk%6JnV?c{(i
z(3Yr~y0fDD+no9Dl{^{J06A4ffNN0(46C*7$Cj$Wy6#=S8U|mf5QRqkH8h8(KNGu|
zN^c+^zNy!vX=A|91F**^FOLL-25g;lLKZb4olaKeAB8!*;;I&3caU4`)k1oZ8={op
zEEo#0K6oIA5D4*9AWE$!*3esi65<|9NCannBCofdTW&-my$}Q%JSzyQ6gQ}u7#X;V
zLh&9-agJG*V1i4MI~vI`M0<ulV`YZRAaa#M^Njl~r^gvb9?;U|(mZ|c(xh#WafO#*
zspE2*F`#*!!Ok6g$_TzRybX?o1q2KPG!SL$(YvXC4GPwz5Ck3+{%oHd-hGY+S1X-i
zWqL!B>QR=bBtwVO?yd!lt}agbtA3FHS0AZo0XvX*1^PdECm_JA1?)o&-TPe!2&{6}
zA1aw9x$ce*192NQ$!vwze#E65F?|b{6>w{o=h1r!=E$Z;C^J@vgPK%y7M+_mxM*r(
z;_j~T2?&i2<_e12NU(XR0b7hL(8H7=zMmqv90R-mpt&8mw7ljlL-lokScu-G)BWP_
z)Ur_C<io2;4x$rI;pb+t;l}2t1%$VC8r0~o4Sp{n07J)1O~dl#YMQ7RmH?)kBezv+
z2fv(E99JqWHS_I{2c?@k-;m+?C|flZz#{lW&{(NOw>%kmBfeuK4-LJs`3fw>T^N{3
z3xkig#UaNVI}#@_)1F3TffWnvvbAAQ5O`-@H@&2e(g|9{PR;~tr1j#1SmKd$H;G^?
z)bLW;K$8BaN0ML)p+J-%e4#+8!`$KD9Gvy8A$wBudUhLsRH5-jCuc4(i?hr9ZjNIW
z$QCmLVgKdf%##FpJ;FO%Qoj{y_J$jw<)_VW%*9pKG+6~oHGw#vBMtkBVn=O$5+wGK
z^zz=O9E)Z@a1(9*62qZX4Y)X=Yj970c&ZF|1lCuB7ur2B3~PH7&8t=$s1MyalHj^~
z2j=wF#Oq@PkP1ox55kk@QzAx6A>{~6Z<FTNXnpQbSjv8O_&F+DR3PgcZ0+b<MTPpn
zI6&|M@{NV$G;d8q($nOU+fN_vzOGKk<Q#vdwHfavlO5UiuhnMbIt&r!wrZranQWv3
zfo3Y||DFVtJ-c=A{^^0HkP|BtCsIKL8i27gvx`NfVx^|8Qiox`v6-XAaRqS%i3)%W
z8h{(~iy!DD-)F4t0|Z>Hl4u#$8&ec*s5C#hBp&?XCgo*GpZ@)8FLl<Ml%vjB?B|{~
z$XhHYn~2g=7hxDZ8hzIuf<J!^8|?dx29mDhE_F7dA=?lvbtV;dzuNG30*3qWQR(5U
zQOg}GSU@g(<Ci_FG-}l?2^@s{CsB>Jz!?L=hs&PR5*RXG6OzD5z_k?;@H=d20boew
zwyc2=<I^%3tUjb0f6TAlymu+<w*FuaBUg_%34aEA-D{%qif_A(!#6~)F<0N30sC)`
zE8Grx{k(kp`=p=39~`7u$91Y!sP8LzGykJQV&qnTl}T8;j~BO1s>_?eOG8D|(OCrz
z$yY`^<QfT(hh6Hy37=T(fvE{uv<z+LCL@c1DT_R#Wk5!hO*Tfd@-j@GJvdyUE`%$i
z7B#ta%H%QGH8m9S!W`bAF)igTFi!xKEY-*?Q+vZH{gB&&qb0d$LXJfCk|nd<Dn22y
zhStG;vBV1Al2Q6v`J=A*2hqE?AmpJs>a|QU!b+@Ox`xli5Vmm~=JJ!tHj)|l%(1gb
z*J2-^JM@nCtnCH1N>L)*A&GQ>ARIo8O^5s^$H=%^z@L_A;THE)*40%F2oGtd!&oTD
zpTU9lqmVGf6WKMw%L?m%_=CDBvR2TiNRuxh2ZdjVG5Og~@;g~WWOpLI;Tpy9h!pN+
zEdy_%s+QV!w4CICjhIxaKMfRh-H*bwMWw5<Z3P7E$iNH?CV&NcQzYv)>KT7p=KM<Q
zMITyhp9BtB*^Aq$CIDjx$Ta0e7GOLwIk{H9CF3GzfGA(BC%3~pjy16MS3hzUleoBo
zX7q1by7~bOS0D$pJ<^8W2(j^l9tFigw=LAQr>Lima$1Plv07zn^LZ$@G;E(+)Zs(q
zO~Uuq?+{ni7Q<xJ0yByl#U4LDzv{Dm76F5p+AM|ismAl?26H4WV7DI^%ZD9(<a<9&
zVcdhw{sNjL>I$>lr?~lF3x&O>t6*La@Mx}2t)Z;;!>DMU=ec*kaE!R;mAwCokEW1X
z3_eGz9B2X>bb$~lwXmhR@|(Kd(4a09zHkF&d4_$A1gQ`s`IWqFE8426&il0}a5CPG
ziV!6|yJWHSiC+u2#8xuTho}&(nO9{8-HV-+#YRPDqyHXNsfn0-BYRDMrZ?9yK@W-?
zR{8gG(wc(ezjmV&nj7PpUZ%Ul59Vh3ZSV?Y4ZIF^Go#hET63*&PkmHVzUy%jmF;;H
zp?5_w7P)e*QZC_Y=fGn+DY>yQ+6u?&|7CB52Yg3Oswakp7b8~|LmO3HQlOI@U1ql*
z$C8S{f|p`@1gW<>%0mMLPO**gsV0tJPRFQ-riim<8LHLv1}lg)3}W%9`4GhLzhk~I
zF*RNI2s|?@Z>dMZ48MArCW1Afjrn{-pV#f}QVEat0q$;x+FXBjx`z9_=ldcU?ry}R
zfikh16qG~=KJkW9LzMsUeu(YjfHtn5kAvW@m5q#in`afR0V%~UQ}OyBFuq8k57JS{
zaFE1fkfC`8-W?O8>{2tXMojgb=vT`TZw>abU(Sg{fOb$^Jw}g9h4~yWS9Z9w-N;{J
z=5_OVh_9Mzg_^^LOiy(#?fYM&rwutAKn2<j`evA$-JIpu6-h#3<&t~51XSp0rszc!
z5FFoVgi@0=DTp+o(IM)H0L3vB5pvb+D<`)~o>N#@*Xj)(hn%sa0c0vlLCBOe!TH4M
zL6V8%(!I$R-pFsr|Bjt0(x=p#MpH5;v2MTXT_bL~JY(^$v09pQSFWrX+}u}z;yt_U
z;bJG|&O6Ez^Zq@$eI=ak*oyWZrZVaXtuhhB0Wc(d^vnhM5uT<ZEe@}>c=bh}ju0#C
z$rw(&KD9YCh<V*NBcy{Ttl&a>JtzZ$VC+up^zV|SI|-uC3h>kN*xHa(`NYMVC`L3K
z5D?y8gjG;UJ^3z&N5#s?Fq=`pAaJ*je{-UUh%L<)AFTl)0d>24Hpovhx5)CB`Dm+A
z$0@0#_;=<8KLgJE_1J6Kh%b#7UFA#P_-_Njtak5S4lPK~3v=BS+Oogr`~#={V7#x!
z%TSF@Pg}(<;?7>Cbx#Q6lItrpc?Jlt(J1hS!$P7_Ns+Wp4&;zmI5P|bD>*4MmKnOy
zhIiRNLZjxPL&2&`#s2-A<@UMwZ6PdmO$@ruc@L@<t~4m-p<bS#5oj_PrnZ(GwhHOS
z#C2tqsSsK)@O363{HEtf|3j_poMYL<G}JOOPS0<@tV8AQt;J*ehK<UR5&;LK1*0f*
zj2(e`8!%Mx9cw%<={#%%{y=-AJTYM@=y_Pt<H#O%`j1U+=R6Adw8jh%ekWKq5iBgY
zUFN}Vzn~R&o$-3CndO$(LVq`FBiTO6w=pEg9#2nW+OY{nd~LXlD%q;zO}|^<4`G1s
z?E!p9*dPC#W*P^f?+{QFwp$C^@U>1d{o*4wd+@Ip#pGpmOO`tXO6(d&(a89@joOxs
zhfxyrWbCkNy|^JfC-MDS|ECbU)_)+!@#ILq6|{P*4+FjBN}OPi-q-<G?b4!GYZDm5
zW#)BBhRm7|29Fl6S_(p2e}q|+N>^X38d{-gWWSJ7nPE(i7hJfkL%hqZZFE$10nG_&
zH?6P*QO=qyHIY97q!**M%U^DM7t}O|6392PTJuW9u|oXSGO{5Ak}Es&Sn@GcKPbhx
zm$XV`x8v^f>~RX}esLDRyTOTjvGuI<)$qEh-}VJ;SwWLj?kvJCFZc*7o*XiuZUlP)
z^EEeTnna*t689Mj<+8yc979M?7Jhh$IBgbHbvpiaqKAG5q;LIV)ckinXJDDGG;X%(
zC^T}Dh-&{v$~vEf*y|=N$i))TwDNQehjwX}52BMTYWp}{+rxST*36L}&LhPG&S((?
zCdCs?y3_}QXSE&#QZo&TDE&RpsGA)UJO)DEIJcpPNIW+5$Zu^=cwJaGck35`bF^<4
zLNV^HB-vF9eBJ0<6eM(Pap*d4vDb|5NuP5!u>S=~quy^|R9Acz%il?`*?Imx-PEm+
zFdkm9#Kq0qvr%fivD8qLKDviWRfh-Kp-q%H&1|E4oUEf?YEj~Ub@G+*ygS{OjqUnV
zZ@c9D(hAq$>6b4^dgdPGHY5oS^(-M>WncbPg3A{V)Fbi^+(U#$b%t00;Sy{Cj`>nF
zLw;p{DH^zXjrj;X7=col<8n!P=oFh$m0$l`?sO<89yF!}G}#|0*s(E@bL)dJ2ULyx
z$W+K!r$zppI}BhNc^{#rR!N#Ga}f25Xjd)GwExaFY`eSm+wPVPa>o;FzlFNbD=X{b
zs7UDsQaw0$|NquAyw9NeDlaWlT>3}3J{mveE<PR_m1^DQCvx#S8AM(`<bQk92WZ~2
zE6)8~q<~3G`C~*dOq<Fr9%_t`;%BCP?-F+YW7z<*p$k*+9%J3nB`Rl>R#-<@`f%<J
zsrcKYsl0Lx`&N-<6Cz*IN&gZQ;5Pen`mkSio#azcf%7_g{ar#+p#Cj3nGtS)W&B^?
z8VWckr@XYYW)BxuL0%T)ZIQNUcl7in{WeT9{==Ezy0IRDyj>_k!B+fE2HK_zfnMPY
zzMZXVG@&gHtD~&kG{_~pEV!4}LystKVh)_GL+pim*zXa+GXuhWpAQajAakNk+V3o8
zV<W9uQ}i7*BgS~wF{Vun8#_CyfTkx<m-m$n#cO;Okj9Qb=n=v7kv?^y=S;gye-xh1
z`jWmmq51oYf`n2s-lc(g-!^-rYn*3DxTtnK2Z?W=$x)(EPc+ncFoTPZ{AMBH-HsN9
zRORq~c-fQ}jkk}2%3?qWU~e4oY@<GRDDS}VV>=4~et&#G2gVr)1vUThWcT`huW|V_
z2rHCt{|Z1`vgVaUIt~;r|Lr^u|0O>`)f2Dpqg>8EIp3jibP*NGj<w3A#(UiG#k#Ng
z&t60oD}v)0;?Wg4C`kR#wza-y=3f@({Ovyr5v6t2k2L9o!z7kL;Ot5^0GWW923{sN
zGX|DTWoXY{%@n%j({u8BBu%jRr0?>E&J?P%JzwEviuPS2WJm6HpKYtm@4i=&4;LO4
zLz(v`A|#FWrj2j%@}D6e#VmBh59Z8+r10BqozG8d?>r|ZIBARt#If`lH%+Qc8L1H^
zq>qWtzoQCmO&KXZAdz`k5d<^)hq0BGl97#dDVu4<NpoiVY$iUTaK<8aFDa<!|3}F`
z1xNbDecxzoO>EoA#I|kQoY=WyPHfxu#I`fBCw4NiXXk(42Twh3)qZ#FS_j?L)kmxP
z>fh&!wJ?b(h?kbU<F=5>vHODizzr7Za*;BS$!}_2VLBJz()I80lLG|~l!N9ZcJ3Ns
z&Z4vix*Ad~GmzF!R4_Bm3Cpb`)wLK!Yb09Mt;8Z_p9vTlw@bL35@h^Jvw@1XQ|L$b
z!+(6OojUMaD_KB|;jB7H&A^WW^xDc23w~A)M?dtwle1=JqlKC?XdEI(1`I49rJG8T
zN{`<?v-?s17*yZw+CllQp^Sg4*{2)tvf}NaHv@DtAT0%Uvwrv6PSjyjeJ8YeBCL|t
zKi%H$UhUz!2nn$Y729k(DBzn-IjnU$QIf<j`hIz4J;I1tks}bNfM4q<Sdn@dLOjI!
z^2JTTx=cO|A^zrs_Fe)R4L3}dyZUs|x*>L+?J-|N)ilQIIOj7Iac~@~$CoAI>vRiL
z9Y&8o1NtXfPjTZjuixY&qehqI_?yRX&F#@=c%NiTMFSyd>YIlXaR9hOQWPDfc$aR7
z*0cSIkM-31LL%pPZ`U-4$(o+CObFG>5*87flmW=$ukg<1nA86aDHqT;wBy3`ru6?@
zh)?9^{r7M!7-ul<>y>h59C04}n4t9bYse{jYGkJPlL|eNdHK^rA*t@cSU@CVKF@H|
z(9#3zvn7>K6etk<04DE$tJOr-CL;B_|4Z?QIh;KHJnpg$9Y~|;?#-9#zRi=TlPtBd
zc#-OAS)pryopkFz;q|xlDyUnXT!N)#dn{H{heXT%ktZ+^kSD9FIx&xj_b&w$5^_Ec
zVeAgn#yV#7{M`K**A7!iL}-96tRRX;3330lv?7}0KVx+Spj4s=$g+0zJL<W{q8FK@
zWuWJf55>y()=5_@D}tz70Y0i*d&7p*;6W-lvWkxmtzmLx*R8@}Xznr6w@1qC=iOpP
zM~r9s@-BO3m)7ggFV^mBoT5-_1p7FMisFKqw<c-}G8YjZCjZGje<}2?a3FD=Lx!Vc
zdV>vyoe>HEl(i0A$H$@*MZON1oe8{RWVU^e6c@OoO0`Nxh_bR($rpL4if5xW)LkJv
z<EvqA(Vhv=xQsY43}-YmZl7_NFTHZsW1?O^Ks3inj*9Y1Uw#Balvz9A+X&`wJPwL>
zx=i;3Oe39;*4+7Y`m7jZwA@8g%@*4AA3_kszjP-9wlSLLf6q^&h9r8!BQANP?SK(Z
zHt9^Usx(F0b^PS6=nd`OMWo6((5AeO$mcW6)xiGmyw_|nj$$m>#Nl1v@bIj}@Rxys
zg6qzmHQ3W(NG^xnUn~&SbCE2lFqWf-Di2yb#Ko<XcgvDUv=*QNG1vDXR1<&9ycZFL
zLFy?0VJjsgba7eghTmwC@c~Fe;Su2PKR^jV;;=nEw>YBYBI~gu^J+1ZMsN6xyLhnL
zb7!x|{4*&z`Px&>LwwM{YacN+hXng51y`5XjJ^Et!YTzj_lAi-C-KNtgiMmhHZPlG
zKR+(uI=PWXKHdj-KPAJrf5yob6N6iL#I2nGKaim(=}nj5ahfWaPed}u8=`NyTaPxz
zCTpF_QJL!p7bKk>i?4M!;O%<iBH7K-x=O?CQ0v(XXiFn<3O#Bf5C>lsE;F$AJx)rP
zqA6E;QuBe(8B{Gs)&k@tDu{dN6XaMSzjR4Jq(haBDy&~N0xrYqXYEKs^uexFGua;C
z$|?=WPjWiv=Ya$f1sbV1e#Q|>f`~1CH(vG4yPWi_MY)UWiw6~h5Yv#S-ED(ty)Z1z
zf0dEXF~|12xS3*0(5;u)0~KQejRuQ;^ZltHK4$X1U*)1wnR@1HwCnyYal8GLV;l{-
z63H(>VEj1qIFE?Hgg38_g+-vPT0ju6C;`r{e&cn|;8O9mFaI7Q$rJU2uSU|%`dwtR
zUtT$|J~e!w_Z2)0k$yP#cBMqO{9;!^B=H%z2qnXxIWX}sT+PmMNEuyq4;dBFM;vX*
z%WWun$}8U?bw8H_gZt|-du3|Mo?Dvbdu<9GakcZndC?dK(GiTWMGm7XEAjxy{3?;n
zjXPwnv7UhL&RE9{Az-nKJH&|!PFO>XjoykJ?2U_CBE2qr0==Hvn>|Xe?|JVZI_Brd
zi`Pew2TcfVtw10P2zS20HqNR1W_vS#CU#lAz>_&W1tm2lym$6%fu~j)3RWX(rgEXX
zyfs4S*yTcnS%zS(Afkoyb+Q>yj^7cFFjxq$UQ;@-Wb_<*!+LeB=tH#G{pR_b=yL*>
z^ffT_n}xyeysq?b#(z(>wqfWj*gAbR)cb*p;qVfU6-qMTK23NV=3i<%x6x-BA*GLE
zE|=8cRk}<>`=jer`%RhwxiNWbAsTqUZ(In&?t7thk3hJrPNw=<%XdehrvqQ4x6qDu
zWF+8(Ze(^XS-Zh{Za7`^Rj_~|4bNxRVY)`N%3CAq{@*~`bzH;F40t%9PtxdOKNK5>
zm+6+J#U(fu<+WIoD0mxU2lxa%!9xooDctyXt>Xh$^6Cu|R(=A%0E37Vu<>prfy8BD
zuWqTRaJBvi#AtK+I*uxU)jH=r8ngt{Kb)?muP{huqg{`j+&yY-IJZ((!`gCb!AZSz
zkfpReLkF!9g(Xkh8H(|4L`~W5wB8=?lR|ymr@v@g7q3Rc8tVNlpar(KNpP9c+3(<6
zC^#GOs!>0Z|F3<m_X}<e&kJKNbrH|!2$T}H?fCURsBTBJi(60tg@dGL<eMtyfCpjW
z7cu0CTr7Q2&~la-1k>uIfJJyr>0u$KNiUfOh;R#b-sm*#$2&>86*CT=uTq=&NNQxg
zn?Oa5noLD@kfSPqrR>KX@jwaY8Yb+%R8DA(3hBhe^5oK9mvw<%Ich?pWM?-1j<by4
zUXtY8M4wzTbpSMyAr>f~$etq#KG1yQaNj(n%ZC}%hvyb6)<K^qnmVdv+6_w@HSAgJ
z@mH5AWU3G+^IeWNqV|&~O3OLb78PS=cLS{ow-3_##=>rT9f57ev-B}YsL8ybTE{`F
z?N%#t-dp#_*6$ZN^JRk)VU!kkSn#OnzF#1#*Oy7Mjt-d4lAIGnF$PsY5g7PCedR|5
zdQZ|^N0~1iD}k!CpEW8UXd1DyixvU1XPB>JgF9GA3w_P34sz<MCHWHHW!G?R)JzyU
zT2T?5e)m#GTgWr4x2ZQfQmtc2QJ{CEjXvM{hall`^n+vOi~2hpkeK**3wF<-CJ4Tg
zFi+9q9RZ=P*=wOpz4v$DJN>RG4k6{l3T`m^3Iw=<a1a9NvH~UFg;gQ=h%pmsJgb%P
z`A4^r@fo2M)^~p>mx8npc~Q(mG{CibB_3F89zv@^ktQ`BGlsEwT^?{{_oge$6VgKV
zbkNQN?@J%24+@&Yj!B6S2}Gw%b+yz8Bk$=$#6Z!IO9MIec&}XGi}m+j?}xuuN~h1D
z=M@r~t0~b$X<FoK6Xbl>+QQ~zqQB+tF-%jPVRry+QiX+wy38sdHPtX$DJh%9M}w|F
zg8y|dl|$}Oy4zPR(Hi1@YElY1w!+EA<q|iMn^9lfUG6E;FMivN*G}ju^57xl3eBcm
z1~?IDFaRkv;N`ZNQfVgvX`+CI;eRZh$|JL2qaf3dK~a&LZ|9pjyOzb4BE&^1I~A$s
z*BOJ)e3=Dcd;b8WC;ouz&mHLsS6@BoExLlM7!4JutVH(QXXPuiZhHOjpJs?hInfiM
zGZUx%jVv}xh_QUt0}9G4+I#Z-rOTGHz+PGI)&^3-#;Q+4*zD%|EF8J&4^*Sw34Dje
z)$t$22bTS3K^MANT82_%VvUS9d&nCk?wXb~#9KbZTNQbnC6pa>_J#ray!nACG|97f
z)wlOd<$Bb<;?TbZ#H-K2iCj!FMmkU2&_{(E43nx(SfDEJlG=qIDzOe871}~IfLl{~
z!yZqS8@JE_e2{=e5E!VX()T>gP}?G6vr!P-%3tQ9lc~R~1AEAQ#1nq=Z6$NzS71jo
zemJiP$%z;o^_|!=L*T>CXO7~nIdFMxQyMiKh!Gag@)`SW!)hdZh0~kQ!_yh$8$2oB
z>1<OE%tXjBo^s<1H;Z5<yOual0r1Ge($}a2-}CmPy9zsr6y7ue=$V<?{s3ANp<WQG
zehe^8E6v(Dy}Q3To+9=7(_ma0Rzi@Vc3+{Gv133!^4*Ezl-V=GrDYaHFL?bG`gEkR
znW-eAJ2`xL1Lf64v+K_aJoSlV!Yc>%d!O!&T26;%=wx0o1HW=;;;ZkN0Fv_672JdW
zZ>2lVnitKQ(B_%z%PalkFWGfs`R-W6Ip6y%xoY_ppccGN4-#}-e|cjXi&yz~y;)6)
zQ^$P%1-}fUPU|i@XV;h*HapgjKj8OBkd${=5%}EG9kWH-MV`U@Hrx5^=DK`nMWKde
z)6&6BlKT1mNf5Jk9p8h2MXv7^p!)D!f2?3m*Jt*pr`P*H^htQ3`t*0uQ4apJ%Wp>7
zAB8@b?Xyn}k%EEz+_}9IboM#`9j1*U&OI#f;F_#cm>e~TMrGF~>=v^;OtNqlO`PLp
zk!AD$v4gnx&sEXGUgeMe4>Vx$EteT&Qm4cna}Auw_f0Y{q_)Bw@WV5H+X#-fPus3Q
zW_0dqVcTH(zob9^wsZ0DBC*$(^8XNYU+&WpI6a!1z*4$EX3kk{+opD?J_h1@qF13W
zz7a+7dGso_jLg$pJ*JxDm@Xd%HiO+g*Eja{q1JSY<$A8{^AJflITVELl&BQBHm4w6
zCjjS^++K2I1o@Oq=^byjK@-U-zg!kMVOq6Co3Y@RMpKQoCJB<Iq`4O+mL^i{68CEE
z6w&`l*rvPgHjtn{m--OyCLpASRSz|)_tJ^_2k7HVau&TdYt*J?-75KcMaa1Z(w2zI
z6dBK@X+C5@Gei#UtF*gmq(%*0uX042>H?hlOyVLoCt>O^WM(PB9f%#S&g@XK+WQV)
zb6-{2ap!7#Zi2moE%iF#oD30rFEjL~{kc35eQSA2Hgh%?zID}+J6_^un}libFB<k=
zaxt=2LjOf~8F52z<5|4T^{2`V11d70%@KqI%ZXdimB-8tKS|CzgVX;-ca`IJ0}Oi(
zSAtQeV1AQd=U!*RU=ds`?RGub$X9b#Ar9Tp9vVckwGixjQ9h@Uv=4Vp&n*-XpClY2
zO$8L6UK1!_`2&u@<5A*`WcD@%5q~gjm-=(_@KJQp@(i@OxI!#)<%=@*YUwDRJ2Ux~
zyF}u!wc>9r?Z7O^r*+Nc6}2g>0*2rm7hx~rzNj;1_i{T$u}CbQb(*EWd;u%D9pf3a
zR{hV^m5ZfG(Ea_4TkiyP<axUOVBZEQ!;I)|KR+{Oy|2WeM#H@<UaUR)66<&4cmN{F
zA7!aC6oqGJB~Dfvz0omy8%fEsH3FK}LTq&9tsnHoKJYBfcCo?GGJZ`@!0OTf@t+f;
ziV=>Fpa%0>5*c)L>2p_Sy^1_LI=fCpE$qLD9p(pbCHw0yBFv$nhaVi#dadx=@gybb
z*u0Vw1oI`j`w8L?Yz4Y|9#1JzdvHElKcxF9Qs(xJm$?-R`{|}5wwevf->lJ<)2PsW
zR3XbXF_`)PS+<AW#(E8I0`@kC-lz03qM?q*xU@f%x|nb$bcx320TeD~K}l7<tvm30
z@?d`r9ve2D#w5X3y)*OkRjq{@Jw%J|8mccYDDK_g`oIA!8UOgtkSh3+U;b-^-eu@N
z$G$n?zl*eOaxk?PZd{VEGW>H6%rirPvU8zTt7u~%wS!~zk+g&t0~%0Q^BFX2e}ps;
z{XMId;JnUh`3oY6k=P_KQ|pbk@^-)0Ftb=Y<ZscJ!zTvLN=b@f4$HpASx*Kv+<9&4
zKh+t}1DjZ%NIn)E0gl;shy%806?+rVw3YocqUI@Q&^jzpzBe47)RL8_*aB1aeoEvZ
zhi7!^LSzG<5n8@V8K~jFRnCX)9TeWzj{-+L;3#z@|99N5wWM>>v;A)Sp`-h!3yA$Q
zAf=}+JMDdNvhkn{Qb}fyo}P+><X1QI>^X`vWJvnA*ZYFa=dyDKyyI+ptMJHF(Wk@u
z$(wUs^=TY%;TDUXwJD*!9>5QgDso*~C?rPh(TAuVD&dNt22PEl9}=KAX~UMMsRV2p
zkZIy|2Y&hf1-}H&+nzY9JjLoARia;PXnzr;#W)04hqfEJ+&5QJ>EcDUBYG#&>ZQ<b
z{}MX9JEN_@JuIqHDYiiDC2M3IqWGl9*A{LHviRWA&euLfrzp@q#GtC92r^lKNJXM2
z(Bi3gH9GBE0H(D&tgrJ3+D6tA)`I8Oy4I4yyP>Pu)2zrRk`<h+_P>prGNz-i%-)BZ
z-<7_-kX&`9=3F1UJZzFouseqy@N{U@sC+#z^(fCzj1x^l;bTtPWl&{TdVe8r%0kAg
zS~PGfKoLO~QSZ4l-Dy931E`aHhA+|%K8IN$-$Av*02ldjmYVI><`Q<oy1I=VTV}a{
z8+H@{$r88?-*GRoX^c8Bo?Qm)_tsu3p<F5^xm=+>;l3x`K6SI~1E((RLZNVb({ADL
zK%~iXq$G_%{n&cjbHP-CVBrhOY3lQ%3+QyX$Y;jPY887_k6hG<o#Amh)AL+{ypYr6
z_jEN%pfBhSeujywclUkb+<JOuI5$IoxP1W1k4ecezemU~@O7Oq2{Z1g%m(2e`-ElA
zG=KD<1ZH7^6(`x)sl42kwVqQ~+qIt4$gyubp!<c3R9(^Z6LzraLE??}h;MS)M?aYs
z#xC8nS)@!f#ps)khP82n^;c5wtAvh2J(q_FJgpecVe>N595$#PEE|PvecCp}A(AQj
z|7_r<Myd|Y7`#=QLQ<+rtjTiKGaH0}3uQ8#Fzi){j%S+6`5C;vW7;x%WPQ<a_SG&E
z9i;r1ed4=W6o<SG;(FaRGKU-NgVGlvpqGf2;S^PicZZHM<m+UJRh}3g5Ltyc^YRi8
zz(w11jm!I>{EoRC!$mo(Y_QeiY^V;2hF4RWa-(~<wb!ScT<nx*P!-CtcoW7U(r3lh
zb!|Hc<5X?eA9T+W7OmOnJ}-2KKGdSh>kCVCkeIRjc)<9r;e`0RWJQ{#!2tjR`#)Zc
z=k}YtsD81VzoW-BPn%9z8c=Ax6+dNR(wW_AvK`x-<!8ZL%Rcj_^KyN(GadbPJf6LB
ztTZZ{wqbPxE<6HqB#ORtD42OXQ|Lc`MmN(rK}Lbst<M$hF9kjEipVU)AIsqSG*4zB
zq%7fBM8LB|bjPC?&17NLTxh^ZvfcD+=Rcbag#vC=@lnLgZp%afCFS7M2_1^kd&du9
zS=zlJU80O1L6>Q;c((6NbLHw2qtm6Xb1)3{JFN60$~^}E4Q{pDl}GR@g6+j2lM!96
z@$u*fTk4)De4Xvo9L`;7_k-3BQWBc=izu-OEEqJH-%z3B6oPv3lwYf~rFgtAcB2G|
zW%8!8O!qiU9=PHB?8oNbCzXH_x9HL;Us{uO(v|Y|&1L?YPm!We=>KE)ScnLG**#pu
z`Na7|tR87YvvFjC>DWBzPA^_74PSnbILbd;JKb3*91N6BW=p)cmo<ui{_de$_*A?9
z>-kg=-@sR~vFD1mCkLcbx_19bqMETsMru+8%DqaMputza`yc{xaNai1+CdgqG4<pg
zm|*}Bn(Y|lX{oFMOedu?mzH9_C|f<HT36I6jrTQHgR$05#UqxV^K)eJm_s)Vkpkvi
zUBq1Y1RB_FVSHD0@^JD}xS>}X1~bEjT3DVyf#Cl9^GdQzVrxHj5*@pm0S%2sFc>T#
z&9dP$XNDkYHbHKgB!zH=PUKfL+V^nvKqh-H)E0-#zc{Bvu(iYA>~pzuI;jBfcEw|Y
zw|;fsY+KW*;teU)M8D95y%)VY!rX3<cWE-Y%EKcNMF^hhM}J|fA-^Y){_<zn#4`Gt
zSLz((QD$JfJCduup(rpv*tq+|&u$73$L2s6Ts1-x%qN=b1-%DGTeb1O0<TgEaOLf7
z>nc}g*7M(pRCHAq&R+0T++Hkyx!I-k)hNEqDX)>Pe0#6&9&XqlfxotoQif_@^8*u~
zX)YknIouhTbW7QJu2R*i8>!&Py^Xo3!AQ!PR3YC|csNpxcN+#}BgOk;_jUp4zVYq!
zjEBm`RvFLCuOsT_x;V@oyBV@H(&yn1T^Qwr?3MFNI@)0hQd(_MC=CM6O5duq$q6$E
z<x#4nWLq&7QWX92J!u=!9-NANy)iRDI{m56^rc;;baRs{OI|fljVQ{#int)$_n@!j
zdA7NLfAPW}-=2}7=gr1fe0ktram&xUXvJ;PHCON4dM|#~sc?{eTe-zaO<wdzYd41r
z?4hFQq`!&v(f`_C1Xe@#XkApk`qUV^@AU3Th(Tpohh>74C(JQ7U(J{Z{l|r?-D$K}
z&kF2ud(#1@%QDN0lMgBz$>Db7q<j9rA%#-sD~O{|iU<w836DKs%n`5ae+su0V}45n
zW+#-w>D~p4$_1KZ)izTtA&z{P-*<^ehp)^+RT`uexqHAv`EnZ(s6IAU83^*fEz2*q
zLD4l!3(#uXl#MF?MCu*H(O*!Ar8602@{^IA;LVd4WZtKn^n+2Ga!6m>rZ95nNUuS@
zqyKHIL)A3Fat~0xFsP$8y1pB*v|`0Sn;BjgYumHyW@o6x)6EE`wWe<gXycI{SNeG1
zX}=B5Fdth_32-BOwBg!pZ*8x1@@uf)1!O)TRb`&K5%ErYE7}%xaaUDX3o@_E%QzZ`
ze%C#%?+{5P{-u7avyL$X-zIRZGR%>uo72{Y$Zbh8(hp>)fRpZp&m62dxq0j*@^zZ0
zC}4o36F4-s1tQw+#Ztke4IM!F%ALi42oe=hvUf*k`uDRW9MAJUY*65=G&+$TU0Af4
z>a`W{lv+SuKyl&Xf9qeTpKHqSzCV)x_+xB{QNw4qR{}&y$E@m6TfFxj1x}j^MQDha
zXtLY)k^rB28keLwMlLSM&WO&4J@n6wGy>k&j!!;#-{Zq&3$?3FRKR=aSNU$t`_mxX
zq1scSVRk1i9_JnZr^4CtTu{hKhg(lW$+<fnv;^*m*`+Ohh%2V`E(!N|Hps{Evnv7@
zb}-~iG*QtO2U?4AyReeBZd)=9Hy7v?)XTsE<ICzjxYL(pE0TmL{Ss77aoaT>wInuq
zUL0%pPzgw%kRPz8c8MEGoNj{R@8=|}PV<Cb;9Ex5ARMC3xq}?WJL1^B!Wf6RzWo;h
zuY>zFI#WvXXs(oGH{jg-RoLI^u@dj_-~0TqjBmizlWlLno8C^^bh?W$In`VtBq6eZ
zU07#t4HhyZ7V=|m&!9KPfnC^2>Xltsi>I$GHYqE><W-A-RY9H`G8kmOiS6SByN2l^
zi?@9V1ld-$AnM*~2YZqN6Y#oH<`RNZK;@F&C6}I7DCd8Pfca_xoFpUJWPux3AZM$q
zX#DUfF_|Xi@3nrwyhAdCB^26Xr6vJx2>LxBc9hk>rG|IEGBYihOhp=d^7@xcLaER6
zgMOM>oSrUia8Bj9He5wsZf_7MiO|xL+)zE~*9wOW17`PBZptBPF_fGs6Mtz=Tz=5@
zSNtp}Tk2WjLB8xnx}z^^golE0jpe&ANBhX;(_}U5N$&M$kjoj;EAx4RKJNmSt+%%h
zSnGR1e3C)?`aZ!Hc`ulXjU6$a35e;FL5ZpJuHSAveRQusW7_v-%-9AVd4KH_&+Uc>
z5<&T|7Q%>b3w2zCW~(Y~;KRHsWrAGo4K}5I*qm-zeV;gP^ihRnWK@dnrbV*nBjvuo
z*0P;AacrHKu9WjMqeo1f3}Xf^z@wM;Obl5Xx~5|OLpKw}pT!3YBooG8RoM1+QsIAs
zs63UoWcEzIIS<{(TQN|RytXH{h*TyINyQa&UTYrfZ3|mZgNpS&dPpj8PRNGiDP#M{
zf`@NW`jM{>MSlx~HT6F)v=Ow)PJuRz?-KULhsl4<TwzYeu<081KERymPh(VFDu*%N
z^3HT;)~N1{SKl#co+V;l8JGYvHjDL7tz~{ZF5*oV3<S_0DhNWmLIFv8b_vNpsOg#V
zcG*UY?yH@AHLC0+{7O&sysF9^Z4j4~^Jtgk1@1fIrngNsO*bqDX=E6hzs0!f`ye~F
zV4eI`n{hu`Yc`39IRWy)hn_WsOm#W+L#S%m2Wq#=2>aehz^go0n)UAGI0gSJ-^F2l
z#1KEF5Kj`$QMHA=$7Lvg<P+Ah$5vTx`lL7}AvoMr=5Ortd02BnJ29S$dDw=tzq=OQ
z6{q_?e!ACPW)?Ri+mc&md>OBzYEXoQRij8HN_*8zPkwUMK!FK;rY$>3jhx@=r+$j9
z<x-{Hqaz+k#KoN!(NV)otqx0zj!q)9LyBhN0-ui@b|3~}!6wR+{-gq$U6#tDOTsZ^
zgqMyVU2-w%;e0qu7O9RXloT`EDbgP2SBU5+E$wC)**<ic5}K9}m0dNMd@l*t*Br^)
zARqFgkQ9y$^y)rxzFu~`Rl?zAL?`r3)L*HEHIeWL)a?q$F6lYU@*QI9Pm+p@BbhQ8
zB-1zSN(WPQ>nq~k|HS3JjT|SN1%)K5TYZ7rzU#(VC8!}a$3qTNSE!PtiQG1T8bp=z
zHK#c5kq3N>emqup_&TTOBT<+Sli-?;QiPW~;y@7sydhp%R%<vx_x;!kY_sVZI*?4{
zNunLx-6w|NSdtW2$AUvY#$KV+Q8(K2_z*M3#g8#KXk?sxK1hj-EW_4LV`iluxr@hj
z1MF(1T&!|qF9dH_>STwCoZR&UztPGe@pZ47*&P7&1%;dLO*s+gs2gUy@oKXs9OLCD
z1fwfJ`bQ_<CMq8og3NI94;Sbo)cW%Ygfh@N4;<bCV&bX&F}HDn{er*4P)lYOeGPw{
zNvwQ9|N9M&V6Bl__xy#7mKT|fzL*0?oDytyzwK1_nM1#O?AgBDvdP)@Zyia_24M9m
zDKNN)M-PK^QjFT*^2ds6rUei|T`D7Z!oC9&)s0X*GQ~x>`9(v{h51DguE{rqjBk8M
zD|G>lQ_SphTBN$O0-7*6q$F7j*K=Z|7^T8!k#0(`^5n@zWNZmW|3bxlB6CcwNI?yn
z5DdPb;FNIu`T81~Tp%T?53%>x+dE_=d2)2v5h1k5X%T5!Ea*(Bci^!$l-mhk7x;%E
z&-xwmiIvR}_4VrpD>)AKQ&%REPTi&7;y6X*xf~p}S;eYPGeB=5eJ~<NAOyawfW9to
zzk5W<HLoH8!FV_gTf$f}1bFrWY#pCGxyES1r)JpgujXUs{LoRK9@gd?6mV**KcM(>
z(>GtAywVF#B_MKr(2OmAmZ=nZ11gLRH&0NrvmLR-;?GVJaQ-xq#k(ChD0D2+Nv5-%
zGb@jxk+y}g4q?F?S2ynYbC)z>s+$+a5t;&f>!+*1S$1U5wa!T2bN%STwAu&=$f67h
zr|q^}6L>soD$dR_ZLnt{Wi+m9t-Z{n(cpP62D?8`7%uC@@y)gxkdvW?0bbplJ<YSf
zDbZOEo)bTL**4ZvU0DbN%)Apan3boNWzU#YLdc?xAKNhqUd>AaV04Gf4sAZ}$yMyb
zucgAhw9dfOqyG+|Ykn|#aB?E7dsprBbSSr01Z}`N`qLme7EComc={}J%q*+)*?d?u
zor6-2#P@i|1au`F&MR6{11K(`-$O`+G4>no@>l7u-=4x^p+!ti1F2~Wk3T=7psN+&
zB93;a!JO|;r{=D66Y2u{cPFX?+HT6u4rMNjzd<@a&u~0q?8<NEd8^nrolvN`|E=6p
zzJAE0+H!|+`}~1_=Ip8dsb<FfLW1<?-x8sop@AXFxwse81eZa|4RGQ<S(amw91?cT
z411WinigQ-4>ZZbFU)7UQyP>6`Ky;T#hp+U^NJ@84O}GDv<6=$p`gG}$-+$1?-9Mq
z^2cXx+0<>Ijh<wyUZH-%Z0-w8{3J)5AmI-HE1a|)##-jhEL+M%>D(YKS{_72q8Ffk
z?uJ<B#Yi$IoqPP!e{<rRb-xqyxHdCDd<+MQ$w5?L%3z*_&X<0pwU*qg^QCY7nuU7Y
zu`BQ7U`!J^u}w9eA2&yyThAH;6=$OF)z6b2+FVgg2zS-4SX9V(-NstL2)kDa8%cfe
zt6bl4)LU{NPkKGLi11_3Kjq@ZMP#Tk;O6FMP`B~T<i8@k1Ar3;huD+Gekjg!H+Y$L
zJ`;fWEPBi3mFnY`Q3gQ_#`gRPnX#K|`C7Roy!vmRmWyU5`!ksac173UTN=>qAS)`$
z^-GKHro1D*Yj#JH@5;1pGHwnMNY;lhU9bXtr6S5tptd{n$2*)S3d+zcBAJ2ec<q=S
zZAv~CI1;~30A>{xn@fob<O#EJKNf@Z0o-HgkdiiCI`zqcx)j*fGB#BD=_Q8m*)e{d
zQn%r2Y1Pri<ZewtNofvY)OGY1V7J;rxyx?sAYSd4%z>c~f4?W%>!sy)4j;?b3lzN7
z!b!J=>nEDf<M+Jz8+&g!L9d$mQTLWV{W3Lu;C&Ov1vpxn1S#Nq@O5X!LGwhad8#pa
z)1F$}&Ss{R^2ic%_z-am2UQD&S3pOI&|K}u&VpW(6u{PlqjlGl7=53)%Xv-%EdE--
zR~&uEvM;?YO^um{P;`}dvu3g={99W21r%@>SSTCQiVyi!u`vv-58#qwM-gj<n97{N
z5(z4M0Hh9jJa=9!%Q-m9d|5ieAgi^^I$hIrimX+X`fuR_=nZdY7+9)ioL65P%1w4`
zt`x2^tNcw7Eh>c+`9brEOtQBKC!4K#6JvHmM*}VrYRim~J6U=NO6i&6*xR~CA#0Lo
zW$BjdW3i=pPD<B5uTia~8N6IwTwn@Su5#Gaz~Dwd`&j2DwJKhH>mp91OPy+sJC8C~
z_%(C(O9@xDaAs245wK~1luY=#aP_&}qtpr`t_gBXb)+V{^=OiHv7rUBkjn-yH@QjL
z10kIzvhdcnvi0Zi_#AJM$oDg(qfOM@)r}k9fTpN_tT%%mPiQa60-V8o!OGfz@y{G0
z03m2yo&vN9`n9i*GzL#Hvd%28(iDBqEg4VIT~Yzf$TW|)eE^$AY#dQ&MrO4?vorEW
z54>yHwDG0Or!(hLgW^0;{m>PN6Yw^CbZWRGkrc#I!noxVD&$Nm4~W$oVEI$jLczlS
zsYYH;!rb6?eL|pQGA5^Zx@n-NXJ8lr*d*(+H;Qj^ZKX?&Zrc#l@kpZa<?g6j1z_To
z4pSy+Yjb9t?xKWPNxN7Da+OTX7~RPS=e$}oHHI2j;-F+Mi+aoW`VV1?SC<`pFOkY1
zeJ5mam-9Of<uED)^{C|$FFlJCpCgQF{-cKUFTLWZn5%)q{hxzo<)OcReu7FBu$+d#
zzpO;}@4@yZi+Q?s1zK3-LXU9IcI^X7t`s2Z5xp!f$QH}}15Ga_2ygeE*Y5Be&Lj`c
z&Rp{77YSPqDRJ*LR_U)Z4=|!>E2Fz@zp-iktsgoyyduSE9sQ3sIk}PoEE4Gl+&4b{
z55s5mcJb(Qi?@$RZyhQJXus~7foON><R?Q-qYb+oej3h=7=z?MxE*P2XII1D$v=(P
z(Q+?V#ZhlwCFyT&{N=wv{Y4Z?J?j@QGfnQ_y#vuVAS=`sslY)EXi25eyygWBXGNkb
z5jBail?}jJaz43*^6ThoFBr8)=?Vul=Rjyy{&nH|_LTZ?p)C&hBrdm=0MH9{exdNl
zk15G6HQF<p-9S2!Zb|p=guSp1b3OB<a&`39fAsyA#+`60_2sR!ymtMt3+l^o@YKd1
zI<z^KhXz-)2JPwJF)5}$Pf|Hwi+D1Kp#@EJ+2_!CW4+@DoOk>NR>^zYntVey&tM9I
z)A=qLFKyrSMWGmRrjXOZfh|)^hk0x~l{*p{v6(n2zQ0e?Jt?>?GVSx1QT8O;B~7e(
zT>ar#=QSw)7&tASApU+nZVv<6lF&H`@#vQx5~M@$A+RCVaRUc?M2u2sU+cF(aU?8i
zE_y}eEu2XcwVLfi_da)td$gN^@V=#49Hw|L*-GQjEgz5|>>zA_wb8g&0R3OSiPT%r
zp$37Ny;mp|1dzM7yxRAhuMkUvK0j60;5M&J^JLHotoCsE>bX}2B2a&9Qnwsagbz=;
z!Htr&nVTTBO0r7E5COyl5&t~OJxq=iHtcL}Vzo%Jx;<51^>X&R-PXKp$LXGrT%CQZ
zSB^qWF-)_!d(SEGqdB>^gPjVluAE04NkN$%;XHEnt4DDQc59O|Hp;4d_+WEAI{@F~
z+Dk_XkcB5Xgh8((G_Pfmuc+#U*#8Qs#xZAGGSImi<0{l`ImO-Wr|A3bP1UW^M6*pR
z1(PB=g3@3>_um#ua-%6P(vB*vU_{yoFl{+%OiE72WCKNj{@<Pclp!DcG$@L{lvfb4
zdfU1NNPNpDMy9IziW2Bj_e0Do5RK@DSa6TJfvx<Uiy4Dwz1{1S<Ks<%R2eMaH!$ld
zm~Y5krkxJ$X#VrA;Ay1m80>9bC#cHy7{rI*ZJX2>T96yjNuH&s`C}7A4TJ{UiyY1e
z8I*ZvfN*#WJR<MZBLbY8KO!EIaR)rp-+$`avU%KubCpbymm6FzBD>*w7rHbxHgk3~
zU3}BO@p->&(5tSQ34{QjfvL5PZbg{gGis{pe<*z3kAqdleCf-u;muW4$gZDuS+GxB
zku5I0o6G*8=fkE!ppTaTGc@^4ReF|<;`Zg~1h{)<6#aMxo2QWE*>LVnex};;!<M#M
zZL>omMy|$L*gB>=<Wgbgpd?7NrubmgPeYbBP<MHdTpAwpoBPkMlDHp|MKx>uUW3_^
z#$B12U=2CbSi}wG4MoKuZO+Tzz0h_D9$VGnat$G$Ly+|RlW<Y;=ZI@3Rh=^#5E%jr
zIPw?{k`dYcGhvdX`q^uC@E*Yu<vPAg0xi}i;;>vfaJjLuQp+EIq8^DE4nAsCc_ucl
zo22=qRju|`f+i&RK>_I6@O&_~H8dy|Y1%(&bQ<484<LNJQzaL<$yAqek9uz%&ogzr
zz}Fxb6GdMTTj@}bAq5Fc9MAtFS-<rJUcTDcks%S?3X1tJ)RV|Bs7BXv;Bd7z8)x4^
zHPaoum@u<zgfvgtCd(!vui|F}rm)dm#^+cfm9FUViiB&lXj9>&Jm%40UFQ_3HCof=
zBxPS_G?bEgy8Tp{aDv;Ya2gtT+oth(Yofmu2FR`EP_($Xd}49nclod|jx`AZ*t(`=
zT=QkVD{wE7RX#%c7K`@hG&6^P-?RLC*X@&W48WjG28e=H9mXNiU}Gxe3y%F%YVKOz
zV}!+?Mi`ltXTcWZy3WEOFs?wd(Cqfdu^tFL?5h1Od&fAuzD+T2-MX6Lcbibn>3S+v
zuxa5UdB(?c%+Z5S#e;B-JA}~!=N1vU{je5N4i8OlJ&*yK#Gb)DA&5N?�&`sDGj4
zjF?1OsmM;`V4`IZ-~my}S4VO@6R<ktiKBwih?#<E4le86;4+I=)NRDB|0cn$xvA3w
ztuEr5jjm#~bOf_IpUJgEV-V${PWRVMAk~B=R$v_g4mR_lG;ngt%~`;&X>^8GsxNfp
z0hYR^9X2h&->$ze>f>pyF4vY&$1<G1skQX({HnZ6q<R=%`zD(=|5#utSyWK6L`PLt
zRw9$5*hGaj`cAZ=epvac14j)qE-QjbaC&qb@ex?g<UV_vCx*H4_s|}-$<FNmiIRht
zuRz-WVacA1HfHgSDECh7YqikaHc<{!L@xM*JId#qbU1PS){KX{YdqJo@?cw%>Ap(&
zwu^1ua{eZSLGQJc$`-s50qUfC-ih-+-q}PbsoM@UY79?F@!JroLEFN1Ptcef&O{2x
z-TLG&gBP;P;;60PiUYfRkzoPiJnjQ|Z!Iy{{Im@e-s;;h%)dQ9o6`I7^6KQSfv<QH
zac*Z<==tg`wELvq?}bJ?V(0uI=HiPKBSz%f`EY0T+2(UH3Irq%*iMfO&9*{MZj!C+
z+yefN4r$;ew&fojs5%EnZzxVY+(4=2Yzj5iH-)O*WkhqRda161nF2vn9i<(f1F<HL
z9eU`iOvKCkL1)6t`(Rw8oRU+REfF0!Y*81kSJTbZ#4Z8^Nl#5zCj#Df3oZy&M_v}^
zrP0oZsF@~9u2MJAcIV^gm=)S1zBW9!4s2iwEDG~KeudU6WYcCr4JWfpK7QL@>?j%z
zvh1nu`XQn;!y0p(rUk&rC1y(#92#B*E{I?jEi^(hlth=+$xW8%+YcTy2LGldK!cjN
zasPpM%g8yg2vuEUC9$`8u2z+!fIjnw7s2vN<XOfa9hAHfwWWcVg!eRe_O>x4Gl<M%
z^t~wL*;H&eT!%vvUrIOiHBT|@$;P|Vkaq(<TVG|no*v<ke&ATK+=o|9LNjgfPYNsU
z&-bkHMo@AAYmTl%KA0b?p+Pz?9&f+Xs_f}`e_fF7+r{Lt^6YD8>b0vgV8mBwL$zx#
z&218IE$zlDhF$&Q2Jd8S&8a)o*w+Qsrx`3@{Y8>@wsU)*FZO6ogoTH1H5u*<s!lQs
zPVW6Z60VvBDILh!$~aIA#d29nac&`f&ur_XLdtncp1B+zK74>I7jIu8)$Kz(6W5VB
zJ0IrbdDJ+2Z`#BxV*+Pjm5Y`?N__OvV@dlD5H0@?Ao>M<0b(5qKDfwlZz;vd**NT?
zh=*j~Qgj;Ckex7gRRXNEW$*h4#Lf0IbReUW3RBpJ+U9<j<=wP2Gih(gd4EJka=lh<
z@}onDtVE-R3CH`g3XdZ1-)qae{N{4xF?rX_AxTf_mSaBj#?_!lc@A50(xI?RA6S8w
z(f+?Fw=50hH#N_Pg1%jLxZ;BcUgHkD+NV+GdarANgQKsyd(AQd$pXBCQIqERJ-~Mx
z$ue;Bs9MmcU5zjaB@HUi;OJ3bN2i%hUSKrw8yrfeu<Kt3*oFoCwRHQM!+jxw+Oxt$
zV$oY>sTsoW+#A=wt`-RaD~FM=gxhsA1H?U`V=DZ=PF#PoV*3*2PuhyH$x^e6g4HiK
zZl3v3?JA8Jm)J>FJK4kD6sjQUZv**?X`GyCD_H2)e|9j~geD#YnV|FPPCut1T5fWS
zAM8&9?`;kL3<Ud89O~Jo-l<;`9T6R^O$d6&oK%IjQg+ZjHu!?hLs%qGzG_+6C5_Sj
zzd)k=e~{?@e?cPf|D6)AE*nX})5wp)CkL-s;M-385G!6#C`gYb$#b85EOumKbV@c)
z+=p9Ht{3<+6ti~+qaQS2dZT{o@PYL9bazA{3ZSL`|71kq{|Slm{}&{_{R{kp#Q(#H
zKL39);vj6`hD`#9<L}pjb2&3B7an@eIE|DZY>**buH|o+KXZd?0h~%vI@379h9eyg
z`4++1dN)a?)6(c`mhRTIBR8IwMAEslHLqXCX;MFObzKT?qAFDncG1-b>vJRc1w;q?
zYgB{`YEb$40%~r(Tb?ZcKpJ+eU?lY>3`%njfZIc@U&tq((=ZNj=m{<VWX-f;V(>7-
z_u6Piiw99W&$*YG21-4vh)L9ta72f^Ykc=0hzP&#HUhU(sE?sCT=dy&Xn4)}FfnVd
z!cVHNHBr0sI!$k)6|{*s!2dX(07I@7*tR8OFOTDNqSAW5-M)3+ZIB2j$)fdVEfgxG
z3QHi3?|tr2s^5WY#_1@S!PmgKQs<Xe$VKEE|CzDY+RP9b2vB8&#?raU2F^p~=}Sp_
z+t2G3sS0poI3|!9Xl0eyRQSb<!}UvwoI3r)zXijcA#l?ra)?}a63{DOeV%J#yCx1q
z#_HhpCSS3JGC9yGY7Z71qA{lwey}O?FE5%y%gIU+@14&vk2_J>-FJ-px;MxTi5&Zf
z#1DPl8FNoeXdp`;u@-BX2#||t97CT`+xYA2ycc4ho8*wxWk9-=8-5hBzh2aEcCsEJ
zMr`nyF5_KFlQ7#W2|eb4b2f?<!N|R^miksQBzSr?C?zu3V}1_eWi%asC16Oim=M&!
zavCp3cj|FE>T7jdnOP-R$tZCZd_^qE$x?W1lt`SE1Gu)7PfRH<{gk8oF+ecQGMB_{
z-9Wnmeugp3K=eazDaYs7=Ir8T$nrj4z=W+7PuS}N??sjz!Si<o9^s3TDYg$ckMTPN
z*|FqL<ziVVWNzXPY3C*l3<t@bp2xdDu{xP%E_^NtRdp(v?aZ#$H&UlCpPVsF$$-!P
zWwFPQPav$Y7^HdNr=_GNS5Wt+;FbDDH#qwT7fh^R0X{>ANy!HH@`7NT|NOL}6%P4_
zmYJpg*Pp8U@`JKK+$Zw<%*^CYd?fU-`3s35*ag9sP`|ppt{>&DgPjW*emWhb+eW^2
z)fcHIfyc10q@fWc!CcHVvfwU8m6%inxMJ}q076I~Atu9q<6&0)J8KisD&fBz&nQgi
zp2*2K+>cu`Of8dHM9|}5-xdzsw?7BMf<2fxpu8|)Ol>dFsa}(>S4ae%+>pyBe|FgO
z54ZE>0>7^{J7ZV}We@t~?!g1@c`JiX3-Z;ZdPgJ(SOXtp3M4q;qN8;w15j~@r76zz
zfX(k2CT!1+V`Bee%z~xUMKP8QyHgP7UqBcgFYM;qzs(#jT{}bI$Z^6G;Yil4ioRhs
z)1{2S=I0QvXPS?d<dP%)m|z&M9+Mf;(3bc<Da+;tHzHy21(UIY|8Fn}NA5f1>9)oP
z++)vkH?KiaA^P4dPH7DIu|0l#xw!`M<h2IoZkUGl=aDTds3WnG7v6Fjb0L*rNkVgp
zOGW7NLRfS13?1R=g=;Zyc1Hgck=<5CFk0I^(gZtJX*GGx;cl57MUEC5dlc2*-eRiO
z8NF2UvI%k)>PqBmXJaKqWvet56clBF6sXZ<({+}A$Co!PrA^9Ss;cj4>39SJ{{du7
zIa6sy_NiQp+bC;-Tlgjwe2P!Rx-m4`@?qB<Og31D>I?D#2c4~l#y%&2iZ0>c2=`1M
z_%7DL7oMDlpztm2(Y<NA0*>D?`~ripe(ZlyQbSyS+4j$WC|U7^lEDA4B(Kco|FUG_
zlHniK%*)FG&=n)EAQ)TTJx89xe--3idr+MtXuigZ6)peSg@1i=Eb9#;J}Pdi^_mQA
z?@_JCyE!-ckqLZyyjU?WLBJzp<-rO@&c=eOuO4xi9HfaCJvT$>c^$M&7Fa9-7vPb`
zt+3|*P_p3xhZaREC@|AtL>X<<S%A^oXH&pc=e_pU$^hkgbr`<)W4gx(Dd=qTCMoha
zt0#86S^5p}jz~4Lli10PhJpzd&iN6$KkPfAcst~DnSH%cj^!iiWNFEH--SUqEaRt#
zQnGbQ?R-a|EEw1W7lf?asbf~xjt4pnS>{EM{>|M0Qx9p$i8U!8?eggp@ZZ>P$T;wm
z_v^_f*>VqSl0YxIiw}4A9+RB;yQDF#xK+{yKnN_e*TBSOIrh88E`_(@i@8k~^gykl
ztf8pn=&9`T5gegc@|zANGU^Ph4%pSV$di!hFcZl>zw75v%UHy%Xi|xZP^^2{kPLRL
zurE;qg{=(7-lz*ytFEWmp~F}2j`_?Kn(6s;kaH;@DrdqbKA{8btlZuUB#mS0Z_*yj
z)n{Y1vgWP=h<4kZ$P3nL8QqHiaj~EXdZHmTSS@wA;tz3Qmn%`1iAs73a2eKY69rL&
zRn8>tojn~+7(8OT_{rz_1&-67c~i#VbrL`_qZ$mcsM9H%Oo@s-z8<UHb_U7fHHSk8
z2BJb;5mDjBh1LT~Azp^Ot@0b<vcHEd6US2pJaI)kM41GLY-7(4VYRCPcr8T9Mk5a^
zihM^OcjXn*1d=~t)hL>jj%wt;dn#Y_7CG62<zC*f`f;Ono9IP+$9Fqi%QL8!qsM>*
zJI1L6T~Hj^rMFQFlVWEqeykUB%WrzM_Q#nsF%gVd#kY_NYkO196AWUXhGj9L@UD;v
zoXCfNe_*m<DZWJ88m5Nj?Q4PtgF`_hAYo^-;bzN(^j8G*e_P+}`T<3Oo&66qm@K+K
z(l+>S)5>h+iL{4pN%s<Vry{nCHLMaqRaQaS9;tRPHHze3L3jXP&_!k6^Ghp?r6$FZ
zLeZ`q0cG)4*j~sYbW(}HBVZb~_3uHncN{N=MsJhhPb}V}_xZ30XRK0tyANzLpvZvV
zIuFj#&b+QK38IuCjJkEVyN6nlEo23O$=*@XFV-#P>)0P%6CKB0Q6*J~*NYX<@URK-
z`fqVI^|$Dlr!^tOPMwn4UOYZ+M+nOLQh*%mRFJutKLE;|b<6ZX-%J><-!Sp(r$nd<
z$5v)TCE`v)ihRpNsETi2O3VUoly;f2BgZ6|w1$<nHMfS;b3i+^OCG69UP&g7^_`Z{
z_3zp4OS-JS?ci>#h}Fqga;Qh3kdG$0YrRHhTvsBpPC05e8npAkG%al;bviHk`vZ#9
z)VV(;FZ?#_^s+2eNpqIJc?_5lj<~xlwW)!)0&n_K3VNhwMN@KVoERECoECaL^|iG!
zu{M)H>^UX{TBK1u_1XmFn`d2Vu<7@HyxT6__|ips&NGB_^@WgY=7k$TRwxs_N_vO0
z;R3c*^U<&}teTr<+-_@c7aNYkMCnt$*5U4pvP3j1U%x~^C~aM6W>aAnH$B6i=T+nM
zG39doe);3wtC<&%G6%F2Ef~E!birn+Q;s|BlDlrcll*-ruibBPjq7Z7S4m!x^w_G*
zbq?NwK4SRGVU~xc$T=Xzzrs)%g7jV|nPRGHw5Hi1sBEJurIX)jH<FOd^iSzWvj}TK
zN1d%hkem^eGYBYEE+zys=c)OAKgk0V<WI?Gr0;~F@BN;|oSCNh<zpo+sc^J1rQkF%
zE0LgP10jzEm<0y&xc?&AJt(68=p*36{NfHmnVIq{dnYU0jcEc<7}x2C5Y;PNV{yHX
zBdTNP>K8((rqdN8F#nDoQ_yKh(#*u)DsiVsr<qmcb0=uB-3M)-XY#lVCkSh!=vcDo
zry$;uRZGMYB7godrOucN<);wCLO3{iZ%i@2bgNuapqgm>%JwEFm>QV0+o}(pTcA?e
z=cz68-}fJ#^Y4MVH20^c$57{{@4LBkk5wfOo!vy3Y_jG?!iDtA4BxxB3CZ;hq_1}j
zJi^oR>Kv7Tn-{dRZKRa>rNbYdB|o9a{#i8IfkY~n4PCR?Kt(Fzn;!05DjL?dvYfsR
z;h1qNXYV(X{ZFK^CQ{>E)x$&DpCCd+qlOV615<u509c8Z<>D_9dEHn2@qb7BU(xy^
zL;N8@Y5Z4+;t15(Nw7C7i)=b`M&b}QD-aBA#*8;S7!nfRn=9<=rMxewrwI}g92QE&
zRT~PNCO5FU6*oUfT(duGpiT#~9ugi7Q;m5JYh;19lB-okODPLyv1U)k!rz)2v=n_X
z8GyutrbLp<*2)i#L^DFqU~$?LQDbbGXxNB^6o>3Wyx~9!L-EAhwGU_m-G=hStHTzq
zS%NSG_eH#6ZxuaR>X9hfy2#cj^o_Y8sH2GWi1dSf#;v$xXY=Yt3`E39Z)ooF1|^`8
zd~rKQK7jgK7Db!)k(*<XFkFH0KN9-@*brQ)hJ>I6q;C?!eiC!F`TD}0QK5?}ovG1>
z8bw-A+%v`aOyU}DQMEiA^>*$WW^4SjOhf3qpx4mNlveU6llsk+I1)NyYOxMF#yTrs
zDt-YgeA!7k`j^@@P0mJi^yE2wJtyi$7T31-F@eah9PjdoZ?_7?5Y-EDDS9w~IP$_Z
z><HV~NUibDwgklpTlToMKbb=ei+WqE^AK^zPmA#swqganHhbl75vn%V2(48A4i!F{
zS5RU>3#|Fg<of<Y-=Q<fJR0!{^xF5tc@Jppy%#w*Jw0y{NfO*2$CavVuA|fP>Y86F
zztZ^NCN9H9AqvfQ|C`aJObf7C2hXMw{v6{sK37D=dUR*lo7zj!{O%Vaj6SE#2tZ_z
zkN-~+#MBe40*YyOOVIW5EfTlVPtX!t(i&OojnM_+Tc-Fo*Zv;E^u5ZR!#PDM_PiaJ
z{$mvI-v~?6C!0;8T|IeY&*GebFq6G5*Xfe?bgR%V{$EwrxQN~YAd9hQl;cC06)saO
zVZFn>*<GXf|HIci2S*xq-=gW*wry)-+qP|69ox2(iJeTWiEV3wiS0~q^ZmYatIoNn
z>i+e1Rd>Ht{r2<h{p_{Y-iui@Qkt638mX$*hEZ6a(L^HQ7qL^R-ysH`>Z!^3o8+tC
zC(w}dl<LT$Yd(hc^baqMuI}dQ6ou);Gk9anPqylcN=C)=n?tD@tGfSTi{MKC3kIC8
zrNx<rzfdaooYX|w7^9>#(gU|4L#PEp8Il?|U1ERt(fNsY@wc~sl<FbP|J=+WIis%V
znZEgu4ek_NEvTP=yj;e3`?)|6K!~o?C;vq!coj)ab^0fc(5HhcO|JlHGmp;PT@1?a
z!{6KAabmB0cNmS)wz=LNZ8iJI@9PTb9_ilhh^R;6Ju*yI$}jSpTLMCet!75MF@!T1
z*d^<g=?hfS(DB3=@{M&<C}FwA&dek_kaAXgAL>}eq97!nB%GNA<G)ARFAJ$&aGmjc
zjQKyqn5Mk%%-mrt+@Xj-RW7(~f`&+K|1T_9Bs@ND`T7ED`@V0VTvBNk0S{5)O+UU)
zXfjDM4XzN|e+1|<9K$|~shuBEP0jrdI#8Ac-4rzK`%Lm<{-++{-N~x&GZ`&aE?kWo
zE|pO;>JXtz$Thpz7=pv_Y-^i2O34kAw0ZT9zmSvssc!qmyORoCRKx~%0{v^H1P}1$
zREWxc6qf0rB7x{q7F@!O2q?sC5Sf<9uFiX5gVU*$_!Pj95c>=aEX0aJj4VGTVK|JI
zYes7&;T)X;*rvlo3Z!USwK>?}?R3*p?K=jEQ^Vx(;dCUqir{cstM$Dw+7r?4&wmUe
zA{Z?o?IIWt0?X`b!jlYcQ2H9MV^B*xzZbyh`N0i{qGxtTNri@`VaNIm+zf(>?_=Ac
zPvw4IdnW>8;C;jvxbM}cF)D20gID{YH{^)timG0bvq7t}2S)v>%|ycJjCyNte()Tn
zEY>-RLDQH`v(iU@TcKH0>eytwXZa0pg-=o;I%K2a!IX`%xQxMBad|G;`?phYj=nk-
znH-4DE@-)$-YI)Z4)t>l*?LOUab)wp{HE0PQ>h1foGp`2T@@W$ICH_{w3&02@J<r0
zm%s%49~=+*Z{#NUSrfZ5VJW6)!A~o}`^bE&m~kc)q#X~ZnPI^#tXcYV&hc{sTC5~$
zD*RQ#ICaTYiSJEnLa<#~Y8i@pj^m|moevC}PnLb{!uRs|f2(mAU?h<n6i%n~zUh7#
zs$`hnz~lk1RQD9Pud_>tXt2$4UIZ3vB~Y>tYv#erc8q)@i;xC2OB6AZ(7IV02G$dM
zax=jM5EY+p<;>+2TVCZrI-TA96kD#-s#m=nCMw8Y7O`~vUw-<{b%}<t>8&jDH$e#B
zDf`*9g?27U0kF9hbx-i7fR%X6rL!5-3(##ej#vfyqbqQb!PARFEeDDs0sz*G5mg05
zSwf3!GAe$D+=RyyBPSDVo`wt<lBhYx#3m;&`R~b(33ha>MX!D(#=?e5i(@Y3N+5A^
z%$ELcgO6b>(hbri#VhbQQQM^Aud8yw*b#Rw)(>EUWPGp}<3h496t0b+D^7J0@ndo@
zi+2-U&PtxP0*+oR|1>o;GxQBFv=aolmm(W3U4KtN$1b`xZ<U4o&-4z;!F_lk32S~e
zAM{eJDaW%>vFAZcsoaBb87NvE#(%jSHL>#g)uvE6xDqP_I$$`dS{93`IA~rCyF&g2
zZ>&IzlaldSoh_kS1@e|p@kGmkN~F(rY|&sl&Pb+6ro$#-zaLTGmt9c}lC%#&;%=Ol
z9s_FnoX26va^6NZ@<nISusuTd)HPh0krN@A5Aa)lDP}3=&==wXCh3L>5&<gyp^0i0
zrPGYJwB?oSTNUJhTxmdGDg_NRyf!hf1#lhB&Qc0wKYRkRYN>M>EeIL>yOw(D=m+nB
zFTb!y)CYfM`!feKa&JFVJ<Infd~HdLhcb=kFS3uTuNqg(@rh>psqB?g3fXjqJTFTJ
zf$QzPF@Ze<g3XjaS+aYCsb+BGpd;BBPz=Hw{!8yh;Y9AlE>^VgHd?j(PlI644N(8w
zoF9LSvc?E-XGVflJRR`lwtRJb33}Ehh|$oVlg%Rj9Dwon<43uO5cjpHT+jkdF}qcP
zTF;lQe-N_Xr}!LS@nRPELDg$6TxZk?*7iFDfj2_hFvVMvSv1vGw3Pp~h;Ajjf{>N!
zpXCT1y}k)Ya1W{Rcx0Atq)L{zeqbUybEqR77y{_K*LW|uXI!eT!A%MrO`6$av``}Y
z_gK1AWbs5;<iksF?plp1cjq?pM56@)BiCSizSrWNga5`?(|;(D49L6$S|n2yFYD+u
zt|TOmsgO4B5Mm&tH{hgiFsI+M2=OH1Ic#qUeuv7+mbSb%?B@ovh_?da(EeU7w(h{+
z;vv5!hNdgs*Vvz69zX7u+`zNT;U;F>XUaQ)3(5V6qVYPX!A_~^={Qp3OHOiZ1Sh4}
zIi9jifyZ<b9%viyH?`SZi>AW{1I+1QheXO37ls%Vg@Efq8P<5j<c$P{<}lYIRX-$#
z!Oa0QPJ^IRq33aPGXTK*-bR);J^ZGhCBZl^`H+kaYYAw)O`+O)*`hv>hn#FKT^ibq
zZCfLpZ=o$V8schgi;$r>V<R;qok?-zp0)aw?poOlfmVJ$W^N8*Z{JnYv5~0YB8dU}
zxEq)ILde_nkq&?3QIIw5aMW~<>otcwn@cSQYti9Pu|!CaCsKhh=7W=~#Hlc31g;~X
z?4?tPH-Ahak|haoZ>eJ54cDA*a-L4^4k$eOu}gW4PkVJ`vm}M+=DhY5;*qV?P$Hzk
z|2blU%5Q3R9Y-P(wSXt;Zg+oKqk(u2CjPcY(-!l8o)B@ug9ygj8hMDL9khqHV=4+T
zwfY6a6~8AQ8R7^$wCHK+{p=)<4xokRI%kbu$5lLdOI5jx?vlaar)^B%IaH-Vh7`fZ
z1~8#tNtzOJ3+^KWW|C0N`S$bG2{cVh6fB99GyNCrTR@Cus2HKRQtgi~sKu1W@w&JO
z`$B&Sg*<#O^@9MbbQZ66Sy}}OzRIRze<#3h3UeiY8ZvM#3Do1_R!wA#D_SAkX@i)V
zkSANBZ2f{09Cte1Jk?JxX--^;_noW6!^lEtHVVC@E@wm?fE;o8@w7aRsqt}X7g0zx
z%=6EmkDGaih8R@wG2D-fi60_~gqbp$g+{;;_vQBU-D}<w*24*H)!D~quJsX8BRJ+s
zN4Siu{u7w|0Sbg_LF4Kb3Pp%_G}m|TzS`1tvMV)%=j!cGKzLu9xP8<uj6S#Wj(7Yv
zVh}dp)O_9FN44zpSW(#9h*bEy`quuT(7bJ`uNw%hEVg$~hKRyPl)=N`QhDB<$27&V
z$*1;*nUB|0;cxZtM1R=USch__&^?AO;bL)!xfmeG-}27_?iuTaqeRo{fg#DOY9+j!
zgN7uyRIBL$%&cAv@c8DCDCu!E@CpbeI4C0Y$u|Fxu)bf@<ax~8TNZ0ue%PdDl)M`C
zOSC^HV~UZ36!Q_qv?xF?T(Hdfs$*_!$@ZJPGf^&egH1v(-D36NtXJI1H?VIZuXt_g
zoYaBf$!g^uFWC)e=&ex8l%19lC9Y4<0uP%2MKL8sG5o33Ttq7ePyzH8r9IIg9Dj!9
zEX=t3qP<eKn<C=cUkHDb{_}$)O~V32xVLGtU4M^douM<?5PNHOJo>ts(y@O`6mE=Z
zE!G#Zqbp^8HS|HSQ<uQ$4wr8Q_AvCvVE`S#tx%RIco)Gy>E^Q<8aG#?RO**EK;tYX
z9Sr}^KYzR|1s;#{24ZQK_G9?ZTY69L&WwAcyGl*t?r0R)Vjd{zP3zA>uw#Z5At7kL
zs6Dt{UT6}M22)E%>2gLxCTcn6dSc`~O90CL2Ynx#q?I!4JN+IoW)$nDIgr5dw%N>@
z_Az*HOPjbqIeOY()O`!UNwqeT!gR)ewWjqk6M*7_PY_Lt`pAW939mL?+uJjjjHj8~
z<()W}1LMch2!pbiBb^Iij(}53a+5C$47%o~JPD^LUxnytP$@+##7GA@0OTchk_eO0
zM=A&V1roSMkOTu`&zGrWD9RVbUJ#j_Eiw*A^sv23{0Ee1VzAf@b`A7<kP)B~x4+h2
zE1uVJ_ukiZxkci?_a4mK<$sQ*mvGFDiPjxfvGZb{TOItAKSpPFeiI|$q=&;w{jOVR
z7#-HQfm+x@x@n@Qr@+w-NU$m-NuDd{GcktT#~ZN5uLYjIqOx$_py$-}S2p}uiU71~
zR_NDp$X+UZQ!n${5Q3~7lx^+#u1Q;bw0{y!xIK+q<))G=fBJUo^h_HJ7A>QMD#FN~
zUP+4vYPwD1M?%rcN*fxEU|NbHBbg<PAR~dak`N+C%(2`Cauqf_dKhC&q}KtvMV+js
z!Rmm6pg@?qzNH(7lhAJ;7}#5%k9@S7XN}D5^NfGfq6~+GpuW95p($RqG6tDJc5bYg
zzSx<rx9-g@sE8ten_NML6;KO+bY7K)^?!dbmusr$1TPXJKXhdvx;WQz+QcnVPpRq)
z4^uJSQ)x2bO|WfG9esuI!{fU;Lw3g5Vi>||(E&9#d#(`0(idA%5gs5~W8SAK@_24T
zg`s#NOII_WP1e!wB6ha?Xm)zVZ;1H?5Y~RRXC5+b5DT;sdO?UAU%Zsxs%=cqRA&C%
z;!3|{MC{P^)4H6)i8Cz4k70w^e4*zk+p5zfOH(~Td~m|*m$F}<0#c2FGb>{Jg?H$S
zU<Lkbc&y~*9hhMPT+jq>JD#?O6425cf^`XE1ZO{nt{P0Luf)QO5$r*DJr|B2U)fzz
zCBtt!vo$Wv=&4UD5kTb?9YuiN)!!yur2nUG834}xQxl4*{)7m4P7*tqv^=GHqwhO2
z+@t%5IWL+ms=u45fc7WqYI#yJ|0tU90ScBeXBIy(K7_aPa1L-w-+;A*3ZGNVCtn%~
z*Xf<i?);fLjD9BZV(k}6@&2WuumR&IsT9GWJ5<j{T>F_UhuJ-Y&{V~QGcXbUq*M7*
zkD4kVsJg(YD5X9qcy)zHm%3`NHKQCI_I2p44TMSfouQ4`OUGJcv5U?I%W+ay2>1s!
z+@IPDN24mYo9G3iD!iM>O!YWO2Fv!HYNg`Jphpc_?{?BSSn<|&FnT~{a~mnq3vNoo
zOA;>+?=#t?n|XM7<Z6o(!XfzD98ui+%N-;+jfqxaFfzl*`_kPlMZYZ(L_f2*M-FAT
ztX|FLXT2c1wa0+~-Wsr==xU=~1-S1xM1-Te@?2>H>~mJ%q)<FVW6ebUFu>g_AczZ3
z*Qn3U%}=NO<ns-7p86{4>fk12`E93`Wv8^y1f$bJl`~29^rN^pQ=FD<d4*t7Y=yw3
zfL`sik<}jx4f4?L9Ok3yJ3%H{;28Xz1<gJouX1*qBo{Q*q^nz*e7!GX22R|r@3BF)
zK#nWQ-EB_$P&EZAXL^&VSgwDjrpn~0<Vi2|>OYQoqxpHs4HRF2t@M`6_Bkq57bDh%
zv4d4}UQYQr7ClX|!pmwX#wc31c`iY8gBy^G)LAjI4E_0VA9r|EWYs0XNBE|d@>OgN
z17t8w)^t<Zhw=^MSNUAdfTe;)=}CVfRE9<}({&Hi6(}0ky)2T1#C{}AOpGO7eW;kf
zav+_NIZ#X#k)IC+iN=b`CW9<2*v8OA?OC_vzgy!6KKuTWCyjkw0v~Gz;a2_ojmRvr
zv;nT@*Blsk1HB-T3$$A#L^R{f|A41?Ld7-y3vKM!9|R+4l7vzu0+t4&BtA;#qFfFP
zUs4_Uc%{(KCH|AfwMw6|FRDD-ctc{UjJdKqL)PMSS#CrTaKXnWrU0X*F#)0eSBk@|
z6#gs3Hqo_1`R}W@4pHqHWtQ96xNQ8(>?FYzgPGQIjR0Y*J&?ol^x_Mlf|whF3vLZ~
zV`9zJAd_MGGe-BY3iwgV%e78xZ8XG76A>k#uZ$2AEUSdHP7F2={eLOStOl{0$RVf^
zmGIS^*H%>L@#{|z9d83&LMc*&gUlk(1&F3@K|kHqi*_Nn06t_fKBCRB@;<)}-9eU<
zisnL92UQ6Kz=V83WQ%(c?|w&rYAqE5As!)3;7LkZYWKi)Wd6!jp3hv|a@MEvOsmKP
zWEf~LxCKRS?W9>&*6C?1A)zVgioEVCp_koHl@w4u#0dcekg2?}wPw>(I+5>v3cnmN
zsQq6s8S9pcD@KW6i7*jsKs6w=)T4~l&G9tCkT4-M%(K+H{qtt&Iwe}02F#R!!DAI6
zCv|Ki?5T@tDO+N+E`Lo;Kxys+p1L{P6d4TRH)ela>jDsEYS*=JP59g3uY99ir!hZd
z`Z>U=dP=YS(qI3WtH3brrPk!`qgaH3!ii@Vjt9C&ux2wsWwBg@aG^y$qBe4lX5~ad
z6j35iQeh%-_}eguP9ujs;oA2CtylS&l{F|CSHVGNvj1}?@pW+x)oZZ>*0ur@dGahu
zg!?fpF~^AQ2M5drD4)pfgpCog>Yx!dr;Kso!X%Xrp+HK$bONdl=j)%ZMj0qcV_V6a
zRY#6csr{ASs?Vu5>Bq!}rDSCrdhSvsbEB>L;}zDth-1;u4&SX_UfF<WdIp7`#{iYj
z_1;c<Bql*aBJ=*@Ko)>0=!n@5)4P2hKQ#mJSQu<HW=bQuwjY#6Kfr(C&6`0p5IZPH
zzRE7!mYb@_hh`<;8wGRrv6#E+P<)-kF}=~7lcih<{?y6u$DX_Uq?)@cJTfdg8ZHWO
zYGuz=*Li!LG+(%Mks$%bQTL@lCkbhg0vb?^$oeI~lf`Ap8s;U$N~8NF0^>mNNC@7H
z$eL`)Fm3v*;=YEmyP6P2y0BXixwn^MUxy62wQ{%cG86y{V{*6~8tL!dZQjxPab~mo
zlJ5tXd}sfIz(-5|;f<AiSAkC`1i0_>;ZlFK_uy0WKa_2M0DJt4?3~Z~VfXTTE;VJh
zJ$wb}10k3HD;Lw54k?u-IsMo?XGEZP`2U06=Ow}aZ|L1Rx3*DL!txzE>sv0yh@k}<
zrdN$+1!nE&xs1abMrvuaI@|{QZY8Mt|F#k=FjXf?s{fQ)WD@tlmv=xiBq05x$`!I$
zt-f(N5gd>&InERfCI;YUl%MCZ|5<NSXTIo-gXoa9Tol7bt`ONY8<X%qT{T5VB@^V5
z*XV0;X1;;J(cX?2s-)RQL^>9Zw6;v}ml~bRQosmKzfsc)PZ;cM%<ykmH-TToTG##s
zqUEcgr~XnwcYNW44ukI}))X=*=s|{So`KOpizcr^_}VYu+FU&*5?Eb(lKJGwgZ$I4
zfOC0b7#L>ABHGJ)l2K?tl~4YMPM61cw1*DMQ7}>Hrbsott3b<+@kjSeLt4mr6m9Yf
zCUD~1vZe)Dwnws7-w|I1ocOj@esdIQO&>dhFJ$koEY(Q;Q`UeB8j}byEjc+XQaM{v
zu}zT%v96{>&%*<wSxTl7@@VRuY{lCPJ)meF3vX0W<TrT6ARK6t=Sy(E9zl>~Uc}3(
zW8UZV^7K#DhxbV9`_fx6tjF-~_(HiK%|MM1e<^hjk=BEAjW8=VOP&U+lHVk#0SJN<
zI>+YFgbeyaqNToK*Krbuj(6_Y*6JBAW?6|h+uAqUcd+f{oU039txIbb=Ar$4<#&ct
z-=ycJM<FuFQ6;+cxjMU7k-G|CbzkenPvPENKPcI%m2o2byS|CHKJ!#rpUP;NFJSQ(
zyv$*Wg!gC6!?Ubn*{djie%qL+kO`3%p+yIg{{AN9tf*Ll|38C@eAT*YWh4|i2*YOn
z3y$pU2fS7!@*Oqc{P~Xmn;5>g(v}UKLoYwQIZRp=UvCfnV>lQ)DGC03DCA}W=TOKg
z*f64(s?kpOMvA^nlV$)OGb(n6DNq7*JkSilfcOK(HQQFzfvytNh_1nXWaaV>4I3tQ
zi-3%(^Z3B0V~IvxIWrHVkZR8L>%p&eaLHnKMdxeN>IvmPpAuHP9|83vHgA;1+1Myc
z$q;i+3Ou4WsjP0H02i!@*<v|Xyq63mpKVByjV1|8lY;(G2n_fUEnq|sq!=k&9avw>
zZwB}V3Reee0BW0JP}{YHnhS8bh&>vkF3(%ux}R-D46*riao&F7y+tTRZ%N|FUfVXr
zTo`^nd!HD(@QXM*{Tg8YEXe#s#w$d9kbEGRBkEWtvSyuxP-`P_p!tZ{?UoibxV1Rz
z(js<E{jnvQQo9Jgeb}S{^kHL@%(}5@=;&wfjG)4@KgCAQTDfE;b`Chi7OxV4$sE)`
znk%6HdVa=P4<dlSEul8oa)2=di+>_E&d#+8IB2vuNmn1~qwM6bK_C#cuW>6Ss9kU0
zQYVVik4?p`l^vct*8`rb)9&{D2%ry*zk?4Boj8(#d50YSA6^AW*DI<BhDm=<@bW$;
zhOq6Q8E^dq`Tzw_F^Ak>*!r8@!P<r^9I3+)J8b+l1nSz-qM8_}k$qblx@HT{B6IpA
z-t5{~4OZtFbik{&a+%R&e43Ayd+?ZJ{UvdKbPvxpozPsvE1&emRP1p4jz)XCXKE9m
zL@;&?a{rD3p$YWV{|F20a;r7uVW#`Kw_(@Sxo=|p0=Pk01xm)pDEFyXub!!!y`5-3
zBl|?C4ExW7n++J+A0^;P=6MtgC-kN?A0^ojB*l^J!aAn60v>L?_LtzdvA7G3RL5Ag
zVS}=cjekbU*6)5TNNWCwoa0Z1AiEIMKLfuA*wWRA`UM1@M!7#Id@XJt?@Tki`8{7^
zy8Bjy2zWYt4+5+&j+wEC(tqMD@Qs|BwEyXwH}7-JDf!lhF8+zJ;()sOI=XzHH*Qz5
zLG(oQq<!%%%7ng?i$ft%=!r|mM{#VENQtB8{Ux{Nn>XTIjpd3xG@Li3OBp|iWB(^|
ztwtm98W}hWP!3W2T8a=o{8O;D%)ma9F9{zex$|7W(l&m)Lux3l&y{=-L{s-=fZ$jR
zm6yL9)%%+nDGWVvdE6Rv2tQ5LpIE(#5U2BpF5yJ94pvk%8eVw;Nau&>F9{RrV&gK4
zr06Wx3CMAMdi`2tuJu4Ubgr>&d-gmx1&`}Gdl;bS*c=A2=uhErL{>MIDyp$L`&Fh8
z_&n0PH2-kM#;dArd;Ef2*$D5U5KjEbqT*SL<BFXc&$9tJ2j*V|omt5w%}Gi=_{0f(
zfkBgH_lroLV%1(m0;g+lPJgg#q7%B=KYWWt`1V!3pL<;eA8#Veb)H@t*7@zOB#JC}
zzvEwRf!SL8(RRwq^G*-Ze!(@6)27yruSyO|4_S{U9K{SLzcpavgv>|z^OnjNm7=o|
zEd>c>ck0n;eYDT7#Vgs|O<^Ws?7&<lzuQ*9tJha>j(@RzbvwdWqf?Hehd6a@vZKbg
z$lH9`ByB^bH$&CQL*8KfX1btLQ6yh-Q52YNeQ!z^RU%Ti_bPyaio!CGZ?AkKwO0C1
zm5G&Vsjp-gE=SXVh=5`y0l#D{W9-63$nAQ?g@Ss!uEhS58#F3jwuKL$OUQJ)Gy)&V
zunz|XRhJjkXA@~k&)q-P)^0$X-iaai05OpclDkl!3qX`@3ers+2(~gjkPY*LHw<WW
zgE=M6;r(1es51{i+8%XJhQu3)Efh<k&}RzZ4J3fM1t5b(Hg0WZdNP0qfC!R)LMnk(
zBsG*9^({@2P$L<Q>5P<dI(I`Rf=e|;ljDA0P|2eH#E!=EmyGA6^8}KfouObJ2eW~`
zOSx+Xk@P3JvqQI63njNofnL!j0?8Fn|I|UU13WRPo1{=~L<4c@P^*Oyi6vw4o853;
ze)cC|yo0Qwzn~nC@*#j8g1C^=Af06eI|Bl0_Mot__9SzO{vvPoTAyf4<Gh3635SRm
zLMxdikwSVabW-4Of#OFwo2li`lMlkIhqqZQq6DOBr9Cs$77Csoj9<%k0h{t$&X4(5
z;@H@+7C88ruJ2C7Sj6w9MixrYmO=bVO3)~ljVF;5Q{(dR@NcrEv`>A8u8+yb^18Tf
z9$XZ1iLA|R8Wl(FzZf?n@W+(59{(US4eCI?0Nyic61)wel?a650&!h668nVgL4NGp
zkS&OTvO_hH8A*m>b9zH;0Kr~RRtb{~Pz(732>`sLLk;6}bA<f~?N5=da3CLm_ed`y
zSDVDWF#CLDox^|6BMC%87=^IrLN3=RN|gPHVy0+ymUg*!)KHP2V$$R?N^8(!>#WL$
zf5c?Ek|HHhi^ASbKoBvlhTAn_FO?ylVP6J_W)}GS5-v$RZf~kTfq`9&D+Oy%Se%?p
zV$QhKIK(JFWn5$lmDs-T&4wPg@70F$KX;zsiv}wOQN|SM^0Lj(%0c}jeUpGrrpitY
z0Yov=oHH(BL2Qo}ijRaswvBCehQ&G|PWmHj;|(^31$Yg!;KaiL57gr{PH3%I=NbV!
zq!9m{p64fKLIK{<*qaNjV?n)GHD_KVMW`k=i+}E}-8r>{fl7I@EWMc@zu}#@w<Zy#
zmE`&1Vz_y5o+py<5Nodm+z?IP%hz0#OVl*WkaEX=2jnMlRPN&I8ev3vTOvytL~E%=
z3f31dK4+V7!m%&0XFolr3{q-A$(+eh;>pLo<bQu3-hdxW{%-CeGV1Kv8{v*ToVsyU
z#S5nk<vd+P&*>B&PfYwPD>XqpABvs@zCV7QoZ+1njzDx?y4Cw~xINcKoamd8ULjcW
zbvVM$A8k*+UM^?kyr>-(OqH!(d!)#4OIvodrTN|DNpsK10oQ<QmW|(?m83Il<`QDu
zk8Ls-D8TyBsv@?li_#~767Umu+|i;i+Fl+q#w)X(wsk?eGRHLl$6<Ng{{-tofW*?%
zqTu%u^M0Z7aEFU)JU&HQBfI`Li@ka1x`#tc@C@h<b9ry&zy~7W11aojxjrGB-Z`A!
z2X3K8iSU8nFWB@*E%#pN5SX0b=Xq0os(_*fP)$yOzemT4#W|L;HFfQbnUMI-Lp5~K
z1)*zM2i=Q?Uo2v(x1QU%1nu$)q<#X2V;!cul9HguudwNwv~WFO$0&%XI0S5CStmSB
z6*RpgCc5~PBspp61(hTb#(wZ9Ny;#15dBPvEVr|qE)Adh#}x#5cakVM`+JW#K>)~9
z*wGPbzSLJV>$&To{42so^RQRoZL8zb@S`W1k>?kI+xoTN6$z^%{Vria>iZ%t>zPdz
z-B^X=Dt};XC+Qu<7bfgn5sqnh9vL{mC5m`=V<50_%3S3yNd3f3RWl#YZmoJAfgrSO
z$$1Rn+M<8E(mffQ@pI*O5(O<OmIiW!idTX4rims9$4!=w01Tj`Tz({mCQ`d4y0`~a
zv4V@A!x3qHe@!uIv7dxdWw%g7Pfchbd``-Jip(h@7iI1*G9jQld9Yf^%ZzJ<j}O*B
z$g8FICP;ic0&AEzLDUQU4+?Z!XE4H_q_a%PJw$G@!f6t|WE2Ay#k&j7UEmX#svZ(D
zKZOz%^4#yDwfTJFS=vdqs0jIb3q^*QbmH)YT%=AJ%3k?FIp`JGy~RfSllR!yoUU_Q
zp&smfKVc=P<^sE?q?pl8YfB`7b`YExK1WM2{2=xG{w_B|kPjvk??(iEm>UYU^2{>d
z0HRINi~T>E9{^AN8o@>+N}w;!I#pN@GPe`kQaHgo{0&?GY+x91D~ra@Zjp3J!y)3E
zy0X0Y!*qVq^q0?`+XG4WB5j=hlp?#EhBy+yvV`qs@xfn`wPJp%hG%3`Gicr+>}G7!
zzD9_bJe<il0TZ~n<A1Bi>ls;qv-M&^Fg=uO?I9MU)KpHH`f$yb6(GOE52o3&-+mE0
zmizBxEYVATBf=8gvHCA-Vl<;f2I7&N^nr(}PHyWqhE|0gi=|pq--)mge(o|Oo4aOw
z-<!NkWpY@9M{Xx~1EBpqWocSDEB}tHz9yMVQdn8uu9E?SF~&IOVXX8;;6jPM5`tRI
zv?7x}2;>MZ9w8m|7x3DvA}K317Z#cPkVk6Gss}_&U0T`vT$;Gz6-T9sOs4GDcXA6y
zHugsG4PcLx?0gj${(wy)5c9wRKqTqML!3Nb8WMxyWlQoH+jwc_Fb063ppyl|R(du3
z*lRFnDF@}ULqH}*E+iNPRC+n_abXZ+6!eh@(-MkdsNB910T=Gny2}-42}yaK<esUN
zkiRXQF69E7E{71uSH2rrtXDpoe3v?`04NKQ#D66TNi$2rV!IBP!kr9?sf`DICh6tQ
z%xLwR931X*K2}HLz@0)Ha~(9A%GSl9^3sR}DU;8r$PwH=OfGSLaSxk=wRId>iG|UG
zrz%+W(v7f70ke;lpR_AGb&^x;RMV^mVY7e@yhfg|mYx=HS<%Xuk{1upNJ<_bozkUf
zl!PiSv_khC+sMLgGZsWhJ5H%|C*khY09hJsq`-XqvP?~gyeT1#6fZ#DlnYH2ZK)Rr
zbyJF1Ah9L{Xet7JgFh@jk^%%YstW;m{}T=Jav)|h(DHB@P3^=CGHLr$0+5tPE+lN0
zVG0ONTlz6gmdDJ|P51qtn2}hZf`yTz$KA<Z!#K52!!7~1RfdxgE#ZIeCpCQRD3uru
zr5vW__Zb16H`ULwy;~w`Mg{ZzG-=YJBh0EVa>EoXx=29XSVNpXYr&{hQbF>jP@(Yk
zLV>Yi7omy0DLBg#bK`O<Pr_8qTk#dl!-;dskjZ}?IfP&*1l(9thOUJE5?vs!<(X^{
zoS1ot5!oe$4TwvDOe%z})`NrmMYU#eRg<_*<Q{5s#O<Vr;rU#OLlO&$11u;6G?_9b
zV@+YG4vc1Kaca2@CMB*_?iwwFV@i<^W&k0;gLFU29sHX58UH$B`Kumz8AaPxh|{{<
zYW=WxbM^GkV>sFKw4gix8cB~3F2GVKaBR^9Q_6sT#NLbcnk9_~l$5q8PoVFNDE9zv
zK)aD8?dak&%UF`m`N;ADg~{MT_&7Qcn`)cv7O$Ao)j8?6;i}5#)^TWG-}(p3<bfXd
zAxG)Z0o&EebGBp|FWE<pS6zRZ6hdg5Dp9#iztE;C)RH8x8i=~%JuC5g4}1<G(`_a^
zn$aqv?7Wk^l?Gf;sHH>1#%}lwC(FhRa*5hz6fM~ICc7?UERKsu_tHh3)<%>|^!7<r
ztBRAaI^Sfa|C)m;J=xItQs$EOAOqFkMGit-n8=wH9=RUDkG~Is4T%;4HJl6ZT@Cxk
zIv-M$bc8BF=>l0=D!<0{wn0UB{Hm4|a1uIYFHk(NekZj2Ss3tWu{gL>Djr~Ut;P=A
z<FB~p2kEB$-L;2llSoXD{9XwoV$UL6AsmcM8o_}Yic1b%7RxNOPFHu4<peCJgyvAX
z3Oj_D-~9gu$MSN-umtGJ-^EU5(r3)I+RlNjrHJ=M<a<R3ehWpaa6rIKfmRY<{2f
z%ythOH{?zc1i{==Irr$DB^I_j+oW^_-?0{oI-eBQ7KW5F6AX1vsCx9m@gx>H_s-i3
z9Snye_5peE=G#c55Q63axJIG{P0Tp!@d@veu_t4%Pus$2`h{W7;SP(>P+Vp8QpM0M
zypmX#-j=5G3_sPc2{}>@Nqu~0K&=zu7_ppDpbz|aY}ypF;TdQ~Qs^<*^$%|ww3i4C
zcD56VM}S#ImP5^w<(CIg7wM5jbE7G{$`@=KJir2N&J-4sD5-cpw<2$yZ~VuJdpRXK
z;a|JEn_5Pm@QI-En5I$hkTz#ISr<uF=>90A!xC2QzW@kUStep_rKD?OmrX^J>m((E
zvDc{WBys%Ue&-XiMNxSK1Uc5LqytD#i60SEolq=DS+kW5Bm=|h4)?KE(pLaC|8b4j
zOyI1XD9y&PR;bcDJAVnje)B!VXgd9-`U}8nykIFkz@;)O3Wa`<=apg|NWTAa(XMDE
zW}^9E_m(jBZ9tc$RqR6CA!7B5XyUIp<u7^XQ*&0WD;>&|NebI6ImV2?99HRdZzb3L
zsJ5S0oVs>~<v!=r5+N1gdn(4o**+yZbwJsc-qrxM5Zv~jEoaQkNr5N7c^@dk;Zc8a
zz~p=5-oA?}GAfgMN#M|Hl0+4%DA=TTq4!}^*d^3{dH(eBB&vT^X5UOV8M~LTsDg7M
zR$>N(_v<HFt7~0n@*m@n_wFwh`lvO|jHM*#ef897N@seLFdW;ZtE+6{X^lLB1RyB*
zI<K$}OR&`|@tAauJuYbQ;b?jGcZ2l|EpEFE$+g;i0<Fozt3%sZ%qmsA&?vQZ5l5BC
z03Lu81O+y4ZSR~e$#bQF81xd{Mr@Y`nj<xXMk=6&Om!gJ+^31zATICng+v?rE}70d
z$c)6xJS-z}$jn&5MHqSPIE?t#3}~#-G@eL51rRRg<Y3U2+RdVX!%L+9o-<4h&P_@T
ziVTQMG*MGY3EOD1Ez?v2vngZKfU?a``9~Wo+EiHgG8y8$&?I#FG>Ns;^A^h3_n>_J
z<x%%N+gH=jL8d>nn3-hm_0<UU<RkX+GyRV`5Ke6$Uzke4TRiYtE4nTg4w&cl9pid*
zgh2e#`U^NVKyTEc3?Ga+d$Mas(S`ngWDG#eVE!vLf+ja`Us{!S4K%B!Sf6M0eL$o*
zeJARNOY<(;Wa4m@`h!_o_`xQ;O;>vb*b#ZV#(lXKZ@LqP#e4TW!r*zi4okhTI?^rv
zstfw6<7DawGUazdd0{;70xJm%nj{|QkAAGg37k$pa>@nfYf8;BIpE3;b^BMgBCZbW
zZRLh@TLt^mz8zaJO(JW!#8RH{PCi47x^J%IRKY=HIDrYma_+o-rG+N+HY5HK9v?3{
zTtL`7XMv)~Afuq#ekDI=U+uLbvdNaq&M8kx6QSwEmPzo_`8lQO0mP1R%3nMK>k@jq
zOPc^cO+oqnxPV}%G*BjWf)gNFinqs{y|rh1n!h{xgxuxzq7LlW$2wQM2(hMO86kK!
zH_Kpj(f5RSRCkNWwT(v(XFFUD_e6p_BpzHh>HgSKu4BPUp_*k$-wZH{J{N8Y#W@9s
zN9gdX$?<q{S1eo^0NP-wcX|TqxwGp5>tNZ`g{=%R5o_ZBVV{{ZBM+_|ZjrVFvM&_2
z;qVHUW+-)!0VjI16Hb3%;PLeI3QCZa-XPxHkivL?svV*RlWRw0j1aaX$4o=1wxVKz
zTn*OFO=C3UwZk0Fogw8KPUF#DGFN8<y4uK)KuPU=mN^>@@HXh@`pKAn>;Pr7!o!Oi
zY);s`)cwh6#1y1b7-E+FPz-yrGaRFM&AfvJxvVH5Po}%$x6^r{c%TO1Qi66~$-jwA
z@D=fIj|Ocaky7JZWE{~ss1J*cK9Wz`fIWygVPhV#%}_{i=tYtMEf-RPHt%-i!@}k=
zFkMS6^-l^!U{Qvp8p%(Ks~oJMJbR{R3yAcAG83O>UWy08Zfy#o>8YH!^=kExIahy?
z9mK_$sQcy|<qIK87fDH-v_83g3J7xiK>GZ4D5)b$F>6HC@ZS>))?Od;5k|uO!GQ)F
z2@pYS)Nm!)ZPO&~YO06uGD^y?u6Zle+uN<BgTnqDz@();W~IcXT<J3(eJc;fN+()d
zL}<WOz>LJu_D9QsGv035FzxVkJ!x|#cIf_ti9CPJRJx5)Pe2)w-pU+)l_~V<S}CYD
z<bSrRo+&q>;brU69LX+-@(l4ryN?RMys|qj=uxAA6}7UVPy(OU!%WKJ+F5IJ#rCuV
zt6~kfekB(I0qALwg#*Rl4O5oi{F}&A1w-rMa>XM&j3`A*2CjB-sIB6vq6>w=%S{a4
z{*N0E#q)WQ)&so;RN8)of+jEqLh6pMtAfME?`N2rQXbfqwHM+ti;q$&;9LF&g3^yL
za~X;0YYPzb4z;S)@T}(|wG3+t-Lb&I8*2s5Dg~+bnGJJQb{*+N+yl_iEkum`?gzl<
z{^6W|xbKCTHk*a5QU+f+C-5k}cmF&3cY&k?dh?9fYvm8SSB=!mO}hxDDpzlD&H~Y}
zQgy@G2=&aUdy#_gpXha@-@%U`0-M|e9FN!g7r#?Pp7#ajRm93PblVQ@APno1PFex<
zjobnlEWdjqyg!p_zY_DqeRkc!61+c)$aoy;H2v@K!VtmnpNcI3<shDa!~h*5p!+p2
z<K<c{L2<q4=sSLRuzD}#oL|;#QPK&`UTc4ckb5Zk!eT?k1cal@!2Pqm3;5+Q1*vx7
zgCBtgAUz;H!VFWZJV1q)V*d(jd^YJ!Ilt_ez(eo1lQ-P?Zd-NniTEs+pZu<_ur6t`
z9ScHr<X&!0TxlKxnD{e~D;fQ}scXA>2-leR<8dWEV0-F*ugGFvvpT-h>_MfOG2@%!
z_XgaAnfpG-3^!st08vp!!t?Sxe|6}J$9L_^m@DOfhpaF=CqP~h$DO@ax7lk`-$;GF
z_bS1?x|m0*C*iihF>s(&ymh}E9Ow$ngn>AuGffzLbmp@Wt=DthT;h=uh#wY_<MC^i
z7D8qrHk#gX2c#VcrxWK?7BWEaMKYSj>Q*uX9}QVrX^uFZI6aYg6Kz4ZiiZ5ip@qRq
zG#91*Wn!OG9H=zhR*g07QSDDk7GwJj+TA@@w{Ztbq-&DZchN(T)~AnAX<C$>0=%c!
zv+0})2gx;ztN+B+Hi(OH0`5oVjqfvqxPI&D$qeWvxo3SR=``q~+nVA)KG9H8)T3qX
zY=oKKba0IvSI6-kaL1DMZ-uo%pl1Y18rvh)qEMJFu>3gYvLyY=`$TRjejZC5T@Q`v
zc$uPEusX8O0@xH|CvA$nHE5_le_`WftWi8tSKcALMOEdQaK;dD7cq2CNXU}4VRZ7I
zaC4X>NV+KbiHtV3n`MZ;oLBU0qG@7C$5~0CC)chl+o@clC_u>jbU}L{d7*;6GiMdN
zYVz<7%v&G!uC%dZzE2SErC`k*8xF;@42jaqMWAfQ#htoP@yaG+!_RJh=^iO(7sz@4
zCED8I{3$E?m|&bLa!)u)2`P(@B0%CQxcDFRQ=o$^fW}XCFE;!db#;Zux1!MFQdN_+
z3)9~yx;o~g(pI*+D{0e6J0FM=Xm=D7jTQ;i^f<A=!=MgilvB2fPCh7iFOT&77ni@n
zWBlY+;+^AnZ%B{NH(Pt!rBxk6h58E~|Jb^wOL5n4`*A$)Z{mj2ThG!TUb^jFo{=IC
zKNpqNA>=S4{K&64E1O+3L1!1mgP#;OerXjHlCVdxCVj~d!w>Caw_=|1vyDk?BA?B`
z-@#l=9<IzIge6JXMVSzVXE~Lysx5Wwp7DrXQO6{*Y>wnY&Ra$@XO9jaiKi*{|DdAo
zDPM0g7sQSe*1SXvKRd;gCwSUb&9t;UldcP)pxi0r&AF?56isA~*7fkbFBrrfP$rp@
zD-N^&<iry&v`={`vrpoOx{0klfw{H<S32Hu8HR8e1Fn$wjcD8RcsuP{nOX?eI|%mj
zFB*yFYFKV3Snfxg8(q7IIbz6d28|Yb0gU-v<Pxm*rOnJ>S>%UaLsF(PGr!H~6W$d4
zvMY&&B%`L>38t4~9GXE|H<a!lnTA$+LC!^trsFaMM?B~gT3j<AyH{#Rt{((}+x}PE
zTLF`pRvdnsS%DS<Cp7mPAU2OhD^CE9itC!9t7nJnw;#lFHYN1I_Bzw0Ih<K;&<z-p
z?}Azkdf-Qv>#pWRXsD2^1IfAPWF9NtM&FtGU>~DwazWTaaTM&1YzEpGt~t=jK%a&o
zk?0HlOx5`jCR-J9y0WTC4W1e}>?A{YxYfIECoOAmj;Rw%w-|3eMsna>Adh6J`xC)~
z38@>=1g0~vq^+KooXfaUCIJ1|;0NfjI~YlN?gk8D?P2KOv&hQ}@`1AKY$<LEu^-rC
z)&5h;QfxuLZau^)kj9r|nqF7TheJ89UTDJC(<IeVoh@LzgyT{9gp_K4E_50gn!pOB
zs&ZXxq8L@NPma^@1lnnWD0!kmbrpEgA|rLxTAQL&LMLxjOOE5eiqP5*i7*YpMiA!M
z>BSmKvJJr-Q?TK%6T+)9LJJW3n;`nt4R!MRT-lE<IZ=+(Nk))Ko+#wm7FcSD$H=MC
zI2uQH{E0`(+Q&2Rn~7q;It8l07n#huFt67ByKWC7@H_e}K2f>SPQKAD`O%5vm*Kf^
z=$8EpB@U4yDjC`0zF@?y1=vojyvo8Sd1Yfh2JwF{k-aB(BAP)b6<iyd4IJg)Ch}h}
zHtis<@u{NX<SQdD8OHJvVx6!>F#|R-<{ZmysaV?<&71N~sq>z|%rf-vsi*WhGADrK
zj<SpA&wp1{yIn$tyvM*xzq8u;2^vVQjMYhtkM6~FM7TmpsTKML0$v?0kmNe_k@Txq
ze><ecm<Yp1s4e%#&mjTqc|eBekK_Lwzk7=&KQm@e-mls`f&T8kzS~{byHJi6lW21g
zuS*cRu?fOhX7NDh>+8H(!}?X0r_KMgORDBB(z@mXcomZfR;%jynAJOnOGZwO{lt<*
zHN+pBR?l}X?d!iTeGk<BDT|A`nAiNPd#&E$ve%?N@r<V9hemUn{`0#Y#`sVykriW!
zHiYAVO-fkV(mwGw56&Uy^BWiaSfP)(YEPl{0EL2Q6b(>UD8RKoI1aoBk^`QQ`NVu~
zl6Wo)`hgY$zuVKxM<kS={6t+$BA?7!lMFzB5&qktq*8x^4Bnbn$N<oB;(jgQ-`V*J
zUS1to9;5ym!d7LFoo7D?Fgl83T^f|_K%EC;ciCF!KCy~~T7<G0?~e|~fH0>k!hk4&
zkwA;JKmo>5qTiXv`vks-*tWa({~g>sA<WlnDKix^)A7ubM!P2)rB+GP?MGi(!21}#
z0|E<@dN&7<1r^+8M;|fDSepRP%Fi}a!ldD=UNtzK=4rR}Q4UD9qEzmE_ypFHww#Pe
z6u6a=TQUC!5B{F0*-%xXR+nl7{1A8U1EiepMqke@20~2wK?w-5DvNp&mDjA--A2Yy
zrI|*K(N7e@wSDxXYyJ<$<j{A~eC34&#zwzAX?6a=NZ2=}Qh@RBO?ZdPEHTil(|$GU
zVE%`neiBcnm)ITGM%81)*+Lx=AXLQPKmJEbh^8!0BadpDc;w?W4pg@KU)*$sTHY%W
zJt;+i<y+FsdaUmxsNcVge68M-2f_b|QN{lYpyxSxJKMAm@3fW`_0iQ!1b&d^F*Vd$
zZFy223BA4>e;F4Hf0nF94$VAZB&{rM(yEKGVR7L8X-@RYQhVl!JJ4_(!Mw7GjN#4t
z1+ReuMEu60F7L-JT7&(d5SSOBFw22wze(<H6fe#&aZP6zi5mVlGJ4PUc$@qkZ!b&<
zR;03q$dH2z6K`-5T(W5yJ?P4%G@eNZOOhZeQ77{7(A#Z{tK#HTbVPVNR`l-08rLLF
z88SnSHR>z~rjlHW;s{Suio#wMxJ8ndTI(l1PL_w^=N}M-)bd}#Su#}WBjk@3;&8(B
zjB=0m^-@UImXU*?JKJ)9R$S=90m!!n9lRc6UO_-RS=GqD8|l%N2)I$BFrve<(_OTl
zeH6|0uIIdel@WsUsy(9-(^tl>lEKNRTT`&d;|&CpT}0ehJ-&iNg23G`yCt3VcI_<0
zcH(vwc(O58KHHC$dC7<3_X9-_CK)$NOiChB#GLa#O$oqPx?dJ~L`kEQrN#EExi@~#
z0foNh&h=eL&ovuyGCso8x6T7u8fkyiFk-*mHJ%ql%vTSW5q94z|B)m5u=AT`al*tz
zV#(lRc(9+<v}A=!%K;lK)oTf(%G`js`qRaq1Rn+!o&n#DZn{0qF8UG4P-2rG8Fa#@
zu*<s!<%p<x^z(O4dmwS{ixS~%TW()6)$3NRh-IRJtYuwG$0pyB5eV0_Q<k$y^*lFg
zez;Af&x1SRaY!|p!JK-T-7S5?5TR?a!k~AdfCh`EL_=}gHekah_7Kic2X-5L1=11|
z^`ra0G7H3I(yvh`CJJKScKf}yaf)8FQQN>52WtsoKVd&M0#pn`gdtB^wyO9S7fn~;
zz#|+GeiUiX??6zQJ|@xvP1>CfPOf9*O=zeMfXu)F#{|dBHAB`~_clg^S(O+4Wizu!
zv<CeK01*!)1;(JWjJYnTO#W~J|E4Ut*(Hx~UYf8UhxtLfLmbiy_|xVcGV}`@6_E1n
zWpZ$9QG+W3@!cFoQuQXNrtg`i4Yr@IeTd7ym1Tu)@ze+#RGPC)h~QqW$x;0ndf?M!
zym{Wu+Kbies*|*fNBWA)X0_#Ux-DYaD*_tvNPEnGfktxD;&2Om{=Hk`6k+Jld3Yjp
z72bz}e6Z|tf7oL?-JkZL12cnvF=n}vam&?gNy`tHV87!2<m>z8v=YN7Y9x?j(hvv-
z+a?b|sJb~KbJ~wB6}U2mO`aN1Qxak{ZbmSymihD|1U?+8luKaJBz#XxM}cI3XYMWZ
zN?qv}4Qx`E;4vEX-6YN{l;Q!TT3NvgM+Ku!3q?WxL%>ijt~`zwQNUE#nZoP-zmrX7
z=P_CvV}*RwLtgIsw{C#@zuYbJ1rdFuo8%G?FAGkhCa!lclvq&hM$u!ZL|HWHq-TS>
zuRcJn^i!D7uS^CgN*|qfp-Vb27la4biS9;ovS5FzvKC*Mo>}WH!A@1*4S_jcl3Z;g
z#-G-dpdY!HZGM^EgKMCbtT@8=tx+NYRg_3W<4yiApAWMw(=s7(DbuDvveunT1Ur3~
zg@UNAoXU}tT)t@Q*(ImnoyR^ZzjCcXvk>{36v0jpbbu!qYSZR{W{*zBHUB*{|Fm9M
zT$R{t!(2*xr<{(oH3s(Mdr5T>7)b>Za>KX=L3<Miukk`7Qq7T3gi^pqiPfF_Sp2k?
zou_Ljn5af`k_jRRCK#X8X{M&^Hf9;j+#{kP-+P^BC)A|*Zv#p=7|Oy2y}hu#-3km5
zJ+Ly_4<HD)c8e)Ux3*XlUyxS8JkT91bT~8{Tt%1_rPRm8BWUdF9|SEehWxF~Kp_RZ
z^&)~oQXWRbY~*BuV44#6jNw{~pCG@J4TD6l>!EolJ5bl%A?qGarKfbNuj-J4rdX;?
zQT!>gGx^2nEz#uNsS?<RP^oqg53w%A0F1R7Lm%)~hWPtC8QeUQl(rTsBx#~LX+rH^
zU_vHKcm3=)r;Iv4CjwF)^7pI)kWb}>8%(n2jj6#+{+dFUZCL$Bif*5c$;!RqXs?PP
z`y$4oYgr%((Ury9qKJj+Gbeh~R%Yjvdblb#kL9N{yP$W3+hlgDr1aFbS{0qbJ`lZ_
z4#n{I@3ht4x*l?B0->o<U(uom%($rJ+X6pSUfNl_n=!h^s5f<nb@?`)7)mjNZ-WDu
zmB^oEs#vcL__PkiiL)+t_o*annuX+&#oRGrw^BbBih9;K#~F~WtUIq5dW?uub?Po)
zn>{5K!0j!>5@*i_ZYhjVj7SvJPk?N7S;1^-|0xP>Z+KLJtOzZw=02a%8>V5|x?=$s
zv!iV2xROqG)I4awim-!GS|WN|V=_pEufI@aiFEJc?eo*xFX|@;ekDRDylY`=2s-bO
z_D@H5sO46^yq54FV=i=aAo8CQnDdDsN(ul_zyyZ?8xknU1dY)0tFbNz(DDY!?ghN*
zrR+vt*n!S3qn(mz3}<y*7GqCRk)q6%j8G*Ws=U2(t)R1QsMU^7lw|Ur2{f_RPuV$H
zqPujmfT1|Y-bW4KwwguS*mc#kY~^OWW5Db9UBUT?-j9#}_!=G@xZld|jcM)lB6XhZ
zcg`Q_{P!?TwL{Xeh@U(FY|tUhX87L=IE5iTHfw2x#qT0c<KojQ@0ZVLl0pz&m@9kh
zbq034dVykq9+7WAM=)wZ<`;br7|O1<(MG2uPun*czOWrZv#BT-;1K;vAx1b?=vf+<
z8{efXXyfK<w`J|kn5UGt=^d)S>fEzbS!(FDV4n+g+kA%;EHD3eAo!8=ITli_NZ8Zb
z$7l2QxBHZ39=!%-N@v&}8JJmItDVkC;-P=$&Gsz=3mz>$ylYbUge!NnMZ-nC9P5@A
z;HyeTOccN;-KEHOE5uVgnSZ9*IgLQdS2W-jY~*dqwd}+>_$?UKWFhk|dS0<iOgKIs
zA$bgZ?YZvuY!vXT?<?>h_gxIQmS87@zO)dLjg`8$+R+ei=g2Gh?Z9TeQ~ayh8!h9G
zzGbPKus3YEay)9tJH^jS&uW&41!aW^D1X;QPP9eCN?+48<~bJ<!h<&2l)F4ZPWk^3
z_06$?{9m-S-EP~hZQI7ywzaj*sj;=)Zf$L~wQbw()^=aNzr4Jc%zu;2B$IsZoO92)
z_fWIIi7TNyvvK9ci)3(46lqj&M!*h>cr~+D7SRJg105)FT+T9<6;)Dpz~`F(zct6d
ztzW^*m{67&+DP1Hje*G+{7Bc@`aVas>SiO1fBE{H?@haHDqv9jH?t?S1p-?NzRR0+
zAY#|x&CqdFwk0?p6%q2AWX4kY3OJFlqZE^)+*-}|$)s<*?59G3BnjP4Z#b}%01Wst
zhdhxm^9I;Zku(UEqAFaRpcQ)dIURVI;Xo2iTx!!eXb^AUjWeID;LK`vqBB$mMxXQs
zag&2biLBu_UcW87`0k9cvhy6bmD8WXZ+b|K$-IN}Ljg@&L7HbC<AgF}jTtQ4etd%O
zqQXy@gM~Cc?qYcSLW;WqezSFu0Ejf-^cQe~Tq6o)LB9I}!V0=!7yA%IW&~8OvX}I8
zkMurUUC1;Mm=7eSAeJ-CRrlY-*l|34mK|ktjF?N(x{PyW8z;Fz5{h*T``p)%WgYp2
zvMZ)mL@Eb&oO)Cp<`%fint8JE)2qJ-{U0OO0r`d7f>4SFq0ycTPCHEi(zYb4=~br$
zaCW^t!7b%$bnN69Dvk|vSyHOtM)$mtEB*|3A=J{OEEk4%0a=V5!fEyG-v{rsBlhMY
zh$}&`ghy5~6X~3pwq@(KIQxs8PP(5sb=;RHl??s7{q`XFPU#8l3E=>LyS3A0I3p=q
z92Tnb;yiq~rGvw52df3duwGC9(r=t_+f!Uq``rM~C(@?)j`KkHsm4kmC_4`8_GRqw
zb=VR#w!RUrSc=^O7WHoQ_pVA;tmX;c9VNc91F&6-enF`OHpI3|np+{ppFbt>@MQd*
z?Cv#@MgnJrn}C(Hl<<r6^<S@i7wp{<<bUcaFj)^LV1+uA;x`A{KTj^4FAp)#=AeY2
zReF;5jhg*Dd%S(94id&h&39E8n@y~QTxgiZ1`94uLlyjnZuLlB!b>6Ni0aUu6Vu68
z?Xz&mZ`@HxAY&BMUzCxU9l&qCDJ(xOTEJkbXrtmlU<oIF6H64AY?*s^Y>UhO+O0^l
zOHuFj8iBZFr0)vo&WN)JpFt|wv(Lkl^eHYs{^%XZ>&tS-oI`B!gA*CLo;T6w&t%U0
z`zj_ZkZ>IH$x$k%7%`v}A&AwCP!XQlPs8jfbmv2sgOhngEW$s_AN<L^K%Y^GW_}&_
zxX^r<5TtFUHNg}U{@}7Cm+AG`>jk^D>7aYl5+T|S+-w3iA;HTm{@h1QONz0*d~A}W
zXOk(7KkZ-x6wQxviFaY@u-2K&@?_k4ZohUJLd!y?ae}(SDwh0%hR`?)$XQ^5uA%GK
z=0xDj6MnJ%tkyt$;M#Pdae*~HRU_5TlWOqS_AeR#-ePd5i)i4{+b^5AWh{v^I8&ue
z%!$b$Irs*E2BN@-V2m(@dAHFm8KQWc;nH3NMWK8Y;db8luFHExVaN^3A_Awzh2FB_
z(2G%wxuHJl#WSV9_TcQoDbsi>W1@2indc(gr;zJnvT)fH<SwhHV`nhr(|x5Nqlc|`
z2*dOYF8Jby$bfmpuj#8V%UGHpvAvi=C(})v=p4YR%SSAqBz1NH@?|5NLqN$91hHkh
zM8FOfof+y3ZSClCed$nnkZrxM?DDYK^~m9_)l}-Pw|?<4^LSv-2N&>l*!UB?CO~T|
zU20RFR{_1|I3hVQv+HJ93vLLlpBU9@gRigeV0fU&RUfKg`27YkNiHeH8b{=DP=n|w
zSrZU+otsVfGaW{If8Z=8FS9D3nRG*munf2BPkCmn#>jQeES=Q&$FJy!Hgg0L((09s
z_VKjrOVl-btSV}d8u=;yTI)=vB<Czmi_2ma!R{ZEgZ~)u>#g6FV7RnBeJ=zsek6}#
z!jac10;@>MdR1cM;jrEPbwc1sT>>6UpbuE0nRWBaZQ-$Y{W|2Qqtm6mF@buoqNblM
zY|Jn@)nwjOmGXZNRw|f$t0Iz97j@rXt2slxhmL5?n_}{wC4P`~2R8$kSN{Vgokciu
zpn7P?Ymo_kp*jg}n}b1-5(_eTkM|tJt$K}Ct%V|M@cEq$%Ys2|E;y!bmk(`P=NYJ!
zg-Ozief(E=8r3d^5hcm*8|Go#_z`mP&%!ieIIeg#aO{IISMQCr6$dhsc$Z14j%4&=
z-;-_4d3b1yV9>&!{-}lJkuFzm7h~2IS7qz_zcUAyT~)8;k`J;8`q^b??w0wH>>u)!
z7tOD9W<seVayS;pO{F*9e*H}_P{v<7=&gc1T4h9{u#AO|T()VTkSosqnV0Qy^>z0L
zMySp9x8|oZU~9obH&|hX3lOzhpC$?g61`P+)jC-Y;UplVB%j=v!m&5ALP}vwnsOPt
zQq(WB8P5EYAl=XIuJn3Q;H%8DXPXhVMi^NNkJX4bNn9JLad1{G3gswHBee!}^Gi6D
zPRG}mrp=Wg4&-W$Vr2cW(yUHo*(>zm_hdGzHP)1wGMTF@RvX1KK&w^~brG}Xmno^6
zFdQRq5`GeQ{I2z~XgvdO;36}H_hh@CqSj*~%4N>EgLVSf>BHa+KYWc6F{rEhg*0Qy
z_?N24czsOxn%7n-ZRLu}9BCS;Bz-yQz4%*%v~4p5^6rycq-k|*-<#=tz5E>ENhW8X
zX08C1aVZRY=D{#`dSj2L-XXCp8XKi|?o}>Y{j6o(ui+h(W3`|&!J)!vzocTt#wwLX
z$Su*+DBtzC<*xC*3Hh?h{b_xqa(SOE-=u)}&DVRSH1D^)ApYM&#YY1m_YRT;Jlzai
zF)mW)d=-y9Hl{&l1vIOT{l=>3c~d_{pBBvKJ3&awM%rkxt>zrDj*01xX08UcHfD_v
zQUaVMh>Az!o_ez*F@L*DD$1G!y4LSzJYVdJ!6a!#&W%Rc7{~S0I7nxi$2Wl{t)^|-
z_zi|SjxYU~%>k|BXV56n@jGFv(3R=-431uvD7C-B6U(2OQ9+1wj$C6jQipsQszq@Q
zBxQ$KVZRRX@O_2zm#bhr5y=+UHsqo(1#bfu{|9^x1zZ0s+6gVr>elAH{U7n~DF+Y8
z9bS2_#`MC9EL;YC+!IO18H+!ggLuYdv)`fX>Pn<Iu>3Xfp7Jh$e<fnOC!aIJ-}oI8
zIX0Ysij8JzR_J~~C-aw4hsu-X2NreAp$XA#+2X7n>-Ts}-JS?kwMFrC?$P?{d?Dxo
zwPAEn|MEC9d<&DvvpdorP?Y?H8@A&I)zkY|gBS9}tunI7TGX1(H}*C3Cx5w>dZ{cd
z5!i1HKIr{$j^R8&Rpm3{r3-kWVitz53DM@qDF!#XBa{-4IOZ#Ch)hABZRAtxBm!of
zlp{+71r%!#`kks#XcT+qK7V;>i;-KHvbYbGi9Hlp{`iWpMv6ruN4|lFn5+o?s%{WH
zcLb+D#IM9|?`x}Kekb{XP09YIlX0um5#DrV5allJ?T}rd%%G{QP05HR_P%^yjt=^h
zh(Njb;g)8WreD#^XN*yXG{pJGb@YiEh493$<yE%FR!z&$30tPM83B?{Z`j5RpPkou
zjSrFbfMSzKw;$EFXV;;hBk#1S_tg;8(A24VqbGyPIB(%{C5b!Xuv#J|c+PHD6><HB
z@U*re7(CQ~5urDlQ3S6wrG?JIzto8r?immrBT$SJ(+M!+ELMh7D5<T?LW9kXtSx=t
zmgi80R|A29+qSjM_xpwObJxpGfbc6|2FJgXVGN)A_}?FUVIuj-j<3dL2}>3U7@kJz
z9`bS*)`^WF50joz@m$pTAcp<D_QpdBRrP+TjesC<!?X)>|FQY}cJFoTyR2gxoS>s|
zU;?p##&)Du{Jol0HDmQqu;g$5WH?z(w9^TerxJ^;UpNteLLaSK`Po@9s~^)w&35dd
zWr039p`|LWT9ubB>d<_M;8-V%pQ#=>G7j8RnvB5gQZ;L1n343X8p`M>F^AQI|3^`*
z&)FhC*J|=;u<UA#&|wDwC^_XzRO#1PuNq~3l(8ePC6$xEL`T&uX*+!7qpu^R`D{RZ
zfB^=VVuMn0sTi25XJbx{tJCtha?;J8Q=6$w&C$C5K9UJ|(uwUYkT{jbl2ivE8$3x9
zJ2BYP`Nk9=1L)zFh`K?3+o(}|zPoXD`+fr;{JAFfq&>+PkmLdLEgqSE+VoEY(_BF7
z4FyPLHBG@A5RY}b8iLz)$V2D3#%3A=OQ}g-B{vSnCYGof{2^lI&VH#(#6l%dx{E?O
z!nST)iv%0v`wY(|Qa@pxnorNcYu(d!XCIWiF4P-z@-ncoLC*_+ba4?R9o^Oj<qCAR
zEFNDl3@|?2aV~g#s*YP*GEpX~=5ha!|Cv)a#yimDqFo8Py2R1JGLNPFSHiwne_zDH
z<iRo%E(BpO>&76@1a<WGMd|<{A=)&&NBrl(OhC|e`j$yGl9UQ(Y0zJ?&>RL+y4K6F
z7N2UW?_Uf*A}>r!Qm4@X1l%dL*8{NI0^c*=6C}e9emo!1#SlBF^w6#WK@$sW<qn(m
zYmFoBx&~bbdtPw>yct7%osE|{H|E0hFOzJwADAXY<S>h5{m1}PzKb_Qojue`m%0<G
z2@=gtHk$i*$y(Kgzv$xva2l)iiE7$vC1p2yoRgtK=v<t`DLJNe9r$^(Qb3C;7@<VX
zKOJ0z3&KjmubksoVhq}Wc4dhE;&Og2C{6SCP)J6?;g8*lX=vH1(U4=Pmz5*(#I1_t
zV#RCp<~I0=+}#BX&i3`-UyD;MOZ+C8aI>!w3{)<ep$t^5zmEg+;WOx1W+26}Fxc`X
z_X4XaT{gRdt;rw}d(#D=#(_6~PzGIne`hkqYiuMYRljj+EX`TZ%s=dIRV7ZHsd6kJ
z9XTp3Lj4zfY3<<~p3B)-w|yq=1iNU^{L9&?KA7Nd$K83|7kpp&>10cAC5`neTt2j0
zIN}B<b3vfUdB55Pc<38!mbkZ##k%YovdaUNM{`AK&=E`7+kshp4d8D;f-n>(7(xnP
zV$Ba~xf~H0aCynAn{u0S;<ob99lm!({OB|+N?eM#w%q;M@fr&b|Dc%zrR9th8mPZQ
zuxN)Fw4d8CoLfS;rF|$_3P_rW7Z%p>qW$gU6x{<dEa}(iC|FoDuE7tlqAco3)#zkF
zRYcuVed=*yl4Q$pfj|`zEG3eTFq#;0Qy^I)feoaTKqRQ2%eJDBJ*3v<<kbhtP0V}}
zpxQP%fJBIESKHa~f`BJF+gvItJ4U`m|GEa|*mdCxtKO(<lf_xgxXYd>(+=cgNDolp
zZKl>hmgddd3Hn$x3PS8(%}G#i@gaxxnb|nqPC;|O8RC6#0UXV^hTKHjsRsVw3GVC_
zx?ECA-A&F*;RyIP)6B+J+e0y81!QoSoGoeB*qLr*J7RyI&KyU$bK#ahSGA=Ik;sv*
zBSd%KA{XC3uM!&Lnh&gdmt~xjZLd$%Zm}l5(`AWOrZKSC<vD2Er{u9Rs1&B0k1omX
z*6%l18X<;60uGo8zwSj~k+L+DA1EPiE5`<WijcNK={9o1-Ovg6$LUsEPY&6vZJesK
zm(%k>eo7!LIJV87sboF~R)A}p`1NpJi$6-^wz-VS<)v^CC`)|z7IwdsY`X6n4>9k$
zuGzqC{fBec={>~<_m$)4e2d#s?eZeWpmc=gr_GVO4q*Dn3;1FqdqA26j2qMy-nJ$l
zKpzT{*?e>-ASs5|7AA7#*A{|*U71HsbmaaOc<r>qt}6ARqzmeV2&X^P8(K#V1K>cU
z(-@!)evc-2@{?eYDikWzpKA7r<caz8_75>iO=y(;(#!|b$j?wK5Skjbi?XT3CWnk!
ztLrlcFf=pB_vVw0RrQ;!Fc!0VP285z@bxEP$7Gag-02v}4afr+7_SXgShBrrs@n?f
z&Pd`lh5Ep>O(7ew=rW9E==_qH1d(ej)T#6_r2RUg7`wB*EA$W*$!KX0w8i7Q;phl3
zb*IHg4zQVY^t68w5&s|#Z1AS$Hx3TO5RkhCQohFvUkiT=!<H)Eant>$SdoQ!yIliu
zP^DOrynW94j%UO8@u%DA>}4*E1KUV42G8GEa*}2up|HKS{ByA2;J@1Cpsz{4&FIwb
zq|R5R0b*(Ke`{AadiG+F3Xvf0h5hBox|%D$?=d3Ea<LimTnHchNU>l-S3C*u`W|E(
zEeZKRof$K$&2t}sWh_v<%==ZjnhXgcE3Rw6HfNz@a827{t<5WTJgW-lL8?~umqNht
z3uF?ERX32K9+4N3FWyQVifCY#(=v|K^S2k)R}3;vXN6DGDi$Jx1Lqk0GVU{QZ&no6
zX$?c)TK`^Im6ym2&Wet{C;tOxNz!x76Gu(m{1NG8oEtTla>lyaX=F%qwKHXSj8Y^x
zJeXzbMOc4Yij2<=I})N|>u;lTqW8<W{{5?INm$E|N;NM(iTlj~uhyY*$EkbuHb8hl
z;v4f$65bJp;Ka9B-$?)diulvNY+e5*y$!qd%H-vrHboZ}a-6=hQ7sRF@UowRO9t6C
z0iu{wSRCX*zuX&4zf!6(I{iZ6Wm5`g)8~yhoq@I)E_%*nn$>7Iw?08auCw2nG1jsC
zKrbtb^RMH3-N0&7v2r3Djf6jY$bH##IjTvA)J{DVjoZ1%WQ(;YU4yN1M^B>O`K!&_
zuZ00O!@b{w4&rOew(+_FuHYllBio@_Dzkk|(m~Al=G^VDRec{t>y)Xa(pK)iVq=qI
zYMLw&lfsk?`*`XYe`>xrEQj1T$lYt``3^$mAth*D^;8*>DyjluQw$_Qct_JxK&R3^
z@6oMcQ)e0hg(c85>dIopvO~xvG0aCLOyKqs;tHCImY5bej^zcU8Mb{?f-=A#so#e^
z=!-$us7j~d6M7QJMJuLWFIwtJl29&%+uF&V&x3neJ8G7y?3D@zjKzKUIN0Hca|tMN
z81Q2%vx-~&Fvl5HFEvNM@vnu7ZoY&oow)Jp&01ZaqC%!N56h!(=3A4rwc@EzU^D*E
zc+Oo@*Rb@rM9=~v)AdnYZCelqLOZ(-ZQC}jT~`ZTMFkq|?6l93by2(tzs-_Mzdour
z&wUg&cr%0HB#<Dkr=S>5qqa4E8DW12EXP<!&!Ve%ns273$^cK+vJEAR$x=^)v1WgQ
z-Rbr{mf|huR;po>tad8#;<bJi-KKgD&>3P$rV&?*6Ho!D&H8nO3x2s=cj>F;!fzN&
zeVguGDpr3k@%4L|Di!=~TV52vut9GD|L=J|>PnNpwy{BZyBH|44f2|wo*tKpWc*EG
z7YM?wy`T8@6&no#ox#D75j}47+;hV%<2#}8bMFaMD1Q@h)g<A5^YAj<iRuY9aZL55
zDH%{^E;j?9loSgy7!61HDij3g@xz$?GFFZeo)cyL%vg9oX<t8L;;`^+n5zyW_Gk9!
zulflTO#+6-wB>#gt&t<Ug>1vX^9jYbg7B^BocsMfV2{K_l2`XATyPV$!F$ns*CReh
zvZR_gJ$H?L)9>h1>u>Wr`Q@_8yrGRxF#u|XpXLHU`qQScgXv^wj^SV@fw3UhOevYb
z(U5o)$Vi6egj%VOF5vv>$CIh#CuH*Za~W#I6(x8$f_rh}RgeIR$?z1~rKEl{in`jw
zKScTdzq2QA4#z_I)UO~KK9I!A*GmN(+mG=Z;x~j%5=iQYgA#-yM2sc`#N(V_3@&(1
z7vN%=5}b@!f^0Gwi)DH?E#T=v{P6B_e$rvBB{8iCY9`6B(vqL(n|cSGiXV23T$9d#
zjcohyD>wFaY`BR4foB}$qc@>yX0|?+BKN4~3fVqEn!e*^a=#*~Y5-m)b?3aVKCSKM
z1FR<k%r+PQXZ+@$;!sc!;HMurRLk8d8<3I5<1=S|ntme_s9cXKe=gOiHc|A(ZoipQ
zvxwEmX^&g=g!R*m%kdxItDv09uN$IxzDipr?de9olgn?MWS1F2+jHQVF;RE15ii(f
z409-Nn{%o>x2dWWxAwhQ$VgS@(#HA%zeHzI1*nGNx9tCY7dC$bS$?xM-lXe-nFb7d
zGA#0bGWJ^2!q%e0tSzD;n2)QeG?dh>sobDP+Pe<9ig`0#eDg${P9(j`WUDwkDJTyK
z4daPyUT0u`oiBc$I`v<)DfdG+&^=LxH233hfu!eLOJK*(!YcfzTdi93mQG#2YyFUb
z_$x1AvotP_yj57A$i_2@XzA`H_6!ub&SlzGK1r6mx+5*GAO3=msq4nCEr3DotDu{q
z>r<4B`LTP2)Nh9KW88{$-$r*DDf0GK<hDK7%{d+P%~?vo{4#O!Xb^o4sb)K5zpu0L
zu#b@ku^gp*LHYWP4TGXv=m{lf1SsvP#MAxkh5vkFms^u_Y0<LWw^p(`jssxP#f$Fz
zTlE#;G4tde=@}Wg^rBWYrma)t{~aBKPe)csrnZwR(3u`&Z_N8_dgi$#iG{8H-fidN
zzMc6gd|bT|H0@v-?{CL4?vgWxdB8mr*;m9ON>ExHlYX<$?YlvIS}88c8SyIC(OO$w
z2Ia_RefOd>Fx~O60`)`tbQnlfaL9)FjxI;fIHB5Ay}2IOR#w#^6Tf=iG(VfapSOn5
z45u~0A2GB1B3bm|K5La=^_p9_saDQkB+S$Jsea0T6FYhC+LwuPG}7;OpD4m3ckQgu
zvO}0fH@49@@cmXZZK#(I^RO<4Ay$nhS3NUYF%9Ha6O}cf-Q;R>OA^?e4!DwrsJROQ
z`Pat;@EE3-!}h&wx_!;E301{4V5M#cD$~BkSZ-TDxnbC0PY?-K^v&3R5KYDOf35$`
zLcY|C3xk(*Cs@o9Vz{O)Ufv_#{-JVqU7qTpe0Fb@S+Z;m*P^~k=tU?+a9fdgq){~c
zC(Ym<IbWBavrk((>=|gX@FWf_q#YDh+Kqwb4t57{V_~^#nR&gT>Ev%=xkcS|Wji|V
z!#Nr10<XJuRP57MSEyB-{S{(}7#cAXBs?0<xX$G@Pj+|ns-z27kT%Fd-~|uxYdT<Q
zIq`(tJ%+Qhl241QsNf)ZgOGaKds8sfeb$1xf_AtA47#Nbyny!3<AEB3rEgZvq6cG^
zH+d`zS%U<!3(uWoIovB}w_x-2rDj2R*<{*H;o6wxDfb2T^K~PXNUk=&3o_^Ggj&23
zu&Kw*CAi#Q`K5A@^!->-<EOMvAfCzX{8k(^82O@)*!zc#GC52&(WcMP(KPQTOQsg=
z6xfEC67EO}f&iC{wHfdnLX%wRxs=8Xhuk|BDSE?PJA1KI^>4p_yjRbAXEPXmRwVpe
zj3yI;Y<c|7kN)}lElfsWScx;9m=WYrnH2XZ<ol%rutbkNd9^rO745Zj58f|rK^NKE
zJ}XK0A&qdyMJnnZVb_Yevm%kN)faEo_LdO-+kGML1{&BrzoyfY@O7$zH+OYxnQ}u6
z;SztD_lw_)-)iQtG!Pa-LL82{k2|=P&I1Dl)1__rm&CDJQr~+bMUM<(GIEK?(_{)E
zUq=S7j;H+VykQ?3fk+T_V<df#y<&5xminF9IKVe%)lIg?m&~HycL+zn9E-I(VRMn6
z(HCWKN(iXqkI6OaKH^{BhjDnL3oTic8Jj}5eR{+e{0hfG+J6X1K8IVE^m5bYsW|GJ
ziD3C=`E}^|?u?=q#DADZDZ`<6RWKU3c)a`!6Az;2dqv~|Be!c+TB?P5+pSPl|4GFu
z8D+r`gx4^B=(OiUwkp$*YQYGJZ_w4t`<l!~f&;X7AP}r_Sj@t@Lo}*RVioOR;R#1)
zqTmTbLM-0ESe?K~D4K|!7@|UmV8DyLD!_<}?<NW&(D9|pFI8NQ=B=YMtHH0YNDB=6
z7hBC-3qS<<r<F%D1`n03Qk!q9(#tR8-w6IC^*2k?BQyGb=<fnT*Jq;`*GBP&v=H75
z&kW$}ap|UOw97t>jRV2>iE#)q2db3jh%VMUN0FC}CW!vIld=!bDg1-$EyQ8>>_)hQ
zb{!(%ih#QP?=dN}&8fhO>iN+)G!+@%PD(6Q_L|T6-CHH~j}ZT+JRZB$3}36c4|7)`
z>nQK82u3?xt84U12Q43-=z3W&3noq>C;$+0FHqJCu5eW59?7M^tmd!4szVA1MVucC
z*Nz?XM<aX-h)lF=brEtW%R2zEIzkL<^l4GA>L$J|w~r=7$WY17LN<~AO6}!64>}l%
zCOO>LvzQf*@pYGd{+n%~xf(*vw%BoFrPDqR3E9t|9UF3QwI*Ca8)Ef=hw{Y}Q~|a#
zk4t**3(S4n8<Io0ynm$cjzf!Y$JB6R<a%w(Q<-|n6Nd78WT?MuR?)U))5EQ8uW|N8
zVVNf}C?nJMCoxQ3P7t4btKBMclX7Muqn0TJ&dT1VZ0|E-z5I@8)Z`T|O&aO5mD=Mb
zB8)~&6av6aI%EFvj7`VbP1;Kd3<7Zs9pZ!!AP8t19(H-E>SE1s6b6fqFkFc-Rq`W$
zP@u!yi3MZQ?Bi~OmyiQDo4|h!^Jn`Q;!{^lN84<ec4o74=(SvB7})S^C|@+XUZuYM
zV)eFFuNzN@{&rGDzP`WF7M#Kq0f!QE_GD#is%aN9Q!k&Sj(NZcgZ2yJS1?e?d8O5q
zi4t%{=Uktpfx)ciX{!9kQQP$Vxf>Nfjg+jyg^7`yQ+cE0JcgXhBXElXv$$|HZ5wO|
zP8e(}7~PXDpax^qva8g6H{C+vD>0!u+$-RTltpenV6P~<BXLFrLtBouP$pjV=!#1w
zO~75OO92zkppw&e{XJ!z<_!>2#gw{p<&^}EWB!4}^pQ8Nr5}`%=qL25BjSccwv!^^
zGYbZL<MlU&zu27k?g}WeC`~QyfetXfpf67k)vy$x#Hvd8yP_4c2w?t9Ypj5Ez~9|0
zqL73ma$3>1y?o3#c(kL6eHGgs#$wafE86_$@+l>t;Z3-r#HM|3r2tr7iJthXGR1>{
ziL<oGM*Kfu@P!N7TK=+rw}y8c3U-|x9z~mAL2%s<r$Z678oE&Sn7l>BF=6W^sX@1`
zd71ID{*$6k`cDr!0}Vk3cdU<EI5RW0X^=K-6sBhE-D1tjls=4@^fx*&wo#qn*fgQo
zbhxeXG4B%s5NInLC#5&wOtR}l9Nr;fFAAPrz^&h_Pki<*foI91y+ec*8cRpDR$QD0
zD$UDnc`ks4z?(<klcnbT1jH3;2n{lH`Is2CbD1IF98GzZ(<ZfBdgz}jNt29NMfAZA
zhD|D$baH^&O<XL>_q;epHTm1x?*Wh~m_rI#^7ZNF06;<cecR?y_1i}eC}r&J_yZ)D
zlVZmp@0Pt*J0BDDWm*~4z|R)Un3(W2>+pnVj#H@5?rqkIi?J+oL{4w^0wMl%ess^(
z<)XyO#T3`l+j9ck#YOxV>;m;J9Y*1#B=Z#IfSbHl;x^w>@<I!1A*rFsVu~kid*3o3
z-KrY`py(Om)BMFk@f<crAl%kU_J{-U^b3Rz<MVBB;dEX2dw1qjL`$+sg72+CMfS28
zIWC!yyXqn5Pxe_FB?{$aRQ>z5#bJlURcfR#`9RKwjw*V7&pyQ<tiLuK)*V#5aEm<6
zD|Bfn=oirnilTLTtVt7yDJ<KFZ&1!}(TuVW$or?Yg~X)E)Rxfw4va5pqoMjo75yhc
z!|>~v;O*ujBhYX`pV6ICWaYD~;2e&PpNw9SFg&&ip?^2V=adDFnGwBGV0?kZ+ve(V
ztWsK4Fe1ztDUjVRx)|rZbvi)?Y~?nQI12xrJcdS@KSwaC;6J*$ialOQqM^ClOXzU`
zA(KL?h@_GfxG6|K43<aTNEH&@VQY<#Ln-{<!KX&lMi;byoR>1Qh}c|20Rl{=%cYc?
zpq=nhESR{wQF(_F%wSj;Rw4DbJT`U)!!}umGS4)>@APew93DI*@g6{M<VsGlx;9yq
z9Q@vYEAK6Z7Db08#ozsd-sk{+*8EEWPS)EN-<R{c>~3(mWU{i8tSF1P8ieP~=Sq<G
zK+Sa}z9JzWV^e(Xw;oS!UynqpeL!qc*?S}dX@?v@wF1nm+ag{dJ797@5c4iw8n1pd
z;xodma|fymn#H=)fGcoS`+olBX?2?t(Vq%*k`lB^88$`QLBkK8-*Q7F9FD#N9LL^e
zcf0;w%yk5Z`r>@UgRhh<Y)j7&z@+zGmwq9F{S~BGZdLtJ*_!lMp;Z)NqcE7U_h)>G
zCM?4SGj{o&TiHqp$2p)5-SAW8KJ#hUMiNZQc~49>lF9iX<S^J=EBrL>;r!k;v_|f+
z70mp<NJo6$-K#rEKR3pGDc@pXKR|izmFa6Aiz-#tVkXH}kmc+$HftfoE^n_&IQk$l
zsSaWJL*x(%g1<kWqjqH$N9Hgpu1*JEL_j|U(gGb+q;2%d)*>jI<Budhc~2b+$-3t7
z@4rXzF+jf)(2B2pC(temXZFc0hdM$t@&(`0aMY*fq`9)$>+KP#jW!Yln4rBMr8DM?
z%QU0HD++#xyEERX#hwzUd<!n>9*UI45ui!`qQk==h+qCwdZ(UxVp3vo{kx%XE4!H0
zCNmjI6+@$yZU-uLMGyO_gI<dgB=O<lMLEurI>klta5T7hqZaNFy{|ypQp-ag&8Emr
z0-|YbcJKAFFd9Yzk-$yBGd#y5YwF`g_=lnnYPab_@6EV7-O(C@*FfU0*{E_E3NaS&
zd&T*T16IgcvWE=*U(7TaRPj8+Ji~Ehl^f}L<`r*B8Ul|nuS{hc8m916T>B>X?ja2o
z`;VvhRZjITMpZ7DV7iD@WCA&aK3g1k0*zE!0=OMJCN=n7RhW~&DC(EiJ!to=j<7+t
z3nT763r>C-B>Ws})un*<B5PZW5jyWU59+%7df~pA=HmIihRH~nU4PHt`LB^t=Yx--
zVx5vBNKW4dlWxzGU0X5q5+=I53c~2E<M1t5v;VgX+D)_Qn27P1h!3hiLvK*)Lv<tQ
zqlpHdGGpsCXSr#>r)zcG->W|c?HPcrG81g8_IbN%`I3lLS<0PNI}COA+g|1b=piBV
zCw2$LrLU@8k|o^D;iM_{M!Ia|R{EKY`(f7L&#ou0m&v<^+ib0xrnPK>GHT-nt1vG0
z^v$SInaE%h_x)@ibaZfGCA*!?gtVKQO;tU;%*<pJTj@>Ulg|MlEvS?6usc0?MQ_An
z<cL^5=<B!RqrgUQOIsE<>g~VKi9taZu2IHITZWXTJiMI#@UInoF<Je6*zJ3B<0`0!
z_NnoZyiMKA^r+o^ai-(VFyi+TEsZ!1*`Y+x+c^6%{rZDPjsSX!r%7OAGkjmNqQ}pA
zp<=0UiMSq^`tzB1V$OIm9xI*d+aRgUiS;&pNEV0x{aNp1bA9uyHc?UYL^@@CQ@A=0
z5jKcBn*mk>a^tTmP40Y!ko~UcSc4Qtfs~vr*0<cm1V$QbR-0ED*#a+RTd=NVlzJKL
zb}1M}XKef}tfPyjM8Af!JS(mfP*-u@K{rHoJ9Ql3S@Yr`y8k#R&<Yx3V%G8<Z(6%8
z+7(IvV-J_J?77{U-0h3-W<WXXXJtKGu%ha$-oX+`VlEt0^vNE<kB4Lmr6CdP0;a>@
z3#y$$DVAvYykgl<kL^hV&7)SKr(8VFDd7cL1!-fQ7B;&JhA$==Z(?aC?Iw*nb;{$k
zztw@ZmB=w|;=Dh?$e(JQpQE=b@^KgpgZ=38G3|2HWJvnb<AcK!@lMx%&j*>4lp!UW
zUmZE8xqnTogmzEnN>b$C7Ogah&*kkGt%hgNJA&6|(mH8a_BPL34HxRer^$M3uQ^N-
zMom@R50C6h+^T%aAGGhuV7E}=Hx@8kB5(mHmwC70d8|Vv(jTzUmWOgcNC~2P!vm6A
z&|-S5YqwBk53-f*c#x}5C3v9xT!7S+Q3UdiZl;;*U|&N)jKm<k{qjh{p>3w@DhsPv
zl+L%e&lH)oEpNn+MAGgzeXyweoM(QZwxMMGZn#8VgSaDhA^Ur%L|H%fx8t598E2q{
z*}>4DQH<dsx6sS?yy~su!BiqrmsV{sLma3gRw}=by%M_hPVhwQ@O>>ScA@MT@zyH9
ze==5#5GIF8##XJCMwtF5Pa2oDFB&~j__`YBVTibL?D@{0Sfs=3#YBun)TB271&w%u
z>t35}$Kvri&bc=M^)<K3iHiPrz!(ts!PA?;uA-OUw&M%FS7~gca5x&UdZc1VmBJg4
zE64u9^sDXn0l85PM4{ZYI=RFTX4^mFs%UJ9kug9!6v!ME3}Y|+x6J{%mAlRxgl8@M
z7{AO;qgT$sUAU@d)}d8rsfRfekgjtSirzY$wk;p^)2g2`mT~I0+rsX{874qEEt#lu
zsPR9Eao6+BPb5mCz-*9!NqOb-XyS8BtsZt2?Ax<w?p}D9%@*FyV#jF0OONKRt2IQg
z4IRj!Y)<Ea|GnY18NXR(LZH#|v!3TS`{Bt~JF1`Q$B*e8)L2HjIb_H5)eHRJl+op&
z{FZlxGl>!kI6HED?Y!mC0IU6~3JBL4H96K3lXvyyuh{0U{Gn3VChG=f1nmct1ced7
z@(s6&_W-uGY=yNBre(hhn}8Fieyz&}K%<o(Gv<LuguM>CQNN=@X?R}Kr~|rSzAtSj
zPUxe0m4a6@ePLeLq61CL<Zt^A%Fr<f{*&L$N1-R&V;<gig{zD<01by#4{19;7LJ3)
z*Y8Xs`G3DN(wUsnhv*ShrIkzH1vOYwpm3Gu@o$i{cgwm>_gLBh<@x4ww^4>Wo$_L5
z890wws>YWs^_eT3$cBb%nGT`r$fnk0%(Y5OKqaYvwvpwztIX`dpapj7vQ-@xFG0KV
zdrchSCbr-*4=bP=*-RJ!ZN`s#vy2L-TEWYX@Gk6PgcsDSSwL<I4wsda=yS`0>Ar;J
z?wF{6vQ8-(uxWV{jp$V(<@;rZV@9wV4uRz<Q5+UI!uVATqb&fQp^NwmbHs<ml?Kg_
zYb5gSuxZ5x>#DEq<*ba~A}VnaSN+#(p(JA`6~l94<~yJyr-8`ALg&7^(d=RS?_=FS
zfO_dvh%-Y0Nv9qmlC_Gf=O^pbhpap=Pu&$V5|?L|d6rU4y>g6!`5F7eAM6B52&f=$
zNfo`+)iS#D_M6`fZ$IC+w-DFE!$+}_?f)b*KC<Pjb|=qT7HEmXhVu|b3?k!~w@R^9
zbZN@#MScL5+*7~xMF3|K-|Iav-_Ih5-(3!cN~c}ip$Y=uvcO$brk?TFxb5Bl*sD~<
zPotfT8(?lBm2$j|ui^~G^TiVPERbh`rXcQWyY$305Zj+RevEcc(60AcJvSh*=|Jny
zhZZ7nR~&sQnDRfNZiTt?6ibtOT@!75uloIb4Uzz%=PLe6n6l}<g>bsjVBMy6R!^=o
z<stIVjWSz}^BEI#$!oML1j~P-Z3|!0gd&w@a!MQirf9F0P4aJ$6H>bM9@Lh5nqW!k
zcGMMjgr@oiCoeBIHv)2*u95sItMp7P%A-@!45W<jRsSR1cwIv&-|wfOqflV{j_`JF
zfwoax#<2Ot#yc@IH{D7M>Vdk0z!6>&`5rGa(zV&ry*TL6WfmE>F`wclnO8ARy-1G&
zw;F|td^3NO?;hwkneKm#ADM}|ODe>%)M4cGClcms5*1H#Iy|P9gqKZJ)@D)>7P09P
zA;SV?#|3z+(<3f~edX4d&ez?zD;CGhftgV)O|><`);%8o8pvFWer#$OIp%)rUXw0S
z^eoue+3c-O9N#t$ZrYqhxa587#QP*l!*lPXeRkd@#3GsUCB356CrYIci;X+Gem@4)
z@@rbPI=J)OM&%Br9~vi{O;RO#23k$X&=F~@o!ysOp+VMckvB`#<FB0wq}Bi7A0>9V
z{&d1Gg{f7lU+HJKbv(56Jao{_|JP*x2Y?QqSs!_eC|>m|MM8+R`lUYw=N(G-C-3UM
znhG`(#pN>0T>RbF&^aolM->BctFeWvdY9QUHvz~zyubdr?eG*NY{>^bQ|17FbQgv#
z7ajWD-!JE+l`8>^TWM7PQJfWZ!Sm9B`0Iq9xvTOEo*fRwLVxRtWHOp<F=vX}uftUR
zdC=S-rjRrbW46s=!fx;UpiJ9_1N$KC!XCWV9L*mDXbP83)M}(&A5`Q{666!_k}@|D
zEjM(p?);)K!U_$r5J)t;J$i9R`v^sYICSaU$fA3MQRe|Y9u@hDQOnW`Wds7{JKz3~
z7P8k9?%m7qVcZv9m~~3&bwY+2xuM3Pd<gu=s6j;#$6QtU)SpqW^Xclm2xB;q8+D3Y
z+?FZ9D!`f2$N!44b3gYJ_u%na{}zjCRJSO2PR2W4;HUakV(jhS@_5-px3E7s^xtq8
znO{tefbH01-Qy6(fTns}It-&o7K@9yr+c3ORe?z-%=_2l1O)lC-Lx66+!}jBG^SVL
z@1RtCO|m+((U`EMfP%3)YNK~%{+1^5ba7Ijr#SStIruPRvK@c0gOjMAMVd;%4((c_
z#%J@v_4}-*)QSBd!)9qKyawARQH)+Y|CSJdXV2{q2{@S-6G(UiIWZA6I&=g+GfBER
z58@Zn@7f#(Qoe~DA|tG^1b_REoyo)Ic(XLKlXky4ytRpch8g{DpZrS3*KxN^y*n@}
zA7VfT=*W%RRa1^b=%hUJ2woARrIn-LM9nl3!7=n--PfJ-EFh8mZ(j0q>|s+jO~>B>
zLOe`AL=`f^&$J^JCj4*ASzN_l7+&5TxozdcJI_ivk+fmRjl>k@cD#JLex@JU{-Y!>
z;HZ7=n}nu*gfFni55!rpA<kuH_8Y@G^VkJH1%erwLH(%vA`Hi1nm&dLcsKW)J*7Or
z(hNRES!2RKtfzDFK;I~+e*28891Q;fYK*jg3zZBP(ZLJl5Mvf`gm2x@@x$dNZHj;z
zrNI6Sf^^lhdMJ~7&0Y91pQ^r7bpDlk^?QGTG}B|z%C+apX*wV5JS6;zZiX0Md#5;a
z7iV<1EZJPV#Az(2O1{0ebKvDrW7fL5aD_A;OHth>_?-<7$)PRF`E(kB;VG#KuyEW1
z(bRcR_{xd)n>}wMJ@l02s%5HH?GDCVSwG-UET_D_bWembW@@{Z1-ic7K_NW3|N15r
zL)aA1mHbS<P3ixRad*;*?hp0TO)S+I+|?|vf8Z?scnGp{Y5WNKu2iOY=xKK*p9H7W
zKB-N^mHIeg<h@(Byh?pz;T#Kq*q$jw5$*SW^e;bSzoN&A85se_Bm>CaDKWxNwSw1N
zE_9b7Z{l2m;Z3qi{2U|%(8Fu^WHw@A?)H}@gA!s9YLFZx(g+M+cOu(jqQfH^vr3hV
zl*_JrBX(+CFww->jAi@mFz;l|8kHEb)c2E5*%fQ}{e$@qB{Zmh6aYjdS+c=3X~!RE
zg$hhALg*{DP$<(776!pAcvKn;e-KuQ!dI~pe6xFEiiAHh`2243)Ae=~#=nU_&u3;@
zQPjb6EDuvSXhsSm#Wnv^s~jjuUVbgB!(^;0@78Yb-!`V#=aH?$M4oWUmrj;08ic0|
z?VHv5fiV-#JTSmNbO9jk?B<vVrQ6>M|BeMY4zWdr_W*|af0C4H)z5%ZF+^T~LZr83
z=;WeRu=Lw%*4>cw!s5wWt)<}0zoNY`jSP5pAhRM1@Ki~HJgt5+-&a_7q{!AcI5CR6
ziYGjkPm7fYqDl`BgPpXCjeEVjMNn<}M6=_47zQ}nzh;Qrl!1_(s$W+<Gt(pH$6r+E
zr)^kvFj`P#cJQpx%c6WQ46N85W5u;(1yQnEIF`Ju^bhlWB{9-L3@mg=?&bgYCGC;0
z&_{wCYS-W$ljTHaWP8=~MarTSJMl-C+InEjMDo@f46XdYm1xkH?Qh@(W=R+lkH+l_
zUCQ<V1{N^yWJjTlmYGCcT5uK!V`PxmB5w<o4*<B{HZTq*ISMuiqz(I}Op^nzE6g08
zTd@$7`}t*&SnL%f#8!1c@{p7WJwVfNN0cU@&*UO4KPlho7+;ofb{%txiPXc4!KG^p
zmV-P3d1s#ZPCA4g^ceNf+AQsl<M}j>>fQ~6jn{>8Lzw&Q{q?}4F8W@+cdR`f1;P^U
zvl>Jj({*}ike)pv$0miw_Tvmz$TaR7l+kMMxpKXN?doSGlEp;vDd+V7d+WfX3fga-
zT~mB+yp?d5NUKhN%THZ$!y_LSwjYbB-hAh|XrPibiO-KQMxOql$*+aH!SAt>Di`QL
zwuCAyT(N&@!(K`x)j>~DUwkrddG*77W|OFOK-`TA2R=-BY2t@4gE-W$l8Qg3!+s|A
zvxKC)=d*+ty_uv7u4NN}G6v$cgigR~!i0}mE7$=A%#m<nrkcF=H&pYt?Y|tNMEsxn
zDO0r>tBIq}RxeJW?UbO}9UZL7I3a_;3b*WL1ZnTp6y*vUQ!f2hwJy5!WRC9)>yGB^
zU)CZXN-VA7U`eP~O0nC!gF?UBs|e9t^<f6L)5*Kx+ja$Rj~$v?BLtO}ku(9&p)Jr%
zlbLz)Al&Ks5X)2L^sd4b+brD&n?9HMoe8(3VTF8=5#2=YE;7<+6SQeGBQ0~FjT|gB
z6M5C=7o?$S@bq`k*KiP4k#J$S2SS1jxJl4U7_&4F{_rt7WIUQ$Zx%4D8~;@XGH8!#
zgAfgJ)##97OPf9b@@izr!&C<u3jc)(cTs3b6<jz4-qfBK{@nXj3&lFUuBpV?wX~i*
zM6-VI(DWxm>@)?~YTTpcmx2JeID;K$8EF-CTf8!u*DuqN!28G9&CtY<Fwl2?^%%D9
zD^`P|{<acbK}dDzmSgsqrESHvfkb&5xT+cCKgYI6>fq0URHujev^eE^1j0$}#-cqV
zdKZY*kHzrR#yb6}ySn);!$Ys28pOB(5VzL_JVd-0^3VB?`(-dL$-n{S+<&bmH|mPJ
zR2h?gJzHS&tKQ_49csJrup?Hhtib-WFG0BxPaFY`WhoNL^%Xi)Tr+5(M~G7kwK_VL
z$c*^?AAkCJ8Vn~4MEBDkUD7`O2#1{;v_0PUmfXY&w?@rGN)NcDIK;-nNm^cX&u7M^
zsw89;J`+`-c@4Kh<G}-37)1op@(4|YIYRrMs@oLcbNqjP-!t-uBax_u^;L{9*}e#y
zwQQkouQ3_&IBy{%^q4rEdv?{k-Yoysd?bOusK9o_M@b{&7x7I;3P;a~ua<JO9-{v7
zcIyWyaVqpphqW{<BeW63RNObKgG4y_{Qy-MXHVSM0OI&VM$-u3bpNL>1ia0)FL|7G
z)_j4qx))g(8B7ch&i9HoH_WoKMFr_^wI&itz5@Kpwqf#asx~Eysq6bDoNH(ktO65j
z9E8c}lNK~%1#+VJ+&-N&;>v5vBD5_@e`85g^u#Y=)EX%nVUB2hRpX<OOV@9s6P+N}
zUB4-j?HJvTnyOC$ep(jLK6sgr;HR6&WqYOIH1mRqDQe;@dh6o{G7T`Smgt{oiQSat
z58pvMlgwb*b5|J8wl-Yl7%|@aU`EreS=?m)DAT28lz<GU6YYQ566R`tm`B(NDCry-
zUd9VTgJ+xa09gE;oiU*~FvW@-t4w(jX&Z!*)E<em0?np?O3i%p^}f-W-I6-pY+cU1
zfGdp2Iol@{Dq%LlO5{3!SBnfIOPxcbZVF#}I2QX@Q`sTf)b(%zv6|cma3WB~P$#<=
z($$ApDfh{~XuEop>UqORfRv|L@rqtFlpJEGLjf(qa0t(L=If$A^F=28JMu5f=?lFx
z4amZz)_(&|<n@W(t+fJY$Nl-}nTFb+908{981<H@m!pUycknMe$6=B?mU)DepM$e=
z)-iE}X?P!hKhh`!nYAX<P5MBb8GCWkztu14m#cB~VHw)NJ~~N^wB_BLnHu=(>bAKO
zmtdCU3)g}LChP4Ph%lpA&!XdTbQRNQ2ZX!dLT~_x0|*Y&@(y>4MGuRqJP3hL5j(dL
zQ8F7n*CTJzEo`6UGRo4;YbQ{hv$$GfESOsBKT3+;x9T@-C$3-6e`6a@`8D2p?Z`@9
z{MlFlF}EwwT0OMfGP?~FIMhk*m-;BwF9TObGxi<?yGZO3;`37khKHyz1X5;}v)L7X
zna2Z@H(?i?eP+A&1YKJabp=##NTq2i8%0i?RM{>Zw;^-d=qShk@{_R-*vURm2p8<y
z7CvhAh<c3Bi44U8Adp^OoVo}7QL8OC3QQ2Qx}oH*vcAC|)H^hVcBAd`?|FXy*P$iN
zG5E)KF-3D!av;%(3Rm&tJ3$Zp?N0fXp*ye=IBU0K)o3bsT<q}s*4;&Ht2%8ORa}y-
zCx&X2mMBHfkx_-&`gmDy$fUk6pvLHH;~K^`b`5`OA-H{M%v#YGUupQ%!k^BNTd?1a
z%>Sg4AjL%^jLg(07&>%&g@eeJi(W|G94A~<F#Xe7(O#K3)mp&n?miUd(ivZqN(^{L
z`o3g2mu7xUE|0zBuoCWOb<4@ZJv6Apn2&@c?Nk5)4%kKqF?B6iXmuaxD`cQHe>U@9
zSG*j}pVk?!e0T<q+AdYJcEdeChwk?8yAD+<xZ@ft!QU)o1rT!G=r!2Wm=Gn`wQHa1
zb4NiYL8~a$fPL7DyN1uB2!n_H8GsgRcC*;#FNdIvpARaTPqN*o9+(7FVS@@vNo+9q
zUX@cT8TD7rpyZOW6gD4Cw)ZE>pW%wqa2D$L_VZ<8_<qy1$voA<3IVPUfO2XiBml5B
ztyAHm#<=-lRS@t}jE`VxgL6jTU4IY_P4k|*ryaj{|HsF(R_Hc~<;$2856FUsUm?7c
z-o-+2hFenq*(K-x4?;L*>7&^s$8IV)l$UIx{%~&kNlta@Ie#07-p+eqFo8IPz5!<E
zb-k`Cm#uE|w(nK3q0)MO%3kJ<1BcqimIB8`>x1u$Pa=IQrfr)rdNVBc1xGO8mFme{
z>-wb0+n@fx{q4RrxevVqcnXW-6tX2MR+MqZRC|f3u70I5jZM{V$Z{F-e&v*7f?WPz
zA&s6r{oTkaI^ya-XQP-3B5Cu{xvg}ggL=%d#PbOAfKxpm{H#l2c)|G9zdb@KB|?Nz
zK=svkciP;l0_wV^*x^43L0ayi1L6mc`W3Ql$Q2Co6bzu;I4a_p4F+ydl`K6GCqcEh
znS?=p_0d9fHCr_-jQ}3J5szyv<1>sb=uZ9YW`K-C_s^IAa`#?_%epcarjDp0w^yUg
z$tGXSO@PlVr}{b}+;iW?1@q^MnFgV}SF76QGH+p1b&<V3-nBfBLmupkID^{?2ex4;
zsd=v$hYFDU-oWfvd*69F8wnp~m9pY~q4&?P-(&1l0aMax(7TH?2$=$6PEJpeER)bh
zclXlF!<D*UkPOLVUTz`nk<iTr0*oWbsWcxkIponaKc8h^BR`_9>yt!B!V6O(!%gl<
z4B#S$Bd6^DPc4xC)Kqq;L&KEew@h#ES7GT`2cqtVnnuUo^N9{UQTI38{$d>5RpQJT
z_hWA-1YxDK?yu4?69zvh+LBa;IiN9yJG}CR#h2w_P5eeW(V;LK9sZms7F9c<C5|1v
zhuLVtpT37_b65FGBuhWOJ3p13xk$KLeFEVxpQ1atP~3(`uVa4GP(Oy9rO~5DM{qy-
zH6FT+Pck80HlwJBK9&tWtLi>`-$<VIBVw^e%U!cVPQabl_24kn-7OF^jlE6XMy4Pp
zaqhWQ^M+CipJhB;jGU>pnm2nYpLXd_SQI21f%4ChAc0{-wbwN>nlbMu<?*phUkN1V
z_tSH1Z0WK6_DYAR+`^@o!nn*S>9*zya$tiq!Vqc4YPE*em?KvAAa+W+ql7Jd)=n4Z
zI>+NU;C8^M6^^BeBw#P|H^1dDj1zM&g~`b?{J)pn-D0dj6_j^sykTmPk|i+rP$MR~
z^pMCV0BS{C+Ram@V!XjWk}rgB!1FZvnJd0Kooq*wY6z@V(36c*X7)TgPv&Xl92lLJ
zi%6Z+iy%=5W8iMxAKAUer*28cOW`z%X`x{&2n;}rtDwm`9Pn*cMwsm7K<T2h^xF?J
z*$JbydJxBwpL9GYLbR|KX;-8Hm*9K;KWx2Ia3=8<?me+>+qN;WC+5VO*v1>%dSlzR
zZB00_ZTn>J{Z)PE;@ow0Ro`^|yVq~6=c#9HLvlRH6jbhuCY(MqBUa@(6{x-)C7hQW
zh4qlhRfiA1d(X*R3!pxQqdtYoIw!)s0yNLjjcdx@A*QgO%Gfp<2AcexRHM+kc)}7K
zL>96CtCy&g4fS&PG4qfLLV>7_Wzw=Zj~vj7Q60KJ3)_sS;BUDMY<IM0#^*69#7;~T
zlrovUl2Ww}Ok4`7Of{W5NAx6(qYAzCcV~{pxtW;o@N)4X>Yr1lhWu)aQZ)lO!CS_C
zH4!ElxWq*w_KL?wSz*@ZyeC_4GBFCVP3~<mrws))*zZi#S|XwDfJ!x4BP6JV&8IP+
zbas*li%AvAc`@}#6>6-VoXxZwuXaB*DUOr%Ot<OlcjrD6M#a_0aw+F05o*~7&ATa{
zWH@KfDS4hw=PU5HDomo5fBQf76I$X<i)En}9(BIFD==F%Ko*P9r>Elli{JVqjUubz
zi+>=(`1JC%D3Zw*#smAOf|r@s#R#I6Jq-wl?eDJVxv*@FMqwEDCp{-#m%PP-$gmR$
zD{OV6rH}MqnC^XN-_oS1B}(f={~y(=&s-N!TLtYozgy}zlD$ZCt@q{?<WtdPCSDfm
z<nZUboFv9W>OU2f<?c%TDl2vl7-_83e`ML>=~MN)<x{_D`Lt2GV&k-_+`C5C<%m$o
zK6M;&v-*1rdYh^mzW^8*aref-*UB5_;b6SUMvpSojIA)eJQ?Be<8Quz(BlU!u15gK
z)1|__9QSyBtqVQ5I_-O_*Hh2dc*39xD7`EzMLCx%eURVycnxbq{VfT7EMv1q7pCoy
zn61x38BH{pVyMnTHW!K)=kM(kk#PJL#s|5*w%~HIj^=+;duCNM_*Y#Pa+`nD(?_zj
zwTMZ!F}812zGh$OC6CUnrKsortQi4#X=ZGbsU5H>j!M@$aZANbG?Cgv5M1|~zeG2b
zorl0;u+I_H#*d-f68qV#H=;u_?BMROI10wVqW?tF+H#wjE1i(_)NWYo_DKu`!kDZ`
zAXJB;(5BH(k!8Q@ii81XnLFqxB7(*wGc)F0nsA`!T^>%*AAG&-%6yzR#J~4{T~tg>
z|M<WzwOtD~V4)U7JDB#mL$}p!&Oh%&dBfVf?3hzb`d4QE{>$b`S0m}KgmT6tIrz`D
z)ag@;$)^qDcd=(a@<kCC>2DWRaEm=Z9jTR!WIKqz@utj{h+#o_DXG#Qv=a8ddlsjt
z2_FnMp6KF30`Be0`k`b4p$s|z+!g#;o|rz3N~#Xx0RK=b{347y7E_FwNRaj)qcckJ
zl<3|{B85EGBRRa|uVsc)V2j)yLOCtZtH4{5Eos#DT{XK5)RMGZ5S^qG4|`*rh0?vz
z>aVc5xBe)MGt<=qTEVpfT;I*|N4s&GPT>GP>k+}$w)W10H9Z%5FK7_}L7#&DZ-~lA
zDCqRp*Pk3<gsM&{EMzoPK4=+7%)*!u3WM1QTtByh79@71>f&N#sv8XJ+K))63)@D9
zGW<ZmVUF6)Oa<N*u#K3KIEs?CSN7@?FwfFV<w~l9hoRZ-kwtCU0OcFnMG2}z_a>Zs
zw*E$y7AN+!Fr@Te(7s>*%c>wskNk@b?F}|F+7U~QZ0@kqQzm@NBP@=$6lj6i#eQNz
zb`iT%dnjD)#am{mV9@>nl78zi50~J}aKo-^j()*`+V{&~k||yPr61NF&&`zi?-evM
z1rvOUQZHEK>_H7ftyB8-#m=lnk#AW|UBXVRESt@l=N`w%fuR`yvBTHOV??yncgv_y
zaCjI8`2vDp`E332E24%OF)(=T8sJ0bc{KD$qJKWxt->@V+%t%bvpo%181eqrovgkD
zix&gqVEOf12j~f+k7t5(1kz0#t}-+!w-`;}k*lSTz{4v{`4@De_iWYh`LFlBVCqKx
zng*(or|9m(i5CI^Qe|NwsYXd|CnTbjIi<=08t494)){`2EDreZ!Vy0-lZZ|zIe(wy
zz6)or-p}44dioT+d$*kVQULl_0|%Dm4mF|8?=0d94n1ASMDm^1xh*1ru8HfM+ne>-
zxg;-stONtu%c19V<;<vPv1wA>-O5+hj9!2W&%*HNTmw_U-MC_M46}ln(xiZ>j&qT{
zzweda0o3mw);&<T?a!$KgKPg%gg0x0Tkm;A0-eHt%pupVtUY_z{dL%Xe4J8m`gb8E
zRhb3fdPhpFfN6y{S&f8B7V3-L8yh5pdyd1U!#@Tlm~8ABKlK-EFL^<WC`Hs_fK(}~
zX<_5WyKSuiu%O#-bSSFoZB+#;ZLXY-jmc0qg#<R%Us3AI7>=BWQhzo*Ca`OoLFr3!
z5H1EM3CFGG9?!+<EI_LqC1qf5VDOeVxS54Y{lx_!xgojfyV-$5)Udvm4Mzx%rm$AU
z>{a3sa~{25+<Qi=-Nx@D3ik&a>K$ypL|_eh<+)4-i1quE_8meN`J|>4iv0-6yN2#|
z;!60Hzb&lP`;j7ue#k^lh=`8FFZi4p3L`lz?EN71=t<O;b@HYYJBTZCjQJk_d()h;
zhcCumzPTa#3pI4;=gD*(MgmdJD3V7AMgCZvSi+p`kQE!=4n-Rk{8Qm%0_%i$SKgLv
z8buorV4Ak5bSA@-Eux&0^0Q1U1mgFB)6zxCst~}JZ1UNWTfda~j*<~#om^UvI=7LA
z!u!FGthWd8XIhSPxg)bXP04#p(e%1iR+^uCKy5&9W6yzzK#s9Ymg*TQrdL^$-c$dA
z0v^#VxnarvPGe&5P>6pn{#MJq0S(+E1kPJB05uZ<Pb3QDZU8yWEZrkYcC|$%y(|fE
zm32TM!0pAz>*~BfZw8bQkI*h_{N;{fQbrRPw>uS!ts1oUk7$B6nV1JWZc;^_LXljN
z^M@(PnkuL`C<Q)s!=MTb8cgB^bD318RHk(=%#30(ay&9jXjo1ctrDpcY7FYAZ8CH`
z0D8xcukqI;N!{2rq)=Rn{59H_fu=M1OhD3c{4vbPs7SlDYA$|`YKPip*H2IiXzI;(
zXhigEc{&nO5>!N#?1=we#3K=b%B13h%DsyTT6D_nxq!*Nua}K7(-19NMyISWhjZPS
zNxr+IQE_|trD^?B6oYR1H`1iQ2OxwtW#JgjAYv)~0gnFJ1tle=l<M99^lo_)i$~_d
z@be=p6hOUz*@3FeFh5RPv6CYDMW11mnD3RpQznYc$?!TlJRb{<hx4Nihch2SxW6w(
z0>7jXX)Xuh^$d&M-AM*1Na5a4&{D$?Cc)aDe9%9@!Z(0GX8dLS>QMfx9lHt<^0+?G
zQJmWKEU!Gw(MlY7P>VqS^Bsst3POpH47>5_(81Wkl-CJ^0sKh{Layc=Y;J7TP>_i$
zKgH*IMPru;NZtk&64sS>;E<b6-CTP`R!LZ^G_UXL!8igcR{pDcR^ogncNSfgW44s6
zxqS^7C&fpA6QxUD`$THGP6G$5J2Wr?+etzA0MjxGMz)Uz6jZ|MVWUHj1L&IJl_<3z
zwzm>Q3^VKDmKa3K<>j;+oy>+OIzwZv9x!`j^%WXP4_ylKtO}>D4c`AS7|%&PYQ662
zw+h^TBIa@ZiIQ*b0*imPB_6F;^Ws(9E)X1_@$&Fe-*}s05$fNytoj{KuMjgta^pOp
z1n~V{w@`b7h<KQ%(pK{W=E&AWXOFMem)PNkvBk#gubNd!vX~dwV33D>MM*))9=;38
z37HW$xey<OA4CPSYce+5PVNZbkJox{^*EGbm@yV0N3cjxAZ{7QQ)<JPe=zl;Ns4>B
zn*5{hZ(SQF_vDI^jhKo4=Wuy5xU*iu6y7th!i)uxRI~`cV`p+(R1=JN3zA_z1;}~p
z02&+_luR0wOdD59CwPWoKhdHaIOFgKW!9^%=hk27Kn8LUIw^*~O<g72jbZ}(@t?MJ
z9~*rUN7Px+QTs3;#(;(eltY8vf)P~TQTw)yLp-ZCJ+MR5FGKqvx!Zbd9|uKYdzYM2
z;V_$fynSFUIfyWTDKFyqAbw^0&u4yNn_Yxe6Jt(D;dssdvO=lvPszBAirCFKK@r47
z1Td{%z0!%H2>pE?-G0v3RSj*#&@%y2TMQdPy!@!sBZ=fv)HuH?6GBcRF!CSYmOmKO
zN2QDAl)<@kkc!{yp~55Se2|>XEe_eprL+INy_tTsKj~f|Z`iU?t~Di^tkH}VCG)!^
z0U=pIhyaV|Fwv`<Geuqs#9+wEyA`5UV<MbiSBEd-Laif%r*c)-Cf9Kzu;PV=1Jo=X
z_5Wcyo&LUsVn<~av;-BH%==Jn)1#xf)Pa)vZHn7%ufSdkLd1B{)YT9KcFLGQqOgI@
z5~d(wc<b^=nGhKA${CYF^+)#z0!aXK3paEjU?}-{G;-UMsqW^842#r)lQv6Re$tdG
z%Vpw>WhC3c4Cd|XFuAO8VoFBEFImeuHaYd=Uif%YsU@w|u}J-bQ;Z}g{;-pW!pVf~
z+v=#kFePbpw$Osf)<24IL*Abc%VOPf%&(D7DfN3^;f9P-(Ap$1ypNhF|E-0?p}xcR
zTa-E3N}_7dXIOg*eP&Ga4PX%rPK^PV-vviKn?f3P8O>cGU;K@Gvxz+S<Dk_Q_DLNt
zIx$g<%iFWlSFrRF_6GMmZAd}6ut}{b?VM34kC*^yJ{4!{FH4;Z!&RV9WGJUn#NYeb
zTdi6NUmtDfU)cDOQVo;lMk;3SH_q&B0CraD3nofV<LWhmY7@<+K^=sHFvy(w@|<P}
z2J?*4qO0i%-&jKKOF1fHyv(aj2f59j8zk$DC<KM<TBIYY6<4Y31Pyyd;lH&~X%Cm{
z3rUtuo5`tKLs48zUa@Jl3XYJ^$C>=j1oY=7T(AvNwI9SyhN_wx6HR7D-`P@*04<_}
zIRFX4xyZn1;xf$KpYWAtQ!Eq1;-_oCl@`G!<6@$Yy_Tcyoh{lTmB9?p*T5UEFrGX0
zHu?&(;3`enDzUSJLI~qLj{F3d7>P9zo!!VF36pY2-EheCEn-t2qZO}irP73Jz+Plw
z1nT=AjgT3W*pfMpDD(kkcv+Tr0FIGnJ=<MK4+4eCk~fww{<AHG=0wT3g?P}uUZ|I!
zWCQb)P%@59@nhIUQ|<HzRfsk^9>o;yFPTFhPHc4eIT4hDkT_*x!dN}FdV4WB1mBk1
zTLs>XgTKx8hWwWcLw~v%|1|f$pae!Jrh(p+wDZ)v-Y#D8jDdK@q@A^!0`yox$)2Kl
zs_<LHo6h%np)I>kehP{!5wbz_SnCyUcoxjbuv-WqjDc76j$Z#`->fIjAcud-$>!Ph
z)qady@o*EOcA#<qqF%xV3(cZ9(r4Tjhv;eOQ{^8x)?&U_&s7+uH9)%l$w`%Cj^9BD
z6rZ4c4hPuRm9nUUG)oi&!_oFSf>G~)5yRYg0nuqdr~q_UBnLydH?lnjsH8Pw84~4~
zl=rZ&+;oqIXV_dAgt_0~8?C*s-o6l&QhMfYKev7H3t=PAV{;URTq-{2vcc;4F5Hnr
z!XrJ+_`TMdEA*d=#;xc}{%VEIh;RP2-Q~H<86x-vD1?4oJX~I5`J%0)+!%gMbjK4l
zBF{+g41oM9V$WuxX{k(o9F&6+mU6`^5{R;jHw&)e+P#seWOwYncE!9W@z{;Aq^x5W
zRs#N%Ln3%B7zd!n@V&;A!-hXD;BO?Ry+K&!X}|bVoXfd!zook>KBeRuz&n7p(79Co
zjyRBh;KmGtCR8QjW_m-4{|012<aL7A{^%ZJT=D*0V#B&J=^sB#C~moC<kER;Bid{f
zZucR8DzqR!0o##^>3ZYMlck-eAcq$l?@_Mqx!YJZp;!%58E*l=`W&^EqhD8n(OER?
zFl5|)BdYadE1s0T<)itbknlGDtw9~)<_PXQoX_x-14^wU+elw9ZB2Z?SHv0lX_@!_
zLTj!%{dFdQM)wR!s$7MkbCofcv-xn*xdWNAwtq6(-UVJr4)xbDmm}9KRg75jXTc1@
zX019fc-@gf{)aEvb*6-l&eCxp+SZ9SZ>Q_N2=em8@lCYRyEGrsye^<@weI6rn0Pzt
zVF**6-S**p`(#3WVs+~4my=FK%VUWjfG%_(G$3NY96AstP*QT>EFB0g0MdhrfOY2-
zehIe@{#TPYNB`N#gCzweS*PP9(M*!#-Jv9YkoZI80>(ij$}c9OA|>^DC~!C5K@gGl
zl&bv~!E5gT8*A)AZoa_oP8;KK_zcxEfgw4h65?`8IwFl<YrA=T9A61sLPm-^FF-w-
zt#}@2K17@?JUivlE~4*TGkEOh^4NEq;=}QJR1*V?*&L$d8Y$^EZ7+#)9fww-`twdb
zf5y9UEoAul!=7_ZAng~I_)VX$M@jn6(A{*|x#36)m)M5~$h&@nJ0m6JkM}Z-&*y?L
zx!?79DO{~CgL@;G8X3%a6cvc4^oH~k^^7`4(8isFeWa|nK|otWfH|zb+ancz#iFMg
zzY6P5KmfPjp?ZNiGvi>2cQe2IfPsQgBtOiiLnMH$qdxwg$v(EuAb@byk!;p}Qs82J
za^SDoUSUU1Fu`(P+do1|<4HI{S3Uj#_2#-pCuJoaQiA+HOW3;uGP87GQhF2dXqhEC
zktL-~o+S+E15wQ&Fi@a4zkPUwNzj1y3?Mjwc~**xY+Qejp*-Y1c7lwx`C{?C<6xo`
zTWzKZkP~-q^pt2UD-qVjGsFVDZ|`yLN`<-qsqKlzRkCPZ*YAB5M~ACc_pf2>G4+jw
zUMmhyaU%_OQF^5b9vDU6go8{{LV@UBbizhZpScq<-haaqMY<Q96C!nUEhSudx%o~c
zPT97Hrl~&zv<PF=m99n>)pi2M%r+FNHms*?$7chRU4S%<AWVSn;e*p@)|oG`cv44Q
zqNVocD;%NQyEQ$UN>Xis$s5F~!jPF}cqw2PYsBw*joPeqd()qokv#Nu`qcZ=91FAv
zMh#Sxc<E)@cpbx+CD1de&?>~;$=XSVDh<Pl2_8a`RqK<yT2&;TO^|uc$_eeI+SMFZ
z<v-5!|65gh0Mtw#S;)~TVKzFhhiJ?D3V+uurtRKMG*f$CO^tNqFrxZuTJ7B$p3xy+
z7?*}rtW6loMHL}ijU>1sO&EziiyqYv_dEIV#xo{Ge%n*rk~dR-gcV9^5qDdqz7E|w
zby+R6;WxgW$~P>3A~Y?m0ZJ{A%Dy_4)><r~IJ^j)0UNlbA3Y6NwoD0A+J}Yon$Hhc
z#jT*8KIZ{!WLP9!N~ie#$aKHx8>Wo|Em8zoLI)`a@ZN>BIs1OS2a;gk=JTeX<;%L%
zCUT0r_?PFe^^}8}ts}WeHjII1{98*WTa@fJu-*GXRA=uE6LO+kx~I8Kp!alZNg-%}
z;5nb02!Pae4w&aWX3akbwlN~r3I)S@_<hg}4i9$ufm!d+$Ai@&o1*s-%Jx-?p^U2R
zJ71$V1&U^V+z-ZJuU`2}?A2B3`2LX9Y^K*d@YsG2`zd}bu|sVoZqVzzYTXF4Y(*NP
zfW~-WSyG&`=$_K3d~hm&=8(DBdH!_VMjz|C4LDR@wN3fep*9EZc^j~v0sJ5ouz!b^
zhmXu>;k*H}!^^N4xR^b+_ep~KOp#DQBW-bSHXrU7$H$GN8c^DFnuK&`_kWv?V3Qx4
z@`R@6DK&#QUDY_uZMpAUe@iEWBcu^4<6s;bA~Zfh3$QEnmP#ohiFP{_8DM~D8oUEG
z0WRbpTGvo4yqtpYR~j8oj#AqCGZnj!w2IX5>mx=Rk@vE^P)k<9|0WkA(gYM8Ip?f;
zY-j&9uz2s+1eN^r#Mk64tBOq}d^Z%w4{R*U0up32z+b>^crOtUkb3&<Am+W`RFgfC
zCwx-WeaPl`3fCe*f%S8P3Y|jU2S$xw061f7lFm2`^d5;IB}d7b!kCq8bq@PL`W7Se
zK}nkD?Bd*JBCEoEkVrZRW@lM#`i;5F?@<?9I84!kbLcMqxN42g%h!&aFG#MX^EF#h
z9jJi#N$a_U@_OhuqR&l`)%CHwh-($ue}zkT-N#opzl5L8G9=U(w&x>x?4Dry0Nj-=
z8k5BMeo{)mm?E)63#F&y-dEjV+Fr_aqOTd;E^KZwy)AcLSzftjXpu#wKI~gba0jJp
z_%yA)AbcNW{B=Z`<3kx6MAZe2HzCjC_GN{JP`7iM_|+wn&{JZUQjRnX*5{4)4SxL>
zs3+o;ehoTtsX!QyN<}eT&ZvYOpb9GpH)N!Ytsm+VFrF2J6fmKXYYH^}#e5d{;dTdl
zuSo|Q^}++$5Jw1|#god3rcu36z7aw_$TFYFRbmojEuop67S#SCWu5gZVJLy!Mj9-p
zsY>#AhPk*@>TKCWj?<6V@w#{EinBQ(7+q(78WxU(w`yX<JRUl`<*t_L9RWMGfRfKV
z!D*RNYrX?B40uKlfJSZfHbx}%>&0eE5e6+%6SDc9pmFOc_h$X)cs_Z`Q|121y2_TM
zyxMw(YL*{{r24fIdWU8$Dv0U<u_xXTLayu@&&S14q+R7kVYVa#*LGL`lnO%wW`)yP
zsOTF;!+4HP%P>jW_c+i|$GDLw4bN_t)3b~PLwrVJfXOJk86tnnUrwE9Q(s6V4_$@P
zdAm~|mhC5S^!^KFO55y3gOm%>@&AI_a~&%x!jkc_RPKF(IMyX*R#2TlD116u42oa!
zMgLaWvL%+8|HHEML)YeEKooWmLTJjC<3E@i&m}-Eb`WL&ErY%9U@ocZ)=%OD+nCBM
zM6<`t{T~n`=gu3fOgR21>yM0cCCPiurIfn3XSq6d?AqbQgF_vNW^rFbwbya*HS&8?
zqJUmLAqFq0^G+*N$WDFgcmJ3;<5l!I$$oh83z(RRqREfZc?`8f$V@T@HH~_Sg4A+4
z={OT8auPh?Av*{SAYVi-onsg1yUY3?s_b7htjdjT_)f!KYc0F}!BLcGvG7N{Z4pH~
zzk07vJkWd*r(N`}Gf7Y?j#S9jaQWI?t(sEGP%gjWIlP=$g4b295jhEQ!?{$z#RY?0
za7NGO4nMNLyKSvk-ZWrhrkLIg1$B{E=u*8M!vnMt@`oMZ`xE>J!qNT+nDy;Qor+XM
zCk*-nRfC}*kp%wAM)=G3U<y$tor8mBR%03xjENL{tR21if1G$j;|{H|-FYr_q+e^}
z>Fzpp6^sY?IWYu6L5~Q5nL&F4BCNhkf7`YT`VERMNU6E#Pt|ksC6Tt()h)@T6)j0I
z$P@?wZOe`q<J|nS%|*MgWkvklZcnF;9yPuUX>^HVaq0`oA$7CH)q3k&?dzAapVz{D
zOrMZUW(3dcSe@wWmEXQ1-rK%dDpd0HLG)zE0r8wvQKQ5i1Ycr?jBTpVpPnFDeKMRl
zIx~Z@!)s^0ahr2J<&C(k#$E_<-#=qMVpB1|9h5u0iWZXm2|4#HJ1Fdu(?4Kp52XE1
zq!0CQRP*<qLT}PE$YWpDIWj{^!xZL8-!uy8Y}r=n)*+~tNPW%$X>7XL#ZQ`Bl4=Jg
zu5nBtNEvWeN_jKl_x^rtQ66CGUp7@5w3WlkL~}>6P1iG?E>|R$!qVT(^K8|#gc}+F
ztKNWf%IJEBMApR5j+#w&uZt~3*{+Y!^2fN>8N4Q+3RlI3<)6z{{{O!Cymqp+Y}DDP
z4z#=eDkx74^5gYn?8+QJ)WJ<h{RT(sP~`<Bd1QlSm;Si*cr~s5&A8a7a0+n%RUTpc
zRW1TG>?|s0ygAD@09&v77il;uEhiWNcw3i?4@n&sTp*aDM6<TEbIWisZBVOn<@+6W
z&Lxfh+Q&fT05<~Mp1NI1e3dV`hoA;g$*)n}CUASYm6^i$O^~+tDZHQoB~)$e^kq0`
zW8qKDU}xbWNk#24p`&ejqD?lLq|GTsYEHPoyC}m$h^}gN5Fz$6V#2#`wMEMh;30<@
zQeV43LfGS<HIJ>L1HdrjauCjpj1#66j{(0tQfXG&=jZ<GMPnmoh<i>)L&2#^GJ3-f
zUNb`G>>B8`YfCGth)bd(DYJobPIbi^;i1;E!b?GMDa;GLN{$QYH$mI)%(c1k?0;qg
zt9Ml%%)h4#jYr?s&?B??DqBehoQvz9G@CHFn{X^!Mc4G(G^EP!0na6?p-RKpCsH>~
zI3l5GVQ*t-JLx1*KCCR2kOeknL*-ii!P%tf2iQ^y_m~4HNDyu3<$V&%etAu(Yg}fj
zC&>WKseDtHacIgrwzl7)Ym&{Xl+f6VIHD19|3mU1(3%PkE|K)-_V57Ji=_(Kt)7})
zgO{v<@K&ASKF~3v?cOTd3wv;8#w2(n!C6VOnXFnD+@{f=QS+oKYFt5ys3}^cHM|dV
zEO3W9{X<l7X@$U|wVqEzlu+Lj=WYh0j6Sv+`YH3jge_FaVgSzPG0x`ix-(x1-1i@2
zLc*?lAkd66ev3FF&;fA1I7W2_tdSbiajOQO6;Dr?PuYbqIum!!SE^ge`$IF4@`Ga~
zOZ{AQP@PJd!UQ+68$J^!;Pu*rE>6;$7a;gkDzLL{p5QhSWxa-MLPk%yzwL3gisd}q
z>BTeZ3#HO|5jE@c=Zq$g83M7uHr)sw5;)dMqci_L*ke+KB@ci#sbh8)hlkLo)K2M<
z@p$(?s2POJM=w_{B@9x;;9vR3)pT{wyTnnUwHR45z3J8CE;c7~{OSd_Nm-}*Ik&^Q
z#A^-8hHAwygQ2_$4v_F*&%1w%@ma~Js&5|&{dX4C0+aRc5Hm-0uTi^Z2^=-BPmrch
zi$dJKS^T<V^$f79d9koikBc;e(L6qM^bK<+M9>lW_!=2d&_TKbc^lso3RvME4f7<c
zneNlmeHkkGuc`ZFBWW^zCP>77VJ|2JJOpSu%eR8g`6A*1j6`lasI+PR5oce_*Mm26
zMmj1^sa(ZR&4^Mwd`@9zX`)n9>u)RJof*u}I-)UvZUAfdb*)LOT-$UZl9vv^-Gr$J
zS~b{Aa)9=QBX`6~^QX@>dLO~vv4GtnEl9u}`PBR-i)=<M>57HZzREP~kl|l|eoNfH
zGhyD1i-V`8Nrb$mnV5Cnz@UEXKF4MpRLn5LP#2z<$+NXX3{da<io+@4DN-mbIYoye
zm^~nMAV6QE*{{(CV*d$W9RzdmA5Jy5@s^?Nyns^7FI?tIf@bo~fp^)Z7V_GYANdW9
zFl9*<QD*+zxNZS@hg#he=xv&Dcg-%F>V@Gzy1^wyu34teclXgwdC2=I;!k&57fZCF
z^XO5p{cn|Qc*Pf2(OTuPP|#%1{s|hqIP>Vz7Qj%Y$HODQpz%VroYKZ6)*TdpP*#u(
z@aMzybnyA=5i9s(jqng(Z75}`?=eeMF}|}My-;Ab=H2af8WjAyHbYfGiC$$N$H=nC
z^|T$?SUWjcSDm89p{<z~!!FxR({HUmdyV74geJwLVjNsjcGhz9=B{r50`-dB4|NP{
z762KZ_aEKC@P2YX#FWe`Xr;uuQYi#0^dEc}Vn2N0@50Wy39c*7dC9P^Z0UI0dM^6c
zQNcam8P_}&pok=F)<UpCgOZ(Ea5mQzD}5Y~x6Jr~dfR0hdU$ql>m4Px2XR7-Vg_92
zS&t2JTRy5=2z#Ibo)VmZW+_h{{D3wofK9Z2Fk@bnamLgj^f8==oD*kr8*08D#&?&<
zIXVIY&!AzBG#zMYJYT49p}6f$D7C5{0sX1!uEoeo?;)|wGrr-VLf;~5k%<j-O?P1I
zuexZ3I=3u8Ja}@}$~2O54xoQ5>5mK)VpvX>=6Ofk_8PrpD?&S6fX<I?XG2>AK(mP+
zKult~agDpCMB>d)jaRs8S8!lBV0+iK_h6<pU@4&ByrTJ|cb=?>*<xlxOzsIZOLEpc
zyoC{n!DmBV1$n_M`cB&;QOL1sJ;wPIbbZ2YnjiH9UXG2{!{DOf1ucihWEVQ33-dho
zd^}e_+Cp8KHPUgORynqyrD~7@9uE88!@$FHWh8f_^j9kn#r7mMkJX3Iu`P<u1t_N1
zYcq*Vq^d}oBrqE##4Qm(VCu)hRt6Ws)(<X)D5g7Hf>q(t&_#8Og~W6UBB|h%W6#^r
z9cKXC6UfD3f~epw)v1SmvRu2dT?bG{sq3{$>iN#k@yhMDa_z(O5aTF-(~1_MSAnu7
z!Sbg4xw#TLYLjC`aa=s|rMx@!^4ye-LuD-8PVJA?dI_fo&Ft@J^^0DuSW~3c@7#Y{
z1Y1^XMFPCbf)m0IsklNU)L~mJGQlnLk;qSyZIMq=n1^ox#Gk-KdnX(Pm2x$3Y|!yI
zf<6s|LM~mHC;8a#q_JFNQQbf{@4$`Qpi_s$FW&WzYGvx{Zm+(^bymLNqvepxCA8oG
zBgAX<_o8D5POAN*FS^M<H6{#1u_<Hd#CUS{Ml$fH5C{Pe3$m}ed#Mqzy>C#T{A2TS
z9C<I&uv=LhjZXlnEdN8!)D@RTM=HC&0}?~gTtaofz1HDWEOQuiA`mD9LJVMla#Kh_
zjb6m^FtCijyZA8DV+6FlcO7iGqRr&;DYa7nVdN(t3ZOOuA#ku;p(hH(3r6fXB4E$h
zADZau<?NjIxDB>#_hVuBYX3GYPPuv!-XbZCJe^DP2M80l?qutCgf1u*DK)5{WV-Ig
zORUz40gp<mX%}Ubr{)fZEiFQu7L{~_x_vs%eImoIB}%j}vt@knLhS9nCuKo+W|g^j
ziVf+Khv5GfNnW<(%wkiwv&POz+HwScL~cd%&<I1IGDteb1WLqE8ziZJG&<4agQgU7
zNQ3PCS3)gUB%<jQDyGtzR|4h;gYW|``;Ag0x@%?Oep_&bS#t?Dy5l#CUhHcM?0oo=
z!w!o2%YSk#+SLDmxI7=b?KZUF72oz=2ZQWBB{PK0ufpPF4|@?Nx>G5~n_gWF1F_JN
zJyNJzlpe>+01pOj*+>;fF^Q%S5BntFcDykynw{dFec_>vWc_pjX_oC8Kr14d<45Gj
z(eyQ+c`tU5SK*DIq4ZT|&DOLYkQEpSYU$EP1A6`Z?h2I==}63s=9<U?&36iq9sRoX
z5g~zHcD+~M6OYo^|8=PX)Y>We=rI;ezLNsAU-KHR7e9Y1@L0;<b{_{nS{!|+7853F
zuvuKV+{1Z)_n+In^!^30Uc;OW9!+JL3x_*%)$zw0aweMzLiG}o2F53pyA9fY?J7^;
zJ6?eQPc+I7kN%%%1WjH!hb1ES*ydJKL(#x%FgGUf(w))B$1a;l*edsGh%^gF)i%=>
zdl5>56dO5d{|izzY0p*1y6uam%epfhEwfV*FKDM?AG5=`Wsx7qGl?i`D3PMB)X9cg
z+qrL)5J@vUBsNBuTp?Xe*_OV`r77=C5^o5jnv@}L9~NoJVEN<=J1Z!v5_$o3xfSQb
z1$WR@kYd`|^I7^k#1Y4_7DebZv9q5QI4c5z3BW)oJveH4KiWa$$8YX!djH4wDY+`L
z%>}!PAf&VVY}neq6hGP~a5a0B1ycp<!-D<8_%J$jdZQ8%(i@Bqc{AN<!h=Z+adO<C
z!`#=F{PE@XhKSq)`BB9~WS=&!LU>Y5F}^~$Iv`LFBGSXNgAtOeHhx15RU5ln+-shO
z3xK;zfX%={Hx|9x&kCE#-8Tj{2|oB`7P#Q@5-`>Bk{uE%u=xlmT`d&U&NSH-*=<Oe
zE{c0bcEv347xd~Pm$RykbX0*}@Q04u4&TYuwEuX|gjo@an9W-5MGMv8vc=XM?rQ*{
zf+{d`jed~zzh)<RHjD>&<vJIcGVaSn0B|AP*Oe6%hWNH%^edwso7Uy_y(R5seh^BI
z8iLJA2`&r1y#|I89G!8=Mb!|<?%kN27b#q2ulj7TX7O2W$P?Uh$Wh;6sMa)}E-r{W
zf;u%g>Unb2pYw5x`+VM3CRtWE#eoDM1=Uzl?PTT@eVVBZBu4K3C8`%sdjX?b0l5Co
z$iy+sI4ayQ7?3UEHJwkjFZFSYCD83K=0jd~fv5Y(B~haGmZ%#1MdH$xeg4zowRZW#
zd#LC5{)RCy!e>MqcGd6<Wb86B8M{P|m~)W8aaQD2g%%3{Tcla=%2o_MC%N%6#1=JZ
zibSmb)@Um&0}|~2_c#FlXL2CM+OU<fBTPcON4!_*TA7SP35E<rKAHR*un{)?0){x`
ze0n|h*F*G*>pm<8rag4sfXO{Uo=)}0r_^6Urf#+3*#2#U{?4BfECvS68G^VPsCoJ7
zVo~`CQrlA4##t~>R?nXzF(L$PVZGbjI29SMB1TdBHYx6K0ETc1r-re01GH}`kA!f_
z=Ifej<2|t#W5pxWEn*<nVkH=w#35g-XDpa=W-$a-Ez-i=d+K9ydR&K*g-MD2f`3fM
z;z;_icxpY~^ymS0;YDS-b9^6eoXW<kK$%A{{#+q$<GVBe80!2jq*JeMzr<)p%1FPH
z<2l}(V&Ys6h(g?C>Eu6!-05(_g3?+cL!bQ;6;_HcA@44m05wix=aU>f;F4aA<0)Gz
zLY5ngbEx?WKp)$!#AVkbJc`g&$>cx|=@ftraE`Q*lBF^P?-a_As&|TAZ95#;Y?PNy
z$xp4yXAjLT45ux+0Ac#&lXPS){^b%gK1XMliLbE+xN$bdTJML=!~!v3<{O{8J^yD1
zWJ3c}BIC$-*pu<uB$EU+_RxDI#3pHCKMWPsCe*83o>&2lO3u6Wv)+>4!d|LcjWDnz
zdqhFzD9Q+IFeLF;C!Z&sdvC?Kv-HyGUXEk`3&Nsf!6zNyU>j0tGb-}LNXa5%4Nzf}
z!g%rn_}45YDHMOYF3u8+tKf9zUx|ZfBmMEHz(Uzt%Dxk-_wxdmdsf-8!PhZuTG}U%
z1VPW3Ewizu!CT|5TEb~~OfGjhIqV0pDfQRUZXCc=uvk?zC|9$lLl}9qYim=dK&|OH
zMi`~#Yi0TWqgkGdKeS=ZwS&AFajH<s_~tV}?8|FsOJxa5a}Ff>T9QB+5)R9|6oU8P
zqqYtm*$NE>g<lhk5nMCghC!<KH=j1Q6Lva|eOvI0d?5vV8BtY+-F_kf@Ov6|coy|Y
zzRK`??T+H?R$7;3OMblUp*36<#&_JUm9D*xxlyyuXJ(ECN2`KADJ%pTVJ&ifi2cU^
zeHH6tbd?XD)_eFagUR%S1K$WGi65rg*)GxE{*DzDY};f?2ay{Cd7+Co72>eyJmPg@
z>t+R;<*)*F(wq@#6*0*VS5q0=`~FhAA~t$vr>}N+3#p}pStL`$E08wV26=~rtngOh
zlfB8xR)oL~C~7&@?J(`eOinBNIZkr`-X82nz2ihovTpCikm6!ealIAr#}%Fka!$NN
zxc`4dxjX}htU8tnA^*AB<O{gQi#dDImx*Rh%0*eZvR4LDr^~9vZq*JrNSI)sP*D2~
zP2=`Frl(ZcefRB{ZQXJMyQm!SNC8NNne1(Y(1!5F3GFf1oD+nVjUEoBo)W;s?#$_0
z+J3ujHY-Zb!p(0`@2!ITT^5j428$^&$lpZF^91a}q5+jCtR==qXX!`NE)mdViJg`F
zm8;+?tn{RMI(Xf34y9KhzP;y=&KnXd%+DF0Zh}q#rZ<@ohgN+<Ko`;m+Vc+pw(o#r
z=rKY+pe%!Tv969e6}26oWDVfClVim8(>j1VQ4SWuKR`$|=&e{1+v~eI1^H65qmL5a
z<o8y{db`-9UimvAXhj1lxCO))l>DlA3H*yq8^+N2ql)gp){W-RK%=eo1a7c5m6v#}
zXhuv77)1Z<Fq8H8%0ZAk%@r>50)gj*FWpFuc0)zQD#mrJ&z)FVmLWjJtMOcxTT3ZD
zYucjdr0Zu0Z}8fKXTiD!q0^@P^eEjy(Y!uMaK19r^2*a6yYm5I95xF-*vExGV(jJZ
zGq~>~Kd^bgRpLaSuJxbB@`^TApb=eZdAb#6oPfsfy@ioaeoJ1TW_x1=)h>n)U~|vs
zhC9>AMFKp;hw?{J@^b*PqQr(8eS(Yls5+H;%0ZK>XnL<G8ZQeW?*}t$YwxP0i4pz0
zYpyqo3WRMkdGmgW>0<|TkGm_ty#`iyIoNHqKWo2+ymPdIP4VI<QU`CwWXMpp;JeM?
zU34|b)OCP-@JL1dOa<PP1yJXc?UITazSng6b${A!RgPE)l^@{z&ZGZ{`RQnZ{uCm3
zavB2C^J95bkZjMZUt&lRK1K$fxXZEB?O^(}hmLB>@uhM8b-o(7rf+!=@>=AfZRKr|
zv?x*an@{PV%$v})1>{E=SZTk6dHx%{x}}v9&l1O)V<n1d#`ukf75hBL?hIvyY|;78
zNG7wVpe&@(uL(ft^CAW4@ac+nj{J;<ue^;&lVmGCi+0zWKGx<o8e0yW`vG%!;H7@I
zj_5EBe-UO*)N;H)NRPGOadN}u6kZIUKt{^#^2-QSvuF>BRn&IC6bBy4m*@NBu|3fP
zk^fpT#NMuuQK~Fgce~s3v*F+-#hVbo&QO59+ZC&TZX3}0oG~U)Hnf!&Ifz}`q2IA?
za;$y0&inu`R4cpF{1|3a$)W?I#fe)-_c-EXqTz&{eG{!QRi7r#TQ^C<<-qXi$HwB^
zZiHRHzqB$C89ODp>#LI&<S8GN*6GZkI++tdNFGHR&?K3IE<_vng$bYdu^@mvIAvNA
zq<t!s?gMy$uQV|q7i{@MIDqypDF-)socCFYRI*}_(;hr5VjO^@b&oP*ZG9Utf(G`w
zhrJ{_M<q#fhf|fb0m2tkbi(}0I_78(bnnAgW~0WbC+*{uY!bwqr*4GV>}#OiROYWv
z`lc33*8ra8ka-f(GG6qs1z0%(J?_DxWr553A3A{WWP(z4W_D(#D!;t%?+PvjZ&BU|
zdBl)^0rJxb>=ub|Mu~SP|G+Zv#bO)BwrmnZD@`mCWG_BrcoIhBEbMgBKkS6ME4tSD
zI^=a{WGs?9vfRhfO>e=+Gl&d{QCvYGbeh@xXYw#P=Vk*JS#!d>GS{eBm{@vV7F^O~
zihKdTtD&8nJ8`9&)7q`C6-~lT<$J+`5uW1GFIrlOCQS3MJnj_2rzn_qI*QWK4V3+0
zNP97jNQ>T4Ref>Z$h-Gr*mPC#^SV9X3bqnuP77pYhB<vXQ+~SMM6~eYbj$j=XfImQ
z)v})7^5Os6tYZEn3zE|P<1OcVLC>-sRSY@6t6QeI;)wBF5q)>SV7cxJg}8&qW)C;&
zZJMwJNG4`<mJ;_zy0zTwvUB2-@2SQoBnxY8G6@`-yjK%d28F|Xy%{+->UqdqLW1Af
zQyI?(MW-iXc^<K>0V9MD)LV_FNld#369hJQlgtABI0P}lmp;MwVb<n>&uzoYd9ee~
zAzl^Q&E9%)|Hd_+%XO`{u7VI`juxwi06rgzw{9{slQ(jO@lSyixN#Vhdy_hx_({9z
z+qODX<bNGyE{G>;RmR{HKg7T@Gz;$R@7V9b@g)?DO0bugJO)AGMw*R2E1mdh&=XFT
zHz0uBVRazGP;I>EfL!QDpOH=4Aa(|bNZ!^qS^d}y9ITn2E|`g$$g#kS7vJJ>q%ovG
z?9Z~?sm|gOPU3N<9ysvwDNGts9LW>3TGBSkAWN}5^{ZFOfv?`jT3{<RSDevOoVip2
zMDpl<YSmHgbL4kXxBQF^Bn<r>{ubvOv`n6-+`S96vnu=1RCGA-;8Nm+pfCpT>VtoK
z2!*_xzb9Qj%Ll+7C4Vm!&@VqK*|sbaD#N%G&z;F2DAw-s)!x+uZ8)M0uDk=I!@W-@
zE<hoi&MxEmp>pr4GKhe=ULcXN$Q~YZg0b_v)NlBZoqIZ^d=q)iOr9Yy9)b)`f^$zo
zeV{p7{w)5tJ180)C8ivG>Pjg<1LnV1n#^@4`tvLqnzm`KtH8@QoINCW-t7~=60mHA
zQ-LfssL;oHv+n=iD_r>NZvvY7{e*!L4{M#wCHUVv_w$hnX~IeRR4wKP1GBzKv;Iku
zufW2o^UOUf0!lZ-<~5B5Ek3Ip$Iu@{S_R_5#6HiO;-wMSae!TFkt2HzR&C}G3cdD`
zdCoQFO>8$B9dY@l^L4T2WMDq2l{O+EXZ*F4oArHa3%DesL7PI?uNGa9tQ`AHYrKwT
zcUk29d86d6y+(9kjzZ2jaiGy&<hK>a)8nro9@Xzw%BfH!fuq<<1uOSHBFAxqimtiV
zNI!xg=y==v0w^U)y;UQq@ldevc|JQl9k9{_H0wLE=`2>>-LZhoV{+AQR&%a793bs!
z8j7rja})Y(da_v2)67n<();h9j6b9}zSC31GBPTL1CR7I3|@NdxD*lc7_*xL;gi0S
zlP-wF#fwegRm9#KZ%i!X{Fe1jc+rZ*)iqO2hagIf0fl;3)uFt)V(gJj9HOALcj9D6
z1E|D|d$sdGV2#T~dcxB`>w;;4_KWo`xAx_vtq*<j1ySXv7KN*7St>2c(qOWLrKcjO
z`5K=#IHY`;@zr7Wqq5DdhpmUmWWYV{oQ9K)ytCS|%?oHg%p|G&ecj{=Wjs(UPf7Pa
z$t_b1@MJ{qENdbg^Q>7OgROSw9C3t9=22SbpMCo$i2gl7UP~0sTI-grdGzd&E6U}z
zn`;6%Q-z4k8#ea$a{kl!SGFi5BTn`RFjSzQ9^XF4L7y|n9IbNgjct7nBwD6Qiw6vy
zUJ+S}$TqjmijMD#7NQJae;qyahjob1uX_AlC{SOO3oK(ZAf+-02|%3YUYhV0NV6lf
z&tyENPw2XV0>?46Qnv?hboJ_Qc0PMC;6ubgL7v~df;Ff1t5d$TS!mqlCoWUQIP|=`
z(yKj`)LydegL(rBx$`!N8Amh}$vf=~*vY^M?Iv=+Q#WyX4}1f3T=J-`I%twQWO9mx
z_b_4ncNXCs*6ujFygWU?)GjE>B*a7rw67*5`0qDT_dO};u&@X2Cdb#RTBcZ1#IAcW
z_v&Aum!tgdcj)d^^hSB1VKr&<sJ$GRn!+SNp%eEGz+eyueY^uAIepI;R0}C1<aq>S
z^0ckcSTmLW<P@X@Oa6HVOAGY4iu1R@{%#bYjaq@&dPX%<;GYUZv!r1YRlH&2yAd5E
zH+Ku{A-LK5C|$KiAj#{y32s8)nZ$@__7p9T5l`x*(B~qA<E*ZWq_VB@XJ9pC@eXUj
z1EkNoYOmhReu52azwL?uS~w^m+Rh>~)U-_XG<2MtzhFmsm}J+^>#{s+P*UvvX1h8L
z?7ul;2U7b{+ta|x`fMUnz}POe=$0?PRE$13P$@{VFFZKFjJ(`Zn}Ooen;4g!!3JaU
zP5W@WR`ZwswNS_Dk|i-em)TX^i2f<v3CB0#q8*w7OkKv!OjuQw$aE!#B~N9BN0-gJ
z-22@ph_XC9%kE)Wr$?^fulL{4gru>VYgNCT!HzT~{eUH_ih*+<+mUiYt{@v_kg$&#
zE*AJh8qh}d+g#~MtXMCoJE8fuz==-Du>Wh3uruMTJWmH~4SSOQokos>Ia;rR-71+J
zU39%sYjut=3c*SEsZC#~8+PavWI=KrP7t-bdXbjd7t<>PA-e+WMIRugY}y^jk}N+l
zMjBV^z$UNla_b(>4dP(jUq98S*sN+Q{0=@6p!^+{D+SoU3$?dGs$Z}H8)Kb5*-Lo~
z9#p~djQcszZR~Y%Ui%Q{{B$6Lp-qX5hq_DHlprXTKx2D79#Pdj&MEO2X-Oi#$dd4Y
z!azr<ah<=<JyMU?&T{1=X*+C;H-=d;Xoi5988I`2kixLW7U`4CTTJu>)-_s#jbH(U
zXlYIKMgzcr<#Ydo{9r#BPzq>68XE!QlSWRziYl+j&%l8ow+{~ruS7jCh<ePjR%28f
zHN9-TBC}}Jniqb(u8UCAIlmXiZJ^m0LmG-2eVlgY<Zx_I8N%yf<nZ8K^sU8=uobWi
z$P9O9&^hI3-J9y3#-Lubrdgb*GdxuNM<~zjb_A$ruv%CYEa4wJ*Ne9*aMh;eGTz;K
z`zXw{5gnS<ud^UqhD_jg;GAQ?jh%{BC=_04J0(a{Gd%LNCMPKFs>~o<j9T5N@_gJK
z)WRP@BWF9Jz3>tt7kSWP%&B+H&lOZTQk=a$tn*ta*)NzW#i&%v^n$0iG@g(VE&MSp
z3Jch-v)kbaijqVE|HB+7{THfDqx<<lcJ%hFWR~YLE*2syU3(O=InlnMVp^C6^%(dR
zM)#eA(zfH`%r7C7_dBMXipd)_8@mRWzLE<_r20Ga=cTybD^W1RKt)aA_P7UBf9siK
zlgon$^%mz2*a8qCq~PDDzabx~BS8<zd;x@Jx>VP597Vx8x_Y#wZZ5SwA=rwx80m<(
zX_Q83yKy14Y`lRc@P6nQ61GjXu@XHmQRK{?HO8&jknv((#=LK1gPp~gAT{}-co|hN
zJ-~(;<S^EM5xEhXs>9T^KQ0)Po#eTBn{M)ODlp$O%IJFd<+(*$w#}D<!{|1negZ(|
zF{Hwh1TH~XTUJxC#^Ic#-`lF-bkAo#hs_by^>jdwZTcKGqivk%Hj?Ll?Q<UCejoz|
zv|e}3cEcAP3s5yoHG?2`5~`I{66uM<kNePg1+bdR4Pp_~5^PMg=SxpQNlj!}*${n4
zy6UDcfSVgwWi1oRU9y=tAsAk%x~%knJ0*)(TlZT{QX>UU^iH4nHqPEi<&8IF@8Guu
zUgV8xo_rJk9Y<Vy#Upm+9DzM}07*4L!~w=X&)(`Mn(8ZLWj_9fYOV{Fz2;y(h*;>W
z2`?eeyXu?_CNKFl5}<RNw)LJ=)3uestFJkEEhj2|Vbyi~e%UvX$z6wXBq}?k+ZW7V
zz>j+yjq?6^3AO@v4)Lh`2x;Y|eH`p7^P#lnQadj_V3>k4&od_QG3-2@9~UF!cL3=6
zC$va=$|gTCC$-`@{aeHZhrC4v^e=r$S+hxDXFycZJ|v9JpFj`V!bo`N37vP+&qGsI
zMWDLtbxAy`g1)eOGPFJiX{|tIY;Cxtz7+Qcd)<3-7YZt#mkjJvI0AHZf=k4sjih=-
zN``W7L6T*5$KakhwOCRoB??m#Xdz(lB(@OUfIv6qUzjOH_=?h6l8co!F^N_KG@?bl
z*&NUQ4rGcmF;*?##JL{P);4*^<AoN(><g3;xu(>Ey62C-U@-CZBf81<LjwDmxqTZ?
zb#U|p4d${Iv(oq{<j1V>E=Zzx`0!}IfKQsR&=ccu?ZWLA-f;h1S2vJ#7o`B~P|z?<
zzZiX<{i9hLO4vF^=id?Ck86hNq__99>jM5J884>3)b55tlr&NkPjJGZGt9LcS^S6`
zmFsyz{udGR*X5G#{lQU<*SL%W`IDHq^?nBmIwo3E8N@>a${q3ioxh_3>=1ATmq-1#
z6Gnu3nBDF_fCnMo3)Q?K>IR5V(<MOu3TU!%EZ*=Rc(LVHJzZ>A2<qsthB>yJgYZA#
zDY(pu*@H<Pbwu7h&#UB4?7gLL@MheE+w{^K)bp^-*##4`4};d;ZL<VVnFERYq3{0!
z`ySP`;Ue@zRT#q@o`LRRr3F5dP0(W9K~J3H`j53Hf{&7dzY{v<fD*tIq!-l%(dKZ7
zVx*}RCa7y|?#C%##Mz{Fx~EI5%hG5~a_hJi&8k?Mwjh3nXF48NYs=#`J##<ZMGtpi
z+M$g(#44%Wd6Z4)_767yv@{7zh(ELe)1VfgxU%J66x?Vy<`D1+-2J0c*~uP8h5>r}
z`&EOq@=nvFTl08@`2+4k9~rA+km=IyH`@lr(?batvg2|N_+OKWzurA|Q?U9TEd@>|
z{C+slGvVxo)KW5S2E@=19>v&rVXXiwsz=gnmR%~gc2G%8W8boCGgOwkdCm`kO?{Ah
z8WCtO9cRi3TaKI;&h_jk>^(zZVJ=->4WWWJP(dTcl`E&Ii~zZ;Ny<M#41-2h$<bvS
z4x4|LbU*=UB?TjGvSIf-@5)4&4~?(xItLDsjYYsyxxgQbiNm}qTf&b)W1lEIW~?7U
zTiiXQ2PTM+Q^x+&`a;_Yn<q)f!zO?J4`Xi`6i557>jn!J+}$05yF+kyhv4q64Fq?0
zNpN=v?(Q1g-QDew_g(+J*V<=Som2f~W~%zj)b#ZIT=(_-+F&c}XGO@{EfDjP-9JjM
z$3UhRFbcdX{ZXz_%o~#Rlyp&_`>hFB#2p43|It{)VNPJm;G`*Y%2<@OEGcF)OK{76
zt^U-&1XqfbrKlwbE)Q9BE^&2*ZyiKf-U;nEznsOmoH_BH`Fek;;4OK4rsqNiUjrZf
zvzaH;z<XySdlO$;W}5crfM_%EB(HgM?GBTl7vK3hm6QwgxrAT<I{EZiY()ql8uS$A
zjPv@@uA^`-h61lMLTUP%z$nA?pw3!*SRic;DU0I3_Q&VFPR={Qhbud0wWtcm!1obR
zW*suy>`6J3RQM;2;qcKnL$jcY_;W?;D_NGuc|h($G3}dD_HEm}<xfqA+KxV*)2``Q
zRFjuvu4YD0RPw}#j)9ersS;2n(%(~pu7AMZbz=_MSv-mn22k5pYs8ft@?y2}1$Fn{
zjn4kgaNS!ZwbsRQeUVg!P>LN*3rUhft1*dE&nna03)}4y#SjN41x3ye_!c#dmgjJb
z*OpOue}@3)IWePVjgl;NhSK!=3*7ecH@1$o<r?#N?l~llFVEUbMjODE(Ik6<W|O^(
zYKgKo*9-yCP%^?<hm2OvGOFtZMYQQtCyU7wWwO|>b%B?fF(0XVRohvuh;OsYQ)1vk
z;@L7m0p~}UkySaeMsU@YbEk<DSK?%hEX@H9q0k}GP?v@N^3)==fSY!jP}zK%*G!~|
zTyZN?T!Q47V&q|o+bW=$Y?DDBeN>E4bnBm3d2lulH=I1*mq{YhQzHR6&xa$Pft|J|
z7vqn-2J!60!(e>8vS!P6&?=IT!+qe6vfji6+cDk@dYc>egsAh=0c6M!CtZZNHFs;h
z%KRi@bN}mnjGdTVN@m6e7<Yrp$x3@_pH0P0nKIhr@hTe0Iv8-+1BWR@ydF^EyDh+8
zTPrSFnPSsTY#1@v#%zU_WUvMLRPZr;Ibak@t0YuRcg2c!YcKYxEY6uJ@qF#5RbY-E
zuD%$;%@YK?NU}nC7YJ#yDfINtrp>PXh26AOtM4PSURA(Z(xm|T?TJnkewXFt3Wf*9
zjH5QzW<mT<`w+li_Q+i&(Q+}d&fOcvQ_C<%rR*KSD%?=KmV$jG$pwFrz{%r`$I;ji
zIYcf>p2T_zStEmtHN;lUX*iRDD8-sTWC2N9(q!NhlJ7xu->x{rlE0ad=%_nL0=tYV
z3#~;cpg1kwJb?N7T*AM1Iep2MUn{8*R?0>arGjQ_ybFlT%|G!u*80cHLfUgM_ZMR<
zXKC(d^Jz4cd9jAz6Pui9DKL^}zY%&wn&-)}{?g^-r2#w`+r8@|qh&d##0I|ju=Eq?
zB;MJVAbDsIsJw13jic{J=)q8iSPziDjnlp^Fi<ed@J(P-?)*x8jylK?X%2qm%@%vV
zvjRl3ls~5GGnoj(VO-8Wt>+uiX@EfxDNcWuJkjA_kzD2L758LVA`xkr7Gv4(sJMLq
zxpjCNgxz<r2@s2XAtukj%;hCl0=r;_@$w-$w{o<0QLgDC%R-wWma*RAH=%VrmBVb+
zK<jEBXFR5Oryhy)Xv<2Qscu;?fA`@rqld@H!vZE)taZO-GBzWBpaOpswdqfq^^>US
zED;5Vid$|&X?x4zj+wRjEL7fezlDPx8XvT2e<~_as!fWa7f-LpN++lmt5ZYo_P;n3
zCzv7#jc_$MFYe1~`pfX_U+5P*YrTct=9e$se-bD%`beBcwCepi6t@H(tGqh-4U`$s
z8vr$4W@#@%XYuC@OcLiZ<kCY@rUt{d)iUL@F+%lcJ^d;d`oJ!17R@W{Q+kBVtXWaL
z-anOF^>xBuT>>|4mLnyL(QqN#@gS;HGDu-q7Hd4!3G`8-+Xp8a%?%!DJacE(+>C3{
zoCI8)JLiOd&z-M0pHHcrpP|>|KTU5G%mC->;y>0pJ?-Tn6fBb}mXIk!7|sXs5!f4t
zvhIZumML#8)Zw=pBP@9iE4NOQ({(cN7J%*4iINKR)hn@kr&P5ssqc$)Gw-d61fgH$
zN#rW1YciKFA9Swke3p2&#hmrX-RizJIj!|2mC?9rD<Y@i$6|&z8-PIPTQAz5JpiN$
zk}q7mEVh|+J||p{Dc2Ax)M(u;aq<-^wG;}wVe$p(U+Z5Q3}>yjC(bJ!$p8qatCqJG
zA0eG6RxzU?Z$HIqonlFIunG%RWst@a9y@4l6+0C}z|BsDQ%62XXD-Ixt>wA$C2afd
zv*=2(SO^JWREkpK^yjURvJkdjB!I^*beqFJ@C${|VZh5Mrdne(HdBAhXdqlNg%ptx
zI)oWMg579N#3i^3Iw7%d$R}dZ5U9T|Hp=1QzJcSKfrUU#(+D6UT@VRZ`GMhZgz|Vu
z_S9Qt$Xxa`lS3z7p`E9N?&qpn7eCs0uO8}Bz?vMSeflnICdkqh(`JSXD3aqcP&xLc
z$cLm>1P2RA%VIc@ng-RF{Nl6SoKxAZbMDjN7mTXUj)x`IhF)YAB^Z1+n~<=Lqr%?Y
zJV_10ZC3&O?I;o^6W9>X1RB9MWqTt&-pfqDxAYt`pI32b_>gy<**$3_FlGrTLn|0(
zn3xJHORyg)p(qR0b{heHm-=1yI$0-^;p|oSt&?S$^>R(ZzR_j&1!&s;tPL~Ii1Ni_
zxBhrg<RTlYZ|y2KSZQm9pFNnY<XfS!&eb@gEUM~;g_W6ezIam~=WY#NAh_yj(u{Z^
z6<(C|Pf40@POJ1c4vd5E5)cFyQusiGgG)EPK%>SPuh@15051v;@N@mevy@;k58t}H
zR<?0r$_Ct%wm%bax^H;RB1k`$M4=3acf4;_rk5gA>c1}f42@Z-)Q(RKfQxIbbb7Qb
zS#fV{8wOe)GWy_q^HC6e0Jy4OE4`MymrlE3@x5s2R|i|5WOYc_Z~F-1jc?v9eil&9
zmvwkXKVK^IfHF;DY9=zn6ymMIND8GN;H?emDRw>2livPlN%<c3U5A*(^DW1}WHFG&
zwxq^lKAm_mIv-Ynhm3HPeVN*{)M+{U<;GkszmLD_a(2$WYB_xVMvr_%RQ#+}=sMq)
z*!&7!Z>PNMdM=5Or}ylDf&VQrp*GTdT&eA*bF*)lA8>)H_T5`r;!VbM(`xrM1Fgke
zqBF&!i=$*I2?w!@ia*ee2(*Dd`Q`JIFR6q?bdIM6j-^)|-y-FWI$CrvC%4ZSPNvFu
zXkvl5a(SKM&(QN4tsdWRE`b9=a=8|%=V)Gcc&tVvH+(&Owr<*t7DwVlBG&UC-w3_k
zx<jq_CIE8v-|MiEqWOQyGC)|me{Oz|b+Z$X0AELtg)hK4WXBibb>-*2l*L(aBcJiI
z=mxP?h#x*eoSIwu?<od`qnhwFttX*Z#OXB#hS#LgZ?F(+OiU)EaZrHHob`(|KOsW#
zbiyYn#r*0#NraZOkb0rXUo|HT$f0e=kZD)negg>i-m?iYAwKb%-rcJ$G_&4~){<?W
z2_zZh9a3$wu(4{I9U;8cbIBeDcNLk+6T6h|FpJYx)CPk|qQ%(AV@dvnxP@N%q>M1d
zQ7T!G)4|aytweKOkN#$&9MX@%cjPULIP#hUWpx7GGxVECA+Ghb84a2%p}(V-!hT_)
zRss<L`HHp5H10cVKS187tL2+q|Kcp6OOc(z!I)k0+hzoIb}grGWvG|Xs&%N)3#;Jz
zp!#9*DONA5pUpu(jxeFL6v9n~*fh%K!>zGU?yny-uk!z(c@v5MljiY4{)grP|Khx!
zi~k=uFHHIWj`Ij-I|J(qcY~#doO2@mD3ZT{nAeCrcV)C1^s$Y`gHG?}2;pn0I{)s$
zqIt37!q9oJ6gUv4mFD9OW~2m2<VAl*z?VlX2=rPNlN2_`>0S*}!wwz_306Qu?4TN=
z8<!rm>n5;{J&1yW$$1_=G>Nx625GF*Xl{UD!?9qSvEVn>O*^upl*lk=sfhehy-}yd
z9$NivoqXi{CQJ3SfM^<LE&)IwM_unKjDoXd4)-g9DF?k7#%Ak7GBp4bO5^akuAKJB
zEpZ&PUL^9!=iTED3b(&@$@c0$w@9T^4@G}WuS5y*GoXrm<xs1V>#b&Fk;&35<4;$T
z$+o|+>KvL%N!5v#Cat<4Whd!@m(S}w@HZXLC+&g`PeBBwpq*|c<k0WJ8_6l@;IoE_
zWPU5L>|PD`C{{!^EUo}Hy~u8iN6wHm2FUTfW3%mNdau;(L^UUjW5ty0&hnh!4*h&b
zpPc9nPnDMCQ*dMXRXi^wD@ucG;H(T8YA*^I^1Gb8%csl8n2_jnOQ!I6^zfFyTcv<$
zuHHRes!*yT{8Zo4y@<VVE*FZutX){Pg3*Z_dVdn0rn_#7EVcqPWqm%jfA6Sf^5_zu
z39h76SvfK!(5YY{B6L@mD^EvbBR{?n;jaClbzV-Hh;V#r-1Q;4oNdB4nyfnKh#=**
z;w4`7Zc_Bjg6<6h?+`1`Lx}SU!)dKCD6OoCjD5|~6jySNF!KI}fD%jNW<N<eaE1Rt
z7T=}*C5zku(Dy{jxzB^nuHIlZL3`(Zl;X33a6{Z-z>+r>w?cxqlP->lC@2J>DN`Gq
z3+_n36vO9LMy$gGMYZv}%uxyklhgwU-Gc-pm<-5DRfw2c-ft4xb(4;Wdh!!aTyo%i
zG}--Qp{#Z5B0Wr3ean~bR<hgfsI1-owK^U9?P>@p9RW=n{-~ZjT+Vmy*Tat;nO<Re
zNIn;;Mw<H#oyz+NkXRWGShM%<Ac$lRVo#gK6S1wY!F`%#6jLlavWSfgc21ywojdI7
zC`9<fMN^-~dREn$iGU7?zTi#)30b3(5$yog*^@@@v^Wmw$Yw6IEbwoL9!ss5{1#jY
zAc!5+TAVNA=1&*tsZbMQq}po?1f?pOFWYXe;2n?mlD^;5!I!~;+mHsLXHeG2n$}l=
zXCI@<i}}4BX0XWm$>Sc)eu3(XlP}uFh26!Uqc}fMM~5)*H(u4ej9;e(Js0$B@>&vy
zIaT-(EHzKL0dA&A;B(A)Q6-%DVrO^`7^Wd%XE@|`cjGc}Kr?0v{pMh8#kXbFQKIu4
zkr+|8mckA9V!LEM%^WE4c$v{}G(f)%3`L1yJfq!*l#&>~mh5)bs>G9y(YkyvCs&P7
zyW-V75vmJ5GUhy7SI$>w3DN>*8)9IM+C`rd^GlLB2x_TV86*T2B4GAH%445^Pjw#t
zs_F{`WfnIGF0M7NX`O(Ky5lRDCHm*e!vjo$vLiKZsvP#3{3m8L7$*sJQZ1<^ygKKC
zdS@6^XMY|@5g8Klo}hyy3|(W!W;nLA5hD7IFNS?QSSaL!{6>85RV}-fiI!@LEafqs
zR}|JVW;)OdK2@(gi$3ojrMor2B>I-Nt#VZ7_BLny@9RPeJE%?~C>>*ejbdwEAttNH
z9x#}qlu=3y68kO5xiB|$8KrLJ!V?|~#o<Odl)uFDt1ydd5>I{8{$1lv5ASkzZ2IV>
zC-kR)*S1HNZ9_<rB24C07eS>Z6jY?FGG9b`{Wri<fi+{qopX(!1QI|5M1}BQCFb6H
znLwhRVm`pH(=p$MY*;odOKiu!HfzGY&AOGra->B)`6c<+p+;Uto;Tfc`Nr+3Y(M_P
zaV0yxv2g|0vzS6Un6o+^I-N_W$%VOfTu0M<aL-7xR8GW$g>?}tv0R7Bi1Fl(`=}q%
z#xu4pMFT<|Bn?*);DBXGPCs%!oa<T!7_{qb<9-&|h(pn!MAGiF>P`l9;VBm&h;zX+
zO5C{2&wX2BpE`jsJP^bgdy3xSs7-HNDn81>rV@xZur``4S*lve#<7I}veppAk|Frj
zYI|ytbEA0=Dvs$85}VZ6mp>+9@%53}#5?4dy*B>RxpnLGDR8PZ%BUTOCzVrxmk($A
z_JYrd>r?}F2kJS~rR$uLeoy@pT6FC0*&NT??B~0-7jpg95HPao-TfKo8$JtH6+6>i
z+Co<9Q%mz0Ouc3sx{ODr15RVM%I5edSHEwuF@q|&icKE<v#+{N>1)gUEGN8jdi65e
zCWcAqIUFnzmFy^sFCMld1u|#KVpfbh>^fN+;qij&@)o=OiyEORc){b+3_`Ndjqhx`
zJ!;~0+cDH!;^xH5V96RbAhaNf$w4+ClqAr{eoot7?#goB0(5L~XCDl~Bk6-7pwLAc
z$2T+rY%v_H$x}8Uya1eF|JhsCh^jAda+*w*EGt@c%&UxwT^GNXn+0~|(O{}LkAh76
zg9^M6jJT>9yS8SV+{;fjtECcR%a6r^H>32et<ltu-0^^szx6ew`ap!%z}jt3`RwWX
zkIx`IhXVf1efwE{L!9rse{6!3=RO#pwfRUF-hNFB#+T+aXuw47i2tV9lltiI+V5u(
zb8b;B#wLT>9A>fM(ah(ROo6iB>4}oHSm2iLG++Ll<uj=N5S7x+eck;v^TE!;7oJpF
zlUy!@)G*JW&Q<bfVeW+1qd-Fp<3;^!4&3DW|Ir9QFZZ=yD)F1kefBFnqbodIe;3_O
zCVytn^mzwFv2Q-Fg;V*8qvH`I^Q~-<dqyFw<|`glSlX;pciKe|SZWWj_d{pocZB`c
zvvIi$0Yy@<QSSAi!Tx5XaNKmW`58%Ha{|+&oFg-YiXsCxA9NZ`SQtZmFn8i4X(FR^
zppX3Ri0i@R*f`40RIeO?DT@vZ0ol~2?@sjSK7S2(B~cOA3W6{a#+s=A+QCzb2bSGQ
zZZb_SG+Ep23Uz`2x6?4|)!2UKR2dJ!{NV6u@1ScG{#RvW`7m*^kdS$mxGjc(&&I31
z0=GmoMSJSvncz>8(=zX2JWwyANTtY@n@vNsUohRmlCi{Oc}UdkQmq-oNs&@oX{=ZM
z97~WuB3vwKv$i`GbxpzU#eH#EbQp<YdFXbHcKB|XU7r#x`x5ER|Ih}c@GfgZK)Hr?
z<IjpAeT3tDeuz4mA{>+aZG|7P*#9Yo^Y9_uiI6LEmN*izr$*U6dy$^0r9DKaN?ug^
zf-av>%ubY<T*Sshj*9`Woe7mkDd}ui$6~vy#lrvFB!H~`YZAaeJg9FjT^N3>V{{o3
zSd&oAKsZd=LUeSRw)pJ(3V50u)v;WL5<!>q*@c9Q!V6Cc!FB<%NNbTMMrowWnf~}<
zK_Puxhj^TaU|ByseT@<fjEtgKkG2;4BPXOO+IAoh4ti`HV|O5!OhCI|KR+$uuSHOD
zAJ(Jre_y}k@28{8b|Xm~KO;jP&bO}wLMr4RTN$QZE1#b~oj-l7UpyiYN}C$Uo2hRa
zyg;UAO=1f^X^dON4@%`s7D+nRameO)GNLz=RjJ(jJ*fl#Xase+eSb9q{ijkMY$DhW
z1EoKc;95y95p-A0uW*K$t><G`yN;z`0q_P_buD$zvWx*=w3&FKDZs?8tB_1KS9sBi
zzesozkb)lepvW3syR&sQzv`MPk93>U<aECXYlQr`QE>hz#J8dOR}%k7!tr$_u(4@i
zNdux1iTS9)l88aJAiBZ3QGYQkKD5A1gMS>Tfh>%I-MPf93M_(0c)}kkKBtEP1Ed=Y
zA($EF)RCoDe23x1Zxp17!?OSbvO(YM4c~z<h%l%nY?JJcTI|x5Itcw#y#0ufI-JHv
z=wiRv{0l435Q&F=A1HNGAKAH&N8lPRK)xYqbBcCe{ta;<kViUoStXoyldT+~HR@ac
zsfeSO>%E!E;dR%ir_pf<v_AuN8%|GoXb|JS;7v&;saD!A&Y`Crf2DX?z;AoY?XWN9
zdt`R7=tp?i4j9Esaf~O6F`~%dCPwC09K&h6S69iEE#$9z#?&-Dy_Ogns38Y)0Y}i1
zTRICGYhZkJt-Y;(HtZZ{`tM^YW4?DY$kR4mMms`0fAhi;p4lnxgCWfvdyn{G3ML!X
zKYF6M8Qa+%FWnz7=lP#_eX8|rzl4fy`m%e}9ayhhlF+N};G9v}I|QP&1A>~tZJznj
z$y*HkCpxQJe|iLz>LDw4nk^3gbh3SBh6P}hhNb;25>4nTHh+drn~>e}V&d%in)Lo#
z%2UHJRc(-L9xqjV-ZriNy*OugOZ<L}0Ijh1*ja3)86}55xiTRa|4(ujk^66@JTW2#
znKapJ5mD>}K@Kn@nAE(6ds)Qcy#hFN#5~xRyfO{`)L*{@w{}_Z_}2XFVBEol5&|a?
zjQh8iXYXt5qM>Pck_?_(rY{L(Q5gt{+9MMz06J}>N=))_u5+9nnDLjz8`Q_pxnXu^
zgJ^B4tQNC0POfJeUmFPyM11<p#3zKCj0yiH=PG5$s&icq+*Ex8G3Y&jkCXu6gjDb#
zHnk{9xKjcvE0!TvB?mWRs(q~VXxHO8tYz`rWp=c~Au<ALPQa(~(6Qp1m~uPW^i3N}
zaMb5!o}XfMe3}$42V^4k8jXqxpraB<j`*-g@mj{_TWT>m%FNUq=Qf)fO{5JeiJ*=c
z0?D)>TyJC|wFxmRez{@530)Y}gNuM;dm((>u>S(Q)v(J7*I+&@zdBnrFZ1n+i}*lC
z&51x7F$dN+v8d1!!spYoxv46i#%36CROrw>tf8aDH`%9~wYOjT=k8A>)>}-=5bA^b
zlML-2SK@IST{>Hn5?*Xy#?(Z<*w|Ou&)&kti8!XF`I2M#%qS5+OeXyOdI4PY>z#b7
zp#PJVqKOtx6A#QQ`LiP9jhZw%DaqAmtCBEo8tCF?>z^1eI{5m3Q-8S%yY&6sN3~!^
z!uZgXU|$T0xL)+|PrR4{&2J`D6{$Ex-$&|@Hx))at~b#Rj1HxXF3CeE>T3``_gVd^
z?yj@KsmMYLq&ty3EDCJRPO=!b%+Do=4-UYi`7=qtKr2|!mMvYW$UpZzl&JETI-)XL
zNgKL5BfVMuc}6;0!}Y(A9zWQI!$=0d?gqJW%_!34T!j8>(ZQQ=b=hC`dC;{J7RH16
zZa@kC#Ay%$A^db79rIkLgk<-I$@vbGF~PHMPL&Dxi-}NNb$kz8TE>Du<h6Nh6N{!S
zii!&DP~fkM=SSuKJyi1`kJk~E+!xME?^<a|9L>Pg?*wZVA~Ul19OvDlca~Y_wB&`)
zm9BrZZj|;5G&06Ma<aweC(8cWgdk&U<_tbnhH{>pYDpNe78Aya+!SXYremqs^}!WS
zBefquP!M1hU=$(8o#nRY#XB3x;FPy*x`o?nP+1n#h`95^WI^aI+$tAi(LoG{Ne(o+
z0rzE3mfUd>RpO5qk6{_tH^;yb{I-y0;<yhU5G2;FA`cm`R?}uOXq@bd#M3}>x$shI
zRd3O)A}#E=?XaCu-e-7$G|qYVn{_)?^PQOXNPSiYO?@UBBh8xd#IdX+@C|Qwwz@@R
z)Fy3|Hr)eq&1$<te@38gEV$}CIm8VF4;Tkg`q;CRDgCiyeEEyZA8Cby?>TDor*(*z
z%2@bRs47ZC1k5jfZzo<%n92-AsnUK%qka=vV|D16VYu9>BN$VsVG>TPE9v#9l%(?F
zR6Wm0qESC%;KzS*pdmNP43y=+l+vn|s#N5}?1EBG8XWE=7FI1lne}|(Se0r4od(nv
zSjJ3|DUMk1oeV@uuAb`ji!I`!Q;WDf*Gp@Q1?z}Mw7is9BYhyCsp;qkX^CUC^Yja@
z$=2G!%4W3iz?*a=HKLW}=E;wj1ocH~%wBvw6yJZ%%Os6_|JX48bi#STG_9P2igp#l
zq!97sSd%Zbdqp*J;(sO?u%2eM$P4_d73`R<{G%1bTGVG3+Z$dNPT+ALEpZF<dfwvN
zQazm<Hf5_3WZHh#Y3^eeA<o{vrbC+NRdL=U{IOuN4E^eEfh`OtH$HVGTVCd=kI?ty
zJR$lrOhzW_D=0Q>WK}-jF=+@<*YuTia1M$^F->;WKIg9wy$i&82*{0u^Uo5Fip+Vs
z6Y93YP*K7L_td}sd@)J5`iknb36H!6U*uA&XwLSdY8wdDA814M$Ad~n8IYBG8WAp}
zsGq{9zAbg5TkS~QwBe5VC@V6|eYjyZE)=IJC*iz<yYT!XDkX&SaXA{ndb3F`J3I~=
zo-$j`c$<$_S9AUd1A5piPEW5m=94&X^aY=YHKgnRrcm}pSiyDXfiHKEjni(^dVS0E
z>@5$RRl=Rj*&5Be-MqcYN)B4%Z7O-qv$94(mKj`)gfZ`lL+_S>wL)s67>Ia%B7^b;
zhQne<0eew&9ubm*Ep$kS_`+xZ#MIYZJ6&&TLbzo%Hh}I*?J=}H+P3ZNTuUZ-e0X!8
zclJ+@_F0#`hFw^Briq}<Z^}PhAzG?uv!CoE?Ndv<HZ#-&u_WKf;9mN8U!eFWh+a5Z
zLY5V;H=mkk+I*N(ycKbvyStQRAyRbiISrRQ@Zm0KO0mKXx1cm0jL-IO2LllG&qm_c
zQ<>%7Mu1lL8+T^R3hG;np`rY%6OIZim8%h5GBG)K`A%5y{hND5zZd#!F#>fI4%={o
z$P2qW>C^ReI3~Ab+|A7%|83=w(?f6Qu_cY?1a9z$L2Og655gO*5@6Bg{)j;H?({{U
zDebRbD{DN`pHC+o=x+EI>!;cZMKoS1DjNYZBby%&O(8%N3wqxen$|p&>~j!lEU~Xs
zJPOh_j_{ma9^U^itY7LY>+py&DESZ8{~uJ}FZ4a7xNhPv*$*KavHuaB!DGjF1X<vv
zG5}3hwf25-3`W6P#=lJ8lv-}_=kYP+lX*Cfx~*Dzbv{tHrM?N>F`9}YWE@Eb!SZWh
zl72F@kVY+ov|=Hl@b-~RH;e!Kyi;%}#eef^t*QSHuhs<x#m23p04n*hrME_z{woI_
z$*qSg@9ZIQj)qq;7P4<9{cs6_ly751k7OdP-tm(2d>4&}Sf(f7wA|Ej?o>(D@!TO^
z6r6Wvo4KZwEPEj?4=YAmelGQYLj8{8j$;g(_XF0P%DttugObVRD~N9>DI&H64!QrY
zOnm(Ck%|9_#An|j{#zi9m$mOsBI7r-T&o36x3>OA9DZMj-;eQ=3fjNi)h8M4ZsMj;
z#8SNG6aO&cDKbX6z1vZ+|21pP0t9<LYP`vX%3l<&pjB9JYY0AALG8tNuygO8*~HG7
zel>PW7WNctd=}is9$%rQ@=c0WuLj#7!Co{%{uc!Y{>9*X^Xk8o#<+-8!ydqCG{eBv
z<q*jc3!nl+ou&RF@DB(s@E-{7qP$HRXi@G3aO6(%i7}5HZkWRFvA%*WGqAij5+VvC
z_UQCoAbakvKX9W7hZBDN{cmJ$U$>M<6}#yUgHUz|ms)$hfobEE?o0$$mT1<rLJ6y2
zA5Z@Ldg8!070uN}TH0;Y#B2r7xSzGbF)}hW7wSOT7pw9!Opos@NtbJ8XoK0r>|yg)
zh_)#pOa)lmb~3Wb)SQ|Q?(!*;Gm&z+e7?4G{ooc#E4dIxP@UP#sB5}NcAXS1PHAQQ
zK;pl76SOo5+;kR_RaAk-jq4wWff3x05y`zv){5Rj6ACk1+J*XvA)i77$VGfJ3-$cp
zX;iIxw5gGF)!U8T3E^*?Y)%C@1UQ(eaM7VYe;obI-&+e57*<MxU5a)0D}zP{$~ctj
zqzq8S{|)VSqzpJxI`i~3A5eVNh~ig~ud(Cpam3p0fim#dot1eK_{W`v1c#@iXRKu9
zfNMKc?)^_}w+7`MM>|pSi;9~5i0!y|`$ppbOP%F?i~or2E5iufUlKrbVY4`s<Lb`4
z2q|yZ{bM4Mo<5{!4`Peja^D$PDnkwBLJR>zyuyCaOW6)L9A<7%X#P?lvBV&f8~PrJ
zM;6-kuPQABk{e|XsZ|Nr7OyMvPA*p<5eXWupU*vicG=%;UGFL7DeqqSdBKpnZgWX;
zlT1PG{)Zx;NfX@sUX&Jh8G|`v^ju!;CC^`F+vTUw@BeClZ2EjEuhpDf1KV%T+E;Ef
zV#n_@ZKMn`O94gk!jQSmZilg<86wn}ry>BOQ6a@==X|QtGQH)qOzZTpmDv_rein}a
z3VledYT(zuc-o(M;}qI!!#yL6&EPK14C6IGxwPn`=$Eonr&0@?K^9w(@yW8oEgd}9
zxtT>jQsidvka3<;itxw!>!QbG&g4Iv7P)PwCjSC7|A<J{&kl!M4TnXJwe{d4URgnN
z%3aHVHpJf$;L4d^-i8e(;ln^92_FqSf~+J>^PJ_qd$%Clsy^;snUg3aBHTKi`$Aq~
zpWy!yTj<TDzhbsYQ-2#?YuuJsD6*QMWYT_fwDmMcdaJfOxunF+Ke?dUU>JNiJ=bhR
zVXpyh^ol2ka8o8(w#-*7ZNUd^WUW@fIEFH0>BZ0k%gu?9BpOdwQ{UTF$o!DReXB{8
zx5!bV>A(;Exav6{^ZAPxt`qET3bqCWcd&ZjLTjJ+z_|Kta*XUpZJC`3)U(-VEeckd
z^{Lzc)>--8Ua&~Xo<`)X$g9ljln`MzML7pRgkbkyUx1WcK$6<>#-mzxOBks@C~VYu
z!Tb6s-ewocnWvyn-M)+7lV@4<4lc2=ncEJHzPIK+`bgPsnpmBek~ptOtVrhBL=vj?
zi3-avgxSRlZ{0{X5E!MykX}kl=j_oLMet@wRD3YN7VmTUueb)f&L`YpHZI^q*W5q^
z=Uo)%nx9Rj(1uLUBYl|Io_Aao!U#Ok^$W!zLN#4I*ZdZm#i}-v&4Y{u+~E5yq=BtK
zFfseP1WVq6H9&}w$=$qFKP#4;xt{mc+u3H#kY$Aksl!#$f8^jg9Pxhhlht4={Jgn8
zrEJF4t}*^u19FT}CT(U~3?W`!!U90MKM76|3oeL)`<BDA={!`#8W{f>U|!;pI^}#S
z#TXG`jKi|dP!d|)#;X|wqvS|8&-7^|TQTx(OsKy5WlbsleK5O(m$dyW-mD3nwIN6C
zMKIxMXM#ulN?FqOB%DI!*zjo2<^n^o6Y|#fv)C_rPT$X=X&WS6&emyUJ+;8(vemVA
z+aUid2vthB?3>bJU4~7w(Q}9UKQemt1Bk+PfN^>c_o8AVkv$qNtmMbKSA#g2JHXTv
zQE}c=7!PfFfbO0iS+4Cs9|rE(K+TS}3P1LYPcKy|111V?VfP;wJ%h?(FA+`4JHa~Q
z1aj`<kI|6yn1stMOB?8-7e1^l8>zJU=}sEF5;4FQM>pHn)@&fgBAYcb^51>*2>czf
zh_eIENba|EFO<d40f=no4K(%reOtu69@`t^F*!fZ5#iizqL9wUp>6C?IsBw#QznZy
z!Wg?NMG2yAR8Q<5g%IW3hEL&&g13WwKUEG%+yl8L$ZlLz!!$5=#$Z)@GI1a`+n+M;
zrk%-axk93=PzI_^T$9R|m~m!Y4!VcQ?OkmPXy@|T+?f4DWl>A(EtF+$4Y>IWL_l{U
zQ`VJ(9Gi9OP&In$s{Gy<=e{hHbMth+6+D4!GK^_X+8*@aTVoKR#M6(O|E>uwVTMf>
zYX-C#D6Y+sMsah9pV(vHDc*6;hTuxo4DqQIN}&4GQ-g3!pvNeJate2IDA+tUm+0=A
z(3g=j8!pn_oM~={61&dhX32y|T$aTNf#1E$Z?c9M#M-47ue)>5umfzmel&r7nTfE_
zYDN3d3KT-T2yU?;4xjV64}&l)(bs;zzX1}pq^y=JL#(b8?h2pIXXFN8ZJIQTr6{J<
z>%=Jy(shp7JH!mf7^EpGKDb;s&GLfEbki@UPM>!ix+01ouf(NW*{@TRI9ePo`4A)6
z=gncEuGG$8hnUf3ovQVeaT-T?TOB4fNxb5!T6wc;`XUAHGV6@kNZoKasR7D!q(Ibx
zlLfZi0Z9D%xe43xWbVQev$4~&z6s<0oKO(E_so2hpcRaPk_6$`x&`Gz?wGInYdoU2
zo~9RvdoI3V1L&Cs&v%$(Tmgk!{n<3D5SzvFtGrNh`6BIcNF5s%d0+DEkU9qJK^7|s
z{G-YQXO-xVF1ul4&3rL7;NU*`2Vf3P%Uys!gGm28v?Efn?a!BDLv?I!u;$|)1g(hK
zOU!ij->X3_QK{L7M|#W|iRIlSsweK0z&fh+cWj@zAby}s`*kbIE%`UD`Q=x<rb<GP
zpKM?~q_VU_dIqWxKeVvKlLH`oTYN+%8&brEtqjJ<HBjX6)~TXd%GQ?=aDj;lS=O|D
z&YLWb-y21~X)X`?@uGKtbymyPGxY`91==>v?>bYKbc+(=WjEAL`f$a{#u1nACJIgn
zPdVz-so2$IusPn(k0mq6kRKiGHKFd8JpAU<8GBaQ2r8+yLw@=dQ$db-i1s)@oewIt
zyPk9F5X`=mTn_E$8pG#G3t*U=4sVJ!Qo2Im<di>lp$*H6m}_WCR9dDyF2n2T4nG?0
z#HQP=%*ekSNo%p^Da0&e%JdQcFPJv84UjJ2?So`W-N@TkIZ85SBC4r{4Y_ph^S#fF
zz#P**U~DHC-?LW-^x)W*wJ_pHmZ6-XO1Up+iaj|WzJ(*WsQWzv`A#L}Hv&G&jPbdG
zbFR^0OPS%Y#bV*6Up?68#mFY0Ba-}-!5W?e(p0IW_^=QaJeIMiA-4N1gFbH49=Kbg
zm&3`)?v2cWYX)Ivw<3el@DScZ%+N%_JZz(7lJ%)}FkRzDklcng$U0n9YJ#ZSU<>QZ
zKj+x=_QvuASU81%v+$?e<AN8(E3N^OFF%jfJX(CLgH#d2%hLuC>Vg%RFMkza8fAy}
z7clD6VQt+ab?QXV8})0aB7{;D2Vr2-lj7+P?C2&OCmNtIAcR5x{oRlhZI@|L@Xyqp
zEHI$)c!=~X?nuY=cQU$4U-@S5fV#T}mHS(xt`xhDM_mD39;3~foU}c4s|C20b678k
zG#M3q-1`SjQ_9g+z!(WlG`^ut)Mr(`Lsk5ZT_|1jSRJP({V-Sj{anq(^Ih*g{8oOz
z<Wh&ghl3L4yi%o1wKs-E61$Dx)w|IEc}b1XSa*u4wlc0C4h_U_AzS&i@D`EwJ)?18
zVX;U>SrXV9&tUGKWjSNvvCqt|PtEJZ8rXneei-`Nm=W`}598;ayX_YH1Q@<o++^R_
z+}zlCorcuMR>gvOu?`^40~ya~rp0`{A+KS1*m%(@cki^bO|JDw!r(dkz)|jK202L<
zV@l3ATV&0X4&Eb^fAa88Go+mfytv4BDGUKJ`)0s&)4)fEMtG6E`4i4Sg1fOTHVJH&
zdWrYV?T}SV)#n57GiT86S$n(-ib!I^8Jzd1bN7@P>qArf_Q?(c@}#<6nJ8JOm(K=G
z$~Iz&<OC4<*OI7uelIx{&hb0JH8-Q?!AlXS^CpNA%*K8FYm-5hAkejy8Y}g_E0|G}
zIS0Vxyj4>r792@Dh6^F!6RPy1xs*r-;R0N*5UzIS@~A4H-;ez0M&aJX*CWnTmbjj3
z-F6stgTUBHa(ABd4jPlF5xsk@4qM?a0$L@$gQ|)BF(|bqQ%Af?2sWP~5saNKfzScD
z7Gm}K1_nFx;>Tr>D8)K~@~&SK<nnT6UM`SjnJ7B0K+2sg!uoA^L&{zkx-Lfokr(l7
zg?<HEmta75y>v4$?S8ulmzp42AkQ+tM&4ivgTWPl+SU6(AKcJH&h*-m?qkDheFl_|
z*`n^MT2ql~AF<d)6*IdH60{w`giBG6G^C<#TB8#D$(KQfq4u;-=$~-pW=ChrC-#7j
zDHOkCP)-N$dk@aD&VY6{Pjy$H4c5Q3w#;zio{k+kUs2eLbIb!4Oun_L>%C@(_wCYV
zp;2gFuUS8|0jBdTOgKg|@w3u{XZdoSEx)DP8xPnVn|<*CU##ayroLDD;NI`{f4Oa@
z@gVOG`;l9waS)hragr={qyf;hSHX<(1t7S|e4Z%ieCj8IdD*#?CP3ylAB-9Y>*(f7
zrh}C`NPXI3&}ooS=>n&k^AYGz(9Y9vr8&={|Iy1arUUN!%QszcBUg`@J+by{2wLp1
zv*Qz4qm&{oTeu3jgnkR+nlnnTN4<uLR2QSWGFh8&Ib@laFnfp1<S(EV*)<FmZiHOM
zt(pC!#$_4%1WD7jTxi{}0v9u`v}ukk?I$WVAHNYVdBqhCki4ie^am2E{cyc8Tf3S_
zO$q1Rj`(;`-v&X`hG@($fzg<SrYMZ9hw+^CDn-U%7Ghtv@B2aiuCo1@8b-_^qDSEI
zV(d<^tGpyRYyCl-MDoB?2kK%seM_kj&l5Kb?qRJi;)K{7oz@?;(dp__$OwG%Jz||t
z8Z!$JqJ_RS1(!MkfA@|>v!w8+9|%RQiP~zErAs%aXMvWm$vl6|K6>TJ+ddyJmpkC<
z*oBokftB*~(H}a8tu9B^rCX1xT`l{wq(WYb1oPUFz^MuI#~QG5C^?O60rUeutBT59
zMHa8kbPAP!BQU)=(JfLo%URnlz9p2ae&y*4=_d%1Gh$cAUAnah`ec(rBK4|ATut^%
zm3D7&;tNv<ZoEj|Z_L7PNdpTk)uN;qu`A}LlgbSQ<4o~-)E&yKSHg%s<{bPbavVSU
zOA7R_WVXN5Sj_+rdoSZz0k5HN#`5lNe+GkPlrj7Vl&Ei*Wf~UFq2BYejb|CBmGOSd
zJhns?9HIQF7|iY{C3yy^hU?vuwPX|QKn;Z|)B73+07yK^g$0a49SMV{WZB2Yy-8h<
zLU+c?HctU_&ydUKeuE^Btxf0`W4B}LqkPxBZ39w86VF^f@Fgj~0^N}3-AYE#dmm#+
zO_fP-)-j{*3wRVm8yQq{=B<XY!hRhGKvP*}8cSc{yb5e9GY_&|$0nymf#3p(P_5UC
z=Siu5b~DbM++_m|^5sipPoohcmSz--1>SNHOefPZ=ZjGu@uQ8jM9ITDQ!}Ta+j>;3
zT_&ctCkHy3;%sm?v8(8l$z9zN1hZwf+h_PjUSz)&OtK{`2Eccrgbdyuavw^qi4|(o
zFm)Gu5SEBiOze6wwP+O0QSSnp#foGsboddnxU91D%4Lz7{ZLx_fad#cm!scXliRlf
zbX=7fAy*h4lL13j+0-_f()lk8+zh3KAsmL{*+_vXwqb0(c-#x`T+XZ3ozIjPZ!TQx
zBp_=8zW480%>pyLS6(D|^aqGb5jalOuI`?6&XC@=`{Ib_nK4=Pf$6~VA4=74QRKOh
zf&9D~Rie8hEt>Sn55MBbPiy1r#*Fg8T9u-X-Ka!unRxLD_b~*e{F1i(o)Mf>@gb-R
zp-S(-9@9oAuo5z}Lxyr=B@c&p74}}Ioh@BVXRvLbkeLS4-`a1s0zxXd=9B~JBs^fh
zbmADVnaN?V+<j;cZ#jT+ON>wnE<_*T=Ch!3dEK*dgXxi_#o+LR#+?EuJxIaW`^Jy5
z(>+EeYcc>#$5ZrZ1nXhT_gT;<uTW0?cc{bc!81Q?hkeVJ`t)6gncuGd_h{r_&+-ZF
zL1eHx3SI`RgW)xvX&Zdd==0x7L9$&ZDH}d!4$eD*1e|Q`fvy6Rt{jHf5bYm|{xn1g
z|Na(WzibW1#lG4;Bh%!}9TTLM*|i~weHV0%vCaC~{3cV|d^1R7dY^kr+C4Dohmq^Y
zg<WykIonr$iTT!V#6=sanz{gcEmFLMo^zCr4`JQHz!Jdfwx@H5aQSpS_mz(exB4=o
zcBXM*b-^_8Bng23W?cNW-hEox-S0(<_+8~XlBc7hSEBj*<^8RLdmc@Y=t6aZFaf8_
zO9)G1lxR`&7L(12Kf3%^_b(0uwMBH!=!;)eEF1(lOxbeC)Wu|u+;)n?Xm|aC3%a`8
z6d6Zfa5)a5m2%Oqlh7NuKBBW0p6uYqfKO9*wH|s(_a%U@lXK1W^gfXVQ@rgBW3%px
zHJZh0d7*AlFTc_+cT=%mEf9`68AlmW@Xeb!>-&!?(B$sV`VecRFx@c4P&6-i-7tu%
z8-wx!>V7Cpgk_zZS)xNOeLTcC93fPs_);NMd2(-r1)mrLzac?0LWEEWNE3ZKh$1Su
zz&C(FRN4SO&XoRpW<h7)$C)PIh~mYts(XEs7AGXhA>Cm!^kEr;SO!UYG2`A35Dhm}
z1d{{>uL7|s2T`d8qsxv6y1lf69~UN=z@Oc|+zIL+*Da~{;c61$rG<VV^}DN@nYM`J
zR3ZmF`ccHDy?`$V6y1vz_)npBpEWn$?ApVDtD&y~1ngimDw1Piy)EiF_0}OB>*L(1
z=aUNz<{Z9oFajiA>@9^hgQYrz(aIC_{(8N!ppYF<FF8LZ!Bjx){Ie-(H^KS6uOK0n
zaozI=eiF~4<t0mUxJyAXKUw>2PDXMg9=^0py0qAN^dX3G2*cy{7XP@gsKQG{Pn#13
zTKe%I-I7>`ucbcnz)xa%g;CCY0ZxclnOgq#Xmb)2kkEal#Tt8isW1fF&dd??Yvv*m
zReYEaonGUGB{38N2g`4!h$`Qpm{wbvGbiE9_^V=2mfKviLleDTK&9RzS{cu<kt5ci
zheXr5@6c87CbzPz#k2%;qp#dohD$wwd9GKgtn+h0zWe%yhAe&z!m}S$#gW^XqvTPK
z9-A70ZqHNOYfR>1iAiQw6f}voJ=pQ7By&xXCMs&b6FNa(E`89~0=PJyG)xldTsp@O
z6NS_sLJ2-I^A|Dw#JS_w3t$;;HdG$N(wZM^*PBcq3rC+$i6@APl*gJU7262tv25{!
zcJ(Asm#zZU$O*kh7!Yq;lR<E94lzTQGYhinl+`O>C2#@BVS=XtlYY#xobP<`N@6>@
znQ|^%KX;<1TKggQADw=tEvfkjX1$&G^i#uyz*RlH>++8(e}@Lfv2m62?5FIf#bN%x
zQ!QZsPPN#uZsN%S0Y*XLZmI|W8EdhbM@-k)H}>24t7wf)HR(@e0;xG9tvTEpIH%xP
zufqTGQTYGMNAZgO>!aZP^-<`qAePVbJS*Qp+-APee^VoDMjp!*b_#@bAZ!MmLJO(r
z6~_Ih#?#-GA|0HM)T>6=zSRzTnYgn7a|t9e?DGG42pg!rL*iOUETlgG#ger*z}Lyr
z3ek|YF)2~rV&uRel6t5>kn<u=zpsV<bNaicU!rM=mfe@tU%f|%z((D0Ue_)Iw$tP;
z`+Wn0;YO#PK;2fo<K>4M2N>C}KYtm1)|~2>v7ax)Nmc4r=%sy+4(A}g?C?mG_a9?Q
z?PYZdfsa8H4C4+YR;$aUrF2}x(`c7as*N;ThP;9)0x5S$<T4E>_?MH6_T&8Vcpe`L
zH2YouC&t>#csPg89|wBSl!1SM7=%S@#v9WBrLtut%9%+*z1-AjH;L1I#OP*DdiNyT
zmoZRAaK;s>FB6xo7x?@NjCD{0fU%m6*zRtaC-TMiKdYD_qJNZ_#bl}L=R-6P#~5Fl
z#d@_pTQdL7uVpNb%Yh?A*zMiFZ)+yMX7+adR?vj*$H=%1Ur_d_TC5_&t;OrXBUwzE
zT5KRxss2MPGB->Gs%U7_g9ZU%S*|fN=p|0{%(b-sBmLWm^Us%s2*9@NLBjJE3Fj!A
z+dF)`^-#M`ACKI2)Zh#J2tK@&B7xhgFMSJ9SAPah79V+w_fPbQDKlb2NvfZFGD9Xl
z1-WJGGX#KZzsAfG^i>>UY`?>V{BqxicN2k%fUgkboOF|Z{_zdt?_y;m=Q_nSS3DTM
zK9Ui4Q`3BI)1u0p1uDVXq^Lx9J}CaR<OjtMbxoRyM4+QbnmIa`wH~hKLx$MuV2Gef
z0v*sphe;a0X86Z#$DSBsmgBs(+KA6(WNtx6=@oFx7ytzGA)pzWH&Pv`B%YJ71hy{F
z;U&zs81})Zp2Ef&KplReeAZBJn*3QErDW*mMxoalnVOuJ41x(9N$)z6pF0&S<wOw{
zli9#HoG80^ca!hefXc))e4U9ikiPq&h&!_aZtmETi$@uurAp%Lt1Dw`5h1!1i{_d9
zyDlH4x&0JDUP)-nPs=SZU|%ITW~TWNzE?K6p&A+TKoWVV9=RDEvQ1X%XyAY#z_|qZ
ztpQ9^iww=m%o+_S$olbT^IsCMTq3}KX>>Rnvx$deqKa0;hbT(~6*p<tg&*qCy0arf
z8KB!9=s(mO^BE%kIL}`*GSi=4PkHlLjWT^FcvJe8O2^Dzo8xf8jJ}rbW)VESq=prz
zS)n}90_n7X#7{aI7RFsbV@@duZz=yNkQ+L@$qB;%@W1RHx#gjigUtGQ#OzZQ@~GUR
zT$0&KZ8?DM8a&0fHpMrc*b5Ie6kZ}w@iQHk10m@8`}p{emVgr7!d=|AH>BKpXv-1Q
z^G~@8<zF(cx~a~W>oNEXTdNGi=CjVVzCl4h4+L;_jv%!jK^?nvv9x<~T~<J#e7aZO
zs=&nF0DP1iJ4jCx2+b*qyX3`}7R8|2ykI4`L-oyo>7;%C86FG}WE~E{vk{E67k^4K
zvQKpGs2qzDb{=iPaGQ`js2RGps<`<%gfP3k)KShkv^BBR5wp<0tH5{f8vX2i(1?vN
ztaJ+8oW6^@3Ay;Ow{&tw2k!W<Qp%_}2zqF|vBW>52Yk*ZoyDjmn>|BJd(KQh^P_@K
z4XyoVl36j(YY!FCi<dunC7jT*e$`MU$`bpXY%}>J6@&yZx;SmscCv0<oD%%3m*RoL
z^q@$h=Vj^$(t6ol8dZA$nWqWiU0`m<7ekw~@Yu;S6NO;~pPdQHr~V=d&H`x`H8})j
zrwWK`oL+IYAo<XQB2}kLNz$gnU{E!OV(Qm;eh1TPHux~n@<w4KI<b3=;k;nM|Db?&
z!uggHTcH6Uy1wjK7C6Kp(D(a~ecu@C5<ozSurB%tNx`Ubl{tzoe(>=1_7f9A85n*~
z%4HpCmq8_DOY+ztas#~xf~3C@kl<>KsudcwtX<iFzph&SNjB;|AB;c^n%U&a$hs5~
zNhtkblSoQ$=dYZ0O^xA{HL~|=CFDilD++=HB8deUSv`LrwbDF3jN)E$wJ<NI%3#s-
zqM_T;$4M|=r2Z4Wo1>ml`GAk<#=ah_Rkc9*>*Nw%<FEcURwv89f^1X6r#-CxJozz_
zhT-o>8oqx<(g-hOj?=yniE}gL>(a58DnR%f=%gdHr>`pYlNsTpvEt(FP2xsfeAEHJ
z|Ly}44Gsj8HZ=~c8{8{&cd~r|*tnrz(NOX1#)UQJn`M#NC*^Hve+Sh3e{=#L&wRN_
zAn{&tcqhJ`4aYB-tfF^;7yvy+I<(0th!&9|`28Wl{ihEA{zoD3jV*NtMz^j)NW+D%
z1jh`2h=_f=I?@Q%*|xGmeT-?wPU^pJB4PH3Az*CI4IDW!k?sh!0uzEtXIW0v+%~XB
zGHA26GZmleat#UMVIP2U-3MpqfXV12qmVXAq=Yowk%3!+@aC8fhXM|&2vXWVJQ(8Z
zkM3;ZiuZ4K*1v}N-7_+kH*>5-%;rw$C*XX7S`{VYX<=Ec*@XDOoeM!!u&kLP6XynV
zN5yK_Nu5N!8y+)i83C=Z@~AMr9TfJ2FUlgcB($~TY2~P`EXc-^>(BTVQ^{)BNea)D
zxg*9|GK1k8J)*lEK>49o&-_g0N2{;4-wflu#at8@56_J<K<UCL*W@gT%fmFAi8K^O
zIgUQ~qA{T=NRWO7ZtCd1wZi{hDPJtEOxnP)!%X%QKLt1Xk<88{o~jY{@t-f-IK<1T
z$-eAOfuB3kWEyByVp)x-dYwFr-aiQzcd8UvqO{H9)-3ul0shK8+=>)sCp5<o-kh!!
z-zF+QJ+OdPM%9+;RF7v|$ziUf+3Y}X_z-tA;_i%fj-Lyla`Yew;QdI4xVW;h);l`s
zTEd680=MJ0Y-#-R2fd80$gry6ldapuKi0nXc00DE*ln0%F;V-d@_#WT{=H#?x4v{5
zr`LP{C&{ez(@6yfc(JlMLtw{zv&{2dZP5M%v;4<c0d?=5aLP5T{@<ZiKF9x#6`0Z}
z=w0zRQp)-dLW0xe+%2mm!8~69)U846=YmSwRpaWwNBRRF@tus4#)$6ld<b~v$4+oV
zy*SzEA|e>$VsGcA*%*ltj7#JI8r#en8pb4tVrMn*1$0vGCo+fVnqz-;q$XW<xl`rp
z1zJ^-L!m7lcEPd_UF1eTGLEM}Kp`B>Y4xR9(E3Vph3Z55ORuv!UHixSvu%c?jL*M?
zU7>Jy@1UHOJe9<4C5XFjO}C7(A_jGE&zM~1$t061ir{{Nsdv2O1*&o@`^KGH5rs8@
zGQ@MuWWE8m&*iu{iBg`$V>_y{C1{)e1P&9fYuE+IYB#SS$Ae6DV$y6gVP(fb_W}?n
z%1<ZC=T~={NAORD-z(OR9K$0GM0R`#d)RPXrYio7C}hHkb?tofvn`;+Ys|b__I&ix
zx4Qj%Tie236`|4le;9k`;7p?aUof_9+n(5VGRefYZFFqgb|$v1i8T{VY@3tJX5RN#
zyLWHht=c+OPyhRLb#<Tf;TOol&XLk%jpu#E!KEt7wdRnw#bEupn#eoAYxg56gdfd$
z$s2pXzI5IGp=7|W?#((y&qJX~Pulhr8mG~O&GZk;^8(NExC8X=<k^LJ$A;}|L(3E9
z=x4Hkq!lU_t55Q6%<i8_%i^C&i}t|)vHdJRHJWk$srmRp7eu5KXzvjIa!*Ky?-y7n
zE84NY?vDH)msa~lD^zZ_J+a=seXYuwI|*VPZ}31W$2sB=^QaBN!tq}F!XRRF$?TYy
zT(QbZjZ^7HiO}%btVUNmNAFvUbqiCo%}zkRtlr^wWF-QTM-DJvmo_dfO8WhiKFV7I
z&aLtM{ZC)$0wy3v{*-dU9ank};vI0=tv+7p@m;9V#~kz=^88>{|Bv>BvH!uAx+?eB
zsk*b`5E;aL>JDO;oOiK)U4#_M^E%zar6|R@)0gb9I6?nhr<G)xjUivdBvUia^HB;|
z_4`TWKDO1oVl$87_+ZMJFWh>&0VgK#Zes&$>0SqyER5+V1)#`4u6)Om8Gj`ks2Xks
zo3dFU|7}`3q$TT|alwdiG^K8u?7wTz<p6r|$b!@PMJQn-H<F+KVp>*PdV209Wybz5
zcJ?RVGn%I9s?09(i54-)tr7@(2`)?j7?0yon62r{O54Ch*#_PGraWDw-R8(&)dTM0
zx|ZMYjm+jgIcS^GtJiB0iGrvM7q~MD!oTN(jxzlDpTUOcbv#lRsErQW2ElRH^g@0N
z46Vh+XRFT?_Ml~H1`zp8_OqM-9$P(*rsbI6drG0ynD>G68GB9#+J8~w%0}5|2HW&y
z=$ep@n)9RtLbsn9JL9=7(|VWkn;|8k)#ew^v}6E!f5_i3s+#xE6Qnup!RL*XuStEH
zSAY;*a%~st=X$F7x{;tD&D1sgOc4WvUBdqSFpkk~i}_)v^N7qYTp2+wMYDTF{?gG?
zCErRR<kA0v7D3PI(`bo?FwpY2^5hX4cUpe;hjjHAd<~g=W)exW-2F4es#*tlEQ;ka
zk!6jbXOZI9FQ$9n4lxO<ea077;+Z?$5hrG=-&M-vYx&;&%Y9^SK9vw<WSQSvygk`W
zcAMD6pJ!z4Ei5AG-p2!bACzy(GhAh>9jEVK|9%zHc?s7if3+^GiL8a|W=xmj2++zv
z?%txgO`^$B(?6u*`4?VDas#3dr2kWaxKed!8T^MYW}gpn(cobI)31k`-55r-()hWL
zB>uX%!a3Xbb4X-f@u&Xn<M&=c3)4Ch>E%lL3kBT%k~C5+h^9N7()U+5wP51(+bYD_
zqmI^IiLPB5em|bb_nlhgFW(yshAb{24B6jifopgio9HR$s3*I8;DGo4!Fht?hO8bt
z+|<koeGz{!a4b3sIoAAq)=T1PG=|AFcWNWzAN~cD;Z!~Y%BH~_fuAdsR6#!Bx<Xh1
zL*TSNPPO7E+r!mA^kx;`jN0iVdFYtfE7XiA)+CG|g7Abk<ibCDKdPjLaFUUFUKL!?
ztqVAby$LBQ>?sZ<6BJ>8V<<@S>)`0i*Dw464cnfi=VSKqpSD4!9@=uE;o=y}rAg5)
z40VJ4wyzC8^8T4v|KnlJKk)hIVU6u!ox>+57}7c!Ec<#rO~t%pd5Lq0Gto+*6YCA3
zHt#*y(i>t00Y@uNWTo}#^9=9R1nhDjC-AjDZ>Lu1Rc@h#4{t`s-C1^SrrS%Yc@d0?
zICaExE8$qCOSA6IL-17Vg{F--FfO>uF(+BdYm~Z4fUu<Y4Yl#(&u&Lk$fn5EH(Bb3
zH!~2aS_-7)F|d^3Y)=cG*^9YbG}fAr-p#W8UUC(Z5BQ-BW7_^3vqHl@;1Jb|>@#sZ
zmG6W`@O6eC|9RhXYN}T2J1Yv2MFuZTIVY$jHoprKrPWr6hP280AStba$zBdg)<xIq
zH(_)AK@xa+c4Us9PSy-4SNqXztW9DG*OD!osMQ-X?G^*YDUScBp&(}vDMXc@#Ccqt
zrHp&d$KmWH*`pqhW9ApYFrnz&jnNq!W28wVamsGSVH|RHHs86h(MBU)<*FZNkLV{9
z9znZ<fweEPk4IGd<bma~H44twKR`bFFp1jLWu%oK|2ok>Qn>y{<ui7L9fVs6znT~A
ziYp01yx<*NvYe|W*MGTXL-u$}^owUFaZO;8+~#!^b{Rj$05U#hN$L_)zE|$}&8B=k
zr&6+DpoAo(-Yv8J7Wljd4z`<Nio!Sq;fuml!jf_~fZzX*0RW2&6V>Z7iuLuA3zQD=
zbTRu%T82^<6`Ud3{m*O@K?_3uNf<tcmw!E5F<oU>n=0&^NIsF`@>rHpA<<|<01s~m
z8wN0uPQgWVjle^lCh3MTHzA(vognpZbV5+HjoW<KJq7Z7?;K<ona2ONP`bK^L==mC
zt)&2S$k+Ge>^AOFwU$V=A`<W!<0*g-l8BsJ)#Z^zVSxL(MyiQX=j)YuWEmu(oZFg-
zT9C)}>GKg_^hgNRbkwkcqelGD#`Q^CTLcUX^%<~7q_uHf6k*NcJ;uq93$|^`++KIm
z9Fre*R7cmsTMJnNNrBISd)#8p`3;=Su{%?P#9S}Y)XUCI+HqoE#Cy2tYX-tih6onI
zcH*yyLq!ij@F=W<;vs&ulwBdriaIn+yL_fw?xF8#bvPh%f0P|?9Nt<{o$GzZTL6B2
z27lotexudHYsJsfNoifk|CUasjUOQGtP+JR#p~b*97q$lZS+a=KNN-2a-2u&4K(!s
z3%B_8c!mH)`d$&f7_zS<PGuFD8sBUmR9aaZiPhk(s^SvA#~=GrIYahM+5cz^GwN)>
zS^b}AIKI`0@$<}aPC9_kpwnuB^~UceFLa43FdmQddtSJgk23uK!{sy6e_bCYGL++(
z$74iREzW~O-c9`)E8C!2<rLm)Aww?k$K~fI1fzkj*kfS)-Sg!}ZBj6?CecpNf>F$8
zn1C9|VO)I=qfD1bt4eaYjXeh-e>$)sQZQQ|Et`2NDP^V=v$nH`6N-=PD?@-^<d~9Y
zY8wl|)(`6l>9J1i;zaHL^NJd;-x?>_#%|i$N+e~)HM6nPa8<TO>FST_igXnttWQf*
zW?c<TAMPdf303?6A6xjbWcg$NVGOm+^Bj}s5Q&727eu9q6fNrS-G?b~g%*oaEN#W%
zyJ6S7R^@AufkXc5ICeaLwPn}VetcCfsAI^Mkuc7~*WFRQ9sFc${>+V`;_Idg@h8Jk
zim4kfhaZ(utXZmh0l`g58}<;L1?`e`9^714d$(89vMC`yK&x1Ke)&5D9*0yf#?aWu
z{PY9~?Dj}Ys3`yA16><9*F7QTGE8BJF{fYc3s)&hs7L|Pe@6bvfB^mC12T&|T3hJ4
zFl(7CmEfP?KZ;xTRJs+!6N*CuJ^m6zkVcb2shz^3h9`e>)ndQc`~H3B<qkn@eVFo_
z_B~m?N7#)WS@zaztL@spqMLYd?Y)Z?`f}5hqan@mf=sarLl*_m8FVxA9-jJuLeE1`
zopp$s_JVvA{q39inD=*Eqz2}jH^a9ae@c2IvM!`-EFvotj#5_~6AJP(FSdc!O1kk^
z8>(!a#Bz@1^3?b&=73?(DatK9vTYNThj#HEP3+;Q@rO>YQ=I4#^9DQ_DUz#BSmU0x
zm?|?_+kMv9DDCe+`|aT`MC%ldUB$#Aqtz11k|!v%@~|?mJ@Eqhw4V-A4xJP~@t55{
zkMchgZ|!@*(y1e7I@Y^0vY?P|;|x@l>&q}1c6}0N1;j@9UzxJKjbuN{Le&&KRx(Uv
zmGAJj!1+4$&UMG&onnl{0=V1rhhf$=8-m?wt7RSIx{J2}db2`5GVziv6CH3KyB-|_
zuC$Y&E8j)IOpiHbtZH$XN@(6vJT@hO^;5{aUC%XI9}RinB~`4y!}g)=v5GHf^l$(j
zZc}^vIw5MY5%tV-X4-76l*=Wyw16kSU~u0&cDR7=tERiZfGFtE&3}vZ&h*%sduc}8
z?Ahs*Hm?S-+ulbgxP7N%voU{Ma5?|*e-1G7AzNgn5Vy<FWTisaA~te>q@oG9x9j{J
zXzh1JBdiSS1A#vx*+3Q*vyb89fqh}!L$EA9Oq1d3LmsWR?Vi<25)|`EpM5<_vE@mH
zUQNNw=8v?1pP`Cl0UkyIr+-8lXOJJB;GMVlB)bD!CCjy8Og22?dOrNlo9mdU7vIBr
zRUX8B4USw~yi#b*+x1lb5y(`J>;DT#x6HRNNLwgPt#v>(JE|7wi^|7{eTIqG^r2!#
zyAqK^Z`?JFjf$wH*5Xk&fLos*33mk1O#E&9IV#ecvFgM|MKiG_n69370t&D-61<fg
z4YuU1x^@G9=VZ6FO}vN`G{z@;wL)&ZIAx1!3a!Z1YQ(JjX?Kb1+OdT4Vnd9XJ?Qc)
zoOs5&&6VFTieoR2CE=TJ{;CtFU!<^D#+bl@^9J1G8TvPK*R%Jn3LBCmR~^Dd_ub2E
ztYE>tmZJKzxNN|CVY~Ggqya$tn`Tn_D<eMbKC4;k507rvG;TgIkX?iu;)%_ZC0yK{
zc0`}NBJQvTMmR*$oDhxe&}XkeJhyP%Z__@YzqN5t>)5l!Om>qL6-;FFEK9V%D)=+i
zYmw>C8uCH^lQo-8#uzp=aGKiF!7?BTg21^U{<dG-mvFuQSJwmxk@mlVcmy{c7|Y$O
z<JKL4IpX}&Hs?HWo_FAV)zN-XZVd>Ld=Nzb9kFZUJl=49J1+8YiPdfcxAI)ZMW`(?
za^MD7^-U5+6N2d(H|nK>BpX79?;LR5pazg?Q?l>FJ>OrfvC~IAu+_e&M1Q0q4wlQE
z4WUaV=rsh)Yyi_v-zMN*YL7pVG4D!W>h}Abdk6h%4sBk%KPWyulE#+%J(#>SIes%z
zN`9O*bk2g>*UU_(QGe*B4>~K<>AL0T(X21vFXQt!NVT(N4&weIv(kJ~P1$5jhu)_4
zVq4Z1YD2mA#(UuH>-;tjKa9reqBeuAXQEcpr@s2vYaJLX6Q&Qef?R|v<NR!;THJ{+
zR3E4cQa?Zz04k9NmTo6O^A3p&tp*mTMi-dWTY`HaeX^;1vJ=^-b{CZC-g7_3zF+=9
zb_ca}_+iH^f%~2L@Aq8EGrlX%rnZyN#lg~4#U_D6jlX4#=2HzO;yuJz5j4DkFh_rL
zip^ofg-L+HVD~CLQh$_59{a)UxXb}LO>MPnw0^mWyu2kZSVwJ8#X;V+Z3>D}8Sd>&
zFW!a8o*cwwv~cs&)}`*A?sHe>Ju~wPkQi+9!x8&dwq<KQODRqw7Df7GFe$(M#!%P7
z?B3vrc%9cYr?xj)ts;98+wHP)kWr8CgrmNq7<j-=*Q$V<K>r1ga~m~bAx4Cfs4to#
z>rfh<QUIG77OUN5g-|MOG;Z*hEVxzpdr2|wFsQ#N<bj_*^zHpZ3P>2+`9T2OhgD-p
zBYjYC7Tk{9_h2}I$6a@2Z2p4DWWQwNQ1h1gTBkA4JTt^}@K1>HNsy+F=6r=q`8v17
z0le+KW!P##P~<6nhJ;31rs1|emCA2{ig5*AsXxR-5;+;8^UE3q>0)ApeH3Uh<;-@9
zexR0OHB4dCiha7x<T^IenFoK>{ek;NWY*Y$aDS-Ku_avoHS~YeD*r20YP*c(c_t#&
zCB{RSwO8$B40Zz!!1LsP+V}g~h#UW9<`)-PjY=5p4SV#gS_u#Lf(H@$wi(vm%Kt~F
z?EX2Wt5M4>qdSlyW?^B;2=j?OKil^^{;t~>D28-u+PvY?T+zewc_G?l?y}00?pW<j
zERo>ruOrI0ufK6hw8vbbnMP902oZdyguVy>|A`6l7veU8^7&=LyRJY+{NVmR<-ZeM
zs~3EKSE*33DD^AXzi+Xp#1mOi=6_N`3QwP`kMYtVsm$5>Ao#>*c-{_q5bj?EKc7Yq
za0;AS(EpvY5jS9L=A2g#Y&*D(fMS8?m>VNA$rjf~3r&t{-o_e}ab=cl1ne6K7Eu2`
z#Y(J2k%_ZMkAMC9ZR6OseaV+SA4kjAqV<f5Q^oKdVK|O1$TB5p;yVHiJ6(eOH<7lR
zv;RuJe&A6bST}8e{Dd(^{4^sP#p-qGI5L4`uL2*z2K{p$%l?LTDUXD~r@;RLKmQLh
z<2DM-)Bj6l6tnt2s*Iu-RQN)xBntnQku*9w1wWlKx<owh_hO8ymW}r=T(gU>98K3S
zpXMu<cIji{wXj01e+hdZAaaK%&%%3;`ZHp`)Bayf`u`7-+Hd(KXp%*{4}M9P!ab}=
z4|PAGKr)Z}iTatG(pnxY1HYdB$+VMkfphX706Jg_p_Z#;>Qe*77b{FEk<6%_IH0gw
z2rz#udre_#ewKmI1Mc}34uV99R6gsp;L<61c@8WrYQej4nSV9ki8W^rN~|B4IBQ_-
z-D53;^!NuN$q#mb5a4{U=%nlZ($KPbP5>HW<24%)+0jc@p4FH2QxSL(xQs#$F8-Si
z`M()%n6+3bp@n(-4eN>mAoyN_9)G3OX)q$osgABOjqPUFJNq!@D_Sd>@1ungH@JPP
z`akfS-gT;VM-v|grU(6xB8b0@QCfE5mDYQrFnxbHPz%M>u_u<yCo1r${2cZzLr|r~
zL=*YtX2uzD;j?YaSo{5Vw?e3fRFUu>>@e1mg`K<f(4hA$seK^wUnf79U->h`!BkVA
z(tr8@jBvnr8Zra6FLruYy|*)od**+)k)X=Pv^q-5%EfR~M|?B3Z^cyROl27#tjYq5
zb<4qnALe`6wVOIeq3Bj&{5DMIfp)2NO+&e%M=V%TK+vz!^y4AC9Nf^7?~34c2jQgb
zCilHTi_a+>Kylq$@06F@%{YfH<u2>M-#Q%i;Zw$G`z~qiZlF-7{JllCVVGYu$Y%(m
zuQ`bBk;%czM8wr^ow%SZWEi|H5drqchmie9UP$eTNZp-O43mf^=-ah`_9sWCR$rKE
zoPmd^O*=|&pnteyFVuMD1Qmj$Its%djs&K*gG?g_K%;WAhW-}=v%lH~DX5b^QWW!#
zmwjJZnfk7_bQez*g4@c3t|5pPg3I;qe~D)7C^(8`ka8M@F%?$*vLYqxtx+;ovPCQ*
zf*>S0PVlWKJw_?>4=xb<4-AIk{e;1DpD@^<ghu&*y5K(;_&#a)`tP|OtjjJSo%TV$
z>e&k3N*gsqjrj-_?ucp=jP2+#tERyXLDAY8GAupF<l*r5X*xYgn1C$<tWaM6;v4}B
zEaj!wgBTWj=pIB8lclQu2zsOfOcE1`xr&7DoeDHWwsPLmeN#!P;fCJI2<k1rN;HeJ
zm7DwaR?=wSj=8ya3{b)0;UNKvLCLAtG0t&#BGdZJ2V7VZlfmAU7^8(@z=PU^c0ojr
z$0NB4Iu(+$rvEfxCd-9<zFnGXMib4tCN%`uWCSDrWGq3lhW1H`v1#phyZnEmtl4%V
z8tRFhi2O#{zs^geYM;Q6#oQoxqTgDQJefk+98V)j)rx|)6+ajn+bjcI19_gv*jLby
zydT2G^wtzts?6P)U_Jnn2o$-ha}+e4LJtBsMvgzzdsRY|v&$k?-!fmm%u$nQ%6iv~
zkWBXPrFN<dzEah?<<u$ynKbXcxp5{TW@gP6^a?3U$qPOMETYj7U);k<L*yWoLEly6
z!$}s8#+;M1%Xpk>5C~u!=GSi!P^tD&XA$l}b?ziPr_ntt9uI#NR2|Fdse3^axbo&6
za!Yr6ImuSEfHBmDRq%UDHWga`ZVSssA^mE#znYvrIYuk_S$F*m6gupPZv(8vqg?OL
zX{cWQsx<i+MG1&qo7sXFam1kQwyS0KLJ0C_1K!7rN651sQ#Sw+#ds~YFNrH8-i&hE
zBKxAZ&XJe1%AbdDf>?H*aQhPNRv#<`7rd>_<R7>`-O}d)dM1PJ-@|LIZF->1>Duuy
zxIgD)Cdyt6HU6Gd6{-R0+)oj-?ag=Q*PKU2qkD|haHKI6AX)Wm;f|L7BhkZd<Ax&Y
z&FoaB7p!r?v(^NhD(~`Jam{lEqb(n!ADTAOyQ+o=C;f@R8=~;12>&e)jih3?(B9Xv
z6X?gZ1NLPl?li{k!*30w_cE2gayG@4v~$*>>{>1X@6S!WOoz9xw@S9|u%-jX@rJ7F
zb%gNv-389htMqkHaRdZuB8wU)6~UCC-<Jp0**0=fhqpi=u(s2jFSEBy&>Iq%Z9dBQ
z#s>Ne-KtC{MykvKZ0tt@GR2es?|=*x7<6zb5Ju<`unuGAv}^pFPNpBAXVR|)+|RcX
zlYZ}~u#gV-F3y!4bIJtcv5`+2ek5LyWOdV=y<;am*Q!7r924OCjr(92-NY-7C(9s_
z8yT4a6#NM|D2^-@5QN>RfD=E-iZel?B#-5A-TB^Rg;MSA<m@!asDz*g0~L6{YQkd0
z?ykhCEAUIM7N*b#1SMGBzJK&vcuK~%{Pz@Vbd72Y>FCEP3-`#X9_qpOKY?aQGVypw
zU2p@&i)<Awr1|@ep}tnyG|W-e;l5x+x3N@0eS+^myEoH9LOF9)u2FSk^Oek_tS;kE
z4MI-#dK|I33`J*ucVE?tNU_3$LEIpui4NuD_Tj}_=XnGuY|8S*4j~e=U9rqv-E+oz
zidrY7+t6gl^`AX7Gjxl?7dkC+?h26Aqxvr2BMsN+Q(x(r94Nodz^{>Vx*IBoyQ@fI
zDXoP9=^$l~B6BjjXt8QGY2w85_+kzSJq_<Q<lC{+K6K713?*+O)(-Og^od?39Ysg+
zk#g9pA$oKV2jIo<Y_h&Ix0rq=kt=e#TihkHu1*Bc_TQSbZNF;535n(ue(Ct?;GE6l
zcu!S~nSQJ=>P~2L^r#4DLEWN<7`is_V{`lhpmf>Cj`Wk}p-H7vysbMqvFes6(d)))
zUO6~thUvk(w=QaCMWWopMvQbfdhbfDKQ151SIu~Mo7W3Lyd!e{g3N!na$NptPyP9s
z=`pQ3RGM`eEPthFdypVkoufogqnk`@ePxgz7|t87>`Rq0gp-87N#3Kq6`8E#r&r_v
z9KP5J;I#MmOQc9X5k&?%MZ3{(Tx<)#s7A_F#13L{6N-}<byz|B(E_?PJ2>MY)9L^H
za1eBinV=hbEtDf2+({?Cn3?-OnfAP(1}66<KC_{334C!i=f#l6kt~{&zaPMfi}Vwk
zhVR(ByS=ZF(iqFt!-#_qH51P00dm0yx+uP&e149&kLM5=?~1gciiJnBAy0ya2?ALS
zslWcs@bj4*B&#=~VR%G@)@#Yd;(T2ydgpscOk=F_-y?Rd1%a$!M-sOHX<!#{vE?S3
zXp(lQv<vC?wA7fQ){6!$7jqb(p-ESg?cz1}rFM)+O^&!5d?JDWmZC?~0_@#&3r8SD
zF1UDo=~H#0;S?G-V9L^ysP?*jS#(KZfBVX2&};cJaRLz?&ZSt^Ge`>MqHF+-$Ap}7
zq{LJuH-z0VI*IfZXLnlypP~hH8JB{C&!mE$KBqA>3*QIL#LM2cRf<$|3|=v0S&p>F
zvib~sq-f>V?1RFwp{D9A>62FpN<cZ9*CTCHld1hqh7lvX`%ToK#x^5aq%4wtVbIhy
zDYEmRTY^c05G5{cglIJu$9fdc8?&AEwuzU(k1Qt7A^Ci_NVz6b&Gj7aZz&0;W^oM+
z9jp|khBia}FN~6w1qGeRQ-nLh?<~Vf7JomqNht)Re7?~CZsjBBTmZ;bsg<mWXVg^Q
z{GJG|1dKhQ&!F_$J&c;7-7eHu&W%OR;Ic7i0tsSdI-ASm>|~J;TAY;~C{so#Ni0M=
zEOAnOh;pwzm+~qV@F<J${ERc==y@Z#NXkTZI?bicHyNmOlqE-{n<hsMiT(Xu1)XNR
zIi>}1Gq6l*O`9#Ej(}NsYTt0RsiVt3fq}KrZihXl+uCb<eMv_SsIYA0h45xW^GP_-
zUy3Miwi1)DZ9o#cr8h4Uv1P5Cw}WS~P<s2eXKZ{0CpBhU*}v(diS)VxA%oBmKh3vE
zF#5-!h_Bm<awIr98J=2~=B0sL<e)N{i(XVM3Oj5Hf*3@>IDq6u-~oG_{`(6$>(}RW
zxwfmyzC!1_x+_h6RGTJ;pO-oXv5ryl+Z%7bHIq@@d5$HK?OkGVI0@Sv25dk{I)Z)<
z1nh{J4kI;gv2n9Ix*}$HFMQ8OAbekf7Dr$5vqh?Ci>kyfym@O^uc_hom~;zamI75$
zq$>#Fi8$R-8jw%kf|FCQvFY|9vnE@*Xq#BIV8on4XT>G`xyP?85W4W=sn9RM>tT|h
zTTqKO*Uv2>YE0eE*%c-Un?{9xx*MfSSFLFnVVr+&B$y*+-2EO-Wu7D9@UiDlb&sBa
zZWb&meJP4WJ-#M7^G35bL=fVXUigT;B#sn{5ki=o0+R6$;R~L?ODdW)jiqo~F?;q&
z50{sOW^+>&sr6+N=9;dCRg7WCX<Nm=rZ@OhU1h)KNh2t9=-~;-R*&c=d4nSm3bCfK
zE|Hp%)j#_$N^y6&&wz?Oi5LBO(6FEH()aK_bx&UldE2v6bx4+Mx-)Pd!K-@ER<NRt
zoe5OU1^#Gvc4(}TEb`{#hU&dyXL5QN7WpyGaz5I0?MCsy(oMvf_}c0mMyxGcEmSil
z8C=X@Vou2YYMX694}k`Rd#92ZJ+kV|CczTyY!@YE^cNE4lIEbQZs_V1Q-beEXnwQ1
z=@G7CL`b4JFv-y#%y{r2^R}{T#ni))dfAQ<0F>{V@WYzSM;~5JzuJ`jlNf1A^$5dT
z6;hOp<@Nb_Olu~^79)o#C~^&Z<tw#oJWi)aiANnuzZGw`=lQc}T7X_O@I3CggY|;;
zmL0!GVGo(#iZ4G*nSLL`+ciyt^q@5%<2k**MO(Y(LHWuOP%a0Hv-QD1lr8vGZW;FO
z3t;<@xOVP$6sj8RXMZLNzxUU8`0ta_^ElsX!ZkiB5psEC&d+sw^_7Ix*%{}Sy{|g~
z+jc}m#$6yG#a@fj8B)lseiie`O^?GADl5D<=R56P8S9EpNJ*PJ#shir7#Xix{1)=|
zYh#s@Ad$q3Hw}}92j(esspi$%4!X&T=zv_NSqTemOYN0Is$@7wZ{jdo61M_L%A5Vd
zp^_5Q53pR^Uf&zB<xJ5(txHzR0Cm2cT}M^xM`?4Gz1CWwYG&enndGHxjoazZ;leZB
zUaS}fPw*b($nJO?jx57wQIe_e>xx0nBD#3t5+xfD@4T#)c3kK?DC?G$eJzJ%Jb-N%
zu{|4_?>kQL01iTurX!jOg5)F4lwDJ4!pxCsFb6&yl!+%x7{${DRD&5FG$-r*hkl!E
zEM4VKxnI+gC!fx4(1sH~bqMmQi^}Av`lcbGLJLuU?J89jM+%_<Te`knioQr!%y7x?
zjLawDqBA%NXh~m3jz`wiEs=a0e*z^zZGRgd%#C;@2wo1#rnW1TmOT;k-ladp*guB&
zu2<XMEs($T%SSKu_sdTPnrlDrx32jWgTZBtkx_&9Zdk3O{IaTVV~HEv^HiPBn|)N3
z?sBd6yz4<q>F3X)ep_>+TGUxO{u=#wm9eT3_BG%I`u)UEHIUD6)7NWp=>>>#QjBpi
z&T&yH!rsflITH6!hEN>grXJ<_RRA{1gT0SE%0qs*CX?f0b@R+T+(qCBuPUUMWD?{2
z0}(#P**L{T@W(_o<|@|)jU6!Vw_c;qZ*H9q{n1@5abo1`M#E`G2ZhYyNdkrYdmA))
zT(x1J7)l5ijk0>h6ai<BMiWq}H&(+cZZtl}I*)xy7)-z}ZbUCIQG4uB>#sNu`ZHKz
zJFMFk9>`cxOjdT&IyT-}bn|Dw`4G*S2=ywDF*+h5csZ!a3_KguWU@Or*#b>{(Zplx
zD$a|5d}>BlM7j4n4Wg>m=H_pj&w*_mtC)}FqHBM<d(?Sryh~ltZzup^3~V1|k|HHA
z@gX+tDO=zA6d&bqOOCBp<~EnCtNxd6UNW)ehUP2Nf|J)j`xN_*6WqRr+A?-OeLOR@
zMfCraxclZ|)9B)(3C$xU=}1RJidB{vWirc*X4BpV?jLg0U2}8|UWYC@OX5$p8>0L`
zQ0Vr7nIX={RD=|*H(>(yEl0dO4Wq?}rBo#R2OkRANzEK6rK(191jGG3#!IiZA2=Ua
zO`g>D<SywUq;$U|zIZQwfioz6d{mZ{LWq}f`XYOZBh>wQhAym-RC%W)rmBipQ6sLc
zxcqmTR*^19xSnc7Ax#t`bvhOW^Ak)#t?fY}Bfq5~FG)vDOaibM`5LI}()$^SkLWOW
zEaPA|$S53X*FXHIoNoF@SDd^`9euO1*_tO0T(GywIy~4H_=z|y%mtW_2(+?x0#uOq
zLMCQU9q%Lmh?`q?{$#yL-!0ai%*9qr-!XZ7bh{=EKg{3D*Z%lU;hXxMC+G!(YYdKD
zVKjHAQ5lyW$p+9(a)D|icGjg!-H`B7{OCyE4Z0di@cDJ%wVNdw=Km|_%X*K64>rH~
z15*^ZNjx#NKm@Dq4bzBf&2qOMudqdIubdaiwRu}{h}u)Oz*gUR^q`Ybx!tYKqT=Mb
z=HB0Z8_Ok+PTa|<@s)~I;-_hgmh^~Iu3_VQSC)4{vUH&E`ze}`2CpC=R;2;mFhST)
zn^=z?d1lFpAsv0QERUEj{bHJGO%AAE3DT=~xMmGxqc^SB12|I6B(dB)f%~y(-B@m;
zX^Bgm>w&#cRIy<!o*`f8^t~5_$Rt2ebbUdd`F2jXqW8|7^h<W9EHm}NlJ>Cd%%b^S
z)F+4y965pd`QF&InQF!NqZ7+Gi94;U$Znm^vKPAw?X9&qYo2q(8X$8Vd)tS!-zn4c
zIHM=|&&!BJtK-fEyG6`F*>mF4RgL3?x#WkTEq_b}Cjs9C#*Pqc<CN56YmLCycefK{
zUJX6&w{a>uFZb$L`D&$_4xM*GW$y9r2OpmhVM#!LYXB>k*kZD@)Wv>(lkA2Z1(b(_
z*~!=1+l8$OQu2!;Cn*-zLg{(h&Wp&^D#uv7*m@Ta#G0{}aN!CTKDLJ6vKF_pqNCc;
z;Q}YjzSt*BOt_m$cG%e3AOS%_0YMoR*(IjAv6Ceo2)Gwge!UwMcMv|IoG?R$fQA9H
z<N<Ew)LwiYhx-SN3M{lUN!{|~m7!sM#MNyonA99ev5+9)JGqvG2~63&hWqh#=={Sx
zc%Xvp;O!X+28FFv97}C}Pf5GwnOFQ_Bd*|^$Wv4N4TTm+NsuXEBSyg+$^gUPzWm3l
zLWV}Gcn)<oG(wfi2jA^jZ?N>$u9;#~@*0qqAy8!zW%v=z7I7yIyL^%`h&KiL7hm*|
zr_a6kO%|$?3y;+}k-tH$*a7K5!?+3gfnbfAp001QoO*R*B2Ii0>W%Y}dBs+QS1(z~
zyUb{Db7ejcTd^F20M^uW4X@6PTFqp*%{<(sm^(pn_egb^B#x_uOQx4!CHTE`H3T3n
zDqn5RCp)w)#Uc}?gE+3bXs#F#7rNN0jBiFL_5KrFSVfjkW->JIT8r;pW~QPC8Oh3{
zHMfi{&wj1(34a70Yq-EQw|rT*4Mne#^_QqKr}stb?)6qm*h&;m-~^Hc1bGB$$E4!F
z7qWF9{3v7TRQqBOP~lf~x*tl5Xbj9rVY-^25OT?)c05Gf>Ji4d!bzHZZXGM==+zC<
zvInxMzS|bvX>x(-%d%O@<>^|nDbaVf9gX)ig{vMYv`*r(`D<&%1HPf4chs+0)6KzR
zuy4Z|YgP+{eETEKuGtlxcDFt_c?8oq*(U2<^x9fMDiIZ^ySa$JC38txbAh3AIiJ!S
z4}Bi9sNh8v3~?cznQ<onKhL+fNezOoZg)3{+U{i*NJ49IrM&kl#gEWqe+=VkF5;z#
zzatAYDGB{BXYAbVR*Gz!@o;Gu5+Ez5StS3RcoB+rwf)hNC}zwaUibtdSv+7nY!+7h
zS#c8Fpk1579l<f=u2wAdbqc67%_~KvVg-{d<km5DMnA^)D*H+ARoJPh#&>KlM}^rA
z6s+!VnbbTf0F>sxR5)2s2CzcRY@mX{?wERTIFd8=^4MExd@aZA&GKI)rI2LssBq&z
zl~AS>LzNJ+63KSAi|$Oj7Vbs}&TM@wQCEg-kb_hx&g+;vj6djKFaXb?!#FKT_8bZf
z1AGbv+Y6;1`iG0dMdZ1OLv@|f4N9FBFM}ElfB&BPP23mNZ+A5wyCl^Zj?hL97Ve&P
zrP6eYiJnZVp1pp2vvQpyUJID}_~ywz#oRbnnj}2sHynJt`(p!h3?^DK{s2mzwP>Kd
zi)7A~ZYZDvX};2$j{*>)1shLTXYHCR9mYYfEPaUJy1wO|dN$VARl>Jyj!%N6Og8MF
zEjEqnGQE|UQ4?h~2t#vtCA*9Cs=_s!Eu6@!pKEB>{GyCAk|Vrll<thW|5KK7^Lvmk
z-nhJ}^!q#u<D=l3GR;^?ipkt|d=bMY#{xpQV57l}_}WUCG#>!NObxy-SH`ygMb3rB
z;TBZo;J|uiH?tWu(QF=w=cF225a*2e>qeCV-)G4+ozQrzQ(20!(-d~%xkqGf4&7_n
zsd%ojUMRLaZ`l2p-bt$OHbCghUH8_WM5SZmUl80%ET6@~G=*Lj-mJ};ye*?71o&w#
zl;Yh0Y$`=Xb7=sB=b1rxiyZ^XZ2b57V7a%qD(|}bn7Zj;c-ur2J)oAi0kJqiOeC8j
zVOAZVqMUbwpa^@?L~ZPGTE^2@9#lGZ5U2A^{@f{j0u`JsL_R8j9|RpyR@H~fv^Rqy
zIQo3d`&!4u@_~e|A^hAQ`-@eP%@)msnT6^kuJdefk3E1OypKx-giR!iqGpG3WuFg1
znWv6Zs@86B`8bV8CAOD4RlJ7a;#Ln$VCq{8@Rh*V-*;)P$f47Bs6K+d@J?ahJxUSi
z65ga?c+e;~l{Jj?q>;j1MUca0d^5>rsiC@w#q%qZL<>2IJ;BL7h!e~h%;tNG;Ehxn
z;|q?Oq5)tBXU^ho*Q**Kw5yyeIr~#dG0k6M1coePiE22U1lXUz^@r*-5|F>F#+`U;
zU)o)Iyg)4vi>w#mPMr}s3!T?w{Jl=9t><-3)0Fv$wR$>J^*zz=Qo-vTBOxx!HN(^N
zu`S*{3Z^1q81uKl&UUbGoQ;i`;E<^gz8D%DdIi!>;h##>nc<Ef#l7;4Psc&<nDmyu
z44tJ0PgS?|Bb?NSKa2@?2{!Y(PgTMTgdJtugKRY!g6cUI;<;+pQ$837(0yiFe6|0A
z&%vEC@z)PcnDyw>#-q7t!lk%YVfprR&J5fgfkIs<NmlJ^H`l|y3p$I{Tw>WdHqyhX
zea=ofbxl)C#Py8t$HR3V%Qe5M;KHXPe69eO6m<9j`aT2xSaxGjV$>;Fgx_~7w4a@5
z1Cikzntb;)F4RkTx_3zAK#!L~*(k?<OSf_Jo-t<mDqFFKdRa9yuZgPVYGc#~(hEXW
z?qaPB$sHf%PQtH+T1<x{8W1j4X0sO-=>h)Y3VF{8F1|s5O~)_V<o$3TQNN})x1EVK
z2kZ>if2f3ay8^=lN2s3-w{PFTYnX?FqoGs#S7Sjl?>=Bfb;cwGfgj}|WL%HYu|UA*
zSOZye%SW$@!9|-0)ZvJ5RHs-@M~tI<eMfKM;KFnrz3X6^B-|K*@qW>_%Ovgud;nMX
za}2o|RVn4%1K(2*4V*HDXHkF=Y|yU1zL`}H7tSWaeaummeQG<tmIA5}1pgd~6L$|Q
zN0{JKq}yOpGF%0RiqAQ?-=&iXZcx64s4aOb^~xcnd_cXG9V@fT4o~#5bY&WRsI!+{
z$<*dd;;tT?2psRqjj*iTOa7a;98j6mrFzgO_v;6qUO!(r)azP9K$zgAHr}Fu&5Aq?
zBa*!K(j|VXV$Z5&vUatX9NiP)!A{xc$p#H_4fJ0H49W<6>@Ui>TRufS4ilVgCooY%
zppX=iDCiN9BvcHCUH890A*n!-!7+YZm9BSXyXhu_D}nd^?M5{+Y@AS`!v_e2gnM}Q
zCUm1#ln$eHmJ4{h-Jo}=t=bO1vhV6VHmp{%FGBrz37vpOIHJpx6_a=GM-ko8CBs~R
z#Utl`v@JNyv?j{%o>+h1U`?&cy$KtoSg+#(s565K(*!oJ_BK<@0CuayrA87;=MjP)
z>37WT0l#<niFX0q@QxXWcEIEGbo|l54Y5=vqfE2eYL?^mG1vr*-o{)A_u1PcjJiAL
zFYhql>9R-UZ0Z_&{Mw>Ha>%~aMdX@G`dYFHEl-cEawyqxC`gZQ4M7oI>om>5&$|RU
z=VDi0+3I&n4gQx?PQ}LR%L-B)m;6cdMaMLGV*PT+G6zHb0aBEVWPoWJ+M!*+_gu8d
z*+U7>%sHo@O9^9dXGMgSXed;H!ksCb{xG1;&nR~m6CRY5k?CTAUCeAn^|--QJO_I`
z(zT!plS$q@zulpI5tK=WD_JVB$NGuYlQW#_n9N7TpwQjhRw*pQvp}_GjFmt^_yZe>
zDnO=7JO7r{DO&+t3IG=oH?*@R$<X6FlEC8k6BjI8Cds;kR?AsKy)-yJi#$Cmqzr8I
z!*3)UX*V<`P`wJRB(MN5R1%`;#!pTT6?bu?+%Yb8u0`jWN-SrgC47h^y4s6+SeWaT
zlnsmfk)RRbY<D2snKKo9{-!&#AAdHExutB<^t4g;G`P`20q2h5RkAN1ubrFilA&=4
zo??Z})m>!y<{S`j==WjmKS-sNkVR=D<O-E~QFnLjdM}F{QSO5W4>N`FndkNKtnM~p
zO;rkfeb`9z^-h~FF$-M^&ZgPYNAsd5p6!;FokqreTsfJ6YjJBL@!)pjZ4{TI3Rq?I
z0xDsJdR26Jz~;AkRGkVujH1yJSOugqd;yl`S9swbbz$i1AbKU}eHCaJSp&uKMPgMj
z>Iw*DSX&WTLq%q&xg&Po-{e7vZaCos*r<cX<~pIFBdr>sFxMi9pvs<%%H$cyDGF8Y
zP)K%iNb2QJDS^c$$@g#4oUfU^xK*M?=&~=OweDnd0M&~8x1Wy{=<W?=pG2)pX|L~u
z-+LBSq$wSi7@9Krxff<FRRDM`vL-$W$6z5V50qyRvf$tRL<mNxW*iXEH4@t9s&1y)
zx^X^qdw=y^2E%j9PCW)BoThQf?BRWwByqgB@KTh<YqPIpsbuNt6!e)VX|Fad-v$0L
zeASb{@aZ`pEBu1JO*Ysr$0jnItLu@%qZ$BLc4R;~XxJs2KI%U*39c$zy=<H49n7Mb
z=fsx@sOAu<U=Hc6Hp;{Esmlb7)Z%B~$K#PdaCOwQtv`%%k`&)*kunx>_}8>$#p;yK
zM=Dc0M6P;lTZ|bLU5zTo2{qV_5m5R$eS`xy;Z8hH$hDLO;G2se@=KwcX|m`rjb<b`
zBgJ1FV4+J>_=LV4lI-hkO!P`5)N_NGD6+$BEn`f*oT?VNyTLIp`{%FRXKGgo3fOE2
zxxCyupUH=BRwlIGZNWf5fQ8Yr$fI5?cxD9Aa>S?_rid!*U=?HFl)~(i?q?Xl{)_@d
z;1E;^5)8*kMz})3plIj5ysN|wlR$z6f`bNQH0T7Ji@^CilS$J4F*8abqfy*@FTg47
zry1cVVv!A0fs0X<CRt&XZBQr-P=Rw*1X9S68Dw|th}<C=K?dw_hGs~HC+KqI6?guK
zSP~q$l&40mbe340QYoigX>jSx<e38RJ--dxEClkEr#$hgpKnDk@WSd4!qbl4?=n-3
zTJ<3K*oGw5e<xC&hfak|HA|T91_vf&|68s73OR|ELf~|8U=5EVY?!$9SNYs9erp$T
zplcG`4LwSob;I6<*kkdIHv_B+za_cExi&%&-Im?8r<hBpD0{vVTdfwI0$)Bvj-p<Z
zV~#xxm9vdlc^r5j%QU(Rc7-I<R+mp$B)ZAvEI#e#w^!lxjxIQ7Pf74s&%|a7gUlsJ
zwsf-y!bsd}tr_FH?biY9b}0TjPUDFKK^?`WHvuFACllq>KO}nTsZKLiLanSKE!JG*
zC#MCwn)aA1(C2AyBb6W30U249QNofK9qmi5Vk=Uakvo@t3pegp@$OyL8D3f@cp3j(
zusm&2gQ{FxXc}YeYjMN`Mo{vSDhqwzQGgUyKcmT^p>J`?S#v(ow37+z)u%j!;giiQ
zWQo|=Y~w5Wo`CkEUg?^}xbQu3-MiS|2avN*8Ff2ipRgFWC@oC@yu+8)B|rj9_w!T)
zjBMbKKY`b~HT1V%F<B$lE8NG^R}PBCND0)N@LHHQpz9<^_8|nz0dh~&A0%GvA<!0I
z3yNv!MC)Dc`gGdOZ-v`wa=4_E@*~VGVj5~)&Y1GG>5BrQC{hOYMW9DFB=k>rMmOF!
z6-@X`(te^-!^Nut%uo?Ybtk^0l^`<#rp_e0{L+GeHxX@hQTQ1+?j4Efk;sOLdO7RE
z=zGmVmBybH@18@#W?tC7U+X1)<!YxbovMiI4Im!y!hdi=@*vX<Zm8M9@`_t>$)P#I
z^1w+@$?jt|m-94G)90F$g%ddMV+JG=(>wXwGt#4dadgFjhwy<BOkW|Lls}7R^|Cy~
z+H_PdDJon_&;kSq0<sw144hT^5%b8Rij&^!r8*<kCy3wf2Y*eEcen-t)8}RbrN4_T
zg@)^ANJnW|hud8CM0_D5NCmA>gO*F37P^bTjOP8HvroAwrw$(A<zpYB<QnwUZPf#N
zlb(B%SQ-0)iWIz9Dvy-s`m$0?ylcf3K}b(+kR%g>ZAl;O1ntke!LlgO^XERrc>`;U
zMX^G3hX4ZuMPXUTZZBI8lclIHcZTq&x0QNqBot#>U}Ri%(wABy=%<`^JSGpdvEPi-
zn=F;tBJgm_Qn`qE@|Rq8y9<4cT&GK=g7TZH-y#6MJ{t=rAyU5;3Q7#Y1zwAVr7Bnv
z+tD}I@H-aRQKs8Lp8LTpMfV@VtxK*9FT7~IlrYRt>C3G6#~-rLbw11o9kOWi6Tn#B
zOfjkxEW|RDwf|EgL&{WHLqbSP1QyXHD0#!3N{>9Khe9j5cblF{w5?~g64Gd%++$f%
zd;prKU76WDE3>dAFE(Tf4)ZwS?jFn`wA6%&;fSg<J3RSL>Ry%NS$Ta=tD%EJYYNq`
zmGPoNU>HSnQia3@g>^P!7bv2T(cs3CX*^lOpmRu)i4kP1A^H3#Pes~cXZX-K@?VHs
z^5C!}#TnG-O)<^n{36&olQushhD4i2z5ty+r(@KPtnR>a?7+lF!3cIh8Y5$3U&z|5
z+~K5@@<$GewzI!ku<X-C*r!xV8E`Yy5J<PbVcJ<yPR1+72Rp|*J0^xBDS(w+scr?*
z(qw*nj;wGD%QrVW_Lz1*4+!KNgP5;%TaQ2}rR5>xB4mAlj*8=^Qps1SQOco)Lj`b{
zFsBBHDtq4uHJQL7f0!off@EmA*N;omLXmxR_xHiPx2;4xC(w#t%ID8@d~v1?J>dIV
z)Al^P2z;JgkJ#6vtb0^XY1CIzW-K6kgq>vg2!E*A(b+9nx<}ml^!&EM{u{u-cm&!2
zj<<7fX5UbGb~=6-$D(r;<<&i?<q5Dl{v`k9QnA%;)3r|818oB{k#(}NI!-N8Otzjj
zdP`!q`0ab}wBH0N+o4;`48qX=ras3&i<>ftpw4`3V89$LpRa1^+mf*S8iaVg`8{Di
zxtjdFFS$!BqRlCY8JVStZJOd#HGM^Y^lWKORwTgLt*;N@58342YE>XbEfirz-;gA|
z5nE{>bV)(L#Ug&l^636~FyA#y{3$sdYzVcdN!P*rkOdDqz~38XEN6{pMhAU!V?AyA
z+q1@59J04>xP(C7Rdt5jm+mi7;ZY7BFAo_a^&W3r;5yHSos^7Zi!~dMO+H*W@4fzM
z{UJcjepdwIsnAn*$9$-?&f|a4Eem~{-t?4SO$NGu=7d|g6TWH@6WkfuF2)-<r?XGU
z6#A;7Si-+H^Z7%-G%ja$zO%5@K%W&V<qPp2v5h%Y<Ln$WVxI)avk&i4OrGCem+3LE
zvsV5ls!Y4c8lHk`Bs?puUemq`#2vL?2kYm;^nU{Gd$Fgzyfa++Jc;?ptY_!*)Wgw^
zWSSytaZ`i|areCRRgN@f5-5E0aHx#7j(@;{speL231QGdhI-o5S$hXxRLra#Zj~c$
zpXpB|zsk#`s}A*K4l=W9L#?qOeMfg*S9d9*tCEO5h)|Y9Cstt6$KgaFf<eY6ni6;|
z6I=kmB}S)a;G$#f>cCJyIw9&w-GAK&snqjR8O{S1zLF@udBdIaU;Q>(ngyG*(n91y
z3lFmns+<E{Vc8OuebSOFEs9A;0=L&z!|mirc?aiN)i@7pvi+hGo>A4w=qS=85<n4$
zy;KR0+g&Y@jRlq_`bB6aFpL}n=jMy;-PM6y*mR3&HpJ^4RACqrY+(di@nyT@4c0#y
z0$-^+Y#)Oq@|*S*Th#Q*J0Hu}w<qV8e%w!QdWgq+9%pJzrp*;2Cg~r2<|R~PdwKSQ
z!g+#qmskwQU8W`Cjrr?rBe7#WM0IvsKw&STPO_J%!cI^;>K4RZ#uKW-E`HbzQD6am
z*%{=gEek@Gt@v3M8c=1NrdzkNgoeV)o|H8iGS0YNu){guGMp0kcW3Ox;Y4rhKTC=D
z5a`Q5(dBYVtk>6y8%Vmb3x*F~ayWz-@Iej~ane4@Iu&k&)y2W3f|<3b1ZnD|HC5>8
z;twq3t+@?vlof7vj--yz8!FgapF@C>JRX9Rl;7J?8o_IiU0%?kW!IuEoYQjUp}Ga>
zE~(u)J%oWIFep^cNyPI0np~KY2ZPExhk-23_!Ezl+CIg&+EB9W-Q2=&%QM_-2{Y<$
zQKJktPe&-43^DnpQ;R0wuufE=B#-sj^C}PK1oueH<5yGkj{CNt#L5!Wthm`7b<0$C
z_QCus<QFWcO^H8O{!;)fw}t5%k=#Mlr4gG9HwbC^nnmuueH6!vD(4f3Xu8>NP-c7)
z{59cVaXKHON4s&}abAZeq%jPa6t^{#zwhaFzd>ois0h>7euG9Ke&xf3WZS&SVY2ZW
z%TIUO1f>S8OV8W{B^Bl;V?n_Q0J-YNKMX1w7B++kC`3%cqnOwsQCn8IyR*}O3k*|k
z*+vr^g1_Uj)+pES&YeA)wF^ih1`VO0CXP&h-UKBE8(&Gs-2!C<jZK%@0wo4I;lnT1
znvA)jXgg`V9K&gOolsiGE!RNSo~Wn@HWF^}g?EszK-L)mYQ%DZ>tfcjFH_*O{h7{`
zmvdDT)mSJf>HZqD$afg?W0}~%r=f9t0b#7A{*h+LqrMRZccwZyh8u!rn72s_Z1QNk
zU?%pnaTBA8*1owLA4b|GumsRia{I|Q!s?SZt#1kmb~$+CJ;?v+JkUKlkW@5JQ*tti
z$Z!o#(QzAE4~Q19-q_Tlr|mavy4F}Fk^GfHtHkh8v1uJwX!3%d=ZP2|@VWNLUc!G+
z>{(K&I@kA?`mFjozV`gR4%1O_P#u~itcyU7kQsrNc5e@vYAP<1TnjvA;8y)?n-u45
zKp8=|AjJ=9obN=9=mp(k1n&X-;Uixv9@KtMvL?%+wL2{COY)@GsLR2iTcO3=t-S22
zQ%o5s`8_BvTm7b|U#f#n`h7ly#D=FK<@?Gaq&~+Q`D$zJH!5&0_8+Sr$ESP;Z$f8_
z0G>i<8RuPs#DFXfTs_we=1iSSs4zu^LHG?*?yv3e@X#DY6`9r^Ynl|NbBs%s<kB?D
zCH@sR9&t2l!XrSEG_o=6gA<FAOnDqD&V&q(1Si(DL&u7Prwq+@wnV=~yIl7-^jvq>
z{;!a{B&FS2@fr)v-5m#C4wqqGbgZ3NnOT`U6#f*O1}@_XU|UXb@`THKK^2b%1A;jk
zT7=SbLFlHi2((){rgOp)XPDp=IOcquRctN|r*c(H03z*>`&9))BlD#MkC=JWymZs0
z$^W5x&7zSRmrWW!pU`*ACa||U7TNIrCAT8?X^X7gM2hLKmog|Zj*H+u65L#BbjZGY
zz%WLGh}UUB_~p;AS61JMAHKc3U+!vbybb>NQR(0+F0(GyEU@6dD>QvMd)ab}rNJ7C
zIwgwqsziD>@@q`yk=vtaAC=mO`{vOiyn2OBa5F8pRQFQN=>+?3ghg7a7|vs$Do5m?
zU^qLpHjwASq3)Zj5k0Ji(a0#≻N50M&^kPyc_B^^U=TykFE{?1^n-V%xSSwr!(h
zPHfw@CblziCbq2!CVBJw?{4i@ZQbgs?*7uB`nfpgo>PHgxMf@0?$tYddMZuPMFDY9
zwO#ntcQMgyS82ys-B$P3t%r=v$DsLo1!QdbCJQ(k-7|x&BYdeBRpX>NGWOkZD1aF{
z)h(ZgKv}GWtA~{8o67pFo5zQhPJSCw-{17;d>`Y&{DIB$>?ca&MSU4dTcZDKo>Djw
zxTeZ)n+}5Eel`5RnQ~+ELszLw@&+?ty1qQ(d~7w2*@wHDu32P-*XSL$71B|t5bHBu
zu!X-l+3gxAtrSVN%54(!8zozVQ3nyPqM-7-#%kY&Nud19d)XJ_2iB7Klx;XtntOz{
z3&&%$`HlMzy$vVtB4n-TrUsVffiZb_aE=-Tp@-;L|1aiCI0WD{P3KQH7De9a7~AyU
zXzySyTKu~24tmrVoB7t1p;kTHj3C`y%d~3E%s6mSACt@{Fxk^c#EC*1AuLhSlBDyh
z44-{O2dqzsRgE7TY}1E-JJ+?98XbOV=~m0~;k&5qUI{6lWP^O8C1eGF^S!bFt!oFH
z>?ey5mN+*r<KfWmr-!>cR4P~Ws7i-Yxw=Qwl^?`eQQ_$ilc$HbF}tVpIQkR~1uLjq
z5eQcJ_lS<CxtNgdwWOsEv;C<S0UYX5<%=%!t`8pvCwHqw+RaYC?rTW#Br$gSnA?TK
zA&<q$m{xiz>`w6w*pO4;OUyGm=~O%+g)2i}m!$Ip+1q{pyQEgTCnwzlnNEf&!gw59
ze6$(~{6E3@itX=TT~r7@K#YjF!r*~C!|6O6dH0s15=cgyERm9Ay~iTXA4;UC6rK9v
z8G@eZA$|1N?3;?^*qPMk)#bo`^CXE)OsDaipSlaFbtlPSNPQldOtY+Z^mmsTvl~Rg
zbQYUx0MF^0L8IU+`T#GgS5Ac|pJYA>5>88}M~?%~jvHcvhEx2A{se1pA;VF=IS^zs
z4PiD#1+e90d!_?}(k$zO*K1a|k7pV;>_(%K{J@MB>L)n@<7x59ke?7C>f<<bk|T|U
zS=cPWuQk_;i`)UX%!{hnOxcn?s|j&FNSK)12&>PpPJs7#2)-SP!B*g&O;;>i9d1bm
z0|iA#j4qC?J;S^&OmKAZ8!!2Pyfl`nUY{nlH~RXBLDm?`4pr^8=Ny%?E!)0x?IyP>
zHEFGi<g0bKU9jY2?Um;?nIh3YQO@g>Ig~zm8>y9WR=~1U_rz9(h76}zhw@G>2hte!
zS^WR<(Q+4g2h0DGk-n5_uryU?ZU~-=Ro&&ck8}vIO+j5j4$ncpc(807P`oSRLv)ob
zmynB~AN0h1q5iYYd=h_p=!36f+h-WHjvfwvp9+^j=j0BGLh$gfJ9*#fd|f9G_?2oB
zz<=+%8)ZXoRw0oJ7J?NNt6BXMHCv5~NUATkCM?4^XnwUNZjYDXO)4a5T$>5V@95sk
zCSq4)J{GPeOXFd4UCsSmYsq*bP&ppI7&U|$K)PIW^U4i2^x0^E?XDO9?P%D_ArtB$
z9qJW5kcR4G8bq+Vq#;?)6n%E3OL+`PyNyMco)^oG&t2GnPZO*U^^JGwYzfxEdPzv+
zcCeh71aMoIWVHs(30Z=GH48ic@JzH_jD)n+iAz1WRiq;(B^8ID#Qbq5*R6cLD9afS
z<2ZF|DONKlprVY`9mz_A6j*?gpq5#a>b!pxJ0Ww-DQ@-G%sHkqjrUd8y6G8M#ZxcY
zpYb!o!ku7$TdaedMVEh|d7D5u_5HCCHsUVXVUx6d#>p9PmswhR_200r!>0B8=3Bpo
zp`h#bu7Oe&C-&c<xMLvhbJB(1=B?p`G<iM!r?JDjLNu<Q_3w|ISE)%%_1;?~<4kLU
z+&HI9nGx1L$mh)6A!RWrQg8s7iO629>uE7sfJyxQ>K_II;yu(Y+Y5x+bFHB}7zMbb
z6hdZN!w3bUkmk>-pz^tvf}^QnjUqjp5O=o!q5Lg^G^aM0Uk)U+%5FJG26!2*N0be6
zWBQM-3#B(a>D#YK+DBb$urajHy&;CEKgGyMmys#3BN1<Y5=Ank0PE9PZ?Pi&^qU*^
z18y+r#=hDLjp6vRTtv-br~4-R{rnBSX9)v#*HIQ}DOpDj_)7f_y%pQbya%<5-!~h5
zbR4FSlHg4@Z1<U!S5ZBn*C#$VXk(&SHv44mpRU5}rltj@qc0uqTR#|EvvNwi!Y{kb
zO<vR5iau4er{{$Afx9ufDS7--yKo$sDIwyn>R|@(1sUs6pO9xnObjrGY2uDO=`+Tc
zN|(1%;c?hNv>rRgMx(fzFH;=r>;;L$f4wZnx_15avLGh!312WRNn`EtD!1+SRu%an
zs<G|4xKuUy#b3QA09YUcO^gRIA!idUt9`PSHhg@j6D%tOvJ`k`6laM4MxrF~aF@>w
zGgo;b1G8f>bfcUZv2aSe7CNPEGDCs7e`qqAL((@4e=L@g1&`x&jaT_N-OODHTVB?L
zPxSjYh3bo<B0K8F5R((`v}Ya;#Ny2)TK)TlpOLEk2#+1aCA!2^z8~hU0<G3mft%4A
z$+V_k($E<WD5Wq;)S+VfqAfT}hYlZS!o`581xJqVgl^&~;~m}4fxc`rqZJCUS6d|G
zCTEIOM220MZMwo!Mh~vVG>U$!Z20`pBCHe~WAm?r+J7t$fNdhQaKZSU*vQpUDWuB?
z8ZfR0j;W_+v7~uzY{<KmC`~lK3tUJprR&osSZw+MAt#~X4Bs_M(DdG_d$>&b!1Q?W
zyK=)-K)7Eo)1aTheMjrm@8FY4$7dJLHMP6bpMy>}h$+j)rx<ZZ2kP=E0uQ%+A91|`
zwnNIIwvo{-^UzW}Hk4zOR>~Pw4gCh8R9HCUMfL-|gUE~$NkMCr2$ksPS*?w&SFgK5
zvE`}&3f`&YTk@Di?fZ89(`J<K(n_~0lGx=#yd>8wQFjY{AHl<Y&8t}P!O{!Utrer5
znd<J8-+Eq-Y~{NT;*fH6#mcBCW*lBaxsTP`b~NT0QolgR5@%g$Y5RqI+gdgKbd5L-
z?q5Ax41~jIXKDUj58*n5Yn`V-_M6cp`owg=mHhdVg=xQVR-S1c43&M*F&~diR~cI!
z#{9)Ve+Fe>t|Hbi_(fAqSO)^ga1YFV;r5QJEv@2VF<2sziQy(s984|EH1uV7Dm9m5
zLkRCjw|mEv#6sC9hSMW`0zH*+-FZ)*wAYHw`>n@sFZ<8;(LzNk!0f+$j0ei@&TbN5
zb8CsJ@cq8k%gO&<@S^E`{Aua(@SBG&7THm|&iMr=WTJIBsL6L(UD_&P&^%K)nuMMY
zCBn+5=Y?*S;_aHh&Tyn1=|%%y6@tOth~y?|D+r<tVZ*ApRola$yVXQdrt>ACSw*nJ
z$u~2de}CQZ3`I=Fp;j<j)sV;{!oTYS<8sPPZuA+)$co1?cc(7peu)sNR#<;APbI><
z4&LXQbi7REaNvgwa?9>;eNtc6JnCc?>Iq%t+2Qn&(77zweGpMdSV>W5w#5zah8<$j
zx&`<p!OE7&ZO};!ev=&Iz0r1@c?6cMu4{nZM%p>)=baZ0|B*~&0CHO~(o|D`oSk{h
z$c<DIvHgas5fc{z{2_H4&|Xh0blv&|D9?|^XXhK9NrgA2khjTFsKqHYuZ-qo!A$-`
zQ&>EmC7HIQA9kX+wm+9R6PX=qH|HISW#ACauwIwZ=o3FyDLO^jCfRTbPeBF*$N%D=
z8Y^c#b(B-Ecb0p#Z7-gQaP#p2Jg5&aUAZ#LwC;XG`NmL1foEZ}1ZAaU^Z7=hN}6mv
zbO@n!2H7mGVT{!}AO;m0Na$Nt1SMRE^dGOqQo|A@9?TC}OV*;df74uyv5QN7h!i>+
z7u+cuoy^4l>1-`VK$IsbypeCNw%J6-7M;VG=`2aNl}%W|Q~1s=cm{<Jl)$phMA!&2
zLtEdTnWg@;e-fsHT1>PX;RQSswoyVqxPN@x^h5Cw<<6ZEaLc1%PdmWo`B}lf+qFIw
zqN+)mTXB?F8sc0p!@-=zwwxsEmRq6kvM*pJ5QeirChLlhNw?rha-sSfu)*4WD|U%x
zta;IuTYjYNjyQ8G%hII){KSGXx0!tYc37?zH`6aveNDNn5iwN)=F1_0u=Z%~UQ?I!
z&c84&uIxHrMUaPmh}MP7Ei8gO={vZRmxK7$r)pW(1-hYh_Gqo=&)b}Q{2_7>f1F_V
znf-ll2X6(F_Rj``TcgBb31<7<X}70J0(`BPH)Cx6Rw|EHmp=R;Fj9_zwI6EX&d0-N
zia3MKeiWp`glLV0&5q+m!BF<gNH5xJ>Bp0n6cu8fxk7KdFYG#(^zdjF0#)k`-EEzz
z5jy@CtK8l|AJza6BXTcVTdwMrR0Ql44J;7`i}wlHp%+VoXV1Xx*3|yZ2%hAgK7&1@
z<=Ql{yeYFl6Kn4SZ3J2Z_BZ?5*ew=A4>;NX;yE@xjNFiIojvG9k$-u6*y)AF@`F=4
zlf*h2&IVj6cU$$#6>A^38zq&3Q2H4I=Ua_L<?s|2Ke-rk4-&XdEtfoIDt%5{+R>tn
z2#FT0!=p$$w=xCKIC-73?QG;AUVjAGmlur@-*Ff3x2iG#e<==y;7m5!+_)lqocENb
zS)gZf+H-LC11&aV(v4dvH>ZT#620)lULx#ZTSb2Q+QF*${*y18o7;^4vpKg&5ec$8
z;;B%Yw8y0=47)iDd;fbpr>oB*@CDXI-6K$(3y}kW@);g<YMc&9QU28j_6tk%+sBGA
zrF3EeYO3HZ@X$wXko}lra@eEetw%0$Br?HdL;l-?>m1L*&%q#Nzxw*Un|J1QCQE!k
zYkL4Sw5eyfMXsxlGy9Pqv)2&Mi0UcaKQZfG5w7ElGi%B`1bQ?vtUqu$V6Mj*(9PU<
zcx_N;em!DS?_|w;hwqge89LDUR(*bja^mGSTz@41V7(7?`Y;E!X>a;PiP=Y}cUFvM
ziaYhShKY*;4E13mncVVjRFL!0Ze@-*(ACeF$x1J0-Dpu(209Q5`BUlCX#`IYp9Ju2
z&Q(7e9A%lTbbEA2bsqvkbu`;s{M2JZbzXct=r$&@tcC~ZjUopI=$+^n7;e77w+-!K
zF>as&nzFf@x|?i_zr5-6Gh5*0lFDvu|Jel`^$QpAcSo#)DChf}lOa++yk#uyybVGe
zp*d>!#3=mHoS}*r@cd!7U;}@F1qH$I%gEflFpSIg+T#y_Scu_|SyiNfm~lCs-W8QV
ztw9`98K3c(RI+N!p}`kDlC3Mb=SoR`<?tdTfR^oCixMj+SeOBGP=B5MrTc|3v>E*{
zaYI)x!M;f{GmXI6Z@u%!hu7r$x}F*PwtsT`Hjr(g8$VfvH{aEroZDuRAkX)(NB*!k
z5VlrVGgpwm#sqgG)Nl}I8)i_70lUTMV<4~Z&f=3FvAa7GfZiprmv6JVPc*&nT|!0(
zAiodG0<p2Z0>(-SC!`U4t|2TD{F(?#(hLD|KS`HYDVDqI_DPmpGMB`P*gT<WzHctB
zEXsfM<&BFk%A<@$b09}nkes!@D!I8$9#8|tVt$3KuR<hQR;q(t?D?7U>fwnfJ`@6U
zv3l?QfOEves<*?-$|+K1%r0veWCJn~9TK`)TCsy`@k69{wEJ)f?5G5zmoNN*&{+oA
z&>-o4@;=L@XGOr^?tEp;ny(*|VfvwmrC#+~ebS34YuQ9T=|#zJ&ksTdXRqzWf(YK0
zd}SpLN(s(132ZAookmA+r+OYdWF#z~1QG7aZQW^7SDEf9=P;d$=Dk7-JLm<lW)_b`
zj~FDX)1H0;-h?hW7;6n?9n>aoi6iugoZ@fa%kg$f6lnPfw={PfoU)}FjflAv+S#8m
zBA)MGF^GeL!0&0LcVG@RG)?hxMS`?VELy_+8E<Yk_25m04i`^pSfr|9R6Q|s%=F%W
zvMMRRnV`~U7;ukUOOciti)I7YpGf6D2$i!G#UFn2pKe<z-S_*EmmLs8^J+w-^ASTS
zsc^gnc2~LF`^!7xx8T(rnOm<HLxlGV>;-@?!h%C3qrdE=Q_3j=`5!Ps5Z~l2ogimZ
zN<ou-CNa6FN;wO!88~=RgGe^Xmy&{|BJ$`PXbw)%BK#!35@W^DJSgBq-}8d?kkeMz
zeg^&at=VxM({J_=M@yk^M!xP6Q);fWTV{~S_phhZ;_g`fRGd3J<$V;093p#qqti*=
zVZm!~azxQ5ky%@_G=)**s;g<egDDnCOtx=?E@jDk?2TIndbPzqQ*i^k?Ntb@nb7jR
z6E29I8KG^-j>5+-W+=b`8HCCGqRN{TuF@84ZARpmZa0`mbN}a{@7JNPF!xLwK9Dp;
z^7GYy3EvA#>wCEdxD=B-x~Ug+>qj9*JKSrEnrz>j_N@-v1|r`PKNu2LuN@KbiT{!C
zZQXh@cpxzrGUnqsyn|+iA9O4jmVdk|i;`Z;D@CzQPs&&{5Dw5}H-L|0SNOt<x^4Xb
zvc+YL+NS?9!NfAcyaofkj-N%HSlFUpGZdeeSz5Qjog+G~I3p6zY8qq0AoTdt>&$?C
z@@)QQqQ=kHq&{YFak*V$ha~0-kSLIHsjXRe+E!}jHG6DE%gkU*WQwTuIG-(JKqV)K
z0WtDN!aF%Yb`j}Asb;@};QY5j{jO1RY0%u{4;LRy&JDqDR(zM}kA@Tw0%m=#{4Ys7
zau%z?<YMWxQzR#94e#zMLhjxNUfSCbBD6FT`h;$@`JOZC3-h<LyVv(ui4NOOqzXTF
zGo2K%UK~35Yi^9cIW^Jm<*Sf{K=?b=%%N!-Un7W2led}%M4{U~ENHV8ciOIJ+nQ62
za=!SE#y#8JtV2^b%s=Q9zmhc!sVw=Lu5Agvz@xvh2Z0KaK2vfWyg4qG;(wRki`KgM
zdf&O;3>{UzvDH^JA{pH4N8hkheHh&8q_Jid;!0#>d*{TXTG&G^%4@Qz0!zXUNc%sA
z$l&)C6boR2z?)!7di`V;dxUo--I$0GLkTOOa8yQ>iOL7Bx@MCz$(^_S(PX}v@R87S
zU$K!Ob0HTx@g8;Gi_ZCO@fsZS-awF5I`b8z)2a(pA*5fJ2ne42A<O>#wUAJNZr6TA
zy$(6%yaJ9`9$@osdiq|x0R%&Sc5C2tcfN7qAtGCrt(Ap6UeM~-6;aCI$Y^B`9z^S8
z2c$%NA#%G1Am}o9GI%_kbxEx+PYhf5(IB3Ds>l6d4yTgUx`_E9{PHd!CB4M^5!0W`
zhQGM@mapOcC*M`KPs}GBr>qXAEXDj7j3P8wFMl@$>O4>-*wpR)0en;@JvCWz2}!cy
zH`F36d@o)jg2Tj@#u8)>tT0$!X|Oloa5f4n##a}sC-2?6-E=n*W|<2}SdfRNY{wfY
z5=1BoN4S`CFCM6=AS7A1+M%Nsy5}6_h?)6!FdT)7$~uGESOg|0QY9lLJ3<mdzTqEM
zUJWqk1zg;V3IuN70@P{bd{3&%3q{_Za*Bjo3Ja3EWZq*2k)~;75M}d|far(J(6(W!
z$HdXl-RBs~j*-q!o;!wd&s7fVy4!lYu*gPyqE>#MT7s2KCmA~p)p{?C!w?>1N=MD?
z6hYENnX`6=;zu*pLEi?s@C|g9&q%mwUb|bovi1BkG*z=WAem3_Os=AP*U;+mZ~gpN
zQImLk5!x#-5Fwl2(IMJ$8%7yUB10|;>p>k1v-{=N9uceb-B~}}k}HIpB01tnZ^Y+2
z9U-DAn7L84EYN<KP}9Ve8fuRG0VT4tChwl{PVR#|-q@Z#JTwjOT(hxJAOAPAIr&Om
zk)*q7eG+jNsJ$JKy(Yub(8&7|hv?Y#xT5zN+X}I1K|MW3`XUejf}6Cr<r(~p)q=iH
zw^4ms6~p#*_=bDc=ozYA1LXk-`dj|iH2j6S+u4A%hyrbDqzL0c;Vnkb9;{u@XnEd~
zdR>|B%_9lcao&ty4Lis6FXOwuC#B6#sU|mh`(#BRuP%rv)|al{Mw79YnE*VdQRplb
zPmYltpRdkX6REQl7cw{#>56RXUk-oURjhoK_Nn{aAQlQGtY{(G7XnjIg4``Y5C%2c
zm}cojM1EQ=2UpSB<T2E&yC-lqQ~ele)PTgVJRLK{UD3ezZ6N(?a*@qt9!ppbe*k)}
z$6^Qoxqj(Q8UGwfu_VzF%l24Uh-16@XcRFpDyeGuIa-3Gx5%!=cJ&h8!81}sCk8O_
z!(f?l*9^=k`*D<cu<N>9nhmxczDu5<MPc(eB0~g+O=J?6yW_=h5P1-z!*I}Bf+B~k
zNX9<Qj5+&ebA6vWpkp-~ufC`GYE$oT*DiowK%`4Wg*5g9Z~~VdMu|<a3#9Ad&5A{E
zSGFt-bW4iXNx1v7*xu01MJM`EoFuSQ8JD-&dKFod32p=PT{&EKR|WpILxYn3FwR_&
zJAob^-DcGEs|#IH!zQoA<xq!aJm6HJfD9K=u6`7Y<})bfrQ>3f53`)m0g)X!wgSE_
zV%7A0=#ce_(6hg@eM=~e^K9Ggms)vzrsW-zoc=dB`0deQoJVF}H8OGFJgG;0ix`6y
z`<q%C)PC7HR@udwsJ5m@_?MsvA@#2_JSWs}G{8@#*tqDgW;O5sr!j0|4z-Fn8$k0g
zb}}|H(ZZO){8mw8qaz6P#n(2bh*j2q=9Tvhg^|7cOy(wOI&JB<j@!ZV{NR*+Bwy9E
z+6?Q*+h9s4Myux1Tq(t)jAp*R*L0oo@eggUv>IJ~q^+>uE*Vpy+WpW&z4IEFaDcJk
z2Ad7lQx6H<pW{Sb<0@BpMsSynFz9X%lD3IIU6&TPhGTM<=o$r@%aM-{kF94c1pOyH
z_f~{k2BK4${3kXtdmDDDe{)%C41t64RKt$-1vTX`aP2P;QqUqqmA&TlE!`$f5TSLJ
zhR&ip-IB*GB3`4_*3NOB6v-Gb4X1huE(Yf8zg&)MWr!g@^&vBKu%mn^dAS|P2_57(
zaE|!tgkSC?p@jq1soaiNy65&o^=9&jy)HY`$F)uRB{A>|t*{e49HPI1E;m)IbE8Ql
z6L!Lo9qY0xQE&;r;|$#w#i?v6Wto~N-$sq{rMKgjFg_O_<fl`r2WAT5F|=dTTh?R~
znCb8}K^D_Y>?*ssxAqsct^L`)b{8x05RgR^t<P`%O+*G<+s+!=`k+Y`i1uJf*eUaI
z`cf)#>giC(v9Q~4sn|KhQpS-h!o+>+F2Cddt<M(BV``X@TtW58nT4pnX&UQd$-iad
zErPA+zd+HsQDjlbDr@f7(<$)Tc4WWVvh6w5XcyGD=38%@Gp;%nro4P$e_tKq*@2-p
zRvr%uif0QjZFZk|@!J62{3RyZGSFGlk`yx8>`9l!P;6X3pBDXyc9IcX9F8nEAkU``
zua4f6L}~?T{b`^3^y8`0CVB;FFY!u5(R;tWFG@~|f)T<~+!ONwc0@3qLn@ZUYKTu+
z^t%D^S~*5VZkoJuG#IHl-M@czWH!M{9jw%ZhvEU4Adwaz3`c5}F!DtcR|E|s4EQiX
z{r-G*FJXi=mK_jS8{(nLnP!Uf34YM#q3Hil0cGa>^F$}|*Ix}A3b5&3O-()(w>K>h
z>e*Pl#S*isYbGpjsZB|n_~7EzEKDbVDb^|FTgxQ5ntjJKH?<(9_n0*QfBqc1qgg-z
zUdG~c5;~Y-fZ~1}%6Df!;qM1nUyR1SC5d^Lo0de_XjZjv!Q4W=1yE{9(<rZ_#eB=&
z<v-v=A0>r3-ReOpj4RE2xVSArKO6Zi!((Jv!a6x!80PLimoA=6-^(`*4MMP`oT>gj
zqCjJ-_A+^IWb4GTD?4k}(R_KACx!s1fZvdCB&K`>F%L2U3IR*Bz~GieX)GiZInMnD
z<EokDv6DyksQvxk>*E&LM6(o24!zGg#?}u0S{AGimVAZ2YQ{dH7azItiX?irZF4aL
z1Wkdf(erzwm+;5mfA1o@DS)!f(2<kK-W;tn`DK+(@9G3&{sQqs7Gon?Y8UW=CzS#f
z0<cItc~6MnAHAO?>ShBx&lx;lsRH7bZPtL)O`rWGiA`taK;q@$<@?{=vC&c|LSNfU
z(ypQrqkT(O8u15o*0ALX`bu9)c+s^W;YELZ)BiVh?Ax>9qQykp5Fw$cQN2W`Nn{We
zdc7rkT`Kd3l1T!CD>=-#yH4ji<UR}E!C=;Q9z0#4ilQ_m5_TM6TmDfbAi`T#qA;(%
zcu3M6*zkVBSrTK#LgY=)#Y#T#sAe|h(gmjSfT;-k{IyD@uinzlZPu#RAUDc*=utIi
zq0<dB+*ch#L3I`aTO|q+%YX5PV2wy%XCg(!&n0pLZ@mFTM?=kzed`kyt*>mDq1O!r
z$<VbJa3`I_m>X%+F0=u$+Yjaa_O%29s$uH9)zGoE^E)<k&yyR3Kj(-Hn})x>bIjgj
zghF^RsBEG5G#)N98Hy%}M@M{<ulLxmJ%#lO*PWznI42~&6U~&DqC>o!Nk3)f>UD~L
zhSKxJKga>VO%1t`5bw51oBlW1!G-oJy9(bgj^VRp#XpE~^r~tc65etdXtoP-jQ8;F
zDAz7hao`YFwLXPrg<cv|AX}23VoO<lSG$f6i$d4$f8W*GjCpPls%j>Ek=$`_f3{D=
z?nm9LU&@|u^{hSX8m?rP?;kVdab%f_wC1Z~1Y&`w-AL4w)7G_3&?VbS9@n)UErWt1
zA_%f|Z^QFFCY2bhz4ppWmKymhI~#1R5_Wm`*GcD>NiMA)n#;SUS{2!T|JPZv3KiFV
zufWlVmZ}G<rq@co@Z0&P(ee+vm(*dRsOC8IG^hMkuQ)mMs$WqlrrG^lRjWSirO!ZR
zy9jQuVFP#WiIO=_4`EdfWMbgYQTm<8w_i;j`;p}M4J0j6E3r>V<c1D(?`B6YBZv5b
z@zBIDs>G3w<Dk0rLe2$D{X=Jb$&&Wo9w@^^!+F~Zxg)LpF_Jxg&pkd%`)}GS@g|;^
z@l;RmX3@HC(s6$IF4KbjZiya@(QH84b3;~D)=_Ah^saP@B@YEwp}NgYa5c4d*Hr=;
zvNK4CB`1nOLgquD0pH^cP3o8fs2gk5)RsV0WjKcZUJUggCpVRvs}&TOFaEj+LjVXN
zkIQt1OD;57VEh;;`NP{llXX_p-=9eDui^x;0`a|Hzj$C6iKF^AWGp|Qel6gKZi$Jq
z1<<=&A`4b5!2ngv=^HLv2Sft*we6mpe~cDXHI0S(ac=|pbVG2kCKh&UC4UM!d>^!I
zs~WA*eS7a}(BR$X`OPx<5T5JX38(3{x^>-bF&)P5`i3<u+!l9c^<^%LUe-}If2rqp
z*A(rGuCb;TuH@@Bz~BBNVhV7BGHK9JbFavXseUW^1?3U<eE#<KLiKi$?3#tOG9fYo
zl^y4crb>DS@_HY%j!!}T>ChV#ed8{vQmSjTiXVL<IAoI(sS0}Bo<{I4Z}8jnkYdo>
z?{DE;$*In7F}x>xX68pqVhA>suf_u0LQb0u83YFQ$@(I8Bx&3*><VayOfN&`wH4Nl
zVKBI2*=<f3n@*&VV@k`<z-yO$4jqRA$@`%~_W^C58L%=WW~1WV2V~FCthq?2wS-r8
zXuQKmVScdI$o_@z!Ib!$<NX<zN{GQ1o+ZE8V28<<k{iy!@$G{xNQCa$VN{E-ORPrX
zU;D7)s6-mIFLW9;9Rr|V@<Z^b)F{;Lo)L1{&h@ef+x1R>v(sildl(@5Oz1K53wJ)W
zv1}O2U4$<4VgJA)nFKmLJWQ;hTZ;DgWpaSTXyHQW$C=Nzu{do;W{8sM*a9TW0x8tb
z0kLqjl)m@IhVw@xf=X~1y7pHw_OuRk6X;CoN{Ak0e{XY;CLot=n@bBPG<enk3%Eqy
zDF;{4EL5-odY7Vf3Pnqn1=3@mcV!O@CdN@O&lvuyF^WI_4vEYqty6!(DdcVn$v;<D
zEgUR{no{y_IBJT$wg{)i-eO4q7<`vKf@HndWR&;N$JQ6!B8F{6$vw=BFYfjZseeFz
z%3frVa}y;e0}>%68u#aiQgdK-H-j8Meot5^WG7CpN5ezV2mL^p>%<hnB%$r8>N9o~
zV<N@YXZSQ!zl5fOtuY?gGxkHsL`RqEmzN>JC9imbM~4L^!X!NYW5!3Dqg_&!A2^QW
zDp^^B3w4XcKw__q1pRWA_Hz+;BB6~w+kvSo?yER91Dc428}jqTuVG45d@BzAX-sHu
zAP@^1O&&CASLACJYQll;v5u_XNgW=G{LUqI+`k-mYWuAMaYFB8e&QEv?)4KEjsmWP
zYxyAZkKbeg?C~v1Tj7qo%2t(D7^Lm~<ih2T)!uQH+J^Q>J26U|1FftumGfoQPzlTf
z3S&vUE?w6>XEVz(mckd?@nR{fbf0B`m+vHRN0x~xer<C21PA$jA^vI=fe7g7bFn2c
z;mO@R&dP=&f@fEoW33-))?-oyu#lq<Gjn<?`5xU}<6bQtxMVn*o#%Y?&*W*>P+-hJ
z=5b!5nNygaH*3=IkyFW*Z7KOI%?*FS^CZU`SaIN0%wu9CMwCR`mgOjA(V$1W(%-bi
z5U@x6W$)`a^Xt3!yM6ZWf*miu_flmpjOpd7;q9}SPP#g7r@;z2s5o;d>>US9RJu>?
zJk28v4Oipwi1crvs}Hjc_NpTzl6pSCL!JkvqNqt!VjbIrV{2adkzoU=-NlN{Thz%>
z&`=H-Ru1Ko|IM9+84`ErB@=b}VsLChti*_HfhK8ZlVhz!le!HN-0Uv2wMw}LFm9<z
z|LvP&9ShV`2H%^?RJvuuN{&q1s71In=E8!iMAcB$Vi^?L3Fun{`yJsTA-x`WQL$ng
z2@DuIuvqQRYlQ(~M#c2n-K(bhI}l@M6i+4}cDK!Ea(VT8=W>%Z5*Y0{x9RORlg0(I
zO_m>bmM+}t(G5ZKv{%ct^0G@cmu@|&5Y5X|T3|5jIU2j0=f}%$EYYo2{)qf9SkgGC
zvO=!#1r6<z{~mf(HI}`&R=^aPXRfx8Xr2TCke`p6+|Zec$8dj6Z;h@iP8@!dfJkJ;
zMIVkNdQXFCT4?<!64PyQP%>VV3tc^f{|imIK;Bx5@}y@{N(d)vBQ)-aa0QXGL6<t-
zF;3W`JKzerKorc<VcTdKP{H<;Jgxha%KYUuxq=;6PLz0z9BM7Lcm?yJ2`xMV4eV49
zF!}yFw;fF)vzA%fEDy?HM~kREE+74uLBN$&kJ6Iuq!Zs~QNB-4a1o*wxzU*%{a=s2
zE455XAjA4R4dHy_)=a@vWepVS8BOOVb~v>Y#Wk6Rp|3kfe~{1w-6h#(z9$|e_|8?j
z@-~2yU712IOQK6*SMEQRbLVpq`2HZW1SU(*JU;qUV(Ph0!Ue*jA+?#VC~i)~3<<rg
zH$BjW6yvSha6b#vR1h6F8maekiDXS~Bkb>Rw8W(peiFb?n@3u1<KEm17&%2B;W8d5
z`x!qMj8a7Fn-f_;kv=^Ceq3eVq%MGRrC$WVP|;PQT0d^M>M;7_8I_{W;>uOtfx)a+
zUlF*a!a5E4rJEkAa>{-hjtU6^=nveLl-fs^r%*Zhux%I?o&)`*jWr+0-n#3<M5D3@
zuT_TPLOpfPoSU{Sot*&1MUI0sSuw^1mX&J*A+67Bt1|;;X5uOLmAWDF&zrqw1HuiY
zGLqJ>VLDvP9)uu8EXXwtW)I2!mmq;}Gqq{uRw<qe<h5!(_EY=78d*IroJM!_uOy=c
z63e_(*2xvMe8y3zLm5Pl1}blH6N7O~(sI7;7REsDkXgmgdbrBSknJ>OG&)T$Me7rS
zLJw+R-|P)*+|9v_`muE{3vW3eH9K?4we1k4RKNP0>aI8??-h%cUey&EK;piL>9;ek
zq+$<TGJ)+yaKg6(xk<5?0z^qD{y@nBjf?ooBYoJ9W)Dz*pk-^2ZI6L=kSV1_qNZ?h
zr~vgfE4qp|)zw_`u*VK+n#Q%XVyRyUn!aX6<`;fDB5wPRQh9@OfzaMCZwa%SXh{kX
zOP|IwT=6~p+EFMa%X%#g$PiGXXdms~%aE}EeGj%0?ryz0L{b_piZ##~^@y{%{O1?w
z>TBCW)M9q^S1hARmdZ(xtHfU$<SuUGG*sJpcIn%(ct&0o&OT_@zXSEs--4rZ-7K73
zzqTPxAzy<9d7e$lVRderkZy4Ky0~$*Vy{%(2Yo6q;50I8toI{qfQ}Sb%fC*wqW@&y
zvy{V)=ZVpnA1Z$xs-#*_4cw8Dyjg1^$x+~%&4=!Tx=2cW!0Yj;DV}$EJ_g$#OA+5H
z;Qt~adlq}T#l-Y!%<?<C-UShS5%_ko#Z(t3(7@{geVBc!e9*VWm7N<@r<$j>xf6~}
zX~4Nw7xwL5gH~!W6ad@3Ab^$o$7LksLQ@BN)J~f;8~rF&<}}BLZkffcuh<6$Q~h1-
z&US5EnDeLYd7$!PJPCRYpHeM?eh@xoF!O9(e}ie6oc09Kbgr`q+fvACJglH9=}h}|
z`=YCxzds&2gjzvBRG_L9lw0?v+0$q~hxsyu8O&?QeyJG3KY*8F32wITv~<6+bY#wT
zvWEbXpV`e+`<<ZFVIoLnPT?e`fRoa!-tsL;Z|)k^mWp)GyTWgjWrP7D@oK;66|8K`
z&VWQBn*q}+(PuM2)gYG>@)e_%uG9`*1Tj-1o6BwHNZ{bo<{^Tk)vkO7hsh5XTDgJm
zZTd`knbLMI0??Z;C&qap5BGl1+2JOuJbh-AK1;Y>9f*l+4HDq=-@QLzU7*ymE$Cr2
zh<DHzx^1bugUaoM^B-R4v{r^LM4Za7ec8`cf;`$#ljlI^>11U|hdv3fpXt<r^#|_9
z#Yxx~%)27%${SVe5b|)m^fB#u!p}^5I^C3ESQ_fyfqQ;Jhvadh0!kMBB3e&FTs5I2
z9($WM;xL*zvtI)=NwU#-?Zzj8XxlrSbU&K!0&F^u!IjOQW>BBVCYJ@e$@mYCT0Ju8
z(|%KI37j9gM5@=RPIrGR<1{xiDV?ib!j&xVYg97?j+Y{+wm^q1$$qwPN&Kq$P5v9%
z_9(+{1)vj?Ig+}qROT<AYJ=7c!5A86*c}plyZ|+Ghrk650=Wgb8#=G{^}R8#qLX@H
zc<LsF;i6=-K_%c}>ICB<y&vU=Zql!~<}?qz1|16_nq5=6o5FVl{2&HUqPsu5e7011
z4qTf4ZBNHd!0{KThKK!Z$grG?cbs(pg4fN-0gaO~9AT~?oPIhm;Uv-=NJw6I$(P00
zvzkgF7ws9ikC#5778J%UHiWp7Qz(8m^u3&Os(owEigKfhi>5k{u(@j#sY__x?5j(&
zx35P1F4NUVBTwq>Nur~RtGA;=>B`$9Da(>z%bK9x(_=l|qF_7cNvh>K^k)iQZrC^l
z;5>NerCCKIiQ#mRH24zO>EXXSQAN^yF$KBD_bGwT1ah_U9hbh<x}tVbC*~VLM*@q-
zVNIM=y3)ty`@@TIO0xP{L_Es`77|K6#gN;2{>yZtv-Ym1il*G`e>Ity40j(xRy~ZY
zCAYIov&dtyHcf(ciL**55#slh`T`6f&!u4e@UG!Th{t`L%k>}IJfI6DFw9njnI{j$
zPLxV;jTW!5U|pozOOR)JB|%sy`V-O_1mY(o6Q)!a;ZMjR_7H4%m|=#myB)L<u)Uk_
zUyqZ`to~6VEL&p)DUc_i*4mPf`~PaW7Rk#aJwmp&Rbz;TH3N0~Q95lOf&j6fIoSm;
z39kYSO~{jf4yHu@I^}A?(BYV4GBpKcgko3?Oo@CvQk!Fv8<Ln~etpCoGsJMA_KGD+
z?59WKL|$Key80gU(ZH`)IOX=)_ND!{)tp`P?q%*PIY#46M~Hi98Mj_a-(<!+tGw9x
z&VPSzziPKClvp!F6ec3~CxCX-<fRY4RAtQ3bLp?L<5mJvBQqm1(kNv7SY7rlJlI}4
zjEb+s<~pXFN_cAH!Ju)iTO3}DwVtaealPukHg7*MQcO9TORU66NosOULXOsD(V)E!
zu{5s<Xm3+d5CwkS!8m$Prb!=TiE>SAsA8@hYw_mMa0@Aslm$4f5>N%xU%vygrYeO+
zJSs-M@MICsLMS6%Tfk!<m+4zSY4I4?n(qDbA4{dce_SwQ;WW+?X=o=+u!6w|32QBd
z)m`2qT3BsP;N|r(HQ~(n->Y)T$Vwf7?EYJaqp-)dMC)U(TJZLciAUH~Csg<OW<$JQ
zFlGKY=R~(-^DUtV=@jZ3n9Y7~@y=eOt&451`>J*HBEjr7KI>xm-I#NTy1hl3C@L5`
z5Xa@-t+*aMa`dt$l9}hW{~MKgX%{Xh%|zC+swQ|}L6i$3B-O16<_jQaE|*$FKr2oL
zr8&SF{CA@x@km7z@5!N(QKq5gT&u)3l2f6uHb>CLq^5V-fS{@b_b@8O4t*GyvML5j
zZMXbXwFw?b9SRpvYq}A<xnz7hnReC&IPAgwh`hjv@;QGxJ;z_}l0lP~PNeyw1>$q{
z{i5`_=)lgMhB={pU|0bO!DhVhN=v5yq~x3WBP@DKND7y@7}P63Dm+*U#JXic-Sku8
zGN)c(N{HCd(MfQ~LlOs&6!SrbTktm559W3bk2y+SU()_<NU0cJeSoKe%Jo?QoGg4w
z9ET#`)1p5oaraxa(zRvJb%k*s=rV070}TtgWEf(TDs1Z*i-q&(a$g$G$)GR-0^II-
zU?9c=939pmyWw0qg6+L;1-bjem41l)@I85!efpuj`I?aUgR!JdmF-J1**j@~fA9~!
z)t*v;28D@iBti}7;-A2#fs{FYpg~GOaeSkIxSzrrW7AuW-<HETODpucI$y%3Wkk8u
zS}la`h4STees}D<P38G<>pQ{e5VB{+<rM$$t#I-Dx!&X0%i(}abAu~<S>=;w+N2hW
zxRpOc@WGS-3IAbU+cAGh6I@s5UbX%q?>4vTiTS0H$eQ8vQ=^}Yznk<7P&Zz7e|~;H
z8ArUx*-Kk2nLG*PJl8S^(xh?Xcq`z|FW9WGkdLyxaE_b~vAtNd?rRqf450$-y_sbR
zfMbmG+oC6Hfue7<CaKbqQm(6lW;uSc`D3|27TK=wSa147r<s!MJL0?JCb_SAw37Bl
znb;*By1yDU_}(0qbrlqFd%us^buFN3-8gHfd%~1JY`S`{ffx{gBN-T|Bxp@7$-Wy9
zfMH6N1lJgo2B%&Xe#vl5E+WDwC5PLSlu(0I(6;)=QfMT4#PjZ}{u^o@>oXxRnvJvi
ztM-28z0<CI%{H_EUcin?DKhCDT;Zg-Ubn>6JCrZQZV<AjnF#=I1y@`Me<3Y3l*t9h
z51xhKLw3x#EQsE7GM*@k2152NBqVcGy&i3!Gg;!Ao$9O(yG!5EIM;-=1JlPTaGq2g
z2j$d;C@*8|(ptY|#hYL6)7$>l1&5d2>EbQ+S_2KwdEa_Z6Je;6l04;bV6-3W&BSVh
z>pgcsO01gv`R6ZSmw3SL9u?A<*-jIZCE~!TlVJ~MB$Zd{YsWtr2m32|HA|lb*q8og
zU$kX=dj=CCwi*<&EQy_6{byMne<zM)5YoQhBg!qN6{~0WdNE~U3U8HQJ=HUr;XjHa
zDaxswa4NULyY4q4i?joBb0N>~r63d_6vWD4y03%cC_(_rsu9Z0Nm(!pHu#t?4~r~V
zN85##-E6?qn@Qh3EDo=$&@1Cmj|tW&lH=`63!TZP5~!IZ!y0!{l0@+X^jPfRjcUZJ
z;-8DCyu!D)vDQ~80!&!lh*jJgrY<%w!rHz;X_X3$qboUq3Bzed^utTHXZI1fw607^
z(&JX7a4n!1YiZG#31n+6j|c3Nlh?>5HhgiprTgm?Y8=@zSSmmhM9|ygb<?-mS@>F+
zP4Fd7{#O82LJdw0zBeZH%i;)r#F(ghshp@y#g6|M2{V$KMPyY~x)FyOG>^#j7Z}=d
z(an`|_~;PeRZw1~q#<oxS5VDDwYPIEaG`;IXb3=?nGVNK8W1@_NcMhtYoct@WSI;i
zx1%5;P-@!_g__7a#XRYF3Sq`9xBgtN!oNVT_!AEsyFfi~1D12L!3AUdi|NK~lWgq-
zy-+n>5!^<6ThDPMU$(sy16>8ONH!f#9nZP)=rNKhTPPfvU_hh$#v74@a$1~~L;a=k
z=p6_#Zfsp3RM%P3g<JVC??DYYFDoqXPLuAgstA9tFXrahyuF3xx^r$0N$jq7_wUL7
z<T@i%TAcv@IwXqtF8WL6`etIs`gfKM65JZO{?HW<DYZVTc7ka+Kd0Cj;WaO+p96D8
zqNCc|d*V>@PP&6y0$|RJi|JCiT&|w1)&<CxOwcH2$9Xwe&r^w63%|Ph_=#%6S;0~B
zEX;gp@IQTAZfA9yze$`XoM#cTz4V^lQGTXPE5f~RWGjHh2N#^TeM%})1kc%<D2(5!
z67Mt-V{S!-uDXp23Skm{-vgb_94lW!qJ^`drVLV9!Ni-%GJF}#`qe8|6L5&6QwV@(
zKG3_b*MpEc9C@`h-G-gD^IId9%R}caLH}v2c+CNn{*Vz87vpW#3~at|t}!T!L3$ih
zSTLprm53~bHG6F1W`QsHqpW-+N3ercftpL|u=p0u+b?FlT#}<E9igQ$=G&X?OURXZ
zIBjh$6A_l|N3>s17ZPP>ZiCv8WdlXz=AD%0D|-pzUBQiOSfL>Va<j!9J}OoOghXyG
zHeyHZIg6Jrn5_};2UdM8Lszo>ECL7g>(q3#OE~5l$N5f5D3(&rd|fi6tY>CoHE()5
zs0a4Lhe`IG4}pKW`pUb#yx@}7Di?L8)w+7^ye;~1pTBAU$Z4D%X|%>(W(5dfqDRG0
z?!DDB9}3=w!efyEgaww|D5%XQtub0SG!0h)5tMi1(v}qKCTBY21rb<SqZjwEisZIX
zuRRd=P<rwI%8xFmi*MVz7mti2KKCSJiK$9I)v~y`qNAu0&V^Uh$hrhFdMct467|%B
zz7km035>X@ecK7=RA<68Fp%G99y3kR<vH+@=Jb7bvc8W(@NYDA_t1``5nCvI=z@fG
zVdc=>DgEz$HbbxC&ba855rt$W2Yr>p@+wSxxN|OcZYpPuY+Rb;1#3ciT?)Qevhu?p
z!|1u^K1r<!qZ^?Fs-o8}-=aQ{UquCk=7>-;7QOrx#J<ddT@zJyfe{8?X^~sqCIooR
zA*}rMntC8<SOy8=Tgh@wTO&xn`M6HeyoWixMk9G)?NrpV8eUg2UhXjLD@LrlgAb0X
ze$OxlOQP`+ovrKPzO_ipRxLV8L5)|EYdy7|d(9}lMSVESwpok{JyW1fnVRpG0s~xz
z%DB|WaH!-jzkrJkK*1*lSRZj$J44L1zxZjOQ7Y&<>ei}!7tdjZ{QCj{Z@ZPYUKs61
z6?-fTyHR3it&Z|BY!b>ZZg8)Y;k*<+w^_y`-}P#gdi6mC0Zs>h3;gTswk%E=Xip^m
z<==8+qt^Dk*u6eqKaA_rwJy!d!QELH^`}5@Plvz@9C&&>a4WKK_5yuEknd5Ct3RJB
z5fJ}}TdIVq7ev*rE6v)Vjiqs`Mm?^h(397-RsCPWQK}aR4{szY9rQ%dH6BRiajujX
z_51(Eu^l`ZwXSC};=;_*jnyAt%%YYOpC#@xYIF0SFD&;I(Mc~oB4{1`m#6Y@j$r~#
zz1}=<j_3evtW69YENgTVCH?fTgDs8wDlgm@=#^CWUbL8!+Xoj+=y?W%Ql2Zkdt|yn
zB}v3**wrZXwq9~P5f7FXGu4xyk(EZtRNrV*)ig_Gwo6m&NK41N@HLrTMO{EXTNK+l
z&Zbj6y|Sk^aK5nDd{Sk_Uz3)D%#cAd)WNH(Mgag^G3D|Wm6dv2?69m#XexX1IdAoa
zMkbgUJ!=uf&TTB$1tgpHs!iUnUbDiTU!UbF`+lSIBcc;h-k-`I*SV{1>49#?+8!q1
zJ&JdDAgGeiZ#BXH=$eK9*s$jRy9EbHOR&S4+3YskUh@EdeOOm*(cKPemCD*7*rmN;
z8qxx0W}O)$U*q-B^FBi6JTlt}Dq8m;17#JLTaebyOO}ZEROPs8{Z=#x^twz)>MVqX
z=RI&yhxa&i3v1$Ejn*>YI=H*nodv(t>^J>>CC$~>ag|9EwqC_wnOl2nG(XtWp4O04
z&qLK!cDyE?seJKt5Sd+*P*bQ_ci?hTG2;QSa&u$t5E>gsCW?3y;!05l2S$ddp_T}Z
zGftmACv}gDo9l(^-E;l1qE|8LJe38XbbAInl^Vj<F240Z>4B-aj)I3ZUNG2ZSjVxk
z<?zaXd%_{ihVo+g%edj2t0=_j9EegsrI?(`&6!(QnU-=jb(BW&3J?6F>}m<T=&=D-
zHYO5?=F7~0QNP6D04%$_5Kw<1b#SGDnrkx`c|&~8vz#!`R_5k<($x|9UM40Vue^-G
zKmIkTsFe%(FxJocT{WE{^*@UIH)P!M55r)*&B2WM(+K{3rV=oq$EXjJzDO7Er|r)x
zbM%P9@qKwd(s4at?lOELceT}!h#&wYsiZ?2Klk$3eh^+n=^AP?8+}KN&c1~?ar{uS
zOX7aLzbu5n@<08R`PM*#6W#bnQ~GezL##j2R#^S5KS9CN6k+u>{#yMqO)tJ2(w+ku
zC&PxJ;Fv(v?gQm)`YAQf>7J5X$K7t&Dou0si>jk)bX6q3B)oPZYf-^^02u{L*G5iH
zEF361Dt(2y!|ZBfI??E+gzRbYc-%MT#sN8Dbbs`Z3o_pR1Iw1%Y-$;58B0kEhilGO
z(Sq^WuDUsei*oN(Ti2cLza>H?hJO_ZyDsewGYsY5(<Yyz!HAM5J;)<<L(E<%uRhn_
zFpq{3OgDQ~OcZVV7zS^iOn`wN5}~WzYP>g)+y$2|UVJb(CC&fV*q*L3KK(0YP$h^#
zM-r7VPFFAcMGVaj(`F|o9!BCWgnoLAE7O!%%lAZ{Bj|;o_bh~k&g=A6WKZ(7*dwG<
zk%1WpzjO%DP-*=k<siA_C3h3x_f1`QRg)6!x#qUO!kuRbnxH}Y4P4I?u~^HVAe}x5
zU)hs3E7N1mn8a8mIu`eyKIe43*1T(-wcx^sd5qJ(B*q5+F@pRhd(|Ip-l5X719PlE
zG~%(~wOpN@b1QaiNEU-NY+ZD_95DNBJtz8ouKjl>L?ZQkKX~BZ_F?6ntQ27j)Y#PU
zQwt3xKf!w1mJ98=6hKhh{G|fDpcChS-fiQLnif1NmBerQ2)#4S)^0cq>{NzcQynL=
z8#+Ic*6v2c_Eer)JL+?|<5c7neaR~aSXFZf845SLb(MWm<WY_U>X9E(Pv=L|@j~Ar
z`5p;nsw}%mjhS5U1AhtO4zCt(nc(3Trwqn0Zq31=!lL2`3Iauc)YlqgU<aVT0xRjp
zZYvq;H=4}D-`)M0nC-sVqw(1-6(J|F-%OGR0TPfogW;@AHG9yFcEkxgkodZC9yDiQ
z%8<&C`x(V*_I2yv84zC~r$XuReHx<SzmQv0yZYe=9|Zdq8O4V7H1&uF8C;Od`3_tK
zC=b&B;VFHA^@*^Wuv$O6l*)GxE=s3nk@0Qr$m{T7)1iZdxxP6zGjk)?P=%h;Yja|c
z?r8;%xSNEiV~pHb%9n~4_VS8LnQgh+%s&$C+OBpR;XiK!$&9K>HQk}lzn0H;yIDru
z&}bDef*<+<m?u5E?R!tRAF5J_yf4YaF)y6}7qwbZ`$^On4K{kkkWY!u%ZaYBFoTbM
zbZmAJVc1hUuc+GE_)4<YN-a~o^0<Wg6(+w5gz&L}AzxbDOyfz??E0~{C-@Pbmkmks
z+0Pk<L=8F(p*vKyYM?u-yH>bbz4(>egU<ttktIQA&ui!PMp(Rjk4xUN+07W@%S8k*
z3xII=*+pn9u6oK}$uDdNdw_G`70mZGH&Ekf&_-nM(9evr#;hz!$IY?Qj9@3ANTu)7
z>+oZT)Rk!`TpsL>;-s1<4*_TTQT|<8kahqrqr*sY?BQ82toI(g<MRJ-^-j^5bzKu^
zY}@YGwr#Ux+r|^ywr$(CZ6_Ud?4(cL?>{%^jCry5zT9gp%&J+dCJqsHBMFK58P(sx
z0!6a%4$m&Lf{pMg+c-3r19baOhY*0hF=}JL#KQg1RbS1e=-RMv`y(le*}s7+f%LC5
zJ3qVJFLyn9Ox%;@dplif@4hMBH(Q)-jvR7*c1&5YAoY6w#BC-girkVszAYR`5Pbxh
z+0;;${#=(|vtXxkKc@Vt`Y0+^-7<$JX8z_rB~HU<m;%SF#cm(9VHeEWTT=ji;+Qrr
zn1#441U!a;Kkc?y;xo8xte_5Hyr^E&q|->4p(%_a*)o9!&5k5sztRPOQyo*)xr{E!
zwOdcpC3KzGIqtsDoG>@+vZ%-L8GTb;2B!}A=_K1t|8#XwSwv4Eo(Td80f|k@-VQOx
z^A?cB5rl!n5@gWFB$OXJ%Y6bIpuph4Vz%Ij1wMh9c-Wa9oK@z#=>@VufqB`1qZ7(4
zQ2$EMn}1%D*j#l?;g5Cab{(?J{-LHtdtk@JjT&ILNPweDH1^paPrR6i;$p5t$03d)
z)km&Uun5LOp9B4`rKm7EB&+agvi!L_>~k3GM^P-Z3lZ9TgqtF6MJK?@xa`xetX40e
zv_g~DdYh5n4-dho?H%TKnu~yMxj3vf%<0ooVVIQ07DEjO)u9AqITG#c6v@lF!%^){
z5>ySHZ}4}#9pSN^#diop?qB3l9<WrJl~7ekLrj_Vqlj;gsK}_|$0%t76l3=YOKG(~
zrARTu{%y(CN*;h_ge$;FSkFLbyvKO+Al8?y&A(Q$kO8=GQFj6&!42zcuqUlpT0v;%
z?eJ1<QMx?B&gzwg>z#k&_{$%<RzV;#Jdl{%AEhd0jpTCju%z!*fI`N?-lx!x!T#mr
znnjc$f4V=0Q<?iVW(dc&+wT34f!#A^O2`8CpiI0Hkp#-D2UUQRyo>>0@3k<Wa#Hs?
z=0BDS$V1Dy>6+N5Py7sI)lEkK@DI&e)akPC{`d?R>Hmx;AuDPHnjo6iArv}C7ti*c
znLSlaZW^vB6$%wyidD6`QaI9p?$6-yfx0<}vkFPYgd<Ueb!pLi9!N{BDs+QnDk6@}
zaqK=?LjAibi3L1r-Av%ea*eja9{=-;w_DJK^t~8IDG9U14L3t$=r~b^`nFIu=Lk3&
z^(i50p-SEZsgRSj4fS{*aS}UiCIUHh{OKw@<eac^nOj4RY0}{p%!?@J16hkmsM_wy
zfzWht>4HR)mq+z%(!_Nd%lBbEUX;8HM%UQo1C94L1|I<9_}-{$(-HdKDfJFb)$G`3
zb$W-m(nuHk0vfC+4t(%f$u6B*T&^bKL4Ay-mHxb49HF|Zf3%?(HlAz1$`aEXS~iY_
zulRYW4~zlw?+V%b1R_iln$vp@zHW&qWkM+Zwwexq*Hx1DD@BxXuI&<OhAOOBgDe8g
zqiH1pw+GNQn|H2v@g<*HxM3y9*CB;j^3pXu(;J<?uuxlR@90tDN#U;avs#ZCYV7K~
zCmOp3LZ133L6c7EYRjPA8tP%Zl|JZn%aKESDA%J<s2|W<7NJw{ZfroW#3E)NY;je0
z1%C5#uhM>gSM!xuEH`?{k%Diz%(1v1{dCJ|R|QaSFB{bqL}INN+w{S)SnF|dfCZO&
z>BBuluy5j3k`t_tPI5L4wc{%yI7(WB@&d}Fk5{hO)e#CRd`lM$G5NVm;#ma`n+eA1
zVgHeX*1y4K^$7#-Zw|3vr3OZZ?JT<2PJ6q{bZ7^;&do2OkM(c%v?_uhFxZF=iXV4f
z;R619DOMZ+EK8sH=N=4jkPx%c0wEPnVM=Up{2tEmZrEHLR+g0DbdKW~)z$GZK&q5A
zmj7lGw_qFqZH1Ry`fcNcO1vindP8-6B79vqt*<0)w0KjCD3H(ftKjdx6!sDL@7)-<
z`Y$G`p>o3Xdso&qLaH&2%igR9ZMjrqK%gU@iK*;oU3b_q4eNMyO#V?AhV3qgT4CK9
z&eX$9kpYh^s7<U^iw=Yulss|+h<Hgft;H6?C{FYfv1P$sa;YJ_4O_bZqXUdh3sftP
zGQEsUyGj55_MP*s)SZEMq%02)1vrD9;YrjFqGG*{nL>Oy1Um7G)3B{Fi%LK_Hx8Bv
zAd;L)JnO<k*zc<(8Eq}+*Jn5v?{Itx6k$5E7QSh2!v=>1`U?7mE5R}1v^M=J2@c7j
zni3uaDxp(?8?VC=J%)fFHYNKU@-;_JDL$<nzR{z;`_oJ+x8Erl3jN(w&1#uj)mo?Z
zgYZCSyG_<|wo`ZwYO}IlRx&^XNNXcl6v4|qe0n$DIH5DMOsA2DO*npTc6N^8a9NY^
zUUL<B$G7Fam3xGy2gxm<Rm4$*UOEK!<~CO|c2bwnF4+Gdhh%W1OIRUJp`ikSIhV|a
zO9X{xB%-8pp311oAo^Tt1GtrTc?>M|Q8Y%rPzV<>p>hkt$?9)4-w5FBuv(k)Rzo%?
z{dL(oCCV#?^o|%gnTAVcgh2NG_yz9mA4Z0wVSnGWYzl>JTzscB<0yL(K?O#a^mh@h
zd^(BOAX-m)rLA!eB@SykWhrz0Pjix1sVR+!TT4yCt)!Fy@2;t|D8k)i8Y`1XQvD0c
zr83jm$T#UEM~Y+)3U5H!QDJaccvxn85adZnmACvidn=&pzU;_N2*Tgr9jQM06$JcP
z;4dU_d6<ET14JP2j`LifqBt#8$V@MmyMqE~=KzN^Fw%G7jP^oh*f1FIZ!*D%s6;hP
zXkN10_~*MMAwDpXAQm*RE=i4jQYK-U5$&f>avYDSbJC8>Aw&RDdR%i5gNq(-lGO~+
z=*J$id>wwF9>XDXbUk~RA-8ST<AaE2)lL$R?H>%bn@r}-<<+~aSv$)XT=#3cs;#80
z7y*sOtKWA%af?~DI0RsP*RpYCdpZV1&FHSgG9wR>-5|eSjCe0!?sU8HDFFE@ne2+L
z!|EI|UE(iH%I^RiGx*{#5t9RR8|we^Ddo}ho9HbDtjlBPeY1e9k?m@CP+*T-?UAlo
zy;T(}C8Eb^@fnlz5#Jid<pCRYpl$GN8K?%hCA0ngk*@8AZPPDFAEMH?gN)jG6C~{{
z!hQj~xRUQmI$~6X7<1lj7387Pe9cHsEGn5ldX?!9qyUwSL%1#MPLi7+a}Cm^bf+C@
zVSLl>0gdF;ng_r3B?}cbb8)t?@J)8Aru6|;Kdw^~GpeT5(UM5YKrS3PSQ5il;(Q+}
z0nucJEhO2_aH1#uij~GY8{^5FQf+FDIrVM)7ZQbK?v}e$lO0OdtJdEpFrzEJsL=a;
zktcnj?SRThD+Df6T8;4UXO;)Oe*=2B44yJy?iQH37<Xz_F^g6pdp+ZGhyLvTF3e!H
zwd`01Il1^pYa(63(TS#V&nzmAT8}i;0qjTdCHd{oioZYW(}w$91|PGEnRi5GZi!h>
z!(*c6JyJ9RKwwJ(x}j*3A?zH>_%#{Yhbr$tV*qTBQF#)3U9SypdUnYmO~3Ld7iV24
z{i214$#p)Vdi|_U==!D5QhNx~YJ)d$pFmZ5a|?IDikKJg5OFOl<;k+x>6m3rk}gkw
z8~5~=qcDy=P_5cN7H*<4{7G)8FCy;d2wTTVzliQs{J=SrE>TUL1(GwwmUH*@2*jGf
zJO|Xt_x6h+@YzNPlU(EsD)rAg8kkc($?5xv32<Q_;yWaNATWYpaWQVXZ}Qkw+lPY0
z<vjW*-hLk(9@M<KJRNYELyk@7>+)-ONtvvTA7t}QFW*m*jOOCm$U2U;w#85hNCn=-
z=p%%fAZM`juF3_t(}z^TO2@jW(@TBEDFAi?L<MkROQhe(OOc2oNR7h|EkT<9j5au;
zUd}b6eW$E@lrhJy!>i>PsFavC+L)VT2AAc2TC17&oHUikK~8QDO0q;#q}J%Vq*1iZ
zPhb1L7N>G6vYNn0Yc(3<%9Lz!>HO*i)#WEhzhBw7T#mk5`~(h{C1MU3C+i&t;{}wu
z26~pj^OE1D_Zx3Ir@TF<_b!HlFrd0JPSDoYnB+9>>9$2ijF`ddT43OslgI`*kM%J9
z+~F)WC3WL9UohIsrDVs!T^E+#^B6VMsc+^SULZdnK^<igMUxHnqoUztvgnzJs<dq0
zp1$5s3?!`&3^)EkA*4lBJpZ=6VgSU#&q+E((yX&)C4>6*fiIYP{iED}W%mjO4?j<@
zx-r&o=KswmEU#SIZDr0l6%5W|JN%-JKqo_=9FgYhQG2}{bdq)vxT{!3UQFJQ^%esa
zE8|CsP;|2}Yy*KbhaU`?0`Lq;c6g|}m!X#l6^ur}yX!7~J<_V;o0>&dAOXx@CwXEU
zMewi28us;IhJ>9Uvk%u$9gT>MzVeSSGfR}@M9Na0{<@X70u%Q;6hlDeqwx}fg&}}p
zUin{4TIxByndrFhzCy=!>ZxcvxR8AnOhkrfm-84y(KY=sHy31UAWT;-a=PKx{!<F}
zPmykG(xdWMm>&81@Hxy^;xWJ@yS$k8J`^MC1n&L?n&}xdhE`eOcYTd}r^=OJp8Et_
zcuA8N)9RyAwkRaWQ9G~0S&Zh!_IJ@rhczH|tLKF!i;(n48`JSEym%{?mhGSDeVTgP
z+Dy@hj@HHGPfkm)d0xA--(Iyv@TK0)TshW(EiVJcqe+NjZ|_kT?Fjh4bD32qLPD&#
zd;=yp4zeYzEA}KNzARtO_>`)K1ov{V7+jdQ^-;hxwsijC_>E=5S0+;MsvWg!URDqU
zcVxD20aM74!h67PZnX!RIQ@vCVx3a*R|F_uo-;dcguVP!iF?G-iCK#(eYM3j=hR_Q
znY#{jH&OS_)^`qI>SR$1`igM9FQ{TY81g*u>x0tylPus1^kb~(D@d4XH@(|Fs1M2)
zuR;1(j;qwmQrPC^5?v|#tla2p#SoZF)}Nyg?=>$n)7rm_yD}|-mh-z(c;tt(O(wXJ
zS=|gb7n%<@X1it`CoxKfG5ea1s}{8aKTT>DO0shhd$}C}2;C;Vw1|u|1p*Uwvu_7k
zVeFpQE8ix)@h^@4i1mAxw?h5Mvffz2dQBGW$A_P{msyTeuk<(Q?(9EDeo0mTZ53?P
ztW&jaxS{tApL{ByS5O7jFJL>p69HzI8$yewq2OX2m_%ACPag@=o0l}boo^oP?)#I_
zgtq)a#ak%}$iuWF-_WAg0~cUX+nK$DFq{{+Se*)S$^f0cTI31gpl~$?*178@mwEzA
z87=Vk?*My{-Z~{PAaQyq;_LS?Y+nP77K6Wg0d}rmh=;+9i5d|bm#n7rcK3bFCs7TZ
z{$o@{+4YM&DrMh8TNTSDHmmRiL)cg@x`SO*WQ9Zsz^g7>%!xl}A)FW+Q8Y8i>o_QI
zYap+0o0fn3O#atiqu>29(nWysqzI5_3Vj;QF=ACPFvN6)e_5;R^_kf1<cM9<>yFpl
zTN8fHBW^^N(c$@N5OrLVv19WGnup)3afE$@y7g|IoIPYYw_H7z+YxlyCT@uz()ywg
z-7N4OP-k&2d6rv9&|8N_c43gdXq*n>iFaPnWkusZ2Auo8_&pHd3D>hSm;&l5A+V?~
zN0$A%iaq@gS3dB%jC(Ra4=G2Tm;Fw5>T^BOU)Qo8eCwL3Kyv!pt&=H#ot@paQ)kkr
z>vK<mpOKMAx&$JGg$fAjWMq*nzbBFoA?_D30Jq*btYp40OD+}UxyN9SW(SRMt5?Iw
zTh9O}m<1?mQu>ww7=?BULK29Kk*pn5Tto!keel;Xj(@597!Di{l1(tmG>t8ijbJ_S
z0E8wG6gv<Kk_`}woA{tkI+N=l97Q_kTNZQ9ms_IokfkMUqDWLdSy*Sy+)G;xE=~X#
zfE8~SUk+2bW_2G?&PX+UIDp*PZtip!3R?E2C}@o~yWx-QVbzEtwDz%V{wYK&oN^VD
z&9lIj>cbkt-Vb||wo4xGLJh5Y_&UQzSCno{ll2w%vcko^=db$_)jhpPx6Q~fA11^f
z-$>jp)Ovr6c9{@2D>Z+X8ndJ(I_F^kKn+vEhJ$mK91eOODoxB=P@O7uPt-3zN*B?@
zmYwfUN9o;Ywg>Y@y)^b7d`*}k36acEc^6qfNQPZ}D&<Io^|lIBZcyjz1Ma#ng?>f$
zB8j6H#=Dy<%&U#J!(qVAtlcCB(4V~#b6fyRl7uLJ7v13Hdw$t`LqZTcf?u%#CWjy(
zo{r^pI>lnTN~5zgqy|KSd;eUes&_fA-L!V{b?m%8^gD-;go=Bn;P?^v#zC`e!LgJ#
zHnZS|P7miJCE+t4RsZf$?<ijj3`XicS}xms&JsoCItrN{>H0}9aeSara1_sSW@vDz
zPO`^yC^gS&<zSu*wQZBt(W|`y33)5y$iJ*m7_}2!i{{}%#KQl&60|-x9mHNLwMm5z
zd%b)5nUs+OLp5*$wJ3LZIAT<28+>qTQ1sQWZbt}-#Qeg5{Y}eX?!Zz;rh1NxQOX3)
z=HtC~Ee3@~%s2AdfZVVNH6^kuga7AQz1Q{&Q1JspiqyMStflxmB8qnbzh<T?Ix{J9
zf9-M{EmdPgh3MkiI7S!O#5-b=n8wL?N-H)buW3!~XJ7H9QA0R`Y6nX&B&!gSsb>sS
za5YentA~2_%tbO~=klkeiP#(k>@dpOm|8@pxxMBmv~Oi@*1$W6JkV`*G|<Z?o}kSZ
z)}z4q@?R&N8Gxg^e9_JWYKc^9|MjXyG$Yl1b&309G7t!>voW}y>52(ABtn#Dr)d@8
z`%endH6s%Dwex=+T6r0OKo_ejChUSKcPL9@;N~mrCJ1&~?|o08h|$21hsi7cefc?v
zx+LF4tahZIbUgod7C<#gR)j98Zwvlt6Hs^6q+l8)k!FFAh#nFEKec5O72IWEocZ|o
zdGzGb1H(dOji#U{V)naRJ?tZHp)u5teSSA^YQxX@{iA@TtR~6rctQS4{II{9h<TSz
zkUKlH%xH+Z9e+%iFftzbgod4aro2-DG}?!As#g!~ml-gkek9ZC%T+5*Ne%GZ(L|c<
zmqU=~gwV=d;KIuQG2}rY;`ZjEf|(kQhr^WqADA`0J8-(i?-xrjj9r1lmv~f3c+d8h
zVWbc~`}QAtrjx@zlsNMyU^C8c1FVap+hW6FTuO__KuVAQG5TdQl#w3;hpNQGm%Rka
zYYjTc99r8`m|>_z&4*oaAu06uFeT&V6c`Gz2#?!?bqYXbwqQZ_wdlUuDTQP0Fv#@T
zmo&+Bt<XZ(dYQSARk?;s+QFmNBT&38EdH$Voh*Qs7N&qLXP|?T0Sx^DP!+)LJ(hDj
zPH`V%CO(!*l7K~~Q;d5yMHo^g-hA0sjWLrswS{E|OQp45XyIX9G=hqxSyKM-4-#4s
zu)B&#gaUNAl27~_K0MOoks2+KrrnyR?OGpOq;j@vh))S^!0v=-t~ha!Bw2nQx~97@
zp7Ai8#d(|dNrPS^k&;>FM3vAkw=;czxb;V3<{7{zdY)ksS+gGS?{`x6EVEo1{hc!W
zu1gq}0^S5YcyR2nfbO)|GzJbmpJo{UDumR4=MKQbW&{Hg%`P2U)9|t+58g7zxq_ig
zP`s))#N5@C6@(!QAu(X$`I%OjB9G@+C?j|ujE7%C;A+m?Ii^o{V#$9>W|T1m0yKj(
z!vaI(NIH!`BVreU{B^KHqfb^fWYGQc<pi^WjY4gaPasEsn=f7N`hji(*<sn$TE-Bx
z$pcKnhg?Ts9ODMsz_=i|%x>FDyO9)PgUUgEzA!Xmwa}Y90`Vj<B$*}e6U0n*-*~7|
zqKZ8Hu>@r)jaXa$5|4UAhXW7W2u>fK!#nBY<LddO(141tP-bnA>~5ot$F=Qy*BgwX
z>P!Gr$=?ejv!`#Zt4K$!<);YJFip?H0Mbi3F#$d2h6cn&nCeW?D{-pjF~xVlTiHBn
zbNIj51>LVAQ<jq_B7l+Y<^^dEeSTx2?fKg(0&w>eGD(<ja~BW!N~##)RL24;phkY$
zbPoVgS&G5b6oZ5ViLwg@GeZ%9AZ-94tA7Jy0Y}>mN@iXV1Cd*XNL5A>{r3b&Wl9qw
zgd|#6gDOMjaZVJaYpJpq0|_gxIT%n=f;`(9O;<4AttdH01(#RkLve})K@;#2n^u><
z|Eg*77Q1jUOK&aQiXwoIv|C5Lr^UZj-EG^isA`Pzhl%vhu!O5OJy|K(GOYWsIvHEf
ztr(KU>BIq*R0cu;AYOgU>WTnv+?1-MbPW4|!{xjFo!A?HaI7-)$n1UIwY=u&-RyAn
z{g1|&j-&0#OdZ`YdL0~7hw!mY5D@6kC^6EX!J=$tGx|<4dqkI8%@Yd3FUjjRp5e@4
z7q8{}kJaM!Er{MgqWdAVRtk)a3!74ZeEI5!w-SA5Q+`nbv5H^=NZfIziJ`wec<XBk
z6c$iiUDTBcNv-XSOcZ|HzM<`jLw+042p58>najO5d{<|N%6Ao<B#63X@F~%GEwS`O
z_@Kv-aleIFP@bl@$QRj;q|5efq7I?gkS_Hj<%O2x21>|!fG}B|7fRF|d9v;DXj7|c
zJS>ZM8D195tb~mMQfg#iV+L(qBZUBgNSxvM9%bPUceR?HIas<X_$ItkoT*wOHt6A1
z2WAA}J4E%<1px)9K}JW!j>h3}_jVn?JOf?foJYdHfeV7l`UAqDus#h8Dxy7_xRWnV
z?&Xq5UC!8s(rXrnXO6-woF&v@iOTZ_s6fO+Gn8mp5tX_Dzpv1X4dc3gQ<3ouV?T19
zJKj8UGRDx9eu1J3H3c4&K+Q)t&<q*dm{XcJVwa@>^VvSJ^xQs;X|&vbd1HLZWz6~+
zzcjF*_2O=z|D9#T7t|>C&`T<QhpC470kx;Jxu!Bcg2s#k-HWBTW217_0`}{VCplJf
z4Oa-OVe7+yA#c;@OLFzxbx4ubA+MJE_i<GVd2R{bM`(66?!klq6~t&&E*u@?1`#S(
z8l!S3jUMxmBBU_3N_G+c_|xW<ELx#j8%#55D%BTSLyrrEeofXadDxyIk^eO_i?lq5
zjset1(Ai<<JAB#IX5c=ga0>4!g!|!|&MQckWC3V1Qwwb68aMuVG@n^8k#;>}dS>uV
zm@O%eomT*z2I}T~CRX#m&VQR*Sz(eE2Vy5HnJv7n(y>U3&|en15og!z(-#TW57b#A
zu`i%reKcMnq8sjOzEf`LlyoM0VG0x~)2ukj0j0n=$V-*J1$`9y+YDrb<0_boOTMk?
zQ~>sSg`oPyXxoEXjZCl>lQT|!fcb@Ux>mMZM|bmW|9seLkuC2mRBN=M8MMcimCccU
z$@rp={I3!Vzj?=qzNEL9`m1;ULqcQ_@yc3@sqZ1};0dnbz2*e0N=lQnKbXfA<+Sf?
z6X`6};5tG8%uFHZMzTZaUr~jfLw6JbHGnO~LrTj=A2)jRAUUG4$J`rXhYQQDQ2{d#
z!8v=lv(m%s*3KV$&CN)`+<V~{9;RtbgNP>EESSZQw*aObD)JrWJYtOC-`tR%+5X2G
ztRDCl>%t-;W9NT&?a6bghcPaU_pCl)60Q@(TWh0aSVwdSj=yWS-kIwWZ!X1eh5>|`
z`DBSGTqfz%31v0cY=;yaMha6rDO4DkbuV(v+sl(PW~?7;`4K8Kv)|%|HSG6ZAvx6?
z$x<{GJkPOrFMJ<+K~;6(C!fzZH@Qx>Ch01sJMH%(;o8X2EL_VPV{a8k{JJ^E@{MS|
zeFUK@hbQETOUb~$KlPWbeN&ZQ^Z<4*^lj&;YuWcf+92~K@RVV{-==!D&nM)_+OpE%
z26Nr}#`GXmhie3byFj-4c4r>Ob!CVH`g3acJ6WBw`UY@0gcp-0FhmGOm-giyp6Nq^
z&wn$b<Jb>i^@~Zi`xLX=C^Z13gUBk$UMCW5$KxMIzV;}GlAZsM^R2_g?tsc;3zV>J
z$J2GxFcQY;JksAXmA*?V%&OohRF**Vz1VCPvJKc!67kw0dYSzXq{mzWS0&f0+nPDO
zpYNn$y`W=ar|sO6?YXr}=9E!>ox+`#-6oVkZdOU&|52&|PxasK!7Nt=_-3yj`)znT
zf_LvSo|b5YYx<udcLBS)fC|m3+`Zz}#Orif%NB2^GP<3@QC$^rr!wa?p&Irm^Bpr=
zPu&b|A>^)-6}#=0t{OW9JC5&d4<jS!*Pn$=r!p{Cly`+&J!xK|Tmx4bEum(%F=|5v
z|0&m}{buM1$Iv09nwzx=)lHWr0xGt2ug5SMV{2|&;i2alWur?8;KDr85^z5$%)Z9t
z#y-1<#Lp<K%5s(Xrt*bfN}F^Dr~QnU6BALf(VE24IrjYkH_q`v;FU<=8|$k#J9t@x
z5ZpB{;K379s#BxCXh}yC!;5}lBP;o$dyt3J$6-81dfU!M=ytu?EM{k@ToJcav+cxu
z(~7}UiXjIs`lZDInCxelox+d95$7)Zt<dg)Aznl_rH8uZtn!iw^!L>Dn97gwOe*UE
z?Y3K$2XnTYP;SI0SdcOoUGoSW2t*%)&>sm7%+G^gMUuA8@cKRx;uGH1ZexAA{vQE0
z9)GTpRFGbHX5_r+f>#%j9TZ7#ofuLC+SK4u@AJD|c2MXMKyoO=lp!(3K4jVCN0NUi
z8!44E5&+JQ&5qrJK6aF*PMBlw7JMa0J9lk*Bejcs&|QjBW>O*;K=FOOQdHxllzidH
z#3p1v@Mqghp}~$6)jG0<-#F@q%FvYYU_I>?&*EvXdTYNcCX%M}>0T`0M3|_8Bv|#M
z8=*-7>$p_=$C6X|aJICz!ou(Ra;?tlPgmc!V}w95&y2KGqBNeP`PmM&pX4S719Cwu
zI<x+tpTd$LXX7v%S^2JMfTz5oAxJ|Ud4-Q@XSX4rc*)opsES!t5gg4lBK<O=vr0SV
zoE`=HhBKrOxRk?<0Tm&MMkPneG6?Q0em5@8iaqxj;IZ18`QIJ2TkkF)!tqHisflOT
zZ>JffVKsfM0pnxTsy5jhQ)tDx_~PKVI2;7&hBNk7FcJzmUONdebkdi+gJ>)S@g;wU
zNRWNh;M+-+Awt8ScParDGt(DKD4+`w!G4S3Lic+4+kmi_;r?a0YTgFKo4kk^6p|6<
z4-)@zfDqkuu3$K`9JHV$Fw(M8XnT?G(BQIt32v-+TMB)0R0?-;TA0z0E}>t&P&5>-
z)oCYzeNHhj5fBn0(SYt4-#L&f5H%citKpJR+TTmA-u|^P?pP79X&#T%kMOwD_aa0P
znPa6x9P<+Cf=Oh@)KY}DY50>EL*j~sf1OoJ0RD4Uec2D3rZ2QP#rBSqU0sDv*G_^x
z>8hwfM1{Xy2(B{PbH5z=VrP)d9fv=Cjjh`rpl*4GZ5-D!PJ(K9*g^MCyES<PU_pUS
zI5<8LcVwO(`cx@VGgYSkJ;AT~q66<6U_%C6o+@AKx7~9&rhveXeYHik+6}xrzqoc0
z18CeyuY2s8YQHKC|9W~=+8!DW_>#P&_8RgoOkn8hFxfhlag!CGTa$`bmDS1<c#r05
zbN97Vsz%;$L-La(RmC$OrsVZk8hyMcRZNp#^b4zqX^f^5IO}vQc>iG30bk{`OxsK7
z_@G~r(pl)R<}=cKlddc)cfp^}GDh&*0De8!rXuU}T698HMO){{s|YLN)|fbmNcpJX
zQw#z<J9q3uh$u`RAnnCKjJw&Hd9vLo?q-TAvfG*ROWoE{K|disv3PysP*)`V|AKzR
z-K;Y7jdLk_UVVQXhKXCO({M?u%!0QR?frx4KbXM7gf#*Y06P#-*heyy3yANNs@n*r
z0>bv6PS6AyY6NEm8<-|Jtc~q=DO+_>K=UF(xTM8}2*qo@Z?HNEFw@kT9E>II4ZSAO
zER17FpkZ*Ta6<Nov@LUmT|cA0Zk(Hu?p%HDkh$SznON^G>5)pZ{arY3uXns>S&!GL
zMewP$`8nvZv54<1#TMcR4|sd92si(sW6yiepy9so0bgGERH^8%*J~cv@7@*pVeg=L
z4gNTRnPJJcnjttKnuR;57Wk%icd|gA*&J^335!}SdZ%6`kfCOZLS|8#l*_Pcr6r_n
z+?SPuS{$<zMbcMjY~&ud!8hL>16Q}w+rMxuG?^+k32-Gdu`|tH0poD00yAD6zt6hR
zC?J3(UxK_Gmk~@M?!`K2$2L1?alC|oBb{;Pti{ozGG*^%iR}N_D}^gq4`;wvV91;}
zRi90efbcvxKY0BE(&m}vos}l^1A3GI*6k|KofzQ+&H<xU^<^AT_Y0lrhWrdh@neej
z@C}K~l1s*t<)PN%2QZt{$jYvCCoxX)K<Cl8K;ZZaG@ytVv7F>!PAe1`Of^;c+xK(K
zz4n?(YJ2+$DBzs6Sp5kk&1GCblRfh<eXB1$fb93(WRM~Tj5jb3Cs7|)ZM_kGtwuoI
zG`HdT`~s?T|1*O{Jd|WZiel`EN{A@?<`r4Idgk7=e)daj5a7;KT5>n+ZnOjmgO9>E
zFdmT{gRC5{V`GRLPSB=JoyO?mR>hz9Jh7DEt%Wh2vD~E|kJOcm<k+cKtt7!qq|Gfk
z8)-(7o_T?XgLOO+nm&v9t6a&ZIPssBWP=4AtLK-no>FzqP4acEPa$o+>i`&dE0wUi
zRTsJkJU1|%B|r$)g=RmbmKDz(;tl*wPS~$N2RNJ|oT2Ryb_YhV`bEpgaG3jP2L_1Q
z02-wMY`(^yB~+*lM4ZvWbbfNs%Jf|Q)=(*A=6RUUs^a0#LDTk)>x=4gi_uBpsOrU7
z2g7G=T9@Ybz6h@vzMHVDD8H`LbXiJaY>y1LpcJ%O9RLKON0&YYh|8Yzo;~xjQjXp>
z_0g<E)57U5H*9|VLwRti<^tWlep&=;<nW|H;QJ^d_2*S!rA}*MU2;PZIVZbZeU<s)
znO`;J#!-CiN>kA1Bo{LNiaPoX&6x`fui6g!v+oGkd!T+m5#o0x*9KK%n{h~5c&|9>
zkm^6VTtIfdWo3G>&~~I)I2T1MT#XGu#ZgB@Lc~~QLD96ZCW`!|zd^AoW41_Stx3JU
zAZvfTf4Arnt)40@dIW{J<B7|$*~Rca@f;QX8I(2Q7K6)6GBuk-2#shOBZq`a8MUXM
z#*wEfeeP^go9CKRX?$Jh!Gq|+c?A`P?P#~nI$&r+<mfJ_NM$*}8!_iOH?>Z=RxJxN
zL7TrII;V?&d*4vNcq|8YTHennQ(~PV2Wb~|@ZT(tm$3h+jN;MPi%1nI$3oBYYuTgJ
zuHlPU$N|EY{bWce8t3!Z7o|VwkyMc&$0PnTu;5}y{9{&_ThR~w=e{6!cEe)D`*=LI
z1yHYZrx9=?=;3%#1y$?Nq~Gz!fqDqoema2Q2*kiz%zac$py(V;>XXd-Ywzx8Ni@L#
zw!}8l_*20#LPh+W!c?23>1jk*?67`!>|608lW(V{lW*1XcH%+PDAz=a(lE@y48n=m
zceF8q_5A0E{v8s1Pg(x)IN96jPeMsV3i#sb?QeQ~psm-Z;=JvmC(*@yA^uYI_jsIN
zx0`$lQ^?c4R+tFHE=+z5T4$)-wzN$cLV3*6oQ5G*>u(49!Z@EGL>3uA(1?vlFoI+f
zRR-PhbD5lF@MNK0__3fczGx@A6cXYW(At?nX-61yHSPS{&|=i%wNULC2DH~W3cyC{
z{Ev<+xEK<V&LtVivCN7NjSd*T%rM$$-_IMF;ybEEr>Uz*&B>r<XEOcCAempsWEaBT
z3_aW_q$^aZG3^hjDdK{Zqt<J12ihri+)iwsFV_xV{(i!!BjlJgQpAg+ZQJ*oycONz
zQ=hTD@fMQ)^$#poGB^axs3jo)Snl%sC;dfZaebyHK+lT&3??;Pn@_n)o(|T`oBxk{
z*j8t|ymu1yZytd!w!O<lczw;k4c0%Um4+5GmO_w=A)aZ?_ULpPkvfc&({gbQS%ryM
zeE+<Ju}R~KIU8ZAScPL&{B~D%d#0^4l^$gj8&>#%e-O|fwj4|cfY2A~NQaC*8Cn@O
z$ALds9ZMaT`RcgHGlw!W@jzGi*9L=zm_+nP^~&b6n7Uo1mp#U$<tt)&Iy-h%UlI3R
zFI~~xl6L<VF1A}NotME{97Gr28y*GUTNvCAgjA)tw$e>`S3czamFm}g&RdJ>(>1d#
zs1>xCI){@pKCmu~0ho=mMD<dC<gA_n7-Lc{45N*9&^Bv1%7ij2$T};uw#=Jjl7E)4
zn<11#|EH2z)Z<=nOlB(0xyf;Z*U<Qz%8;jnPvnA>RJK#{Wv!4a0zr?RJC+s}!w7qJ
zlRf@Ra^`~wxt*W)K$Pjf()@LtaFH_gDh_-v-EjXr+$2D$RWC%iN0H9Mx@*kpV<=g1
zx*c|m%d=n?W<VoXXbQf63$d+DC(_<n7da12tcwKr5PT495Uqf5<24L^5$q?>LRguE
z28VdrlvstV@JkeHR$iXZ&^FU%O}f8nm<ch(`r=v*Ek4}%AengGn7T*xd?~+$M&?Pv
zr>eQb0YPmvcEY!38J3T1yfY+Th9e*d%=T&BAAiujoYVagnXUhN5dI46r#Bg)Q+;s!
z<V-ZWlr-6PJGgE2ZmxWMGY~8fz@=_QaR{ES(>2PWNBK3C$ajbZnpEp-2RNc}xsQfy
zHC&{NTf=4-e@v|0@`(P!R)!QoFf~z7z(mo>0I1XSeEI4wv>b(Hn$aj<^Ne?v96RED
z%W|HU&ifM!we{*cc~|%8-LLhCJ9vBwH=4gMIDj7#7}X->fM-RrPgJkTJxAlQbf#ic
z8H64K`40Dpecr^1c~k9t#)!Hns2|V^;tT$*0r#zus}KLY>Imk7ea<{@nm_frVk*Cs
z1DN5p%wDT3X&1*I5mF803qJJlsi!tj>#wYy3b7Vc-?S*v)W&=A+@<}|m}{I~9~oqe
zp~y6%w>xAfjb#TvseBm4ch<U~eLdCm`s1sYW?2|1_syQ`%5#j>k|D5J!6Xllb+4AN
z_=KgVrCY)yAaUUT=0yQV>1KpOlQGrQ2VlZH^W6Et$*GuOJh*2MWd#PAgUZK_PG5Nm
za{v#bvH%wx20Yj3K+b)h$K3RqyIVgUzJUW;Ru2Zf)G0BGHLYs6%7O=v(6o+$Avm2E
z+drQI%JDDP6kt~UY;i!{T9ZC2H{<4xJFU+DN2JhwTBj?Up{UZWRh<m0Il%#}o1EMZ
z98J>W9z};980JQNOkLr@h+ZXMJKd;5?pXFVuB_U<5&-Wa!pUO%4yEa-M@jZOOIFzl
zV=~9{Ufb#QHV+MTniu;CqD1oi8Pe}cA|zRS!%!JP@$Xx$Gdc}izS~a!$nO{``8?6)
z>;J9;tKqd0B;54saoJOX8HoTEi6v@^5t;c|>~5(+b$r!rqW?>Q<yM%KABp({<!krb
zhEA|6GZXA+GZ7N9?#9I<M`pL1)ZJz`cH3}ibsEkvKKNZl`)<(XlFg&voAYwKqtiZr
zp+L~0BVTo4BGApP4=sQ@6DGz%x+M}dpJ=IGSLZKb&q=0JQ!62iIso<#NEhWiJFG3x
zw!;qkwg}c=)*bD}(~ng+1=RfTK23a3mLU*LWHJ}Sd1LAy>PeUpzJiXJylB+ov))dA
z=5u+}z+mvRY3VNg3XSZL18?UB7C!h8WmmhY@;aluw_UBu2**dQ`Q~J@^hNZ1+aW((
zeEZs|j>zriY7hJ2I)E_f?#ws(=57<<SHquBQ?!rjVg9=wx|`*8_B;(Q%>8r>$X<TB
z1!)HGyjg#I{py+76pS2{$R^-n9p(h*11E_S*)M9R7P&Ma%$N!mg~~ApkWFkVi7U?&
z7E;~cQ#!Sc`in`iT@+E|TP3`0dPg64c2S#)m0J}dC0lBq3cz1C!JSjAbq)#)j#Co*
z>QAou$gQZFTzN2J6(j!Lz^i#HMBOSqax+goSuz=7jP<P5rnTm`Qvbrsx5Dliq2|#$
zVZO(hTN?Sc+=9MD?p`@mm4x!4w<&&paok`IS6T896(wA9Rm+l;8c(hMVy@z6mxUK<
zI{qkO9Qq`S0SNg9{>a=7KzsaC5CVrf6H3a;p0Xi|Pb3jj=QByFXZGqO<(>w~2xmm_
z52eG1K1c1+w04bd_FGnFS9HJAQJz3es;qqAV(agBhmiCA|C`@iqW@bLT%4X~HSVfY
zORx&bl1za{vnBiayRqC8J@D2sO^Iat$r9m~R6sE0Sf-JtON|Mn&gWPCZp6KKiQ8)R
zyW-6*UZ7LC95>l?L=$%Ctzs7g)cFu1T3xvrKW9h9-BBc(({xLUpLKrru$p|P(FT4s
zVnamO%Ouv?|1dUlOG+NqrS}7`%&o+7vik%)W8}#omxAq=f*xE_uH7`E8cTBx5hW8q
zA|heE!LdPdI}k;tFKomsi?A-MMX9jkp-Iz9-NQ^T2nRx6nT!ojbFyRF63VsX-~{`e
zd|obWS<cF(>oR!)wY#G7-hYxxn_Gx0z)5?EnX;^TtZp*;6CPzmi-lJa9UcUfu~+!O
zn{yI(Zrz7T%{z?@{%jqAvF-CVGDaT&{MGbhtE5H~a^G+k*5m5PYw9lb3d|iaD7X;J
zlY~azneFrk<~8XC?^*IIAZ)*0Sg>`+;5Nt#RUx7;lcRxo8n^Q~k{!_(JQ;MdLzIhe
zES=-@*+xDLP~G(%@VS>vxdV?Tp^=2&uKo#R8b>_OOW^q}7*%9~*dhL$pIiYbleF@c
zPwl(mwYPA{eFWwCbej`i6p0Uws1A^VG&bSqYi^_oBSFq7IM9SHzNoCB-aL%WFrrqm
zF<#nSGD7<gP~!u&BuzMAMo}a;OwfmNkpZmo&&%WfxTGDN94shPugw173}Kn^1jkrh
zyTu|~+^GUi^vj+zvVhOPPk<A7-&R~{;A`M33=k~S|DoshRZC4UV(~Ib4jWdIg)I+R
zD%$^$Gvu%V8ZH<;7)9iFB%HVKH+hd4z+jqK5P}w1fVrPGz;FWuC(ytQ*-=Bnpi{-V
zlOpm&mZWrpk?}Y`B~Fb7k#%Tu(LjIvXJ=KZU;`0b`G<h226#dFNOV;S;aZ6BUj9ew
zx>@sFcGLC5dS&l8`N(+fT~?%{miKR5Nv8MCC&ybmgfQLyT=SIj^j)jNMd=`{aNZe5
zNX_^2mP$hRt3Sqj#VEF<u!ph^Lpc{+P30YDwt!0ArI0P2Ps!$Sq)TCiPG^DjFy0Je
zbGxwXeVWR24e$g!d%6G${FY{v5Q@otF&~@5-fHs>6lJfVP7!!5v-vEQqkx!q;#Sx7
zm;8H1>#choC<!Ytq<`cPf~ZTJYKjthjyZCmMJCo*W(KPYRLq$cqs9@pa#dDE@82Fd
z(WuuAIagRIn^R=F^S`m(HIA|vT5muK*N-b+`4q}-10bid%>x<16)zk1)uxqQr<Mt>
z)h>8gdXU}w!so@@jLZ-W#ejBVdj4&Q3B=^d<)u!)v}_CCwV%4aRFr}g$Oh=j82l<m
zR~EPhF+INw(g=lrCF6h-oCQ{(5zEx%wmc-13RP{4UC7<}G2+_k`&EcEWlu8=`1sO&
zauDr>?MHt}j=dSmWzNAmq0jQ3xwACD{{A)FV^=Q&yhT}96=-rFFuUhr0X>bI-eyYW
z`0))Ra<n?zX4Dv~%`A?_JPLbDScM}b@;<NW`k*m9;rSMqC+Tl#DXty6Ae(C|oHsw8
zQ8HMB+#<C+-_X{>M3QB~rB6*&k}4>Bwjpb_29Ovo#mLk$$WfUx4LJWM68Di0ssgTs
zXvbk*uPO>JcITW0Gkg?coOsCE?&)uyl>gf}aPyJ-4Ri}~i-OzD-Ufhp<R+sB_5t=0
zL)1UC0R9UBulS2dNS@H%W_ONURDT~!LB9hUH`$G!g?$>@EDL_Ex(iPLrvIYmTw>|5
zP^s`fNjuAwXs<f;lH`P1r9FGHFj?FtD}C1f1-NGxJS`vv!3(9{)(kifAxGjk1J1lv
zIwC1n4UHX3dwjGZ>-!x`e!z3YZ6A}A)~OoH@=iJ}Rs#eJP;7-`eThD*<=L1W$b#6w
z?x*ZX#hF9OndNlOaq*&J*5Qkj3Ig8C2IEEU{{_m%<3Zn0crNkYmLb=)+^0V&n?4xA
zL!EE-UvxQ7I3rSwO5#_U1Xpj|4miekv|3(m6IWTk3%Kn&3@e~Pcw|^cAw4GaK34{0
z`wc(BW&*4>9uX}@Q9tNEu)MwR&M?~Q^XJ9MIs$}o*RMQt#w9&WOfQ>7*CDwL0x=|z
zQp~N2G&`8LK%Khsy{lUyzG*<F$VyD_=fbwFGC)S9=<y}Nr^pgaQ9|?U-_n^IrMN*K
zlLA7R%*^{sYRHi^n!FktNTR?_M>ThpF{#IPP00H(syTU$aHo9cY5ms=c;97?jn2<$
z2uNBd#&Hh|F8vUmpoRLl(<U`R@Bn{j>cj>|4YRxE{IOdtos(})YMrAnqBM0feP3y+
zoNylP2C-;OF2pB#@#^#oFb$pQNYH-U>9Z&2TI8EdXJ1o|-v8=!wy5uZdQ@6j!Sh|)
z2-3!}Kfe;~;Q767(!?<eMViA)M|T#^TPZRKd%SI@YbbLKXw8S5Dx`QHfdsI!#Ynh{
z(iUIHPe5()EG_4$rcN~Vbml!*J=kPmjZYTxbOyhOQjlLx2Q8qodya)dH_-abjqm2(
zxj5W4XWB0xW+vewUNOKgr^?B7UzyL&vE3f@hk1t}o|P%n!>1(#DSNo`nFqHFu^qqJ
zz;Kw~h}qq#PFZ7gN{q=`m;vbGEqny|OdXak$)`-$tfWlKFUsDWdg${CP$W_tE^Bf1
zoSWC$%;Svo&<r$;)}Inmpe#pk;Tl$3r_Q3o$P1J-8b@Klh&+KEK%ZD}ZmL=@j^Qm-
zzY{TmpyPjoe6v5=F8$nbzp%`GkcF#sQcWLW8)E1@XRj>?2<}|V*99nwHK;lk%g?4V
zSJ!gs$wl|pCMdONQmM)aksJWVGEmNc!yrRL(W7EU<lgsj<m4_2?_JCQRv1$7N5N|u
zw`Yh`?&WZPi;;jkf#s1%1@u}K3N4+iJU+{{ah?Bm|0v(-`i{)x06lrGWWoAP)z+i)
zcZSa}e{sA`hvDQH2ozwp{|-gUaetEkDtHFT{Mj)DcHF-+;MAf>r5<SF^Vh~}u`9@4
zb74w6Q4?z;h#RZirw;Jt))JKjI^m*Zn-facCGm*NG;xmGO97v>pZ6Ih3zGicB;;HL
zbjq|TCy6HgmK4az=$n906sAPFcO|{>imF~tQjf1IpifDaMh3WSYTQZ}vP7d@;r(zC
z(bvwSXIkKI34dB*FU9Tr<kIU*9%?x@q@L5XGc1fYmMK0kkYnDxW^NH8KjrF)n{~I$
zqOMNaZfTP0adA2{X*j+Y3o+YAxT(O4-de5GL|DF4V}ZrzXmHZ&;B>ot7o}Dus=Mtj
zM3WwAJ=dnb?*$0TSZ@E;bL&i-fN4t3`aNRRqn50Yo@Of8n_Y?byQmjffbroCtI2J#
zo{`MDEu<>3^!oUKT476TYn{^gP#%Y1bvMBiA)r#`MlZM)6Z|T%BPz+;5$WjyElyT=
zlFWsUkE%!jDKif(J0P)yBq-weO==_4(@vfFq*y(b7!@#)HN)(1;QJkzFNTGpzwuqQ
zIoXyqSe1)QG8vkAFiBDBsz3EMHQtQaeROPedLO<$bE8EvFL6dOULN+-sON_;8cDFx
z3%zjgq5Iqm*|_0XDstMagrNZLHH~3`zpjnLQVkXkrl8_g3jVIx$OA}??}Y~pM^Uxf
zBqG+RV*rvIG1n&>M#=v`w}v5dT@YAV-t&&UQ*tm1tH4GtdNrt8jGZWs2vvw(JLoXq
zG)U4Wy&c63<^dwBG_{hY&6Vp>89Md^&+jD;c#;(=C!qywA(V-7(bA9WGA41B=tTqa
zERwnGZt2RPvg6(Zu=Q8&GS6Jf5&L=<3X#8d{eUIY^mIXR=!+QS#im9RX>pq_O{{gR
ze>)}y8>xgowa!M}e_jHmzRkITx+<5iRMilHo&e6$N_9b40StWVyY9e;%m_4eEW?G?
zYsuYfz>u!*=hkI=e_a@``ylnvmPMpr_HuSzU#b@I$-)PbEibngLBryD2SXs0ex2AD
zXFlh@gnf$$=)aBCDs^k;vyD{F$C);U|6+8eT^M6A&QjH@z?7e?S_;~UYN*yB(Q#TH
zQ+6HJU>iD}0Bj=%|79KY8!19=94BRJIE{&DIRAJ?CFcIxt$La|Q-`F)A3hGdS^ReY
z|1&d`Vn57`D2aowXd>$WPRh1DYMc|Jg3V|Pk`-^Q@PY#D=pvt<`#164#FtgKxD|wJ
z5zKanz$DCz8pXFs#|0<Q+Z({l%1SDCy{5)+nb<<Dq`y}Ha{#kgxvb4MnFZ=ksGw~Q
zqenzLO^Y^wRk$Wgx2W%Va~K!;p2Ai2WRX6&DLr~y<Bsk+p9#Uo>B2LFYY<B3X5cg;
z6aG<ybbNIhk%RKf&K%7VC&wP9uh#50d_PHi@pKFE<&+%=`_tmy4X9HV+s9<S$EbYc
zB&gYryz7sy%igNx?7kwX3H&ZtY%(tO2)6yqHAgolVl9^m=R{N>mKwsKQIhWfbyd+i
zPR3ub2JbbzOkT+Qc(G$SXCZm)wG;+Q|1S_w(m`<d!$k&AxZw%wBf{93fb)QU0_2uz
zRbAfkKt+*;0^mpq<OA{(i7Nn@D<$&TBZhz_QtV9#5%Riv{XDtw3IvUMX^IyvuRiPp
z33aVsxcittp`rYmBALZ+hy@M=0fZ4jAPHeWpnew`qh%tSA;}UzXvc+@X|F!zngBA4
zMy~>{U^(n|Nd3OWr=VD7cIY%`G^da|;b2y^d=Hq%Y@d$k`m^-!_|Rb7F+eOzX(Bt^
z*4cDerJq})pT?9?(QDQ|?TLoI`jFZnFJ~<PP0oY{j$kmr3(C-d?8;1ia3|}Ch_vAA
z`A42_nGF$;@DCqB{N|S*DiA495!eHro%#6*rCAef3t5#B<tJm#VeXP<OU^&df7XVZ
z6<`2htaf>S^~NrFi<SwV<%s}tyee|qmSg}Sn}h5>DH{E%2>18Bu}sdnSz0Q&{XuAX
zf}{+8vskD2&RBBIygodh1HP+P1)<cig)Y{fD15SA>sea=;VcvLj+Q0wOPT3rh_ThI
zcP)Hqe=r6(%*c`@PXH~@v%<wXR%U1!Mje3O!OKgEo)`D{*kXFRHf9Tjh?fZ!sgl5A
z&kbTfgRo~|r>#V%MTwIO{V$YXd*kYTi4R{Y>1&|M;P4S><2fjtl$Wb!2h%ZK-?_|6
zuW#|%hFO1&?e;9~Jr!JX2ZWk*j5L*_WUQgBy0-RqEc`<%Tc^k8NkcqMU2Omu7$E>!
ze=Re`!SB>7OBWs!bEHgwOjPR_v~0Z@FVE*=S3ibdkE93gC>45&-TrTV34*;P*dbYD
zMk!&7qzUt1Ce&A5$sKk#btiekw%tC9&m=2>ZE1^j%q;z}!Bx`1o2V3aRb@#VN$%rv
zq4%KR)mIQi+b85Lunok@5LPt`s7nCpsz$sXF#{ju65M|jk(bgEVp2m){fG|JJv5~J
zR9hONCVQ(nMqeJcSzA^S<fr63KCbh=n|l!;A{1C=1^DA8M1;cU+regFI}Gh8AcqX<
zGMG>;&Dp}Ap^;3sUV2~vtH+O=4vse>VnIoO@#r$N6a5BS#JnMx3hM547SI+_sOo%A
z0Up4a;>9d$rt3tyg7(sW!v7>VWv++!<+K;TWOE244Fm`wrUo<4206fodZLGp<b4_T
z!?{M@_ctbu-~CRr!0<$6Cv?#wzvYA*Etl}NPpnAw^Lwe4+TW^Gj2C-+vo*S?PG))$
z$m1_%_Fhxrz-@6V!6|O9OXD&IAp*EBIzJd0*NU>Hf=`*nub_nqNg&W3P=E#xC;u`J
z=!DT4-sYBO+RM&g6Gj9b2j#>)YoSmug42O7aNf#szVUfc8A^@$6w8>tHr#u8i=Y_M
z9=SVC#VpR4!`zID$^zwmLHL}#^gHaz>Ee+WNNDK!|E3#&s4BHDn#hJ>F{b706KW*H
z-y8`i{|{S#6%<$3y#K?5;1Jw{yK8WV;7)LNclV9EySuv++}+*X-68m&`}zLfv$w9Q
zshL%~=3vj%Ufrud-IqiizhR5~Uz0?Y-Y!1xw|{@ASgV5TdvNaQo<XSanBsq1_8Drv
z3w(M>%_Iv|oCQT*LUl>dZ^RNBgVXZy=EVxo^ly?*Q4Ks27B2N4W++v^O+SIZOb7UU
z?s0+ekbuFD+XNr$93O8LiUO9_qiFAU{6Mm;8q6lH#@}(x@+FaRdy11J|64>ndd&6}
z0l&&gf8?{JZ2jO&OW)<TMz0iGVOHXW@tf^Qby6uD?3`bZ-IiR`{!Xhq=W;Bz|7~;h
zALDf`cO4G8;Gpufs!Z&dNG5K<v+Ia!*0$%Y;%mp%tfz(gT#IF6KuE|R_l`5XQ%CtQ
z7eX&Ho;0KP_9#$;q|mS~bynr2w#~No=Q2mg5~s?0`6qXuw55)$*-z&UtK1$Ig-(|j
zf0-xbMoykWj=z`hbpD8t^n-(gUL!G2@Sqj*u;d4-!K4Gou#ngz*Eu`akp;E_LW(1c
zg{*n$oaS^9RL#tU0RCFb%y!(qM{>qAiHwsNNupq;gp#PI()05Pm)~`!$a1yoFF|xS
z60*hV09Eu(a&)gPZ++3?sdtA6?NS}Y1ffZAnGmP}ZOE(DXY;m1jHn~VOJ>Ob!WIVT
zfBb#Y%*CfvOd~PU_UZnHK8EaZJLt7Fb!gbw1w&<ntG0K5GWCkEK)8|AaWwQ-3xODJ
zbg20KPF!|(XNo|@dzc>Iv<Y6UXkur!kX2tnR<M!5%T2{qdHdKH3L4bjNZuCi9D)Lt
zv<YvNQl^r8txoa0nI4;yj@gIXX`7@~y`r@;DETdJ`TXErcju6d`Bt}46+7y~5qjWm
zZS3Ti4f`R0E*W3ASOykINZy_<bjl{$_9ykz(t-(-92RA*&O5nE%MN*-#kHBvQgW~}
ze4Xvb+Wf8*ulAF}qo+j4H+}!Zn7|`+i3!ICjse<PFof#Ybug{cm-E!mCPa%5=%<0s
z%LD});RRn@{UbxY^%P_tw!yL{H4aU5SfQnID$W652mQ`Or?NouMU`4v7H0j9i_YsF
zw!RH!x8{A*k|TF`MJ*+*szw2>72S6(E2wz3B8x^z5Qd}d!jzauYhotEwSJ}aD7>+q
zpFEs34Hy0M3<4i&eB)Q_lub1}TT*!B*}zDi<hJq0yo|-g@~F!0$kR$tGT33^0;;qn
zn#D1&**^!#s+eN-GwWY^-r(R6yI4c0SuN`{*|SAanqk0fM7lFsF!rd;LzexaV-6m6
zJaqU`jqx!Uqs^vZiV;&RqewCvE*r<lWx>x)*b2sh%)97#q>jCD5rlpcwO=e|4crYU
z*?XzdO$`;z=S2QtjpXQ>`X*DWZP^alDnIRjnk8y?Rl1`yv{;$Cl9zn*QWtFT=wX9N
z%;>qZiFXtd#@Y;IE|zjE>nWE^?v}8a3W#oA$GYB0F0{J2&va58rsv?ak)xHV`8Dis
zLL$t!kegpq<H#MvmI<{4NpbG26Tn8o#G0DeSuPY9>F0OWygb}kCEN<Jq<f{~sP?;o
z2P9F%^;rY-T@LXw&%8k-i5Zl`r#XKKe->rj1H6~}(>WvN%w9HK887=gjpnkdO_|zf
zdpRF-YA>K(oQkEqut|(^DP#C!Qkbf5s-5YIZ0BE|tHLgALr)Z50<Ko;HD3vch>~&s
zOVJ&9HQmVdvPIbw!u<igXHb3)<You(z#i5L$5-1s5K+PT0aB{CC3xUo@L3Rb4!O_A
zKtar<(;`b_h}rE%*)bf{IPO6(&forOnfaAp<w|(tjN%c5+Krx_4Xr?%@T)BvVe}Ef
z3lAl5A*adPh>-`LW}{+}<+hBv`$^p-d}KNt0cw_moTVf1-vXO%p(UA6{e=L=e?4)K
z1KYvj&Jl<d5PXmNw&LfFItaO#Z`=0_4VapmhHjJwCJePESZ!ky@q$`v=Z3z5;rqQe
zVFW|b{$T_w3hZ)7_wcgWtLASwe<OnPNFujp1AZC{VJ4iL38*g!s`-jANurb3diBfu
zHJ()Oa&vN>H`=6n^5`hZNNxkr0~cSG@6rTG6N39!fuI)DBdb!9cyohLy87G<2s)Wn
zx!PK9=q6<fR<JCGBTlDRq3^#WOj4Lv(#?ac7|E$gt%L6Lonjnv)I~V-C{FAx^yV4s
z6pzlx;Rd=Kej?eWlfhXL)i|k}W}t%;*xscSxReoxFn?fqp4`j77Oep|j#X`KsEoM0
z2<Z>jUk_MpnAQ{OO5#28Jk9n~N1IN`o-ti)E-OXAt0U8vZYtdWuCqeQP_}uaMM3Vp
ztax6di<oz{BD#~Jqp|*RtYFASiRkcHW>FZ&zTn#IF?E>^vh#24Jei;#=!PDcyv=TV
zmtSUOBts9XR8IKpAPm6l3vV9`4GOwPZS$k~%32^`4<!yvxg;Ixd)R<fV2s;$??)BM
z!Zf+!tjd*y8T0ob=}ef^#HNYoL35-}B$s4MECdhO(dn;Sw}xD!7<qi&O>w0k_iRuJ
zS&%)hD(c67v$=^XZ)Nx~%*YV^t+Kl@ImLz%a$Zwnv?{~0Y!Se^8wD9!%;wr7tFLJA
zHpl)&lVWH_;*%Qbco<y^Q;HEqyr4e6{%U=Xa3vyh(pe4Wt8!>FbD@fIZFk&5{aC~|
zfXgv5BP|&LY3!*qWVSGR!YnOG|4X-+=e=8ws@Ipl(4Mz;QUXzeQQU_)#}?vVZp5EC
zC-_3J8%qvr(gzUnhT%shzSO~I$D|T-b%pxlN01|~x{UG1FDFHlo!t4eo4L1s3elBp
z)G2b*KKhvW^dh2j$741pYg+b0__NehPLEO1HQc~%KG?c%s^>gX;ns<ee(_`uR<5hl
zU4cv>Q~DP(5-~2))A@#{SgBmae1fWl05MrGzH*#!87H9SynUKPO2(Re4l`8rkmqWb
zwE?(~^0+#_OP}0IE$mWEIw3fngjYWtI^l0Pa&5_fK7H^%hs|!y6e^CNFLZgG{3>mL
zPtb1LRGdzB$Ny2aqF2$$BsNPgnJ4~^kQwLvZd16_{!vCi2B&s^AyN9o(P+<7CEJuL
zjl3@Oun8PYs1$PPf{vOG%5|&>@)WwvZS-?2K}W{^3wbN&kw91{$XBISJPi-G;h>B)
zWa@s#%G|E`s!85nr~V)(&5M*3=Ais`a<h+@a`PZvKf6s3;d%bBKqJ!QK*`kwz0}UA
z#zT)P;duvs&rSJ=H|)zMo}h&@NFV0t)vnW5t_mQ~^z$zba!M;lexkE`JcOuFN~i44
zX$|C_Rb#PUbg9&Bx8=J8usn*7awKqO2t@MCFXJ<~$r%8DO^2X-v0n|c>q$a89?u99
z=I}G{@T@s_NYaqeAH7gUFFYGJMny;O`XzBu{P|sKzXm9x;`0&+wuorO;5bQfiyB=%
z^#aZqITHCP^uJ-T#bM>%{PyS+xcyMRXMn*@Zq0}9G1W`9Rs9o>g5n@wW?eXPBn-dL
z&CSG!gR9<TtSHM!Do;$Ajj#*)6KYHDjp!W;lc^dbs%X$x0KZlrIe7YRY4M4z$8Ji7
zaw76F{uEcyVSapk;(Kh`?{Kded$~etQ3jN<q6w6#Ylp^CJ4`}W@O1b(m!?!J20lXI
z7|Q4u4{6abmTdR_O0r%(F=oMt6e7&(Zdoh0Pezhk5NzyR^>0`iKZ3m=sC)!6EE+vR
zw>bFA^{A=sGCSBt6eaoc9g^IghA5<KrM2KQwbxuR&kB^P&;qf&VnfQ2rt<5V2LSv(
zZPMnjigO~BIF_C|l|>9PcZfb~f0k%{ea2jRLVw6mK?%(Vvb4pAOJi6}X0b<iTAoJ<
zxGf$*h4xX~3d1j;jSwAjD9ER?|G4^-3x|xl({v6wBq1C3+9RnYr;3m7>(NskN%_{`
z??EHcjolbh--tAUvI&rL_0H#yIRL2&(Y;Xot`r4WeO7y%zsE|*HexAa9rF#j?Pc47
z<w?yCxkTSD8#fet)COMetZwA@iYZsvZPehXBNgs`Ewb`5hU`2dYM~}E-t2i1lFpSG
zijMi*J!G|S)(o79+2qQN!Y=eUm3G}z8TBF}S;Ea${mHmtb)Dh6F?@N>d<A?5bcs|?
zmf}lSp)*s&BYY!lxERK&eR}O;nw^o895A@7hJsk(yx+&43N&t@3{Hr4k0iKs_gR?N
z3PNtOwW34VH^Y-y?4*s<@RxEcv0F86YGb<>&2sIq&kiLNxlmqSZz&<pL}>rQ)nVxG
zu4Yn$FI2lrbUbkVFM8$$u7DHPS}Th7+7p|ON!=Vr`EX7~|1;uS-QP)@Ru?jAHtNj>
zC$IXl^KJR|dWbA&T-|aRnXCqcXVPlneJ*POl8V3kl}-77k}AW_`JmVNU4<5UG%<Qg
z+OQ$W9sJB7cJpWBtKGeLjAF&NcPFPJ&Z@iq-{u}{i?#AR3Il)>?nzEco(QvA^1b;o
zd`FtDrc=ujsmO*`y4{m|1QCz;qs=B!iUW~w(VT_<TbucIH;*!Fw#)wVjXr!|B$gw#
z0#v0YKSUnf4YK>OU~N#?=RuMiUNgDR`$gEX3&sTUbD0q05z1W=>h5|+iC%mu;CBI~
z#$-73QF#YtAvusHA5CZPCORxz^udn_8np)&mQWn<n1;T`P$N%`Ci^|w1;$@oCL|V&
zj79Rm_~4g}Su{KOZ=Sy&>!YPF%VH)UVrmZIXf=f&hTdP{^XAy9M5UrpmHb6m2xdF1
zN*n^klXTzXG6Ok&^(%)5FX;)~E(#~j;u$Z?+`<v64GzH3(|5+JfALtY==C@bXwYDM
z27KM^LWcti{u8jZHA%sf-L1WF)f0G+XnJTZht_=!GZzcE$B^IaGOCu4xCE7=HTBSK
zx*lP)M%#!MKg<2o7ZeH!WAMo*AyoU<*F$*XZ0#W368%%Zvh_fZu^3hjWYn)LsMDag
z3A^{Z5`Y&_8C2RfJf-nlc(7Xnr!?o*=fhW#x_IgrZAS`7#~-0BioorH{S2%mbowwS
z3E>K2E@GHA<;k?Fv@w5jT~T})I6M8R_8UaYT-8#_T0Fb36m0KR74+Ow$Cc0n4TyAV
z=4N4&xfeG9yDYEk!a324bZdv{mP_|w=jtv>Ek)k0MD5o90*(ByBybi(=WODPxAnxi
zGymi2-S-x)uq%H|`+Y50CZ5SP=YI8}$ggw$^<39fNKf>4p}tdGUtb*5z4vi~?75Hg
z-AlP&3WMQHb26L*C^m0;)fS@nDo+)yVp(K&<&t2oA|-85e|xWN;2+PM)ioDRCfFkb
zopL?IXYwL7yM2FTUVLh=1o~&ZR(3}(`eZ`~Pzs={yzT7+6eRsLm>lgDTwX5o&3wjZ
z_>}+n#Ue=h)#jO&7fc0FQ@|FyXMw8@#jl(}uqV<PLjswNQ|+5AS5tSYHzJ7Xc=})Z
zqNq8tT9cwIX5&XkV~_Y8WIB!{L$SJMRO<xI(!)EV#!y?~ZY+!Z(xeWfgmjt>iGE`Z
z)p&s9Wbhad;cl_GVNgs?G3mF8!=Le1jn*!snXMoh`+KyG!}l~{VkbeJ|565pPqMSb
z)m^Cl*WVJlL|gGkQS}MVGzXopF%r4Tw9u>9x68PP-%agDRmXX0Cw0U9LlcNYV^}O#
zc=Dn{#6`BQdCXG4KKFH`IGvqI1>g=4&;h><mJ+#Aii?+<N(~0nPW_Dp{pTMOmqlJ4
zRY+HE5M@?z`kP<GKe0>Djf3Arw%5;e*7pKV{eyw?&U5Zpw^L_bIO>7;xjE04FNkF+
zN%!c<|J5cQp?%95n+uPfMX9en(>y41lHkLHMj!HnvdJQ%7$Ky!&eiX7c$ooOAJAzi
z9nbm2SSyU^O@@9IS#+8d`9P@lVA9bj|My=D^<LE`>xXX$dF}665<H!6&=2Z`^2q9w
zK|vyolUi|}ow3!na$!pXc1zN>x0t+M{1aalqR^_(#&1T#q?d3%Sp1}&`%gFdA4+r0
z2A#VqR?5CcoX4i@2`4>Kz(-#DVgb?b%HB%tJdd0(ne?bKl942+oY=d9_8l^Vo9RoJ
zQ>#}*XOsFF39t`>4x-?;SBQBaOxCC1bV-<s*Y_L_1`cHMP7BJlhqzro=Ax|>h~Rd(
zjFtt&`NdP%xh)30qFl(p-a{7k;*`;x;c)k_e;tm!)RtlQPCM6@8^cak<^u})dcBz;
zhRY-<d$Ie!JVhHA<Yi#o^?z6+gQtb0^hq3VV(5UVfT(QxQBQHaRlJCHfYkh-+JC4X
zg3$b@;Eou<Z{bGKC2jh%(X|y|P)tyD-G1DV0zqJR)*iH9X@<e=u<NXYCRmB4%{jQ+
zcBslS&$k_IH0wd&JEAqBHD#a&EBFk=Op-#wbJ<aEz^c`zy+S3bfz>85qr7H;iW&0P
z#F9HRA08i`aPY*2-NW7-K&Tl!*%|6nVAUn=G1h#ABXS)@3)KOQ|GU!StTBQhd*?ot
z53die#*8SAZ_~n#EX=PLo~wd%$Y+S>q^;IXtf{YurI8zu-WWTuD_47n4K<k<IhX&<
z9mJgk@m9CCJHE6iD%0F4H#&;n5}wL)o@%^|O~7BkpRkh$(G9P3QL*n1$E_)Jl1ag~
z6)C_IpIJ#sqA-nLe}~n;s3Rz_Q6!fu>r4WNV%ztkLlZDyE9GmI5%IMkjhZIsNhR!o
z=(`jweWeFfdW|UmcXSD<c;zXbGyv_44vFr7&WG#}6%_O#=Gb{;^K<xw+?<c*0W}jl
zX@V<?2+uLq&~TT0(bLhmHoJWW>pf&`0@d*FfoYF6&;a&-#aX;~+7N*l!IvX4)bycT
ztnf#}(=A1WlJ?reb}4dKYHg?EUwfpijFk&ORJ+HE1l^PXc|iY+amQK*o}7G6jF^ZR
zF7bxy@MuW%o3JV~G~PrkBPl1A?;T0Y-TeO@m%a1c<fv0*E4!tjVBwMAag!P|Yz!ca
z@H(@kcvDd?c(ULxDPFTT=oT&<8K3^%+Zk8|bLkmCc+SYO6@`>7Ud?!ZAIF~^u|{EX
z)HgzED#`s>KHrer#sN&_@JR(sysiWcqAym9$v>kx=U~p%?9!H!`iT07+M!<Vs?nXp
z8c*kuAEkC@v@%`B&Z#CC@nS~seD>Di^&BS(WiocfIo*;i&axdB9ucOlhF~N|*S!Vp
zrH))BU%P^m{#5y1co*sA8^_`ckpq^qu7NNxDR;H%j0Jl3l+54;iV7Xy^R(tS6YnjW
zEbRtL2CTq?wBnRaSsM6R<xOOEsP?slbBe?rC~N0^Um4$a<EIyAP}glRWL0l-#DdIs
zA`lr97j54(ozSsqAQ^M;oq=IhH!4i2ZtOO_Fl<;KKo&jb%PCB{GP(aU23*D=EOuzh
zrMJN!DZ;=Ro^9PhKErt#rTJ#~{Q0~!ntNzG4xmK6M}gO<j@KFQ2%B7`u?U$BIdbgR
zIpdjAo!z>h<fX2%QG&}+GAfQ9ZQWr(+kaDMi03{dV=DJMuQMEP17YiPVCHV4`57EN
zHjx6^g0(sDz5ypjjHicvV{|!&H5)<xg8Ai*vOu`l&tlQjAX6ez`jX@i7HD58M+qI^
z37AR5l_B7(W<G<J*Py_g)-(r&;5agfuT`wtUk~@0Z_npCNa&*TU>3cwc?qt-7mW;3
z*gAI4gc|h}x;VDynwi`;Q^pb=+bw)hfV@U6%LW|F?BT|*mj^|uQX@MvB|F@Nnit-H
zA*P7=6Y{qSiE`NqM2;;o>Zbj(+nA|=NxpUm1^d>P{NVm|nm*HYb&a$}Lm_Rk3_}A<
zRW}Uk>^asdQ!1u|_OLHzU9ay-*Zzt$t;pLtNoeyP6tcAjL;`8-N4)Gxc=PiDxDz9s
zqC$`zB<^E5Gt8DXljf#MNXpgSpQUXJyAxLz65X*rxsf^O1typn7YOsRQ)(|SmPO9R
z*5os|9b0F)0Pcb)s6fyW5L{Y6(FDBmA|v|t8pgDB+3#f(SH_RC$>+hTfxNk%RZ#L|
z{Cj27@>d&|@(YAySYCZk1$xMD@qpdYzd6~jzE<nz^4*-#Tf<9(ZpqrSvVU)m)fUJ`
zI0OBlX^0gXrlwLw*tMA_q~kRPUSjC4CFi@R+GR(mK%0-b(Tb+?lQn{>!$;U1TYbo~
z5cXW_)hWd1nNP7#@vHI0n<7kb+NP<CkJLBl*7~C}YlF!%r`puQM3GWOD=3muO5#Le
z%E<SUT702LgF)}NROBU8<AYrVvEx2C1~1`C;ww-K?s^qN`{EW_!{8usdZ`~?{g81t
zoXv#kfUXGcxAYVaMHd5os(nhHK$;h%7x3!uv>k*gwhTKnDc7Eri)B#1nkRl1(&?eB
zXifah9(CrZ{<V4|cOqBZ+Hfg+@PeqUgg<MVnV0!BOvMf``D<K@MW>zcJ8Q{VGmwyx
zoA{9I4W#p1XY|j`=;pqhY>Am^u9VSC0xebcCqOAy6>e~;z{Lg5W=VoPknC%uI_Bkn
z+u?cS&c&QysX8K5pJ*A&!Zr6%L-mSa1}Y7z+iTAb@|uROl#q*|^>p&)v5AX`02B0&
zGaqU?jO^kyRA~mhzy;w7GS@&OH7H+MMTW@K$X2HjN9!%lAJ-w>#Ok?qxd~@KVjt0@
z0sSc{9`Y);T)8&zsP1SkC@c+p`$)3?@FyYfnw~8w2Mk1SludLABWMC;9ISsn?xWj9
zOVUYWRZ8G=<bwy9j$Unuis6ZZji@AGA1;3p|0eeafGF~~DSd9G%gCDL+B_U~ohD0s
zF2K1+yHIXbEak_7el>&?^0PStE!d_&YkS&e=Lsuh=9x8qqsNkt{GD;my0#O9KlMSA
z%UMSYIh<mGVCgahCKPPt#siXw^*)#BR8S!T{_ypY#3H@iAYlX&)=k3IL-kgb^H;-8
z3{vWpRAF(Ec}lZopPAL?z+RcgcphqEULTJyQVOI6M|?@EdAX|4(d6MDbRK#D0=`o-
z9a7U_$o3XSbU)%&L(>sLqFURD%$csrTKmAcR&ENGZ`Z^uY{So*cgyr3O%vJWh6&A2
zL3Nl2!g8#-<FRy-x)Y(!v&YPL_pQ@0EJK^(h)T_ybDT#}jQ(JS^r{~&T{o||&a~xi
zv1Lj@cxiV~^5EN;{Y3n}B40aTp?lb4fwpO3MH!hsHuhDU@x%)CO63`I5U!^uoD8p7
z#Zm~kdB((<0SQL=M?sW$CIEUZ|C{PND3@+Qcf>vTUbna}$BASP2)S^9J#%axX$Sf<
z?MT24DN8Uf<o?;Qa7%y_sUGAs5mDj?3H7zyZ>nUJy>0<QEtXGX2nKoJH<b=+jt@Yj
z@I|>nS)#^{Hb!de=7%W0upRjZf32#^ebYou3D-t~W003SBFnLbcBM4d2O5{9Sb4qq
z4f+hhn}m30EPr1bKUJU(e^DwWrLbD+O{|wfFvPGj;*EWFxU$adSm?_R`O_sCO=Z2}
zJ~nFki9X#ANd94RA;2{W=@1F{`=5Hg&p5eEBRCRmsam7>w<Kb)JjP2<4G!tX^Lwzh
zzTkC43-vT^*;?Mw$cX9G{Qm=0RKW5QI0mgo{tksNGXrn(|L$b%(5S-|PNZPWZ%Chh
zs*0fmUdi#6QjsG$<w8)3sl!O!E=jSB#8F(MG&2qT=sW;WeeeDYSIRkMN}dna@Vz^}
z)z|-t%~~ll&diJ~C9tylZc%BUkEVi53=e}Mu{m^e9RBCEu9jr1mG-6s4S`xEgCuu3
zLEz|!f`0U_`n&2O-`bCP@m%HVCO^1@&&=;!Iv5K~qy{NOyT#IEfBYfZo6=Kk`(9ZH
zG!L(k$KZjqzuS%xU%qBfa>d~kb^jEYe}Qm$oP4|QghfoYQ?=QrLHue%+V6muE}0Y6
z;d=Kj_89rc^Di$4Poeem^1iA~C-IU%05wWXV=D=d-SZxejyr^k(&Xn{17ru^J^st{
zD4%oW?gD5frE;J*O}sFTZz`<yRZZTlM2D?{GWsmA%N}|;pt?vKOUrR;P|CSmo+i-b
zdkNL+g00uJI|HS*ADf~ogrcwZFH1FCVCv|odhx#wq@1q~B>Yy%*DW45c&xvMu6R0l
zti3cDZ4P;;ge+Gmp<lnQxD0x{+$P2UgIqcX|NlWQkHKFfm6-PEGPd^v-KiBj?KK<L
zLE1eWoEi`AjVh>j=kiQdNccPWQ~6Vg6v<etF&n;&KN=`utaihGi*Oj{srt=xN))P>
zUv;E^F=Rr_;Ue%Z4u+d_Vcs@YebbMb;;`nf=jN^g?Jr+;n)B0n$jRDHh{5zK>jB$C
zrDp49hH-$vd2bjvPLU3sTegYHa&RB2XNtDR&?}V^m^UG@O>SEB-pptvCc^l?(Uyy_
zzOl<0s$+aA9xmfB(dEHo%@qA}C%VdeXtX3A4Ak&nmbI(<3C8Q?LKTIoDppU1j=op(
zoDX+LAJS`ZFA-23Hff0bB>W`eNFyk!LtAF^Vb6fOrq=xK^kp*qUmv;u<QLKoE8;LR
z4b=_ReXP{_L)Q_6=*n-#{?2+DbQ%;2+ZhhqJkh)m^N+L+4b0WqC43P!KMcQw*q;tk
zPek}+km&zp=#-{kKt*&(6R8P~3i>~}lcU-P4}!pagAorCROCFzn-T3N$+rRiW&M)!
z18(Uhem2&Mi<kd)jiWa2AJkKUNpU9;T4P9d&pWP%{?8-P`edJ1^ootReJz{$3AYB0
zn~D-CTpWb}{r_aSg$o1b8vg;(|DO!kHx2v=dagi+|5M!J6&2NhzDB{Ru}K3pKj!}l
zW!$rV7O)QaHECql%D9dBe?pPkfls52eBvd;NlzP0_+UNW0P*Qa#0Sv@(Y-2u-PHAn
zof%7E8rS>Bvd&Z~#{O6U7uHUxT%S}{iw_s_hxu_&Ru2LTyq0FbV_67STPg=mdH=oe
zEu{h~ONzzm9A^a`D~TWrlHmK7P%!#{<q@TB#xK=2@1(G6V*mg0Afpv&-*|NpG0_F#
zzN}Pt>#&KZnQucXTxJIrEH}oP&;O8*3VcW-p*7iiMhuux7+SUA6_#?r&hC=QUOn~M
z*2d%FUz|j)n0C3GW5jH~y!t9`qER?NV^r}BUpHTnnJas~MiZZOhytJlJ+OU&`ex`O
z_^>|j@tmwdMLVSf@raXP1|xwc{uwxtsVh!WPM`tg#-M@=-!Hh@e{kRBv$ej$*Hp2>
zkQeX%z=2g+RPRiv9%8bUvXw}2_2ftQ@*xy2PBY22rZ`vqKfE%65Z_ixr|j{9xMT9v
z$l7Yp**I2KFJ~NJDN!(2ln52FhqI_>>b`Qe{Mt7ww)W_w12x^8evtgHHH$SZ_KtaJ
z?W;BGF9q*AA>k+yKKnq?!M}YCHj8wY7uQfJ>u3!!nzi<hz9i1}LZpy*QsXr5d7{`6
zl7^v&j{@&aqT@J+m%k4m!h6|@uRF?ueVxnCoajw>NPx3W1>8oWpuW&oXV#Bde3FJi
z3|G_g=w-5DZ#mk6QceL?{94M>dl<EI)%;)n#sxn^{2EbGZs7qa4bN*&h?OAbzwRgB
zlOiut2d;{vRxS>YzwseEiiTWDw-o)4x+(mRy1_6HRF+oLL}ju<HqoE>&)d_Mi8$c4
zKOP_x)uv#3L_;W~(ElMHu}c{DABTg~hxt6!e$o3*fjQw}o>}a@^ZeF7dredzk?$@y
zsOXfTT$#Z;*YL{;&mqlx<wVLjxh^<fuu|I!ymtL_(pQNcDt}J?l&KnW{73D@B3P>u
zbna_8H4T^LZi4nzYcQu)!mbfO-{`HD0C2AP3SPw-)JaU3mwMU%GYsyB@u$z^HQ#bf
z_n&lP`(=+9)-~bY?7_qT1AOB?aj}+B5eD6PwZtQm&f~#Ghf{i9;bX&$s>FNY&uQh>
z%mZ6tdBb$vyMSU)I@=fcfvfp``J1zR%=X`};>XC+)d4{s`VzedZuyptl8Hv|boJ3q
z^dR=Jur@*7CI=Qo6p%t~cFL~b+fh(07zS$16#fx4FZjyFQU(Z}W!xinqKKVyP_Jmw
zBEv+{{r%W!h-1A^{drewvNT<G-XG?L{;CHIiIzOdE6q*C$vwfKc%TkW$veTIC_r4=
zYmAyXSx`!c0T#-OfnX#JcIh4?V<@j?9npC>RzK5m&^v#2s(TTY7hmvdkoZO1|06fJ
z<v)Le_qx^{yA0j?D~imeMjI$iwRXvriW~fo+nDTcle`e;1y-Y!&5amzeZw7>Wv`64
zOZd0NCrMCeEE8K<E(yyCdEoIw?I(wWL3$1Mh&*+Z?~d_9{HBxk1TNfjov9QY5R(%^
zKnbwuZHgnqp$O|$E3!Zg$%b{c*#+#!XGTw4k|#nyu~-()oz9&9JJnWX&w^FaM1`Dc
zW|gx`rG^)NZ%O>~@c$4IbXZ>nplX%^j$_AH_LMF8XDBE>@CAYx^ksOqU*0LdpTIFW
zT1C-tH_Mq9l}}M-rDSgJIN{eyohOz}Wj))#)*fct>a>qsQ5<PVoS?CaA?wW1|GuOO
z5db%M>{zjndpBuIhzb>2(zno;H7WlhJQ6^|72r8bKuy9qG+Ia9$*hB3-8)sshSbTD
zLxOh~n$ygHjp0<V=)d5I#a0)bPp)66dV5=pwWuyLU|!tlT#k$a?5>D3|M&`9vc>M4
z$*!ciy>u^>Q^RHu$+36`1@fe&!f|7mznWGSNQl3-xQ4HBPjcSnPYKLNmYdTlQYk<P
zixBy63i5_v3K|rFK?I!*=VcaOTfDU^ea&Eu$t1uhBXS?b{F4VU{@Tr!6>If=>c;O|
zQ^UE>L>|L83)ykC0h!z!7rrB25A{pq#zh*R1MRx?kB8^ab1hhhUG}CA>a_cH+P~5v
z6U-y)1z|pOLROq7lE;}Vr&;O9B8WCW(`$wCxEep-Pls7|^u7}=*G9COlMP}{&$w)I
za+U##2_i}bjhub%44U~S$eq3B2oQwS70_`hs@q|_1RD+Sv~}oi_k{+Mwb#wy=Ylj$
zag6Yp_;&q%^azYR8L$)mH}hx+PYC<iH>~S~@Whzf3p%mfSaeQH`>A6K*#5EvtWRuM
zaL``oN4MaPw)0r&tWM$M<Z8?IM5yNCR~!H>ZgP2OSRupiUuH;+OOD6_hQFjw8AT{t
z@0$*za_#0iz&j6$YS7$2jkeb_vi*rYDk=GALE#o3cI#D~Gv%|e;iB;U)NxMdkOgx1
z6QD=Peh2Z*ABf3vv0)l3kw-TgKw`$)UO6vYcqBD%*RF)Y2<QmVyn;0%viO;NQ4S;<
z|5&}cSi7WsQSyg#BlCQ=Ez6hH8iB`n<p@v;7+ew4O(vkLpup>(D^C8{?@5nI_F^&A
z`BD2~KAhp^2lLH#!`>!RXTLacl#7##Tzr7aqAw2>eEugY<TAtiyj4Qh{dRTYm{TKi
zv_gdcp@Rz(g!;x-jR7N>gTuUAlqNv7%gv=S_S4o2LQ(@9{N6<=u#jPiA=jjKi<|J%
z%ItLj)F*2-EO?xxoCtjcHG_yOay$C`yK9c=K;RV)F`1j55j$33c?yfX1%j(puF2U{
zJF7T2g&__XD=e%CI6Aa$P3&e6Aw|2p01nn=L$BRsxHnAaFm=LwEx29K69rK1uSvXC
zTa`6H2w7b+!XCzDk4(iJ8)=@O<NKlr+GHkFcW1?V5EiG(Mvo=&jeOXO)jnOi%rtJz
za`uAl>(k1L@myIcniZS)bcE{V^K<N)C0Ho5OXGxy2;*+bTJi+5{ZW8zR(S;Llf1rB
z17-e1!U>_?PRvZv*Gf|<;2*&{r#ETT@cdW=pJ>pS>js%7R=e$wDhlele-eYlEX#C@
zI==^-IOuvTt{~*5IiiHh4&7Z^tt{6j77Xnu@kiNS-^N^`wsB?zbh^jWwKZp6Hy=Y+
z^i6*)3@p(huhufc5z_{#@1PR8oGYlL&|E9r$$vF!WG7Rou2<;-IB>>ijJqX!?~Lfz
z)!&H3?Q_2D#zKLQyIbI53q*;yO8U!b-uXRhiyLCH@o1&Mj4Lr6zSWqXmo1I0`S{G~
z)y463YOU%K>%dzyn+e>w_~u~AIsPQLYO%%g7t6iH9=elFPVQvoqBPU~3P&;%mGwui
zQyN<~%jR>5Pp76C5UH_A3FT#`l_wkCzZI58Tz@6?t4gCJ->Fl(I*xTKH#)WqSIvjk
zeC&O=T~Vbx@?@gfGEkOD_R$$WS!3rw*|n#scTAz=xAZ3Z(9DPoWNOgToQt<;qn0k6
z>`y6wjI%<QB~-WmRZ;pZA_VW&m2}4x?v`^S^Z>ov@hr_p;0TQCe(QF-j;G<vYFBCT
zZ&5gz+FqLTljzsmszE1>Ia*mdjf=cdqNRCp6Zd9C<NX*WyA+-k?z{jtGcDHP-0YMU
z<|R~W9oG4mP%#N?8AkgSzeRGH5I^*wI4CX8CY(ReQOn;Pvjx4f-qom*P=X^ao*1(;
zN*X0f@>Tx;^F_|jFg#Sa1Bw(&rTLhoWr9Ua0;P3ww82j1+%ZJ1zQX0n13w6edgGh=
z+;>N!;jHHSUJ!=k`G4)!?xx{YouO7U+>7#C!o@7z8Xi50RxuW61quv$<vT&_L>zY(
zD915o+nE}0<E2kIn3)d-=A^3aAMjJZvRMUf<|@YlD}ez<Y&-k0>W{1wWABBG2##Eq
z@9hB!ZI%}$pc9bXf8?w9`1IwF%%+)KG9dLln^#egez0F-g^%k)!cI&ns8weP=s9cA
z*kCe2a8$nv-IRu?MWjU7PS#qluUyykCCwp>qmS%rxps?7Zv4DY`m2C3S&mrBw?p1q
zW!R<(w3H02;+l%$aCb$EX=>K_4f&$2nDtJjG>l?Uu47>5{j48VtDUZ8MuRnY_wkz}
zb<;N4WPyu~{3=j|2j@FWEE4|X9q#ARC$n%atWVF`T)susMS3q|2Z28nVzBt;ooEZO
zhr``tnQ~S#Ofyo6-)5h?ZrLg~J8m>OE^f&TgkMN@UaB3Q8FtDXp2dN%aq2wL+&Pu@
z3IDf5M>rW42%lOV|J=IlszFjuedlEJojnC1wOUelejD1jPjk#Ob9_mhIk_M;e%TmZ
z!yv}Ntm~hm<8nN-)4|OLeZ5op{7I8WZM4RzcxlT|0VyvS`ro^PCVQDTS>TKGd0*H>
zo4u~(7N0gX<pZu2twcwwR(KuXPf9_zv2NR^?IK=vRK#`EL7=|RLmfwB^=NJ>C~7{4
z>pD<@PV|h21Ix;gb(Nu3lB@p|CHTH;o3O<e>Z6B`JyIyk2q!L*o#eaBIQ(9WBPZh&
z6MU>)NQ6)wLGP^FWChd}eN`%3H!)zzu(Pn_SQ81HSKs)Wm>sW0E5~}YT}k&)i{f1t
zVIW?&iOPMG9_&Fo^0*T2^r<mooc0#<PGzpX(aY}72^I#QHP$h%q(%_SghHD(Wk=WW
zflEf0W-=7GiB+I6`W-e8s*sYy9r+3x4(4xcxqC$9mIsG>^E+QOI~k@3JQa8iA(5E!
zrkXvnQ?`th$?A56q9Jnq#Tw~r@{Jh`hJFJLFR^Aqg%Z+8Hf;FOK3w6r?1Y;Srv4`I
zG(4aH7`H%epm4~}UES(i2+5c~E5kmUiQqy#f1B!^0NBZEeQq^QqFI`BFik(Wh)bXx
zpe#Zv8Se3EdYgmpd5IToqydfBRq=PhwvgZ`E$bPNE$2tB4NEa6lpm3hyH>gPDL-Ve
z0#&*Jx?VaQ0?sB}^JVmQBW!e!Xzm%%FW=MX@jWA1gHaT7niocz4>Dk!Hc#WKA`U6&
zbrXXVOs@d7yyuFa3|4~^TKkS*G%dp2R^9vx>w!nKBc%0QdI?jV0my;Ozd}g6nYz!v
zhNUoxB|H?(hqHuSLE(JuUv8kZAC5sHe#&eQ)6(?|fX0dtur5tu9F)KM2lZ_Z4_ym4
z{r4l>iejtbR18?pgd9KKVmRGH{-m=5;z|XnmTSN-3nbLo-q!JU3XP)dlwh9#ryq+W
zftZo%METt<Rlj8GzMLPD{=lJ-M2M2CwX{QT>+<$1^ifZ?Uf<Wdwm#R{Ikg!o^#2xp
zL>sB=O#Ht;xW?%L#$JihrM0--=(?73VYtn^n$=F=h}7_U=J<v<xAG35{^ezRO;vI=
z1L3e_n#^OUcQtqBISQ>mVXCU<k8Y;gf14$+tK^(NmATe|>ye1d*Eu5Q*6K8-TDT27
zy|%st118KXcF|E;_k5HgDkdR}eevc?4CO`h#usa?Y(@KC^{2t?kc1x?426cjnmCpL
z<`eqh5SzDHTN5;E*}o2d=9>e*DClI1WG<zD3Pa=td-07&K4jdqT8-l^=DZF1P}b~a
z=@#6L>I-zh&sR2^9hfV$yP`xY@+{G=iS|gFllm{W7{?!{F>$-m!$5(_8lTAWvz=tE
z&`P|>=Xs7J=$6F}5(n4cvsL#-M^F1a#j+1>2Q%m?GSBPJN(l<f<YG*z3NCFRW-NbL
zC+CQcF;phj=(vB3P}XIgXA=dsB|u?$<!EC+{pcTX$g?K6yvucu-pQJR7wM40$MFGH
z<u7Od9HhBfMse_BEn^yc*R<#Mnv4ghN{&t%Wm^CI7gPIRJgZQk5nyNOfq0X7j@CnG
z^%nx(3HlX2XrlMre%v_WiYO?vOY2bs^6gaL-HD|GRBDASdSR5Z>E2__%->j^YKXKD
zMg|XP7kGgPT$2zoW2Hlb-}slnrjsuQ=wm9OH7e_E<@5CwWMpLEIBbB){(zABVZK>U
z_Q|*%{vGU(t&jUPa5d6d3D_Z0unvM`joL(C#}FWV-jsU3Uyt<cc?fkmF0b+b;>Dd@
zZ>baTXDYO6DYgu|q{Pj}!)<xt!P09sjSd9{Hh(TpLj8H1#lm}LW@U<&2vgwl<C#A%
zz93`3H~VYAYA+JaHdYRNR(5gTDbY$X>Pxi@s6_tA90(j2c}HmD#FY&msM>t9TI|CR
zf=t^1Plv&$Cha~AfBHq$;|M<Gn%&I+f{<}A;?;7%&dMD5_X@&+_mlSRV4cFfh_Ms+
zC57Vstt!?*B89~$UthC=%2NR=S-WMv>$d2jde%-zUPb~Vc%p2UyMn^$qQ*|46n_z|
zQd_U0nswn2*q)1cBB!fCW7RMq?pCd&kv<%r#$9WKY$#dFC-z{_Rpv{PVNEUDy+cnn
zp1Ruk4)bwKi07u*t*Rw@RrAcTyEG1{<h}4{!eiEcyyvH~{TpwWf@rr-9Tn&pnsWY{
zq)k457=QE2R^&SQ2O^K%dDi=m>)hTL(_P(YCrvhmd1){B_3^=;!BAm#2#=s`4_SI-
z2*Bg5>*yK{5#4Py7G!sW&sf>=Q*;xT*Bi4I82Q+{nt<v*w~)o=`7!jcrFa8a;n9mb
z&~FsDK;>aSoC>QQyUkh6LGTUE8IA@jJe)s8IsD4KBV%_%ocy9bJn}yI*#I4+<S>bK
z%2{bY+}~Ee8srTu(~EdSiz1ZRcDxuN;A&+45QdL)(|3{<j_XPadhS)o-l!~MPH;M!
zOVoLq4%MiE8)D~!zI>~52!aQ$UC)<gA#Iyga17k1thIrT)=~7b(5qH)TY>TH61Idc
zvUY-|vu_?vjWZu_<{>=BRmx@R;pFLCQ67F4n~b}7oP50uLj8ICf(L=2NCA`yAad7#
zm8s8K%UC5DoxCH2Q<7iu!<p|>1MzLL&6j(p%>ULAS`Q<uY`Quuvq%Fk%zklS-c^{I
zD;++eS0d`XzE57B8J7Q=^zWeA+cRZ#_$-z-)=+B#96h$ON%PjV&rlyiE!YeA{@4;#
z&1SQ<O(yN~9Ubrq1Y7Ivgx2VA^yL{2d*eKvL;59HStKPU%~D)E9`8w=&;Z{3J1+>^
zy_tEC=qv(U&)=w|iOB$nq+A1SilMcmpTz#|ywjHSuSBhOFIh7F8l<Ym*@zgC1l-13
zasI3iK@a>mqo@R)M7hS_?8cRmSZV{^EtxD?DJ|(VDRy)n-F{2v#t9xkqZvNQA~B|j
zvCrY~jk;kBhTw0<(C2(l<_wZq;o20&W7NB`L3eU+&#t~udnN+R5sa1L$mnh+LwtS*
zd$^Bc-BukTzV%HU6o-?><lIgr@JDz+J7@2pdsJK<iKU!(Mnt4cQLs<=VIhyfpmt{*
zGh6x8FlR)B2zTf`7of)Q=9Xuhocg?X2J#Z2#qyB{x%`F*^!k$!Srgrk%lhW9sEDgK
z8dSBFpT$%Avg`%~p)jk;86!~CPRFSB>lpwX>IV7~$ediiQ<i&Qun1*Ui|HNY`FTW=
zEux`VM6S75g$D2<rF%^*>r>UM3Fhq|{vNVWs7{05UD;}FZ>^GRO|#2WMFugncK6d=
zAckX!&3^g??}G-!0}LK&w|BbW4o<oRn0`p?nJ^LY-q8SV(c1FWKWqGXe{BCFey^P@
zi*ytTM4#u7Y=My%6D)xi!i>;MrVSH$bZ8kRJPLa#yOrPXx)GD>L*JHS5ovlwd4ySC
z(WJ-`Xn*qd;c2PgZmxU9l{PL={pp@p7d`X7&@A@BH`Jy@d-E{?Rov>)n8ImPP^t_>
z_nhqNd+r9h&;}suGDzIeuDf*Ba4Q2lf29F7p0ljObyvvOmQbByKFs90xn!5W<cGg(
z#Ow|o-;LDF!uBnhc`@O4P{(G=YqN<}nF3-syT=TzpuHNlnyVjq9ld|=2A87PK0mas
zSZ!%;<&8~Fn9RSeE?icvZd@ihFSWpKCgAmU)`|g-_NF!ELcv-R7EuIfGAV4Tc2Qd8
zlEu7c;PB_?-G0(k*t<qEWkajv^ms+n-Zq{e_Ko7d@kf!!$s3L9TYpB0HH)Nxv1zih
z$3vBoY0Cy?-)mTX^I}EJ7*95OO5N=?3*Bf!BUNv)SLj-XL|TzCd$}WqK%;(JKSDA{
zD0l&y$VF7lCcMimbDm&@5crG`I_%|@=6M9<On11d^K>4+k#!YVtO|775I`@hVa0d!
zY7gTo&Sw(xKrU8yH=25<+Ul8d>!?1gLW12LW9U!;I~%vzHz<K$4%3)f|3(5WzPsz|
z{#D$uQ$^hk$p+m9dA(QAJkI7;K77dI;3Wj`^mtW(`rHO@X|fvaDjc10r#dN)83#XV
zI_GCTOlagTODs7M77dt`6U_DRUaPZoKXs~GkDFHh&hk&aJXJSty?JO(Oqv=bZ~4-4
zabQRsOv#mk82vNJy4+;Vr@nJ}jqSQ~)~HA-k|Mdu!?qpv%o;x$`R^rEI(hsE8jx0G
zvG}lz;fQ(WPMwl;;+@(;>_C7T^~{3o__`)M^M6-ka*7<@8S=^=-XTvN6)Lk_#teVo
zG(FahoE#T9DIA{~eLF7?5IQswE>bu&84#5_H2EX@zvT+CXSHL976gJ()XLyashP0w
zOsSc#_g`OFWNH-lhzlq**pJ=c*6R8F(YWql^$zf}SiU2iKcwO>qQEHRarFLz?PpVr
zS%h0;#j7lfD4$42Xqp3HLE@$4jp5{4NHrb7jJ#+4wGYV$24oJ&jl31$aO3OV3b=cN
z5#Y(j@ePmPLi+UZ<s{qwy`21ceVI$5$xSGyt!(<4G`ydivI)F}^dlo`XEHn>v7bpf
zUAM!RBw$3VXx<@ozL-lA;+RV21~UjnL%sRFvT_t&7sy*Rf!X&E4fAO5s>-?+MG!oE
z_lqwo4&Hnew_yTvqQ8i4;EUDV!y>hljw(A3wCwr^2Pz~CPYqrWr#`3&cG6LpL~_y*
zDqVHI+4Qi9Edg|BcC1G}B@G0`@$vQAM!F=#4E;pMi+$G*Z!tqlqIH%Ax8?6f71o6T
zcz_5ki3A+y@S>gBb*Mk@oh1fy<FUnzO)mqTxGmwCvVx@*+Jj1@SVJV;e#Zyq)Py)S
zk>wh+*`JTn?|adpoWlnQO<L+PNdyi_J`7Tu#y?0fXaHM*O66+JQGdQU5k%V;^8q?+
z+oK>Ox$o&vTzw-Us;D<}GR;r-)I%&6e@RGRh3f`(0uhD@0U5a9@#DM76#>pWM#8SM
zZw|B!VifVOk3B%EDo(q21!2dObctSTaq;&-v9r?_^~7*xO_PQ}0X@OcKK3A99)2#1
z7}V+ELjX+$(zi-8Bs9Q}GJ0}ydUjVyeF(qq^k&KZyHmyD&qn1G-!y>SNOHI>t;i5=
zF6z4AE9MU8EV>gf7@mCvA6AgobhXTN;$*{}?-Y@$A1|DNIo?KfYWRF#y)>;Hz9vqE
z4jE&o!4aiL*GjJ{)%5PLc3xv!D)9>EZMLyl1D3Kut}d&wPYEtw@GmAO9!xG;({s=_
z7kUrR>U{;xbas0EwT$XYc_zWjp4uKZI5qx3KBaV~nOhW<6XB0xAR4}h$KW2F>?;vf
zv&`>>u^ymz)oit4jkS~36^(k*^a*+@--WD}h$3F5g3;M^MQh&j<5+(2I9l8!bROBf
z0J47VREK)!;rc0JZakygs@Cdf#ZY_cVuJ_=+~%_jd;141v<>HWA|*zueW=Otwo-Ln
zC0gOEDJxiSZ!NhppT!&Lov7LDAlDRrWo;7^fV#!sJb^#KpJcwPKZZZR_ic8zqh}Tx
zdggRodvqi=wL_Z8V@BTe<;x`ad}MqRSUcd`9N+ZwtQ<QZnU{rN@~CUN#AS>wZh{XJ
zby3whstqVhG$ee$kqnrO7W1sDmspQew_pu4h-@ic_zKj1KYLPs1!_uuEgSkxNBQw$
zSvTY|wzotbXoEV!lj`2)V2=$|rRLF2B8LU0&CJL|+6pNJ3|7QaUtV;a;{hmuaYq*e
z#z!yVKgdtrfE4-L--Q|C9P#!|u=xYvq4(js?F6M4mxXe)EJnxDK8><7{YH$!S6Cb_
zb22|%=9?z05ag2d|6b|4%6xeS?nCz@MzxO6wnn{-;qoc{yIZdHAj3L9T34;{albrJ
zzUE%{4bBJCoTBYO@l1dQ?TQ6}KM$oe0g|uC$f6#Hwoujjp7JN*-}noB&kVv)EW9uv
zC6a1<GMF#=);+undcMz;UUc^k-K8vl_a5~X9*ODZsXqM?#YK_R$)o396|#}{>5Y_L
z^e&cjrQUp}L`@J57k<u09`N+t*9obU;i^%DBV8vyxX3DiCxcky{yb?EkRs4VX*p-6
zn$E0!tw9`g!+rk|J=j&@G$DDQu9^8_ATR~KGTEa~{_N1Zzd%J`)boiLhKq5!nS{t`
zov;wc|9m$LzPezj%_=crFOI{MjKzqB9^sFQm>ZEBpCBS8<D<I~=LL~eSn{5w&Vvpw
zRpZLRG3<^K_5zlE{t9;pG{Q^$WX87N#4;ZKh^@(|bRM|l4o=3lydPq;`$xy&kqgHu
zz(=aQ2oJNQyu__hGV^?oc1kyMRNbuQn41gH+S&3LHyC+_{$lU2cG>iqTU&i>p4~8M
z-hs|{$^}2N)kN#D^_%;lx2C-SGfoa%Z>x!|SiMXURhV5nq&9^Dc;~KVB1K>nG-Rv_
zejc$qQ9$XO+t>HOXe*01tKWx!``g)RsN9>b?s|Y6og&Fy#E6U@H!Yl{3)#eig)}s6
zlI$wMNV_;*sN|nsXQ@FT0;ATG3aL__IJ)q6HnFWBz+wZ8lvC!QbpuqfrOcd#RHnpv
z+|$-lWD66{UJAnl;KA76{ln?FQnqYp)@Tf)_l?6a;}2-x-7UnMB*e--S#MpbPN|sV
zINda{0G&}#a@y1|(Rn0*G9<DroFW}QMzdQ!Q7-pf?%Mw0hGm|ixKJcUKJnY}5>aK$
zP3?3cP3Bhu5g0cW3O$YMj$`jm@vF~~<MVILo!3S?*l^yOuN2Nq6PT5&3RtBT`O~)Y
zSMbwR(`&(c$xtkCE=Nu0apg(V)N_o#+UR?jN7W}ACgU}jr@3YQa|y*m7~jjb{0PU@
zFh{i}O_-L$@Gp<Lf1klAG^cUU6;HU2Ysjj*wm+M)mRyx9Th%N3M@+bi#I`9=^^-oC
zDn9i&yR3e|0LL-!eA!zC7p0D97-%1$7-MkZ<2C0Q6sdYEklf5MdS<Lu)i$jqZVnq(
z43}VSemNA$?>Dp5E4}i@DosqOyW7<D)Nu|?5iXD1ZA5UFZKFkoKPTUjP1NGCFY-$z
zD%HYR@c$Y}zxAo|3JYY+V8rkQ9mAU|J*@jS>%|Lu0g&32<6}zlY`IS_mKt%G7RvNv
zVMR($a>-W9FJswX6>$)F1_+LO6sGk&7RHGTv->ausjKD@=*ujDy8n->cMJ}s3%hnF
zP9{!fV%tt8wr$(i#Om0#IkA(8HL-2mwv&_ReZM;Oom2g*tE;QJs(bIf_I=%Jbt=V=
zXaTl*UqR#+sd@^WY`;)5*zP1Qt~<{&{j8e@9si{P8dgc9=#+)3s6n}DIyLuXw4_=6
zwf}^yBF@Ecu{S6{CbS2A`vF0_J5nPkC$>p#fHI!Ex9}n|pv`tuqT=sGIx#V$2!5>8
zz}VZh>9jr4e1OdzatArZ)c1!iT!k0_l86)4uqdLg|C`bzLTJ0~AO3#)lxOteCt6l2
z-V@=qd7i{NlOH{1Me`Tt;}<fAIY&|juhbdgs%^_hmX%uov|QeyzFz)6Xl!=c8uNAY
zrlXglh`*bwP3x;lKCPrFAhoDi(@%9ZOKdXSF75<ygHEPny;1@-+09uA4^bhF*AD`#
zT4EIC)VPewrqEeGmmAn75Q?omSuQuKyiJ5m&hnehCe?=PTZw9DiE)||J#^Z8TMvsE
zCcw^#IQnvcuSOBDY>l>55~@-{yyW}09i{65<t<f{gqK)xr&W~(9jdU$lV+Nvz6`g)
z!UuNrS_pcJbfY3xZZX%Z^%vVL+P45ts36wT#W4nfr#`pk;gqp~rc(cRBaWuWkKaF=
za)ob)H&QVnx$cnQCQ!8#rlOLr`lR`tY=KqgfkHE&^HJ3GG`C8-gCYldkTXn0;Q2p&
z9+z(N{FlEQhRBLqSGn@XRT+|#pd`TLzkAdy)CW#s6#E>~C{upXq?1>oQ!+>{Q_H%L
z$y!2fQdrxz*B?oRnu($xgE(XYHYIV9>+W%vR2%Pk2ZMTBVsfgjxg-^^!Bg{E({&cT
z;tHUq;E0q$R!x`hM_6*e^dmj4-TkNJ226}si#&hTWN{r<$>Aa3&l&FH&WWvoU&2)R
zZsXM9Ozx!n@OSF|$oUBUGvcH)iosQtBBI`wSE^ywE7b=RnnA98vUXo%b>uoGcSS=X
z{-Rqi%=?LFd)^k?da)D}N!+pN_(^oV1qg1MOx4I0_<Q>~$2dK{X5xO*Sz2}_-L;XV
z?U?FTDD8qM`5$#`6V28@`>(6oFCd<A!9Jwq;!N7AvwfdoKQmy;wZB8moR1ATx76)E
z7zLy2ciOqRWQM$Qo;<?iLz*M*yTK#7lA?iI5?JY*41xG%lJ6?OE;5QfB3zyO52%#5
zlMLI?&F|m#Lo^SBFWNijT}JWVuhX4jH<vnEq}neN=8INY!_Bk9D##D095OcpVIm9-
zzY6U*ryRy8WX~PJl0JN+nSiJGZaZ#~;Wi{~3v&b3yzT8TW5+mCgBL24mJpQ#_c-sN
zei4Wywg1vO)PUPc1D=7sQK!g72}Cq~;1HvKi3|1$131Si-Jg<g0SkY$Qo1hVIX24Q
zbkq+*2Nldj_n`3N&kI<W*^~*mK*HGgxkrR<mY!~!RcBcooV1&{5`Pcn^!s+Wn;mT1
zTd(~6q3T7C5QyCvFbp0@NS%jL;z~vhA6>|TZNY-w(=uT7a9Nm{4Zlcx2vq7fBEc1>
zYIvv(k~xSl-K{=&k3uGVH$GziKDUAGkwy|~7>Ed+TxycPbskAD;Ja;;;(~G1NBB+;
zcLy7?h5r_Cx0XG1O=2#2DmGVKywR^?XPElFc4s@lk}tL>7!}W7W2?4Ejw6j<A1tT3
zALe=NI_b)ZU1%V47RH4i2lU&hFfHF}v(p{<Irz=`w&$MTC2}q~Ube4Rn?c*5{e^Al
zEQtEt5AVs4GY*$}dqxhHBJ%hO7vdMgEs#t7o;Mjm8%okE9-PMYYtSS1jA)s(@4H4M
zJ<YNAS;ixB?Y#?Z2Z4rwy{az}XMjIfHyuilUJ=Lax2|T3to=M#1fYMC%%_q3?vSxD
z)E&xfCu+@ZbB^x=2|YhYO-Q*^xj|9;TxTcu@Y#%>9~~{O(lIXzwPCtF6sE{+!4|_N
zk@xThI-^k}cAIFlC{ztJ1{X}p@dW=L7~+e?`<4*hGpglq+Fl;$8|}LSKUr<?UFT3A
zD)#~RU8Vc;Thnqi7BHj7ZS+0o6B{qeEN8E4uS-W#vc@QN+n$Z8YMZz)FVV_D#k;>(
zEO`GgPvPsw2voC18P==EpKsbY+raWfmX**i=Dpb3Vbeb`9+S5*L6lK4?bLcNCz7Z6
zWI9z8DdnmPlB^;G#3+ti{tN*zHTZm$r}2b1qPElNsip`@CjdqV3@I#!dmO_pdUc9c
zBXsn1D_EU(KPI9;7^cb{h)6$#(x|_absnB_Y0WQZNgVWj={4da*bp-zIn*V%H2VC;
zeqDo+Y*z^o%pdDU=p9$hpGD_2diblVn?aRHMyss3xRP}ll+2D(nG7>OCcJ)|&|@m$
zqmHs@r<20xumbNNFPhRCHsqW~tJ&^bhZMp_ffxlO4h%W<OdAn)G=IL~;~aV&I`gX4
zap(<Yy_f$CwAZ-#WT;NgU4DPxNw-O9!F$1~r^Vs-Tr?*X)%Xc{|0@J>-?YWUof0Xd
z_d<R!Tm4;T0rq=IVhuC3jmmd7<BhbR64{er)ESJsiI_lfUP(sUEp}i3-aTH7lP$Ra
zqSm<cl2M<l>C1qAY=%qLs>`f>xr&wUUlKWCOfSGk*eDU%=Vm!jr=yTIZ!gCBCd2^Z
z%|P95J4Um?XL^ryPVhQJ+Cvm#jJTj{*)RjDJv8zYdEa#Gcv?_9lcuoEv`<`n8j9@X
zy;DM;#|MbhH!$bgEGw+jdsVAa)ASP2ORr<#eLNI{qtYc=siAl!t*VPRIK>bTLt54-
z{)FD=|6OX8m}E|1s{)Tl9p_+Pi71}Kj({`K(~`s!>JiA><0pMLEjl-qKy?%o(AtxT
z0nuHmWMfrR7+$5fFfVm1Gf{Hn28k0u5HkORoC0WYkY_Q{t25P@n24<#h*zdosU)3|
zy<u%=?z9aH$tGEM>pgerwB)c(7r8kE&+uXBeg6ARi{DYQ=H=I`D(@#R6!s;w&Js7P
zgMk(7G-GDMlXz?sf&njaM9r@c860c`(LQSgZBw>|)m)>7jgla;ri0C4?b<6ZdKr+h
zaYx`*4)8=pSa=v*Dij{3tt=bg;Q`CT5Vp*kkb!Ebo)$3hATyutAy(vvDlT~~AJ{mx
z%mP@drh*pUWoD}s9-FY|Y<_+EMA&`}wWjfbpPm=V_!yePId+=oqri>p))m&nJ({Hx
zulIX}j;lzm<0xuFjUzhM`S$auM@Lz5A@ChZ0it;h?+~=*)7HCb0_~%_Bj7k#LON|u
zm^Yc809TrdQz+%Dj9{~GV5%v^Q)LcYv9}p%_OW5B`0E_{9VAoeYm4jh4ARTvZ$C+5
zANK0K7Z0^l1MMQLjgA<`74<PO(fC7fa~=*^Ff9#r<|p6{)6vJ>5blbpJBwE)2_QeN
z?TkLVl8VITn&ICqe@}FN&eq(oBUQ9Y>gDLvC6UG9T%bh`4IWi|HE{^W8&}Bi6yw$H
zmMLWsIW&u*#0$MRRZw~6o94*OfUx+ixBh&VO~9<0|2+>{_CvnKS5m}9gNmR*-4GpC
zYA%Di)wED^wvT!$wj5As3FMuQ01}H4*h(pGx6#Fj&R|GoZOg`C&8dj2xNFoYWgIC-
zz|f=P{ss5%MgH>G9)!sagW%Em9sTYWKJY+{O?2l+h=#@f0F9gc<L8Y8{(M3L=qu=N
zraIZ@Vu6IQa@3=Y;>`r(j4qNEPg**Nx|RdlrjyE>Xzj6~urHhzsZL%)0AMaUU1-h`
z<H`XGrX@9J^wBnmM4bs8L`#M6i~sT*7hJE1GniaBOqDFL&;-1D3;%DVCjWmSwZ};W
z$$in5nRfFhyVy>FZI{An8nuymNH}*d$!rQ`l4ChDD3d}}I+dlzqe(L;@}v`ojxq9J
zfQJPGDa&%Bfb5g@Kd#37#nsH98WMKPKJ7Oby4Z*GyZYQ<l1nELR*2c#Xr9>KU$zqh
z1|Nfcj%%9yti3ohTYj6I@@dQeO7<KfJI<cDKAw&n?E?bW^Lfb7Jni(kh^3dM%-r@S
zmxs8iYWcU!`|gllE!tqt#IP<>;j@A6v+>)LP2jY8!b=5KDH21=_FBbAs2DZGV)4^J
zt$6piW@$d-tJU<$@Q9zeU0-S2s1RJ;`#M{cG_`6A^AZ&Ua?7utyM?|9r|9Am=Ax}T
za}t0BIDUm^TSU^s&l(pce3(wXu6lBlT^T${xS?_s8_yA2r$*MQJZQB_F@E(Q4b@Np
zk=NHrQt#AuE1y)jzkclhadKFypc?NDS6oEECQo`X(3w8@S?E~tzsyaVlZ>tg_dn*g
z1=3K*vP>-GYQ2azm2Vppeb)VNN`hhCx=yA4kcaN_J9#!D#nWhr+a5B6kuz>b?*XE5
znId1Qy&u(Lbh$QhZ4*!`lsXT4z>rFAx1e1VBY$~azix)=_r=|Aul|R-MZxnpsOP-L
zxiO<czerS~Kzqn=ls*f=^YXZ8zl1TM!lPR+EMq)3?LvwAJ(^!ZuQgg@pnby6rnfo#
zScB;;)~r&&;BDv?El}aBP}Up_K}SC;Tb`4pGHFmPwN&x~);!mm@a;URCQ|D^T|$X+
zVaOJ`j6bg*&4fK38j9t(X~&(Pl-;F1jUPWG_VZOc6XDH?d62(YQZ8uw6JjVO+N<9?
z!qeMAV5u7(LF$+rKMqKPHeH3F*u1{&1lMa_b9=JHOhY*LC_Ie|O&v0WE7{RM)`N52
z-_JQ6t6G9U7ro`K_kC|hE@JG%HlI*PS$+kzjBHA41|d0<LXVddE)y|iCNS`Z(%hZ-
zpmy6ZmRg)%|EtDpfFPvua1BZG{%|0bsTb0k-u_qW<>sb>rV#;!b|@d-k113zOV&z_
zm0hCd-2Ifc@d)OpE`k=`wvL|@P6|y<0e?Avzqup;>oRo%8dc>Y$3&gk`BWk|EDyM*
zXO5znDCCE-9yRm>SO*k)S`QLNFbJnT+bL4=^Kyv`wc=TSp@Mhm?vsDwkdVo9gutxK
zvRgqac@!AB7wi~<2V3CF$8Y<yCx&6u2w&mdg=@HOCF5iE30cMmiPx9I%vus35E7rm
z#0-!DnUP04xj3E>;#AKiIFm9H(*xIY2%bbEZd3-0_y!SJRoKcThU~_6!Z<>D^=XAF
zm6@5LbBEKgV83JQ(oYe$70>vUOL_`<vK`FK-n%Y0yL+#|hL*`tj*yasWh>4K1f?M>
zF-kHdPkDVCt;_rRWoVNnJqgXT=?UrN(|V5qFX}E?y7Ga(`ZhfKKO`Pzj+F-{EvlLR
zXKRWHs<n*Y4@Fy4X;w6An~9S2G^yG4HEUTS9V^`$22gNRzpi`}cdS(RD4^6A`Z@`&
zTEgF@X7dHD{u!duKM_Ef64(Tj@wVT+RRot2hzx_;_E)Ha*FpRf7BBvPwX-8o`(#q9
zt{v25n&ue=Mu8LWIL}MXMveJ*N8LZ6{K0UCOzj*3M6zmSa*7HOaVsDI<`xQ*^Q$2Y
z1m|DM%^~9L_kBork6EHCLpTikq~c$B2QpHtzH9CVbYHcHzSqYvMkr*+X7Ba(mQI^Z
zmxj{=zWP9S@Q$9o$s6NJoyj~P_CzDQ;%5k2EV-4$;S;U0{o&JmBmtvb<xZ3)^F3#Y
zXZH2qkdLNiu+&tdG|DroT)xIR#3V_L7NpCsSn$P4!T*#B5$z<DahzCR3g{B~A#<W)
zkg;@-eQE?9V_w_eVxg@)Js6K<0sJM|3Hk08Wx2y>pTi=OoK1o(v~0kRy-J%!cg;a#
zkpMw1FatD(+O5_m4(lZ*O7`Hoh2Lo=E3EL)s|ABm;hDloV^6uY3-biu-DwYM_f55N
zi~Et{?Q$PXEU1q<Vhb8pTDa^0mr;^|D)$lcmCn=ffe@9|!JX*jYabXGko}e{GaZ`d
z1ty7e=J#(N{y6qu-xL5*%PqpmC~fDNVt02Froj`oZPh?gZ(nu;FuP}QV8s|lFZXF0
z4@*}ep2V2(T$4h*?^YNc)=NwXhL}hERwV^Od-1m|*D=76$E@Lg{pm7xdPQQ^7)~wk
zLmDgS+Ma$w-|J<k|HKqgP=}rrd$PzlTh91#?$AQ@^ZeCx3>}cJ>f;-Idek3Pll4$a
zgVaEi6?YRtYc}?qBRa^tB}O2w`8Mq9-C9I>W!YUJRwo{`^<=##Gg^t8Z*JJqAY^;K
zbXtik#l_ff@O5g0IA6cEg8cP1wYZG;ZqTSUP<qM+O)6DUqLE!X!tsWo_bnTAf0pKt
z;016UA1rMR0J(&zhK%7G5v)=6UaCl7H7YBKI!ddA?Z^lP*;o;)X#SsJ{QDf(KWa-D
z8JzkU`pCRD6zBF`vgY%%lg>GN>O9y50<YIC`;*|eDt?{U#S_tqXoFi7q3Syf$ggpN
zP1?!Frok^Y{yf=Le0~ycJz^Hbo%B5n#w8~P@3(l50v_Z_QyLN!wQb$uSfI2DeLG_)
zTuZXcYTiv<RKBy}jCW-Ne3<vi7#mjK9`?ELhzb=v5)z39`n`XzFWcu6x=l!lI?uXj
z{DG4Yt?B12d>XdbvnetDkjKGwp77P(VE$+8Gn$I)(f;esvjk*vJDY30Z%!=a-wDyL
zzp{<_%D{+O&lOexD>=CNeRz?oB6Zm4XmIfAx!1g7&63zhrarUT&sd>>o;h_|rN46t
z3ig5DhwR=A_Jt7KC8%++Jsh9A7M|^*>NLa4(@79M&%rcNN{tyJ0x<M2E08I-8PdZO
z{|y<##qT+3IZBB1ic~Q&NuTi(JVu*RwgjP2qyWfjCCYOd@^j=%#yPA<JXoo0*>di<
z{bw_y??NzbcMpNl<N`9kkpS}VEs;AY!I}8Ne#SH@KH&>^A;d;4kFf-pqcBHn#<Yf4
zP8mc-EpyYvMjV)<9bBl%#x#$~lz}bC6Scsnx~yO}q@2@=Va3v~HD7$yQ|m(AR_i~m
zi@-^^qF=A{qFKiefiX`B#n?yj*z4H}0y(?<3K-`n5@}K$^=01AW<++8(`}i=q^@Fx
zQXRJbS^YMbEs;5Wd7K|RZkZiN)IPuSWMu2=zU&KcARx6%Gs8||?G}7D>8kQ3sf?DB
zMRR!9WvW0?GOwc@+~py!&^hBLKXL=q5s=mlA)H+j&;doT)tjg`Mpi(z8>)pF^!kqg
z!k9TTQ>>SC$LPS}rd*}dnv5#eC1;fOs3s_w%OuDm8^k+vlRbfkYsj8d!)42As1CT}
z4Sy3H*cSm4kDF{9>cY~UG$^+^sU0DG98Vop%zKo-m8VLqpF*Y*;EcD++poTH2Ydvf
z6vIziOX#q%B{Y=Lk;Yzgc$P#rwbTNtK#Vd<`-xf!ID?4?mx?hmh;g-s8pq7cGm8jT
z4e+bb6drwg44aNqqM#eKjUQh>u8kGrI(179FlGN9sBgplGV?<xmBv_^IP?l(RSD^A
zKniKr2%?}9F1C7ThK7<lh`z|Y1cVgS>+gthbo2JV@AM|KW<EHF(P6WPdCfR2z|9?N
zlO)+#HP{Hbq;dZ5+)s%BRLkI|dM0y<i}vlRZ-SLPfd0`Tbs?D(<UcQ5?}LcoMzW&+
z7q2m`Ci5rs!@t5<?<$sQ!%Z2jQK}_5=)A5*2VAG2RC7bvbzDZ;UI|!ef$h7ko8aMb
zzSwXL`JqIg95;BOYFxnXkS%6{ED}C9;e{Mjk&i?<r*K=>Npwyz{PG>yAmRR}|5Z*!
zwKxZVT2(a?dU7FH*?~Am{PN7uazM*96eYr%quzQW29y$^@sXur%Ky(DA%6iE%Xt2}
zqJRms6qy_`IVQ!G>no28sUVRO5ij1UYdhq{(q*cTqF@F{*%rjL;PbMlVgs=|_LryF
z6+`d}MBQx$mY3?4zbq>mA`dr*GodGQ&`G`xj6&hKv+8pqF#R?<Q_SB($Ui#+4Kze5
z{XcjV3W7!d1DVyj+<Y6EGB-eJuH6E#z>dXUyWx(CFPoEucQ_#E(Qc~F__f${zFOt7
z{$`fU5R=<P6W-M@)(6$LnLsoDdqu$YKN`yU{QuHW)r>NSqUrx=Xe<I$K_0Wya&zlE
zQ2?brcx4`*+FuCO8@mB!;A5;K57JGE_6kH+l8L6z>=z=oxhFWVoeiVX4E^{zM2gtg
zO@@WM*e|prd+IYep-US$voBNGeomBs4PH=~bV{B_XJ5t$v<=xbx9XH8CPg?}{u~Tm
zhL6)CLK-mS^jm{-g7(FQLUR|UV}>Rffis3_b>I7=FV#H}t}jH2FmpNlS^r2o7DY7N
z3KS0Tm<fd|lnVfItG`#tNnjt@i?l9w2PdLZ)6OXh;>X!Q;R*O$9^W37RD8iv2#jGh
zRE!6!1|ew@lJt0~nPCwGQdU_R2a>`JRNgqt^s%PfEA2|%>AQRaZX@dMv7Hjn<Dic~
zUI3u_LqiDljU^?xqJUe~{lrKY&e&9>wg;=yI2?zm79W6qlu6NY+&4Ue8D#|T=(d~t
zfIaXWY_;OmVJK>P;(vF?|A4JvKT8Sd-Y^Tmb=F9U*@lT~k*MIh<n=n6Tt|!YC0c7B
z0)H48+Zbc``F@zRfG8;=EC899&R;4~DkY<t-)tjqS$EF*Ly<?QR03J^o1~1(mg{!{
zJ0z$idIX?n;dj4%=_5KMw+ghwUSl584u7apK2mWMRWIJlANYpRjr-?*`BYm`X!pjB
zV8JpgXm^4=WYLWgZ5_U&%!fSaUp==^FZ{?C*r<Kwr<-L+YBr57W&MHEx~nsqZFT|Z
zqf7!@bdYgK%|crjv#ZNdg|wi3^ayiY>$X}p{vJRuKQ5^AG^n$^Lxksa3dDh6E$Rc&
zVr0wQi)E9$a;zY=e|{XJY;hQI&8%hP<D!CA;SdX%Gn5v!PyvKlq7xXMyFh*2t}O15
z@+C%N?FV)Uci>MPeR7in&Sn9{`bENxv9zY5v&6;7hJThCq#B?XJt!U4VXfy6|Hx@~
z3d;e_0_Q&G*PGUU{KHc;?eL1#yAe+|i6`F0;KSHply-%yp-LMdNB{6ODFio$*CLV^
zkLBk>3r)`F_10=bWfOLYn3!ZaY6UZTWMbChmfiy7*{GxRZrXj+uXi7#UW7HtpM@W|
zSaWf;2FQO_8=0AC&XCcUjrLI18%3Ag4{iaJwL(33birM9*knTWGOsbLzlo-gcl~2j
zv_l5#3stMfxwA*)3y>on!YF6#4E+rig@#1aO4uQ=cTf$vA0iGVktGK#h5_EXg%aa!
z?1?5Mb{mtAox3o7i&FMi44<(oFlf|ksCt13g!)(sg}={yFXKyXXFG-P6-<w1b6SC;
z06+fl^qD_Yqd#D7UxxGDt~dp(8%h}1_>mxXUud}~RtiBm+n?W_lb~}ATy~+62rxh^
z&KbIjSypa)aE?>y1rQ#Pxm3MRrZB+zM}zQo2cGPw&NIjR*`e9}BYNaBQ0%q!Hu@qg
zQ58B9_58R6aA5Nxa>(7wz#4~cbu58K{k-ra3zXd;Oufy#jpvjojkAE3bxW&&phnN1
zn<i%_unS@s414D$5un7-e7k|@SP2)AS6}?RA_>}`y*tblb>z5uYypWHVpVRzdblHs
zEU_1qtSW|mDA)A*W_BFi#?AX)P@Dp+a8cYk4}2Fx13!+y04u@hj^An0WeKqL0n?}-
zrghL3PBS@LDrmo!^I{+i4#9cx3*;H3+a~X=EPyo=eyXmMijtp{tvRffD~&up#99Dz
z;AYzCsx1iL&B9D+RffFV^NLjW$K!IXhpFW@Za(tWHOC<s5r4=k!<iGL!$#|cNH?zJ
zubw||5YhBYJC%S`SolST8UZM_bI%gG-O;45GiltFEI60qWyQw5X^*KEgG6l?=2jSO
zn?p-K-M^1j5=N{kWW4gHHol`foN+ruknEi9DJmM)XH!wSIH_t!^Z9=2D1K^n3P{cA
z7Q)m@({Q4>qp7k5?9~)OMv1<=1*Jzy>I+*m_xLOO6-Gd6-3~=!jR8#d@j(SqC5YNl
z!udCK$#ET;ie#h#1?@7#>O)(V17DF}*P3Dn9W}3vuJpRJO$+N%6jc!w2+f|JFwkti
zH}UGkJF(MhUScm5bJ3sqfzi2oVyTr~#`h2f{_A*uCzZD?GWD$jc~&*XtHJ_?p2@^8
z`7v#Gp+~EJ)FrzVk^!k4KGq3b;)&j=D^=a;AxHS_<O?UdkPQRQPCjPK$}q*;b?xhM
zy&i*9S&Lei^F2pfD4)hP8?hD3VWNLWEV3cUhZfJQ2YE8QdsOI5<WP_{oPv5eU=}XX
zgRN3bsmkN|+A*@)u-kgamH(V|(U_J^$%f#jmx{CvGWMd}=L2J$FC(jbeGU3bs|7py
z`Uwex?LGErwvD!l0j1EMYp|xpTHp}sdtpb+2cI8J)+e`KXE6a)l{M+bQnYNam=0cM
zDH7mO7CZXw!M0|2GmO~aHmPLP2M#5LUiub)7%tCaWDEE*1{CiV^9jY*v=^RMNH9T5
z9-}H^*#EhQ&;bpn8Kc`C+$5@<cBqI<oRIn323A=NJusjLmdMjbJq5PEAhtZh*lnB>
z>G;kWQ-u5QAKS9tr6^HqN9#!hh-O&#VtQI7PNi)2<mHJ{R=8_2osQB6L7&nxilOa=
z+Hk(&T=T?Hp#M&NSmufI9WX+T&cw>NSWxl=TW{W*@c^t1?v9k^^Dr(#Y@!&x5`qb6
zyN?P|xS_&wpt1dCiWIvD4US4wb2J&-sNprIE3JMSO_^q|EjvMUv0?R;7TIG;MtkOB
z086{)1$;9n+we>(>GXcXbqvlPG+if?rFvuBAF6-cn2F~96w+9Z!?uZT7RPcCYp#at
zrS!kI&;v1cVH}7cs;`H9B+116rkDK?tnH~hWz~V%o#wBC7RPJ&%MYoS`gEhATYqWU
zJZA&-q-Ad;BB$ZUIa?hp+<ZGEf7U)@Dh+1Oq+noP%=Jw=^Jbx98W}Hd>mGSzO<DY%
zm)*G@x#rS-3y^vfQtih-SU$(Vw<f*>Y$xa&K!BGOuJ?vViBZOXW?|c%=9#QXv$}R%
z6tPBfawI?08-L1y?3sDB-qV*qkypito2Q-z)@N)Vf31mE-?Mady#$JXEv__89xg8C
zG}xtV_GEbaVpD)F!LHR5{J!(d2dej=NoUBw5M<HE!{sUef`{vS3xftde0u72`b_Mq
z!2z1^mCM6Xp-;Ow$lZl3!e>qCsX^%4FOFn!Fk~4C$;iP2zCV1BlSfdoEXSE22#9l8
z(Ti9-yvSDFKSS?EUR5hqDjoVrGMz{ef>bTjRiu+ismu2Rq5}=5-`erQTSeh1S?7ju
z@*PT8JUX)Z6;*}K_|uED`MHO;J<3DA+`Y&2m~n*>$iygJN(=dC`ak<X1Y406N@=KJ
zsSTHv<wX8ut@Bpo$-+03jJ?J%Nm!f<DMRtLNAi085Hx#FE~7&qlB&h3sA;X$4X#5Y
z5D4Ch`HlIVE{sxWYj@DgYa=pC=r$RLHP>D3s)6$m(HHE*RArsTay+$FA&L71@&<U^
z_Z{ay-Sm2K;(hzeNRnIa8ult#&odRyIpJs~LF-e8xQN5TX1$;yQkkE_%=4nuP;mdY
zuEi+d<+ITIBU4(E5O%!DS;J&{$%V)^m=D5XIXr*fj#_4WwA=`4T`9cRXmSym-q67!
z+o%&YVql*lT$krS%PlKG@Q6#7UK%ih&Kma>BzEL#Gs@P8bUS#iK@4*b63bfH3T?Y7
zFGtahY!1I39Nne*ju3m7Sv7g(Tj{TK?Ah6awU64-)9Ao;&*r>5`mdUxoRF?}=|EdT
zoz*o19G$+=dqsYPNPn#5XNrMUIG^NY8>wpepI2L$B%-)Fn5%X0`}II-?=S$pHMBf?
zAr)!*DgL;iGSd_ybA?OR1T73^jJvBH-pfx~FSpzU%WRYwY}VWFY1apDu&a*`(+lJZ
z!sP_^>(-5w?nVYrB8G?^ldK&nc(?Ia+o5hNv?(hzja~(acU_x$dMkt&G}&c=Bcec@
z4k9OsGL>`6b<<aCA7o3F6hI#JYbL7uCza`Kt?m|Ikfuwf-j}SRu5O+2J=hoNhco2&
zyI-@f&phddi^P%;ZO$$}QYQX9@10dfvlX$$l#QbIakzYERCRD($XGJ>6|_@C43E1s
ztCG-z>Mo{Q4KJ^kCpo~vltAIjP8CEA_>2aqi~V-+g8r0kQ$s(v5Rh{x+E$Q7dxe5q
ziQ>72U%8`4h05Z8j|_u_knu1k%bvL#FiA7IFFq0&w}b^75oe>+jl_?vm=n<Jijge%
z$h|dmk)|CGfO`?t?gPPhuVha!0L$EoH)e)@z<!SPmml*lKY|^KuRJ6{q)oA5$r=A%
zi9yMSC*l=hVp{A^d|-h6-kdTg|3l+cp1N%D`2I(wrci>q>xQ?9bZuh+Z3IyXgYxi_
zq7bU5@KTjtNgzU*%H1`wW<4Qs_VIY07UH~Zb%I-`@tZ6HYwM_zS!07V@_oxD{m9c%
z-0)o0=~6_o9&^TIz4hUSQf7NVp^mPucK)q_DYiY)dQWLt0I<T{(Q|vza3%p3qGK{e
zQ-*1pA^y=zg=u<dCa{;QMd5};waKz1<YK#+T4Y@mQtX<0OMSQhc5>qN6lPI$s>k9K
z_IG=8lAb<T@9*~g&C3~2q}+`)m<>kid6$@iw0`5~m<x}mCd_&mUj<)9jMV26E7kdL
z#K7;~Cr0yFhJfck=SH=WU+DFT0SiiqTGV2MFX*ruO<BrIQ~EtMNP6>BTx2iF1YTAs
z543nuu~K-U0pri^j<>Vhd;{dZ)<Q=F*hI!>4_%($z{NAZ!v!uFh;d2tj{o_ibz#`f
zVAQDRR(p@@qH>;{O)ClwLK5f??s8#@>eBcy$4Iv1?gMaO`CjKU;_+Z@@Nmhx?bRdj
zT%qpil|U_xB}Th>5Ht?-2%EQUke)hWkBchQ$3^5qYfbi?MOSvd_t`IVyR$*IYp!1h
zk4_M?F4+%vY}bAE1+7wWgu>7d$A*RI6qw}iuduGDu~o*;w!Jy^K0rPZJn4C;nY;Nx
zu4r6->2KB-&&%7WV2@q_w{SNyun2CZ8t9STVhq@c!5^%@8MU`H{75?1zsL*%+}Y*7
zkBL(e&*E1uq%3qwJ-h6R5N&ONGg><Ou6wUFjpF%J`sipd@iAI)Ywy5riH$h>oo~Q^
zqxRboSLbn!@?}!!T_4)Ds?d36HW{H#W+_su4X^^2>(%h1u=;P3y@_7ULa7W8220NF
ze%5%Zqev==!}KDlQ^1RT(g!e#SeDdP{vLy%WcLA;Pw#CBXb6{BbXlb8-tbNJOYIOI
zQw$s?mctWA{W@mZ{)EAzV9uw78CtkBB?>k$$y~S$<a$S%(qZPUn(cXI#H79+EH|bB
z#FC(l;Ed8MOGMjSj5mFKQ-z`BG+d!CJCd=JZw@C*O=M}x$Za*oAon9Xicj2aJ$SOH
z_8oPO<Zmie&~Jjyx6m-6d(V0gS3zoVLoK-f43N?2sl6F7VlB5smwLIlQp%v|(guXp
zlQZStg$hjA9Im$fj<U~jXs{M5mjl^>VA<*?X2_!k$j!UQVrHMn%g_3NewcTmpMs7h
zg$r>r7Pz$7n@|0rRWnb7e;IXta*y%3BRZ)Jv%Ss;c98@n5pBcRFR8UiAidlv9WHiz
zUb;&q`SX}KW^qIDNETD5@C@6VfIX|p)y0OyK>a9$mn3|xE~}xvZy{1^DjWg>zynkA
zgRpA3w9Uk}h}wZPW;#wQs`SRL*$`B+>qzP&DC1^lE7NpPxgR%jj^e1c0%-}=Po5D(
zFbzs=CJcdIJ11xc&GF19o?uV{Mc|3x2`;zF=?z$#%-8b6Pq<wbLtP}lJbMejeO}5O
z*Cgxd*mv$ZzBf`vZM~c)cW=D_Fr})22IBdPhdgp&Jf<mdqr|R(IuT@Hq7P>aB+h3G
z0k6*1=?}!1)2oXqzg8OCh3$#|JfH*|bb~((Ts~2Ti4+r_2?pq_G-AO?PZ7IRyv*fU
z(o$owopT8d9>lAzMNdf4@LM*m1kCn@X`0aTk1Tf~(|iS+K_D;HISmT{CuOz<Mj8i6
z%3j>Ozp3{A_#{oc6Sejv^&#}O>uW?*W77J?hq7sPHMIYbwi-F`77cNi_#uANV=piK
zT^7ipUg9e)Hp(E&Ze44-ZaX7|l7p0$G%~%3R$Kzv=3)jZ-SN|1$pg!FKaRSP8YB_A
z^kx&xef)@m;WQv7*PQ}@oX`hZJsxh!vpf{2b4HmNm2}DKSUjq3N`ou65bYOG&QhFA
zJfJx*f6M17Fznet|I{m<<$e09lV|F*BL6AF49IBN6H15eA#hl-eOMjWF%QUJG%Y2+
zUR`qOtd;E#c|^4AxQATHH}{w7SxmB3nwWes!MbPc_==381bGIyUwQhYPsMUEZSX(R
zGhL2Pf3hAp7pL0u9In#3z`tWhHF?91kaWz&mZ)y_kb0mh0=i&M!yG!cEqdbI*vk(x
zg1x*k-$!~DP>wGW3OPU(30!(x&I;vc?Ba7Nu-erhBcaZa_tvrLnBXG_#u6~kzH7im
zl?KkT^b$uO^%?_bgl9JzoL)UueDy}xzNZFu2RD<<182M0(>y)|N6Ask@B_cCX>CBE
z%&t?6V9~z1sOith1^m-0wK%y)#e|!tkF&Eth-PL6{Q5${QB#iCOd^aXkJMh_=dP0}
zFehur8;Gk*3%-K1zolBNqJ!c+NYsJQukgY^yoKr~<OKp+n+BIb|9pG*T#T@r-g_dK
z9r4HpvG!{IsZ7Brm4hEjHiUl!km$=eo|&$=!Yt3lnnW5@qTOOZ7%l;h^2BUpO7Td?
z1V7VrmQmII7@SBPC<x~}(3<R$y?*Nz`OE@OH!Z!d2}~Dd_tBiWW!4%F9jIxS6vcV?
zW0{Cj7E=$1TIGiR;YF*=addDn!*&m9=fbh-v`t$|2d8iPZyALiOeZI0+E%~qP!uX(
z9k#D%L0T|$^E)8V$a%OK+TN(A5ZUuNe%BG|%?8w8N74>n5xLpVwParrioNZWMz;Uj
zS?pPjCL?&rHuc~3{g>{j+)rpfX(34r(yx<#1EBwfHsvB^Aeh{bJ;cxq{CR<@sstY$
zHWnSmc;E?UE!k`#t$scJCZ<{rdc5|&#rfwqVHWJMw)o*i18vKPNZZT-=6Vv1St8qh
z&3a*x0L}`Dk&5UxLyZ~YQi3-FWzC_@FvmH32>N14^>xv#$;|gZU@pI1#%&wfdYT_=
zGJtxSE-aVdf8ywu#@)wJ1Rq?9`$5_`{AV=E{=CnGRq^5@ddf6R%P7iwdLwxGeJ@Kz
zkM4MQV0vB_T(K_K6r&UukH`)6$VrN0#CzpJox@bxD~;6KmL>4$NBYbBQ!>l6*QEGU
zS^bFrBYhhh)@q1wHXJ3pr=hv~L&B6f&IItZ_L>icgy1y&U4aJ^MS>&rIA&&y?6IrY
zSG&ALC7se=Xvr2bv0+?p6Xg+H*|<LRF4edezWHy7DbDMZ7D{BBRB!)2dgbIejor9N
z{N))b>34u6OKQ>B9s^mLRtbub>xLI$64OheYmf&wa}1qluN+nx%ha=Pyt~iJD-PhK
zZ%oFVq^o8H-M5xtROI`H+DnnXdl-;cX#9y%DRN*Ao5wo1&>!FTeV?p*0m53ReS?(J
z<&;RM^MON(q{z$0^VC0ejt*n@dt>(OA0m4PNSy}S_;qy8X#5xY-WU2mY7><X@Eyf^
z<*U`5L5cr#?XeX>V^NlP07BAllzzbdpV2fF=<g4*^udcfR2gIKL<`jXpTxJqkkEmy
zM<o0N+7JBf>Wm^5_ld%JtFK4R);-AuvDJTvqjB+F+t1NWn{+P!PZ5M_)thZO{ga43
znem*f2nT`Y_1inVToGUK@jz~0WvRJl8-|Gbw`+^<jAs|?dA;>ItqX7f1$u%8dF-i#
zCb=K&sLz30W7?|5tVPxlw#1r6mN>nW<|T~b{)i?78v`46!u~J+U|}bp;8a)UE^nWp
zk6fP#eCqZ!U_c;UNcI)lBH9LvZ0T~-&wDM(%Vs(EaLC@gY-R@SkV9k%yQ_B@TRT3Z
zV_Oj#OB!qU`wIw>w;wU^E}(^362`Ijl$EQHs<(qr)G9{oH(*jK($zIQXqRO6O}@Za
z!$ZT%j)k**7Ke!&qWwsUnHxFz{FlSC>2ZIeGK}dl=@hkRuoy;P6*i*^3nqk_Tj!d#
z&TDZfi+}vjxi-<92D}dKmgcX)@^=0vS*{S*u|PL{|9o71blMNhKmWIvri#bo?0CAl
zKhVThCZ;9+$f5FH6wh}n9u$4GizRFTFD+cu>dP<Iw?R6p80O*6t=Ezx60?-xu%?J}
zXX@mq(TTj`_A#4u!=u@t7M6|PVv_=KX{(ei)`+a4_Rs#2_1K&4pavofVYqrDoow>b
zM;L?{<zO=iI}1>kqG?F$k_y$4jZv<a;}{76sx1#~FJ@jy@@KXj%_Ev`Qzy;;aN}ET
z3X)5#E3g#HRD=PBBRk|#qblNC#5n)2st+H{;C=B$lnl1g43!(F^BK7@Fj@cJB{9QN
zmmbTqejdTZ+eu=f*<NwUV3B@=ES1S3^GEIBPbov_I3SyT(Gs^#Vo~-!IsUx$cG4<V
z4S76kzL2<a;Vs&mLGOKzqLV37U*diK@=FFgpT5E&R>#lDvdZZsIPdc#!&1Chkn{e&
zv2Mx9GJgubRkD0t0;RQMqNzjoDA%hxbKg_3wrT$IflK=+dHMhP2Kb#1GG^kiIDsMU
z)D{^FaIMzg-SuuOVYnnU=Z0g}x<b0O|IocHbQyc<-pT0XbR8rZC34iHzaG0i|Ld_Q
zsL|vGl4o@m6eVk`^K|8PpIu+h?R|sMdvoty+$z0in4s{(*>t%zIAd-@i`i+ii|s0B
z%xqCrPDO)iJP+Um4$zpIsvkwef$_Cuwi++kXMdUv?8@YdGa2slrgck1SyUOa%m-|Z
zF9V1^v|GrML^IvA%?tiu_H^|F#_^-@=R|Z=|J@<-14F*3Q8?~ce(U|zapNl3$XNSS
z7~e!5sy67+8LJan>I*_E4-p4}!<;p1hE%M2D@Ks;Su`Td=L{Y{QR5#GK#)lA=0mA_
zb~;%B=!+KNkbgYQ4OlDWF-h;OR0bk9brEqYXD#kuB{0G^DG^2c{J?gdzxjGWx&7jl
zX=th-O{Nn?Lh7nF;~kv%D_rv(La|naea`vBD3A+lWT9iKUbcKWx(%g}KM;d{EY$`p
zp*5iLl5s(dp(FaZM?O6saJ-J|lI~!CqlUOtoz#hlcUw|=8Q+)AwI@*&pecn@U5xBo
zX5H=wH_IN+n0muw>GdFpmO}Dj+UY$+iK_BgIl4PkWl{J4D&Z7hcPXj<qEUx1Fec`v
z-+}sU4^8Sobbuaa^<oexnMKZu9fRW@Wf2~&<s;<Gr!A1xvzF`uN`A+=T+w$SQB)@9
zJKdUgJVH@$N$EhI5+le9YJ=-Phi;UI=T{t0(kG69dil=rY}TG1flsTEC7VUUXX&Ks
zUk3HmPAAL51PL|&)3@Y95mPzYv%o4*I9q`KlCpiJ&GlM~zbSrG;kn<G?3Il0&GZ%h
zsEyD4F+a6ruRZV&2p7V9WpSobI&<$cSh?DY;q!T!#0T0Bomu{wI8bWxIf?%L$2MQW
zA>3pU8PV?P-wbK=x>TkpB$_pMtfmL#QMz!um~~BQ3qx+%OlyOSIRD|<9-brlZAu+t
zezt`ACq5^!m-A)ZrL%1;!$1sz^<NJ1WDaGStRH7<l@!_q5XqW>G&lw3T6FU_h72K5
z5HH+v=ss1p7|wWeMae)1Ys(ccyzGd_QohM_crS#g(~@}3qX1bQ<^`AehdxFhGaRF4
z_y{QNP00Rrdg+K0G&-SVhnYTeIz~RVpUkEEDd_J7yHYB<R1iO8aeZ!~Pu-lA(m}OK
zA%BKM(*@fxfM&NEvs9k`#_vekgX$rY6EkUmF@|lh&uRFLzy&QtKBYjR;6n5sKS`3L
z>1$phXEo`*_0qIns1xlIf~!a@nh5bA0orbTL=&Iqz`Uzf%Lf_Retih@fn*fYl4|Vm
zysCxY#ACu2m%1qZta>U7<(5Yu!MGh8Wn}uI6cymWeze=$n(2!|z4&Xczq(nNCf%Un
zY6<wDAl;z9T!fq!=+@cBWl2=%1=3kvXG@I!wTa`|lanu&q1iVYn-ZH|bAg7m*emE0
z@<RjNPNCQcXETyCAtH9B=4sC@Ji8ViEPGS%!aLWkOzj=FLD4{2xbm4Nu>~xCtAwS<
zB06dyTN;4ur_Nz5+=<BH_bP;;`;44^0?7|j6oW04^QlA99N9bUU167R{z35PAHTB$
z!5jUkLxCHSv<d+#<hNyMFhMdnFSMaRkVBVr7-~<sg1yXO<<!g@F1aATH$ghM9f-6Y
z2-L3%Lak^YVL|m%^xbmp?i;$m<#_#9E9MNql19>Mds>FzXeJ0sIEajMZ82aaO4s;T
zwA>+a<ag@tcB6;7IrC`lqZzN)Z({)uY3Xj#R&UY3N6ROSMPy3d?-@M^#uD*{<4Y#&
zpgLbdxb1fF<eWSC7L$og>^bFfDJ}4LM~n`k<a5H1H&*5nxd_{^XpP+;Y{!O?+3&vr
zmm_B{E51E;VUvP-hPruB9Q9@+&#m4ZS6T2*?$^eAD97us)d(J9a{%ICk{cJw7tSHh
z!1xz5E<$S-^JX<_0N%#U4?oYOsMG`W-UUKGGnvEX6$JKVg2-zw_rU9#<&9E?@5F&x
zDx=eo(w~x!zjB(Jqek(IQ&<~oqZ@O94x5m_Hd1K6F$Vfj3uh*?q`u-S9NXi)K+~J)
zUMHWE68;dO`;0KF5->JPy5^6-!zv(PNJV$QCHx~6n+_@tQA4mqUYkRxJ9PPR8T_*a
z8o2cCoUWRBW(Z}Jzx-eR78Q)Xcy@D7kJWnOJw4UG&)+7@^Fh;q9s#v^3;g4;3i}?l
z?AjpAq`$SNxNAKJ2`g$k{qtG?!g%faz=A~qnwKNi&xV{;@37M$wQW9B0HyP&(PF1m
zYAI0ruLoozxBp*v1lohwK=gWV!BV5=KIdg%pqkT1fX1$da;r$g08x2a`DjW8?r&1h
z<=j*^{sKN4J~l3IjaP{v5s=QgUNyrfg_Z)&(vKf9HudE<+v|I8)kGXRxa87yOmuGd
z83lk7PyJCo2<ZLwn@K_tDJZI!z@nTwjOqX?xuA^yCU&0gV3kRK7qr2C!UVOlXt-a}
zTyoHj3o&j#E^y}W39E3<{@8W}olXRbc6t5X9SEn+FW7M%(F~#nu)q}?;gb8frBH~Z
zlZ90VD=V+cIHQ!)zD4j7v;@Ox6>hD_iF;d+QLZDA8pr?BK>K%abw}iWgQK0JAL2-~
zxCh;^d4!Yw-NYxG>S^J=<A6Y^u?4@81a&I}ri$S3Jj;L-@0r?micMI+9|exCF!$Td
zA1+fj(fNABZRoB(AlJO`!aa@GKztY@%RljM0+*0yJbsTeUSnh;7Zt6L{4#)Nxny>)
zNyn7=CwntJ&KqS|0BKUcW@gORm2WQ?0wO#2)>V7#L=6^HDtSauMO}hZzq8B8UPX~)
z4*L)B;p9RIPa;N=eo=BrKT?(0YFI)Pro%#LLU8osRkJ%}fb`OZ98|w|g=<L;S!vo1
z4t%TSh+A((sDRIOSzy}`Uv~8pmgq2Gjr4X}y0qC;x<}xQ(d~t}-DDSslx>jW)Tv)<
zLGY1%(*#SO1`fH9uj4CM0}IaZ06(6fjKR;fNqb^C<nBdcwQ$RQ=Oq*j+(^57f!Xe9
zV0jg!R*3iyFk`#(yY_2~>%*TIX4K2H0EY@9`n{=1sq%wM?BPcK{DCt}XXXnZ{emL@
zc8wUf;EV=6!7W$4TT@Pz*_%u{gpzaQ$lJa^7t%3#bZ3}Lcz$!U6CZWLS(0}>HpA1~
zSkjUni^VyFXHPoFq1KC6hwEe;HBD|)qVrJI0K!=SQ1=|nTqg-G9u{}ySEH*PE*_%y
zmK!z7pB76d^&&CkbJ`R-ub_oe08fc;Lj#HqHdK7QH!?z}r6IJNO-Oy-x|iBE81Uq7
z+UwaX(5=(|c2|_TdhgFHNbfq9tn0t(z5ecI8b|a#FIoznW(<?GK?i%&_E34{0_c$*
z<{jOD?<)bI-gQqGa@Z>up%*z(crds&rPs-+pYxCRfc=TO8;-;BTdl=HY2Ehi;%!Mr
zvpHH>fdBpGFeV2+A5pI$PO8e@^3ngdKrF*TSQT*@Ds|}@6R3Xo$fjx;JD#yCDSzAk
z?Avhm>>k^&N&P^F04FsY0>nc&?8wWTgnP8$5-EIr^v*_y*~W2L7_#e23#HhK-Jr7T
z>b^o#A{e^DjwxzKcRQnt*vU(0R*PL0SbR=Kp}aW&N0(_a3*3a9#tLdQ^O7r7PhY)-
zw7&m`CTJ780RG*pR@!@0YVPOD^byj^`8e!h<l5xQc|Gsp2$N?3Fz1-qvF_qIB|#q*
zWIeum%e_orXSU&l28mp>?eTNgwAgy4dMtNkVyLD@OxuHUG+Qt!#yoOzL**#TxiuuH
zRZz=x9gO;!4s27BBrLNyEeWkds1PXnE}VjIuY^|jIRvrDtuQ>}u*RE*oi&Zikx@kx
z9c*7bfo%jlI-g(w$%82K*74R;?gOL2=y{Uaa1RP!`j^&Kf?g-mm&_+5fUG$Y5-ZP=
zkoW$#AKCwCRU~~mjjX~^1A!L#N7w9<BJMwL%i0aX-yA&3@b23(%#k?|Y*0cURkOG;
zbxohmAOCW&>m+7Cy|RAvdl#~B_z4lR(!VSVg-1B)P6BlFog~RBhZIinP~XzhEb>|=
zsKS_GOCNIlr4LA6syjMYU>x;c?eG;vP&<rRbrGh|p<X~Yc(w_@&#T;OR<mM)@%=#c
zw2;*noQk2Y)yVI4SW#{0-RJyQ7@<H<;$k-09=mA8X|X=@kCjkPsccR$OSzZn$#!HX
zp8-`}%>t;pR+QtjmO^zGSbm<t5y>4w+2bS>Q|K9w75w?$SgH>_R11Pu!{~Q-&9%SX
z!hYr)`s;T^cvoAJ&{5Nn)qtjcLq$9dh^s-EZlkbSh~=Sut#IeIQiJFzMXe;!M>8M)
zSl17YrsJeX>-a10D)>o4cv825tvIdK58m#HM`{3(3(tE7!<*auX2_RRju+3Xt?}Mv
z_v%R8BIxuCcA#wwf3(pnW2NfR#&U()%w80ue!^1{0f$<p?^ALHs@``*$Q2=b<107e
z3Vgd(jAKZ*Y}f)6=GG>do#%M}f46~Vbs2T_Yt!zH)Zd5V3`Uj<s#61SSoL<(jb
zP71hVle&}R@%ojeuFNK^)!h7@@+64%wcKcRSFMBlJxqV`v;|VuiB#I-_N&SuIR8Jb
z7S<=+pA~Td_>_xBO-dumU)Eqlu<UBeRm3OP_h;9f!%ar;=`IcxtwH6i_0<UuC5<m@
zz{XbO&&gVBV`H~B$F@S}c|7{&G`s|IAVv`6SS}JP=$(uIh50?tbIydX#x-9qYEbx*
z<x?Z!lM!+ptnX|Fq}4?4om{N(Bk4rt#T5ZD){bR@JduBoe*R7!pV!Nfb7g9>B$~{c
z<BD-{r8{Y|@nhZOWYaO`cxT;n<L;>!O0A0=^=#psvi69ltet$RfRED^-~VR@kcy8L
zKJ)w!p|uED;Wc?8=W&(X%$(V+1|BwnbG<09xo@5r3aK4i1x3NKVBamzKk(l#+&BT+
zm-TYKFISM|QKabum`=E_n&eP<^~Y>Ue=@b#z$g5e?mx4MlN~xbjxV>C+|>~LBKHU7
z+u)McTKd4<7=z3M`qJYN2U!B3CHR&4t8M~=KPLaOeD8nxX7cX3TT@IS7WMHu+vxbx
zS?JX8*ioSR=&d&0N?Yjs)tg=XIqH*P2plk}8~kQq6~rsAOzO|XXf<kkSSmk#*bCsn
z!{&rZNuVn#17F&4yzrEL3+nm2QG44`>fV@lzLRR@#eaI7TfMpSd^iVEQLwc?ky^Ud
zOlvXD$WttNoL-;$@e~%e(#&rA-LSY&AM{@QkHT&*INWdn_d4U}-krtb=u>Z^2k=_E
zBg5HWJNPm96u2J++r(#e{|{mB92|M{eT&A%#I}=(ZD%I7olG#XopfwWCbn%m9ox2T
z8#i;m_xI}e-hH=j?Z3KvSJ$Vy>zsYo+WV|!Qunw6=&Luv5_nR7dadI`>MZy?d?W=!
zd0$SF(C@V$=Ti5=>Be|M*|NaE(7A6t_W>9I!<kMCS1F@@`%@^!cg{Ri+;E_-pkRHs
zUx%QI8LU3K%RJhg?ythJFDvf*a;qa>GJLIc<v5YGo~<|jMJYA%Z$Pa{I0YuTbWEnN
zPrSK|Hlt98O2KsEJMyTvLpm8S@FE+mvEX6Oy>}XS2s-Nzm^(LJ;QK$j`plC%z`kZy
z<zGt3(`w>ogFS9S1hE|aT!R%?=Om~e8A7*t{2_=Tv0)RDK|(|8;@-5$EMfc+G9+TC
zG9v|fG+*7=1`3BIcVkbwY2smnB^Q!u%1)ErFcv&OSFs;;yb$_T+;-s#hlH~FZ|B$1
z3WXk&6WmWN)?^W+=|T(HiYDPIfHjr9z6b`T*2rqpLEl(jZy$+8N9*5TmZVj!sOfT3
zG6i!o<zO{7e|e2+<OCr9ApJ~2KVGn2Hwc<_=3tHd?JUGYtSgc)xVZl2c>rPu((DET
z7?$HACiufZIK#QN3H9faeMrCFh1;@49MY8^aiNXW5?;m-!m8e~<8jXy1pr3ImNbj_
zkQ+n-4}t`e1<rSgRL8~EeStttsUTxSw$apHZHtf!%Z3IvUay}D`w#_syf4Iah6g?~
z>gyE4=kEOSZBv!ke|*;=PbXS8WS)&*kjYLUTHgu?S^MC7%a-0Ic8Da6ek$9=<HmsM
z!O-OD41mCzlLQ$l5P!}v21K45kV!XzjDYU5<fU6Us@Yw+*}6fLA_YRXE=4~l;ybLn
zd`}enrJT9hnQRL1zH(y)zS&{IwcMAlud14mCr1|A?)g{ypNHM(GOgZD;;feMH;{jX
zWq3}pfoDT00O978&*pFjQ+OfWp!nd4h!rTT<xkKTBylpuGR@qz0nfhGoQ|IXOrDk@
z^*|EB#&ndDkJod*>aTG-V$?o2!+|r`V%HB7<4$0Bzito_CS1=)u$Mf?=jGJbdfF)Q
zpBOZ^XAwc9rxcj`<0^kJXij|aNOp4F7Fv`sgHH~qfGWzc(Z}$vs1D)tJi&H-B&;3u
z;(jwWVeW6eWKe<!XqQxe58iUBL$jwJ*xL%_P-n5pV)~FJc=aP)7yocSXDDqBB*tA>
ziQ%5Gfi|J3<wtOfsXksYw`hZ!wPO&3<)IcW+EmmXtTLWmbW?$zU<U!&YgkbE-575p
z?`6s37}k8UDTqJDY(+JAY~WxQsdcI?0-2#;$gvjWxtj$BkcZ3+Z!<LXxCXt&^V`Ns
z8XOxJ?anE^u))9*^Jr_d>KK*h{MXr^ABcPiYR_wo=839ErN0ma@VQM2s#;O6tvo!!
zx^(KhFI^wqo(-K*7L5wBzwXpMoZmiQsrzi*hl3$5s9|#zr106R<`5m3rg3vOJAH5M
zHmc!dZ)*ZH9xkJ)D3*;;6isuLvX$k?x(pF;t`Dq_kvQL_8M{Hyti@|{%%oI?vGM{5
z`K+AMI<LX_lMPVOu%z(=J49pAzQs(Ny&^r65#u;v*Jf_8e*55@L3oJ^sARNd3kJ?t
z2v?T}5yQ#suck=}C?Z?!`_zTua)&^b-Cgp1Th9Z~j*z4?jpnubma(k&RZ8oqFb{(>
zuS+$IO!&AKiE$209$jZX{ifI7C3yh?O_%cRtAF%vLMz$x4<^?Fcjk+keT;tlkI&|1
zg9?L(mx7)u^JcPx(l@)7!zn)*8hkt9!%R>fHyfwyW4==mp9wKX|H)47{H!oFbE3HK
z2Nn%be=ej}W@~o}J`Yf8crs6>Y^C_g{{S$_kryWL=lLGM!*)$i-ScjWS~ZH)kbODR
zPs=Aj3RyF5`zf<Tq;dI;@z$U<MmNX9=A;@{oJ_$Ydu}9qZo@^#7KN>XD4eD&6Enqn
znC&I{99-BBUK_xg>a7*!dxJIIu2txh8ITI#a+=f%UtlbTse?g{Tedhx>Si#c*++rx
zG{Z~_v(f7I^V;$m5j9SK5AgF?xaWuRgFQ^)D6v9Pm2c{k9FK!3Nk2KdTuPtzC5p02
zl&<bDkVe7Li<0$B8?B76$(rWp7-XLlk|17tGZNKVEROal=TRzRJYF)D;g3Rawb%ie
zv54^L8_hN?w^zgL1l7-*l~IYd8yGFNkuQXD{kp^0j1@?TxO~Ghk&S8J_wt6JxcLK-
z0}+V&P3kAK4Gcwo4Yx2jl4VP*l(4m6`V+`}<d(_kv2>ASrwo30x`!hlmZMG8yA>SW
zZ0Zs+X%iuTM^Z#=XUo0&0qI+4R?t2`cEpRP*GQq=OqJER*M9N-8D9OFAQaO%lFC&%
zFoPEtL9@Lr%k^e-U}80;$)My{x!3P?3@%YTF45c^(s3M*aH0-p{1pw_r!RikGBCre
z??OQ^&Fu?!nERLPJtbMumB&;rB|b++GCyvPSKH>drploFh(>R@6A7nG|7HcKDYi^m
z&lVF3D-Z2!GeDSO&Wc5UAiW9Sk_vDoeG&Der9TpfDDNP&Oz8-nT)h;B5F3uB?qmj}
zH$9JrV6y?v;rECkpBGrre}4OVAz0Qp_l9BT^<!M$z^~x(`0AWF{C@XIO-)C5_b<vs
z2qJG7S5A!_=_i~Ms$Z`HLkIwrCrqrLApZ0GuW6PV_2=pzG3rlg{U9^y&$#{rAu7cQ
zIntaKgmLk2BIvo_`jJSogMMmk2_w`<r;-HwVjGQa1!Ifgrh(IoO^MXQm)Fco-xUO{
zRgbL4psGX&;RQCo1#`gs)_9(Q3+!@5Wb_xAR5}vpdHI7s41Rm)(YOW}>CRbXEBRSi
zZ%yaKwrBJvDFq+foU8d`q`hNcMG1hIe?j8R_y?w1MKnX;SI@SC=LF3}K2zE7r1OF9
zx9bn+jdze#c(z4pcJ=fT(>!OF1(c8xG}D$-p90>aI`2v6Tg+gg9Pu8%a9l2vR{oUK
zzFTGY@53SV8DrouF0uCj=OD!L^1?LTT{vdyfosRnkD{7<hbhF-4?9E@`Km?CjLNqd
z(Kzg*2~;<4wz-mk1L$!eaw>pZm^}9-B%UolKLVvo{abA-BEVC9(R$XU79$`34MkIU
z!|44r!+5e-YwHb8)kJJ_-aYQpOA6kzH2P)5wnO?n`Qu3_9r+PZzK)_Dst`4VJiib8
z059fJu)is7<{iYIA0KqPWGHU#P`3BN8iu39Fj!|LwA(snM*1hkQyRFd*StDV$%jjJ
zNtn(yWl(F$WHHsBaK0Og@cJ?8b;zYM>9`GT2xXY4k?g!$qoF7xA>5SRPdxZRm~TeL
zhZ__?$?u7akPNtkMFvm$L2LV9>@L2p|4ffLpCmVzM@`fDZumN*ufHirZK+U8w{RXk
zP}_)x88a(#G(1ZQL$9uF)iT4Aky+~7AAOVWgxMz4XeB9DxKzRlM(#Mi$)>3h%8(}v
zb%@qeDCn?1(LL)a>Z&=5mH91(>%!<UU=nAsCg=j@z7wFK#hg6mUl{Fe`8P~^cVFT=
znPCu-c*Ax;j&j4Jx!a?;$b15`N#5AA%A3LDegVaHHj*#_gW|F+UFrfu@JPkU5MV;X
zmOkr3WY<d}{<VS{r|HcL;nLC6YM1Oz6X?kPObjJ}m!w-#Lmn;ZEr_g|Y2J)GG4hW0
zMf^&<#RKdBS`2~``KSa%r-G<B#o7h9@jesuWf$ArxeDO7%!`epSe_QzWQ{cE0H$}b
zkJD^J;ocE_pAzNmYAWZGc3+WoTe+0rQI=H|R;A5rKU%&;#=excjJLvSC^D0iE~tq{
z<_Gv#jpu#7Oe3{!PMUV7*~Nhm-{7Me!xPQPH9$P~X;b67w|~YZ?&A2s(XMdaHd=xS
zl3u~HDnKT3OB=8Qv&idIqPwYs#L4lpT7B8I-_)>yZNEMORWhTI?qD=5v*-GeDqAr7
z#k%4~P2@mq{nQ~c{3hCcy{qiq11C<cd_s1k@OaBO7^;U)pK9=|$vG#Xpca($_Fe8z
z6`-W%E!t)k36ZV(hpg+>9{?$BHsTb=kJT@2r$)EHG?2z{!{dX<kg7GjdVUuMG>*D1
z4M;XHT+S(PzRy27Z*MGkQ$D^;i@VksmIAH0FO0#Q3cesi8C(G4T?XJS)@nI;J?ILD
z{YUQ4QHKu@b>Fh6^=2pb#JvGOLhyF$Kmh0eiyJ5K2j1#Oa=R&_oASDW@PMpS4#YVp
z7HLoZT;&0Q_ia#Az!hC6odGikt=Bt9?)Sxb90pP0g;H%^1ne6Q=1tzNe^7d&!OIT~
z21HE$n~Uz%4*&y!WqKN8cC+*>N22B*J{7$B`@Krg4<7_h%QhO&3{emH6s|SLwbz(>
z1StbXiB_L&CeNxuY)J$C4z1zIzmRA@@CRew3nkxg1)Ta~=vi7Gk%|Y<;?P?qvX;GS
z?$mON7%DBIQ!U_6!T@Rw+6iLq+i0Cf?{qqouv)4OtpQ%%F;@|>6G;Lu1Vy~cGh~Fb
z3p3)79|5Z9Tl@<U@*S$Q`3F=pm`=yoy~p~78dNcADofhODCpV<3?~QsIX1sh2>VsV
zgo5;(c!q@bIr$E;&*A6NE8#(Xtkq`KjOfEe5Kb;}mv7vmiXURioOF6ibe(4tOG-R~
zb#Bym=D*l!>zteK&Q0WzLLw|bb7Q!a*#~Tdj%R_lK=~{Hrq>p7{}pKF5<UJ++4iTG
zhbMbGG4NsX)An=xYBvt_vDtghBK`4o*$y=Gh&~lwW>m(0lv>JzhTQw^_~z~9JY+QH
zm(%?=Kli88co}nL)g~fP&l_=LO%X92VP1{3w&}K})Nkk>w3l=Lbl#Dd6BrGgzJ9a_
zFN3R{b2A6bhcCF{TUbhrTSPT|uPd=R@tRF?g~jBzQ}g^{7s}3U+7m6>p(l(jsW?V>
zcgb00H>Q;&t<3gbgp8`TKOZ(fUl^hW@yAi#O>K#LHr~0xrlP2B4Y~~DSdaAhb}NVt
zX1pXkrk=)&`Sgu7l&co=%(X{9Z)!P3u*S!EEpq{zCr&$ybilQQdxq4!gl=QQ4uY03
z?eEwsR`$!LjW)|Yf8NcTt1`h|w;|Z|-5(q?qFHj)F*3tzoeaH@3*-os!xhqN2I>b>
z7C#&0Fw2ZL2CY2I-X%Lafu{k;;R;NnRj<7f21~D3eXX1r%#0lQ3;j9O^T5Npd0=~;
z0384vQ=2=ESMAKz=e};|zS;yFaX+7d!SaFe%gVr}9J*h3?CFutv_P&oe9ktBym%A^
zkh&4zuB;U#(7aKXgx=z~dgsk0Cl}R@8g^UyqGZi>A~IPEFjd^P1xbP*wU?&GtV((h
zldPN2hOerAQmT26a`aih6dJpF;{`JHmH<L0J;@e(V-|O8C)lLcP_H5VU24Wp_g+Is
za^PBu>BiBBR4j<Y4HVIdE42e`G6Fs4>ZXa*e(A7|>s}`Jg~F3n28NCnqG`8rzCVRu
z6_3JqWfEE#sx&0TF_r$E_UtRR;OFW3Bee2|xOqNs&__`YKyA<kk}XzA=J7d2f({UA
zxp;L|a(gJkvS_1w%TdYne%kH1zX|D!i&?6Jmgx^hyF6id4YG)~`Tpds=|sxFf@R7l
zoGpbiogXT<uy>##wl|a)_qVxbP4N-DT*<TE4h2`cQPMgnDH2}r_58k#<XB$^fsvL1
zoaQV~2#jU`$5TQlO;fhFhQrnc-WbqegxPfzb<!Kjn<Q*7iBtd5<r<YN6?P#r!8g10
zUK}@hzePB~&QTS|ypL={O_iKI0C8WL@7_s&@2!3LU}3+vygk^)(K=@UtH5wU90_ZC
z{dR#Pve2Q-Uy^0lfF*aKrXE38K2#dQ)P$hwY?nEEztBSS!7D3>r|!-?JOcP^;HkNA
zlHdds4TmftV^nTZz!-?}n<Mv-9`m)eym_aL*u*2~W6s1+AbW~O6Ay$fhiMkxY)cvC
z`5^oRg)(m;iO)u0CgpFZ^a@>;LndRl;(W#O7Wx@gwz9beHzhhe^{ty<hs*6wl|zj1
zwPgV|JY1|RJNaM4ujPgU5C})$M&MG><I&>L>S=IffRPuAC#Fo{UvHtKp`+DPi^~Gz
zH?aOI<tr`j`<BLUoEU$9|G&d?qu6=b=|lMWFiu@YsJiPrn+#HVvEnolI=LkW35?81
z`Hn3S&Eu_lmCnp;gB`gCfh$foWgG?IX%?IPSnG>r0BG5TMGA}WH?7lCES1PV2#jGf
z3E3M9G4m=pJ-qB%3hu5ErcDIJ3;5XhdFz}wI!=P0L8A4S4S4AA1UNEO<=u!BH}2|M
zi>=XXz@W?JFsYTOmBe@IisP1(Ij2;4omm>)1k2!VIr?ab?JpXBf#i;hfQqIikQkWP
zDdfWq0L&C?nX<C3K_jM51NG>)s+Weix`PI!f5szxb*8FIP`A2N1Ykvug}-8X@B@Oz
z3!UgIvZ%t4!;!vXqB9nGb6Xr%<SSMvZvLfxB@TZ;?2A(i{0sSLF9oN6;etxA#9l(p
zOU)Y+n=#L0`=9RtP(kzW^CKyi@as9uaKl>26mjT=n!YsL5XMmBq{Ej-#NlN(D9wh{
zCZr}#B#NW;sqTx4seVK;7sZ%7Q%C)aAW7rlKt!hw=0>$Rix-pg1SBk!v0vGwyrE0k
zsG=0XQV0py9>yVPX|;N+zNrZ8I2N^3VTcyYb4JDejR3_UEg7Gh?z3d8P=hcUS{9XG
z@|bukM#X&!VL7Oxwr6%m$_uB~3UN7<qOuR<HNy14S_lnG9Hj~o!;Boo{rUdjlU+Hu
zqU;Y&Eq)Q=zZsiF|C1572lXMjku2I4eO|jMkB2hoDw<!L#rl!1=}SInKc~#w!9ZmU
zxdw=cI50Q4^53v<>|K9*s-hF!8Qu_ccI*-8a%RkvwQlI2VP0`^^ibnIcH+0#*LUbN
z_#h`+ghtxsBg5+LF5<M^(Cdu#rD#RO(A>HDew|%?tXwHEME4M4D2riMj^p)f4)ADC
zEisrXUG!GkCx5xYk3`Kns!70BkLLr|NDBCvB6Knn9j;{2Xqvu2VgRW~$oovQ;S8DD
zis8oelu+3x7mCLXCACARYTc!U%20~acP$NPzLl7kO<D;rivhmoMX0J)hpIqEIIC_U
z2dZ`(sbYLr!00j~^`kBYyP^(O6&5`XpQbkH($=6p%R<`$jIe_Nl&?J~LN9N=&Jj-E
zJL`EijCMJdgMvsB-;2=-IU?F4J?*z|MnX5acR$iDlq%^560L8-ssa28#05Jo7%%b0
zJA%Qxz=QP1K-_7{bmTxqO?{A~ddh|oFf7tBHtp>_dv}C2U*>?nO=xq8U>}idEWd#n
zX~ybJGAucqw=^-QA`#c5;@M%pp#WT@UxLyUJB2P-YNo%t9>hcl($n>vy6lC_HE+jg
zpH9I8!ZpUR?ZSJwgWmoXLacb4Nk6ed41y#M)XI%us5Q%RMT7$qwM#bFw<vL!V8sGt
z*70;|G&&dfDCHH~#<^)&Vk!4}9$FlaHLgXAEWtAU&Mu5{o)Q(ygrZALK=$Szw4vuG
zL_U!59#`{6p58AxY7k1SrMs9Gn^_DsY(U=brAaiN$S5r6b2-(R{k$#wiB0q8-X_8V
z)nyZWMqBH4yO;YYHcdC;nO;3KmV3~fyF#;K61drkOsp3MPrlWkx#0{74l~*>nRmy(
zLftB4W@h=ICLd{V<quiE0O0e=ZSND}@qTbXGR+z>wNaHw!x0$yq>b_Koe9TOly6r8
zjY9T_EG1zPa)Ss}l>$1kO-X^j6BR6iAq^OZ=dS<)9UZuOR_k8F0MHv2r1>SeO^_q0
zHZ<lJCwHIshh7Tz*VuNz>-FyHC@22qsebTYI0ED6$HPhQ{_FNWz}ZiWi|qG0BmdW-
z3lfX)S3<mFg(kZ~2QUSgVrKby&0JWOxRQEYx)gK$qA|`CZ?I}nHiZ0YQN^$05TU<*
zhImkG19>JUuq%*ggIZI{zN57ljCwwh{ifatemW&sQh~D<%@fU0faJN^Q8f2!)3eqq
zXzc_hr#D&0CF!9G;O#`)6G4+SMj1ncrLbjJwUVk>_q7!ah*Q1?;Ux8$&<-#d4nVRp
z$TuP;+jR#<d?Hcm>~&+HF=&nmQtFJ*J3-4g8w@MQCLgLlB^oZr>6h|YiO7uY5241M
zf``eodGY}B3fmYh>(eX_-|!oQuA)(2PUp=LaH*mUtX(w#=XUX^qpTl2pj3<t+J<ih
zFbg-kw)Ww46UKA~;v!1%=iy?ZSa4`?%E3M4(YLv5IbUHyEGzQoNNk%j{P0Ic9(rjM
z_&lfM#0ul^nqh43xc`!VQc*duHPM8NI+!|`_e)U-7k{V#SqFh&z<%Bt+Ww7ehYI48
zqpkukU;{7@L!?pXyL+;$aOE!0BZyD6wR<wa+pV5L9O_5VibE)`^gYqmP$qeocIshy
zyGF-V_jpulvJf}Ryi#J~jlt8|XuO%@FeGMhULvE|d8TxJxYj+I)&;5q?Z#|dwg=Z;
zoch(ZG(Hz58>S1z&zxv7DdK!X2e^o%ixAFKjRTN;mHL!vrj1xO6LoBkE@iTCl8`9?
z!PqD9nXzzsU4n!1-3Tvdq7eMHt~l0WNIqxQm|8L6t{2sN8*4Rvh+3(#ID<ZAa?Igd
zx-Ha{BW43mn*`p9uxp!I5v6OvTGTxk&z#m&n0A(NS|%@wV5=a1uJzzwNzxiz+;L0n
z%L%ak@mRfQfweBG!W5#yM8~j`<#eXCCWtVxm~zW&acX?2gU}lCqoU{v+aqk-1h?z|
zTf;}&ZkyJzpA`}DYtfrSu~%eea=|za6o2HDjnTkjFU-mvzPseDg6C%-d0!e3vWe@W
zL6|)pzkR4eTFsR7WCA#ga(@U4N`_0}B(Tt$bV=b<zMwuPBp@KO#h4RL^Hp~FAJm8)
zpy)_YI0>^_u+04`9#)96AAr#KAZ(Rv)oC5m9V>Po4dDu!jcUK{OWnj)nZ~aHJM$K2
zML5m8NsF#|Ky=oIb=J-vzzj!oZmrOyvgTV4{@Tw;6<Cy`p#jS-^AZ{r2-4yds&rI8
zIhrj*_=bWLMvh2+9+G4ea;Llcnk^_$^vfmhhMO%spmKsF=V8IX6Y$jqXA*n(7faX9
zP`L3Ssw&L{y52&d=+$%m)yC%wR9l<7{)TIjn-5D=Md`0D9dkIom&%`*llvk5?#cIp
z#X#y-k3dhNB?71qG>0{*54>_9bqhyGP~)i1o&hVZ-%3c?^<(B^3br`Ym|G$YF@(-L
z$VMC{LR%!NMNZg*K=d#!9&kxzDrH`>h<r8)r`jIOA2e-b7Zv=D$EV<z@Y7EhVeaD@
zmKgG2ss@HZF>K#usb$@!LA!!%&dTr)h<T%IrN_v;OcWr_rEV33{FPJL#0n=7QT^yT
z65a#z54H>PBv`{JiEg+EUF^HYQ@t|}7j#jh7tCF`uZE)b_B~DpM;mRv2hM?Xyo^2b
zB6T;_iR;{#H&pfl>1RROr+8nLT!X|1QOb&wsft<ZXgr9mkHYgi&Maj2V6O<0)Tp%#
zL346NOCSIb;@N|p$iH^M=Q{T>5=P&=*Ahf@@l^JbILdx)5Q!r#Lr>9#$#Q`qh5%iF
zU+_z{>z;Mo{x`VYy`+ho^Zj;d*Yjbi#)cm1<@d)tPHn1$B+2#U9NZTNO(3?+tGcZ~
z2a%p$++&u2UR?K|JNQ~WcjCv<E5IvO?`_lN>m^|5ImDHya%eN;aPe$^*ZK40<$NEH
z<%x<l#wz;~2v`OPekodfc)Ja6!lk$VedbRv;#`(8lHT5;qc!ErnygFvw{>KEuQ;iE
zhRWKG<pjA2HkDu0Fy>am!qZ_n)Abw4NSc2hK%&1c3#!15LB2I!&9i9?vB3M&nY!{K
z24KDb&06l^KlyvAvz8}u?!W{$lhCo<0<NO7%pIXIrGaUtcm%95$&T?CfTQMkSdaAu
z)F@d(>(p+{>Z3$ny%-cMY`cO0h^0+{a$4Bm&GA~8?+WQQ{mVc`VGm51akolBa=Lh0
zMOO;l#QmI|ezXDMb%bi7Z8rb?Y%>dh>F#c-j7dKV#~YFm3dd=<FI^-TrZl^&{=eLN
zQCu5ce_MKzNtMjE?BEi|`l~QEG+ooLprp6MoLsXxc<6tNdk{DVK#r)-r7NV#`OFtO
z87QB(JsHU`RkwO#?qDiJ5&|!wamR}=zIXdsS}jvvzK!88IBX}Zse?0^Yp4YQkRTTV
zz&fVP#dm4lT2M+^R7?iTzt|C4GN%vrLv`G1&l_}WrD$B2-qaK0@vhfQMU))U@W=;`
z_l0U!^hj9n))Z5^^u8!+Oav`xB?Xf90AW5W&`{e<2NtY!otTtI!*v>&_H8(hOiy7T
zC~7ZZ+@=E5d#%TKP7W`_0DCbYO9Y|z6lHBgJa5&yXJGBL?yO2KkJ&t(%|BsH`$;XC
z#9{l3S>a8B#U-LHNmI&Ai5NU7$K}3dIH#eyN0B?3^`1LR=*yWX$o;z%;hGH3UN(S_
zK$Ciq{yXX+itfLvcunYR7KMrbQ4oqujS|wkdx-&X&b<SeGY-J7(H}}Ib$MzS(N@@b
z-ZLLV2+(Ckpji5p<vNR{+YU<c)wm{}Bxh<pN(qS1*9~Z$5IWB=o%@k5v=D~1V3Rk0
zWQGM-UX5fY#pTIIhLqNotVVapsPAQDG3b6Rkv9+lZ^;MDi_FKeb$UF_veIdiIV`jE
z&fvWhPglZ3bpp)brthi3XdSJ{9F2R8;3UOb<4i1SkA7GHGNuHxK1*oN@7bwAyMOr8
z4Kf9mcPTG(-4ZAe%|u)s$G|Ek9&e<W@@C3sCQk@n)*ic`5OK+KRleU#INn{WWBR$_
zvFXgjq*7Qkh~ANL!<1B^Vt4HQBx+g{lzQv9@Ojk|=>>GIISF8D#Qbz+W%vHB_0pVh
zSBI*#ySTD6|2d~4wPxEYim;w9#!3|knPL^l0FU}Jq0H1sE0D&Lk%@wDP>3x=$H(Kb
zQbK3cHs|39bZ~yHg)FmZ^Jb8Ki1S=t%7&V~16lclv4+W@fYxjgPO_m)&uLrrXK>d9
zYt2CwG6I0MeGJAV8Deek8eY^`{Q;YUByn(Wi0}{Bw%lyWKsDuduiNjbNicKmEFWWl
z0M_$c&&<`gsR>jQlqWa#)U{mV8540H8>*GcR%7rzl=j0@WhA$mI{u>d%8^V8K?3}M
zmCddS0GsH#tXWo?KB@>N)6|;^stQPxk^TL(Yd?T)Es-EHv^AP)Re94^z_w9NY?=sn
z8G{V3<>!r~pO3b;0)k~Jm+|=0k(DBwqUW&L3S1^D*LoJmwG@LLRIdl`A|vMn@5wVU
z9-_)!cN?1}R*tSidh7=Tlx~o`)BDCmc!1nvKHx7E-38BcjmfgO%fOOz<_P!{M)Y(?
z&?7)<Ek(H0V%Jr2yzhk1-B&Gz*QMY=^Ib{a0P|jq07FSK1+fmqAP$BSwDy4T5M;|P
zrX@|^B}%>a4l_SS-9Kn|ba#@}ft@^s68etOf(u=@d(Q2Ei_8f!U|5<<-U$-9pSbn`
za=H3|;vPJvU-`HU(K-2?JL|?o?Opt@=L4XpYB?6337F{*PrSc~m8=gfI%7FjWvmb+
zJiieKjiy5HTgh~=2p?0Xm8{R0F8h`F%Ey0TW9&n4dwK+SK$m+22NjoOmw5z77%uF)
zzbJhOO&C!$!)KkYcUQ-3S}<Pz5<&iW7NlQChJ~L1Y^B2*+8!tztt#r2q5$Nw0OTY#
zWiaelKBs<}B>WMWs-P{-M}}&V9kfvmWF$pUAarvE)5Ov6+Gw7~p>_>9_IO&;p^8Ul
zvZpj7|8vUIVrW}~lMy@mVQ6#Lbvw@TN-L^Lke)@nrJaxi{Q3jCmAw+)!bX<VQA#J8
zH;3*(`Yo{ZDLhl?E5N~nkaJnT^j8<l4g#1~osAft7_&Ay;Kw|1DAE+wkcOir>g5oU
zdW&oIFQ_Q`tHoAu-b*?bQ91>K^zIo>u{~YEU>KIHdT!#8F#n#O*AqA@={i9}bs0a@
z-(KFp1|y<IH~G~_jS#ptV5`%hy%ZVE(%2@RS`U;(ZMw@p4hCrLahFs^1K5^_URZw0
zU(03Wba&zOlu77)L#|5MesObOJ>1wlLu0>Dc7tGs3ax=ocIJfjJ9{4&yiq2_`~FP)
zgz9O<x_!~khxL8*qs?A@7iFCsS?tX9sPI6kXd=yb^7{`gg5Cz`4F#RH94OnXik950
zr>Onb1%eteUOSeChGkBrdBA}2btKK_2RKS*1_E=|ogUaXSK?of(!ais<}k;MzY&02
zaAUK58?eGE3ks%uZ}n$k<6}XSPBphq26>D0fDddaDc#<td?htfCrrBywY?($u*1%o
zV_2XPAKTMnMM#cbiq1h4#S32QZZyQ^7e~cM*fGs)M~!PvF*A?$6bFpMHXQG%TT4N6
z%y&jGCfcW6s%7nP5IJ$V)~w`w)rIc!jGnWaJt!m9>bLU@CF)@Z^In;G!q%Et2<U*L
zfe|`rns)zr`T%l7DAIL*8$FI;9H$2+#;DAV*4po}xnk;!68Aa1Sdw;3p-5TnGk-?5
z(Y~y(s@D(+Fj2GfzGUgVbisBfl0u$F11JfDXw;JVLWm_(Q@=b#j0_dh71A@w<@j(=
zXk<<|?j|#58&A#&4BO%4hpLOZa)PX?Y>ta<w`+S834ac-1m)2w3q@`lruS~suXT4h
zB&^kd@4yE}`e=SP;hx*CS?)9zqKCE&PE-2#=;kK*bu8M^0McDOcg=?plr~fxe7=y#
zd%b5zCTg-^C<H`oBj_bx!!`7OkB;YdBQ>UIvt=ExwhOf}ZN0TUfM~OZZsI`E1WIC}
zkBO*>q?%kZA<D`Acgx`0x{CWgv9qr2*fO_W3%b))?QnZ`S_qEW-Sq;yiDo@HkGos@
z(`U96&zJQe<AIy~Pmkz{v)yH*_<Sm9rM2aqKih&liai1S)Y9@v1xEq>czXB2F^gi)
zJ*@rQj?`YyJUgBvD4$6?!8hWPHq`QEJS3Y6ov_3k3VOMLMsV#|JtQ|$aHv(wzpbrk
zo9xvkc63t~BwYUPHlzkySz8H{4G=TyRit@CEWf4!0OUHHc>Pc@S+JcH90$<`W;XUh
zjYEG1=+&pil(U?dmwRF$Fut9((8~8h*BU|fi{#+N`@mAT2K0Q$rFdmHV7D9J>P3S#
zSiQwwCHZP5oHIaQBwf~5X3HQx(&$(wnp>{mIr9}akT<ZOI`f^6$K{0*7(~?hIK=O2
z*DgQ<NLK3rBqzpmhd&f`b`4?f#m0n|+3SCs&wk?_;y^Sm`RN}ZT+a>S${5FB+?zoU
z5_(|-6e0WJka|`k$9J)j&kSCnmN)Bz&*#9pRW;z3Afd6DB)NE5Hn-nvW_zBI-Ey0+
zAq7Oo`0Is`wZgdn`!BgTB&qD*IAu7fiSoa+fZ@s#e7H)DMK)(2oMda5w;Au-ZFBHn
zu2Rx@N258@jtc8XNfQ57O^Jqk0ZBsJ2)oA2`uuqy*+IdbZepe68|YG(u&|<tqQ{s5
zg{1Q7Q$9_4fP|ljpHzm_@<vTg!$M7WOv5On0rc_Gp8~Iht8xMC;g!=5R~adxK0Yl#
zl9eEMr*4Fc+zMF(r0;4CDg?H{tGYk&0CCtVgg_e0PrD^xWujy}c=HIU|8SbS!Bw~H
zF#hR&{(4fMN6UKRAI6wI#V*>J>vg1R4WzvVayIBDr{V%9&HW&f;gWcHIn;K59X=2z
ztV@Vkk6Ci;9NNjcZa?K*Mju~QBbpR2lwhFyeB*m9YOI}O5QYEq#~s&XWFBv7hb^eA
z))axbtJ7gg(f~o!5e;bcEwUCVL^L3}5*Y&uqZS$iTfsocZwW&P5(?d24jUW}fto4X
z%^{9w%pp${E@#5WGmfWR8Vc%=VZz!RV%686pMa7=DlMKmDwi=tVp7F9?*}g6-|AX{
z@rSS5&+M{}I_=t5omL#DgpatOb*T`YRS@{mfUnhwf5Hdd5!Goy-rY^y2>DkXSBXM3
zPkSLKejdexeyIv+|LC^^`f3%!uaIcD_-ai8sad2M^x1M7`e(fPiVGMcD=T$Z1HvFi
zLjVcZS#`86)Snh&K2Ic|Pa8mSKWR7Lb|T}urUok(EPHr|PuZ%)UYO_rIroZ{>G#q+
z`J}x_^WdB*!GlyqD&sDu66YCOESc%T7PXL;?V1t{!mq1DcAdDL=JP|#I4zwoY!Hga
zDPRRFC{5|(RwUbLc7!1OW0YojA8j$cJnmD3*x%$J$>XS;=R^qwzx=n&Y)UVv57wAU
zkw=mtJ7#g872DD}NRh+8=v$OU!qnZzT6H|O!NAW>L|zxPh0}&f?V^-W*Sh6n0+IH|
zAq)axblk#xZVM+ZDiG>z5*p5-1|Typ-)niNmCrWmDs{<Zi*P7ok^{U<yNkC3m#@9>
z6K=IL)G13L^$~HAtr4Jhj1|a1pEO3mL&E7){&g7Vr--&ZHD1*C7FYXapkW<hjx-IU
zO<QR8bq`#+S)cBg&?S3XC;4OJg_6>>ec4=+%I#Xq55j2zqZ+|;&y1)T;O|Bbm*qJ&
zQ+Bo<LgSmj7XoK|YbjV`%+h5fZ4t$}fhRNdtot1?e(UsK#&ZA<`b*&g>FU>ihZG8C
zi)D{|(iw!iEVG4hMUU6nzZ?ONM9mg_skvsbG{cHO_YI4Kz5sV@y7g^~w$}QQkoX$D
zBluyBmg<OusquVn{A!LnBD-zcQejcUG4Y2x>lUmYqO7yFD36K_YP`*gc^ep4M3n>|
zfykRZoPG!YA;4eOijgnJWmW69kM{YWE0wn-2_dzry9$xn7i^;kqk-po&zNeBZSjqC
z`#Wl^tGCd8NF3!QD=-DaZ{#?i&u<m(ZFxu!P5A-Bky&|56rbL&k?FCz@N+Lz<I~R|
zEe3R02ACB56wfA#Z{vPoQOKr>OcDN?=wrD<5SXTl<Nz~<v0Qw8Op0_S$v+A*@n>w(
z7%R0>>6Brt5y>Z`j)}=fy<(fnsNO;4A2#c$s5aHHHrM)QajXOlJnx0-8_~<P%~mPE
zpAM6my+F$eHfaVf1#WhS?hT!(%TjQ5Z3gP^TQhaiGnGufRvJIh(*=*@5)4H&(;uqW
z1D~%(<N<L#$0cYL$vnH?i#nBe`Iws1NeGp`ZWn0sU3y$B)%J(c!2JB_>kGOQF~lk)
zK}z!C7a&$~!H(@RVZ|doq<tW-8muQ-&QYfFw^rL=O4?#w6{$-KJjq^=TREkDT^?PX
z2!Q+;hv<{->3e~@(+&G`>5w7BVU-U?8$Th%3=F_UGC9<@aM6FPaQOnZzSa;YIQAw5
z)gH&vP1yorMY(dJ^Rh*_Har%bY;OO+Djh<}ysKvL`z5V=phAwz@+LjIrzfX%O3S3r
zK9{%BXo`lN3etmDTxI$%R(=jX-}gyuCc<>yd@qVgXO=?bup4I<r61PM;TN3VggI=*
zIemcG-mqhimGy2&5F`BeO=$IdoL7Q24(#9QoV-|is=w5b%(qwDHPt(A<|FPT-up*l
z=~&b%`99C05etJkDidi_Yx$fjl1zbV;k&GY7%GpQf_f<7VbS6nk<W|=bMG$ZW3@I-
zo<bhsWs#eFI1s_Pi_m6mI|esZ10WHqGIjvD!9?!qGjab-Sj4_XBThea4%;D)l$hUK
zjiLZ5EC95WyLROFF|#-1Ln1v_fg+2^BGrcLHg?w%T>|y8n1&$?y>OipZ#CfDkYmGx
z!)?Yv67J>?%-s{|hVGBzHyuMkNaI>JQqs9NpLIR`ho&;vB?JYk5{HCmG<RPeie$jZ
zTs|36!8M#ZdcR55W=8=|@cudJbpXm3(RB|tVO34hr0cp;PtQ4eA|jbPx2nr;QuoN6
z+id8(CpMUM^WDp_E__HiJH^&nr&&k1J?1+%>eDSuyZgHd7ubPjah|IuSc<9)YVXR-
zIat7NPe%I7g5!?j4QbK#^zbbH>pFHiAD6GL6zIldN7d`*fndLSzwOsFBBKCpCj}vK
zCIung2V`vh740^~Rz>8a;Bmmv{ZXQ35;z7p4FC}TN&YF^?bsLb`0dj567f=bL%sdd
z6Ztmt?45N3IIKuh_w!22PH&?|34P*2Mch4*bcD)p;iuR!r(6u`55ywz8PqUYt-sRb
zTfdTB?y49hPO@izd|D8=0lpxl_dw&eFiJv<Anrxkq0YX|;pyA(C=-QB3UA|5#!~Fy
zB;W^CM`U9!$VOPo-|a6jmqqqrR7pV5ueX({*lD%d#&>(Emu~IPG@!{t=Y++)n_kpx
z!$ramBa!IR5sc1TSWc(BT)i>FykGA0t~4>=Q^#%vZU2*}6GA=6h%q#v$G3wRBrl|c
zSg5;U1AYfVzU@$^;H+r9ZT#V?z5^fP7=RZ;-@EfC91<@I_Y7!Y|EOov`S;_j7rscZ
z{<Xppgax=GDcG8Q^+vFonidD~2!cR~^iJkg_r-O1`26lrFUr?&e6-rnNVLkRQJ#9K
z8{un*u*E>IgVe?SyB+U(Yi?j`w<AzER!@Fud{BkN;xdGqOT0Ne5|qh@KbFDc1;F_6
zEAvC|`@Lk-=)U}i@KE#}A-}$rE7gD_teSL|4gIT}V_6FSlSCG+YwG*^8SwN|SDM9T
zDtlF>gbwPmD=`<_BTn|`@A~OeT*2aBu{6%&T~)P*aW&4aMJORG2a7tRNc8;?cdteP
zrbYK$q4R9p&cpT(El$n5Cz{kC@&NfCjhMVDjFncf8+lk*Z00psSPwh2smbUa&Dmko
zm5@c_re(ZC3u1{?*pvF?EoPJ3@|}eOu2L+5K;sSS)?+1YtD{(=qKh(FZBX>el}bqc
z6Y<-8``D<tm&JVy%3^uLKVrjoSR%E@aEdxN<t&`{*{EI>KzMfsu+MolCIF)dndK|V
zF*o^Wrf5j30`n<ab&U?h86Tfx@;V*n#(yUSNgu4PsADlC&pI{m`InkV5Jd=AAv0i2
z&fG#4esGjTTRC8z{pNdN2_*6|9&sN>yLsSI^UBUL7I3LEP_tT(@(?bieI3x!=z84?
z^3a`s|4!oa0yok;tBymCN8zEH#Doow0XSbjE}OTCk;t$@@czvS_4bcn!2ba5ZG%Aw
z%%4S2Q^T;+ELTo-+wr(i)BJ>}79IA>O{RSoerfz19#^Ygl|tDvY)WwJxVjsEBy;o!
zbyYW95m4Y#$fqxT5HFU3%&g7Ql(P=mNhBMkBogWTBTl5rb~N!cGn(JCh<>F2aPwHE
zT`{0uV%V*$4aHIrmGR-aokYtG=JgGh-i5xOvjL)MPU89e81UGza+Gn9kKa9-X-nyE
znRX>IP2HW5XsDe?EqkgIqGY$9=8o8&pAVtCdHAbxvgcwGt-H>Gq1=||tmKZ9#>o(T
z;m^ho!lX=@-MsnwF*|6<g8lat@YPy0{(S$_j;Ghz@)2|=n20L795ryzN9Xz6&%26n
z7ZCxnX#dq*tj7I+Xh}H1%`-gQLQ`WwwX(PxQ*1bHveme{bogNJhDAaK&4JOymw=XX
zZIS{uvv#TYP}Pd-kwHAEdc1&e(xKr%t~+@POQ@E-=seky-VK23B@O%X;{yr{f<S>w
zn=8HBklksMjPDPIzi&ES5uk*oZw$p>T`0>|l>H8wn@qQxmUXK}o`5eAJ>x@FSFTC5
z7W*Q=82b{QrrZui<kXeqstYQT6zn6RFIx6^ognuFf0>mbmu54lNY$oNntRX|<O0yP
z=%#*MRsOaWi5?JRaf+x*a96ApVbrrzJxj22<l36E%-188w-U1cpfFyds_PesrPh)F
z9a6eydhr)_9^pZE{awNaQQA@TuKgzgIimcWA^ZbL7e6Leb<)|?F@Xu%Dnpe)53idm
z(!ks{Cn;|Ku!MBH_N0kM3Ql6#l2fU+spM6095e<gsXd^GFVI1^zhsJuLOvgpiNY*r
zO{3CsJweF@lh}j`gYpdY+xah%exE}XYfeXNiVUjW*o4tsF*E~U65=H)VeYHA#V5z8
za!i4>18C4|UW)`<eDuj=8X3Pv!qKFcX#xe9;X?HFlt{}gmEW^3^k9VmeL_bZF?$vY
z%jl`75iKBPBq+xC*5_kOA)J#5?zK=@dJ&fJL{=B&xw=o{O33>StHfIuP(^{=7ap?K
zAUBZlX}pRTXg}~0(i`i4kNi!+B0<Kt`eQpR(BQqdvOM@6d3CGr>qAF#_){X0h{$_n
ziqPlkj~pSb>JMQLvFZ=d(7o^Ii>Zd!;KV{pO@Nx4Z6?BnL?)Ko5omz0zi_#i@7TM>
z7c)spQQ_qq0$%4r(G9w_W84u|K4j1VyEFxUDho2@Fn3Rx%@8wO{WG8{N>CCT?Vvfl
z6!lDp2D%Z)Fg%0K<U~ia4^sZUj5BKknXw94#=<lcWYd!B@Ll?5`(|Cx9a=hp$%_-t
z6_9_pAj<A0cALUlxwOhGdai0^`@^>XUsa2<=B|$y!6MO(g(q2d3bLd{uD`jk{^@(E
zZ0~KgfSq$b)0@7JnHZ^_NJ*M@Sa_iK7f(Gi_62q3jb=5-?6Jd|O*EMP(mKyUi1GJM
zzSyJXC^D0&Vl;ik&FOqdciXn-O-$1w02s>43L-*~CR@Uy@UDNFfGb~AzBY6UY<o)s
zkPgv!>*zxsw|FL%N0@m_h&sD{a3u%T4yedX7#P!1iek8hs*EW<_uqsqm3}`@br?n!
zY1rq7u^AEXykfVm{_84$nC~S~87#IHj7q0TPK%qDxPna(#&(%ne1#=>v3!{an3WG+
zkGOT5()mO6@o_U00}vZb#;3}-h6NedA-`)ozSmw?Ba_Exp&XtzWi?hL@NWGjJWv}4
z7QqYBJ69g((qa&f6l3u%q62F+_ICqr21lS+;!nK<8Jsz{<}YW{btprFa@kbvsl_$K
zHJE^0$+wJG)>zy!+ruRr;OfO4VDE4#nQuwN)8K8->g?gC{Nw~8=_H<oak4Dyov5Tq
z=0_)x?`BuO9GDel8ipzIKO5RASQ{h_{{ekq%g%;vl@Pt!qli7Cv&w;W`Z{Fw_&+_y
zGe-W}ts@B(5Ut$FNfcExJ_}5ISdRr?k0C#DS^}g)_0#esiXgWEzGV+g{{em;1+m{Q
z{a{DX@LC>L4l8E8U}*Ahk#>7UB83wB+A9YraFiv{t;WEZ%UDUYP1QkM+1UTD5D>z2
zYwNm91nN$&?pWHv_LPAFgQ!W$Taw+@Z9jI9_f`vVCjY8uw)>yfI|qAIie*&wVDcCx
z^gT@=EZDqN2t5hWzM9D*NQ?K%`F(q{i>+;LJ%1g8+}v!i*8-PnGJat`yT!d{i;MZv
z=x)r!1oJw>UDn|*^t`)=B8u~iQ2<EUwb|DV$$Iw1Rj1lKQrvs3j@Tl>2UW2Om)?jN
zKt7@+DVtoh)muCcB+DR0w7ceR>B`f)ds*zv@QrYMC}t%W+#I%ihTg47Pd~v;g?TfG
zbmjh9dv#9pyjSa;ZNwB9@^p2fMPBmSqZ|YT6z)>1Cul-e!<Jwcklog`UgzV(iz$@B
z2_1S*zwb}1be*LM36v5|GYELeUHl#<;3tMTkhO5zvF=y4fmj=|IQF!WSlYlU>;E?R
z*R)`y8)<)=oC#;mZxU^H3W3JrE{8nJoE+WlB(pusAVYV?(Gl1z+Ic@lS?9cRURtV7
z63iXdoF4oip!az<rq<FXL)mvo=R3+jy<UBv+_b3*fVPm(9NqB%cjvLM+7PPI_)!`2
zhS2pTZ&T|&TNl)$e<J2M*>_QKh%!d->UbIvN=OL3qHp26qnqYvI;$x%!ur`T^le(h
ztWs&2?*b0}*3{oq_kRO(t`i)nK$Ae*I;jROTGl4C^gjun(k(L>NlAbUb8o!E@d?PY
zJp2@@ZDZLQpayri=s-w9bB6OElp)EI3(8PdEnkx-pX-hd&+FQoh!2=!^Y<MeRvyf7
zP3B^5XCMJ?U3nS{{+~>IjuRtR=l&~wLhYEZqtc@RJk}R95+r6yiE-Ts62k)*Op2SC
zcQ|1E2)RYRtPfRgI0Ven!atR_NS~iHYu(o_=B5#D-j>e^l<8Yc&aaG)bEfKB@NuC3
z7cv^-9fs?i`;;Zo6L@*41?s~03;@6MZ;?^48GkDx*`YRdNEutIJ9V4r-zjk*jKW7f
znc~#_%5)s+FnG9oW*(H(+(xl%V-nz@X4sF3Vd5AG01!q-%@8}~MFAbb9gB_&nRv<l
zQD(8x*JdJjuJp?Ik9=1SXceLkj%%sN2fMdWJ1N1L2!Dg2Fcn|P8la=MK$=7$sS@1n
zS$1V%=G&_lOwPygY%^M47VLMhbzseV{tiu2zOJM4ap|9oemdswdhL94{#<=4dMbUA
zMF@ZF2mI)7>I3hq>oe{LW>c~kf{{<@H9*0!YXMnyVz1Tsy}v!%t+OCkII~igk0ATO
z@Q>=ZJnUBKVBDxzVm;HT|9wlalzA9&Ch^?%i0!Z=7}jg<{Kfc@_eS-M43|l!hZ?S$
zeJ!B_m7#bteTuRvRVn~B$P-D8Y=Z2742Z$u158{kiGii1FmJ<Ftmj~-VUJs{g>Z%j
z5Sa%~(FpNzfkq|FNC!fRLtckIqMZDaAwtiz691*36h0`*|C=LK6vXX&;2&!HIXkV^
zC>K{dP+v^FVMSDCcMr}zhf8=U%U<=KEOG4R+gOijuBQf^Fbm@v54!4KrA)ts@PJ>l
z;eg|X*9P_XW8K=d4-p?qOK?1Pp4wX}|FBqOzl3`m?~+(70Wj9ydz7n-?uU<qtfw%Q
z`786+QLS@B#nLJ!>n3$2Y7+|G$9CuG(S|8DQCg$t&*(BKMLI<*al6L)f>HzqXai@`
zNIa`*g!2Qg>YYIWKV$Z30(Iu2c$T0Ke*jy2AOq*rAGEf8q~2mHL=JHmO8>_mJJ9eP
zam4`CA9BO~=5|K)r}H8ma<A>HdPqC#ZMF8cViD99T#}2~(Q+Pk$j0A7v0w++amLWi
zTFWeNZQC->72~kaKy-E6!g8AR1ab<&!mT<s?k}{gI4Rp?mwv4!Oc<@UV9AhC#sU<t
z@bdXNL<7t)Q9UpMx_70g#&9VS&TG4T$3Z~zLdHRayMx9-A_|orOntG=7sShe_<WhV
zEgG9#4&Dy=q@E$q)3bqx3rh=Y>Ev#p3QS?gN#dU{m2cOT0Xll6J)c%Sk}Mo2;%%Ee
zAqG}rRS^BDwl^dSINIGM6sw5TG=Rd-x)HDQz;g~J4J)q98F_ckO;nkCvIQGySoj5-
z(V#*5d7DF9IToY}mvtAyMr<prQQ^qE1?+BpBCg19w7(EUXn`tnbJ5=;YD6exAuU!(
zm8gEx)^`beu3>2P3TvN_Hm&2C(8QFAz^7xMThPjTQWd>NUOduS>+1qy_lz_7Lx4*&
zg^}9T4CwzEP&E?#)#iJEEQuoF|HkF4Z=EWTy{5j}Kza05KdTPCOil8ozArn+@nP|k
zMK(t*QKV!{mY>V3#YI!h4&nS-3-+L@GP>O4_iWvd#Ewnme!|YA7z%J)z@4*3cjjka
zQI8H*LwEd5YS!*NSh*XQ;Gox;sc`=0vSb<Vh*iMi`w~{WXU`-w)$<>FHY=fLNM~Q0
zywyJ1V-6ms)P**;2}al6)k*TscbOLmScy*TXDOU#qq#I_e(d`~;UVDr8MU2%#LR~4
ztUFiZD-~9S8pna&!ApS{Ant-O@Yn2q^2hxz^O}z0U*njf8Dce$`#IZODwoo>{2o*l
za#U$BrwSS2GmNj5D459y<hfxeff9l=y1D-V-`;|0d;LZoOd2KzC?N-Z_#(VKqe9PU
zZnt^bsztV4_D}7(jVOf#I6f1>y7Uz!NZ%icBLab5BoP9&#YE?T62^`<1TR@oWXSg9
zeMxDk5h%w02KiX0ino!nw`ZP&`5RU*#WEF<fR4YkU%v2Oz{x+nr#bsb@n3kasLY^h
zi|PyS!AAa{@t)Z~yf-KCh4)PJn$g}q0RIc{{r}-T`k{Y#k9&B~K&Xpr7)AWJlD!N{
zZI|<w^VaHIP(W?_o!(2dch2f;(8BpkY)F*E?3ZU@qGNY;)x%0J#i-k`Hh}4WQ1(ty
zmaW~^W@Om5ZQHhO+ct8AZCe?(nPJ<uBQk7fRP3|QIraV3>Z0oAUDvD4Hs?E@-uoCM
z&5fq0?BpW)lpLv2`+zN-BiWJz9PKDQ3Cig-Z}EF(sIvr~M3!<`*Y&oFb;I=~i794;
z^ktAh7ym#~=ydtIy}66YQ~}VA$f@!ZLZWLAMKMH^?KJf6pr9y8Dks>r5Xd=r=w(x{
zIm`Av26@;k6JiwVctG4K(+~rjq2rAWoae;Xexu1+p$f9;%Mbc(h@Sd&2m?7vjvG5y
z`b9Hy0B@^T)kNrRp;Znb(d(C<jd$r#7!Rf>16q^pP2h0YS};}RKb-kP4?9=zrQq}~
z9C(OvrX_1ionl2@5X)lIHSFFt#e(#dzy;2oO1>cs8ZNI%Z)Pt)c@`cm8^{hnXWHpX
zi_2;6avBcN`!jWY(q<>W_aJDaoY&4{Y2i1Z!{)?E^HQ*b8$lU4k$5&g!d%cOE^>a!
zD?d5EvJXm8siT7vt(0eQ?%GwP@%2gy8^Lnb4q++V_!bz7;UZj|-dW84m_9>s%5Z0C
z%i(ajj+H}nCTg)gSiO-C+buO{)tWT&yzdZsunTI)(LHtH=juf%IEsIeJLf2Bg7Yrm
zR{NU)p?@EpKsVDu_Ue4<_*MLiqKsuh1BBuU^n)n^WE-`~gfeL<>(DKFGR-zvVR0rw
zn>TUfVLq|{g@ztUb4**W5T(e3q9aj(Ph3X2=N{n;%vj$+N9>IRvdNmZJugmW#{m>z
z>B~eY#R8mMESlzZz$i<<Dd6~OtcB*?*k!UAe?6#A_uPo8PnYIcyD_~*vA`?<X#38j
zK}}sQa!kGa6*j5bPz)xi8pyKGl!jdAD35VOjZ_C=AqaXy{)kcOZ_9zcl8dk%_p4gg
ze1Kf@^?uZ9ymU>d>RO~T>*s$mA3yMp2jRcW2l$3QlL~U97kkr=;&0}Av=bi4{ARue
zs(+Yof*eq4wutH0V0R93FHQ#_BF(CC)b<E|GA@GW{*)~M^PNiV!A_J-0(o1VCKTSl
zQw_XjZ-zA*-dMb0Z<a8cs;MjZ-CBOMHcZ*LCP0>xY-Z}Bh~`&i1aF=gOPBO;z6AJR
z<m(A{-vug&ouX93D*IdeC>>LptZtA0u%q{PN&x9_iN8cM#Bg-sXo`P9s*5*`{J`eZ
zeN@#enO(jkMyV)KV`*8+rYR_qLAKUHvjbYXmVn$Zq`hK`v}(n=NuGqPDtNll_pXVW
zb7=c+8nq0Ag-E-zpxGlE;EW55ju>11y&SUG|DdkCh_<@m;mvJDHSwSF_$Nou9|s#x
zjy@6$+XLY5d*K8aAY*y3nef5VbK$VEZAiGPhL?x1l(fw}IwOh*(}_%x*JSVM*Ksnn
z5T%q0t0=0ih-)TGFCxHaS6-cM{4U%6DTFJAf^8t>(@hB=Wg!OCToFtLl*(~srGS8z
zm4>r3Vg1u_g%p5?of%!Ze=eH!ac--7h;$iErFbr?w&>i4OvlAL_WKbxnGkJ!tN<-T
zIZ**A0HrDz&!6{fY_mS<Nb)W^yej!Paw8Kt`Jik8yMn9U+=*3(Wseyt4={$7gLOe7
zg+<FCI-2k_`5Vyh78k!sJNjjDt&@%-kr@0XO}E|U&%=@GIgtGchOj0Vz42ts+xh(H
ziWou}T@q515KPco|MgiRjZ7F1WFn<l$Zb3<Of7oELqo%IJ{7@FZ9>q6Cz?NZ|M8n%
zWXTS3S5hC>nJ%UOiGj&k#k7&)aA$qvLXv2$Ui39C!whg@72%a@>!Z(IniTNmuX2!P
zw>0jQi?sV8uzm<>p49>Nkk|)5A~fR|{)2KktD{;)N-_h8Y;A$=s{CE5FOmz|$r#V=
zQyi7+WrA$fRDJr2JNiCx9rgO9{V2S!&o(L)g_vBvs}tR^zXFQ7A^c;O&Z2<nKtaVD
z)@@_CR0crP<RamPhdoR>I31+{{d&jM`?Z0M;^L1n0aoDXo5d_tQ{I>j2@+P0xC1{G
zfN3%=;w>Bz`nU~rG(d6_<cWAPD{is#K{^d)wGWIx4g-=Mb_WkJ-WO*Q^Mew9{Gj0X
zVaXMR7`3q1HE%KoRX4)v?;FMK5x(Z<-Lf2%NfH3<84RA15f|&B1R6ItJ2ZqRU)I@>
zY*Q3Or)S+iM>+{uAALO;x-^p8vosw>n<V#7`kTmG{-vDyRoZ3Jk^2)`q+w+G#MnGC
z;Qc(=8>d{&4!VrBDnz2eKo(7w&LdBuVJPAL2_I2?&-f6uC$~&vqsx7EQmmP#0DD3^
z0E-3&^pQH%vWq7PFoSDpR7XRxbn}3yM8F&;-UFn{J1lvjQke7Q&46mckm%Fo;@#UM
zyE!ecM5{chakp~e*ppsoa+Yx=`rP_Z^Q07}35Pzame^*Z($hhyDstwL=0GA60adLf
zLu4_mS~6Zi+HZb*ixq?n1;oKJD75d))U<!qAN0bb2yAb#+H{V`6~3XudGzrR?;_)b
ze*7!=#8{uF0@$FH*Ifo!Y`URw8y2;xah534Fq0kMDBnimlFDF>#1R0pbf$V@Y+J(B
z-7@oeU=P*$&gMqeTAM!lxf=SaXSey^%0j9`1|M$n)+ow55LK<HgQ9!KD7q6(0B!Qr
zM>xpM%j4g(7LaasK+<U+g2$A@JV{$m3BmgF4t{1AKc44?-$fG6Rws_*R_B*EM7-QD
zT))9w(`&znfXbg4$?iWNeh9SRgLv|UE+Dp<Ohm?CW=SfiR^+&6q4<XE2K@8P2>E~b
zZ1?taK&u=_Cm3C`;QWeucDl?0koKj)PsBGKg18ykdryVc0{Pso)?UkWN!(CA3Vq>!
zEdTcFkNqjpakVXlH~W^Czx3nQ?a7p6uY(f0R9Ut`&8u^jGto*}r)!6ZYeQX-Y7t2%
z!UuP+O%3HYy3dA2V%C&!v-%!5Q8jo0MLT&P)UjdItZ&C<C-^=Jy1xT1;3}UdeJx>M
zV%I6qCGcA4#=V+4f)+P+-@KL!Ttto5N9R820F*Yr4>QT}0y9cal{1NyqSfzpxx-$*
zgM<w4(xy#(s@cbMb)WKAO1%)mC<ZV@Ednp=ILS0H?8T3G^A>L4vlG$}TqqUFV46i8
z%I4x>-15}%-NqAj)Rvw;fZgtHegJBM%$5TB>SMF$ZTD|(=gS%keRYT$4lS*#>GMWf
zB85?nz%U4Svg<lGlGCNX8>WuEy^lfxTfrJ9=MLy7yu6NO9QFPh?+!lqdRT4TmW`s2
zY6X;d0IgJLy3VjBB<8u(AN;>kBX$+u#Jq<&z`Y_$)rrj2He0cb0OFQu`q0IqZIA)F
zYJ<h$C=oF87M4x>JC0QQsQbhZNWfrt7?U-Nh?cQN&n(S8B$GGID!X}ZEt0YRWRP<z
zV@SR}t|(nv^eB6jk)UF+>EX%+syjW8g}0VHm7`e-2K92Ph?`jLBB33BU7)A(;@Da3
zc$KmA%yC&n$H(Yd072dKX)w*+dX(*i+<{28vfpD2vLYv2?ZCl%GMbDEx8NJN5Z_>)
z(C`}GM&lODJBgQbz$uLAg>E|OK*q*36Z#sVVxaz>Kj_q{i~C@vP%d=5xH?UP1)awZ
zx=J)^+oNR+XH`{lC+J5a3w?J-RCGOwmKiMADg`v@x)cD*03NNY713e9JTiq03fdbh
z4XB{eFw_DTjc_AGZ*u~kK*t37K5KE3`73+b{Hus96TL?)@L1vW)Z7Ttp~`7<rE(70
zpElQ4FQgVDi`B`N)l263(cBxNS56XJkPMFptK1qsNdsM=n1?O1xt=p0=Po>J{dx9=
zt0ACIeG8d+0h8jx17Hbqi}fN?SFNUJK4s1!m_9!mn2f@<<)3-<RCa_`*J0yx{Uf8R
z;vnt|<9(-LBbb6TCf-(exp$LHpKta(xmK2p<)SN*`RmOya72_PrZ)#`wQ~HhtriW_
z#QT7{2-lXUT!2cs|Gt?cLtjdFVruLhakeQ>o9mK90dg)U^H?m`zhr8*MH4<=OYcW+
zhgb&6g#@f-4&$n$M`#{PWT6hZdi`mbo9Yuww>-|qMF_%t2uS?}w^Dd5rM@hmiogY-
z|CT`YCqO>X@kwi%WW+yGn5863@uw4)X{B@I!{t{fllK`0W|h{AkfJ+gfVjZmGl{|n
zMNjw90REy|Rf+YueL<SPT!rOe=NaADJ1dL~ZA*D^|D@DWE?@iBBrH>bxi(>p5eWxW
zS$XVar3NC&y+e;jPe3tegH=oqcEVmz+ApoVtYP#MB6>CF>f}5<9{D=$XTRR-l=hj>
zLPveTy_LVpJ<@1X7MzNpWmIFOQox+0ixtsn00B}^e<_kxHg$}$`yYkw-gd7QwsIGL
zAWS{gPT`#04uE>#H5bm>v;W!yu3<<#29{g-z&X&JU!XZ#QSi9@`TX~4jte&bb1@CP
zb?fJsI`#wj_5k0KU^|(v2RhwCJ2oT8?Evgfk;6#{IdH;7%SHDPVR57;{j6#KD7qkT
z4uJG3f}30M&EkGz;F4Y*-r%dw_IwjA-4_01zmr|u<SmUAy}<8oGz?D#q`nHexlI}5
zKu|>33J{@%$AO_Z?adQ!DnZv;4b(~k)vO+Y*8nzfHREf#@@6kkmCnUv;<3PEns=W@
zhpbi>z638Kaj8$&<k#RXfM=sCa66Is{rw+w!v!|zz3837D}Qu#S?KC~Wn}3@rB%(u
z>zLimwIdAUc`*tjnw~fJM~~fBCPuq`4R7sUW3NBHtmbrYDk8Hx<OMO=%ElnsYhH1D
zl3fJj7$(qAyI_1;3*0aAcULeWYZ-r(XQfyr9s9xGG;CF+`%(8ptSs^-$KwK>0YLr>
z%mnCmoUhWovA{USQDfG5Sy9DuB8itvlp|^zwk+n3gpV~MPH>$Tb?97^M@$U1TnyG}
zyi7vI%w?x1O5kT5jpi!M<na^dV&pp<6A~AG&kv&qi9McVYqWCoIpH)Hp?O`yLII1&
zh*ZshtMwAEtz8icnB+L`{@|MNbU=nX+Bwf8^?dH7l)E`i#gb(-6N#M@NM(+4<pCoB
z4kEeg657b(Nd|!x$rxAq9}>~^BF_%ya_LFV-m!oQ8Nxc|FaOV8Jns3$LF>kOC}zdK
z6cXq`anN#Om?hz-XW7HFBG<lR9$q~qkZEU~q7A>m={a#p9=)IWX1hQ%8v!rEES^7>
zH4hW;(SJB$g{8<Dx9qO7kzSj<240c8Ec^LhY;PfYKle!ILlkxHbacJ*%zKdu(cvaE
z{iNr0{8&$GKN3<E;?m)0$o=!F_paBe`zmo|<j1zM{4n?sZ`AH02%P@vh0-oZ%)mCJ
zZIv*8#TjumrFoM&nu_~Wj|zZOteTfc7XjT>6gxzY74IdUg3gC78Ex2QX}EEDM<ap*
zTo2qGxm1D#$JUO;Z54ldj`-pS$Rr-$4jGX|_qX#<C<lZc`_UqW6QsU?X3jKH&0owk
zh?f(puOqRwa3zOGmUO*B&$NY`;Qm&00K3=7ZAk3|>hbC5+DZife*eUMl~L)$fd`{t
zP%ug$Ln27Aovrrg2aBnotqKE!#l4IRCH)t<r3$nsuJNWJZ8$8`L>PS`jtz4kc<S8r
z9`kAkhvK95YK%QYet#%Fh!4&ZkLXl!hu`hT;vjxh8vPMLlXLOk)U5Y+c6XY<QJ6s(
z3K{^_ovz`#?T!Qs*anOib0|$o+*-b-#z3KTC|fhgYdd+wsi*v{?{BObmY5CQO+??x
z2H7I)$TsaC*J0KSb}wFxuKXp1>&Tqc*TDZ+>|BPxdLobU+0QNCD5Lg0$Y^4<UY1Y7
z1wUIlSqAV1LM0&jlBE^gW_Qw(Lsa5!_pV$|<!+~Di3F0T^a!vJ3#D$skbk4@rWhs#
zjpeR%3${+_vJ)aO+SgTOh>#`R$ApNxkrhn!;1<+dAyf1*sFJX9K7S(TJR^F*KVcEC
zC52xq38aZoxSe*MUMV;tGuI5#V9P?@k{1mK&K3|#H>R?E_Ny@j7><Z*s99<@5%x<x
z$#C7<?j|KMu>d;FUCDtw6SEkMUUMW2UE3|Di%2iC&b6!}nz?;^T=x34(ISOeQoFc@
ziHI(oz-f7ZV41O84iBNI^DGa+iT)9jNVB3R{9CpfgyEZKO9E@_$JjV4mabfatV-W9
zF3;jQ?9|w>!iRZ&?0@0k<LiZVU#x~(1hk7%aITFGQVg)@Ze3e)d?bFq38Pgyz7+(?
zOg%_~aHV`YDAF5RkRcu8vd^uvbzuxnB}TY`9)v<ckrJ><9X*1d(|aV$x3`hVaF)(%
z*<>L7t5#1Dgo^!y$75a;Rz*?NS*w1xpxjqav=^VGJ`Zu`hR<;?yr!<7ALGxbM4L~I
z4SXmn^BXW)RH{9ll+Y;31|pY-QNGt_X$xXGhij+(c5xU;53b4HWP$FN$4p-91A+1q
z*c-nn&}o(Tl6J#vNX21zJy7HsL7Raql+Gi?PA-eD<BniOVp=!{;ww&v(wFxTrGIOy
z8j|qaJXH<<do*AOv^vh)>46=sS6py9B;I2%lpf#=cyxy-vSs16gY0v?>)ij{#&N-h
z-}kabNaC$}BNhL<25xe=CLo)=kSf|sY-~O=HDb6n4V=UW1FHe((>71I)d*;sR#rzS
z(>tvNoZfDY)2wUG9Cuq(c|;weny=bP<2CHJ4I;lIaOem;*V-8+a7tkYDamYQNcRD4
zodI0^eq1IeTvLDL3;R@ME%a;)9AA2Xn7Ih9EP5FBdBB*OXWpe2u;r*w?g#)kAa<xV
zt>JWKHnQRE{ghm8-s4+AZWSVpNWq_Y;1GMqsfFSxEO)4cnmBrr?E8%NM87dkiLxl9
zi4&`bPE#T`dl&%zGh3%d6n4Z+(d=u~dIktTzgEo!v;p%806RsY{AmEj0p-bI3yztj
z@+Su-XIA})ZV~Y!vw<E0CK>AI*twlrftuf>3vo?&h3@eE%^8ICA&l@?-{5G_>_LYX
zsZ^YK*DTsla&vGt#0x!*?3G|m)EksK2~KB~Pmy$!m|(LPyp|#1Do<C4(ta|C5e*p0
zZ_GKMuXCv$HG-gc=gL>6RUbkY3ACMpT|d1hm|bMd)6&}ToiSz_J)RCIE+g$4J}Zl#
z@H1!}4cMFkj|<H@W68(yY)_94jIxjmKZC$!lR&;<6#H(k9eacM>y+3l0d6P&)5u{*
zoLR=PZ^f?SmAe~fVCk-JToZAJbqnAUJuil&xac6?We}{tTbEVCvTtvy%0LMwHnt}1
z_EM!uZNI>aTtDZsm9MOuyw>jY^6=R4P+79i?XdB(of;9V(UGy{$h&F#yK+U+cDc>>
ziNgDL#>C(aocr06(ra$z*a7+U359<ur}9ap0Rs*mPdx8tR5;N*&Oy8pS0C_F=g-Z0
z(Ilo6ZOOtdRDM(sCVTPdXJHX#-?-R;Y5_4hM1JT|x@6_<M3L6w_}3e)JI|ZfWS8>C
zJ{Aa*DJ-LPR^c+5Ov_kYp0bo>Di5rcvzU((tyvjyMc?BsXQ%#jE@8M7I3Q!mL!+Uh
z^Kp&zw@lz_toyInk=#T)BRc>N0z9sDKPzV)1;~wQ9jMjM?{jpDi@{PZ^Y~n-cSRH<
zRcif@7t&T3R?az7sAp!WN<HrNA(%52hC(-mQ-u%mZ^2lt@~w8kX){dx%i}rw?}G?w
z*bhygU}S7!Y=YYB>`%YaDT~~z%Q5TShEANukoAOA*ip711y(o#z<pH&%VOnTzSn?0
zpcjT-As%nIBSX!<u>FMI0<YB}g3TE&VWdPpjwvurTGOkFY1HFLLYStX^Cd5MPu(I<
ze4I&^c)z<SO7uNja_W9)-*n9qMXG;awjf1LXGYmPb|7d{G?1DS%G-FnJ;h@8>P$!G
zPr~vU70H2e3-`bS;8)Sxx-Tf5RENtS2p%boPbai&+7tQ)7Zu7x(zR8JOsJe+Xb`+%
z<=sb76C(4z*P<JczY=&q21!Y7AX!Ua`B?>I#fHXwj1H<IE;7oGat6G%*tL91_ZEx|
z=rlkJ`r}u5e0yCviaC7To}w3muc7mDrMEkG%hm0cX%c3$0rBZcAG4$cf8J)OSOL<>
z%;TrbDzu$RngC^|+(ENKm0qMy8RdF%NF=l1Jk;$PrsJ9{pMxpSrX97zsnXh2a^$m!
zU6_m~%&)o7lFNF2KZdoLwn^Bx4qf`iKW+u89=EuhV$|+LXA3l6z#Qh6G`t{opi@Ym
z>;+6preC-2fB<4?sva12t^p@hFG}c|^&LHOv2w)<9+-=8(}ubzS>?-i>Ljx_IfY>M
zsGSnu@Xr>I5f(fDlF(27eB!LB0JwcQje(Y{6pUG7y*D$7ZI5_&Y5D1K-E}PSIf4cb
z4lFBnW-XCy1G_;sBRK%0o8L$I*ntA1w)=vz;V!liU<koS5{C#S27i~uIs@`=(KU9)
z4Vns^0d{&>MIljP(*-}dPpE$Lu}14XO~23!2VDK4e<ttT6ur8Zap`=8>w1wAyqCXt
zq?|nl&(XPdwREL!4ES<XQBL=Vcs|l-4{ZF-ZJ2eJvLhNVJ{vyCT55JZsAI*)4=4&w
zjaBvtK%i!yhKR3Pl}49Ctm|tZsAy0sOVjWf^FofXe?a2{6ZYNKM9hJ+Ue*hf5Q1%+
zJZ6_m<%9s8F)(XoKmu32-7hC6qcIfL5cOPF)c%>#vEkWaf~gZ2j3AZALPE%tC@NLf
zj|K0nQ4niw@KDg1xq^}7ov({p`mPEPJ4nV2z~wEqNWx`C1l=~^pfZ#x554>CVWTMx
zLX4a-2>}z0z8W-69bwv1k?{0xm~+S96d2xVIia)7Ec5Lgv@hXkZ}SWNZH^<VpaOFx
zvLexdaHZEa{|s?Y%curD3bj1l#({35%X#*#R=Ao9@=^F)rjxd*Cpoy{p;(JStw$9M
zP_B(E%~ns>@+H?DV`gNI)DmbtqR5LKRm^58kG8f*=VyCj=I31yHxe1m>uPKo891S%
zdWFXjDL?uyQ|h@VD0FO{;f!!_5~y!7Gm%Ph=o}s3c5Oh*m!Gah806bfhYmWh;_wEU
z1U_w>LwsOoD?etyD|91n^?W7&t*Vj(*19!)w;1W(!Ov6#=@}btqfgKGU_uA5{=gBn
zvgJ%h{SB`OSecM54mXezT2M1PREpn6ZdP09OZ+@$3~*Qk^9q4s{4v)x_%&xCgWUA7
zhGZZiro6<xk{ydNf|qdMCOVHvhm^JCTsg|RZK^kPPK-eoIGb5}L#+F!yC&2Tp!=H%
z9$$+;_kX1Z^v){a6`+R5d@1y~8E#ALj*6?<p8BnFRO0nEX`yrqEcx*8-Ghvl@!_9+
zb}7(Hcem^lT49!4x=mP3fr^8_N8v4r*d*m|LCA}2Y2Ku%dryXz@MH}d=xb~*SsEE?
zajXsL`?eM?FQuzNOTPf+vjPrOUr~%G;n^Va2_UFCWQ!-%gpaVhv)=SquXpa&_japj
z#y;tW3kb3uvl$-q@{axT;RS1-sk9L&vd71yC8SS@sQJgol$aIvK^jB3K#*k;wU!T(
zG3CtIMh7)<XQdh;_%+#N?QoIiYl=xoP%H6rSY{uwBZ4@V9rplstRCNRpok2cF?CSq
zoP^)gr(3*!Nmp-Yc2Wc+1$?TU_(F~J$(i!~L@0|2dJ0+{MRry40x_b-uZyKntOd3Q
zJN33G6Wtbt-^?O;H$sdOXG2<(5XhEDFz#HST&~{;azuRR4qZ{vS64G{u3}2fJf@H_
z<!ANBtX3)RoB~jjr}>aBsTp5M#I8r)p#!%pHR^`6B6hYYHrejIIc4H6*=LWxg<Fhc
zt|bN%T4RXl$?1x~HvvzQ9gw>S__O6?lW9-=?yk^+(z!s5u2W5D{GO+1+}YY6J|)KO
zmGJjnrIo_1SETXN=3ac=sFXxHuyjs?zBTw<UiYkPQv<L)WH#4~x9qFyR4nYIC$QE0
zzV+TI@sG|5zOLYB<+ySA>`*M^u*-L-9b?0`+>Cv;L6pQ@e$l;kQ)I3dUs&EG6EO@k
zh`>YZ1YQJS@8R3L+HV1K7A?@fLM>~MewVg8-XerJs5oV@-6xJQn#J9&PcAPY_TMIa
zeK&&N2TXVmFG*0#9sW7g5N*{fi-ZkH;z-qR00H9@gw_q5+~R9HV<YIkaw>!4(A;3m
z@eX?AR<Z?!eVL3w<51@@>tQOr*2rXGuDWyCRRNw@`edgq)!!0PE6$pl_05$-0{6H1
zjI;4$C;#0Vc%58P?H)k$0z&!XiezqrgXmt~1kiNK0dAkR`s=k#cW9Aj9fCrXx?vn-
zbX(X!mK*6v*E(jHW@$XG3*C+PLWPPP+@1vc{a>RU!MT;s&yGucJX}C&`+k{+Swy(1
z*T1_X_cFK$^fwOgrcGgYuTeJ*!Jn%>ymtJ|YKiJ~ahpM0$(zbbn7;>mTD&r@)V&sU
z0Wcd=E0nq44uGQ;c=sJ7^X{Rxrr0sR@q;7Q!dk9Lt|Ht1mntz{8EIxwI-5%2g;imV
z-i4)zb0k5-h9Og@yr4!Ixgvic$xhf@=hPvO^_bZjr%eC++4*t+(HZCl&%CyY?}fSP
z5FRml`vM2~O8|N&00M^n9Uc}OM)m=4BFi*~448dzmQ7rw1|#6sQiN7gNS{^1s%=*p
z;eLG-@A`je6Sr{xt4-9{k=L<2dPPXDOFF}syoHCR^Xj$lb1{jP;I{n+1*W-2SiM;-
z(DIP0VyDR=1UUjYn<iY8%_c`1Mbeh=u(|$*ZUJmrhIAeyY^muHtto+E;QmVR?s~^R
z-3~G5AOkUeu%5^8?U$K+NUW=;6p}|n8W@*d3$$6`L#-9WbtUL)y*e;?z6uRD`>F)>
z7VLPh-stQh!Rtj~O+}(wnh9}_-H~K&jf|qsw>}qNZjA4->9I+p`5Ntl9Se4ep<eDU
z7yvk*_=D!%lS<0`%FGhgw~lO1WIaJ=xDs+04r~Q|Qk)5=sj~DpW|MfcK|t~sNoiRR
z{e@IF3&ULS$8EReKNJd`Pf$oagdE)W@q-<o`QNIQlFh&_tS3ya*py^U50F8{S2by>
zyBpN}sgQ#RGQRt0)PAXK(qMr4dHY5?B6w%gjwsIIKA6YvUW6kmn|3E;XV#<Cw=zwp
z{iH%mCS*c=-Sjp_F-lohpF|myG;(@5BuNxBQr2-46fxnRWhrPhLJ7%f3`i1I+<#Px
z-pPKNust3|h?nDwj7{9tj+&ppi*llfFhs^}M<*`D97JL1CY{#QAOK4Gn@%8s05(~_
zvO+8Pw&sSvdr>;up3Y&g@F<R4olk@B)aanzBx&2EA_PMUY?LL@jsHUmR4_-0{PH|_
zSqS*lxN2&)&^b0lP-<vU1ZA@g|7THS0PEIUpJ>?4Z&a_>u{L28ok_%sBZNb){tm!{
zXb*Rn#we(%vm0)Z_(RLgAlc&PLZxH(lM}o9#8f^}TPU++FBwAbjfLHL7sLNX!3;^d
zeU87`?!eEZrpUM%H(SEPnpM9~MxqKmn*=B}>t=D1ysXY-tH$GZNXV322;aDa0C*gl
zJ<K`o0PdGw$cE~(Ag{gBK~l6StqtJ!gyDQrU~LddUDKH+*X<I$!6o^%o~;0DI)g{M
z6aSBc$+Vul;zwxxmphJWtGvlOk8LH_#+3c@ouZ>-%H<yTTN}IrZ8HY1Z3+gblH6cw
zD-5UTpp%+v4jM>xA-$P%|Mhobn;`W+&pryu+yw>{5BNWlPQRzDaIZL%-BJOzwQ8by
zn|=?%+SKLMm*YN&qDnFvQ>xICwfcQ81SdEFnvo0%74yR>C1pRFYz$D4AD+C>e$EEk
zl`5eOQxhTxm^;elN=j^UFsz<OwsCu6UL6Q&rbZ1hr^UL677<=OfYtN_W*#?RkBOqF
za<7WQ315tgrRR}&*W09g+64pbXt+*cuf7?Y7ep}@%aWDp+NPkKwPJDDt5XYun{466
zb0?%RFc^0?MFyb)`4YS8RX!FxOg^)BV%O-K68w!P8|THh6kKrjz4M1`A)vUV&ebQb
z2HHP{nqY?zv81{Ksuj#8<WG1=GLO)`LL8aH1zj{~Ve}rS@>_TAiGKln-`KFqma87N
zUs?DKx$#%s+U5JQU?ZWyBJ$c&1c#1oZ5bHjEqB0npcfIwKh)+B{nx{jj}}mlPsrZ2
z3Nr&r7^br!NQZ&)#dxr0!2{%n9*q6XM10KWo~cakIy@Ky7$9M#(ga&MvA>Y@!D-Fx
zFFG-%amMDxkiNg=rY3-XdBkAW#Hb1>O)W}kx1WR)$P}RaVPsF~!eBPAeofa{d`IrZ
z>ZS}@30z0JS?;2X8mwe-Y2sOWaV3gn(9P+d+~8&GB}~SajhVR|NVsR2Z}8co80N7{
zt7xlJsMX7rD>-e`_fZluMi`@_4d2Pa$QSu^Cp)l;-L>Fh(FDNlcQeHwE|dB>CU`_U
zQ|-Z&tXAJCWb5_z|MoUQ55&gbT&1)5%l@q9EW#BTQ6OUeVwxS<JuZjQuNPi27|W^?
zIs55Pz4z?FjT>`=W5J$Ih>`;`uxLlR@p4jzAgw4P`@@~YO!7muth?Xi4o03V;hq14
z?|nTgakU)4meGizG-2Ncv$q=tM=8vPIM=ASKyOtzz~vd_kgiWkcIx+&EreXXhn%@#
zpNsN){P`^TTROiFi@0oQ`VJe1jp1^*Z}athf|8YjeP1s&$#)sX;*z-Y_E2AOo2Na)
zFE$<UY0oK!e10-O2B`z9uXOr~wc3hBm0Hm6vEl^CFxegOrgmF$({YO>jjVRB8u0>{
zW(Mmzcm>XUKz8k>tg%Eya96T5IHXNY(y}DI)Bgo}8NSN_?oP@le<M`%sh*6DW95fF
z><y7Y5oy+go+WF3$BG*b_<GpcQ{{oK;3AzZB_UYLo-rK-*pppkH{3J$X|8m5l`TmZ
zjFtg_|At>~SA-9BD3&cA9A^2I$XLGRwz`=#^3#6;eI2D;C^sZ_QP}!Oz1%8WLwyag
zeaccc&<=}92v~@HT_XP_E<IwJch<6D2tV^B7x~cU3E%wb?c=b&Rqvr6N^9ltEz6x^
z%X!$Y=thM&hb(_Dsz8`urJfu@;F%O!C3*)OyDAbQX?#e8h!iDG!yK-0#Upv~I-rcD
zh*hKvjZ~I%daX5sy|;uva494(3S-CgIdP@Ei))^se^0uE0AeRb(u`uj$F3m_KgE1r
zO8IE0rl>H~IS8=sRJ_(5iGU%~c7u=r=#EQWDT?s|jLzZFTu7yuyQtU8uh+Gw6!HPC
zrwA5E@q}lE`p34{A`{pJH80K=B9tU;M$-m6t?gpTPMNIa%S;%hL`XUuB0kp-m8XB$
zAYqDP$c*V$ILWtqUkx)5IJaYQ9fp5>)(|`J`wg)toGzPptgb6<hy91?s{d`eMzZ>f
zt%DkvgZt^xV^}|SD53+lo*{r6&KXelcy(SOa`U>l!M?^OqRym}6*kAXY7bz?!QM$h
zI|adf8b?!u1rz9OQP%3LT@;u#2AwJ|5HMaI&G|5Vl+8ascp38}ZG9#>T)ZMR)^Ht{
zK*y@-{S$&z=807~gl^uK(Il*5hwXw7e@`p97E!*5JROsh#AE}?1~B)RLG@6Fv`1Qx
z7at&0Rz`@f2lEh{E=r1f1+R%eexD~m889*576&-VWywXOPF$hLiPdV~S8lwb5z+pY
z6QO0xCs%N7jADz>M3)Yc&Z6|X1ByY-+N5ulUb!?c7xrhuKAkGL1l_JXrb>lA@}>eg
ze4daMs)BqW7{UVl50tA~0rxXu3C1V=5YdR)Pax6z$@oLb)R+haS8gZA7&740((4z|
zPS+-<{URIHBPnD6+)90+hrQctsTgawIUC~}d5aw@Jw2^gV9D%wQlZ#~ZB>?v<w4x>
zeMnSexwjHx@Hkj`@JLT}E0jrLiP>)Z!7C@~>IdLG?{leG!)^4II^F2mGo@hkX>t(!
zs^?a1AYB2*!0*j&eIke$C4CGc8ahooP@)RLf;E5f|0jy`qsx4wxWfM%iUa<G#s3q<
z+X4TdP+TS#Bq#nkP5!n&AaHVt`3Qe_2CQqCKaoyh^avB=e7i@}Tac+y5d78D$PF~K
zcIN?!94>o}TWok^H9N7#GjJCWGk>*}6C0E(H}6~j(qW|i1UFymuWM*}fHk)=fB_rE
zvWO#WZ^NB|@rk5%zt!e=HT#-`)tUAp@MbGTm@2?Ce|pMUfpbzZa~u0Y?9Y%Ji{fDX
zZTr{9v~nU8LOmDa{=?TXZ~^Rghi+fze}I|5{@cjc$NPt_0UZ@@2;cGUT0PIa7$}Ai
zUUe-mG0!36;`GsS5GQhxHluVpq+j1$4eQ-hSEJ(ivxVeRbW~RFi<@wr>qSbnk8D))
z$#Norc`||O3}wrUpyiPKq7CdHFJH>H?BG~FZ7{UFhmvI85T?@z-3o*Uc))|<Vx~OL
z{tu7dxot3j4;qIO-6_#3h)r%pK~|9av4Y-Vw?00$Z|3w_l81BtYRUx5BP_g8Hd2ec
zO5q9fKCD$mGO+AB`9S2fUFBj(w79xokUG93y+*{0lPE%5t$wiIAHwB8^UK586s3^J
z#&R86-&K42fB}?D{im^l+7(mrrpcdwAv;bN0QWO{_Pcy6o~?o@Ccn4qgLe)U6uqdj
zTV6jSob9kP4i{`ZL;MQo2cIe7XQ91_Q(B3>l5&C@WJ5g14i}wiL*}-}CJL!RvkFe4
z2c^ZaCb5i_jI2jd=Y+uWJt)0T*G-V$nuxo-jIQ8Su(g2XE7H}2l~hl0s^!uG=LX>e
zlz%ODzB8ua!OdDlx-k`+PxEHSvz|qC-UXWKl-f^%cJ(18rckMRpnu9&YT7mjo?Zil
z&#t$je+giG9*TYar(16o7UGf&s_<(=+a78KFX9_smxJn>OKf;>b%)S2s8LrQ7xX%J
zx_W#-SGwLE{`Kqu|6|&V)GW2Bo9I>g{T-cdP((g&Cd3z5Qqv~<xoQ-_LipMEdb#J0
z(MlIjq8%a}u(&MRhcr_CXub_+q4N2^WIbom!qhm{;qZEBG%ZzsnrdQ_iT5&K@H=~{
zmyGhl#=EVMOHB;rID9UKnZtsNHi0`P9?&Bn51ubJy;0Klq?WES3ap{Vs61?9I{#<)
zvjr8`RdT9=P|S6a6M>z=TCx8Dh>u6nR&^$TNI~-8HFCwe+gCo_%x1tSjUru0R5JvY
z2k2sPI+&}0RP_{DrWXZ9ip}g$vP{77LuF^H!-*n;XYA*JB@x=3T6?QB!v;z#KA^WN
z!5-ty8d?G}s6(r@I2}w7w1(w5ajhP2Q|4Ho`8lyY%@<hz?)h#z49rC$gJ+yKQ<l)<
z(`29A+iNR}O_v*Zd=o{#%Y!+WQRkv-OJ?(G72;w*9V5ri(rg4@g2P4T?MFG6J^4W@
ztFnwVO+oxH+!^~(pNC1in_RmaA3%Gm70tC`NLhUP^D~=LO$}QmF9XImCm?pwyQBSV
zd&@VStgXN}`;CfLYhYM>*M7CU-6-8cglBQMX3TTWc)sb$n)dXT(nq>;m3oB-?U=JS
zY?;$8BZ`1MAF|mDzw{8Q6sVNMgL#?mo9Hb?u#D`^)9w28glKB1Q*&~J0Xn&dW@Qyx
z59We)$pmKwGz`F2jf$wX6R<+oMj&oE%zxM%2#R&T_X^yOAh?5_<C-?KPmW*={hvH9
zA@mQ=iwns<oX8S&^`3kP`G6WG7O7Lc%#H;rhyNQj*bCB9E#liqx|mc~t(?O&5ntcO
zQ3vo8t5ej#A2|u8&khUWs^mRmN#RIM_o)&PA=})#9z~uEqGmYnI<uLMCIfFAy%c6@
z+NpC3fscP3juK}n6;NRfpd#_vvHlm#|I6}xZZ;vN5^%bX(c*s|uYB@-FgmPqVZ-Mt
zr3`M)eEm>(`ABVP-$sig!;QN70jq&F62#LwK>v|8sJ{TD2qH*&^tkyvdmO69gT|PN
zN<X?3<Q#A$PFC+1LYJ*~Bs|g{Uaw!TAB_t*BiQV>D-CILx@unG#_zi1f?8Rrx+IOt
z$knSS9&hds@tuC>4bZ>31vp7+>I#2EF5z;C*@Xk)R)p*J1Lhz#Pr))2!BkAOZm_go
z+CL-XLwlZ$Uhk@Z0td&2$;bK~-Cpxq=~t)~X^Xbs@rLGZ21C{7V0s7lDSNEQP~X_0
z5y<>Q4EzDd_qNiVbiT3QSpP58XUmh&cY_v?C4(Wf`WW&&IKB%GGQsPYwFbV@C}>T@
zByI8mwzlng@>t}fUy3NTqd~(lFv#g;k)V@ewr%O<bTivn{UkG0E!X%i5J@gz1a`rk
zCPZ&2K8?e(m5@Y!@iOU?QrhMewaxwPx-TApgbQQoe~5X-x;O^f$>vMS0?}Dn@`Isi
z_s{KU?-bTwX7DC4wls>JsOgCPC!aw5olgM#Z$KgD-+;nZe4i08y~Rqdr@>I5pu>ey
z(g>=^NIQQjw{i0>E~)G9MvB2=QUkRI34zOa&-#J&D82@yCL=+RF3i3t&rXF6+d!?o
z21Dl@3aNC?9|`pvV$5eOk><ZFOwFeQk{~6Z7gh=$7V7lq*kPRTns5O;rzC!%Q_A?-
zUKmfjs%oET3F92cWj;&Sd{QTZA~RmR=Ao8uplhDLGEn4);4f+FBnw9a41oHB8X+&Y
z%1iaDT37F(P+-^$R$$cXQA+)9V7KX%uUv%2a%oebZwMT%ze{31yJpge;$Owvv5<<i
zh1d<IEcl;X1N${(*1rM7Pt?naGHc{o%3aC|-hXuNeOFe?mT4owLR%n3Tol=Zw5U-d
zQ`&rQINgPVT-VTJda<c$ooVW$RzYdF_*H2QsZ_*5$fbUA=Yz)u>604KlxKz=O0s9K
z83yM^?v(Bh?P@L+^!R=dU&U*4&+~syPHlP<7WoJ2P2?4TAwvPC1=OkF%I-PUl`qs5
z`zZCW&)_y1O8SIBOV>KRh?g*6pKD*A9#$dfpji91KJi|@l&|qCgSuTw9`AN57n1~A
z5jx`F{D-2g?e=LVIx2(*3Li@hyJnN^l&_rz@jMhUjbHShVk0UW92hr_Jf?AA?}mqI
zL6KQ@Ux5hnI}HHzp8SGU3;Zd*K=UfDjRSouz=^4tTZEMi4H~o-R*fqz&ll()X;0P^
zZrKyBK=l`7_rgQJ&<@;E*IE_pmys&eo}khED5F_2I2J!TzC$*={VYGmz|Mf&wpN8M
z{q_RpYa|Du<R~6=S~U=UeMd#ua{rBrU_^>c$V}D&z~2+Hv>md;fuY2sQHFGbeY<DS
zE)W2w3swS&&c|!L6ci@r?w3qSK)Vqc{x{)$-6?6CECOD?I<E_B{t-Ne?|b~(s-eXS
zuw>*258@G@I6=^J{i7Q)+z<AlD)G+u!HFLWe@!&;K>c?{1T+P_g+E>ElvtKZ1&SJg
zp>rhaoi!4Zm4dy|r^x|PboX>L5tYtRy-x@Rln1G|Qc+69cW;9wivPghHOl`%B$Mpb
zw&GuM^m6#a2jypWQ>CYO1nK_p6iTWceCVw5;Sc=)&J7mPd4)_e=XMcXS5HL6))ctn
z?D<l<XP5{IuetvUiNtA3B4{Z{dH@XYidqf{t^YKoJP5_&#b$XG*wQe}iudMHqAQ}o
z{V0o({6*pmL6=ErW;GUBGP7a6|8GKMTd@hNF^RSX({ODzSL?xVJpK0-p6o6Ru*iJE
znjJ?F0Ez)5)!F%JiuF*==j;%&e^cr^)JXv;ocSca6t70BO0Asa0-Oc-uT%*9o8`4h
z(s$K4>UqdF4b=J1?L~;o=!9LXc{)?2$qzAP?M3A$5Ap>jnv3V*tc!F}AZPkLLM*Tj
z)*Lxh*#`urQ_`J5%B4`NXk(p*kzyr<xpVN9L%BZaw=8kyUe>!t{qMh{a&Mm+)fYR+
zrg_09g5*?_f${J30Cv`|+jXZ(qe3?*=+)(Z*m~a5EBi|8xQ@>J(S-K~#pFM8*~>Y=
z<&v=f0hmew;2P+6T#Ju+DphL)c9!d&Ma&(O$(GQxRI~r#{_48^m-{PBq(t_-le)S-
zAEm1d|1UDCBg=dJ+(wlG>^xiwn5XtIjNu!ax%dk|e%O|E=s~&;W*aCEN5Wb~OO*-{
z1MNtXLc=FSYQ5k_#|JbaK_TxEr3a{fjXQovPsRgeh|{8Vfew%aV+#lJPZ8F#vZ7gW
zGOCE|<&sd(hnC&Ev}g38Yi;yw2N+xgA^#mu}%BaYP#U}Y4y;`J%gbk8f}@?Od`
z*qW||e7(u~zinRP{bt84`=wE&$ipAlRP60~wlKmn)VfJjIB*Z<3*kpNeL;eM&HO)N
z1ys+??g9h@Bw|J+TgWVEm)iVq#J~|zO?zjT`aBOM(+Fe*T{ZbYaB&uL`Yk~JTmija
z9RY^_QK4$(KGK2UaAsuw1xnY3Z=j6)PoO;4Oxw3O_%gY+`HdnG>-UucpR^fdz$or}
z&~@Me8m_hD$%_K8y_+_Vi=)M^iWU?_&b^g;0s!>rpb#a6TM`soy%m^4Q-Rc0YR@Yx
zD`jm;xDTt@`0mXCw&vMOHqB31gFB-NSs*|Jsh410&cbm#6SqBI+Xh=N6-3#SWcj`W
z8g#BxTH2yupB!Fp-I{#pUXND}SptA1wtTGZ`0^W1wR^9}r7j#Llir?@&tPbTU1kdp
z=M|Auv!8(eKdVnCyf1VQcDq#gaDGhla3DW_=#!k|&fqfvyr=_>q102MDZ3i*u2Xob
zziK>=w5v6)^We3c`>Yz5mE@1~8(r1|teu+8HLxXvcJ>xP-(_$Ni?IhnuI@FhuVjy9
zS7;a5Ddg2yEG&+LPu(sQC^BTrkBw{wPV*5Czl!5LkClJSSnI6AW#qBEWnx;yK7le&
z1Zz5c{L|BvlksTU%aaw3`|$GF$VFbIlol&RaWFEN*=KO)e+qH574^({^)Og{)c81p
zk$UOde+=6P!@PDFak^4D6Ab^X{)LX2n{}i6AQA0lxI2kgFoU3Q(M9@Iy26e=c<kfB
z)`1*5-jlM60<XP|8`D3XMe3pL*K$HPB~qOrp76ahAv)>g>IBYGxi_sRn9DyJ#Uqv#
zr1Mdulsplm#0XTT7?LO(^xc-%v7t_SI;icPgvZww^a`qE$p)^W##!c&{zg3{%90t~
z;gZx+k>Sme$<cPz0baY@eTa*31l_!jVi%q7u59~y1+%a;aPw1EHSp#l_h5Fix|RcV
z1<?v%Mir6cIzN6mF|*e2qFJmAJ-%O-d>LZKe{K@Zx)6v62YtmUl()jk#D}@W!z>d+
zzx$xbzAR+U8@&!4TrGS#*5tHWV|aRJAD4M4Bw*_WQ9FBd2{*@pgv<+p{NZpn^?3AH
zx&dnZdjWrZPb>^XgEG%YxvTmi7;^?LF9;d%hO|IJV&cc#hn~wb3{&MdO0|j#v6IJ$
z1c5$}cN*0gOut3AzVy23x`=TXn_EFj;`$JVPjTP*OWq`(k@UXH6bdmx&6SJ5BtH#O
zLc|UC?m*#BEJZ)8UG2k_mZ<?}1Whdd!Q@(iV|j+lM!l+EY!F+dh*=9o=w|_32Hto5
zVam=@;;~76Nu}(OEVWg~YW2Y^B-(cYr56J}RbnB@EL#zYpQchY^BLT|<<6VM$`O{M
zE>P{Lhu!c8s|wyUi{;|mTK!cYL8LGsv?E7zH~TkKYz)2_tw_VES(pwKry>^OlCIy8
z*rP9!nE;QvGs3Bu`z3=H)AtYy+eiffLu0SLKESTE13M*+w$=A)aVPLcy@K7{=Rs)(
z*-x}f8;nkNeQ}Ekqd0g@5j{;HBeI9@EZEZUHR>Le9Wo;o%5+{#ZtT|?=kbMFW1(a-
z{k^(2oBmCs*sp^1X}H_c@GQ2!74l9y{eL*#gieIr<9}GCrO2tSV$Sdy`s<(pcKHHD
zzUaA|oS`n5EdjfG!8`Yv{kX0{m--ZKr1jWH&O63U0%%w2nVXetW}&p@xGC~nh}j-N
z6WPpEmGD^^NJTSvykr8OvUcj%u1eX%_AA<n#P!n>*ylxM)t*ak>Mh;Ptl<W~)^TDt
zYv2Z8Uoi~ibUYp`W@zfmu?jN)GMt#duu7gTJLNAtCtS;HTw<*sLMayJztZB1vC)hN
z=823KI0t_agD9`yY0=xRvf*a!oIvturyw$`dHkfFbI88>7`0lgy@{J5Tn!c@qzzQ8
zE6u!02_=n`dfwj<T|LBhSwr5*3vWEXg8Oq}pQ!g?P;~10&cc3J7>ykR$nCW12iJ;i
z--WhA{84X7mkw=_?R>z^%S2Ns&QpX~M?+yOG_vUoj1s4!UuXVTboDKFBnIwcImk`c
zq45k%u8Pjf3H=z_Q|s>UY$y%lZ(+TFm}Kug=<r=b_6bh_h&>X{BuHEtEwTIGbyfG)
z@^F5JK2Ip-lg-GL4l^EC0L1DRNtj%fnwc?-LY)}cT7}|V;Bfw=7uX#4%2nE`Cd(L)
zb6!^9=?hi383*ABg;Mr=+62W7W#t|fy9L5cz@56&t0EH>`$WqSDbMbQzkPD&d)B#a
zJ1z5J!*N3-EZ!kBXvia1#>2fVmws()XW&7G^G6)DZC`fho8c=bU;%Fy*cCA)?<{q(
zKR8mc!y1!8v^hEry9_F+j4<rypOppi&d{I2hdVonw&q$hb$<Sp3)Wz4W5o+!ncM<G
zpG(kkQ%q^_^A@6RJbz!)BltDot~o|Jx)=$u_yfeTTC8*E)j&5!Ve|_|??XZFoNarM
zIKMZE@%z$9<)~Yi0^C)g5S3*nda-F~ci=^iQkum>kqx_Qn>vGLe^&O<pq_8X_j=nG
z(P}>D#y=L+S=~7F{s?VsE&=;qem;N_oUSWa<*^qeH4O}sZQ^f!--&tCNUtB|Z`3n)
z8kUO&ukS=h=s^kmqrY?-_Xo$g!8L<JGSM3~!e3&5GBRl)J|J<AcQ{k$bKP>(cbI>o
z{<xh>LGe%!lAa!P<@B$&qsQyf8+wd=x3756JI~SHpS!CoX|Fb~8wTid@*?bW<x8L(
zf%C@z)X3XHdqD6s0#E#le{#=6UXMN*(}pk-lcmh3wxy&*5u(E+f=u|0bZsC;n$qjU
zWKruMyfV9%eE<v;V_E@PBI*=pr7*WN^Au4+=s0x{CfcDx^KGy=26hf#jJ|!7%~xsc
zUs=`nh}lcwR*P|>)<2=M1i}EvGJ&?Jpb!*O`kbM8A^UfT>C}ds(~xBW{>oqp2Hv;t
z5j!wZ(I5dg3;1F@f7n9%Vu!d+;$KG{?Lfv@ehydsQU_4XI=8{1X-BN4PM<eJXKEia
zH^>5I!OB#Xt?m_uV3<|wV(X5zwjv-e3OE`p%TS;ls&A6PR$q=H=Oy$G>sPXsu{5ax
zl!RJLkPzllFB2-bO29<7w>D|Ar6y5`nO7(<&Z6$1-tL#Jh9#D$2Z+gC1VIy3g;?C`
z{Tt&6H~{h@#+@G^;D$;XvoKyzyPN?Ts(6(26>)pL(f7)M1eE;^22(Nj!4VfhB&BMj
zHnsC!cMRSV7h5F1x}W=M+8np^pMlhn(shMcZE_%T(E{|6wa^ZiqcM0VaMY3+95^``
zh=*`cpiqwaOiw=;K_HN*&=dAQ8m3y8<h@|f!T?DyL#mty`KxTx0!qZ<RL03v{J+DM
zA~B1@m5#A9h{e4ye~X5atm!>7fF(>4YtjY|aUEzOt&#-}O@c*f8OsE>?O1p&hQhTE
zC~_1u?-gW}u>DNOeA|bE7#3p%z(5dxPqeV?CeR2s7$?roBj|ozW_=Kk-Omg^K1nA{
zfCJ>3Sc|fIUY+eZmJTfqxfMiVEe#pl<ftVME_?rtK-M{LkK2f-cB^mxsY@UE&o6z;
za*}x4>zekL65`n8^Ku92)3TPSHqy(`+A+6-?7jO67mH`%FR5r}Dbrgh6%qpAI(k_+
z^bB%YxV|RebZjW){>*0eR{gNB4S6USfIbqH5iNs;Q*Ty5^TZIntK$Gxi-`bn{n1G8
zII_;qcJG&Tx*q5^c9$>yKg*cQE5dokF8Eo;!z06UlHDD2{nJYGe8m32onjAm?hTz=
zO^lE2op3{I^ZE&6SHPvPfpuUOReN*+_FZrLSV!%0us<9{(jA8z8w&W1gX&;W0j9Du
zSh>V3zd;4WE|P@)d4Rw1jNsQ&cwwSY-6f`4NC!yl>s)OdNqjMXD`(#j9h`2*poTJM
z6D6bUP_2jkUsRn_a3^8c?PJ?cCbn(coY=|4nb`l>wmq?J+qP|+C+~Z{I`y6Ei@xiN
zuI}3V+3UC0iY88uU$djKNtO@Fkzt!kJr3an%lOS|)lq|{Fab-UMx#HDe(;@B&z4GB
zHnM@^#HBW;THIik`Th-Zuo)7p0Kg8R3REyWp3vEGe{JV?$%e|4+tW7megg~kNZ2ye
zGVKh0;F=pF3vKJYg&ldJ&dU7A&`bP?I0C8Cys>cGC1?)TaKaq#?!4O4E1!9CQ9Kp2
zjKx$==_p$sE%ylNu#`MdE6^p$fw|DJwS9bzZr6K$9(RJe4G5FfTiQ9r2W<XoZXwhy
zX$r79MA@*99&tQ9aFsq0xY2y>Jm`Av#tc6DyFx(qn8$0Pe+e|ZAXP9Q>uWR=cTF4)
znEIN%@9^thut5JXP3*?Bo$(WaFH{ih^dg;0rcOx!Fwm}e$T;gWAYfPem@k-)=!Ml(
z%k?*mkh+~Y<l(IRwbxvm2Z(1g?ZM~I+TD0X(((T9PJs<W!8cW_ad5M2RrE12ozwK~
zntsc|#dW}8b?xl7GJA3|&^Z@hd_Q*nQc5}NpCwG-KxK=Qlap)2QjzL4M|vd9XsAxn
zNnnWNn5F<#hOfa;>zRkp?O(E~-?*GxeW)P})}H`BKuYCietm`s6GaS3siCUOcQiaM
ztnpJ)J%(|!gVl@}8Ap-9Av#rFoP^>EjwTYK00HaW$(xr2{pDR>gR=z3G16V8?}VQ(
zt#k3FeZ})I><|o?y`OK2DAaGUJq?BqLEh&<J2&Q+K&k0=DZ<)0Z=P-<%@)6@2d}U@
zSSSY>z=^`!hXO9{KboYEYvP-f>poQC=k))6hxy*BRi)wE);ptrUpoa&t5=Tn4J{4-
z4mIV*wO-!ClBi&%^^KrGL@+W(46)EhO+;)f$oo)y=jF^`3201?9>8Dn>dv9Js;J<U
z_qsm5ByPCTXn=2ayc$B{5Y8x`ZUcTo{XPLykg+?pLxfhbl#)>b+cP#zMwaow-=gDn
znenP6$)<wJ`nGHk6NKhu<C8otM+8xxjcdr_kUTUgCsCsgRkMi^t=zEO2ctG{rQ2ul
z>l43-7%MH-u(m#Ggv1I>ZsL@|3q@78O@jwuuc2%b1<e!wVwUN$IpH!>;$N1Ql?DTp
zh6FIqymat0YKx%`Jw&AGP+}qeJy7zPZv-T~Qc!e%QGY(+w!@bDOCKgE`$GlK*`To4
z%Op{LMsL^zmB#N}swTOKJY+bU$hM?yZ{Z=g*DRI2-oIX9M=vmK4Vd~Su#OfmPo%4-
zi2it0UU0@jT$SNKh-$B~BnTeef`S4Ru(G9j>KH}BS+tjDzg8f@Y9x-pxD>GB25SQ^
z1rO*DaO2c+vB#Nc6KiSckYB^dnr1hgX8M?O9+?OVhB4pVfB&wIp@+?(Cj-xnRtOn=
zAndC~U5EB`qW4^f#`s!RZoYxwZ=k5~CZXL`rvxiX_^~xOfrRKfONg|r>;4Ap3OqIw
zZuQ)k-mB_yw;AT-C_u$Ny<BD))(gnev}HmB@%Wj#5a3TWD`WTJ%259?i?+CB)mOK5
zZ<MV)1)G|^Tp*Y{zrTj5sjTQ$1Yb<zcJS1_v#KX@-jSYqYC~j`Tcs=ly`Ar^50ct&
z>&D5~pJS8`x2~mK9tiSu7@Gu4rr<ic<LfulJEI^nkb~laq7N{^PR~Z&Ex(WBnvP(i
zYH)ke+(**2#fz(oKQ9Ngk25Tv++lckb#!)ie1VPMU73S#5*@y!jAt+`=OmrC!zCU1
z1zN_}cs5|B0LVr&F<nb~s*UN#x`IE3KM3z*w7+|?;s$8`Nf+85*iZn_!DjEN!4fQD
z*=$E?>K1IMX&PSsB66pWb(qHP-T})RdPkpBWD^aG#$c7kKm@0pxkXxkMA9t^vRJR5
zp{*aZ-Q(|by>65*YxKNB|70pV#k)@LKXcr5tlH3Q&Rne#2TzH^ee(4xGvG2pK}(&%
ziJvo*<Hka5$^WX2Q=kM8#X(9|JZ)hl^Q%;*Jm>%7UiQOhy-7KbJxs2KrGC(L&7RrV
zSnH<4C<94gVE(feMD4Ate$3$cBnjbA$#Y>Ubmdo)e%d6bc#`eMFZ3<+)zjDTj0b%X
z(zsr{9Ycay_w>${1Sug!84lBN4X;q&nYe}CdW3^?oax}D;Wh@~$hrDe;~7zcAawT_
z>$3i)`6VW_tUsfOJ^HNSdV=|SrkAJDX)o~R&{=GXbH5^q#&aZJYj0~`lpX7p*y8k7
z@4Df&%~~+uXNpF@j~lQ-96Si;#2jrXRG~9aD_oR~2!szuAp^BM6HIA1N}e|ZC6Iyy
z4L{|o5+HAlurLQO)VU2sVO(mE{<{ZGU3L9ql66{ER7XAlgG56_uKL`MEF(zk2f_dx
z?!#N<&+JQ7Ie9NFh-y?)?I1Vwa{CwBjW<N4{Q<1{Q4k_U1<T7`c3#pzJEBp=x6#q>
z)1UVRvI0M2r?|sLz~UU4f%tj&=VXuL=HhQ)AhZ1E@G%wu5QyxVR4P}dxz@QAkT#Hm
zZ1V6X@2wa~p1{j4wG;I4)qMBYBbg-$e_&nu*VaA3wLU>w*Q01dDXTnpjF4?qt}1lp
zuFFiWHnXCGf~=x;S?1bVh7<gXFA~(-p5gh=|Hw_t+l5B`MWm}{qypp__?YT9#U<Eh
zAY*HD<gGTK06$CJbA~dXRkoy}^VKjQTvVA!WP<SbR^_34$ILj_Gzv^NA9n<bmR;n*
z$5QV4;Z7eMH~K_<bP1&+@Ez7d1jf2ey-GFwoA;8^sXDdZv&_@_@qkjcCD##aDvsyK
zZ)`RkzGP+X&9AgU1Ed1w1C#j*Q!$|s=zjS1H~eIPrM`tLB&gdHkjl9OM+1V>jf?B}
z5`O+REv=eiDU1x8r%V#Qi;x1yPojB;?id2TZ4Oio$8F3B0=DqwvHXv!YkXen=#uJ;
zHBU|qg=xALN(W6|i=vE_Ko#Eigrm2_<JGW12sY29y(bkdP}Jq?ei1N?hr&LcjfZ<s
z8hj5hnp(VRY$4Iw2ae!yK=#ocZP})lWPumaU1^Xm_Y4e0q#Ur%;_&-wQ_(M-bNyq8
z8)osaR48E>>`*kg<#A=XU_16{i-k0Bni}<=w<pQfYp=Zx%ZcVv0xEFXwQyPKiknLN
z50f)^xs=WFZg))6nih133fRp{klh^!J_R5Xa9{6Phjvk~K+QfCTqJ&89oO7;e<M5e
z%j0b)6;TE^M{E@yD9pBe*3NXUgzU0)b?n<GHFXKpBuF5KUuN+Lx$p-m!VIkDN&8I9
zS<Ik319gb=ek0M@R@-sO(`q^gyP8BMa2nbF*EkqUMQ%Op;ihV%>q$gkT`pw!4-<fc
zLP0Ui7#1yjrr_y2#Z?^9d9_SK5KjH$X|aO(6tx#p2_C<Rsa@RBB4{m2Sv@2CMhP6=
z&|7X9{Z+w@#ih^|1?L?dCmn#z=)Vl<H15-h+s(+?imiZGxc1=2T)BCeU81vXz*AP=
zaQb}OP;d?Fp`63y5eTBUwL97zC<hR2eH9fFW+%X+U!Q`{j;ReA)?cK$(*FG^F3_T6
zCxuoJg9?`)yL?V0v2XjECGR?@mGnD30llcCBcH2K^Z_+P2mG~}@K2Pj_L(gQ;@w$v
z2oRkc4|qBUsc%y@C%rHGP#JJbKF))d9?gi6A2s9ihaWY|9T1CclqRLf2?0PZcah?c
zFdoxJMOhSnNR6X{2QY^fZC%Snz2}^dVcPWDN+mU@INd}X(uvV!r}-szQTLOzb1kFZ
znE1}LrJs5(D8*Z7(Mp(B)C1M~rBWv<0{hc3Ej1b8bj~v#6(PIT<I!6AxTW6QQ0%9X
z#xZJy<mC%_gIB*lkzkI1p$s6z*EK$nGn>+N&|ls@H~Zy!;dLDPjlsF;1-erqlQl!R
zyjwmCbx$-D^~<o9Mn0TaP90=`uS;UO{{nAkJoZHg>l?TBOc4>t;tX$1?!(c7XMhe-
z7BjW>=CThPzz||U&gBZ#fe=X3ZQzVYH~Lz+Afy)h<UCP+qq??S+XfhLai!AV?27iz
zxauGplHt5V&3rM=8GHl7Wv5>sU)6F~ozbb(VyWh)yoD&9Qc%=-IB}bn&e-~_*r5Ki
z*cl6w=k$DkcxgGo_kJ?kmt_8NZVA&LEgcVlynB31;du~vZgwrHMv91;r-GMKq|U|e
zYqiPn5Zw)(M3~R)+y}H3TCnq=ZI#~0_p3Yji!>V*Ct#nLWVV5fx)OVf0iEyk!1%+w
z;dX+k*Ucr0pFd1`*T8ILA9r<06>&(EowLQXQw%lAn}*lHQ&B+5xu<?UZQ`DGu24I<
zBhf;d`5LXh2UF4#j^9!fC(Hc!FRn)GKzBJ#KL<gJ8G<jb)&o2d<iIy8eL<#rPmHAh
z7_3B7!Qr5Y>L;1CZ?5teqQM95T%tj3DQ@`^8F5_u%X^j8O2TYZr->=&O2W{3C=(rE
zF~ILvLY5aiDpnzmo3Ss0&xkNR?mwq7;3*PjZ)jjK#S?`wJ&HULhsA*>h>$YbPDD!@
z@y19Q1?uB2k^<aS_tk7VepTOwCO9Oj4r@${yj||cBw#-jB~Cw3>`PMRa0QB~vmnP^
zYM~KT)OqX^A>ptcn<FeiPyS6nI&-*XhZdDU!-DEYH(#`xJ$;KU1pu7Y4DjqNAv#V|
zL0tWU3m_=dIErIDWPeSr?mwr3C_HjT)cDuWIoMso|L}*T@`(;nLGmt>;Pb)qD@;Lr
z<og(woEdRYy<g#Qp0x`D$`N{&kX4y@Yv2lH&FkTK5bQ2(JNOtX*1R<l@oCa<l9PD*
z91eh7$uUWL0p)zR`+hjmd>l)p7^86fU|c|%{f_Y=P!{`IBV~O6#gLLvx13e{BAo|k
zap0wh0D$r4@jbzIr!K<2jY5hun-upXmiKs{7nX<jxkV1ud1JZQDDQsi0jw6YS5N+c
zQ*(7e7TLe!kN?NCa4f74{l+D6`#P{&VerKSl>hGob}3$Fc~PlwbTVXNcVmxM<T24N
z;c{T_d@Cno(q8t=n!%OE!KchLyvQ?ru1fW5D!>dX{T@aUzGwQs{}wTql8%=I{%;j_
z4|W!U+riw*bj!%<vXcY7T=D-FGU-j;^l0O$@cvx+f?35mW}&&<B7f=V*3jZGwy>In
zqVbsm3aCDhA52VVUE|8-t>4*jbCTH5kA^E^!t4mwt-c2WJuW??-51{E$8(gV6kyrG
z>eVJ;o%)jxk1}V&%2esr&LmD<+t0UG;0JksYy6;T#je!4jIbX7_{VW!)K<9kmNjd@
zH!`+l77oLj1&P4>w5v{+__LVTJib%4al&dzWyl3z*0C0?NAjn?WDh{6R^Z>jJTOjk
zqIwn&82vucU+JwT!qL!g&X^EK1pE{FSmK^a&RC^dLJVb8qsZ~kZa5k^B&<RF9r;y%
zo6YEc&({%_^GhL@iA7{98;O+~As&Y1w#DhhonW1Bl7&f@ZL~D0fy}#e>&lGm6tQ!R
zozy$HM@mGUcQ$y^)=Z{yZ4EJEttJmHILr{aL$}8tzZ__+s$d8E0KUz)3?N>=b!nT?
zIT42=MV#><N;w`M0N@Ox{di`uw+r=g909pzKiPGv?TP8=*-viT_ygymicl|0%E|uD
z>sNRru&)x=H->0L+B;^)YUNye7ZVkD3?HF@awg~tf`l~Typ<b5XbME8GE30NVZ9uZ
zb%~MDQObvNNBm8??Ec|P1PH^4ELb3n)Ckx@`ZUXfz;;MrcP9CEAQqy3q*)ejrjUAE
zL~q0B$O~;GY5)-U#_kj*K)s`bKpBvl*1^(d`fn8MdQ&sY8`RVj4J3#Dh96BLRTyOg
zfvS~8A%^z+xbO&hS?xsP;6Zd(7cM#gmd<FuhBm4NK@)?Wp{NrG1gs%jnpe7bh@K-Y
zk}O2~Dsv{nExOqL|I>ipN+(8OOkm8sSTNcrs=~>#BG%`(?uFGo3XM-UwmZRJdEdE{
zJR-IvQ}K$*AbRv2fp(;V$mQV<Krnfscjcj>W6(;XvCnn?LOp&x_u2@fL;JP@p_WPd
z5^n)$Q2uk*LhsF+&{bPlpB2q}_B$t4jRQ)Bt$7AN0sDq6a!xE_wVtuw7Zv{XQ(1#J
zNCo4J6)WV=eIBOVA$)6v2z3t~RBZn(3NV465hFCTdCW0zqG^Q0C>4><zLqlB@!v77
z8or_G6CvogGhr8A*i#GP2~utgw|Kq#0f@Z5_ujq?XAkj*1z6SLO`!w#8D<XiWlWBs
z6hki+fx`o^Nb|t03jB~dbb<VkiLB8~f?p)_*C6FeXR(6}iQIyWa`_Vvn$~NC{)B!Q
z5%oBK6JikX6W={UHWX_yIm&WhnEk5%*s|g8<J3=()KLfxRz5?t4c*<jv~9%J05H$n
zAt3kfQF<f9B8?CU7l(FXp1nMB8iX7|Pux6LdmH*65<&q}WL0>-`3P_Z#*k+^V?uov
z&FS5HF(837s}Q+~C<*Qqq|u!V-nb(6n1&3MO46kG3jKqj)_i*yrTSNki!`@ui>MX2
z;zowPb#Feq8{HIGpt}&?fFBV617J~!t6^h}`Pf=!NQeCjPWZYtrajbQ{Om*j{-`#X
zb(O~wqDMi&Sj=aWk6+>%bx~(d@MXH3Cdx3W|I}o-VShomLWl1|KrpjSHsO3>?@#;t
zRyuDLrImJ8%vz|NpH4AWvK2+lX2J$$=bFH``KuP^t}^%OSuEjDynM?5AMlvZpfv=m
zW4&+)^f-EU-q~1YwT6XKDR$?Zx12qT=Kl{U*G+Blgk$*{4W!7PXH2*UN>LFw0h9=e
zkPT5#a%DgwG^h`|fH;=SV7E^=Qwu}T-Y<5BA<QP3KAH-av(+BuA2T@RmR$hUB8jR(
zeDI<0H56^@Z<IqOFv`F!WI&B12<&W&cn^)38#Jb*?_MVturM;Z26EUC8;AkdU3mKA
zo8TF&qNrbA2BC}KUnn&vK}Gx0@Kn%2Ep%|C-S2)dcvW~?0seKbFiW2rH7ePgNZ@Rn
zrN5|f#wKxLY(pD3o*MnEhL;XzD^NerCtd>gunJXg5zO=xZV|j_1qhZSl^9@?ET!E5
zQnd`vC!u<o#no5Z-^8I+Qdr85kXE5U&@%n@yuV`Xow1aCsu>tRtF-3)J6G#m3x*Rk
zNV)xv1c>^@w8M<g?hmp=Jp#-eb&ow04tr@54=a^yRL>oz901f<a+j+rPac%6>EQzi
zeAU;F;HuP*F73xpfMK%$D;-yQPuD$SjS7S)#~?*CQ6R@EF#U5DcGUs+`K?qnQqFN&
z^L>XHFzqx`E3e>q1KRq?sI$W{RvF_`!%0IblX@)%#vMc(QDlFOKWM$Mv@;BU0fGHz
z_7^uC+(+GwC_DjL^RPShv?8b+YiM~)E<$uJ9N|-2B3E*EKm4Cs&GXB};}O*-h8?Jw
zvbZhi&4&tbip?>^^zcgGe)SpJGhKyG0b%5YMVn1ojj$z{P~eNCM3z)CH>QEvDXOjt
z{$^%(*WXN})r-&zd)W*b^V$b*SdnIYyZU32d;m>tT&b4!8FfM8PIe!f;QA(~2ZVAl
z(~oeFyjdIr_=A{LaZ8su<z`K-ERvk+8nhYh4E1#pKMrT3cQh`+vHx$+<N1K@6BllX
zY?eE_%0}jMO|XO*qZ1OAN5)FeC29H`_=Z3a)C#=Z-5D)y@2c1b%XJW~(<c0_K)sWH
zO&O(9OXRfjz14O?`^JUaZfj<!gU&V!(+JLU+Oe1ifKj#FL2+m6&N^wwviRn5UXm<w
zwJ=O5BiGi<rC~0bBCwIBRlarkXjxgrF`XA9sUZ_ca`}mAb={Ehs7F%s4~r-1=I3Yn
zFy!`@e5sroMB_n($VTDdJR^~!qRz=(MiK?!{EX9D1rYh>i|2z3xktP{D3+SEva8eb
z6*oKqw6`nC(Dt13oKi<}va^4SPIA(}t#Pt>R*Y=+Tm~cxmWfnqsYZ{EG{#vekhi6_
z3>RHYv^7L*Vr&2TP^gIO(T_ypyx|qp+VZ#_G)q6I0UpmRA7NG`y4-)UO)s`SnP{aD
ziF)*_cWy%J1Jm!+UiC2kB{XTJnw8MY+$Tf;?1Iku^zbh1bfe7ajbzAIgWLJSI4qk8
zA0+Q>S<MiF>Len`3PJPrbzfx}tGrwOqVAU-S)@u~aSo3wm&Z6>d%UNR(d9i|8zD~d
z`}644#^hjXxowR!;P{>?JDadiJ2gFJeF;6S+HOgIlc|%Dx+Lp7qHX1IG`TXtS&qsD
z@M2ea<a)Jm1lsAZ>m*W;Kly9U_4T~{UXH8Jrizb_tk|_%j?pIm0iW;smC?biUqKMj
z;qlx&nbUoSGXrNgqUzRB_S*4wwiSiLsQbltMWN~jl)}}X&`wZCd;gBAKHMa5+-sL-
z2dZ%U=OiM|7V8G~21oPQ?J0Cr1sP2RkXq(DN3(vb*E;`p?V(Y%=<#F3nX87I$!>*z
zsCvc5Q=FAPXR(y7Ye59>6QY26!C>Q?WC@4aHf>zm%96&STQ9qw$*DG(nU$6hR%Hm_
z9yO6>L)3O+YY#^ECi-UZ9$-m<aZ_KPvmN=!oLwHka%`Y*<N7t=v2Tuxc`{J}pa9)k
z6+OruzRsCsGK^|-lI6x0AYX#}iiZ>xpuE=W82(#KVM~`&DbQ78eT>R(4y6}6O0FQ<
zt>;bd)YAOc+;g)Lc$_a_Gyw`~eg)<2<7RYC_A1cXR7=FZ-CjPMWUu;F38o);O_ung
zgctGu_!l`AcvM*ojX!|JH!#?*hfYJ`e!dO-f{_f4p;1tZOHdqM!%N(^yA+%27|^|`
zI5UGq;pniU|1vSDz|0BNiA{6K!FfrzcOA;g6WJp`1dzf)q6TEuU>6WJ3F021$MXmf
zS*aV0Q51E0R!;w1Y{NGE^f}|M5CCZZ1kJ0pt)fr(S0HY;07N`9WL63RCdn)H%5ybL
z%D=~MTnn-gV`Mg=io_g8`(cQ6D*V4A`SZ`;%gcYqo0e`O21Yn2WaSB{W0u_~26EG5
zr62f?0rM%z&xx(N87vp2FbJ5B@4?-_f=472zu4CCs4*o$L>i_r%_106wpmGA&A{_T
zqID3@O!Bs|02KRALr=p@r;xx-Z9?aG%AO{@hL#2xyXT=kyOURn(ss<1I~Pyb(l=RL
z(KFZF^0dkV6gD6Kmc@VfgXb-CEua4kFa>CqU7Lm5yR&D%R9p6huNKppMM6Zj^6ylf
zD^!$Tq4BJ;S6ru(1~KFd&N6KO7A|QwnlV(?2$>av0Bj2o|1nt<wmdZ{atLo;XmCV%
z6fhuaVBOpeZwR>{T|@E&eeH^TuW<7Tm+~RUNI8zMtVD|iBO!A=6w<jvw@$Ej*~p(*
z0hU4NcKf;3i=?3MYsA_B1#wbmW3`WU=^~}N#y(^P&?d?yOL#!#C^q4A5s?!$HsvaK
zo?nSv03Y?)Lnm7;x5$Ult7N$n_qmIvLfb!`l{@H`*+0U9OrU#TgL@xB$*1m!Q9Ay`
zjfIT_FY0h;=<gBaWeSEa{5<}(qFN_L5HlYQee3NKtn*lZ>v)NVIB`i;XX^6Z_@M8`
z1)=P?us4W39I*AwmWj8G&k~v{_U=7+X2n1sK(i?kPs#iLU=tdSt1OZdlZFK;v5zFo
z3;1Z#pYA`$NrL=^92|Y3B6TZ{wcQy5b1Feiyx7w8yfxAIWEm9+Z&v$Kf9ZK1*mefm
z`?xQBBW;>N%<oDzOTHtDfBqQBuN5&u1#X^4ZyS7U;64|Y-y86i^SVyO{-IAGfJI<x
zu!V^!y1(~B8)z0E1wBFv)R}tsL^b-FaPLv#0<7hHZb4_@KWC9JMra2^l#j=+J8|s%
z>bo6=I|{Pis+(WfQw6_f&X&dXN^OeZ{cJRc5h`@aR59zyiR!GJ0DLX&9}eCT;vD+?
z1-+kHclORd22%fcH+Ww#gA7pp0a4Dq5T3L+nT;E3(Ma}3@-4vSzY<dd=kspUPxyDm
z;1`3Y5M@#ozvuy99!Ue@DZx=ivMCqr_W>G|_nmo^M-h!VY&mw1t9?CC--^4}xXS=1
zHnljpMkaxYojn^09P!Q}#@`+^$0r<wt|;&;?rx(r?fmLLiK~8Vnx+!P0CT;6_a2r0
zMaiu?H-j}J;fEIvvwabD>h$nc2kYyQ&`&@cDUWkf;J^Yl27kN+dj3ls9MI^hc4V?^
zH%U-iBZ0=bay-M-d%;N@L5E*Yw$?h1GA6+)Al0@*bz*J3w!Y_k`F~y5(D!1oo7bw@
zYVPk1_AMOq46CiUe=r07@DekIJ|$ECDWb=yiFo-t$%MXsNXQ6;Q{%5UA`@Hqm)GR#
zKnYxERX9h|B@MfLB~@FK=*HC{Ax1-noKLg0I~k6!jc%Z6gH-vHH62BT4aPq`x{iq;
z_rjw2<pFz%Gnt5#2E_hh*TA+T#3ZDXFt*P92AI!CPIt{M3vC~8ywC>wv}mw`nL|S@
zEPNb-y%PGO^F^pPa4hO2KXB||ERH-qfv9?k_&ftOs=W_K0sah<mbWKIzg@Fs>0&ie
zYq@^7{usZKcT%ZI9llXY6>>_@W11Y!A#ee-3yMSi`L2*dUoCJ5gu=@$ULlm<VrtMM
zEw%SI*0rxFs@?+78X*3#<&+HZcZ1cj9l8%v_38J(1U`?fC9>nl=!^4+hEN0PCvv)S
zb1#nuA8~+=Y1ExsXBuNwcF^j3HFe3=59+tOv~C=<8f?kK;l@9NK>$!IZm~0b#Qn^5
z{XIzM(ya2p2o0z6AoHni*TT;~4Z{wCnq9PhJc)>_Dmn%bBc!TgA7Pea0!LxMFj3pl
zGQmod8`3RtL5C?!G%qXQQ)63X9D5lS&Wb7DuiU(X*YZ<t-HKIS$CU|CzTuF_{HxH_
zrY0}%^QrcB$>D;j4xThdp)KYbDAt=K2J8ZQFo)OYQqsIORzjkPWXkh19YSGMe}vO`
z$%Wj8mI((0#r5E^M8t`x07DT$l~(oaYQ<~Kpn3iqD&erQu}F$b0fWT0W#|Mzki5s&
z-}dw@&y@j*Q;k_V=INZX&z0F{K<FlY);#`!39#9L3A!8tgIYNPYf_vb3DwGLhbeE@
zyIzGNvWE@~vq*A=MnJPpXOU{7DUL=gY5=Q#nV|r9=0|)g)t470J>2?)yCmiL9jhFd
z^3ICf*{$XYr*Im}fKN^^T}=De3I{2k9)Nj&gj(JP))jl8jx1Fz#I2uSuEI0nbi@lr
zE}h$~VUx2kjR~{f#(p^6SR1OdE#4XdZDq?=1Yi~OI4b;o8A^9>yg!7{)f?8hMl8tS
zQc?qyRrvcm;nw<x4_(|lU#^2CvJ7hK=6econf$?6-yf+~kEFMoJsw|QA3s;9DY^Dh
zpMh7d0<?C#I>2wXvovHPkQQ&(WHe6~k9IT54;POL)Y}Z}Bzntz#Jma|5jKAXFx9PT
zjAi^R9IeY!L&YPN)o3cXHtz=p5X)>Hu?GS=dR33;0(ANi4MNTxMa>_<xX3*M$qoOP
z()hd4l5{!i8f$v=7@_M#l*mu2`S=+RNJtb98$!5#UA@iu$d`!OX6jQmBcNDSQrWy_
zwna<OpGvZ1$%(7q5t5G9yGV<{P9ng7*D6wC+xZw8M(fqaaP)J$5@Q=MdMH@EiPyB6
z0P4qOHY76c5m35W%5mmA#=uUge9SOmO?%1zDcfuV^0A*(Bo$C{Av0enBxbVVHyaeM
z)S9MBz_%{fF(HCwgvxMBuJ;vo-+Y9Ew2ydOBSq#ytSM%_P_+NoV%+G*-s;r}aowtU
zxI8Kzq%zE^;HFpfq5cC9;a}CAo5vh~U41pfaJSg$e{i1HZ4=$4W3I|^AFJcF?LP8>
z%V$)tYMtB#xHWInq!!k-sH!fw_6qB9k-m_#OE}-@y;N$adbU`K^%YeN06Pl7&b(Ph
z_kjjB*4pI#Yxfcwy{qhLpBvpfJZF<`kqV^9t$jdcRhceOZ<h)<AE__SLmlRHkFyF0
zU7!|A#>@=Rvh!R^FXSfiV{=-;hMuK1VhS5{>KQ!PA&mYlXT%n^=oGpriitv?AtWXQ
zv_*fAFN}KkSzg%1<m06S(l9+cH)Q<q;$H#D+>+V16JVS!4x%qfD^d&B-_%i~--!C{
zYx)OuAgLK%)P@hRCr0<(QZpd|IxUX`XKf)^*+Fo39fI7#W$9;im__p?@u@XW0sWDy
z-nn?T*aw8)Aq>13%WYFViFCi)5!iCsuofWh<JyL@t^<Y{iJq~k>uv=ppjUgCX$}p_
ztfwDAsqCjn-)!i<XBf0QH}zAq6$#vaoW9Mf8<h$5*qs2)OL+elr?~lkU*)k(gFm`R
z8SyQ$qBsll9s~&a)Z$SlUYYs5e#(JB8_t+QC)iOLXZGMc!c7oWOGKsTSC~!&hv9T|
zmXMcm2v{ouL(Nc_K0p?(FTTz(V$u5@8uRO^G7V=ZkhUJ?%_1vMhg2-Yo{VHCjY)(p
z&Ab{y!1MvqXk%H@?iZGu!}ay(#Dafg?rZZKpLl2~8t0f|1ka!FfUhN3mZk?(Dzh$H
z-z}AI!E}ad&U&}A79LbVPt!JItt@6kJNN~0=<(WZT>nWQftJN5;>D0RMj*=g(o?Rc
zIu<1uvNB4{`9e942pA|?{n@ud-{iQQB}wHEk^}>AAH9||piZD*$!LeMUJ-k!xxm9x
zVn>CRr^84KQ$+v0bD1o6OB<(NZ09yh`Wm9>ZMXJ$FOq7Agj!`?O|9Ww9Tyhw&=5xx
zXPnSqYOQVPoZ6iHl&6Bx!_I>I6=w<yWTzz}4-EC|R_nqVW0dpBLn>{_-j275SyBj0
z@?`*^sA!G%>&2PO06-#NYk_svS7ckGjUVKAGq6+0Xi07Xoe2juEekG6kNB{}lG4Y_
z&@qSZM+lttkV`Ul*<t!Eo4C$JFNCRRdZtL`y5?QFZhS51+@Lo9qjow_LzZC|gYDb`
zuO0xi>d?baO<#>F*G%)5u5jX}<6I61Z=C_A`dP4-CtE-NN9)wA{=Zr$TU{)K#q^}Q
zpMOZ01)t!m|M9$e=^C$pgJi~z|BZ9;1K6SYBYF6nuZ~I1SLp94%VzH%u@kprCBuU0
zBz9PaMuSyFb|w1p=4V&(DOXEHBcM;3ft6HVKH3lkKWHK62`I}^9@BFZP_4nMYT>=3
zSX<rUpI_4`SodyJUV?-f>TUX-EDUv807f|nQC-IMR)1!2z<bKh{J<qn4R6*1rKb@~
z^A6tdiII0;#``zca=y=_p`=T@$I3h4vSezmXuDWp=Ui_l;Vp8OM0M%e(z%n&8U=?G
zEz10~zc<hwadZf@T?nZfAp7o@KHV02vap^t%iTtfR7+OXiQbPQYh95RNSFK{u~PzR
z7&71FPR}}mF3!!mzKKk@hVMtzEQ_UHvwDr{LGn(iR|*L6F)(1y@lN?Qt4}9?Z8XH?
z;^6-k5MNul_($-$rDHC+gBM0-Bg{Bd*~;mCLSrJ74rVWI<b@a@)7JOf_~eYpM*UZ<
z>Qd?nNta!^f&5%^6~chRtOW0b0+RQDqisbf{~q?7ivhaBC-|#w|NhBXE3kJ@;|Rhz
zgae9ZIx$~q-Pol+ef#FLT{H9~pIZK=``xL3tKon+OQ@Jjz_GFQ?yZJ%_&bJr_bg(#
z_=%kkRX%cCC|@@~x-Pt)D`e08<iEt!j=_tyyl5X;uWsUvxfA3q<9mn^AEq%M<V@ei
zUYG9%i(%@Gb?OOfn-Yo!HA2hKA^|gxDk=3Hxr@wztrUGIYC9S9>7jSCsAUr_9|qEU
zjJsjt7P(>OMI`Aen2}8gssDnXm-pu(!Xvr|9;rxxIN*ZT1RkX50=pJqKcYtegxI2~
zZ}h#xr$V_GRgHpY^zEy}4U0~0+o-)M56~7AW50&kx=QAC4mshduSyhkZN~Qv%;pb;
zki2;!1sa{98Ice8hY-sOPLyjJlWSf{w?c??YdB`SS++;mcDaIUwX-w(;nrck>h=pf
z^O#`H4iJ7c4Bfh+g4Y%L*&+Y~jXGSMO=!Vf2ek_SLrV8<reu7Fbd9+RB&=fVAsq0S
zLftz(*Y|BUbaY*k90UChDxSTGaEKuCD*yPrd%Akc6d%|iz!?<kwKb0Yy`}<p@K4yx
z|D~y!<_MAL->R6G_CWJC1M+;7o!zL6I`5!}44@$D5giQCv5Z8DA+aV^xgi%mf~xYl
zZ%`Z(m)DgG<G|gCt!?L$Yv0|>ka($VHY^H39NSfCfY?8;N~>V$icOejGH`_ixA9Ln
zar}SSI%LAo`6_P`m(%sQUKl8lgx_%5q@p0S$A~?*ZoqtJP3v6pM{*^hV;El=iy_Yv
zfWjmhTV@Ns3O^bi+6~NrgRr1qHgsSG(X#6PHMuqTFa|tOuEA;r*N^1-f7Yc+<w;ZU
znsQeyct-v4Q=>NpeuhVL6Mt)(Y$%_#JKq#7LYTPSf|%PX8~**GfVRQI`q+MScm6sd
zbBw%oRMXay;x2NtehAq>-#%YT-^Ew)2Q*_sLgn4}BXDEvOS$Lc?hwpo^MYW4VEO`?
zH{FN^A2a9VctSE2BUsyFG;#_!M$)c?PfcjbphKNvd3hh&-9_zXk3-R#IYmzl{wKqz
z1$DUA`8z!Lj0MycY8N>k&0D!1>ep#SY2jIH;q=7WRRWnca5mPuO&YJ+)JO3<0QZWa
zeXVX{US1MpfVtgF4l<B1{;7F;4#N?OO!pRRA7KJ)+`t40jKJ>V>Q3+l_KNS`v*}%t
zd1LqLnq<gOjC4Cl4Sdn#(Ei!*rop%^^17$gp64q%9+$_!@XE<s=p`$zS;Tb*-Ib?k
zkS`;g=O9DzJa-8g?DtgXAe|xuz@Vh9Q_ts0)%UmcggeFV3Y)WL!@b-xNPr&yoz`1#
zOd)ggpaJl4c)o`sDc8a$w0h*@4^=sb!QW??zIW_79)!xYJ0n*_>WGXm%*#!T&TO)<
zYd=H|Wp7rttv|^!T|d0=VcA_p3wi-u7gsNMJ8}+A8f?c({DfjPJj1>QfLI?f^FE9d
zfTJ~zJzluKDDJRd0GzzrrYJN~9TLJG+(NFDs{)b5oyX8J_HGWF!YmP`HucS19oif{
z1o*wor%+Nc)W_T>?X2*W_-qkUc%=YKu**XK*jCtO!DI>&vnC3%NT%U1!aX;r+hm_L
zG`qJ7BT`~E3(<dNtY%gM^pLs;;3=V0e1oymIM^F}Ud0-|pVo1O>h!r*1i~H%jRfn^
z!N-k`B?x0_cXKD#gcv}64OuJ8K?S0L_L}#Oxc1~?2pdBySlLP~da5e`cRm((^B3y*
zAM1;4&i9Xf8KyF7B<$*fE3ij!|JYvfmnW-n9FU#5mwL!A6bjz~iW@uU!>%&UMlTNS
zF2jgQ&$r-90j}tZ0!SKKM@qKU5Ys^;!%A0AYiwtW{nI()k8MW5EA+u=pJ-nhdFOrj
zm;0Zc^NKLtrngocI5Bu(V23*6csE8d1>Kj^^vgOYrmuxAYw5_D4E#?^j&u|)k5mA|
z*Zu`K+V=}DKiVn4ISoN9<7@;=mh9$>L<yKP?I4P+A2dpz=_?0d7Xp5O?xskiBV^?v
z?A4J&)Ynd?w|b}QH;4})no55hfUQE$DR!Eb55Q8|hC6~%pS9V;c<F9y2GnU}C&xv<
zZ@O;k^|1_ckRM1VXtOSDFnYfc70G%%3`DLa_lH{C?%)6#f<K&izaE@FphqaOFVvoJ
z7MLKa0Q3i9tu8B;8qWoUsTSZf9n1Ws%a(bbyKF1QL??og5~(4~ALGdT(84G==QrTu
zzpBn2Q;{3`t^R?K_(tIQQH{je7S449pW<wLgY1<N`#?G0zaqI1s^-f?j=!_ycz}*`
zJ`P?pTCe@^$BkajQ9KCwbGz_NQZ@9@$z<4|?1dd*rFOU4(~lvnNcx{&Rq6G4o!aQx
z1N5}cnAa}aKmtn#*(dEM3Y^o^-QS&dSsn|}@RZ0kUi}5U<499MGjQKj{A?Y6n+J+r
z1twd+ZLt~6UWki?m8@hdu6Gwr5Q}Kq$U6drA%y{ZwDMPlqtBZmzy;(nM_x}<cjvMU
zZB2(+T21hFzS9GCHHIl+deC?pJxzJO^VPhHIeiwN8jm%a^nfaY>xQ$#r76AF+|K!r
z4%Zq-MO(s)q-O-0O6l`$`)0xKP|%7J?>$ZAC>l@L1g-|S?L%9J*zv6my=ms(tXW3P
zG^T){$}se18XmOIB;5^o31IKTu36;!dKv-Pa&|_Dn_3Ep@-^3DpI8=_G`N?cpqXEr
zA+XD6g?R-%NeQ<r_aQeLPO-ewYkZ|wf%Erx9M|%;&t3$XOs<kU0jburB(H8B3`59z
zIJzK{(XgNK47FKl!J9Q=PlO&7{Ndl@+eH8@qc_iysJzGo6nok?VO*$1H;+ZPrXtFP
z1J4{p`7)2`jNLW*6^cABYQmKBu|==>Kzhx@@;TlT!sP?iKfJuy7({d{#8O?KY*U`|
zUiR=tE0J%Oi-z-#Hl9rla<s(_@=xW%wT<KTbh0S5)e<4w1bou3xkW}A)G8c$kc|K%
z#_|Z%p3Kn%oCc8qXnysU6ej3n$HE08om<1cR$}(h8ymrwSdZ*fZ!j)@-gcs6VEEj^
zHdAJb8vkN*g&e_5k}0;q)?)`)S+xOfBfj(B3`x?V)}k&DahEgk0+9$kf{`RcETMn-
z?FpqVg*xo{GN2p--60HZxj>J!g~9;L^aehgtwQe6g44t^{r@WdL5}NGC{%J~>;KLv
z$1gRDffjiyjN-=%k)VsSp<^;o2Ozck0S_XTK^IOdKM1kaK3W=$;*++OLC@DKLgYoS
zka}Al)(tT`kFLp3s%#UhH2!+S9FJSnwE!As{6pX<Dp{@dVgYn2CWeadjoA#4WlCOI
zZ<C+`ZOHyfQzoloWs6TP`EC+!Fmir{cfmS&3uB#TV3*m0SHBlQkwK~7mgvi^%i{*N
zw!_Ua9M~{yF{IQd+Efyx4JpZ|@Bb&?C!%^18IEi|026AnSqmmR(*Jj72i%=|kF-jt
z#9}O|ZGn8@IwlxIvxK}ZD)}2=>q0-xFkPicZIUqs;VhYgypFEnUuS@D-;0@YTs?t#
zCH!WtOrzKbAXE|{<-FC@4}C*<B`4Yu_??MPh9!$RWVp(XJ>w~!YNJiyN|$g0{5az_
z!YSKgwH+e->!!4y&ktkRjX3+-#fBZs0BYW^T||7!b_y8|8$v|+Qh^ijYOmw%8#+yX
zU+HNUpM8x$=3r)U%ScG2^l0{A7#?<tr_H14=nvycL~Is^#gEGFC)K8)NwPPMO_Je^
z*}|~+XE9oGTa&R3x8)hT0G$ej8ZFJ*-h>sR(4-r2Zh&I%jE52}&26QjXdrTG@lkG@
zLG#L1|2IOx)zvg|oa-V0?80Gt<?_NmaWhWY1u|VGE7C7ll76asRAbx#;rH4!YP)7N
z;nk#St~^(*cGdGi|D9Ak2TY+>6<%zP2=YHZ1prmzh(aRlaT-_OUlbFzF#A+2w^@<#
z+TbOD)zmP#Bc@?A^7_G-iBMN2hh)xCMh=`d>B*_pNwN1k91(MXCb?5QuE|QG<9KR#
zqT}Avh*M^#q=7&ws&vFL_?KbhhIP>41543n$U<D3_gDZmyno)OWl>rh<-rfgp*+MX
za*oxT@~4^9A~qxvSsXzo^pAZP*hk<BQ8Mjp4kkg`j}6&SP-!u#v`DH0@I4);ZC5!1
zSI)tosJQEZV~aX~=Pm%s7##r!slc-{EL_;l(%%`aPk3v}+SMvHC~7vnz7-GS?)Ciz
zPg6@GnVqj3IM($tPIhX55t#Vbx@77_-5;d&%YXna;^Tx$Oy7{L5vtx&4&!My5ST=H
zjvY#0OrzPufVa`DHW%{>vcTd>6NM2<QL6Ut*!%1aS=1`PW~{0%(Gh7sWGm#+dzC_P
zq~k9HPNt1tXj<x{>!k}nx|Kb7j$FkSto+)sOQYH{l^#9;O6+Mx713cRY`J(9DEJsV
zv-rSE9LryCz2E$qhoGEjXD6-w(kjt&1H`Q(&XVcqRwHCncU22#A@6g!8O?C!fVkv5
z%Xq}m3R6yi%)`;~onC!uJCOI2({W+}=UVJ5ucKLje8ia!AvZSY8ElGZ$8k1xjWp2I
ze1nIAL74wfgxFk25Ry2fTdo3gBRM98w!KaPO&e5lIjZ7#*ww8eM<N3pK~luAK@<0t
zTW1t@c=@RW)O5*pi+xOs(>{+IlWDi%f;Vy2)1w^V*r8r^2y3LMRJLF<`b_7I?72A;
zZ`pFVRx@f<pLAAD1!Xp_kz;!8+u3o;A}98Ga!EhK{E|%lZ~~c^6+zzWFqsiS-QS-U
zr`6KO$v=1vEW%e%kW%m^nJWMy%Vd+rss%W#EW~p-2{lSGu^F$QAk82(WXMy(YX`W|
zhPofHVvI^yhU*)JXg+{AXcyda5Vk^({!EWND;T_u5NxM3uD`cn-T=g62%Np8(^NIx
zt*aL|Kh+F7*Vbyg?ZemI^|jV@xgjplrNmjXa4by}FSeVjZOQ4$G(#t~M>0-4ZWd<R
zTR4f;f0VMbvhq0SxoBBlm;X4pTGRiZrVc>rMSR)K#?S@OC8rLlb<K*pIfqPfGh?`0
zAEt>{o4Y)XLdjDvSC=>TzOw-ryZHDMkVuL@rKz|Vnf59zWbFM#!ek*xb0GOe_h^9a
z^QGkVh`&sDL=(37@kn}<GPn=zbk%+|K#4Xwk7fBb2cAYM=RJ)RLmCAse&4ERrv_NZ
zi7JwG+ksUH=6Iz7r#b(p;_dto=T%7?LwalY@43f>ROBLX-q~>Dl6e*o6&_G)FQwH2
zH*ssUj^+ce@nz1%b<WvAHUq8Ma;BoNqO2N!3gI32-7)4<_kvc^{AZoG$@q**I4`I4
zn0^`BXsic8#W`PkY#cMaZprXWhc|#4C{D@ClAIrakfJxgdM5x|5sKnepsrR?gV)?i
z#O7R2Ofgu}B(IZx(RJ|T@tQWxQeV^VB-*seOWAC-FFr9laT}CQ=*{>Tme~+vH#IfI
zFues8Ha+R_Pgxh9v<|CK<Gui%G+DZV0TdAp0p~PXA;A(ER{Z3y4oF^fBLxsB0#<<v
z%(AHm4Pw13>5>~-J3VpTFdgkJeti(ti&>sh+?18WKlU7&$=<Ek2phBtN`@Jj0}{yY
z6}!_ftFiJ6bSw|XDJ3Y0-6TuncO1RK^r<j3#_uS4%zYtuCbRZ>HuPnB9*z<*4fp|3
zIxVg$0Ke3=RJja@g^sC2@&(|v0RN+<J}>YYC-`MBpybbM5O{GL@)hhM+4`Q4k0hwE
zqJJfs`_qZiFm0OSyHxQI42g}pRnok3<%0IG<S#>~bf&W)t<w~J>h8}X24i$7di{~k
zEs5p9Ev?5)i$8YY`d^dQoONtdP5!i39Tq|Apjz;0H8{+=f5FMLNml{Mv3em%0Y_g_
zcv0qn(aZ;Ko~6j4dVGjB2<wy9K5m3WUga)9BfkzVlryFgm*V?Mktqh#q1J7EBWtCE
zaG%TtdF@j*fe26^Gc~`ZzzIy8!pBR}zZN>g;N?koKF+%CQ*4o}zkHm1R^MFQoqe1)
zyR1L?WD@1D?8Gqd{~!a(?Cjr#OR_Vs2ssdyGR{im+;==)C4;`sKF;taKLeU{;vySu
z<4E<@DN#bTl8D+!`hQI`d9`bDZCFhUNwHO#e{B&q>Z`+ef#|O}gl2B~hp*08(INgV
zERy1EH|KApFr96ub@K%!jEUYF2^`yvG!%ByR=#fPNIy#5)<FfhFDLAp8NF%59S(^p
z-cyEV5oYBw(Ub5Fg-mS2#G9y7PRJfNjd_(f=S{|6Uw`B;{$VL5gu5%~sdF!wMHTe;
zh3RQu=jxT8u-%mhzI1-_)<UGRB&kd)vTRaQArPPBNMEjQLhTz<lgHO~ef^yeq)jeU
z;+Uz6JJ>+%QN#kM>YBPZ;vD&^718NT&6HZlkP4PsNmbpQKtnD@n4fpk8K-xMvMh|h
zr*Zh}<LvC?4q@&7sgmI&`F?kNKh-=Q@pSkIk#rayc=n#GY|86ZRa>`_U-K%C(Q3!5
zQ`M1!Rn$KN9XuY_K`5yg8EI%@%{QV|b;SfR#K`zWkZ1@nxEQopi`47u!GahnItle-
zx(t2sK(8WWW*P|*;(vsNc%2i`fV-=V|7IFVu_=1LA<~z;@9h74+6@o)-xXe+bjAwB
zr%%+~4VP?W?b2Oj!s-y09GHrc=gji~um%SBY(YnLTAFq*&S{HaRb7IJ9eO2=4<CAP
zrLS{Lxa<M4e|8t<z1+f<!h1BNgk=W3a-Di$4?C8*Nmz6m*09428BEAciqM%=dI&%d
z$bYUPHTwkW>)#;g;D**+NwQ)dd#}cjy(XJiP`R5>G{!DDp#T~j3GY80@H)#!Bf%rV
zc1HaSwA9{%1b*<Xf;v81f~@3p)I=`<$L+cq@BmmaoFmrUxExloWf54N%z+(opWT!b
z^<JpDyx7|@?{y5nCJ*;JNTb_$mO2}EsDAD6`%-HB5YNjdABnXo`#0_aqRWIaA!l9l
zo(dGS_`_p#@3Z!-Sac|{b(8i0?ms30{n~%&F(Ga`(2HPa_Ndf1NNtw5kot@8&XA7D
z%>m@c?A$GKCY$(L%%at$hb1K3Y@~!0uTZeD1e5eTe#SwmEp*dZYkUlHh3YN}LmDr3
z&DiDgi0j-kF2=9C4JHCTnIqX`<u6V?KpsXm+L5bqpcu%s(w=vz>jC7PDx^B^ANlBy
z!XGaU>b7jlstGQOuwYozn1N%zfcJq}W&znNw)ggr!2#;9e#vmF<*&2GXuopqeIWv2
z`rPQ7`|iSRb5paO(u5@NFjCtF>_f6`uvy)#E(ZhNe9DVTeShbf#zGOW33x;jE5%7C
zsMaUQXVVbgcL*zRO(wJYG&DEhYu$yc3sObxm!qP-XJh-FQ(WPTdi-+}oaJ|4zXZT)
z1AP=)$0;r5Ac-s-5rq#j(H@siMIu(9v|vwivhL4B%J77Ak40wSm$^V{-}=2P)@kN&
z)eDtGq9D`p7A}V%K+*BS4=Iu|!K%!hXgBlwLyHt3Ms=Zuom=C0C$kuIp(j#iY`VlN
z5$Df_JiN^S=gk)PVlA2M?n*}`x&TzE`_htfu;=gWe{<c-ASPqOv5zFv5W`n>>*AA4
zO)0!PUE5!qWC`jO(|-~%Sc;YCEqTsmh_61RXD1YUC@ZY+-hT}is<*S?FmE%9jt{ia
zpoBci?75RbaEwaF{o%05*$l+ZWfq&VVGMwrI70;aq=je*&=?M@zYlO>Tm$s}fvo<W
zdnU!t8aV49EaEm8Z>-4{x!%iBotC_bz%N6stCn=C$#6ooL7pA#7wKGc2v#vI2(3`Q
zacr@|UBXs=zJ}lSl$UW;VpWfFNXGVr(rEMSmzJ@~)kLg>U6s0?8Wvc7QgkGsG`mOL
z_CQIlBUX&wN*e;B<Sfw5BLFBk=zDaKtVYB~s!EwRfU{~m{OPG&+M}Cv;^~TyBw~~#
z;kgrx8Y*o^3wn1hM=RApNFxsW*g!*P)`UX1$nVo$vz}IKr%$pvlm<63K-8zL0uvDd
zNv}UPxk3XUtNA<iRHVj&-n&90p~>W0j`PBQBZ3Ss{4-;YqSeUd@&WZ`JE4a+PN!%k
zpO((w_9{@@E9n@z_Q=)OJ(3ft>Mx*q_N>u|DJunop!}w9Q%t>P!)=zL<a3E$#F_5G
zpQC3f_Q`LZR%xVIj3`pz&Ch<S`rmK^5rQCULiOOjwWg2(9o9vH1B6Ql0W<Y7QK2al
zip+m<@wgG09}c~$HUXzz=dt%PC|V<^^(~SPQIS+ZFKJv{*A3ds+uQ((?TlrCuu@Jg
zJGSa|3P!4BfyP<bg_MTERj&5dKl*wJ33it>4zHK#4c{(4&fbeJi^y&%yt%G|BxzlJ
zbS^)OMmpSmMjiE2l}~8j@gU~j`yL378BR9({A}o~o9rN9jsPMi;@edbcf(OtoY=bF
zl9951fBz3#?-U)#^F@uuwryu(+cqbf*v7=@*iI&PGBGE%ZQIr)6FXmi|L^7AyYA|z
zu9xantEx|(z4zJ2E&gT+UA&2tJ_RwMuGC}>B>0;>?+P9XK~6F<Qn?qn2saAjjM9{1
ztB+oFYGJw_vFO5pl^}o5MARt7k1#wFBOWF6%yJMTY37PQJTEv(cU9Vr8gLPDYqYVV
zdp}lo1I`aMm>&wC%_dD^fVkPiui)$&d-Nf)9#x^l1|Ikn0b7`qG&00w0{XgcY<snT
zY`qzs7@bt5!WMY3&?GloT#XIZW<&Tiq%Q1i`>6((%&2pu<2<LE-S)mewtJl7HX~!c
z6y#jBp_fh13`saC1$O-;-g4sKd(*0E@+cWj*ySym-s!!ES4fHuh^-0K2d;b1+P5Ng
zBW|ptTC~fEYC9`;y6M%}=~2`e*~>2x-h;LnuT)Pucvl1G+Zh*r$6cyxnYlp-2p)1L
zaO_$6l|>e1w;g5q2L&JdN{%W+asvWqw=-0Lleh4>@K1*y-us-9utZPIKgJmLH9bgn
z8BuobszSx{Yx6sSTw<XGVgnGbA==JOAEDhRXnLFtbLF}?_a-}d?Ph5=c$Q8ERj>At
zyAlNJeI}JtbTou-inilM5D$zn)M|hC!ZmG@_`}WjTEt<_XILmvOh}TG#xrXS;A@}B
z^*-2J{h`@$oSykBpyRyBz`drWQ(34x(t0z4@tGU%x!O4Zc8im?HJl70LZp2{d3hi>
z0R_@{&p_c$c~8Ck^|hirFh+WX9WB_uv(Wd!4P|Gngk#owkucWQFz@M>Q@l#EVqppM
zP-nPSU%GU8{C?@9&9>k%j+%;l*pG>JXd}9{K2(H4tL6oxOqW%_>cHi_%c5O`gOR5O
z`G`taZqymz8*WHK!>vJxYeRI^JsS#YQ*hrr(xsx~RJSc%QLf8378^{#9b76%xhgOp
z&oGCF_VVdyCbp)HmZ?EsJReu=;A12N%lA<hTAg>uWCRld3&<#jOjojwOUiqz5{6M)
znnQ$Pt8WKuWPuD``7gTo`=jRyej#*En~hm{Vqf)VSKq)Q`5h5ug?J{qLT<vvklM@i
zw!fi;rM&GPj1(PDWiOkHZZ+tL^OX~m`l>1rYlWe)CN~AYunKxTN;6jAjziDa2Tzp#
z8fUr~o{zP3*j((A&CZB1=6;#kI7{&PsWsB@p{pkyvE#?oa!>dCOC!h$?{I&u#qUE&
zU>tZ>+f5yFpn*W8_V;i8iz`WDiU7kx$JzkT)v#<xXyWApNv<BYw&u0I>pA6kQs7rp
z;3rkb7!deE7(fz0V!A7Ek<ig7Psucpn#gjqUHb`3ny3&?V_1s06*I)We;sr>AEq0}
zJE1tmP7;eEg(k(ODp}2M#eR_MbA3=(C<KHIjps%}xG0}eQyt~q!QQ>+%KR;s7TpFZ
z3)={}^JMcif4jhaK>HL~BXuD4-IC9>vd81mWNu$UJPmmZzVk%%{WuVS%Pr??_rMtq
zw*B9TbAS>di)=2B2vVwShr6s%pE@-MG?XopJ$nZ#@tbWO^z?m@&&p3?-$4&R^QA0&
zk8AU;%<us28RqoXPcTDwJvwwe1)uSHI}v(SF9#4mlGu&>?^8O6iJ}f*BZB3>X8G?e
zCvt!-FE8vd=t-Ai<As{w{p1~lIMT8!?->ln@q}3%(uXFoi@N)+ldKh4Xe%nNR#?HO
zRk9FUg<Peuz_pqSX^9kIN5MnGgCeZ?ne8E2;DiXX%Jg05a^6mAWJ|wQTS$gM6f5fA
ze_T;j^(5N(LuW9gZwqJ>XiKNQpk8Cet-a8J{=t^rf&NtB+rk5psS*3^k(QR)xqu$2
z$5H!c2lI3!ME??wlh{i~wxAh;4mteS8jgw@Df}i^O^;C~uL9;I=`{$!X(T3QD87xx
zKo!s)&-9=LN#Ye}hJNBW<>e(z2YBYtAGO(|wDjxIX*L^L9bk|gRADMUAU&>!X<Yi}
zQ6X0xWmeFp#r|B5{3~Y7GvGM6s%D}5=1}z%R}m@1Xh<&jWWRId!uCCXHud3TTNIS6
zNNa2!W&cIFPYW>Mt^~7l#aKe4G8t+*MZdL}cOeFcpLi$&!BSRL96{99W}FWsV(Dtc
zAy%D6uV*r>79^cRJxv$qMWzY1-9<JgV9Ak*ehl#@XBON)Bw0bdNQY=YcE+^`o8&wr
zHzushk*U1>->r%GMy@Y3-A8PIe2*={V;q-x8%{PE=xs|Ya|ri<XnrV;;?EP^S!?o$
z5A<LoicN=77pE>5@1{9IIm8ujRq&iab))}(gnOrliE#Ns13)F@6?9G^*Gf}VAuz+#
zTEE_eX)-G?5DXR*YVLUQE`67B0jHFu_=v*=4;*<^)HUEv8<9M}TDUQfFGKD(FgL6V
zla?+6j-lU84bcqP5>GA)BcD-wyZ&!wo2@giR+?UkjKS@=#BEGpD2f4J-%ti>&kKy~
z<ncwF%6GO`c*WUo?dofboztp&Oe_xNR@Pe`6+5HmDnNBrkHzAjvxK&y^EP*}v+W;{
z3&;b%QaqE_lN^G*?xkASZOoQSzat9)jLVOU{L^lAz8&4}Yl#jDj=u!U0Qjj(jK#L3
zvSdQNKsImk<EHKI%eJ^$dY6yYh-~%8?x!c80O69{)7wbBw`*5x2U?YWQh@4*^~Sxg
zJ4LNS8`!L7+|$Kl!XaDgWGy%qVB=jf?2yykOecEU$UxI)nPbup(+!a@rK>|vSvG|Y
z))(b<c^R3E&kmE;O(gO1rdCq?_eUGY(2#Vo${RVEK&83Z*`8Jm;aAUy2s@r=CM8Z5
zA@Vo~##QPsQr7NeV6fSL?P1=B@(zc?KuP_U#`3GZ#}kH?;EP-527l!s5aaiN9V7Xv
zkA4+Ip!&QKK~#OY{P-xCx`<+O6+ff*^|y>&KHsIl=KLper8x{P8AN)&R2a*3wOqan
z)6J>wf+eo;ym{WCgBcM1ED^bWX!J(+)eLIrWOuj2ZoLDy`ZDk?LCx7q2kW!oRChI)
zaiqDdY{tW5juZh08D^vpNJRM&`14OPnh(kYAR{|JH*+Ad*FQFP3N`QRSlkVcP~3g|
zZ5W3+hu$B%S6es`l)RXO!d}VmPMf%#SX$|biH#NOU)0kNyzacRF!t*zDPyNO{8dhn
zi`)r!8U2p*ewOCF2UPOoA|c}}>`0T}O!mwh`4+l;W#*L5h?RzrKroyZ^ES3S^5A_#
zx40K(<6IQUS2{>H8T8^*2^3O)PyNbEU%U;wP0@5GGmrY=TiZAchX|h^mMz{ue?Tvj
zgie+sTi0DB79X!Qz5WTNCH`LiYem|nDMW_s;r`^>M&Q`G)lF;Gl2kTuo=`J7ZYoZz
zKzyKLr;1V0-<|3^KzSAlukHH;9{d>TWgxRR$Ma-0Q8-B11V{4ytirVq%!V?Y<NJ<f
z6M?8#1!<?LW!?|I(yNk3y&IgYc4SJcKQ<M1d%b#|q-FK3I8%|L9HLOJm#FjFNDK&Y
zhcj&l@ov{dI=AV4t*C_XyxBfW(tkpYIc71Ik`!T)Do_-W0dRWZ)8IMUq&X|Yd+f-$
zNhH`owYI^eX?YAqnyf!GS++KokSIkh9nUYAEU`G#Bts$IJH6?gUwqN?jTP}_G^_Nh
z`h6HTR2J%<3`(j!v?2d;W0_N!CnmerAIn<hb$n?Rea6LEEs<~wl)V&}={FE&zqL27
zMaksC&O{M!fYLBdBtrMo6zupHoen+B&9}e}I9m{IMi)_k3+x@r<U^Qyh(adRv$gcY
z8y`b^W!jk<J}%2N@~Qf8mu*B#5FDv1bdenG_<F+wF%8iF=&*{R)^Eo7omGLHJY@W^
zy9?|r6>%Wb)!8?~z9k3GDd2$wc9;-eSWO_hn~72Ywo+lPna1W%l|8g)$V&|P6YZ6$
zsa|KA+(bJTnrTT_%udpq-`6!<U`Eyd1cX<LTGZ;BW5w#eVA?np5HP`nqh|@Yp4#4h
zR8)jf!?c>$S3^i}?btaNp60d`J4hwvbUDmUYV{Ow$u|-|(Kp8-n8w!iCl6`co9iFJ
z;;a;ae6C<T6fQ)-^2(;KhX>6+L&LemT_pGP`DO;e72n#wU$T8~H$P(Sq0QLc7r}bk
z1*vsd__V(?`3?J5j?LIGamkm;!y5qtn0${|#477AB^{u>VLNhn!Ga+Y4!%xoLAS=X
z#WSOj{GP%we9HEnwf0kvkB_MP{<K}~&CLct4Y9gS`-dXLB20q7c|tnhjS-*oXTM@9
zB(nU^@+&n5X7^eI>k$U5Fo9)aUF{8%(8ZveF_2m^%V}7QMTy@%tho>*JoK%8K<Z0?
z*-CWWZ%Y24RH)t=MbM#{-*S*c8QJxj|G`n1k%Ym)fBJ$G)U8k~*iF04#3^?}z6Aor
z($lGQ2~HyBB9#9&31yq79gUD-!jfPXS|MDNP6T`Hz|ZQe3t3Fv;tlpXTp>2y{O#L@
z%!6>i&(z3Kt`5O?_sm0RzHA2uNE=4__)ulSDME@$BM;fQ;o#?wtI+5%H5epN99xyl
z{$1U%e-CHGKn^PWCR()Fzl#mS4uJ=}O$5cF@J(18xvyCa3l_THNc!*0unwan6Hy$h
z)lzJ#kl1;Hq_r$GkRPjpRf<n}vx0ul>LL-Z928^0DUjlz`*eCibo>jqK&!SbchK^_
zJ*n6(Qo4;Ty9&EgR~{beQM?Z|aK=}SBcD*~{qGw$oH;*C3&q7uybAOR4URJ43YTye
z(V}%flo$Ku4ca8p-9rWAvz){CjM!=(HF%5h&A=3?1#6HAj8=boT-5?mmGBi8jfB87
z@psXJRsekf46zss2PCZ^gVQM?{Iq%!i;xjyFqMS|X&`bh*V1+V3kIwFOm7Z%I~r|0
z4nn4BN}zYO$(S40++UOIDs>kif*tl-eo)0inN^;D|KjI_dbu{O!@ODkOAf#9n`M3l
zXiqtXVfaqL%or5mKa_*{%Df2>?v1RMK!cpnK7zFU=$V&rU&jc&Qu9$}31c39!-e3K
zRs6^~ag?EelMq8$BN0j>)TnPwheoB=K|dM7UX*rK`sm`5iB;9cj{YPA=a{EeGHQJd
z5YVG2$;@=3qYNO5286AVo>uaA<>G!RSwA8S4JeTfpQdgK=$tv4I&Sm1X~LG_1Y2xF
zmoXxpxv%|}Z}sdU6WI#p==Pyu6hkgi`+R?O*Is9tMgJx}rkR38T~^-vA;OEcFp9#I
za1Oeo`UAU{&1cLE{a?rv0QV1bi7nF<Ml>KL^`-%Vj463!P7}lfvM=q3=sw+qpEMFh
zf;2K&-v$P|$^jN#eI?_MSzxA5$i-V7@muMgkG5|GhN|NTM*$5X4PL5KI<}N7$S54s
z8JpfX<X_P^XqI3cRh+;;;n7jDRz~^`7?z;4^uCFraP`1DSer5#0GFbJgDi-XK!8I$
zL3FlAa6TFLCna8qn2Y5li9UvYAFq^})@g7^oarC=s>o-(58yKur^#kRK#QO^Xa_e)
z${eM*7aa>DV}-j<szi+H8%b=f$O_F$&)PrQ8J{pP0h1A}Ap;T#iVA)d$2pCJ@@GR=
zmbR=lHJT)M<*6V2Kmp3IrWModU$(vHznx~H6_^McDkOYh0p7t>qH2(23T#7mg-w@&
zr3CaC>jC0N9>fw#mZJ&N+@qQHJyERNqldB;q;F;cDs=KGa)<PgtOUJq$^I`QErT47
z#{|#N4YNcY><Fv~0!GLEXjbkEAxcC^&+FS#)iYC`!<5brK!QgXIa|TC=19;>7v$+?
zyj5Q_cK-)ky)wmfxt9P%(z9@9ciz^vnE;NqEp+flSjXKfFUcyK0X>Rj4CrYL0XS!F
zu)rBu5Ni-<kXx>_;-o`i0=cJF7x<~c8BU*O`kQG>wZqnf?f5u09BbpZdWYM)APE!o
z4JIIagFz-4P{Wpq<k@4`6O<ZB4)3@r@L6$reqO2#@vlRe#X|ItQ^QXo*SFo2TP5$A
z3yFx=wqhX*$bv<JDC{TeFf<IXRV`3Z?aw3Up*3zogyM>5r)CP2Z{B5No*gJ17(v`A
z`6p?WTPGDYZhFusJ%lXc6t>5R;hxkdJO84ErITR^OeyowbEXm#r<1b0U^|~C)-on7
zl8aJIRI!Q_k<MYv6*x;FinebwN{RaQF2ktK8WYZ1Zh$&Rk`{o$RQE&9{@KZ~$?Thw
z1FdyzH=^@jv@+QK$*3|zLX-?Cgro}#N@mkN<sBQ2IUpO1sf#mSugx=l+PZr)yHX3j
zlp=Bm(A>R&z_a5!k^Dh`B@c!Zl@JLLWEDilv=qxCQ7m>UU~Ogv!?rH~aZN^kB)~i@
z9<ijRy5Mh1g||_Q|BwRt4IzvY0Q--PIEpp_e=1xA97eBA&{T|nmVACMM<G|%S#8P|
z#{yQ1@Qb}o0$X6}vp%{((KJ#6tDG5Cgs$WTE<fA8fHkF&NAvTV$@&M@M&Bhw#JhY|
zP{dz9goQ~+3M5E(b2t}>!oztkoxEID+XO$|JS7zQwi%_*kJ@}+@84ctY+jpO@ZZA|
zA)L_nin*NaV?f#drqG>-=B?l<3n!wm1f-=61z*JyuwmX4&z?y`(yReC`aYj+$rWvY
zy|_<u8cefKsTHq4h-&a`vN<T-Il-2A?cHlLU11jTIY3Z$2K6jk_{V@E_9WDDogQQH
zsFFO9@0`8+Q=G2VZYwQhXZ_pLP#BL`23|l|k2m^t4wKl6th8dPAynOy%k8-WA2vC4
zRML;ff>wR?#<(G+hxAmgn_tGbAJjwu0&=dGlKPA^AyQ<Ae@87HsjW`KRUh38N(2mL
z@SNA(vsOpM=b&E(mM6^|N?66vt{y`lS1+1VM2im_WBOGTSLq=g3<bnoD8VtdPt1#W
z-=aznDI-_xxX0jIrCc<(L{>FH$v^I_AUi0>sONTMB8l+6q=$c#tE<n4t5{J0Yw!R9
zMb2{C!ElJb3(^`DH7KD3b^SZHuY9PbFOF7#Z4R#-%$l$Hm{jnwDnW?N!QtdtL$uSp
z4cOA2^D|tF{?Bkc_v^%3s3uFg)U&;FJ$c;*`LGds4a@Mp6USG*=aG{P83AE5$uBzs
zty{aZ(S1kH1(L_f*qX3JuKfT&QC3o(F`V<UqCz>8yg!y9qP7p&@xZOu*j%SEWUyQc
z6ee7lbdddjjHnp@7`Wz?ntmW*q9Hw=D3aRxRJ~LjT!MXDFM?ljF(_5Oc9n+WIi3O-
zv{fcMnF1{QSfTRM;TM$C{T~G?J$Qq?071+Thq>WOr7_W@j8DQ2B<8bJzq1d8>D_g2
zK&dz2sD>tJ#DSq0-rwAovsPsb@#S1~OlJ8pkHmqEAH_&2<=Gf9*B|h3uks1<NHch1
zNu|xFa_=)pYLm+(ApB5+f+j2DwTQ8jC<f!<B|AK}D#});e>*pOchlmr{F^&a_Gi^V
z6>C6mkzJvz_NO}raDr2te+*Z`k#=u|U9`{cn_F+7Dr9`a-GKiG?oMUu7T4lKVBFWr
zop+n};4SeR%U=#8r+YA0Dtl6b4TeG+lRt5MYRL*~W?4v))tbixYHl4&xmgAYQMVNp
zx8rPvJLkD;MPh5ILb>i;KlB@zYPc+gE2XJwNFjdt4Ym^z1FY`@M_<@4dDUyhZG{W_
zBl6mTKKk#{5EP}3N|3L@LpR#7NCA3+waKwveQzYuawmq{rZE)xK;<V0PA&pF*m@o1
z5?p0&IMM9D-c^BHBa2`<W-32qA{>_iUwR8(gbmBny*K;-e_;>^LkMJYkMmP6bz~|V
zj9w7^F{ltwpri5?8K<Wogb3=o0byrQJy;G}gvIZXF(LY2JHuQzPPguF<({V>aA}Ng
zes?&KBq00lFc7d1(c4CE_5weMm;X-U$uRq;{t+<(VLC^GfVyV(4+G_A0QYb@e4Q6j
ziQegyqcVYTh{9l&yw;?}U=B<LD~`%07LWvmS@{E#17z@TSagB}JHiY8ZIsOF2u2?c
zne)vHg0R!uK2R%l^8G?~1w)Ywpj6NcLSNoz<9T|$A;<I^|7=@ZLC2(*Wz)`TGk%Ex
zuV`m^&Y7u=B^9!c`sErzA-iPJbm*F>$7d%EFTh(Z8`;qM^+)7K<mpYRw5iZqVZR<L
z$PEerHzJ)<StUMwM|V_xjt8gx7uC6&6v!nMuG22h&;NhgvK)Wx|3zEwy!~I=5?(x;
zi`)CTh_;PLXJ?NBit7fxTf7-gG9GbVbJh<E3MV4FuX1r*Oh7j(RXxU0d$L1W>#L*V
z*cxiV<5;o+xyi3_jbxg1^?XpR_`=$yP80#oLuGny>_BsCjr9LZWkrd}h2oGj7N9E~
z%0|?7+{|7Sz?KS3i^ttW=NpRlQW!lsn7c?)?ZL)>KYIpapeaxm?-UhR@J|Y=&o!ED
zw5+xwk3lga_Ghkg$C_xttknU;7}e8><BF4If);8tACb8RlDsVLm%0f~(RTxcSOXKL
z!`I~$^lT&ybNRmHO~3FH#%wEE49(EQ-6tucGxGWglAb5t6kjh^EbbJy($dsCN7Y|`
zOVf2@=Kg5kWF)@Go*<XuDE1u+WSRrf<K7_m3oj#9=qr#nI&SLn?Q8{FC@^m>)d^)%
zTXL0wj4AjbgrtWe&rOPw>?4WxcsX;mP5u_@I9OK(LmM8@b2gz3DfFwjPQ0Nuc`>lT
z`2)X##7~q5ElHd2z&h)YFgYc?q%3AkvTXuZ3<kUE!N{z%2)WssbXk7F&7S|Pw9M}W
zRWWi(-J;2&gqyAqE=!<sBmoFt3`~EI+rj=uR7MZ7RM%r0F$hu8XnordOHe~UER=*D
zLpzqD_4`{@kKcEhKi`!9{M$|0h=fE>UWVI@H9x+9I<rZL%B-+)-k?%&rp4nuW^kVE
z_?z_^15?Raa-nl`dGzu+$-&KdjbXMP_dQPhz9XmpCAmdT0A~wO?X!_}o0h$w&^VXf
z@c!l3*7s0+43gpOr=L8`WOo9_&=B)C-0kp0Gdjwm87go)*4s!EGV*_*{rXvb!TC)w
zW%g|bTNOzhA9DUQftuI1Uli{(+Lcz;*oiE8f*fG!Vnhz2<vS_T2Yv^BqwjykWR@(1
zx3-P9rX^|nAA#FpvMkmX(uB7Jq)uX?jLq+ICBY5*#Niu5v3uI+UhiRtZ_o+*htFWo
z26K<;7RLg3Fr-M@_Zy!+k=i2u2XBrn+FCU5b*n5@M(CDbA(hz;1Hn=_*fGGjR#d=c
z1(J`n#`3oxi0f1e4O4jfR5T9!G!>Un#{TjcW?IW458zaRT-G!}6e%PAx$vQTTeQjB
zZRtl@<zUe5Ix>B^)cEA$x_P13FQYOHV9S<!dj6ru>!)YT#QviA+&)}h(fnc%DvM~d
z=48skxQh!@E%In3GX2_&dA(@Aneck;gAD&<FD~Q~gcAlSE_4O&ISZUQKpxnmw7@{0
zTp(B!1I|PGx7HhjUOr^3r7%iPdwX(-lLg<WW$94p7wszs^hDr*d2oYH^}&vhC2Xvr
zjt@4fUtMr`w=YKe0NKHpA*+&oPcS9Lw-T!BAhZ-M(IfKTBvjuTlRwjH7eduf>-t%Y
z(&54#c#T?}AfsT%J^YKa`WqFr&LAD-wT!X66>K)(E1!dfS9C}WiJ)$ods71zi4Do1
zxU`JPpwL-xqaRDfCz6wOS<%^Cp^J@4Fxdn`T0CvxAxV#n8`MFy?$sQcGE>^qI;C|V
zz?xID2Ej>y%kNDb4S{ff8D^2OK65)Aj1&oEmfq<({t8g<AjNp%n6Nebf`PlC+n6v7
zv9-@`E~|ut{bZnsp?F*Ayxn#%li8y+x`4_@Tdm_85;fE~Oued<pRlqGCDub)xi8?%
zo$@e_G!bSr@oZYAXjkbmp(^Fw`puizOHE&-Fa1FdSgFRH)#UtrR3#`Y-kOe|e(RQl
zjTVkgeO`cIee@=OKv_fzNBkZhmM>FTr)k${{}h+0Su!RB?h*J;^m_mCY|(e0F*?qi
znQ}VE=>9hHa!umRai2fHX`pmMAk59~s-zo2SionD{RxWja7=2;`AxUZ(JM^3ukX}f
zKr+f0fFjkRrpE_~zh+T`RslDxs5?qnV?)#!cD((FghJp_N;6zOUsD*sJwH|9&pY%_
z>3{teCZUbH?E1}nG;aV7R?X<us*5xes-8A0ZP$mlAoh0`Okq{6K9touOs~1#tc{~R
z9IsvRW7?os^Ut0s21(*U=uQZ=v|mh)Y$kGTfFoQ3<zX!`xwrw^l#zXns8JMhr;)Ca
z*?TXO;{c=8=G<`G%W;~gY(JCuh}q5Q%syq}F3Xv0WKzKsnYyP6+Ox?o^yNjx_Gcvz
z-5$VM_PkBUAwCm6F~~%VLdjm|`zKl1`Pt@nXWY(gMO?C?_$02!9UV>{%d*N1XKa!#
za5o+VjeABk^Slyr@)~8uNzU3%qJ`Z_Fu6B1G61ROE$^-#VqKQQd$To(^65%iW-nI+
zA5r8IBWEDmI`Pw>0PJ%FYn-Vb7;~@Ss==>nKT@z+jj~z7uTren>?B)kpIN`vzG<92
z;&PY1jcvepZ;sI^dG_8vGx&AI``Nz-7$suerpt~8!;B-`iO-f=MR9ZR?BMN^JRgix
zr}zkAriFK}Y}c?6im1442T0}mwNnM~rwQl!*~iV6<dT5<z7WpX^9c%QF|2kC+C4#b
z>o-SRT%kQ5!tZ9zWj;Ng%sUv9COqA}-N$EC3eQw*yO+JQ)O6`zs*g&vdu?g~*T>zO
zfJ6MlByo`|;Jam+G$i!5)nvPswD~gg#kUIlbS=UKfbNF1+IxO~Jvo>TUkDDLe!sav
zF9)ApmNFpreUGV2PEHdx0L~3zHc~x=I<N2Dl|1+t|D0<~K>7xfLwj_YZT}c|RjQK@
zVDpA*3%AfhG1`TZ+rdp4yq5-K)3v1EPu?I}=Y`kfc!0nN3zI=yT?F%-OOyj5WC!O_
z3qS5oRa}-yncpGK<f7@9nAo=RffqrSw+v!8LVY;BKWw^wCc{HXUT~3UA()Et2>tAh
z(R(gs1Ro3mh5xEwkxNeYF)UfOul`%3>$ozdu!$jUDQw`E!3g&Nk>mzs%*qi5*O%F<
zJ}sJVZ0yu(xP~DPbet&z&eNt&9UB(xs(aQ&7&FqIF1?mT19Q@}c?pT%<_Z!5B_p+X
z@MtZICh%w_*j2amnDJ=q*i|DfizuQRBEAM@b<q?s>TWjEXtY9W!H>d4I0%S5Hz0TG
z!p_`A8wQEW`>c8-%YhTjahZcS!u4|>yQcF27ycaz{S}t3#Yqi|ZyOpzM4VbNu)~Hy
zi-!I~dI3g3i<_kop4a}+4Mz_+0!gr}wYcL*me@{gShj!g21UuNMpzkesAm!(&xap>
z_<jD=O=-ldcC**?gUGC=67eveUo=fX#;>cik(eq4<sn)@)&cTzi9@VNkf~;-bJtDo
zZmCCvg#hS=D98A*qObb5)?I63wUh)#2>j{-3u4M8_f<#C*Q%*IT?uFd*%wN6$b!r`
zd%?D6d=pw)mH_*YrCyX4rhk53yGK_!*DDUL%N1z@-eio|60vbv`yBSI&FNbvSm&E_
zpnT0Zz8J)Mm;jURp~jTR&}STm4oeTQzFu@PJ5s(WrPc1_)N&|xZD|wJ)z`6_DniSL
zQ;Cvn*S<aOa<0_0Fd^i6%wG3K?e};1S74dM8HN2=rBmWL54T>6D}v+AEUq6+)Tlm8
zcK5Y#S-FP{qN#8@C!Gc8^+JmB2>$U*M5HJvFtgyGK<D}|L0heo5(#v=Y7&E`sK|U=
zP0|U7WG00}C0muyg@Q=(mqi7rrUNtF6*S6iZx30~)wNJeVPkyDi;PV!xsv#{VuNs;
z*N_5jFrC&IMZ-=vI)x8nT&;-(RVvpOkY$mq&s-(Y&9V4oKFqk&-~T*ktF7iE=?QXj
z!|{0?fzXA3B{62E@Cq~hV#wB_!|$bs@qg)`@0#vwd+)v<kb<+Lm~$JN>Dw%CbZNx)
zWMxnA@Oer%N^vR+fBduA&t3E?T<WOS7cr>J<b3W_Ilm@$&$e!#xBZcL3pgFB+btqL
zOFYJE-nThPF=lpi>0DPksE63?Mhiv?bQDgB0%MqQ!{1DO22v=rM}jG`4thf{w58<*
zFFi2*LF5HTq5M}Uaw(()7bi#+q$TNW`Nt>}?guc2{RtLtV1iaFHTtA&bwWq6q2QQs
ze+u>wkd6IJKFz<#KU11;M$<CS!v#NajF5n$fmg<=92OwMF|PPTBFv+zVb_|^5c}u9
z2Y?QWrFke5G6WQWo};Q@br`5s8+!LrYd0Xya0+|+H}(3hKt0TOOs#EfE9Uw1NP|U!
zV?bZ+5k>Ks$VsXA7l26g=0t(1v5C@;D5G42ko2$?<PGQwK_;zSR0=LNU|il~#a&!L
z;jf&KL9J+&5p%0+LNZYxTkMsUZSFFz09w0&)`#jb25$Csw&CX~aZP@oAN0osAgctY
zWQBr~rwF)}t9-IN0z%K{e+R)IFg77gd~EL^ge^Y~lDY9EJ}lk360e}U{f-~yLlL80
z@26SN&1ax^*U0~(1@d`%ZITPSfuPtDrg58!8_g<)ZeN!D;KLDCu2wF)h;WKJ19r*r
zBi$+-nwRW#vcCk-ugNZ5#68DMYj1=ntXri^U5Vr4>lK-bMsxZEle)pw8O<>9$lwr@
z>jR|Vlj|`qqluoGANWvqG6stp-XmTpEfSD7`Pow7D_s6e8C6G{-z|0S-$`Pebb|@;
znqcB{L16HjK)s`WF@s8b0`<S&fwE}19(J3olX^lyChXh$J{j9*A6OWNRGDc|3A$({
z_$RDolyEDacxd>4XQ;VlzbzF0y7IPH#&yvg;l&`j+AD=(!E6wFu9$ZT>ZrAAC-I7X
zOWaBsi$@~qm6#h%tn2l1stOXFo1M4#7i=}^YI9#FC=`!(f;eqRYj~l(0MM_QntIW4
zU_u`fnTgMcNufc#*{N*@(55k0UBe?($lj~dkL#<i*I6KJbf#~Zu_aXqDgPW`f$?Qs
z$dNXAQSCjsia%T54v4?tKm0?Z4U(qiRy(4}0Y&yvL_-FNvlbT<{;5^Wt1V0FVR1Iu
zl=khO$yxW(=TKM0VW~V@5+LOKecKqIpVv{8t*O1>fahN1I!_t{a9;7c|GIi9>Lbg)
z%dWxNJx9zAqACl)4%-VYy(c*1+Fi46elZJ5M`}gLH#O&FXFOb&zha5Hw?FmI*8MtM
z$YUzh=$??6RAQW3(siQ^R;=dEP^^&zjvu%rw#sCN01HH)eE^9hK>ymFG@gAd%L7m+
z`N!KaA~WBbXPg%0(5J%RFmM=NRDAF@md&B$Mfabym%?<0_^dm%CJU8!NlA-aa!x=t
zyOU9v9%h+cti&1>cQ9X8{~k<P#v|zDd#VPOyT82!*+4reKq?@>1ab|<=qj2JZBv0g
z3_CA4JbE|UNx}{Tqd9p82(eJA3gc+q6XdqL4yT!y7ZRpAr2METL0BA?mE)s&#P_$c
z)3?+m9H)8cQ<aB$EZ;Y?+?l}%N5B84i?jUzfs-R#pbLA{yI2WJO|}Yi6~m3gF__!^
zea6rK@4P&J9Xa+NbJOZB9gnXnT~2CbLev4ng+iR`FgOUnXwa0uitO9(RZHY`aeAvJ
z=-to)4XbsXGlOZ#Eu}3oZR_2*dN?ZayiB~;D1SDpV7*$M{&A!osFN<F&jXuQOL(3y
zXQ$!0qZr|B3c>ALLz<u{^|Q_SQSSW`F^@WH?sf7}Ef&zs-W3DGd_BO~=!$oZGkshY
z75Xhs!bt%T{8<)#Q3GrIhpRzwAHf^VEw)INQO*ViP1uq|8|(A*dy@jPApHbtw{Oi=
z9(Di+rcvhn<BSfGT4E!z^c1m4EW3`OOX;$IxigW}*SKgr?eEsa4wHy($0^UQKxRLu
zl&uxGO{RVbKCV#~m0*TmhodzKKMD~ee~XV_5qJ|&GLp8wYASxf6^t-L=^w-OXZyEa
znOg^g&EfGfCJLtsU@kBrWfE>{gi7jWHWzvh#iS-fc)cuU{ijdFob{5NW|0w@rbJ3w
zM)Wl}Pb3m63?$Gd{%`YQzzK}haCrGf6^GaD8V?HE;ryOTT?<G=NEBOq8XzUy*Ny=$
zJz)o&a+0#toAR)Sl~-t6_KncS-?uTe4G$uq8`HVby*EEMZ$nM_U|!+hcgplh4{(rV
z`WfcY&#Zxi1W#-!*nVx&-^RP580IwrAXjK)B!Vy_bVFK??N$d5<6ckC0t(g>qL(Fk
zSIO2EQVVqpsTd836HND6Ylxz5t|$<wtU&+@D2$TrlS;6x*#P@+{}_RC!H+Afi_@7S
z(^v`Db_XqT(TH(+@x^=v+yFb6b%<pGy2Yr3BI@Qg$kBpZ?d`y=a<X=<ovkFJd-ZJI
ziBNFb{e-{8$fTI32pSd84d!nF+25~x)$?eGKtVk^hyqCbdl+cP9rFr$`PiakPEi3o
zM)u(ByZ`QzRBj7NHvAD$-dHD6S$T{iB$W_g%arpx)Z~k!`^o47Lm2oog#+J-2Z;~5
z)!)0r6RpqQ5uPVL@#9Vuc4=~>Fc_Q*O~p>MBJZa3o_-y0Aw@Y?Y1J$~LSxlQ^_1JM
zu|<${XS<z|uz@E`^xlltxb~8sP65Khwp;@|r=5E3R<7lfOW7#J5N$EH#NFR({^Gkk
z+^@v`^Q+WGTYg@%U4F*W+caqI)2)+6hU%KK4ttyJXhZwaG>tEO-)`PHo5QKv-MM$w
zP*F8LB+6vHP~!bt=~$w8%_3ux>Van?{vngQpGiOSA<KgrTp&gq{A>4~h5#<Qd?M&;
zH|_codIWwDq6q$9#VdY4<F_Zx2|5sFt(Ndmxhk_reKo$5P7uso^Q~ovv%d2-TN>9T
z{2Q=MOQ&@H*{2#5O!#KX1jk0m{bF&lR>5tUcIa@8y+8`rh|A^qEcY;g&Gx!0){mBp
z4Pb|2<C_c<@a?^C?;wicd;l5(oX+TBRpn3U-XnvS1S31Pa!64t<RICXkCzh})h-Ki
zTam2WY;%hJ)RTNF=JJA{k3G+m_qn5!gU0cPHutgKn}~m{zJG36Y8Yl^u2W$ts_0}1
zdF2n&pwj-Kkv_TQo!?3&xcGe@)r1D6$|;3>mm8E+@UQ}5XZpFDV+;I|NLje80B0>a
zFOpWeIXwF}u$2=s`o`x(5ZP#v;&IZkazMANk>n`zg-bmUh4$%l)MT>JR$0-}_fOX7
z+<#Q>mNYT0aV^7Zt`P!@{xo5!5bah!-ff)3_PbSdN47{l<#v-NVkw?Sej?@B(0;d7
zXQGIQMu$CFl_j-7<OJH0d_JQbFtBMCy*gK|wi>$50rmAN^K<8y-qk#8ZV%6oT|P^C
zuy=7Jx-El$7xj-5o|fvj)KW2!Sa*!}W{>0%y1=ujdNvG6q`vD9Nz3mh*NG2a1EDnR
zizDH>j6Zv9Ke}!a?9RG;MAotFbcI1QA<4h4zA(W&4$I%S$$>@-yz*GgJ(xIHeyX%<
zy<OD8Hf-2*CdwYeN?YmPfkuqlZ~-u8r}wVx)wqFsai<Gh*M8^a<rr&t)n$DwP?zOb
zNGy2+2Wz$@BNj?l-x#F#*zceY>x!AD0_JsF41~2VQL|G2M2;_*KyY`FfVAXa(Q#W#
z@qa4%5qK0}zSN5csZ7Sgimb4lF_A^+;`PBQR!oH%mQF^|*6075TJpXtyB9|zv5YT9
z5On5|ZU!QBjeWYcPctd!4#jCNu)=S?1>KI;n(NYS|L;Sw26Z<6Y*HVxOyBun?SVen
zz8$<EY3YclT1qBoYlTJ)9+~nasPm#NNBJdp1GNZ1?)hDkHI#CnGQ%yL?i;rli7hyc
zwUC4#Y1db}N(NcUCH7N(wvJz{7$u3Z2sNjD^(SA}0f{&WnNO%ZsKv+*0xs-nmOn0Q
zsjz!TWXsAbOfU`zX$qSunNVBw3=CU-p!rA(b{#t0p7@UG^iS9qJqOt60w(4cbteQH
zGPXzno|SwJQ|inWcVlZQH*N5q4%jVyS;1K!&@3I&`>ZHgiyVAf-_t?hhAGTp*&#sy
z$Wb6rHV#${N!sW==kEMJ$7p{Gnkh^={<-xnMvKE#Z@*F%MF*i}yN_-ylIRaEKGX~?
zFY61tto!Q$qAbo6*#QcZ*Hne{vzv4|OZ^_8N5S$i-QBB!p!0<SM`;a_9rR7;3xtG7
z1JSCz-sC<@WAUopS=Y0O^l@sZG9L-gprRJ+W^PSqeeHGo=Erg0SJO}2xH(I7>Y!gX
z#lA=FRb(bj+W)x|hv50+5&TUFdk5TCV;)PWf3pLjE^Np}(*%47^>y%F!Grhon#>BY
za-F%*K}63mPRtSc+5ayzG~6fp)o+Lxr4ObA{d2bzj^%~TJzjm<kyTvHH5T)TMjL~J
z#<~EVBT$Cezls<N>#|JV9gcR=<4C?^q}ge+^><y6#oP0<HAg=j!>$q<1*>T(Qm86A
z0#@yJ4E(Y@szXW~;KR2tm7Ab1g3|>+(9J^FNqw8L_`$8|a0?&lZn>lTF??750e!O#
zXUJmJFmA`d>izM28TUKCa}CBTMEYBH$R@eEX5WlUs2>@g10z~(4(=>!Y-~6fRLCPF
z0^AiMPUVw7F>_u9Pw;U1YkWA4M?Qp+1O^s3j6IY4Hy#ncRk-dJ{hHjQFVhdeSH5w9
zLKIwcH9+{UNa%G59$O2TK!QYUH*>G&i;yhcaX?2rR(A3e=HoKL=<m-xz@_q`M_)`_
zF6!h`9Q9EO8llDAgz?;eu?e{E&>GJkPAM3-SLn0Oo}$9@(wXAWIPMCEUsyXp=W%#b
zvE!mMX5~D^)$<eecw4O{GO`5RDtNQ~H_hM(*P|!;xD-McsFK0;uULL~x}H}^7DGX*
zXTfHDC-eE4DHPBB!Q8YP6?zwukh6Bh76mlIw5x5t8%r%}(PdM*M<?X(r@0+oJLcUW
zyYaBaAQ3h`j#0KxA_*YzsM}@Az4seeD~qNg#h?iH<Y3wufhwvmm2v}1_cPxOv>fW@
zguOV>bXeidhv_3V4@xhe-D{AsCIt73X@|^Et=2`>40lSAni&QGRC5?MWb;maO<dN)
zS#Z>>2Ju_ey1Fd9J!x%u1c~+mni#>qul>@hi&$*RtjbU&E_}?`;7dJ6&h2Iq0fX`}
z=Vrr3ScZFs1;#wmk{198_=wEaeVJ=3X+3G_K$jKQl+iXLSgn;ije}3F5IIg<ql8Kx
zR}Fm^aeTXC`)V7-y<~~XIr~}BD%yKAy4*{=v6H=xf)`gi_;E&+V&P`Fe!~CF*g%^O
zd0X-8jjvU%Q+{kTTW23BoFm!Jqssl3Y1B~in}0>ghts_V@+?u<Ra9kCvJ5J^xi$i#
zCN+P7?>y3_;^A6G_-&Cqy6XjIeacCreDGpT*3Ld{{Q8$ZV7?Mw{KKm|s#jD+vd&xC
zP}&rm3`(~zPyI`RfKn{%e{_tFh+6nWjfFN0N|>o=IsNMM!@3??_nk|<mhWP4w9lr}
zSnApi)#ug)7<3C{w3Q|R{}}@LVz4uE55X$GA3vK7-IAk|sGk0urm@!>tbu(6mSkB;
zBip<cjIoi72&vidS|77dlpB7j6)Ly`NE2lG|1Kbr+yaFVb}1);I=xX6ES|$1fZY>M
z3t}oKWK=|D2v0F<2u^c>!Ge(Uw^R}LrWA)SFr(A~<jdI@)_;$YX9-|iB=s}bdyg2>
z&?Ck7*7Jo!V8f<)h>^gMl)ht}AWXFW%Pmky?bgM{=J}Z3UZo&0&m6xkurWd3i+vzo
znBzITz%k`Pk#`OpZBXBoVeu0iD{}1EbSaR)n^G7GDLT0M_57yG3;zDdZoP(5$3xPc
z2<`_9;AKx{D+c#_*}JfPyc(QpC4VFHF#8p3X<6t?nYR;}EPz4jTxck^5ce*kIBJCI
z4Y4CK_0`L$(6eap6L`Eoio+_~N^sjJevD~XEF=o+m!e^Ir({m@c!G(k`%wc{m;)0G
zi_-GQv``t&`c$WMv;EWbgLA%V<QI_*&Q;?cFzeLS1d_2Pk*~)99bo?928ws3ies6W
zCUE;8MKsf5f?@{>uNYD}bQb{5Hmb<&yH#Ui4yrPqo1>H$X#QGl*|l-KF;jmb;d%tV
z%BHOE*2e$S8~aeCIm}w9wC&IMSgdA;Lh40<uZu#<<_hsSTx~11ykM!v!+{?NNdt=u
zAX=6~3b)vB0bS~?!%v5`-+jRb820YL?SCHFSS}d0+kEU<x6_+jy9KSuNxI@s6m=n)
zGcg-dMr#!Ajh9BtYJ8V?GWKGTCEB{ClQ5CnYt#NyxmR_{6@UiY4~-)wN^+q_Fls!7
zlksf)P9SQcSQkg=a#7-IAxYa474}yPSoEr>$P`pxZrIk43-4xrYwdnREZ!g$L?QME
zaaaBgAAQ&ShKx)LEhd?b%5l68geUoOIvMjE5F(QfDMeyXshWykN@0+~pn)cGw~YT6
zeS7bJsbyI6eLUHJ|7i;2i+X!3Af{1R^R~GCwbw2KyrqAd0{fg}aR{ht$j?9w+-FL{
z*}?N70x9`_ccH%rYy<6jPYG^vw9lRPgi;8?;VhXr{>)HwFekCx;X+7^iZ{d5f$C8d
zxqwvsMtibe;y#w0X_TIuRTh>a{4wzs*X@+&q1TLePVAXhCCOB+!gIY1P8_+lz!uMf
z^Ha9<)p&1$V|tI}r}Om;3y`e3d+??8<h_5;W|nkBdq{e8p#9_(%ZvN%{%Z1W#Oz^6
zKJWRZZCM~Fo};)^;<0y;A0`o4<ZR6rehHNr2#a{C?YyQ1@bSQ`wx~StLo6jjQ-4N%
z)a(gm_Wcag{$sast9m%}>{K0({aX91;W1r<-<)D}2Us(_aY@>h0M-@kGqo_eLyPq%
zY}90cl`~$HUy@!Z4`J5}_`frsn=|VhIFP~5;&h>M<sZ0snn`lC`-P&rckEZ)y<X+u
zskbq6PW$}spqZAMEolcMBxCTEP_f*%>%YD`hA73sQ$F^vbysX#M9w>2g@m}c_BcQ|
z4Ehd6BuyX2Jy$RSfCfLqYwiW^Se1lup7Z>&2JIxO0W;5oy|Ru{DoKLIVIZF)LQ|M#
zaP&j99`4LpG(F^I_4OnHZ>%;Cf5R_Qd=);MH2LEzr+<P4NS}*%1YUgHb4c*58~=90
z%XvTNB@tJGH0+XUf`tm@;ue8Vi_f!(?~A<&hyJfU-_y(c07Bzx@J}v4<Ewq|puziH
z^QlsEIhakT!IP#(!`O2}o9h*o#0ujl(IYwrQO}nN$&n(HcFI3~(*}!IZ^2K7z2(-D
z6YFx%d~3WnNDqZdIDA)JfG+loe?%CrwDS+))vxxdPueVX7an^C-|An^jnh)C&igL+
zccpKwSvJXWfU^|!vrg9owok@w-wKqLN|k-NV9c$QTqchy=6G#PSCU5bZ2s(31>9|O
zl7qS6v60+$r_J{W=o=tXivQ<#&}nO3+03iye2+;x7Af4maY2Jh_jku5#5%V5N&jtI
zMhuz9=d}~3xtThX@JWwTUGk?<;`k1jwN9_Nefw!TU`sH|Z==EXI1!f|jOoezmKx1K
zDffwYL-$>G=Z}4lON3Ct=|TFfdjjxVD<i*5AUDvMNn4?Q<A<}p9?Ha~D~!P|15V0G
z=fNlT)uiLh#TDag8<}eOQpAC{0ir_liEpD%qaH7lY^Q?@p^|f}i}EG$_RNgzevfaG
zkLQj8;77*6l74b@MZaO7js2Z8a;Owi@V<vcV}6F+#6CSBE0J#+`-T|VP?}zTY2f2o
zC11~4Jm4KM-;f8P9_IcT63baaQQZ>@hrs9NX7Qt}h68V@6aH~)lC&k2o;Z17pHlyn
z@($3oNJN?w3b>p0LU+7@4DJyW?ZH0|U^UDIKzHB+(1V2wNF9sIHY;TYY2|jFnj<=E
z0+Kj<30L(3>Rr!LxukDta5vUV?GU7r_X9Q^6&gJW86wClnIzEQ(O&O~)gQ-&HpJY-
zC)hkRTYXB5`^Z&X_YiO~A|6`OE$#*#rn7v8*knYW!^_u^7BQhnnWcZ0y?@6+IYrq7
z9J>{EfueDk15zxnqPz>~oC`~XEdwr{Y`$cYE9!=B%xW`bfk{G*3gb~#Wuc-6JB?Z0
ziCqM++CNWEG8Bme?5h896#bvEms(kmzL*?e%1%g*d1I+$X_|yU%DFc>xsd(+Kc(PF
z(((~c<AUMkkpq8ZiR03K(<$Gw<*Wd6ePdX&^%%{KgEHUvjM;F_H1uW}8}A{Uj4))J
zGn5L#1HL7pn4yjpn2wR1P6)C;Q;?|Ej#8^0T{S=749<<#p~OXfi3<oxEjs1)<dmlh
zvSJ|q`nll@CVn&DMpq-~zhUDt1Pu%>__01SghVbpVew-9T2N8a2a6LVRG<K0bcvUX
zCYk+98T=r)b@zT_8zU#Y#$OPejgf&XJC3Yatr<x?66V@rds`YAhh-MPU^x>^a4U4#
zwwI7In_!euIF!-Ez3Fy9Z}-#xmIIv<G}*n+8Ro<kf{L%r>y`2`>|4KCnzU|xd9^q`
zgqOapfW{x;W_!x-uCl%KjNJg3Ec-WpM+*95qhz`~=v%S$m4g2bPczeKzhWn<&d8q9
z*w268>5d3Qgv`7ML{y}sb0pwuR8*1*t{><_lt!$6;!W1SZtmc;S#++hOD}wdowW;i
zB}b>i`>fR8)~_-tgeEI+)G5aX1D}y`86W<D%oNHR@`Iw<DwgSB56A{q&N`5*Ev!Uf
zi*%$LOR8e!V^AMNcaPi6KqdDnzC#dKM15O;JM0kG&g5uw9;wFyMixcRL-m>}nhlTz
ztrzo=qzH-l-U^X`8YF<IoJ@dE_&w$Y|K9!Lo~ztOeYv#yQS*nIf(v{_Wv|D;Q-QKg
zw}H+0zaC=Enft;U&nf_g`0Xk+P!Y>cv&kkoRd)V*STLTR94XAabIJNf_0m)+{Loc2
zIjs}I;qWn})f^vtOs>JlUTB1pDP)NcGYNs$)0edF=%Aztn{&c@q~|>W`ZZU&)5#C;
zT($i%1cHXsxg|i6xJqry$R+2OS-)`$*|MV@))jMTaO1$ciw!KPVNf?ZlL8qi694S;
z{pm!=rcg9Wjd{b<vK~bc<As>Am%vNHhVejM*4q>icP`@yDqt{S1Xcg|V>mR)v-29k
z%ki_7UDj^kt4OTExJzDE7OyE<ixZlqX_OZKUWt;q*KcZeoO3a+s-3gWPwS?UX;95P
zFNq3Psqu1MmIg|?eaAnaNSRE`Mnve|1J1Y_@SDUBE>{qv9X_dSp%G{*n+PcoXyox?
z<gZB;8=Mj;z6@3DN)SYi=8+M8YP`xFsp}C78rmbxLr*btxT#N8tG^IPJFJ8rzRY@V
zLgo_C>y6=!-kv0>)ya^{ghof_g!4ZCaqe|ajFh*!IRxOf$J_?*pHm@McngyhzW^vF
z0cnxRZuEM~$KobhyupULma7r=7VHa;S-9zAWQAO&_-{*5@UgS(YnPdvR!ySxs|DJ7
zM}r4?*AeY$qm)KwYxfi?;t1!qq|0qf3OP>#R$sycJ+7<28C7zI<5SB}R!NfQkLIt%
zkS=Lt>OcxUTDcJRT10nZ@>JBiy6_^)L~;#j_xbic6a44S8fY@Vm+UXO-I?_{T(I8b
zZxB`z>Moe$T;<P6g+mYwej8$X8)ID(@%--)#kJ+=brn`=&I=*cRT(%Q;@)4T@6_fh
zles~rw~yHtLq9)S-EWt(9?^S#;_ybaAOBmP9s$%Tm?VcdAFgq5lOnnO(Cdm+AizzW
z5)E*0yj?O2X60~j`g>Lo2w9j@K*&PxK!4e2<6!r<Xra0O9^x?h|A0)<X}<Al%*LUl
zzjCpguj0~L*@bv{?6WY63VAfMHQ&nT5mT_eT=!xEqZ1ZKWR*ML6JP;qKm3y1kRcUP
zK(Au7iWEbP&A-67+qv~tkM{WL657ps3xFMrZgxs!Vp`DIeHw$a^aRD=rhS%tEUo%G
zLV#`OCIz+vpAMUD6*NH+YfS!s=z6E<%AW9Bv}1PAaXLnKY}>YN+fG(&8y(x}*tTuk
zW=AK#|GDRmdmiqnmo@g-yB_wc_0|05oNE><;n%SC%>XC<tAnnD+8=n)VMH-|w8d+W
z<WaGP3sz5QWaT)O`zlcZ<zvs~3q&AB>ziHKK8IO_T9I`RzRg{qLo81phDuUnhmPKd
zV<5d{m}6;2Ma}kG60-rX_wfL)mS096rj|JM*V(U$jQ-8BGM9hX*~70a<)ki(hiQcx
z>i((Cfd|f>6<o54-x?K=zE#=@EB{R|*WJ%UQeKO!fs-fj-ikh<_g`p~0vOB|^AVyd
zUJh7>AsEQi13C7v)$&>H@!9|LEvYdMSd7MMDA{9bIiCH>?MN+|wOc#%N~2vCOW1<m
z11JX6dX?{d3Zr<lvMo@@3bz?rz4LE+JLm7da9bzaW*~+Ii-2FpLP5JFwUo0B(Y-u6
z2S=5PL~V|BQ%5SRc&Z(cSr#KYu7EaQvbGm##&qVAseQZ<zIh4BMD5Xb)5GS`rVu9y
zoogfyyP>T#dHhk;5On;J_evSyFYt-+zNO6Y{GU{IKpB_4mtu{c8VGA{yGNp%b);F1
zj#NwI+~N^%T0OT4Z!gz1*J(@Q_%Igu!PjkcCt#qem@Oyvh6=ctOz~c(>o&G8>Twz_
z<W6xT6ZHHf=A&2aeYF0P>5QNOqNw!78te6dY8a>~*gxk{>FQ5Tj^J=HwMgve6^`)#
zWk)DD%!-0p4jsKDD88J|_oX%bp*$?pELBYLBK;O4+kc3dQ!~$B>EJds5bf<Ip5~&)
zifT`j{8`3<2*|#*S{y$H7;;5gyI}rGgn&v4KPnVEoVN0`MW6KMy8nI#A``&n?K5f`
zgZ@MS0=J3MN|Skvd_Uf(nt`3Al?c2O(u|VenJJ99Sl{|WKfYznkG|ayv0ryt|CioA
zUD#O0Wrx`ReHX0wikW2gL_5-9b3_5~1UrT)zW2EU-V^t-l}(>x`35=<)hhP?w>Z|T
zKJlM8_Wz@e2|VE8bRMZL-|P9n3Ne0d)P28v7|%EZSMzlJ%)pMbBC<C1IOnm}c3<at
zyf!GYW`B#cPQy)ef=+z?a0N4=VC6}N8*;JV-O=l72SWZsR0PoVU$uv_#%t%fSzefA
zI1JpZMhXi4H}X*Q=9l2KMd#y5A@Wyk9HR;C-|ArP=5TF_L*c4D39d2=+TZ^qvcbqf
z$kkpV0V$2J2qD2$i9!TOFh#~c0&;^oc;n_4DTXpi^8aX2ev$H6oaXrvx0Sxmi6DUT
z`w#$;C8Ha9`K;9nOT{1Vm0tD^RJadEjZSs*i0k<5Dj?8yL$dJe$#$D@=v8BNe-us?
z(Um-0IuXAA^;eUR{Cg=2m4UC-#KSTAvMw*!S?8rWPyDt04^e0x<oy&ftq^7Eqibo=
zpLlxIfWjstHcCOBZ(Q)Z#Q2yp|JwRy{J#O}&Cir0ju^9dzbMh3f|$S@z;I}|#y&u3
zVnaGJPRd%{Bgn%hOdoZyM=2qp#xc3Sxjql=?}#?q-eih(G>0VX?hNh1nz|QxC303}
z(OgHm%4cN`Bdz^qlU!AP_3t&eVOz#^DpSfLoe)T5&_uc)AU(DWHme_F`cN1a>GS{|
zw(MFhCv-e#|F$y)46`Et6p<7ry7jrYLmIT7rf}~QMaeWIahY8-oA{MY^v|pe@dfvV
zJ69)^(TMo*+Nh+1waj?>JhE=(WVS?(yXvUeagsiu##i4K*azti-k%TEz23|YrRsUU
z<N8DhN*!ZnNh(`aX3e(-tyH7Tw7me}2U&c7&93^fWHVW+wbk13|H8pVi)l(yUWY6d
zh>Cu3*8Zw1b7ewj2G{|PP&Q&r`egS+UghM!GN+-hqAA#0l4ZT76h{?-HT8p{Bvh(>
z6rtts-rb+EeQ1@?%w3uf`D>JjRrDiScK=ol72&7e12i}aCO5A);er$zNE88?>HpXw
zYF8sWO;Sisp_kG2Cot`Mll0#XrpE-C(Wif&RZ{<La9t>THrqMLuS+p9Zd;l^bL}P@
z|EGjLyecYr3U26;M{Cx{QQ!U<*1Fl>-4oRu=R8t2{J3*c8ZvO7VbdQC*Y0z}kw3ec
z%uVjQQu5SAh!zrt8@{K70uXTp$<btFWF3ce#)2Rw4!QHTlVhWi;XXlS_t;HS(3wNZ
z1e&w_2)WC0Y1fZ<Ed+CVJmqZZL)nZ*OHvuO$Z#@YEj<;b)YSsWC0qE4#ywxptL9u*
z|1UE_)7JL5@{!@zY;4aK{O?s<N!-F&BZ#plrM0i1aLOHB9ut-m&B0K)+oROe=u<gS
zZ-X~#$wKXkiy_c0v64JY6~xs-Hpr8|on>`+@`ID6`CCm}s8}|hn)^_wSaIQz7A;OX
zi5^>x>`O)zFX)kQwp(8%_$Nq3Sf?B-pi5|JtgRJ9hMkWNz3<0g_RIC(k!cx(gTy75
z?r=7cPZvk`@1Xoim}<)LO1MSKW<P&>v(kj_x8g&fVy(#Nhj#a)4FZTuryRPoZRmMu
z?Dr79UbzlFnlSo~X=Et*KTN1?(P(GJ4>T`C9*RUI$n*R;k4?SGE5Bc6Z-S(hi>6tX
z4Of1L5*TqW(+X%iXuS%S#0JC#9H>ZcGZ={i<5<`obab9sedd?`K>5C1aDnUbFxS4u
zzD8KH3Kl2G%|k(LW&EhIBdL+L(FVa{($9O-mUHtKRE`<co+uA-1f25LJ{akSib}Z&
z=Z{S}*^y9Ds3#Nt58)5(V6T=eUfW^RsS~Uz4;4PYJ+2h$eB@v@S)xfl^iaos_gQ;@
zhs9J*M<|wf7zNrv5*OaI&>f_Z9h5takM((LdIxiRw5kN!4!pVIve<%CJ?gt2S6q$S
zD6|u}2dLjBvrg?~IXQxfWom&)>~G|KCnF0B-m|9H$$b52F<GYBlL*+3SL9>p`#{hQ
z7mt;ciUC`&IS7<o=zDOmcQn6fK3NW68tpNjnXo8z3>|3~`i4YceY#c~%tY?{MX?f%
zrob7FUzSkLQSI;l0+|`&qDb$cFGvJlNy}a9@=|%uN=AvB6rzz%Lhr$e-kp(8BG^)J
z=1D>#BO-tHH_sxy!(%d*(|d2j|DA*fq2|r8E;8zqqbkMAJepabz1>%b$!Y`mL5SOY
z<>4ivB3$f?67L-#+fE6VIuk4pAm0k~W@!0D@1DN@NpTcCxEmqRnwR;zD6^6ypH%0R
zm7*a&r!nT@(7k#<NR>WW5*QGau>N-T*jAUy*m&Q)Snfv#jjFBFm-~=IPy7uuCoU2O
z%5RMrkzw}JofXL^_)s3f`GEznFjS0bvMIOv=jT_IE%rZ+{TMqTer$hS+%hiFpg9tk
zx77c%%ll#E(fY_AUC0Oc?)++R70tBFw*$Qc$^Gx>H9xF=RN*9JI9_L@@5B-__=;z^
zK$y&y7cm541&C}J28{PAN1VpgT_G$C98A^nbt(;~#OxE^sGA7_hH4Wt_8(4`$!9Do
zdzgRfB?Z$8+&_?}u#6gP)QHRCfVaJ5K}=!$T4pENi>EY*C=)z(QY$20pK8?%xrmqa
zjEB%%6w~2Q!&JKzBrB~XPRJ8Q;RU@=G89(1OpEG9icUD6?4OhpLDF=)^Q0_}I3Fm;
z$#y@H==SE;htTDL;NS_TUU>$@8Z5hsvk2!{dec_|t%5+wCA(l69X28@=Y1(3S88Iy
zl=U3suA)<|60@>*ALR!3W)k91q-I_Q{gTzr_AZ-x4p;oZen`WxuzQ*&Y^liUwm^_X
zG~E(AL6K)G@*;;MWASqjNC=gpB~SsXS(In1k9H&n4u^lhDccuEAsa4w5Mdpav;w(q
zY8w43A|U`qElw31iZ&E(kLGhML2qjwZZzIR7d~zRsmweJjQg*LsZARQ^+Uxfm7vnn
zjLtuW>z-^-C3XK9T6P)+J~3RFU$Xs7q3kdhoNxA2mz~(e28>Z*_K#PM^3ynRP<ZRZ
z(CIN+ghj>xS1C6u)9FzFpFzMK$PW*pmkccv#j86KV{C*u+@y$4RRj72`5{E_f={UW
z0J4OgQ~uvpMjX|bu9G3NJpbO+6%H`{MIXW)2x=H!sdLgo4I-s1*oi-=@xRaJ?sQ}a
zH*vjjG4i$4YT*A4P}xV6bjD<7`?dqZuNv{ru0q`iF!5;II-6|ORK@d8n5Lu>i#$|p
zpfCk3?JxCB^<RpQCks*tIB&2!j%7_y{$nrd&|6K+b)?RsLvO%rq7zy7Fc}k*CPI4+
zK2X)}mV=kPi0;wdUQGR+@QI^nQ)7P4XX0%kRrc*EN&SFSCuP;fe^Zp`zY?CRkWES*
zfJp=x8elh8y5yqnqZz0UG6t4jre=IR6|XvuOl+mCg$YVs<S)gNRQCk@;~8cOBfc|*
z;B`BiwWL)-O2z)+sH#W3(vLihB$AXXsa$3H=c~tyCzcaN^=Rc3pVxM=J|MInvm3^Y
zgcmcB7Np&WtIrvB3C;~kLm9{lr)ObCHhTbs6=BcGC<O<LMk?z2!5dT6jtqv`N)y55
zNyUEm$K>eDuHRSDFM1|^I<L__WDOr0#XVDvOD5qKkSmA0C7P=dZN}1-84usd9Xxgm
z-9ksoj6+7RteC;M8x;vEz?legWH6<(>g3#gFCX_j&}}XnF?KA3A2F64D?BhBiwXj2
z5{F{L&wu6c+1IQ3`4u4JNcnv0H2rl2q0cT2G_Nkz{pmCwj%+oc4QILJcRR(4-!<?o
zYWcBr3obK^CLsV1os}9#TwM4`)KK(Whk*04SmkzQ&hWM?lslV=v%miN6&mj6A$P(2
z^?b=i{r~NLmeM;O15_*2HFMK7za`KG<S~rsEnlvX@M7Cc;iPME4TyXk1=Mb{0ZtxT
z%OzAb^F6Vj|J1!K^S}d|)Mxd`@5RDz#*>h=s@*R4Q1%Adk4g&MzjOxs=Z+on!<g_F
z1$iQ55WJ)^O2OkTx;%=V2pqvooMEs8jnaEj*|Hp_WkGWSpql=hnC8UlNKPDWd7OfM
zapE|$ZE)d72JeohF@qQxlhgDxzo00OpsCfr@cD_vk*j+*t1b-CwW+kw4;b{Hd@K;X
z#7N(emAnqM%(?QRBlr99CEA)LIo*TOyBELK%*%Ug2O%1Vi&#l!1r#o)C3iU_E+><7
zLBfO<ljHmq5TGTa<w`2=OJ^QJax5ksrYjMU-zy19E$IL^7m`e-O_p*y#2nVowvcsF
z{vEC$eUOQbkvg5Rt3>i)DPHlFH9kZ~-JnG=;FMKCj6fPigaadQoFQ#fG3l>JrTU)_
zk#KQiN#4$|qWS;MmPwrxscrm}piRs4SI@H>noN8L&{$0sY<G=tO&>Q{6C_~FYFGq_
z;_a@_FM-ci%^!zENWbEb=l)bu#~^RtpB%0KyKl*%Y^bIjNLUE4FO1#$5wTgHCo#2$
zwfq$gI;aFe?f`md5%rA8AiDo*!ouPNiV-3jkB=37M$KIKC$c!B6XRs6y^z-3l`*BA
zfV9#dD3(%+g34x4Ev-<oP*sgJfqS@UIpmfyl+d}kCMV_h-h4OGo-i-K*iiOCoAqeZ
zur6R8Vf`*MvC={`lJ+K=@QHbi&-k(Dh)@-wpafS)U9@_2d%|!wESZx!xl?^yFg1qt
z><PA}k66UQnf~|by*<+JQsJ=64bPcr7S@Xm7~E&46xDh}euQ9s=6DS->zxVwZP+V0
zVbZh-T7c;uRY^Z9)gddpYhybXS*_r%O!=@%E{%L!$_T1~n0Kq}iOc6gU{AK>(kK|r
z^qhIdy!5PC!ZmZ{J6T7LArsJl@w<83H~jSH!jpOVRms{9v>QcfPf^8|6bY#ENmUze
zz@7j0P<!ixqkXYFF<pCp$MfV`uTD>{XJrS=z9W}q3Fnc2)HLC&oqjMDkqT?hZu07l
z{V023_o05Duut*oVqyqyMAL<j%5gijTydL%KXQTKWmQ*oYgWv($pIXSSM!fy22I+T
z<AMVS5e?=osD0n>6s_1lag>m1wA@Y(K!^@oPMMOLa#_maATuC^NXE<uTlv$dB=5Y|
zU|Tyc1GyC|S6k)#&YS}yUfz1-_0EnBq)YJHg}Sfhug6IEW33DiT8y2LQtNAox&+QE
zQFh$*lSz|0qvQBGEghm-Zx>2VyPAJFWx=bieM~Ro)LDIe?<YmPiG@L{Qry-ilo=XZ
zSzY^G%S1ad9lN+7vBI&!Ug&9-D!M3{49rtsCnU8b;+)R<O(1e2ri{LVQ39X&!=b~W
zJj6V2WW+o;0mwH7xU`E)CEzKm<1f05rOc0)E{ee`6GOqHqULd@BucuSIA+W|pqm9*
z05D0J;ADBt<`}`NWebU>@({#2-`?fJ{cP#ldh$v6(WOo@|I@6XemZAYSF|!r;|~1N
z^q5aDGloMKGPCOv&b0J>1%T?`Pv?7S!!(x=-P;@9&KHfe9@>SP3Dt5~mZO&*4%?;1
zz2AvF{&-{N!4YPYYwr$4#-sSzB}yj2F3}RNfQJqqHVb>OuT^D{22VSPX^tFwncplY
zni(3cap-4Kx@aXhx}okM!$dP?VO~|N(#ID_mp)i}ft_Vfdn#OF2*9(54w(wq$y)X{
z&{aR>m_#|}qLtuT*z@m|#cc_X`7vOvCW;2Hko*7Smw=x_BoFl5iT?-)r9u@q-pL9L
zN&1~(!DUF*Q)ASnPxbsw5f@y;a-vQ`f`TxUk23rCLeBP3?JTi}aF_Ik{2C22-%Q|L
zzx<7e+sk{Z!T0rZ7<d?Bwe_E>D>j&Yn6$la^Y|<I5ozjR&+kwkWkK}7-=tmHxF(<1
z^)DLfeLL=%7&%Y3YEkZ<alhDfzv$^HrrOp+`?ho8CHw0j#E{=$c$mHxrE?afeGq2f
zY7aq+;8kR4U%_;*lFG-CNI<#5KWF+sPxqXU9v|Ay5cf|HHvrAq7q_+7+rx*8i;T^b
z4BDNH6>KmiFm=a1GZSyD^_iZfw_MinwT7O#DG&>@fZ=z)epJ?uwHQ{<P@Z5r!iY{8
zslJ1y=;Y#`=xg^WX+%1uOp{1AQplg?Cly0*ioL<f2BY1x^m12gPo8Wc!|>N>`gC{a
zGs98)!Q0$MP{6^8z9!%G?~v1?#!)gNgolk^-kyiQ`!-&4Npwa=w=&d&OTdo0XoG)F
z`wc~G{~1UKoBTNabDV6$^Cs@z(G{vqn3Cx`n>L+$$QU&x4f5q$h}n5`Q}5#)(WH*(
zC3(ckYcmoPg+#P=&=Ev)wiv%llIb9uq^Ig(!?KS8DgaBDu?rnyWR+e5wZn?}*R931
zC)-CDZ&80;a^<rIqx<2%;bz@-e6S$2Rp2VIpPhL1Fa??({&~1fZ*osg4*ZZWw_^^a
zN+fT*K}59Dsp<c~7`s%mJ0`fRi0rZQ)8<vnHLhVEi;fo*wK{M~*mZJqTKcrEMM}h8
zc$H@*oPjJyyUmI^fI_J&oeTsWNbwG5g)kVa>C*h*vwtipU|v~jrmEmhmrlV`itvfQ
zpUsDTjpQAV@1TO)?RD4tP)!%!bl{hY-jO^~WKUk3zOxPEt({+YyRj-IJT_R#Q(j+X
z?CJ>X$}(8K^Xq7LyjlsK9|-;vhHoXic>2CF1Xv6_tXh5)hv)SZlOg{A+B(^vWqs85
z3N_CHl@^fNCgRBWn1$t>YD60#tfGd|y5rwMH1gWLbBc1<AG~S;m+5J!(P%6>sntgJ
zVU(`sY|Ezwbra0YFceSGdYn8jVybmII(?F;8OznEXE<mVH0(qyQd`e9=za>}a(Gq8
z0Oo7C|5R@Vz!DfKqa?0QaiZyTZ;4^mMm8*}RAjaGRU7!$1RiaO);dh;V9kHI_mitL
zi@ZA@m1cKTq8H4x)nKnKOVg(Yn6ns)plIJVrP;U#V<}38K;y`&lND{d2GbyOkIby$
zsf!#NJyj%weHsT!Iy9@$7$ecF>}a&m0p-M6@qG>M*<rEb2rGNei~^(pk}2uWZ$XkS
z{VI;kNHilm8nT7_-Fn~9i`X-<b+!k=X$o^n2f{RIrx^4{gy<C*PYbrC*7e?~VSm8>
z@IVUjuTMOWyv(}SyBHM~y(bZ+%~SeDpMX;rDuK{1<uJ|yYdtSYg%&<rOj4Up22kEA
z)J}UQZ<0CEsGm`X!^-m%t(U}d;Pwk8=dOam(Dm9ob-VHZH0+%SXStPPvcj9F@EaQH
z|85Sj971qV)^~PGG*zyRJ^fN6!SO-Dh4JdA_I5C1jdy7C*$i5WGBVaoIPm>?YWW9L
z1~44L2ib$(tUS7Sro>Q+_;wGJ09S5}(W^$aY2UVa?Wm@Oaaku@yW;OobK`R_N=?D5
z)0^yG1<Iw#)u+B>2TQ}M;kiR&*|0&f*`wLKeP|U#BL{Ho6|t)8w%btkBwit_>iMq8
z(?V|h{uM3VxmzZ|eZ?Jt6$Gm-<z6;wD>Bf-zdQ7z%js9I{}^e%9j-I*0>K&C?!hOW
zUhD>usC$qsZ4E9<raYLaCa5P;h9?nmNM-yABq*L_w-JKcEr%q2{-xc;&c;rC&wFQE
zN_s)S(9e=Ohq?tf9O)LL9F?uOiZf}2wZtfW3250EnJ<iT?ZK}{#I|Wm)@$sQLXeQ&
zd~7Fmb$Gb(-Q8O{)kGZz_zI$29vfnp9NO@$kL+7*Cs1>`@SH)C=^?7-jcuyt^>Xum
z58v~C)r?t>-F&#Jw%ye3y3pP-DG5S}U&Mz)9Ip26!9|RR9d&<X!Y|w8t>CSQX!yJ_
zZTGe<sZ6U{{r6WO?)qC1It%0~`E7^FOU0t5UBnd$B%<O<6FdzzkZq4YRuj6s0E!<u
zTP$>yk@#7L7}8((G(RKFkXuzA^TKC*>Rmi<u;p3nRoT3NED&Pqu%f;Ci5JQ)vR&K5
z)xl=vFW<a<SfJW7{RS&3bcsm9Gn08^)%yzdMqv3z`RwA9y>^gg5&97d-csll&;Gk^
z4@|esBO=fZ5$VMWv~zkMzE6v7%-KJw2*Q0^?}$<pWBQ4i&-DT~a@QW?SK3yUIzVL5
z(4zOKUB&kyn74V<Kx}v#`o#Lt&0^jClDAg6#1Vtuqly$|)36|Ya-7udVnY$yg}tad
zLXdqY3RcT#A>)ek4&oB|Vd~BBuDs4EL@xjmu>H3JgpXMdpyQ&!^wDZXfn+Un_6FzQ
zA=bs!%~VZ4sHcEa$Z?2Bm}_-^pr_dVjx|kPkVV#S%j?rjdVyzNN?N}ZX>F_HI(*eJ
zNv)`K{<)Lz^&Dg@eP-k)_^Gm~GxD+Ufa~mVId9Fk_mXEWa5_^Bj))wSwDk7moO2w;
zy@CaVSFVBs*EKC>{?8i=ED^r;CeC}~CG|4u=DC4mv^y}Nu9XTlP8{3)p2kg3jpmL%
zXqsvi4U!KE%^OhPY940E9rk<^nv%Lw-^~2${I{91?J$?P*(Y$-kt>W`Nr(`^f?$Hk
z`SpKfy2_>hH-5zaA{28fu}2x8Mchb0EsGM7dk3&gFH(#SX9-Vuk&edlIOj2@8l^D?
zkOr;^=-UhQi3vH542~J1$ErC)1BQP0H1x?T4qawADG<E%q(k8B<KqWqAIUz4G{HO*
zblu22EB}Jt<=@$Ml&jPzSJ8KTDIMH<cjD+<s7NzEOcAZSO5$vI|F}Du4ru)h%+@(M
zk^>-4cjz%|FTuO@jk8#UXoeXO8NG)Ux!fiFJvpweOiM-iQ%TfFBBF<k<1`Ljfi|^?
zT8da*7~@ZutXaW@efRuDYv>CXgW9~{bge^5Lgs?_zrhts*rsWDStjRQvj*k7C<UW!
z|ANxJq?jk<c#FO=Z&CEJl2y}*8r1hUD}iA;HWJ#teYx9I!C}o`?I-kaHlEcUtou)k
z88O_qr$tMx#?>}}rFH^?xwh-bL)%&kjsXa#5pkErhV|ThaaGL`$yl-+jC>ulCKV-a
zT8(6yH9xwYCmFBjC5roYvb_Cr4|~_h0f-f%;!X0ukgPxo)hLbMxxUC*fwj!tvViP%
zf4Zjm&PM%q$l(#oS(FSKy>2g;1JF&^+c@?-6uSeP@4I_*(IX?5df|Ea(3jWt#qL`~
zHMaL5v(P)Gnb4IE^3dgL*pVs0O?)Yq4dyXXel}J8Wx2z;sQTn0D>F#Bg#LN`(@1`%
zQ<C&%HUb@-o+W?Y{}yX8I>Ipx14nLssW1!F6S%j!T@y}Yj8dbNdNX9X@&fsRA}~?$
zQX=O6!6h?hXJvWR2;U*5K~pmFBUP%mAc;6BTu*GNW8ou^OeiP3X?MP(of6DjmQn;y
zO-HEK6DZZBIzCKwVNR!<&>Y<O?E1NSLBZFy!Z6kwF7Q!oYK$B;r}HhqBK+p6y^tWC
zm`T(udK!nHQB<pO8Yh=sz%lR;dIC8;pEx6b1vLs8o0v(&EOHt%m!6}R-XZb^7cY7m
zFPB@u6MPDfzVykagt-6#w{vNz{zC4({Fz9DVu31Y;=_APkT^(00(Nv-EH`ddQ8%Al
zfx4^5j(eiO@X$Ty>(>AYgt1x7e={j8;sq5I+fFfG=}RnG7+CATj3tCPN`C~fqD)Ra
z$4tG2q#{I^50o6rP2{EroVlDOl$u8}DPAd}yxZ_K!oAxtiJ8St6OdcP-nZ=JeNFxU
z&fNR&(=aw2$)$>Q{xH3dM<j`nOYJJg<%n;FuMHsPH8*pO=C|;7IRd#rSjDa99D9v)
z5%rmsnY;Y29W#)i3*MYYg}BNg7f=i=A|r)Ctc<p{@b>K7S46yY62*5*JhQzolTXT}
z5LAW~NSa_SuMT5rOE<mNFM~@O#E)J=)ky48ArbUkWGyXf+eQ3)W!7rGx^)t($LOpS
z0N5)8Tgds$CaMM?mO#pX&%>4yP|d+sq8c1U%R8jxg|7!?ur5c33;79QmPaTc7#Iwl
zN9bSPq{Z=-GD_xzHm%I33Zf11zb|&^)(kn(V+3YfivPs*oH;PDm|0CPlbKlcR*@_t
zAUVp<5akLBL<FLrG^3mUEcKsOs03CnK^0~te-lkGe<)Qj&|-_;6Mds6^FTfp&cX20
z0T<dpJA7sv6ml$Kb)>35<rPZHbpGn*AsQKeE*DL_s9Wim&eJSN?znu?eba2vs<H1L
z+2b;dG)M$fQea)kwc*}(Hm!h3QX{~aNf{^u)qZXQm&jA6FfAnUw`P_Te*@UAnzOhk
zWSy7t3-c`_63wF?p&S7<Kn=t4b8pHa!&FB&jljbYTybAGQLtn|77pUIWtcKr+WRzI
z@gCnGHDGtVE<jdhaGY-0d2YU!uD8rGoq8uN&9`!wKTBNCX@(JsT>O(e(Yfrw<pBrh
ztODAy3}^QW5uT#aBH0vc04QW24rS<Ev!)#FkQgo}IBqxUoVF66j)4vdg^4wYjk^kV
z9STt3PrR9?Q_9!o;k@czRHIzXQ7=o@7=fToR>HEAVDsc6rqS&7J#@@Wq2*yvanNm_
zEq%$eL=3F2NeO;B$l^Y7-0gIGOi7c}wNm^pkiS%Z;lT_y5c{|Z5B!JYBFBasrJu%z
zT{_uxgYQ3Hyw%Hb`MnFIufw)S>9zP(wo5n8S={_|0%bpwb4GU|4a+#I@P(J-Aa}3C
zi${`E!cbyaWSg_@3ai@&x8d$kyx>Mj-wV|9<k*5WJy8{aD1E1aI3DX3Akpno0Fa-q
zp_PFs20ep0u_*$IWQ8TO8{>H-V#%!!7of82t34ClT!SU0`tpRDf&+0GR!~}p{}Q?*
zZ`4@*2)17iIT<Eym8|0JbXO$E`ZBa6^K5&xvN{fIF^arRamRALXT|drCV%RWev-Pt
zyJ;*JXjNw_dwPK_5tEc(g+hy8i@hOdn>DM4Xd1TCp^pRh%p&d^On1TM>K<99RcOP&
zur9PGHv?5){zqLe&kL66fl2vHltI9P>!W@Eydu2}wojZy*>@*evESck$r^Q_tNEb<
zhBqfB+J4^t(=cPsN*1r;(Q&oW?P~kz`!SMUnW2im^;l&qezfsmefIDGb$j`5f_KC@
z;`&NG-|8HYiYGbp5wjknt=^t@5oo44;<%7_RUdYa``u*qvb?#1^asSF@>cyKPSXpK
z(Jghf)7Vb%KC)OH8y&fgLaT_!vFYEhgRav)Li-bLXQu?IlCx98)7qzwt98xRX1*pd
z<2E-l4?#OP96$CVR;r7yg}2Y)UL-y}QplPkG*$!*%x(Lb8s#zG@zjKrqfNf%02~bo
z7DDumrL@s1rAuT@66EC7OvgKfX;xBF*yR`l(LUQ|7!(^a9gFT#5Z?_JH!RBG$eXWK
z`~Fd4?K`4uI7YH_dK@?n&EPL0Y)@k`<WnXoC4bz{GLR`7iQ8g`ZO{_A`P?hGGI#Nd
z1l|Jt8GwiA{cmACB%RXj2B(QrD2EAKCgckgm+h(635i0&Y0%wnO+;VZ=fX9;LOJ>M
z2<h8ikdEcOG)`BR5yWIghDC-|BroR}y0;}+A5fXl4^t;^CCJYc)hIWqATZl|?~Gxh
z1;{$Oi_^)Dsg!z?j^Q@#WY#e$f&GAJw<y3OKTe@-g^EV@9Y@1csNYHN@xiJ}5v4=J
zp;R0S50^*XO_c!r*UZTg6qCflorebbnfUIMn_7n4h6eI>IBhrS1y#{scq^D+@<iuH
zQFqiYYY|76<|D!ee;Sl&W{pz84|?aduYxrMPp5Z0y`{b!vt8k||9;$A7inhc&+-8l
z4vl>ucu|%*m1Q;EbsmI1mk9$uN!hm#J?b_u3s*rYq-}KQd=EMK+EK|-SY+bsC3C+>
zINW8rIFLILiC#2v`{KZN(Kl4tH>h1GUi-v)$D6)SW;cf{;}WKY%;oQQ%K3|C#<Tsz
z8^p>iW`%M2P19uUgZ9I9HxFM?q~O3Cr-jT&sQP42rZrr^h^6bVHw+htMp5E{N~BGg
z!G=5PrJt&?m)UG68T0tYdEdXgTAU(covu(3X}0HyprXWOKaAWj>5cOHm|$(0=+l#b
zO1*4HZ&ga<E6SGOXJBLCej(jIF-5HjSDuIl{qj~n&+f%}Lgyz=KU!$gN1%atykwtj
z*$~a5zP>@1i$UC=H>IFA2a8pmEL`3qF#6jf!hh9{kDorKZix}TrB5%GG*~;;Sf+MF
z`8e6~jneJuo-ZxD@X>354Ey>%o>66TU`^`{-!V$e;s>L7Iv62EpTy#2_-|y1Eiuw_
z_D>PZrJ*>k5=m~&fBlQS&1nN<-V8CYu=A0f-NcW%d41!!7@3LJ(vEe0SGTQBXFW)#
zKe@dFsV>SSYiDl0pdI!rpr$8f{5@U-qfr_Ma<nZP)=H-^-xcL%USJ-f5JZ<5ZCk})
z8NM%#MUcm%oNlbV1rEazhJP515P-M39m{RZ6Kx^OD{^@K$vNZwkSGQ=PwNC%YFswY
zStjH41>tK=#_T@}izH(d2~RNhI3I6YFi8sfH>w=|nVc*Hud4gFD?Q9vq-aRzGRHDE
z)tc0ZJ5*=&O+mqO*4cL_=y%OgMU=+3p67y5_bGqInk>6sH`70ty|8kuhKf`Fwm(tV
zO6}ArchXdyz0ng&^#KE#D`3RYCnUc>!#bm1vVjyi8!{Fm7Omyr7!zbB65h`fNInRN
z1Mp`5<m>@CrElS*6xMnO+)%#-GqV?_G)DqADUt|+i`Q3Jmq}h$x-k#-xVrDn#o7A7
zXKStFbFJY)e51ya&nL2nHmeJqtQ2C3l>sroEakHy#T;4v7&Rbj<s>N`&Ai!NSeV8s
zjHND;&S}E*;lR<+n`Z)zN0}$#$moUt#nG=_&0S$5zZlXXSc$_U4F)rDyaY2MjajDi
z+Pu3v6T$sc*;eoH(LDXfiQK~3LY=~1zZ~!UUsx76NHmbW{dcXiP`}^m=6=%X&r1Ya
z&RdU{3dV56Eq#<~EF46+io~KmO`47>l?^`+n9|q@7210^2yd%blnk$$7Pvs*Z3pIg
zvsqI#et_cusE?uco8_$c_VVy_EOfunn(a5Oo>ot$Z@?C=UT-x;9^S+idfR^_L`Hd@
zhO-awwB6i$euO*bjESPi;(A(e=2=2}N=mUF7IF*UF4T5Kz<zRW5KF=R=sT$V_MEY6
zSS7-<ylWd4#?yoEL^k728F9UZg<Vyw2^RM_r5hmwfG%3i7Dr3|Lo_<}yuxlVwrw3;
z-i2#r4erUq`lRX{{e*~bI#zqal+XL_$9<Q~+e5h;DmWCG8gSS2-j;H`6Nbmqltn?r
z`t`?X8n0zazra1=w;Ix$t+zA&JktJWP3o_Fh+N?i_O|O9=>hMGrkbZVuK2u`9P{9w
zwC+&>oI{MHtgyP2%Zw+nCs(QvLlHa{34LcjY|!t8#R6unrT~~e=M4p^$nZ|wBWjWE
z`&Zt3f?8PRQblv8xl?uY*pzFpK~m&ao3PTcB69%a1C4gys45xgJlZ_qrbTC)p5qv)
z@u`p`RekeGV`Mw+p`OZ}WEMbxb1-#a+qDDkvTB=)+ALTyI+XG`V07njIlOV}c8FT|
zFM}KkqU3pgZJ<ld#<^H=&;f{dPxDBm<6CRBu7A<-O+K;;R*l_X=v5>(y>(@y`a5Ji
zt5a%a?Ca(>t)z~lR?0{f2?;|}r!&qs@vzCe12+YqK~OxhYqtZd-05<kN-PHhbyfhR
zo<JQKb{G_o!o(~1ldo@e9QOw5jF9yNZh|3Xq)vNQC<@P7xR|D2&S-*>N?9Y-1X`jQ
z-WH<uQLYY?q@!f*>X3l5Hy*_+qS8m;#!hxqxyad3E3EB+D$7f4=tuGU3qkJdM}hqM
zQ{1kFK(~nb2WOvG!7!F<R}U8P8!_NxM$YfgHpZ(Y0_ugV+V5O0J8BAg(np9--N1tp
zig-FNVxM5TzaT4#(kPVuR*BMtx1~M4j=fxTah<Z;F+j-^IhMhm>_#!07u~@wyO9{M
zmRqcsS%-A|JKHmwav6$ti8sdi{}897Aj6{0tx2M*s*_T1VSeLXGyo;*4I682YvF%1
z@*<1!P-m;~(qGr)#EmcQN?E%JQBgn13{_M60lQ;xCaM?jCNruhAri$4-)}0BOKu>M
zV?v}jfT}h@c+#T5rX1E+I8r`&j~HdrPIo4)4Y%bUL_E2X8sxaEKk5#y00kkqB?`8t
z0!AV<HURg|rB3S-&_kO<=3pWjAN7+w(o|b=Y2OC-G2?nD(=J`0_wOu1^4E^DYUwE)
z@hIGsy*s()yU$oqsJnSan$g`vW1so_>(JJX{YO>O7-Sme&o-iSb%1od=fj)zQd{m~
zEQiyx4*nbqXRJ4xuF}ZOFN`*BHfF6WwxKrd(v+#m9Xa<F0Ge)K8bWY>|GKNZA-=kG
zVXpb(*T@v)nxW1X)mqxV>K4Th8i<Gt2FayA;z(4~s9~IuKEKx#Jc7QWAZ=C@l%|gc
zG+sQvq0M9%?81TBAsP`5Za)H=2zUIeV(Ttoqh6arjwg_CS!B2#v*pSXQ6Ar=`(;p&
zt>WLbwr6hw0S!E^N9Uy!cvN4iC@gIKzaPrPZ>8<4A@Jbf;yB;7M<aE?w4TimIkLC)
zRTaU4%yh3AUxE_rO}rVs6@Bzj=Pc3OJg9f>f{0ELk~VHm<b;#aPqC&pMfZQT8xLHc
z6GF&Bv!TKac{-?<awd5?G@n2H4^*jQvE~zs-YEw390by%VPp3L&7xYk4fUJWSv(3g
z%761Nz%i5b!d)A&R@H;UBCHt*d=W)$IWaYD2_Us1l={2j{hq0}kkS*@f%dhd4gcIx
zW^v{g2=Wb5aj}!3p}*tjHH1{yrl9msgXh5rp}3LRn}PP|Hx?G#tSb{)3IFg$1J`;+
ze-Z@jlJY4I&#up^%00@H$Irh9<%|7y10eXzN;6c~%L74&6+1>NN~Ji@P&OFd2nLO6
z{AYs3uqqmc{)o><&UWun@;CIGN=ht?Y?4`s%tzL;lgghuWCu{P5y)<D!ww?oWtrqb
zslI^l7AB5^2l>G4iShgLLx^UrrEQF!l>sYOj|45;`~Yk7A<0@N>2l<4s@l)|H^Zl^
zocOb8l75NedFt}m<k+Ah1Qs}i;%mjTK!gTQo~;Y|m+5MVxBB_{dw;6}ir$4F`zmH8
z%={_OFD@J8H}l&ODgo?o{x>cWn1qZR*`%rRg>rLGISOn^HannD(N-T-Rscm98}KPg
zmZKc0NJ<U+RL_<Iv@w66APaK(|F&1(o@7HUo0cMN!Jw%5f)LqwD*XC75FG0%1dHFB
zmz;u4Sin@Tp$7X&F&UMz_ZHZAtzbw@$bP_R+86=%n=;I>b_NO&G_I#~d-kWp%fYY6
zZYvM}-0L!m8YQ6Z#BFOAE?!eQQ!KE+e*&k=_Y35wZx>N>IAP7vaG#U;*+1V3jfa}V
zlG`C4Kf)31CB)f=3yCDUlQ6n(z&qV4tACQghL{jQX=C+;982=;Y!<(Ug@NhVB#&B9
zY~6b%;ibW0s}?Fyr|n1)Vi<jEw-1lU!yUw&pqi-Rdh~a<yX0dnWno$`E#(FOAOc>y
zyBW`i?4jGfYd5WxG$~%hry~s^S}DSf=`!2N458B*me29^=0kIghRTFSl()<m2yl{t
zA|qxJNK8YVlxbHbo$)2&;4?dM%mz#wdHHYgZMQ4hJZXGGQj5~=OA2omas7ayK->Mn
zd1MEg6?gq-`^Oql7B}f*Xe6$;M-!0F{7RJH_N$Z#d`HH9H$YjemMB_EF8(LZiI8oC
zY=D;yL)Kw932QcjV^DGND&&))>V!wNlR~$Mjjxv35mx;;CgLrrGy`*M`}C;I0@>E&
zVgxuV)5&*;of<0IR<YDUqgm$-|Hh4w%}el*&1*t3Qd^_gR9I3nq>J2B$39?!(ZCI=
zaRVolClls(i*k&#tbh}0*d>hT$8t&}E2_7b@{1PPk1uj+ZfYp?&~dz>9JRN|m~!V5
z`StiI_Yf<f`faYz&{n~p=pV60)x0?i2X3kcjr0ZjdBmGS??4ahglZ)#zDMyel)4%x
z+1qvgB8@jkU4u)FJ)G2bo-y$E(&EW!dqu{MQQPgD1ke6~@W+q$Uv=SrKCRr2J<Em0
zOIC2U4xf>hMbN6$qkf!^gJQdX^{?emxA!<wB<ja+ZD8VUb(zCh>(rqdGhPqT2>OV=
z2HtQDcD;0P9LL}G?YXuvXgX+o11k{;yp6sC-9UX3MSb$Z7vCOZLjW9k*B;Y;W$>~4
z8Ut~UFd}!j*l&mrneVaXKp0B^OMoj&L@N~?1fhH^oJjqBZ5Q)ZYO!xMZeA77_3!){
z%jx+veDcIL;cXyS<-M~nNs~{kAs^<edm3LBdhmOWK#V>wSC`(hEuckr9zqpN1pIK5
z8r}+(^dbDO#(z5p0-8np;*^cLmA3kEBS?+^s$Mt?D4l%w<F~6PGvB2*z*c+Wa39K2
z?@N6-*PG^V^LHVoY$N)Nk1)zt_kV*|lpFF5fuQ}*onjgTwMm+1|4R7LOKF)l6LzKe
zU4)R9#&<?8FINgICbhRVVT7V&J|i<6lRMW$g}+(UOVl6uZ^0p8$ihIkD&fKEb;z67
zuZ4Xh`xJ_YW+HYKBnm~$r1x(gjXKM26c#pI3A<z;)H*HuOljCyl<C}{{t>}`gkcHO
z{VFNjvRJ}eCFh~yB1q&V{oa%1?oK1La@gZ1#5jP)O=@qt;rRFWpiPIurIJOZM7o#j
z_H7dmtPpT^EOl3ixZ%?{T06J~4F=)M8j^k-ynK`UzA+1qDcl>x^yba-9YE<l?IJ^s
zk+x^3FgfLF!lm$Fc{NlSzn1-CA7us_(WT`cz0I1lOlIkg(Jvs#&FM3z{Gh>XX=yd`
zUZc46X8OTb+VtJzLJ+pj3|C!BK9s%2+Ta=9BY=V{ZY!Zl-|Kbs_l2e->?^8LXfT9h
zEyf2Py`&1+q9XcTAvr}tIsePpZuzj*+Y}tsO=I!wyAD>7c_(tUj1vz*>nis1&>^mo
zKRu$KZ0u%wbaT%+@YsmkGs_*@cMgS%GUm<dKj3_F%_0w=AiF9}MsL2uqPrR)J&B#{
z#Q~`sZ*Ko}3#E;HGNGup`AJc4meKZ&Ip^<-p_p7kSwm(=CUsiL)fs~IdKf;SO`kJ=
zn6PAdm3f8=d-(hgF@)vfnCClUD(N6~nG-Y&uVj$VF!q#LCNnN{BvR_b+-26-FVtDu
zlSxR&@##mFn~P~~F3E1vqGCP|s^1=&fY!00YANfGrK9U6sXl|NnHE(NUFuiPJGYf$
zXIY`F?<ttcZi-&x3&T*FZ<S$_*{{HKuF%oV^!3&bq3#Cu^mPMts9K$PHV4fKZ*1;W
ziRFHB#|gG!1nu!ACS-)m%W!rfX+pd3Q0q|W?txB!mit(CYI>t64u?rUj>p|3&|m3J
zdfP3tUB&W?KITfJrg`{36ouns2wvkfX~%47lCECGwex`?bzEHZ!55Og5YqHX9}EwF
zhw<@lgqte#QQ>h~wN{xt8I6au24<f!be1-MvtrS+{Jh|284~jYozvGU4?M%SJ{)ZA
zZ%MxWlVg)lFlb+6>;lBd0fHQ`F#>n0-cS{uL*zNuyV<Fnep)3e`4W}E_+oF-;V$T7
z+wrpe%D1T<x3=X`u-0nreY_EWd4_nkRsXSaR_3>5jF1W8gx)^(dHd?|pYYJ@+_%5y
zJ&w#}S%4#zA{_tPC(y*FBQZnG*W}v<F?0m#e5E?gE)<j#2pC7+egF-QKGX=>1iUdF
zx`wyrv&Gzfp0ueocn)EKqFpF)m!W6BK4)6%haMh4sgH$s<sJ?A`bOBw#CTvo5U+^6
z5D~9n4;H#^*U?lgNxNWvLYl^PFt$=^9bRp^=dTA%bA8}fW}Gc8RVxKwkh;6RA7+yn
z7}zc<?Vel`e2XB#7zAoHSv!Wlv6y#wyS__+H#xR1Nbk4!oQ&~*ok{$P6nL4mSXp_t
zm;rp~R#OVI;cvVd*0odlX*rS93E|qlAQ$D!p}LRZ&-x>9bKK7;<q5|8cX2#j+&SU3
z<jZ;g!w%$^F-|5No%>Jq9#aUUhTP#_LnbKgS4qo5eXpCS065x;Ci>&xtIkNN`-y5J
z)h^T${#K(WD;Eut5cFn~jfSmf)46(;Sg=zWC(T4#UFR9r=6HnVYXk_1IO$vr==<V7
zzSWLR<{e|{rYkxEXbytsm@OSwAL1|3Tjh1*ikoox=WZkjZ&;-<<~8w@F!_K2#I69p
zAr;b;Q;4V^aP9j|cw>i~r`HKJ{2C7*(W3MmjN=goZ@$}n)NGe%GQ+|y=iZI$q1z31
zh}MnZ(ot{y`-*bIYTGGj4X}a`*}G_@Zt!|PjxJ@uZzWsqhm?x<;Q7tRN99VTP2RV<
z)6u|pET)ovG@&cjE<LD_ajL0C_O=&$+aCD@ng$vVu-*Y3mrV3yR(bvg&TE(~zvv(C
z!(8jSd)@&g&!~4SQV6nLLiU`cY&)H1{Y4AS5lVQ(qz^FttkV2)O+9X0t!ZNQ4TTJ`
zE=>ERSd=spAwxk4)2$HGgcai-$nw4(?&Sbc9}WGexbX#b<C}ip{l(j{51Fxhui5@O
zb-f+~uNSwmNXky>@x_NHGARE<b3le9a|J<ELDz=E0~g)z%1+=8vA5kW+B(H=9sSPO
z2c7JSIq|wZb7~TLZSG(p;Wl4$#eU7V)rDZR09aX~nObqrU0R7O!-?5>=|sb*Biznh
zhTS{_kdmk)0p?OPkT3BnwDGknw5TJ?e@zvEiP_)tq;H#5VuwYWM_s*r7p@zB<b`X-
z>}&2ngb@f3gnOOtzGBtMv8UPDkh0tuA>k9?`iZf2e^e>x#pg-0_9o~pc0m+i6ci|<
z#{IEkT>CQi>NKP@McT^Mu(e=}^5cF1zqF`3#5VVa|A($R{l)!LG^FC>ned_(5RX|{
zf*;#+WgFb4L?y)n84JZll&IC*o^Zxr9nX`eaK^Tt53&45=a8^}<lo7io&IRDZkB{Y
z_vodHC{sgVea7Kr$~l;A)pPFI#ES1}|DKP%;n9>x(IS$U&YW)L&9)w~s?*@LK-rU<
zEV=8-anxwkHw_k2fgYm#QnpC}6rQ${L?)e`C?+6SDbg2UKGEkf4TjggGJ4{vB<;0d
zxdy+Y{cjpAW16rQw^mzZ-rkl>(?ycm$Rp0!TcXz=&;SL!h;=%tL9|-G^;iGkf8yJw
z8@=_RLGh9hCc=@Q^VDmlKD79WqlJmZ{Wl{|T+Wz2x7C>_PVNN;2~dBU3_QWWyX=ct
z{cWAozuQ!te+bDy&bcQ74i!=&h&n^aqhmVZP^$3tURorI5>?Tq2kShx!vAL@jT<4u
zl_k@3@Tt{QBJCiG+;s{qJunr{gZ}m^@^gQP<dO+WlxX})&@Jt63Vmsk?<1g>@$HFs
zXk1uo&jg^u>?i34kCvuzB003TQm)!hW2>nyTgEtED7{!HJ!&hRMdnS9J~F;&=lAh*
ziw-eudnVuh^D@70ONM<rmm6(V(y#q(%4#MMS~&+&A%!l+;S{IvhRoGU<C?7LolKQB
zMN@fu!^$5um?_$+Jj2;p<s!3C12fJJ?O}PQ*cDh^XX>^5S9-Y*$2*~Qr1)Zd;K*V@
zZtmt1);sJnoVnxi0y=|b>GNu`G8x!$tT^}KlM~8RB~hrsKwKh--1ALTqHGiMKv<vZ
zHSiF@DZ#>PUH=bmdDar=?b=^!nA+<<5mBU+5YSz;At56?hAKa#P+&7O4NT-D+-}hk
zkjsGSd3+&}aI%gqDQW)YouA!>kDITuauPrGa?&^r{`?s9e6p=(J-g5m-UgqE4JU?U
zlk||&7AaKcyjnOAJ#|>Cr^u;ke}6uFe0XL3^!jr6Xf<#&UD+ssSkQ}WqJfNdT{KjU
zZ0T_Q#&3nVy-fEyuHU@pX(wrEmgA<{2_FG`Z5AmPumc(9<1VC}suNt#)0oF!{Zefi
zH*Vn*TGJ(^Ij-T?C{V=<<(}`kbbZoT|1;7O@d{@pIQX^eWp9LPfIw)j!>9h6FQoI_
zM_=gHk7yk6lUkX2*MHYWzIk45?ew0w=aip}rItz|5`%#|&!~c)h_aBeVGg;;ZU#2?
zJbzHh__%Di=HU-|(rpIZWTA%1_cz<7WtGHEf8(Gl`{DfB*W3JaaDznA0n@dw`6r%O
zOr<s5rHhdP-DR*ntCFMn9ns*%u4~Mdcn#aj%EPpI?@^{h+mZyhp>r_|R7L>}hSOwu
z<*##!KM57Ap^KM~;~A4&6Zr|H9fbhe7R8%sgGQ4wME4d~sTw{)Ll69J^LbMH5t03W
zO7RT=g)u&v5dA>3Uy*oFv=l`<R|W`c&D9F;`4<JxHI!=SQ!kpou=+1ywlnPSw93yv
zzxGHdz73N7icjGog;BMKMSp5QlRb`6Rne>A$YCI~kNX}YLcu1*x6<vJ+6}OyLT$YH
zgABN8rFg3|Rr9w0y*7S05!C|63FT?q-yl4kBjiNVZbG<t*5<Koq*Mk6itC{=NO(mt
zw<B*4$Y|dVU>+{d318r7eK)M*dVKvz|NUJ9m)J0<Z2X65cuB)&;6fbKuv%LECU_-X
z?YMRpKRL8^bX8Tw6Z`g>;R;-BZLD|txZ<Mh1gCbz5ek_1|H|*Zva|0Q$4YZ%Tn6v?
zo*8T*g~!~8bNEIMnJQDNc?t5(<mrJ-D_HQIs>x??e6acd+{1xzg9o~S5^z2PRQfG7
zd8(M-pfld#Y~uP((D{NZBk)-~`welJxB|?BfWdzsm1G;E%uiwBhOcpV{_fe`H(j!t
z<Bv&7aV7^R20|7Slgbwy+TtQJVQJ{14U2zpy;H4}E=*WxCviO7F$FQ`SJ9oMYV@Vf
zPLH~`%t$*ISl22Em;Yhsd^s<wTc>G~K$3BE*NKnMXpyP9E`3Ym`u%e78|@MNj|KO!
z3lzK_rQ6M*ACHdW9NhH}VzYpya}<c%u~U?OOp3urBcR#;@23aW-avxlbg%F&QhCUf
z#=#wSpxI8X&>~MlSLLYLg^>{|yq@H-66-zPkfr*WiZ|9oiYornG6u4Fj@WDKBeMDD
z1LW+^(Dp`#HU9%dDH*6$WKy6D=$Z*Bqft8*^zBsI-(>vt7x&nMck5ltDQWm}u@4=Z
z_$DW_W8m${B2Ma_efob<^_Ed_bWxWu?he7-Ex5b8Yk=TRaQ6a&LvTXl?(Xgo+#$HT
zTW}kmcfN1d%(`oJb=9gL)m`^goxSfqX9ov<ifU#OyDKue*Ni=f3`P&_iA?*#i4s|^
z-aH`<Zy0uezHBT+sj2OGxSS2GUNN_ev!$kHv2v0;W)mDkPPs2-kG=_7aS^I(Rfqc1
z4<Pd%X^j@%YP7>rG9saVq_yhY>yAQD{pl?vXYGU?OtSx-!VAyb2}Ey1?McmWF>m4C
zX`hGR{qXat0#TdZFXXOeS;}bFEj!P+vNqU;wG%ttbF_p&idSukU&MUQkOZC(7q%Eh
z-M+F6-crBIZE{Q2PS%`T1s4%Er%U%;w*^}+A9In4lK-&PY>!Pe>2PiuR^{|9ggV~C
zDqojsFZs5S-#{K<kqn0V0PPZF#+?hoq-0kFVFvSx%9%v-ih?i;+M-RA(8!S?t;gYm
zP#6!OL@D7u&-~7R0Ck>>>5704TklJn6RaaN`<1BP*Noc*v`;}LEjo4MrEnC;KoHr=
zl~S%lT80O3JjY%45@uGwhfmIc_nW^~NnjpodU&h;r3c&w_Y8xc%7zq0<%~$4s-UBO
zLJFX${6^a?s1X#?#XCSJRs<=o-`RuW3(r^{KFQuD@w@7ss3k^Krr9QFtZn1q35@;z
zy@E6<6BO5)&!45ekLr?Ro`c|$!z!%B)WZV%Y{vSB&}uwTOJ9l=7C$~TkmEo^92VOp
zhYTZ_7no+N(qAUDivNh7Kw_Q1BbEalFs{$CPUyJHhgwe#L6Z(!O~!%Q26=!ojOxWm
zdF*~@fylprjJgwv6>4Ciu&I~EDtDCf;1A#n3Qccp{=0&0qhE?FnBW3yI0|5i>%k@v
zpm7dk2wds_x(HV40AmQy30ow*Q8))u0V_Cw#8(3XgNY~oJk{x(o+cnOD6J=n369oS
zF2z-TD2Zv=4ztLW2q%EF5HXR56c3++<@2wkjH!U{VTw>o^pHEIg}24q%rSCuWbyYF
zAK21mX;lI(QL&bXyu)pEs@+v&MHiGqdG7tkxW4|T5^qm0(&o;-X}Q-tod`4d4)8Gf
zdDfLQgHBzzxe_+5=YJf20n4z8_9Qn`fcgGBPkFZ7wxN>#1LXzRoW|=X4nIFDnG;KR
zsWTm-7nqC!qaL!P2ZA>W%9v9!cE9`9{J&DGBTc>@&(g3MYg&EImHodvWW)7xJx5Y%
zFC!kz-^?O=Xc)6SLQe-<LMv|%G=W<rF!s|bcV(}`^iO^Kze@79f$p2RfkaKDs91)H
zx+jq~rR_BiYIJ+)$bXcK9H!FQ8`c>}<}9ib(Bq1oQz6iVOex)`2e$b*5Gq)p@`s*)
z7BD#RlsbT$D;^#NEs~a2s3m>mZYdY_Gb1jw#3Pek@+4el9wD!&SxA#01c<aIF^{J+
z8}hr8FDwUgKdY!v3`ZLwW!3zb57l!#mxc>BZ!r5ut6SFxKc+TPSZHF*PT!LW7ym(Z
z<Z7NRUDVJleLru}Y=K#w^T=z@7ZV9``%0;c(UiL1-94T(K#x{LnVXmqnOapBSHsp_
zzpoRKOOMTc_1KVJG2DdG2}l?J8TL{eFYT{t{geXt%YR3BGNU*E+qEiB_4fVQks|*P
z6?v<tyoi+MWkT7LZ@Db2Y!{`?M7B38CX|66$w_36awo3oaP#EMtG0#NgAW>`WBU+5
z*^(EQ;|XB)1DS1Z4q8Ef{V@5&1t2bL%;E1}V1}jjGvQyf@!0CEfVt{U;jiC05Pxx9
z#VH~MqqoQ&7ylMdM2bf+%Y$HIpYIhpR?fWqf*}L1f+eY9aZLaATfT4wImdq*6q34R
z!-D7nl>7J*O&fRt<voa2PdKVLQ0PpclI-(uD+M0~y6LHc$AA~Ld=@(wb4LNBc{3$A
zBwX`w4kBhtdp3nBa3YWj8Ga1clz*Sio&U%ZzqFI+llwdWZdu$QWJY1}kM^5_|Ie2>
z)WyM&aTzMzNpy5_=t9Z}+_8h<VYXt-L6xB-lIY1vPkO3(-)5${+X2D^oU8{6`Jfqr
zYh+SV-agwu<uzTZj!FJ?>kcLhaS28c%=l?)0(n_~q&AaefdoN#%n{L7nHxD7c_W@S
zz?_KW;r?_!ZU5HU#~=NTPgG;ePVKq0NBOs0JMOx|Ey74Kb8=MWv*HQb()TduGdJIq
zy|wSLGcM{VSngmzXdW1Z(YEo_tG#WH<5O;3*L2C=SX7jV8rj2Un+mVL!B_4d`#k9W
z%9b@Mko$^Gz)`yRdmg{=8!lHO`F(Pyi{w~8O=Nem0!he)z^CQMx%~!~?$fva!x=_X
zFA;ff))o}jKK16+#p850OeMtWdt8{~P|tofnYL4kY!xO6dYs=wW@PJ9a?R;3!*TBU
zYetUrc7Az_LvQN_$<Foh*@nSK_mTM$jQbq;3NZC=K&&jzS%GVOnJGrgb?Wwac;uit
zyl^jsrE>$(N>Ddlvl<Op9l$jNsl=*~->cp~-SqmgjWVP-Pe#u1=>)moP~=c8ZouQm
z#_u`K@;yOj8zpSI#^*6&Kjc9;Iat0EwP5DGaKnzl_=fq*=`w}m1A8OwO!S1=c;npw
zttq)aFco@E4zFa%%QkJmp~sFqn~r|CKq<JQUCc!_i8-WV{=tkfvo}ZGlrJ;~Zv0S(
zIuc#tkd{eN#!4D`Tr*ZKNFaQ1o^Pdzfb%r^N?vpTHIvjVd+hJ?QE<bB2X^)_TY72n
zg9VK+i$c4@YF%P9JmLwKf>5(58MgB+A|8w!Aa42d7Lv;CEV+ghg;x3H<%C>UGv+vp
zflYl&7S~H8u9(p#Zv|z-^0~<T>P$FPWvLVbgtY{d%=g$`?q3Av<?Ja64}UotsHv57
zb!G)d{yU9%5ntM4=`234ZNNIR@{OTjj-MyfN8%X0m8v0`d<2%<B)*JdGAM=@orY-#
zFln!YBzfQxW;hJ_EBV78eJW7&->#in+ot~$A+6jps=duBZIc`_Wke?BwjBmi!qQiw
zT|+sW6}Lc9yJ~m+=>Yx$eH`2YN2`<TW5lo`Z-&9#QrSoj!ZiB+X^ot;gWY#`!J!kg
zbMGE?x%bC@mdal`v4LAf0PBsSkD`hfAX(V3cj}Ppee08m`P<?&)b(3%-<l6KsE7n#
zNp<aE>acN7CfpBJ>febCDj^o^KI2bO$~$=enJw_*Kym^7chpoEZb*$CwI3I<AMl(o
zHqks+aKnK*AqASl+0F7QIzvRbS)|`)?l}+*=2InMEsa|OXSkMfx$VM@!ZuuC04Sv<
zac10lFxZREA7hU&Ud$=}w%r~wUhM{>-1(_}h~<5IxFG4rK0;eNTny&cVH|?PIM)#o
z0+e2xgwt^n^#w<cObhI55_JFJ_z3fRF_;Ju$RANV9pt_OYPAWb2n|}YY&RIJX}l&S
zk(~F0Zwair*i0;XM;FTA`c0<a0riGFU-{&bZt+QkteY6~)cEqH%CXnDXd22GX#MmK
z`VY-|>ll?Wwy$h?FpU~vC|Z99CJ^66zH@ohe*1!{hRI&P%i5_`+HS!7<+c`EOF{t&
zc+LGG_052*hY%Z$4oeilmsgD>2!TVF-!rYQUFCzkL>8&of8*sc5TxyJ13Wq7;U74;
zrv5@d4G+fVpNlHxK0_J2DCg6JNfTWD+EMm2D@jGo&N4b^$Nw4mUkR!izmSTau)TrI
zKKx2NU1x5XgX;Q~6$P9<JOX^@7gB4~25Tx6UEBcO7AtK0bMJyOwXPcrhs0Usub12)
zv@6n{sj+=6LBD0L-W^dQ04eEx9{ihEEqSlI9<C9ZEYJOS+SbF2{2HTR|BK3sVQj-{
zaC&tmT^N#NFr_j4InC8$HB*tVe2--@R{eP=^`4Hg)2`)Oqd#%IPQI;tt1V$~L~_J8
zUkzNn3nz9o61#<iyIEu!v=}dCPp!IBUW2ttk&P-8%g36jpvM}A2l9ZL;-yBWJg^M$
zG|B-hU!uXsv^9SA%xC879A50N2c3&(%?0-<ZE@HugExcQwei#3#&Kx}x8EAGE{PVL
zmoirZSbt6jhw@oOKCS*-uEv^TX1ZC(8zO_~_&S2Ikv*MIC}!=pC2Hjwk(t?jUgOx8
zh3c=L9`k43QA3*j7ohMPs;l+ow39zV4yH)_xsCVBPh<^~2hE|}zY>-yPgXy{iEKh~
z!+I_oMyA<@&LHC>wuQOfmb6+P7`uB{9PZ<-5o`CicZ_j3N93sSH*?hxLVUY|U+(RD
zHL5l7YaS{cPN-AlP2_K$q`Jr@b^{68EdRzeJM^Rs?a#x``T}`xmd+2t`~z(BgZgZ$
zikeY8)Oe`{Lac7<YyIEo{(?l{%_5VPzZW?&7e>?V6ugZpmii`cWMyAA@v6Pw^D9Jg
z8bv-~Bu+a9Ebs6*!xFJiKlf$HCmH&|25BH6{$i2{M_}xSQNQ!92B+NXgu~-GaR}?V
zD4cE@4bj!~*9G8`4K}0-&fJ>Cgv5uHk)5lvJ4}>aC&;u5ZXci6BVUMSZ*f^;!xfXb
z&DTYAsJ5L;b?J(mckjDmG{cmVaz{^Hwzx15#`1qd()T=KB`}sM<|*KYd~7`a#Cj)N
z_soUIa##ua(H-!GLXTha^N!XfV}MN4)dP*|3+4}SMS+mm9cZB=dxjW>1nGYd2Om`b
z-YR==C3WZJ(SH?UqG)2F-(v>*Q45|K+5YCS`%+aIIa*MB)v(Cm%*C07e7o@NvLHqP
zz{vA2`uBCNDcLdxIX|?!WOqba2C-NMk$n3ta&2CCvpF72DGp5PH=maphd7=MpQId*
zv0RPDYA&#ryB$L)fgqFY1`3uw)oesEH?f8(3!X!R`fOjQPF;4p=jD$9t@mdI&e49n
zbA`Rf%+Vo#%k?q(IfS>0MCR>Yz7CCMsRoB^W8=brAQHt8I&ZmSE%hxC>8qQ9>}JD~
z9p)S^i>Ie`2Cef!_Ub&@fny~;-)$M|TTTM^eGVW*L||s6yLT>q_sj1RG5dA2E33CU
zVfhC6mEZD0@nXR?#*kf%epGKktz0)o*8fkhVMmj#b_jMxvRR#(h?`vnZ-&8lHPx>!
zqcxe@z3m~F>gkv$`55|`l>bJTHhsU$&!xvN*jg#!JB&K-Kv3{Jj6F5;9Z0OrvJ(vv
zCc*;_20gTK0HwdSlJAM%t?h2Q<!W~-B85}z5p@021TChrI<+(1qGg?f4mlecXYDwS
zod{lUu4lALi3~Ndmol0U8A^zNF&dK$9rq=|&Io==&npSXnEUSwsX0CSxtA30W|!3`
z*Vr~G`;697K?e*<u}TT<2}_e76ygLnehYx3X9^h;L~wiNRg9OI+_<E~?|(((2DP-T
zEo$2z$Hz7w%Ro$Xui1?$M`MY~7lo4+-SXo5e|X0;U3`rxU9kY){KC`a8&0D_k&Qx>
zHRnR^D|Z}DmDf_{pObR5`=m!`%(=fU)06aj+txOROXcy}Zhzd!=W_h~=(E)HBeVe5
z=B)_bHQR!*ABVB*L}wd4!u3aS&9{4o1uIcT4K1;rqBaZoijw6;;F@kQ2#a<14Ih+U
z(0HHLJNphHiz}^3-6aG2C34ewm>g8$DrlL!MKN3!pZ04u_TF|^D+iCBJbe}-H1mxL
zf`1GIgU`80as&M{F-!Qc_v<g<vR8r8z0mEWdik#VR2qdp2Z%(pdmAYzRk(#|i{*-q
zcGTfZx8MJ^m>e@oPa-u1ZYT%q?8(!yiec3IaQ=+o)iae==;!rm9=;~*o3J^3g=iww
zJ)U)SNwfrf6}kz`bMQAQenYoDaf}{?5@Bd$9`8{h4{H|Jdk3DEd{B`oR-^-WYMuCt
zjNRYN=GE%(zR9*1W{h*o*X`gcYf?LR^mkoJw|+DnOYHCR-leiwBRMr$I<o6U?lMV@
zEon4x?@0HXv#X+EHFuT_`uBoVH7xAmWu>SYtgRe5owb&KMHPc%d{fDA_U4yfQ(tVz
zb2#zLyVWWQs@be$o@oY`gZvA$XA^ff@S`5B@qk$Y3?1&=I8V-kRkGUGIoaRu)Cn&Q
zm%i5KZlg1Tn8CB17k3sy^KC*2T>hERj@uwa)-np+F|t0WW2qKX=QXuPZPg9vZQo!O
zU>gNLE1yj-9@5P>nZl!H65P4eI||GU6tgmhg|pi$IFhxjwfnFW+wFb!pUf`hqwP@f
zEKK9HpzSEYFvxOCzE-o&OEO@1^L~w@odt`Zg>{}k068(zUB6p7F_KF_i9<^Ap`R7$
zbzC8#QQH)^O}?`i{r>BrTx(X)$?_CJT7)lA#g1nycpnAg!-^+%uP3PhZgX7s;b)Y<
z$8)pMC;N4MG1o|ZU&MetRqu9+3It|!t@C9(;;Si7Wy=xi;bM;wj-4iioz}E*XQ%!B
zQMtx{`nM%-VsPh%(XD{aYfr_tQ2T_%%k`+WPiwR&dq?aRMzfebGVwb0@--7?{k~qB
zxTGqba_rtQD)yUcnT~*Z@daAuO|aA)2oa(IZ@a{@`5(-6P!O=_IqF$?i2pkx#cTJ8
z-&pMT)oZ85KCHIl&J)4Zrichu#J=*%G+D;2FXjZN!<QbkO_X@p>kMDY3CSuM(Rr0r
z=K0wLcsvEk0`t;HNJd9JI97_7cEn|}zp|ZTyxkqqxE*M=*K>y29;MQm+9!Tdw{0ym
z2EV?_j*gw#3Q7R}O1v`V3I^lxa__KJ9DW#~QJcL;9sl6k8(x=vwL27b+tgp#CdjJG
zuHc!o`dDr}$VRxwePnqG9>Db#Jfj8E_Q_MJETY={`6;nn3@R6xNEcwM|D`1$)inx<
z;&KMY&`2q=*zEWR9Z;vG!!ydpl)zux;Y8brHtMT5%P|7H50zRv=6|KS-2K(T2P#Zf
zXpC6taR_gqKTR`9#xg*^ua5el=?io)d0MFjp1dx4Hjg5~GFIvmN^`a})5O^{+|s>g
zFtS`SXRyjd?45aZ;1CSdTV{0|(Uy#L*owyKDXaGw%oH^Hsgp@4cgK@=VjoCf(5F!J
zJkl?JJtP3-^Ree482%qS_SV7YAu-4*1Lq;mee2iH1$Ws8<hHCNMh4={7R3UkT+NN%
z9n{0Qf4{MNijLo#;9&v9)bEtG`WgclB*#*4dN{dPV%uMa@!7Eq)ZnXeKllIfp&vK4
z$vLWX7935HV9$uv`kMK2NtsPIM@*uuNZcYIXaMqrZEvD*HdDU2d29Q&LIskpK~9!5
z6@ID0X5tD~=o37ik;6BXApv-L7tQj2CPI&CFWR<eN5NH?|2%j)R>!j5P8&quvA))t
zH<5DRPe96>cgyhPWIhG6nfv7hDtc=e5>y5sfTcm|LCIWq=-vi|^!g@@8;kmou0ML-
z9syAu$ZH+_5D(?`sNO}l!Hq}(!)QP70=B7O#@ZB8ZpAbSVE!iXntx10$!5NVEQ<(d
z!Y__IuhlJlOA+0G(y0N)U&H?{18X#$GnVZnqV<RqRp*pQ%V7dDim(%juBHX)6x2_y
zA-;{E43|S!pTv+C<%W+Ao^S8SlCv+87yzNz<CaOaRVdrMtGDby()CCoeEQUm8HWkR
zLZ{|wBDmkob5cp0<U(&W2-9Y|ce4d@r|A8uE0ZFd*de6KO;`sB)eXSuzRrDF!d8mh
z*Mk+{-`gHZC%7y-4g8~k<t7!fv8S<|t&0uNG%~NGRCZy$T*MnY7sJ=rx03GM1I?bL
z9*MOcU0>(%-L^1)?P@e{yo-=~$~UH(H%lCUWexewO@BZVSIe6-HqbG=>S_&2+jYQ9
zj8b%p{}3QaVvzamw}80sW}j9T<v@{|#f4v6orM(I2Q{yF|2w)+VCP_f@2_j-gV&8q
zfPPrk7<E`j?>9Hb%5G0cpYe)~LO?sGOmOw?WxaRnJt_v5<QA_x2$GjyckFVAS=V)-
z(xBI+`mEh^Hsx(dLm|qdZg+6a>f^IJw<p(k2fm$Ez6|_LDGHB=Zl|D;V5I@*KV-AZ
zd@gK_if@C3Mqxz$%NVJ|$5X=tY(9KM>}76QOM<4OZV6(1)KmAV{M-4nG(h1F^hqMo
z^T0X@&-|Te@O=_|bAF0z<PU8hRbh(}o(tn_l8<#?<Y%cD^$Wd6f_x5gVvl;`!`wdB
zMmx9QTY>KEl4%;pf44>dldLu9X_{2_)H>;!rL<x**3vXoBU}0(YBdfe9VIOX5{_%T
zQIAW;`xog{jmV$x<3lb2Y_5qf%j|EKOw+Fw05fObn26qcWU?tCGgkQ3_XqviS`~VE
zQljnYUD#PWVF7o3SY5qJ@9jJl=|MP5@AR%3@!|`gpTBvsS8uJbM}Bq5xhi6Pe;hUA
zEJEYn88o>gL|4YT7LF=#HE;g)t@2BHNSX-ppEgE5F2f^{+Hvhc;HZ}qnYmMhPcKX{
zYaU0=yF<0JS%3|MAZ=Luj(J>?N%(aA7Z<}q<fhy6x|l%Pl6q4w&@k(%5rJ!jkzVvE
z<z*9OD+y7@U*|J_hT=KAy`NDH-A)me#S(?BN+DkrA0<YCqe?;1J~nYt5+5%aSbt=H
zV5^OoOkd#LRooK_xI1oq%*vk}K>HMNP9rVk-sZ?mxe{i5mc4q6N2DuTN7?c;<6(MO
z@(Cy;pbO(1{3_t}Y3-+ds^96~c@>u@5NypRZyuXGXHKYXA75hv{RU3hj^5XVwxD}f
zTQU(*_S{Ci^tJ7Z$8NL#Z_tRE32ano>Dd7yVm)^+t#{fv?asDiA6vfd-zzVhrCrl6
zJdvZ;qrmkEPRz{GAWi1brw%m->DK+(il0Ty6vKaZ*^Ku+H^@kSp}kZ(r-V%M-(KMg
zwYKdLbd;e~Va?*wx%{Zp&$s0W^=c+`>#drPY7~QSay(DxbI?4yxKKRL)2+IpS>6Dm
ztrG!vx1#$Og%K^kfG&bMDr3TyZ0D+##G@*>c_dY?UX-t){a%*|sccN5FkHQ2g0bGj
zX<5vpj^6ORC0Of6LEBt5n|@ZJ#J@Abphac=3|o~KTgw(3m*Y@0EvUxi5lZrUUzPiu
ztC>SeXl1u)msaz2Y|7_od=Wkitycpkxb?PR$*OR748YV}|E2qcI17|2v$4gY8-f>_
z@myD^B`>vcZCzw(TeyL$^SWX3zfD;lKeW_F2-Xv&lu8;(Ijt2q`0=45qaXO$x#t^H
z<NRoZp_~VgebqzKAk}v6tqaNEZaeDlwPD7DC4A_CHj|Drp79yhbh@=~REPss=H4;+
zRru875pv>~4~$AF`sLEVxRRB2$(?Vu``BkcA>OK+$RFr*ZwjVi=FQSp0m#peI}4zP
zr8(P=hOWAeSf4|ZhS7z{RTs!x)Qv#QO(O@@z!$p1GTxAhu|<{-QeyNP30)S0rF?|D
zUn^A~(**krY>(@J&7Wxj4G}<cQ4gV%+^cM>uBR~a&OaT&tAS44LW!E8j0HrTUc-Ik
zO~oMp^Yl#`^9$M1PbMdm<ED>y&sDqT;YIIT;3RwHVFZH7RQ|6V1prA`LTxDjcP1RC
zAcC=aw{byS$DKpR{dUP(Xl=voVnp%b0?}GXsG-}h%g^(1?fviD0RV&tp+FP4iVjJ0
zh387)5jr!sI_k}tFR^MkckuqWt=fTi;Uf3LDtHR7fsDHsTl-~Fxg_=2yiE7`l+K42
zijrL@-Bl?-Dv0JBt$CrRKsq%VEx~5EYXifg`z#agcmtQgV}RQJfE~a%^4uBkiin3h
zjRHeK!FMV)8n|<t69T?#H=j28@~2fnB|H*CKEPtAHheU{@1Gt*n$R2I{??&6{AESE
z^9?0^Avz*m;APjr+P7fGj__(99-l9WE7nWV2gNuUYi4U<j@wOO=4o#{qKm^|Qr?k9
zpxl)AZYbS^F|GQhcKf&WYB>RJa|qc_$VZ|wUc-^*V5z}hiJ}1KB>|kSzImgTQ1#w+
z#m!1ljEOqmFjn0jO*PlK+-qfc+Uq#M%M6>skBrj2+hxMk%2!{g1<aW~mjQRiKl<+L
zUFZ=bTzX4VFkrp`hc)lItrhz+*B&=G>rcnhv!y;HSRPxM6lcJo+x(MeO^*ZX*GKYZ
zwBYWagNBG9V`2cEG{QUajViftJTKz)cnfmOH!2ck`2x3QA<p~N%IpCh_tty*dPY8k
zRy5Ys8@r{wJnk-zpB@eFIB9Q<S3WVyPI~RSubJBXUz2ld2qL}w?Mk77^cnWNl809O
zsQm42g=}<oHodxgHrMy^Tm@)xCZL#ae)Tt#rshe@5NHFuWWAbI_EdfToU<WPjRtZB
zY!q5M(BCQSKnj5oF^bM)RQjn6zBX<}a6`d2<5S(*aj8eb#W|gmqzqJ9_#OzPUPIW!
zDiM!Wd3d8LbKD2}GLd-S8|+WAFU8lWopYd43mKC$8J}3ya*`?jhfu&wkR(I)uzR);
zGI=yT1(a?&uMM1k5mVmnKpIY$;Q-w7hw>rmYz4kj+q#KJH4g*d+<A+{bL{6QR|71*
zknG_EOHH-w`iM0_Y?6p0AhsD9Nj~?CV$(c%Na$^`p%pu~PJgNVV7dA-dY6#f#s3F7
zMN~WAr5JbjCIK0|{Ww)ZH<mZ&kXZ16&Y0!j8j%0rSJoWL{KI#nYA)&u=5v-!mP@L|
zdszMTqZ1|2!py?2D*K2dy#IEkv>ztzWb$gjnDS+hF{HJd&jfyP68~t)``}+d<VV+R
zzj*wJ;>&ssj&Xv^2#axYjm|d}K%YT{rBtvMUVDoDwf8u7jeKjs<_K(E^}){Ca0A4`
zer@Nz-rTZAN3cB4xX;T0XyU@hDr$P)_kRdEBfXv(P_|U52c>cIDUC>N<ch)J+D+!8
zCye~v-x?j6A?2tEGBJe1mkBFpbQ<2;*TjuZ-(JM*)k?#?b+<K7EGH<4RQs~SwyptQ
zs#R9rG*|d1K-cw3lV~PXa>4RC#|7|{j8hV2QeWZd?qTdMNqoII7UH`|--GpzPsgg=
zj)J&7p^0}O1bKz5P@IQai<o5i=2RApolBv8&t7M+X5STSt@Kmx8Wu;usy6E#3B|5H
z5H<yc91e~coBrggZq*sD@s>6r403m!mMGKwrl6I8#n3`qY4L}Z+DG%Ncq&knXdV8|
z7Q3u6y$y;Gn_~q2xy*lsy<ZidwN>@~?wS@p#<`#+kSuH>uqbd^jK*Q40F(N#H?evt
zWt-!zfCjrJqLNu$*Tdox8;LAZD@F=42M2Gf2k{7(EZqj-u1yRQIT63Glstw1x)L|5
z2Ts`}%lCjl;8Vy!ZXh(wg$b}N4|-9M!Y`vGvV*bZY3RQ!-m*AyxQ+Jq&&)egW}p51
zVm5F#fG;xwOWKqiR4~cn0|Oxk5~T3BAJ0vt0!h>{mDVV$;o}6C;$BfQ6`2s)X!jIZ
zDh&2$ApQ3d-5^J<FeoxT+ha|T?|@b}FCMCc!^U$%dHHX!QdZ#)&|`)Kmy>B!j6}#7
z!U&>o#S@T_6Ix?o2jd9<;eo+Pjw68uVpzZ!S%~bVf;*2-L5M>wl00r*3z6_JO{!hO
ztO|Q+FzISKV%5}D&ZvoOz4v?@)LPzBkL;rxOgBx8evC4XSj4RAx=CFxL%?@jEUP?9
zJ2(4NHV3FZmsME;N2?N?Fl&KeIE=(EUGYft_MGPR-S4`Of@IyAgi*us^$PcAZ;9M9
zGU4!hp3zXFO~MG`;L?>BTv%W+x{XI9>JxeL|7QN9abY4F#Ei}P$>2gt3_-hP$lSKY
z4PCENQhytje#PG=rShfO8ycYCk%iy}HEOlaL^2q1#C!p$_d;TK7g9-|sLjG{<MxJ^
z2vvw_ug&+<_Ms4p7+c%qY=~zbx<87*#@_jLpp%Ve<TX~4qD>;_EVrkx2h0wQhHkYz
zi{I<rn|&6knu%CYqD7qQcH&2<7FJ2!r<r7%-yVim?c1Sovs$so0^+Y<{=nH2^G}4W
zwxhIm(ij5vmzNIw**pw_uWQu$vlp+sO+JfHvljvZ6ho&4f_Sp0P>x~0OJ2ibct;-V
zKDc+ecWus5yZ=C)eAj0&Zo6x;jd2_*+X)jAFIsFo4Y_S8tHFSS^WU<_vOud1(b)kN
zNJp&9(R~#2;i9E?smL#kfL5CL^0;2syb9-yraFQ2R4v9A86Ko{hzOhL)6XtdUQZn>
zLd-$^gm3*WACAE>BFuSnOj=Xh5rG34gP(8RJj_o%ZfM@ID}}=br_dSRfY8|d$1WwZ
z17K7Uq#j^lg)l{83P)Sz@|V9R9g|7!OyPd2oTNM5F;2>4mewf_a%hDDtkh*1%@)<R
z&;d|F-jr=V(eMvL)?w60S6q^q=5|X&bg9*U@})2wdC~Lfv8ZnQG_gQ**bQKIIK}&|
zLlnuv^IJkr=BrW#f1Qb*WIF}86b>6En|kp0-ZVFCEBxo6L$Jq)mViajR?P!FQKm*B
zEk8`HOF^!jD)L$zCX_0|7I0Ji<#u=C19ot_d5cO{heL-$VQaTJ@8ofg2rR|ZRli%T
zi<~i6d9vmTD>K#?&pG|RLi~YWldbdp=JHVcabgv>BYXLA07&|*8;MKiuJne!<{>gJ
z^zL#?R1UJ=Me2iZ12c509}ZutT-xi2_PCI#K;TDdfk+?glR8NoG1YJM;zS3%0!v0l
zYwmfLt?FBpEdp2_#NwLIXO)^0S2n{gWWR-YxG=GizQgS9S8A7!BzGJ$nJb&oN|xJh
zcJBG?c2W{vH5ywC?i?N~+lL{}tcQCRQrNd^7$?T(ot5qtD`BqI4SPhLYZdB_6gP{Q
zyQGrS8L*j0@xGUmRm;yz)QFs{1In=if*HPE9RTSsuF^!Uw_GLrwmW7!XAkic)IFE>
zNg;-WgZ0B9=FAt!n>u~sij0)BLOz93pSIciuOnJt&^z5g(sAXe1v{68J3W4~!cJ$d
zEK#|{mqkbd8ekaKms63_o{OGDf=|a}mjCvC2AwLJaK9t&izEEjW%!r<$$y;lCmG(d
zja18sX5_E^aXVTTpEAg?yi}&V)gl$P`h&_q5ns1NE2iYVv{;^UQJKwu(NfINW9Ztd
zcdxzrbc-KAF@G)o6Q0g=E5MfLVT(-Tp(CbX_lY8eSOjO!m_IGrCVcuXJf?%vip<e)
zR+;o@nBPg7c)?DtPYEM%2!I<$f3#<vvHxN-2m{7SRsxx^Uq4vjLbcv^c=*&0rSp+y
zI?y%KO-OCiJBBJh$)GXH30L-`-C@onDlmWXl%3j2P|wf4Xe{vWos<Q$3dHJmDQA(X
zjrCjo^uI;5TKy7{{hDN3Lr=zo1DgC>LW(80OAp;@++HPAUQ;LnEZ{qPv{62W6jg^b
zlus!7nR+fPwMYgNC(L}_&+3zKF2*vUC|K0WZ-E#-YZzBbD&3M_lzPzo5LDsoxmY9s
z+h7(6lKmL!{1<Y>N~3TgJYBf8lz*6Tp+~F|B*b~B0yM_q@9l$7p+~XCvmwDpmbiIC
zXm3tSXSSha&kgdZNWhF~xPf@ThqYeHH~}s*Ba&gTwuR?%gjD?()3Q@C)>G$N4g1Xj
zy6F5WWi3W`M9s5^DyrotsvUE)&iumISp_x_r6~N*+}sx^H6fE>c7v9khMrEIqrBTy
zed%DgN11ChI$hcYd(y}Bn8LjsWGZ)EN`(nga2Y9I%6LEj0RW$naq;1^dwNJmSAZhv
z(VE*pG?fA#k1gtx)Sv^Ilm(v2s0dpFS_|_r4VnU;$P`^4!`F@#7V&oukFoQn;b|}L
z40EF!MJNOu2BbOh>2Jm@wHmEv47Ru)TBLpZe9Xp@Y!zyTaFLLuYNS8=)v`R;B`d$#
zaA2ffMTa`jCIJ`UTgCdgoNB)bDkuF?z7vVeJ@MR*9R2DUw9r}}6=g_A=}<bR)`=LK
zopz^ms04*1y!keJkr0-BN|X>;ra$k?DvnAE1)G5l@j&$kmC*Dlb+AhGF^*vf-9j@6
z#-Z;hN7rKX?SU)jpm$AnXYETP4*CPfK-;1u%g{qD5)B|j_92+|M}WW`;m8|q7wloc
zaA5ziI1Ug-O<o1l;`H}i_kBtJfuGJ%jFy~VUONzrj@u%HErr_><1C#xvk-r5h4h_9
z95L`E!5^Zxhf|6RGZ=>=9nRro`>q+n)WVv=GCI|c@h(q)#8Bksv74?*oMu#IKOV2u
zXkZwVv;s&77iiU+qt$Y_`7&R#%2ClQY$QVy^DLP<*h2V7Op_J@yAyo;TvLM0fg$Yg
z%CK{H_GLdo0xg#s9A!)3cLREA;8;Lu1f8+)>;T6;L5I@Z;*u;UJUz;P(-78OZE#NL
zk4oMUp(*WW{72^o1Tj#mqODn@<_q`0q2&uR1GsOZJ}eSb9o)~5Vd_fS;KpkRyIMxW
z_`6!9g{O2wm0LN%d}ut!#skQvUvQoJ!wIC;Saw2*v7=i)4+XD>qH{;LP@&*7GA@g$
z_sZ;p{p6?}htFrTC)-m3FHiCq>2JF*8vz<kIdNJC#qlZI3Dv5!J`3DAf=Nvm%^Uzg
zL^Vtq-@#GRmsS3~UavtM(jv-gbH$SwP0JZP*|Yg+Af3dx<zJj59S@(-h+`?G4LFmW
zgQVJJ!*B&pZ2LMh^afvSws7Jq0eQt=NwbUlcS@D>&q<8ZBZ)&GP2A)maz`0T7$5nw
zl;^*v58=`chCVfVFyVU0#yrA;E+PPLOhD%O0vU!D><{)M<6H*(Bc24TlcxYD9~1lW
zS=V^C`Rgoa_4v*Ytx#-!nJn-a0GfGDC0WSlhbH^$^f#GmT*oy$(@bnR=Ml<Q!vyHx
zBh}4c5?gDHxX)<tep;?O7m`13qHcfV;UPHyV5KL|?QyCe;ll|%-yh~`32}h;W9Hb0
zK0ke$%4hM+HP=(amA#A5d6MBn_cx)LUJ+Gk2ie<;+uf8&!8Ui)Rf!yOT>W;L!Cno*
z{~{8_^?Zky1!9rO1V!rC@!8t+|C`=#c_*AnQSZ^~EFv~Wg+|e%<)2>#8#+{~#<f$z
zT5dILO+x|5t>K$~s@sj|*`A2JCPDqy+F>p4i8iHvVK+bE8>z*4NmB`@w;}`M-mL-h
z)MoN0QFedQjK_1ojLogBG3qRLe*=t6_?d5t3aTVK_b}4Kb#0yI>@w~z9ou3pXO$lM
zs|CBQry(0MA=&)a%TfLp5;9K%uWL2`-~9mSZe2|8CHlzEyEU5W;7@sp9F6)QrJzjm
z*6HUXk-<#JkN-7)`!%c2JKlYOTl>+~r3NZFUoE$Ea|@Acg%NeLyTse)=iN@Vv2^f<
znAiy*gVkLm=u8g4SVHM;=ia2u!8iI+uU5LmG_WGcI*I&^4mn`R#GY0B<k=v1JG}we
zPN(evjRjtjcg*y*r^T@+GW&+X^QxkAI;zESkV*6X9t<W-f~>U>XfYubAhQgG=RIPE
zwNEe+Q+hV^mllh$g@LEu3pYP^U^S+Pq~06oo_N(~C8PQOBanK}NKDC7OsOXj9u*fo
z9A05K;;s=@emK=w4261+FbtTOv;f5=x*rF^^Mc~?bGO?YD&-H5qj9i99rCVB{)Dbw
z?cmA=<81#z+K?#@mI9@dnj3PAtMa*nY^8T*pXIc3+vyI<xp4w=Fe@q)yhnT1!V7we
z<esRL(=q)S>y*KZ%#`cUcju=vmRE+SLoIooEkH4KkJdH<1PA4{*gydc>W&+P*Dot(
z2orDt6dB}dA3{)UC~9rQNr+S@ybP$54!b>s{$64+L4i5K6L0zKQz>(5<hi9X5`>ji
zNRgNq_!&LeSB!!<QaH&Ltw^#C<4IlSD@`}!OEwQHS1iB6jt7i_{6W}SNWy|2uu~JB
z_Ch_B0ph5iTQCFQbIsE%>%XimjHzDIzu1zy_sa%iltc5{Z^-(uT6VhHccrx`ixQ}a
z5TcMamO4V;yJjci|7yF9)X3%WLzIni(w8P9RPK1$l$wdljN~(M?_YMEu<e8FoSVw=
zdPi5k4xMN(uq8zl;!R313EFn_VVqNGXKXDYvXGYCn9c#TdYb;_YL$CG0*>?uLTLvc
zwXt+uEqzoSf)ohL(*nq&gQmI(u~aRxZ8;m0dbM2#3#CKz7*LXE>pUM=<e5sQ@nGS~
z=lOS4jES@5RE+cb6e9Wvbx!tVXOPTGjtS^oYtY3z;o%tO)1{}=k7gjYXM&K^Y%i*|
z?kbK{H}QajF9|&hoG+U6`VDe-d<D_B;C`yE6YgvTUbT*Y(Gq46l^ST)Lor`8sG=^L
zgg(ygJ*0h}Hj!$)cLb#wWL=*Zy_??)-=;4hw7ee9VQ(A{%P#-yLLRThd*5!~li1yq
z>U})l+9fuue6-j#EgPbqABeXnM@~M;%Gb{&n=ApmB&9r{NXXBZ=Rkbj<kONd8*<s2
zT`_aYOwb0!KjOd}Vp*{YndDN#sk=PcWBZ)x1&?4(nZuKxh#`AQax7eq5{{8-1zo(H
zzSVoPA>9dm?Y-`Jb)KpM4o&e8q$&ZYA6qrmPlfE3tyDI{ndbT*n}Sa(o#z4#dTZuy
zS1v$02TA+7`(f_`f(7yWQ{#l+atv>#Ldlmr<#O&uXZRrWLq!GCl?LHPwq;65PYqof
z>B%}W^>Ben=o%2DWTIvFr{b2Q5O-HGw*a|S&tfAvVKeSdzvVS71IT~rUv~K80D~iW
zOHlVCnxGl{=VV9l`P>7ynDc(0BHgBa)Mtl3n$V}L#_Wwx+zeeDurTt!Nvc~{&;EHh
zc!`90#4V`0?{+l~lm4SPt;Ol~-ArZESO0XrGeLPATs>rK{7?QI^+}p{9q}8-dx*R;
zzkeCT|Bw%HR(9-Yl_pAqN+K@q+AYd4)7@5<*uVY9A)o7zcD^JJbAgXZ*$(_Ip>t*^
z+Jme;4lWh0Itn+%xEq{@8Su&*NI^Kb3NIPzhSXCd)u4{rBM)dbAHrEKB{3XG8GSNQ
zBNwvSCMyjWW^6TJDn#pGul$6$PmYUkpMKWBW2&9xX_Jm2!e+q6Pjy~s`<cMIi{c?>
zfqUu&n`NiP|D$08{<ny2y48NY@ly8iBCz4(;r@O5U9IH-y1xH@&6dW@{hixQKXYkk
z;6v{SPR6mw7NG+pg27c|ZwKj8+ju}7<Wz^9T|t(Ki@^EW9yIUwpBoi9guI@bAYbQj
zBn8-TPekQZG{>E=7n0Ja*N)`sg<xw~92a*X4ql(gbztO>;7M*{bA&d6%<&Up#?C2N
zsbHO`K_Xe1HnF7E`=@4Rl%qOw5aJWXsL4=TAyovMsL1%@oG*O>a!=6^(_8uRAISHd
zQ5q-Zjlu@rcM))LTfdw#rx1r0s$IZeUW!h+Y^%#~lA}%}Gkn|_@!==egl{_^p2nB$
z!shiVfz?kYVAA0a^PT-+)2;ROmUHgP@KZ#yj9~j=edbe9mcDoQp!~HZLG?)+M%{8_
zlv=BX+=qqX%d(>MUj2aUh}I4{iCRwQ%m8d@C#oQk7D32u6gAK6@?=G$FP&lRg+i?a
zUypX*$S!2<GU7k$g9@*w>-Y$wmtC6V6B0cO0N^XKl_e3jo@~Z(*Jnq5NjLRRU-`Go
zmCMLeqGqG9Gs_ZjF%fWIM4{ATyIf=s5TyO(K3xsydVN$=89ntO?F~|f(vI0isDn^0
zBrqvzb4GuEZeFUrO^b-IS*~&-(Pf^is9lUIj}uf)t06koQ*La0=XOtW8q4D;&gPH>
z%<AFqxcJ`HxK|sqv1W@W9uQ-Ig^Xec4ElSo(!<goZ?U=;zS1VfAsqGMuLZ&~CBLc{
zy7D_qP^cyoEim=jX}1Rj&f@4s(_U|&6SnhGajA{_Rs=y97pA0TivOi%mR_7m>D0|F
zz3v*8Hc^*zp|7VFd$NAH1=6M48w$XagUL`pq3)yj-0|q)!ac7*>?*<%8U|gSzwsix
z*z)~L@NNRk>LVmwu736YDpf(S9g~&=&=V&eS>Tj`$d($=AjZMSaa8s7c}Sxk1j=Se
zxHP*+cW<1w2{64#CW`qB-_<6})({p34h|ef*z@^IsrThK)Qk{doV*90C;?s*3Pg&{
z%*<iYabj$dHkHo#$p*Z2tzs<oAwN6C$MNBuP<qA9X4pp;-_zX+9Dh@&Y2T`B2q!3L
z*lMKD;;egz!`aD#r+h%9SyQ%vctqPN*Y7t;9NN#ZjXZs+U_AGRu!|oiOp8Y_B}UE#
zpPYRFVD)M5f%Q1S{g9-29cV{!2nUbBuSUhqi}-2<zSZ`fr-nh;yXtF4)xNb{{s!|$
zi#~Hk%4*OSc^GB9zw>y}JQ6t=CTkKgm5p+7UhA5Xi0!l3S4|pIo7I!>l{1&a<S^x`
zs*Q;@hM{oUM?8Hr`1C9IMNEp0k-jKm>VGVQw<l{n+%)rX$msuA3ISRF+%P)V^n@At
zBV#et#0+BUs)6Z#s8IZwe#&AS5ma@`xZwk1F$YLO&EQGO)@?~kwYDkzwX_h#X{&#B
zc_;HuCAKX&@^p%7NcwWQpVXYYSPDj4(!CqqQwE#{^l)D4j518<;IuMzahJHT4rN?U
zHWCrtkecKXO(KE8KOnL21+39W@RxI~R)p;7KpnAvixD-S`orYe>aZFOnYj2&U?@t%
zCA!0r2;zE?**rGi^mL#*2vSju2-foJkpyJF#gRm!`e82SR}7iyh){#yrlsVrc_YZo
zd^59t;SeOVO{rQQB~%1VM-1|bSmK5{M_g3-nbX>#U2F<mkigYF;zYDR8w+Eof1UcQ
zhe(Fybc>8-`T74V@{PO3;e(h#r;V}|lP2wJ7w$NMQ+R+!N8m|MqM{ByF8T#sd;}RC
zKRykM+om`|OZ^W#4;9TDwcf*ul?N*{ONb>pI7^7P5#M4!5(1vDZ(7}%L!Q)F8n*<{
zDW-vEj$u{Cxcy{Jf1by%dK~gd3q7`|RA{ZbQ6hnYduj4n?_%bvDR@7v8G29NxBi_{
z3WD>zJ(I49uCTvM&@v^<g{Z-!eS|<g(m%=7uPB^*bLk|c5=9pO=owA^Dt*k*V!yfd
z<&NDlA3Y^k@`5_B9LBYB$ZFEd1nvp=NK%laJe3@*_wKtQ_SPDr&|w1&d1%ERebZt*
z(1Z3IVxxo-g)`M;m)Rt&&{cTJH)O)ro509J#q<-Wq8H!#i<<!MR&rXttNppCzx<Pu
zodsDA>2@J%V@W*<!$l^8hMHSp%p?|#+a=EMyc~syXvm<ZVFvT%T6$aXPjDW9YmoGl
zPlc#42W}?^a*)mNOa5UYx*M@DAho9q>qX!s{ADd__&Qv``88sz3Yoo(74u2OCOmA3
z{oQ!ld3%xYm<G==@@Rb%ftQUz214TA!xPM`eouJ76n+BNCiFl>7MEg)`_ctb!XQ3a
zIZiQ7ap$KTz8ge>VocK4pua8v##NgnZ0Jd5a#3j2vXYcG4W{40$HG)go;U`zME>&<
z`Tm1#%%W2~1G$pjZMH#n!@D|{ahO14`oJJEd$Y<$P6uyHM<8q{z!wjVg7r{Le&I1-
zXgBA~rY`<sDwo>xFrm_o@H^fYn@-$hbUGBT`lmYDFSGU8z1MjB4pMmF)1&7RoPY5d
zF)xZx=Prkm3NsGIBoBOqA1P16F!@AVd%)<!U089Y7Os4I<S)4#E^`OZnS=X<uIQUv
zKI6X*-F7EY1<qPnSe*m?5eRP|{m_5?vB19Uy#K=gfXMg2HMbRMLq`ecO7*Gun3a)u
z60Z|hSkT*ojc`LcDsYZ1-LCKOuF&Su9oJ5<=&OqTtL#6f<sy({W$!v`eWI$N+t=@r
z#(II|WUol?W7c4kiUrG2W~%JV(OVSlB6wg7awU1-!HYcQedV9QKU%COVP@ZO|9K(4
z(H*IYW5|paHtk5hps!oe1n{@b6a*#8-cWI;TFu6G)e-U20=3n)&BCQ#iz|-Ak5@kH
zC<sqBwM<4Zwdi_EEX|n*`iJlL@44&N?=F6W(>U5ZOHpc-3-lpYZFr<ZG~?Gtk`Yk&
z|MF~W*j6lAS1qmV<)IJjjxhf${+Eg_FfP5hxm@RYZrD;li{yZ54>>O_j$cL>-+uou
z6_vH3t?>dLp01;676(&?)e>TK&ZG?#5)w1Wz0EDC)7q+(Y*|_3;pCl($!PABqFCjd
zSGztw&h!yoR;3c&88CmiNZMJY6>MxIRLMlhX_(71$;Nn$ITeyR)8;BbAc5)5+U<95
zj)RfHcS&w9{A6}ZX@pl5?e4xfK7My~^P+y>(KiJAT8$SgW3{^x)#=w8bk9g!_`~<P
zEwfuz8TjNOVigqUdvNf#qO88BW>@^mpQg3G{m8rNQl-b;GdezjlUgY#VfmzRr+uY_
z6ObtJBxeJ%9FG>)P8`ELgzz3ZjvaT~pLfP)V;RgyDa39*CWl4p{Pqj+WF&PHZC8-h
zPznIlZ~_^UI3cV{&pr>wiRZxYIatL#))~~6WT&9TJkUDEi|nttcw*6szQHY~%wOLZ
zYn$f^_C6|4(;Tx;I=0(nYF*>O50LvQdMTh?$1nTyc#m{&V4**tK1V?&FbgAM+;@K>
zLFB<Uhjsak*>no>jN)L!HgV=OCb*#X13c>g2HQQRBW>!K0^-WJpc-Tk<+*j{WskIW
z_|@rX4L@Bo!s&(>GftEvM@v-txKtNVMvcBUW>fJU`T2}UX)F<K8cKnWHr~$oQn|9y
zl2Rum?Y|2USwbITY@(mS(w|_vojm`i;B)J?TjiZd6I47^<A$faCA<$S7pQqxAS5RD
zdu=$3c&mjXUaD%HF<rU@P6*>XTP}q40vpA?Jm(7GoqUU1cw!Pwc8RhD+X^ACrZ;m(
zg-pcE!E-EE^`S`vFQbrtl=K%W?3tWMwjZNCmLogtJ$97?sV}8JrvrjH!SW}x>);k)
z*{G%JdxP%oq|>Jw>|>?UrC+#Mffd3Ts_Ms}XIru8ukxzKY`AtPG;u+gA_=Ny$UR`U
zZY{wA6s9JhJ_q!;M#N-SB|59TSF3esY;n95`8>mfCA12o^4hGmLW^~GMDL=fTW7Rc
z)PWo{f4kp}xXpEj-ul#*^r8{(^<dZsK8BHU`UR`f0_JJTo~{3VD>c>20Cet69qG?e
zh$hdNZi!n0q+%$#`d&ATo~&1|9^Z<ta%-FQgl{W}R|*WAnE={S5XHYgHH@q5Nha{-
zHz3ZlO(#orDqg$4F_XK!AYFRdPR;dILGnV%cfJ*;A71ASFA;8`KeOo8Wu@=R{?5Dl
z0+}gk?svZ;m{<B(y}HVGfKdwgH|IeTq*y`R9g4*xm-~X(Wp>PJmOXnNp4ui>*M%wJ
zdg=PbTcF9*y?t%_Hw_>w!AsYA2aPP%HBsAK_91tpz$WixY1Cd+r+5wA^mlFZTBNsE
z?5&Kugl4$eg|$w-38kP)P1Wld1^aQ)b<`xC7AQ7d2Gja7FxJ-t<*UNX>1UvYjadAV
zAAQ97x!8X@R~@H*HCggkJw~jaPN~27s9;XaQx-K!$4`HWA}?scLe2H_vWnfnPe(p?
z6K*IM^FYWI#HpReIZ`q`yy7Sdaj<p>WNbRrYf|)7I~>Zb7ZP^;i_V9L&&OTank!c)
zAg5NIFg~@^q^4^EB;bqIE%7K~*!5-(=&aGV;r*1}LibD}Ar8r(M^VyaHQ(bi`4u3{
zQSvQ_Q^@(U_o613fs!KOB-h+~yg}f4=}0pF{<$r`UsQ4RVp(l3G8c9s(<UFEO_;Y3
z0arRhp!<D}J8M7JNv~jt*iyJ<++m-Wm3sxXuHJsBd9Sq&upY^F<aL3~@Zi#Du`%`d
z$U(4_WX1C@D{%>yV<vyWn_F66#*q~hD~boxM^xH<U76-6buDmvdg>1ogR|j{5A2r>
z9mdN4NW2<onuBlm&RTY4LAxZEKNWmYeJ!=TR}|F>^H7ZP=sWS`drp_wsq$4sHyc|K
z^uhgAoCMVf1kifHQS5TK2nwN%T1!5Jg0o`5v+{@_K@~s`fc?*iBQsIqmNh`egW&$+
zACBp<WB<C>D#l&*3@20F;yqhSQW+UZ;tKtf4_n1L3rZ7euZ|O5P~1!odl<}VmKuSx
z%#-&k9(IQr8#AHT`BF_`?6iQ-76twm#ODYCWN&7Bp#7Z?S^|Er1t6+~|1YlI0lJc?
zS=f$k+qP}nww+Au<RqEc6Wg|J+t$QRCiWzM=6&y7>;B)WwYvM9?%un#d)I!dx=LsI
z3|fyV_4|U~^n+{%M2M&vz$$YQ6cbSZDWDYqaRGCof@Iy|`K_EU4?=WDl{)cqj@Rvy
zTX)qVGL+IP13{o2LKp?epalR22#fUXCMQz~0g2=fU8cZYc(KXdmOsV(wbK!7bUzz7
zf9oV_0dS^oi`&>BWABZUJ(y_>W)WOpwwO$S<pD?7^!c8A1^F>gg8Jq`i>*7p>(Wkd
z5lO3jcxA_27zzUH^miH-D<?NchL7aUhG+l8Am-|FI`<=j53Fuw|5bxsK~Fx5P@yt9
zNbzfbxe3%a*lw_powz#QuX4NkW1L{}_8`pH4sg)>+|a4l4~?XD=+U>36dK_d!>4Kc
z`R0yIMs?Qx-B?OMk>{bA>F?V=@(ZC9(Ir2i5I@4+Trx7>6Ha4pP_K`Gv$C=#rdVhc
z1hF(_HAT!<*y$G&aX&3iu4lr_{aY5{_@q7#i#0ztBT=S5q&FE@z4_HsH6ldYaKZ(f
z0k3_g7OCqRf3*2Hw_S&_#kM<0r0yUiTH#Zpd^JCQeIb3VOE>&pk1f!1k~15aBJlp1
zUz+nJE1={c{ZWDNr&01d?ocZHFrE=wHgn{<emz{Ple^NUEE{Z8I*A8YJkPa1GLo(W
zF(sZgJgtr2Lr9~|J&UrLq{cAD&;s`f6@W`hdwkXvL`W{<VwHG9!ZQ{HV=5Kbs#{~b
zmNS8Ej_k3+&5xdT)CSqt!>W8c9)0rEr5%)499JP>WU(?XS75UF`KF0GW#NU-)Hhby
zVtU2;;9j1VZ<VXJ=MC!;WoeK`ao_;YS%t5Cew`78&&o!x)hq}H56nQ{|7A8)9uQ|d
zgXobF<C2E~W8%T@qXd&5$*0Vn^G4}BK{6-rc+;>udYrr0<;BX==|$Qna&URM|FnO8
zhj<}-@OHO<w|a|7*{OVZ*PQWkH;wy>dsyp5_@Nmzn8RX=+iehn-*`U~mZ`4Zu2JaT
zwkb1GSfmoeD_~JFoYdFyQ+;zp9}qVwh?9-rVjePaTHhnJ9OH8xd9p)#lJc?>b%ljq
zN+!>FugV;r8t+-!!IC?loNoF3pt`!6W9Wg-CASB|pH56M@7cPW3l!3#>ZgbEa1Y_1
zPKX8Fp=7T&(cWJ=mbR}3OpUVSP7?A&-td*ux{%n5vl?9<=3_HC`R8FH0rH(pw}n}a
zVx_+)J7?b$Dxsx6sxUyz)(jcIuq$wBMEp*c)*0Z?P0R_;lvbuw<s1u%kQ-PB`z2;*
z*j*tUBV1xZ91=?tEQ3y1*T|_;CwF17Buce!S!xjcJ{F`<_W5-9IWe!aulyB5zig)t
z_r}<r{130j(64Av@3aBotl`a-4`L)9)71zbo+pV?Y#9igJ(Y~HIND7TL`xk}D?&u4
z>>|6OyHWf%_yo}1JVLv=XfGh2b9o{*0qgYjxmsFVNd}EIexM*c^AfF67c9%wfXPn`
z>+T*vS_KN)l^WS|t5iPlxDAlQF9zeE7T=qMY+Wgj(yOSHqyfToi^6%RcM;B8rGDiV
zJrAhVWD<OhYG==h9=p7XN+L%(ekjUQF;evij_<#Vr2{9wtc{D33pn{(xm{o?QRr(j
ztK0fXRI|?xNLHicAcaBXt#ky#mo*=!-MCRc<*!8Q&=w%rR7)v^Q&N>RkC6Yj9gADJ
zuC%RG;A8_RC*Z#SRv@scj<^TapAC0w2jh?D&QA+e&g7;5FXXsah$Hy^z`D8}`D?-H
zg&7*BkeykFwN^o`F)z>DjjDfDu}R=&U_7sQ<S5+`_wBd3<Rlhq#!IHDoYf8+w5iC!
z-oN{F8P1&V6b@!8qI@z5x0oMY+3cx2B96^D31DH!)*K~>%jy<{nH!cOQ3{X%#$g6#
z1_rl6%Ns}IQ~vjRjoGEX5C}g^>U(Vn3#(k~*%9?+Tw2Pj-shZh@F~_+USPvMf_F))
z^fYCg?p76kb|r3OQIOZxstAUSbRQv(niTw+3b#`61(x58UFje`F)X&&R?hm~c_d)6
zFh1k&SV!!(rGw%2GxFX@m*svd0{3<lKdi_FW2EbNNWcxkcf=66&5b#$eLdjt&apFZ
zWpz3&1oCE`l+;*uQI(7N2X!yAONJ?8^K+KQtp)TQ5Qavoh>Ib}U=Wdxte@R%<#Cn5
zFPJJ4FJVP2`O;B0gm5|%FOn~m_%HyD9GT`;AAf?`Xf=9S3-6)X(uqQAb-nv9PbBFW
z-qt1N&h5Ipj_Doc9v&TP5U32gU$v*u&{aIxL5b8485p=aC8L)NKHj!`@-b#Qzw`66
zw86rTy?v3|8lZne5T=j@6@}C27hbK;gqB1fbRCt*U2*;Bq5e^9)>Mu>6tV-LZ#8#5
zspATfe?xmZ_5IV;jA;1nS`7Y?Z_1(z9F1=snI-Uc30Pu%FY$B(b}PPm!T)`g!+%-V
zT9J5@-hSL|?mpdO?B$R95^95P`8z4rM!fomO^@LD`ct0CHh4hW^mGoAk~K|#uGj?2
zcIu0rIzc*>kBS%;D)#{6l*t!BZK$#3?@!*EhJH!_NKF<r-<M^vK0zat<m7&)sHXog
z3_)%GE!E&E=iz^~6VVdQ%w+Qy+d$~{LEhYGpB>k>TYW#}*!b44oOqntG9sn3wVb+x
zBaZUrh#YILj~w1&^;jw;0FF$@5c1z$7Q&|!H^B@LidJy+P-;p~fXne+j^~tRBP5Gi
zF4t4xWR>xk0@n(Ol$r@w$AV5FwLB^_^-os3nb4nsp5Vqo{O`YLMVp{tmAv7TQ$W>+
zf@umqh&5gPYKjV{q`Nw;rOApgU$}|s5QP>%#3Y1NgckUuD?Q_rs;V&N%-0Qw`6yYS
zb)-duJ%;JTCqDWl0d9Uf%rQofSI(x(V4KCJ>C)j-n8^kr1$;bBOdbxsB^?z44w9Ph
z8VtdvnR+_9NlEPZ&2k_X<zVlzjrp|B=QTia^8zHA^oGr0{)>|G9O9Yt$l)k%V}{^j
zCD0^MXFCp;j~{G|_Y_*QrG-%Q1_3-UwH=c6k1(~(then~fN&yX&n<KE#sOM$ilo8q
z62miHC-M%cuvw8cGxN3p1H#U(W4(^~t@YYb7h3g1E{r3@ei!R6*VcsC8cJc10+s@n
zAZ9n<xS#cWr5P||lh{aL<UAtqx8iibUbQ`_vuO`gAkP0()1pY@6G+MT&hwsRBc<Hj
z3CqzR(e9Z5-5vk2dretmk^o9kqu?nxD7AXO<aMg$`IOIB{}Ps_9^I!xbm?P5EEUuv
zy_J$*N>Cb?1iY_-roSIOf8R_!`D|ZfzZWrR=6C-+KU*hw%%d-!3_92Hv|Z?}MVIQ*
zND;=no8~W;Ny-?P&RM`(#2hF*;bK+m1@p>BLt&x=c#eN7Hm1P3Q+wyf)<~5kGxjTu
z=(6@Cc9fMIPO8N0eg8DKbwQ#BOAQ*f@u1BSL)+gRf*jGR+6W#`Y)TkB-poD+mO3ro
zrG}Kk1<43yQX)Z&Tz`s>n^QDWHj|#i2G5P$nyO-*IZG3k1!rm;zPjxE^+5tPrfw-9
z-#7r<HXS}hpS|pFH|zlpmaP>f<s}fO1qfGNG@lXEF471y@{k$tkKW5Cz`_k`I%hxl
z<>yBdOQakW%)d+@AiqpRh^JfjUSI3!a(VY{aW^1X4vtEMW~3c>-UuS|(n+Rhe~mds
z5(|_SilG!X%+14+*-EX1liO1L>ia$}Yh6Hg4tW9<m`DT|IZBe05{b42?pSncCWs7~
zm4J)^a)Px$!lrdNJCSvJST&QtVO@udtIQ8#LmL6IT|{^qU&4Z?n6wn_+o?~h!m=!*
z`IFdnhHbvib7^ZIRY!v#*BWtkLr(cKU?2IyAxi-@S4nd0vn6jmXf>6uA<G<$W21mF
z5*-sY8yc2gl10wyi9UD9N`XNM#P&dWhs9S)R2}sBJXh>rNr2>5W$<mALOss|^YhTa
zy5iMc)`F)*9j%oi|L9Wacj)S~)Nyjn4edBUIqnMRg;(GPANWrvpTfk}dCH?X2MGXF
zfg;*=0%tZj<u6@TwaMukzk^Ogns@-Aop&QVM^QOEH!H^>)Z#7a{3)V%V?NY!9r4;6
z>;okmsTayuUv}MLfk0Ae7yDPzv*%&?U()<Bmvt&antb>8Sv6g~Y{!L|T~yVrMn32V
zi#<c_*d!!07eHdW4iGIT>(`&CX(TtuFGDgAwW>yU)2+1}3rQ=Rn$)f6fl|OklVQ>Q
zwSVb~b8pS4)<8Lyl%AP?oCfkC2WGyIw^>oEfIl3FDGOR@0HL}h*o|pJ1_v<iSlY-4
zJ(AD##w+&iS-P~y@V^1#4Zp;i2s=S7&!R-~Cl5b8V#D|mazDG+`Pq<I`x%!OW3y8Y
z-WU@G{_4RKnG?QZccQ@YGXU|~$ona8f%RQgwq<P6@~n8EH#GDpoj(>oy1n3x_HTeP
z7%9w8lB=V%jP;q;hKz@+22%~wAJ`svo79q;rXcM=?a{v`FOOV#&`wWnLv;O`e90h^
zK$1Wvfo0}#iq+N`oeGk;Hv3$~s1J3hEzeO=W5!)_lC`01O9<$FfdKM9OT^ji;rfot
zHVy}^lW_!fKfpF1?4ztU*F0h}hfnx^8vZO9K3TO;x4$#a@W1i}SoW@ymD)QI7Ben!
z1ie?2W*~h=4RvDD(z;LkOkOga62zeB*Ir)|5)=u$4!8^)yNEswSmrfL&x&PIW7*Iw
zp?-&s+4<vzaQ%$O2nG-r?zDfvtswL;Hp^^JD+M0a(C1j%+yNy|>yb-|%{mB0h$_!I
zJua5fnRV>8b#McN&sH4a|G~5MQq$A6k}nd#P!8)H!(ewX^VxrN{+oBFk&rTjIT_K#
za<eIy({6FZ_$GLaY!GM@cr%emW(Fr$;*86ADnIFi*AqthV+dd*$(<DcLlif(!`tu2
z2~~+Mw6|K>Db!axik@h?&|4)2m(}u~oR7VXa)_kgy9E7KPrf#tuwyCmmNc|PqwalU
zu{cX@4^aX`sv;0#A~8!K%HmRt{;0p%N*#dmv*R`cR1E;to;e`8pKt!jUbH1=2nOxY
zH3hdVyI5n*002-~ca|eCjEVknn^KD-&PJU!kH=?Tw^sGW;&hhFDXfxe8S|11mJXIK
zhH`8kx<`gW@^^2~9d@X}qZ;_#b#ljNXCt~a9mvu3;%xI`oa-SsnDNOwhxb5Jpvl-4
z2{RpJVSiY!p~o9yAY99FuT{yAer(tn(B_h-jm4f5O#)EDWf$lTZxDa7Sp*4co9nbr
zu!6lNtB1a2dLY4mE5iB3?yXK%ePO;E^@D!I_)y3Z+S+4_k?sg*(Lng{rz!(*rQiak
z@M!ESdL~$09CRyStUljc`hpBkv*0H+x*he2r>fFY(F!y`K-2^;rlnC=ot!yZVsZhd
z%#e^a793#8h9gmE&4wdibK+OiYsR*5ii;JaLFW?}Em6sc7lUyWL1Eg$vF<PDajlV&
z#zZ1*GywerR4$FPz%{G0xHzdP<9ixx!0rK?@4&M(kL>>t*?~yP>;XG4*?v=I=wEGm
zjS=J%+Mk_d^pMj@Zv68Dj|-M7s>KA;IxR5=1rTf-In9o&2__$khsKg0av48Mk{c7k
zaoFfZO5=Pgs>wBG4U9*rMTHHq3w84g`en|!3mfGDDqm7^xe(NC^L*JKW3%%<56Y$s
z`#T}i$H3B!^SIZ+pgipJ&oc4h<maCjrwG1wEUu$h(3lJIo(iMmef595yT1Il&@V%;
z`T%!de>IHUIUmx@6%pEQ?A&MRudxU;{e~}V^NqbM;yM`-T8miH_H#z)TMs2%YI_ow
z^unNo=PN<~_2ut(nIe$yA8G=;{-#ZvTX#>Q&ZfqFb8imoaTwcN4?&`B5-yg#J+l0G
zuhKZ$f%Z-HQFxby+j3C3T3ET^s;pTV0CGuOl&vo0=3Qqhd>BWA(EA2x44F`HMrl1{
zpw8n#hJ6!qLf|N$4``oY{s97)n4o{XVY1aU|F|(Ttg&(*Q-PT@{`#FHZn4#fwCaNu
zmxdz^9om8&ZVo`kTqg%gPU6pSi}F$-<qbp^@Q;O)7($(=naXKW??FXWk6T!PsnfqA
z%C+#Ou$Snt6cY>6*HSEcK?ff_n;(up-p)yRzkI#CeVp9JcapR{teW{3TWC_|Pt2c=
z`8)qs5BEe;#_~7u6!%!GL$v5x@NVu?zo&jE+zfeUc;3b6>+Q7E0DkxyS!!0}x_h}z
zD%3Lkz{N}zrkv?b_k7XjvjO-M@OMD-cuCCme7HNk;quz+P5j~1ts)ovvDGp|1JJq~
z9(q>LT}`cS+>Kh*G+aA6iJM+|gtyge-T5+(cvc-sAt0Q38jQRx7>D)qZz1drdUqlD
zTRi(~W(eGcY_NrMI%eNY7s3DXF*=Y|%Hl_`zudHpz90(6ETX#y5;VXtKKTufw2EJM
zYIdK(uWUWJdcD}OIr0NI-^_m~2>|_}sLoF<o2;r}{Q<g(Yikr;Dw;uOnScAul`|a_
zqg9W2YUYbP8cjvT``o*aA_Len+R3DP*)#6XhE&8<_9H#+&y3wpPLmAc6{a~Mzf~<r
z-MwYmjuN|-luc&#bO2r5!GY=VU9U_t%f)}j49V*T%^lt24YF$uB`sBKg|<5AvzPZo
znVe>yJO4+$$h2pEWOYx4$LpNdPlW~ol@Fqz7rwL~qzwUo`VW}-&7tP?_i5iWN<X_1
zQByP5Y#Xnhj1z3UG1^0gqA>6Fxcj5>1_p2iImo<0;33n!mu(6oHubyL1H7Y^=Q=m%
z?}^Y0T&L<CqHsaZQT5k_+@u-+Q&?*Gi-VDS@TV1Kl(Q0K75>Y#TN$W@c!=+%_bQ0a
z|Ih4lj6Tpfn0;&jO4A@E4)Up_jNJU4oylA?FPUE*?-u>D`P2UAB_MQZ95`LIAdT!Q
z7XgNdd%3nA&z-z>P8#H7<?(Tuy|YbEnP8@U`Ew&H<tVELCwutR18&>5jjOZmBiXOL
zjj!`<z~A<TUTC#?<m^W^>mw;4u5f1DRK!OPQ94-!<(L2F%kyvy$ZnCF@k$Nob9Li~
zInATb#oO_HOz0i1Bw)*Wz?rc+=Ev_LC{p$_uCd-K{aJ!&8>NPWQ_u5UDa?dEY+=)d
zAi~Vw2M-jN)o5Sv$St^*mY5AQk719+PJYx`UoTO2u*og_O?=I5S<6GIKX%4`69J~*
zcY9ITM3c2QPakRM@QT~ELYi=z2e&m?7|4>BS4KyjEwWM8S^!yc9;4|_s|R;+5~YKB
zpFepe;vU;CBD?0$XS*e5=x@6*H(6QwzA+S~pD-_^59BhwGmB!i1lG+r<r#zijz!EH
zjoVSpt8I#*K<u2XpSxT_oSYFjn%=JHJt>*!ge;i3GAoxz{pt<lu-rstLtxl@=22Nv
z6~D;!8+Q&J3NS!-3FeMcnSBcGaT-YnRFQ#=5<nSXdG{d4UZSkh7(+2bNt4$`s{<QR
z)HNWU+@iPm*mhLT%k?L{UC#ot^u1X}rFBd-Q9BhoxcGQ2ZVKn=Pn#;3_W1Qz60DFk
zl>J|>s)#kY)bd{FSreV$YJ30N<Yc(YLH{)S8Gpx7c7R4!;yLjm3oHN1NhtZVJ2r_)
zI45$i<j$(AkA>t%I~`u75t0v+-T+DO!4^$$Oow}JWhY9H4}F{<)Svwxb$n2SKbwM{
z569@_5e!7V7i)gS-NeoOz#$}JJ-Y9^oO9tfqIe$^Mfkr~XKs9Bbd245pv4p#7MXuQ
zF)0B1n*#*TT2`x#V$gWhYf@D!l>fw8P_$;Nh`M>tIYdn_wP;vTvP%I#ln7vu^7Let
zYSI(MS78!Ju~6X3Jwt=Z&_97bi4Xt*Cc?4P6?MZwRKgiqG1i>#*x<<f;K+hA-6aGf
z5AHU>2iSJ!$#4ka8Iy|q<r;Gpx+rV>S$uiNzW_f3COIXpChh!9jGGv3D~dzM+kaP(
zhk*Mu{H-(R%EIPfC2*_8(m2WX?9DUMOPRUT{<~yUmN83_W%(dUdwK4YzJ2@8OtkfS
zW66KrrUxH`Bkzm#6{Kvw;hKKpICzkqr)WVjgfXeITiV!w21t-IBELi8Dbo>%6G-gg
z1>l<{LHbv%W#o4YfW4k*j`hynA^t&Ca4!~uhA9D?cZT2f9>`*3fC|i|cCSm%2m^Pc
zYT^k#pr=eoScjdVZlHh!j~VWcyc0Qp(J2KSG5a0ngk>nEtRzCdiCW@x3#5lIe)r}Q
zxJ$#b<;WE|XY0n`ECcw^d`GQH24D?w0B!ifUwSX3YM6mHO8RtdAElv<B9KaYVou0N
zWaaxLTXMj|J#{Tm_Ea?-a;X^du*u?T7|OC6RSjt!lp!pK64(|nQ7+di(o@vEO(PnY
zI@L}Dvu^J}L2qVI(edfn{)}Ai!sS%~Duzs2MtAN;?)Tuk(EnsXfX!I2o(>QS*|%nV
z$@C<rFvg+{pOM~O+YDR=1Owy)IfZCL2lX6M?}I0cu5=oh|D!e|T6Im8N^z$um)E((
zs0(TYV+xl-$!?kZ*8$E4u)t?uCee>b$0Pmxji`NoZ7Jn!lBOXDx{<|Y7O1KY@QLp*
z(naz7=kxirC8ze89Vv-A){KC2|90O89i;4Yo*^=^vf5o&UD+-xbRzTT{PYoPW*}^1
z40jOTh)IQqAYNDTmKAF(Zc&gMup7CYoOM}zvmL)c-bp1Pr&J6~6lqU~&&%+S=P5k~
zR{j+HF<BAUVve32H8Aa0%{CQ!AadfS$AdZwlPH7E0@}9MVNHz$%XEO?6vww&>;Q-M
z+Vq@Q)y=137Jkxv=O-~qs3+>JPZ#9OHzz~}Fu;z|U=0yO%f<GXqP*gtXb{_}UE|zL
z`M1&M@qH5U6+fSRm_9mCeTGopFxTuXAA$*_W5#y5ho0DKy}|$=3k7Jg%jhH4@3V~V
zi=k?Ko>H&W%$x1-^8=#0**nSLe7K~g7-lY|!lbE(r(@Y%8mtIOAQ+m*Rg9&Wis;~>
zmqA`YUw(3Z09SBubI_{U-&yI3S8RwuFB~?dIhJnvMk=L~7^K6Dz;qQbp&tytZV0RY
z<<j_h#`k{s2a^^DJzJ)34SV@l;eGf0iol~|nLgX0$e)x^K^s62uh%y`Om|qz8Z(<b
zZv0Z7z0BMnzN(2NWl=Aycb_G24~lKIJkEGdS!dr29fDwC++rG7daj&Kah9)rhG@>-
zmLri%*dh9d13039jG=FLo7}_{kX2w|fdN)ZfpvMAyZ>vqBppt~yC~EM4iA%HoCp{a
z&V^V3z!L;Y?kf9xf<du_ImHc;YX2K%A?N;IxaUO76(PpmAa!)u4i$dbj5Ty3-dx`a
zPzsrjv;Xg9wv$Y+U{Ovl-<1@v8aHNp{H&RrB@{ptQJRc^!&6`pBZ&j}xh{zcRoQPU
zP*&|Ops%PB2eaeB4FLe!7`r}Cg)gc~fIl}i)=tFYQEh1ISq!1;<EkhPK{`vyM!Z#~
zf;v-=FG5tM&1W)3dG?Ra3C3pg?oFv?Be8)k|Kxxi#ftaYn9vH>ikDFZY=c5RNo+HR
z-;;d}0mS!(Sf>(YA-c^_i+dl)PWDMN2_TY;+s3OHL?hxWmml3(d_n~Vp|tZ$-~qyl
z6mdg|gx)iPP0T*l>kDWN;#*Q8SWDnGlP;*h;D8)RDVTOSP?uZOvr^zlqc$lj?75Cg
zh`9PV`dai-e}#de0UFjImL!Coh3tjw+l7FsT98q(jBt%IUvBcem3WId_ech=J&iZN
z(fR)xFw)L6s*GzTy)6DZ{}xkV_%`u@)wcc2IkW%q<ss_A|IfI;tWmI<Ss9Be2>PGh
z`ciOSwa$1#w?Zym+j>toGz!Vu2e9HqY@P}Q0a0sLxZOFvzLq<zx4L0l%k&L@SuzT+
zk!l!F({ikJadG6cPNcPC3>a*zP2kfaRlvxSk>m}NB`Kl{l9YN<)?Ms}kD;6ks97h=
z#@fv;)3yx&4)&?qNkeY?pd!G;`mq)qHY!gPLKzb>N@ILChO`)H#2=z6P+zjaB!d8a
zz#idVt~=G5EAGJF&h=|Mja;Fbg#0mWWZz+z#?gEeP0cHS59)a&TlYL|yQas@n}06d
zu<N(QA1F|^?G&Kd)TKk}$;gslL`-C&jy8{X^cFgw9}b<Bf9z;&8hDf^dpX4DxN6fR
z=)TP}KO;jzD|tA}EdLINf08bRzT}by0g8atJ0vz6v*GdSJE&85b!@iLHky}bD6Jj~
z_?+#Ttb(LFpeeM<as8nTIn7zt(6jH+ln{ZZuOmum>(EJK-B8vl=y#AzdJe4hcvof-
zJTiRX#pcD8*>)9K)Woc$RrI}E_{+3BgxyO1!`8>Yy{8anL3SEWl#PXQ;Kl&1uIl65
z%1=d5EF(#<MK<+-x7GNKC&cTv2~h#oVR8$1%#W&vj8VFQ`G%L{xn#ropix^#zKNSh
zc<z~tb1rm+Wedd|nx{F`hP}N7nN8G{jQpeG<1~;sGS225mr^w}kNcmKB)N`=p_F&V
zzuy0WbZ$WVkRC7P3`puT7yy>*$gIDfP~QUn-Gf97P1vj{&NRw}SC1QgaBs@coOCO?
zcU;2P0U0CjA-W<5##fEcw`t4q%x${ZG*r5)!&*8E#_RBCc~udssY9c|VIXf3uecZ*
zw2!UNwPJyWAdP~y65Lh0MH6Z^0xw*n19eZ|X|i9S;^ID*`azvI2XHh}aV96nLvk-(
zoEooC76(loH0?`|+LBSa(dtZ$muaq+<?mUKTL^GLXjBXG8(p76=WfoD<yr1{v+_oK
zn)}mPKE7#V;i^;zhJVKi9B&OWW`eVX`YtS%vFPK7lpp_URw#k*rn&ud&D(8tBmHH|
zT`Zc3@46vDdoP&<6)>^qu*mpOoffDTtm;B+*TAADR2!#n;F7mvb$z*dfGzryE5#je
zk=86*?Zfm;hDRd<47?Au4>bUJ37jt!`7Xd7MuyH(2ua}#id{^iUMQmbZHC*PYQDv%
z1+Cm<?X`dYv!)}H4r}S8^ltqYw{ZJ7rIqmRV_79FfWYN79RRnCyuq61oqx-(9=hT&
z>EA>${t62!Ur;rvV~heL7rAY^%ISF7N$>|ZXWhScvj`@4>4>JO)sUZmvyfn-$W>eE
z3a0iuJo+w3#Oee~lW<Rasak{zEBCl6rs-?@6x?zjshlKV!p}UK{nEi@^F1OAiL8^V
z6S@=~i6OAGEI@%y#&&`+|2t&L=bUdysJ0{UU)!tA<IyxHnuEO=f+S@RbG?k5X4!kf
zRKKvdxW@l@;25S_-L=|#s#7`RM0TC+Kv3bJT`lkF&K$#J8{qHZPDpY1O0r|`;YMaX
z-^BQ*9z}{?0=Yzuw=bTdIhd&8(?1)MRq!~e-r^7foVB*Od5T-1$k_=|Zo(kYdk~ci
z+2tpzJS_lC;PN|Yr!J#*sSY{%k3~)0)P2M}`Rwc`!uCk}(yQ(=B^{h<euVQ8H-=<o
zY3+D2dNlpvdJQ1jjtXzEmE!n_e$w`O_u=!2HVvHFEfiClum2hM;|A~F0te~|lbq0g
zB?6#NGV`U$z(viX%JX!q)u3_ROK`}<#=PWlRR00-ec@Uii#}k$*xIc0Z_f;-<T*$G
z^fZ|oT`oO`=Li@k?9OMODdXq5cg(VB7cNVGwH`u06@=}+Q$*n{y+D%k^cj$^I(G49
zy@?ghp7&;<*ZvavkUC;{`?1V=V#Tyr0HKy7fGPuCyexJ)H2nBSUH<FaBCFn?3|;)|
zn;FW398P=(Urj^3JncdvbDw0$+v>i%rLa4-_TyX{!Z_cTFRZ`)y#=%8i&|wl;&mhR
z78T$U3Z9CJOmXF-DU?oobUJjCm4ZR=dRKJ%Go;QD4P?*h<M5te?<SPN-oL&W0VsRi
zE5GEKYck{;wQ2hMzHQhVbjHIqK@SGof8Z8n31Ar&6-dzbN$z={B8ghtUyv6_NxL8C
zyBRj;Nly(PYrDa=Cr{4GJWz~!klvSnRdt2OC2$@xotIn%|Cf0CoJ(U2{nr5B_d@XM
zS+^i3duK~Q4<4A4F+v77B|pCVzU{)<qC3d4l;M=ZKzH|BT^)w{t8xT|Y9Cx$y>(9L
zObWl#my@MmW}ugEJOM_iEvt@-Nv1o3p2<1uxf`Ow1|qVOqo3?-bhDREcGMH-@`k|T
zK;L5Et0v?Vmuiu6UtLv_*75B)wx&CRKk7a)Pj7%6`5Raa1R%Bm@k^yNfcR`25sO8$
z76C4uaS|vLEOaU~4;oI}ycbAxWX9~D&k%I^tKD?X0(DRvj(#HXWx+%0wsd+v4Fa|~
z5h>W<KE$BY6>vRL!Ou*21t}!uYw=(qtt}z0bZxY9gj!t<C_z~vnRw+-^I#*bEhDa9
zIz$*Eu8xgk0GKG+0(sEA*UuUMMrj5lh)cSs(*(-t%Y-^%jLOs@_CNHEokpJ0NvGa!
zmR~nS@CAhz1v8cVx$#ejZNy<x=6TJ~iybyzKRHDBl@oiv(?*vXz5I3xfImLU=<iBB
z^;i;0JN2L_xSJAFg!gQ*ktyH7Af)Pls2V@wD_^mx2Dn;?TMQ<CM|Z|z{SIwM=A5X_
zGm`!HCo6IMf8cm12i2e5b2d9maI9rfNDjP~pBV>WF64vpHpI&t56e->Hrka7^gKa>
z%b06Q6dZ?&I_h=&iFWwRP*0+PWy)m{NKX#-0lKw+uJDGxqZC>~Q}d?EFo`%8)G&8m
zCAEYCY<;T?)b^Bn<}1t||M%^JyoJ2Q3UyrsR=6Jjw*C|C8=0>^sSy%Th?=30U!69Y
z|1U7N@!;G_jm%&q_E7bd;%l1f4NbBnL>CTHRW`$C3_hT`tC42KuSrW`QSGRKP)GHi
z2bKUMomdK2>K1u~S*~QS)C3$TKy?hvcLqH=Tx}J48B}v9Vf=!F{pUV;N;RMgB|stS
z*%Zz6EXkps*LZ$CKd;RPQ*@Bi5FqG^cKTz-<+UU;b$<y0e;b<HK8;t}CCPclPsd_!
z^xIAhZ~OMg>xONv8t-i7(5wS{Il~v~8?bjvHIgLeq<GUKc7Sht1AaNkD=26mH%nPz
z#CssH*)AUKi=h)68Qe_{qpAnDmB-`M4Y$Qh55SJpC-}FoDCtV~p4L1R;fSk|e_%o<
z#zQFO`upL7W2F89MIbBMcIvH5s$5S#u$5~yf%oBOUoPqye{vziRAmU<Vvbb;8}z2w
znh2~MZufG-pUu_O1JDZyP;8A7V@_LJ@bBd@ys*7AX&LAX7Y@9#H3vDr_bUH>a=4Nc
z92sxm?Nnbbi{0F%rsjKl-St5sx3Q_1cmDPr%E2OLBwR#2L#;U8ufZazY*+j19hrrq
z`&CjHyj-_svUxQi-H#LLGdyS5X5|e*=VB(M-rA>Y6>(1BLUvT=HMG^9_~aJd4vyH7
znD}&ItFW<ZHp3M<&DabFmx*|)ua(-E@SXb!oH77;U<Y^?19(^BnB}dUf22g%lN*!Z
z(MjJe`a;BUmZ>cf1x~^Z*iybui3N@;3TI})E~q^RCD@jxz;;0{>=COcS`F{}2?#>Y
zCllF8D%*^=xa$LR>2{98qD-?*@R>S7#1{{brbHBdNq|1yi?gc``VY8ge?MM^0Thil
zTv7qXyinNtvy77v0+h0Zq$brF_JSk+d2L~oH22xX3)&_5)n=`*h<XI@B;@lt*bzZT
zD4^hPhd7Zo(%|3fwrx%D_1rI9uMa8&W1?7({Z3z^8eB5=WsXDe4!zKKsU%(MxgT#k
zkq^WDy^Kj7YFP6(=GefGvRu0wfUHu8d5jSI<nesO8RGx+TJG!qF=Ex)esmQU9X_n{
zU+v-a?HUu$g(HLa{Zap%!T0k6_gu^xBGjtYc7PINgmPbrS~E;W3g(_X6}=|jjy|MD
z%P}ihqvd}nzU1*nD}vY=HI31de1)q4{`7^Isv?&JmxQ-UG2s5Z4V4A7=pv?jyRdLR
ztSG=WLG_6Ya_|eBvR^tQ%PxoT@_KHQIsu20&~lRreK0(MoU~7a>F#mfu?h$a5pMNh
zc_H)uqB8FC<NoOEYIvF4N7Z6GfB8vnvqDQEH+4Pp`SPh1<DDjl>uK&h>8(T^d(9#J
z$}Xz(((sPAAAovby+8gPg~n^M<ForVFk8RbVeM#{p0~TB%U(|p;tKrBcz@6Gw(QmG
z;grHk5sUx*ZgcqT$7Pcfb?<Mx`jn3@r%j&9594u6)Ze`J`i3ntL5v25e#WV}f*iW5
zoK;TS-d|zlJBrA<-o8wmh_n{Gl#CazFRcFQ@1%ydg@9kRg$DKDq{f4vsa3CuFG5D#
zgUzd~5H&>U-N<ju>?Bw@XgRJCl-ZB@7Mkw#sRQ_RcaXseG5ij@fQoZz?TRdCnFGSL
z9W78(IxpJ>@str*S>vy4(DaLXdT$26gJ!UiEf2dmR9foC6()bn#E*G=X2Dz{h6e}k
zu_aN*L;(M)nfPg{WE4@&dhAOLDa$Ur-~Im!>(6#%<buE&Tef1$$ISxq2?<$Kztk&|
zwP#x((JyMIo+{LTnf}FWfzy4m*S3~eZKBCpKN05mG4|0ti{Gf<;pM)ZxbVcWX>x9P
z$$-n{p4rKO3Ive>Ca)8;2G|VMgn_**)g%=3>#;Apd%GGQ#pvX*B>c`AH{QW*BJBW`
zV2S;wNKQidVA|(Za}Rw3G1Rn)|2oNgkarX<Jgpb-2gdb>+m;~z`)uG1wtMLzVUQUA
zr0&;hdP?~xLr{Fhm%V{=DC9f?#!()E7H?Svp(H*NEscTdca+*KYrr*o9#j+>@$PpU
zMM}4DOsUDE_NghGd^NoI%`s*ym5mtB4I3%F@QIxl5T&Y*J=w2*2a`?4=b&$BsNWbh
zi7vjIrF&I7ZLK|XmOZZ>?Qb|Y9TxSy!f-t`YO=AC@v=4a_o>UaP1ffMkE-BZ{q2w5
zbV&;V|C2bmkfFpcK$rc;`@TfK-9iVG?P}oriuNaq@msf>-~FqE=xYq=HJfpJ>dJ`z
zZ{`6h`8Sw5M%&d!$U>RKOIn_9|Ih_nGDnFy|E6#MP!E=FXkMyefN1d5$ohInmBn9U
zKWjZhm=nRzf)9~mdqny`%WH7Q>IlZ)UB7yTKQo^D0pwO*0Gh^@`3rR|t^tR3SZj8h
zu+C~TJt>;b)3~^!887lW`{VV?Sk{A@(Idhbg|PMyNRMcO3zt>)>vq~4pSUdDeV_GB
z)`bB+uwwZO(Enlck?}^;@v&2ydS7s2vSB%q*3IJnY!33R*JjOnTs>oc`hM;{p=u1N
zq}+xNfM%_``9p<|*^)=rr}C3=Y#RO|T4X<Zt7SEEn_rr})J~P5ED8Su>TeMK%jtnU
zjt7F+ScwJ#Sf#d!;=8(dSb$i*VSTpBH>{U9{0HkR7hxmFe829+(0%m(L+jt&{tvB3
zB~kewS`YX?xIUBL)w~%bLj_fNf_$EgN{bN5+C;Eqa_M()6es;e3EQ(?`2WTB8cpS=
z|4Z$USpM`Q<?D7uyzt+tbZxTo?$4Sd?}7bX`?-gL@n2TB)cmfeFLl3d>-8bbb+cP-
zv*zsi`xDS#xT;Gn?F#&=KYC4j1z&yzEr&yVx4p5=7AO%+Sdd9Ud==j1Uclhr4w>h!
z#{3(MM4sa;jS~#36sWf$*L4$0tB!y%>*!F_d9qlnrlPYfUxn4OK6^nv>p|!YBa=F=
z=H}+Gf?JAG1ehdyjwO3;YP<jx-WnecI#i8MR3vpI{ks!fRdWuH)ed*$`&m4e&=0Ka
z@>)*0OjG10OL74fc2y>hn~7K+r~J->3#W&5HJM>kZtR>NVgqW@FXJp|hmBw<G21Me
zEfTuyMY+RB<(699=&K>fj#Uxb3w?t9cOItm&A`H=%N|a@S&95ZJC2t@(*;FRG<Kxd
zNKKVf|Ck%Fb>m5SDxU#eKiq)!au!WzF5rAaSpi-`+6ml1J794rbeCviF%+Qm!s3i6
zc}bfRXwCST_ZB}a1cb!{haZ|_y@vQB#7vza6m~2WIPogS5g0t`aX#aNcgZZ_sLRYz
zD5<9cJPsyKN2omPh5FsPx6Wsc|5Cub6OmI{Zb=?sS&C9TmVt*iMAw5)QaToGS#QQx
zS`_ZUf(R$odz=JW(de4^{kaLT#f0x1jIXdV5vs57(?c9l;u-a#<`qT`K8KW8+daZI
z;y8>L3l|ZMIQvw_c?dCe<i0qY$h~O&5{P9%ND@y5q0uRmE1O{6YrYf%QQ9~xV!#gM
zg8&&|(rbJ-fXYuD76Bchxefun8j;B-&2MkYzvI5YY3?Op1QO4^;8s{UW(1N=pbt}y
zmB_6ROf#PUZ`2#X@zk78t2?(%tA=gXD60kb5!GQa*S@HUOx9;{JncpM!Z#`Z7HU!k
zZcyAKE3D2PldyV4wp+A|7i)tPUu1NkCjdek;gS~8wbL8i#cJwj-(0Ka6*|?WaU<mk
z$}IkxA(H`^Xzodj2}8?y)|m0?=Nl+MUnrhayzQ<=DW?8V0mEcIIA_oM|M}*uVCCJN
z1qikS{LAyFR}ojPIidAdM7v6BM~W__VnZ8CdcJQRI0IqAqAWnh7(rH(<V|NzItAOp
zhJZLR61%IP<%f_aQ6Fy|>Y8+wJIGc#BCO6uS#lOJuqTKoh2rexun^0ip1y2zs=T&~
zmG9QxwczdU^XOy2keW{5Gj=4d+M$AL_=`TMilqZ0OGQE0fzY=ixs43!zwJ8xcY*u^
zS5UIq=l~+1aeDomjg3U^!FZ3UZU4G-8FWs6HVU7k-x2#D_!G$5YUw_dJ4rGC4N(tC
znh24aCgwldt`3Ep{JzHDNa&!(_-%>var)3d_2-ods}jSEuidaMdJ0{70;7mQdUd!G
z$P^+SK%BQsjCU6^;8n)#-2zprjAyi(nt9nL2YUg?08hxePM(oz8q#XnBF7@#5w{<y
zh=`g)uvu*@m8Y-fbl?7(>MTvgx2sR*d5lz&t*Hp@VnEv+=u!xiRv4bX__h|D3XwuY
zHILW+kF~%J;eV_J=kjQDP>&^WRzlqo2=<i}(N@kPifz*_0^()m$hGC*pt=SCWtAu=
zGU=<l3V%G_J+~a1$LA>Ym`DY3t&*7YvH8~>FWX#it|H&FKv;Ai1A>bOqy^oBao536
zC5lb6GnlJ#E|80KKh>`GZFOY>3TO6}f12nVC=Bs2Zr=a62NOFrS3%f^&z!$6TgZky
zu1YR!%sA-$VjH+7Bee$nbI-O5%^CRl^XK9UGPk%Cyvn!tv2aLiyp4b`iY2|ri#OCY
z=`+D40&WM2`8@m#Em1u|LxL&lvJUsHW+7(s3J*d%SI_h6G2id^uS-98mkp_vYs2|?
zuM?5#RA9J~obn<Vrl}4Gj8eN3hpU*@60~^x<oX*DFOxqzuYUlV^w5xjWOY`}Z?1y6
z&53D<lBKyPzV5k;TBKFK%GhD96l#B4<Hy!HTK9juy0aA-tvqGd*HPRHR;|V^XNJ5Q
zVJE)&GB@KdHtLyIJKHJ0CIpRue)ATQuHU#(wE_sYjhrDr^;Z}|YK$7+B>92#=;lZi
zJ^%DNlzTD6Kq>%YT5|p9TYprv=z8Q>G@iRS(iCgwF57(J13q7F6>X1ozWn?iXPdDr
zcf)n?UT+#;@lDWPBtBiIWc~DscxDL(SoM4Ff(57^jiN&6n5gj)nDQ!c=X_e;c-Q7&
z7T1|BZ8SdlfT(xAgY&BdV>ozj@m;C%IG%gcb-Vjqk7WTo??WyckI(n-m8ekT?IPq`
z61GL;>r{cxrRw+odI{d>s$PnlM0N`Wmw5)KC76t!Ar<y<7OjyB@T%CUP;<bQokOtk
z565oKC8KqQpz&f-r^p7P;f`a3KuSc$py5)tOzeF76kL}1Jk%}oBHTbY4TQ<*5$@ZW
zV0%XsC7}Th2Xi)`mxw-YP@vDI2_r0<{3U8IIL6OI!1V1gyWa`8RJH8_Nd+bSdoFND
z1R>ASlqLORKf31V9S={&hot`%rNSTxnA{P(yT21F|It2;{mQM!28{EkEATTT*YqnZ
z>{;Bu#<xUbhR4pr2DjBEanH7aSJd3t@<rO46+j)0MxmBuT?ABXa`PDoO!gKKvrtFi
zb)rzbXWnd;;p;jH$v{W4TuP;SDwW!NU2fLV)dLe-HvKDDMT3QuCh(|F8+EBcX>!92
z-f0D_OjQgY8XP5Z+OuynMadHi!>veGMg+NZ^5XmcZQzYdQSC=-NmJEv=ft5+77K|4
z0_GA1KuPnzmx4ulp>E}Q(O|!mXVucf;F&2>DDS@^S*EZ+uD99^R!x?Xq>RBWKf!nY
zU=Lvc@{#W-31wY;WT3OOp|3ilyk9wIq`Mb*AxRi%t*wCxLm%D%+KqZZA_n>cdEEHB
zDTCm6Lr%y$O|D?*k$8sgXkjo(zS#{Mpgx3hmcP+s{2Viw{D8z8#uv@+j*zT~GPVxc
zv`kW7d^J0#QOS48w_ziSxpm>86GrNsCK+Xk8Q``5HWyE_eAtVc6a#}gcIVZ~fMUNZ
zKvy;Xff$wUFxc4`T^HxglFL}qV(OiesoJkQJtAET46do)zj`XF@mo;tgaC2_K={t+
z)^TV>sJk-m%VB8;zhbk*&uv3k|KsgOKnwIcO?f1Dx>l@5lioER&}CYM<VAVO67W?}
zWSs49o+tTir<>CB=L<u4jZa{3VPPjU*JA`LD4Y~|QD)#^xVw!Rv+Js;O6+4Mb|!Z6
z^bCwHd@3ytaAeoOP?Wiw@pg|j!1C&D9vYrnmP5yA*Tr2WEsaXnUsXK6mO*r`Ma$o6
zl_Q8FcS437kCNRMy_`tGehLfxtC*jXRUPkKg}&5^)fOLS0(_q~RU(fCSQ-^V>QTQP
znxlpfdx7b1sI={DRZG;UbQIZ(KT~@8@rRgduDm}5tzLSz^c}o<53h!a0TWYLDjQ6V
zcIAa-CKlYSJNl2~^<RI@)G3!O&5(7Jjkvjtu9vfGA8|LCEg?r>rSf>8ZRpavAN|>J
z+Chu^^XB=not-2}Izt}dlFagmInLeK_-oaxfIb=f0RTS!-Z#H3&d*Wz&ru<}0(wrQ
z%m6Uy`tiWq+Dr2YlQ@$UfcQp1(Ctj!54c2iRJhFuO{|aHT=*|(!{}ZJr)@Kfjv7Bx
zUDZL?1H$T#`{Ma(Y{eS(>CNa6C??FFa^$@8DufLIg5%Hkkq-*1cP+*QW}L>=lTQFf
zEnZGq&K2NtnvR^iPY72eg`S->(SW!Vu}ZzbFxeOxWgv!%9~fN-0B(;Jk7wx5?7y~9
z3>OAO3q+e2c|)yjp_DvX0&cTRB~EPxb8L?1MI*gID2v3tTKFSY56T1aA&|xZ?k{#S
zqC@563JZ#i{+k_lL4|*KX(FzqR4(3f38w#pF(e!?9I&&GJXT+%my#zHADRIoL|$5|
z3@qCH!yw*pSAn=VfEjU^uUsRw+MI03`K->iI;-lQ67d(q409mtn0L<}5vm@@6wnk<
zB;i?;(5<N8l=26ajvAVv(nq|F+q#Z~NVzQGrKOE2HPT{fXfxLYm@JTN9_}vJBv?$?
zi0Fax>|_!6xmVfA_gN^-(rTzJ-7C!|{&!FMR0Q5tz{EI?eqPGej$<0Mp3q7mF!=WZ
z1F!{=7I+%)ziNbTL5C563kqHW_ruiw_gsxcRdHnl(*)S7hK(t-K(s)%fkurftoYan
zaN@iTDCm`DMfnwlRRg>Y>I({TD9ds%U3Mlho#6K!WM?qAP-vX^my>wd2|Ovw0K3Fj
zAX+^HW}Ont5x0R;S83A@q%T|?q-eK(fFOZ9MgO-oPW1!GLHCQ9MLP|1ZT(7(_aEQ+
zW+wtXk>Dvu+{UQ<o~G$eRyr#sRx=bUytfa+3q+I*mmu#_o*Zk@5)h*?t$3UC^_}vs
zXg3Thm!oqXx(#C_Vu4(0weDW(0H{?2Cek785R(PT=E7TfnWPa4m?E=oAfK?iE6~@3
zgVh%|;qW#h84W<%BsL;*?*9stZZl=i2Nexn)CHjxn?FBzNO3&wW*x&@6drG|5}7c9
zd7QA-lY>}+`m;;*>)O#)01J@^F?&_#q&biP6bX61e{1ktZ0)WG^#PsuT7iqyG}wNZ
zMtRIVR?05?JU9p6;BcwYs+z%%=~9N2Gj<EUJyXN6G=_axNL33+%Xb$(s+M32OAl5)
z%p7v3?y~N3Ev#Q4T1JBKooOwMT1V}2(1=i?V%gjl8@gGKCyP#1VdXV3ED3@-#=_Bd
zhaadoY@skV-w7Y|J9P)tozb6Vbo3{JCjvLyrE29?iR}&}<$p7s|CUh9km_9uh67m6
zmxKhC;|0>xR7-QW8#bba&}CkW?AMU%FEIJHtWwROJ1JFmw6N!z?%W{2{XFxriSZoz
zuhs_Ym{58!C2uPJb1a-y<~fl7BE%;HeP!c6#()0Rg!8XvSPb-kvtk|nmXu#g=L?Ww
zu}$@8Qafy*c;mP_PynMl<Nt_qtt!5E6H;FjsH2;Bi1nxjT#HE^`p;@CyJ0jsnidl<
z`{ALjr#o?)ch-56O>=r~-Ce1&p+_~<^Ci#S>^zi`CRn3a@AxYmb;^;jkVmHH-|WYn
zp<jyF!cDsyH78%9p?}!_Vg~)^H8S$wH;|&->=Qun$>N|nHQNRsKdM2M*>j><)dptF
z^S)H+jjc0g_dut0mYU%7ePz7!#=fJut|WI#x3afx>U}rLYYjt}#Re^jV}T|q2t`td
zZb{3drsls&+^{Sk3@~rM!XYrZL(r)-r~t)Tn~x_OUOcNbvmMeMrdm`_IgTLKw{`~j
zR?lqekd-U{YFZHx@x_o0!uw-E_N@^1HZ*kJes(K6UFnYYlphB0G*O3b7g{t9LjvYU
z=S?eCtR8R+C6FaZo%&(WXy+3gG2)z#+6m+Zs=}&dxWZl!*S`E4DiUp0ED@tmybs?|
zcdD1!%u1l3nvKe?%uv}5P%oi0Ru}<L?w5^exW9!rv@Dz+FGNTe-H0P58`B-j!*%MK
zjjVh66S~;6aqPF~ju9h~$tI<aAG*X^`%Go!++{Oc(aIz5?m}-EFB%~`X0yD;>$n=|
z;3<`nrx;U9+l+GB$O0wI77fzD$cw*Y1uQi5@CfN?4<$-806Vy$^HI?NId%Yo3JA!8
zyUHM)l39ug$b>2yoEfy0lFJ_+k0&s(S0;aQGI9>qT7*HHER?S8U+y1&9JCeSx$|z1
z7z;;|41!c;2r(ZkK$N$#;nT+P=*(-9ywr`ceSX{?WcooryM#twew9^nGdf~9V!h=f
zg!L~P@Ama|42IXv)-yVmR@MP{LaZ+<sh6$Dpa97O$iC8qNqaA)>@M`AXGA>*yh$Hn
zx6<iy&h_lA`!XWS47~B@uAP4mVxcByX_;tfNl=VpH9N3@$fSulXK`SVpo;a(j>o3l
z#;3R=0uvi1kAycjKT`~W^9vw?>{sN|UCglQql7etT?gG&g(?g73)BHvlA`4@5gL0-
z#q)LQX;m<PjzdqgL6;wa%4&R|7^(OtKR-IJh=HUEgwC770v7$&MCXv)E{Tm0g>%O0
zPa~5By4THh9o00S(KxRlJmq8pTSRonZD>ojgw*pZ*zjm6(3CYb>_nr8YMH#4^2@Z!
zCuq7Sf-2;AuhW#oBnSY)%*M!Zv4tQq_~=yq0rkz%!78dkBe-l>#x6>tOU!SgLEPXs
z<Tljk2pN5;R^dFp&?8dA$a@z|`xif%Rg}uJ)nQDF$o!N-P2aG}!#^fw;}<B1PKK~@
zX;~-RAFnIci8U+Nxz5<po{(upvl=u-ynbM(!%V*hM9Z?anQ8*!94dMHp!%T7C>f&r
zB9XN&5|BapD1~iC6kB&{B}5&Y0Ba6l+rlITe;Z~C>b?&z(l+9}hPb_`PT55ItSP)U
z`3?j8Q<So)7sw%d_w)AQ<NVo4knb(_4Ok`2lMY%p;0-t$=vx9(B=R(PnTo8l5~$VL
zcCepfjB)6s0ZsTZIv(x~5gE<{lmm@>f-_OKt9h>M>FG7Lzt>ilnDiaN@NT(P+)gIh
zDlPchSJZ;IB}im*n0-ygb{T?O&#@&c7BMkBh^;~yu0{Hkem$z6#l$Ko3~8#s5cQ8>
z=%^X4i0?6*2j}s%T^4{Tn@>zQu7l&FkAT6VK|JZ$16mU17bzL=oNoU=#@;fjj%Qo<
z#@(IZ?i$?P-QC^YVIcv6y9Egj!QI^n4#C~sU2|9R-}~Nu&U?>z?_i7`>%(Gob#>J<
z=kKZ6U98c1eMJbixRxhWeH9e%*u*2^@lnK@7Tzd(sl>FwYiOcn2!PP(vZLw97V>C9
z9A|kgO0@!KPlyK06&W2};0HC#GMWC`e6TP$99kpRv5+zH$ds~Vlf`kjPWZ759eCT9
z(T>I!_^}S;uCKO~P7@H;??V7~^jXlp%0$0C*$wyusr<yOh7~)Cf)Xmm2nATj$l+y=
z3!&%?@EtVb*vM1^xY#n7GSbq>5qjh@(m@$@g^RcqXd9viRr>(TW~qsfl5F0ndPhKH
z5h&w3^@T7v*K!fjH@(vY+tGm2w2~cs!u0JEZREJymjS)ld4D$jDI0(qU4K;6E&Q&X
zof=YfCY9yRvTMmjI`Tr0^79~0)_0_Oe_i1CvAKu4v4^A=lQD4SAk2zmnqsUD+wBz`
zgdpAd!dtPI?dtmL+~ZsLv{&S(OP|}}6P1bTY}>>4mETLYT?^ehP=>po1uQn1nUovX
z+iux1)q%UQmK*^30hYj51wKg>diwZyxO)h6d3?;Y<9v?+yncv3gua-4?0IclJs1Dp
zAA@^EJX?H2d#CHIHV?U^Jh0Q?+`uc?7#`amB7WzR4evZ~*4=pW(PZ}@zF4j!AOoqg
z@N(TPH%~XT#Ck2~&i+(&AKrLXUj#koKA}5yzxmjuow^G^2F(DdfNCqa`mgcxeL`w6
zT07$ijsrIxWsbGTRq1aztNR7ViMFUJ|3fu<!eU!{8S=^KIhhIYt)W59$ka{W)L8x?
z!0j6CaPY$7$NG`ms~>CV?6^vT1*HzrClzL>j?a3p3oLR(IF=V*1Z#rEWb^m-My0!#
z_uNvNNcm_Ef=uH86A@?+6RErgl2qBk?fNLAQSnF5MFQI6`pp5C=(5q0^g_`X?UN4n
zAAr+Gc1I&D8Z(x@NM4+)$(DMofMP#O+_<RL?`*(WHimZiW8ZRo)9#Xy+4v!bpM`+w
z&Rg|#CJxbW&Uk$J%E{uy;ACiiJ;^rfS-Yk6s*lOXvw{nNhTC;j-fOXRwm;9t2&n$v
zC0XJaV)*#9S1ajbR2}YXBXPf){k``0A0E=-g%QB8kAwEjoe59J>FvIKe@wYd3oUu4
zxNu3?9Ue>NVZE90-=M=*5ywa1_>zgoIbYbVTYm1GVAitr?omI#{rN%E43>v~4Vzt=
zL5FKbjsdV`rK^=^urlxTI})C|p5N5c&C72MdmC4lo9m=q4uG2RTc;gEm`NOS$jlyt
zJ<^!t&TU_5rubWNS8H+67THq^GLfdC$vaAI90c@%=>}y$nGg0&kY)Q#h0<@@M9WcN
zxre#2%Lu)WSkKn}R?eoy*~{XYme`C@)c}ZG=``RJ`}zE0FrQtk3+5E;w9Z67s}&qR
z$<BD8NL{FgSte*=&Em82?RewSk7BjwZ<cTL3a~>WTgG4YB;Q2|@sZ-*G@p-BG>Vq2
zw#Dkpav4v@YeGajEl(>KLm5jjLbYFqoDQmaBkZWIdboVogV-$Vjl!ABbRh6T?Z^_&
zRX+esi#7aD@=x;D0j}8j2|S78*U3Y0l2&|9AWZOB2w5*Ao+ngX`gH*}C0SgGxkOTr
zYggc*-*KJ46S6k%^rNXAc=+8&qq5LyPxhU2?NKW8&i^8zd}=3-Hy?hq8?MKy!epa3
zY%pJfT=2){(jHBm#7MFCgd15baN}FeFI5G+OaVdo!Br8D9af#aplR?obpU<U+hmJ~
zN9==B;hlf+)staEV3F5^&r%Ds{B@4(OX-0r9JhA`t#P6RQV0$)(8~5vh1qG`?r(>{
z9GTq2Q7NtbRP=}t=+mx(Jly1L$E#|EFFa6tZeL#B?A}{@t_ELV<P>eXdPaaRNC)UT
zOK-`sk(s~=GhO99#*DQ;R3pY;9jZO&RAY0t!wX2t-CPT`;%Z_bZQ6o#gk5Y+Hs^F<
zz^!tXzMpBC6bJ5*r=oWV7K1$+no+%po%*>EYvC@*+x8wi)tkr<6}uB7K^UW~$BfZO
z_=u9n4ZMT4-*!_ihA6to+*~|ag8|@~MmQ|}QsjPYE%(JvF+68QWuS-LUv3s#*<!w)
zj`S~6vECdf-N`a1P%wz19GK}U?Ei>g0R#DDB@5nqF~nuZ8Ep8aoyF?>?9z)<V%eiO
z_cNj{LL?8f?5Pk4y=GcG1BXN5OfNs&?stw0XIW2NyHb-$)E3%HylqG?KxIRYfzXL>
zuG(P+Sn^6Q3f;y}txYS6>AB6axwc105Zclb9L?OQ4;hv!kXm@V{I^<J6FTxWq8)4y
z>dYz=oJbkJA!dvRBn3RAC0{Q4kvJQiu(~7zXZ*1Imjs#_2wEy16^jvgRs#Fnk9g*4
z==fPItZ*E*9!%A`<n+W)z!x~xI<&9FQ+z3`u22S7J1(XXgSpPFwHR|s+t3h(4tl{V
zDdapp9E&@T_qHO!koG?UPk%b=d-uR;SJO$>2aA~d>1kFG|Dsz=q5E3YRcS-VIRUO2
z>XgFp0^Ne&yS|i~$7JL}t-Oj>LnmOilbzclU06u;@RRR|3b#)-7myE0Pccy%zROlZ
zb3BwNQ05IOH>I*V12mU7=IW1R)9s3vd*ISdtr*5x0z<j6sEvOj;YiR?7uI_7KGe9E
zye<$F%iG{~f@sQRe(BZhFo+b!$9XFGG>(&AP?%$~^_}r0DhME6$!LN-Xo4Nxt2F&V
z5+M8*90YKI4+{dE0SLxRaB3c>)5d`v_tczZV7x`dnoTY;jg6Io1oZ)`az+NO>Fqq&
zBZ%-ODnilyD-(8}QLMj55OpI0yW_b|D$#>%G4XwYD$MyU2)D`@)7^w|E=O|izCxmr
zwWEcrn|X3bV!x-w+a*|Fxy@8S3tD0y+tew5XZ&HN(qgw1FpqBh`a~}lNhg4ZJUN_Z
znWou)z4e`9AR`oR@yA{#$xG?#Ia`DzT$@eD=>$=^EPJ;{RraIpZ`0~$o!>UNelb%|
z>Hhs4V)Or?{JRAszF=_m(<S|H#LcgF3W`%zlas@XlY&0jyY<(Oo82Q_O5ex7<1^tS
zFWz<gX#<SAgWkUx<^w368tK{7pL!kFy&R)`h?v6h-dUYfk$fk1A8Ls4Fvif~)aN|y
zzLuZkfI6}FXW|($^uO`cFBwi8|8imwq+_-m3fXTvsN8?K^xdQ%yY3DLkv|5Jr|$S`
zJZz<4ZA@Pf5OFB*_yJ!oZqn5)o<hQO$RJ(@Fr+r8S4;rY7Pzb<;9=hqsL`hgNjb=B
zz3g5_whN&t+wJ_?xG~YsQPF&K0^81*XY*!6-@j`Y5a%-y;HZSs9Tk7w+I1ypGQU2&
ztTUg1q1N~fbuHlvPF0Wnc()Kv&O5H5;ag0SL5PPK`P`&L15j(SQJt5879#u((6s~L
zT(mRzd-J&{=VY)HHW*thtJF8^U-ls<SBA2`n{DX4hnjCd8zNNcXJ}7WdTRRrNHf}#
zT?s)@tDBsZWivg5spD_iVY~J>zcI%v!YQ6T31*fHzHGNXvG8JRW~Dnc2Y>2@_#M+o
z){AX@8Fka}M2r$p->sY`^zF<wNFWrjF4?LQj2NpSt1Kg;R}u0!$HlNH0Q5j(1mc@f
zZe?@;!k@z^@3CW|V=lTN^S%s)avZfX_CRB*_f*2K`J_!Ya^8$!yAK-N5k>%CX6Fu@
z+^lWJb;kC>^(lf!R3JYhrQ7aTcM%by`8pjJVOgVx%(w!JauEdl)i%&5J=g^Nl%pst
z0wqdv`<nudWEA3Xz=#wHSPRqElKgKLhEhL%yo{0I;%`j!txqbD&tq3Cmc#3fK;C33
zK(a=~a<6-LELlA#@`StT7Kh2uD<Hjnqq*)`2`v{&q)mh_iYv#$HnqEm9KiKnI(NgP
zRa}o|#b5V;XOjf*yVi>ZE@cg%_{lTZ1%ds+Kc0OvGAbv(u7fFfsUnb<u#XvtfPttw
zf!|dLf`MXNaXp3?kV0G4WG66hs8307Z)Bf~{C-tWSdezd6qKTJtme}Jf^Bq(UC-(}
z=G47Fe+m|m^wVy&a83wO0xX$Q%H76_$0B`o8M6sfSpeLl$Yd3ZXH55zD^EIk+u9{;
zI5F{ox)jgNY$8LN;GmzuEba%y0ee49U@qeZ2<x)Ml-?$d8OBChOu76=oraxK=lTiy
z6X89-@P@Jf@>*jVhx`VEvBnY|D5ocsI;W@V01q|PZtS;W`>1or9Gjt>(NxbIlP?vM
z-~qZ+nhq$<AG0J6f%DNW%t>hHe9`yPX6w!@Aj@G}p`t3;^Fq@-B=ttCL`I~+0oS-p
z)jD5~H56=or1y#Kps~Lo_Nwp8j3+bp+)(grDN3T78bYW@NUZpotq}-w(pHo*B;rLe
zZPBiI1HRGhRjd}!%MzsJBuCu$&2+~To%}7X&ln&(OU?$ni^U#BfOaT9k}!~7ggHQ8
z8Ewa-z(^B}rA3CTMIk&5MmlnkiZ*oDLl?wxY(T@Fsqi58<1N?}tai^?Zlst7efz1#
zc@{R-FrE}FmaY&*INwx29Bv@?ydp{ruLU_vG2^UgbfUirj52!&9D2c*yT?`=Y2x6a
zbP;g&2=%37&eeY#Fi_SW_jwiFG%^yUU-4`_eAqB%zRz9n^RZ|VjLLH}n$v_~mAYa~
zt{cBr3_Nwxoi3h5Dua86j&?8wlk~x%9iZaF2+KbgiQcZynrs5M=J5esQn287{^1)7
zqnhvpLfnQ1CHKktb}}_7su-!wCkQdp(=h<4j}ra`vu3oA3G{f7z{$Me69J2)Mg!9f
zFz>g1$a^WqJ!l*3XzP!mSz0k+QqNGIdR<^M#dla2q7G6LSym#|;0ez;1><8>?d!tz
z7}?sH5BtPOI1iRxiydH@nxl(wf=2BaLbZbXu&=h`b0Wf4dy^L9V-xr>X+%(Bbqtu+
z;T^L){DX+!BNqC;1+Y3+e)O?khA`~}K*KT~q%8PdvuGq&V~Tj`I0Z}{mSR?sCEW>=
zMfV%4#TkaINB*6cQ)P+1=vf8w%4WZCW1uaJ1gVgw#vi(~9hw1aG$X&UNX%UBBgbz`
z2Eex^c9TqhzmAqEl>MEW03Lhn`TBpRrFe{AqA_Sbf@HHl#DwzHngl8_2NjErW49mv
za$q6yPba@@?WP2DE&;>aL&!oE5P#9%UycWsJw9FMXZ`=qOtN>L?dAdcVSzke{m@R$
z=V2BzEy%2AKrwFpp5w*Fa1uci9s?r(vdi&ZP4wuqAC6onrBC%l?zdG%vksoT7z3^B
zEeWPGIP7>PVnd;>T=s`PU$I(UMk>^S*iP}@jvi>lEi*`0d6}Ko)U)OQZ`#RiU_8#b
zzv8M^95-M2uxiVhFVzUc#y-RzT3#x}$Kh0*9H>78D6F#(Re%@VM6qxQ5;(Z~NLJ>+
zlnh&vs3>G-#XukKj@6HqZ+l%$q4i~ogq~__boJxr`~JVd?zIo-8K>*^uQ$j4IUEE2
z9gpetIp3KKHI^KWx)0Kh3}FN8CjjOqlNRo~oh&E*9r#1X?ZzQtK+g5mibsqTTHdAI
z*Pp*5SS@J!A87ag8<7L$egx8&v=*;{AVz!kKQ$~~G#&wr@jn-^o*x|ZFiFV8)18CW
zICCqDr_EhO^bbi>#fMiJVEwQ>mnd0fI$@)C1d7lGZiVCsv%gVREBZ%IQ)imksBJgi
zc~X+93tvQ~YXV=yLU<TPH0UNW@=*J)UC#4!e>`x>jbsjA)DG=!pup51JBRy9*N>!F
znnEo|Ndq+Y73hQ@+}CGi;3+!Ltm@b4o<T2l`BZ7Ux6*~^xhPSLCs?lA@31~3w`(Y@
z4>G?vM}0CSh`6QPlQJ2fg)>Rjkl-MoO2ZI_uSh_sNF=rbMKyDkg$0k;N)gI*t4{tV
zPv!eB<-6gcG@Bz1ZX3pgqYz;k##~CkItIkivmviCe`5Y55>e9)+!GNTh8)b1rVB4-
zgzt$NXCoU^D|*zgEB%F~#Ikh@t_95{)Oqh<!fkq3pL&|jY*S7CkYXvtdMJKl>vAS{
zMu0zY;n+SBt7z*~1r@D_bG%k1D3N+*Fg(Q@LsXO`1=W9*sPg73CBa3()Gp|$VkIEX
z7eJKXpPeI70^><98qUjR(BLT&-d;FLdXMC}mmRFGeI`cJLx*FKG6rX6N<&_uLRA5`
z=%u6~lqd{on~d2PI$c}g6kHb^_9?Co4BJXcV@b_?FWW>N*Br8Fd(axvtd9C)?-zk3
zEAqv)2x*iU7W&q_f>|b_S>4Kyw{d{x)xW(dMc;pUQ~222jETg1ErFWAN2=TvNPPP&
zDm%c4hbQXY$YUs6)|^Nh#+@@qoY~d9&Fb`Wg(=&A{FHJZC}U;SF4`8Sq}weR4xcj~
z)-ZYRjb*6QHmm<;R?lEo|EY|xPMdL>ICEXiypR{zz~zeBD*$QcU>Cj6j;4zcM|n%H
z7zZBXm(=Amk~(cfb;4zOY^(SQhceb#g4(iZI}K|uZvCHdmvZp8=8y#`=j%>Z+ASc`
z6bSoD`BBIchocp&`TE$Eyr{~uGU__N68EsHmQ(XmAyTl<bChI77+@o)Y@`FR-9zid
z%99X_<7;KIYXP)+d&2sl(omQc%9=3PB4ceL(C(1HbGt(_PUlwD<o%T5gDht~fs^Tv
zx`mT6SAHLTJW`zO{QO(v;0^?$X~bMs8-~GknQX5PczrH^t}ks}$()>b#2l1K(5qHb
zflRiqs{+`4X@9zJg2H9_PxqO(kC@-9KAW)uP^6uHlo>6=I7_W0H1u1LSTEuj*a;<#
zK-GdW6kH_pbr4?jsywJ*iHs$`<IuE+(Z-!y0#A|zJV_zXh}qwhyvvbD_8rs%Pm=V{
zJ!1WP5;Mj4QJd165wljI5IGOgl!O!FB*>1pt-rU(u~iLUO8ORXWQDLWX-4`T5nE)~
zuW4RLpgiv<+4p@g#G7SAb&0+ZKdh^p5bS6A7<$-I`jk5G8|^`P%A3ExZ<#q_2^<4^
z^}sP8&<^Y=ulOl2gMcg2?}H!o@hhT$Y0fA+tmO*IWjpxg3Si6aQsLP~EI8G0kCE!M
zd&ZCr%<2IcF688OOPS`~5P86{A`QGwy9bp2>D#OVJd>a1lnCv{NmImW%cW^oC20S&
zHwD;Ul|SvR3Q_yp-XHWF7NBS)KHw_aMV00Y3~^E}qLtIEqvyn<^UQ7{iDQtT$XoHs
zqUixrFXWsm(Z-L3+*={uw>604utpoWZjvNwh0KTc7Mje8dO$0$kSEGB_#z*hGX8tB
z-Pb=UO;^1^JDjyAgG0<p0Tz{GWj@y9OZLJxjE){*JCUU;6IH9?YNf@cn5YY&Yb|1J
z23S=Q%(C^T#>my;@(tO?GhpH7%#x}CKqwcwiN{*BXSI1D7iVq^RnX01h@*hSFinM|
zNWljT<}E9n)$DObA);=m*&20~AY0=em6??k@p=9Z9;+e71;np<fl@S0^lPv{$PDQ2
zd|%H_&#`0xi{*@Y92)oNyb50ly6*7XZlxK@#SSWG8Xi-LO+)0iA{+m53)%p1NCS%%
z1iVS%&<;*t_hwAJ8IgcdskBPET0HY;+Rt}2eC&gJ*-j9p68)$l3IJrS#txtC9-LiI
z>UAxadZsJPhBjB>bbQX-e$X>~B5F9HXT6h0+)7w^4NdPHVAidPp)W4lHbjCjKkPF@
za!Iw4wO8fZk!h*vC<)z-@^lAWd%O0>$Kne*415<PE0WV^Y2w&pjTT%E<EoP?Vv)eY
zogzi<+s(T+oBbWsS(ljlELfw-;gSIkjvI5Qw?r}ltAJb8xiT&jfE3N}V>$@+Ky(>?
z&8AV3z#Ey+5$&M$Lb6l`C;on1nuXYq4Gir#nHGgCTZBg(a;Jl8-w6R=OmI7JpGP8I
zmQw6GWFZ;U^gWr(GXSDQax1fmkxqIODDnNwEO3Pz2KVslK5Rz~l39K!Ju11(hQOlE
zKUw((BmuvL+HqTkB!{uZ`pW}9mGq}`iDuC_a>!4Se|{R$RA}iNXD2L=OM%R3p$;ru
zSp_TBThWLA%fsU$z<~EIjj7S#cjDEvhJTUv+iTP&KzidtnOQ|zLcyZ=)FRPkmud4k
z?JBK#4qguKw`ee1muLA3!@|nS9f^}eXS)r;6=LI24x3__ep#50J_nD9k<izjZ$~9$
ze~MiAUMu&7QHwBkqt;(EIap&fUY*paS+m5rnduLacRRNMT~8J5+D|oVw$NQnKuh`I
z^a0j92{!Xb)5M*qzMye%Uh;xcm}JyRo_&yF2=H%pu7m*RaeTkSa5BtC{t-8<5xn-}
z;;E(}usiCr|DQrMBD6Z;s$q}AwH#hnstvzw320FmS3UX&)P{&qY|}hYrY7ahv$RtC
zf}Ig$Wx$re)!KX5F{N4CdFF`?q7Qtf?_aR5jEZ+y{<$SEdG<8Js6spsT#!2G->n{^
zRl{RAAE;G<EL#izxl#Z`h#eCNSy`vP*m|oViS?v{2h`VMPfPy)Qf0K;smago6S)9%
z^2t{H+nI?jPnuM_Yb~*O0Sh=tN2x)<=9rN`i*noSOtbTECC5fRcz@EZOMES?Dvg_a
z>d-syQ8=%VYk>j{=rZEFyT3IlFQJZ&lE+Pqk!zX9{^B{4x4}^SRKtIECr%C8?UWCp
z`15$X8Bwzp-XjlF^sMKJ48_*pN^U=cfB;}ICuCA!O3QS1kNx0d2M9#aUP@wb?Bh73
zTct#b7xkslxT4aRH$Q_vT`~?Ca4DX!zAW?P?kF2jRE!U2pfI6`-rSsIn*69{X*+uN
z*O~#N>kH8qQ2Us=+L0lvNw<x%u$zHoFy&e5^QqWq1*l#6fl|#&bdC=BvM>8Fd2GaX
z&Yt7r$a{zqXQDTM0IrQbEzSHaQ8W<jFbBiy-LhI8VQYi%X6xjg&&SIZq3lvfr=%{I
z;x%AP;bL=c9*x*zo2ch+`fP4D0PZGc)Y_MdM|*@TROeKP$U6Zg6%$b~AUx~RLhV1^
zG>#fGQLHvDm(OIpIB#N$&TkfkJPy0ywd>De2|X-e9o2S}3Mf(9PE8QH?LTe}k*^{G
zt_w+e-8LP4mT5c21#c9{5ac#!h-kg(GmQ3<>x5rY)^nQWAw=i}0YeyBS+mglYIQv#
z5+Tv3Qi)%ow3n;0x?p)tpx26N?B@LOc(2QC)2+vQS8<=(CV(k+W1jF8^2;FDvXB^~
zPylcbJTyLTe!$s%@OICYbDe3&@yHpTN6FHu{JgN8FG`G~ksv<3AkK$UE6{NJc>--?
zi)Yr3UIO^_sgWB}Qc7}=pBWqvnm;A%kwG2WC8h3udl<|Vd4pZBL6Kn*n(CGPVvF~$
zHH;)6z7&SkWrRE3TY|iF<9XFPRS**AX;Ly2E%6C}oMmG+-Ai@bZ@b<riPhxS&ar1p
zV@;3-+#>;7su8`>`n)<>*hd+K{En1v7CS$`;{Q?|X!0vTe(@W<@90cIOst)b){ohQ
zbGohlWyjt61ZSTB_XsF^TAb9`-1DUU)A6je>UAiY9IB&V5+xi;fMq<)I;$gP5gF+W
zjp%Fz{d=p*H1#jJjT=u<s2{izC6*x@PqZ+~#%$7f@^@>pAZR`p(k5X!uTnYRowZ2a
z8<$qZkhQnKU4t?CAJO<dz4;&e6fsYWEoUFsLO=&uLHHl@NOKQ!AXsKxndIr?5g~jo
z3Ou)V_rAI$pS4Qz0BA3QaF>o+HSr8AFs7#n7*Jn2rl(>89MqS>0xp1kHdTqdbaWW@
zJxvICW~!4@^5tBkS=EX?;xePYXLV!a!-1dAhp8J{*#I^mJdS=;qv1q*ijY@<Lx5Aj
zpRc>k<E3Bt@^=4f<=PNd(W+KedvWsluzak3F3<VC*?36@;Dn2zlGE8{`#U^aig!qx
z_gSn*oA*-~l&8X;6Qf)D^TQH}TJM2!yYEU#j5odBt!bb{Nv)oAj+vAm*Uk>h<UpKK
zs4A&UjcaJsvD|?|lb{OpoKctL#SO7y0kQZ2TtR$4wNn+hD|re&`{54I&jNUqla?1S
z6{SVN$6MP1^tEuGB56SSV{9%Z%%#SJ2oirSXWg}WoN)Q~?75Kc1_jwEK+>AY+nRkw
zo(qGhQ}#t0FQ)VH?gKT_38x*SvD*9%F-g1A!LqL)oW2|K%BE?HIq&tT3@D>Yx7lfD
z#Yb5w$8WRT?R|c_IlHmJIln#yyE$1JP7I@}dXHKM#Lz^-RotW%0FDWF?C~f@C4LvZ
z#dCzp$53Ff)yq4Fb>{FyeP36ZPOW&M?IfG%Q6BqTTyC|6Zdr94+ERhO5*a4o?3U>;
zU5(^J*IVTp?9z?u4o5-AA`BQ}BrRz)r^Ds%v^v4)62#*D)moSKR3PN4)DXW~GByp3
zmUOoPC>(YXJW6axFDzjJXO?DCgOHmg6PA;krF(8=k!C_eNxEZ}M)qdbb;&CBrz4sn
zf<M_SniS~&F!E_QMAo5^B~Jl+6Q-l^&zJ^QamchN-=Ol9iVAh;hn-a_7SZ+YJ0;rJ
z8&uIo&a4RuQ3$GyjmFX%izB5Hh}LTfhEqZS<P9Q@qR&o7lyAtF%pPhNaCVQn2_=6w
zHo9j<B=zz8@C~hn(M{Vo<5e2{eErbQ-+ASOD}|^!Ohoq#;AON$x7U3R&)fMF=O2;u
zXB~^67J1QKT0+=;IqbA8*u2MGA6R;P+O~B_F)gaMn9`+lJ)l=X#@G<h#FEUuis1>6
z2tQAV?}OW{oGm3PfhhSY9UJ3C94lMy?6ro$dJ&k!M;}oXHT06cLrfu<OFC5r3Zw7W
zS35K0;&As;v3<Dk=-q!>A7;4ZJJYd!uibz8*?Je?FwPo-{J{sqNt0+1m`DPbIhM(O
zo`@i-a;RB_<Xb^gJ@Zkm<xKi6?CA!euSQ>19%1%jG%wFgN;F3>UGr<yKKeQo6BONm
zdkZR{;{e~`u3|R3CRJw3{2Azs$6;6er!#&hPyus9xAOoRAGzMKN;+oC<9TvnWwMvi
zFdShSQI8Q3LVoThrjSm7BwQG`e4^Fgj0Bym{xVyALaDCBXbK2VOHvq8EHcBxN0_I1
zYVAgz?+xT1yh8wgV6y~)Itz^q1ob8V8UgA%I{;K}qaK4hB)=?t<ybcGhZo68<xEAV
zFK%jE;Gb)pqvfr=(0z>SI`vlLICyRdwO5mCE>@aj{ree5v2c8yutXw1$mKId|1Rj0
zMG-p#W=CfL5n&i=Y4%5oaH*j0(l3Gfa)8+p#X2PLRTx2JK_-0@%}*t4@@G$XR0*;V
zU5=K?TT0&ZqstO<)3Eibx(N0IDp|(FCo%^CMrYF@BG@+!1Z+A&IrB7P*vmC(($F9@
zr$kC%GA?-yMc9dqU#}SzO0)K8A6Fd8#SuhsO2_H}n?sglrs38((`B>0KW$z#RBELP
z8fW{fz_sX~XRD9fZ`>^uF6byc07-OB1UQaoObRC)2Iu2NBFz&y6_xyh#-_C{5K5Jb
zFZWZ-Sv+`mLz=>JMK~@;WvQK{(N#8X>SIC60{ZjLVxt+-wUrA?WSb8?Sw`aJjbc;!
zCgqKQqyECS<BYe;1tXz61fg>GQW&d7N@2MM9a(gEGw>jd(<6c>F7AAMOx?<aSU5i;
zC=Q*<`LHxx$cAeP*)I=-4cGZGnwU*_F`QaZ4c8~&94lVie+GNQ^~FR&K>PeThK~G>
z-wJ&hxh`_t#@MBBvS#q+HkoW4W~I+U{V*UlC1xr{k!Y&AD*MA}`qhuS!{be}4_)U>
zZK2Wi<N4&~%*{&g-5!T<rV{~<X3Ai_@pr#MF@Yo9Ng{7@;Ac-Z=K|kOJm5Q!cg>EN
zZ_io8N`of~JRTbq@ou%24u-*E6iiza`9|YgoCXF>vqlAEiUI?9qX*QWC)$ERRwe;v
z7gpd~ZWsICF@MhFj4AsC@=qC0aXaE&k*8p+LJcO!fPS1cq2|VRmEzpkF`UzZ8?vzb
zwJ4vjQ<vSiv~`nCK!w?&#e?><r}eWRo3w~ZR70tx`;~<XnB36VgRrQgQ-e|%P#fRS
zs3}0D^zW&X3T6M>Ry1~VfC02o@S3FqWMCP_1u39)>dYJoBflYfHm^H0d_k~^N;z$h
z>6eK6U9}&1x4x2w7SE(JLZ(0chOMs~b=~i%O&3iRpbr;Kq%o>Y^(dKqo+Ost{guMa
z&7NCFt8npibPOd3+K@!j0D@@&nRmcYudsscm<FrQ%Rnfd5?`&q8lA1=XaI&4ZX}k=
ztC8f3kjdBk%sZ_MX#GSi(F%48kT9+A$Uu?>-~=RDpb+E?(~<-;`O1z-#Chw#7`iX_
zU)IQy?K;?5&8$D03F|`PXn+GZmYWnnZ!!L((_yB`QnZk1EVB@p0P~ysm(n9NJtV4S
z4!YyAd>q7ED79J96nyZfG6{jYRS0%2vvM@5QR6yOm~Y+!T6E$FZTh>91k|PJ!#buN
z_JW$(9l7HUvH9k*Zj6y&87f>)QNf7xJ&nmTs4%==u6<wm5Z!N7U~RrF=`55N<GQbx
zixNz}zy0{7Y^f*Q0k|({*sc(ddfr>(y>(P*Vp4{le|Iq1w!+$5ItkPsF~;H|;@Fm^
zr9sNXkfeots$VrmcfTvT>!*YqX}GV6V7<b0U4n2>^W}`p)Y@|ch<`&|8xRqK0v>)-
zre?eNtPP4l$gd3sw`-Mxf7KNTL{r)cMY}Kw`Z=2G(h?O4fKmEB{M*;>BaWrYm?Q|R
z$~f-}@IV6nsmj<JeXy}5tf&C~TT2+^V$IekEWLh_MddMXL|M>c9+_3PYq_6A>n+w2
zKB~c)7BGQdGA*P`!ZIyL|J+)sKeyK9nMx!1Fw)^j@7lO%35pT-;(|tK<ox#R7(Alg
zcLNw~0e_-xOS_~g)Ye6g5~`CYGOWqYMNZ;(JkLilu$>D+9iX;y-Vr0Gav)<hW`5D;
z%2hBpIdc_7mWz9M>zz(UlLqE6%xs}uayZ|D^Jtt3evx@8g$H}J<Vnh7q;%&=5|sK?
z_D!(gv?}5MP~cLjEJeET<5v7^ED)~|?R5^o5S9%^dg<=R0ddE#g?Mf{StY!~r0}%y
zU$b^ol=Rb@<>6rrrPv!$`gHKxv--Cn%Ru?6PF}XZ>qiwK_|$_>fkFKy-q$OiEHk|a
z@mAoP5gm!%F1f_N9@=<;n~O5x`T?f8alZ0U%6|eGYTD+&iWMB)MkRSp%mcW)sDQTZ
z+3+=Fp+X<Kj!l*Ac+9U}TK`kZY%|3CK?UWdq7><6zy&#PzUpuYpRCz78%OXJ-Tl-W
zV!oI0ARg?^kMS=9x;iCm4Ok$&Y=El<Q|;efm3MxXe(d?;X-^%t^b$h?t7%{Ab}LqA
zz&QMqF3kRB0;I47_xmM)=zHSjXVvDQ8QO8;$xobb#WW=^h>S&1+?j=C-`v$5ndvEe
zWvIoN@6_4XN#LQ2trH<(=h%QbxAnIKKEVQs^fnzk`}}bFtDtB5>kB_-X!KOLfQw7?
zR53+ejINBT%p7S!dsGDKRM7M$99pnqhD2h@Vg{MNYHl4h1K_=Kb`j&AmD4>VYLu{7
z=tZ4Xc19lW8Hyk;+V+>757H~5=gL?0{9E>{cxc7m%l#YwXV%}`MjOIgSAH+OIaCAP
z!A`VXx(30DeFoa@Syz7RFvW6veCnR*zdg5<f0bNK-(>ymCJ)w}2Gx1Q;}jsLTm>MB
zS74#mw84O_<ouwUxPQPK6sqh%5r{e!Em%?3GuLT1Xm8m8bX7)-{Jy?9YrbB#N>dIx
zqDh4EHQ*N)_%O6Q*vpxrY{Zj<*2vwJ5&<vKqAxVCt2d$eb&4b!m0lqN!C*3TbyAE9
z#r&6FzM2?8O6v;W#g3u?kn%jENiB{DqV}F~i{5|zA28Lpi<Q1x+rN*66(lGC_TnZl
z6col7$XzTj2iagj8xXrpXR8(BKmq>iMo?T5*-zy<PlV&r7T|hYfv23kSBA2Xp-O`o
zS9Nq3`5ncNK8`yI$g#jKa{nF)L$tRJ2%oPP1)bc0S;ywKY>{&5tf^G-6tmOq8Q*9A
zuQxJRynbfVP5J6mJ`YRIb=R%e$Ggw-jdMguKHr(b7h`|Rc!@-1|7`P^6g1S&$wrDP
z??&vOJ4_QM*<UVie*i2+G??xqN$pe+Os>GPvZZHd^|wwNAFP^&D?w&To?lph=QyZ!
znur4he99hjq3}*5ao+wKHOz%BD1xaFOWQk%$yoe3gLnBuia;(TU}bGzI7dP_avW2F
zBwTq*&-FlxmfL3XpJ=Edt|4hC!-S1ZJh#wR1JpxRkTAUYjhOsGx{_DASv#X#4DJzT
zB^g}mQ{kLAZxsyGK|Y!v8H1KNzm~n600Lcn?H!*ae8tjf#dTN%3X1Eza}`u-nzYtg
zC3Z0wc7Z4|)98ld5oT;T>uor=1{Ib)N(?+G3kIfC=wGX9pA2G;stl{Zw{5^fACCA|
z&*lC8tttXyEJeUS&RmEjdQvmjF}6~d5}hO9ybXt;cy>II(m6~7sS}nTx+VaS2os3M
z$#|sXPkj|mD;yl*i@^;{@XjMOG9!dFF)=e3or+2DP6))w2v6`H30Y3u9k|zh{wkrR
zM8)L(kZ}S%SHKUL{;W;<17i!>hw^yrOOEuNTwhFL+?Fr_*v>htyrch{9xP*cl3nJ#
zi@I_&6a*OUbDmDDlNv0YOU)So=bL#>ta@@TO-X_FqOo5SwKtj_>#LS+bfYzu9JhIm
zot+R>AcubuFPmyPy7!2ZQI9|zB_vd|2-Y^ulYM{1M7{k!QD~$r-vfohj!zkl`+$Jn
z0tGHmuJ9O7fmv*t`!P5N<7x08VrE%p_ix-;Zqt&~vr;~fNz&&n0Q5gdZG7zhdY>zm
z{%^d&NBRX`*5o+9RzRLpYLpY|$0+T;c;k5Jbl2s(Ztuw2y>lzm**^Azi(Q5F#&+&l
zccOWJcq43yo3)av0mZvv?GEHi$i<PC0aHyuyz!H)SDD=c$nu0{rXvAeySW7cI{?iu
za5H(xsCe>g%t1-jyP1#YIEh8HrkXLKvE7|f6k|)e78UmtMz)rudVa>S4|YJI<9|L0
zX%bn1Ga(t5^%{MU^amq;4i)L^5~c$oVk#5}5#w!tAY!Vmca0Ya5w4d&h&b2)PT*Eg
zU=GvjIXeLOLlGDH|E37n%l}3Z@p3?lXj881#BY_QJE`(NUE)0ZX&zJyZAnsigfiQ}
zPxiMYunq>z6DAu!+e+#W>LKL;oeGjd<fTD3n3y;6Q|6kOtJ|yFSqhILXPLQySjBLZ
zR<p}j4+&T#K#FE|mi+)kCp9m7@w5c}ATCsXmDvTfgby~1e&L+2Rs#rkk!Cn!j<IF}
z(vtE5PcB6Ao(N45UIL@nwSY|Nx4p85m8&xGCE-8#@DDvmfBT0X+BHz=v4&3<&vl=9
zK2tqncG-9DoiIWVA*~m?X(z0+m*b@30oIc5XE=2W>Q>VF2`%>T6dF^{76tpY4(+D#
z3?V_|Oh)p8bRTvqzVX)`*%pT{3;0hsi$y?_37V$y{HNP9F$r{gA|@vD<6KXQEpIq^
zLqJ1WK_(`FZqITC(Cs;+|LgXA%l?<!qh{wJ!9#l$1i)Rg^3eF}_DJC|pnkHOmWq<l
z0lGaBE5JUKDo0+jDh&PW_DHq=AGhaLxV|*e?*D4_X#BrfJzh)Ctki{kz5w^d4`kj9
zkB-Vx*$XAG$+ek*n45D;k!%E7)!>Zh2OQG>uzT!oO@scc-D5^nq$1Qq5T;TVVxcb#
zVsFXn*#PRq^LV3px@R@DPc6`sb69hj)r{zsC=+QP@{i%O3Zo=xsD|L(l=_eY2&zv1
zB(PTc;DDj-&Ao1zU;)0%@LSM#voDLIuJTpWhu2kMn+NFt@j=Zm%5sPucCyz?kXE|Y
z%xbl41dR-qYUA<ZcR;%?ngnrBPvW#qCHL#N8N`ECrDLbhy$?t(=&-KOF2lA7&bAm6
zv<jE6nhN5;a)D88|BuwKX!#h}02wY{4E6K)vaou|AP{%%GXunGrs`q;W-b_ay3_b=
z2o<kVNY$oWt=S1DPTNM6q6Db5E;h${HHOCB?sJ{|F7L0~DbjOKjS1Ln-TmxMTp&2m
zAhN*}NFZS$DG1<!A)+~3NTH(bj}5fMekBYo9_X1;tksrWEw~_3zymqJf~#e@=nzBp
z(PW3=u<-REC|WkY!;Lp1HL^88c7~N)e(zx>@+xS6WtR8YxykSsx(+^Bwkd26%WN#^
z7LGMaaD-_g^4kWN#0mw<u+%P&H6OHVsww<x)y-E?V<oo-%<N{lHi@|;-{)Qu3}_5m
z*gfcQrJfSlaHTC2IJW|TV?13<Hln23ca<cCZffKPQ<eSjr(U^%2WmBxktkMQA9t-=
z&@5%ShDys>bMYC1k{Bg(!m_QSz#pHU(j5uZl+Eap<bwOjO`O_l1-LXLsdbd!@u11o
z#x<%~Hax!5DUxo-uj5P4VIAp_L$Z<=T>-acs3ER`&NvM;mbAG6bowua@*U11nDJiE
z;vN;<=H%~AGQ~@}ndC&H(DK|M1Y9)-;M?)p%$iN!6SZBdZl1~R3E$hdE19stFSncs
z)z^e;e{DseP{{i!hAZPDLJ{;PMkkseX^3PHmIxxRJEl|^#FW8`;a1S<S<nJck$J%+
zT_4o01Tr1iuC#Rt5Q%Ur9=4}mhC*%5+v31TU)s(=6oXZcgr`^w3oi4JQt!8?77A>`
z{ow9+8Aid*3lxIK&g&6H#Lhb--)+_UqA~@<6Bx)+{vmN{G~<XUr7fe-C?zy1R_l_(
zoF6VbyG6X{yH#o9nP_YGU?Akc1KQ`c`<+?W$f5|A>r{a2tpBUf5QUUM8=|20(tRRG
z!s_AoerPn$&a=Yd<p5<xE9XVoSPBr&POiwnb^{?8$_pRvE8pEJ>@=aZ``~tiEBtZo
zSO%Z=pJbB?!ZXX7ZX0!{kn)Nb`_SQb_)4YZ@kID_px-I+>w-i%rej3nwv^q7<HPav
z_r#Q6Vi*B8LW2!(oY`hmp(dbCLZCH;JTMj+vS0_6W31StER*tz7E&aki^ak3_}<Ki
zs8A3i=$hul{-m!ASbpNmysE5RwjQ5Id#^ocez564CnVxvFz6e99c3-#4Q^VakEq^-
zS-~*S>KVJtjqvHF{=R&64R!8PvM%4hJ1F*jG2R?7!8R`-REmX;JG|GOQnGw$*7~?3
z&RR+?Z1cO5rcu~_8Vb0BHpxh?_d;pJ(R4teV#Ao=t5cyIrCKu{Vch2UJw-d-h$8oo
z12ilEs8kkddW4sNJ8H7#N`6X|RJ9v)HK}ymugXD+x{z$S3D}qC&a08??~0EINOXf>
zzY%f(o9xWcSA{H{Z!WghJ)dw|bK$Oor)*EfwsuCb^D3%QM<Jups*p!vow77~lqwl)
zud4Zaq+f7{FE8I`a3t&d!R-4n9_F{FOjhhntsicePmc}}>l1H{eIN&&1_P)3QsJo&
z6mok|y$?Ps_)HR`nv~YtwSi+HWu?Mnk!Pg>;L$#3rJe$ZgY}b=gY!lUQ^pqjFhn^Z
z|J)*DFvAVD`|(t+t>JL4*q}O*lYi{ZPbWFhi~{6vjwdqk)>;^ba<XJXj=lyLCQI=&
zAXtDAPg%j@V9w)HiZDa*7ED~haVIl0Uerfnm}goAc#B*{X)vA6Nl4WOb_vvx2csRJ
z$oV-$eS`br;OEd6k+XwH`IVYX*&5LH)*En-hxXR#DN|0(HRyKHQ8fi*kEMxyWr)N5
z+RU-j1F)#=08K7tbUG(Ydm4&{HO?cC&04Dc6ePZl?4g_y4JSJN{_MHvrYEW<59zs;
zLE_G?H?9tRG2E|_*#;TC%sC27VISFmVz~gPn4wfAyXT*&8DLQo3i<QqbR>qpgbznS
zvy3w>hYIIR=v2k6XHDUzH3|l*^u7bZb~dxe*!w{qa%FEQf>u%f_*NIfnZ9Q#<05O}
z`jIFs{fy#iIjS87aQj$?F|UcRT~2Z}CTS*N8?R3Kkv|rV_vNQpH;cZa#aBQ8jMDN8
zV3vwtM3`%3Sw7eHSP<99&b&kT>v81c)3jq;bA8)@cde<FRd(=dRqjFJN)y+@jU#Yc
z0;P&-Fow=D0{7_IhqgNl6czjq%0GbA!Z{{YqG7#&K0!ICCw6F&csw^A(2A+N0s5xE
z*8rhA6D$u{?r?DM67TYB`vjN+@V+~|9<=?Iw^Yn3`)rVMFW=pUYiRiJV)iDvNmJL2
z6_IS{F{AQ(LiSQ;Rz3J9>yuzj0E+E-otc%B%p>6>WUSc*AukuLmcy@aok(`${@b6T
zKxOU|hex?WN#dP*Mt50vnetpLk_mbyDHMtYSnei1qzL82>8>}9d%0pn_$U>stS}y!
z;aCo{Ofkyr_Og=DfiZ$g2TPmKpWCaA3-nvlL1+<BLLr!eJ0!L%%N<+Hy;E3oIF)Lr
zrVZ^u5CUp^**m&A2p`ws`LgVwzSXr1W!tBMc~9EXRoD=k@TtzS1=#zA+g3Ym5hAUP
zWg(}GTW|Gr(#HpZKi)7qDRw=PrrT`xf>k~CeUrG_fIoz`Ac-D)4g%iB=O9M@5-kW6
zbVqKg^=4!<z2LnfU9EOEhBw{FR1#~_X2qu<WB|Fs=QZiL`uF^t=^eIyd|42M&u$o?
zCAh&Ta#<E$TaQE5VT?gs6DTFPJKX8@_)*)QDUxT(5`U$V)d}^lRIX|xQ>&qkAUD<b
zmB6lQ*Jfj`Q$N8FibI2VD?<@fDM0ZBjzvT?-TSpxP`B0Np_ZMA?#nMG!_qm`0K^a`
zI5dW3Xmd&CXF3k21s^awP%30S=oWDfi?f*vfA-kg^(X|c-K9ua6{S+$jEDq-UV6JE
z8uif3TYE)5RT*nYAZk8*SBZL6-=O6RdP`2S(R^`=(-{&|GO+IFw+TjUkz^1w+9hPa
zyTzRr#c#u#WcY2(uP>QprD8vv2WY<+ZsZto`0af3{$sV!6{O@7I)UwVAW@pkj}=-@
zknKL?TBjMUi~zZTgtq|#ln$Qedrk3i2mS2Tmp-EItKlx;lGhk`_qv_8IHmNq3%~-@
zOi)62O@^zis7PF|5;i;Ex?Mca`ay1reh5Z_6crjc--g>9`P_ZU{`x_t0{}M%N*-bh
zC)Zz{t^BS6_(mS%q6^}w6oBfP<OM0`_;ZpmbB-gF0VHD!7Z1a(OIX32lbenNv+vkZ
zLxl?9ieawh(q;^P*weMm&9QrVYl}Yd%jT;)7VIQNlH8ONqxK|F?Gm{o16@af!Ecgc
z>>st0J@}W}0W1y@-T%m4&L6p>=>p0fynkUBXRfYMeHwVPTCUX(<TjtFw7p^+-vK!>
zN**p39RZ@hA%jTw;p>U&e~L-^{YGs4kTz1Nss{JK22vX0(rN-xP15l>PR*{S`jSL6
z2s7BoTEe9fYsws%AzvrKRxB`s#U_AuDs?uWe#VSBB873FLHVu4b9rE@>+3}ZsaMF^
z5LEMf3bI0>9bONHjO}C(Vq~kByk+EN^~X&hi5b7%g>U;`X^e+S>O*M`*=_oTk_KJj
z<!JfAGZK5jW`46vJ*mO!B$<T1<Ida*86bP9E&xU#%%X<?sgZaLSP(%0?ayC|tIMmi
z=&OIZI6ddg<dMvNP(EHuvRki-0`MFLjXC&oVTFS&|3aWQSmFgnp0jMoh-F<L%lEIr
zpOcO`e(m@lHett}N<|_zE_+{sbNnjS$4Ljqn*dP;8KHc%K*YvaQX^3YnBIY~mGJ?v
z%%Gkyok#RL=jE>e7c;xk!Ck9Jrt#jPFrC)zeF@M^c6}`4a*3Qpdt=#Rz1c8^j9|p2
z#0;XES_<@cKd3ttUDi!{vTU6qZAG?B4A;DjOzC39&#(I$V&2eKyw}rccH;0H@8-94
zb=#QFH^(+_@|Sn*=b5J;oZ0-&v<+VB3!LvK7JmNAr%BfUBL5PyEgf=|fMs#ncg|`#
zTYK4)U^<zeZ@lR(LA~(MW1`&W1oUA}YBg)6?7jp`>92hpSn$aw6mdz;aH7P!fl0&W
z7OpfY5D*eLe$>d{#9=|z?j!OJp?ZE8UkakKO8Dl)?18V`@}GH+FG4xPsZ&-VCjJ%4
z74bm67XYd0NIiQa6QbYf-qlRZ!A6XUhQL2rW(XSSG>W}aKG)}+EU~O4m`}iskiNR2
z)kn}zN`LvXaFaDKX8dU5<?qdZ(Vo`u)nH+!-#TJT2)@|v5H&J$JN2-kQ^fxk<?}#d
z@wM$PuO8kPQ9kFdRBqkDqI~r@Yg;H&^!dR6QNHa!ihh(SqbLGBb3YJLWS6L8-XHz!
zaau4r-u(UB<sd?R+vUV8z)yeVGywfS<#aWsMlJsOUve5J8G=k68@&E+HKkqxs%dPn
zI|is^FHlYKA$@^rs$2t9Q#dTi-oI+Pss!*zk;Fd>%&)IL>M++~_?ELahRl&j^f&Xg
zBM(<Z=SVaq<8D(vLgkA@g1E_-i_j-1_W?;mN4GzHcF|ULgWe=n%Lu3Cf-)+D5~$OK
zR!^S&NuIRD9Dx(;PysfXLK<md{@cz87PCIZHu`q%#%ClarqCMRjWH1nH<9ZJKm@z0
zhO!86Cy<EaLe)1C@q0qVUBH0hkgjC8VqONbR!tUd#yYjE$OA5amlcW2ThHPRk#MqR
zNc8m=#7NeupxbUq*K*oO$jNFG)_@gOo%4K<g)wqHifgTOmPks3Nz&dI8L;OF4Lj1f
zF8q!qa$rQn%9H{(WF4eDl5o8QNPKo;7M%&A;W-^c*6_h^o?J_trr1Bl!9GhJzzRjg
zg9>g!7#Ss&uRTMZFZoqC$Mh5YJYDUXBw)G=W>alT;rel+FHK233-z&eAvpvAS5gmE
zbHH6uuf9Hw%(}%>dnxDyK`9$m$2za`9|~dq32Z1faaY1)`$&>KKx^f??p6h3Qx$eU
z=u+@mXDW)^Z_~bU<diX%0goa0w|p(pqUGvX@5&R^w(O-G7W>jNKHK-erq(>MDre7~
z(zEd?g-^FV#z)U0eB}jmNnrv`#XcLkhB-X}AW4Mi$&Q%>Of#dsl1DTK&cg$`lhTjp
zb6DAh;*`Fk`{=|jfY)60C!rms=j=N!U|BpY15(OVEd+2~SQ6E#ln*Evc{k7jj`*~f
zXkSJ5olE(0Lb%8FxuHu^M0y|WF|z%|4;e)FXYTe40&1on1OM=w(RU`lZ#7fcuYNQC
zn^#2l_kp~Uck>rl!YeI-xI$sq!=@|k4nP^_dxjXFO7*@;G|ftz!7vWR7(+x_T8fVv
z3hNk9_ub6W{Ap=9F(}$Gn8;(?;aF_uv~b2I7QB0|qGH{EL`uOzt<%<rl4QO_a0`6F
zwnOT`qH&L}l)hlPOgHR?i+Zv)Nk>NLBEvJ@FDOi}^!9PpM@!?<!H3(D^&gsL-~eHl
zhaWI8&YMMUUebZN@IS@uDq&I2)65=ZJF(R>2Qz-0a8x^PuV?(jH?sOx3aFM}^3!~Z
zmQSlz+8Fyg2)x3s0s151uhen^doYGz&)||)(HWJW+9(Es^OSy-ei+?)Q!|83T~O&q
zTYk1>URV##!#-wm_ouEfEX)L~+f--%R0>WPTfy89UTk-_x=9@Gh<K<ZWBF#&hpCN_
z$VVc1=&;@OW!fQ=0OFIPK$Z4f<f;>I`9j(J*_;=umeavI!<-{w?|mhN)N@4WmgPxs
z%G7Q6-6H*#42M0Y9No04ySkf4r+}QQzG?Bz7wS3d5u;Dcv#rC3=!Cxjps+Lt2}NOx
z)~WkROzRt~4V;eTxVe$<yy8g$xKnOAY%$-(-(#0obB8B=2sc&fj>=P;hpDcbs|1jq
z>K=99k&#*JxG7P6Fk~HZv?qx=7`V=YZo?;Pr7Kl!)oA+&g|1g(m9+ja%WbD1Jh!+1
zo<A&O84o-&g(xEk%K(5X=_xSOy!+*+Osm@hV^L6Yb-dLypY1QP?@8}c->uRB58F~1
za-|NOe1t%BsHuC-m@a2ODH5|JBuIxX6EL>a!7`}b%yRYSijO4<TbcgQV1o<4?g{?|
zP<!j3`)D!((OWvmvRDj0-Ip}l<$i2vUPy((uM^c;$C=_ktpkKTnZXi`P7Cpd`Jla)
zx697Wd^N|xUj_-X??t*g$&kSjt3kc_PQ*DE$31`_0alEb4~g{r8SW-Al3Ps8FF{l-
zK-pH7v{sEyef9hoWS-hd^2v29)f4duKik1IdTQSNISUCzDwiM1YIqGJ8|^Na<67F6
zkcN7VxpnlBC@6sGaJ#xgL)OIrNC$TZjHi#&umKc3Z(A?DODU2`!X>&{!sAHxXSK2>
zsPoQej`k))wv)~&twwnrtk^eo%u9sb;UtrfmjA`pJ4R>r1#g?NZQHi(JW+=o+fF*x
zlMXtzZM)O4ZQHi(Wb&K;J2UUhnzh$D=lfY_*RK1zYgaX5&Y5qil|&X9|3!6Bd3SW-
z=c0N5oSj$Bm$v#K`AFt5l8WPMe8RchJq}Q^mmXUCpLOUP$oRhjotZ#s%t}M8XpLJ8
z%CaDt)z(x|>6J0+@Jd5DQ>IW|>{Vqtdh<g++<nA&zQ<5G%J(MmK1)-_xZDPoKIu$4
zM5(l_Iw9RY4uoERrQUWc`i9GfJ)51;b|Zc%=d+}#2iuXB+v5ij>W(P&z4rtq|M<}2
zMEtP&_z8tWWc2j|SYbS0F4dw-2a7A+d0`s0I^hM)wY59JB0=DMaEr*}xvtnju4g1)
z^SN%gormVGNQI{^-y8W5D7wlnK8HY@_B&mLw_*{m=p7dwR?69nHFz!jWI}d@fGTn7
zG_YO~Z<GK%>GN{7X{`3`00s9?5>BLL_gRrw(+p%#i|iL2uod(i6v0(MLSl5WfM?7s
zAy9M)AsG1*<u2d{$iCZXy4R3=%184Gym<Yv5hP{2mIYh;%e_^b-yx?H>O(sWV?P8t
zhV{onfvrchlD5&rwc-mA=Owlq8rH?w6u<a+snJ`dn#~mu-mMhZFSM=OH}2sm%-u$c
z`m_U3Lcvo9R8rIPaWEE`IOOtfD>@gf;fcuZm4+wFfyDJ~e_1Dvu9<_>MoQSJ>$Onn
z$t8B(2+SGP?^3>4-2Tx*t4BfZcVxKeS}D%ysU~mMt2fV{wfsjm(tD}^)z8G?<_v!5
zS6<6#TFWfP7?H9sK!NY07b5?E7-1G7Zb_$dIWU`3+kYHK2|fQ~luruzRAMH4AARcN
zHgvUKKLTY9v5GdPO4i8!#!*N0N`AnHmVo#xM}chZl3V*j$i-)*Xc9r7!B+8w%9%b3
z(2=uop@ll4j!iVqn{e`nCq{wLLE;iIN{b_c#ef?99{VYh#7X9W=2y$pWe(#~T$h(y
z1dvDDP<bFz04qQ7v%q5NKolbSS-R~=G{o03GjjtaTxwB~NKxftfZbp;%gpju2EWVO
zP01}80?wCVdHOE2n)wh_q=w|d8%3y@fbtc6d){|%wnO`btSfgLi;coT|H7l)jUW>U
zzy};?)H_WJWSF7)cnEKdj*`dW3ZiKc60o^vJe~<-<RH+#wgB(T`t+$dRa#LCc{5HK
z=);b88TmIu)iz~n)B`sdRL>kTQ>#syN^$L=ZK+IyKZ`{95lAMn;)_8wYB@05Wc=m*
zi}&<tIK}O0>T$>~NT&|7djqao(Q5i^h#Zm)k0V9cP(oZPu=q_}<DdDhc74x8Es(#n
z9Zw&<j;qO={c_8%l^{m=!9ludGsN<AdS2G*)Pp*4Sh6HbM=Bzvap3H${^Mn=gvJPe
zlT>7I#l~Kj6F`)CIo&E%&m~jN$qU=N=w|kM!fTVrmsR*GrlJ@VKx#Y{CAV#QWyvmh
zpx82h3Yss?O@{f=(+A5uJkyf*3n+es1&!L`WRHx@K2b|=DKO)ae=`muXeqE`*uau)
z7}v_P!?~DG7Slt%CVQl)n)%f;U;$6Dtp<fY;T?qQH-7fv+yK`3PJG)TTeH>j@_gg`
z%&KQJ2*$+IdtQQ8Fg-33tW1Mb=(k16bupjbfJRc{F(WAaVT^}dd@BJ}4wR~zn>ji5
zZO~*+R@$hRZvLE!!F-jxk2uDjVJ2FH6BRoSFk&lct*TbbC!@lRQEtuHu~q||e}VM4
z`tex=X_p|wEhr|fL|*j>av~ncF6_`P^_`p#_;I*A5ibmu6|LTXG%~5>q7)NE7@1kL
zUWa=NvI>hu#w_F7GS?dj1JA-F+j1$ohZZRcoKl2YN?}|zvI{X<3Y*uMXIb}PV?9_^
z88VU^b-%>GFU9Km)0cV^jFpHY&?V>K?w90eN<`n9QyyCN%95d7MH01r{&kri-zY#1
z4H<MLcL75xG^zJX9W&hoQ3esMh7c3$grk@97On4fYVH6(;6!A917*L`D}3-q#4y@-
zOlMEKVzN+TW8Q*!t4}|)_Y)UVkTtQyAE4jrSen!)_=LMG0{&4V*gjfkWt01`(8$v&
zCHGe_rvuCa_k#p*RY&K1va57q<3)O0<b`}lC~3ZX_;yE=S2|iHWn4%x4~xDCk1SJ~
z=-L8?SO<bi6mra=zAGqJ?Tvc{H$(K_YDAK)b_3>EKW2`tC742jw6F}U|9q1t18Fsa
zWKdb&kfbp1+T(C9-hD(A(v+qByMu>^gKP33?|12(NEzVs)7Z>HYz>a=4BU`w#dYz$
z$YT|A^PZ@#Ebm$>roh~r!dEd+^Cd#q<k8xwDy<6cWAz3WxC1+$-_-Z3VDxG2K;yzy
zI~tv$6&Za;2QH=W$|JJqRKoEgwdS~a5BwMB=vS$HNq_#-yc)-?1fv|nduPF;62SHG
z7ynrbb0yLbK7j-a!NQNJ>{JsdCc^QAvA)67;9m}kubA=^ZxUp&!m<1P8LS@~Go|XI
zQTEu4-)(==fjy|0?{rCgMP4I&7ewpP%{fbE1dg|21(;L@k`~^@zd(-aN176Ys76t9
zum;BBKpw=*#LU8#(`;YIr0m4LSGaga5ZR1XTpNvrtOik1`1OKO1bA=a=Eg*eWlloN
zfFhDRbAP)T@)7d;PL&-_dRHRu0?GnQIiCKH8lgns=dQISYZ!ecRA|BywY}JNQ<~Mf
z)%{88H8eeuql(9Ox|AdQedS2?TGr?S{YpBGEEIa^7OVUPWlCDbRu2twwVJ1Wlc);M
z7Y8JWk7x5|eFJEXez`k-nPy6Au3U!}8)!RIs?#h+1U5Y}FZ^}#sR&lvO(wo>N%d#o
z7uyps^<9zX?Ew-nqmow?Kd;gwN^7aZgCozSP3}I#<-I#E{nz3k=sbQe7HUDI%U~+6
zF5H8_@jJ)mo__;-!B?)z7h3uku(a$$P&Ci<Lj-W35=ba-jEcfbC*lt;!-0)W1HqE(
zr4xjNG-bKgoJZ=hoJ!bvA~(DLJE6#5D(xSTHarZ?AUyhKF|VX2uH$7ApdcwZs6@jU
zOp+<!gAf^#@(vD2-k#xCo-X++*I;Wy+pVV@;9{kat4s5bXd$2d@UtHhjak%}YrafV
z?Me%B@6*4Odg`|!x(tSS#6Ft8r0Gx?K_k22u8w*fZS3hg)LMUttFGOpT>Ka<#IItY
zJMXFc(9iElC(IW&#F3aC->hZ*%5#E0!_&W{8$n!dwIGJO{V4pt`gO_FgOln5sIMPW
zGbL#dG5wnnR9MqzrEYne!^}J^ki}s-j-hv~+epw=IOJ1-z^(~9!%->acLtK?@7PLi
zLVTOF!Geq={eiyJc#L{7Cmj<{3bQR>aXSyQwCIMfBnR$xK!h(O>0|~7u0_S66pD6-
zN<D0i!qQMLMZ~%|W&eNwW%IaghkO9QS|q5bt9?+-co_q^V!u#mo1*oqN;^|(_lLLB
zv`o9KTe3whfwlo!F(;37lhw7nB7zKyN#^;!jrrbLZ?~Vp)CFPoPg#0Eb0%<qa6%%h
zHW6&6ti}I&{j<||2q*1m^W@UI)W&7Biru5*@6PBdgSy;>gqPWG)!u<U!n$R6dCQ$r
zZ$B?`4HMt}iaM`BiSa|DukzKVsq+@=<Hahw^+`rAJoCh0sD+9wFFO47_`?an)yqYP
z(dKRR{n69Zfw5f@G#W}Kx-M|37!F1tqF^HLj}7+D9I>;6R$}NB#+<Y9ZPf+n29<Y*
z+Mj4jM)YL6%}v^5z7{iTloI$L8?S1~-DjSn>fyeG*EcND6oM0<vyY0wFSUyy9auu|
zenhO>?)2~5QFUz0QgC62vLXAcGBq{XDM>13#uZcH3Sl3m+7ha+01u!zIKOB3BGSon
zE^V@4{%c~o%d4bkE@1+la6dBK+(o1`z2lLDiwP@NiqoIXHHo_Zc^=lt4O$TIj{r0J
z$91%E)hpgB{o#UJ`4PgL;Y*}^V_P%myrRP+U-*I7s6#DwM{dN-e&zNuNt|!Q=Jv;J
zh%*+~!c{yuh+5KYum_lW*(NlXbqa~-ODspaAGYUbgw*I|;22V5i`3AqU?13y&!)nT
z&Li<i+^b5Qm_euvbMFTJt28{n9S?^$n0LSV)WsMBcftT?M8ilh2ajf%rQe~C_aVtR
z>r>vYQ6CIK3iBteBR|y@4F7(005>fI?6nsxma(Vd-$xwmC@0`68gk0IB?-{*M-Sz;
zuv-1`rl@*}zI8ADRg&QkNFBTYyZ3UZrbG8&gW8g<QBS_zKz!SbbUtQ3p^u9%UlZVm
zn~Rt{-yBXcncR>NuXfivtlF6=j1nUU=^5)mohCOO6xa$>8s8z{oqrjIQo>BO<SXW}
zw9D=XB1d70?F^7L4R4;T)rdWiBLlZsK{3oYVvHy|h3hh`S79MxRIL}ZLuN4M*I3*2
z^h>X79!}6!QEiU-@AD=HAJvO=2}A$HEfA|V5)QtJdN;#_wl8#feGBBx1Fgw-fQv<g
zG-62g&eI=Qp1GRjre+V`+5B<(l&P#Jq~M|`zBncbjdS1~mo%Z_zYXbKZs^jw$OPH$
zvby?4BU!G_d=-m7xmWvLY9C#pR__>RXCP~9EmhWwFD<6erLAW4=~5-xw>-b{I8DTa
z{0HdzlVrTPJ42X#BGv*q(-e~HN6pKs{^7z@Q53l`Yv(ssNBq(Zf22}2e!icm@&}b?
z1J|`|Q5*#RY<L&!rL?Tqr+3dMCVV%*=9>4V@T>nP4b9U$lr_2L%+?Qqj4-~HJti$W
z+)fg7_57l9EVu&qj|F0LO^?>(9kmJ~6|(1GxXE!cWq^6aKO~xz3UNV`W+-r>ilKrn
z(-TZA8bb6A!5zc!Q1*182MH{4=suBA?RdR|@(mpLC*4p~Uq3proKwnZ66~G+t@a04
zA(us-J0ZOxD3;}I(Z)yphtrs$J?|vH1LsPo(hnnckJq*%&JV~YO)-gC<6EB<y>ksZ
z0^d3AA)2ImQRL`MgUST(qm7vL0bObzea`%Nt1-{_a?vHZ2sFL-Md_FbVC7@i1px`{
z@`e&{dUMy4S<6~zOE*Tg<a?Iv(%xf<lJUkF6<8jnG&*c}z-ha(EISs4UHq<o#pZbt
zTd$xjtv=jaI~g_eEkE2c9$4loxijZ^LItRSl<x+0V1-i0=dqXF0d%=D(nr;kOmSda
zeTH|NZOwe8;a@va5|J8#z@@rAPoULW)dw}OQ^9<sqajdyU6~GDtd@sKHTNp|MTe=_
zs3+q#b+TDMaj&Hv;54VvBRLa1tQ+iOH)Lk>;nt#lMs4!En`wl{CmukUqm~7O749_R
zo=D@4ckn$gfP&#wUO|#vY5uRr8p7N9=Lj_=@cL=uyi}fl>#yOS$M+mWRs|~)M$%@W
zPp+52G=kY_71O54w^O3|GyP!Fkh*RVg+#CUg#nXla?xTbJNR-F5fNmpXmA@rrSvTB
z9eH3rE%HK?t+^mV`WCw+>S$F%-djk=a1zP3*tnB{d$opB6`BPS>0{?(A^s?~Gbi-%
z=gbm)O3rGsX{@xT(y+m=q}5e;CtWrm{4g=5YTEKV1K#Vd({mzs>~rzITAG)i$Vh3s
z6-~VfBBR%roBEH*Y03l&64)~nwg3bz4ETukKD90te)K~h=FW-g3+yPbD|QXL1dV>Q
zzdzO6*&~US!e7KQL6^A~XidtVnQwPLk)@kVOjK(AO-=%$TE{H%a7cf>Z**}17x2!p
zTW<AC>apKu48MLn53-`YZq+eoqUM|mI{6tt$oUa*0}iZ*a0Bo*n+^~LNWb}lNA((g
z*ZiJ|L1_8dcYBXgB-S$V0$rx0^5G`EmqnX-Zwtg}pF2BMh9ti8^}{*`<Udjjn2s+{
znF97Dn{m~(VLyS{31L#02?mBh0h~fXc8-ZHav~ayq<ZTr)TvTnq~V;m5z<-skxR&i
zm!MCEugOcL5YJg*9tbNceqNIowX$0m0of?eA~}5O7)vZ2&vm5AWymhVA_sKf<?<=n
zha@#wfr3!z$UY6cbK8lo9dDw2DT8(CJ09Qennkk0ty?qeO&9AtgnK&>#0UPRxVI{+
zh*M#2|CYAVdAdJJb)ET~!ecch$NkT;a^Is?*`obWAb+V~w90m8^Z9ngR3&a!=@Z?S
z2URrc@1NN1`DsAU|3L@?0rn^lVa&?js_-D%VdrvV!m9;42iJS9tvm(lGWbW{efRPr
z4a*Gi%v8P_Q7w|$Kw9vLX+~oeZ~JfQ56h|Wcro3n&8=~koSt;f)S`bRYVczqgIm9H
zi^1D^{9@~}q2)o7hw3QE0pOdyk{@HjHK!kHe|*&!zv~$K7#RAf(@EoSe{=W?$Rh>b
z2|Z0j?_<P_b1x3CZ#;j#c)(w<M~vOrbj%_r_dDZH-w|m5(g;{65A>UFE;c_Q5lAq2
z?OiVLYUZ|x{lc`Z-VS{`oghf3_J_b_C)%N3wONX^XwO_9foI1d<Ii8$`oq;!rV0I}
z>RH$9ihg#0`}s@9<#!u(rT4-KOpZ<>2F4g0Re5jIWWDq|G!`$;5&e*)v**<&&P@z=
z#Ee$OqAj&LFfcjtt8rmO*93t_*xmN)qQgpH!<nm?V}Od&h);iB*~zfv{a=fIsjc#B
z9~!i?+oDD`raJoHJ)-{LDP5do^v5jP6u29B#ud{(v+@4vtA&t&HwPy2e)E==Z)@@*
zfi=q}68n=`lXj!CkV;T;Hf?QR*4N`-nEkR(V^BgekW;}tE;O%Z4T=WCCfqEhRHz7H
zl6m~m_5ZAnpuwt!=fiF)V>b`Yp=s)mxOa_<BtBY%Rkq8&Ra@xy@G?%{S)MN=T`p2S
zNNaUV{NPtD$G)ALA+YbuI&XYV;A&cSXkvjHz`^iBz^jsnF_{wIyMJW)Rqo7eSnJ-9
z$~XtKlRVg?{U3OeZhgb@3)P_Y$x99wBR>`SuH?b3D7f9xinhUKK~cpFN0#L&$iUqh
zdmBx__57Y~Y<ls^{%i-PoBvijvS9_yEj<drTqX)lt1Xt3c70D=?dc;(aR}NO9E`-S
z`}BuPK~Q91tZhC@<A0GA@3=W8U`|pUzy{V!7<g^#Yzb=d=OQj_&h6#WwOn8u0mbZ#
zS}bux+9*F#gTa^nv5PyUz;9|T9Xk)h2-2}BrAQU;{`E%YcO=xr0W!`^>{M>BM49E>
zeuGtDhLO9ECuciV;LL)>YzOX>gRoxoP|j;pfzDN*OA)Bu+K!}z(AwY?kXG_cgj8%o
z_|@TrA-lN*3wwTTXQ&d~wD9fA7WjEf3rHnObuUdE+junwFB|s^X?gPBBDT3_{G2eu
zGn^$@$?HliNuy%Q%w$4!pOI|k0`KS(ZSS+eoU=~d!+umL(85=%CqLIYfhWAVN>=9t
zWGSkMbH($P6bpdTc4Q;~uUQ{N%pBx0^t#l~QrPx+$S8ctIc>W8thNw6aF`D`_2J~u
zJVe#Z?S4hhf2wIsvow9oknL2MdRWyo-)p#zX>cf@n$6i$2I|O6`He)LHkLDoAHvoU
zZta3=ZvfiT^RSve0xqQ2vbBy%N{}N1@7i|zJ@`J1G{3Uh6*sbgoowb76+v<CZWxH<
zR}=S!`@SG#&iI~vo)0K@F{uRC3muPzr60KuDjm*HU5QQ^TF0F9MMN~H=h-=atWnUa
zo~Pe2YjrUxEKe!v^n||Cae%|N_I{RTjM3xKrV0R$mYr}nmlk-GxL3w-ZOW0OU9Pa@
zeb8K@6VJLmPJ0MsXsko|g+)(x7fsV{pYSnsQ+*VL*GxgnqK~7L>9H#Sf$Fh4{$8;c
z#vVaESyZam=-tmRgc<yW%p~NI-aXol`F|29HvhmDAJ@qazoWzN*1bDJF2T6`3_CFO
zIRF7vQz*Lc$MM3A1c;=K-%v&w3EPc1+@I^+1-0!3Acl`2cL6VrI=-*P8Yu#Yv}E(7
z!0I)FWo!`NpZO-v-52KiUH>CtfiFTPF#j{Q%S76MI6K#yfywnFqv@m~_n};PhuB`6
z5zoC&3WDEQ4dRop@ST@d7PW-GncYt<I-q!9uj@ERYS#{s5$MDa1E^^wnP~UBINzS0
zXgBDnfFipC@^!Y_t13EBh|TZ=bhmN0!PEoU0$9}H#hAx6Y57VKswR67FK$KJMDw7^
z)~sI~u?QWHC1;;EH%m1KS6e%Di?F<O1XnZOF;3!QB-H5KKL$T}j6<|5W*pRiz!?I5
zH(j5pWAsb(%CHyy{@)3cFCYR0xb=BU%wNG=qx{k1$1rZEUOzc?l}2I*3!c%TO<hNb
zCHo^)MV}$;r-3|Y(zkJbYbsA$^Y&<a0EmTvD>k{^0jCx9{Bi|%dj~f0muj@E*h4(3
zy?$dRJLCTyGyUI`3G4p<O__lI8)h2HBmN)Els5E@nbQ5A{wySn+1$R|zP>s?$L$Gz
z7AQ#-3ctS{99uuP2osc82Jf{td8>Mnye(p{-;h)nX4;LzWuJT$1LuS*-v@G7y&Gy5
z$2EfeE}ny~A<W~bUnR_K6aE)yO6*t@I}(rw2075`c6aQtCYPi4ESPk1zeGFWyK*hi
z&(YQ$vvXoj*xWTOAN~JgCLLS4-U!;1Wql%5rrhoutgzufFy^{{O}kk3{w67$JX4N0
z8xERTZ27}t%QrXY81_=~Hyig-Dt-(FMXUqduY=%|1&V-Sp3cDztjU5cMdcrG{r$QD
zWT?Z$GS2^nU+VHT2h1q#qV;1pBk$0#JX`|ho*I=h)$$a9&C4FUFIq}9bKOBuYGw{?
zqg!>1{DbT+$tFY1{$wy|A8ia~X|Q?tB-kWcDYL&P%=itk--{N>4lP`wMDjA<!*71H
zNF~l2`1pnm9YovFmho#wi27GV7Qv-{sOK2bB=K{f&Ke5`{5Pv|Waezo)fZ>_7+9D#
zM*FFWh(cYXv>vqWz&_VJcRa_sgRCL2U3mRE{3ZnK?II}NPUu~w-i%?BL_v-|xInlk
z&5kKtZl!?xr|0CsxTMGV$^ru8>ge;hwKyXM{0=N@G6Svs5PGL3rD)w}0l3r(bsp}|
zGyBS{Nnr{KN1cmlI@diwS8sVg>co#mup-OUN5@jfLY)rg*e=usy3=0@^p%rK-^o!n
z8uhz}c(@-eV;2N;k)26R&o=sDNT!(P<G;*UCFd)-XOU|(dAX!x!T3IG9%GCj4_RD5
zNmk`uey3Y8_>w_W_a^+dpr3S68ro>tn^CkVEii7T0-F5))`u!bfaE@S*t#KZHP$+h
zfBXZ325`&Gbo@nE>=a;l{G)%yn4QEYiWtdQ9^^?{FaR)qLbfkouly+K_bcrGl{)wV
zifDXG027TA+Pm(;x0)t_`s2?(c0pxx;4<Qx5Wq9C7f7~|TdiINPt~JXTS!B8a_k{8
zG=12K*F5Rz_)^B4B95f|@*ko?=-M=E61;qBPsXL`zQ_)aN9%t)-a~Y-C+n#A<iypF
zlKsj_S=u}MH6udig6RnUY9EVKUgRg-HW#}z!cU4WK9`tAxtit49HYzDHbe!qc}9hl
zLL;+l?qCWY!gzxVtvDrtq)4OR5c%xR4Up>|{9AH!Y+h8w+FHnxWi`KJmbuqB_*W);
zkSzWGQxM_!HQ9+11zQRBJ9MyLofIdaGl#^mF8_{Ob|=bU>Yx-He*ZfJHbXo|^tE$3
zykP_%tfLMCap@%(iNPgJGaDwiuOpjBP;mmr)Gel5jPcv|$MFMFJa{>L=2A=&nxs#(
zZZQUu#lsu!8733sB8x13>>F>*vOA`qCuS?C&(V8$Av!iKEKvaHOJCC|c;m`m@s`1A
zL(ezWLgUwTe(R<_d%!8r9?IZ_bZj1!ZV?(pePY%Di@`CFZ=K;H>$=6!eB2phK(KcX
zZXWSJGI&;vJtg1<_x%IFiUhzcH7M+D7-9+yQ6-oZpf=-&#(M?e!8D|JfZ#^gACf2q
zyE>yJwY#7w*YOjBHvlXtNg4`?Dfu2y(g{)9XyK?j5uNf;a7Xnf9eU*W5}L+)`N||{
z5ginuN`-(w8TOk7k^gtzG78@OdCp0cj8JePNfH(hQ4T{PtGFGD+3b~;jnY+APC0ok
zI#`aiyd#dL&N!W^C7{lzMd>c6Kw|SFehimVxd*~Mq1omIu^ojk1PPb0ZurXZMFCEV
zLhVQG`{15L(a$PAUYj3fJ<U~h0`UO{+#{(#grCZWp?7`({z+VuTNJ@daeEpy^R^m>
zCfWN)lRv<>|EK1raWg@Tl9R-K)<B=U$Rr!`peH_a9y*ixse>hYPNonC;d%DNl46sB
zKlwWZ+cP~+zs9pv^=S1KIJ|Oy^o<(4ugr{mce3OVLm9`RHh4|1d2VH?-CAOI1BM)_
z&xWHhQMWj;Be+S<h4)ve@{#-pBY!-QT(NDtu(Cmy$-WVt;$=CXRl+N}^d>cEl_#&>
z5JDzPK=V-Q3T&cAbaV${MYkT6aF3sBmJnl>XFW!ayoa&HD+-rCUa)&t<F&fld{5Pj
z$o_+CmM8jPU81kBK|$S}@tuGHg&lj$O!3oQy&Y1U4lfdgx`0gBHWdP~{@a<`qpbI&
z^w$j0Tp@VXdO=FSVN|o34;EGVv9OWvI-zr}l{4N$;%0fxMo;e&&C~tlERoDh^6HZo
ztgyX3@p)6Cp!uyPnAUxnBb_S)&d(BS;Z4BtGB#n7=WkBw#*DO+cFhx@5B`Zen%5ru
z|A$L5!|<DiSL4)ZW@0p3QMHPu*0~`S2OZ&r8prKSlqe0grw!6td;Kp4*`?HR)O8sw
zKsS}Ru3XM<Q2gH29}OzUc=U!P{D41EZ|-n|V!8?;_96bCuoR^+>7wR|wqzN+qJ;$>
z019Rt3Mjf0A<SKYG{`PM((cicpzG+j;EwK8@=o5`QjLlOYn8GZis4n~F@5HRwx~j|
zqNX|Y*t!BBrrv-&H59-<wALKZyfz??7Dm<ZpOzBU6LAG6Nb@G#Rm^vj{9TX!2G!P4
z4=8}V^1>C*l&pO_WEqNbeZvf`10ur(p!|nIp^Z#V-OFcYYM#<J%rOaGg^~xi8PTML
zFR>2|E1n=7#&`cS_3r=E)~K18@O%V-CxblP5}&b|AqZ&KmSDbpmK_e4PN?mpqxuw`
zge)h<HB*(J*C}yC5~zl*Zc?_#;1=i$C22;h{a)Yuf(D%iz1NtH9_TrS)Oy}Mwr{*)
zDTN)SQK44TcbDOS{MQUK%5F|Yx7(;2?NVq9mD-vlx~;#I^+~yadr1BgebyoSWJ_!J
zW*Awf`RM|thXYx6JY`m3=8}b_BQP2d<kT6Cz1S705~OwjdXA3<J#|sbjIP+VR|0&F
z4~OxPPTPf?R#r(Y6*$<#F<Fe1BwDpW=-(&6`J`8_J{E;U;%gv;@(t0?JX(4=BX0eK
z$`ho!y<yzsnghizpM$avzg<JBEHsPA{bZT(g$+|0d3H3N*b5!MfwhlFZ!v{VT$|ls
z8-foP2_8YD^^WIT9_PHvLt^_Yg}oE5j)gkeAAhBZX`qzpfxGZb{+-<N-9oC24vkdt
z6n*MhiD`?+$d6wB^<uRWcJ|$zHaT*9<Fz1cLX+vFYd%h@4Bye2-Nvi1EZNsZjKJ5d
zuDYUGAIjX!O%|iB+m^sZYsX%rYJXF!2NZ!`+L)C?WiupME~-aM#bW+hdFlTtMcA#3
zmIWpe??R{eEHBkKy1Hf<LH<ecz4>UvSJAw~ZK_-p#!$65HNnJDSyKd=mRGr7N8us(
z`AZz~=pXpcxqjY@3hmFyrBA+|sx#`Y@}w&rwASrye$wRfn-fja7i*rxbHSac7Mr2p
zf#DD;ZNlKxv&UC6TLoQ8@+|+orn0*XfZ~I$XCu=I-3#9e!HW;3NNQ$&`Xu@Hl(UG%
zmm2lF0kWvV9Los$FFzkCY~{<y7KQ&1$S^^ktw@A0t2p_gU`6r3XQ7a?^N`FU;!&L>
zH57c#%N^Mp_R9Y5?o6Y_;>$^y8GQorAUo$3=bnX=u|U=*X9*Uuzg^a1#)20xfj53`
zDRsy9#kb{EA7;#39q7=ZnJj3QPeyUjVSER&<md2>;#cIoJ!|k^y&ZBJV}6&!O^H<F
z-x2LLz`JmFIk3{g^_G+P_XsS*eQ%U+UpkouGk9R;IS9f24+E4B%ZWVt?-~UK-?M<!
z#aw->B#@JynI)l*NyL-YVs#)72f<!&@_PE17^QlXK@z(dRlT}>BmQq2+V>oiRo3=v
zedbo7)P)}R{N>GiT#H{`$4JMR$A+QKK#o?txG_>*J}p~M+S5^VK~m|wvf61!iFoI4
zFYPktBF1+zAxjH{=|1wu1r6dRwhcEzZW4F<ye3h9C%d@B{}5l`)cO2F^T&y5*AV04
zbBxe`B`?f1f5aaP|1V=M3pf|!5jtM!dMpSj)7-QG907lYO~Xkq9%mSh=8@iVDKkWH
zc*2&g{}0N7lUZzezf0BV!*0$~dCDNz8_e+?{&yzn3%vi|_Op}+s#uOABht$%>m2aE
zGe>>_`a)@GM=(|HbO*ag!j)>GtpJ(WqwMc?wPulrbeMAqZxh=8B?#K$YX1`iosu_w
zA%aP@u8Zdj6068+pDgzBB6NAe7VU-9LY6!9kP8g0SGGMk2Rft1th_;{AG5w%EOZmj
z7O|s3yUp!35Z3Q30pv5lD%!VmzQ2qQrqhM@##QE%-q!=N|8&>FU%pn}unzm97}mRH
zdon6Z<ye^ah7vm<p$L{Hm+O}|&#Oi%D^L+F87as({njbIJ|3+EMMyvgWv260+rk)w
zDlhM{EOMX;W8M!;))rs?WTYQZ@Jw8W@<bSDfG`qpc8O=i$_C^D6O^gL8VdSH%`M8c
zN){;!D|HtMSs2}jOmSa)t@tDcd|ulh26ag*bs*fgP>hra*{Kk$WoqrWXJ$86xPQfJ
z`CUzA#vvLT&njj3?RZji+uwq54wO_kp$9;HQI2fM-Yz^CycBi^y?<WlcnteK_}bKE
z*(@w>Hq8+F&V@Yzg`V&lWa_p%<GkgDICB0hd+lbEwyB=|WfTlJ5@Q?mJiOUdivduV
zlmuHZY*>WoAen!P8j$kB66PNo*U#w1-^KpCwj?RZS~0OT;!>gUEIOtZAn+B#5SxCi
z#60bAOfc#7)pM7PHxQ2GZ;`Ap7vYvmQngR83+I%ps5R#VCUSQ89_nJER@^_kBny_n
z{L6-4z{iil+Nld;2!BV9TOcOk#EOVKoN;P6KPMsp{Uh`Nrrkz9r~M&|`e}%*8aY1s
z6$(K0=0s03;S0FMSIC)Ag+rCpl>!K@k>sdHNSX@|`+7I|loJv!3~X{Q%qtigc2Wk5
zbvX7@nzq{mBibASn5V*&b&*0}gY3UrMD`<Eo7oFbA#`H%rn!~vOix>JHO+ZSFzBL;
zW#JOCkcY=D29zohxEVLc@Q|~2?EKFsCs0kX=}}U^S_1{K%;qOz<w7BYGqTuWzt9F~
z#p})_G7BQ{Plo4ywl?QW?`rRQ9XVaN^^WLf6AG~a<JUrdrXGhVH@h*Mj}-u_TH1V9
zj<8oq9??0a>nqSKKtoh!Wn{^l3t7hnT%<hfIN%0&rxO7I+5y_OKXeY^`)@Da(Q$vf
z4xjy?sHChM_|w4t5!l2Mo0^=e81`}VwwzcT(*Fx^ZJU7F=UnUU!V?!P(Co&21}@u0
zo9i?SR9y<fgU@2|<^JduSprvU`@FSBYddL~Fm9pHhaSQ?A!-%w^*368Q+JxK%J9v0
zJ;=GwCZ*OvAb0g=tW<e~<8npsT3%?GXN<Hx)RQ)Mv<uOiC8ZcJ$BNCosDM7+b!npd
zo?Hf~Zzk<XLwcb_dk|I2^!ytj3nyJ`3FW{Iyc~|u>cIv087&gTUEswp(4i3C9k+(q
z6A1w)MNz38qH9)QsG#?!Bc!0u>Aai(JCrFyWY3Z&;DfJ7?m46|?d0{`12@EVC=@?b
zpfQH+c*I-Fj9G*z)hN`+^4NTN7_0g8xB?gmsbYHH0SW_N8oEn7)<J#od~`cF5o4>s
zMCB1>!h}h(nskMWW4|pc;|``1E2uhUm>-hVbnFPt@F%S#9vlIrdwPW~gcp|xGl4AX
zug&vK;r$CfOseJQltvi11tfK~6KkLk@;2d(0GSd^7SnCk+1oYoG3MAFN+$60jFY3X
z{1i~5-7zq>TYvv3sOPjN!Alg=IMRCHX&XsHo-~e&43V*(JHBdd3F?};5GhxzVx$n^
z{)(!t$GrR;94*PCrilDO@5}X`lUyy#J2>MJQz$FX4>93~6KXXH6a0TjZ-3*>p6hAr
zad)UNV~&@}JgIT=YN0p5A`Lh)<mHj1)p<oDW`_S+_|7dD(N2ZWL5@ASu|M(yvkAD}
z+$&E7?pkQQ-*Q(cdxIjH4?~T}sMpYb?XBUWj59bfjlcJOxwbh0oRaw%Z4zd1*#`TN
zb^wVVqb$9GTv=<L&Cnxcx@&BK(bVBIeRT>w<AFB5##J`;BkQpPmS`;8g&XZa%PpA`
zvYTrp6IlZPwidLpn~l<xv4>Y6LMMGNpapMrjL(CAsL7iaZp$js{qi#u96Hwh%d6R*
z8-EjI7SnmxJoknaWB9%)3*#Fw3(;bgmpHjfMo15S>sNj;NZo&6TR%am;Z|I2=eP*k
zwF=)bpJX@4p2L8wBnR5a+$k&_4@eEGO$O5keZF@$cMl&QM~5bFGjGa{(B0%!T~*|C
z!HdKkv#NQlg=B*IVNOV4;PdZF7h_9P6*zwEp|O`$7>@ij+v|S&^${@f<<2Uln38PP
zTg<hT-hju%^R(&FUhM@qBo3G86ifn031Ah1z&Tx~B$qw@#B1Ne!o$QuTMt75bT4W)
zAItYBhOEqsxnU>g>5>kX@W<YV6?AQ;N8eegy^q`zINwW3Qc2ANK*t6p?f3iE?=U`^
ztkU_0Yz}P<Jqy4Tk*7q9=u#A=58ci-W-}Y!TF8DW-cMRPDULf8v97(0|3o<e1ikAC
z+BH0E&seS=uOiOkdY1RjLqS<y<RNeU{<wAFGN?PGlvCbJe0$yraRIChH^xDB064jL
zYzgvp$i}-?b5*m@GCsrN>T3FZ*Ptl-P8+Gm-r%G~z=83ZOm5Pup;2L^SNnqa<66`1
z)737U29jB#tyDI6@wz>U?Hf1Hh+j>yd?F#K1X&j291D)&_D!>X7z4vYSuFkIPq$mR
z$csW3%e4C0gqC94>+R2nl@#}Mc3PA2lgAaU7*MM)T#xnAd3>ja*-(>U*etw(n^Qj{
z)m52zaq)Wi5K&wujY%9?u{hd@$th*g#HSo=!wLj-ep;fX@+cNa%y(0|p;-^E289ut
zV#Jw*_?_W>3w}kzUe_z~xe^PR#<>W1<ROl=i#4v+PSIo{NCIGV_0@FBSzFs#27T_+
z2eZPw)I<i*pvl3(@L4@T#Y24kweGt6QqEv!Jx~Gipp;idVCF63Cc{bS4q4aab~I&C
zUPE#-d4d$i@UWwxKbT>w5>tszftWEm?pQz4YzvJy-h>P5p8k*CMC(Olo(=1JXq%pC
zf`zdWA!3-_o)EBZ<>)>bCD8Y3;~kTwLOfORb-&Gi_OpR%!VH25H+={@WWx|+QYk+!
z?*)WH>`9s;)38FANDO=vS*$>XPJ@|wU=gLfEJH-O)8N0OP3`*8EouX)l{PNbiCgjd
zKvJ6?!N3!VaS1f!clMY{*Y0{82)3P@wAiv7TY_mZB9IN8vPR80VV`P5A!D0rQl>Bi
z8uK6$dR4kOeeCKY5_HwVa$LHwgY;}h>D9NexU7ahRrK#pXZ;T3&8_$deQn<N1<`T5
z(9`{Je3bXKVK#m~Tys?2)EWPBwwC=3<e%GzG*Ca2T9n#r*rc|x|DGtQ!T<0P4lurx
z*`IgYVFB_ArenSuqkwFtAq(=lpLOBhy4s3U*$xtyB#e0W4NWS?<sk6AtRD4<_cFW&
z!Mw4<k;a3o^n;!N7Q%0%ixVL;<9L$Hg?Dx=J6Gii#CQb|&qho&WR<Lv<Qf*0aa2d|
z6ZQCUl1^xQ?h`Mn`LPBSm}!-(d>=r<s7l>J{D5Qc$)ZtfhlH&y9nvXH7jAdo_W%{Z
zH@5QeUfnP5T)`Wvg}G-GF2;0!`x+_{+%(*I-}aa$<t8To^}sg0v7t+SFiry^yqa&E
zphOj%DjSJob&YodppfCo8Kwhdyu8{ZpQ@-qWN$ASJXMKD;hH!%p0V!xGbEbw|1}hp
zYXc9&>S<L@yarDsIMq=(2s<Jujj@+$Nn3#Ry39NQV#Z#wHI@m#<61N~0XsF#sM&)(
zy!qQ?y1XpQ@B~{QprZWA)#e8y73QkmXAIx1JAY43np<dcub2(Q^5JV`7$u;pt_q}q
zjk0DdR7D*7lPD#pbug|n-du_FlOmX-aRBu|nx`8SH&14KqV28eX^JPZ#u5Pw3571x
z#f`Rrr}M{%*TWEj6xpK|1Qt6lSYF<wFrFx@&!g8%BI`f+OBaU+A9inFx2tkUH!^GD
zXkHn+miEKyhL`~asgm95oUVj1wnSI{r<<lq3x!lsdI_Vd4Uo$l&t=-X{>dfQZ9u1T
zDkaqfPNw~`bcz~^8l3%&EAcs+yRkaK=<*_sg2~^sP!@_)(`d=x86PKJ>&lW^<ro|y
zZNV%<fq)YRP8QXQ$lo5nzdfesnfst4;BO<~my!G<An~A&381!Y>0n(sU$R?L{pUTD
z6S;-LKRY%!hX}Lw>$2R1r%d#?4S+J(yzIdy6Ig;)7|}5UXYjbW#+iER#n1FEdk_Xh
zhUQk;Wu<ai*aFKQ84!;zE`haq8&;I}tFTQO7-(yN<|MIQ@#Qr8*A6?sFwd%;oh8Yv
z$tMZ_Acx2VYYtw~uS59dtCkj@*o!Om^+>d+gbro)ipF~-!=11)f0;*-Gay%MV)Dx-
z{f2ACM^0kQ>4EA_F1q2OChG2a3{FL8A_FvRk}PE-Nvd<L=1-WVc%UlcOU1+PNSVil
zM!~|BQ=V?=z^uLcbrOgGQtX7j2qoa~_Wb&{Q;4reaG2jRi+VKGV}%rP&Wh^<$GNn%
zP_&&h&H$h8E@9G{^Z^A9HSn?{&C1|?nY~Kd()#`V?iZ`UN`OS9caOK&Qp@0cX7<aY
zaq#x2<joQ>s>?O`JGt#jbV>&Y^!aUShdvpAZBuZjPGN1FiUjM#Z@^cj3C3^NAjmQC
zVY#QVScnbrER*;*!m>@RY5f`tOw-@_7o=Km$rIls7^E#Fo=?ewK;RHY`8(!NK}krZ
zH$@w>xiy*U@8!_xBYpUjsvwJ9C}xea$4FXHm)sSZC4_)nT$o?No;n!7+eol{T-;*e
zxY!MOFASp$<o%sr_E}zSBZyoBF1B|^R)QYu_;}kE0emdyi&-~4pJ=~OisSz*PI#9k
zj|P+XfYW6}WdvQu-~cN$y95148LxN#EZZ?<$4QZqg4=4wQZ_ahhNq{<#+rMG)kpA|
zgy;SU7~#MSobH#WAS)L&Nw(pbe+54s7C3iL-3>^j#-VF06Rr=Dug>J}<%qe2K4;Dc
z^izc+;&q07;jdx13?>+L=7PQnzqQEGC>wqKNuOv4C;{9adjWaa7{d`8Gy1_0ch9+0
z>3z0RPub+*blxjnWx0p^W|uoi=+!J2t(+hB*3ZiRaXyTXoH!NqvS$}&!-^NQ3Ge2{
zA!|*tb7*<$)psPTEFX5!ipwvE63r50-0~S0%Xiu^ou?|(Cq>jfc!!yM#ICJtGZ+OG
z&`zn0_`S6}69NfIB1(V6!TA!jV5H)dX1DwEhJ)H*Py6HasQa!%igJ8#{?z67aUg$0
zGeN@;-rVCUPNgAU<1JK1QNc1|-tif(8!Y@?%olEYo7BMGRb&dtFm+ZQgeFylEy=b*
zth1*_sQFQJZGI%UPj@{QN~;-2w_=-|mjjGC>GHg90{~~B1&27>^0hX%9=fYQT6grH
z+@CTs<@^j6M>!?ANX0Ys%pQ@w>lmJc1t<OzvO8WXwPf}(^A!+p@KZUI#O{Q4@}^nj
z&cLeQ@%H-|Ju=0MQ-m}VfMtSqa}IG_Il)VpjnQ!tMcs3oCC8<TG~PX~)ta1YxkF@m
zTkK<}N&>gJeWw0ix#P;`;V?l5Zu!q-t~OhVyJ?%$VrQDrk5>7xglxp;N8SN&Az0B%
zu4%yEcXYQ2e!r*0%9EsZ8-X3yUfh=*_PQXoYi0Kmj#6#S#H8O8kOz5EMmfNUt6D<Y
z0J@dITZj^;^aZAI*avpO@L~zfAud2v5-L<)K!A2XC)^@Y^YFdj+=FIHes6Q>Y`!#f
zv_aQt7V6A8K5sQlAuG(=f1d5MZrJ8xwLN^M9O}(o*?Wihb##6~Zmj1^)vJ3yk276g
zt!MRsV{_MIsv&e4-20h`t-T~q>!7L#3D(2X7TlMga!}@z7-AMiC2pw>diB5wCStk%
zs|4=L^Q=QI{z`RD@Ys_HK}4VN%z$>QO?9*u%=CA9jE*tlyc)*~pu~zeC~&rcI=GZ{
zD&^6W?<%cccgJ)7AxP-7rrW}J*E=;8XHN2I;Zd#-(y~jKb&V^v!Sr{8y{<Mob-`#)
z=&qkXs8b|);^hSi1`n5V^_DJ^W~OzZqzYJ|Qb4A^;T*{)9f|E$io}_k@l9weI4RMB
zBzjTUF~Lq)u8KR)2e8re{?n;Hk61+N?@-cyk{npMwVIq$r1UYNQG5D#GmGN4IeDtJ
zL{{*N5uE)_VDNiA>Bg?yLU>QS(sIOUb*!RJ0?<~!uJA0TO9G3-+)$YeaU^LyfGoI;
zpTS7^ZJ?W!+y@QH)tu>I<mu93>XhvL^RxPC7O6eS!CU4>6|j`S`4%wXBXm)KJ?F(-
z)5=x4%zlZXeJ}#(@vX}=EYS-<XDsuzZ5Os}L2x(H7RpgR3;tiQ+7bx&NsCw4y_}S)
z+Z!h%at%Oy!E(@@h;b}(T;C2j5S^Ney)3J;lCs!8rQNJHq}l!5F&ggf=T7V=f%;z{
zLq(+<2o(}IkqB3*&I*W*F?E4RJbJZaK59YX1?tVl-#u~T-NJF|WPk+~DSIzA^kHiB
zAjcHvAMv*wWh%>ex!sb;XGQ#v_NfbDr4dKj7Y@~ID^|%s@#@_b446Y-V7P}yVquM8
zT-;>!?YXAX@(U(fyLmp?lHG9b0>T1WDz-S@o>52#dt03Fyi~XW<VjI}-_p7lT<P+4
zv0|-4)?Z`~*D~`0(me3`2sPX84#I$8`rt!(JsvfJ0Pa2mFmZM0!iQdlh|BrF-0Tq|
zOTlJqyh&@1j*2Jk*s0AK;H&*=YtyVLq(_U&HucZ=D0fls3UnRe>$(h7?a=4vj7Pzg
z-|foG4JQJ3Gotq@gy%fb@r2j(BN^?f+bjFoI#<IXFptQZ5IsQ?>i#=Cj|XGDCx6hi
zP4{WE{srl%08z8t{WS@M(>aMtoXFO-W_DGDEI38y;hXvgpe4_Ufdw!MPxC}@En-@!
z2bFQ#d<E`cqcqvBL+EDFJUehPvr~5mt4Bh!m3KRd-LR4bWZiR9fS<aogFgrd_5)M~
znp#F=Mp%$scPrz?_$5f47Cyth*s>q`3%+^-Dpf%>s!vGh;G_5mx;VAeQ*xDBt-9sQ
z_>54pUHk#@+*-HRK*0ePAaO>bhj=i$Vtz8|ND!@1ak_m@R+)#y5SEj8ED;Pr7gdN?
zy5i(UUpQ(p_~8lx+ule^ns2l;IZZRK^q;hl;y~tPsU@5PCdpu#bP%vO6`u1vN02-c
zvw;4w#wu|w#|FeWWDBMAEFcvfC*u_XI|spQAMdEh9f7bOP|FF4gW7^X@}An6s9z`C
zld9w&?_5+tl{YS%Fr9KZbGrZVv(Wt7wR3^n!oOHH6ZlqXK$zMbj%iQgFH%YWAZZbZ
z_KbG>m3!Gb-JC+0kxQrDX&+f60uKW5`ZH}L0h`e)^qk*Po1p#ujn_EQoB8Q7Z)5dw
zy$PMpV}qu=z-?@wok@lOBB-CJ#YQ}e1*#OZ3{lKf)mK8((+=k;mC4Dh-KGuy1jN?4
z!$Y#s7?Lfc3IIjw5wiIMb%QMgE>8Lz5!Ho}WQMn4a*32#9l-^E1dtQcji>|MRNmw+
zx;!tKRf#Cxx$zSj=F6*gO~L#T0O3`Oho+fw2CSthK*D-n5h}vkE<3kR=z77VHaLg&
zVqC6)KLF=rv)nT9c}~wWOg(S1;XCfa?F293VRV(NG2P~mg}3;2u(&0v0v{#zTB6Kf
zqhnGp^5kyd@(k(ed?{K@Z4PJ8rMev$NYt|!ru~Qc`ZqId6<@}>GMVOE?|FZDFth|+
zLvs`ufsLBFi;)%&vf4K+Oo)Ng01dklS{|!LfPZ>?YkVue+y+mhrTfB*x5kB1V@0Mj
zY{#NT<mGXOV*69eQmG^W@+oRvhp+gBqknOZZ~)89-C7Td;#2pS(ka`soqM5RcI3n=
z@04k#$4lZf<83NFv&pM&UcmJqS3UILph?a&a44y37p<9n#n(YWh`beCdh2Z%_jE$3
zLhY}mJ=-OJCvl5a-l1;)1y-KPnC9~OS*Zq1L#EHa&=8a{wyHjCTU>b;0zT6IY4_{K
z**~2l%Od-jM8-D7zcKi_G4MF~_0pw_C1u?g_3y41_5;3ah9kPo1W1DS%3<dRjPq*K
zfDor0t4+oi&z9n0M{BF!l$LaAF)1+EYbO-GkCWSFfVS_l6Sh+zRu%df&3r0<4akIY
zmd|vFP{!AT#W8u6QOd>Ux`DvJo~a6#B;~~ZD`{>A2vH~Oz<KV~KG0A@W=m#-g@(%K
z)Z1knM`tD_%M#`c{_JKS-Q9$ca}EXp%w!jleeNKCM~~enFl%N{h(lQxhe3e3{M&V&
zi75kdP73x6QQB_gwP1q=;SB=@(5D6Op6EN8ivKJt{(@x9)&tatSdi|8u{rG7+KWi9
z?<>Sd1s6CVc%ooKm%4pWYN3^oKa8%yQRBJJGbGFPTGWN)0K-5{dxh{!04Q8QkHrqo
z!{0OzVHSzqD>%NS=Ur(C0YvYYCmW`v<@`oTZ?(LZSvy3dJu$I;TLT){v{2q;U64IW
zmyBBb#FJKn64@)--|`PSp`B8vNX}WxX*yMDZ$JYmKmn!a3}R5Jp*RZ0$&hw;+n<EZ
zs|4W>3-2?Ab?~?DXJx^P3}+P3!q`$oc0#0)mqKRXk*<%wybAE}muPUcM~TO#E#+Cf
zQ28IY)J#po<HR?~%EH(oZirM=97{?TN$ZJxZu;}dWro|V0^0wZoK40NNrQ$#v3w{A
zjQL;2Rs>2VghR@yK8~xwoWXTQlwqzQH*wTa*d-7*_V}Fz8x?6|U?&<_m1#Uwhe8RD
zFmViZ*McJ&3UvGU9MJ_OW(jJ0;zU4e72XvIa&mt=-I2RH{QPk9x7qqhH<8bQL`J%E
z8N(7QW3Wt7$*+lL<wC_0^{(GIT#O}wT^iQzn1Ug%!cwHdBB~OHA#V&@EMRTj9yFi;
zxraLy)Qb9xS!S}}43`_|cDc0|ycX!*()Hr5uzQ8lpA+8mwv_;;4A-v)CqX@+7~a28
z9B=<<1m{7t;6V<`&8+fE5wGJqkGk@cG&l)EqYOi1lQ$3JXFzC0jR*@>ZF?UQewULH
zIv|3Yx<EdF_h3N9S#|_-0+mdTNyaEyMXuH#=h7`reZWVe_HQN-N~3dFWtg2c0G3G6
zD><q=<epudW6jDXNqPy&yZKxDWFkKJ;7=mSdd9^o0XpYRf8b52TwgRh8}A~%g1#si
zrs9CV49S#e6@zL)zmx@Wjh;7b4W%~#dcy@~H)F`ZiN<H1Z#}q2tQpAKkT~3gI~A!Q
z{nS6Pc<-6YnI>Sq^3s1ecG_oio0{PDvWVt*%x81IXtRr`dTj#gRlDuo;-b7oOl%6F
zhY-pk2|lG^)%*tuJ5PetZzDW<3i%*3A?-WaN9@oKop0FswA94yUb@iv5x_)znGusy
zb9E)mO}rMf8Nv^OYDHj2%zeThP2#B&EaYu7P7se|1jJ*V++P1*yj4@;o40OmA>7~(
zI5$v=r|Mui!k4QC?3WW^k4Yo`_DE7%36Qd3POna{@dVhmnrgc?X#XE{on=%T576$5
zL$Tm)!HPQ+ch@2XihC(q+!rT6ixzkHhT;^r;!bgEk>bH!ZvO9mKizZgr)*C4<U@8d
z^UU*`onhtADHXP?ey11HYWW?DK5S95{F^M%^#!^tXx51<m9sLE&q6R+w%+Ng^b=I}
zXEnw^*(k2zth55{#uavD4Zw4*@=4#~wAw1GDSDbwO(D?cQ<*c~V}DmmT?}?;m*)qu
zN{B9>c#(HW;Roq=(F)r;F(az`8+>{xI5I&}dMNEQu);aBzrn_zSSm_pWVY=i7{1RN
zvIHmM4A76G*mEkmTi`J9rn83)M<=t*F(myM3XGRmi573{(YA!KvjMQ-v%D7Mx<T2J
zx1yPXDKqa6)GGK&CccTd|1<QB@5ml-s3DAZ*vWkB&~ch+!vYK<iQW2pO=u|5bChJJ
ze0pDLc(YTSxn}r2^}l1D`DUTPvI-1-rC7xeVwfDcyUIV_dH?#N{NG#Ak7p2t@3F0N
zJfBU)4@l|YPVf<s8Uz$)HF$XKK5iQ3o#q8h_R1DMMm>EHV<RpkmY`=vNu{`8k{Tip
zyCin{L!xmg*=E~ZtD-(?%98W$t*)K>M~EW*$8*oP0Em?(8}z6bs586-n9?5lDJ5|D
zvoZtW!~^V7U_FQb=Nry3^>|5WsGdli=!m+o1q=Qki5g(uOpMAB%n;6mbD3JT-7w<z
z`D26F9+=Gjf%ozL7tAywhNV89bGv6FstF`G7C6k$ds3&;zpK1|TgM7ob&|Odq%$c=
zm(NRB`VqUwoD+2KEArtp6dNPipwg18R6sWdW%u>HBrg=&AJkPhc2+)gQa7aZE@e!Q
zDT=qtdJ$kou0lfhCIr`WSZry;kNz%{+A3R!_G)p+mLHRl6Aj<O{7g@bM?A=!&dS-h
zDN2d@BFe1w^24fAhR}&8a3&|64(C|P>IA7o?wWOj79U*!t6o`G4FkEBFRmwP_*vY<
zP3xC8iebnX*y|o)-cGjHm(~H^H}(u$ZHOS{U*5o09IW)mM!T!jTdA+MPk#;$?sq?-
z(bCC2uSCh+^P1_a))K~f+GO?5v05iCvK3<4Dpxa7X`jv3L1yd=3_iU^?pvClW|3vQ
z-&V)<hWn#|lVFN~pG=v9HE%&{^f0p0lG1pj=rhV6O}a^EuddWE<8t&WGf)qbCR%XU
z)y*Z(qS#uDX6jWKrI4?ydr7uijWpba?a@_awa7!;?2MHDdo|ka7Ux4QVZ5F0i0Jri
zrtQi-gyokZ-Svu^Hw{w&g$9TsW|}Ph(hsCjw_PrJzdKcmH}N|v1gZKR%~+*Xxc#BP
zq1f##-b6Rt{Z&?U;H3GWfT<yHX-`ut%;zjKqt&lh`sGu7kjl@GK0LGQVxH_#-RcO5
zy%eFayx$ZC+Xej2vo5Q=qF#;p&OZz}vttP7JdMs?vlVfsHcU=z4pjogzT~O*v)q3c
zKTEdRjasD=QrAPOwBH}Lg2_#`ef2B1BgJ6&uy$3Zw?Q6J7VtSRglIc6zTYOcPDYsa
z#Y!WnN-o~y3~sX8#H~9|tKPHE*e%iImAn%XVKJRp#Q8n*u65~@OGB8qXQh>X%k;0N
zMh!gyJTaYy!PeNbCdq-V-fR^6lXJlaZeB`!$3;846Wo}kJDPWYHnU8IjGid#6y;Ei
zv~3P+Yt=6%7h~8v@xQgZcEap!gt%X)70xYwgLu~?w}3Jx@D_K9-z~YhqtSYE_eAy>
zp9xsJtA<0I%HAGuZO!N2-ZdJY_jvXtK1#z#+N&|k){dj3sXhRYrAjzqDt{H>S67n7
ziNj3w6uQy_TBxOo{?H*h#ae)nUl4Y6!+TZ|BSpj-!Y?83N6OKb#j>8RYi^&~TE3*n
zk}-<vKz_!sCjOFBlZzYtACmLGASX#KB>he85X@>FeJLtbg!Vw^@4xE(TJixOi~Op2
z)sH_n1_86@N{NSa5uSUG{4W)yD|U9`sKt!7o1crcUHW(C2Y&R1hu(c|7is9+<YjPT
z7%nJSipZ6Pqa3E+HS}>i*3);wtdR3(I<yvmU0;iNVC#bbe-QPf^&o!8|3x`psfe%@
zncRIUDAHZm9wL+(br@Fm2O^@ZGLNsFC`^D2S|&rp0<zuUoC2>4Gu(w^TPurS%8wMF
zG`DokLeNK(s@&K2^j-fy)B_((eM0c9H%nyyR?ElyrUdGG(o-7hHb=>lyZ1DHJN`EB
z4(&US?~bYiQx(`HELzlH;bL8QK_=B@K887^G#(%p@||cU9qnXHAaZ>rm}*PgS!F1v
zK1YY{?SUxy*BRcClwn|?cw$j7E>Xr6PI}}I+l2-jQuTVwFmHG!Uqc1#Iq6-8r^Gtn
z9-$Vwo(wBj=z6|kp~IUK2zkOoWv=r2f*&9q8I9Gu7+f*<Lp{Mi$6~tSr=Wi3Wa+$o
z&Szj_g2$x7vPHE>sXoz}*N8rfTh{qo#-C-vG{y0v(7j^CZN2)68f!|`u~CEx9(OA#
zcI>3+UMEhA`M^Xg1M^)xMMNr)b=B_2W8^ejWokXnq^Jovh>b%EuX*-S!n%K_=c_GI
znWo9eUzq=&>Y6fuwidc#)sjB1cOk!b1-Qtmg2gGxxr}o<*h(y~+>#8qie;-ae`5IT
z7#X~o!yF4T%zM>S?#QswM!|f04n7;WWu8pj$rU<vJRIN!&r@M=&)a%4$oZ?-CXF>n
z+lHQ_467$)F$g(@X1%3Pbf(rN%tl-BAJ%n`(9GeHdcgXNU^3?`J6~jmeWj$X12i*I
zs0bRzVsk<Cab!(Hu$>~`+RRJoGV8BiOm=-`uP;irAnu7E!!Bn{R$YH`MRAU^bk)mv
z=zn0fLJp$KT5o@RHE^?q<w_ha(qx}SYGRpqqQA=Q8N12+EH5NIf-=(2k*kMhnSi?P
zTlp2Vh2G@$27RO)$<ZPOpDkHd1IXlR`>>6nt+BN1%dP(!RSy4Jkx+)gt~i}fI~FJ>
znWS?^hU{LQ#fS!ih4zn<vc4tTF&Zf~Eh06Y`Z1_gpspP`u+EoQCQ#PmQf_)+C*G#r
znuU@is@{Poo`a+^bA*2VoXGiTQCYofU$2rf!{n>EJFL1^kFbVh%$(yE4ZM{2_yQw8
z8K6?z>W%i&U;aJSv9Ic6xixx>)gTv)eeM2J#4n_6<IJs*T_?Ecl4(Ppu)3$Kz*Wy<
zyFe4;&7(CeOwl1(AoocxM61g=XG6KjHGthT$IKjaON}A}QFiYaMrS*E7(10}&VnbL
zgjgG&Itz>rCKN|aIb?Zf0dN$;3oA~z1$mWby>-kovov*RNp3E8AIL>dIQ{x3KV3!b
z;Xe{G_DbkKAhiBxY=1q`?UcQHc)dJB&!~Z+_OxGD%ER^Jnp%*32CQSoWThf`XD}8o
zY3AIjhXmQgf%x#X;kynih810T&qh-*VIrt0({lS;&Tp8F1_yu*vKdDA)~lr>6x~8=
z&Pf~^(r-t7Y-*Q%kri&v7G0k%&oWG+kaz!=e=w4_$Vsv;#N%f`C~nrr=JLd^)eiA0
zXL5lc&RsSm5Y1;UAI&`xeZ!&s3>U8VKHrTAfhRm^qL+TBaE_H|Fn9Ckvs`29`UTZ~
zbO;53U?V^+x;Q_lXv_STobYWRlrC}HxvzR{L_19*UPGwlCJV+)iOY0p^_u7*Mpkr2
zc3wPbjf78Nt3t{#Jhp8?#*a<9w52CjuTtFVHHTEeXq7DpS=_XvG(RW|6rm`DH4DQW
zc4iWfsU3ePJv}+p03!b%%mKhT$5iN=(?B<zb07(V1nsF($cdIpDByxgna6YyBUGy0
zU582$bLYV3-RZ04XRrlCfQfSK<Yx_p?8@jXf$~#15{)<T*HGS^gCexQd_~7E6UUdp
zc$U3M+S0_-%?<-?2sQETjQaIKCPK|p&a-tbff2W?pxh{>0q9FU5J0BXh+Z%D6K3m?
z@s)~T*vfpvHK2<-DV@u=Xp7j$F|lk5WH@#=US;KgZRGWOd3;o*YFpDWX=+`Z*Wp*a
ztpiog8!6W;t%1QS^1yA%>OA7A6t_J!b^FflB-NA|8`)yFA|zy!e+erNuPYY0)n12<
zWW#5LCJ_0H{msDx08i)o6!wG6eg#+9)X}F7p=VMT5XRraZ>=h0?@@(1!j||TW($nf
zA+_%p_v;KRO<wbV-j(@`O%u+-7hcWc!R=;goj@{YdrbpNZtY(silcHv?=_SdnT?!7
zf`s)zzg?6lE`-?X%=*dNspcQ406pFoj(er|x^_PyH73Jq0OohKxKUP1F}4-qwJ+{#
zqc}YBMJFh>;BuF-cOc4XN?jrcV)qf|%Yf$LddWEBUKYo!Ab9o^E-m^YBtc1g>g?9G
zZd~z!buOZ>d+^R-plRd3J{T-JIciwtL1EYjml(V0pwD)&dy)VGZ;5?`dm?M*B>6*@
z*>Sl#pf(H0@XUf4`Y3%nGLvf_ra4OPsx!RiFq-WFdmR|0$zyD-ZmHATBsNalMMl<N
zIHMUsYSoX1hp#-_ZuFNF1W>p-qLYo*p75fK$aJLlg<Q}I)qXe=Kb4b7$^hFEFvu!Q
z_Efmfz+p{$UtLDA{FlFd#Kh0g^$_j-qp`a`Klp(8OXpYq7v~Q~XKmjM@V$bAJyl;A
zKr<-euJn-o`ps%vXvfP@u`P5!`8wXl-tqH^D%P(*13gC{)`ZtVA><jvjL7?r@w(OY
zLj#3jKgJ~w@oiBIP99+b8AmN{#!#0xOA6&)-YiYHc1yD9Y4JzO(HWTg2JJ@HPjAZw
z7|#K11p!VdiOy(>_+KryzN4KyswW2aM$Xjm_u*CrWG{k--x$f*R8ZZb;oCaW#rx`K
zd@U&r1igu@y`B|yKmD7oS>LfQ9*tXNZ5)}lSNA!84*nLMgc<IcPMVV>YIsRN9VbBt
zLAc`Jok`JOPgsAE!THKv($AZ(`5VY`-Jb+r&NZJOUFz35on)2X*l5Y*@W`JuD5Ggx
zD1Gqr%feCa!IB!tkOB2iYX;t95E<V-o=U$ekbJjCV`0u={SFe#G*)VgT+2`U>vElT
zT!D0d%w#8_Z;cZH_+oO*3w%{PQ9s;w<RYBOLud=ouvU+>tX@;HaX-t`t{72$ZT$$?
zDQ4%y)5wm_Mt>`ijf`dl!SAW|S1iN~<X#`_l?;C%?Je?^(8Y@v26wlck`hgr;4*9f
z_5Yb{n4yD#sTvwtm_zrSnE%@+kNt`72EQ7%9`Upf=XJ89hm)-?k7Z@qo5gCT(vxEA
z4UGypob_5;RDRce^uRLvv_b9d6E0)|c$j#&a=sNUjYH#-(I8CN&1LDvebhxTv2#P8
zpeGIXC86_zU3Ct`Njj62>M5gCrmg-_CqsK{mg()?SjF^7sB66mmW4e;btH{743)^(
zVg8(o9U<_LUs@;pEu<2KP*7hv$stzNQcid)xhK^2uV6hykb{xrc5+2uawc#ckRXwI
z^2j0`^JiLvwrCy@_h>%Gwz+?x8~>KWr{neOM=By8lfoi@_)H)AIlmcV4+7$z;zyKC
znn}lZprC*dHU&~7MJ)?7trJBV*nrp%#Z=p=c8il4y0AI4w%{*Qx8vg+NpzHCy@}f~
z-pZmNxwNK5&R29FW1ni{o6_-901yK<uYJOCGCPEKh^aIUR;w<eXd6`1=<?m=`{rQV
z!?JqOyff|2?)9(6twkc_uGVIo9>Ss9d6N!bL;n_s!j)A1hvFsXVa9O@mIDHThr;4Y
zt;3Zv<66&5p^>Dr96iyCspGHMP#$m70eX=ph(nTZG6m7ONL{M6QA1`5;MWKS4Kl@k
z`VsewK;f#??Cp$}n3VV^CFkeb!z-EknHY>=9`>t3)yUFbJ=`s`N*vNo?FEyF&aX-f
z6ne7?mG+CJ8g55Xbo>>&)=y}S>zo!y40bR&yOnU->Xlvm6K6uq-$_zAnw+TL;wEr#
zqi$dR(6!U?=qB0KN<RK>0L*)jq2yHj+yhZBq&a!Ss<2-{Qh}<G5Z4w<0T=vj{8;V>
z940x8Dq7CR?-Fo_4jq{N0Hrqbx`>WGfHyCr_@GW&muUZ-)xK)MUkGm|B@Q?%+^@XL
zEbA}5`yQq%uNNcS|Exd3JBk*>xueC1*{Du_X^xX+l;lSQqW2C5PU7F9-(PKwt1vKW
zM@FZKt-O5K5G&Y^@-a<qECP+Y#yKkf!PWH;v{aucH%>K5Ik0`Zb(y}wiM@NQsv5;+
zYHKK+Zc{DrG(S*B3Na!@44x)ripTpCcc*5KWsGC*n>2YAOYVQnDQz47l8TkM<r*R*
zbNwJ?VP;grQg1>7bho{+Y1-p8;>#Upp0VY`Zz@kBi}M%HY2oX$y%!wg5PToSt?X|r
z#U2{a2A^Wfa4sfJg^2o3tu)2eV4>`v|BA1NBDz~qN2+jt=5&2+L1tzSY|mD5qaNMj
zMa0??(!Q%0HZ7@8s*bDbsEC89c+?ezJ^qU-=g~19QhCJ>T-m{vdcoiW{vkOW-GpXc
zfC3Ne^CnB#koAf~jc9|eUnk2f5e$l&!cUU38-i>_ngLQ1H8LS5jE%6ZXd<)D{uZAk
z6WKtyQKG+iY*p8wX;Nd_>X^gy<2}4<`gSXTtY&MrlEBsSlamtXpZW`|_j;C7mCXbc
ze==L`+hcD4*>=)$1+=uWN6XclSS$XOm^6`IowX#MH2g5n2`X5SM9Ng!Q`lq5_}!1G
zi)bVS_irQwt3;7!M=F%g0|=EuKuu$oU9z5tjz{~G6dY!;@Z4@PmGUR_3Q8~J(pqk(
zCqCCS@vT-|xrWvLrjgB(s?aIc_~SNUS86spAe9IdL?zl(swpRM9<-}R(avw9@4J_L
zp=SUwfZl?#8~8q#$;iU)f3v0`UnLHdjBOY<W{p21M009z$vOUe;+IygU1Km98A{R`
zmZyY1FFaf|RG$)1K@YtY+u3!E2C*^1-_W)^(BpuzXKX+t$=TgYM@ft#{8x1EdL^_K
z7VCg$L=*D#*o^@~JNE?hVrgrq5ee(GbdFwUTbIQDNzBz&DYcrmay7Smh6@7(x?dQG
zpJIs6&yWqJZMVnq;*)j-WZeg%{pE+<`nbs0SXyZp>%2?~r+ET0sIgm9-zpBU8{K5D
zlj^89Pfi%8Bziv*?%)RH`H!wn&!D0>0M}4TE`rDkaoK-&WImG_-XxjE7stZ38Js2`
z+lLB~<j^v|vKMAf9!=i)N<=Crs@gK|qj8>waZ-5xBJ@=MyLxH)LDMx32}7S!kBM@*
zI*X)%OMNeuwS4W9HJ{7W&qV|Fta;=R2beP=vQez2OHWc<zxKgXRD{S>+a?pI3-F%h
zWa;v!H)fnC9h-3!ZzP#xJ4<Dp(~7-lmb~PrU-3$Myn*Wm^&2$%OykxI`?rn0Qw-@d
zw9@r4t8Sa`L`2<#e2%^dbR2?}Uz0_XQ0qPqG1RGl&|68Oo~c}}(<6$-V;smLbTMvs
zqCu~~@`^<ZPtdY@MKR3D)9NbZ0Rr3+KL;Is9b7D}f-!}(cw&dBw3@D~*?Oof%2Q6H
z=hpqhc5?lfIi?{V62EXG5O0){tWI<>N}L-VCdaIhhf|su9CR?NS<bx?rGzEJo|G{s
zce^zAa}*~}Lv>Iu(JxuUjU98iG@*n+gy9#aA~@>@55a*^ga6+Ibr3`2*+-zbOC6((
zlIzxzwZ<heF*hVG_FE72L^7WWu|#`&S7(RtgOTkoL&E4GlD7QV(G*1QZ^fJ%kUfF~
zE?eugU%l+?GGtLVF+@?u0cy)h9%|D{o;-pL8-o}BJq4g+>o}r>L{IGTUnkuWC{3n|
z>;f;-XI(;hQ0B1U{qcnf9}pFZOd(i^q~@#7@{vOQr`Y!Kx1SdZ$_6)wi}05E!4>(*
zK+gkFLJAW{&$2FLg13ubS(}t`ILP!#*cP`=?G_K#Qh97~$G2bGSZMCW?tA4lPeVH5
z^v{n``oiTG2guBanI`)!X){cS?=v2DKctcB>LAEx+UR&Q{|_YvkQS+9vBvG}G-46#
zIG=AsU4~N5W;BgcSZjrP4U=g^UZpyRmxnK8=q)oTq!|=Hq<jwepBC+RBBU~SGw$3o
zDh-#*_1Y<=>M1Jcr5WuXG~w;%{7jT7=dI#!H%l?1X!M}O)mG=r8}yS&<->9haJ$8y
zU6%_a^c{iX1a*LNPxDKx%SOyb7grOvT+zDxrmK=M_-SE2lD;cT;)Axq8_JY<Huu{E
zNAa5TZCf5|huRrkIcy3}1v~3_^OI%cCzo6B!*xqCClwy3^5rT*n<+|)rfx@8oh!gA
zr`Cueb%23Vf0Cpfnia-TV{OkE%MQOTK*@kg>h+I?c%USZ{dkMR0nMI9^<fb~p>7b>
zGN-;QxIHBC{bO}gzV-h~fvL3JHIw7Y=;d)}^YMdi*pIIHurF9$Gh`|S^>CpRMDyn2
zWvMN7r@`XQJzJ)=tYAeCs*K$T7DjPZ63%`dMHq_DuOFmVGKSk5Rcx3sp=v3&2)xTn
zya4WDCnFWrG~37g1v4N@-?a$cs{L?rV(aci_njKj-hE$Zk_VvWBC-0f;YO{>vso7n
ze`nTNfbyY}_Tr7iN9xO{L!-QBYiWy}kE=|g*}S6VmOh+f=@;fHQ62ic=?)~1gsht(
zYu|6EFQ!}HkFo2KZc4eS#(wl4`CX>yI|<YuW90voR-w7NpCuPfcobwHt2H&$b!M{A
zFpP3i-0x%l_BsDMw`W@J(h1e!#M1O(hSX=T_w-PQjc3`~?%yBOxG>At+ZG_ddO%jV
zKi(ZsWNx7s@#MBt?Kr<s9Z6z=kNhZ*%YM=cX<TR9e13Hn9fOoU^L@3}@MN{obWjth
zh?_@!KRfMrpq{56b&JAZDB8<s^0Tmo_%W_ay?6ar-d<I2FXlyTI`@p?Mj6{!iF<tH
z{)}}!pRcFyYjYNjl?JBTMCWKf--%w}(WbrH1fvWkq%F!r>@U*$NS><IV(UmGot^K-
zUmW2Bq;vz3%D%jLHDQb~#*K+TKw1<G#Na4N1s@9KLW4z6>2Wm?g$l!d3PiVSD#*(2
zSE^d3_+?7%k5E;PRDYLBmo;@2v(pka?~2xcr(|N@zC@4v!Sy;{$E}HoHiZ)-3>)-R
zJA3DPTUis-CiL-m(0#kGKX@85n!DCGqWt-flx@@IAyK{Cawza)OkR%)U*hEez<%g>
zNAo*UQ<k6fYpw#MiMD0SY9K3nGj&DHS4@($zeL*jH4f9rAT8&lIp-oRCzqH%I>K7N
zJD&+>v-325B?Q`#i4Bjo@Wq2Xw<CPIuz%W1pSyNHY0myIL}k+-w2{J(uUeJD*EB3Y
zstF(#p3yTPR-YuqZ!e4n|FhNxP@x}BV@OO{cd~i0P2UOSIM{sG8)UPU@?p?I3$x`n
zqElDT>nL?;JdB$wW)OIi-xhOH8=1TGoxAiJbZ$oDMFZt=qiw3&<7hQgZN2(e%Wc?F
zkhVKfu^vEL_%*DGP(3<<z3pnz4n^y_z%<&_CH1p~j99=`+Vk6P0o<;2AZfp$sq~)q
zaK$yir?6S~c==X=HTNn|NC1Xc`mD<_?wK}KCNH(eHWNZE?Tp`C8bn)&J!ds93$^(D
zKtfutYd9SnZf}ii9hE+h9B!)0Ypl#nO)?daOUxO{b7<kFobq>?HKHFqCMg<o1-Xgw
z^-;x{_?U)eX)g62ry?v9p!hG$(3nE3m0@ziSrxv}QMwIfm`*Vw!<1%n)jQDNT=^eq
z13`osiK_fGIX_2K%{eR@yvEue4~qr+3W-GnCrOaMFv(diX@?>!&->xX$$=xX`;JI(
zxfKrsEZel%Toe7FXy7Oc^MxQ8#mW0%IR36)cuxJ%OYLuZ`(vB(KRu!@Rl&-DwogWm
z;lhjz>uotT!H+qedgtB*_`zeo&*77`w()A6iV~QIK@i~nEWER1&C=|?E(NuXlE9^p
zb5H-h6xGHPW*QF|m9V)@0|j9dBPc-!+rH}6tN!KzPE>nt;J(GR*GoH%@Ghj0^Di49
zPo2j|<l7T@d~qRw2_Cg`lxtcj0|}enw0xEGcP>5>V2~$YEt4H=1W@Uj%`gc17l@+?
zeO9L9H2|kuv3ztjExfxB8=>>9!|>2sR?|e1^~s?;edd|p=>et+46+_Te!PTC6fSys
zCE8gE2DlHHyduiD+&w_Edn`|rm^D{P9YS3JF8{ilgVCz@c*wJ#t*{=)FyAkeV*XvQ
zqCA}*9P3mqkPpLbZ(X8O5#PPMiUv=)Fx9`_O^UI89NS_JK@K4*%)%4fCc`h6ViIf4
zkWQ1@mRYq)MfToekB36=5yZa<Iv_b%&7V9V$eRME&CC@tv4na4B>PHC{mx}#7+nf>
z4*?yjSsgx%ZxUbg6fD7A7^0993Op@294gHgfqx%6MYwH0_EL$VeUyx4uN~NB2QB6L
zxhi#oa|!O<mbpSCl7=qzpd6)h4>uRHdNROqMB_#BAnu0x!<=M}S#IYi&Fu)WO)eeE
zYkdIUfTAmj?%<4#YGU0C0^bxytHOc!z8e~l;6Q-za7=PB6YL0u|IYdj(Ug$ikDo%?
z$G8YNL)eEVpAN2ndw>V^8<7iKsDy+Pb{_istj^cuz$E8}UAc6?{o!x#?ii`O#38hs
zRj4d>%}_+kxeGs}=B0c~3V9AwJ`Vcy3FyLbAC;F_cE9N5^JyS^GV`A~6|j}2ExIn=
zZxfZv|793nH>&AujweaRuPbM2X2T)lZ-nez&)jG1n~0Q~h;A*P8$hS=OjsWEHBR&`
zZtiQ9L}}Vjw{wVPJ318BbGfhmD8Ojo22I`5S0Q)q(zMLa4G7m0v!-tca&!N9!M7BS
zp6qkeKj(En@Bcp3K`szbNr?rM6xK{CfusRcwJm%W(1!1?R}NaKyfvU1Y+OXsJ$eF~
z&JZ#Bdn{ddGz7N$L=22cr)OHpn=#y@X**|;#N80lLv?Op6XeoYG2I?9do)qm>sd{(
z)VU^=PE*5T>swSZCMigJ&`$%<DFB!rR1q(kN41ws*{Pts=pyqXy4cqO>*w7ar>P6x
zEd09cA6?dJz1ICY)^$MT-o8eWIgBz@Dhk-uVG7K5D?b;!@j~o#z?hCN%~a+7eiS*D
zE>R~4c;y$*%%@oQk?h&%{ir)RGEIM}sHz*hAKOL3(dTMn18x_j{7X9sfQ{bA?q1+u
zNO~X1z-PJc(%TtC*y(KR)<!S$1GbWcDN<peTk{k7)9aj6v%4PArxB#qJ}9w4hgo1^
zxADx-NVo94#e@>F+jp~K<q)+N0r`B9EeBlj9Mp>(d5Q9{$K{ZiawzkL4!cZZHB|q8
z{wXZ870UIts`Fis|3X|Cuy`y)Y)y3a5{CUSPLu7j5k5TWG!V=HPWWM1Av<P-?54xU
zv7tWq{=0eKcPxVWP^pxO5t?=<ymAR6gRS<DUe2>CA(bO~OmcntYQInsb+W^{B}@cn
zjnHkpX{oOFD+zH8ucV?zW1Qi;PoprmgKOHCQ2_s|*M{LGJE$8q0M}{8{Gw}8?37UY
zcI$Tx@fPrcgK1+9?`n^zw712I-_rBTQfqhaO2+l|L?4?zL~XIGO?RV!owt3tdg$N7
za;Oxg{UEY4xU(60{9*HQgJ`Zyo=JBV;`7@-k;X>{tTkSHD6}~)@0f*NefYI`H}wiv
zv%zVDhTV;pvuPy)NWVG3eE&#I=7tY;1t0Tl>aOz#Dk6U?mKR)9@;5&Ha$VdqnoH=v
z%>Sae?YqY}>JS}6x#-SJw0BH!rh*8@u2k1(=mLc-2O8>ue*fl-$niXXrTN!-3KQfT
z$W<|Peh?v@ub}W=EbtJ~PXLU4zpOz9iX|>^(Gv7g0D~j}P)Pey1vemFEcN9!JhGOV
zL)vc5o(Q~GCIE@bo-fE#My|6`wB>4dGVHv3VTYtgOa1P*_bAFky(c`rM84_W$7W5@
zpaIp3(o44=>@|dUa8uXjPTaOXhYTV3U#eB|zO9!+-clPe&Apr9%0ZjB<>T6JcDa}K
zJ{ibGgU<_^T)Xs@55m*pa}v@ae%M9vtvBNJQD`Rg!EDz0I>tdz-D4D_)#f%Dx(vQ!
zbNyvTHXY=$k#3Bzjn+JdeQcxoA(H*tiRx3l%l2S!JLoPk5b<;Iim3IVb6#zj3(%eQ
zv#3w`^B~e$!jxLD3%7)1pS*Qg2R))tGGqZ412F^mTXpd&w<lZcs~>zy+ev#TNkQ4S
zfSPaUbM#kJw)vCo>kEX25)%tLk;7TwCGQ!C6o<6t&yUA^AOx6f@m&yBVxHNVxUE_r
zELylsDia!X!n4zzEBa&>NHBCJ=m>%}dS;|Pt6}}U_&Oc4>8atbn=<usap9#Dzsid5
z=Y9<Mm{jEYosEE{+wBO$=RKhkrbPWl6yh@Wr~dcFdcQ5uVgpP$bdV2D{n44HxjH)a
zi_6X~-hUXgzVN-I_0awjk%#d0I(PT5jZp48RB+!=f&d@rjZt$kiV>4gpv2}}--wGt
z%QGQvU5_~t)`!x75y{GPRj<5KlrG?*Iv#+D;f0`^PZE{lL-@@?m=Q$`!3m>y!kvFh
z55sSA$i*2X{^SUzJ93@-=CMGUwdB4xd;6UXtPp~^W97n~vG8C*G3~Ws?k@q~)R9?-
z{iqZ84yF|_y6n_i>d$ZQ&})XyO{FLK?j}$ZkBX4N11oyq<lL2jz*TKzk1@HXc;NRQ
z({Sf6HESQXpJ}TdQLC|!C|x8fzU{wR>H_6T(=yw;Xp;g_^^5O*zDh)>YS#TZts3#6
zj~amth)bRSgTv``Z#vV`7+CrivM>d6TMig+8rs^@1CyNO@$xQ?*ziL?688VYkw@*w
zJL}!o{&h~p*)b*qIL&at?*W|*Y~Vx~4;CHk2sLAL3mnvd(v`omzJCL!b!#v^w2BvJ
zfv%vs=pjUkCfho}Ps$RZcs0f5$@(0~FB%TM0YL3g<Q0=*uN|@+YIcACQy;=z$d&Hl
z){{cwf}OnF(80Iltu0IN+Tcv+tz+xZxj-;@@ZY-405j|i?JV#wc#UlgfT+<lA5EVs
zY?y~NzEec=FT>IDB`}oeWvJVCir-?hd3Xs{ryZC?$FuMJR2Wr={5XU0nFlYxnZQi}
zoPRiN^d%jEH$d=xxU-$_-l>ON`-wHKGIy7ZOaf-7Cuy1#st*dC(@vhoXr6uz{A=e~
zWafWDCZiPxkqqqVqIM>M!}q+a4K-LBOC0A+U4BM)0n}Q^pFaA}dR;p3TN^);JGtHd
zd8yzsXs*Kvh%1R|Ko>ilr)!!f_dtkslvJz{TB@vFmmv><Otzw)5mkecG~B4R8@%o#
zDQd4|VRj5=C^JQW$ex$eP1}@_KgG($!Awr3pD|qg1y603S0$`~S$`Ff<Q~g|eZO?O
zO!5%YbYX5>ynL+G(?nNxm>C^j+$po;B-fMV2lzgsq3`FKx=a#_qPz=#C{unMiGG)7
zS`_V=B0t`_$cUmfr}=2r!@@K>;jj?%Xs&m4e;dSyw@1Av;?pgAQv8YmaZOG<i1kt6
z&hS*@A^1YQ|3)YVFg~q(P%>Y;J}YWgzo?gMnsm69sLy+$mr2#i>$_QmI7*`FOWJii
z3F50s5E!0fN5+A`Q+?lOMZyBb-9o6u&VxI+&nIt2k<W;Pq&J-@BhDF%_-p4C_f2Qv
zDl0kQTYRTct0`k~{D(b5{g|=n+YAjb+R5=WT*VeSwfqAbwrCG{wqLqMkF~zCo+D{m
zUZS1owDN+SsHht5A4n4hi0}4Z$YMLQ+9_{Q`TE3#Y|%!Y$7tk22Lou6d3VvyDtSnJ
zhOBz6e-8Qko8Tsr9Ia<aoQ~Tf%F`zvr{KxqWYf7(1Zox!T|QxgwWN!Yt|Qd8*%ua)
zZw&eRy={T91Z1~B0dLR{-K0XBb>s(9&`7{!x#fAWP%T7u4<X1tD##A+5j9YnNqa(c
zmDk-G;yE!*N6kB^*ySO^=pO$WKgebBE`*T*kt|M_Wc3QZmgC9daWjp&<zLW#*zFz{
zhrvRZM~}wi4%X5l4P5Fbps*meZKp||A^^V9guuTvp~V16VU!@MN0pm3DsWh*^qp^G
zIxVQ=nP^1nX(Z&a7yAD67P$Jyb}A<U6`PkuL(26G8*T*8djE@gq*c9l&{AhU_ia?e
zOZR6%tUtSllGGLgPgJK1sPap4wDT-~{TJm+VT!N<F9c0H_6?(gdVKz2IVr0T*1a94
zKo&{$1anUS8Q#<yiM_dz#QEz-(LnsQOwD^*k-Kf=)_7*kQ~SKfZo{i>TsJ>(CyXB5
zpi*<)G^9W>;=Bi(u<tj5#xqprIMf;=#Sg1MvBZ(y9RIU_uB+8Z)an`unKFV@kW0Om
zE6IKTq>Gz~z<f6wbXj9>8ix*7=Zb;=iP+)u#gi?Vt0SVbz-wn8ZKBeS5K<yuG%ZA$
zPXj*RG(&3KrEO4U8ZgNfehL_WW-7-fu|Ex3pQGEjWf~!B$M8FUh;16pvQHcOd+~c8
zJS9H0f%r)koS(~f9;UQqM0^|UGTvEc5I)~Sae*psj{y#L2{Y^l<32YoEIa^cAxiUX
zG-Df*E@(RjiMVx(E@quLl5=+fusWqEzHKU`37;i$5T?%{!stVaKe~e*4qW2}xU{0*
z8NvcKQ3L*=orQT*Z@kdE1#0TQXoqH+<F(XxlCokV^GnDJ&J_gg-*?{7YAxLDzYX3<
zrVtC`(%dLwI`77L<`qBWasi<IbZ&I?PbqR6H4Nn~cOz(L>rg40Se$jl`AvNb&UmG;
zy={G(@{hPgukTSuZ#ZKq-}-x!ONl)@K=Zmd>e{^&UQo*g!#|l8Oln1J3yJD70v8t|
zSKuimMz=p+rgRIq$PK-6oR5-hVL{NLS1UF<7y`CzlAAxOwJnaTWq?LAkw0|zjU#A8
zLaK^TK(a~$5_wn}#mb5|g!KKj$X=j@UEe37>dfEdmtxh|VXW`)49?he6@!gi?De4y
zULwCgeqUV6#nQSb-ggDp@X%GY$l#Nqe-w@J*ia-x)<_coE_Zjgevin#nQ7S#VZYAg
zMreoWZsxA=c#$eai2yL#=ke&^_)Y>n&J(%Dx1y5f*fCNmJxt<+i8xE0CQ>&B$o#|i
z-^s*MIX1|0r6X8u!MZt}E0N_Vam&8oO+A=I2W#07E-y<PkU+L9QnLs5S?WaJNMDKL
zZ&S$UQ@F(AB<6`rJObvK3Fz`uH~C-2bBC)kVss&6$4ec#K-f^Yk^3YrtM0Y2sZ~uL
z=_yw6L#r5=JLSVR62)~fQq%eH@Xd3q7Ns(B<*{ZE9j6zsN5-`d{;GQ!;pwFut+Ve)
z@P{TH^Hdqo-p{wi@eGKsSHTSm^6k$?l1{-;-{r~4OrnsbUtlgDd~mI%W*7_albj+d
zNK1_H&eyXYfJg0YoJ``UF9(gQaw0X&JWSO&XTI_yLAHf~BAAQYc7l-a*~0B3<wKZ)
zM;SEPmf)!ex82BxON9%1ycW2U%pbjEN=Icu<WmTfn%J|Q$^=!I*KDl*F>T2>*m8N!
zPu<+b6#9w2z3l9p-?g23+!!ZU4m*9H{s5Nw+-U>=pp}*@B<N-nV#;o_z+O?ilN9@X
z6N>QJ3eJ%Z-s!DN+f<<Y*=lRYEUbsTd&)id_X@L|wR73?CR-M1uF=LOnpw_EtrzNE
zWUR*!U96PzO&p0|GH0#}SW5>i!)9%}*ZY%KYY+O)?jQ%(8t=r|HEp%EN2QH-X743r
zc?%$=+vOVo=P%Q0Ke-cRbSi$>{&(Dr$4jadkn_o4Tf&yvEuhRvx5-ZvObB`OBdC!0
z#fV7c0ER-#Kp_=_;FSd(2qn;qde*rMAyGsoOq7=}W1;BRZ$&vfJ52m^<Cy@CZ}yib
zPCBdy)0Zmt_&G?r$SazK#ea;E3KxVxfv=<mmq>^CCa|3^gg=?(0xoa!x<&Rt(}W?c
z{fP=!!bP+ahir(IednREP;O!bO&Mz+7(FCj1NE%AerW43Mc$N#guw7wg31lY;>wo8
zu}jtMOD_s<z<4%#Icmf?S$k_Xoh5O^uItIRc83t-Okf=EFQH!d<glZfujNLU0BZpK
zARz-SH?A-Qs!xpYI~h+MIL*MB6)arLUGG&|rzYBWec47h1L36$zC90b1%vzi{iFld
zg}=!K&S`C5`F_E(mwH)H33Z0jc2>$4_dejA-3Ng4Bi`_~4>0lu#y6msAI2vo>vs|&
zbiK0<fouP3jp4zEVCjHVk8j#Q;G9YEgGNvK=rNm(;>Qo&vlYYM%cux!HFHNbd$baq
zOo&82V97;EaOb-G_8Hz6T32w&>$B!}A5cdN8ur=h<%1*~80!LbRkrc>;ZKDy^eNLJ
zmI3?Rfx5-_&nStzRjl&eqJ+MGN;v|1zEX3@*2*Yc2<yXxON>}&jQ2M{*)AaCYwlHc
zXR^XJQdaxWWT=@)e0%{i!l81%BC}!&-sXm}I?8(&xZLRqX%fsn6L>OznL|$LlI}jJ
z&Tt;jY@2M1p7N_A+wF)4U0;c+qP~WoBWDpzNrONbGV$=v;Ua|?Jgc7+{TOhSF|<|1
zs^~P}a3AeqHcf~4*$qhCu?mw8cnTT1O3b5EoLumO=)N6xfOR@^qpy}V;?`1rdfAky
zIlueEF;qu?e#<#ayGiyNA#sy-Q+e)A#=2la{=p;kKG?Y8Zy9zBCG-}r|KGU6=&*h{
zgde&}(!YQYQ8TlrJHK^K!Fx^i%+U}Ku=Dt_L|U#i1?{N${Tm<-KDXFTi=bcj!+2+!
zVzG31#qX08WI2hvSxgT0yg<Aq(187wkLT0$htT@wM#N`h2Q41jnaz2^vmZu@Cv2z+
zZ$5XgICHID)dV)ElD~=nIuCR{OKl^;ioRmJV_AFc`U3H3Buzw}hLC0)Pa+-5!>y)&
zhfng<_u>~}SxvzBp{Q4=PtEqU&iRkMkBQg+T^rUo^q{+q+1YbnzC`SKok^+y53+UB
z3}*DzE4rV1-W^{&TkQSUC8EUq>`je)!WXF@1v2_iiAQh?*?u<ll<uXR9Ggl<xP<Q!
z1@8WP_##IQg@=M^S7CP7*=sNO=N;4Tv7NTG(8-Z_K;q}*qTE3IBw4iM1`4}d$N3^8
zlFDbkel{kL<|q&S@qvi<?`Nb833ul6i{HKA_{mhoE(B{fg%3fjA3+;C^fJ@K=%shf
zAjZt+oVb9VT?)pYuXrI0x8mojn{6oyllfA4??9)wR)n=?Xkb%7cN<Cd<F_rl_F;?4
z&LG8N;BFa=r?pB{@VJU@|CW{j+*2g(x)6{0VOW+V3@Qn--wgTM+5LS+JDfrm`9wHS
z_=Db0vOTDD$Ies-a_((9QME<}r+G!3a_3UF!DTvEjYE#oY0z${yNALuiBiOEj~e!0
zLOxHZIj;=s5_CF4T3WTZH;^?M7tC94ZRXAjXa{;Lf?7yy$vQU-w?p`>Q+L?58l+fP
z_jH^E5>WZr<(aIx>>hfk#8%MO2OSBGNDyFwkut&wI5)jV4c}!K!iKJ1?=LgtSCg)w
zO-u_*Ny=IG9E&mQ)}n^gdrx4!@9sB|{D<)1G=cefC$Ad2>t7JYt{9}<^*Y9<CKW&g
z@`itVFUJ_0JyDX^c78AjcVxhOI}l`Z|4f5dNqVbS&P*~4s=E7)+L~>a^WOo@s$|z|
za#QdW!0Hy*_;lNdT($#!r*OXz3H#5u-gs_AdAKC0XBFQkJIXPOiWc7@c?d^eq?25m
z2zq(6JxkdI6Md=+x;P_8f@{8GpmozolBJLAym+^5Pd>=hOl>=&y&eLiqR@2K!nm;B
z+a_!^RrL_F>CW{ZL7L`Tno7vR52-(hmP1WIY^2tFx>HfZ8~Gu&%n&pn>DedFjCkEM
zn*?Pi3X#}8yZp~k&!?9QXBEkva(acxha)Eev5uDVy4#PPmmc(9o<srw^<R?kjEf|?
zO}65_j&|4MN5TEhORR3|kKv(IsJ95seaQtv+fBxHQ5Q=H6&GUStPm08b4Ma6!a0)a
z)d2ap5J~x0p9@?tqf5}A_8`8G(WoKW{aIYAye<rPA;ud|!t0sc?}Hp7`JQW1){ULC
z#!u(b&Kq2wXtOSxRJwpNy>eV%qnp_QIh<QVmocZS;^eNU8Ds54jNftK&Qy3h^fZ#E
zesrMmfq`A)I;YnZkt@sBgXsR<<D?I<&(JZSDSU}WU*v?PCm_w5T;K7XQBiy%-^l23
z1Ho<B9F+D0EsPK%NsE!m+ZoKV#WaF`%1&cCiTfQIpKoSCv_Q<o4z47U``gg&hdY%B
zzZa+)1P|DWxNp`(V-G;8&b<AE#&%qV>*Hi*Vono~@@c8lZ=Zr+tCl3hejQA}>fh-_
zM}6*DSJ)Nwr}XJp_|U#kxyVx%%HNT^=kyK64aZv5&Ls;pfPhIZS_~xov>Tyz7BhV=
zRjz0*e@l$M1fa2ryIz04^riZch%G?MW-`Q8y&7QGjJ~HGulJ^|L*+UX1FUP(rh?|;
zJut;VD_O4iK}6vX$G5G2V2JUM(D%$uPQym2l2&x<x9=k`sKE3+Tr~vMdwOQ4Ir673
zz8$uqLufxT_0sgkD62f7NpIYGs7*TjID`^Wy`K*9B>>vGZepW%JgD%_diQbUH7ZqR
z{6JLQA$jWB!Ubf;IJ|qhXQ)%)7&I&=NK2<~HuAnXhjTSkl;o|6moum@Eeat;MO1U~
ztEgEmu9vc_)e&$0fhUO;KY3ekgJ&Yi-=aF4J)?ykJA;^RHRePlJ&QKe^2-$tY28oF
zh(y2I-T}Y+oe98KS-W{L?2u$w6J7idQ9Y#NxGL7&SNJ_$3ct&ZG!UK|7bbi0mp}z;
zRJLL9;!-6l$aAR*<+<JNSs1(~{(|>_ZPbO|C`$MSAM+(rOm_|=qGuSmfn+!I!EW|Q
zAQ98|GMU^XH=DoxS`Adop_!R5AhC}Y=m(zq2l&k(P2Lxi{Iz86u@*^05q?e<tgcMP
zvHq!{p@jm~49@dZ3B`%gdF|s@<^frX?q0?Vf$Xy2T0dxnurPwZh2SIP?N1e_erCO6
z1iSX;exPHfgO|6uQ#=G2mqFHDJuBZys)J`)vlq&G8l>^kr|{}&?S-0fe;R78fDAzh
zaQ_AQKtxj$gex2<L%DxRQ$zJ-E-xtmVtWWTvZDqG<Nfb930xS6WsBW`BIRz|Ur?vp
z`J|eakS|RX_WpJ+XK=U4hO>TgVVzq0T=e4$`W(%w$Di@&ZZa+|()l>_#cd&>+70KE
z&BnF%KMY-Wy`mi@(o`cHRDU^}vC0`+41f^tB5BMa%s^N(JEUOMjZT9TF{KNf;Ls+*
z@rR?I+nko_AVr?`fKCQh=C#Mwn!;PvY_sRu>Li7%g<Oc&+@Uobuo6nScuXXAwqiE)
zvO4wt(mI&64Beh=_3S?}jOji98$W_s<g?!Rp#z4k%CNo^BpBZr1e`~AtBdLY)SWW&
z*1qbo87gO*9t!q&kCG^t7H|q^h4J^wS(a=_e-#1;aFcmT@YJ^u%Wme+PJ`yCnz%1;
z8FuT(fLDGK-($C4nqDE4w~<?bYoYG~f{~5CtUJ^8<<)}onyYNHW3G$Se-Qtn=sy*p
zU9wx0@2(=Q=uGp7r2n&_sR3~Jx$R~LMsh>qKSvWn#hA1MBRYppg+JqIabY3)i~gJR
z&WY_7tF1`cq9+9tOyn46b>$S%E<EspUGpYZY0H=qFO21X!j$pIo^gmXI_XWfuPVu<
z12kl|XYBmWt{aDf#nT>&?WCxlr5~xp+_BD)W&U|mCu|V6N;nAAp8#0k-$v`;e0=<V
z#VL7e8k3BI0>ncnPBRKT<ay}=_(<LZ0Umkk4VE{Eta{+V1@35LD3_orXQ!ZXvA@5E
zf@)_MkBUU6nZovrH`~%b3i+5D9=W&$gzOR=C<zB|;I&&MdGv;1Nsze?r2FeT>*_wC
zk5umi?DMbTg@uRB0r38mwFAtZVPySaDxY}xbZn?_GYwAM7l+FrIk<)Vm)&ji3s?P&
zNP0@{O8A$|p;J9w_~{zkXGF8`;v*Ebpo~wN!+k#C9Rly4OOlWdEm>mhp-Dv}`6gI?
zx_}V$iWPCmAaQVzQm*O7rqq$#On?)yqgWbFW8vCeJrOvr2ZQrr_+R;5^XBiTU=`eU
zp7Vm{c6v*Mr~c$0%jG=Lvz7F`&;uebQRO3{Dfl5c42fy;o@geBh6`WUuU|#*B@W)~
z%L{QVy68#p9wi9}Mg)Oq#EE^o-+#Zj@<iapFZiHp-F@J+f2KzZ&YKKDLBMSt#yTsj
z%hrPS@dHLoMrz3I@u0O(mmCYPwxP2|<oe<2kQc>Yv1VvB{HN>kr%+Eb8ZKV*<A!q?
zkCGImp!ZezW{kYuMUGcKdJRi5gujxX{W_!*cMfN4N#C3{$@b-<*YwxAgh<(FzKOo)
z^uw!b@#E)3yt^}e4d%+W6qhD}`^F#NX<El$8v`Ag#{4|@)Ny1ZL~=o<$6)k+Ex(}d
z6QVOcfB5fv6GPcD?e81##bEoYRWK3r$5+lG#C6RfXS}n-dorv1y_dVkYDqk;zSnT<
z@m+QucE8SRKmo76SI|DktP}U+`iXWHx_WUq-lBH&+%X|er>sZ8%s_V`gEKb5Q$jv}
z00t5S(asxe;?J-_yFpu;VSUM1xc#DZ0}ld{wFLgUZxDqWNNiCi%O|l#*@{P>>=eGt
zqVl$zW)7V<;z`r)QsMUorG0rkgjaxf7s9EuwUsg*NS#K?a8JMZ-nd-pLQnCfe`A7f
z6r-yQ>k(@}SJ5}XrfCTC-^v?I%Y+X=d*@$1bRU`m6q!6>S3YbL+h-3l&mLuo1C&t<
z{-I5{sr(KsnWsn+(kYv`V}}cI9FqF#wePegOm3-ritQJY{t2C==k-2;B|=eGr=aHc
zMaqTrrY?2hT~NKS0_uvSwb!FMMIFD=W-cZj+2YqCX)zu-3&`W%IQCuyrZ5_?!Nx1w
zgwUDwsNmy%f%M>~wik#;7wb@pAhD^oO*eUY&-p9iYjBqRF0zZo?q9ao@cMd(2{Drx
zyl`mVe{rd20md%;S^sE`Hn~jup6?L1UC^W_VLxVcsjHRt)Z@eXS_C*>OH{TuUE)Go
zj%gX#Y`a2k>ca9JN*f9fe_#f%-rfU2ppC@+P0zltZsAG}G@)5aZh3|6VEB3n#6~=K
ziB2m;Yk!XZJ5{c2dtm?K1L{Ru6)X}GEp7Tw*E)FPmJP#e3(8+h3+()>T9u=2sh4Z_
z`-X1ACbvoWFgzP(r&R_GCAp(DqT?bx`DIp=iGp`8UL!MhvOs<`YX*d69DwsL#cgn9
zskfM;5IT*@9<N=jYpa5gJ}t>nLuej<VYaSqv~)@YJu`juiMHNI2G<49yaZ^6nz8;p
zypwxfeaPHXt(G_mZxBdeWcVv!@Gse7xkd>^^KlsaB#Bw}s(uIwc9|%IGD!=7;Q!y>
z58H<iz^)FtK>|F~0WX?P33m8rI?Q3g16Ol6UcYrLWOP9r{w(J`Yj6Ixjlijv7uEo^
z876t)0oi-YL)Y1u;Esz16g$}L8~2+Ar0J(Jq0{BuPF}IF&H~HLXx}in83&&$QvJv$
zQ9GX}om+(-6nQQ=Id+M)9o*B($JURUr+inPRw@h$zTU+Gu!&xp%&0n^TRSj*HvG%N
z%#L;`6~EcV@jlIBw&vPsPo#vt;dG;mr$glV)@=QDcte|Z^dj|78>T$9U!hG%&Eq^`
z3b9KV90VST4#{Uin{W{m^O`-&)qGZP&%e5~xiX=K2rXdt&#bTSImIIrFHkH8%F!ya
zAPRSQA0bWx=D~=~Lm|~r^!}6$X`TXo9I9$6VWwxXTkX)s3&!*2F1fQ?E(Joweb<wK
zeYD)$H0slOiLtCe3|rM0b|^o8*P9d+ZQ9C*o!4bAQADg}CkgVm=fO%_X%>iyLzrHV
zY0|!i)*%B|JIBA#<ZxE6%#=+Z47bt9;-Kt)B1ENtFhtD{)3ay)TyF#vR3p=V;Lz}|
z(<=~H5j5liJpHpvKnr0Fqdjdx`z;($6sG?Cbr23kw2rRf5?61-(G_VVIYj5E$gta;
zaW9yt!Zq*v9Bo#^|Dou*<Dvflc+TNuo|$Bwm6e7)?vCydqU=q5uaH&Q^W#Vk6{n(P
zMkIS@6V*u}Bb&1~XP^E1{QkOs?s1>b<9+Y<=k<I&$LrI}7V`eBgO$3B$7C62JIDaa
zI1_#sjfqr9Lr%Z_qLn_eNdNz%Zb&Z<0G$wv9W;1{ueImDpI8P8AC=>>sJ@No7oX74
zY&;%-?yAStg=3}7c1uP?kag1LBE@wgN?&d0)Wa0tc}Zwn6Qim-m)VLEj83aPRVfLB
zq&o%mI#J9&FLECLP^8qqyj_npqt}RebB_Ja0DJ3#gmEVjnZ|>O@E>7k8rnhv$Nrr7
z6Hy4!6i&+CI1X<vy)h8IYG`Z*MLeu$4j8QuU;j1sws6rtZcr^Hs~D2unl_=$d=awM
z^8PZ4t!^-)=d2=g0(kwA2s1DA7_$!MTC5wYi<~;+%k-4p(&F)qwuqIiKx>3`pSRda
zM}3W~8p*U-$zfIgajUa{{?BUwxC8gyib!qm;Qw-Hj3k$V0`q7G@9JMu`mx<Sjf76s
zxyaQmDJC^jY-ZeDsG6P8pF53#(k^#n3@N}XJHBQP;}hZOkc$T|vs|Z)%le)VTEif2
zJC)I4uG!3>vqkPU4j&(OUt28!;(x;U))nC}r8-{{)OZnI{>aEol@q{AuiQ|kE+U*{
zG6JV68>9knUnepnf81tEJyrG?8dd&_Y*fAZz2y&!=fVfA|Ip7gvOhQyswbG2$q4?v
z?t!Ige~KJk`B}|z%YU_Vbj`ENITXoBxy;v!UUKBJvBU71V3}RM@qaBGy;q_xSHb%0
z9IgL`eV^PzCwF_uw|@mN!UX~9i#mn}N95b%J$rNzP9*lqZ0`E+JE~8!f6|e5{&qO(
zf4AHtm^s$VUvJQU4*MO#71)pmQcFUFkclq`pX@@KFgbzzw{QXV`nW?R579BVJG`kG
zMy&*V(r!Ja6=8=qw>Mp)VL1a5nKEQVLpn<p?q@DF>gxRsZUe9!C%wv*{4iWsoM8s9
zaCt~b@Zy6bO=-yCuINOC-1P<{k?V~NCvYJ-UR7j<-R!o~A&g_QR82f^31Jc{*hJbo
z4Is`03&fz6mSKuVq%^&KK{OGoUUnMdcL<X^{5l`dmof#dnGdy_0iN^1Gw{l%&g}3_
zo1FmcgHkoXqc6-2HThpqLl?AEt&w`<qtVTzVS5%>k83~YJ%{}muZNoUsL*W?<GuBG
zIr~-Vr1~a{15J6zmCvMw|AIu9Gmc~Rr?%931JYESSP(vqyax?f)%#zXWKu{w`_-iG
zBM9com^T9H543Acxi>cnQc#I{rT@g+R9`B%YU?DMKBwmb?1Yqq+bzGp7;3p6=o*PU
zvzcX3eool<bM=0-paWvSQaDCIEl*yM33{xK7~D6$B!%noPwdJ+6CAgixDU4$CSpR?
zn_z||90wuA>w0AY{PsDA#kTP+Meh;6yF#HxkM+emOr<9z#Y{qH?={fx?V$R8+F^xx
z4xeuyTprT{RH>Kvh^roaMEWh^xv54dg46H5bmsyz-b{q)u2}HO9q4rf?DGvV3L)dE
z^z}xzIe3PbR21ScwNM!b7mI?qP-7l`Z<vZ{V)`!H^r2f%!Pu+8L72khV16xOIn+3w
z+Q0%Qz%(HH+Cjt6B%IV-nB;s<s{`rYq4oQ-b`;>t%0H1$==L~~Jau0o9jLA%gc>po
zPjsVwE((Fr<3b#1b2Ca=y6@<#PFT_N@Nj!nc@t@antA|~@2e7+H%0^6buEN8@2U{j
zLbVYYB_9Fk&R@H_ZcCowOqSU2%HMQqnXl|bt+HxagYta))Zfz42Ulv%f7II#^qcYJ
z0YtQ`D<%{ZyZ5P=%2`6&b#RBWX}T>bze0oA)-H>LvsgiVY`EVyNxp#L0zjVdhSBUo
z=t_3`+>G~GdovwesDt+a?`P}tsm(NRx!@P3E+a`|YkRZsZYB4WT*DyCY^15<mbkWW
zad!W*=y$}Q{VsOvrl$t8w6&%-+c)U2Kp66L+t^9SXK_c)WQtxk^nese$`$CX*=RX|
zS-tsmhTgs+pZo9@%Yd_W9C|JZw&Th1J~MC-o!7MNw<lvGx?==D8;TSyN?nDrj0?zY
zL<+-7Dadmk#fLNJOC4QKC9(O1-0so!?zd|(zU#gFBzN=Tt-d@D#|7!9A)XV!XS&1c
zSgumuM|@6Td6z6m!g>{7>UrySa4=DY7<9gH#yO<UU6&i1;3)-B4)UOBI5PKROwI7q
zYU8C5(qHR|wtyZo#>h^tc~uS`^XctY#Ucc78!PQogLP3E8!C#pMXmlX*jT=*uTvhV
zX@B#$WTYIO3X`0827DKmng}F4#`c}-kc2qbe_ET#LonQTlC!65C@kA5*f3^FR-3fm
z9L*`}J7addCjGR<SfoM__Yp?=D>22ik@v92%+VhPu4%KN$v-WNDhISlC97gOVqCZL
zCsSEV$MV%^``H%xHF*g`=<%s0juPgZ@=i|MVO%hb%(gV8y;H8c2-Kq-vOSs$Fd?k3
zN#`bm;uH8qwExDYXMU(AF$9Ge4>}9HyfVZ5PmV2vNgCqqt|QOA{=-4dMX~Wx&ty=P
zsy%|Xa@Rq*Jt6peRnpgIlYP>SQPn46w8R<jRNcY0*0Y^?nPVZEv}19b7Sny&q`Mou
z?mN6c_Rel-N4;BtfhO477fbqRLyVv0LlJzq5*YOkriDRuI-LJ8=H0n#Iw{D4E%>CU
zCGJ4&Dtu8!QHPc6D^z=A_*+-_*Nv?m)$p13(-hQ$h{{V3`j!_{4_}Kz_Qo!wC_S;{
ziqM4N8^7q-4sWpCV|ny!m6;En@b}s>+i{4Ah}J9pA{yhyeiq<sh0k!LlFCKH2mEyx
zLX^JfWo`G}OBsJB67f*^X4G#YA1lu0_v6r{-FMf*&9933eCEDI4o<l?lgHgNgAL*3
zC9(>l;l~PFJzZaq+n?!%*zdk$3`&V(RvEo<yY#Q~VXKEyAH&^e??s0$+~6f}|JESW
zr_xEJ#NBxon#4A{3l#pbBofFM%|7RG9|(7PD2*|A;7=t?8gt$m-=;NL^w<}2p)vT8
zbymskB>0XLo|?g}_oL)-OGW9oC@Van<fVWNJz=&P$>gwCtkUWw1sOSvWw<{r<KWm#
zfnjni&$HjM&ObC9)<v(_F{;VO$x>uWDmRCb8K!5j)fI75fZk6}Vz~Zk<Y?gBx=g(g
zjo05_2{W=!Ll&%v-3zbd>`1cqbPxX^A*$8|W@_H@FVIkphXv?A%T|iiu~2yLiKPHL
z(IGLURLwV1$ZvqD+`o#kX7~}R=O#wj|7jJ|UiWgVrC*#osnsg2+`&Km*SfCoPj`mc
zyXzOGlHfEYz|kufh2lZJ){Vbu7Z*zr!UcZI2_TqQs26@I7fh3`CQrE<gI@pDg|b<J
zXo_JMWeL=gvNE^u7fbdP8;6&l9#>u@vh4)GgGSfT4lSN20>g%K@<o_W=cRgv5fKE%
zT~o`bW1YJL`p2k!@b(F(3Mv%;z4?e{sr-lmnZ*fY1{cG2XhNVnX`B+&RU_t17C~#F
zup(o-rtUEn#@zWB^v`i!;qIxNWspderP#)X56+0a;}d(@YWyVUsBM21y2$~j=%g~g
z7Qia4uDnK3bo5GSYegYJ(^3qc3#Le$6>Y&kkq0q(7LzABIa*1hu6zwed5R=!m2v~J
z2T-%7ZX09w8Lp!F_x%dp(^O+2c1M+yh<NJ$MK;>^nUbg9_B73&7wVAc50_Z?z<1db
zk0%Xw=&y@xEa`F&BHH#JvS(>zIjT{N0M&4T<4!Z*%}zRmee-do_hF|{k!|kpRY!1=
z`^j5IWU-@lrP>^WGP#7uCXy#}WM%O37C>>k%88zhER1oCCr6*SGLn<`piNzqD<#J5
z{>AJ|nQe~-gb|b#6oEN3i;BE}Kv6!q0xceU9Xk6*<m@XV2nL@{`O}u`htqV%lw=MZ
z6X4o99M)d4n5?%R5`*~eSa+v(cMdl3QQjH#M=Ph*WADM@eW|A{^;?9=O~)UAfC*wK
z#qmHZ_q!!iz`5p;5MBc!>&OpFhOrdH;~SxM{P)I3A}(x6PO-_1gbCY&yJ~KQMqyx(
z4}Tghoz~RUYI?SmyX!U8B(jN?#$<YiO9#PY!Iy#URGrQIfO<{+pe<|T2b^jrsI9hN
z84r76xhS8;6@(`OI3El;6xst#OW$5T@rE>a{+W$#JU}OXh81Z^C5Qwv&5ci#x1fSt
zlOjlL8l#9i|9QJTlU*6*SgWyUeN?{|={_}j!({~#7W7dMA3Wvv%OAe^Z=Jc{%U&$F
z$P}6z_DA^kK=Ef-e^WzCTFGWe`+f>L4U;wQVDTL(S1~gz&Gd(QsPQr|QF1)eA6yH1
zSdf#<e0z8pE*m+$#z9;<nce;*+PX|ir8SAzm;F7M$&+4ZxZ%oWy<;FHjk@`=$!(cw
zEodBz3udy0N@T<wyErPp8<a2!S+&mzl)DYh?T<q@Yr#XRn;L~&Pxo&@x3VX*M~C&3
z`DkZtV+~bM*DeV;rvTUd<R>u_4Qm&YYbJy?Em}pJ!i;9A=Xr@4gLfVp{$1dS!DX1r
zLn=@Urb@@_FJ0T3cs7HMMN4OdHuJ9K&iv<bV}9PtnK*T#hv+nBVpnZ4J$(`7`CiYu
z1Ti@H__r0UUZ>?0o3TCHTDt{V;fHb;D`_j~8L2yX7;LZE(t$Dtp<D2n$K6XGcljI;
zZEma}05d;-=ZW0)M&~IwRkNfiy=h@VH2ubs%fbF@JzJoCA^wz8zQDwt>e@HDh{)J4
z|MMnd|49sHGRjUYumGLYvOdsEt?SWkkTihi{*05IEHa$!WhH)RYnEgaTH_p^_{fDX
zt_Ewn&2KXZFy@`l>?k8&9x~8tsQAqzX~M}-m_3_I)hBFup{QCnM5#^maYk&ogE&OZ
z@ypfd8MshcAfp5Frb$i!Tan)OUOWsJ9*5>>AnAZ`;u&ezAzn&x!Zrxe8O(xxRbuhp
zTEF(}lWNaI$KiH&Qd>z-jM((bU!l#~=kZkim+$`r_AILOHvh)zhNL}%?X;(=(rti4
z25lxcCQ_xx#H~v{AU(pbBvl120Z7L`(vZGq54#dvuQ{$YC#$Q<_ncJthy^@ao>MB5
z2`ABYDkftxEE4182|hu2bV2Cfb__##wSpNBgtuDL#HTf_Z`6>z6@+L`)M{pBRE0MG
z8W87!6#(<R8#Hu`0K_#JRcvOt(GS0mHRGIVh3Tz2y!hZN`}6+^A+DO<nT;^ItG(s$
z2>z~6+=Tp4`rbErxhZg6rf&ewF+FVtxNo;@r|JJ>VcwZsk4rk2ZX?f{ej75nLH)?G
zu<w%9)Q=D_YIMKk2-?q=PWsoqao}~;H~>|ryGe)Jf6H`RdS%|6$U?a9S16Qg8^j}%
zZ|C5vWQdFP1WgE!)l#%YC4j8Z9WA8>Il~E<u8We3A&L$euA79Cqj!qful#QMNHvf`
zZ<<4886O>lh`|%DDC%a^h$Mz^Y2}1B6!SfN8-tGaA<o?}!;s=kw`o{D@hn~h6x==q
z7xy+O5t)0-El8GOP>q*WVPw>27HO~;7T5%(#rwo0gxc_%&BUPBK9Lq~bO>$+QscuV
z%+w9}xunR(g^g|Yj_OTOr}(uoh>vVQ`+^X`Jz?{9O)gN)L%Voi?utsD-0h)FxvBld
z{v$<~5-pB|x$MV?1P|1QVl-ftX42C6Pk(zrdYi9)2X%*y3QH}tKn9_>KZYXyEL@H8
z`T8Km;Y+iDg8=GSF!UwrWs#N_J^UT$F>4X+3xdk*DBgX(vBY>BQ{<npX<zp#*<901
zh>d3nxnd4|*q{mV&nA?Zu&a_sdrCGHFU7a8E?qkeV?_)*qkr#nECPK<M{}j%Urigy
zuvuFq<n&8=cDFdYYvsT9K2P~nODBU)(Iv~`;a-i&u>ES-YP$nR*vO8mI08ke+Pz4I
z?%SX@)_fxr2Ua$%IV%j_w8iS%^`0ewrl<+R=_^L~WaAyNUSx~<SO3oeoJ&r(8GCFO
zY43B?Ew^S5-)$4h0$Sv5IG8dDU#+p5Jv^VIdY?8>^Z#X)JS?~rKB3-A*W5|!8NDkO
zxhn8pBfKO=ER?NZBIZtE&}`(ScS|}y-(etX+HGq7GUL{r=f58fiO|j8?HVj=$tb*=
zS?s~e_v$o>&qQx3|3T$J#nt6!H<XvR`fg%a+t2Hq;C1JHWr6Ujus=TGW+R)S;Q|_m
z41P0j=E)gO&uN-=51z@g_kMJyl7Ihshc*roac%az;n0)*@H~4_z&(?Ih&O2PC8@2*
z(0hVb$|+B!%SFB+`xd;l7MO4W`Nt4uil&0w5}d%*eT<j);ECNLgzpC=={K864rTl}
zu)#uNGPR7XIRJ{0xYLPiDtGR9tb1vK^9fiWy)J^J6eoh=>63>?t!IT-TTiX0A}dM}
zh6!9EUIl~_74{{vbwK-u^e~*g+h!B3)1PRxK6tpx6z*(5vqiLcwBw$yg`7+*W}eIr
zy?xdv@}CFyI|p&?(@66xq!n|~)E}0<tSc7r8Q|OoU@EUQ<OXr2;Qsuhm_4ITVL3|~
zvaU^$_QYCZD@%RNZI#giIZ~IE$I&GpBHL3V%lb++&4|WQHN$+~C8U@Av+~;(89vKL
zd!d$sG#NcXeJS{3O0u^7;uwv@{aCIZHUAe*F&oJmjAA!E+h5(Ok+~lFAd&>}T6xvO
z9kl)i4IpCB)&Gq6J~pZQmeTlNY<j)WE=cRWa8CKK>%#=wBO3LalLTF|5+}_`Q1V<o
zgm4@ZN*VM0X5YNBXD@`!+<&j66#-&d4XsZ0jM1Va>m9_Mm&*QZ<dv~Tyx`fRvnj^I
ztwv{z(xQ1i;Iy<BhB`w<2koLDJX?(1MoBQBp>JCUMw_LmG8OKB292#vhxs}oXeaiF
z%$v?f?X-z@?f&~5A}bWkqNGB+Qpsd?e2Z;>YY9deSP=U^kP%4)uJ3fQ$FMJ{sYW7$
zo<`|kY;+^!q&A=>QGD~A1oVS662oSp>G|1#RQ;6we7>AGSYx05$YH+F;$Yr4V5}_A
z=A2hL3{!?h^K#;u-}Zgn5EJeV#g@Px2naN~BT$ZiWgsVJ3t2J42ySUR)<3s7@HXEM
zn66qI<V0RNY5eG5{Fy0HkEVW;mw~z(rV#h!)CA9cBEDEB4jB)}W+qpWploD$XA=MM
z+`6*?D2DqI=hU&tY4tFI`!SHtlDJ`snC(CN^|9v8Wt-Cix~+RamG<{nu=W>WzTxpy
z%X@xD#~qN!0aunSwlEK^SHDX<9OLAkqHi8l3F&P)8Ygrduk^gt!GY*=<tpQoFfVrc
z1SUb|;r_6RH+r<jGrSCAFCE7oGXR^14)?YVSiW3G?L>#ThJWAs#Bvu{l^6J&d>}@3
z!uIJFyrw1nExE%(vdm=4Ann+buue)KRM@wFFSL6#y`&q`&}Y+HaWXI!j3%HRmk+9-
zd>M`fkL5#(@Qo)`v_Tt#q5;0mMuH2*&a4yKC1Dm7dXc7QS>01Byt1485WO7!&vi4;
zAr}W(1jN3x@P}wCUjiuSCeVJAmyE~e%6_FQUY8!17x%kSHKowGlPeWu<CEL`T@f&x
zvVF=6&?3)Ik%-NnWDd>8I*q)B`OL%LQ&F7DBc2XUB9jeTH*3qqoQk&TJZ6$$4iUC$
zR`oT8lb`f)2d;MU=%g;W=sjEft+$9b9#7q(G^TAC0<^q70hjlRc-<F9Gv!~wtW81k
zzN=B;lLIcIu+HPA#gGRoA*3cW0qYwap%FbOLTgd!KUuQ~?+JF?U{+}jjWFKb6Y4)B
zP-{vfa>`GH;P1e#vS2n#eKE%EP>Y14zFX%UB2kDo-Aqv6<F^xe;5DZ)ebR_{>=tY2
zxZSAzd<hst^yxNXlQzb82sPT+Y7`1_Rr5(yVner4&fn2oL1mf#`QW=ngMHuDC%A3)
zEIyNRj9<s%#CuWr#=h$Kt%m0t6AQvy+o~@S<mk|L-9R6Z*(i`n41Hme(=)L&V9SAM
zTM&%9lu=dVa9^rWivbvj6*;WGo71vSXeo}h0icKJQ}wyAJ*+Xfsu*{9PHfoYd5lxh
zf5##F-MS6pW%q4_Isc0|-^qO=AOc%g<BD}+T@RW;lJ{pb20blh$AeiueMPUu>1_uQ
zNIS}zq^*G;2yoTvctU*NkaiD5I7(jbxws*3Zhdm1h9HR<`wGQF6*Zte`PTv*yiox1
zgqgTjV2;ULm=>L|6y1t6&CuRC2w9<1Ikwd^^F7tM88BI=$8ak>$CNtx`rE=0W6_`&
z?WrV>kTtg}Qr#Xp9}*YdZ9BVNJ+IF=U#0=6d8nVjAhosa%dzsQ2eG-A3JYKXZ=4m&
zT4Zz@BKS+6F1I>Xb|Z^>I8!x88yI;`=*|mh2iv$>q+~#2?m0|m=b}CfSp2zj3?9+q
zLLPN?LqXMaEbx{Y1*LgF9TP?Vht4_F;38&pBZHv|$Ni`vrjtXPg5BrM2OAw(Z5_Yv
zQQhwm*EBE4801q-R(-Q5>emdMf&))dlzR4$$*IflzI6np$&mM5{ySpWf89rue_WT`
z^nv~o5VR6H$3d2zHA!Gj@T!&V(XE0uncT$n<_)ZYyGDp40MB?(>cD&6YK5nf4J&qV
zk(D_8Ul!u392%M{87rG1FQI+tvOeyNiZc}Q=B9DXBW>0~5xBEA=WZl`D_jEeVF7vC
z(^l#0FOM&h?r%+m5*btcW8%`)w4cvPGXl9TNXVhqEq1REsLxj7Md}^><;UvD3gq0-
z2eO#3ru}VEa<KGLgl_KaA{Zurzh`vi@nEuD>yOKP(mGfrQi4gQ!ACuYZfz;f%9ke9
z-A_63(n3U`un~sd^8Q7cnhDr7$Jbl*QGbiS{0w#xqOr~G7oy5xT{g@>mgAH82+&w`
zKnT}kaQjk8gDmcq(A(<GWFrSHD{5%PBaOjxcd0U;Ji$X#l)E!G(^^1Z2RvGAVu~VB
zst{-G8#Nb^3FeAQ0ZfXgA!ZKGDqTF>GCt(dg{RWN7sU@@LoXG19Ssg)F2;(+l%qlB
z*Pk)q9jCAbq`EOa9vqZR!lMCfi5|mAoxVZ#hANrq^Blug(Z9tTFN8M1)0Mz>1RGkc
z(q=f)4axofMGo&+SiR(Z)gn~<6rD*sF8Vm4pi;8=UrBJZV<e9#wE;^$sa#kU(>BiJ
zT>bPQuJWc0q9mH@K1ILkQp<>kY6QqzT2#)_r57flKz!<$;ZbfA0|*mS-dqbnT?ZEX
zBv#WbuIN4Bqif>q7e$)h%88KtYQD7{g9~_MkD|=`T$t($I%k{9u()_g?{#>I8GiUl
zlex6UG5A3<y=Ds-H(9CRvxpyX{)L37=lo=>ei)x%UK{j<{fM2W+t_#%WOI*O7KF8%
zB}2*Hi>eV<{&`1y0IC<JDg=aR<k0QT0ibC(jI<nu^D3b!Om*O2^hygljdjlee}sv=
zy)Jj?PEVY#{}RXi)fh`o-P_x?CKT*rj*65s7KE=K&q}~Q+kjh5r=CQ&@e4+?#ik<+
zelmvCh?=P|-BgneWi#C7u_jwzxEG1JNRyji49_yn0Z435MRksal@zC=8gH1Klag3r
zD1@vj^%UrQX<Er9;qBsyqIyk4&ea5BmJkS-MPuwT%Bpo@*x3Y4C{ukn6R;=Z3~A5)
z!P1;3K{;klQJ3h|vUsW1(ku$8{{8u?CT5oEK#32hSGA#)Qo8v2h<%vsePM0e{3(LB
zQUMO26sO^(ygy6@!r#z%Wxf8P7{uW5pARWI6pACgX##`M%-;uZ5n2jA>5}uEsd!@-
zoyF(kCZXalNfGwqag4`$raE}XmG@4p(J-^~#qXK)I^XloM*luOkXl5Dl{hmZkF0aT
zEkvi)uM+Mk`0>0`tbZsTs43$UXlpvp6=Lit!|t5YeSEIB)fmbD0Q09>`2G#`{`+w_
z$Gf|PjdD>)8e9->_(^J)(Jw^OZC~-iwsPMpQ$wbV8Oh4_UgQ`EQ{9WWuaFYBnTA?r
zuJo*0Xf$y~lXg!af}6v8D|E@%GU+b^aDsD(+9zL`9wrR#X40ZG!?ZXeA;0q1r!=eL
z4aN$KD78)9kV$SLJ<=Rb5)bfHqJG+Kv2R{xYbYwX-Sgdnfzv87ot*o&jPW|=AuH9>
za*~S5*yybG3wQ8D{I%o-Q76Iwb7p=lnNzmsyI&?19G}{5=l_-LI!2UD5&*Igr*MM&
zsug-`P0yMWL5}32Q4jdZHr=XDN4h$}xZHxdIFzUE>&Pzy5%`W}T}(oN6-oF<vp1zN
z1IvM9UQKQNI99iOFMgi>ItF^*RNZTmQ|H;Tn1e14^^<t-dzq^n_Bi}pT_?>Xd1V7q
zV2T7O%pM!d&w3%yr@yO#oa?I-Y9hVj(<=)Y9v4?z6TUFlMyJqKooh+sN@i@Ug2L*L
z_n~!j>wH@fzq|`u>!==zv2fZk^s_W)ExR1*$xXq{TW{dLCXQjgMruyZX_uyoQK~ne
zoxC}xKFhVZFP7L!OFfLHzV_9pl_TF7G$T%_Lt_m|ks4-fm$xE;&&N_im^!0RgpqtY
z=`Th_WCL~I$mGMyMYl6-Y)J@Le&#?by{Q6L&FT#Tl#p6NgGQ(g^N&Xu%JN0);iq}d
zBqH%2)7Ih6`H@UmTA$HrO&>4LJEN)f3-YAr)ds5Pt#R{?(NR=DhdgF7`(_TFVTPA;
z=t8qf=7<5bArNR*X+@rVXDQ!LMo>9{eC?BO#);h<6OJilCh8};nr#pe4#8;-_6&qD
z*f7;dp{3-knivwp(}iStsu>`D(Bv2&zOZ8{Z8hCoF3fRI!07W7KH^msG3f|Bj%fy0
zcDLZ_l=q2ExapQ#$9^k&)`g)G-=5Vl&4}I8m8gF}5!<AIF{m>B;6-2pn-b8`C)&@c
z5wkL>KXeyCW5;G3{816o+T+v%1!-3BEB(`?$Fr5pVYT5X%`SJ=`<%L$oCOd_y8@2}
zEi+e6L^Fj#qPA9>%3X!tyv*?Xzaqd^nKD>F7bf*@DCGJGxzB|s8@6)?e#>bow3m%M
z42(bwHnL;lK$PO;m>1A7C}ps)Vy3mZ%eGa3X(>KaGt_iY=2G!iz7sX~)^$2#?fHy)
zL=iu&Sw$>5>g*+7A)EE}#~;RFrV-dz8;M^%^_`Z+m0w+e4Zy~rrZ~`L?GqDD=g0ZI
z0+UI<JP!>k`w+MBR8D)hRhdntmG{4C08vRFwd{&r+~@j|-O#^eI~)0ECN40oM%x%v
z33&TPY2wcQWo+<>YJ=$pNWfLk&IIcq+o%sq`oozUc(A|f^rsDjhFq>WL9@Z}FZM>S
ztknpjMVoq`5$uG##}8#UG<oppPY`|1dj#2LkR~i+106c)da8u5X3gm}I)FGE${J6t
zp%CJukkcnz2;qA|n>?RjztJ4SNTlS}!NsZ*`qPo^Z9)l@Js){wnDbXYG;#>f?C2!{
z=CTjr^56YP9l^F(>C^7`FyrSJ#{~2#4odKMf(VE&oqGSQ=p;g>1^M}^?*oQk+Fbb0
zZ(*~HYdzT+?J|jt-BWO2!;**3VT0Sci9egP2{wD|Yom8dMoV6S&GlfMg4bHC51<~$
zY~=<Qvgc8CFy_Lq_Z_{3wQ#n_-~w%RX15PY)pKVxy%A1JF=mr9I#3Da7}*RKv8PU!
z-**l`={Z32?z45-{KnGB`wkOqGUZdJG2({};Q%DVBC{=557-Ti1X)X&4XcWMY}}=i
zqd0Wek|@$`I?Arm*c>WxA<oCkE6(sMa6jI`mY1(7ObZu`ih|)%P>@7<x;h%i{{qFD
z;)aZZRL<OcFNgG7LJx6kZ!&-Ah#j)tMszr<#~N-F3`;S5H8y>P%)5&_YG5JG$c4e*
zeLMj5MzhSD5?HYywoI7WpIGJ$PTBI%mw7NNrz1tKXybykFypm}0<cFH{3Y772cD-!
z`3J4K-2RVRov*~Ak+pi8Z+#IxRY%G2TzWnL$W)z8nDPvv^E!P)pr%CYhG^E2z@j-F
z>Wb(YcU!p7kZ}wby2~k*xjm;6%6lG8i391VP$ODDucV99llid5D7z5w56!qH*X!cD
z1uc?TpzTE^lmE|;miZH*<@%_fKmTB^q$VcQGaBD1h({27dd$4jCY&Y9LkiKrEN#(^
z8mP@n1>TG^koi;ms<)j^25Da_sB|PIOt|3sgTt5*%Dg_Q$;de8=#m6+%Ddc;7_sJm
zyy7%H=^BH|P?d+AuvII4EJ+V*AM^o>c}<ytgRqo7k~|5=^OwF^p7911mfrspzCX_X
zFGt(;sCCLDF_<(?n#=;U1(|o0b%-kqL#w@z)WqJ{k_^saa_i-_!^NR1AX1Oe+ckfW
z=(&l-QC%iQGXs;6jSTkCWo;}gDN>vB1mwRlqvjWp9FzR!vkV)$?aoz%kj!X83nLHS
z-==L>GuRJ2?e=g-@qNMu0wng&l!>g!;)_|1d9c9YM-c#HA)<%Z3>WEZ)kot|I#15-
zJYp${aPDM>voDo$x=E~6C!I@5hb@Lwg*=XM!>82SR}Gck{0zC3)A~{bf@F~5=l)m3
znqkoB_$#%U9~+YNf%H+p5y+(Lw+L-;fJ#WmggBM;8<oY6aIR5}_96%yZkM+XQr?Vf
zQ;5eWBY?}Hvm4&?1rMi{Y{>%bgw5wuAao~^N2bs>?#@qexj0H9V^B<TSfOB`T{jk+
z*ewO`Kr-BcO1lcuEKR1jvfz$X>PU0~)pN7n5ki}wa?Xtl>13I<mSZawvq=Z79WLYg
z(BR^?(9;Qv^Lm{YDc!uFSA;dLWz`N|R|8;M)ZR$b@5eU@Rlc+k>Aal-_9?7QDHf7c
zg6#=<t1<oRj_aMdiVxLbQpmWnM15x?wB3Ig?&<_~%*T=Aw6ijCvKb4d!O!>xkHi@c
z2bRi?maB@yDb~6OxVsO$rywZemoCw;S+_hE=|O#Eim3Dtig4wxUzl~i9A`!Wp6@pt
z8?q3ENM9U;kF256(O90LmyMSwPj=70uEpZspW_p@uPhRgm{{2~N5)f^jfn=qmEsf=
ze?}9lw0|Ljv3d*}IIqYF*U`-F-96gLB3N3KU<glx-d3n2_BvHq{lQ`!I&?-bE3)Z^
z!_f`lo_)9YcjzDSLn1n=)92)YMZ%ynW{r3qv%X{P6cc|;I}SLdw_X0b|ANys)Ap5i
ztqg+wtx(<a5+jn-3&+=J;+}sy>!;y4`+QXki#vGXM~qLmk>*`Z5Dc|EM+9~A%mZu0
zYz6-bgJ1<;dC45yHPf-1a`0ta^FK)Z{EIhJilT;K!Notn&gqlM8Iz3o0{VCGgO4A6
zuI0S76ZRAjJNnJJ`s00;&}Mp_@l>$GYgggdetLM%2l3%-KoZkthP6nB9ZvgFsM6w3
z@$158PSr~lYuHQ+2i(z%Zq49u>fY)b(Hp+5t*lF2MeO+B-?=ZkeDMl-3oLUhtwwSq
zzt&iOWlR^QpQN+Eg*)H-&N9&eyp_h7V8|!3OQn?HUbgi^@q&v4cV}w4-M#qYv8gw*
zOO2;##SL2DikTbt_5g2BVb1kKb0rtW0=vm**OC-r%l#_l=No@lz$1s6^f#G}__H(8
z^6jkBNIi$8N68|a`t1=@Wy(MpJJw25%O`NDzYUoX!4h0)+;h{yQTPO~MET6;K&jm0
zB2?9{2*#W_)&YOn0O)$3z(w5Py5%X`Y&hV<@GkYr7C!7m?OjjaT|crx!OoGB?qX4u
z<9R{Q`QUmP!BF{f&>NaAtVp3-g{IYEl&rGUy?Qtun^~G?u;{i)x;h{h{WJdLTP%eK
z7jtA<?v@pLhj%RH1CUE1Tym^$@IS({0X0m2=N7(&rwtWs+^WtXvNq?LDt&<kH2a;k
zh*li!`jU>Goj+aSqz}-<koz+(kh|12X@P0cu>EwdP@x;d^LL4_vwl|UMZcRM@~rLe
z{@tCK-0S%1kAokyM26{Sa?9~;-80Ur2|aICgM#p3@<7uKfJNJ+*NHSemL7hHAhW5i
zG1T@WDhq<3{4+Z&haWsS2tQ7>BP6MG?>NA*Ltm(`Csqb+phhS1-zr^*Q%4&FyVBe+
zMz!S_+>l6ff9vVNEm8W!C6BW$hT+?&wSZ=S44R4h7JB_Iv~b+fRoFi)N;*LhUsLoN
zYka$0ANmN82~#}s<VA2#FE*<QzuGLo-8x>TUy^4MI1Xt(q;(uA`DT`S;>I<1P{opH
z+#5dOdL+EDG#FVz&ztq4%26nxg3?#CG&v?2HH2#2A7MVH+SL4x#P(6%F^Afntmfk9
zcnEG0gXDB{dr4RFy;muE&rQ!ByjuYKc|!!57c?<ayLG*Xgz(gF$?|&0H7SY*A>97#
zEyS9TkrRhvaul(hmk}Cba*RqBMetR2l?e9uH^m;^mb@@w$6x6`5}sOE5}K(Qx)x!v
zy?Z`As>D?M&!9Z+FeY3e<YxFU`r$xw1t(4XHVac>L_yp?)b2}~>$;w+rS!@svs%#W
zVt|)jLUy3(*0?xUDYfuJSmyYIGS3@Wk?lHpp*~PsZtoA%)-%sjt!Fby7o4!J|1wC>
zm6;?K<-;N~(-61F!ctrHfc%nG)5A(3Dg@KQ+QgE}u{zipt`|9ybM)i*-1~%>z|rTA
z{*7Mg*BApQ7kjHe5Sehu3?lE*SDn@Pe+60a{E*~GUEwR<N4q0+ze|>H7zk}vzkpc{
zAg)#%J-ZNc`ay56W|&bbs2-C;efLanHz}P&u|7d`c7i6M6A{VT#a(IQswi7L)$S&7
zT2Y<&f4!0Hp_yg2VI0yPoQka*494c}PZL5;lT@>`S&l@JlXLJA8I-o0eH;oHMYI`|
z={EckyMChSug&DJz=>ZkM3Db*;?*=k2g`7bofil**T|efn_v{*$4R9wxq&_q6w`>y
z4+cah_)lT0)C=)7*NhBXgCnrgq_lU!+5xx@s-YnfOp_6ru8Z-!OBC166Z=QjY(;d8
zq7W}9Qaycu>Rk>>aPc7EZ{*&&0FI_g`T)cH#1s!I8NG3(_o~17!V!A^81epJ5jes{
z5Bdlpv?<6;?EbmpcoXON8e!7Vt%akW2Z)Tu!jRtUx8nB~6u$o-H=#0oeK>BnvcE6&
zl<HHztr`(xpn7mjOI)g7CA7Sl*j`>av1Z@o&PH_T!UZ?0<N+{WZ5AV|!enB<nEY=Q
zgwx-D=3<ft>4i4ALnQzDAc@_BWvTtyCD*z0!z_M_HaX106ZBM}J*(yY3TII3!3w}z
zwFPq>{`uqJ9H-W+`FB&vO<%mJnq{qWuf=~HH+4M;*k~eczW-g~TeqJtC?LDh#*3f0
zjreoKiB%h70uUew9BYU-fPQ=|$s3aQDUj_sXhW4CuC1hj`Omoy#qf|$#lM0@d<tAk
zZOAL4PmiCCy&l~e_rfT-e5q|!;39V0W=Z-}>NTdb=MZuK>g2!vc{+*ox;YzuY2YPY
zwIPvlPj|3(Z#mFz;h^Ngsa=+Lx^uru%cW7SCl}3{fp8Aj4_Bo)#JGQ6KAa5hPP#Ds
zrw$3r+^d>AK}Nq<TxnQcNV2Or`V`|{%KAFGYYV~T6Uzlj?UD-L4T}+o;-qOFY7i0q
znz5*rr+01SKo8N;#ReC#ev7yzHVpMM&(*8({*B4>B9fjrrkR{;lotQo2_N>W8Z_kl
zID~+&0v_+XobKi_dz)M-tLS$4ajQjL?y9Su*pTWd;_7X^oCpR{?suZXy=zvVN$1iK
zS+`}f-1oIpl>>hyapfzi%dfGHmU>q^`%QF>d>Hmp|KRyBzkQE;w&z~=+||_=RZ_cu
z_9SBi)Yp%-$s?Z;cX8ozuV<iM&yo~fsDrilOK1UshfXohOM?`i%C$9n`TP4X+4KE2
z77SL?EfIgZ+IW}9Oze2?qgN%-&I$K8HpCzI&L0oKKfGt!k>wUSkoeyE3KJ`)QsMLW
zaeC~~rYcdtyGWtF?q<$tpeurDwa_kWG((U6%%yl6KW-I2{^Su4?`o4Ef-c9Pmd5%6
z_0a;LHSt$DbDzgp(usKOK}q{GTOm6mp#J4D#Cr;JG2jVimDJX-J4*4nU%mYmBb{o9
z@Vl}o?pO3tk2X*aVp5Cg{b;b|*RKy%Ym33v2dyPCOt=QEBE=Dr|Fmo>zx2|3l!}2q
zz)tR1w6LE_rw7wHT*P<BUo>p5VN@yosLMb#>^DOLbZ13d^BR_Q=)$g&6@B8WJ3V~M
zgV0iIFYsHK*OgquRrVhVkx-&hz~<MTZqULnGAi(|3T9P2V2q%ti;NQ^;A2|yt!a@f
zr7f&HGzJ^5`XZZoL>Fh(-U+9ei5gaLF8R&c;v&qFFwka$+imNNH$F*6(ph{Ofdd`v
zs3JZ!lP;U*_Lh$sUHr^#nbK_q$UyK;Mx&09<!Y@w;%ckg#FvLh8lf6KQ9N0`6zY9{
zj;}eRGN+q&q*S~bD;ilbE!R2l&L_}!52O>xvT&N0D~w{q^wL##GkCdQ#=F&v{BY3P
z)2p=)AbfuD4nOYFlxpES7JH|H0y3C)D-m}(z#$~l*z=3OjEnu)U__&CHLoD=@56!G
z2%4_iJY4<OT^yQgN-DZ3FDSv>ST4b1t3TQTW=@DPR0qs(_>%ej4}02Czte{=`9w9f
z7vF4xp4pUdr(*=`54SleI7Q{Pb?J&gHYL;T9-%#s`Wq1>$WdMtvsXP3b-uhNidArd
zCs(u9ubLVzFm|6I-a78hA7RMTyzea6N|WeGkx5yx4zpR3DUvumA-p7a?t%Y)9kdJV
zhfN0?e5Wb%=Q}1Wc)}h=7Cz!W)U=2E-Lha>G%qr$>#Gx~9{sHC?zbSE>tE2d=C;CG
zhm1}5E)S_ysWC}waz+CW?fe;X9_+Iky<cadey__&Lsn?bET6XI4@c4#6fv;LG)2S0
zItBGAUi+~OJKN7+(pY)77>w}w#*MGFKyVsW*p7p-KL2J(aF+%9vGXmf;{D5$&z09C
z(>Jv`sI})}oRszJtG#U(is+-=3fSy@J;;pfz7hV-gLLqvDR5G^d7F3w!;Df7m><xc
zr*J>Dd=5Bu+utvlnBqTGOcY5T@|v=6<$Pbtp-aisY<w{ahKu(VAg+F*Ens%(!Tt%y
zuuonY`o)8G8izJ`UYZcVIBsmmULy1)QXKhNC<k_+glYK&rxCaXEnFsEcJp%bD}rHQ
zy3ei8Z!3KSK6NXQb}#UOmQ!@J4UsmbCjk?&0mrAAQ#|5Yss=zFF=2$sI8P~P)OS>~
zVcv(eoG(`tb>PD`t7BNxvVt7PlM)EyZ@rD!LgTLB3$vyA-ovbQ5DCgEv_5F=K(g%n
zEWAAVU*qV8kIrWdr)p+V_bi(2ef}tSj65{m2)LL~q{w+N<NSPB03mo??xDnFQ5T8=
z?@nz+sd-nqHGxAEn134*i>)u;zfm<u4nFUr|6c73J!xGW&g)$J$X|zDHG!?Z!FzAi
zVPaI17Yn)pJ8wq+sf}QW-;V8(GZl;EX}n)ie?VZa|M;kQ_Fi`p=K2%y#q)CId5vE%
zEx;Ft;hQ?^zqoeoRgDe}Be|P>RNB=4w5@MSX}j%Md}-s9qd8VDAe$$H$k$x3)W8C-
z`YaiJDb$=Ti1dXUOw59kwI^?@=UDMK+aU!n%b$VO>O2Gm%Dl&<NKoxb_-E1{|06*^
z0z1Q4)`;(MGWuaRE4T?3OIiJ_cHX}O(5B;0TCAIACMZ58Q59j5@4WL~q8%dS#yN_G
zo$Groaw0T_4roxT|D<RKb4{P?MM!7Zmun7Vo*{Z|j~JTr9+J-S(yaWqk{RVzU<OVR
zdmHgHkT4b#o}mr3*m%-fjN%pvKBAs0*r&o&#ebqQO)nV_xc!tTC1@z*W1%R3^BzyO
z-t+B+_Vbc!F+I+<IQ(-2q!x}~FH-5izd%W8RsXy7X3mepP_yZoh2HZ(p=3048OQdw
z>Gf+(CIjfhM|O~~q1U`P<+O&@pW>h`Lh9;I(;T$GoYd6<a4N>O8FcFr?oVrnk6Y)|
zi1hEuMrZ>x&yOG-J5&}ihr15ES6|_<t>MDw_a_g5a-QcA+dmdqkE5+$v;^_LK@s_d
z#v;c!T)lFSxL?i+$7rMoBya-JR?NnOztvzM-@0c?zy<xw8c}WS#FB2p16N<{`F`+g
zJ8Ac@oYaeY?$Ckw+ttc0wKH#0F_VJ0s_tK#>TWG=;t~HsuVzII^PFS+uX%KQMI$zO
z$N>U3Ua9M6NY4LCF9Zx@Ko4e&#^ZqrElF)S>20F*q{m9=!MW{ROyU|5ywLB0x&O+l
zQVdon_&W3MXLNNsJ0B_i6X2%pmXjQ<xdQQqV_d!;X{;Sf6CbREtaN!sJKwpk2RMRc
zmbjn8OH!t}d6NBYvGO&w!MOrj>p8bdZ&6HU+?lxFK!#owaFVoA?HR%CsN9f&wts^y
z_n2#7KMIS8WG0O6{N0NyBHzm;`-Tl`cKnl@EAxN+^)_cXmpbhRhe~9mm!H7a`PR{q
zOub5901me9`PF%MXUjZ%h+38cPG5>)W(S4r_rAQ*3-><auwuA2bh=`&T6c-4XDlf=
zDq4Xv9)T+YL>GTixYSM|r&Y1ctX{BIFUPuqq<V=CH0Xlk{a_=kV8S3m_~fSGPW5*A
z_9{kr5{QBf)FN8f?PU^Oc9WK;YPmrsnYo8@lckzFC}vz)TWoZ97jYG@_iWb1hdjW2
zi~64BNK*fJ)L8P+b6=A8y<hQqEXU)$D={N`6iwhBxEtV~;a0>$-`Id)&qTfQ`UbGV
zJE!W={F!>yS~XeyjA@bKnzsE1QYq-j6AlY%yI$QWXR??k@%dkJ#4NfAyX1wrzF2zq
z8{NNGl?nm$fx}gE$wPdJGQS>2smpZ~-{#2wf30JXE&co-|G&>W(!pLoQ(U9^UBRA#
z70^4Iy_CA!L8tD$88uWdRtvLnewuKtG11@8GTdM_HN$|caeHF4%$^8!zV%UR6!G(w
zggqoZgR3Ep&!5rda?oHkXr#b%ibm0U*<P$2mo&gdO3Hk9*A>JFs<^)H7YG=AGJK+B
zq5sv-XP7d^vGa$s!!KltLp@F(`p~SA9jJ?N{b@Y7Hge8bQazzjN+>}cQCVzI>K*ov
z`xm8p<EyRkpFQ|w!dHPg5KbD+=nwQw6wlHazKM}C(Xz=^bKf|vdlhsPj9GPAt;-Hy
z+kb6eRqzm-lEy6o@jXWL(A~n}*Aq2=_i>7NrTb{rnNE^5{)C0^rk~KON<%Dc0an%q
z6IHSLj2f8XMc#bB@uNhk|8)PumZ!0Sves9ulhNicwMwT(akjb#MvQ)sjRw}W%REIC
zG}8LfCBIxFN_5&^HOljO_+i#KhFXKtGd1T@UF2~jYC}7S7&ZLT(_pu940pv5{6-^o
znaQsRHQSSzFKMEpy=du$>#p4Z_vhDEwL$8%(;VvYR1OM!A!W&Er9AjsA>!JjN#iH+
zX<0x_ny(mS(}PIU+pN<2H8vVF<@PvQKhszQsX_-Y1ld%MMuI6$p}u=e+y_PTNChKd
zMG}{-_i7g>qb^SkT^lq<>;RnB<R@1Y6zzm4e#$=#S{ST3e6XB&PoM=s>;58<R!<Dx
z^Y)W}q<=G)vn4n}H|k8fhAIRz<B!9svtegxj4krVi|g3=(vdgNsLR|>dj{7OwQ!OC
z<%iIVWzQ~<BA59!R+iNlMF!Xse$_BQLyRB$IZ6@HmI8XOVq4T3+a0oZ)kN@SJ@*R4
z=<Kx^*79{2o-o7ffpQ};?bPZYkS8q{)vb!b@pA<)Zrcj>kGinJaD6edl)AWZtaTra
zCZ?JJhd0~N9AM^@`zN5ooP(x|BGjl~jDhz2bISrFnF+6}zEq1ISDk;2{x`$n;ujxz
z2~A{?B+uMKyw@vRd4Q3|SO}dBZ9lz&C+%}y=A35)%hNl6{5A+O;i{4;L})qKZg*K=
ziN*Cm(ydwxj@nt1sGJO7&IWCRd~)+j*J}+bG5yWzuh#~vV<iSH9DqqpeJ{a%Xw;b1
zDGGVrkYGky4A_Fi?99?ml;oKu#31~{Q}k+N6fyMRcS6Y;p^x%*E$$_@ij#SRr2`pL
zvh5<~0SKn^gFExl`4qPVQBg>JSuKs=jvpr+FH4|RG9D>X;HA7)b`<%Ll~wqt>1LSu
z_MNGxjP1yj=u?s8a6@;zQ7#X}ii7|YP<LDeo_Tf`@MK-~i-YGF#PwKOLP5-~!Ct`A
zy5aixVHMX>$tQPXf{h$1^GVg%S~?@*j)63AlU`yi21mixn6$L@<9@fOU5?jWFv=M|
zS-paH$$%{m78$A+I6@D7i!CDUt^FA>Wwy7;H13Hpd;31gB9T=`;W@1n>tSf-+Tg7L
zLW}j}$@2ipn%}*RD8WqsIz1Y>Is@H^ZVbbdT)}M3Te>;;@4Ykh&44lM&X;rq?hx4R
zpORf-{IY1-{kO}7_m-ct2@#@(?dCpaKx7W=jEUG<?ePwIH*%4p*FN>uUE!X!m5(2%
z^V`)=Pc-t*k;jh@xkXUzl2z&aZSuy=Z{luB7?v<-bhQOiOqngUzOn8PjjLnmNi|PC
zG|{SNsxpJ;U!dt`?ka^}Rv+G@2IvwJT#?!ls|#@HfQ!*$QH`uRF6@pM@L<GhH4)C^
z(k|*w;dt@5C5i9vUo}Eyv2IRbj7SHArGO0Rjddr}aM7HN5^!fP0))FgbIx%yh-xJ}
z5WyP|^`$7hQV^Mzf><n)XncmA*yhmYliP$2dekmYq{|>U7QYzeNZJ6IH;eS18<yXt
zr#{jde*Ey|dG8^tQ{re!7d|vd_wcXt7kd$jE%Lr$&-Zyb`g2cPx>6=0LD%qMSG&sL
zPQ1PU*N3po0XxVb4`TMUaenq*pxVuvqp+Au^&h^~WxpcdxYkmf0T0?E+~N5M1M#w-
z|64^<dbUH%<l676G$G@_-XUddi*|49H_OMSL^-ln#`^}xiWYZ@0eYd{W59#&x|b6@
zDcn0aiLo{ikpNs}`t8Q4ngAvpwu1hh^Yy5(pw#~k67JJ>zJbw5-p*{6V-P67(zjnu
zo_%#Pmeh{0!|!KCkCk6IGD=3Ve|Hf1>&SFf$`Cj7wi<C&@&5S3e)nMWnD7Ok@8{CV
z+)N}4ycB6+SE}v#=#oaH*=m@p+uZrhjJNzn2@Nf%x-EM753q4pd_Fw62Sln4Y;cNx
z5k!-Mp`YC-yx!{P@Bsg*=Fzh;1ia^-7nwQTNN%T427LAhCa1R2i_zedTU+T>#Q%#v
zPLS(kE|z2Mf%v6XYGcS@iP68))0`%v^&*DWjpz{1)HRa@ztLX+TB6J27^JEVp1pV_
z_aO#(6PaZ%+8A{h{-AP&zNG(&;J>aXaD{>F|2~}BM2rm0Tr5nOaXS5`)}$rSx!JZ}
zcv*YpaB+7ics*0XM;lt}{f#zhg^|{x0cN!!Df@W_MUxASyF?0Cxu*SQr=6i|jq~1V
zzFu^rFr;p4t(?|N9S9`Q3U;vDt`ow=!`3fjgjtqf82LFxrC(ZYdX4DeKIEtu$|2Fh
zFT+b&jdPh#L^a{XMG6Fq)ZVP~RIYdXWk$XU+uo0$jyts-ShPe@g4)`ce7<@|7YIaS
z)7CU*d16|=aCB6g3!Au?WN}?_{+U)H{m_!$gEy^_-zyji1j{!ltlyZ7fceD1@gP+a
zZ!aU?>#pvmK8&J(L4a;2qDQyN*Y+-4Qr+CDkrD_(<0osbLB?FK_;GnodS2Vn{8NnP
zz^C(!a=+`4N?aYYAZlCQiCz0%+rGy^FtgdDdwR9H;7(ta*9M<e_Ab7LoK~YgPZ0Y*
z!m~1;0JJ7yL{B%G(Ic1Oo_)FO%KiBLKgb$v%j>F8iqC+^aP<pPxn>H2!QyQm;U($6
zj9tW^e@d84-Ud?Xe>z96mLl@Jqm60tr?c|V3F9x-I<k<w`Mu%cKvu-WI-@AZ+VU#{
zcQ9==#K(>pUsn!@DT<FpOD-9>%YO!ccNYPS7UzvOiK#qj5TnzG>?I#>J{VRz*z09;
z=bLsb+hvD;malDJ@05U?(h^<w`MK~7T>^Gd-s)dH_nFw=Q+y&`$sX}RYu5Dh1jI0g
zYqxTS56ltcd++Q)_a3EfPsi^5Lfg7<5&gO(R}Y(Z_!Y1Tp7;;`nTv}i*a#}29t&Jj
zmv@TRv#}xc{I9An4~Mdg-)GE>H9IBy*s?`r9lH_PM#;VmvX$^=8DTPp>?K2Q(_*Vg
zN%jnlHOf9jV(f*GvF`@+ectcypYLz}IoF))dCqgrb3Nx==iK+_bKglwX-Ro>@wetj
z$<!v%5Mx~)I1p?$NNcz`l{R))2;WI=8<R{Vtn#`#=;kh~x05ZWqaHH2M@pXP@-G2Y
zwK0*yk0R7Sh^L-`=(bSp)R*{Sg=Q(PMrYYSQGUK+*<qXbj>DQuQ%-|;mQaxZFA~|P
zj0NE?!xh51E0L{BQ0L(9rjKeZN`khTeSJ8xrU0>SYpuN@9+wEiOJj#=FwyQxsUaA%
zG|F1F*mItdB5cnD<m+xp_=nQW%|N-e+1_>g_d0(EXQ05c<n@VI==9@<BK37ct9@p8
zR^F?>&(qHz457=!cBbLiMb~O&;j}x1imw25wt`OXkwIf@xz4-=RL2R|@Q@P!&;66}
z5fec2KefJ3#%r`h)w`*<l4`x3_qs-ZLhB)8?8RFX?R<|YYy9$R)G?Y&UBPkESnNf!
zp>BIibR_9Ng#w|vfY=#9?u}7td8%k`SoO+7gjYxsOW!8H<jXzaywE@q3WBw-Ibmz;
z?-oYB1?{S71ww{UV5RvRVjB=(T2D!$OZzh&QzJG=7DBb~ICr0Yr)dgggVoP`u?`Sn
zrvmY17VYae0NQHLoNNNGQKJ-mI%^OT;n#`)v^QU=M=6QFTv=)GBzz3*KR|Z+#R#&m
zGb<D{esCPBV?t!dLv^D8|Bd{}+}W9KR)4>A5fb17(zptI>bM+vf|pZX1_B;nTQnZ+
z4rAJL|LG_JQZGf9lk?f-cvs!I@Sx^4cW+0^#mut0--RvDEUMJa5v3k-l3}e+<;TPK
zX!6q&>Rx7`rJk<=a1|r40ujy18Bz7R0vE*2qvs;SKlsy(T6T)HnguJp!i~uKbS9d;
z_4p(m)@tY)h%WDX9l)(Mwq({LQI=P`YB}K%25xc!*Op{wh_VaZD%5#mecrcwm`olk
z*4`j>9WaD`;EdQf=6~-{2M8LWY~)|A)j|ioGi)P{`WW9Ufp|Wp=O_IY(_SnUMn9_U
zN=OaZ74yZIMccXPN3`&~mxwAgL?i!rm@>44oe@%+r}55?n|xX+HPgKJIEbV0z2o!a
z4MFMB+0vU;z%i}xag9X{sm2pD)bf^X{E7|JZ!&TA%$l6>?s&NlZomHTyQIi0al*MS
zs4O<l9aOMCCy`<4+t#4$6TrTn^7$a+2I;SfFJ&Z&#wzraY#K>|jgZTy9$)r}b@1?M
z3Fgs0n?EKYcHD7m-+v4lR!B;XpO58@m_Hl6R5~0Vc$g!bYY0OdG;~2l_qxR}oR6P`
z9dTF(o?Z63sV9^0@GNfZhrL*3UV%x~zB!g8R{rtIxQhlZW$*penp|!1W*c8{AY4n_
zh$O&hc!!-4!>Eja3p*6G$%nH}BF~}yCqusD)`ol-R+?@NwB`A5+Gp6g5Zz$%vY^=V
zVjHKdLegdkrL8<v#}xBw0dx)jc9<hSFob(1q#+R#6g?&&3Cmflz31v6GbR9sVR<6P
zCf+bjo8ZnbPn8l$rtYApLBm2E$l5F8k2DehsBufztSd#uM6t-krm8DmV_ekwgg2U+
zrVr~cADZ^4tGN%EGMqJD7xO2gKw-5srns{_pFzu`|CIcY@$?;kPJ#8V{gDcjIF$ma
z6<MWINJZvcTOT{w;2*3U+rq7|oMn$~bM_EwAG2;4UEkS>*BBX29&T7<(DBzd1tcrJ
zuEgF&GfUtDN11`mZ0ve<$FE>}R1}^m+Pw)K>yry89ef79x?j8pH&W25c+Px4CdN{X
zQ#W_@*Q>-dmKz-zH!2^7$Z^twS3D*}{Q)}VyQ$`9W%(--s8*JFE+K5xEI$wF3#v%)
zUdS&-@lh5g+t!KE#O8p6ay_5IpbQ>Zy_D?6m50yt8^vEVOQ}hmhyPAicLvY54O+!3
zT0nNRcv71mUT*8v9Ks!!;`tCT^UbWYs+J5Mclh7^st1>GQBV@*<~Uar9&+i8qz#Ls
z#eM3uD@;^+C_)wDQG^|j)PbGRYa6z44joobg(JB{XUss($HV5NYPX_zF25tl-oO(C
zh*!^|Na(`oP3TfIpc5gk{^gDeo<k8)4T0IIYsUPn!Vx2{EMerUjT3V>jerGnWAZw#
z?}NPl1B2n&C(mys_+1`;6<OFH*t_zaU4O-WHIo;{wVKqpj9s2*5)!S^Cgr@afIe~t
zD>b<=brIziQ!HL9zGK}<bdk%IsWI7KZgNjZT@2%*mMy$H^49@`eX5PKXBa>2olF{>
z+Z*P2z6Z!Ip2sDHEpobk8eQtuc7PRTC<~GP8Ic^yOXd{Y5r78fDJpeAd2yI3fs83D
z;3%N+$)`jB%loQ@?%Dm_&EiRS<8{X8^}K#P=|*d#mV+c<i%?cxD|uR@7OlEAd)Cd|
zwINZaYU@F`w?OEAJtZn4S$W3sYSvV3;Q3LR!j_NMCBId_+pp=WhPMMwScq~<9d?-X
z2ZfK{-8=&gwRr9s5<euLAc$^R0&jP{qjc)Qjf%8nEFLG)6*)8A<^ls9E$snk8L4b9
zhhWN!JvK`I^d6^>+ViM$-0I`rFc)qW%CGpi%A4^b{i5n!rJ}^CSo;T!D7a46C_K9B
zA*uW0Mw~Q$sX{qx6SniG$f?b;kFyjW5+<gXXgz|^>ObS?6T&?(4pI1RmaI?N<n01c
z{1UK$TDXfhNonKHYPF<k*r*q>xEZXs`%;#iVJEs!&NaWkHZGy%GfD=NkECMBp72il
z{Ps=O{7)?N5Ro0Tf-JAl0(j;5E+)B;sZ%Q!57hjS&kBQc89d7O)5Idu!+Ra#TSfM)
zTG?YrQDZTwsrpg~@Smd8tjK%W%zzNKCw4!auM(HVwbd2!z~Nu9jpJNIG`f1QZTN!Q
z*_}LlkaG;o%RzkEVIzXAZ%IIYmZV>tP_g|4a+Jy991P=*G8n-Xtp8w%>3Xm6#qAwe
z*EHbiwVWG6ot9PUXsnxauauI9eJ`g0CPri{DkTXv^7NE2X5fRr6)-Eh*tE)r1>h-}
z7c4te9-MPq@LykZ?M(Ks1{!7(BuC#d!DlQxrFynoXAJlfMOAokE@xqn^XTZ8Hz-3}
zn(wQTviLg%YOwxHg><C}c7^Iy0O*@A`56{fCnAuBM!5P)Jg@7I2r1yb_u5x9-3U;`
zyBUxPq?!>~elTYk+04tJ@1J00*w;#IV|(hls*%s<@@fF~n(|Y`Gxa2*F$ELGvt%h{
zqKLTt?`k;r)jCln(9`;^+FfAEPl{HEayW74<O#`TufAJv&^J%Dh99YDd>^i{ZOfcT
z6%ViZN?oTO<XY71YT>=zAOLcP5l6ii{l$HCFFGbozrnF+j7J#nk99K1XpOjp_x?}*
zffixVd<D+zQS?P#%4cO>PMb?EX7Eqlx?H^<ki(tTa4&6@Vqh#h68E}I4lebV_dftO
z8o9{1STkz8kykz*HQ`y`kx9tq)}tz|q9&#Vah|ELiM=!V6VlUh6OtyLC0$igRKl2G
zuS^}mrZcHV$`lI>_orEUp0PBDhY{boCzz%LzY=X?Td1El2CUtKrVG3zK41g2D+vlG
zm<a}Q>Z($;lONtrbxf5W%r~K+`G4_dIlR(^!{>Mhbxl{%eGXt<TDsAn^|bQ#y!5f3
zKV<4k5WWeS(x4chMo!h27E|#92Z81)X6eDWG;0#p9Js&<7)qD`?%Y^f8wLN{ipp^n
zAJU0LKAc)3!PHhC4y;_LQa3<ai`mvZE1XEO;6fyej_MI@-YC!bBzr<(>`G0~Boz-k
z+$>m&CR~)|eu5X}_iDLutVzry4<H!wfsbgN7ScDN>7SHhN|ZuELA23z>Cshy*lN$m
zf?=SVo{LpLq^fGuU*5_CzOQ}=wi6AMSCX{8$Wb3dh543h&~nj6yq<;k;4mxSP9DD?
z6~Z(4(_R~Wkt~Bgvv&2;7f#zM%4Q7yrjvWY69pjO(PraHQLXV>5yohaw@Mw0RLD|R
zs$RuZNkuYKXKy5~vI<7u39V=8YXTy}$||j9FZXi3bxK22YSdWOIqmC4{z~)Ly3^9a
z0&uFp8>aa_kJ+Pj$5DpyMLW9ME8(dT!9=`JmfwEJkl<$FWB*{tJ@%@0&#t@iGcuj5
zu11-I!WQ%K4tY7utC}-{vPS0*xQeoe)h(n=!^nlNSO!abRGqFB;pDV~MzQU>UCh*P
z6AcL#M4G+3PbU`f*5i9@#G8NA7TC5m04OzBrFyoHmofC(-L^~?0G($;v}T0EeFp;y
zJ+Blc^xtq#6u3T_kmoQ7MK*zT?p3GD<yC;QpUQwq=H)ED0jYYkiAvsM0v0fKM;}(<
z#zG>MY5JD6SML*N>NE!os|q=p3^}_o2`eYox4m>K8gB#oSz*;PJ8|KCB3E${<g_$f
zs_P)T#zRVZ$f*5nm7=??ypS*EmFHZ^k4~~oJ(vs_g`j`P6#hbiLP==SFp*`)$zVhy
zmbU_kN^r8>?IvGn1nZTIn9AM4vknv<ow@e)yjNyC%e-ouuQnEY`vO{^iRBhi<8qdd
z28Z_9+hgb(r@WPy5Up^Rc1D!zBXa(kyCEqY_jF-9Law74nyE}bEl%^!a|I@9IE=Mv
z=yR22fv2p^-sXv32y3;H@QM1PqhSs=-Qul8f4Fh6K7m8s<YHC$GejY+w(E6d-s3#s
zxU~%}=TI3+<T?y1TI8pki2la~-Bi;B75Dx3HtUSNV(x@ihe?q!+2kX;ZT}*tK4Z2d
z`k>z5=`gpGnZ!xu-MDOEknoj$;JcWd(ONWWWt5p7k5&W?Eo^d?9mpB&Yw*0=02uu|
zOwrvo8tdThsor;Ek-0QG2ia_6r?}rtpQ~AxLXmgS#B*3YV*?nO07RlslR1GPZtg|2
zK2FXuCjjBKHWnSmkZ~Oh6G4_o7z*94GONp@MdtsOm|2i#y!@nufADtx^|sfTp4yTT
zAeTQBd8W3sNcPLV#YU^cC}+tZd!R299$%aP{sA<3ERu5G>nyd!u;!A_oLKZLXT9kc
zz4yA`9Gi;MD3IiP8~uz=9&|Ee%PzJ<6fCSr0r(G65}T%dPFcc!VzV2MbCYrmi3|jO
zPXJ=SmUy3lH|bd(FpykK1{3;b278g8738voi~nA(Bw#u@{PgNK;w0757$34Y=n0Ag
z&}}|9*@0WV?0VK8?Zk>$mf(a|HhEwA<rmln88u@CqNPswM`%~x9RdIP)GV%I5?d?@
zzU7u~q*P1tcG68#8npZDnQ&5PMqTcKspcWUG5qOSWq;!a^DJ2_7XtTt%g^cSBebnz
zE}*~Fb}^t_M|<C{Kj&Wl8`yyU+o)vH{3u_<S7_&nM2IX1{j?c!c|HHdl+}$Q(|l|x
zOYS-SoJK)L{bs@)C$3a}5}Oqm@G2FFnGo1IG3%O}Twywjtl)n`jWj)d$8s>2vJC)~
z7vl}@PQDV<++^I+O+o$>+^&+9)*hP5lxgI+%fho&f5oB_edSpeW4Q+##YXCST^h?l
ztocr1bVj_2*W@KNBi_J7sgOZs^DS&)@WUM=AV|)!f@R(|6LGB`@AY(2h+_Qbcu@0`
z%wd-hrUM4>SXGK6W-s*Z?Z8Kc8FqrlUF^TZrNl|u+e<%^0}n4e-!sAW#Nv$yJ{v^|
z;QdCE@~^$J?~$3EIH!Tr3jG0fx$y>FDSc%MZT9>5MU#vFp_h%tya)+eH1>$|^!u`0
zN5B7t4Z*kM7@AO@vzJ5v*oPBDF;`;f3GYEZ0^M!R<Tzrj%5>G&*nrw+H!PEFlx)2k
z3=r8<+V7nfWM*6PqvNB(1eeGP*{f@!Fs?DRnurZZ-(QI+zn=^Z*X@rO@vM$ZVK=6#
zMEuAqS`sg6F$|)&qSHmov2JV(zBqx;35n09*%--6Je7-rk$3pIXCn!5pc0DAUJM_u
z?5V)hsrHq!r8d6CcNHXy1U^$REQ-*|XCpAbWXV?wcXZ|MM%|hiIWi9e>6O5;?$%Tj
z<u0ZpNan9tZnUP`)nfcG6VJm`9g{hiVH4%?`X97Y)j#q|@`eA41v@S&a5Xi6JWLRR
z^PLqhj3%5v;Q&(;&nX9H3;L?X45b|BPxMFu@*}m0L}9eUwoIWt)vd!Eyf?7Nsc@lA
zx}b@7ebD<r3?(`G4~_AtZBN+tt*94ZL*u1B^{Zjyj|szvqs;Q+^)znf8Qg0Ey!ANy
zEGX){{{1VOe3RH)eY41pOr?Ml==mO7CbfUjy6FAP@J@Y-T+SNbu>*WAyd5~)tN991
zx8nP2*{xgogwTdQJWHF=tR2Epk+bVfQr=#nqX2{B0e7EJ&XxrNsyj8_#p`DtU+gsB
z$e8+KN9z|wIv8~3i0H?CY4|6Bxgor$ndcVwR|7`>2Y|g^YR`;;OkLUMuPS?`y47K`
zvA-X@8%blYUNtjn&U@nRQNY2+Heh4*cHl=yMMn;oK2fc$R1l|IX|LIF$#!>VQo%*3
zgtH5+hf-gpuM8}08=kh)LT6fzwo0%kJG-dG_Q$BW?bAhMpVN8^eP?eZG_Imx2H8a4
zqam?k*ih_oPYZTCg{xrtSov_RWWD7}X!D^ECXW7hinP5v!i8$<ebh|f+npE?AAM=q
zcubpGN8LY~tHACPW)6H$cXnSI9uLqD+zcBJH+x?q=_kV}1qP?q^sVKQPQY?`_a!Ry
zOG!by>Uzt4+8hBJM>|~^Fr@FCM4=3K=RRPO*c0KM;l~y9(U$u$(Y7rH`7IdQ0y1>E
zrD0&P0CjTES%KW;DrnZB(-u=s+DXWnlYg*_`xORp7}}QN4E;NOjkrF2x`{*`hwiqZ
zcCZyd@3wpC7<%UhJG+X~ai8bu;t1D1d1s>Kcy6>Hslt%<awY?X6~vyn935e6c2)+q
zMhg~DSjXnmvdMyz<qGi`Ow9gXry<ZXeRF`eZHT$Izt)bLZaful*`n{fDG>j%!bRWu
zWkp}^O*#B^(uoW;z@Cm04KasT>Gxh0(0BgM2{-LJ88*>gf8GDJDBgIwz}#|344pZq
z5oS*I`d5gkINHkr@tNj0EdB4yE)*;DY_m7CC3bg|%kcgoX{L#Ky4Bf2X9va-UHrJE
zv8#o)(F;AGmzxnz|D+II%Vth?Pxi&9@855sFDy;~z1=3PPh&8<%OeEfT~*WfzF>aZ
z&^M6d0Y{^~19W7!ZsyR&Oo#hlt@lTV{i!-}d_$Z=3?6`e;pq%&@7JOhsyVO<bJ|Il
zV<0eIfC><=jzf>nLL_4sFuw?vTTmOUO;EckNAh~&X-Xj}sAS6A{r8gym%`Diw({~h
z>EtoCI(5x&r!*1t-%NO#xk-%Y4KYLNlcpeX4qe8>r^avdzL4vUTUV^C*t$%fl;`m2
zDz{3CeCW5KB$>*+;oxa9Fb}Tzq53{w7{JEJLZRznVY&hu4gD0U_ip5_@;^wI*l*Ep
zdX=GF+2h6P#;xb(Fol=jU2_wv(+(me>&GbGG%>WJ#=eP2yj?YJ_W33?zf@*<^<`e6
z>-KSo;fjI#-1*?mKP^u)8mytrDvO(Zfeu54rT<oVUOV~U@e)2jm^leK+t4G(83p~)
zd&KM=s4OG)yJfp7fATtH=GA<&)lj#l$ysAF9!>tT3)4yS+KIug+3Jha)3t8!l7zvG
z%k(G%TifgctfSu+qrupCKdbnq?D*kNRUsO3kH1FhN?kQtc&HG>9A2GdL|$0FS}y!5
zh-<$&r1ecw79xk;L@RtNd5py}F)q<E>6)}V2Pz1Km4Db0XC_=cGr|`5ZR~+#e9wg&
zwE}PZy`*D{zb96^(5m^)*}QtdCIam-&lPh9qmm5SKhF!hxJ+tp$c?8kod{7GLNs4~
z_n&-Kg|#gpWTdRUdmp)4cA5LRu=1b}t6E8lpZhH?wjh;IsoQrSbuWy&EfpHv4sa`t
z@4a`XLJ-^c)p7d5g&f~#lJlg1&Pg6oM$F(&?4#@eN^N3t-t@iokEW_>H!)AEZ}rq9
zAUp2C&czDV>OWZzXjoyl<_(ZQ^R48RY%Vz5wRx>_Q@;1Qru@Nk?U!p9!@pc#ZqFQG
zuZXn-_JLoD3diE8FAU;Rw%8dM=yWCqJ1ZtIKLeCt0E03U6rs%A#B2Ld6a>OO@Ai6w
z_~{s`!ifE>RabL)8)Wu1%V){kjj1De;Dxo^7vHN{V}+E_$92L^6!Z(*d8yq4#P#Dp
z#p)CulON=A7ac`ssAI#GiIW%lKTzcgPtBgx;lH^!*YF|GQqwJS6SI2YAKdD%EARGm
zrpEodcull0=k676EBwzSTUdj40jX|;EhH~yt<?u>ppx##slrl+`_qGbaL3nObQ613
zxFqKOW|cwoZ@$K4vT&M2^S)Mp+1;^Cj7jvW-3!+!iE@It;Wgi_L&r7A^yq6_n9GMb
z>^A%k*Bl@N<|(0^Y8FX9EUKKJe$B1sOU3R@2g=|-D>dH|aF0;--%s2rE18Xc@nu;D
zZZ}|zox5s>|FMAjBqJbR3K=;UJalf}zyh24TB+1{8N88~4b}X?`Q4Haw;0OM!bEu-
zqZnDSmHI*ILWE?-(b+Sx<Z5v>lbH?;cu!%mndaL^WG`XyOR26mCDyvd`z|Yg+BJD%
z3KAluC8`HX(5{O*sB*8n#R#98-g%xbdimjXK5w1()!~l}owdrhhw`VB5#gAmqi^(w
zT+L|W3As^oo7Z}8_UfGXV)5%kca`+po<+iV@L{H1U|`2EMmB!tdr-o%)u0C{<xlQ*
zRN+ij=cZIfn2zEShgIaaK>4@gvT;m=4mxz4&CvXvkcK&gVJ!aNcg#r8g=_yFGMj*|
zW)k!nnT<h4PYCh<mFM3O+JTY@Qw$-9y@3c4L_y37pkMujt0173<0phrJ`6caDEhCq
zWsEQg)PjUJ34BaIDYZ**WMXy(MT-$C|NCDpP1ydgOjICff|=z+;<D_|-1@-4(4hrp
z;9>Z`Klw|9XfU%rsBMx^4`zlljjuupcUPc%1Xc*M0+_f3Wgs9opkf3E2s5|H|E{IL
z-42Wl4CnrDD-2CL1RR7J0Xp6#bV8VA{@)5j*n}_(fWZ3%7AWw9nnQv)l-Ynu`WQ-R
mJA?`l@_<Pp^1ssoXzqZCfr0z~YZ68$bVHe8Pzw+<Q2#$$w-<r{
delta 375660
zcmX`RV{k4^ur(Unwr$(Cy<^+f6WiXgZQHhO+fH_J-}9YQ_s>jE)l^T{RQKw&s`@So
z?;#eiei0f5TC@VZQw$tvU9M>xIu9D6DpjXx7b!Uq)f|8&2KH~7uZo^Oz;-R4!>cqG
zTpvTAfB>}sxX?fWh?_&VH~bmOB{z{Jy=)pq<upSHBYONV?y(#)e?<0&lR74X3~T;{
z$Th~<gHjgxJWY8h4>>wW<MTG1o<tgf#RW^_GlaH|r-*6)Dm92iSIq2mMKZDRED-UK
zWuO<ZGZ3Kv1KWw8WHw~~uxN$2M3#-;H=}@deEtYdPdsdwNK~_FcpkVVi5T(9A$kxF
zz9<x~XC9G$U|!wy+$pS<Skxh*AXL*<18dXK=O3ZCTG5%{z!9hf|C9&5@CqXXwLVHR
z#Il8M|EsAx$8SKh)qBt$26B);i{K)R`36FwJ3K(zC%ZX5v6)k9b8(4HwpdSO@dwB%
z{jm6^gYg)1^XeBS6+BYZC+4Zu>IEyEeC?sc0b=p(@!2otnV%w&{CRwx%rDP0uGh|n
z?&R_lw-`Ne=)C~j)JsM{ScahNShe^VUZruIB?y6#u;`S^%LAi&ts7hE=W@Y$5XU9E
zA{-!n^DzV=r;lS!M42GDX)@670`<hVc2vowd+FJ{C+;!<IsOlA!cUH8isy0*>$;=V
zaBf3Zle-ET@PLvxo;%fOfnD@7y3rIgdg^MOMN|`CIcf&?^(WC9u5_Caf-J)DFZ555
z5jf?1sMpe9r9!qPUD*+As@D8FF5m-0KMYW4-4K9zIi7&|fk*aP5=n3bmU!MK^NdB+
zPEyFZCCE4G#flop*o{98u~~W;li1AP3zyhWuJ<{;AK^6bGE7a123>X>58OguE<NWE
zW{#o-o=YE$FTgriHhwS5$`{72!1N8x@(qDaFodHyd|w#kj&Y8^34DAD=N+#LeFmuP
zn1HL2ypa-@?9|V!W$ty#uC2lf?UnA_b@nY1aD^Y?<4?C~(Prw-#`yW_i!kaMU5Y+w
zDvv9-B+!`*&(jL3W8QT7Zz1qY_U$>bl<axWv}q=H+Ruaz7d-n5h*6qv`fGuWY!mIx
zruQmCTJwhhd8$k5L{^nMOk6?93jki>8?4rQQZLn0zI|g6FjR5Zv`hM1I>(jnQW*#H
zh(r}FrTYV?)-O1^iG|5MuxYusOJ);|U`AdKkum3-i+*q^m#q$SW+TOA9N2rqdQoum
z%N!t9Rjvj|$Ys8D%FeX`FFu1`bSXU~2+Z9|i^VQDB@n=(ADFFPL)?{02W;AbxG{*P
z`Cq+YSuT&S(H`2sMP={X#Hj9(8JH*?!$jb&;`~hS8e(B=ylo@&-TaxG_TQo?5cG}_
z2wXBa2EF|Y$KRe#X$$9?O=K(A)+sz#Ys-zML^$|rH#HqpCuNCJfHb*FD20eztCbZa
z)bBmK1#0l=AzF2?5`B_!3xGxWXcW^Q$e3~>5nN@arIo05AUE!5Jc(GGh<KSxI2vN{
zRj5I@VH!scy6#}}F(x|V^2yD|$yTsKEq*a##{$LHEQaXQGM}Fbz@bCeA;}b#TeV<0
z554Ly^rU|x&TVk$8sD61;((wev2jRkVdioYufB2ip9ZoLWrZTo0}$30xXSI>PN{jP
z1-?ecJF`nrC^)j0!fcRfnwTC&wRhfW*S>+Otwl8WVb~_OfJ+AdfxJnSTy^KCK)V8Z
zzTrPr&O`p3M7Q@32l{e6FwDSn^P(BcA911)93gs=SK)@BuZ@`W3*OY`VtxJ#f1!y#
zWjI_KKaG$a?C5pn42aFHV+@)eiXm3&?`O&Nb;=+A^=YY(8+f#hbBac_jgv@`Z{!~R
z&&LcsrQt53a4hz|YjO)cWaCRmE;Ks4=>td}cCCrZoxBRy<LhE~Y-03Lw;r9L?KDN?
zA6F{c24=rC#1Lf><iXjHReY&0cT3&^JE>xRmK>;}$T7K7z#n|b8cNzjJ2<QCV>`8W
z&WWu*aaPNc2OPF*b7Fd6mXHr$(*c+4CKhSsCX90{ShHLMD>`ZNl699ip6U+VIxdS<
zpSe|S6fc7)QE7q49eiHKmIq{B#`Yzy=2`pI@id+4G4#5Lg|v+p-$>aRks=-0VIb)l
zX?txKS@m+`fNz4vdwS>B#%#S2s==LoTQ$$N!h4N2WC^YsUK0_40!R;@a$rEVJaeUE
zki4v?Sh|86D(c%1&VAYtOtPeF&TPEEf7MB2wRq8Qr5c&x+&rZAd=Q@CnMOeX7bAMO
z!SG<q&0-(q`6F;2egK1f)z8kwlt^veU)2no2$IGK@Kd5Mb4iHVKCw=TZEohHH-cLy
zqc`#%mtRDI2%Oo(QqTemeTZ*53&X{55Do_yv|TkS+AR5wGPL$W_xyvx<3#v78w12y
zJ%!39;VzHDCUIe(l+WC3LmZ?a*cdp|YQsH%7$3*BXZPq??keh%A?aJ|_5=WT*m{O{
zdG`+hv@i$RB(*3ews4MbVhy(Ntopke9Ce_x2yGZekM|`=TJyQ_dGxQ?FP#ipv_y{m
zU}+JsCibPDI>a1j?%%O!1n!anq7jGzbs>GCGaRDd_-ie{p^nss0vou`1EG}?+ZGI1
z&+v=N5X-H5OwBf4BxCaoWREu}++prDapPTp7zqB%P@Z|@#BlmUG6wA}7U}p;#qzMa
zkRIlB*bSeq>)4Sijj>&<#(L7`SR-@PX1xVweDW|lTXNHa))S)6<w?m9z(Ntk`z^5#
z!&ELi+<T^}qMG+og<VjyK;`pltFE>@VF~AbwLZq9Ko||f?<P=Et&e@Y(&JbFPars8
zMJNxu??n4KQ81f_BG4Sw8jD2URFZ0f#HCm|M%eMZb*mhYQSlBdFzmTk&wFgpowm>;
z&n<tv3`%1{>``g~PvOS5O&o3uKclhV<+~aAoqrk>j$muhLW}__Zi>-!;SB=Vq3~O-
zJ4@<Y7FAFJNC63;KBk1*9A!I3lava0A3cz9-?{1{z(Cc@W&RHyf|wgbcHnrjh$<1Q
zCb6>HN!0%u_)LoE>@u*1xQyZlVk{s$L5Ci?&TA><bhz6>Z&IY8wu}7pbLydF3T8Un
zc+`bdGdg}Q;l@A0PF!l6y1<%ir%?dFcG|y|>v7o7^pQF+&TL^Va7}FSNU#A4UyONI
zqv^+N6m%GSy>vk-B#S>_E}5wMvkBX=K1>zcXa9-#{4<aS3J(*ID5q1b#<J!Tt}e?6
zq|SMvc)uQkX*frs879<>;D=EA3#`}9*q*f%Ii7TeoD9J>JwkSab4zrN2(sA9<h2V~
zjvEY{*x4q#^9O+eZ|Y(qh>#y}`sKS;17&3qP|1%0#~02t=h9PzdC8sK7+*{&!If4h
z#d<>M6&m)HfaS(`z8HY@eMlO7<-1};L31$0$_;^|z~ZGr8=#+qW<z;|+UTgxfKgkz
z;PZI2XPiEqOq$3Hjm5DEtmGtBm96q_3ddMn2bAXw@3F7#JGfh)PW%N>)MfuZCVvph
zF#P_72iS5`g$ZJD3Xj_)%Z7X#xa)9m&fkfAA){3tv{HZ|?^=Klc}xPt=4ToI#U`tW
z=c+TZ+uPtIp(BXL_?gB>mz`Lb5S-<~wf^kP*lq5Gh$MaZ=Nj;Dv+1n0alTnqg(0*j
z=B<F9;w#I#LP>MGQDFioPbSq%w7UD9{<v<}!ki>meJ&nxw0gCH#G{W;&s)-ax$cw^
zyyV`3UV?uor%e!o$GcHouS<Rfsd)Ob-N2Ga**}L)UOY%w=%fg;|HCD@HU7peS(1S|
zSyo1J@$3~hy(~n|-jit}bSwXN`sXO|k1}tT$mX$7h)Jl}y-^@wlsMC67Bj14^5PNZ
zI)$6D?Dm?Qk#Cgp)YlV)p=>TF9F^HOO;7?Um(Hq``eA2&W2C;~jxZtMt`;F;CR<^i
zpNNj>5z>$)Q{TtPc`)5+uudz}#uu9k>R`}mNwXQnpDc6`9IZ{?|2pZL>9B^Yq<lTu
z6@eUH*!T!6JzEBlp1yJ_tu9peN|fy*DEG>B9uV^brMvcum>2z_@?$@WQi#5L{JMvt
zC$1x;lOYUC;s=51rt<1Lx?2$GUF7Vgyc$RmBivsMaegMh=D8A2F#9%fvoxrN!~j_q
zIvBYwc{PBjsR{nmhTxyUV;|+An86yk{q<bVn*WMzRM!kxJM-!54`qWyD8RSomX*<L
z@Yn58`r|ORVgW4Ni2Wku$iq<arnlw3myY|RX$AG?z~1hrl^U8B#Me^iZ;ea}j2Xff
zMY(g$<}0K7{S~pLG70|CZWKB1w_}tkX0#bLS55s2$xtExnGreHKLWBBb1;9*08DeE
z$H;9y-+xg6kosT{AV?5<TyjGX^8B+*xQ9Kx?-L^;nf_ZDp*@!`yx*$NwJ!>oegv3+
z3j_kYv@-u)XOLAR{oKvguR*z#DEu^-HpEk)ZT#zcC`X*}aO3_8Vn*ZOL=<;T0=!I4
z&#H=(e5~t6ta^nXZH$O_F&z&?D)25Ua2A(g5I#p5-8%GIvD9HFXsKOf6eN8S=<#~6
z0D%O^GRtLT4`HZ%EjEO=FT<*fBFy9!B0>O08OQK~Qpj#`S*VPkp75kiKmSZJ2Qjn^
zbmQ|w@by(B7Nt`)IQQnz6MP$eG*%#Vuy@XYn7un1;0CK!l77se6g2Peeivs_iWHD9
z@;3zwmH4gV1Q8b+%uR@h3~-Adr`QiP(19~iIWP*@i&j6M0$G|O9aSR4nMuOhs#~%P
zF%R%daxJj{@W136Vsa2et7J41B0%)9Vi;b2wwyj6EfiO@)~pn`0MRuzLkp!qYC>+h
zsmfS}mwv1?bK<OoTRwM0*wl+=oKdc>lY6*t@iOyz2D$)*ZicOlf~{M=Z0x8q-ayZM
zUZH$!|I2ICS0^zi@2QYB+nftZ!!|=cy3S%s*wp!>Ue5sMONU>tc$*!HIbadvk56y-
zopcHJxsES8S{0E23cnFG3@vd#%cPtZCuK|or0x*Md4IUxpXUdqyR;+%>F9TO=8Oi2
zM1C*LRRY<6K|_GarWjoor#H1&d*&H@b{=rHPM_XoR%c273$Q+al2l(}1FnR|#<ikA
zi=Nw?yKt775=s5ZM7$TiPe597<4ya(IYNtGxC5adLy_H@iPz(ZA8@3i0YFFtCw)ZX
zL&i-YIFhF-9dm5Nn!GwL-<~lXgZv+G>~ds*>s7q{{^hcCoPpyQUb2JPv_ta{1D{0R
zFy<0c-d{L5@d*KkU}ddwhhmZIjvFa_EK=8hC%Dhm!`5Z!tE|y=mH-f<^8bW3!dd~b
ztSmMG{#ej)w2dm5;P-Vq+jJ3FkwsFa43j6tj-UT~kJnP>kU#&+ca`@5C&$c6eqGU$
z3YWlSgxF>;)a!u-3@!!PQ3*XHk5LJely3>h6`~5V$^smXie_(UZ3`@3@uLhg8&-I#
z_fxSI^cg$TacFJLB7g`gOsKpkH|#uPMe~H+?9P~}DJ#_bp|bKD7%nhnG{8aYA-OMI
z^6PoQFISsiojrrnW3*Cq?5(xKf&$USoLHTkTHWSqQo5(%R>Xlfr}qY;1aG715iTlv
ze3GPJ*;!T41}jbxHdHk<>|lHS2!9q|Y=?k1cM0X#FZJ!uF2LC-x_ojB0tEt2{lnH?
zU7Tf%4M9XtfHZm(ud`dsxqLd9Byc)(x`f^(Fdg+VEeJm|e0nh;l?KqFQd0?fr6#Rj
z_ELmyhKQg~+%9#iI2612vq%fiK+|4sDz|QigvUU;AwlhGM=_V66z8Z@4~G+wgPIm+
z`Uo+h?B#p}0T_1lY6_jh3SlU4j^Qh%j~D&{Hf@NhW48}~O3R^iQ_Ni@vxMin3C0}7
zQz6>N(IoRJDn|M0OT2Q=kYI(w!xkDvIVb@zO;aNy6DXNrfg`+hCB7uQG@JG;7e7md
z!nj+G$BS0F=nnEB9;W&AQi3z2RBq3kWf;6^1LY1X0}=}r$uK8H5MC<^6rHYV#O}0}
z4w6yp$I9)%9qu#M-%8YC{rID5Q9FD^9x*<A7scSgZkn|#SH`SbRH)IK&@coe7~Zr0
zdi9-Uc^ghLgM7)eVrN4tjC8G+%q72an0>nbP(CGvs*edxZa(WI;#Gnoy$}<58Frb@
znI!1P189rudCP{IbLk12H~6gFxX)g(+_;{iQb>$K7f6u5ZCxd^yDe^#MWJr6m|6QU
zx3-=L;U@H`28|Y5E=sHo^%e}gAC=wnr16bpEQ(||baQt#L9B#+AJa*y7g>Y**@z($
zW#_?HAWgZoBR#6ePGkz4I%_wFYzpEUT-~Ec0iL-Q0Jw5~RMMK${GTNU_fJs%{%v54
zojpi=k&JmF{#7P^S-yumxndj2v63Vi<PnmjRm_6s0*&Fe*+AS&MWftJuM8QdWBGL9
z!D%849loZevJrD(FID9f8?Hues8rk}=!<yHn>YTq)Hi{Qoj)fOu58hm@vf3r{<c&`
zfc`-V1$@Gh5d*U;|BM4JyVXK^L0qPV0-yghZ`Ws_nfZD3fQN_3z-3pjLtrN{a9o_A
z5p|d!*dzAi28eIkwONNQU7Eu3#YyvHhm_|=nC)XWN(TL-Anekk<j1Rc`iHr**7-mv
zUu+s_xqXI51MK&>ZC+$}*b0kZ8zqepVEW-%d+v(MI6a*XJ7CsPs(F(pt_WpWsw7Iu
zDp&?0*o9x2_q(*$Q<?0zOz*VJlo57jQe2eSAXaO@m5d&oH^uSxMC+0M?kOak5g$x%
zt?9?~)PXfNaYd#^+kuL#8<6Z}G5Yy@-l@oc$-e`+4Ez2@|D>vLJP`u2kmoE25a_~O
z_Jqko_iv>>*YsD}R?=gdZ69^$iybKpQHU-m%#(7ZiSiPEM4g<Os;}0ZK4Qi+kM*h}
zk!sMeq%qqd_|zeQd%_`%VOcp76aU~yF%0nl29GClW6uvRKPB8d7^#Qn&p|l!uToa-
zUmq(>jIw*mt~V_a_B#!RrH`2hWIe|O1q5BB+KoY-H>=gE!&EY|m@0NAjm!!RL9-KB
z3+J@eeg#QVmphfpCT|`XX?%uA_NajOJmLyb#RRaB6^{Alxno(P{!-E@gVBOwky+5D
zr^g;0O~I}N67aMSs=U8T$VqoH*T^^nvT)Muz}XD<U;9J+23*1+R*tm*ox%BnP{p0>
z0gg~qNmN&pH#edWe6<6yl>uky6d_+sMr1xEJ-su9A&Ga40Y=JW72C*qG4ufw@zLN+
z8W<>;{V`GD(^<8(o@Ykkkym=H!Z;onSc=y`SQJY<5ki8ekcTO98jW$?9A&;44(T;<
zLvLdDH4jO?cdYo8pXHZ;GDpdic3pn;iS}tXjAAixN1YIq@RzPo|4AzbxiY%uwUAvL
z?@q*-7!9BjF>=IPSO)N-qGQzu*g~L7QqpC%P-$wi`Qk>pXz9l<-LJTR+k?YVq3LKS
z^LZ0`O)rifP1-4Ly3|+ykq*6&Nl+Tx5vm4y1F{+5cu)o{3t1XKX!^^cRYNa|GGnSW
zrIaNudGolpyM#oX-=N0v1NgjG^d$f_8QfQMV0z@AiW1Q>N<=^pgvFw|ADj<agkDOx
z<(q5Td>?5`m*N3k=GFBb?AVEThp!W}JVKI32wq6K$}tf89=&Dn@7uv=t(SD~3huAQ
z?{;?4^EuXj5_~OSGE~Bk2A7Tj+}m=LMjCjJhH{3AYw0EDLKBfQcJ<|EhQFh;>t7BV
zfzAw}LEWF4x*0+b!NM%rMWIe=FHI}M%XcbLPMlHYX>m8pS=GOnnZNqK|F+h=T(Q~C
zsiU8g8k#9vmS9hEgbup7<QS-OBHyAcPW&FL2D|xq>=s}FLWvSFz9^F4kYew2Vbr*Y
zg>v1@bsh(>_*2u-l!}C!9hoDIG6f)?6}ss&(cyD>bq)eY-rvebqoA!Z^V~k6w?c<|
zJsMBF;5Ky;2gsw#ii39n-#k9#ytViSj=k!5Q<li87z-)(;x><HB{FtcL)IMy36~IA
z?v|lA#Kt86YtA7R`#e|JL+V>7f0AA`-e)nsM1NcED!HI@7H7**<d{dVgQPMrT~(ET
z2WqNdUCJ`dA#&*BVbG~6mp??t*rIgy50dqb04VgfeR6M)|6riP1kwbRLj}JcKp>tK
z8dqN%Lw$5*1XQve)<!%rD^{H$<s_b`H$^t>MG*$D*-GQiWwa$INOUP#rz;>MC$qYI
zFDKI<D&#%X#T5q?H(9l=d@NXZ+=)4a+)SfP@|uECBlt2g6ei&e?_tO-7DD$@gsX#z
zzO2NBYx}n!C+wtV?tf%+96ENM{BG=M-|{xDIQ;lD+poY#NM-SSp<aRH4N*0J^Q;SY
z`D_RP{1Rs6{Ka*hQ?ik$g;L_)GuP;|(NBvyaZ?xlV`gxUHahNK6>0&-qNlOH9&SC^
zVZ@OX`Q0fq8&e!y4#QQvp5$N9^HV2Akam<kv7Ak<0JTy2pr!6`s^p{)HTJT3!iG-i
zMQLJmQ-wo;g=(gj9q{sLznia2Czj_RsAU9jplBK+Qca-9^6+T$IR2!_&9-LYTsQyf
z(rL`zy<1H`<?D%IuxeE}52viy`(GAD(kHmxHZGpLEPW*VuXfjx{chYKD;5LuX^Zx;
z#p|i#B@}H=>Q+eM?k_G5TJW^{rBHUld@ti+*0`4GlGa`3%b~QHp3&mf`A+yDpt-S5
zi^^g}7rEMh|MiWe=CJE&=Y#DPa*c}>9z78>>8HU@0K8w1#<}&c1%y~SS#phwj%~l*
zzQwwn76NinC*oLHLL6!o#gKHR6-pjIYy9Q{^i_HBM474QM9>rMSL?)%06Wy|uS?VY
zVS&=>!AS7kMTR0iA-i{CCqua};C`k(;^5%T+e{zd3W-6mktXK|q3VrK5kn{^4@A)>
zbt+Q`Ih}tcjlZgAt;v_FqZ!f(*(ImGZvH!l^4i9>{Py&$pu{U~_P@|f^mNq~GB;Xw
ze0#C3tR~IXA(6^F=-pkJie_)kKRcX9&06PPcU@RyQ+FQCZVuZkmBKv$XTyfS0mp@`
zK(h^w5FB~WB!e4Q%Rh2kWt-2FCu()58&3z?IT5%2RQf16+m0K()B*3CB!C)#pJmeL
zj9H4|d+zqGE!Co3YR<P-WzRM3nP*6_Bp2Z+M`)bm{CdNNxX2zU9-AG^5=BO*ru}Yr
zrTQi2U^=ikNk<Ubs3bXnn;E->{vB<6Lf%h$K&Y#&Q#NqLe=M-iU6&?Z#wO{~?Ky(S
zVX#~*l-S^U?CotraN@i(;sv{l0@W@~FT^*_B;9m=XkA>2H@%%rpcp20SKnHP<Get>
zF{zL%J|HKF*#_=YS-<oFcj#IWb#QRZ?OMl_E3>(!oRr(s`1KaRWM>}QkJ>~|`O_Q_
zyA6!ZTaGomQXz4>Ah$Z#>gmD@18(`_Q}jXK%n~HbECaq#zIRSbbS(R?PQI2>R@A!D
zH1Tj>;qKn!Y-{IcWbO3sW!23D>W3e&f~HL15+meFy-li@Mtc>zqSr9Zh&|@)tIaK-
zQIR*QJDhu6?ED81d=MrC;QKh<1{NAoKp0pCmFobPWP1@pwWW+GA<&X3!Yr(S%rvYK
zqf6cLE88?fmgPS^K{H_$se)`SQUwuHLlp}j<Bce1utqk}M3&?4*X>6&aN~RX&1-1)
z(?P65xLJ^o)>$X-7InZM_J*vF>#j$-#s~W2y+*vu3*-hkW{)(y16p6i#VA-Mxa9V(
zR4Oy)7ARu619FgedzCo$0QzW06$I&p4HyezMY_+KmlmIgH|_z|DuYzDf_(e<`OPbe
zzjURf;j3qIoE$+nv;y~netI1LttlI*LeZF>YU&)pdU`x(?bDFUn>yVco2EPZzx<cM
zgE<52=Is)XtE5Q$Sc7=<W`^)^<$0J0P!@}CO!)af-M`S%Q1H-FX!vCOh-BX;0g}%B
zJ}b1ZMVI0<jN43J-`-ae6FJ=3n~#ke|Jp)rMuEX4&looTEf#C;WxLBNN+E6fbXQ@(
z#5D{3eBD`qseA*GLC(8zx2PCb2D6(w(~bs+2b1V--OTzT=SD21DQD7JqbhH<3Op!~
zQGhhB+{*3Evt4p>w$VE;kQpKlP6x8M?E8AXpk1!8Oqiq$Ata9xrTx8`r+FMEirUy<
zI!0&?XB<J5U=>SOgf2REMYI0wzMx#b&^gSvk+dkaFA3Y@uM*348}!<AVkN0Y9b^W?
z1A}SM8cHe<LRITuqP^c(##2j}R>0s%m`=!IC6fQP<wDfSXfc!6gt}A3j4zt|*JKF%
zcT)fV{}3yS|M`yK+?Ivq*|Rl02_Zp3YwJhaDOnL$lKd3`B_ACJ&&_`ssK;lPJ5
z`RbtcD+nzDErJK<(o&}epleB?g*skk&$Ms-Kb@i|;aoX1iF+`mtX96PWztOE@ZHHs
zDC=4x)QBfb8X6k)V2limE?)c*_T~kU<8`~%O^2S!qfp>WVjI@7w&1PoY7SSLy--5H
z4szP#B2kQ^hT5ZCevvf;s!X#o90vR(R9#TuxP^Gz?mxw6?q3AJcRyhMB<qd9CSejC
z)(z$;zNUY36aU5lpH18&ycM_A%h@*A^ZnT*i)DQ2@-ABGCY`C1e@OhCA@At01HNW?
zm7{Ey?3W;*PV5WA;BwfwFs1Lw=)|g<-!MIjX6DJtZ<+|lqzcWJpLeX^Ah&n2KxIQA
z>4tLP*pEFKhcf}hzlwJ0yJ~B&5OiUsb>nx@SQmw@L5YzJ780*cH6Ti)vNIE!bA(un
z4~DEFG`o|L7w(c6WW9W0bo5vyZP6}%E|L+?ImY5-gwpN@$s$12j|Jw(f(}BeI}bBK
z&&{Vf{?=@?_jcOrC-q$UJ_R)25@a4n0gSBAU0x|?GPD8y1LzpR_>RBOCP?WR+W_Nw
z(Xca2dnKgZP!1|)%G3sanQx7D#Se>Xm1fyUgJho+074sOzQDc&8SP*&F7kFnb*O~l
z5wlqz^3((N4I>fXi4dMFOcc`PDsYFwwVp&*($&{O*l`!@3+V%-<Q%emAR+f*`2>Xc
zooUmt%>p2(qK_Rx;q{E{*5&zFXZvhl0P<|@Q<G`?>>Tb+?$_pCCat&U;nH}28+X*|
zxDa0Xyn_F=;Z3)N+y%or)VipX;8b@iCVw_--nL?pVe78UKpidnpWkEhwq6x1!Y24G
zVcARq{U-Q4Lwy=GLe>K^q@w-i$OQ?Z@WqIu&lBK9kVDF%{b{#ZifY8X{b@9ZbGGiK
z4!Qkl6-U2!BF+kknjkvbTljF6m~=3R7jm1sI+SD49ib>H=nc?i&)&<%D1F6ooN`rl
zOkB)mj>j8by*E)b_%JZ}gE!z8mb)GP1-EUn|E=>I74!vHb0F`{G6_GjSkcNbHF6|l
z#106m?{IU)zaUIS4$zkD9zxZB#GLw=;)QIXOi!IU`t-)=Z;|>jtq~s)p5kXEb?m`L
zA3ztkK3D2jZ?E6nZ^mi8bREZj&al9qhaS3lv3sS_M{p^;2;J*$=|DpnCxGXZ4UpiE
z?;U_mR4cn#APDwoJ%PM;$O=<s!3^MlX9TR%!UZ{W)bmaR5`#R<^6dWnxm!U0ULxRv
zXji$CXt)d-62|7b)@DGQ_G;x@yI`iwumY#gO=FnSgaID@TD^F-<D{CoUbE_0uBO_k
zBa8=~Qh=0S@G7wvm`4*u{<JlB;|0ZKpK+A-Sc~C-*rCZ&qdCds*~#Q?5;3^AHw9pB
z7q6gCIEgNX2!*{CzC2T1Pf5g_Wer%G@WV7^J6>#oe@7Kh@ktQ;6^8t|Hf^&h$z*qZ
z-?p1-+Trgo**S{2{1Z1Yu19V#j)~&OuhYZ}W&W*iA_^?&3MuT70G=Z9a*+2NLumHg
z>5$a7+9$v^*UPIT{OK3BE0*rcECHZ~&)`e_nEh*xnV+RmB>O=PxjsXh5*3V@TVoZW
z=ebX@7b0nYkaa#U-<xXP=u02xemu2s5a)|H#`~Dcx1U%yf*kM5&2tQm%k*8zCj-@o
zM?i5M;5g$$Roq@bGD~P4T)7WC{B{m@n3JogCX)`|;>{rmNLo_tP@quIZUzK7VD{tQ
z9pYs5R#KP%XL*Igj#Ai~zB-AweW6zhYV}LxeY6-)8f@R_D?~^XZCfrc5Bs)gDdMNw
zw&~jrZgqZ;myI!3m92h@^=pc;2F#BcV3$9?JZNH+Cvx`bClr@gUml{x<9n?I8gyRW
zmEB6AEz#7se{~IknKb^`4FY8Dw^t5tZSLPPOthA>A6*vPl8l+^{zPkx%TJs*Z+$Ly
zlh!c4WaZq`uh(AJPSK0`?5A0fu%9t@T2^piO5PH4VLP!Lnr>S=4C;Mm|7hXB*6P<I
z;2$@O+w!#d&j|;K_oPjEOa)u^L3Z`7nNad4!n<~BZrlx`<VxJ(rvi$4SZ#=kwialt
zY>VtA3+Bu!bZR4)<BpQy|EZ)XNiri^wnn&J{g^}5iQH2xb%&9Yq>Vpy4t3YSt>0<f
zz%6}WUEHT8(JkZMmGk&HLmHh*O`}&5cvIW^8u=^DnC~@jmwxXlzse1{{|lsKeEn-Q
zVMSxSNGExNm+~;3{0+GA16Rd1_d$qLFN>QrH!{1VdlN1!#%9t=#i6hLYh9jUl_3Y0
z97*5bb9vO4yfN%fRFjxXUC}&dC&>bFA^7_&M;rEp=%iK@U&=&8YRmNxX^j4vnXg57
z+wq4uZYu+@piE5?&Whx;Ekbxuk1ISh5zb{_5Ko#87lKembO>k<6&MU83JOa`{GS^o
zPsD>^OaIRljW-yEKO;PBk3CWB+#t2*`u%^NW9<INvl-HuSO?GeysN=qFppS#9c9V-
z_2um$@c(ol%U8ScYC3unCPvnxg@27Falf%D;fVKTyJ^=;K>OrKsrhp1XXLP|t{eui
zOd+7ezgN>Olff_XsDrzHFbap~y!IkjOA`F5ykF}=SJLb@Lcb<|NQ7n!eYd3?O?I6u
zF_2~RKiIQ^rC2ReiZM`Wq?Qn;c}`uvz#r3r=&UBxnj-Si>(_ypip7UhTW`M(jw#cr
zDrT-Ws~?$2!^5mLW5PF)bMNbbAno29$^ulI&AZ5#`jH3Yhkx%xK@frQD$N;oyIek#
zR@Z0gj0E&Ss~##&5!YdsZ?150#hbl5AmWi`AlufWnySKZQH5Y*&?(7nkHTO;jX+MJ
zn|2}wB%0S7Hh{2Im{%4AkYdpG8-qWw)_tCvR0O5milw8eYd5F`uJU@1IKjnou>e%q
z*p5ORNJQg&VN3BotWBpA?~FQx1LwtwiZ>)laIyW7^z`OO_8rMzy(e?vYoIh;i?mO|
zK!S%_{u=ywEA8LECtFn=Bj2il99Bm_$eVS(Kt37_+%HqyL^svM=r0)Sw-saJ!y`xn
zrbko6`_CF0dVxXAdx8_E-Dl$R1#}g55zrgI6%H{gsPak-a?MP;0w1xZ<MLurnW_1*
zG1VoFwjz~j_m;0%7e!kJs@Vz;3-!hzU#%))AnAwh%a2-3mV?JEFXKM{P0SQQ8}{c*
z3Nsaxb!Em_dR*eAWGO2dv}$RXEwLR*E<9@Jt#K5K9+qNN>i5J4n?CBK0B9ccolz#E
z&@QE;-QbcAqva}Mk+!A^>_^`C95rLZgDcvmoBG8de?h%f#{Zh8X#F1@I^J*1%**m%
zQLzms>*4?6{dA=bpv~OY<o>r%_9yG3WifT`6!kieo8l?sBlgJue>oP1c1@t9=S$cd
z|CJ4ve(&1b_76bdgA$s)_P|GuJmo|HNFsBfF#Ipnjo2N+vUxk&jhH3nCfAF#1^FUO
zM|7!{J`|<{_Kk<n(|=vNeLcU!W~rREW%<87m9I3e1Ga4>OAq1)6%D&~E+Upe3Yre4
z)2HwT=P$($Jbc#Y56f;Ct)hOxdzjs2I~6hf<b@aY6<|{l>{=7|dWIq<$=Kgor%{x(
z+<9&qO1FHW09(a-S@mL}ze$W6{nG0hUO$5^|DzSMMD@K3$aR0)|I%M*NRCEz?Rg1e
z?=V~8&l2Vyab8LluFjHXt5a0X#ReqyQk9`MIu>otQWX@s<yt~kj)6F&ZkfmT>A}@n
z2MbtLB;b|Xd{p_`QjF$h&joT5wCx)w$OZZAkiwglraB;58I-Fp+{gJZyTSG3$<#ux
zZZ9UUORN_aq}BNFUs=9Lbi7_UiS^#C8^^ESG<%0**F`5#V{YK9djo${2>f?BI0a*}
zI1NWJWxBMgld5xt)@f{7R3@zQdZ&#%=q>K^F@WknaSM#xr;655TdsUy#0O-B5Epx{
zWC83#WA;u{U#z!^qrwm-F*A-rm7orJNX7iZF8!E?ZG^cS!)lcjFXms54=5-S<YsVw
zeG-n9N7mi1>o*$IiC(7z9ma92s0Nu_$)XwaQ*{-~dCTA~-w@P;Z=!7JBKgh5608$2
z2Y}yc>lknHKlh@Z>myYK<;0UGC7GJ~jK8-+{1>79pz=^)^r?3U$tqQNIy4I^&?Y4=
zAp@=@k{;*tz17saXM4i1hdkh!=@zu_<#zdA+$i{Lj745v8(yhsIBbj-Lj;lB)toSl
z6jvgi$h$AK=SES^IOMmR$=?3CpC3%d6o9$A4#sITt4S9llBZvzJDY4&%{af`ROsF;
zSaiZPL|ofEojw$zFQtYD{g}B9uPpj>I*32o#K_W~Ym5`WKc}J~&}4zyV3`GNG~8NH
zN{YXqF5aE6u%Kj1g5WbVF0!H|M`TCnaHBo|TbQpNYyqk%6p<kpOHTsf;eAeYazN$r
zNrm$ieJkTROar)V>%^E;o4jgPdxpY}_yuK9lbn`m5xM5oxG5vYFAoZjyC}h3*$R{E
zb(&-^5j4=XUS7`>=|qwm;hM{fcZ&}1l*F8w2$A~L>oBWr^e4aB=DS&<h*O02cq34i
zv#eZq4*&FZ`gc8Awp#*Ru&Gup0RSz@s7p2sJpFiWmc$&$j3G)-#(?I$cdM&cd0}S9
zt*vG|e=ju)5*<B#Df5br9u0D+fQ%lh`g}UXmdA7{)TNaut5^OCZo}2;7`E}Ef7flp
z2Ib$z*eVjQ=lkCXx2b_m&es9A=o`A4YE%&RA(@)mKkfgOp0n&wwrUbHBtRuBfn|Cn
zSVcf&GNBZXoGvh{YQbZ*dqxn-h&$wDkeUr@ikJ@aV^ZHwM2msdXAQ2GGS0}W!s+cB
z0jk4A7Strk)Hl!)x9~=npGs^CMRH3nxR-U_JTDLn*Vy)88EbeQ$t2j<tWIE$(8MLs
zGMf#Ta_b_uM>h8>x{W;%7VyC5?l+uIqU?lM(y2V`u2IX9$<@P)voY_alO4+jGD+q=
za(ro~<f(8?##|yQHVASDj(Vd&HbXMlTJhXgX>n%PP?E<r!ePevr`x<lUzuh=#G(Fq
zttKZkC`Vc3z$rxc5r~HYV-u5*Fk-!(u#Vl<|9bt?H3nLmunzt%5x|ASKu_rZd_7hX
zEr{^2^!&FmMYmr5&B>HEe)b5d@8?m6eobI8JrQ^P#2F9@<h&o*SN_mht|Zb&A}1iG
zYcMt<j`({Yv6R_MaYNzNU0t}512XfHJw`dh{wz7;by~m&J{>4@*w~DgJ|6X?ZmwZZ
zZSY!222@w-Gg@_u3{X58aZI$*+Rh_M3-H1lMqaqMK*4cj8ITYg?sy?mt~iz7@#rE!
za-oZmSFbTkU)?sDCgu|MG}9u+Ibs0aI4SJ*!aG1kK<SxsG?K;Q5;4*FDBBXHtGSG;
z-YqDFie0!)`J&TooLJ|#Bw<8aI@4zEjpZBWBXKKJthMk|1#EUCZ8st{23!+<zXH{3
z61^(meN7CT7>sW0QZeb8hc$&57c*HF#LI{L+e_~Zal@%8l_K{Z_!%S79#l%0Z(!Io
zi{QitSv`pXS-q<0_ZE1$oj?20yOQV;$9Kzbg|vp$+gV7ArZL-@-Xdu$oRcx=Am@DH
z%<zSr)FSc~13*rE?RanyTKcaDSmiZRK@Zk&!JpT~bG<`hUx_sPV+uLd=kLn|=f)RE
zfci02BaUn0OCtc4G}Uk@+K~MHHSz9V$1ZvY#Ry_$#+30t8{!ns)ePI0)q*q)_{bZ%
zjMh=Y(;=E7vP=uauZ1bWF%PdHIv|3<T=G$fuE0Bz2FTBpPAg3B?Cn&Z-$tG4pGS7g
z1>fr33AkF$zLloO%`~7&JX3zl#j%{XCd%lS_YCm=8!RR_UsU4f??IXPq^mw6?uVrI
zOFa<M<;~iy)lrhPw8s9;e*AaJ*vvhwvj69^vuJXIuKdrJXKi_deMoM6LI=;Xr}<J*
z7!cS*0KhQD5|B6I1o|AoA`h_5#(7fB-s+!VKqdOwG&dk%aID-4hvEF}zNL#wLTX@E
zir5#0W1!%zkQ{GJ41GY_D3Xf9i?t1}#qb{Nl$g`IUukTR6*wUFxs)#@xTOBl9BOIe
zF+nIl=oy9an&ty#aSnikW}TB>6){Cv$c>PV1YkQ^Ml9D$j}xXbqNVd!JHo%(8SX)}
zw-Yi9_R72NN$9G3SGkf^8OlV+n{W((QA&4oB+7wRH23)Isp@Y0dF(eu<Es<`*NETo
z;W2tAVl&n03E;vj@`QD66qY;;HZ)Ko1|kRfAem%DMoOopm3ap@gQZ>5+U*W>j{ROt
z2LwCAQwh$2BIWCX1^@{G5lsgo@z+h4ISfd11dQPk{D^U!9xrTY2}nEwAyd(f1B>M!
zgC!ux!%=Y~2Z<^<C)xZ6O-S!bha{oy8~b-qnM^=$Rd*`;tOf6#mu?y7Ow<_YiPSz>
zhU%+%<1csNPBYqXUSo*tAhDhYO&rvS0A%3)-?Jem0st*c_yi(KnEL(t00M<VCCHq@
zK(`O$oo+u@p1u&(DBi0u_+AamlN07jFUQW4^HhZY!i_)#OXlMAvxRR64$o_`*mkae
zjrB{Ar!g%Ze(7=S(VCz5$H2@AerFeSgY3uN8lLf!PWiRkz;-F!w4yz%-UHa-Q@(U+
zkJ`<(&=o%&*4I-X?+akw0G>7A8kl%`0xH0vhG=P#DaN_+uF}@VRb1G@&h?ETKd>^$
zh26ifco-HMMv*m{afu;}z9`vIxm&_#RrCH@rgRsA;lbM=9o8rt*N|XLlNo2H!!P=g
z?og?{Hn@IG3k7ZXqHZ{1;RLu~`woy21f6niZt)-t@+OuvD_)*y=FeM<qY#QcY+ebn
zMP_c&@W6AUL}&~b#<J}+<$!!PapZna>il!^o28A|h-4=5|Iriswhj0C2wjYV2<S2x
z(p?vylMt=`d){@>Lj<TnGUz#95Mc_#E~nEbcOSqSl{)%q0Dp+hSO@5khQ2*jq@m9T
zszq1N2VUP1+kzs+KhB4jOLQv&NkZX~Zc*c~n`)JNAFMsjzJl%F)AXCxii&uBSzOO3
zKaQ09#rBhl(U-K8<gW6b9nNQ}*)>AMDpTCW-AB+0`?ouH27BWCsXJFC&?G-vX?NX=
zz>2{O`e$tHf%_pWOa>UN*RMD61OwNcoaylujX=CH?8VbxY^}(L;u5L7#X^lYM|wc4
zR_w84FL|C{Zw;6@!uImyYnT$?^F+Dtdy9|#lrDi-U+3gV`|`>+0LBf7C8A{=PO~`A
zoS2m$=g>!rq~HP})is5cGz>Ar#|zZv0%`IZU$tec#i_YjUjsZrM1Phv*8O}&fD&`*
z(7?FOKS8!<7%<c>JxOw%wC{>loSfbSvAP_c`(*6~#w@fsg7i<u3}}L}DX|UHMw>w{
zIuKLeJK~nCB~B%gpL;>mYofUsb|S2~+ChbAG%Zxr!;1n99%$3g<tk9DN2b{=tW(gV
zl}xGF<+9byf&r2cP1K4ZOBOAy*>EUnMC-?+fGY0~tzsI{EAR7R5G(IZ7kw(OUuV=)
z#H<ir$iN^!>i4t>EepqT0$*UO4L1CY930ovn$v&KmBxKQ0|1zkEbks&=#KYy{e4>_
zECB#W34-sg;MY41cC-y#qSmtme1<JX&{vs}#f+O5B!KjOe6V0HU=8;B0qS#r&|>PI
z0nqKD!e0UFcB;)lOUCRVa9ac<Hi<fBVcnxs2Mrd2hsPLJ3-(n?vLMX8sqA$)3pEJl
zAswYFH>&sJ%L+Yzh&|xFg(&wS6K>F`P2<03zPaILshB5+=jwh6NM&!c)gtg8RnTUY
z|KcTB1o&w00r?v^jchO;7+)@X>R_>;ID4eb`DVt=^^%<`uU6zFYn+IYt=8~GLSg5A
z2cxm@B^*2~AYv1=!rfilQ2xADGR5Z_oBOp<`&r0wq0-`4{lxYMqRTaYQzz^X1VO?Q
z*zRVd-y4#iLn5ZpgL>9``MKU3f_p#5`b^}s0xl;%d^6IiC~H`^Em6a+9zZ6`42o@(
zPpWs@QD(kMzL2dxOLh3w6}h>@SL>3TZbiNSvfa~#^Hr&k66Un`^tZKc^3_BxKipwy
z>ug|?_KH$t_ldddYM|+V-AZ$S*=<EiX&~8V*Zl;&IE=sL=&5CQV)?y0#*At;Mf8Pq
z12ALo_*UUw)A>bkaB%9vY}Pd{KsvCRc*ol=GdN5nOO0;IgHL8LD!VJaB`R82s{m60
zld?l4pJ4<b3`YP3_S+HPcO`y`>|8Y3vGF{Uu_*d+!HM@sx-C(#GtuV|w^)!LUWwW>
ztWJwwV%TKPA3}HA;tsIuV^y-}i0b|^K#yz8rcvGy+)>8*%S2cd6#bthJh@#O1-7w~
zYFf00?Sl*Usia0xXAdjR@4_yqhdT`9^Srb~k{OWqtk>zMCp%`ox;2H`lv!R+2I+}r
zCcgM||6LI>y*cP~L{5prSq>?sMSw`=OWNj9iC$GUCzy;&zdh?UhlM=XHAlE60Ng#x
zqY<utD|<~+vA(8x4(5jTF)Fy1WNr?1>DBteXKoN%H&M0SzXRVXxcCVNrF#6z*m}^x
z0RMpzO2bPhKF%A}6G2p`iu7-LRKI9U8S|v=J<@<E7$P#?Fh8HS_Z|C!H44Xh-gz=s
zs9~vF8t>MnmOVH!MK+R_a-CQiU}30s?Dbyq7<;vVbJB7F;~;6+?px~P;o1_x>wo*j
zop0Hglk+`3a-B2J=yAuArFvzogy^?Q`d!$Yh_F4ER2#mtpYb12oq3i_olcAOG`a6y
zx}w}?n6JgtRGe2T6*0-?3{lY^Z7ayB6*#-=#!AHhqX&sE9{{Q^sCu|IR+BYKTpE&Q
zTWE2m`G%gLICb8ic{2I&ecKA?Vy<|C7+&dgV=k6HOqSM#CunP|S49-l@}vJj36|xA
zf-pFC)AA=%^M9Y<6>!zAF))`{DmL6zH<Q|bif134{1f)GjTNf)@-=AmM^<#8W@**x
zOHGVxpPws)0fkO7gYF*rFG5!*;X18ZM0Ajh7W_#<SAnabHErC{LRXQ+FVC1W3$BRI
z1n2zoKKXA55_f-Jn+j|wEPJ~7K6P3K2gv`zSS9MFHg{X5Q4cwi^uW1!V6W`B@t~KK
zPBVz??7(R$c~Ca0Y#Z<Fh`0$X?>zjXir}c6(dA{Y02FZD5>0)?SM(yvsX|U@hEe~9
zD`ipAn-}OZ=8jVRDbPine96~Ur07YXw2!=YnU9j9S@MONhK2b@Mv~edrBC=r^m_lm
z;s(17uiDRq@OW3llb6`C^5vV)p<Fx*t)x`0gexgG1K&_vq8DY{{}Xa`Z5)1><|;J0
zU|Z-JACQI%WdyfqksvVD6kwYsAp2yoikkRcJl6_en-CE}3ZKP6sd)PFrCQoD`WK8Z
zUXKuHkT*cdPNNbC>>=P@m9K79dk9qNy<gw(Zmww!3!O5Kzu*E+=TH3rmiGL<al@A5
z)ao)}nU2V>gsVHaU+j;qcV8h)$KbCw1wg&X1c1t!$Ktzcy}Ilbhq>JcUUl~$_ZA|N
zIr>inKx7)ESy+RxAYfptjm$r!G~=b2$;AcVo+F1qm2y}WjL5k$prD*+^(<Mki%&Yn
zmSSubVx$Q^NLMW~)1aB})4KYUCv@||S8t($he%_VD-}-5oW(ju)q8{syxr+=UeaHd
z13+8&m3YrDYvUpMD$pXAr;NKvQ;PmiC8W0*y_(3Jj90<L5w-61^V2aE<cGhRh^+Yk
z))F=rT|0oSvZbRj@p=>IP{-Og-j0!oIQ|7Cs!wd+Cj`d9!b$r_n(IJCRI(-vn@qks
z0R7)+KRTFJ6g{voUM^*2petkfCcv8Gk2XH!Y|!lTFmR_|J7IoW6pwY6Pu2g*?w4F$
z?C*j~el;VWhE9_bC|uu;H9)-{;x+i|X>iYIIo6*lgML9vl~H!vDj-!PlPn)8kvHVV
z*Tol6?ix1Sn;Xm6CRB|<czSsEVs`Tt-aeuk`Z=C^C-5XPDdg7ZfouR<#e=OFqO}~m
zx7YzV>1%0;35^6U6u6dr;C-Lfw1ExVW<l^+;XR%!!COj%+wugfby$nY*h-aVYPG8_
zL^C-)mwvGdo;f{#X9`>t|289!@^`Uo;;d~P0fB<Wh4xwTlG?D*MR%Zu+@=y9O^`Hh
z2-FUNqTF|IPxgyxsVbliey!*iUOeTPF;hXe#flC~cI`%y6iH&I?IGE&G<P}PlVc;H
zRkrO1I=LQ`bVziMJzc>(%Ekl<4#T3pRUdw=^}PONaB#!+@@(GVk43G0&&sX>?6e+5
z{{Zu|s^uzf73&h0f(&$cR$B3>Mc2E^5Q+#!bg*9G`hTj|GY!D``4dF%KRo96ueVdq
z(!B`NqYbV#7-XcrP(hxp7nPKw89Q122v^?2^3K5&26r2l2#FqQV;8#mKQ>TP6%a#a
z`hG=<z_Jg9fktqb0m2nXUTaz72O`TQ-u{HDwZ&G#qJb41mhahuQOPNXiS?lH5}XCp
zYR+g6lFfoDH3`^H(GF39%0FrR6!e&c^A(zyuYXZB&u2eg#<_z4xz}~ft&2e{f+76}
zl-~Yh_*KiP)Y;>+tU6EF=jNdHpP(Q*F}mN<>-z$%%)&@2x7T2DiM9n6M-S<9nS%e9
zB~g?5I@ob}Q^#S{?k6mKO4tk(!NVtrRaW7p`-?Wyp#X|~FWWD@nue`{o>_r{q^Z`u
zfS7-zq`QwJYK0}Zz6VME^(F{DHC^enRMn_~AeMFrH8X8puZ!)l^P9R@m-**#?HQ0T
z`R8<T^ZO)V^M#<mI$5ANi{C|ltsIcxiJ(%)1uY$TLP+6fkc*pqyW$2#dmo1^5kngo
z^hs9*xd87PxX!+0K?3^~|F)YpyY%Ru^m#XZdv<Vg_5U!XF8^A7yEBI?>;!zK>02S;
zBCKN3IyX<x%qpX`#qMg2`F%L@N<zqPm7)<_+^%esPg%@bqtZs7d0hK<=d<g@>VU9E
z0?$HFO{&`@_$MNN4SW$kivli1L!5s@<x7^^6F|@<a0oO&h%opB)eP+DWd57*8)XCg
z$oVh6{q#QL)z@?aL}ZfADdDymYD__CYof|+CrLP@NKMI{cRh<oBT@Fz9$4su9{1NK
zm8U?P=;w|DEIr_)K~}9MPvfKE0|}$j`sO&hqJ6|IdUtkprVmZ0ZKCJR{BYk3BJ6Bm
z2QUI4`AwtT`r%m3c6iG@^*mNEkxEPPr)O8lgNX<IBUqD@lc#CT4)}|`bOicD`fH$b
z>w}(s*!MSMRuc1S$ryU%N62+U;fk_sMMf)x<7-9p8jEyw80&9q?cZ~C7r5AWGKNLG
z=}o*RBLxDMBwp?&N}h@~-oz#2EdIos9sp8=eb%R}?SGSel5hb9EE`Al_lZp*dcT+^
zhB$s_0V7{}vS^3<#Pkar(oG#)yv$}hByCS_N9o6UZS1D}XR{*t<NqQUErSiR*BPLr
zE1z8nVH3O&IXSP$keVQ3wU(A112XMYg^?XYQf}=K`iYi@(Yyxk1^QyX<O(i$Zvo6*
z*KiYNa}t|Xc5sPRTc{jaE+PLPWA7APS=hE=#_4qIbZnzzCmq|iZ96;c*tYHL*tTsa
z9ow9I-~9i~)YKeIy;XZ3tn>9;&wXF@E<Cm@r&G)6mPjauRLioSx6H|l8NJ%iNf9|a
z^AY4hGq3n<if-&z##i{i5#JoC{W)SiHo7n<bc+@Qk7WV0<1O9~dbcaq8!m(O!O8k{
zI(r4nryvRl>WcO`UTet7&oR_{+ssZE`aFCG&<L{oWkb&1K3IRf?y6qE=bP~5+8|<~
z46wHUq2Kij3k$!wW`2?BqYzr=ag3|rU1v>IH9mzAW5Jl`5avc_{pT@OUAx{=OdEr{
zcZH7G^d?v_v}8@?KJ}8GG<&r+iB?cv^JoK0vfPha$h%tO{lj^rke15BlBsw>JgTo5
zfE<Olbh~Zy4@c)SH2urgogwkp@jS(z<aCvxKI_Qj0X|!q&k*L<+jGIJ@Sk5U3BDZG
zW%0h_%^kMagUq*J_7Hlz)xK!T@1KUp3<vO=nFx&R;f#qTc_kWU%2V!|(m(D=Sx%`?
zfb>bR&wOLpI8gfL2>mlNGXPTy(*%Psz>#xl3qvYyGKKmKr9UyK8wTMZ(V`kO;s<^y
zRuA}aU{j@<H(q_JxPIQN`d#P~WA`yPrI67;Jw)7}?Wg+Rr+3<Y`E?j(Y0)Qsis^nu
z_@yRyD(ZAXS*o2|mV;qZ_ju&=pGBk&$dW#nE%<iph`8osUmnMgiwLik4PWKj0EWyv
zjgVb!D(yuY8?q+Bs}D*)(8YF9e-Xh&_+L5JbYz%-v`z{=3T@u|wf_WT_&N_)whC_s
zxqMsB(1AD)bzfK=Y2X`vDM_#!GY819gW(>bmjY=AU}&db)Xh0|OE)xo$O>gy4e9Fw
zLAaKJSVt6Su+`K?`*4`hdv}Qd&Of+>^y~v3t*Y>`2;wQIC(ipzH42rvdSzl-I<y+J
zALDqh%jNGF^f<nd1xLwdtwZ^5qMvip^K{8w-i&n*=bb?{NnO5lNuT}Qg-!3=Z5Pj&
zR-Zt$!1hs^FQfBdX8Jh_hCC9qSI>Lvj@OKmxC@lW!t@XxuK;bi`EE!6m9)|O!OSEE
zPv+GUen>Z?e`rxw8bDfSdV4xcjkYd;GOM!2w2jTPPvB!NG?sJw3;wb@RBzPKC5yQ)
zb{YDDB=_;A_AIF^Dog4^j*`^2GVm%hNgb!aFcPVNpWmx{>Wcp3Ove-MBl!KE0Vs#s
zIOHYPrtWl)c}t{RsS5$fxvNQ%)HA*R3^05c12k3PmX)V1lhi@!GOYAAEQtthH?nJD
zzUIF)av21R<|LMu+xsUJCx~Iy`&f91a(WgCbvfqd628TOd+-V3H9pw;Y+mRDPJ)Ts
zi$9C$h8q6N^0;xCpAf;>D>-2{K`lk;nWSy!bK6FQ#fURO$G`xr7xzt1Tpx#@#EWaG
z?)B6SQCCI=4$D3d3j_Zi=4EMeRxF$@R=mrisOv(M%hjj`f#owM%e0bCxjvjN8!1yJ
zFP*6UOy2oT3)*z_k_O%Y1rlX!&heb1)DLrb<!*D|%NZZX3Kc^X6zB13cKWN>%Yl)Y
zhZ@7Q5hvJ)r6~l!LZ7#c_4j8Hij9<}#y|Z}jm9z-E(iCpt^}YTt58QMI6Ptu1=|f8
z=Ia}U>!V%~%Rn9vk15TbA`ny#=-DFgKVQj);Bqx<^Yr;$niT8(p@6uLTYKwV?>?JF
zZH$PZn94WaaST8h@u59jO?cpXd2J(Z%(`OCcQ8vEGNT3LO3ag%EAnzHVCEj^O_+YT
zHYo&0mwvF&4yZN@v%_T{`<8@hkc96dwLYVBr-u!#6n#)=-*!D^UB>#pLPX<m*rI`2
z-D^cc_IVU&mazwc_f6&V0KO{?!)%Zh&q8R9I0On{m903KA03QPAV|T4mf($ejGX`&
z7YH5?;x(Wj+#Q4mWoOrIol=QR(1}!_RZ|eJe?IZ{G+^12uaoGXgX+fh@oLkri$8EJ
zT<Y4p`rtTUxp_dTvv&;mB);ywe}tdo1el6~-ufeb7k5&<ANM~yIbnu89Zs@#3P=F)
zV)T-+puo3bYwxxYe-quh(~`)ObIf~*)R)@vTTVb7X2xJdwL>V%Z`PFL7&P6w%T)v!
z>n@pE-u#sJ;t=Pp8eTH$;>uQ&FU|i(CUs`s>f3dCT@zXw(3%17=&Mq>jU%Z%(#S?L
z1iAcJU{MOlA27dcSJTtUxdoNSMBbb@n{<Mppl7B3->cHSn`Htm0B*ZkdfplshzB{N
z4Kg*!vRu553<B{Yood=SxzD6ZBiu>++ZY5HA<iY@=L%h3?8i8qvR;?~@INJ8B~#Qg
zMn}OlNPUW*Nc5o4mM8_WvC?PVqJ`=ydT<ov1O`$2I{RciPgmY)JoJo1-Avc>^N!*0
zhkKR5ozd{pDjN;pW&RppZ`5UtV|NxvP(uIN5_{o`yE;WO>Hk074}DSp>Y-!zZ>Khj
z^nJ%F@rATA&KS9LtK%49cE|1<Up1<i%NL&pm#=+V?Q4G5AaPRt4oiIAd+U}F7TQRb
z0M*;`*(PL>rPu9xv-Pn9(*U!=R&Q&}DwDm@<rjEBb%}>kiVp$KX5jYoi{;V>(eW35
zQD=AmU=lQKzI-ndGKc0#eHAs7Ux@QHmKBr&$)o)|s{E-%l@nJ^7i_G$-+)TNl11hZ
zC>YO;q-^hpJ0qSi71|l8U2ipOOtdw@Lk6HJsXz&e<0#hbU-G(~mjOq!?$|FC$%)**
z`2MT_41=0;WMTq!?1Te$460kcd$x?8{g}}_QQ-8^ZeqZ&uCKbG6CMx35Nz5rI?zrd
zl@-nV#6eg#dF3J^@b|fFX_KzGzerYu{>h7|on<=>pEKpP_%RqgefrLsQ%Amyq6{bJ
z69sDo(Y@7Jv9UE=SeNm8rd<7jKwz&}S=<-E-llbJ<;~T~0FUe8PO|a~v-86~(uWNZ
z*W;W{Ab^1VCu^(yHO+-Rqn!d<qk3}$FLn#<F#;8Ev$Tdbl(J~rLpydnwANJU;L`aX
z*>k949`}xGWkm`TJ&K8*02Q(%gRLrD+YVoMjSCV}NRsXqLCG>t1hS~ftyP=~fBz1k
z;G>dKJvVihB&t&g!LzubT;-q<ng&Y40j}m@R8QHoSxX1|6<>M~xHJ1$73HaD74m)>
z+Zfj6#IZ#F-CQCEldue=OC;b~I}JCkelE%GIQX3?gYF^|5;Fzypyi$+Eb%NEZ}oD_
zaNK$PyW@^Q&N;d0T2XQme<Qd)fb;>BSvdn|>v`|Pv)I{evdtOU?ozVJe?mFr?>ZaS
z8}u|KsfWuhw|dM~s+{JZT`?*C?H9Hq7B)jOR;zj~qn1Z`QZ(|VzjK+#6M9jffDDfo
zQQzp-By66mY^u=K2$!bcoZG<;?r6|!6w5-|HLq^=gd{a+F(bjeQ;cAG4O|0gzd&v{
zN35kVQ0R3_lFg<J<QuR}2QephkjF?p-Jc`(VKcfARg`dI%?5@KM0OHc@2wsjNAXtM
z7mJlOTY9)u6Q2P}Pppg@h#4<{^0?vVqs<jE<VWi$CVNQ8V~r9+*+y$QNIpMPu<us{
zvJwA2fTF*~Y%r=-EeF#B1aAO#S*TD7hb@9Of1hcu48R<o-zNTGr-u$imRXmhEtuO<
zd4*u3t4DNcg^9NO(nb>jUed0oP^|$+AA==fS}#pFy9{IV=B1&ngwk_`1Z)l%^BH~Q
zb<OT32wsq|Ow-`+6gj8f`%?7o_cm<xo<%%}P!*_sAnivGqwOJuj>iD5J8>PeXyAw=
z!3RI_t#Es^^{=2ZBJ#4>R<zNm1nBK`BzOewA=Zf{44E$-`{Vsr^3Q`KkY#-5tcA}o
zoytc^{zcbRQy=Knw()mvwX57{D8>`t8tuOf?n4K%Hpv+7Ih=N>V<!&i70*N#x9W9%
z{vhW#Y4LwZEVw&Z3n2igsguq|OFYlj{~bdLDQ-QM8lpY<+4WT~m>F~y2ekjxQ1_)1
zlCdAcUD1HyoGM@lX;7ffkbGz$o=aVAO{v4-xQtX<w_1*!a@dtMrf|;TfH?873xPl8
zi|fjs@exE&f!U<kO7T;WO)shPpkgQ0XNvb5F|r$fZ}moy=9&W>iW8CHa$$yOyyg}9
z59IB-tI`Xv7l?G^wtl^*9OLv-^j$s`7xbkkNfzY!U82iB4be~~$nK>{2hzH=>0(q1
ziMB!AJ&6nwnO)rK&tf&&`jH@1ur$-A(bg5~GoLx>zbI{Ang|(V)M45@PgH)q`yC);
zLSUWN`r&XA|N09^+~d*ELE&wXxDRfTE<6L{|B8jLds~wNJ>dg6*?RGvpAvSfx`te7
zgvLH*3_cZ!%jWrjEl-xzo1LwcL%w+J8OmHqei`z@BXfuywlc)?l=eH_yVsNNyd`V>
z*tDrm3cwvtVdz5h*4fub0QqAM5|m)c{K-(>=T<;yI1B>7u5U3RSx0WM@UP4F6p7#D
z=||zfqxuy=AGlwgml@PD;eXSm{{stUk2QuLW6e3@Q`o~Hj_)ZbN`yjP@d$4|UWC8u
z81e3i&W_Z;Oqwj^Iy2HplL}F?P1UM4FeJC|DJ;5m`bI>1mKUz%!wSiNzSHV=Zamfu
zLyv@ga1T0QErq6G#@m}A8!4#517sK*zLd}({<~Pjg+q$`9;IRFk-0@5JnBEWKydyu
zR89GaEo(_Q|HyQ-o6B{+V4Mn*bba5#(bC<nGJ7T2UD05>;HAtV`kALdpfVBn-TC!x
zqGP`o=Rew6zSUSlmvwHpf3$B^t>MH@Lu`xM5(5BrF*DPIMK#2SN1q5jmlaP3stXr+
z5#3sMa<mwNzC`+ZF!we0?F<LyGEAeM_jRlT2>8BB*$)2|O9ZasX_x2HE1TteA-mT8
zN1b6|qRbVf^7Owp$@__SKDEn5s0`!B3(E4FMOa$wYWL`9vj-%KutOANzLty-5Zdnq
z?WHtK?Rc{tW%nC10t-amH)8s0<q#Pf?Tj+OI9Y7VfVP{fg6*A+*0G+)H*Bh6($#xD
zX@na_ily~wOB(n6zULZO+bGLd39#^$O$k!*z3}riO}pGb%H#x9me~Mstfrd1<eYdM
zOBmmk^)G-H&rfC79JLSgYpX8+k0RA4kMY=K`>C<7l&0o74L|91&0k9+^vUvji1Jyb
z>opMUkMsPsmqf3iW)o2S3>yUURZJ{S=^rK<gQ-uqf7JawTJ*H^^~1K>g}dBoBsQ~x
zG@8Yg?j(y;I427ue9qfoWbqfp#Y!g;f;T@{MNGmaHz4O^DmfcY)`3xg#4wxmM@wij
zRYloF%T8Ca^up(}cU0)iebx!vojNi!h+;3~ojN#-zm*FGtpfkT_T=vxNZY6u%_vlg
z)-Xa&$Bm5h<VsaJ$yw6LJv@pwHYzMg*{Bz%JX9!Ax~-yBvcI#N7ZXB3A@*`XN}@As
zQQO%f)QXqYXx7Unf>dq*y5y7V*I(7jo60cn?{wLqCw2kJWN)Qr$i5Z6W+%v^3_O>K
zwOK>F-ZqAU{Lyt`z{yNoQ0V+LnlX<L<t)waVz%!x+A=#i{%5Yq{mxv|9jLrdNSLa~
z>b8#Xhu(qN;fM{R8}nD3ocxT<8M+n}Ug~d_jj*!LJmjU0vOopEY;A0HYihvYLZ;b7
z*XrPzwYRbD(Sg0)Kr+Ya-F*##`Q$vWJ0A!P0!%q6eZVHjv}x6|T0y^B-&sD1C%2H8
zcoaC-z->hELjyfxUx+3>9eRiW>N+dWJ~~PN?RRp(>3Ej-$yx~RQ>r&m{2@Iocfh-M
zte@}zEDEQ&4j~5|;GObVL=SJ>;;g;0>RUso%u?>(GJ5u|)vD7cs4Y2u?ScG4!=x~4
zl%?oSP~Fp>x&)eVyNg3t)zXqW{3x;J48=)Ht}ZxeTAZH;zDLVwDZ!12GwR^{`V@?`
z4%ywwmfyzbxnHCd$Ko;<s1xeaH`KD0M-xz4<6^~OElLBZtg*qWr;f^uh#b5XopBB-
zwgiEq2dc(PH%WaGl9i>Va?pt3&ywZctdM(RC&WHadJ*Nfh{D0rQ?(R=dGai9QAn`3
z7jf2QJTg1*PcoeX0}KI<U^nH0#s9u4m*hX$>=N5;8q1ZZ_F0tgU5oPsaRx*3^WGnj
zaX-ZJ;O_viEdBFlP5#-YGt)FBRjOjnw-ilc9o9n3YMGvifvbs&!k^lO9-pcE21vns
z1nN0VE=1_d4DBR<lmHaYq3R0WgxGb1kSg&OUcqEMZTh;b$lX||678hdyjD>P-g!5;
zwm%Rwc_228j=^-1{Ud5mDkvRKTT##}<w%(i60U&79O{|`vt_t)SI9?Zq00kC>^+zx
z3(AZNiOmJ=a^jE?uZ&uot)e-J*q}yI0bF6+SqS<HBtlWAO88&vRVP^7e=qT;sS|N;
zd7rUb=WiGpf!G4mJR)04Sf)pqZA<2?=3G^tor~e%GWU2;*k6h$=DAG{PolQ~{5GG)
zKt2G?MK|^GDfZh3EZjfQy>eX(n!0_T?HDk5yzdaiL~C~8=`~W$HB;an66LERMAvJk
z^K7%A;&YSEk{@niIO9^^zJTwMAdp>6sL;Vzu?#3XZxGr*t%k>nKs>wdu@|l<84#L?
zz@|rFU{7LXx#?+qcMvWQmWDhuI8IRdRRP%W?a~+~322d^KN>IFdGxksfvrzA&7qr_
zFJouMc}s)jwV=>0;_A48QDHbeOSDv$tD&w-XMIJe8QV!&N%1p<ocx*Y;#B_!7)~Xr
z4GgD|^_7AcErg-PQ|GF6)<5Z)^Dok>bp~$nd)Y<-DS*&rKoKAXf>(^W75$cYB7i#U
zc4T&_rIfLkw8fX0!8!!qlQQ~0mPfuWEoHEgvN^is$-iSR3RaP_N&ViND?Kzb|LC~0
zSy}k%J>5FUALMyq6^%>3GC|=%4^Zw&)m+Ixti`F5yhhJcc0LEGAv^qO`@4#FQj-~I
zFG{OI&dzMFu;EY%-|>Fv<tqOD4bZd4M|K*N1;wkwfD>5PV;PW}H=~v#+#%cvq1$0%
zlD+291yjiFuUEI5G05$AEbk9e(NFoW@KVv#OXvn1r#*W-hfe+J|2&mZiDDDqkb^R|
zO*Tqna^`Q7BYcA9ouN$H6L1GR#lX?M^B#U?G5Iu@U8=MAvxx3hH@6G$$H)H_>>&v+
zA-Va<%o6S%>zHxID0g_E=*QTYV_IUKM`?>AQ!RTME!PT8ExW#5&z5zx`098npzqpv
z%Uo|L-E7vg=C0r;E>1F=Zozz>bZsbYMQ!a08#)>&DCHO$+VOPp(;0(#ek*=88f)`J
zWv}|(eM)_Gnkn*Es`(P2mZu@b6d^*0aq~~7WV3CbLmUDsWm1=SneT@rtzcb4Ei;1U
zm){~^R*6GTIYDk596a=wlVz<#3@D#E_)Op(Hi;G7rN9r8A#)KK<0+~(TRKH!!?RT{
zjeJR5u;OfkHcU(sV%qwf;@@Wi(qQX#yaV40<8rEPz6wkZUtT!CxLWH(?`o+rU%GRB
zi79x&qKhE$cD8>%OeUX<0b3gV5q_OE;jd*kU|DPvpL#}?FV8O?;(LAge2urLDExcU
zcUBmr?e-fK^#9IacnH3MgHO?`+wIM*Q>&4ECC_3re&UE&d#llKYG`|sAE+~(jRs!z
z?sj$sSSkqy++!;kGmFS3s_)qp8405rnWQTjN~Ai=<WesJ6C#VX&kKvQIQ0Jpub?Ey
zniWt)Hc+6>r>xnBYI6l@GY9S_(R*yaMc0Y&rGUc(hB{6w%R{{XtMvmpDIq1io;Z*z
z%H4yaQ(4sPo9%u=)Qm@__?B&gP$xaI=jw(4{IF38nvJuzu+UK>bEi!fO40QX<;t2#
z^OChqIF+s}r4j>?-Sz<qV%x5re6w5)k$cQHa+(NBU3G}$?mQ44CQw}bsb*Ckb2=T;
z%hber>pB%!z2J=CjO&!FC=Am0pS%7n|2Z~7>8XUf25IS(>6R{tilxcn_&LRBYXD61
z)A-cA6L`)TpyY_ybdwC`0MdXx31i2AfDi7cz~Bs7C=ESV=<9oUj=+0Om=S-PySCtc
zUjXDn#Q%mp5<L%lbKJdRVT84?XDqC`e_CVxFvwjx53?wSYzPqUbV@~v7@)w9ncv_6
zv|yen`Lke?eKf@)mn`sK0Jh%8P+cZkM1CXBsAc<^>_5Zi=5k@~-=55Qx89QU{HJUK
z|5ZZ2E2)ReS;_H{G0aE!Z;w*qAwmX}(Q!=`X%C<0Mx(M=i@bIw>nE6xZ$-`b+xIM_
z#WG<m`de;=K4`5#m`6$84lOfmFI^<<$r{$Pv$b1b)&^hspXm<BgZ`h^7481zvWxrp
zQ&#tptJH0I-@v??UA|`VQ}Ym=a57)~&XJgbJTlMm!?)=!w$5pdBZ>e#9YBj~sUl-Z
zDE(n?Uw1e7#GB{>;st4N)oYA<yfQ04*V~W-ufvSV1gAAdU|si@U<UDc8|?LevFJh)
zY7fxS36w2b&3rofo0a53)A}Ban%NQbo{3vXUx6~|nc+2b{%8Hv^z!z%XJE(HW70s^
zYp#F%Gu3&}w9UDeE>kfp-&L}!`r#im@%O&p@<EF7dZs@gX&kJr=NHK7xg{qM&&3h4
z(K2x((>%7zDy3nzO?Z{(>Z)+Dbgn0Y6AJ-PB@gfDr8!NfHzkd?a9Q6`MmSOzpP_tQ
zsRjgo8onG7Z(Oo1O%OL<WAAR+itIv^>ccrt6UgI7R_PPjEV9jN<pPZ)Ib>~s(*H|#
zcK%=4nM(-ue?qSQrVX)a-^{k>eZjJa{SjAu3sAu08AHvY<Hf!r9b3EhWr<WKBF_yb
zD^Ynyk^Fbwb^F`Y<n7QkfoDs$L4N6v6Yswp2`BiEI7;el^%6f$B=K{T^IctiT}^-U
z(W37qMsaCp(Rtc^ktebPvtOjxz@QMBS&kO8HQ?W{BToi50Wffb-jR1QquK$3VhLS>
zZx6r_O+(zLYKie}6<}Na4R-cw5hYlg|3>&q<l*CE?#Nm6#?{iBSjlJJJG}Pfbzl84
zwYFAVuJ{%c)3+4A)ffFHM(R1m^i_c%m&cc;QM<fyp5Xnb^wEUafs!}RN!QN#%ipFy
z1?ac+HCOwYz3Q8Ur&PW!*&gj7H%T=M83A7h8{h74sgK*6#tGtaU69v;s^`6WU+&k8
z6%BY}VEEGejNnuDoGr4al$CWjOvobZSZz-g%JN)w9Z|vyp}?;<N>i>GT3F5!*38{J
zUZIJ*kjwkg2ROUaUE)rYEL9MYR9>i64gLOX&xDh?Q1Uv^9QpO0-svM{7#X{X-+&-I
zPeU!=`)tSd{6ZBLThjB*b2nL2jB+DW)gD{Fg6{y3?;XsOO)X}tOX2bgQDru~pWL-c
zlBsOBH`sUx(Heh+=Vy>R*+4<L1_1o?U->>n9}FAw6X7!9^3HewLwmm%!TZHs?-AU<
zF3QyqRoa3d11uY&x5l-JK(#>iAOIBXxxUrs9;eZb_gRMCcjS2T7C|k36y6=E|Im{G
zX{V~#3Wa?1MFHUzrJ%q(Nj#11&>zHywFFIEnCCzbX~i`Zoj0H2ohg;4toiSU#1XY#
z^yO~dcO@T6I5T=t@RYR0YJA}|O&+q*Gv3>s%TE*-J%;@&896Z&z$==p3UD3z9H0%w
z>oNT?<@5s;K_{peGf|<s7jxpOV4HtqY!L1juYemD6o2e2Fd7-ui-=>o7m>#q8H!(C
zv=xf~cXy0$1vPJL3sC_H$51+CHerDo9y1t*kbQO_h5IovTy}1xBYd{LFF|KXq8DB3
z3Xzp;th-<G5j!0C)sGI6Yd~}lH&`nM!H*rh!^he7P8$0dh@FHhd=XUR9a8@U2-*0V
z^B^EaMrbX^d9dtBZqqL$)z=;mG81?=HOeL0`MZpA=Tf)lfJUNBEhsXGE|TZy9quve
zzP;LS7so>*l9g44e=p>2BzNo=M^DZK9WQv1m(YiV9J6;FuX{VVBtTr04C3=3Z=5ys
z_ZL6%heRMj-?&&phGvxm)GKbLq&+q2@rkqD;+D@-wHWVs+IavdSQZI_*NHHH^GM=E
z3d3TJo)j}h-OSnchX8lY!<rj=Fzx`RevBgsRw%=tP9bk1IIlbTGZ2eBTo8d*&+WI6
zb*-W=_S5k8`5SeoEkF+*v_BL--23nA{$vrXQz+MQ-BM1-#>uVkao6;BrMG{2joFq{
z{cVVEqzMeeC7t8Bc=G*lZsZ0kS;qS5AG-Hz(9e4nj{|cf?LS|#4Zy@t*OImB2;(?P
zss4kx8@P}j!`7BkDe~l$;JPEup_qYAQli#xMJ#>a+%7nH9}w941#{dN9@7v=dyclQ
zF;1;PeoLC%&L5OKZW4~~HCE->ZI@r~Ea2)*?<=Lq%rX>D6lE7ZTpTw@6ku}}5mGE9
z?HqVRhATH>z?=b>%ONjc`8eV#vy$BC{C0V?i@}3)ZQTR(2r*LvpN5LWa%Za*Xc4pT
zk!1uSf+g(n1=x3qORm16r^wZs?JIe1S!U)eG!^i<9c#Z-m|0uFA1dT9Wo2{_+!L$?
zhIC2xo6Xa~(lszfIrIJlv208^zEP27ySLV2BsF!OLC$}t$qO@5$Db?C=3U|uqQLK2
zXN05yEPuzl!T~vXQ96&qz?<;6r+~N%xY2}llE>i<K&fr%u$*SpE9af=AZ&auye_6Q
zwDv!GxPOGxWSPgNO&a~K?4B|4qyKj7au$6O8qOxMdo_+M6T(L2pzsSSLh$qCch>0t
z<Cn)7(`uTt5>-muSbtTQQ~Y~qGk;`=KokD-u#OB*e;{8suxxE0mD-T3Ar1C;3}OM}
z)eZUyAdz{pHcQc8Pp;P+;b9P%SUydYuTP+jVH8Xs1hTq^aumew61yg#qx!%~mOlkl
zq1)C_c0{TiNaJ0K1I|HG5gv=9gIO>v>e2Sy)?Y-sHm`M!4M<!V71e(&X3g2N2@Vgl
zZkXkzkK?75Wae5~6qioupZ|-PBg5daxGn?&9ts7fN08&>=#o7UsIVLc>3?5$UC{|R
zH$*4e@cvTeuE@C*zeiyDm29|YWu@^%`*HI%z18@eBHHMzJ72=Ict`i|5xV#s`|4hj
zwdLH2uApmkc}WD5s05FV^%9#z$z|4;FWD*-jx|0E2jA_<S~@%!iqy7SHW$|#RTggr
z06`8xtyCyi9&JRp4o{P?yzFdjpm?L7`0Np>hr(!}P-Jjog4&%b4LtuBPyhg@gPW$E
z*<EToe*sU8w4Np-EoADr>M=+tLfz2Je-LP2j3s;Z8o9^^o!SOjxg^Oe7;z{5MFMzh
z66MjPcrfi(G5Gy_uRj}>{npEXj^3035F?Hjq6-uw47(gWM=3oH-Pe+SM}VfoBsn0Y
zoavjyYmw?1E^5t$>fCa%-}CqTJrs8q@4HiXGl;NgY&lZ5zc0A*3>HQh>KDZ@$67Qv
zgvrAmM#nJ6Okl0_S>70UrD~E(byUJIsr#k$RGc%pl-hHnHEV)WIwIk#Ltm}}0GXwZ
zCr)Z&`*8Y^swA<%c&Si~j(tuU7irlOjdDg2j7L^m5cFh!GR)!a3z=jRZWb6tHurE2
z3!R6WIZ}Pk%@!vP!D3BZ2TxNM7KHZO5EbTAkG28&9M=dH=02UseX`W7@Mm9Lm@xI9
zUUHOQNW9j>aj($B<S6l^MzvuB0NXN%vBB>rw+7)dq)cSJKT4eYXyj1C20aSD;_7(l
zFF4_>cW<&|BI$VaL4xUcVgt+4@SCpHb;%+*>p@GpAy&D*l*`p(#*2P^ULJI|SCfb%
zf3^??vnPTny0_UAa5G9I7!8+9rA!I+vMN9~e{AWr1D4Nj_o>=+v@gEzAEY60sC3Tk
zm1&9#EOC5f&e5UytKOW2ixRDuEloRVAIju1QMYGao`1#wDXaRV)9*ZvAx7Rf7y*4T
zy19m&!7<Q*$-n_0{*=sAT(htb8(8t+J{HN%w!WI^%!u!2?5U9Yo6^)O>oo#xcf>GN
z2VZ2^SA&YM4~kKv)zjG#fLqqoe<dE`p?(UU+V8F(UE!M3o^#4&*Xc6b(RE>~2-%Fj
zwIz?cl{?V@;oiQ)Hnrr~V~H6hx(*39l-Si4<+qV`I4WmGisTf`92A=yezVe<yFz#M
z_7olsO<3nFkcc+x^yNKBq9O{}=X5RVzq00@Z}r45u-s#AZ=%`*z<E{@JCwX=!H2Ya
z&XPH&>C0&0f2dR!os}uI-d7nc*g9c}UY|rd-3r;4PwB7tzE?7E6p<t7OAYAe<l;11
ziTRH=(W8TUch)xL`!z-h1K&+Tx`2by4e@aMmZ75?E1TTFd`I1Wpk~7JBJN9^uEU~E
zOUqY3>rktM0tUVdForR}7I_uYS_Utht#)0*;}HFq>6CH<Z6EVS$sEr@`F0QTGsvz6
zefivtMYfz3sbrnFVCe$|fsT**ISXP<3q8Svc;+z$?6|mM-C|S@*H*?a%g(?Ng^;R;
z3d}BFY%#&eq3~XXCv}61f$JfsvW63bE=BOTqS*F!5S?HL(8wN#8os6O+s>B|+FGMW
zjCtbNtjLXg>ZK;G1~KRK3p@Sjsm{=++XZdhAZJ0;40><*HY*w!D6p*@Lp@AEQ6xF3
zWt!3+vc>Pn&kE(et{<H7&<0<mogK=L^bnEd^hxn?OD_x8_ruBpF$RGE*AlUCj_BxG
zuAD1k6Z`!azzwo&<Hb`dphIL<R=Oc0s$j`B{jOuVEqFq>Y=Iij!_+uObxVtVXC9S0
zv4NoMHOu?cC-^dmt5qrgRJktHC>}b2g`&7RBm#kitZ1n+g<d{wq7`x9pR>}z_m8^g
zdZS_^oNo=>O4^)wt+E(%ls((y1@VcG`g^=dTw<Xu0E5Hh#US*yZ-}0l0sXekkek;s
zl+9Ex5qhn@^j624@h@t}UPW|f?7AScx^M_T?`pCu442`SC+Hbr+nHyaar-(M)zHrM
z<^y=UfzjQmQa#eR!#ys_KscRcos-9tdxfc-^C21W>PeA+kh8li%kf(OyHP+jx`vLj
z<1JMyVCxX?N&>U%9akmG`A>xVeIn&h1jTo49VHVHkBnzgzIUxVA-x}kK0N1(Hh>w*
zF**H^TLAX}AN@cgM(8xaafqS793UE}WC&09b09T3METwy%fe`^*Z`Y{>>_#fXpD@h
zINSjBG0=o>Ju--73(kj&4doI$GE*{Bi3gVtK&EUYn=}|`IF$~5_Q&3Gbrz);*u2f-
zBq%<`a-ZTtD@-miRjB%;H>vxSm>M7%{_xwR{4XzKK}1Q7yH$FbOn(N+af9S=kZv<z
zxm~~`1;@k|Jh_RB#No1`d{LF==f7)iFf%(!1hd4WgQ6D#H?>aN5O0enfSs-@`wtOJ
z0LYW=L|O9M<SOH8a~((#uLTBHah-s7SUi1jq29)3!<$-;o$aW|)g8rL*O?pivf9U|
zi`xv&vmN<3RN`~tYL8T!vTk~c6o<1=uN?O2Wp_qSm$seZ>a0mfB#^UV{9s~y?qIBB
zQPCf(ybM@Z05%ODJ|x?L$Ygg~spN|e@c(G^*8EZ>T;BBUUT-(4iS9bQJ1BS?=bMt{
zJ9s_=D>KZ7mbf=XK_BrK=OFiPY$7cl)(eL~fFN(W!t(+9`FFy(t*U)|ydpzNIm)p@
zv@Q&%1n2HxX0z(9d}=flp#%{ZyHMn>MctUOYO;(e7e^40<{WN1(NtMI>&jLhpm%9j
zf}%dyp!^q-=8})6<ko9Xsz_zY0dvtv8rWxKa!YYnLoqxS%gz*fJEbx&@W-i99DC;2
zECK2u7P{Ec*yZ-5WTUqOgn98sxsLeXZ1d8juh+(Rcxm9-p+qebcgpayrfau<{o!`-
zl{|N~TMWt7wGc*O4kX6Mv@*~g09hvoSqCI>+Cb8j{V{U&>d#rd#=BGT%f*t1Bc`>i
z;N<b-gqLKl&9*j6vtmCUaR@G<1KM{uhHo7&89pBDSH>lbfu5RRmp}J(yLn13&sROz
z`T^FJL+fm%TRxofb5<gmm|LUtC{{7vE>*RTF=!27FaUFlfvS(N3v@&gz}tH4EZw~P
zwn3uC9%PX$Z}}HezvM2y;z`HP=T}JZ9Fh9n8q$tNBW`OOSx&XB-HQtj+|}oHd!bkx
z`&Y-QPuH#)2kJr}cl=Uw5%gS*Pj4`EyW|+>5$V5*E&fRUNMJC)JMyF=t;w{F`q9Xv
z+sIeD;)PJFZSvOP`!gXCfHP9hKrfBv4>j^wp-UqRDCCOtf6z1`iPt1{?_wq_#B4tK
z$@vM4jW&n=Yv(@y(i8)I?5tjXwk{$IhA?`p>8ve3whzH+Go@ZJAh!3@Zd5X|6(0#D
zhH4Xt^W02Nb|qmqO15s<_BSD2&1tM^aqP@TJg8!IhX-itg7`YQ0em`-stY=+#l^#2
z+li6)g8PE{8iWiSj}f7ZV7`w9<{T+RSakS?a4Te;23mR1>(*|f(wnlgEKxtEA-t(b
zH~28i!-W?}K2|=~&w7P!zQoTR!;!auLkv7&7rs3*2tc;_k$lNM&n`GDhWzAY$oupi
z^5-rGsG5`1)w3SJB!Cd=_7<Z&7-?Em(K;;*%q9lmZnHq72oc;*9h{;qKMOffBA|Mu
zgXwW6tUTJ{M)_?#DE-M>oEi7^&&!Hs#<Jz2q|Tpqvws!zVp1;!8A`?1;o&;>Jyg_f
zwtUQ9=ei2Igtk>EUK$wf9kv%A8t(lmIc`Pgb_eEjlh^Wqj6624goC#5h`K0@P|n@R
zqcy2XykqG1QnqdIjdq9@&8rqAOO~CBfFSu*GdSe4Mo(C1Z_i9-WRWUIX%~$+%AxU%
z4~$0djg1k`w`KIN3mu7$Wd{po6uUjy+JVE)Lz8KzC<Hdu#Rp?ye_@75LJ>l#8<%4J
zZglr`C!I9FO8)11l}&sm4~={LFJ_<xs1_Ki9Mdwc+Eg-FK}0sExP+&Tg;u8uAD@3S
zIx<6Llw#w;#<(iPb(*~=E=mTkpQID?#XO>$^u-DRJ$gCdIJMl%q$fmb@)u8?L$(<f
z$C19}JR)p{ZlKpZ_H|Wl$#(YG;8;kJ4kM%XdDRl&5dUU}!7>t&L#jm@ZQL~r-}F$m
zMxwnJpCK;4)dzLbwX#>fj@6|{n};1{7$(6&^5E`jR{Tn*_|*Y^mQ0Lm-H|`#*=*7S
zcVzfQMoxbm-0H9M(~DqdAX;Qp7&~)S;dW8gMb~IWY4l?uM<Pc|x{UK=Zjc_;NQb+F
z#Y`ojg>G1dnQ@}W#DFJsKLz(To!=pGs){Zh(Bq0`cB-pT8s>GZJ5q2fJQ|x*ayJ&b
zqGxE(lR^L6el+Bt)j<CZ<Q+bfqmTXsu?|zqG|Uaqw4#x^B~YQA3nLh%lXBLc!(MzS
zd?>Ml)6)gDUGgE{Ir6xqyVz1lreyP<s;exZ%&mNK%CH^O4ryXx8bHjc`-3=K)G0sB
zJ^MZXSNAm-?Poou?b>0I`d;y7pm+8=!7i`KzistKaa!T?1}?v)0$Tn9C%?_&8(i(d
z`b119`czkAD;iIX+=>S5!-N+ELvgb!v*@r^<AYcc%BNxxi5gW}bScPu&Vw{at^*1H
z3eCFHaY2!{70r(#B+KOmsiB<<aZ%b+X*v>$Y9W*$N+j@@?veA2->Yt(wB`jEcs>gF
z=HvNeZZ{(T-Wkf*o!_{`6P^F4&f=~O#<4vjz}4pP2>F$cN4A;p4RZ=r|8sIsuvs1l
zW9T8me#Q|WD{<@{2r#u(dRB?hKJ%sn08o)`jKc1SE}e2;<1{|ZpE&=*9hylZ+y|=R
zHoC*EB}wWO?jl_T&8W)YMZtq)8nd${nHcpma0(ldG{3HP{l?vegs_`QshvsrqbEqL
zN4(_+`p87;#1t?Ep{{JG{*1#o0zh3w0`CZ+$_T+H;d*0RKQ7Ee%uZjFd~P`eR1>7`
zpW5K`FIE^%pH-OfMbN|PpSusNMj1oGFX(nc1SD7DADR=lh1y!q4wC((RcOd)8AX|l
zrdW^-;sPwT$=GILX1z|&FXKk24$M}!QX<IQy{XrAZF7Ym*2-BH0Kwz<_CJ12BS85m
z@s{J}WF}IJZ85bwIvjtXmXzZHM6@J9Aqh1ojA!F)>E#9pLSF}`IKL7G&?ve#gba&C
zXfJkWcLRkk{FS)|a6ctfC^is%`1Y#_leqQlBx0NGj_BxTi}N?FaU08Xmu+%3>A|N2
z586O;vgPv%=@5Xv|4&+cj@$e{X))tWr%V&5+xEu<5brDI!Koc{2i`ZgUY%5a7(rUd
z?q8MbhP1xvPZqrqGB0F@DRI@mmqHssiccaA?$OCi8tQ!%%kIwpi6Y3*Ow=TD2+y#m
z6T>{QGpSW0odW>KFngcd)lzu7ee~EmuXMa7(hgj?Gzz#9sap3L6>De1l_31GcobIC
zC3Z{)yk9dgUj3I0oBl(Fh85_eTK^@(t;5ZdOPMbWm8^|F5w&;!7Z-ZKM*f!zXX0AU
zmM$C>kn{-ihar~(zkeKsZKJVDHi#`nXDGwZV$|mI|3sd+k7Q4G{3|&u>%|$oo46a8
zg9|zMKcNt~lcEkd0<P03vOb)w)E@Gh2W6sU?(JSv<zYLO)8eB!E#ay|2Pln4C;!Oq
zv^h*e6?81l>#j5+Vv=wqZ*2Rrd~7QjKvPa{&OLN?ct|qf+S7D();u&7togQ&H*-72
zqnMgOrMT1MINUTBEi2H2ou2Xhf$V4@%#u#GbMG<j6}o_Cp)Qv7Zgz%toS=MIKQ9WO
z@*-DYD-K3(1H~j*KMPwZgTh=<?0rOghU?eq-6UP>vCpUQn%l?a&TIL0_a%ZS_}8TI
zsvk-Y3(l4!0R-21Ztmr0`^(Ar19>u@j_U|VhETioR0pk0wYMf6dT3rc7Y;Y|^#&pT
zqCQZq%@n}RnZ<?BcwTcj|Jn@JdinUaxkcyYa!)tSAc@v}zJKlqZucTjA4`4Y5i4v*
zt{}iNBme~RCcGEf8&wzBwrYZQ|AT)X%Da<oc5(CYaB&K=RIqKNWU~+Sj7y1s^<<Po
z*v#VkqHkWk7#EL#n>S2qJ@CQDoG|$Y!Cj3xT)_X5U~A7e2$mp3gXWoRmXsV+Ax3Hp
zcuMjqg=bRj#2Cn+t>+aL;q%t|bE}=_%RfvIm&S-B>G6mHq|uC&anocsv|pq%k}O~L
ze#}U>s<g+#bsjy~u;=_21#Qm;oG^6%-D2sDU$S3E*tqR-FGQbh?!SmQjR3&pX={%G
z#p#0;AwJX{uZsHj*q~G%7x>a{{fD1eT&(z+!y5~;&Tsek*~k`AI4j2sns_PRzfH0D
ze1HDTT%D@5jzt?;ah8LAYjf<C0E1M@*843VaQi;~(>j5qWB5=MxE01Ezw4%$jJvtW
zL}rxg8dM@Q{C%Z+M|4W+bsyl3OayCy9Ymv?0y8W6<k2(7>)`b7?@1KN$y$(eXI<Ne
zF4$Uz{aSf82rFG2FrJD;61C=o`NfNi#YwW^_OB=Uleo)C{u<95yXTilh{o&a!UvlZ
z|9dMv``zB^c*khY(U<Zk#C^nl>s<l@qgTt~s<i6x5B07;?qI+XR388_bCcvj`~T43
z{}&8~P=CYVzU}`%7`#|FF@mC(n0QJF*)+hgU3$Y-x+9Vlu0|xyel|1RmW<LY(za_h
z&<b_4$UdjTy*>d9deTJdi3rd!fONOEcSXbVp{mElqFuO(5D&lx!vtG|^c~C)AV{Kq
zpLzO1i61C<=&fOnu!cZNWvPeGeyWnkvg35YzYGJ|6XCwl(H5ucPhUce!i;Ff1-?F%
zAH9RIPY~mDT%o7l15!UedFm$eWQRTapfPB3(=!lXasdBG-uoTEs*jSlZo0=H1mF&C
z{`SZ;iu*NN`0R*~pN3!Dmon!6Kc}AE@C?bAx1Pl;ZjT@N(tVh&h1vUG=cFpgnlxVb
z{Bk2r9+)qtydz$M?-vc9JYUlqc9UkKBIZojnnw!`7AeQ{B$ZOI$RNGv4jQ@+J0K^m
zh>zqulxqI7K~3&k<+t+XQgJ)or`7w8Mr#5f5lgU*3-P3v;HAi7yF<q`!l6TQ>iad>
zn3B#tgSY{{0VD2L@es{jBnrOuDgGnLM-;Q}v$EN~r(~EDRpF9)SUCkf%&=nu&q?8M
zcrq#z2++~k*-9Gl!YdRdyCJ`X07U3*Y3V-lEFbF|sSL`AqKV)4{ZNFs#ndzV`Kaa3
zF0`B!Mn~PhSxzxJ3Jjn~V`R=qZX)}ALkgU62)3pA-FBQ7m{_=s(YdH~xR2_dfp*ik
zMyVn-*=el^k^@sROjNfDUdrM&!}no7yb%4dtC;s15?$7Y3FF<eklBlh1Ni6h*WU=C
znC+vs7LToky$W=j)a&)4OuN!seT@ONe{joE=Mow5Gk@sW4;VBqTxY#}VxVgKkluiE
zeuF|)vd%thJ&X;9uB8M$G3b<FpZPnjSeWk%pYCzXq?}+GNyo7eNcZFv23X1*_V>=*
zzPE1TsWlaM%VAI!6z^S01-LWMOTOx0Z(U||grAzrg$sq<m97de!%04i1WrU?s=-W-
zZ8>VQqV~gS6AcA&_&yNTK%s@4Z<H6jKwFDB6kiSor?+HX2J=wMCBnT(aM$#Nevn~6
zb)0B|r+B;Hit^y<{k|e1S%w~>og)i;%S7T2Zqlp>;AU#909T@%1w^30hk%C|SDH(Y
zW)v(pH5*OYm!Pi-LCtP!&3T9&*Ob2g_ycRor^VHeeE#shFWb1fs-za&x$}%VdQ*;}
z0GffyJAZe8Ir$u{mT;3I;*j>6fd(}NhT!1t;OmE77Wda*X&w%Sal{u~{bqJPXtg@p
z!(jjR`j{UazcHzk*#H7S)iT}0=qHxp0c}~w5#m~}`TRmPn&j{zuvy&O2{6uE)G+iX
zPMmy5fE5j5(Tvr95Ha`ae-V-1|G$Wc_&IXX@gd@pVXHS8b_@5z$jKa2Y;A@6vB&pz
zfj6I?`84?j_up;H=j5PBqYeXX2O+fSjGg-IC7|`^Q8T?G(FZn;%Lv38yFWI+9Sml(
zGzjD`LJ6`{+{+?%M(lTX!`LtK`d)})tB>1)98)1@zT9z~c1>@ap4N^vpJ*bX7mEq-
zj~(<7wCw=bB1L}D22^1!R+rX=ghEi+7iA`zdZap)quB}X&=T3%m61x{CVs`>Yp?+J
zV}O+Uu~aStQ5*xZT@Q%YuXxU10%l;e)T!70H^=oJ$Bg+vJ7AP7_UlfWMpY0F`uTqX
z0$l+V?M~`b+Ahr_;ISUWP_$U4r?l$K9~FNILi5p6g=q5ph~r|6O&{9XgC&pIvagc2
zbfejDj*a$7f-JnGimV@ytvM71R+j18fZ_0mk0k}#5FbUHEWFHXy>Tzd96Y$!VozmG
zzVA;vo#Hpq&m(jKE@AYXczz|SJz(e!ZyaAUrcmtJ34a)7kN%QFlLE>FraldLut~HE
zwcs^T8R<;foY`WIbnbohY+;yOJQ}_qF$k)~uIqr=CCU2|>ty?k$w8%zBs`k#14>mO
zJ8j>rhj430;?AJFJq0V`n~oc=U!Kt0&`u9z!^)Kg_wdN@rROqJ+ib^9Yuu&MFZTOD
zm-T&-1I@Yj2)OpX)Fj-8;pL(zo_nelt#o}PG4`G;vAVTFKZ@+DwD<E>uSn<a+|63r
z@wr@l?u=>I@X1l|G9Uz_2Gbxl0GpQ!6&EOk3mq+Csj3?d`g4q?fphQ(tJY?-yVNXa
zhoYTI&_SlX8EIY<>s~}tA7-GNZYRLDiYS>`ji{CPpA=8!Tu|8tTp|p@JDbb2T#R%f
z=1>Zu)vM9V(NW4rZDH)Duvaga>(AsGvq7IK-a@RsxUGfn+CRYRIRmXt0o4a9EXMoH
zi)H3zVTq!W<02Ns^eIqhzJtQBoUcMfsKeWel1-{$hinGa)56Lna>~sTQ-N8je?N)s
zvfvD_$mzUN+;ZC3Kh}-?lSy1)ppH5AY1t@cO_-;E8~8_^^6AXvljw=2iKfnd`kV%x
zwqMn5fm~X)iJhCUNxa4RfM3?L4#R?u1AwD;RFTp1o7+nWQssP~5~chM#PsrGk-3J0
z4LRbqP;7)867}<`ufIQ)9D~E!Q2eTeV-*qOIcU(T<3?{X5W+@9tIQ^gTPjG}>62z(
z7RqjRcqB^{Odk27z7hEIhmyE2-jBBYQodPqk;zurNvewaL^#S(fPHu?yoc#F+^%<j
zqb(Mj{t98r&3=*1+%NvLeyAF3g*JI2Ke*u{6^0~9R*n|ql@3^;(#;y4AF%MU%>@Lz
z!{@>eNW0K!@J_04QA@sFR)`sI8<4RJ+WAW0J@K@5+P{;y@4ztn<7ij#bYYIyvuzew
zNtsADV$t`<4x+-10c&cEy5Vc(rm<9jPhw4>AAIAq11WITG#MElrk$y{qep!co#=^m
z%Dez;a#3oxS(mtF?ucVYo3<ry8^Sj2)G=sc;9!`=_kH_sHA6Td7|#g#%RCDaA;8?$
zjzoU#HzxLXd|z}H4s-i@1AFH|SI-43#vI>!5szHD2<q)Y0Ol;30JBKa4<%&*C5Sb&
zs@5zNnc1&gs`vr4LAH_S)eC18=b|^~yqm>XuIWV2Ih0&kSrxcPa)nS{ytRd0&zW`N
zhqlF|2p~LmzxH$0IVz1Ml%FPFCH9|PO%(dwy_>(Gj+n(&)*o00?jNO|8$6(PiY7HZ
z2@u8b#Xq#J0lME&JNz!$-ProN)3l++VXnjGNK2ooukr&#YDf%**0Rjb*Wp$}Ccfa1
z)ZOm^-{AZJA;-#NFrtPC&@|&z$nntkPcTQe!<;h>YhJ>3!z2e_By5p&8>^8DPL~l)
zhGi0L(X$^7P7y0EQ1zp9mJ&R^at3P7yE^q+z$xYkz$bFVCzZrFcT_jCRg&YCvCNRy
zDdvS3x^hQN3i??2gDSCOtr}x=Q#7nbID9&SsQthwC1WUgGAEJw0ZDwq%9orQb_-^D
zNeZf{A;i}t3d|ly#8ryYNLNIbJ9RM&EzGS9m{=Smj7s-Y6Rm;r%0l6{1|q?*Nkn{@
z@NdmTfc0U3PY33FZ(Z6By2lp0xRPb!nHboa$Xw7ia^cP2SQaYPgf2TnO%hAG%x-08
z4JXC;)wb0^MqEka-1iLa&7-%@cKYqIHgnx=*VfzgGEU|;ll~22gpz0uLtFWzwI;((
zT888bR1+t^LgJotXQa`p^ks<(MH@EINvpz5z_wC^U47eAWt`X6Q*SNapgcP#Ofb~1
z226^DX=*_=dP&SAdO|@45e<EOUfU8Ebe;>@s*TGYe+tj&fZ&R$`2b^LH@%b%h_oKK
z0}wD-9)=%CwsW5Qy8@?(kdb1~cm_OR&jmN)b}SSe@^MmDljzEc($I?7HK^bdLH>_<
z0Mk6(+5hbIW<(LY2vGD8+IRwf%ud^!U#XNWceFtfY_JYWN#z~|W7(X<72w#aT{nfN
z^T_kWpYI#SgL|~Q(o``?N@>m#M?cp{ZUg>i&w6PnnN@QzQNrfBB+sa<{^r+bb-edJ
zMkEj#HKsqOLhdpk6#JJFMQc@}Qq`mhP)KHdkFKrUihU*<Pj1w%UYe;V$3DDo`)X-F
zb>Tjc0)BYi)qAzG?o`j{Dob6MI<eE5CytE-2y5u4iay)(NVavv`#QMIU+VdEwUS|W
zPUncfRd3^I>pi?JZwlD$bf?kc$nVWl-Ec>yE9{{Tx4d0_*U`%vsrNd2k@cAX$|;9c
zWGC~K(KPEk)t7s72hpB5VggQO^q`**%8mvN`ATxpXE8i4DsFtAqSfsmzRhlLNdm7?
zbbaUGNvVi_gCFm{mT=1d8IEFxkUs@E5-xC~G0^CaKe4|5@&sXDr!dgqNwLL`>m#91
zvi&QwmlPm{)OsuWYBVuwMF6nj7-(3>%tj}=;<#E=a0{kTE;QPs6VvcyeTjyt_4@lW
z19@SAkX-2;Ai$Bxus~RaR%gv<URgLa^n13M^%We!W1HM&|1Bs_oMW2~L#W1is8{S`
ztK6EpUxGD{tkjdDXN4$-RzceiT6--*dnvg!aa?*rNz3aX)*~~KMS#Ou*n?cJSUuOz
z|AVe~4vsAB+J3`{t%>bSI1}5pCr&1|-7z}0ZF^$dwr$%d^FGh})pzRDsr!$vuCCg<
zs%zhCuWS9TwQhOqe@k^_95N2K8AN~*8BG?I-B%qP2A7)d+=mhFf+g2k<Q%wyfvM<l
zu1TLLciPW9Vn?y>N?!}qc8t%F{Zm&L{pXRJ4^1fms-18ywfL6;UE_4BC5{=20!NEk
zpqba)4>ejWTT^_jhOPqlqScnxHb7fTv#<wgO?&8Wg~ujw&}vCA5S#c>RI-oeU41xE
z{*LTDQ%}&p{0xV}755pYN%dA<t8#cvaDH0aKm5q{r3*o$RuuiUj3TlnioNJzL9g{j
z*BU>=@nq}!Kb|uH@gJV!eV3xRCw(2S%lu-Eq{N*?^;byB!XACf2k=kNS1!w1m$RQ3
zDyTT}i2Ya5(^i34sD72BNcmYBQZ`+QvEAQ92RuJc7jGH^{&qexg@SPm?ISV9ku`7f
zFW+4Dxg4oRLza0$xp)yN2-Of2?(9(A$CPgVi6j;*gBHO0pS<lWReTdh-`(e7^L$`p
zQc*h#xB&314(`5&hU@;^#j3qn>ai<YF#8GW>#(&R>(=##Fm8JU-tZI^t1zVpoST{R
zFK8v{mL{gian0tXNzCPh|J59V&9e!%CH~y?gY{sB{NWzy=K1ZjlI{4v%K!LF`TuSF
zTb3X-5Pu0_V0@5HaC=xHBKE(472r9|;d10^fT_}djmoRDLZ)hp)g^sfZz$Zbdg^r%
zQ<k>G6J=ERL_yqP`2)sIAG0F#Z5>t#hOkFzXvvCw@QVa(=(yCmxILD5%u;4uSik1a
zlaI?3ULt`yg<JD(MyIu&THH)pw|+LO5n;)EeBSKJCfmmnY#fy&aVZYp1Kc{7TQ2A)
zOrxSYX#rO<&FgnufUO&j))fW2F^P%C%MZp6#tF=iYo-ISF)hIZp+0XqT_VrNk|7tf
zzP|%B!{@A{gAEsc6F1);Ypr3F$M9n7SaPw2%gr;{DX<lLdMQ8hm~l|jcon3khm%)W
z9a^j&I&P&)x;q*hJOk@J0RdC`sK3ttZdI{AgPde3tQQKmPF~}ia@-W#gQDj(2v@o$
z9xM}$*13p~1>Zx_k4s1Yhs=G|n)it1yuECBW9@sIyCe&|!xO0m&X~ZTWR?2hw?yBx
z%ztAt!2!f+ez!A&t6xOt&%(8CIIr-IS5Lh18^$?LvBdi?i2<$?(wm{-K^2Kowjuf{
zguC9*3^E1Aw!6Xt)2L<GCBFr;)5BpQmS+b_^Ok4#!>+g_{VdOxsgMiQupf(s62Uk%
z^3`kH#&b*b^w)S1Vs2i#@eOU5P(P)^ZBsvOiy8!U%7~Xcee<DR;Jj3nF|4Rt6Kq_x
zD)8~tt6;I_*#ayGZC%$Ky?N1rsix=qyp4(+V$g9;hJ-~=IKD(ua*Sq+k!7&x&TDm-
z9#FZy3(Oh&@DG+Ac$vbMn~M_OR>X#4RE9hk_UPIqM9Hmeg&->qm6oL8@zrqt12uHd
zw9mi>ge1AM^`DfiXtGg426a&KW*6CBFWZsJc(1($0Kt1Cd?k$Wh~+P0!w7E3QngO-
zA7oQ!p|s>vvic9(sQaUM{YR5;@dfb`1%=i8ZYJL8)}1pI^6sjFFF2DL6REXKbvQyY
zIWLeu|K`!=1?drr`oHWZTOA2}nmG-Pf;3)T)^0@d_6<i=ihlpnK5wrB@Lw`Rx8piz
znQQS4jz2lYB+(~g>iHd-jtIq%$0EX}u2{sba62P+TpOkBr+$MR?D<RpdgvH0W#1*d
zMg>Ioy7TozE4_Sf9ytyo6ymV@G<<Gq-lC45Qdy!!1F7v%aH3<@%i3?k?{y(6fWORz
zA__#P4jwwuloc-ufINjDmVJkqaM2IaPY!TDy2FCgA-KIjac5hMrWHttBsu3WWmP#z
zMDQSET2vJP@;n-lnNh*p-)aEXb9jisvQ9RunBcS-l47_>ov)ODO|C1fhnPuv@JHcx
z$_k2#<4rk7lx1>L2Chz8kqCCo1b;ay(j6YN6nwucwK-`sfIIkh=j3@v!R{=y-e@W(
zXk}b;z?@3DJzr0@@@%BgiLy{z6w6SI76=v&_J#<iB>=iH4IF^AJ<Ro-_Ho>h`!#4d
ztB$JAjux$;`ItZ;ax8M2_QG-9JAjRP*i>>8DmeNlD=gBdro`fpsgh7ZUmK+qQ6B23
zWh%2FSkwb9pyX0XcL3h&=d%bBO8I<fzyhJL+c#OSyFq<vn7U=E4yumiQPFE{@~;nX
zB#nOMhtKRQ<Ee+1oNCL3q?7opzKLk*K^Y8xk2(43U-9A7zVR<GyB_g+@p|yr3Vw?c
zjC07Fq&x#7VT$t2RNBgz9u|XA0$&Dn*tQ%x2?0I8h|gKFSwwLHF=$9yX|ieKLL~vX
zAaVhA!SbOW@*Z<)Z6k*x%HCvv5Y*SHG+NM^ZYOygll-hCsgx@r{lESxB!?$TmG{;G
zRS5*+d8XVXRaoVRtF{CtASg3I=4jsfp|P^gp0%b)jc1-u(_y`qZ56-tk#)FN_BGJL
z^BTbVi~ouKzV_nr0#&sd+O0MkY8sQTwZ;4uP6owkR2Oz_O=dhj*Q3+V?IWQCS$U!r
zSt;KUNENX{sWSPgsY?Wr#21nZ#hIbOSoxiyIXg|<dpcZq$anQ?($_Q+nG*N2n(zr1
z*2=&5B>38)te1Q8|0*Z~T(0|z7p|&qsAAYqZ;W%OTV6?iU4Sr?A;4cFrssSU@a{W|
z({hHGjEk3BucpFMq#35biy@*W3jt^m1Bo!mhE`$h6)TjZ$d1H7(jI<mtAj$7DRTmo
z*(AY#eF%~ep|nAeJcpV(PbZ(a%l@>uUhfQ$lFF~YY}!rt;}B^8z=7Z19k197P$9sy
z%Lz&qyCotz_PV`2QgI6*#C_Y_Z!<>22SuM%F#vA3!yb@X*jRI5L>cX+SvmTMoN7H;
z#&-MBQ?0OKyN+3<EvbnZf`C+{Fom#_0XdH_IH6Ch+0K3<Qml1x^IpstV$$<Ntrduo
z*pdia*j6nqp5o#H@+P;VdGMV^h8};~3iNps)R4j@Y|r4$vQRaq#H<#)*e%)Pt{$Py
zmTF)qEL-D(u*4*2MEoJemAKHXQ?v;E`5R>V$vS;I3U0tdNW*}8gAPOhYzt>)e1{)>
zoi4=spc2K0_Agjz#O)&7POGLO%eE?*kh=_3fM5!9j<oA504I=HKi}{|y(=hw8I;$v
zg8yd7ko3bS<}>Vs)oT`8`_eu12czh*ta{63lHvi%6j-X<+Bk}jF@{}=aP_A#@l?H6
zjToZiQ#HBI+R4GYO&b3YS`eO1+7QXfsv2hQDm1i?Bkn9)8FV9)ZCk)=M9wV55qmBc
zDAeXque+HS;O&n@fHy5P=P5IA8C7ogCEos~XT1QJiBVp`_N)vW<oo@b=-12WbvjwB
z5($rFi$_uZ0Q|A?qzMQ2pQEnlEbby35>|=RZn?L>^0$c?6HszvssGv)f5V63xJbG;
z>E(*f@Cr6bc}>z)3q$1f2&{urU(w@{sFoD+Y-t;??w@QIARA=6Lc~bOo913HZMq0L
zZrsZ(X`>S}B_hxDW+!moHIkj_kDf=LHXs7()B6y1aDV<y_5Iv(Fq^WU^VN%+<L3r7
zkzdjY!n3WlJ1nV?(HE3a^ep>%v-|#BvtR=2Z0mo@&ckPZVh>KcmD{c?4|6!ZxXf_p
zBF_cL*$op`-+od}#27riXcrk2l^n?44SQ4Xj*7t3?S!P!5*u~$G>qB+NwCu0@9ig_
z@V2VRm}>BqJy?8c!m1FhBW?LhgpwPwRq^%jH1X2AXznR4+EMBT82)uH-D2eCLX$lx
z>y6fiZ3wrfU>9`#(W;P}19l_|q9pV>8Gis!KDHmKnb*6Dd4)aEbHHP@_eHJX=%biz
z1lwZgN=}ZK7RlD+Z>WBh2(?^gTYK_YmArwB>$Mu_9e*VjG_ExiOVHdm;)M=`-q<k{
z4-g^QqIRLW)mRJAO^u;Udrg_HB&T9EB@XUm3N??U1d4<%c6nJ?1y;fmNnn@{uA%`{
z&vIXi)ukUT?VKhMy9d3wjzU51g}3xN_8j#b(C%`kf&~d-$B!&he$<<$BL%W$<RAr_
zwaYL@X%Vsu>qTNbhwlAsHI;VaF`+W#Ec3l<8TrvUi1gQBwhF#9BK+4otfY`{z;o2$
zZ+U;%mJn^r2?Xv{zcuwOQ6EeJ?b=@e9>!|FiGYOlj>uNA22HdnOV+c8{c&2<7HIQe
zKKf#Ob!=t(W3pZ#xF_g+qd0(FYqNTBakV+JMi%(%;g#Lav_**@!se3tsH-{c*LUPE
z0ROa;f!`O6yj;_Vs9lpm@Jn<qH;-Wi+)6|FnZBcGA}Y4za}h@c7=$GP$-@XB6KsPz
zcszr)Qp<q~^0a&QBluoRnuw44q)}U+rkTgshzVVvKlO|BfqexDh3v!*!URz(<p(d1
z<#Raogf@@F*?0f}wxwX-ov|Ma=(cJOO>AWeI`b9JVZeH|p1bxE9IXf^=x=?{WTcaT
zIT;vSef@CMJ@CxrRU{N6=^O%p3oTOyMbD)x`GkSE^;mZo?m1i#tm<0w^})r9yUb5K
ze&<JrN8|R<1(3twix1=)nfK7hA)8KKc0xB@pBM@FSz;kc(bRM43Stk3rycL_1;o8G
z@&aZ{syNO~1;7t}oHx)a_F*dt3f5&(98-SOms>T<F+ET4^nvNrja3prAt%04FK9Zf
zhY-ehMGxPVea*Z%%Ywv}qzaKM(m0Fi{`)nw;rTkS;P|0t|9NJn--VSIcC<ID9x$(}
zV&8em=?YdU5m0oFQaO;3ln-J%+9D84M86{$MGbsMB5L?@APXHOyMiE68Fb>83ocLH
zN4Y>hRJcvu%c&Dh-zVV!HZU^er?;hub*lIr{603o>bS7>!v;E7?H_&HUE2CcuN4!x
zlS9iLnT<`p2RK|iHQ0ke<=8Isfap_(PIh9f%~k!r#Z8_qE0vix1A%ydDDcm0Qax>W
zkw8Bvsm#8?WVmHh8^;}iTR$8AOl5&5IIyK?3AMn`g;l3jOUrlwU<xPBHu!ye311Pp
zoG7k%IS&vJ<K&;Q=<#5xKzS9iv93Y4*(7m|Cdm4B{##4>2FnJS@HbMM56O3~4s+fF
z<2)y@+eF;e`t)UTwmPgZFZR6auy{lEu<FoB#nZ$zv3c=*bayC^CU8bYT-Mvo5hj5#
zrNCyN)ysG~9ZSzsz=UeGhY+*}Q120!Z9;z#c671pf&o&#F7s-`qM0T&Ca0LR4f+!C
z=kr{O^t*8EmsF*2UI{EKJpo$9u`mM^h(3}Z7267wcDvo~_>MYgqL+*Lw%T9&a<89F
zo@qZcY~~2BoL@|Ka~5-S_cw%5Uc3xUkA)KUkyH|2sRO$K0V}Oj89O^`J`S<l!KS_f
z78$0)&NbOZT2GS~`(u`UUMOdue6vOuzu_K_A-u)pgU4~Z89ezl-^o+xzOqUq6<ub(
z1cdVjc{vUx*$tl}Z@)m<j&HD$KLD16#eRmTiRf_8dTzf%u#f>#me85XIP-Uzz?Fej
zdp3D)NKzdD`Wif~n(PJA28f!_1yXIvEq_56$6H*fmJ0(S1{kR~hxC=w#&>llb<@^Y
zP&6CgdDtw|U5i~FvT&$_Ce67N&2L*2Pu2&>Bex9T(dC|Yc3`ABTQ@7CM-;7A#Goxr
z!dx08m+{_To@%%E0*ScahayEEw|Bn=a}#18g*U_iL50s=<?5R@`2F@vlnp3p(xIO^
z1Y7?%0(o<ChTNTc6}u=J@acz-6Zlgv7+Qr^Zz{+7UkHLyBK?0ckOI<?JFAiJ(RE`c
zH)YJ3o@AvZtr@&<M)@_x0}Imr@6z}UP3jDXRbp!L6TcAvHvQ2HRL`Lj-gS0!Ksr5$
zI&0|@=~3NS#S#@x)Yy6pH6HtQA2~TH^%?K{6PKGWE@F=Hzg#3Eh46p5h?(0L7eR9Z
zW$30w)h8|Og|~b1)UFkKj~i3{YT~OTTv9O~FOwS<3^xgj4O+S%J&qNd$hZ@u0Wi<#
z_@dV_1s$XFNEI<IeE|#&qA&f7^2bQnp;I~KGgWLqG?QW^hwQW`M#xXdA(m)n#v+U<
zY|E`a^UVq`<k4e$oFx2siGQdLq*?wzu&Zd7xB>ZsZpm7%1LYc?h?mJNBB1c-wg-m(
zk`-I=%fM;a9Vo_W6opvHD-KBtU_6auCrxc1ZIxS3`>?9zH@u)U%NAjNj1T{3xo(os
zwzshh?y3*@@?2jf><)oYW6MjRfXgpa4ukxEqx(?v^y$N;J6kd6-U5iuL8h84ZE80r
ze%*a6>BsHc*(>$*IFtg1Txw$HBnuMwN04$fARU$fuJ3btxlmW{lb-SdpKZDLzI^%T
z@{5^K9~X6S46BYjLY<<tG7>h4|5i7iyE0NSL6dv+*kLY{))ex^0q9|fI7MFS5F)iF
z@AqEpx6x9;RwP@ndr{2$AoG*5yP(cfpHfmG&=4^gHK!+d{&Qct&<D;XB)Att)iBIC
z0Io1&K+FK6KcTG)GnS-wqj8jnYr&j(yt#J%T)JY+DLBx?0T?=AEWyDdPNe~}|0@e9
zudT*<h(4=(Mq1n(ZU0)6kejkht^YaybX@^f6y*{ys${M(cgu_3L5HUiSPolVirQT|
zLQxK@O9w0;VJ>TXaN@0clAY@TjnjYu<Ypso7fnFSivJ&~z>GGO@>qA;5b9w5Y%t~B
z@w^6WFK<3hF@ro>^XF<!D&2zhRL!pvThS_`<yPY$I7x!MEn2D8ek6>M6kVOFM_4AP
zd^KrWi_mDG8fjOWY=P6DB9cW74NgdgEK!b*VeYQv7t${w^cA!E=R$ze?U1lW3i~D+
zG`mIWW%-Zv;c+JW>z%iPg%5#~dYpSsaI6WMY6;D7|65P#GnplReMGEEkYL&a=7^LY
zLn?PHkhss(&eFZ%Dx`@)1Wc0f2C_1x!hKX4FXfl8SIRWYA(4J@2dV1jn|QHwU_ND~
zthyU<_!nAy%l>?3F#~98$A(Sv%)Fc2zEm!EzIZxBgNFGHKt#ac#GQu^uXsHbw?X5e
z{iHCJo=Uc~mY%W(X~dm2g0>>6lofmaJug*gs1VsjH|)kJkfxhb%TZGMsqKU16;m4{
z@tEcCV^R;^)a?_)+->frYU;+em~ia}f=R-G_lIYiV_mwUI4NM{lFWcCIzb%73StGi
z9>v#adJVd~9tEX@q#nhTbV^h_n0F=QY$YURg(0<<ujGUd&)*Kpe}T9NyObCInG32e
z&0}|lS{x@uqBm`@HY*8HE_7P_Vaz+9Z(|})woKEiA=ee%TkL8G7r2eWnk}0y)_%cv
zQR%*p_&j6zxdCW9v$j3$0^6!=Z!i98130i&_pvIDGxpWT!L3gflSmSrVYskf;jV7_
z|EaBiWVK&jN@__gzSA_^pJ})v$_NQttuJOm0(mNH&Ux!MQX1SJodV@e2cj7zOdB00
zJRg6nND#rmWDAO8Onw0J8UsK|+%q3d+V3&RI?_=)1qo45b=e0pIvaTb?V1DmRj4LI
z_S5WKJXHd+y@p0K5pkmJG^kW(D*_2@(TL7!8NfOU8q9mI6ZcJA%9SXZ=99HG$Ck~d
zGJir3k3F`JKRB%_kLKL(ifDt=%BolMFsQubEbNCe74tI`cW?mik*i#8e&x`^Tl9*>
z8+ylH7cE`}w)It6_oEwlYBFKcoqii5#=n<OIr}b}ojsc+KsMkxC@h8&9l{MTNzuJ%
zVy>4)dD7qyX&L`G*R|O&jAkfRrkTlWN~Xt620yyL6moy9x`&j@r5O>~fFf6#@82EZ
zW4-{AkPOG<=nQ}p>b|a}OBJ!wWCdAAn7T@+Vd`WY{7Q?foDGo-EE*D}@O;+eWFlND
zQvl+oAYSkMJ*mx>k#2*KikzG!r-nO<cXC-x(jdN$6G<5<H|Pm2qSMt$9R*g$k-CE@
z+LclrIa!h#Z=5**17P7)#ZbkU5nTi|l17Oj=B{q@eGY)4{A9dfUXN5lBU|k3xxQi;
zdl~M5Y2&y=nbV9ES-arR3=x$?Kf^oCQ%L#^{DJMY#igy3_RT0T<|=9SsO}7-sgQN?
zW;aPI=}SuCd5&)YU_Q4nHvfJL42dsNB&$%9xJ+4;fYf+TAm>DQ*gxYE{-~e$U^n_)
z{r%AxJ_3mBP$K90-1)q7UZRW`xx;$CyyU)+CfwP|A0_|`^{VEcu;**;YlO`l4B>r#
zh>&m3*HX(PAV~%4jXz}1E-mJ77Pr%07rxxezCO%S9t;WDl~pwPpX+m!6=>~4u8!K<
zERlm__AtqqhP<14t>A7toOPxBI2aG(S#Bot*akHE#^~*(V#CDx8ln6^)ImEqV~}Z1
z$cODxt#;~Jn7=^_hag=X9E!PkuGd*MJ6D!S&+y2mJZg4x?@-q_kF&9$|Fq(Mg%<UT
zu?R8WP`5x*v}U6={N3vK5KB#uS)EQn=akSG0F8KR;3EB8_em-cOL?{zYanh%jw;lN
zkptM2%|!|?==<)DKQVBUj&KGVr0V;%G{j)EXtI~35%5CV*ZUrvbi6j64f}Mz3;bn%
zkmGtl%0q9BVoBUwM8@jx?A|n~0W9A`^lw}DGJa&me?YVG|2c}<<C-EYm?v~&#)U4n
z<XW?TLF#mi3u@c>G56@~i<kX2!t=AA6CiU$p;kYWmdr5`faQ-5EuH6O>*-B}4~{z9
zqWR5eA_^~WNQDYTsqX~zId1>+p^}KJI*8aCUJCo*H`ubzZUsjGg-PnBYcY1qAC)6Q
z()EkQm&hzORo<Dzzne|!Y94`3;qkZ?aQ#yj+Fqw;=U$&1PfTc>BP`IEG-?mdTL3ny
z3pRPyciL-wP8@Z{Nf+n?F--HZ)4(Pv$477OJui}QD0A)(3Ub^`G*4&84PGd-g$hp8
zsCJTheZ+$YWZ0YpV{VS_M^qtc#g{A3PN$B_ysAatU55PJ)kzG0i8ic^tNSvu&k>g6
z*){gb^e|6uxy(<7ZN%d{G@9!zLxA;xJpND-2viC=V&z543k$f5HHMF)-o^W5*tDhj
zE0)u9RN*Q)52vk`WXrf_yJ{8cETvqF*boDm+FFt25awII>W9#7u~V|yRNQ5}<+hQE
zExO^58W}Dxeq?%T&Z|Bm)e1VZoPj}fg1haq;PkO`r{nPcFa3O3qwep6a)5)>m-{G>
ze&FQ+LxwGW`~17(YqNuw5bvgUEAcs^S<GZb+Qt56*<PE|ED6^7!Xcvmt^KUIE=K)E
z>qtz;4-p0LhWgAt=V!ZnPm`a^#Sg9TtzM}wQM!hH+zN)$Ar3aX<YHqc71u_2Llfg@
z53*&~Im;5wklXzYW50<nx&YKR+v0uY7bu(h7ogk58J{JOuYGxZC1t;j#Cf9nUkN-m
z0l9O=Ssr1YR$mz)9@3cI`=fy?%lqEjn#>KpuN$2$|3zVbGopk1@V&3{uoe2zM!72$
zoVhmHgg_hfFvJln(mLufX=qSF8tuwNpFERzrJ;(0Z`bfQqrxs?A;4ql@t#Aiw+jL`
zr_oX3_>|eu6tNIxfxBRkb%QrHvm<aw?Hapf?4-2UdAIjE#W9(Z+qAa^kN(BLIr8N7
zUX>?xSZ(aQ$T`P|%91Rdpta8sNpEUoVtM%WccDt#W0s?NJD&@@LfN2B>Ic3OF~My~
z8`owtuE*WGkrG+!7C?OiKMekIt(Euv`NAZxu>BA~VU3(*vCh4pPL%l;ms(Nk@-)FQ
z!rK>~bvg}{`@9-Ht63c8;NUWdQWJkaYzaMhzR;F?l-`wfxx5sHq@o;4ls6zf6tE6B
z@CZ`{pG&-nz+<VkzB!@G{U~`IEAnLIF-UQ?r6;O`ID^fw20-Z17a^ZX6f_RLDXtST
zIAn7;L-i32Iq|U^ibX2KOBpS25Or`Ecc8{2G)0xIw19?g@Wa=TAfCnMSl=iay)tZ`
z2aa{8`smz)!>MT#&ZMeO!>J(~BaVFA0}3G4H_CRs9w7VJz-)E7dsJig{8;&#e&*+<
zUb)I!I+!@d4}f5E@ZQsh3IR13vwOjZI|Jcj-T_Aj?@PY4l3wlcCEOD0*#!B;|5eI?
zd~G&<$l>maE(n*h+H*ySh9J`rMnxb4l~5#OO5dKUq5SDKs|ZDvTS<UoN1u*zMY^aB
zDbyAH=O^@<O0*GmY?n5q#0(kC`g{n5t(riL2CxeW6z~g^OjQ6enp7lGq6Pkg`l-f;
z&{n29>L*{J8FDoE-jE`}8cU!ZC?Dn(_yFXa_-*f*(kdImdMQUcK%VqO4+`i*Qism8
z<Z$z2cng|6c4sJ?IY+_|tro#*xRA6vRC+#}X^-GJju-EpwefuaTK~PAbOnFlK(Hlp
zCc!C_AAltr!(GSp2*LRa1TYWd7<*FvY%(=GFrk(nmm;BYoN8-xz~A1`%3=O<bUdBV
z%*gz&+@Q=O>+%uy-qObX_IHLI&2MiYd=w*wD_CHq;Bz6xOn)F>;mpC;#LFUt(O^sw
zt4R&S{Jyh#K%Rqo=QoJuCPCCgEuqwq3c#ne`vD$6_-0l}=jlOSz~97O36;#Nw?r&6
zp&N!CFCz%~_%Y-IS%_L6q2pn`qjIMgcz7qRc%96AsJ&<Y)h^`&B264i&alJm5XwWN
zkB~_!-$3V<CsW<nk`$px^r%Tp33@a6oI+M4F91u{%WT`aqDq3K<qau-v)PjT2|hH-
zKLBpo(p}NghH>HZewHjDV7W%9e{jltsfb3TL6WS;S?rWhxkEo<42*z5nO9iEt;G1o
zIJ;dc5FDrk{H072WoQlA{=z{FW%~%xo8R_KQkUCCh(PR@19;|h59XN&qLa{{*~A|C
zbx8IcRz~=jZIG^pjhXsMqm-h<L_Rj>ZUCmWfnfO_TyrlXrzf~~Eh|GvG{dsI5j^(p
zoX629&iu8-UYC5zg=y;jG7-uq3DWkSrg#mAlXXKeo+dE%rYqaQg2TiJ)T+YYvJk~&
z$EQf236C@7P7^X9M&yPMn3;wod792TI5MmsrINIC;xcHk!>xMR@PC(!stivIuYJ|O
zP6;1@<X^{S;!iJ>3uvy(HD+-wvhCgAn#BxoG@CV4asIM@tq^+Bwj{ZLH(_tFxEt<2
zzh55TPddF*j?Wt7`r3gM3HGrQ_7ybGTU>dP`6=FwfJQl`&ua_mbjp$NQ2Prj50#!a
zf-+HAa1+YLM{G*|qWLG}-^H-1CV-wWi<Cx{hslv++`pp;&1Ps!WgdMl;!&v9ug_WT
zopRhSaV;$(L36iv9w#-A@05~O4;-#Rl7n78E@gNCE0{rD|J|y5nZVsn$jHubgM^WM
zcmVbf+qZ4(i5M3456Rw9JirayFwcBWId}?3*S4X-bS{p?c-i>h>LjSlKLEip3?QLk
znV)Tj8Pi8g|9saYJ0F*6Ccab=pch*^#f6j<rFNZ3!H9|N44(LF&d%Bs9n%QeY5|w_
zK9-`!p)Y5xg8qjur#b&eDY#QUgu{m&XNu{F^MbVXP+KUPX|$LaK|8#^5P_E_zk^ri
zNHKsk%cgTZ+Bw*fe%VX=2GHy@+y3T}kltvmxtF{8vPNu~Dcr%sL%m=0i}}R7l-6iY
z?UaBhuQJ4QxHAxpvtBP;%^`q@C<*!|0UsXv4}BanR2xli*~HC}W235*{TorDXc89B
z(yUCA{46O;&8xE4Z(x_{YCevB%!yCc-jrdnUCJE$mWa4=i?30DFrdaizHC-00R{aB
zgJ*SNnr#+pkuNE|;$}7p{fc0>sA)du0ZfzO1av(|4`~-IbsKD;SdgJiZ{87PPu_kY
z5~dTHj`Gyk5x6Klq|kp`{35QzoyO2?2h&kZ9RkFddvGV>xL?=R-n;M?wR_}X=CC2&
zY^m((Dfcw~xEWtp9|PvkYW`fK_C3YiDyR*QHEJY>>RG{ga>jsXWHJG~Je(ls=0(<L
zcsixdd7aEY%e>`Q>E{Xip_Ie|ZNLe)kd3$?`U-^B>%*W3Cj`1Fh)f;(v;stUdl;T|
zz@BwUm@ldvnYaE}W+f?jDt@oF0egsN$zeUaqS%t@O6wAgUIBbBQ_>+G&6_kpex90<
z^>!Z3WHfmCXusYSb1Kxp-lCRg_1lsr@-@#Zu$m5j2VcrPhx-8+fX_5QHn3!A3EEkw
zr?Dw)3F_71=B4AAf-TR0d5^vYbe+`41JBE^S8!)))3;bh3Vy{hnbGeMy=3`%+k6+X
z<~3~0#v9=Dv;-8!&mvWk2v^bhz!z>E<+{063OF1}*s=53hx@VcH)dHFcT$8M#+*xz
zWBop6baOC=sx8egPxXi8+ZWT(_+uC8mz%om_?t${xR>?#%ZcjMm!qx-Ctn^b&!pi8
zk^mQ}Mn_(`aV<~ELd}7JOcu{6?mF)dDy$-*fN?5;JO}s^d8;V<Grw^;bV$+l-B3K-
z86%)N`b7lZfkTKf+5-H<wX`)L?QQRH%+)n+duWA0F83?_Hpl`@B*oFd;CA0$)xPAv
zUHTUMK1BQ$0pKOB>;#|~(LjdH=&zrPNicLjh{xsnEl%=xP?o!y*3-W(=nqmM;{!bJ
zi!`zTVEX{97T$Bc<~VKqpxuO_55)*blKoS{TleNf*%=h#8Fy#J>1B(GNOZ4H_I7CQ
z%YfK$$I$xhy*Y+bl@ZKb)MUO2WD|<g_>jj*HC7)E(zM_HHYb=0f6alV&Y1F43pW!_
zO;UIeDQVG+jk8ipFzo@3@tq-_O)M;D3=0+l43uAEIN#rSMk;k#Lp^(Xg5%R+O?ev$
zNQM@c&vq|xBkcpN)q_ygC8^y(-u9^a+h`n_{mYH7!7|1qsY_|{UB8ailpx_BQW5Y0
z`hu?(ZMi}y^GC~{5P_~15tX9+Ae$U2qbxo$3L3I@$m2Jngpm(9lceKUm++e9zJXB!
z^tK98vLFi}gJz8B*(}@~*)AEO6P>vPgH)g)=I?C(yBbZ0r;s2GK`v0)^Eo&`*4S2>
z?eQq&j0_YuCPT5rmF)MQLpsisieTukyV!nIfG-tNeO-+i3Qt>jK+suGd~-P5g4B^Y
zF5c_NNOc9pN`{GybA-t3Llrzf8*r8vmqO+Y_0@=szm#!a&vj(v+(&4c7&t9L=JThh
zXH*T$r&(BY3{QE8q3=JbN2~}wz`O`5w`%~zjFNA)1}P4!2ScfQFiD+N*b$Y8xJKy{
zsDefpEgnz>4NcPcBD4Nv<ja%Z>2xQP@HH2svIeu{f41!K{rS8KpeDQbl1_EuQ|3pM
zV3th=FJRspA)6lMgt`mkI<uGko2Z0Yo-$QiM4mhN6_`#RN*R!p!c2OHQ%=}K?`o(=
zPj0+Y=&wqm)dv1RElp5?hr&EM2M;+{$41@JgoX2Mue}sR=fynxQG@TO=lF>HqxNl-
z&o>nV2Xb4Q4=iRwK$Rr7bFkNCLmS-s75<_jTZpn!O<G1a1#n;DK`?MgB$^H(mZ*2u
z@kbJ=hqv<-H&^Bqk@^M?_ohvT3&ff|bmS7dxw11@%yLOA{<_osA%02{O!nEl>VqJA
z@13CHbubR$h!?F}Ox0jBC3^wPT6e|ITf2rBlO-Wq@51>gfKlxZ6Nc$=UJSWY+F1V&
zqWs>Xk@NWL?A9>?1sB%0AzXSwOQ}^GEj;m`>iX6g;K&>PFB{-nD%cy?jq*Z%wia6^
z*{|B>k0nyCy7wL{&LtmIyzG=t#p8vDzw!}Ll;~nO=b|~|nIC+x<=5blds5Ok=d#yg
zILEF~VmQ}u02F&gh_|U2VpOC~>WJr+-+9IHB@v0*pzhnDy2_Z_DsR#2*~1g#z{T|P
zJ$NhBoHgn0tkRdxDvR6T9hV3=P2rBRsilK$b@r*bbfNIor`Xqz7Iv=<M1waQB24Fz
z_k#G1lJ1IgqC}FVEe9z-z{x%Lx*yhcRJ67NAA_uK0goO(i>ky6q4EAkP9PO!MD$v7
zjJv;@@=4LhHYU+U2Wdgr$6)%hZ<vg~_wIW6CZg{SC&FgZ-}qH?l=HMTwHphBe{$ze
z5DgKe{FS>bY3+p%{K|Dj6|<%jpU2u#q+uik?e}uvCI>ioGeu~dGnL)f2vfzxoR+bL
zKjQj40g&5L>6nI`K-^qODsRgvQLlv-8&#sPJFr^FgeOG(#&@HK%z%+FcD@p>>SQ5g
z4n%DOcOzX=p125)t)}UW;qmR13xN?K#Dkyp#MAvE*0%ws8W*4~9q6>8JthY^=$8(7
zLGs;Z?xN()%!k2vX?5(4Sa?2(?-EuQl!vLw0Q7Fu@;KetcPNTTxyV=XZ*<P&a)qD#
zc&qN2A><{(vAD43Cqm1$Lwlvv^zG83quRGd_8ms!al_9uZ4l!(Du5e&x|+l&<GB6@
zx6;OrO;g5sWbKvI;=+w$!Ub~8O-<fXVgz^gO*=%UB(Jmm+<P2uU{o3q?BLmOy8o;G
zOhln`A_|&vOB$j4C+J(E!{qG{M%c>Zm<&%wX%feaG7~`CRV@Ee`USL~m?1g|+wJ*-
zw9%omC3aiE|Gn+AO&hhwoW6t<Wvr|?`E{&D9K)DSp|#c8?~Ltq#}ysfXcxk)5IQ%(
zpJ*`HzycXsks!_dLQK_3aT>W8{XsSX@K%f}kacMs*Y4_e1IsBY0sBJ%1J(vY7zE`n
zXAMeu71@SHk6(bPi9G4^M54TqTZ-U~A$AIIf$lag@Ywv!U@*#vc&|YD#?Oejt+pa5
zWXwok!d?J+Z&Wbav9_9Q*@z5Iia$DEW?TyM#)V?I=UZ%GwtTqY5Q;A}5m-nxz);yf
zE^wvPq5w-CykQ2i%%6O6*b8NVNkd-kZNHn{LPPKHU=(Yn<4u^p!%^b!^KYf~Z?+`e
zN|&wxD}BNI+tF@_qj$gi58$>608aS;F9)rtjRg5lU9vJfnDhGO;l#Yg2gEa>E9>V9
z*i)%L;e$SFdul}yI{)^2(l!_tAk0gZ5h7F#|NhS`UbJNc8JC8!oHwxs=O>hF4l~17
zF~s<RV~v>F^;5doqhOwAYhChF+yApeMwM4mzZamUlhX^>lpK~-?fCO@iSy>t;0F5X
z^*YD1rLqP%boZpgnWs$J6lQxVXG>aT1m&rU&dgf>nHc)K9k{@)j(K|pIN%=3fdTe;
zZqyu0pk2kE*<%YSAX<yzoB8ASnpL>COP=JMj5ZOold*bHbis)bKTj~{g4|OEzl^fE
zF%06JRH=_0(r_Hh(rXl+KRj>y77ph(yX9(7$H$K5zcCj7f%W+lY=<lBuIU+%m0zVL
zAuFC0{+&K41lb=*PwqDZc;eD00UMl)*KVC?I=rR&q3zjn273}F9+7SliadaQW8T~9
z3uzx>Tl2k&S4i1>tamu>@cXC=ALF1Cn?{ec<F_DDEzUWa)~6QAVkh-;tAq=uE(^p$
zS{~=9B5+r$(KBLad;aLGl$8v;31h^N^QS&QwqOm4_1M{03}VCsgne%r-@KeQbEx|x
zOoWmIf#9ZCWYkFw7CcX@ML$D=Jor0;h0e`-c2<n^l*9B%ogAxiaWDR9lJo?=FZWW`
z)p1)^89VMJ{O_2_!FYZjR3>H~=_1p&H0qY>!6L<_v2f8;u}MB7RzDi&AxlziU`)hs
zM{50u1r|~foSUU?KsheyB`4UAZ}m6X-gGB;>sSM1>hDxBm{_1NaY4&MM&Z%Cdup>E
z2@q8?P&z({u(QO=6hBG$(-Dt*V_A-?M|uOJ{pk}C4Knl<V$@ur+qv_A9KVz{J*C!`
zd!IKfL;g6LakVcupt^%^l%r;29*h^B&F6X_74Q<<L>%{z0iHD`sTQ(Xz{SKg6ZqtI
z$RWuHe8w`Hpz=ppa=C*`y7V!Qf403H38^EiBOeuvt0SiiH2*3!o5xcW`$ayH9RQ@s
zI;|HzzGxxGfVxi*Rm|(<CoUn6o7-=hD%V!Dqp^U226gE)B86RhZ`-xQ-7M&*862-d
zb<kyk?wt+$1-P$fPAe6-ze(_JDzD*C9z>?E`Kp$GLk1Q0v7j5=Eo)E2;o8GRPJqw#
zE2G4aPaOG;UEbI;BA(xqi({vJHWWR!u~n`5g)STzikCFbrF;GO87gM|vJFfulxIsh
z@*~X!`(=l<DV5Lo6(Nb`Ym|&7en?p1(>P1=LpQyW18^0#E)G@BsMH~2SHq<gK}_JH
z_uL+;7mBYluoQDRp)Slk5$BoQsYXCh_#u!%t!>T3ey;w8Lry@+Ezn$V-wH{G96@a#
zhh>eH%F6MpYd7j=FKgXb6ua_;fn_0P?%3A@yYudL2VlF}StgcWtw>L`cjZN*1E$_@
zw6CG40HALX{)B&dA)ikCZ4hbRy3l9Kmfa;m=^AL-B+s@tFv7eN^N>Q-HWF65n2Mv%
z#-*HhR$RUW?aVV@7hHlH3PIH97NSJ=q+&pe*dC&*%I)T_r^{8N{k%K6H<x`3&bP;k
z__MwS%YRX$l9ger(<ZA$7yk0!{vQfpdj>@Mu(d(Bj)F4WZZ;DTh6#<vjoR!O-TaFe
z8qJcb{F657latoB!4&e-RfmH?-_CtzPPXpi=Y%q%?ffy<K-&MUujfh%pVHIYHQNl-
zIF>Qx`R6g{@a@+a-d}}oNu|Jee3pG#55zI*z0!F50)a|0S4Z}ANiISb#6y7nsniu4
z`2!nPZ)=aI{ciCFNLygH#c}3lfpb<%{ijN2kyX-3)Q*3ehI)Xljd~VC3V$^Et?kXi
z^nzOCmj&X2LW(bRqH8KdD!9cAOP2vGP?y)kdxMvgyJN_Y?HO_F2VvU*kvjBInh*iw
zt)I^yzgjE-0-X_GEfi7$84<vg7b&X|hb!{g4%~`c2bC%~8oHt1DffKWw`yX&-gp?8
zt2{vdcmG6h$6+R4aX|qgI^slee=F!?j(PN0d$wiFHk3@1s;7}18WH*azh#H7!Q;J#
z*cw_aOAfW2y7l9rI2t+;ZD%`Z!a6@<CM(BAX2EY#>&O?EvwO%?^}_($G8K8`!mQx8
z1Jm6fP@p>yw^xdY2f*@<FZNI_91>xI0=b7xA}JEDX5z-$^aMmHr%u|4ZHL(y@J9Pe
zh`U+`LPoG(^X&3_e$3wXp64ZAxb6KcN8V*9C2PU#NLNjg{wR;#YIsu8LYKm6KT6zE
zNsz7-#i9r_-7(KhLkBzw-#;j{W2e6nU$+g4jTN%uxL=8R-YgM@9vcDz(7v~^w0~c)
z9oUf4mhiG_lK8P}Fc6`+5P<`@`B(8&gXB|8+PNAKMQcYCBWz=7X8$2BgJAqp)7MD$
zEv_lqZr6x0G%T)LSk2HsMGGkBU=U>JkeInR7hh<?@^r2a39wH6oRNPw#4A{8E2@#`
zjuj<KjxNbMzWWeB$Xbu-K<}is7u|IUTukKjFqJxADf(;V{NuR`Pc7-ia52%<#Zb@x
zkK2a^2rz+xUKY7(XX*zljXa39w1OqDcDL|PA=T9bE`Eg@-Pv*3fz{nO;IHW&H(TMQ
zVg&)Of~(Y|9e^OX<@ITNQ9Enh?4cRqaB~#DV>Y7|caVLG7mskU!8RmE?=0vurhBQw
z(0k`7a!}4;?ukTUG!mW0L-<W{0qb^`B$#yl7tbKYgwK;)B=k<(-1zi&E<e$2=w1yF
zl}%|C>FL^;HBL@blgwbc@wOU}e!D#ClEu|~@EkeSa{!^yMA}|2nq*u$kNj?o=kqC^
z2=(cfY>2+XI(diC1^x^UN8*0*2~VIB%_QbZpJ5f1y_EU~W9UdFcE|D3u-~WwmPpu`
zfVfzI3ZHHEf@V@ySYT6E|Ku-0JQ)$a!$|tn;KP;Fvo##s$%iEuw}oW4edVX!{-tK8
zXiZ_%2Edfw?F9%uJ_nsBvsG>wqkl5po@=%+n&d>DB^}a>oV#K%QMQfnc1xtmNw0OC
zkET_z|2+fALWj@2sviD#27vgvGJ)ASJwn?;F0qXh-@5mO)}|*drs*$FwG<!B?xWDF
z6_AD(d5bh3oRe15;?|6`)r(&cnpW|>dTDAb$AH3d3VhJNJ0YLJ#eq`UZ)6^oXNICZ
z{1oc74qjsuoR7In(e+v_*^U~VLuc33gWi4E$mR+t$g*X$1&}wpSvkBIYFSh1%%-v`
z2?YlH)ImxQL`4tdm{OWInvF{aT7fQfq=>jtuSV|mOPp3q?aWu4K=LQB$dwLrs%~tl
zuk<;O{k{bzeNhd%W^Zod%9NbaF*~wvLKai2^s6CRNzl_@pmt>tukR;XTSOx*jhlF-
zPCUM)wyL!0Y-HBgJW+Qoawd~RU#xwqZBW6QkqUJct@mxgl2q$`FFIN1*Tp?t-coKC
zbn6#rZSsCbc#LxMQk5U^H{kSgJLLy+Z-79-L~bK6Rb*B9Abo0c$Od@YVwrNi@@%{o
zRj}BJqm^%t6+JDiF59LvqgEsxHNt+_A{(nj)7-7ue%yoE%45|uNtXhNVt+EYu5Aoa
zo7po!t`iuw!(5%h_T7%Vpl|3>xJfA#wKHee3zhaAy8D-cpli3z<bMg~h{;M7b^>w{
z*JfbdmpGLK&eQUG{OZv!chMe1lCWBMgvDSjbfq5O=o?<=3f`+;c&u8L0qBI%$NSNP
z)ctrF5Yai=jYMd%4r?KO?x`R8>btygtrb%<)kTWSN)1*=ZXSTn^@-Y?VV3{|1cp$C
zl!lxmL`7GK+4Hx@^FNIavK@aXJ^(8(a~iqBO0ZY_It&k|CKU_j>OL35B6VHPXGX)I
zWZK#iolWh$t<7)t?-5##ot_jXjcggMWus|nWfR(u^zL=)to%Ep)Yr7(QN-7QhHK9r
zr?-n`f23b%?1Jd9fDSr;B4^PKse!wi_kmD_l8vr&>>T1z;`T$O0fZOs^MK~&W5-S}
zn{-pY()`Ivc&E)@pT$3A5g#9hZ1B{)ktu9{Z1*z2=<f(e!(z>q`ZR2@i02fCu@2c~
zz&HPB=Z%D8R<5!v#Y)Abx9gA^TLydn(GJVX9Drl~oq+L<`HmT)1=&|8-!JbGQ8qHA
zdJv*ra%o74?A)9zJFdre$_#L3fpr5dlnEQH#y`PzzABeRU&Yt$Xq9t}s`6ngIxkfh
zCc_S*x<MuMqb;5mVx1g(FeBYx_NAV!<1i6k6<rm5L{*R$HV@`&H6w_k1$qYN5fJmv
z;#gx>gKt~aBSlUk6&UgxXfCEm!2Kg?do+z9)VT$0|E1Xguf{K&jrR{3z)Aiw0>8++
zm=EypgRi=lXK@@Yo#GR1S+Ba!Skt!i%>QB4X=g_?EzWbGqeT^@b9(aI#O7<z?LZXa
z4XkRyneu~k^+#^gS3BTc)mJ;9sa`ktCUG)hN^wr>qv5pWqHFd*-UySh0p_16$TIJ*
zNxK}y(Yk;TYd)LW3s-z_7P@s>Q~!5Pf=&{}uOU4Da^5rBM%%Aj05C^)f(H~%uehZ!
zzZf$e4E<MwI<(;z??Al0eP<+q%S+ARmXH=uJ?Z`<mh*o|$4TEOG79xA*KL#|z400?
z=2lv#W6yq=>f`MB;lw?{v*+u1P(6SvU!;=0a~-`nP4#DfV?X5nBHzq2SQ&7@%S1Ue
ze)96+^1<wc&pSfmOjwu6Y`op7O49WpcYxFr1$8p+<bQB-N;SFdum9l~V5%t2!^hN@
zx75Z?QBR1+3b~JTBcF4xUOUrSbMrr#^D~P!m7>}&`4r6SCfh|iwk!Zi>tt`ZS0wR*
zR4RA7W#0xZg<Mm%hW@7NXLAvjARUh1<|PV?j|<!1aC0&v(zw42u@|S=4H59k34L*h
zqgY7_1s=v2NkN*DwI-5;dPzGmniBPA$yhN?+NYa0MaHTBVq;<PIlk|a`?ha=XL$3E
zrxor$<EoNZ<<0?uzijnMx9^t72|PgjRh=hfH#V77cH2Mml}_KA4MTd69%k+?lXb3>
z%g)xHC)StpE-A_P&-@JCT)+q09WGz)4qLM#QXeuuQ<?;{DxGeBI{Ufa$eI8B*x3H9
zOZAFe7g8xk%lqD|+w{197Q&-LOfGavk$Twb!n4j2!L<%(??l@dPG|fr7#hX00WAHw
z`Mb-szrd4}hh{-WXZ>nCt7Q^z9BP&{(Tt@tF2&6lrf)eeZbjiB7`8hz9>IM~WB%RF
zzFE&r4Fa;@)=P02Rx-ijG?s@*M~+7qMp6yt?SWHI-0`pQXs@p}^PEiDciGPdT!_@y
z2;*CR&O0=KQj4q?!9SY^vMdi@k$XpJ9KDT!!s`YgnU8AOXJnY_+o<<qqnaA>Q3LeP
zA*7q$)L}tyjEdzkXW0;zJ~u8)>|&%78|r(fe@cCx<a5xq4ksGl8h4GgxzFW`5{Fy@
zjnQ}Du75a+YkB_q_e~l98bNCu8#K`b<ZTFmy?+0EQm9E1W{N`677t;hspm$Zs{tA`
z3hx`()guwqH8KHLH`!#3)1u<w{+TUWzm0<qDq>7+P<^;pkZ7|7eMA()@e?Wm=Bf7k
z{W1@jK-i4AZIhfptM-EL0JDs&Dz%mQnMhP27{}2}ICX2hT<T#~T6@d!hBHy24Tcgx
zzC}1b=z)6vHW+_Vx7-NTRs<`@&8Wy>YJB$2{9&n<#?x&P4!c%atqntyM271v8;@{{
zF;P!iLj-Py+pBp^ggg*EfF2<SV~+bKYXTgh&Iw0!ozsII3|YzbG?s%cwwG)qV@{bH
z7*l13y7y>|I2&yO-D9EzkJ*%6ZAAmPN?v7Ekc2WM0@Xw_j8=H52}kUS9cBhfFl(x}
z9J-XO7tx;s<}b~@T9NXewobX?{7EDgjM5$=hrf-7oh5O*w$prq@8{*;EGNqRoB>7K
zyblugCkg#VHNAu`7%CWwnJt``5dx%|UL(bisp!uM-uGuSnWX$nZ~28{GtvQHEH)#(
zpXAfV+ztH|s!^dLjMU7xZDt6{gCBjV6#sXX#hHqP!879T#v}}WkrTqbZnaXQ-twj4
z4sz5WL~A-n>aTTN>PY>Uw3ysjDiZ}iT5W?Fc&0F{f}5Kux!gxD++-}$YRnLX1$Vie
zY%&qWEesv2kT4cWwkP*u3%L6&+K6g(sd6r<DobpsLyr_l@jM0*TuL^ax)@tH&tcgZ
z*y74%#(Ng#-6$DDkxu9)PP#L&Bw$%31tpM2RidDi!mi1KtE`M+zwm89S%d?4*&V@b
zODY=Hv<fN&<{nQr28lAlA7C3ygeN;iN*VA6UP8F?UwCyAnWkL>Anwn@q$^gF8i_sB
zE|393ZKR|bDPF{2VO+mJP_PIr+5rX_C>x&e&|_UacEfdx>{E6AR|}3e=F&ALAMHwW
zd1O{X38TIaF*5meo%}p03(Co%n8~0)6+=k0NPw;P9<d+^|K^-Pm!4ora&~(tO!fEC
zDWL^%T7lprLh5gT>Wm*hYr<|4c4&}ldefIZ71>TT<3#js!X=x?1gO%M<D@cU+IY3G
z{r<w1DBr?EE{o&iOea(KxGDS|3+8|s`7WZYi4+t=Qx-#`{Msc(_kAPj@D-#=_$r9l
zXG%cHY0l>mMvl}7ntQTB1ySQ;&ku9aoP7rPG{&Nd{}-<kuu?wEyUAxsKp++D#_o2Y
zYB=%6h7q+G-!`idslvg7A*;0XwuhDbdm*1k()hx-s%498Lf(;l$HKGeW@WJ;38Da8
z{K?gZDqbgtNrCwzPR;--!3><hjht1fvy{AbS=D4AU<fYNKxYE=y58|57cWt$Uk3d~
z<LhKd<^V59fTTTHN?kH`lp_uPdB3WJwVUnatdq?#q`$m*N9}5AJ4-BuDIBSvrz+B5
zQWuFHv8>7wTxCiRcBv26RVBQR$ywAu%-_G#le+RA<?;@2%Te#ou9613A(05=Mcaht
z+fOw(@T&FIFi;bph=`(kYJOPPqHIOiYBf<%xR3^$0Al)l>!}$w(>R8tB*xXhjSf8^
z{HmJ1-+u1nY6uE}WygX+fYSWpMMDAuNkGp&xvQ>7oqdJ+<@rtHMe(hWsu81;wvB>e
z2y$^)3S)~>j(pz%jW_?FP0Fc+&)F)}V%ZNqkZzI>j=$!nzh469I+h;58fIN{tN=Ns
zZ4B%ItUF<ii{?eRc^tR>;?%rF7^f85d@~6XopdmKp`Sn5Mhk4b!#lJqZjIXXJixEz
z8t26gFaCeX`ljf*!ZystjcqkaW7}3^+eTx1$F`j`Y;3c!ZQHgp`TxP1HFI+=&-!lm
zi|2i?zr9@he{X5KcG*Go#>Eycy`LCoALiMt+=M7upqMa+`c|9cjCc`zIq%JU_irNv
z!(KaEX3zM6&l9|}3*NIeyuc1z=*31%zmMUAqY;ia@firbkNC#WbMxwbzE~T{Kvv4(
z4E8)2x7^=TuF4edq&L`@)u`PEgR3Ay$JzJeoR1-3Z)08@abrd=o5fpc!zs`_xEwmI
zA#@c9ke(i)86_fd(K&Hv#v9%CCokID$(#6yD6%y{#X&vqYnR~T0X`c4-y)a$X`gcR
z&UdX-WaQa;{fod0*vo1ME}zemZWCI;M%{N-SvEcsO6>n^Y_i9s!ylC4bpdwS8_yHm
z56<tdFKxpq?ab${HbR&}nR?_CW9?^zGPsq22y&-Kxg9yIw!^Fa?LjFcG!tuA1noh<
zWs60`n0tbko8MG6uRK|~-bea=oCIGps-e2FYuIJ1XGpzRv<YWKl7yt9F<Q8_%<^@0
z&>)h#kIRn^P4ApEU*ZB+*5340FEW_G`0b(AuqewqP(qFrP`KB*MxCt>m_u#uxfA>n
z&Hc7V-0Fo>V<z+VAQsb0-+|nkv)oyLUu^E`H!&bB+n>g^^$+8w8`xSdOJ4_rZcO<i
zp4PvoZ%2yODCHxdSdl=}3|7@_MKTIl?Sex$&eGaxSAq-pSr>kL&1akGpJ3tuhb>a}
zJF7E!0AGVn%$YHGsc~82;zP}qU@Tw5H)IZR9RXi(8N4i`MNxN{zN^mHVOa%4k$|k;
z(fw$~2aAGp_8%C(32q0oo;78#?fA2^-YS21F&inOcWRME6CfqJ<$lrWKm@{|JT9>E
z^l=R-ZUy+$KGuhvpJZjfQQiu)7<rhbb=pq1UFay&v(#Y<Li}pNpVd?oR8EK+d}STn
zgGjatzHM43SE6-9@GYRkJiY>S?!g?jT_sL$9BYuu?3=Y&(EW}r@fR~H7Sg2`zn1pp
z>Q?>YlVt^A*hhhpe;4dF>OqZ4va9bE^BtY-A9wIna@LtuV}Fik`8KuY)^b95Udd*G
zAFXmUrOflWqFlZw1Hw^jJn|19dQG9c1P@C}1NA}T%}fTH=0|MD3QT~Lw&WPd^|INW
z&*t7pfVZ8;YrscBhcFLhvuMC3w9mht7%}OXyTP}^P<*e;iEew>(0vY)7@8xMHeiWk
zY5|?10xV`pSE)<vJp`DK40$AR6j}dcqj|aRia3YUv3PQic*%E4=UHWMSHEdBNAPAK
z$U1b2b+4OrHvngm;|3s3ES;dPUUND-F*#&qi-u(4L87CjqeRa}z9&C@YtwiIBD;Xr
zM}UOToWy+$Yb?QI!tH-e@Mp8szR^X*^6^mTFa7RQyTX(y+5}w8IUb$(N?>`YVMI&>
zPHJCPBiz}SOw;7jw5gC0NlQ6BG}k`6YZNtI`AC!wfdJK7ibJa3X-$<|L1&YQaC72}
z<kXMW?=p>~=`@@p4Hq$5A!o)`x5~)uY4fXcmQT@FrL*C5anLCAsUBr<s%Y2)J)KY(
zbPxF3?GdS!gh#~`PI%G0+w&3Fb!lyrXnSB07yg^Ul7W(I1W8YhE~C}q%@XA(bx?hi
z;xLLvvjC4sxp=Gi5&vLoe}c-k2G5!IeMQS&uBxaib2P~|@4EBs!G^bahtTGRo%TaZ
zTyHz@uwfh&xXje8?U4Grpx59cBLKvid{h&)DNyil#O>sF$H2abZpM;0So?ps&Z}Ec
z(xR!CV9wC;ho?|*YKPf5^9&GGDB)jbV@`bmcR<F4burGl{g*ZRHEC##nv8jncM-%K
zD3961(mT_krGF6l$Qa8jt$l$3&N>yQgVBM5;n*c|;E(SLKYQ^m&zbZBh_04cRQ(W{
ze%v9hUsafGyWAKRxCxnBJsG8-tfNwK)tdghu^2N$$B8Oy7Wj96914k0W|UA2MRjL5
zbqIuL4Eo~o$c<la>=N!$W`;k-^D^mI34d~2yg*K%dP+ZtN7WC-dSv^^J!3*f9W7&$
z`83&j1WSpJg)ybApDTQZDQ1J@%|dO#A7o|wRuewdRK>X*+E&;citr@Z@E%VRU~^`h
zJ<uqiUG1g&R5S5Br&ojV0e0)PLQUR^D?p;cvNB1^W8-dZOnSibd3QBfH5k2wm<|gf
z-2|BazC1@luv|6VKIDk0<S1*D{9}>YkK!bCe&^YlARn|LruPT-Kcp>|yHpb{-y<wi
zZ3>c~s{|<dj@gv5=odWY?NO%9Qu<mwvKT&|<~q!}sE(9L^bR9FpSJ~1f90hgWCE&e
zR&yx7zh&`oUXWlu$DOX#DY&B!Mtl@pIN}-15JpHsC{S|V=es#st&E}1@*XHgdlK-s
z&lr#VxMJ~kq<Evseylb>(ySq5ezJ{nk0Yy^$8#8zyeb+2JrkG2ao)BSwW!5uG1>I7
zKy2n8nu7;|hZ7!3%v!7D+3ae+^vCiCZ@=h@nV?n`R(YD0ND=D;3K;6V;Hm{CYxo0&
zljye7!T-J8=lp5OZBdp%<+|`!63=ZK(MUt0tdpIpN+@G~AX8!>c)Y_uXqaz3Y^_T0
z8zvUcwv3>A1)g?xa)$SRXu?4wCcvqQ^k)eYj6dK#T!)KO9xIHxuVi>6%K>DLFwc!q
zHJ5PC_8eS}RJ<pSK>Q(w`x$4B_Lt1>XR*>&(<`w_cZ!e<f%(x_(BtEsz&QifFH)3J
z=$&IaSjOx$h^!eRdB%D@bUl6v^kQsCR9Muqf-vV^-ByNL6a@2WI;4(4r^O{RKXBnE
z2>@e%CVJkCP@ifRKAv)I4Uk3qsl<{0A@C(HRlUp;r6!1~b|tPmIp(+M66?GcL9Ajl
z)<gAam&D6hpYS1N5LdA=R|@B>89;xyGnLdyeZ_gk!w_^Qu}lSHZiX$=<Wi!wT-<qb
zXgck0L^EaXcUEaj{>7Y2>KR|=mw@y7vzJ!XPt<*)E>>4reVtwB8UWG7L#YcAWVxPX
z)ML=kVI;WsbO})Jg>yxcdBpDTG*m&yHz@8n%1aElCxPm(Y7jLr=-suQ<fibMp?)_u
z-CWaNw5!eV1v;tY=~-`@rY0Sar=BGOurf9o#>+%Nt3mFoH#8<HmbzR-bZ~dkq4aB%
zN@30oH>e(WHgf-^1NR3Xf8=>=7xgXcIfpX|ImM98vuqB#Z0Xyw9B&$&9~KviCq_MU
z%d>H)Jl<;+gaxaBHOKMJi!L5~IHrIxCOC?b8Cn<X=!cl-de~)<%vRPL%fjPIy*Rg3
z$R!w_pB2rFD$=Xf1Ku?+6g@RI@?GW{f-+8tHbY-=OJ&~XKqKzqOz-F;Vi-JISfS_W
z^lJ&1ObPgX!z<@DR<Sp<JHuv_$GBLMrvhot@Tq<gCz|fxZn~hVBAsu+g6^Q$sdR0M
zY3o!(Y(xF&kfA)$g`U|<`WsnZM$H$PMSe@{5B{L16bbmN5!oL>n6nrP9m;U$E;cr8
zi~V_#CN1N_K$Wg9pO)0ueNlGNP8No=wFp0a_0I&1lG*1!t22q@afqvMu+2s9_dp4r
zPk!uX1H`p*nMV+26cgDzl}aOx2~@y@W5Lbb6M{Md@fB)I`cEeG!V4;Ga~>*5X<kIF
zT_U-Ts326o$`A>dZ&|E6zPXar@f*oB2$BKmzQo))U{O4K?sI$*`XS4`jm@@QMi0<F
zfb&(CxUmHt2H6RT(ryYF_H*=QsZPI<9XWw1Qpkq?Wt5-#I0@v)zk<v23Lkqw4Nnmp
zT*ZLU<6>0*k+5S7lYu1-Cp)Jkip4Z-#dM_5KOR-hok_JF0`p*sgoHkU(CO2cQP22}
zBGN|#%x#ae_Vp+}LH^3+fNnRdGF<^X{JHk}tB);9;+)B|SRUrby%wl0MfMANTN|#Y
z(0-V1k|65n27hWuV$FNIWIeAbMTh=q)i08M$6v=$$0|gEub$frMJ0=*FbrqikP-Oj
zU=Lk~>Se=+`usONdcWAGoX)E8<4<|q1$#fwfRChG4>h3?qWxMa&Q0}^fE>Ho<G=h!
zLdnyVgPC}5IFg3-K}k(Q;aQ2*VZ34LJ9I11nkrE*{XXxJ7QRW<-wHWI>wK{f{iT|S
zj&T$2B34i9GDhhaH-|(YN6j2uJZD?lTH9N`L~+iSdF*;6kW-5$o%w;CbQu`n2YYMS
z0OFim`4$?oowiQ&rqU;!v{}Hheru2PA@-`8Qv0$+Bchs!J!z8;)GY}jiHpSHWyPdk
zm2TInM^L1^LqMrl<d&xYVLKI1ud}4fK;!cy<2tgLNbRDS$syu{;b47$F!*yE-o}Kk
zO6L0A=khDYVSSv{J5U&MHqosUe8uDgNH44Ki0rf>>>2WKYuQ&yVOW>r64k`p?QBkb
zTy|X~2f7ECNF1?{Hzd`#8_&OcA9}yswYV~_$83enXq2Ry=~3=WJv)5-Jk(QSf7Q8`
zqtn$$kqQVF;wy@ed2?M&PYuwJyrKGTw(n~Rahv^Fz}GxShxMj?@tNP)&^=WKC|;qG
zlT)IH_Zzl&Dev+pQ;W08&cl6TLg-bE{3{+M$e;#EsG=Iu+q*o?8WL5!jqiyM_ueR1
zmD;aU44S2t>!WV@q^+Cep9Dg?MxoQlq3oyg@!r%y@#vg{$O%AA-MeoZl-ooikx2TE
z=CLIHp3#}=z=%NhL#f~y{!9x7G7uMt{?$--`I6i*>78rW_aEqKUQN8OZUx}}a${KM
zY3t}P@Z4V8@R2XE_gu}JU|+=9+%SFW&;BUDxy+Gtqk%s#o}ps#93CD+X=_!nq8{9b
z%q_8+r55%k3i^GgI~Sq-y<(itTJcY;L8iio)M;wSO})D^x1TWywxL}dpyA=PrSEm{
zm|8|MvjBhcgRVqF4v`6?L$55z@DC04PYV1J4aB<WMoWG$$UU}z>$OVUfyV*h7sL0<
zl>|=u&=HB}W(}~Pp$N{f2!#Pwl;S?SP<B##X1(v{YW7P_KM-C`Utfk!vub5V2~0Hp
z4(BCqiQlmI>`ovyvSQ!L0+<=mbu|9_!ud>u?%+}?QQZH|^6d(F2_t7pxvi5XTkRLX
z)4P{{XnDjdcrXhk<y#hl@<r;-H?iSutZVVV_01f}P%Ms!7Q%%VmGbg{tLweMOX>2(
zzW0m&K}^$3f}LhPHOonn_MuNs7wVT=yO>}F!qu49t{hOQH2KYRfyYQ|GU-e5*4!y;
zoYveU|2)`mI`h`#Fgl9X++1W{EKH|+A+vX>W--NbL5|rF-!wvo?x0avI|Ohlf?kl|
zkr>W>P$@23rN)8E`LfJ&lLR<{szYTuq#ABSGR&`^wp$b*x3dXh?Ra%Wm&3vwLtjFC
z4z#7Lkm8D^c-%r@cVL(iZ)1dAY#Oug%x*~dFH=>FEG%t?GeR@gbQP|(u8KqnnjBG5
zSnwRqI`2s|^_sx&j{sbdV5#h3wSq)4xj!Z##UyU-Ds1zydaC<&***o)gJTg1@Tf!U
zat;;;YAm?j|Bl{%SdBZOfobRVp_ryqe{Myy=Lq1I_9A5hatP}1HWt6*g!|fw$h!Nb
zn9{CMk<84foC6<Tg_)=%7op>ORgt!&dR1eU#eXir5DQWdaUmKP@q?eNz|Grx&f&>t
zAn=t5ujq&0WIXMJkDc?F)f|!zhYV(8_K}9&>rYC(P(nUU`w4?|9GWzc_u?81anHzq
zC|pvSY579~%^G3jCN5i8@qVKChMIKGHp`m_7>_=OtDPtF4yX03-pH4;+FKuSilvFG
zcFpv9RZ3D8YhF)w%g;#h+k+L7HQqSzy(VV|w~tWRo$SH<(qrS0>Ik8#4cGrbhT8&E
z=`ssW=T^CtMT1p&kFXe7uZ*Pf6i|X5GqjC)`YwI~5~<%YbMS2(4+jZdx^mP|N`%9N
zz2ajzKE>QW`P<p(_pGn^SRQ7iq4m5xmf!m8pd7j5D2i_TXp^Ed&Z(Ssm_#j?E4_4>
z`ACb`v8h!#5Y3+duF<G$b-UXOL}9J+oLaW1|L$obg2fcgTMuj^FvYN5cY;l4Wo%KU
zwU$u;1ei%MSNaNZ>G&GzheK-*&FGIV)|VPTVmMFE*gqP$g<W+5HtsJ{H*;HBO)7!>
zsRY@GVbZC$55>^~sA&j&V}VO@VO{bxtpiV*wRRh`=ip~V=nO*jW)K);3>(EJ3YxWV
zG;7@lYr)vCc1Z!Q`K>mV3+Cl3Cmn+L!;(q>CMS(BhyZlL&eo{BJ5!l>pyA%}jl~5X
zwF*UIs9fUHQ&91;rZP6H!WC2KY&z;ue%EgjTpCUZzg^R_>OF9m@pwk(l0D_Vbkv2+
z>VrU~PBm-dP25i+10|4ADFGsX9o?OOMaEqBQ(sH|`qHzgAqYIIc@6qhc8z(kXx$#5
z(DTOAo~KdbrCR!Q&%`(BdQc(DD-u^;s*KAj3QEn;rC}$CNtz<uZRHM0?H~9<^!v;M
z_G>6IjN;wz^oFrexNyW$j=o<^@J;*4;bP~_+TcJhrEQRACAckL%jCa@9eP+ABelzc
z`hVZK9rnJVaXTb0(nyTJmc7`nX?C&$AHtxAJq{p1_Vxd)WCeflPcZN#p}x$poHQS0
z!``f#yWS;DZ8lC%NiM(nS!+H}Xi?S1TD_1w<5Jdb@-HnIYa8a&9&?nLFcZ2F9T)|7
z2&+W>hJ8?|#`z@PD-76L`muU%sHFujhl^eaVxId@N{4rZm6ZUeHNS4>2DsY3Z%qJ@
zP{sbfQ6=s+q>{ShS?`4<boizSEt<St@^&lza&@})!XG!Nk1(l3T~w1D>BpyE#ZaWB
z%Rwbgujp{u*rA11UwCvzdu*}6?1sK-WM-&%!8BerwJ;7|cWj2H0=td{+0<gM+4(yh
z=2KY4H+v=gaYWS>2rD^oKr$+SRuvu(?4XfV6S91b5UWhRj2N^CjYH{;MUCl!Hked5
zQf08PzcBgxQn$`uS;40blKFW8*mO%Qjf2pVgIor6zGYA|4IaIvWH_pg>G)CaXT_Hn
zbUFt2_%Hp{t^4Hqx!7s*it33u`@%9j0BlVOvg0piyL|+j{V+uXrcX8eD~+W~e-Nm}
z*Ph4pWYG}LCq6LN>St+QJwl<dTD??-+00`5n8zr;V)$gU$VZQzGpa|6bihn}8$ilN
z%xb`g6{j?6!h&uLZTE)aXK~pP%-7*n*f}|f^g#bDET^!IhC@;}<rPuD@lDa~X>yz3
z4PxgyWUd2OVNf9i&=O0LBCepwH!wy+$O+<Ln(G>0Y?|`qR==dRg8Ux#0~?uuhPx~!
z)aEAE_GIetBMa2Ww-%-LMi^L9bO1Tbz?s8qjah~oXb41D8mNKk_=!#9&QuH8ewocp
zEd+f{P=D*FxYjQ3R`ht^dQWRvuU#1yLtZ3Cv^SgIH(Qtw)Tm4;FHZVD2c6Igx9k*K
zvy5utGg5WvHi^@5uX>35+o!#m4Qx6}AEHMbOoSd_o>BTZJg6<M?`8=u?~|qvYi-mH
zHk8@ado7wh({th8yGi$IZSU+n^V;%X`urP9pup9o_t2UpCs~~HBNv%+X382G84^EC
zYt+p@`YzxcXkOsJO&!s?UEtAs>eVyK<87#v?pgGqNg;1H`qSMOEs`-|Dw8`WMLwT8
zrbwLCHAZa;(zeutXHV-O7uI|q{{9yMvhR=FoGScIaW(!!=WCW?g17A_O0a=9jMZpA
zpeO)7_6b5JbYD=EnfMPGgu83Pf1+wg%j?b4Di-AjK>24y*xnkEACXh>{P4<3VCxTv
zT<7Tp2_r4jRqn1?2l*Uy36$*bc~$sqg49(bhHdqk{p=O{7RBh_se*XaisNPsCfWPN
zgvZgnp*bdOT;d-;|5ZtwngJ%8n<zK_0knUnd1`n(Y9X@wz$|I5+%(E^zt?Kq=&OB4
zsO|<s14@{?%{)ZT*J?Z^KO!^YHaScHWM6P|Wa7tE5<7V31hoFr_Xs<DT=Ez;}6
z*^vLyD$Ccl57mSs(*fBEH;I}`CY!E>ge>$O>pgXxOZ5+j-gugsjw}AN(qP%)mzese
zZ{!`MH{pwc*-P{hNN(FadWxR}$7uJD{jH)@5l}{$YCL!In%Vr%mpAHE^aIwP+y_Sp
zu4rusd7$M=T^dC4n`E~pwB0cxrsy}(EniWA>Kp`^HLR<*LXx_lG%_3=Id**6abdzx
zdmm{_uCq18u9YU(2Eq`D_$|hV{*bqH$B1;?2)l??X%pCp=lkgVgpXfJp><rRM3T24
z075I@L{ArnZ&GiYP?tPO)8#m8t1NP1Cn#w2%4M78pt8N|8^g3bev|k(NW<|~9ajN-
zM(JiB=6h|PTw)5AokB>jZ!kY|7j8hNCpumfZRX^+aUKO9Z~23LyW)qxE2RgDAL*_r
zDTwQsDx9{m6;GJDFitys9uGizAsVSDKrErL?7>nT<j&bm?fGj%&ac}ngbg2enb7YA
ztBh_I=}l}-n(Z+~n(ooIM;s-(oLmxxKq`E&*vLyhnxyM_f!TjzkD&TlOMHW;i+D=k
zAk)0kl)mRT2Q(w?cia_4IJ2I6<@(~VRizu$2ruVBXOUe7Sv+_6sec$x=NZlqz@5E{
zM81gkA1U<AIE`8K`XwtVRVTabxPw8Ds*N-1FZZR(XH?{^R+KOuW=wuBK{Ij-=cL~=
z#}8pQgZs{(CWig|tsPyC-KJz}nf8$o7ryI)p;)?mVW!x@TrMJ|!LC2(D*b+z<B?3j
zhMZS(@X|KOQGXd=LvLM5y`*(W0bDt_a?*6|{;<a(-bUjgdQLd1M02M}H^=rre|Fos
z5@rtrfB%)^X@B;>-aN6&WXL!*==?^<wYNM9#UzsyUnX+ym_nx!I=#@!CbZ)sOr!s^
zSc%6jSH}1!MLmYjU7Vmq#@&paHK$RR0{<|N#9H1yJInuBzwU?A>Y1Q92T=P<fF0YE
zYf?ReIP<7~TQQB~QCW#x+Wls%;{xfvP*j{f?Dbd2NmXSDq&>^;w->Gcsg4JWfZ^hf
zm+#zUIa^%FOl0YF9V|+>r9R#=HpPt(Y{EF~8H!T3@gX)!GgQW$k3XXa*!Zc&yAnp(
zN?boA(s>Iv)HDlLpKJ@4$$`(K4ef?@)E%C*r4m5#R}-zz1@+=vz`yD>Yo$!am>_qW
zF(U(l)2$ozq!as&_3%LO)WPsH@Ed}F#}B@%M)*KAVz(3>w&9g`wGe|3Y~Owrh`{@f
zbIq2&zBoh{lZgKy3RPWrBmJNls%`DYLEvucxNtrX2p5;*IRv+5Z-DI%*c-d8qlNKw
z<ROSo!*Jd-2t4y}UXA;B>F^GX%PTZbVInojL*Y*nN<$yko3~PBO^n;J4>9G_c^75j
zV}?x(dQ8J#t%f{k<QJ*2!fvij*Qv2!xOH`|-8(E9vCK|`{d22YO>XB%)L4k^jQ7DM
zSrtsIU~bhA5+x0xEI?0y3Au%9o+M_<qqU0RxXTe4W!5w0rSZ9pa?3VSYp?LdQNc|4
z9Ek$O5_Ub}4UCqe=8d(|4xHE_xDN<<QrY!OUP_mbj>_M4xFq`G<B$<$SR79Vl|!XC
zw*%vjNm%*GyyP!ZLcIAhlcmOA&4rjwg~1}g{;!d+^?GD91Tb>)gHbVkJkrkPl~*Dg
zLD!2JJ#JEB|NC=1i{8T8JgnTVJ3OPqQV}b5G8)UP2)Jt9>#<|UBIY=wm4ruHMS#<4
z#xz(qp-~RhY;r@U9im5C0Y<eP`(mC{8G?L)k6QY>O-%LY{P6sK!%u;rmPc$}R3QH?
zM2dgNo;{8T6Y%vcslh)C!oC~;7HP3e-HeV^#X8DfKu~ndi357tXT*s=BtcK@2ZAq9
z)}3g3>Y_yI#Ej-$o>h<xE=z^?5!+XHC2Au*>^xzMTbp~#>Vkzb4ZSV9hOWD9RbW@t
zkYc^TYE5{I@9R}d1%1B_!dR~sHar{`IC;%cfOf0j0IPlowAg;d>N6131d5VYO?w<z
z0zSm8N(|)GVjaDIDBl@P{k;8FT!j94)0qGCPQcO4N@3`Z-x%hl_C|`%@yp{S;P3UF
z^la9Ly?W9Bm8($0$iY5%xU_qNt9)NhSjALnq;?`a9=6{v(1;WxA5kv2ek8r061Oj>
z8xSR<016qd+n#PjLyB!1)mfH$>Q+I-b0acj=UB^$x}K;KQm{wvkx9}=8E|RqTC_d+
zP|itLr0P=)@%6fTR$mKQh;iDp;0RXO%x9n<zgre+(n$4U^7yCU30-}s`caJ$ldFob
z_1)5+;+)oBAQZ*V{1LTBKJICc5E3I>s_Agv71)=*DB)0nU0asqAM%x*&)e-l^!MdZ
z`La5WDYh&g>sD%TE5%6q+`#wy6+j{EyhgwZHCGfcc@oQJRqBJP*9R)2Nq_^(Vi5t_
zlr<l1ucIC)iY<4|rH^62o_!l*4?;0UF-EBiC1NmujZ7)`SENzOOYSX1E0#VJnP|I)
z;frD|$9g&6)`N}w{nD@J%Whv$Sm|-P=p<KvLs+8+Wa2Y)wuP20T25}ou;j<ohTY$O
zx%?i|n>&_?9Tek0xDSx^hEhj@$_0h<ziyeD%`W8%KRqZ|EOWC5=R7#T6QBMOHgg;o
zNrgB`m1-d5Hg4E0QK;hH<7ArX!j%tq2m-LBN>>Inf4|0+Vz6p2ZEC{}-e@v>j3rm{
zM8YOic&Sx}AhC7WD*yfn4S9jJ>E~43sz&VkBE?cnIS{|bDk;(Y=#vemLmSx*e@sSh
z7#+f6M6IHO!LGc&+A~RC3bDGfTI?~4%%7llCR(7Gi6zYo7*&EBDM-6%J5ID@9e`Ol
z<NF=t8{eYmy_}@pF^Je1D*su@+oa5UIN|-<l+uY8?|aOBw3FDiG#hy^Weh$9LgX`7
z`6&&_YAbdZ={uHi>m7#o5--xv&fS(mNfl&X%2Y3nek40GeBAUI6QT!%OKl%KMm>^4
z$|}OZ4r{16VppddKj#7(VfIGn1mM|aq0ec*=eJAsv6E!W;=(_@8|sO(hLmuV?DC3V
zwlv#Cmm8$@ffmZXcE*Ehzm0kwV6Aw|L@X@iKQe#(`IY-?$OA0Lq>U2!`y^8nwWuhv
z89Ab~<U%&IMhm~42-JaK+`cmdeG7x?`rvppDTiCnCK*Oy!AR;BsCW>ApDi$g=}zHS
zg*IZ*^q2igx?X%<l~qloL*NZLZDQPScV=N@_(=^-Lxwp|Dvrd=EuK_52mg99A*d%}
zz9_0~GhQPoCv=23%+Ch%2yxvWzuUgKz=gXBOXX)<!C#eL&28e#P~pKb#BBm7d)hoc
zjThhMm_+cCWsjx&r~WRm-!H(EiYZ!r+z3{+Y=APExIwT+aW?28(Q%lCL%}!0l=J-u
zpDGQX!{D0&*Pmftf~U2`EMEzg{KJPHpCqf{3nq16t;|d5g%Dl(r>{U9cnW2ls#*(K
zmQ4-?pso8@HEAOwW-aI{tzK-c@udn2{u7NK2VwZjI!zvC(w$GhC=w7;-*etMSK(K;
zm06gvwCXq&Q1Vy!ZRDPc(%AB>|GVAcZ{$(y8e$`@_<{o#3P{#>!NBJa8sRQXL6Sy@
zKbKpm;H&;W=$)}A1^V`dkUWDnf+4xYnYDX#iO$gDxfb1;JA~*F=o<g(M}#Ybr+T<7
z&H7Ohcyj5sFjt))e`^4sLQo*YmX3&^JD2GE&QTOs*sW5!B?nKHi5sQ*%A)oSU>c?H
z3P!rA-21dbMJUjz8+M7Uh^xQ#>^z07TQo(S$yFy1J-CRgY}?{p?b}aS5%`h}ma4fO
zw!ufFKpkA+GqV=1RkGzxBTIO~+s~L+*j%y0@mVzL9lXuh)hPkYFmqYUjB&*>+Bn7y
zY>~oQ){pAUjW(OBSWPeMZqRuJnHs+!MnX=~&mKC1WAiFNNb$8Y(zkwh%ol$2tAOe1
z<9LmA6jGqZ1I>5eSLaDv7N2JM4jB6`wvGq>Jvn=3-l9LwxQRG2Fm#^lD<qQ99D9@(
zRG#Y2xwFt(p$q{^uQ!&%O)}{qo~&obX`O+l#HN2cIuBB$V%6Agh^}%MPoANU@N-?O
zarJg=zKKZU;}sg3Bljt<ZWpffy1)uvfd6XNI`Y-<=QSZhkK|cbF(}EeFO^jAlgZ-1
z-nm4KRI!7!qCF|{5~Y^4NS&mYRg1>7Z`p+Kh(lhm&qo8cz<I%e#~s3MkZ9=W-%)iw
zFrJ$~W0?I?T3BtzZ-e;X8qnRjmtre*yIhB!Ncpc#N_RWy2&dOSlSTS2?*EP!O<{+Y
z8le4RFtCU@jnG`K+-Gr50i_PMIode3WmsRxKk!{~y3)X(36`0+-HEW}5L7E$z(r;)
zCHocEhldXUfPT4Q1*ALFQ!;~?8<RcOA5#R2Ksr_yN%CTuC99lKcRWn$ey-d;Jwxj4
zJ2!E>hi^FJrs-yvwy6{xm;8U{+%{|f16+}IeCD>f%AGl{Gi7lvwq|hLz58y+1>-3`
z_?vLJA|`vAZNeYTR>>6#J@bCrUx4lG3)$9@>;R~UA4Qd6Tt!McgQe$gACC*O{NqLs
zZi6Ui(6DspZJzVf4cr*kZdBo@2f37SqvyQ`XC@_G)~Bz@xW~Q8ub5$<m2EMuH}=^%
z_IIjo6x9T=7p9gk^7VHH)W;nnzV-=f^V;tc&Xo3q&LLQ>63~R~W9@E?Zk$K5<QkRf
zVSoj4Q918q&ef0TzX-=`+B*AB?WR@u$NON>Wv!^<Z=Nq;pNHAQ`7a{FyoE)N1~@v}
znpF7gtHhaoG}4A=kre#b(N}Py0(eY~_i>i|o@fT&0*(9ZTU^#`NJthQQr%F%!pIQj
zX&@u4WtKJ<2<GfXsNGW?w9LgB8G^9W^?@hvUp+>|9?aW+Nb|6_p_ExUqyq2Zc_}{Y
zRjuc@mIqp*g7k6oee@9?9yG89_a$5KP|J4Jk!>?}m=Ba0*cYngw-Ve*;R%zpqu(Tq
zeS=%#yl5T%)sYdTfY84fNwYQySg+O=CQw#amlNTIqoZA%bIhZi&vjxjH9F7IngcY6
z&d7O*LdNPgf(~kkbX=CnEeOfyc}-B?SsK&v>28IRQpSXObQ^=S-Qp)dUQpGcymDMy
z36<Rrw2J(#F*<G(WGn{fNl><#gq@LCR_0M5W{{`Q!PZceS?q}$Qoh0yB-D#v9b$@Z
z#Fd`c$j^EUH1y(WbGKq+?l;kAN`U|F)+6^lGHEj!-|bj)L)~~hcU)#m=1N3|Z-vu-
z^F6FM0{1f#tYj*gLpnEyha&7@OG49GPI6%$w(|BhZ~6jwFCx}OXkv5rb41QduNSTx
zsZKuG<qTg;8;xDx!MD5~ul<j1^?uEMkf+qm^n8d&*9=zP=EhX>bVGR{8&Np>a0yhk
zncgjhm5Ypto9rcEh~T!2osf)7jsmLdCOJ7xZ<(L*^;DY3{=%fojrCPio&lPB_PAcr
zf{oa+B>7IZ5=+hZd5}E=Y>eOVgV9H3<MXdJ^(@-neZ??CD{C4?7whtumlAf=xX<<d
zUB$7*p;n#7rBkc-jC>z3T}r#7Zsx&EM^B3}U^C49Oh)Dh(a-teDr`8?bt~uWY<*Ck
zX{mH`Iw}0f&XnC8cvzI(#joi3Ar`cx!zX=4X5`OU6A`9^MkamMX#&fLd`H#q7p9YY
z+V_UPLWuMr4h;PlqSEcWD0bZW9O*>i+sD|~hF!|?rhX({2jsd`OZ-}}Stewf6}bps
z2ql`vcehfaE51f};%ED1!!DgKx7UJAQu-W|Px9Hx?DcfQB!XA-9Btj4YnhBVfku-(
z*Fsls$Buqkp+7T0G%Xz+YXCMfBr}%|ifT|+vvLUm>#0E1H{sEGS+-pX8EXvm9EoGh
zq$KRdb<!dgfWNfKPcI~RnG|m0XJ86vWb|c(qwDH0pZdcu*37fm);xTpI>@rd-xY?$
znEnHXcY=Zq5aBNS&F(Pf<?Ab4Q1qYIVHtB?D_lri6QbhjkdFUdZ<&#`QtzM%#4+ZN
z$a<v@ciVz81X2-*w1K@L-g#v{r6dcKJZ=$h>~f$q!*Ht9^lx-NvsoBNI;(8+S;PRD
z27D2}@vvMV=YW=2Thj0SJW$K}3|{_nbQG%JcUOAOw^cm+`+*YMhSGQpoDXb;p;+y)
zsWP$SHMMR{Z#T}52|Y^X?=dhf7msw@gq6=`3KE6r3m$SrSvH@A^RfBl9?&iMjNU57
z10BF`-hZWL$X}Lat8c8OKNCjHY{n-$Y<4A|l`r+!W`BA?(Kun^?vYtwO)0DrE9s?*
zNBt<oE5NjUq2{r!E@)_)g4Z<TTL2MO2(X<SNrgJ~Rfhu4DgsN$?nNQ&X<-^YkrvYs
z$B&6J+o(<;KhHpbUSV<+!E2&={;W`C=z0Wj-U4X5-*jOj@1ws8v%ZC1%SD3RdbZFD
z;i(ogT!my1GWyYgGBtAxqn*B<cDVK#zr_2=2=24m<!B^Ei{eRSx(QHDq;N7Gb0u)=
z8SjqL;doo}=gxm)FROeS!;A56UoJ|3Bpg)=DddrJG2<`kmp=ohGdkA4n5xCB?%IKm
zZ2coue*)<~h)@}|x_^SgF9Y~i#jpq)eAmNn^iZ~D&l%yU&HD~AUU)jw;1#sl-8ya1
zWpeX2+<x5Ds;e<O4rzV)%@*ij(9t^+7VZ*9N&fZFE2VSaDT`#-LM6i|_VRXf%0aqW
zV0t31go0f8q}V#Dc*K2F#dkN0Y?pwKs2UET)AK#`7W!i!kB`RPhrgV;Z1`TPA@%=4
zZ)CiXg<tP8kV3~Dl!12*3*AQgB#aiTOJ2Mf^<wFUP-FAXTsL=GSug7xI7IR!e_qf(
z4+yQggZh|G)e#pCwMFpxL?zo<E$_c^uAky!UFqSv9&<O_6d2$<n5k7Sq*{R-f&hFy
zS(nJ)UTJAQp;8m3gx}sxl)oL(=XXVZFuja^w>{ZjSLAiOPUD5}HRCQc_Cu^Ye4%@G
zYoqpvjM8rt``w!>%1`!A{Vd+G*mCCe<6QS~)|4Kuag$0*zZJ`#tu1RaNjE1b|GiRP
zhDOYK!L|~eiB(m7{vU-izbOYm9NSgaILxMM%F2qWNe?lNMK@W6+H2CB?X<M&n17dP
z($34SdX9NdwP5)VteqDV#}4zWoi$RerdM}WK;`i&p5rEv<kfCMm6oA2zSUTkt?eRE
zpU+#iLC_l|XV!bZf`zbLsuz@dD}^{orl>K4jeEua4D~?Av%(18w8sIg(j3;!)#T7f
z8%{O3p1Ud;AM}EJZ&xc-Ktl?W%?GG=LfFY38z<oA-_FIi1T|!Y3Aly)EZvKO|7=Fk
zMGDJfQP&4r3HZ%A#>?pr>|C7e(8=CF2Y6PaA7vG7F6FwWFN7SXq4wqmYp3Qq!kB0r
zWeQEB8H}0<B$O%GZbt%#pis9{`^?gRFDJF4$p1`8N!iMll5iUvhdA}mM}W-3brlT#
zj-E)hY_*K7F=&~+=zRXI<!T1p|6MT$%`zvjYBa^hAg@^Z=h8Ox0p5n6qyB^o<>j8q
zJki`;qP(?;?HScU6TAoQw|zA+t-M(1LMG?0b}`?wSx)%Y_6HY0R(Q_4k!iZw&lbys
zX;$U5X*i_-v4{G3{q%=Kg+0s^{$R9(dYZoy1A7kks7kJ2641`Q>`DZ^J8Wc;eP<Vb
zx~ux~$zs3v->lKOG9`ayo5z8rjT?VA`8bcY?42yH_z}J54hLOHw+*h&5=uP{QCGm~
zt`zeu1!beVO|Srz4XWfWb1hji5<3r-=uVY~VP|5~1e*wzD3^(xk}bbn#l64$)&Sja
za;W=KlM+;j87q+O)$Uge_Zp;d(LG<<7}Z(E5N7-gPw}0M)%7{jexPr|r=V6(PF|(m
z5FVjyY|lkHCZSpThuktx6(}y*j=by&$EKvo$;ZdZ8I=Ic*w&hlY972SQ#QYZn)aHC
zWk{Cj@<0qe_A%vueE1@R{&Su9o^wfM+m5H8wOH%-)2c<Casiz@8J6rabP3|rFVb2n
zKU(;LGT$WIJo#C~It3xrrWNBldZmM}B0L!++GI;+$Tqt@e0|XOTTV1a7A_H5dpddA
zqacOzpl&;`yZhrTti<|Q3i3A`Mhb)(L`$Hz3m!A-X246jTkdMmxPz*#s@zUe-QUti
zey2g_Z+fr<=6&F5NNI+@76QqALdZB!35vP9+*w{stsNBkY!WSx7s<-hvi)l=?dhu4
zJ1i-A6}wI-?-%(0Ct|EQb|_Ss-mZwAKgFj<3Eck<x=Y-U331{cFEn_yi~Ys13;ULB
z!qGgyOabf<wO>kc?syy*;Hb@t#6Q>EI_71FSeZmOW3&{Dbg)x=D&WI~3g`T-z%drV
z7spr3=6CE;vaG!%rBcR<?=Qdm6owCx{IXNVDdHJeuqtgu3z+4#0*)ME#tl306kg+$
zfx6QUZwdY@gURTfW}W7Vzk5xy`X6D3IW%glFzKu~w;G>q_3?>cEl?3Mw|1bBb%^W4
z&5FltxtOi$q+Wwd;m4Xr?0@=7#1YfRb%=B*7m5pbmsRx`{=bOGklQ;SNaMO8|6XvL
zkO#|E``-`zR0W&oi&ei450r$D<7_g3yQI?()|Pg@-Q#y~HM}^VEap&skX>k73Pj*L
z{EA)8QalSJl&HGvu?S}+^<Z496tvr~>qq3**Z7O#!_Y1gV%ww1A9jS4_CT}9f9{H*
zC{Dyo=>i~q-pbFdEJjSZ6ZI<w>I)oakwG)l3T?!hFTyc5YG0K%C!1vAA3ea3SBDRb
z{xJk6z`2^kLqmFN^3U<fBqTExYzRRBrQz&05m8qREV0&d@$2CKLorb6KPKWvuoL-*
zJE9KPMGzbR*8S+~?_Qp-^p`a{`c~&9&2$CLq-JoX@1q#KZza3+f4)c2g%kzehgjKJ
zPZEn7c|C=qzRkjhplJ6HeBrON*k53E(f4h)x<t~Y5@@p_vpz+#Fp$_$hC0{I?pbWT
zyP3r*&YRn*_ozgSyFnC;&h$_dm9R|xElO^~v;pR6!l9vBO1YDaGjz#px4Uk9v7974
zSzAXGXT=fnB$95~5gxs7hu812X=&`;!w;5}>KFJR;TmFnZfQKzQh<B>wGi(EE>2QC
z7JBeVBN-Hh_N(j6q0ltU8OZ*;DDNf;P5bcBJ0TCv8;)WD;b-8ai9Q#VDaUo{iQv)s
zyw61~amb;#J6$Cn-&}fXX3P1N-E-WlD-2b{5n><QttAlq?0rO@yBcCwgd!Ut`kR-m
z`Txcb0d30zEoUq`n>z)w%ihuNzIsF6#7j*675rDtF|MDPIjGq<2a7#7+CTF?aqa?W
zeK!dp5(jz0q7%L0P5eN!hz`GERo(s7D>X@cB8txxo(k?TB){K+bZ0)l+`~%N8_Q3!
z;mNJKpc399wkacm?u~CCKeXjlHQg%Np14M70V>L+{l?_Y`01tJA=^E5{FZ7-(VIQZ
z4+Z?9FlSjiJt&1YkdvgP-}O)>{|buF?IpM}NY%QIhEb_dC8g73Vj&dI2zjj3X_RIs
zl2}ShYTB@lRyQ*uY0cVTF6Vr0ZpPgs@L;XwoUSe1Ats@3sT@+b_otZe)ND;%!8pf0
zz*Ix!0>D5*g+9VF8MPj!NEE)PAq8=`)6Qun#Dk{AElOa00g@Y~6UrE<y%dzn5mN(u
zEr5pMLR6&qR&92S6o}Y{geW|CL)WJlND^v3iZ*XVvSZ-ezAV@yCgSwy;$vwglBlkt
zPm<TSWk|B~r}=a#G*$EeydS_qkTN4u5HT8HIWZNSR9?`UpRt$GEzt$JAA$chY*gtS
z)WZTp+;c4Q<R5lHZ}0+60>zTFZu$*G#_vItBHz#9n0T}r-v6&R(efhZQVq+HUc#Vl
zHhZ1aTc5T<rd&s730bI4^*=E-zh2F^uXkxw8PXC|^i9Y`HMoVTKmZFnq>z}uuV)Ox
z8{Ev3QB_KnsaQ2Q81a%Pgz`cMA}J9rIGtR-Lp|bwj(DXbf_i@wn3Rb~?S17YV^gY0
zM@Pl?-C?>U5Qt5SNV}#CqY};xPNKj>mff;bTx~ovUrCt?k)WR_^G?;cnzrEm)~y!J
zK?3YVG9TZ)C?aY_#sFTMPI+EV|J{^-B+l+ofnI>XidMrTI*Yw%gj>6XfZUiC!tiRL
z?*`=mNtm26ApbMAg1`;^^=NMB!rr@ztUyLOq^Pd)0m2G@WmTfDr&AJT#fc+FVJqQD
zqnR4133NDsDc+?1?%>3e{)yPLZA0Iuz(y&`46hU>nM4H$J-bloT|$y1QU-kmy_76N
zi==6rcMa1&{fcPj$*%incd6hgV9oVZj$qMSbi#fudYP=4!gawTx>Od4t8t(Y<*E@}
z08>lcQ1156XFGT$@^e#<=Ls`Z{y_SPw0-3-WztPaH#(Z&8%-sF&*v9CqL$?2Ya50z
z>*i)+qe%gpTYt;n=r8wSE6J$AUz&*Fj?dH1a?-SpS+lFR{+DI(PBa>(px?K(UhuvV
zg_7q%z{?qr01<~rJw%jx9ksioQ4l%cg(<$+a{v?j<HiE9lH;Q^12AzZLNbqMF^}z2
z3iuVP!xDNKRUi=XQv!)|Z9bJH)RHg2;!|6<8(6p<p8BDob8+qDB^HRDrhT!gg)j;R
zm$4?kPILzFm^u(1v)D)~41u>7WYuf;Q6kdeM^*Iva&8dctUuR+6-f-LJ(e#I=UOJ*
z)80#SQ>2m;J7a$!duME{qkG?;=-^b<Nm-!@yi);x{d3hs|GcJFL@jUA!+@B+M#F5O
z0f=M{ckHgf`JJ74y88sr&k?9hlpT+9-cUnNpPZk^%}_PS${IM3hjH|5<~T`pTgiIW
zImjG4xP0kRk)t7^Uh|aA_J}7Fua@*b1I@Gf5KF-wbCw=$M9&P(Y-DWRvmYs;FzJMF
z2Pl#<A|+7l-Vm1YB8uqntpt=X3rj}#fRA{MdV7cEI@aR+2iwM75x!6uQb!rnxA;t!
zXo+7^f&Cfd$=|hY_^fUzlilOK$Dx9>1${#$>iGfD2KV?yg7I|?WRh-|?G<f;`pQWW
zXM)6Zh*_Z{hH@*?3sNIhjtVN$uxbP#u8s%Yth62>@t+xR7vvX}!32|GjcvJL0erVx
z4W!>`HI2oGC-P=eFg2z!V_U}B$bV!Y>X9dBaaLQ8F~FxFdTR&%Oh4#MxU}eq!T847
zOjAV@)6sUm2ruN^m#y?>Hn;w`<BT(HP^1qSW#}Su1;KjsJ1OlHxg~@}Y~;!#(q=|{
zwmkoJ4UQh)NBHyR_kmcLX$q~gK0vtdWVF36i0Pym_ZO0|^;6o-2n-+Te#0Y=XxK-)
zoQ)c$l(X4#Vx6g|YlcQ;q~9%+;P@u{?}4TZ8#8<*^Rw?uhT8Y%yx}{3C%TkT)T-i*
zk}uw9z+SB5Or@R<b-X;{R!5qi5AFW&o5)V#`(u(PO+?ttK*jN%r??L-GO$R@hR+Cg
zn3ZZo!$MJ0aVcWNCa=zK>c131pz?q`|I>-ZB`3(|rw~7xzZGc4wT>&{ArHtn6InhI
zx=5tH(n~Jh;M6)|(ypbJn_fut^~isQsdU!dI^mZ14i$9U$tY~yY;!s{S>^1vc#6h&
zmew9OeYr$El(CITfP^#<&?Xpit)!&wzzhDKO+?LFD$dFi2c<|fwr73j87z|A6-!>K
z<aSJwn&1CeC{XD<7fG#gOv_0uXg7Dgq@LZt+?DQe`&SXnb1bxW((DUDozfTUKbU`s
z&+s|ZEfTqnlzT1ejTwP9V@WuiF8Vq@_;gYUGi4|WNwp*vG?9Gqe?>39t2Btye?KJg
zR*s8AELn@F$hU-Fb4IuSxozM`%28&<KiL4rCq33i<b3&GkPZDLM`E;=I4#=j;D%H)
zFs(ie2$R?%QGv1o=}g>*X3bfmKigb)k)cO>@5NANP%}QmFfP=tsE+}mAr}~OcY!I=
zi<>}Q_bO<d_y=qd0SJ&>!%iQK_fIce+4p^RFeVl7sOh20?_nGewu)R^$yO&xFKpRs
zU%8hn;i6m%34FN-26GcPF(}v5Td^HO>$e$gcTA|TYSeqf?{#hpbygXcbk}$>JlG2J
zd!H#6`6no&EZNW8Yuzh%dZ_C~U_NQ&pF1B&;q|SA6aa4}jngU<Cz3-P{hh&Oz-_n-
z&oI;y^-JZ7P%Hbhs6%QWE0r1(cS}C(X>YVF4Ob1NE_F9qY<rF&8|{uxIZ>LAAV1#w
z6$cOPUGMXzx`)R&#tnQzFX19#?H)PVD1@USmi~pPI#J!u*HO(fY5XdPqbbmlb9c|C
zAQS!o0vPIFi2~zn_=yhhkB(CQ2J?v$?x!A?y;c3pYJChRQky_tZ9!q++w0<=-$d8Y
z)H!8<G$0q5P0obi3^kar9Ot0%A(d^Ms9UuP>ut}jD8dvzCRZ?9`*G=hDSKf24lN(J
zRHJwPpGkd&+sc)*J1xyTxxszKb-Fz_=2?{#uvg}>TuAp#3{1^t4Dr0e3>`!~MSSKE
z?J>fRT9<hJ>xFWv@iFvn+Wb@Z{QA`zO)ETa%U6<PUZ(_ObVkCgO`Xf89>lMcN>f5!
zm*ghri`VP?g?X{T{^YO{0*|DK?H1J2K~TqK=wYqkb}IKnU8;u8PH-skKpCGyQWt6m
zjCd~pb0oQH`u`Y4>|!s}>m%zDP9ECU=vr~Hzec*DbwR&XIv~1^(o%F_R&+`|22{eL
zwF2Up_vhizI?7H8N@sa_?Vx#Q@t$cC>cZp~Tm<R(2;PFfogVYQ4Z7`j77)6<gXkYt
zHXrXB7}_-E$pZ}Pv+CX#S<k&MVbYDxLEqBlTJxEUT}H8w4a=6wg>Ims?n<`g*<Z7!
zM_+UqE)#hCGSbn`zAiW8v9Oc@%3pj&B!lDVnlwyB#t&4Bc+m9n#v>nD-8F1JX4v-B
z@vGZ)*EAmcKKIWMtE(5ih_rp_LCKSEn>Y!C!IQ1YWdM$Js+f@tt$HHo(B3yQ4^?^<
zV9Vm{tp4*bPVZr_LXEDVLt@#^+9&(%i8%R6rS;jBl7s0-VV1w!%P$Uw4T@sN_~|bm
zwi9haaDaG{(j}o#c)KOKJsIf?J3hCe-}j>PGbz#5laAd_b~jl%xHm1Mi1QwW66L!b
z&Zi??QGjd$k7`UJlzA2r+wcN^v5H}Ev*PB+Dq#ge@2<m5j0xgL5A$8j6hDd;QgPu>
ze9B&ANm1E=frmu33ZIAgwi@q!v_N_tI*FjdWPUcL^wmWoz>14N-U_j{veQzM&Q-DW
zG<-iQt9R^tqE@hMO(`-qqfW6!bItX^C6aj%257tdBAX;Z{hB(I;6%m~#+EUtlWulK
znYKVDWA1;?T*iM%Ih?gp&ZP<)nk6}4VR&~)M{by{*ijLdzRqRE5Z(Wqx;P3>?v<)(
zMBM&h)WS2Z+)Huh=9!^LpQUWns=&8YBw~|nCYWx6Qj^5#<Nd}<`o({Kjycf>{(C)`
z|M4-8hTysRE?Y9(nF1=&3-gh!M#R(ejw)ChtWHs>WWIG8C0U3+EjV02PTCLM-bMet
z-HD`DQP}q{w*#tmSj^M586D~7*{NWFNki44PP-3+S>!zk5r*lV(jApbv_#^s!}qGO
zSyo+?n$R&Zvu)D0vy=SXVl7}ls)NMxRP*6w8Try!7&ASV&SBbQ_TM*Qd~T31t0F!q
z4I>BJQ?U6$GqW$;&2ek3Wnn$leiR2hfQ0kfXFdKO6srGm5uNJK3_>UMol6rpXUaZ)
zVo08&BFnKCL@h2*kPJai<>~a1o~obn3Y<6D<~vK9K-^dS(?K+gH8-%0`gtHik?2JP
z$r^HD^f2h}FXdzqHkW{rlJb%yP0vme5LR*H&_xmIt_$wogiR;P$yYMSg2f2ZythT{
z7_6;DdeAXNP%0C6k0iT;jck*1$iT#USB^bL1U6bAjs4uU>{M{#b+s$lLJ+sw$lFSw
z@Mr^r8CVjO!g25T_yXJ@TCX01`^qM3Pt3Qr;n8ZD-PAXXVr8oR)1e`_9|<x%!X`e~
zESy)AM;*-qn^9&zdRak|H5NHtlyY4%3Gupc&KS98y*vY_WuP{#So$AaePeKD(H88)
zoY=N)8xuR3*vZ5;zSy>%i6*vfJ6~+u&da^GUe&8t=g+QFb^h%-t9$iY-6<0IC3$!G
zHTwz)K=U-@NNIz!wo_NZ$SR$20%q7hd_@z9UOvCE${w5VWOJUbO#x}CPq*0I6Au+7
z#s*aMiFb7a*Md@Cw}~QEg0$J6O)1V>WC6gSNQ>^OFkr_44ZvGH!>3q0N&ZYW%xybA
zbP7}Qgl1$V0z8J(&}%LZ+7a0>yY~J)SnWeWsv0vu1?h|7e3be}$+C!Bh#?bQdT?Yj
z0P=){Tz7MhhLwJs1W*#ElBHrdF&qvq9~Wetj|X@Bxj(xulVE!UuRG>us~uw7=6<rC
z%ysL0oKPr}!w_R9HA;U_(}8-JRw2Sl1DzD0qju6lb=jW0yG>^eb~7GuC@FvkOUO{l
zjik#3G2GbCme&b$qg)J&b#otxyR=UGJ#`xOt30uvt1e`!5uki=_1|8|XADy-1<u`_
zIA`sH(7L4bIw8ZX9K|szvr1Iw-&@T8c2b<D5SAgQgpnzIoscLUNEWePMN{6zQb-49
z*!`%m<Ff+&%1tDavD)?=(F;(WSlUXO_NZg7+6=EXb*ENljuc|Yrw|q}8D9&CR(4FE
za{mM3037}LWBcK`K8E^J2P3>JO?cUv_@Rids6{&^Izu$|+>q=eUz$%AH9R8){58k^
zwPpx!+ttoa-5t*pOwEAFQ96pOVpjbSHLo;SOrdf^tGFzYq^TsSqZe-KVi6M43@;c7
zlFrpBA;28|_+w{vt1^Ly;GVC~OTcb>bFjBApd}uq_`WBAG^wjoT7SkvVi;J)8_DBv
z#lx&nTF#%~v?)@a$$+oCX7_PdsdcOep_-}8#Yqa8aS6v5LEpxb2D(CcA9C8lZ;4PU
z2bH@>Y_sx}+i7<ruOs>TuEDlHg|`y>mpL|W-_{WUaXvAdZB)e5?DDabQj(eE2Gf`o
z&@^Ifegx1@w!Uld=V}U!@|;@7G8QU+RV!01hw4>$4~%bjkkZX2*6plA)@`HYEKv56
zq5d#|ACEphlG*B$P_>l^UW>xmVDSZsy^fApuOuk9)!nLO*nA<JdyQY3))6__d_et8
z$i6Hdf-W@#H{gz4&tu$y`>ngITUK}h{x$7c;AON~{vsuu%OqqjP1ANb^Epy!Sec_r
zvY$Watgc$982_5@=<K-6AKyH<FhxCGWp7@(L5(TVG~R5wf48h$Rn_Gug$&GVi>%>m
zThV*a;tCI6e=Su}!GWs5v(<g-2w$n+0ad`1{@y?+GpEhbF6F5MtJvVPUCQVK+-znJ
z?MIsxd#slwLnwplfiMU@AZAZHI!{f71ffJX(Kk^X04q_*5eoP5TrxGl8pGhvgqLf2
zZPy<GrP>iy?mftXo%xY0JR+Ax*YMhIg*%69eedFve{O|HGZY7Z4xd9O3*5r3P42^T
z;1I|X^>bK4i$^JrnphCr++U#pMBev;fVyQ$WU>Og>q%7=lR|M-6$&M#PkUZ67lmgG
zZ68+pHm;Dz!$K}=4B!>QH|bPvo0%o{6)O>ScM|9NQSDa}qDqz}eK&%Q8*aKv^0-nq
zEQyNXHs5L3CCpoHu{WqYS8=Lv<YcBmTlLMpn}1>dsPRuAgF-zLx`O)v5^74~1v3=#
z76B*5hghj^G*w*De-H7wwlC`jlMMtjE8wo+&+^6eskND!@j^V{j9FAEw{<(XGN_R)
z^TR7M(KI<g>wUR`$Fk}dg}%CNmAX4jaqJY;MrW^pCC;dZ;JI#%N<204MkG4?8RPl+
z!LpGBnB%@8QJa`%|9&$9#tGDJfCP)zcSWn+)Xg`b)R#5*Z`+5bw<&gBi@g`bR4(AW
zJ`R`$je<fm3(JV**w=X(^-<VT+$WhUe4!w1RQ4o~)-5ov6#JHBF&J<srTaX%7!_Uw
zeF1DwE$*mHendNbZk_)^w_rrigX@~8$hjU{{uK>+nqh0JEQ@?wtYUkH5jUp`F4|Tq
zOO+#fvSd0m3o$P>ICCX~Bd73#cY`#lizM<bK3=MAu$s{fTWaKRg|#f`b1h6UR-J2;
z-1Di!1wPL}cb4_z|Jha@@5e)fg+>g8uKze%L~o-EdFV3@@(D3~*$5z>)b%&{(ZuJi
zjjHsuh*ms%kS$dN_=Ho&=>r<AA2MpH@TAWTeoU?E<#DHD*=fr^@;Qg~_RG;Y`KUXG
zQp@_X4~c??1hxE~jiH>y-k~_puaO%L`nf^tvRglcmJ&+b;Jd*tMn~E)8OyA|UxSqz
zeX??Y{@zD2FY^l`hjHC7B}U@jyswgAvJUtpN0(@a$pQ%tST1r3SXg8sIISFVXr@6R
zc)s}xTG>0P>!m>%XSfX}nUg6}6%&<Z9&5jkBc4}KrS7GD?15{N6FN-eLhvOL|3Nnv
z>x_!a;QG^-Xv*62*aNDwOUbuu&80g*aOb?&(2Bx|E~x&6ZA8hbvkjSemfJR~VNj-B
z$`;q&PE%n95a%rf#Ocpj4n3zWXF@@ZO_%QKQogEU@B&nL7KKOWZ4VwM=hNVuW<@DZ
z@QZ3MN$-GuY|<Y<=^R>hL*dYPPY~?d*vRDBr4cc7#>{N%kO;AX(>-zs=$|{|wRZZ{
zt5v%4qM%q&f?ak%M<foeLe`%J4vgdzLSB>RCqQ}uh9)r-Zv7Mv()3L3?t9$Hm;WK2
z<u@3{A7?Z!Y}|Oat`+LE2MtbymOd|OsU=Jmu`$zW8D$cF+t_O;edwSNi;$WuXD50M
zjC07ni=3i{W5i(CmtT|EnH+nn7U9JAZ1?m$iiUUck}H1NeTwXg?gsAlYv5I_Zl|K)
zqTtE^R!Zso3LV3a<JNSCZYT|UxdQ}Iz@Q@h!@oBqOyD1!)@j3oz{CS4e@eeG(MyH@
zfEL|B_CB&D%N6VOFYtS#_o155|HX$#Md0pz&5ei}6XpMW8h`#E;K)1w)Jzc09yCM$
zOxoHpt>f#4cavdjj{8N95E(d^ZA?ohC=gHhoj)TZmPi~<R=5$dY{t<i<EVgjDV0s`
zk`n1D+MUWH<H$J9Q&G5-?8p~fBtRdiTn`V|GIsQDjlLf6`7V)mb70@4XnH881KT2%
z(4i<~p(5RK??>9xjUb$!>s;>0<W61k-h7bJx@=_zdJ5AAi!t|^0{7q@L+zqmDf9ar
za8UlxhWXcZ^*N3bhL`+1fr5^X10MX~xvLmTMz@A>WTiT3+j<4b2tJK4WQ1*g!dU0D
zD|=41_|xl(?&`P~=(Jc{SZK!F&PJoWAYJU!7fCdp%mPil`5q=L-ctEd`%f7rLG;Hc
z-=Zq0IOw3yTE34FLbzNA=d*vqO4oTf0Kfz;0_RbRB7!2aBa~iTB%Ssv%{o*{d{=DO
z1WexjvOFj(=ogpRdxv^BN_c{OlF(*_SeV!^E)ljI2x?{$W|WxtU}s>d&^Ih~t?2Ly
z1eZoq?07-OqvFd{7ZDR7!z1BC*ZALoPyjrMi5ooaIox~BJ$dL1GVQq)pb~GWEKs$G
zLtkr%Vz)aP^L#@gZ-4uPwZ0)g49T)F)+*bT4{6NEGLBQwND%~@>fH-5F0O#(#*FdB
z_9~o&*p}u8h~zIXvQ^YR#NV9LGpw~+aRNENIR+`jpTzoALW!)k?&G7236VKiemJsP
zi^2H??cqjq$x9IybYtBw^D_RqK1?Ej&OYw<Ua0CrAp?R+g@HlLzk~B8CEc}dO=f<&
zXye1rm_i*yD4s8gsv#dNMG%FN;_H9jrdkk#Q2bnk-FS6qXJ}^3?|@ND%_0UN6V~ag
zuWbY>Dg399v_B;^k9JGj`xE9>{cWL@ol4tWdqq-7SnHTF2?WqiG3Nhn)V?XSzLq);
zDNi?8h|xQ6@lORI0YL?+N<}9D;RVEyQ8IIS5skvcp%*yTelS6(lURXUcV~9YN{+L-
z6=)3(H<p)6b27^QepPp1vDNEyuRA|Q7w=jwlNyrRuuV~D?+ex?Lv*#%d-S?z7$Lrg
zjIJGRa?3;6`U$StWT^lJ>%AcP;=yP)UGDY8eJJ@`r!tu~VQ{pTi!8^MYYY(U42H+C
z+1tzKO!I_Jw6h*<ZRHKMgjhtmdC+8I*JgRqknra!B#jBt_Q84SgiBIa>@^GvVRTv^
zM92Vrs0ziH0`J}(N=QFt8kVJxjhOqe?yq8)y&|c`B>&+#KnUr8C7nFSR+qv{u&Szw
zyJKu|&vw|<@O^-yKG?+oahzw_t0;3pBn>sf@7S5m7R3ZT-hz0DR~~ZCI)EBm8a;~|
zJ<G<G!U>*!$WOHK2F^JAL7C;M^SMPqH8qSBg!WhX$?B#Go>l?D?Jq#{%ID>F@DWuu
zbksf!h%umk3Hi`qw{RH6cf`JR;}F-1x*FuJ?u&^_gu+8Js*8(?sKqL`R4B~m9``==
zjTA%(P@Nxfd=S5~{pYiwsMRh)Mk`f8RQW*TwhpzeE01*Ka#7Om7a?)zX^8(n%I^%_
ze#o6|K7Apc=Ov&HQjocD(bbZz2tEPK(Vi3vSyB|Gx>UdOAgt2c$Jr+a?S9eRA8OF7
zX($z6VBo-Da#v_;J=-020-3_a$44Dt<E`ouVB3tHc(E!~cagNGmE3VaMT#dfS<$t6
z{vNUb&_l3=B6oHw=76F(t@t41ciyn7phm%|@645X$`XPkDjLuT)k+Mu{@j|pO9t4-
zmMsG7`VxCZ324#cLw*(xoR8VwS?1ZoK&fPoNy*9mQXsdVBeYD;&J{I6$Q=tv9SOVw
z`~*#P=MqCOG{fbX`ELb8@XGB%0%fi#gH47<{pHS|7_SA~qjtsn=w06>io?*@C1nga
z>l_%$jnxS&0+u>kvNI3L^PHs|Fc0S#n8CbVnT@Kh9+(wV`<Y;D7?Ydv@FVd!E!a?0
z>6EFp|5OSyF8W?u8^r@g)$rA$y8!OuS9l&O2cwuA#reaDREa|_QrM>#|H7c8HmJOZ
zl_BB+O^levLXSUf<g(AZn%uAN?Z!}bX9pfXh`VI|<{CsP5dS3=PUP|*@VpCjz-OW2
z7&T>xcDD=r<L9Wu8|+mPJv1s+fGfbSCrGI37VIAVZ`g~nY;LuDPS!OwR}l~w17$H0
zb#;=q(w6en+chYfNi}HU`R9f1H)*#3eH$fwjChgeA!{u)>!&+U23j<{<y&O<WZIc)
zEG>2l`%O_k-t4xD*28>L-hnl0<NbyQ@M`wkQ~3nkIq+_SH)iyW^V8oZv9@;<^>ODN
zd<|h#<rWqJ_74sQb^T^`W(GjHa;Y7KwIc22(DaHKbxCUK2Vr(%XjNcv_PK;%5cnFR
zBYPC=_tN{4?70E2Ji}Qb-z(}|F9TZ?pGP)QbV}H_kS>g2=85?Al(V>Vq0HY2kxIVI
ze#N2W1NxYCC)iO9#^x0YD(`Qzm6#on_EQA&$mEzLd$z$_6s_;E_7>nOi+4+6ie+Ve
zOn9Y;dtUc<eh{b8nv$#Q7lTp&;l<c>`uPppD=g_^jcUL0x7!m=%J~EW@6YoY0p`#F
zihf**ar(?{%Fo`Kze|d;rjr#rl3`csS&iQ1S?XfH-dtMj+}Ys#zMRr$%`Crt-Qb?r
z0Y;_bwMujUZi-Ac%>x0#y-4(EO199Wkh?d>?4$vtMswHgsq9zq4VnTraG2YB^14wW
z7kwF+T;KHqxm<;aMU7K(NnY=rpOV`k+c>1dHa#_}X9V9=o_5$q$wq)0N?4AYf25>z
z%=qti>ZH+XB_~d$*gSB4Ms|iQ3cgxyj~=c~s?YxGB%SoK#FVc-+jw&WQF>i!y*Q{{
zWCzhX2ygsv$4370)pGcsR=j@@-Bvmke-6((N|{wbnj{K?X?h$pDTP6aU~X(vZKyyf
z0ToKb2ZJ~_GTjC!B-LV>66F{a4d6am=x;T`QJJuair+9;g5Hr9-(19M!G*8o&$VlQ
z$mI;5-99u<(TKtmkp2BdtP`hDU~+=l_jg;S=tWt@OTPE)`C?N)ztS_O+ew=XRKTNW
z0G!Fb2vwzJ;~KIG>#)_{C4YpdJnc0rV6^?zJ+ozrXHi-+Z%C{c+{@;MrN9d;V|#3>
z*^A}$XSUrNzS;W7W^<pG<v9EeM(>At7-9$FKH(UIpgSA|JmkF+ffiEobI2Lq#y1B%
zW_0I!Np@3lNzB}db_{BuccanLJt*?TgA)TsEJw^r{RkcT#><G%mk2&PLjYnN@xEOm
z!`d>bA3sdUyEvz1Gq`OdT5aTS_t8`TQG*Brb|V!t^iG>5=sd<@C)V0uurIEKdKvW)
z8O<^|pCA}b1uJZ)b2p)Ic~$<PNH_$<mYF2hKFGcY#fc;S#9E!&7-<@Ei=4u_cC^06
zmJKPikL@M=M0K=sYpIFUaaT1MDnYKPJNkRVgG6-?Ppw&xs(zj{l(zbM)^rF~0NlPb
z>K$ZUi;W02HdH0otJH{Hbmu`0L+7b4r1QEXg+%xBBV|iG%ynutlMrTj3Z>lSe(WY?
zYX}ScAFng2{^+5P6hL_sH`n0k?ML~F@$O~(eNuDfAfx`6;rJ3aE;)0XIz<CQ4WgR*
zOasCQP$nJWpAGU{A3B$>))Yf(n6eVP;Sm+TuP*!cZSr8mXW1FY9>J-Fx3h?}3|fvo
zTa=cSa6d>B%qbh3%p`O`!dR?V)<~L{;Oq2GlF%F_Kw5u6hJrrPYDzeI>$!tzsx?H~
zJBNS0ZHlI5wN|b?MPud93MiG)erfSs#X#Hul)a>W<F!E8>F#^!mzW9R6^D`MP4`6U
z%MT_b<DMQ3pM5)4c8%0zBCra-&s!mYxb)cKDyz{}ME^&;>CpqMbGpV<fI%F`m+5eu
z>Pt6u1mfd3uNou|+Okdrho0;l|I#w<1YF=`q`3I~U5VbyUMvZV!)S=s=JpbDETJ|)
z$EGhR<sVOJeNZjkS(|k>Y1EA;c;v_ks$P3=HnsN#FxUQ{`kC5|>?rsL%<hj2?dY7|
z79{Ok;49nDK<*E^TQc`y`p?PpF^ReE9{!;K7*CU3iM*XOuhrJ{(2~g^DL244XH}q6
zBfsoW8m9MohkHn+wXwxfQ9+X!kJzmFKjwkZ<UCTyxQ-N3N-l+F3<V{t2YGF(5NsPT
z1aKBAA8sLH75ahAgmP*fstF80pt%jDr~+t-3Nap^$oMMDcPJJ+=v{;;aexI6M{}`U
z<j^dfAl*iTUJBUUjXf0|F5^;?3B^Gw96bn;Mv<-Mb|X$FNiA&#aZ2ey7C4pQsG~e`
z>}Es{2Hziu*MfOT=Z5HL>GoBdlwhMmI|VxqA@QR_zKw}-ra<s|KD^3;&De^HKB&aW
zwrs6-V8%vLJ`}BEY5@KBO=8$nZmXZT`w}V&OFHXMuzJh6Ar<VtNbaL4Sp4=%oRO}&
z1}=e|)M-e>%RIt1si0QTjTkY9l0U))ufM=HN(Zi-!`y|Kl`jod)xJ^pu~PJ*h1nHj
zn651R_vbrpM9Wz(uWl8lS1DY|PCeRKs#V2qRsQ3)I`;NhVT2JdCZk6wR!DBNmXiH4
zz}`B4@0xLkD~OG8SnZp>f}ssNw5_4j(yr+P8|aOBwo7kqLSHVb2-#vO!3Al;K;&8c
zsD60=ZrF|iZ?^rEej#bi($R}pWQzJeAUJ{vNCd{f$xGyHVic>y-?PJ!UE}=q#s1K_
z!f|1My6EzpKW_v8MCl(M3CmYSPX+B39xGRcC-tCon;|W~uFiwkicRibos==?nBLq}
zbQU`WTzk{wBjcW_G)VoCXs20Nr}qrtJRD}ztGQB74NBOd4xq{-Y`@B3(;NgB@Fyuy
ztu5`WVpIvY0$QWNya)Pj$g%#q<;NKNySZ<SS0q3Cu^$7Z>g6(QW?`h)tj9+Zx+Yva
z{645vN5IY%h&d=OA@B^~%){}*f~g9jAAiG{kw*ySkdy?bmg3{>o|jO%4vhFBvf;#<
zZrp+)83OAJl8^lV^SjB%4KDnL?hi6=T0cvnQyq@QBiE8tk)5HXzK3j4JeK+-JK*4X
zSGQGo)CS;Nt<8`zZPygIP#0G@9;J2sFXHdHA^qm;5ko5jh#fZ8O+TAO>-!&+2%fcZ
z<mAGpcY!(BS2Rw4=yKpCJ*Js)w`ajGd1<HE$T`o5YxeUp&`Ztr(*rl>?i{bIV(8q9
zkcyXDR587xiWQ)1<6nt2b3)iQ(`kP;qb;BSruCPaIKR=mWYNVWvZKNLP-Cw4A*_?g
zOOAQ6y2A(!$-GFE+oe>N8LF&C0#6ZaO3u8GSqZ#7{LBzz6uEtnYq!dHvskhND4Ra~
zsFPIYst2%^dwiyAX3~6Pmbd^4>k=s=cZbJOf#gIVWv%<Z3<~_so{1=m^t#YiNN@yD
zx^8>-EdBo=c;TQj$2As21nA2)DjFms+z&{fr;e43l_#XSHISe5^AvopP7;htKH4fA
zO3$rxF=m*;DYgOpW+eivOY60|ywtg{KV)yqc^5|C%8SGF2*Q8Hj53bV1ivZBo)*WN
zG)^k$AZ(Dl;QXaMBUa#J->4(VQN^JrpOuf3QuYOkS`QM+$^Cp?au3MlMP(f1NaJgK
zdK&+9xoTf+-!NN(E;Qr+$VeR{10f`uKi|Rkk~V2tjuUl><5|d<fW>~d{j|-geh>6Q
zvfg1#&143VfD|F;6oRQvJz@qS0T@xntsJbNPP`g?ZP57<Sc&^F3l1A#%z@>zLd~1)
zjwNd~t|0|;No3p|CQZaoLXWd1DsmErSE_7LO+b@sG{ixRpc{4DYOdlW%U7n5iMktW
z0M<7c`n`Z9%oy_c%nyd7nX>Q_FG#Pr)#Zr_nj$PNK~zyI5|b#8bbP&7fLswI;eZso
z31M&~EaM@RUwI9uB^T~(@a8gbE&sibdetZ37(z$wNAdqZrj3gg)9ZDg{YM<0q`N;Q
zCs(F^rW>B#I45%1an5fhGS;I7$w-|<XcPg*8!Qb+O19$b&bd_V#S-p6bbj~rN&;fS
zKVZ;jI=WUPl_RNUS|^usUkFoc2_Yk(gE5O*bCmT(U7$Hm_a(hQB*ZW^IwQ_gXz>z4
z735=eF|en~jgA`H0j4~JBOzg6b?s+t+A21xqSAVPj;&Bw&7}NOhMjKl+PpN?oE3xs
z`fu~`AB>IXqts|t5GH_(*7kqeD#xBs{=eF|>s)!#fKCZu7+hv0XDJ?|&{5k>Zs7uz
zc+KMMCv$Ouww<$o$TLtdtMMj?z9J2Gh_UrDhcZJy{oH)D9^+O#mdW7VyA_|o@QH`(
zs0(6)up?Gc@ls{7-;ig~)s7&uNa@uy>Ld!&%4wzIOrXe!aZ~x&K&SzQBC;9myQ#jr
zEdN1|eT#-tGsBv-h<|O?Ro(qE=0@2p8w%D>p(~Y^Z?+41>rde~@;&sV3E6(d6E@da
zJ=d3OpcXMvENZ?F6cb7CxXLylB_V7$m-4$T)BN(z>0CYJ#t8OtYWB|@15eGB)106o
zD|hi-uXLup1^@H&$0nfTT(AMm()r>m@57ZjGguLgAm|^09&J_>9m@ZpASW-C#nM8N
zkUlpqd`B1(*_ranDGlY~-tUEyZDLvfZ+e{t=xsXL0kU^v+P-6IF(6PGIuO+`)rh_g
z;Gtx~+g*cM0Ko-s5N-d>Ld|G~_lz7xz=hEqF8LUHF_Hj!(VnVi<w_!LtE!sOt1Ft5
zVvq(Av@F|SjPnZ4HW%%}G&Jz?x;&jWy483yrO+hS#Hn5>(6!8#T1_|h*^llLL$>_`
z^f3_hCOq$28QtSsG&i_Do%czJ707?Fg_qOhv;H2XFg{Cvo^|)yK%C=!)kszsHM6en
zoyz=%ve|&BKQR-=>0_RMWMl%s<Vf=j_)C1kU#xy$-p|-2Q#lNM$&67ZSomF1s_>(j
zp_sw+pQmaj?6(GG25EcSw?CV*W(C!PZY&Cl3k<Md9!LOpS%@A`aHLMP<0G@)YuWCr
zHP`OQPz~u3ynwFD4V9RH!ZWj*zERKB%#B1#T^4|V_>fiF6iT|+a;X#dpQpu4v+56A
z%d+M##YF<;FOR||yF~=cN1o$0cEvY<Ipca_+2r_yRIB<`L!yqEQBL%hoCEv)>k5Rq
z1VccmE&~rJ@gplN+vvl=-BRkyu45U3tCpUkZ(jNFV>OMlfh^|$+u=`s)_+86>>mab
zX)gfH=jQ22G>b5cbJ2}&d(Ym#kFL!gmDWw=i?*S(@gEf7B7!Xkq_#WHC>@%cp32=P
zNnJn44W#drVLSFN_4LYG1KK4HTLh6F7JFkW*46=74%%8o(&O4%o8P5#<2FTIvuz2Q
zwK(Nx!bmz)qkN-s3>3toRx628TtONm*BgL?mMM0$PZ1-9XJC)BMg_;%@~pb=oB|=t
z2^r;A@mPG%hxZsqc@~tpCu3tYm2997dGSC94LEm>_4b2%(3Gvc!n&L(v^44d-ml8`
zfRBA$31*px$VwIEU4$ah8M%TRN?A_fDY#XPYRafD79uN3aXRU4M=U82-vjo!$6P=}
z<*43$vaQTxfTX%p=h8FHI!7ds&3Dvy-_tMRSA7&PTdzJWuYl;Pw6EO}ZMOR_2E^iC
zr}Ib<?k}dm0Tw=EDINsU-zdA=6dMbi>8}mV)-Dg57UE^j9<IBg9@+5)ZEruPVz0F<
zmT=gGRkY?;et7<mcO0N;G$-`|5Qc>oZ9B`aKi+5%Qnt9W!rD#o=;G)g5!TD+5F~%>
ziHs5*g87&{uEQkGmx?18EnuXVR}yEX)J&jwESAN_Xfkb~FbMchkaf~+d_K5^aox+5
z+}R+n!RL4knhRt;w($Gsz1H_CG}SKF7$%u@k|D|hJbJ<^uX+Lq)-?D4S1Wa+%f|cY
z>loOH+cV@koGyA~R=%vNGyd1o^;%`@5F{O8*GM|5BoJ!DO6=Mk+DKuY*wvUW$LG<c
zY092bPtJMyk3oUCQk(RwbvB}L(KTstfk&O@`ZL&fmj^^|__8-mtRheOl_f!aXk1Zk
z77@%?Iuv_<t)=czzY-b&dTi8jgMBmT%xK=}e>6X74sM`Qx>0b$l*^=xJ&~l<t__|7
zjetABZRXZWQoNfCkx-9)%>20VZ%}{#)b9(xL?EQrE*YYx#s9YQsEWI#cxD|dglX)$
zS_$n0(o}ePmv&CMXU>$_*xpF7%U3*D$`P?+ohPDXsVPI4YeRGcGCLV!7UkvWD{iiJ
z43!xKrooOV--Hy=4q8_Egow^@5H6--u~dJ6rX?ec9u7FUB5Y=%EDn<qa!;v(l6n0A
z5u1p`n*3k;C?6G_=G5Uk>mfmL6kb*5E7YGdCKh8YBcs2AxwF|tXbhMx$~KP2V{y1S
zEWf`I)Vqb_2zYlIeeUZ|*kL)~NfbD@`Phl?d?!&v&%pm#v2Q9AyId3Sx=kIxdpPU0
zIiUs$ek7Gq+-bbTVJKcRe>hT-Wg5}{_seBN#IhRf#JM?jYng>JC?OZW)ddnJ7#MJ1
z%}Pdx)Q@oHh?l<HJV^`wET%D;>7OBvz>t==Yl%9<2vCCT?pgtzYeJ9Pa{3MN8_CMN
zWjD$vlwRsdj9b8>r^a<9!5#6AQFAMy=Py{;)Eq^TaurJkM1a*gjAy#}3xmqOk?_>o
zl85#zg<cg>V!myjzWd8foJS`7ai;M5&HCvAg)|U1{%NGsls~%mgDr36@41xENZ8IL
zGPErBG+<*xORLk-(NnAHR-r1ql~asCq&H+?cC5PxAFi{d>&vb{iLEK*_17{}QC&6P
zaje>b{k6oYQoW797OzX62*ZjTO$k*B^({PeqcX3{_gue}sVQdJK0Cv5k@C$Uuy{V-
zrwN~w1AB8DYOW5(ch}AidC~@C_9DZx4$ZXA1;7k}DI1q{_58BE1A60z;}^9L_a~5H
zmLU8wh{@9oXRt1wUF2MaV~I*MLDU9P3Ca*szsh=Z<sj1y_k?r+FkSSE6Nro9W0LF8
zZQjmPF5^PBsC0T-3M_3d7n#odUS>PS6Cgiutk8LiiWXClTK7QB;!~88ES*H(eAwbB
z1?VV+^NMB&*F={F6VNSW{SGmtk54By-IY5L#=PA9+utYM4P--}9Is)vn5$-^vKK%s
zvZfUW2heoTVPdE7&7a~OT(<NkuNAY&M}cOUWkuSf;63?mJd}F8@atlfV4-FcMShfB
z?`|uMH<~FUe1Sg`)D6p?1o5y392Z_r094ugCh$B!_69bb>OS^%o1E@;)p#oxGa|U)
zyMAwo!;J&sO2Xhj!i8WrovleY;if?kZwS^%GnZd|1`tRfrvu&kK0v7i&09iI{-)D;
zjdacGeL-B+-)#v$nV0~<qr-oWory{+I69yiIGh9CZgej;5H=_OXxJ=hAJ|ed0jkBS
zGU4t>QOE*l@WWYsP48GJaRmEBnpw<-f6<y=Q9B%l9K5%PRN*)-j`Wo)#aD#*fdfs%
ztc@(iE*+n8QI0mOf^{%h%0ikX+`}jeBPrn=@qteC`#E3sDa5kKK4j3RI<?7Q3>#sB
zk12!!#*Y0Gn&E>xlF}!qtS2}!08A>$((296<EQ=dhgmfl0(_rW(BN;7^~*SLfX8`L
z8hxQ|P3N>cQlmmpTW@yVg>G7%vE`yiGu9LlXix(tgR^e3lE=fbE;-Nt8c!reNEWrh
zAmQCL9fSEH+7tX9igNnV(gw;q($I%WN+w$el^A|7gRaX0CY{FN??^KX0GYQ?tElX0
z6B)m<?{RMy`9XSds8Eyn^%OhawJOXx`M8y_wSXU!qJed3>ZCS)VP?<=d{c`1*{(%~
zmX^|r$3RQY-=47fF>AG@r}qRI#z|xM)uQ&TINRA(FSLJVf0>D~Bl)Vkms%3p`v&#O
zJ${oCvL3`Cx3X4hp8yhkE53)^$tzBc_B1ws2Sobf*@UV-d)>pySf()O`P35u5F$V&
zl$(4CO7tS8hk<4M-NlEcUPLz<K;eDOb{E^35>^9ZGB~tI3rU<lkM#0q=E0tQ;f&!^
zN1$0^hkGmLN{9xJLf4EOCez;oeT^QuYM|}3=x6WAa}IyFBoZ_@=lY%1Qyfn@hvgDY
z_PjjbDiv<Z@$LJ9TCtW?Id|Lo?+XheZT!ZEy|w1?&E_y@JLmF)%Q79iex%FESS~;N
z{wQ>30!#I~ku0Hh9<r~@lXGO14%TdbvZr|{gPV3{O7v2p@`FE94!`3`X|Rgb1TXO5
zLj8&@<dl2e|I5f_N@LMB3FqhOPv}x_1VQ)!RR+v5r1~qx5LFGhVokV2T6_p=`L2%t
za;|*^Q6P5ndnmp#PB}FFK;0fqJhp1svPtX(E<wQbAL5xHW|t~}*aP0*v%u7gKab9>
z2mBf;i=9fAPRS0U<pM<f*K8*WpzDSbNW=gr_N~ut%ExAS$M1N_;@DqpplijOJK=N0
ze+Zy*;OV;S&3)%v{!(WLQ&adVv*i4??w8>o4r=byS4o`|`euw;c<Le(0|i!+kkfrq
zvj;zXy7Ax{yte$dngMD`gk_!IK9$v8iVbw)^mvI4z7za4U#l7|7s5L7+?H~;U8lj1
z7DwOC#RT&jtTq=e_i*0-6veiC===*~y@oi#J62WYE$^=XEdRQ<rC_leCDbe=t*LcJ
zz0sr>+9>l1xb6f9Z=)cFju1}|3_;?c4)-F0qbwc65mUUaw;x!I{zE|GX{Fq9OLPr2
zvzP-*l@dG$@!TPM;(q#8JP&pp;;}DJ{30CJZ6hn61dENMIUNELF)ag`ypz|BQZJh_
zj#xs0hSU=y321}7q#(QG!BVg)QGH^gw8?X1#ZuBLcyoXOh;VqB8nIR&cqo;sl~VhY
zir*u429LF%{a46&yC&5Yek0x9$Ae@smNfIe0jci-r3VDFa~}O_xqs>8mAh@1K1Gv@
zU7Q0*Z*(%c;`p@P%XB}!JC|%v<NGS_r(`<OHE!^ENG`p@JCly?*{}We9Jk}UsUY>_
zE-ZLI<~zWU=#i~D6j+x(KD6Cr>p=%536z;>%?7I=C(4`m%QIpEAM{%pAF(oQ5n32E
z1UfZ@GAx1?EXLivD>@`+P5g!$3JUyk0sl!7?#~;9sFbwi{UK}NC{W7(%H<$8-9RE2
z-EM<Bo34@_!`S8zLQ9UaSPaVamW1bjMNVa80D++F!9RJ^DXBpgKV6k<Vf}fKAW-<2
z@zCI^a&uWB-XxLR?r_|9?azjOf{k;zb8PP$q@?Q_e<&02$wB%4-rpjiYnx-iR8A<6
z03m2jN>sU-a_b9jRH{nJ49~iPmWS!M_`f30yO0*ge`e?G3yy2@Ey!}Ma(zC&iCe<m
z11`nTa!m2fFV;up61VB|F5C1e{Km__2rq4pL++e)=e%4V-k-Psl4$go5r|(dlmYcA
zPej!3&Y_y2tQr&GYcW;`wGVc8Sq-?>?qz6@vhMPE5E~o(uFe-R*np~Kgda{baa3pm
zg|ZtZacz2IcX?bsD~^Jx&zRW&flX7yyY|1e=Ao5PMeU60eKIWmD{<M!!n7v})Hs=J
z?zIvR=CUvav2B6y8)VS+leqP}sTCYY!tA2+EP}&9{1Z)XGeCF|IuC&Y+%a}QPkx#_
z{-+Pm71?_T!40eMQ(V%qxJ|8oz+55W3C1qO(r8EHJ?CF$59;-f#+EujDq16n(Q}9x
zsAM13nJoA3{wIw%)Uj@A(7-*Df3~M^$S(=an!EJ(BFPhz6i$J1hCWH5LRih0@3}5_
z)1axohai8c5d^DP=b&r33W7F%31?D<i&$h0dSfy;3D%Q(j5k)2M-*pMdeC0*jjlJO
z^VXer==p;hkELs%co^OQq}EnTRbv#$+EGoK&E(|t(^BD{q?cf)RtM2aA&@`IO02D<
zBq+y^xAFVO3vH3Qc-<W0g84;2J_NzhihoJ$X138WNl4Y3_~`g1CjCtRCZjaHN$}WD
zuibGZnptXrLo!6#wN?@JQ^j#i1441ODcuduC=xwy^H^lOPiffzD}i&lwImbj2q2BL
z)3H=Wsqz{@jr(aKpE>R8JU5M`!iykx)%P?c7=24HIu|2okAZ7a;Q_F5tHk&(WPyYw
z$Juc5YX^5I_|WZlpNL5$;rS?3REt3GALZ$l;67=kTOZ3UNgcRbdE;J6TBJu5WUgYQ
zXPY*ahYl@}8u$X}!oM-{(`Z{yW^RCiqh-b;>1SshRH3eQ^gvI^CS(avsR}}U^7F4=
zN|OKk>AE<BKc<4+VL2}jo`ZPYScZwbwUl!wSm)>rF8i#qV}qw{+PJh&772p(U$Ri9
z0pzM&!mfWzE_b;&>;teW_0`ho-hf+7WH_sjsY%oe0q7Lmk})C<dYe+S3V#3nqi&w(
zI<|cL>y86^hBNSauXV9Zj;5n&SV!QB{>&okq6B1+!&Ko^R03X58Eos`<MMK8?ld`Z
ze%`Thcajw|D(;nhvs=#h#fAj)eNwU0n5(URbOq-Sj_O|0fInCsV#v}VlwD2LYIy<m
z+xfc@127q4e$#|t57#}!>3a2S%T|ZYC?NZ@aeb94!&G*dVz(2gCXnfM84Z_bP}STH
za+=zBDT$8LF_Hz8YX0`0)jpi6lTJ1WVK@z@D`F^tdb3odDwSh|(R}Nq#P_C^=cYjM
zMINVJKq3^)33Q`Jdf8dG&F>*G@=JIN$X<?rkN`vsOL@Bfnv@avq$I7omUVt1(*zL7
zg@S_y4+u;frzYmA8AjUX^WQ*-bv<}BISZQh6w0SAV(=U6lNpTvs=2Tj%h<T}pQu<r
zwSiqiYXm<zle(>~jFmeFFx0v^b*DNTSCP!!28r@(t`vV%;abUZ-X9!t_~Rk2BX?K<
zs8?|JD0_`!{paf`xmI?pNiOQt>-x++!Cas0CdaQ%>YnRyAStMqCsqFLCi3LS_+5Fa
zD)UkhwUbwR4EUXaNwftHx&Mk|7ddp6wbs2`Z<R+jG)YKJ4rR|YkPZ<au(Ik;rABz*
z??Fu4@N?~k#)SmTe@&TRUYY8TG9f$xXwR#k^1VO_w-Nj8eqSPW3O_TtsFd(<U_h#f
zq>`FwGrp@>Q-sMOnI}l)6DeMYM%ZuNLmS&|)iQ2(pGF5GkF;Y<mpOHB;6iwL0?p|c
zdyd```j3vg*#Y)Aq7b#g8hI|o>lX|VHfc-IHkyTTS0>fau7Bod9r8r5R-c0ba@g%z
zHPi>e_n~c~c0jv>XEiHV7xs40&sw`2szz38%h(hj7)n~{`g!h3woc9w#S;5=J-Z~{
zN#m*ZQEZRm4%JuMB1P5H1DrR4i)bG#Dr1VDioH2F(P0v5Pnv;WSU*TZ4PYIQ2|mvR
zxEmU9aeI5^$%i2x_}sXM5M|o|?udUnkL);d_)fJjiwMJp%n{+-s^~68LD=nh%&j8`
zsI#+tK2HC{nnYzMuoF{w*_i;1Wq~qpI~AP9#(P%9Is%`t!QDUwZJQs27CfjNjuNrF
z<C%c3!tS<fHT)Sn_$1^m%K|HB%X+m7KwU-ssP*sj)BY_pfvsbl8F)s(imH1D&09yC
zL#}J<!S@V0X#WqL!+#2w?<tdLx1_zI!?||{cHgab2?x95#41~?YdHJpb5@ovN^4d4
z!R^w4A9J*!o*I|`KHDAV=u#DED;Y05TR)1`ZA<ktN*ae!B{Xdo#u`zJD;vJRa?NIP
zdHiUsohjO;jf$H%QjK*1_4LCyN=;>WfJO;YUD8kMx>nUOf`%5RcFJ>ap-+8*focY1
zCiUT^S8LC%TPsV{a{j)9u-Vg%ueNK*+_V>+9@<a~Jm;_54IGxKZR*Bnla82k=66#t
zvFa=ryG{G`6Cl}`{t;pB3c`oPKjr{9j%91H_p=+;!hW9x0Jtq66C*0}E6k8#<N0)R
zXD<}lEA{;Ay%WOYxAz5XVa0aKYvu`mYecs(J&JJ;?)l@m+)}maRHfQ94!AM8tyS#G
z5DqI*oM|C)yAfbr(zP*u2YahvkuEdZ!YEzT_DSm<#N6oEc~DEj<6mtkCShjmz(<n^
z%tHn=EBR?gWtt!WB3cSd9Na2afZDed@{1me;;qb!C~R#;9;R^t!1i{Eoi;U<<YPN)
z7DD+Pn}&hzdRJd&Oogicp-CuT+iSLqkpUYGWA?8g*1aj6sM?*6IS)C`uoKM(gW5DE
zK^kZpu=hIyp~Ji-eVYAT`Lbl;gXbv|C@TP}zcU`r4=Dh6CS0Odxn$la5WQqpf5IAm
zkSKIm_1lGs^op`<oc&&-L&v(EBM{~0=jJYR4;gi++~m4>f>3J0GH_$CW^U=EZpI3*
zW`1_Q=*0j-i&{78@QOqx8FdULUcPpli05OMalwQvG5fF+?5gNo>uvw3HH~4B+@9@T
z&R}{gG?@t?G$cZ9Vh)`=Oz#=Cgj)mrp|(n&9?+9lOG-;k+wu<UkS$ggWY7TTq0x_y
zT$9!1aDV<z?0LNyh&P$m7<GME55|uW@y0smdE^H*;ViMy0c*KBcktpCA#1XQ2~P>C
zusob<Kks(>S~XwssGXFR@)H7$BG;rYIE-4u08IiQcRx>zK<BWhFc*4IXgWQvOLYv^
zsUe_tX63h^RKDG2RxerR^$7Vp6!fS|^9chS%DWrb${54A62%9h77`(!IvKFansKP9
z7QOB0R5k{2p3B|eX&{zt)oCVsB$_`puo^aG3}+xAQaV82#(hdC8)q?VA1d7GFXDN$
zjavYuJsQc%GDQdO$az1;G!i8IZI6jghDlB|o;H^9oR%*ND>gUiEjJL0cI3>NTH-@G
z^PhJ8Es6Lt-Tzq<#R;r$Q+W9zTu|iATw=AJKbC_>53}XmYur4JPYN*n3^M1t17c~?
z>i&(SvmK=lp~!<vLY=C3`JzrbRz<0lWuOAIqTUqw$!FX{y~**xWUKY9=+bIfI%K27
z1`<J-0PhvE$h9>7xv!9n6Wm{t`zE^*_?YQ@%C;BKtsTe9z|cA%U8=&lU37pHOyrAV
zGi+s|*71e#AUtT+FQKgA)jDg7rgpCS+LL$Z<p}c!dQJs+y7FTFo;bOU?cyh{M!o=u
zxBm3RyXV@MUJtz|?Vx~3zC{;=p4L09rf0vZuIAmavX|b3HCw3$*fCYZ8#{I~yq4JO
zQRgAQ6npx@R;j-xYVGKOzjkaGmM@-_kYAEBc~vQo@Ceov2v?V|;_S`7hi8B>iC1%q
zuN(C8TQ=y*B_5~gAx{Ov{;`%1+_MIhozmCZEG@8LX)@Irn6TQKuv@RKG!TpgOET3O
z81)VWghzWO|L*{Sn)ykNUmt#uVv!0~COCox=$Qvmfa1fyc2>SR`NzxE0LOXf^PhK=
z#W@IMs6F2wL!gewhpRbKrsWJ93+5=?8kqG?1f)Z-!kKZPnqj;W0?2I4F!8pInOYI+
zH@Y$=S2f>~c)_1XaOgonPx^;xMAU*p(Ik<1LfaE&1DNS<=J->ZHbQQOT4aDUmgS~L
zf9opwFwr&V9iQK1mxc#xckDj9>eSnmYju#TEb%InJO?ze_j26lQ*U2Fp>`aM0ZF;y
zaV>PI`e!MizM_Zk07#x`X4tL#Rjy<lQz4LCDyADkr^NwUMFX4pIy}zjkx{UtZ=c68
z4VStd?&61?>dZ65YQv9F->FF(wfd=PO7LIbE6kqltJcHGf>Fl7Fs78d?)Tznga2xa
zgm}&C!4pboIkm@b>a+OTT7<~;asVp+2VY%lLhM0p?yCqa0L4!Y<t6|cdU^u0Eb`6#
z(0sZusGe8lnQR({rm<!@CT3Plyk}=S!KpLyEM3<EJWS72YIN3C>JJC((8jI~?@Ft!
zq>g1XT$k75rui$kw&kR)Kz*_WQRSy*`Ku}!N==H=VA6!8=3=P1YM)j(!~*Ft;1JzW
z+2+>6)<fj?c*H%<tOlHooU_`o%?oG&%mj(teeJ{wM|`SSzLM@^l3SM8lM$V>jEPLl
z^Y8K)EVVo5h$CoHkD^-toLl!Oy7vG%O;HtV&0Ex_k+Vn6D3{wVjq%j!%5UZ#HadSf
z_qivOBMQlYoim&o%HKzaXP@h!&yjnm@^3Aab$u2D5G`G)$qj}^r-by2&^E8uj+Xa}
z2BHj4e;qCKhjpl+LLJ^NsISU}xv_1%e5)qmuf4(xWV?TPxpDu#MyNBaHMzAuvm^n^
z4}$rLrquUR1h=WwJA(U+#-sXOS8b#imKn9i!`K6x5B!rW`D<^VKiJ913Yxf(X4Mor
z<Vu?a0b?%0IE;~FVJrCUl6}&$Bx8)XsJ0ouKgTwqVY<M`$@yWr9LqK>IRXZdz5D-B
zJQKC(A9)b@JpUe~4e*-66l2Xa!~nTP2Tp4E&GDEOL!jjN!o&)7RG98GEbhv8_$;Wy
zPv*IslF=n+6l;x=K7qhrYw!k-b;@u}rD8@f0!ReKL+CgWmpZ#!f%&9H#bF9H1;xL5
zd<mdYp;dzVO!lrgexZIpDrj>9;4ra~3u3j@C6d&txHcjZPOukFY|YnL(Y{1ZEywsQ
z4L?Qeh2VwZ5mEYO&KuKvw#u-yz#Up$pp-QA9GdZcbJB-?_0RJaU@p<+C1~s2UY2M8
zjwcO4tM)ncZZ@UC{dGq~HHW&iOdy&qe75DtSZv$ve?xYkN#`O$&X~Jqo>j8bl9C8T
z${%BkA82xz4h|hkfp8z0T4eG!sa&!5-P)|q%iE@T>mwK=K_Edeh#MA9uX#xYWUkqy
zp1OR}-1@AC&m;eI@zLpX=;Dh=_k)B30DqtU70Dlrfg_B`rbRq;&QO!MQW?Q+PjIik
z9Lgix49Z@#{P&cSnzoFSl>k%^a3zB!OJjmZqs_nU7wHv1Sst2U^DwK`Ba_G08|tlx
zpp-N=d#r4fHQ14&pzEh$R55VwWj#_($b)622ogd!!@&esOIxp{E61sv|2xB#dZz+{
z4``!QtP{{3|NXYWfkr{Uudqna6?ayiuPwTUHAVk!VN>nK#XLd00pk1eOoBF^%fS%{
zJTS6rmvHbO7E}@~SQ3wLr^rk}{fpe~%`2zna_b)c2iW3o20XMU(=KNri&#-oLSvHr
z8Seeb{~qF|d0G{&fPHlDF#e&uTNcYQ0U$Vhy3o(UZz71l?9!pV8wEEKH3}<KpSwaC
z9R8`NS-Xe{92)Uer!Okwkzj$D4Z~EZ+BTIlj|q-#tNp_uVM`&5mq4$6IyP0tOYSm`
zOQ+p*gz?7bA|iMR<(Z<-gfrWJYWslqLV*G)=HLpAgF!*)E0N^ofP*2i*fb@_4*)@F
z5$jV`4|il53XMUaNu}9maa(_iXIo`c{m)ewFQs?Y8kJqa+TSlhkTCH&;=vtn-YDCR
z)=$mg%d`;u9m~^fz#=>)){%mbySsjKsCWdMWZ9H-dN9D`M03GHp4A>$t<HYxNh+0N
z@|ZW$sKiU3oXLE1b>XW#(-ma|IA&aFh_?ch%HqmB#fTopXgr}ta;?g^GgVt@%^8<0
zL9agj3&n2C@e+^GoXZut=Pssys8cK|b4{w6dvHsAHU@pl>Vg3A(~bc4m(RZSp^slc
zV+35RzqpOHrbQT9kH;?%K7>l^;JBM({~westQN0CPh*(f-Y}ZJA80&O0NmT!JQWo<
zV9(vFt?Od?iYVyEEr<iThCCZ*M@-D{F=f)Y(!+kRLLceggLEp=z1^+zBX;Gcs;k}H
zk&1Qsm=wW>#`c-E;J_Qjm8gh-@y+d}!bo9I$uIJpeJ+k~#gh69t!zrI2~tW)c#45B
zb+4AkXUl}nnYM}C*)4bsQ0n651X#Pkj9b8dCwUmLnMxiMt7Lcy(klrbu*IEi)ppy5
z8pc|2Ci5zNh+IL{1}oDL0)KOk5BzGmqpFt=duS{nz?D)!-=n>os&xne8Y-QRxuxMR
zm8Q@}OUKQa((Vuv>G}u#0`7H*9&+G<VS5-dgbTVgzC?Tg^<O3cy<=9AKRRi-rrK+5
zY%6`}U&GfAbqWc0ua2ek*X8+YwQN;)%i$wuh;-)IEiDru$dLKnMM{ne)X~ppiB;+%
zWodEPKNN#xrOXa(u39;lT0iF6-$))FHOt+?S^I0u5bGmNGoD7+a5eB!;hkzJABubP
z<~uJjEpRFr=~?{HeQq-c^sYWA8fVdWo>=f#^!LFY#8cZfK*Wg*g)`skDjMs~WTZdl
zLw~Oemc8btJcwB6s|hV3%(?QW68#3jf}GjLtZmmy75EK80H{l)Zx4N7=D2x=n_ZPd
z9oGlqik|<V2N4x4{zST@h&qW|X3G2-p0l6hYgJ&bX*5jObE~J8XNmsM`uE^Zlr?t=
zbN(g#+$OcZMGER~UPm>(vLWf!KybE(2R`&UQ`;a@HeY1Gog-eUjFrIi{4W?v4x(%6
zu<E!Kt}xqmN&p8DoAlBHcLTm=rIxL<|D|HN$0bT^Za$LIZN@;OB5YhtiC}`VpPVzd
zC22}BnUSQ8%2H*xe~`%}NlW%<hWJrjD6YVqzfR93UA>o+cVOZiPlI`*&=Fx+yaDl|
z_6dkABAM72ftzk0`cs@!0OZrmFhdRBls|p3Fr`CLA|P*EY>6L!(uqlyPkTd(nuv~e
z`WW=&?+yJQCsTl^vl{6v9ue7sK<oVu6vxnd&g2D&%pvb~em4dsg@3HoAgsJfjEnJM
ze*_^$)?9%f-5Qx<gV<R_F$f(AYv6b2H!Y9+vPbFl^0?YbRPmC~bD-b$Tqq$C7l)BJ
z#J(>6Hh{Y(^6tC};rF`_RZxNK9R+kg)_jNm{f;p@8VKQvE8+&Cx{J%Ncetgd(R5cR
zgrm$I#WIa)Wz@^ddh)yqaZK=(qo_6jPYk62$+&P>eQ4dGEa-Us#BH!UqsNs8GSoq<
zPXr=TFEPm<;-fa?-{Me0pxzLN-!trwu3I1LxqvvHAu?O#{cKZYJ}=kxx{EqL?^TQ8
z4w}ofr0Z)Eu9VCUBv0l?g#Dp($uL7x;#cR^Y>-v9<kbcBvRF5N=fRn};E{1r&M~2U
zGi6L0+azkzzP814Zozq8#5%g9)C)^Zv~fJC8-Fj)bqm_C@<L_3_iMZ<Ex%qzzZ#$=
zD}aHBQVUIA0BQT7Ui$w~_Lf0)E=;>-a7!RK!QI{6f_rdx4-nkp!GZ;McXxMpcXxMp
zXUKl{ciuBqXQpQAsanNfSkzi~Uww7=XeSAIExuRt$KQJ6h&bQ0YWIWyatx2Np@vhi
z#Bj=OAD#*e*Hz9<&@=9OL&fLidOuP_8+^HVw15r_w5PJgFa)d1zr`7jGtW7Ns}4s~
zSEsBA*Q8EsKGicoGBAecOry(6M(z2iB3Wk+63bbOLS`5ENPXg-!;3Z3gF`vfFYpNX
zbGNu7xKBNJU(=*2^ox<3=p8uIvl!~W{;8c~&AaP&$yWOX^CF!RhPzKoFcoku>7h6?
zeYAZV(~mP1r(MMK1yoTM^H|CQY!R9CS#&31@CMoQ_$4?dngfFw0fG~F#fiTW$robF
zBSJ=20e1jCB=vXSop<Iv)<>UjkA#MA$LUphQL%*5j!AM=-9>G)^3hU!)YW4{oe7yS
zn<Hv^x!WU->P*tL0@*#!CkPM~v1Y2F7YR$QgkGHdul!N*2&PQnW)Lz7&5L&-i)Zd{
z=O^t52uB9$PVZGzHW3ca2t~uy-{$(^-@cu7&JY8mpsp4<RG_QTd0djT$OHS=Z>d#=
z@fUF@LZ*6ztJsIZrWOTss`ZoYZ_BKe(Wwvfd^9=V@^1UFi9%O)Pm5It2qgs#92g(x
z`J|;(L_HJ6q|bOI5o_O|>MWS&m9FkS=KSLhGtzUphfnzYjSn~<X=h2#g{FbmdzA7V
zac1CFoX@!(vWTSqSz~1uw**^UuMf!0#;O+8lOKF=8%bRW%NPRpH)H2((4`BB{G1;L
zfD0B}p(T?PCapzxCL54TS&U12^t+g(Nl)Fy+VqzjD$%13mVC^SC=&eK3)~$nGAdKW
zwXpXV;wQxaAtAOyt|dUJvOY2FGFK0nE&y(6-sZ^-nzrd5u7n|mPJY#6?XTfH=`+QC
zLxvMOvuN)>(VHD|=DU51YK8JPhf;tau?}y;d!1#_nfyv;I(Opn_vwV=ZB5+Yo*#=V
z>Yiz4)mwl3VnF)wOB4?on!tk>2eGXfnYk~RGnS>NO}T!acbF))XW{^9h=dfMmk{Vl
zhZQVD8GYNdF@vqt{&^+M_HkS4YDi3;l0}jwuXxMhr&iNP{42`m&6s0SZp&)csVCRb
zMt2HB`7%t3qxo!zZ)0ql#3ukN;f!~3ec2IH3g_0onsJR5Yjv8jKkBf5MEyc~3_BA}
zc|>IIJp@54Yhi&FbP@Ik=+3KD@dwhIHNSV^@aQ?(`XHPvC|BA&o1-;1PF2e*?UXyI
zgt#N*Ah78|2e~5`frf#tj*Jn0Yt?On`4a*|ct2LDq#6_icaUt(SB=PMU{Oel>t<73
z8Is(86fH3~sMzLKd=0Af&ODQcws8ybHG{HIKK7m9<B;E(1D_hN$#)jpUK61ECAdvb
z+D_3o8$R$xqXv7(>|qzx8D+xMifFYWvcw}AQ2pbi8|}EM)8gcZ>SCo=q)_mbVCrw7
z(8|CxQ)`dCL1E{7#A$6EiV!6oB1Qdk3BfR#m@&#y-;Okofh5jcBzhKJOwn%i<*Vea
z)R{+So|#BD1=W5_geY1aMGAma$M2Ps6FJk5@%r4vw|6;p$(dU%p%z@sN)-07`^;!p
zZI<nc*RI+pVg|~FovE+TWI0W3Kb=>tuEd=s5SP&SMB@iTUiurpYp7X{EK8;~2M-nC
z%G%~p6B;hVF)7;r*^`+RPb>D$CJ)h74NvKHyKf9_KST$HJkYYA^bHuJe(j;px0B`_
z$0Xm0jC~F}ND*!dc;rbJeZMmVMA8)Bxmov!zMx-@Jgw*I)2M+#5XetKN}On|uS=|g
z^oT*xFA)jXO^Gt^w-??%fZUot4Z!W&TKS2FzA%!eU~YqvDu7)u!Mb}AoLksgIw@9l
zlBA(dGfG=-@$FIDodU8LE$XP9ZDS0_Wbc&2p{}iINz;|j3uf=0+@^H!=sDkHnJu-2
zQem2qem%)|22q)PDpfc2m&O#Cwx_UZF_gYO?QI-elt_wcJ6g{C&Y|f}i6&P_9z%6N
z^hfdVT#QtrSdlu`&(^@^gVa_-Wg6i$H=Wp#(R7lg**#RuGyya~@6(X-sm3RK#VWTR
zR7SKNqv10>kC34A8)46`{{#{<+HA5fXKt%tsRec9DKHyCIH}0Jqdv5SRw<YU4p~TU
zM>*_{2Cb_lib<pRs?R!l6^?YjJ2CCGsy^pqp3>c?rp<`x^vsoS)z*l{IfiapeGii;
zM8zg(``BGu5tPCVYj0p#jY|BXba*<fTYlkmTIFpzT3yYWd*#PmvxoG>GF!lMrNb?N
zG#{N((g0=oy0_hMvkA$zcx`i_hxHp5cUS;B-X@gnRjK9GwmMuA?D-e>wc6}#md4vb
z8trv3!8KpA(Np^T5z!=s`Oct{aGjkSZu`%)IoV{u<fY(j7(9HABz~-UA<|(tOj1H{
znh43@hYB@*W(^hB7*4vMe2`9lz+P!}Z@32;o?TmpM6RH-3er%m`5keQCgPC&bZ*?b
zLr9Y4%j2s<*-p8ajORQsFIzBNxfxi}Kqp68N2f$#*`YKVO!-MA7K+xp7`Dv(1OcIp
zG9L~Yswl=}g#QJ+$LiRutab!)-9p=d_aMR*$uIF^N1{&pv-?2N+F*-TL~II?r$(fI
z+lIaW@x2cts<C48%8*28AnOy&7D$*{3`1JS7^o2VS44iHPlzCYe`uM@^bYg_v;2f8
z;0ggv2+C^+&1$U+Miwb5Yt|4KT20}pIu8(M%C~K4F<rXP6#dKajtx6Mw<3NTfuRW`
zMXy2+{XXOeMru@sfegzqeS0AnIC$t2wlg9(M?JHT)6!&~Qdh=)Og2z?XJ;$mM@8V)
zcL4T?cgk@n%_wZZoy0d8(OrKe`}NEZDQOC_3kTW>e<dj4$Jeu*!$mzOb`M|=$1z}f
zoOl68hVKt!A95fiK~c!`iA50Qp@@gMkA$gm0yRA^HAmU-G>_{gq+<elQf>7eg%k8e
zB+FQhk0c(iGGfs*PYw!&{EGUyp`!k0$(EGCxKK%~Nu2eN=9to~JatDa%H<)uACW9H
z0qx@-tzT@cx6L!EpGJvvTPT3`EW0`djb_sj#{EKK#!I*F#lt|kZlo7QJfF8<Rmk&~
zgb2R>ow%unxM}ix-?ER}u51vIA*6&qc~j`jiVM12*7LnM*u8GDl<+N%H#U}<&pkf+
zeKHHcQvIcpYzgJ=USdte(>I<?w?}LkN&>+elk8kdmd0DAX&2R?uYjwinc>H(zxMV5
z0yE!!&MNmMlQyl+_jAm9Y>bkmkZ_*iX9b~`noE1*2(NiZWMnI6ye8R-zW5Gz5Og4)
zwv`TKixK|kcl!tR`8GH5H7V+u!|Y|5_Vq(;1y2OeNI3-Y)pgF+=Cjfn9O8oC*-LG1
z3a<FCOYzMY_(eG>mLs-lN4Ze&baQW&-vc3PCg#o>dpee{AcT+;dMB(nicKT`@-=ub
zyMOtbF2XI?ZKWa(L(yKOC6IwAawXP%qvFCXf`&sw`_nX-M$)qj&1Ln?zfxxH-u&^Z
zaTY`nios<`YYD&}up8(g>Arrrsz5Z}TuDVP{9NdQNBqH0k!1|8&3Jp8UX@LmDQk^)
zKW899eu?K@;Wt<(0#AYKaC~s%doaRy%3B)mTgfBEzqT{>sNvb70WfZmhbEB_p8G#z
zA<x-H>>A0&m@LMiVkMKg+~|;B#8onoMG8F}7g=qSJ%PB0#pZ$YM^J{6C_S|uhpAG5
z<4_L66GbjGx8+mF+PkgeW#st@-6YRoemz^$$~=3uR5-f*5oVqALVL?eBcyHiN_f{Z
z$A6T$-0UdN6Q8|90T@bF7X}h&VCI^OF>Ao4*`{2@M)i>AaL&2lihcZpUIr^d?ZbUs
z=Aqw3SWG@K{>5*GQ#}O=BirsLioZ5nWxZr>L$=0iPUUYb*TRStCow8tx)b^Q3G$)1
z=VS8{3HJUlTrYZLI^p-Y<eB5O^_7ze#7g-dGSx|WVqGU>KLEy$LQt>j(g`aE=HsD&
zHPiR;>lZf9I>`IOMt0WVXP(_UWGa*C5w?ehn>~qO2#?OA;$Np{iWW!^xGA9T9MpP;
z{lJC#{SQ8E{TV*xFx44+_c+ruhM9Ac=uKA=gre=R<l5^EY%7OUN9;-;)tW{Seg&Kh
zx%{HapOOt$pcJ7cNT|;e9e#MjN$c>s(@PEveJU*ukCtRUh!rwT6A2<`5bYGo?#c&s
znO+uFEdCFyi>cb|D!fU_Mfahm{bo5xCQ&8*;|+FSMhO2mZ${T&Mm>;(#*#Q3v#~c(
zD#~j7GU%|j(&YI07}ew&!uPjXC1!nWNEpsblKry6z*VZ<$dmRYD8htk#w36CWk+sp
zT5+PBWzuC1WmR2Xzy74yS(|0EV)?a`#ALred$bhQ6IFzD3SGE9X~v-<YaC_<;|<>M
z%V(~LO1CsyM~v3-py-9z9=B08P?Ag1lai0te0@t#%bc#0nFh&YBzc5x7rk6bP9+{1
z>|G=~!0Mkj$n@vd#IER&2|kU6odc7p`aCiS94&3f*l+<P7zIP?VuT}x7zufarzE^G
zs9$G?Y=yb<CAJlDOuNS)hutl*52AAs?o1U;$>`2Z!8GMk#R>V$c}6le3PW=B%RW`+
z#e*01)hF1o)S5BARb+Ct^fIFo4|6QoXtb+*z#Ktmz+rZxG}H3)KH6K2@+M0P#$Lpn
zN2|L#V(_jk7K!eOZDqYyc0<Q@dI7=Z384#jwDd5{hT~YG<?P@#NtV$zWRg*Gccew(
zPz1TYsPK;tY!>3H$NLMcHKW4jJ#yoe>3*7f$)J4RA$qrjb*H88jtj0rP0zNa4Q$AP
zz?UF`Fm>^ENfbObXQN(a9YZXlXSLdhyd#W3Vj&`J?na+@G7ej6_nA;(UnN^6n5_tl
zN-)|I#83sX_Lj~nBR8wcaRsuQZO;soM~H01Q<3etk?3E9di>xv6gO8!2%<21yoUdv
zy(Yu2jArF(`(aX;VmF&eqX>|^8596dOa{jGJi>_I)Jg&d&MjB~W$mo%>giDFY(d@8
zX+n&K{;>z+rPQ#;eu1l#f_}Y&;%>A`JOp#JC06o{^~~UTJ31Q_;uE^F$Y+^S%xTc*
z<%37BzJ<Q9Wh<!xyuX!0&UikNMcZ2gOMyXHj_T|zVRHK+xz1DZ^vMf99#G<-Y|ip?
zMbDcHj~br30*nM6-mtjgAQsOn;EVJ@RVu=s#k60Y_rKxTPAfAAUKTHGmTZ={=2<BB
zM|+VXKP}^SmCR?R{HeB~u0}4l<RJ04b1rOEbm$c^*wslD!&LZgT^QbZs*Etu*J+NS
zM9@s<aswEA6e__?8Nil0#!U{dwVFfi!@7%e_&F-;Sqt3)ak_hDMnq8~`URI(E8C06
z(ahr^b>E^}X7sW<)OWm7*wlI`6c)~|V}rQI9(HwW<RZy*)IqxT1V?mHzm~>^`f9L8
zwua+d@zG`Y`bMzY3_7U{e_um&5*{@P2R#MMx@EN^y?|BkHjtmF%Ah@P?_3EtiCYxE
zn*%7s>Q7==eOn;zo?(yE5W<5{XU)VqYqB-)ebHm0OfaYuIRcwFf>GDgORg%K<Xc!B
zWbFZYXVDwIMJQ3jXvxbqaIYxRvC9oy8;Cm^*QA&r(TlVwHFcVpOPrgdkeh+I!bV{U
zP56wJrJ?N{3-lhaE|`}2x4bZhpk5nFCE&kDJic_l#kDJ!X^BtXB!LX4*RW(dkmx>l
z-d!7h>?tXY8y3_Ev}^sr0g8aIFSRy?hu2B)DFEL-7RET&(FJC(f(0IdJVT!`TQj65
zh)RiSOO3g8RX(-qx<H&_n^zj2K>n`b^Q5ZU<qTTdGSD52aQ{&&STXjWlYIILf4?^!
z63&B?sXyYd3?Y5UFDK+&1uEF$_xOIQHfv40zw7)Sg3bT(4aVh$Oz9nY7@Ym+d)gT9
zbm_Af7lkM7gyLJ8-K^YNvqy7~C0o&KQcrRN@%CNM#RF8W;AWU80Qr`~CENMfKrg!5
zs9^6%4m3<V*4p<@3w(?}+Kr^vBAtARGu-3PFaY8!aYBb$898;{*0|sqqovzVvvA>F
z!xn~^kV*=JF(qSIS6uAiiC5Id(%j;xhcgMu)Z*~eBkf32>W*I^EIY30|7=*jpUf8b
z8S(Owr(^6X)@i)J54FUU5;9CLD55%IyKw=OK&g7`(ncU3xN|e-lr>~yMX0FBzSnUQ
zk=RfUo@qo3$Q!@B@%#RG!YzS+pN)%Yd;Oq#eZ4G0zqPm{bx@LH3~S5dHB^uF#31|v
z+=(=l+pM_NWbp-ARvG%~p$z;1B&cgL(b)4cQknlf43@=Lmx+UpwSk47`fgT7zm8a`
zfEn{!b>>5dA<GR<<=<RjD-cM+f`lqIV!KD*6z-uWE&m}6{{amnebh9<Hvf0fV8O7%
zrj@q!U(gVgfD<q#MK34=^8p%myXVDf_9G}c#mp#|zfV^fVeMa$U;KQT`Kc^hek8Lt
z+7^#z1%d%m8_#Y9LQW(w>uta7?xHB`!B4{)b#^vS;~F>Ln~F>mY7|}905C?dGsXK_
zf$#vR0)209X~W81Jn>1=X`{@j;gPS>O4glx?#{-TbVmz`V%$p7(GN=ShOlDFrfiz(
zt+Fo}=oU*wgg*mWdpuaBZhsk0?#LeZ+Xq`-L#g(MX!Nh$zJBDp8v$RhO?!NS)h)a6
zj-8Zu+n;RTi;hsF>l|x#H^S@|SK%PFM!|vV|CDs74jL$9;BqsGKV}yHSSJ*lhs?s<
znt_eT;Lk0OO5`qn_NjeGh5UIhk5M9!O;SGjX1jhUZh0eY@k*&J>E@jI^RtuReUZF5
zgj;60C`eDGcs4(qy*vK)&nP(x9q$J_+PD9+8v*<yN9e=iM+x`$=OVB*45n>=cPd0_
zI!@xMfu+aZ?#G<J47C9@IV|5gjZgOA&f*t}!9uNE3x-5gn{lP}Xb=!em}3&rrjjAk
zAA;EzEcE(-p%V~DR|ibZyG;8>HjP2;ZSZZ2u8Bk|9E!)jAtsW2zCDRLm93GGV1@Gl
zkO6DfU*TmzzWsY{H{PVyU;}xfM`}q3+I8*$-uu-Jtnq)VhuW~cI9%zgzaL;36ju%8
zM#y1OPaXeGKqoe>GP<we?IXnnBGav>d#+UOL@;A%iYN~|dsVtScnD`oQfp-!yZysA
zY$%=MdEHA06&;+#DCBbhzZ$V!<DhsZLez$u>`<a#A=x<ci_YR%yRrLOgg@2cf%L(r
z26k4~C9bleTGcXW6>b)%?9=b%jD0+pd33wi2nb!<BJx4$!YC8swCVQsJQw6qH{=ED
z{Mf-_?g8f8isnL0%NTMI?66WV{hB16yGXDW)#-A4)mP_#>lWZYg$oRu<qli;P_sh7
zUR{ye{ys8j4vv*>tCnRQ&c$VASLkN{3-?o&F8NPtiV0Bb5m&_KCfiZ*tzc;vdz(tu
zlPD^PI@w=-{&MKD%L0%??-H2H!L|kogj@Nc`AojxEo(cZY%K_>pzVECigdK>I7GLu
zPCPS*gjs6w6Yd~Tz;O7#Zh*Ob+g$ChqBSab(Nx~1)0^kiGnU=^4Rn~>h?&>Y>E`39
z>roi<&A^;n%R}#*pE_q<z6DXXbqvEUMfXmxPOti>c&#Egvw|APTBxoXZC<9J)ndpV
z)J%2KJ0w3$6>}lfvB@S|G2m8_7K)zvJ*bsLLyBCMfJuG7NbV{6S0&u=$lp2@6#uWK
z%lCW?!GGO=P<puivlKBSvC(IIL@~THYr<sNYl^wkdckaUG#L#<)V=oj-yiRH5$z_s
zj&?!4;DwUmge};`+3V<Nn8@_s^e+67-a+|W;=|=Z&;ZRG(qK}tRs++15=DEAJw|?v
z?a6-=dl0q=Ujzp+`o8SV)&%S4(e>}}JzsmQbr@QDv^~h*WO^sEa0cgbgp6EzrJ>)r
znL;D|^Ie4KO4RsC>RWbJWy>d$yTLUFEnBu)W1&HtBLXNs5Mw0jO5UwETjW4Pt3}xP
z8Gu_#<-v==pT5XZ*!7fvmLij0lhwg7vuOd&xx0w|#v|ZLixMq6s^@+X^dj;%NXL_g
z{5U^Nq_&0Wf4m|$1}$o<s+^H-B{=4by9+{X{_J`$&#D5bI!K*JY_*2D`b0S+J1XCM
zt!rrOYyY%vXtdmZ7W;c}7K2=Q<skt0?c+h*7LsXpHKGp@HMR%*w>N$v_O~}~&~?Nk
z|8H*`zVwyd5;m0S&DLSFOPx+hLZ7;kb4+Dx$BX8_3Axt5qbaWPuF#Uagw^Y%aH$*J
zR*S{{rzuu<<8QuFqu&9@-Plu_m&5(&nW}F!KqRYi{GZm-rWFX>ccCwrAm}jhi5BW%
z)5sG%qq5P;sodIxN+dR5qgA7uVs(l}Vw&7!`ywEAGZ6#(>)8Th&D0Ln&nrJ{o*H&N
zDK6I)pIs7oM$X-Y7`QBW2v<EC>j)BHdi;BmMN4zmbG5WqHeyjVJTj*{PEBS7lZp(4
z0F!hix-~ul7EvXpsZe1h`o=_X8?idoAD_8?ZdtU^{i?87Ow%~Io`HF7z&lX$>@*cZ
zNNIRJ*+~||sHmEBJuoIp5!e05$#I$w-+s_`aYTPr`nVM%yzvc2L5*an69aDvr$D=`
zzPthJ4e1>OC7$+@pFje7BC9(^LY4t`AE;SAD3hV8ET{5!O}W>WS8QhZ+5Fa#g2J$$
znZes+WI^f)LLtGDeHCEUpAsvn|3{62-U-Bh5Gi9@N4~^uHjJ{$iJWj_nJeiU%ppP%
zt;bP!zIc01fL80FvnuGcv2fq<JLg-h5!4h)=O8|(t&}0uV!2a!nE!6)E){>47$8<N
zAf`MH?=)rN@f_gay8o@9E@;>SU6&%AsQZC4cC_3>gVoYDpOf?tM|Qg8_?2eLXDEMp
zDz>pPiP$7b6H*KUE}XM=ODx=g(@f{+rJKi^kvdje7nCRIyIlPPx&&%5k)1CefyD{q
zi@d`!|M){Q_dgv0_0JXwXT?8(^xa7j{R=yaZFvOzHl<6jkEM)}*$@&A0wuwk(==mD
zikV~n_;Wu;VTmC6hqEaU!OSaZrvA19l>es{unO(@_q&~Wbm7#2g#_uUQb8j_DoRjL
z5sT%x&K+mQlcB9>cKist_p`>ds<xINU-GS^Uo%?(K6iBX7~tI9OVZhL0sq7$qgalU
z3WjYWsRYH8y3z1p(Tb8DX{U(8Z|#Ra@kJx^(^s!&(E*PoSV(ptBn1E&GfBIyiIM5t
z&O(0FsSrF}{Ty(ETmQnF0fV3@+TXKYfVVJ+7o56msP>{w6B`!aJIq7pRQ{7PoPm+l
z9?rr~dU)|U3h-#wIZLgvUvkIfNY*=AH%!U|4UMn~o%moPM6FNz85r8*X5lC@l=9qE
zO2UyeiEwxM2Pyg?jY@4U_Aa=Zu^k8sf~<m!efd1H+;(laM<N)UQ<um#@Y)qB%HK7i
zuLSHY2%gpICaWzPtG5%#y!*M}J?+S%8!du~kIzyV0OeLuyWfGht$|I1Q7oem{{3xA
za!`J2Rn5i&M)6LF+;v2k3opeMwdP$)QVQbRw%aMCy#^PEV;pz&EZafq?}SIHGt#K4
z)8QsbmiQ-jCGEf8aCT>cnuSNKlE!F~U7^-47TWcu`D;c4tPfoNMA^HNu<(!nKIDk!
zefe|@7XQ;31P-A6q+mNCm@0Lwmb&oSaZdBc!A?B-{AtbCOT}-v6evmx1b9qO-ft)F
zj2Mda1&LDLh9lnNX`?l9slnJ>iNok0^R?pkEGxyeDCER)VieuaVIpA}5o@UeHk8EH
z$zf`ek1A?q%Eg+(1jP_?8NE}ZRI+-N-<tiuSvSQyLgWDD6*j+RiDhQYxDR@Q6?eaw
zcFWHa<1r~Y+_j;bDEOKR$JIaNnPa}fA}OkCN2m!AHi~`RrIl=PLsiSK=Y+Oxj;X>e
zOe>R~tM%;)CY~%9j9cw~!bo<0`7g*f{tw7Ed_)krUKNNGnAa9bl9<UA#FZzd1QY;?
zy1)WWULy{j33HgsXRxP4Og7ey`<2S<vR1IRcBSOYxUBIs)@PU3$w%XfTn%>x;wzW7
z1tgnQ1tk-B#z4!y+1n9j5zMr-?AcUxwfipIV5oHpylsf&ByJ5zW;9F#NslS<uit%g
zXA|Lt$!6r|c{e-k1S8M4L6@RL0j%VjW-n|k;oH+hzeU<2V`k{Pu05KKOo66n3VV3Y
z<e4Tcxl`lZDGNxw{&(2!FgFTtuCKIYVX0{!do5*T&9mwBk5w+!kJ^(qt+*mSIt(dh
zo;<J{7xGi&6JI>SJ8`^2%Ebk9vDq8IdeVt5+ndhoAXA&I@U(8vtbQDO0{CBkar$<K
zGM{{DMOXL~NliMV`DE>zeDXWvSE2#rGnl3s_rGEE1$zN%BXOL1)piqiYnY-DXMc@S
zUdx<}v71E(NAv5z`_RsNZI@am)gD}6*cPE7`830>frEryKN^n+$<AtApgkFZ6VVq)
z`SN%C6cs!;vF^H1Y2eR`l4s8Ownq8$v+{Iqwe5p2LhUGhyqmQqT8jS9e6@&Wn&4Xa
zt+-e>7XB4o%{)pS_I)Ke|0qBl47mKn*EMbj4Xk1HSQfef9;=7i?U*CHg`wb^8x;AV
zBW2Gi43@2dkPor5A-xPXe$+3F_t(aTf`JtmyMyN=AJ=S*fHm%?7M#y2v97f{JMu1$
z*z<Lik4EKD+4<b$8xTHrPi}?$Z}H>V2qiEn^}XSJ&dfm8-T5r!w-(8;iwnI$o1$I#
zn>Ih=(-My%98h<|P^LEPf27#jjjWe|Sb)=?{+l#LHxivmmC^Q8SY}rX!<l?t@CX}$
zh^7lIWfhROdvTvw68^`a8@MnuDz(qmZzR<Gi{~A_mwlw(Bg_AUUZQp%C4RET7Z}4_
z#xpiu!8h{%#N8MW2=}ax9$-XiY2O9}Lb)emx1m#12A3W07EGb7DB2H6^%w9k>K%<a
zmwvLpvZBm#{NgU?Rnz^Yq`r7`K^{=IDY@{|*`tQd{1{dRO`0+|KtJf2j{PZvq+&5N
z{q%xFH;z~Fv#Gx)`oDR1h&48gP=V2(g3xoxnUhJZ<7aaa3|q^(oL}*amfrt^5o<Ji
zF6q?}(ybm#t#iEOG~Y?(`jL8p|6}Zh!TtXkdk;>(TcNs$qhKnTd874<+tc+4)`L6Z
z@X6sx&LU=;ui=#7hcin7vHb>yo9Gr(6fz4RM}{nDT<ZWW%@*}GcPj_42?41z$M`!u
zs4Hwn>%j}|Hf{?)HFl?cA<FOOYfwS_PZxgC!ujKDg4sU`S*T1th0>e;?Z;sTu-xp(
ze5&hSf9|B~o?x5x%zJv@P{=7`lKaTSG-o}_qL%qm<0N{8Gt{#ZxQz(9UVLzfl=uH+
zWb!s3@CD^t)yfOz@m!UbXpaP9x~{Nf<fmhpF7PgtO!`VIaeV2&526?{GBG!;JaTWC
zId-m`Wz2c_9?H_bsTBv#%gAO{9$<>)zlVnot`Du>P>!LkU;csx<m;Ov_Uz96lafd8
z-DoJT1GQDFy(Fw$2Ak8hGuZ-Q8lr$@=Uz7|L~uuf^i@tpy#DPG(-|tkW^5HW-}-X>
zU+I4~e5w`TmRURMvFc*3Ev7v~jWisSb|?18tukY{3}13KJ{RQ0r(9R>1k<9go?Y)$
zG3hJz%l4w1Sc{4&pn4v3nscT^v!cIU!Ml>9MzihFL+(0y*-9RR{epci8iEDT-971B
zV{G=lE`wE@rHf<_ILy^kj+sA$B7Q;~$bqi7zM;ARI)g0}s7SgbhwCX3dc4zUN^qc|
zphZc4cY}6*`gj&lLgJZW{WWeUXlIXKW+NUGP)&jzg+oC9&FNv_#s-DSPr`Emb#?Ly
zF$!vhMVV*x<TG|^FrX#gq7uz3hxDs!><GmBUyId$&5?5Jf1*08H|_`VuH+XL{hQRo
z!t5A{w>K7jqO%#8o^jvBKa#rDZU7FXg!d%Gc#5Pr=Z)^yn5T1oF%ijE4xO(&h_$QA
z9f!YTK77_`xQxm$1p{yAe4UoiMjUUy^jSbcsj@HC?-?35-|s^~GC<Ar`axFXTRQK*
zCSINeMCVuMh<Nu_Ott>JcQ)88TWXEvDQHmvJ7GXwvOdkZ2rkHf7NEeZUh)6w)?dn_
zoKTa{Z$7o?j9cmCA<HHAzjSI1n~CEW)K#ray4GeoCmxd=yY^{I@e(^F0QL~2l2cQ?
z48n!e4|M9&morc{*y->WuqTI8+xd(j74^g~(k)pt(dQFAaQZn8nQ%D>c*Pme&K{t6
z&bawmp$V<TOTv5vvy?}6(#we0A0v~SH9d%q(dlH^{ghWw=sdN*n`Yocv=y{&P_lI)
zrvk9Z4fpL2H@{J?Iqw-20Ll)N5;V0})X7!)2G<+(P=4GBf-+uu|MCR_UYGyyg)-7~
za0})`KK8Wbtz*JSnlGqG-2RGj*I*4e!fq?yqUujoT9z<khVw&(LO~RlT48YI7I9N3
z-GBMK0C#QvTsA~%&*k<Bh6pD}ArH)H^PWWfch>sbYCwnhS>nj<Kvrcf5iuc5xW`9b
z*uF~RjVR_-Nj#HPh1HJ+s`xtNq@vUx%y*9#hqj3ar>_&%g(HEW%{z)OkTZQAZhMWd
z-PMgCz1{+b-ptmerO!Tg*O<MK#3ffQo#N&gL9Nc9yaMt2`EH&WYgdFRN4zNDC?G5f
zkrper3Fts3x=6wLEccra-{m+hM-^Fy=$JAPG=JCQ(@k{IyO4)OcTHf~JDV&0r6Q5Y
zTwSHOiKOw{Gg#@phASLf|6LmgUVmy&>CW~Xw5^f_Gx+aPXf_Vwu@b}QkC8Qgr12hN
z^EJV~h3>?iCW|HerF;0(E`m9we%x=?$cTO;=YaJ{<$c07&cPie_!@J{dzQ1|T0Zo1
z#mQI3aUkmL>#}^Y9g%kS+$qbZ|9Z}x*GoR%S2r+Ma@*_nK!23xQ)Ju8#%!uJF%SFM
zN9m&}mHj;)k!~EMOSgF9h7jI%Xcs*A4*#KJ>r~Q#tsV*;?Y&Lf7%#$O!^0Ur;zQk>
z9>AC*fPg)S7}|g1B7PU<mxM?uMGET!SN3T2Oj>EXYo#U%5l42Yodmg?J3pGsCkPX!
z(&}igF6p_!*pS-jx=ea+0-)$F;-Tp#gY|EvoA_H+Sq2O{?2Db)IJEoIqd0kt%v>0p
zVLV*xs6?+3_ZvD(MxP}Er3Ynl98*~g(4&$(pQA(;j{YYrG-;B08GCIwuIO_Rt)n7}
zQ9`zHkmlLfvTgw}Exa~6MKQekm39ek2L=Db?Z^4EdIRs<K>PdNA%Tj*jRi0It%BGa
z0no>SL=|O=Tk$%EDm*1Pv!bv_DM99~-e{_a+nIA}tn4BzQ(VgEWbv;7hJSGUfic?+
z8G_E^FL5->On*GmIP>uCeQ<ZlLe9(xS9o+1-H~Cy1Op|LPbEri8XKSoYuFo@a&Cs9
zgzO}eqv@ONe3CstWj3OaNpBX6dNDTHvJu=!sqPxwDFXAkQDq^Sv2DKdd(H36DEwf8
z;Z8-_MTK~-LiXQNFa^2U{*($&1vrGY*?+~;BMUbeJ)_a6HDS%W&ifsUZ{zeUk9szj
z)tSj#L<Xg})?86xSbv3BnVd3Vy>$x?lU6;rL{FVV<~_mO-!d9HyrR93*B4Ey8to>z
zsJ&m4i`-!w&W`iOEbC~bYd4Y>ePqxbn~y1=An;9k=Y8?rAIi4-Xf`4L4v41|WKhoA
z^j8zS4$3SxY%KO0t!uYwrm*U%6Y5`lC)HjT_pmj(XeM?Dnh0EFMH*PAxh!j+)_jJZ
zr*sEg9Q<A@N6Ysm^lVIa1-no&Nd;$uCg?%54ZTtXehD>g_9p$l$FcTYq81;;bg~^?
zmC;7&c>YX%>7!Amd^Bao44}~<1*=v_uHL6lsxp0^9z5pMv#=ov(IQF3q69m+Z=dvW
z)gMqP)WVBJ{L)&PYoKDRS#nucoZy@<&lVAO8AxiO^oKApSBCQKC+W{cU&UHsiBleB
zL-qL^oNPpbm5NAlF(v7(K$-l!%Me_K$5&6|YDDtj(pEW<dOQ;}J3yPp(0Q26;}l~j
z51%a-d`oax+Mx;TP~vFRPO;qyGTrJNvcTD5kpiv0-*k?E2bJwj^N?0d_V>}5fGZ`<
zRcy!$gbJA*KMeY%8#a2xt1D8C(U+_=^B>{p+5<vhEuVgwUKDMYSS<_@J#AvcxZxEp
zk<b1djSykXwYf=Z<^^6Vdc|-)+OrPJsisE>kBMb(EZu60l-zo~Y57<8G?S2dn<|zM
z*Zo7YbZ6g|uHtBD)^dsNQHejz80#kW43a!bjuh!yQ-K{uU3j_A^F>6_>_svrt5%E8
z^(gc^4H11gMz|yD@Bf;QIr<0tDUrQ?qrfZ4@j)+I<PNaZZU(lVDKAhjKD7AyS`+4E
z^CH4!XOs@QKg#6|%Z?jz2?w}0N|ia)%=(h3LeFQ{4*4YTId?640xj5LJofYX``+vv
zdDJfX@7^Q4a9xeG7qKthR{{elWjHe_ykG0huM|O!qzx{L@>=?T5UJ-z4x$FOzS0pp
z{(?#ml(DTrQ~(r;iOJI2TF3EV9oaTtG7Wu75j7UOp8`!&CQTke*rcfegI|6Uc|5&P
zl!?0e+tNpIN{j?Zix4{xE&3hL$|50q^f%b+2>aLECGF*qop*D;cLrDm@w&{xJ!u@u
zA=6E-JK?G5+hb&EV{r$<DNSsd3!eH43RW9iVq9xn_JArUH#^0vln#q;Vl2ANvPEXG
zRXfoj-*WFu3AE^~T8sy<ei!x={6Z;rcS#K3QaG^83slzedpVexi06*}%GhJjJyN8Y
z3}X^;BIQ`cv_~+RJ8g?9#K7zI+WKY5^4f@33RWnZXs4|}r`@WEj-=eq-48_SRpNZ6
zd~^a<zY<VFe)PVhHz(ztLHTGDm|9>WvqlkU<AJQ)I!hamq&~_yN0~e8T?kHXI!K)v
zHEL6^f?Q{$WHj@>(ED(XHF;byNe_p8QRvKOQ=}FiBQ}%5w(BPje^8^1h&NfFToCEx
zxg854$1BY*;Y#dglv>6as+>#hM5ydqJgT+}O9PUnZwKz?xvQ=YDE2l@b$6q!c0shm
zHjKV4_Qqi4Iq$ph7RA15uc@4LK3|)lt>112_-0+asrQ--e-_WqnS8dJ51=B=w1be}
zL#1JQrZiiTLyQP~6`uvvWjn9NgWRyWuxf)P#Mct-`Vx}*TK&n&uB#nMDXjY#8oF6M
z>i{@fvK+}vhys_jfQQjzFQ{~4dlxqkk7q8Nxv8esw$t0m^YolFA%v>K)w*+a`DL8l
zDvP=w*H#;AWZL({m9IJncCts)XmK*=ld?wX!fS3s@UE$R6N__gP}auqV!|6@==>x$
zO(vWAo?1V@FS0d%@-`T1sdB<0f=g2^@&ImbqZb|ZzD&TZ9D{dok1{OCBa05Da5$mQ
zI#H#pM`zP*S?zk|h_yU$a&a$T-}l-Uu7njy@xpfPrINRSJ_>6klMW&n9dhL3xH8GF
zBnuFaemFpG3RrEB#3PNyHus>m2JC{vS>2k|yN$-op;Fol-Y<_NGGC6${F!+=kOB8p
zhLN$J;%+2NsHB2v(bZ*;Bk?odpqV`P_Ck3tAZGGgtd7Yt*#>lHEy`WqY@n^NR6){<
zR&VPd+sKNrBS9xDQB0R2{7Q(N&xrU4n*I1H6^4BmLLU}nsym<<E4qhkf#(zz=2@nM
z$srmkMEl1oFs3tps|g81I2zv-A6RYFX+mz{4Qp*<>xSbw?+@Ux=f(9Fo0l-k?5UyC
zJ>$+ff4b;~?i)!P+j~||GGF}3g7`8~)!JNct5zB;9JZn7X0<|yw85KpJ`s=zpVPu@
zlaIZC8=)7~kn{=d6PE1E$Z+ZSe)}Yj_dFzrt;fA5$61G8BdeRLljjCYCBVaUTVgPN
zZ^?#~EC1bnb<`CBOs=J|^Sx-0=j{^GLNoK2ZnZ*4147rSmq096!k6Nub<T3BUWb_*
z_($SawEvK}ch~6Kqz$h%*5jm;R|><%n09clupC;(n7X|)6(#dSMJhGz<`ZH3+Yz2V
zmhvm;du$^?cwRZ@Ap`;G%!V5K-a2|Z1988}>}5SF6KR+8igbNq%=rinBxt9Knq!^D
z;|26n_Go>Y<vK?3FDB_CHHOsCdx8gUn>enKR!At&l6xtmb1Jl;&zoTMTbHP-b2k#W
zi=sZ{Z6f8i%w-+613B%F-E+bIM$u^8JQ!jrm`TQ7zOic;X2}4J6LQYAyt6vELdN3M
zE#X5>RLj`*I&?c~hz-HO>;?{xdi_hd)W~Rx%^pwJ+C{jgHBo^azOmTJN)MckyW8Zs
zYGHx`CL%x9*T&3^fk6zSKj6NmIWqJv_TTF#Uf~FjEG-g{K~Y~NBWYeRwNSSv*E2W8
z(r=@jd5f$TtI7af<uFS={Tm@F^h3~rq&`}DPKX$m&~wnIWbQ=IoWYkxA#~dtR@Y(u
zRl}}Tdvv*4v&H+V=EHpg9KgxSLxk;-Mn_d&M|J5MPSR5+&@%VJ_|Y-t)VOx8j8QFA
z?%+LlWKzDu${<?7^4H2c92IjVqiftR>0577G;@+C_G|$CUNBb>?lx_#MARktG@e3)
z`JLXKbW+#GbiVi-XoFrD^(_$C6X&O1nP9@lG=Og_g%-J{pMt1{YcSpOaVCq*Ha}HB
zVbPO&T*|ZXMSgYwLEs^G^QC}y`!i;7lK&uEDAPcd^mg(!mmAU@*;Bm0?xrvi9_tS+
z=(MqdiE)5=QNNw&CEhdbN57bY!wT!}IcHMQxN0*g@KzRfBH9BQ*qb_M;p0hxgs9rC
zN&NmSe4&|^)Cc`4Pgo!CUKWq20l5#i7;o8ph^&l<YuOG(7`}78r{2Lf&jv6(k0{*Z
z(q;0VBe;@hHoU-)t^Lj)KpMHMwj3)EVQpi;#sj2Wa*%BFN`%T6W0+9B&%+?g&vA|?
z?65s~_az!f&|e36Lh5ZtN?65~Qn<txT0bf8|3tTF+?X_aQ~a$TRFQtYh!>YQx0`Zq
z?=t46-#Aeqa~d8VvQ#Z!C`STJ<ue^<g`Ca=y?BXM(GW%FZj6i_#H{HRvo@O?x*S}u
zYXg$?VfJ4Z>LsJyxQ8;v^DZaM(oH<c7!^#hhAaeSJCOrNZV$N*CD%;yEvOiW3O(_g
zM99W>-5HzJ3T7#D6|xEgduVV&a5>CVw2D-*YC{m~Xy0ny)`RvzU+pdfrRdn3@PEId
zG>n7|m!vTGCM(x}({<EU9{+mT7r{#eO9rfAT7Kektax|Wds_Co|FQ9UYJp1(vexf)
z|DMvwKaKB#E8Y*sxx=~*sdb{Yx+=CRu+GJa)Z<2$voE=cR?iZQ8kwCv>3fxqE2Ely
zRi#xLFZm)TUVOKxykuN81+w<7KePoU+bIzjA>#T^MlQePWydSj5_Pyw43nSIb0pv{
zy@&TLCiPd)V0NU$;n41nz5U$dxs$a#>9rdY!w9xlkHcOVL@kG+LRgKc8%GQ;wxO2E
zH0Fk5hkL~r;X+4>-!q)B!S5F@6~$Zo0iB1uH!TgiJESUCG90YZ*+WkUe&ii4QPOD>
z!C+c$0!PQB5BdSmKRmORWmW5gY-bD(o@L|K8?^kFMr|7pxwt!@VUf}A<s%xr@<MbJ
z-mc5O%c?z7*Lk9r=e#9@q=!$C*S)mv*4u&PoNVoZu1+{D7+gcNQNrv_kp%MXZvigz
zA1ZtOY8#VGqXS!n|9ooahI)Kw1PBTs@ACW3*U%{tlxW^r_h@U*`7>a^h!ZDLEOqkx
zu}aYr?#CCIaGyFY49qD^?(4Ff0O#vRS9Cm_x&`OH#YD{$$U8NXW+U()YDJQkY0OI6
z`0rGaK=GqUnvTjAvsT5!vuh*w6zX@P6V;*2aBMD5p(F`WB1J8mZ){cqa0l@KH)P|s
z+Dx3t9}9m<78d*zCag&$iyTr1?kh#%INQ$ryv{ynd4`b}qKh0pnNal+0vbQg%PzP~
zcrwEr>KU~#^9KkL*?lPX<>#{3wz$7G#XDh&Qrljn|A$bA_E>zwa$(V+*jw~9!}eQ-
z5A@Y}@+CM{4r&7ms~FV<=?@@6Sxi{{<D#XqR}HABKj;)}N(6`*1iS$z$ujD7Ah>me
zf*5z)$4mN)fDDK&Y!D%mEo_8>SBhZZ1kQJvO_Ko%ei?6~?;soyG6P;Ipy&*tQKNx^
zq~VPEL9f)+>9JDqM`-NbXttltB18umh~Fm(`a6n!sC|8Q;F#nhC;<5gJhgegoW};4
z>&kQ^vAfHJ6W=ysjM~azju1gkX5{+Zz13Nzd^c1&rtf>4+6e@9qcGp-W&NrPaQA|&
zA=P!=*~7Eki{K{uM>`~9<*NxVY2!?3K40E1yO3RjVPcU=Lsj^KPC&1Z;}&3D8KVyn
z&0hc6K)1>1=%fQ}KLGG=_~SpFY>>f9E_BMKzjQ*PifSJ+6bv{(9Ofc9t|4`WcoNWC
zG}%BA^1Rd5n4&wDKWswpM}R@u9O1Wovk=YuZCpE9X44OnW)y1sKBa}nTL7(yVkbi(
zCnEq%pM>-EqYMb~Z}lqHS!k}vZ*^l?qgVC5d^V4a?f3amO8_)JStI>U5s?kEnHc#>
zMKs!C>>m;aiNi6OA3-ix^K2M>^H{|bf-0PGQ4_LATmFNxSH0*xz{gG}*{g%nZbZXL
zz06CoTq03bfK5+fvpL5_L+x~X>IN5cYG0>BJzc1b&IZg^yy@~8n|xR$CME`5Vj6$<
zPE3VR7o$M{zZp(@MK2Hjd`F*#6vvegLCjl>Yb?{WiE8kn2pAjx5gEUKzli>@ehPV{
zSn#K|F5Ua8Evh0ov?7iIw;Kg1jWtc`3)01sm44TkbFgF>Wqe6yo3}%Z!ujUYOc`{<
z(Jk@Oh)*Xcw>5j+wET*Fdb^(#+76aj82I=k|8AZD^8T)#P(eBGB>Nw|<aF)h(=0@t
zo!HDn#})taA1}V-!;6=ywwry<f6Mo5yF&P1eu=XNZTN>@!vBw7BL3l*u$PL#S=k(0
zrv8a6(V&JX=;@bt=uu_~umD>%AYT8O@J1OfGSDTv5W~XU8Arkg=>BQ30{T_S&fUV@
z1s+oE&u{rY{Yu*WeFAH!lI3DQP-ZOvZ>WOb47wQW9O!!z+|P~?eOSTl^8&l0vZ8|D
z9)+z*(2kV%(x0Sr$v(#`2=&9|DT-EZ6NPMVa<=ie>h6wmo&*VQv>jrV-$!KZA**_%
zPu5-ar2Cl?S3GJ9fT<D%vaH;0`m;EuLi9I&vM_#sMDPUpzW(tkJnDa2%;)UOB!^^P
zisrO+s%F_c@p-7+0_IPzaiw|+-97Fn=j*TDw&%3i`#TU<A6h!-HZ@!KTx)aNS{4zZ
z7Gk*W?`A^YIb=T=KQlbNDZ}A2L<Ah<{THU7iDc5a_DVKLATTjB-7<pd<Ef?(LHmID
zlycj<Bk>#}nd(?6)_mC|9{$E?xS7xBAEx>=%KtLewe_97_e1X=5`!~^*=d`PI4R3j
zR#Lt2^7CxUH^t#6;i%B$2HUig*8e&$z8}s@St}#(7wGF5B`niw!Y^K+ww1>Jb_5QY
z;oCx|B<R~VEVO&_=6px}X*Ls16V5OO>q5rrNeY8&l_`pT$RI{wUSpJzldEElCW-8Q
zhYdPDa7VC$rJ47>GGJ<}+{!<h6}2b;&PHfB@8j$yy&V!-82&J@Q@gQKi+gN1fVG|Y
zgJaoQIX@+w+<A-8i)06dP~>_gu>mB>k5D{APbN>>=VLZ=HBLS)#6mfhD?BrfmD>B|
zguT9^F=nYGuY1kbDbb=XefhoQI5}lho@#*0Xkf)yu)w*D1ljE||M^rfVq+Bhj0?%f
z=80ae;oZJ~xcES`Tc>>1Syj9s&`TYRJ`~BIZ8|sz$?<D?X6ZEb60Gq;CU#klYm1G)
z(7u@!xrg65Wf>4Cg@B=N+DNpc5Pz=w#=m)i1}|>5Nxu(1`4&7{2Wp;)45_AC8~<_C
zkpxJR+N~x#m&#kp3d<-Uv4XX=P;q?qlO8jI@W;2*$|LDbIDyJ#&vSvE+H(=*lEbJk
z5jpzQnz%ND5n4k;bys$FAVy|yH~)9cq6aMe0v2Zc7@i>CVHqbdl@`Z<1b(8HIO->f
z{omp-ZYF$^5BjGH$x8*8hz1F&1%@RmkeB8=xp_GhAMertr=*+aTxD-Qa1r)}sTfuf
z585OGT->xyH(ZcM+x&_UMUZY?u-i~y%-4U=L0+J=Po^`TuIS>Hsz&;%+dA(lg|>#>
zD&@wcX1<#BW)VEOsHzsT=|^d-Iim1%B!Av8#Bc75k|T<4_|vJ^u2jF?l_oe2pwoGC
z_lcW&96T-KjLEYi&_$(Lu_(2V@@IdXQ$V&yi|Di1Jkl~gF<lX8qmLDx9v{Y?FUK;%
z8B)>5FI4^bEE+OLS`tCA1{R`U&sZ@G*W7YGul;gzCD?;f&a+-0^cD6l9LCNop2&F?
zaqiT|%>7<!OG}dU(^+AU76Wq!@LFIE&@531Ov=bdXUf%7hktc_L<@FA>YjMZp-G36
zIo1Rxx8DSBq|-6Igv-i^{ZzB_V?&Xw^XwLv)ekR1p~9_I%_b(;hdc(kn0+p&N(^5K
zqY&9j52onU26^*^QmBA-p-JlUpd;eK?+EmlgLgF^TrWiPkEEh4pPQ-|X0l5*m@mBI
z1$1j-1*-(i_gwX>U<%lb$R<G}oC@JV+g}rcxFj-n1IXFduEGjLY9e%VH{$0KL5Kjw
zi_;cOd&`EPgA5-!sn##3?uC(bZN-hB)|@w$`qgg0<V$^cr<oV<#L#|N0J_>{{cx-S
z<uV|6)Se`uNnp&P2ZrG6#e6v>PA;jd(Cx83feWJ~BykyC!XHv7kyI9xqpqY%$j9oH
zI5i>|Ro{jZ?_a(ht^%JLK)zB67C-|Rpe=)P%MmbQ4SeHtq`{fFpTCL?deK2re3RSE
z=o0viw{uCrh6W-Kfdy8LU8;76ll7^was~`x5ZQ~3+gT4%;)DF8pz3QU??O5y1e`+B
z=$$q5y2~LRjCcf^O^8^4>vcI56T5gL=S-!jG-g~2vf>#dZ@CdTqI^Znh!Ft#x}Xe<
zr?qA88?rrQX-ZjsPmBEdl`G6`;af$m(fb$h|G7q82Lt<cfzt@_ClqN27cW8>m%npS
z>iA?Q%sSe^mc-yR$C6x1cObop$DMFq>3h9@9Yc!RgQ4HK2?C0|ZRbSE<^p}dp0!`P
zrza9@F7i|HKBQyg#BzzEW*@K({CE{+j{c5^Ncw9?#%Q`r4U2DCBVe=UXy;P@N6inW
zze6Byi}JAqlf7ZVS)c7@s3V7Qx*XV{3-RjdU56rhtr~$?=cByO)BV@w`BJ!WC89aa
zK}o!5G#)DwPx#*MGb2<~bx*|07>nUWznPW##YhP-FSiC!TN1wbRRI=Y5^NLPill7O
z`0#^hM{8O-)p>Gxn{ifu^)|H+y^Z0u0sV%YPtRVK>-*wpg6IEN;$I~Z<I8Y0r+wnX
zdy*&wZohyarBio;8|h-P^<Q-P8K`Lehn;Lcvbqt_<Fh-$E{G-Y)YdANYQnx)`gnhx
zu0Vqa#JC}zXKcB2Q6*!qjKua>`ht<qUdZ=tz6*OG5MdHn64+dU)p3-mlYxvD=!U@c
zp(HbYDcnwpL#vVrLiRs^AVh#Snj*I*&A7_XVwLgNsLaPD!h$igK`8=$5k6EYQHjo8
zBZ65r*;ttBzrv&v9rscm7$~G|SkTI*Jd0Ws*-J*vX8yt*_?XW@C)Uqg6M|ng5YB6~
z8l`U|KpJZyZ5gCt1}XdSjthfJ2qX#emcPHk5a)ubB$i3vXxj}KLmky9d^s0Ot!PUs
z*}mUh5eq!i0l6@P)W!i!RQlmlxn3Qt8I7AF$-JI}Ubbu_*uWy!y~(ZfhJ869+2A&I
zSC`sY8GUPcV0U0PJm(y3zLn6ZYRL8LT7s-yd{CkfNYvQp!88|kg!-2f@zm_`eCEI0
zgM3fWV0%RWRW@-OZ>5@53R=;;^)*L6w-e<}`<V4~nGQ3J4;bgKmDbdM{ZgLg{0n29
zwCb_+KcoJl`-w<)$Az*kY2V;K=c;pBGMnzlRF8$}x?8{NwL<hLVHG{8OHLtCW1;W4
zO~k3O1}r!|cwA_*ReKO$?Jm^NGxmvz%;_*K_5^ar6!Mn0ILMUus1D`MX(8%rM4TMN
z53CNA{a=*5V~}KxwzXTfyKLLG%`V%vjV#-?U0t@#F59-PuB!g(-Fu&V&xw2Di@32O
zaz*~h$c)If)_mreV@y~hfY_ne7H_xmuu-E%jLBq>`xaQep6v+pDYNu_I*1tE@~rEh
z3y9`%8b@=<xjLKqmS#xM2xXClt|R<>+0<nQM)~`DKgIHt^%PDI0n!ssjupZ+QLx2(
z&Wct~E}7JySnkhgUMKtB1`_&FIawY=p$%UeMBB{_zOlA&T3qbZfHb3usY5mSDzv>s
z0w)GH1bF2Ln)k18R}-xDALRW_#1|cVTR(w$Quuk6Iy*WOI)c8i(yLzHH;sxkklXV-
zdZ$8lTdJO%RJ;cj8Mzd9v#6vY?aV(ATDvtcwSC$9ODo&N-I-`omcdUH)hCN$`Ncyc
zPK&YOf8dBkySf$21M1|<^oWV#f^nO-#p<-rTD18r9ab)SuuNBUldRHEc+`O5)|jxE
z{V;tj^Q;ar0$%UEIyLDzvixkWd_(G+y{W>hg^Gs1Jpi67|9Aj2|M37weR6<azndQ!
z$vke+dGTNfDpA?de~(**O^fUARwe$Ve~aA$y*rWoV7LKvR12N7Y)GVaZ{U2-a!C$e
zOdB}R!FvzBMcrWsvA%QGG&u|spFKEkvsj=OvehJK{fB(hMmY4U)GmZO(GgxN7EPzW
zOIl>oQsmiooOP3wU-4!C!QU^|ow1{Y8cgi=GTFET4Qj-EoYQIT5Zw2^W0Fp9^gb=U
zKE1EY;o211WA(3AX_ym@ZmDTYtzg0qCy&AF#O?AGntf!_ua?NBP9$jW&8gCYqPQJ1
z+-DOMcNT(lQcYF(P9`Z~w9os&T3o#8xdxv7(SDQ@oNRi>zO?OpU|c|wZ2O`qZv`{3
z3RXGuPtyY4`_wLQ3#M74{9(Z;GM!ZEf6m=k{XdF^=j*SoLhvd%5xM~jshRHT>N(eF
z>4&W>EH6G6l#NrB8J%?F&2r#7g<y7K?57`?Hw8E(J_6qP4dCL`!%n{aW=>q;3q*SM
z5F5yj1)8?OszOPzfQpo!i$O@V0BWOYDw{-?4|9xAng38AJ5&bCF1<<Alu)T+=)J~I
z6S?2}Cl$ZqV2V=AzR0wwN{lLKb=0^D7Yk1eO5pgseyP*9`h~Kn2c}45*wBOa+fIB!
z$fgRw?G!j^CCSV5f%`ZGMEuRHAlEYjU!G4mZtHK%2JjLRJcE93CS)3g6tva=)DbkY
zXfOS0Y-BPlru)ZKVA+Cqd=NAM*Skx*J~01p^dr}^Hmjjg&nB%gn3zinUS@wvdjH_}
zh2)9e<wZL;s=hMM@2VxGHN;zH9yU%F9#|Mjs`$<v+>n!P+xyHU)){e!ZV>%(lJ}iE
z2<YDUckp2Dzk&z#{){7b0x4pJrI-o)gMVZKH_!i=3(o#A7oZJ$-%1Jdk-7pd9_>0^
zJoNv6!XKl+tzXfYOre98Cm+YSBg)n{`T?|QwVnRMeiD#}JO3N|{eRK#N<vPkflhw1
zNTxkB-NBhdsu;y&kB@e*lb`o4!iIm6T%&d8H+I>PRpTbmGL#X#yKkFe%qIlDGX+rA
z8sIBPzi(B;L`}Kt<`qD@f<+14tQN&{_x$Tw2m#$0Bi8vI<SlAUA;2*;OS)*GwKeAv
z`6$u@jAj$<u0EGgmQa=mCg~p-BmX{18H*#JQSi~qBMP<gmK9-^v`W4{qZ^*FauV2u
zUl_)4=_g*o{>wJ}FBnVN{)RCkJ~Oh;b3*D`fcoa*SDf-f+sZ9z*l7qFY`KX(!_-Zg
z`stJqNR7#gUPkY~90)S@;}wv1lq}d!4YTxOY#6tnD!&@a$hF-NFG&8!1N(6VYSdx5
zqs|h;r*T|{Gn1#3N4$Y<bh!v0ShXq7!Is_-J3tsFL1I0%U+$@Z0idR>@aGVo&aYBv
zh2XQ<VxA9sn##>dYj(QZRl#K*gq$E{*kdR8)Us2f_MuhaT=R``Fivo^Uzc-wyomQe
zc`X+~Y2_2@8?&A^HHB{S-6=|v-ao1RNz_roGL3<y_jS7(bxzzYJR{Q80Q&aIjynmI
zNPj~NW$Tl7UmJeb0rVfjc@xML=#FPN!DFe+bK!lQSk25<sFyLJ;#lPHR2DLUiePiO
z(h{5O=V+4aZTI4nQ|le&k*c4!Y-sYC84Tk?R5Kv4FF4uI!`<(PH`3RLBt43>eub-E
zlWsQdDoVHgQ*H`z`4<IHULY!jWzHu!ay<@m&C4C|{4s6>xa}hlrN7p>W~2@=tszP~
z%03K*O%39?<ulr8#w=d;Wat%WSf?0`dj^AXD7KD5RIYY*6b;hMPU!;tGf_|wc`Hdt
zDgXTK`r(bzfehI_O~BLvW=M89_+nwGBcaqE>WF7(#frIt{4W8v)5B!v3QiKQI2yT4
z9&Z+#*$V(OsZja#L#YAJH(NrWF;CaAgd_-1Oc?V0joR)3-{*jVHd8busGC3>C8!E1
zBF=WuM{o1nvu8Vhbapgk&+AC0_wPoZWUz;`sWMR+QfXvRws6<|ktzNd6%dJ_JkfJl
zS=n-ZlIe^hQ`i@|LMg#odzE|yNPEW-qr(^tKu~WlpC9QEjr||!6xQ3*b+}1Y-$m;c
zgDD_$>ue2bJs9WsTiSFP_)$hzupuLi%g1`8MSS~w7Q!(|(?V0Crjl|z*XW)N=_1rX
ziOiivBLST#W+=@=8nvIo!0>=8*8tVAz|I2lxImsrcv^OZ(13(<Z6&wY)W1V%Jfj5g
zhM--HIx55*r!=0XD!9A@);t78&7NwSmcNGL(4<gjUY1`E9RLgc_M?dG!_9e)EYGrs
zvcTs3f}6tsXG%xHQsDE06jhqSTDf`cNZ7tn$q_5=uNXWgSaW&@XNuhK44*@m>ZMNd
zYHrUi4OH93=<m?RTfKh_DK7Ozg#gMyTsE0Qr!WUR1mW{hy69Y|;yCSVeX`sRs=>#*
zkBj`f|BUH)x$^%N?RHGNrcD07kLfsM(`Jno<3zD2j!%(>i!=;R!x>2_o!BI<{1e+~
z{!eV9tiUxm--H6@)&fAb!W$0ypWMb^q%`)86s-d%N906HM+bAsGmI0_<2w0`kf3zW
z<KT#2C`Sg-x4EYr)WgyhBY5w*WHB!8!<yfH%zWJZ*v2M~`~D$ix<9hQBpW{CTA|Ou
zTs4!MKjDU`WZQWgRAMUaPacCd(`r3gQ@$I{+&dC$K2lgp)&NN5+lSLv>xnu>xxvXp
zCvV(>{@<2ZhP7IUc5+~hy`OM=4B8U<yco>d+6qOiIOf;4s~qKS&^h{GIfFbT@#-*<
z73=5vjc=Y!I|Qmn3C5PDR;;G??+4-9yw4N3&X7oGc>a`gNYKN7zusJKuwRiHB`sNf
zHo>)QRsASsqX7xlkK%_jms&409EX=={hNkviR?aZ4ytXSXQOi)uC(RK+lmA@W@;sL
z%nCBVVw0M`lq%f5o8PE|?_yIyTyoFi>*_0SckGSV#HG8aWlC-@xlUlw@$}-1|CHS>
z6u@r}zw%}k0DqzBh}GZ2;4#Tyi`Jx{ZVZztQK`xN0T^FFY_NmkU-*3wLG{*@Irl8w
zVu-~A#d(e+8@^{R2J%D@W57;5gkfYbrV;422q_TCAKuj)P4>MO9$&nmOYDvj&FI||
zC*6iVS!2hoJ#<;@?aFv_#WbHf+$HzCEV!7`%`WQ}sK4nWYW`K5+wu)bBaBv-dIp<*
z4ZZ;iQ&QII323rJE@eLWS9|jH$F<TW7Y!FiNNl4^lM#%gO-#CF$yeD`!8HL>CO=5-
zlS|*b?iC-v<1ypf4?AVReBcW4)5<X}0>A5BJq`r|<N8%vH4?1$eO(TrU+P(=EvUFj
zqpZ|c^~V$6Qf|58e@VG5@%-OXE&x08c<M(0v!+%rUiVp|hP}#$<E@{D_C)6!jLtPF
z_aD0lGpkI_F&;lDq5e#@y6_tz{uf-dhtp1*^AM`HEOsTX1CH-2S7^rC2l4XMt={<u
z!w^#`Jn;UmCkly;@i0Q9l3L0xuBSw=`h@Twm2k=4(H01&^-$v;h}5&d8-S5=_9Tx5
zWt1v$@FHlA5`0EkZ~FNl$=lU-ovrbJSq>7#>RVKI@^*v7&$Z4tpndJi&kB<Y7CW);
zSYl?a=CbK@A}dRH^0Qj6bz|o<cz$Ynr!$B=9zFawe?JWGE%}#9L@nMe&fgZ5SnOT{
zrQB+08SpILIb6@Tj6OpRe1L>AE<{n=bPXo*ADRS)_K@V1{*T~W*Ac3|H=5S@AkUKQ
zkugSMs957v&)4)DOI}=+BqxOi_g6u*N&y?1>xodz^X5NfbCgk>sN))_XxKeb6riSK
zT@1d?t3b0-(1sZ$CKuRNtr5v{dZapa;dPgt!aJTl4!c@uh`BZ4eE~)15wAVdC&%}c
z%99SAg_$;Me6p9V5fXVvkpIM8#B|T~jP=l$$k6ES5Km02d;6gY&|u#o=hi(7>yd5;
zMU$Hhk7J`D>nb!jRrVye6h$GP0o4(A9zoI5ENQ8XYKs|)Ewfklm6GDrg}jY1U?CQs
zl@F}IKdwyH)r$?_0epu*h~F*a8m`WnBO61?b2J;!?3Zn?@m<;%k=}3!(6R=;M`Q@5
z-|L$Ad_!3e0vO}fzAcY#Y`zhKBI%=B^G=&E&lfCeW~`52f6J{2&E2)}7997lv(PE1
zMy$n%&SSFxUh}&(E8zN6Ta>>gmFaM(4w+0-CfvKYQaLF^sDbw3?g*qVU5;_F_S)dR
zuM0Ut>*-+NiLwLLcSBx11F&7gu+*l!fwwf%h^oJJK&rY%(XQNnHv}t=-$!N(ahhGV
zjAo7u;Eb@GV2}<_{{yplY{sgAuF1`R52v11F>g<U?=lWefp-+#IP8Wo+^R}ZruiU#
z0skAc{_mVsimPL}7b>RY4k&%U(2V!Sb%hv5_#dzZL#;L;2KWtew(y-H9zi0hLYwc*
z!8uK>NwkBu^|#2VhxKpDhxWAkYVDE_o&HST@x7!vTufK1Fd~e3-oxC2zFcQyb{zUg
zR#XRZue5VJgH;L)gK|@_)}O>bR}?Qu)l<bzqkqEyoRl0&$^5jfkh|ud%uCyTc0{L7
ztmm$NuGR4v;bca~<=G4!Bju8Ss)oNFqobvQGy#_23oxZ@j-_Y|gUJ1rfvVtD69isg
zN_c@4M`@tEgM5ASfhBU$CA#NUklwJ*>~hcSB<AVEB^4^~JfCTrTW<uHAgkNS8+wt<
z8v0)VO`e!#cYbR#`vus}5E+^rGw;dTFFEb$bmP%*F9BL)MHdw0X+U<hndEnP;XsgE
zr7n>l(r+%ifzFnUei;odm0OfPneg1)MNjBDZD9EU?)6<#(h(`n-CR%ZiQnDX@LMQh
zX6G%7U7a20F17onW~V^W7-q-Ab}h_HR=O4d30Gc&Y&CN5*mr(g*fU`cUmRqD)=9bp
z`gdu~0y{#R-O@6k5kwo@k&j?X)`6Es$a63M4d+8U1#Ts(_`Ki`(o3UII@Me-(+OIm
z^-V<pwQ?#8ggq--1IB4gXdOKK7nKCa_lvqi03Z{WvQGf~FY3pF3ACv?7$h6oKpH>;
z63y&#_$r^?Lokiv5Bf07lvR=LK`bcmZ)|$V*BV3u6)@@;t};~0vix4l=g~al22<pC
z52h449E_Dfw4Q*@_-YV7uE1qg>Ns$*<tAFc+==&cMrM#JwvzAAxscz~63JRa%=)Ep
ze0LoTM^5@ZQ2M?1G#Z}C<HyX5SsZ{PKTy6L>i?A~*)w^b2cld&VtjZRhgI=LlaGD>
z4(hY16X*@T2YsgJtW}!j<itRe*gAq<TuC=xF%l4<n!SRXGi?jl#vjfNVH)l?n;myJ
z4NL9Al;Uy=NLWj~7W04aoBREF;|A(Vy*RA3@!gN|)R$VFw$}KK8fORtt$?FY3IXk5
z%fYl4`eGgFfVD$5h9Z{@;1)lBvhyxq5EE_mP+}nk;_>FgIIV+4N^#WS`V+#?;U+gf
zqhKrizoR9Gk~>n-x0<50lZG$Z^L#|l3q?a^j<ay7^91;+_*;=z`yo<`?AtuMEUFH^
zo%)+3vyXOGl;C%?Ey}lvN1kRrQ2Srb_3@731}O=jZ{xa+s`D88ET8y&cj#H|y~Vec
zD_BlbI~ze-ueI0sSJF$jL=WMIe8<as-uD>Ucnc>}9Q_ft(ZJLeX@Yco$V=V-BMAcj
zn<n_*B|*FX|0xN^D6YR?s{A*$Y%#KcI8tv3{-=>fH=X8<2-F1~CfM%&KXLQoyT5Bc
zhn!FoE>Sn2i%jruBTpgY+UpvGCCK*yfDy?|C)`sE$fWF8GQ=zSS-o_<Gec>eSCA1R
zary<GF(@YV_aDnU2oo67rVJdzZPz@GfHG4P)TpUCp6Jn2WY+p$M=R_X?A$}7?m8pj
zl*@~fw}0hSN%o~&uG2iUq6isaIQ<zBGK!#Gr`plMH_a^4Ltigh;7sSzk^n?A!kX*C
zX0i8iXm<u1w-}cilWsWKz$YR9bY)7oYR1201qu1i&9IPw;0Oup(a1P=?wH*#Lvs0?
zdyq_nh4B>q&ak4_Q9C_s%FF3QxHaq2S(rwgvxCOKON$%V)r?-)ko@ud4$jR8eFEZO
z@H%Lt>!ho1YP&1)wZzVEI1FH)l#Mq!P*?R<Wj<wu8i4w}@f{CU_urnE$W1tLe#J#w
zZA(9eaL%F>*^x>d2f{P&@~InAneX<wWu4clptYg-?=&%|{k3hE=$F4eKQ&&<yc5+0
zf-C<8whOM=-(brJ{SCHTs)kdk8;C^;h_jt80CVvgQUo?yjO3wC5A$}&v)OwTrs^-d
z?b4Ad5=Fu<*uiuo%ZIn$!u@_zB=@43yv-UxJqtdmPRDC~{{+{&T`@)n+^)(BI(W3%
zH{*L8OW3=-*h&VY7t!G+&ZCn|Ll@?xXW4?g$dSP{)OC~!8sn9XfZWOVc;K+`fI-^>
zRHD%8Cvt+l(Rrnz-B-d8sVN~HSbsTv?wtWMx@5Z}d^*IsEWA#Ct^Ues6M-td`$t`L
zlzk3T!d>8vv47a@#-ohZRwH5Sv@cbsT-GYlaKb0*@7@FJZ3?9RM`tt;`hRuC1UTr4
zFM*q(++e#Qq1p$DXnG-l2K1j>J}u0Wbd~Qd(MIXm@ZwY>H2*SY;IFy!fGR;!8H432
zLjqUcLME3$B6D(u<f(j86q~TT+XTv~5<Qa^@lKTfT3wy|<+At?M_!E8LjOaZAEg-U
z-vFBV4WJ^KL~NF!^aYhGmPEuo)e1%mHt@yxq<Dm9@jie$qBEpYzo24~e=)8k_c!Ct
ze>1LrGUd-Biu~U3q%(;}tsy}Mh&A9bYC*NVK13Kz@G)v!eYF7f4|M3@)YYqPgzZ?0
zosX>f0l53TmS8&uXvfE~-_{X*8~WD13jB)kTwKIuz|0he)NW(h9|Yd&1KFI3C4Z&a
zbFcgXwvqd{QX$>_L^?z3yWpE!EK6tzz!BASL`;cwzqdC2`aYD9B89<~3Q{JiJA-}i
z-+AkCC%~cuOa-185L19Z?`WY1AVGlH1BQSDXA^$d3c6;Ivt_**(~ze@Nu-C=mk%Ub
zwvT9W(Tj3Meu^4^rSxr*lHkxeo(y}wqAi&K29S*n`Sy9=BOS{M6NsDUkwq~Nux`l@
z=LE0Ep_XT}@M3jBK<p$g`iGV)Z3Fr4@aeY^VO%}}hSU}%w@dUqSiqmZa0B7Qiw{u|
zRWiM>ASgLMb>3Bp(2kBv<@~DL`IDy2LnvF_a{!6#ofVF9>+aGedQ~*4T#2+#BLxWn
zT|Yg&IxC8wakJ6u?me`;(NS;AlWG0<U_`-h4Fr>EhSw%s3v^oqOlrVTeaJPPz#y>>
zF-L(8!Bq~V8~f3t3r^P`g~TnhnHgukBdGG0-V(Eyo2k(@WI%BwMvaiCb0&pKA09Ka
zCV(!rim!UBXU@So0k)F>P01EB?B^hWF@Ln}d%Z&EfdnB{r@{I2U|$pduP}b!uiNb3
zhXMn?GjLvyWG`$B_JI{X<lwoyO`K*|B%5ts>SD+!+Geh$j-vVWNCVwg1daJ$Mbw*{
zw?UJJYr;YLa9K%}Dtyxsyfx3u&;--6=fkF6lw%=Tj)eFp6%<<eP#f^$FMG5AxX+|t
zh1(hEoKQsF7!rOfyQ<!?&%d~nITbSHT&D6xP3f^4_t((4$L|wK1;B<FBW|c0vod~2
zsARET+uL~@Cd;(-!M7A~8)pIX*!^q0$rR36$aJCd?z5;qP!9F{HW}#D;}jUIlA<tW
zqd;=BVR~_q!>PLQf;9=C^&X)C35295p_L=SL2G3<rTpYdZlb5Q0*te&D>f7hD~l4f
zh36U1xC_$cyI#Z&%<nFgBan&Sy>^1KP{9$vp#G^rHa0cs*<>FOqgqi;fozHVln6aO
zb58obUc&uZ)mb{!@T{niPR9G6mae3=f``rRBQKoY7<w{SDOg8wGcaX<Y!4&a877Gb
z8kCd005XHZb|2RIrcsIj9^e2<-37<V0+Zr|Q2K0Fc6g&DdU<;Jwxfsw*nYw~UNbmy
z+mY*v(5e%!aWo<qL?EO1aro8M^LUhn^Ts?S+kLAiILo=*w@G@%<`>bnJb}7NqBF%o
zLv<nz8!d=t_QNEdmUsq$98H3l<5RqXpi0l2*bxR0f%=SCX5oX__aZyh)MvhNy@;ye
z_sSwO2z299v;%_anOFrFHb=@P9Zr#1A}$Oolr#@5J^IXklk`KEk2}a>SszN}?5J+=
z)FHNOsUC+0gN+W1VNKJ_wwM67@tO*QrB&<c7$jNm5+(iNqv|2RL;x*M@dyIk`7C?$
z*uGTfcR9Pr55*lJQ<z{im3T_xWqcY_z|qp{3f5KcsUSul8R8-!c@-03PR49ctN!@G
z<akD$RVP!@<740=Bp!h%)>CLti|ie7ffM$aK|ig%Jr(JeWD6s+ClSu_NbY_vXU75=
zyKU06kJ3H9T4M~j+T51fyDV<0s^H<#Si@=WHi`Wpu1BPTEKwSk`eNhbyKV|^idqRo
z@67%LFHjHClS^hFDKPmcB$}iD1!zS<|7?0yhH3N@=!mK_+!nb$4lu{Z`E05q8_#_5
z&XQgFIP;W$RmnSS>9G)n=`a~kGUb|ABVeaYU;V-L)Orty7V5Vk@^j)9<F|NA33H4%
zz6z~|1jWE?3!gvTCShW<t49ba#p{9;qIQICR$K)s3xXo0<+#dcIEt%c!v)34f1CeA
zHb(RJCwciW1ff-UUo7y!r(*7%oV>x;WfvfJxaUC8H{0)oWSN!?3xpn&bjgU|#~kF!
zzA5u0a`yq~<>r>0oVPVb4LBCm!9g{)Ium-N9YPFze<~ziW`U`m%5MYIF^@(<-S~78
zMu<z3bAm>Ddrb{6cDRt?p1u64mW4v`Jgp``8UK8P*pmH6&@@^;kn2dmDIB3&sd+5?
zD2XTf8C`P>5+>qR<Xf>d)2_=Y9VYnEy_BW;2`AtN458i$#Zm7S-=sgX0k*`=Cl>>1
zj!4mq{^nSn8lmJE^6j@+W0eJ)0zbej*2xSYef%4PPJk=N;BIzgyh)vN$N=H|LU8aX
z;NfFj%+>Ohc-dpk5Jm*I%Oa?lJ)b)$*vuTx)igx8<)^c;!uPcnGxbafcHz+<QRhWf
zi~+E9_088n)$W<h1EX(Ib=74q)j~qq=k1_5Or;)AQ4J1Ml-iU&txAD+W*%d86w^wN
zmS<LM&2Ak#sstUFYgi@eL>8IS5UB;>5c5+DD^duad$3K^)j_RfONfls-Rngy2C;5J
z6C&FyKa`FFf6feSQtPm!QOq&y2|^3ixCdPB=+gujpIF?Om29|$iKu&to*znG4Q5bV
z_1@$jbW6b97;NRkzR=sa)z@^xWa-jc=o&T-aK#5PNu@n6*2|3t6sRGbR|hw=fgQ?H
zV!#(U%@qB>k6UMHmMRieDN=x2a1^7MgBW6^)LDTWpUfIIft?t`v~*j;yd0c~n+8x!
zYsQtXQGhNOO{PRY%ieu>?gih(N4{<5eBT9K>!-WY;B;iXPt@-}=!k{PNL>nSIJHoW
z@5Gx(zj&1Hjo=9rF~&Rj?uDu35PWXeKn7kgHa%lz%-_z}+(ss3h{w{d<9@#S_pQU>
z27|n=?t*@)B{znPryXsla$}#>J^+MfSu3_Uh@*<H0)vC<`$v&Ud&|B+JVV$3#AYym
zoz9;5o;XzOzMk-Y(-_%4Pp$jHHX+L=KxcdB)w*gXto(>&MRs=|Lk303K9)I-dra()
zN+K+1v#B~FHCBmUwi~F}pLH)-5pdYUV2(O=NBR3psa&m6(x-pYZ6ys3?gM~X(IV0~
zF3OyMkWZ2$8M1!wT#O#{BYC59T0ly>!X=YVNtOFDM2Tz);><@4O1r~2R)WMT#H*kR
z#AN?cG#nj0vtut(6TgE_1tnazewKd#tb$b=z64yTCJSf&7qD|rR@*%|cCgt@Oxe60
z)0_XYL=&q%D-#Ibwz00l06ZXiuM9$Jy?5@D<Sy(k3^>!%Y0wAw(a#^%6<y|%a<HW^
zjd!Rk^Zlf~asF(G1{%fdIfqjk&X9CC)#9AlP2tH`dFMw&0adF^k%VP)x{Q*&f58F8
z2y-ZRu`C$+j{Q57`N}kx!D+5!a_`p!o!82Ryu#MCQkDaNSMJ<)-U8tEB~zmz(yUiS
z^}T|g>t|(>z6{5UG>(bZ2}@C(%wOWhV+7fz+N1UnKU;MjCJG_4HOJ^U+uyi~o!oL>
zOJes`TkVC!o@ejRTyBB4yNF|Hnhi(t56w23Kn*dr>y+a%+)h#`>w;{0M@tb-4)|at
z-la8FBv~MHlKWGpR3(7TH@g#|QFihuWOO3ZxU2M{Vgl;`%!{K<0fy^$^a+Iq@RFvg
zgV~mdd#UL0D2lLOuAUa$3F*PbKPJ<tRI8rkyVqU^U#l0b7I9V<5FsvT^ZD3E`8VbJ
zIQmvlyY=skR%LU`J90CkpEN|WCmVL@O=U?Xa`g0sarhyPU?2duE>yq=8+pUV3+7f8
z+=gGF)%;3X-8Wr7t?FiMbuA1(^rK(@enAs>QsAKYkqjK{u_i2dZB)bHF#lQ8ta0<9
z#zpFy8Wgd?@ONOf$A*45B`QT^dpBIpO(%uXrpV>Pb;oetpqdt%=h2~EeMkUN7Kl2D
z9+&&W>54^oS40Njcf<O%8Ljl4D5H82V?A$aMAaCLN?x-39iupZiqyxEs3D9}1|96Q
zBe(u>HZG=_&oY?C;8)JA9KMsvUHyJ@=v+akzO>w%thPWW!}F7xC2@OkOt+$0ek+sH
zCFl1_U;?;({tHpU3v!V7k7zZh;0)Idsh{-WG6!|x$yxaTeG2mCDKlbRBOJP?)@2>i
zFx!_wIq|?+0%JzZAR!c3d86q5G5nx*+ZNRMHyx{ZGm(A+IK+|o!`uzR7TJS*^3d&_
z3$+SGvL&yxjy1}yUkfaR?A^7I2Epxou1d&mZmT4ZvSSJ|s?;#ChZN~@N3BVR90@i1
zv*FE0WS>6)V_2X{!cLW#-5Q-RQ-%u5nJTx>`YqeEvSr~Uy_?BcxK9<>wf&6*;{b~h
zfCnW1pO(Ludl2cC3D}tFmRz&4uz8zmo%Sk%zz-USCVgo<wmAFT@XzQYcOJM3)awsx
zdN7u0c4+oJD*-6!5)UT&ozo^>QP{p#b6<EYe<mCNFpPim1@L649Uv#O80<5-^_iw4
z9Oa{(ucaQ1vVR<8pCUx|RbZYBah_nS^l_d9ehhG$59#nkIMdVN|NV~F1GN)6FOzYK
zR+Miy3{aGhcAA#p@@yvF2*oM6LTtD1T#6c%y}Iwqb8n{PldVB0DU}=E1%^NzA}qq;
zH9*k>P^W$j?ox;pnxJJx1A}6N)4-@jPZ+7FMQ;Um-{C7^-W#d7fw)9Z^rXRHrs&4&
zW2VHUTfZ>YfxD^<v2H(OYO1s5=4CJI2v3vjZ#9}CBRke<Ikeaquo&26xI5g-7Fu~-
zUjNx*iZc#g%b<apNJA?TqMq95!~qw6$RY0p41B3b{>f`?l`d1QO|~P>1q<IdLWB{0
zl=8RG5Rb};K}>Z*JV%N--grK8lYhYD%u|;zDd$do&Vggq&CAOU#-+y!LzJbd_;%_s
zDASBZ<t@g}Jb;8)i|kWm3OT_-GdWVGu+lz|jeD?&@yBmwAEV`(sbluUYfW1OVPU`l
zgfj{v+!6cV4jKn41fkp}LvX7(5fN<?!e1#X#}wN5UC4_ss}Cic(-%%=ZxGra@w$3S
z`OPkKt$fCPLw3V2$eH!+-^KT=7q{^zPQ&+@sIJAO@D9RSziVjJCCbX)!$Hv}rpWU_
zw55NlX(AZt@7et|bYfB`N>T(}$$Y>}81q0ghOmh!zG?6yGx7H$&5Pps@Cl{u=s>50
z7<Y;(Ksy+W-gKjPhLj$E6_ba)7ByDESvl0L-RF9LgZ0HuJaaVYc=FfDQ8tMT3ohfi
zVK(ENCSELmLf)9H+zL1Omy%Dib^F7>8MDvib`Jvo;p~U=hw|^dl;GM#HVFV@-oUYj
zVexnyiw-)q@vvcxjMMb)vq-E=Nn$7E%(%ddlFqF!rqfLauY%6P*z8T_gIVW`%}wCw
zuWd0$gUh>vVo0V?n(Y#e%I3^g8OYOLbM91#K--sY$?@lYjVFK7d=)wAFPr52tk*3)
zajZY`eBfi=<vENqu&{90{u}^$-hMr!C~k#nMCT%u?noRmC#&l2pGc3Ij}kx0grpD7
zdR5=mq@4{@dV2h>l6{DET(L$qMT;8Q#YUm><}adE8{*tX=f^w<M=wUEpos69OI#_O
zUS&?~FyonB=_{T?f{42nAC|Fymnen;3XDPo0#N9DdOK$GO-HF@e{=kTjG)tIVYN+y
zc@6d30c;Hf`+BS5hUZenUtYYadJFywz65uf71myO)eG$nMY^!9iYJC2Zy-JlV4U<n
z5b+w_69szP#?a}ze2=5+H&F%dzq-+CwkF32qGQ64n?sE$GKfDY#8cnKwp*IbD|_ib
zg-$AA%?Z{-$!6@D6D$Ew6>RQRy`V6V8LfWyaY$oh7d)$l4+l2wO)k?ho7$v)V8n!B
z$7Hbza`y6OEq$jW734z$!X3$5%)A1pd|Ah@juze=VS?JUh?J-sMhE+qwcN0f6VS2e
zn>rm30X|TMMnF!vp7$|-vT(rqfI)x;I3+(o$$I3dNHOETAf*8Sz4TLf%5Ktj+Q3~P
zAs{*lzX<9)#8^yDPPj)|8kXYeB<OQO!?<!Q_@xjS`k{dW1EG~r?C>BP@;FQnW0~QZ
zTl!JKI@|v05D`pCI+z-l>MX2^=>@nH{bNv9inXMPiVZ>Hx?qw<bK7vyvUu<S$kt2&
zL^fzLYgF$2b|iq}go%9-LbpttN^eZ+&?*!@RfGtPc(|CSB7|cuWdS#yl132ME<j|A
z|769^MSomyMVv&OHjyIL>T#;zOWNkSAXns%iP(|*+aqZf&sxIRhC%Ph7kCz(WSRB7
z)C{Z5Iu6hJ_rE_mkR%{lR{Nm&I@l`ub@|f41{HE%>I0w?i8(YEx=22rMqyRg%bLX~
z`pBB~tUQ!my(kPO83Lr+oP*Amep>d)^OTTUpb`gQR8c)tHnHK%oOTtNX`-#HB4i?`
zfbDn>w1oJgR(Nk3?O8Gdy2xUHz?D7qosW7aZOq_3P3eKnQVDIM(k+l^7^N7cIC?!z
zU{eWg)d8%bil?0)!y+{+sOEGrmI%SUmj~Lx6%LA15x3$i_EZff$oE^orAz$qHZo#E
zm#WaINJRXIUU8zjZP1eXasJ3$mEh!<OQ)WSPsaKOftV~-@=RIVn>D?b(cnu|zcItj
zfzmJ!Q<+Du!#J!fJ=N}qYO9<d;$Tt{ma(&qw>v<LMurgvnN7H>v0$T|g|yT;-Pf{~
z^=q1@OxDqXJ2LQ;V)Wa4A4j0Cr^mb;R%6G{uk?HRadJ;$qh2M9-)iw|H=W=Tx#aWi
zH0Y_o{P|2`?U$H(_6ybB@|zvPOPRzWs;O5oPeb?5eY{Np4kt;Q?|_8dJ`2|Ln6^oW
zCMf{30i)XQ6coq=v(;FuM~g_Vl*vZc0aEckN>uYGSvwG4l9@#>G6t2E2rI}f+wo$M
z7ev;q5AR=3J!iE;m)w;Q5X7s2bv9E5KMl4^FoMiboHw0E&_fX6*QzgBc238Lqy|X`
znRcNA*D7u!gjkeBadljBZPdMb(v5R$5$OORSsqtHjFG{gPh{$o0J1y~`_)aR-xi@s
z!TZ)h#Szt?Qft3`GBsR^S)5)~F{qxUUTF1Ssh@jmx6@=9U|@S_qWCG+VAQe`DRHTL
zdbA@7c}SCcXq#(p>{G-cKMUZYuI#@jYd(m*b*Q0-HYIq&>1!r+9i!_k9yRRxP)PvT
z(Veg$a%IeL;pa#>l`SIoNyPqe?Rbr_y3)~i@mgAO7N88ZgV7Af!%s~C;(sVJelkNt
zeDF$_w#ks|AIg%UKA2K|XCX7^rI%$YKI3LyLc`)kMY)28C{9AN;=55{2-#Ju>`zgT
zl>|(RhAtAoFBshv!G}1>N^?QLe^CM^_T@;X98O98peYo`*oHR+-8wM~_LxwM_%*VI
z-IEg_sp{2)(&b9wAZen1zGxUb(aNa=z3Q6_x8FvYEx=LH^rmP@BlBCQ$8ua7nz`*p
zm0k?_HRExs;K7gl=|ypWq(OXjW|{fJr<V+a@rD?mU{%k7XSH^SQ&Mzxb@>LsBOkAu
zwE<<KM({A)KQsvyO$eF+dE^r@{2990P3KctMp4qJ>#8G0?&Fd5tL<LmXk3ml=VPxR
zSW~6tT@v~TYi2h&p5L3;t;{Zy#G6{fOXs@Hba)8gJp-4CP~<x@Yz&lfE}yj7T=+Oj
zin(pf?-yh*Tc{f1_r|8NYj;OL5_!jSOrnk(d2=?=%jL$!2#{#5;k|h}FBNLNrbXj8
zy^Vodgbwl1gl!{+x({lEH+ER1LV7|ycec)+D|(km2arr}R!BE%eIku1LwxO^57T!m
zxPJ-RUdl0S9@QLMd(#@L@HUn+(chA-1$w~>Wv6`3+=^5veLFq2Uh^7o8bsCA-WXS&
zgz4JFQa%=)NO;+GP1##bEt6kDou&n<gQaAcI<v~+vwS)y{bPm2Szwp#!wdY)A?Yf;
zrtE0@*e}igryOn*d}e>wHPOr9cH|o(!x>DKh<QZH;ZC|+Fa2=Zbu&)Mo;TdR;kssN
zK=UX&j32)R$bi$OyIVcr?U?+}W_Y@65*W`aKP~Ss`-&b11+Aw-e>;Wb7pl@&xct3&
zF<+K0J+84XFWAX=1aA!hT+@X}mGNkH$E>kO&l$umA1wMTI7X#V7`V35-=!j91HS&U
zZ#!Rg{`CMTPqxnTMan%TQ&@?hRGC#jq|N~@F96?Q_RFMLx<(8Dn>Bkw!m)U6HwQt{
zStyvIk}q#Ag_aTU0`)6_$bHlDIo7JNU_Ky@+u-n_6nFN9*T5e8uNZczHEG8k&0moo
ztafpKDs@gSfg_D{$fZjjT$Sw-5fHSIvzIBd3%Q*EjnAI5MQ<g#lrt`Cr9acQUFz`?
zgw2aJw&w<|z)bppMRjlHCb7m>NLCoK0|?Z`KY4dWFPALc_Jp$^oZZJ2HtIUyy+$L)
zn-<S2&&}1iu46JU61YD2390k#_H487!n|!BKcTa?Fypyy@(7z;Z4MvGe55IjjC)sK
zA=~W5YD_}S^1`TjyF)RBqCGH?S2RE+zKxpTWst0xUhPW&aDjBvjQ-UiINGKg`UueW
zg#KaSRuwZPEfC#a=bflRAh7@n7|c{c%Ig!=)Dg>|y(z&NkD61ph|Yv1)H{(rP<WKE
z1soCMQ|CMl3R<oTf9W61qx`cc_!!AVubyN>>_*;!wt9VFP+80956l?3-(&yDR_#A2
z(^kYMF0XI_RNEy;#rZ*ah{a9Gf7@$Yld0QjRKGf1s(;OacR5?+*W$rioF3GaFl{x4
zP6f~J_Z5p>^(dY#X2q~qd{=zMPCJVDcIP?MG551xw>zSWfo+bCpi6Jv6zhLdbx4w8
z^Eq*U3MN7^1*Ah64H)&v0PubqpxiPF7JTS+ba!e3M%lP-S$F@;;2=YgXW^DK^xuIl
zX0hZ>mV|$r^t7qMR}N~s6Z$2(S@W}z^q3N4<#4mu=4b1t`}8@~Bv!BvnLxr3p-z)2
z5#{7F4H(ZrSmjn62UMhWZOz~*SFB^%TgWziT0OWYQej%H9ena%%V|DyB%w>u!734}
zPtHXF7WMmaHdlJ%H+Pa8(=R?ky~}E1ZT_mQ{LBotj1<dymMyXu;3ThWFndl4sC75@
z82J5GE7Db@=bQGbQ^EI8&iBbCCF!@ATh8ZR!N--FxKIVAN*SmJHT!#If~eAHBU;Ih
z53!w;L%OV}X<grGIH^|efVA^G7CVsXrQL=d0B=72xaX>6@D6c|h6@tFbvEyF8@e$x
zN>57d{!*++J_>E~Aj*Uvi-&r^q_Zk&bkfjngy6A(vHYe}kAyV8KTprcq7q6Mf-fUJ
z<Qb4H30bMNH(>G{@h6lAZ#U6|#S?X{%8ejM-Pt0Dz5(M2&8lql*(5vyl6uJtcW@32
za0u8l07-AEdZHpgX<cJ=U_}xITP}qNTx)jLGwENDIQ*dnU&sXz9In9I`wHpFE$_kC
zLZRGfCE8Goh9w^%deTduaD0|P25&D9%+9sDCe1CdkeI_qU_iYRj(qm{*>s}V6h*TR
z`vCF>44fV2(qT5g4_)5iC2_6kB`-M(AS95XP*wCOe%8%*7E`UwWDrK8q&o4kvhp;-
zM(q+jgahNNxrO$|T12#(m(MYziooUNg(NMc1LJ!igM?`R;EE9vluv-lJ_EMy5%D}P
zne`%17zJ{)kp%sAsZ{1#3zY+^YoFTA`m7^6Vqa7fOC!ErROR^=M4|LRh&_Y`01Zp$
zk&yGx1Y!fW3QWl~(7Iq(_KDFFbbu*B7o_vo0&4Ed)B-xM2I2O{(DCFZjN=yk`R*i~
z`xTGEDMaQkZ2&6`<m|=d&!0DUhL3DhKFA5z6OM$=vO!Ht+tLjq=#Ll)=RaT_fZgwt
zZ==-A;$WfeRL<R928T<QQ@kSqsGOU5ocf6dll5}~!>Rd40DHe#Lqm&1PF)mRiru<2
z_0y5e^~DOS6sdAOW}<(l!Az9{8!)beZ{Ipl0Ww0&I8)ybpWC;;(WDkk`NKd^vk+1R
zsz<J#eq!4h4i8l*=D@}h3s**ted%KC>!joo@uc$(ytE<lAq-h46{zi908-6y+15q<
zD$^^^>t8Nfy5wr+I}w`kZrU5DNM(0^Kl6)t>>JA7l^OZ|QX@lbsCKYE)9B&HD`{1m
zMJ?h4JKjfcjDIBUKZaLgUQeW&YOPjCqtLo-tri0oF)LcfZM(N>IBI~ga%NyGkxTBZ
zyO(S>sbl1Wxjkhjdq4$v0ADD+9HCAh5w<+XlxI;70?`7|zuN&0bijDGxex}0^p7nn
znS$|jp#7Bi-JUX>lqH*Q^~{h&H~aF>bb-Oc$e7NVLa#NY2Gi<kD9u&M9u?1jhDv7^
zm7dNt556*fr=@9|b3e+5H0AsiCnQ8-lnpS9v`8BnM3pTtOa>`J1q7})V3zbziXkBq
zjebbislh23#P9oq@jy_uP@ytP2K1c8Igvt8Awl6nVHSsPftVmF-6s8DlHOy0EkYG?
z#J+(XrUVzKI==H)go#!%NEQa^k5!Dcg#nLN%6EP*!u2#zgt339qp<*nvKz#7gCLSu
z+cR;3IYAUY;LPKx2B-`zlkq8Y1Qh!8Hmo^4lV03hL3Hw6yagMLY;g9cVOXs&`j~Py
z@!sUk{DNz|B3@0vqF=v`r8Q(QAvY<dWBx$^7rgn;(FAU{A1(R^>!%`8cr#*R#hAbE
z-yxyX1v_`RS^9y9E8lrvvSrXtPQTzN#8}I7_6DMcJ#rgB`L}>#<VR&nf0pwu`TKjx
z*Bv6a`eL4mXfJ5cI$YGa=R!S4_dYn<U#5o18u>7)eLPKW?>~=;00EtVtU-<5G#!u8
zzm&$venhSY%ARIGPJMGN!=mV;A&}XYLAM)`BC&AjlXzrNkYLdA3jDgN^!K_<ilamH
zvNAHf3%SMsExYHn96WcftZ?}Rh2Vx-X{AW!YeLh_d8bw^Ian#`+&mJ2s=+ymeY7~o
zWhqBZ$@>e`+eMf+r7II2LF7(<3M+pbXiOqizK4C5FIcJDfo$KX&z#cK!N^$B1=sr7
z9Ud~13Rhpm5TV28o-HENkmRvi>5>JT1hbgUWt=C##vg5`BGww7JSlz=ej2)vMV{hq
zdJrH|l9%XnXl-r5m7&~Ib)LEovqf69!4V;QKY%levp?^h-dLpu>?(w#1Ib}IgieRR
z9prcx5ie&J2%elGa$t$|pRED5B*+1=y(Xc=4^dVg<7n3$ZTGiDEo5*N(gHYgL~h1o
z`g<h+BFFLdiyig>QFBu+r=mx)7n5MJkK75Y6u1KfX%Z;%iO~Jq!h@tH8lMiqSUMy!
zH{w2i6?aSp$F+TmTKE3w@&(&X{*u&d_cr+zPl;0PQi|1TEO?tX(EcdlBW>G9q(B1G
zBX0E-E?55d87Km&!2Gf1xht|QI5z|&@E4%K9F9J+Yw@UmxJ&17aF@7J0FJ(9_s?44
zVIF;u>crcu5>rM|bb}}N>7K!<V&SZDh2oQNv>%d_*764Z7PX$7eO)0&kDd5*$&U|f
zGX_}tD|E3xM4_$-O!5&#Jvlgt`!s@ZcT!PEzvyF<j?vJdqMiqLR9dMF>1*{3tbPI1
zbt11SjjCWCQ-)qrqAh{NOBO9y+lz+6*yyTf{oMzisrnW`;&K0gaIAOwFMj`;dQQ-*
z`$5IgRg5;<U`gvb5+o!jfB&_fKDI4(72*KC6q2?eQ=W#jD@>;ZJInFd2kydSAbQJC
z5dtb5=Lfe-Eqb~4<F=}WXe;xs7ZyNeO%_9w!lqgUE7PfQ=;v|z0LDXt4d)u-6BYJ%
zx0-<YmyPrv3+)2iqX1A_nsyr+gBy{2Jur6lRrUe$)ajWC3R6e+JL!REZ`Lryx#~D{
z*08z#Jwd@JbGwTDcsmoO;heJ!V-nVsNFq`=qN|KtO6obTXd1p=9xQs%BQ!u>H^q{Y
z_jF*BXnU)5aKWZXJBEVS^S?VUo`eNV&S3jj)tIT}!LsDyD$O<v2J?J(G1$Il_X9FG
z^oXh?37OKqsPD-6NFn?%pRX{!x;R#0NYom*NGuJLbEF9VZj6zls1ZaThf=*5OUybu
zoP7ozZ{<%ab&SMD1zJRaDoX8hk>t6Q$7cy#^h~;*xn?`FVs3>jj9r@_gv{u>P>}I*
zCN`58?Q)IEoL=v=^cdhfP2535>Ow9>vV89YhL%Ty16#EgGBlcgE?&6F-pN5xye<|f
zguZF-2(m)sO#=#|AuJ!PG$Os$Eyj>=n_T4KZq*y&CX{mH$n|G{@Ct#SE<;Qa?>1Jx
z`@D+beGU}{2+xvFTQ_}9$dW4LQHLOXLk&S<?9xm~cPHRqVyC^s;zMDwL{8@K!PDId
zGE}rX8CKoH6uDnXn<?CRYicSNDM}I*1EntfG^gN^GV5nQORbbGlf)ZEl6Nky<U3y>
zi!Hej!|`UJ${3shQF6NK67255YhLP;2WZm+Q2~wk?YSbKBf1+~2xoj6B9hCpqtoL2
z(Aa2eEv%`KR9P)XcsAEE2nkHJ#x7=<E#-4%=vO-xaZNUgPdI-KcnOJ(p(7ohPf7(j
zYt3e+?~OnM+R@2p<8I{HwWo7UK65tlgu7yX_;~tfq*#&z6w^9}o|*N^mepWDlB@oF
z9uYl`ytUZEL_onuX6&-VL=X(beydU@V|B0PD_sZ8^l{r{UyNUj^e12a&7xjnKh?02
zXm>C<?6_h)eKMd{Yoio<KW~<5u*h&JAdAc}*Hm?9Ub`P@gu8ikJxr(xdCvi;pe0eM
z?eW<z28|v7zb;H7gqY<GJ#fO^@YPFqFV>85RDo@laijkl*clO-$o)8IKIY|ZyqKMN
z-%Up}&hoolxPC=0AE8h0_+(_PhP)_H$P?f_VSbCdbvlOd<64lGwfZShY1&ztGw!re
zkKa*Q{e+($=>EHQuTyEyUE#)iF1svDYN-YAQcqI&jC-Z&O6i!(nsu5f1id&|XMU0*
z<OdEUHE<GPi1HkB?FF5mRL(B2Ne&J!W+&<E6?9cTy?VS;2ETiuH=gt^E0v}+*qt%J
zP(KVgF^>4Fh^8TMD4k`q@Q{x<d2UD|DGEazJ{WRfFa(n2p{FA5Q80n2v2lRdc%upc
z1m;f<Y!klq*=`U(`(jJ`eiJ<d75Fk-ivvDCsw;EBl4UBjDuHQLqv=KgDVaw21G?Fq
zD#DE&TL~AcCdYSm1htR^7f(dhc)Me*3@U;^k*rYINRt$9Rxq1lDPWo@?3r*`BnJ-n
zG+q`N0s{(<@7zp%DZ{~YnRWq)03R3-40JDCk&N=)m2(=sHGY=OIc!)%ME9Le<XKX3
z%{gDsvnwJNQKnho@ZLe@+b=LTqth0SW0&;8>xz(-VS7K0hs8gJb5b1iiJ4@>;ETNL
zYG_tx-G;WkT7n>L!zs2`%EBxa+3c47F-0U&iQPie3ZJC$VP_OuzJl^mylw}u$koM4
zUyk?vP2=hhta^~rWJEr0`YRIM`CMy6EHR3wwSX9EcyIt_;)?NCI)tpALt>?^b=*kc
zt)4$h+KKc0N1Y8&GLNCgd)7Joa1aqPJT82XNyck_7qOw(R2FY)9#h4h|F8(#a9AEe
zj>6D)Z+<^X&+FKsSlKOzMh_lfwQ+NdhnhJ6JGI2hkb~|NklUns@AT{j7Dy<VJ|^YY
zcW=0+f|&v<5|RZuFl$G;!C-wAV`hOxId!`abfwO7U?IY-zC?xH_-o5t(4d}CJdQ~r
zPE)g31`D$Qa@3<VRpd`}P&%KPvRlDlKODFWajnw5KF1945a`Pfhv;I;8Tm2(e~ugA
z9}d%FO?;^nXj>voiHu`>Y3t@Whjx*yt4eGygluVzTfhuB0l@2mL88>&_|LXu++*DK
zjR>Qt&dILp#()2&eQp6$fwZ8fQEmfA!Y@E#firJ21<_l3j&i2;ZUa*QOQxM{0~7Ht
z5iufR`UBnUA{+-64hb57`4_+^elZ;(DNNb>iNCQI5FKD05epB(K5$v7m+5uo%wWyf
z`X>?q2a{5?g{MjG026?m{!X*p0j2{cPfPoL3&4j73tewC;)I~$mXgQ8fHaJXE``tc
zwKp2*ho|ZVM*X@)*q^+2pJ!~&jeE`8W<)FoGM`9k6RAS4))ExddDkuBP5voLV6OKO
z@o@W(inKDYUIe$(KOt<wSu4-HkCcyeaj6MSn->(s&2bN>qFi3uHg|7FNH_(aXven%
z(7a*oxb$TAPC$cdgB=b27NLKGKVU;1Mu->1mojkF^?k2to4^?Of({rNs%@9oNF4QA
zXfBls{2-7jR=O0cUL)q1I>r+{Aw+?^>EE$cvR~wR`o4uXc)#JL9Qzgx9=O_*92Tb5
zqDzA~GDS0*5KC*!ccQ5zB9qB=tQ(sF%C&arf1C{oqbg*F*dR@aUx*XFL-@49J$}8>
zkscQ<E4wm`%F;{i_Q)mS9O<;E(GkfOOEFFxpG2uAFar3VmPF99{d<Hzy$IK)rY{ML
z99@UX5FeDm@si4CN|((y2l*dmy<;uKNZ#D*8A}I77zNh?j>{#^`|kZ8V|S<lljD<m
zs7HkslCu&z!ZmogNkK|Al+wqMv25&KQb9hV>k-18sBa3SYH;f8RsDMgs4e2pRgA?J
zqLu*OOijjg&fESacEi24pGs|1VeEVKdp%c>p{=%MVln*97Jub^i%p{{cU~d+;`mSH
zZM5cct8#MHQrXoAtoRl$zs3>(KeJ-@pk2;!c3tS<I3<5*|8ZV}&L6PpFEEu^n6$(^
zyDXd)1?{H5Xg)?5*~DUdJ}D5B%-aG(Hxq~V9L1D7_LLIu{!b@YvE_AEiH{uTLT*%x
ziKgx^OpyzetS*JsiccvOzSO);y1f;n=mI|{LgJh!v*N7`;)ur)px_4ueDaJ!cG;Wu
zX6xR_;!4KWYg<czj~;b)&fS3N-3Ns&#n-~+*QFkfy(5}3W<Tv={$EsPeIMU!lQ?Gz
zX?#bN&>!{yx3<XY{R7)I&50v<v@}1a|3lR`2Y2>FU&hJAwr$(C?TPJ)^@(lU=ESyb
z+qN;u=KI^-+S<BRU9a9hud3^H-#+J_d-^wQpfhWQ14h+2mMXk-r`9QU+QQefl^8~I
zf&-emjZGf>pu0idMK7Q+K^ypZ>Nq0vMJFpKN3~u@xVQVA&a=c0B=NqhVkr%M=KZ5x
z&VqL92T(z#olQxcfnu-|ko=vB^9kzBaQyWGLTi6O3WWY*)g^D?;Oe6(W^pdoyx3w3
zVQJC6DX7b~?Nd4t8)a-5od0KsBPk(t{<>B9e2B&tne@#U#1pVV8x~iH;Cl8IR`&z=
zY2_~Mw6-3PXnTiBRR%>CphvXuW5imxh~MSoadI<?D24uBWu+2g`=~6STPHlD?nk$i
z|DyIX;h|)@+^w+yYLC2Z|BY)jYyPam0!Nj_H-^$=rXqC#^GKJmS++!iu3H1yOdH#p
zG~7>Xv-N}f7y$+g<3OT4r>84R@jjBb7%P+eBLP*C1u!R^7eCCM5PX;BBRuVowSOoQ
zio)@}_K?=Hzh#P8bA}!{y5yB9Q$2=cI0iYYM!}E)rAH8Y!G-U*l5r;)5G<?6QXtW6
z*UH0{lzF6ZouVFIZm@VQ^B}oCb-g3%25F>vTt<EYHh{YhCWfh>64Y5Kso_Cc9`_4i
zZf<T6-r!D~3=^#F!f~-Jk;hM+VaH~s&=b39*7LQh>~WN2hR7hw*w7oBW{t2VxthtR
zm94`*t$UH^Z-hP6ZQXp)^MY=kD&93;y|MD4ugEH;x-gF^Kquub(Y0x@5fGHGGng%y
z_7$Gi3E)iEN7V{XJR3qQxIoA^=csW?3XXIe5EiIUmKu)xECXtjM5~|uCdq!U7XHyc
zkKRG<8#rD$LGF2zXk=0L>`<X`)(cyza+Q<Qt1PO#-hnk)oW)8=5_zPW=qpT^djOa6
zcx1MGsDs)Af`I#-(6$2WKMY{_-^6suy1c`O0njJrr9qqQxQ0X$gSb4izb<)|`dQ8e
z7@=`Y+XU-z1L-lH!`FoGID6kDVgKaa9`?L4`04N@Ab~4<iHL|}ijd!h_B9@E;8OlD
zSlq2!p4oq{X(zT^t2oyv5ZU26Y*3)u`k6osZv=rp+!v(MDbYzq0@-h4n&h~DGQ?On
z;JmC&6*{X#cTmM|_B1^0nW4h8^^bg~pY?A4F^j^0*r3)=Ux_LbQk~e~y<}jfC4|lE
zAK9(HBE;#Ekc0!D9UL-EJ0OvqybBfyJjE_vJ5(;`QxKgsY&tE{$w|r5XWi9^^ny}9
zC>l`7%3V}V*<w3yS^qL-TV4@o7QI*rxXkt}b{0v(0l&6-@QccaFXy8PF*fLuNuWXb
zFBcsLv$CZ*7hFiS5+y&cZ|}xR5zhV_crp&;m;|NjO-K5>2d^wobi^kyN^h|Q0-}5R
zB0FJRbhW{JdL4H0nP4iffS--;>^XbJ#VeY((12G;(Sue>5i38FFKznYYrq;!hbUC6
z-Yp_G-|6EZ_W?z?DGfVz|M+!vl$k|hzZcD0wr+dfC9B7i{_z6XG+Pg8czlHxLxx;>
zqrd+F{8XA=rsWQiOU|S(YzX1!Y>r;)2^UoJh+gI_j1DRsq?vW0s(Y$xS_`E&Rh<O+
zs=uC(3m8m8+4Z!kQ;uqAC%~dzzC8(cs+)awvrY~+tFz<i-$O>6xyfNVO&~u+TH$|+
zODMP?2tq*|QuAPnG4X88-sUF@Dg5uBW=&#cQ3E4GvYADkA1#?PXgJUaZmjqs^c4x3
z%UmXB;yofJHj-5FniUm?-E~=&v#&~0$ZztMGS9ZcnPPUY*e)eC0QdDKhHHMlZiB+V
zid)$^(cJpl$+&6rAzKR@x1<jCw`$)%plc<TR_EP%_PU83W*zW5h~5y-A}@Gf;7y)^
z{tU0~UoWpcyekdO1d;T@ZP8<t&&^kj|39)JEGb{!d&4X(H{k`<b*ROX#}zXnWCj9$
z*aL_+Su1p%f~pMOZ2#`tWwb|BNK_DXxq&vLP-J!p8qD*+vew-P3L0vq4uk%|r_;LL
zZ0k+p77{4`9bgqd5)-=3=5h*qkkwY+Zhtfp)^^AfL6>R@9JbK4fp66U<)+=4*|V)q
z-wIIVeUhxiMsP|2@uxsDukokKZCjmffE&ugmH8<?2Hj9WqT$jbQm@gvpa&uAG=6?|
z{7saJQhch-=Nflctv1WAVPLe+=gfE8E|F{L>Up}3mZxPK{x-qkDRz96U^+*)ztx{J
zUGrs>3vgAe_BVE`!<(OpjjLbm_7~qq9cz~DUaQ^K>?HsXo_B`gC0M*eve?Ofz~-Rg
zx@($$FDzL7g!UyfV|9^hr-b{BffCW}GHe1(PSICFEq9AOc2X81=ibIMG0yX9!PlTi
z4`d%q6y{=#P%u4y12*KdTk_`50W{gN5(sU3gV}>d&L*r_4;erj^bN64fR1+-HM5%8
z*^YsL&&>;Eo8>v0gDP#>h~4i8@Ecizsvkt*tubS3Fz(l4E+<G$vBVPV@uLspQ+3H+
zpOz(WQt(1oJDJ`QO~b92)!Y#t9XlyJ`O598bd<-Ddj1~I`qKp7q7Q<35KH%gMnHVw
z8IwKmeBC9jOiwhdi9uAC@b{qO^82eq0)&(Zcv#XRL|*S|O}PL3>^DFjpul0+$H7FG
znxy4jQ$9;6$d=~*_<Ply#A?7Bf1TnRpM~gLX~0vwT&P?sDZ^q^e_pP|pq?|!N71QF
zi0yetKt|+=Cw=%P*`O02jnQ!Vz`KT6TN#}~11;==42ZGKy0;*5ybAk557wqsoB50g
z?`y<YHTtZ{=f0{V7P0FVFe`Rc;ap&yyMz%MC6Q+xYP|BG2u&!_^U3M}2)=o36ZSQl
zXMOYV#FMGSc+h9e#u+?hO@Grm(0s*eB>nqYS*uUNaf`55v#$XLzJ4Nlgy*J8vdDJ+
zfU5bs&Od1qu4vXhFK1UWtw>UkCh1)@@-ZqS4g2}SjtFAGY7CwQV9F~+(zMK3p1eU9
zB#o^4`Gf>PeJ4NK5moTX`d8pDW*Yhjh!dk_`x^Tb=1l$gbWtD2PF8h!>gQcwV&v&#
z(cqh0593MaIr^vc`jC?jKYgmgIMtXfcG6VxILyQO9d+mm$HY*fyYRnx`&3^y84?{1
zz1eCnra_pXp)SiVz<U%er^Q3yCEDJg+-;VGj-^ZTaI`Qs9w+Al;q1d%f}4K^EX`hZ
z;qL9U5ib<S$`(&#V;be2h-8~PI>P`l%tb=K=U$N2HExt(ZMPjhyq>bN)EwE?W@Vua
zb-E0;!ilm#)1SGnHZs8Hj~AMS>iJC@pfzN>XvM9iXrb#1AejW%W}|1Y`;O(CxUtdO
z(c5wwabYr&=U#di5+N)P!DcL}A8e|}M!l69PV$*wfGp0p0`P2X#6EVQ{gxLZ#bA-;
zWO$PJyW($hd7oi8I`r3F+W4u1lQ8Q^Bwn_?(cuqkej4(&7Dg_Heki8nTb+WHjJEY?
z!T?B*2D8c{z+zkSm}_vsqtg2m)hd_N-*g?kTdY0a5T8rn@lqWhOXa8IF7>3mb2x|0
z_j#*wp0&!qrd+)5!XM<M%It-R*hE~k|E&W`mj7eQvEb(x5y{=f43lW$LmKWwZ({(N
z#y|SMs9qn^g#2}M(-(jA8Z|#8=Z|I<n)jV5I#G1j0mShe%t02su1oGVHGzSxR&=`<
zHvs`KuIoTEzn}Ht_088T2|^qIYm0sv>C_DICHE4Am09$L#tq8XjSCgFl7Z26@8ZnI
zQ%ySW!K#vd`Io{oo&{p_I31#HVP@&%BCJ;;uM}9hBzix0hVf%86^Uc0-5hW<wJXNN
z#KeO00s3o76P_aU2$;&!`pRnC_x_BOpM|knXcD2KM>D{P>}6t<f@SZ)T%h-+S|^JH
zzgS67#fU`q@KY(*DV^!T{u3`&*5Lg*%EE-OEKiQ3%+f6LR>f>*DQd2vqY4yJ38VBz
zp}WxL03Swl4r;KYBHvIxd<sdFo{4P-maqhg2C(d}D@^)pcY3Ci6=wP%MgljPWZq2^
z=pVX?<^y8XMBIK4ZYjo;Hb`ulOHYw;g|Hi2sl8sav+SX!gr8Kj6;b8mRwc+yl1D!s
zDrcWmtZcSQq{kW#KaL{f1%pqp>O*-+-y}koaO9Nz6NZiSSwUISotW<9@|7}M3BT*_
z1=!fWb?4)<UCeP|hCOW!zC*{isrt<fhY@vbPu4uSB=nebw^2sOdo~@Q6nYzB5K~A(
z8Kv95zBeow61YQOxp4ShNLQ<ukJ*Q&+;9F4!@cLNzsgk+!}UHmHxkR-D3)I?>s$oU
z;P1jvpmyqGQ@9yITHH{kuwJn?T*YkK2N-qUPR2-thEDcj6%Ok<zzhprffSOH)D9jc
zaa36|)oPCe!<-H7Elr+bq%s)dP?W#AC+-cR=a79R(?$h=LR?H<P0;LJ#GVe=v2J)i
zy_}xaUY)^28>TsL7Mta+H-1au&6ddh^tK4S=mHR1b8mGslh<MsxPa!cd#*<B0aDon
zEB$4BBqcMum!XY&&j^))28kWkSOL3B9o84r&X-t>@i2}-pk5F*vFCCyX2eUxEijp^
z;EZP|A36vre0lUqe_8fh-4i5^<MP#ZxGJxUSN7#8Lp;1StMREKPCc!^%62YC9h<9}
zkWO`g+w)_H14mEdVm9R4c2ko;0aQ_rGs}B~*JtG-Y!jC}FZ-@mQII(<0!9%_8O|(_
zThhmln8zf|WUJ6j_f2GycTXJI!Xj=WR&F9HsiYu5I0szm@(~BT;AGAi6q6DZZr32W
z;Kvk`YFpswdz`DESeIxBBV+6+>f(1GFt_>!Y_9^5>7-I2(wbj`3!sJG0LNvwnB)(;
zGnaU%bQr3Es(8_zy(;^xr!F70n1=3$mut=o9((j(<~}eosMGK0E6i6QI}Xb&nx6r-
zLD{PW??QGA{M3iXi3xE-h>&SC7+$}ArpILq=y`ZFr3cWJ(S=jnH(Y)-`sdJt93A@N
zx}l~+BJX4gZJ_-__ZH>?a6JUJZRPSph1^D@epZ5SjXngtfsGFLudNpw+VYrnDH~Oi
zVL7|s6!-78j-v2U<%>LJiH|K3r$4iCfo{_p-lrSw=6JRCP(NK?k>i_PxWgPMx}D+P
z^0OKjy4}&uR-FTleF+*&OAr|}Ov_-CY)3B#?Iv>M?@k9913C|1fS1ok>*INyyX7)H
zqg+mZ!&3_|c4+9+@*}YwsHp#Sbu4GzEP<JN;0*#wnnJJM6&tsuF;}kO`>s}#TrJ(-
z-<XK4w;}0k;n@3SF0*z`NJs%%d<uL(?Bp!A|MX?R&EU-A)^=p<8gUJq2=S-wq3>MH
zxmWb=6S;B!wF@`o08E{TI57{6m@b>P;&!lmXo|`yH4dNOGn}#Y=3A{2Gdqr>?`L?9
zcdn@`-<}JDKEDq3?-O+Ob`q|)k{>(yVr};GF(~peFxV%3ZG+*#r$Utca3VgwM$KML
zwc~gH)xQiKno>02L}23E!AjO$kifSu^L@?zmv8rh`+j!<9#2kRFGwKJ!s!D>MI<N#
zS9^{8q@93Gj8N-_>Fe;l@es;$8mh>#i~D#amCj1KOYH6;zhvnP>>>p98~r6^;gm*}
zLMtkD$R*xL$j(ukmVkq8O61^f3i+38D!7xPR%z}G9WVFYz7j=LhnFEIdgOalY++|E
z(FcWTk<pC+nI!?%Sz5pT;lHGyz@(+Bf8OKHdyP^F`T2;@*U;NByy|~=kJbJrsk{Tz
zX=Ppb81I$u(x5xxW3L{`Ab-ezAM^r~#@cUrGmiylPq4m_0Hp-4mjb>KnM|rCa8N%b
z5;hi@MT7)LXuoPDtD!{yn0A^>MEzQ%h3)fhN-GcrKo9OAtW1et0DSUXu~Ru}PdF-$
z+z|(BVmU|MeHY<wl}pj^13A>)ZEz_PY%rl>Q0%6B!1w=k`hth$?e>03BXbJBp`ovZ
zl*Z*PuWD823qW~$yRGzNI&!(VhQK1z^rPm8ooHbQ9KfnBex(gcA8W|hV<(K3uOgm&
zeEU=YlEoK~Rgip~72e&j7jau-52J984Njnu5+MhHq@~Rm{<^ss@>l(eIZ*FfTa~rV
zOxnF&k-&BE<2LN|OCTn|T?nE&M_42P4$I5`MOU-iHqYvJZ?Fb(GWh8ExHV)ZA<S5m
zS>)u2`RlHMj6Imw2c+)izr47lyG~Iv#PJOX=ze%c)400Li$S;BD7b0kC7JlO+IH5o
z1@cR~MZ*#^1c87dh8q_v>21!g&qN=|Z!8}|g&qE11FLFJqR;;6cm5@bt+G_Bf(<wG
zDo1@}oVYx{D1D-`uf%J+Ec9}NBj)@nHoIGY5HsP$8z-D!Gc!W7vIE7pMWk>hh)|?L
z)@OP4m`M4CVX?O#|77#tOp=F-lfkwVe<ALPPE3#VF`~Fnk916DBrV!vB`|3kIdnZQ
zst#ZoS{MAEEcHk4ht-`m>n8l>@fHXNMvq;p{u3}dfDbYT1fx5VWnM$C>7J{|y8Aj2
z;~S;KVHlp&xagh^cFD10@lqx8S)d$ry|Mf~THrM2CR%mNcNuSd8FpTSo=%I;+&&k!
z{KMkOqkf#;qvvpsk~Kmf-}x`YZ{kpdgsZ3Q&@IYbp$@WI%qI$eV~B{fexYr0V`X4e
zV8!@OK<<^gjgnFQ7Nh<wD<mDEG-?Y0Lg!EsH;*1Jay9Xr(3kis9)fJyW-r#^tzPA!
zX)KLI^n0VDJ0ka<g6NFHB;9+c8Hn?s`<~yn8b$nQQMhsfxn7CDV7~5|Yi-Bj{`gJR
zAd0W7ZisX4Wp=`{i;f{db4{%sy`5ch<??qt;2cAxSxxFOvv2-BN5y|<t$j2$LT}e`
z6tJ%=?>a^TlgCOu!X`$j&5?^Jn3?hiE5<DoZv`7%8_bt!hWPuFeIs%fpvkX@jpyIE
zw%Gn*JBVTtb0dj=n$$WnX;Qe=?WZ2}dr|;<t1_|@p#d^5(JIehE||}Iz>BVbCF>mk
z7_!%!wu!L9By&1;5thzs!&2lhoRd|9?Ba{kL5ih~;^7G-hJv3YPWpPLRPurdCRp;#
z<wW%?!Zwk_lfY99^||I(u}T>A_>iMyi@FPHJZxPE65JIG8@@_=nMM$L@VN2LzUf7Y
zDjUBGGOAI1t70~&kA3?rvyOET54;Hm1SvP}wHMl85KibJaOQ#3+$9!2Bszh@e0}6G
zd%C<hvf~oY$~Q>BU94$z>PyHaF(=h?hK^yhZ~|{olxWw7CW0zb+EUt{Zg$GAA8&NH
z|Dgrhy_3y`#OW?3sB_}-Lw;tR0#5o0h$3PC)(O1Zztsd#zS2F+co$~U4KS4gl35Co
zp4x(lj0{^OW+3e5az$&pBCKzu*bWS*EZl<}=p9tZa_;`yjj_?yNF=D^OVvhCOMmYE
zGjWJ4AvacOpr6rcI+ikJ;T;JI85KliX&@U(5rs~7$2=HxuU{z4Lc|^0-NmPre-R#F
z@*jNk)93|`&N+wnO}bN}uj2(s^Ny26k=pE1YM5fgyn&hI65Rj#jFD&~-K)!u$iQD3
zF!sE^b<N<1CYTfXk|yooCF4xi|H8S~ApU}A=u}~+NcomX0Y=y`T8wv-8Xjm?@`?_%
zWsyxfdi{HKC-qdc9;RZ|kr6OeNlCHj&*(p}U-gL;uc|!aLR3wciG2)+H03@liF;E}
zwb-_d{LQ%qp_ZFQ?NxO6rerRC1qUfZLm_cA9kQ4bWzO&e_{4QVrl#;JqB&T(V3mlm
zq+3O>M4@{8P6eHVl;a<`MF)r`sq#fxZ<x1L4DTEWQ5a#2_1?GWqHVH+>u6a6Z#fV_
zX827~#}Miv<7p!!wPXRCt)v{|S9CDp_L3@Rx^8ImUELj450}^sosal!o!wue+Lwts
zUt@u$2y2-R)y)c??|H1MSN7&!AG}s(FEAGfxqw!32X(8u4I0QaP8zY}#TrIx!axl-
zf=uaDN{$u`0-U&o2-AYu$iEie!L{zfnKw-5r7l}n)T!P+qep<D5t!fB(NkWvX}o{=
zR38W|S?x!w@9!2|Zudrw{q+X49iz`MEt0^M88!_(%lCFY&<3ziGptZyt>8B$(mr@P
z?6>eT&5AcI3qu)*WsyTfCH_9FiO3Ln`f$U-#h5b8?ej~{Yo((q+n(HnS+))OEM{w6
zpbQux@F`724{`y@>v-Y%)5fqmvN+9RAQr*z{hsL08U$wWk8R8Zvc`wzf)2}DqhP}h
z=pDvY#Dd4mtz+e?4H8~Pd6e8WNWTTmj^gV7f`=LkM~Odcq)R!7Cq0i`T;t*`!efs!
zy$2D8z~h4T<>hyhLs4gYb`6c-L4gOz!Ds`4?LEblbg=;HaS145pjo(NWVM(tydehS
zWW2rEKW=hoYmzQeB)%0kdzBr7j7@b2Wv-`7NCNdxGb|0Z4)@iII@#ImY#87t;snl$
z5L(X<%v-7<jVjG{umhwQ_mo=~g}twnB~{)E=hWm)CqRbQnbt&WCD_m)!Yf1MmcC%#
z{s#go0#L2_=7?u7&fsC-WfO@q?$LV_?mV<XpR|6eGU{V9iZXBlaw0{iblW2i#zhzJ
zAIW%!2g7}Hb0P07CU|6L<)Tu1Pf|P7_HhwdQSTKpfp-g2kt$A4M08ah1IjgO@M!u^
z(VXD};r|;3debgvM@tvixPMM?F~QESr2+a`MP2s7dFnXp{?=7Nn@5H$E;m&Z7)WZY
zNfWEBek-W{HX1<s3N)ROs^YX~*f?wk!>jzs(%3dtjdlaOaCT{u^S{(_tE?B`5{J@W
zJZQO(`FezPml};P-_uo_Z<I~S6CJ+kf4%V<9&rRU;QF8R*HHKOJD$>n+u*8}cmW_f
z$Odcmb^D6gMjx+8bKSr(c?!3TwN=9&?*FizG!XQd@!nnhRX!Gz!4xpHo7&NIP_HFu
z9^dx|!dnJ2G3Z;9$H=;~j7>m;7*hEszkB{FbrkzcZ=v5j;y<8c75ef^L*o?ZQHiYX
z+-Rbc;JkPG?(_MuN|qSX%K$2M6OaJ~DC^{QA}73`ZoxUosUC2>{lEC8<_FrW){}?s
zaNQpNS<$MU9}c$M05koaNu=NRY)8d5GZa@S=^z5-J}b8b1&5FfyZ<&PQEpczS6^4=
zK4gq9s~bI^>ZR<YD4SRzIF$>Bp&NzXs6HK2PnWX>GLL$6+RDTO*w~%hu)xQ4UN4yE
z$t?~i(o|eOeTjE#E){6*3NM@`+JPx-rp&?NPb|%>sYWJG&tStXZ|0Cllt?HI8VS)?
zeao?6$PmJtZ;>5eO#IH23$3tY9^q=lzhUAdilXQ_L)N>VW0^}QYv9-0(fyS9({jCT
z&ug~EF0^LDuh~2eIHWQgptN>s@lY4;+=Qk!Q62U3gJVy-)^+R3X%gtnnP;La4xB0~
zL?)Zg7;{wy#l{u)I`2kskPd0%_``Y~;BxHb=5?n;#F2~Ilk8={Fq$ZH^gB0cxq6;F
zVh6(ayzqz!I5s%J12Lzk83Byfh>8_)JQOmz#6fVhYWnXsfVTp_I0<@UdfQ@KcWMoc
z^!{OapdSv#Z;%BD;}5l3P>GW1ON^#520U1wfj~~Pw@9pdqiz7qbwNMnR6WJnXbxzD
zF!atC)>&nZ_8Tvs_@hg1=c`8L2UrH6?yq~%`E$dyVh$chnT(A3j3NDFTzjm#0HjnM
z2i^W>oLO2SV1JH0U3+AY{-O-b@D+#Q-}`_~e=G+u2XpxiBo){-Q0_1RiR$5dK+>4P
zxx*uk29DHA--<Lj?X)5r>@m<?@VSZ{yTmC$;=QJ|A{9o+`AdM^xdG^NczMd@-eDg2
z=fra!9y|3I=K1zyo8ZK^M#Dzp-GO?x50Vk#SoY%zKmyl5_Fw<0;~zV&WzjLE>Izj<
zQ7L5UpIie;rp-ze@RM3ZokwZkVZCzGxzRD>OeQ~i3{vs3N46aCd7>vWSG-~=7l_B4
zn=h-AJp%uj*ppxCu@-yEnZ|e>-o=IK5hyv-&n2`Y76r`*XGc~Jlf50ZM9Ktfe2GyZ
zWsdv;gp1UU#y8g7d>Ya8_{j&(+INr2YhEAQ6RKe&e)VCl{p5-0jmQfcCy%?ke)IFy
z&KKmt1>SG|^1F&Lf}*)jqL%Ryg_XWgl32L|UU!=-;*NuNrGM@$RAP^)EFsI|bQRv@
zFd}OKLW{n*CYqpnD$xN#Q~n>VIMKtR2m9-Q5fQK@sjvNFrQ=YGZGKaIFI9!T(Yad}
z^C)rm&3E*%?ZyZoVRlYF{OrE6{bVIlPF-PJfK<@{NQRXttn1en63pBBbPovjXfR2i
zIZNXKQrC3SaTUc+TTP6g>MXu*hAB)XM=FIa`URSlCVAIQQaUV5D$(jIs#8!9ZX6Kc
z8IlOeq8q;t*2n}dDhxFAL}JU2?sL{Le~pWv>M$|mLwmCCd1VGlEG2T-30EoVa-6JV
znLx?gL07=C@n}VgX`I59$mzMAHyZ1Qun&B050v4+tc{p@@D?Ky%85*J8_cn3V}o6t
zHJKwe5<$7#YPI$h-X>b9m%40+4EqGcnV}#@26nQTy~9r1;~sViVGuxioB&~IP6q~d
zww&46Ycryh4-0LVf9!Euk|3<ZLVA$sZdcGc#=T@Q+GY~s9%k#)S7<Y=V8hKNX}|YP
z8XlGDDY&4p$dR!NEb$cO69;QB0^HTz4*1Y5lGRDSp}cXgyD~||{|0f|u$2ND>2fOH
z?dq&#RT!Dm;Im-=6kyMsM;_$*xEF|;w%b1cT#vx6hSzR$TUD$0fDjOG3%A<N8Krs<
z%vodAA#0roxq&UNZYj4A@Kvw-#VntKmg2&xrA}>5fW}O1A_&ynXMEF{2zMJ~ssW;=
zAvfmSh-;kIgeKE%dKXH3MKT01#iexJCP5jZt~DaVGXMKR{j48-)<s<tH%zE$6GOqB
zlpa$%VrA}cLU`Z^<G#;-bGhYC{7b?{l5T$KAW-DjDh>=F-J{!)OM>ta7^)vtl33RX
z2!l3Gmr|CojvL`9K66`7^iJaLgn&Z9cw={%WT)F}r|tUQry6IJsnZoOiqi32Ayvau
zG=Xoib4$@Gf6qTbJaxq9`af(7GZk++vRvt*aASpwZycJK^$MR8m@vE_;uZNmjdCLV
z>k!LL);|9mS)QF)^4f-k(oxx@Oec~UPmMfvav(Mcu)m++C^%xTh<OdypXmn!Qbc4f
zzo0KgcXiu8=(%_PzX1Z}19P>9_M)M(d4^EBYM_H+JNgbOXc}jli~O8rUS|nF<?)fg
zXK;Z2;+3Ak9uE!OPWh*@Jt|tJJcH<0L45lDD460ihZW=DT#jVJM(yN8vl^eTt<}nG
z`)rGJvUj~7H8yw^w-sKaPJ`J1e*3@3{k#o{C)V;bk%)y&6~LKut-$-57=J{iNreby
zH}?_#rZ-W27El@udQ#TKho}l%VIzc_0=yL@<Qv7uXO!~~^Dr(v43QP<jnbg7$p3f(
z-ohoV__gB;j<I(YQzYBg|CbAW4iL7&jZz9S?a09Mn%P}2HZBt~^?oasFSF265YB$N
zWn^?NCkSOx1X%sq=%&*e#R-8BkuR54@uDdb`(U>mVHg#NTvr2HvGuw)P5UiL;xox)
z(hRmk&FpsZ`B2GmfV{P(>Rx}RS%T=Jpj1nZ0e?GkmzvCADJ|hw9P8gPP-?&4G}7-e
zylP|axgGx#(Dqcl#kFI18EMwKRrSPQUtN6Lf`YEw1oYsUdt4&3T{I<<3*h1C75w}H
zqK1y;luPTsIN)fiGJl+F?Fw;ThVl|$+LrV()hE7i-dY}MFL07*b_65ETuZkFHY)FX
zptQakx;}Jdc&hMovy!YorRX8T@fUf%cZdb7x&{%G*m**>OM{7}K4DPs<W)f)CWdQP
z1f0vB0aSIn=RZXEv2qli1$jC0tMKbtI`E`tjQ4?wDJncaK<~3j#IF!u(lC5|DlpKD
z#V$N1$?C#^dpndN9zaDQ(^%QDtI~zKB!NQi8PTcDZxD7oi5yU)Ar}sxh|MVSHk2{L
z-JOe9W8mIl@XqnTkd84qY23xS<zeU!=MhK>SQU3qu@+8vK1b@G7g_zIEW@}27at{w
zjyV=1gQDX~9OMSm|0Y3`%jGCub4|~H)Ci@`BzU0>r4>~5S_2X}htwBl;NJEbsMZ82
zkE}hE&@uRq$x1~T7m%5R&nl^Wg+Yl7^3!lQ-8A8@zP{unAq1L8a2lhk!v?!gpewuy
zD8Prm-(tWjUrsi**WuhTa3!#<c$P6^h%gosirhw$E`C)W`>0K7Y{e1`=ua6lE!GvU
z6Rkr989a%qD~=zZ2`=Ch+8y3VOqA~`hnm&d-<<Wrow~<FL6t_8cC7jl)p(_dpiCc-
z*opOB6?JMfz{0E!r<N{IH2@Neb&VYXC>vq2%cBkKF2(ER#n92TqcVdr?5<~nN8_{d
zrlJpT2~zR%j4z!o?_$EwQ*)#=AJ=J|+@nJNzy9SZgJMxK=G)5PAW(U^92WEj`Oa>A
z#M}PMv>KJn{{=qsxIL@4lJ~OpG3L?KhE0N{*?z%G=R$U9#d+^efRPwVVxPN_st5@}
z3c$u7XfSmQ(eq?UJQ<d1Tec`7ouRtr%YT_*{{SiSzm9%N2uF+XELFE3A}<`sb;!SE
zAo~S=-c|LqO%Q#(G_(|oqw4GQ@b(-pf1=N*R0{8yOSj+Jd_M0lpY1bWN@C?SX71K{
z>FQz;t!Km+M}(*S3{!WOYrItw6&>6D3^?7nk-ZKXM*ZlXr|etvEy(yc6w&CbScy+r
zkO_nA2CeIyq3X4JG$W1g<h)_1$XE`ECxRCpl`8C*dph0Uh8=Cz<U-0D3b|OMSo@VQ
zSE7xLy(?r2d&%VST%5)<XH?6RY!^ENzeHP7S{kJuu2HT4M?UNG9L9EUw}?-t2PiE1
zuV%ytj<`(W2kUaFEplvY5lQH@S?0WMi3Vg!!T287qpC9JkG4C4^DKnYZ;y~tU8m+;
zZN793slNPYbic>2W|e4}{lkg94fXHTp295q$r8JS)=b^LLtg}t=}t=zEUv6(Zdv)j
zznR!>%=_jmA+V>%K7JHcIaA~*z)hOPO}aquwNn+5%M;N@VvHdlOMxDQU#mcFbzvB=
z(6WJzv%sd0YK<Keaa`719;N)hT|h1)kL|_IOL*lotTb(AIiija@OQ6C#^5qrf4BAI
zqKKiz6BSlWUv^$1LL`}!VZ9DrDx+TCqePEr4K^*d0d@42DGSCyxK=klz|<-lVgV6~
zIb5ij<g5w&-{m0Dev;t(Oqz5f#dT18iM6cq+QKdj?}g6`Qc}PR%q-s&Mn#4#SIh|j
z)p^FvVUZ}3R5rUD{2S$~iDr^iur5{pl1QN*dvbT$(nd<{`0`y-C!`YDvYcEqAHTKj
z*<gR<I&4@=EP*!xA@fKcz{74Zm|fNT9tQGD#ATD7?PIazpTfBxyKRN<0YZ&U(2$V8
zBa<%eY=9$!zeFj>!>GzdcOj5xfNaOoD@Iq6lY2wF7!ty#CU&HRbK=Y>wXl=XrJx{_
zjOfR*UnmBAqz>zfJ5lU8$_K8eZvCn0GbamTeot)8NX(h|FSa~uK>im>Twa`V<5S;3
zr#hHOw;)x;PNDQ2g8lkE6_sj4w?3QNz(-DA23YB85w;8NtF^91`*Q3W1~VI$FUja>
z9HiPG^Bc6!?wnMakZj^(!;ZKJSG6tY{+)LNe^7p<jkw?26rSNJt-jx|<L@J~D&JAz
zz=q8ozU1U@p%|tA06zi(KA5_H7qvx-bn>?iGf2BzL;V~kor6)a=)tl}8_psXyL{jg
zhOsnRd>|c#4oe-QJXq*h0q|-34c>C#?@l&~HHQW$UIiMg&&f7K0pklgA3ln)x1Wy)
zm3bMGuRd68eF4cYMG@OXo}v9CBV*Mib9|F<k)7FYB_oXzfF(1tgnx2SrfY%0=372N
z^BwpdU(xOIOc?~h;VBE8p7&~bmj=cJQ8t9;+Qj2-E}YD#RGZLUMxm-~FVA}lUKqPC
z8W()BB-i!Kuqs;B8WYlLxYYSQZLRIbQs%Cz*BfTDS;FQS+E*_wv-9UwgA@K2dHyi)
z!Pqny(4_jE02OQzHdXDnlU1j3m8=S<S9nO(w(=C4v@=k(0IyA&oN{OMUIbZ~^ClMt
zfFC!$%jY_G_a9|jJY1-v1LKE(WpsZ=D8{q&xWhmCsGffMcmRj;42ua~HFlf~<Da@y
zm@jK13i_tl|Ncxa()<Y=_>ha!&)a#c{^KT|@E2U*0|4}6y_8uPe{@Q3O5o1WB)lhG
zs8vqzhpr_79&Tsar3?@bU^!Qm#Wo=h(NAvC9l!rn)$}Db`3+~?t6HE$jK_)3=oWMc
zm>m*%YV3~G;JIv!1xLOa?89MpQZON~H>{)h_$%>`4|kVhr__H7=lE%k;(2USh|Am)
zmFq7DJfO66_hOMkjpLOlWNbz5xzcdpa|s&RC0~%!OX9~$OCD=hcih^K>stD7!H93J
zhJZZ=Y~3to1?4XgNF{VKug){aEF8%OHe&@Yb5KDt@X`1#*{I<>!NVsuu8ga6QTgZG
zDRrr+id-06`Yj9Dew0^?jGV^FNsEw>AR2TTCV(^8L~&(9&kO)-Wo#y=7?P*JY4;qS
z|FbVDAHT6}IqNPhVrE>gwhZJ`l*GtJ_6$7AX(>Xl|LS6T94>-Iw*}P0yz2#AiHhnn
zG@WP;uLcXw`*^GW-bHP@q8SWZ7$R1<&%<#zpLi${?CDP#YP1y^6?%w3DV<*16xr?;
z2c%jUZ-ks@@h_x0Afin)>{Q{K+V+#jtdd1TC&Vi_l}Yq~=dcuHeBCA#;lIVuc{9$b
z&z=l3c<1o4*d`32pW{4$za4|4{{En@y(zEU)#&soM0X)P@68%Z5vcj4W-u+cWFnGZ
z%|jCC;pYo=QPq*W`@)f{i>l0G+no6t0idQ?8oo`BKQY}a`BNr;QmKOxOo;L^o_lle
zUMTtn#TI{ut=9*)kOz{n=%~{H&LrzWcYr1zuB-Jqh$e!N^$O8rSXYn?sk^{O5XRi#
zmhzWCtNHm1?~J;|ZL89zEsFl`S&Jc8rphzh3g3paiwOpYCD^q|TR<E<r5NOc2T&`{
z`vgxZj$=6Z#J+DEO&%U%z%pagVMe6PED@m*eJ*6$&DTp@MJVt3i{h)?9D%ZdWEKh$
zEDSP^Lf`!8Ado}HRH(qM?gUM!mnmN>@pRUI4s8}GUErE4t;y*skDtaHL6F?rYdR#s
z{Mhf~f&?)}^YUq?BeG_mJQutA8({0LsDAn@o14`nYDzRPBuJ<Ehlye#X%iNhg6c<_
zgfc^d%0NA%rCinG9540MwN?kqSa8z6oq>d?h<3j#F?1JI;5g0CN4(J$z|t6y5rdO4
zo$21j67Ko&Kp=2%<aX1j{=x8-IP+)v0MhCXoujY&{pV-pwwrAQe$k6tfck~sP{+nF
zFU-H|CW#6k8k4HcTUJ^6c$}1(LMt^c`bd&o+-cDB?T=uZ+K|V37L2fmPaMZypw>T?
z7LBVF#Tm-h%N)*!;^UjiE&RlW!fzS}2c;8g%If7L!sC>`c}4wF5Lavik2z$<ErSoF
z44L>SvOQIhsE&X-Z2F050r^LxSaZ<H3(_9Bs^dos(4P*P^r3LouURm1^r^AIq4RYz
z!}%BH@m~kbw^SuqO|Lf0Wn7xBoeF~ba+T53pRV2sZ8#ob@J#5Vz=(|Ky_$r!o{Z>-
zD~Lw4gs2@qPk{=UFku6g2DF5<3)62@&|~f-ZmQ^ckY$$mEAsFF?tR6?Jo0U%_gg_W
zUsPjoCvb_=`Z*Y6CRe8rV$t)D)=6Pn>qHiqy$ugDqQ53-TD_AXX6WigzgS@2RtnR!
z;{1Zsw0=JFGX%(Wv`<hXGi^&z13(Up)WZqVezaXQqI>)Y>ajaUZXq~LRv$%tU12T*
zNRtFVAhF_B3D7-S>a#BWa6J(EQ>CgE=S<-e?e$*dr*xA)1pSjHO*F8lb|BH`A>o@G
z?cwjvN$)4>dldB>5phH;ZlC46uh}4hyD^EZhyGamq)5MZQIrD=gomxB2?h3kHjJC;
z#3D%FN@@*zEq#fb|GW__q45v;Ee{X1X8)aZj_h6~7Xac@`8-)#=JjxlRb<^#cboFw
z-<drQs3`&w76v|!FVGp-Nmi8E763tYkgs}(U2))AJ_gz0tqj>Xwm8A@mFE@nPWEjg
zUn)?*hZy+#tEfGQPuE0adX+7k4nj%j;&CFCN4(<UXaX`B`*(LEUG&XKKloM`!P;_9
zIYdvNG1p7Tbvs1Y&2~eaULbki7(r6|p2Y#*bsAYc`0_A7s&B}2?RZt_sca3ooF}J;
z%d9MLe$**Q&F&;M6&ZvL_$74;83Y?ZGc}Ots>cZyl;Y9=^}~EIl}Rqbq7)?oQ5{kY
z{Ci4`I#*C8yz?$%R;cYf)Xleu;?c-X$mg*$Ec<CM`BjpK`a`^0J<Ise)P;j6u9xAf
za)*K_L8AfTjkQ8{R1WQ*%p}+T9zv_~sn?*quHGoE^-;Z`gkvR>DjEDu;J6C_ddRq}
zJ27%+C&|m4?a&29H0u0&qNm*anR&}t8R}B}Cx}u~8RqLf3=A&wi)+~yyB9LoVo<YQ
z2lH2T-CxW#Ry04LkhpnN+?=b~et55QbdmwQu9n)}P_q(r;{}E+O!sTKk?gChP!jpW
z9^KF1%b2x8{fk!GC(!eG=5r<hmS12fLrPALS+b^HK16WuCU=vu|IGprSXJhV5TPVQ
z5aa~?DO%bDp7TQF_Y^RHP(5<}z6e0GzGeP$F($duld3Dbz{B`-DH9bRbNN^jB^0`g
z7)$4QZ1!BF8jq7k_tzC#<l$x1Bpx)fZh`t-^+<A|IA3$R+G`@(q6Oe*L!DNQj))`M
zAr;W$#?>N|bNRQK(nFGs3Bu5j-B!xUAEw&AbwS_Os_h=bo?^GBx`cn@#6?rEnwvX8
z$km`ArSW#MC8l9n+i#c8S*h_l?FNt4uEmq2$iCXvVYQ{N`#U}o=l*X0_;~*u6zb{9
z9{LRV==>$}`e{kx=p4W<GGi&?>3O9ca+pT`J<OT?GfukdTC1jM?lT-?JkZ`F)G%4*
zV%w8HhPR--{7T|VBF{!ar-m$6;;NZCmUKtFXDF$+!&*e{0Ts~Y`1N~<I0^mb|5be-
z!`CZ>K0jLr_}y};*w+5wU~tQlLSa3BCW8_h3@ZT)2p3~6#vRo-2@Sy4&WZwT4vK}7
z$CWAvxWeNQ5s(u@ZcI!rL{90N*|Fp93Yc<uFRn?L=;mKEC073#WAmf+wBvJ@udKV=
zJqlVx2ZztrZyQt)s3BCSO*1f2Do<e;vZoyH_?Gd&gZcumtD%k|Fu(E5f)Kf5O|L-$
z%u4`1V`YqcujzA2C#w2<8(k%nB(|H@+8(u3--B|laq5I-PEsLVsGE!^XmwIv#v8@9
z1LQ?>{~RE3_&Iv}eQvk#epEbx1M<GspEYrn*7NGz(H#>y_g)A0@_V`O8-a)K)P?N6
znWP@k*~Ix(<aW?|C9@k->Lr=NnDeDnzjXjJx5gk0WX~n3lRNInwJNo28J*l9L`cnq
z`9DfoippjeG{{%uQ3rpV%WNaP!mD8W*A2E}-UkTRk!CXd;>iD5u%g^5abz(l0vZD&
zYh+yESco|hRD!<*e@P|*{YYY!z!bq;HNiOFEqlO*ftYh+CKCklX}#1p7D~Ie0hqV$
z(!=mqiQ6!BH)}(8K=j)kDW+98)OnYqWIQL%^I=K8kZX>ZzR*hFQki;pD$W3cHT9Xi
z0b+j`B%MMuvsChz0kl>wDe9*FO^QSbtg5%*g3hk%?P<+DCyfWm!_DVC#>mHl3~gA5
zUY-vXiQR8A3Y)3Bxb^nt76sWRKvCU`hRf+7f=KGCeD(_VD{ZHx25ArS_?M!;auRS7
zu%ij#Hp>IJaUH^P)rx#Zc{BcYTpDCW!{CzIBqL4*C|>?cPT8^cvwqh2{X0i*r{bbo
zStV)HilTZtisP*t0sMG`Q)?h|z0v4VZ7gR%p#bXWPS|zQWSt?@CKMzD0ELcyzrU`e
zRs4gFmoQS?T1Wgs4c;|E)pitY#1hre-Cy3>4i}6`AM>^SI^M<^dX8$c)W50VhMv=S
zmP~gG8oE4Sj!ZVV0**`V;awC{K2IbJey>vBoi7wI@wfmhGs0WzkFPM3)`n$lMYR=u
zur<vYPZG#kaUOLCs%%F&pfuQnp@5s0-vSSc`_`oa1hIqu&DTrd@ny2Vr~*F5O;`-^
zO$sl@_i22$Vh&*iAGKbpGi=L6RI!t^k$6ne%PuUAZ`+OZecQ;1Y_IHj2rb;Qk7=ij
z_<PcbmG)S*NuQ3U+%;7?S*3uJ;9+~QNFi)7;PLMIIE58$9#PR5P?G(xw&U`jP7hmq
z^iTbXjH5-*;vTTIC-+s0O@Q{=#*__42r05+{~enn4UxY!oRx4?D%fi-MB9l5Q+JU7
z@T28b9R$r_i&v_~OGcYklm#oUp<~bD=(`W)@BoBs15Xe%i$b!V>0CA&0*h`>KG>Nb
z!2Lgc4&eU?bSSTo5D@31uhsXiKQnDoDGLHR9ugWbWcrlztwh#4%%i8et$M<&0ng!V
zAeF#o5<1M@g_{*C*l$;)$ce`&sSJ5`rU&A(rEZSf+slx#^NymruBdYe@>4f{O-R#9
z$Sis&4c2#~1zWyjm*K+w@FnK+n|xFpu>koPo;Eh(Kb!LB&s;D%gW&h9yV^!C#Rgae
z_bIlhDaq%s4V6Bc0<<u!q+B_AAH*!@rlQrMdb`Mnc0)&rcJ0oBS9%7E+nv0@V>T()
zQ{oj{JM3Kz2GL#+bmj76=EhQNuovmi!9&IcVIE!7a@>@BcljiP1Otsxr3FGS-T?Ib
zi2gye>{vyxo-|9k;(5wksKG`>VPVbgY-8#pTf-bqKRrVkrE(4GYI$6*9~$fVJvFZD
z4benOhElv@3O7qk3^jypmThUYMxn&M!l=kZBiZ2ceUzMA7%6>&?j5ed&ZL2{tVqlF
z&lr7<!<T8+_yaSo!x+LBu`~k{CqNfd&QNBrF+QV=y3X5E#+eaL6TAHSz^J$mg={f1
zRh8?!0z9;zTV8HDd=sry9Qvp+LjpzPSL_#*Rht@9@X7mjk?kRqo8idfte0+Jfk4n#
z90KBFs3^HhPA&?P1jVoHBGuNq3Hby?-WA#b5cE6nQ|Ix^9h16wP$9i?SAf+Nt>9m+
z`{lC7Y^{l8+VC=(h61*wVfv#GbH@PR;KHp9{o$lkd&l6YqBJf99KE;J48`ge8Zl7|
z>upjzTIp213&#np%lt_<twL28+1+(2wA_}&NZ}Gy<^+J_9+Eu)Ay3PK{=rUj9bR&K
zU?~pI!AkO}+m>UtZ4z&h902>re`H{|<qFznHt5H)`ao7Lt*GWgebJ-e1k8R~5a098
zoJ=7*a)N7z-5Shh>2IP!^u`Y6$OcCBaP>UIwn(fCE6ROc#s>WGjom>Zh~Cr2$~atG
zdpFs5&yQToUcRRnh}@)ew$bwo*c0*;uPP+9u@s50$Q53hYLcN>6|iw#yuMN$RpnBV
zj8a3n4WDhR1}JJX#rK<*F94hHuiyQE@H=m#P{xM_6u8bfS=7h}(_CdJt1HST&{p>F
z?;F^j>AOr*MWkaHcTakc?VEJh&WhSOczGs*WaZCX*5<^3;EDwtXcr3iPwrPq$*?fA
zy$pRC<%s=QBouGm07Q)~dH-b%`8Q6Sl*fFG^MXw<F2)b|fHV^iT``P>%w|uztf6un
z7g%8yOASPjp{111w^*8HhFRXzL#o4W&Eoa~u25>8v<s>9@hMo`hJXLh2@>UodEG(Q
zKkv^tA-gMP9!Pn3(~DF_@&U?Sf@;1g+|hf?IOI`?pw<Lnk`maxlLX!iI2!#~yY%KJ
zjgs?5pls27r@8SijyYi)@Iwz%xZ;@^Y;`RETg|`uXym{Ll34?vaJVL*{n+llSt$LT
zLtT9C^~&T7+M|0;@5<r58{Sj)t8R#ug<aWp<e@$lU(X7!ZbhaN_)m%8AKj1cQto!w
z86#)~F<=9LD>m}bKD*7dy5v6>t-1EitZm!2T&d57YyRzt&^I;6z0rYXy!kCzxtoZT
zZCkRgWGF`4+lq`zfpU7X>%wb95e_OPk47ksQ|Fr$R6#bBg#f9GoD&7YFp>(9|G59i
z!KN?`8Mcl-M>V(0+nWV~A6tge!go=rSmlL|bUOe&eYI+gHXgpb5jr`GvmK?tU5sFe
z5eP1G{ZqbGk2|6vtme-$T6$Naa2*CYJ_11|L6*g>(2AmCkAkCEMqj)Oq4Y>I%Cwfy
zw;6zii!uPR?I7P}Lh8c2Eg&nv+Wk;v10vge*JoP>8<y%?u8B0q5rLrzu-$xw0c}X-
zxCd}n;Hy?r<#WSD<Z(i;hgQfySua~tOsP^p*oEbplM^og8JqiVxO_~psG>$6Lm<ZM
z!A*^+zr$1B9UCt<>dS9|a(fG14YY+rGJ>i@B|0`k{VuAzH_$i)S%~<h;;4BTB_SC4
z5Nee<IhT7gLehFGbF#!6B{!y-v@|SDeFNwwftELcsaOY<;Es=HGPn@x)@b_eDpY{p
zx+Z91;?NkB;|AOGCQ-)0tK~a9m3e5h)`V87XJjZ7A0Zdq7R#>oLDoQ#T|6C!LCM}h
z&Q241A>JqS6X{)i#uG-qS<dHj#o6Qn?4o1PA{&aYO0NyEmtcv7OUDESl=+nT0KDw3
zt{pDeU~++*ZsDGigK?$8?U0xKRFLF=J%#(#X=l(0Fzfvu-R|G0uW78<L{f*uv=bbT
ziEr>Uh%9Y~<w@s;dS^v0A$EbN-SZ)2^BX^)04?70U<mdy4!G{_GmAhhfzB3(ok(Nl
zzjNb@Qd@u)EZxyUwVP}HlBO6y5&Vegt@^G*xCZ<1W%wxDsUd83c~9t#b197+M&dp4
zt=wy?Q<b0HjzL=`Wan>gl1f@jLqD`E{Tn31`uEKC3L`}fspWSvwwE#JGiHW8Kz7yM
zfcC}dWY`DuW~y@wFGd(O?{A!@)aun%=C`jJ*)NhfAY-Yr2<OXZOrpPl`unk&@$stV
z3UiRXO<U&|J8p#<YYGbDx$>IgyDDA#rfu#l55o>GC0y(k?Nmi%Q*L)cBmxdjXrl(^
zOWFRud&ZAT=*k;NSJM=|%BFkBi7LWmW>t>BPSY5j3u<ej{X@!j@VW`y33iQqkb(N6
zR2ynLupCo`QFe`G;1y=TNE4f>-rQzF!Td$<KRV0tD8X`*7*iZ8UDEwkZsem?=rMp!
z4;Hd?aZu?tiDj~_hA4bUjX)fRbRll$B9DzoU?X(v9T~Q^5U~((;)%I#2Q^MJlhCW+
zPaq?AKV{AndMbzKaC&dH-{HKj5x=eBKvTj~Vxx!Q&X(o<83Z_(v8#OwHp7u3clvcj
zp0SJR5mpdZ81OD6jZe9F5hnf@mg9%&#p#48|3!Jm_*I3y5aWNbfI0~WzMRjSSlWyI
z7){6;(7i^~MWzPF7e*aM-3nnTeS|(j3a?H%#6xjs+>;iSo&};Tr#GUKnFr(ixaH0P
z8hin;b#3$wQfTnjMme)kj>}Lml(S8JH-lbUWFKBKvOeH52|&Pw(Q&ci(KCV#6NC7D
z7pAKPCUA#P4+?tBdh*i#&(52JKE-#nsSqTiWrU8gEvA)D;fUbEGamX|?%`Q~O`JX8
zKDVgXf(cq$)PBO)+0maB^iFMxWJJ^_z~MK5;!=YblQ2{9A=-tNs9BE7aN#8zLm2kG
zEml(hHrtl8R=3*!u((X~+nd9SFL1SttyVtxpsmHdE_@SYMr1_-mYNuC#8JAiT7@J`
zI};9+RMdF$$j?B}_{Cxx(<Os&hH%FEkjoF<F!-bX!NChACVVm`oc(aDF1_YN^T7ka
zezy+j()XX3?Gwahml{p|+i-5gY{JR1d`Iz76-I4>9)gFou+xI|MGt?-Ztdqh8WEmV
zUD_%0zo>f0Aj#S;ShQ{1wvB1qwmGe7+ts#h+s3qQ+xE2W+wXVIjT<-Cj~(@+DxzXn
z?I$zW%9WwQ=|m?-mo+jg)kT|6nnIX*w-g8(gB%>lxM?8t0+U}GZSvM~ouLyy(%KW@
z<X7MtVD0$iZ4FSm^aEU`=&sd^8=l>c0Qw%85YJnNW2ukesC#+IRk20(QJWFwYvwa0
z!~VUgp9zimdvJA>I^nG09*Ml%KRKd@NjTpLyy@7x>6+zuudn0?+|_2KPAeiBFxqPq
z1|)J#+%UJX8wmJ}#HVdGSmN|J>~x^^K)i{blcbYKsUgWsBD#_RhD{DMV8v+y0N@np
z6m@RH3ksc<qc{mY7!J<cPc#_JHQNjtDm*68<mdT`JqCKopx<X5?bMc0v<N4HKdZ=p
zkIRM)u*C6}lf~kPg2WJ{G04T0Jvhnj+e3lDgUPMJ<#T-iGxM-9!#b+Wch+!afdccg
z0n5gfxuae~GMMjQl2l!`)8dVM0CGC+Sf?*(=+N(%F>xpQI4t7f=o3)AcUKe6XQ8+`
zYSD3sCrAyLD-|q)@G$@18xF}PjFBXN>IVBH0rg%8?_^Dc_7?1-h-(#cjMVyJTUn!@
zUsA5iZ@Fd9;ERV~-1-KimufBGQzRZ~1#|qoU>quyy1`h@L5=dWV?PaPa%O_$de#1-
zX5o*r_Qo*MXQ9V?-loK{!QlQYf>Ec8VY9j&d#&K=5z3Qk!S`dqJ~8!s0ZPW;s=jee
z=Hsy(H9<JA9kpKJ&6frKDlLN7fD7?soK`UN+r{XsO(0AlHd6T&ub^MU!n)LXBc7%o
z%2At=9DA&GH}B>>KwYBM?Yq!f>H2yKfvT>`$$q-)8p4FaZvL=^$(S(^H5mE6dSwzF
zqNKzw83PN#dJI8F9n751iOBAt3-l*U-ex&6;==<sY!6)dA+8`?zj}8P4tTZW*P#fm
zr&qG2f>#7~$@D6TyCU0LL0sK<#FkrF?M1ZbnW*duDfTB0;4lAvUb=KKSivZ&L|Nzd
z7IF+HkBsGK(|QGmN&LKYao->GVkqQJm$9b>C&N&4QI({{G$;qBwu>WWv>H*<&qLb0
z@CW+<Z&bNl@tf;EjT;{Y4@ot>cn2Hc;{X9RyE#2fpDP+Ual{>txGAa}*ok7OkNTo%
zR}jjm*Kt8BfC^0yxN>IPdeqIs#8L){g%H@_K^$0g=m|^X3XhH&J+<^FkOx8f3&MX&
z!_A#7h}nR~S)O_~*U+?k7dJ)>Ux)eVLDC``eM74^HQu!x9*opmgRFIM$XkWf8@SPO
zn?Ym5VW}BUOYb|dGF4%my{8s#v5fL&1@SL*$0%z+zo%`3D9v^4t+f^CVH9+RwwVDX
zswu)VmG4+B5TgJ0KC0Iqcqmk8Ru^W-idNkC1Cgx9vgOr|izMG?vN#iC%LUR5b!d|Y
zSv1;<OGjmsr`wjm?26g-wKH|iMzxtjyIJJ?dk4v6cT~~bTuX(Wulv!Pe7i(;2Z6G$
z-Oe-MYR1Erv$Kvsv)#NQVCj*rj{#HNh|#LGIFhZu4inr<N%?p8zWT5D1kx-Uzb>jZ
zk8aDr@o$MR3Ch26+qun<?8#afwH8J8nO9!N`np})SQ<3Nnc@ZPs?3F+qFJdzwnMa-
zA%3H(*O8Wd_2QYJQzR884AHC`u^YfgNbgMmj^2B>3Q?u47`Y>)zsn~$_^x{gDlnFy
zWKtInP#rgREI^q(b+5GOKny<C1Gdy<?yK}ZC*z%Nt*T}0nvZZkdJa%kTb8hDC|~{J
zWY(hH6Wx9PU8T`KMyj2kPO-0iNKuprP=JWLy&bZA27YI?r<T#n(FZ&Gu|EGissY7v
zs)@?K(^=nYxkb;@%Di#oirZffDWtn|AvTrPD~pG<)H(~27K^rY2pt5SWJL@fp<_q!
z7_P-3R=<Lhj5ZM`Uw&m_$3`7oxa>?>7Uoky_i;ftIAXk-&q>^YD?8DWo(|@V<v!9!
zpI;;o-x=RiZ6Y)R=@_cpqzWYmr~p)}{FJasVb@y%>mf|K6tp#Yi!VeWb>Jz8c(Oor
z=mV=mk|z357_e?;ACUuLepW9yhcgplV0#xs+>DdFgaeS2Yx~XS5GX=Z@>C-SE=}0_
zGW=DYkYwkIhAd=oqt7M`Xe}MK^!j7d27KVOB5wV$6Pt~$(lvb&)s*z|^a0O8uf0`A
zj-(~TB);fFzD&2on3fZuoZH6XplIkkBoa+K@F}i!E;fR*fNj0%*2OVa%4B;fUE5{>
zxAm?kJqa`z^`YMjMaNMh*2#Irck4}^Afw|YWR9~PB4$tGc#&*PBpV4^703aOj*gz5
zDMiC?aU%^8r|%jarxvIUivU<7T#F90)+EP9X9~P+l?f$}3(_bg`%YF-clVKh4=+KQ
z85#HFMg!OhqSW&AYQZtc2H;Mk^5YeQElRnHnhmJz5YZ6$EP@$(X`|I$jc#bUjftM+
zWcti5R{Ks%W!ZY{U?_w*LLLdMeKEfn=2y7izTsI8`UJhx(|0NRV*qiUf0#%RN(n4E
zdZfPtPbg)Pd5j|T>82dQgwWxYM^^hQL<qNC>D0TWgB2);N3j{<H#hXVc|XD^3>-*R
z2wvwy+WBdnhL>q&hVhxV@h-DsB2fug2&_;L2%<1^uDJhu2w#$<#cV)e1A)v5^|J0e
zq0N4G5`yIs^v4rp0iYtRXS!bcvFh{CmfwGta`C~QLhMk13ce(={}G!YK*0pFj|U<r
zkWSGfyNT-+Sszk>ctk>g?^8s&BGL1U9fYn<!hKxGbl;^fj6M8ua|-VA&tr7YTD}&o
zwIP^z-{7A;tn;5g8Z!)pSvg@9FK|fpa3!_!@D*=ZZWk<X0h$Arn_WI^Z*<nPm-SA%
zeRfoSRSTpv{1d+Q4xP`G!yy3UyO32V+0n5mY$A8bXPYQx{|zm@J!HCkebC{qBmX5h
zLT8hQ9ad|b;}U&^8ub2a555Qz)HKoD*a%-P%;o^ui05D$mKfp_uT5*$%z^V_OAsi>
zT>ZADA@w;a1Aqy|cdx|C?dqUG3+KXgf$nDy{0t&&z2L~jC3?%yO6R7<he8-(>ELc=
zm;)zFdk;DodXT?p;F2*)f_k40=c+nbM@^MVf*>(2{5O^Qcz?DZHQ|6q1aq<L+uYV7
z9W`G^%3>f(;3Z<N!C^br9>=Btd$SA!No@ne1VbQ=2EZ_~IZ0=-Jl>{?EmqfCU@1j5
z$xNQa_@N^H>L@!_6bYmJovcM4D!Wqh&-Ogo>eF;=>y{xxv|0z(j`I?Cvfp+?q{LKr
zi#;Cbt5WytPf+D}4Mv>Ku%rKKWsm*!qrl7cvCQ1^7y6o5s)YBWXJtcH?=wS0BUc_g
zQ8t`^C7={u)wr(FWZ%;QGfbtSZ0xIFO_#t&7=6|ZqyY_W1tCHP?`a+@O)kwhQMD#h
z6PiD07^L2T0Eh9OM%6H%O%csMc)l0@Y7z1Ryj7FkA+LVlF#P_M2UCDUzOq`{q6JJ<
zYdIZZC>EM(&^eN@Z)K*cGngxUzTx>`KAFYq0N`slykfG5i>fLH@p?w2$699ESRz&u
zjOV$Pan^~6r2?{S!9g4nsLy8pk22el!+|~M``GWNdn=a?+g#~TvqZCW44p?!Q+QS#
zD~V+C3UaVCh3P(OR7$TTD%z#S1+&Tx_m8fqd#!e7meDrE9TpGW7bpryoFU3J`w0Q}
z0&tN}WK7uwg!Tala(DOQ(DHTfTO(#-)l8i!)<=0ocLXy><ZRw=2x}|_!c_^VzqB!-
zkxVf4Mokq#))Kvns&!UD%9e>g4~SpjEmzOjk5$+T7$m}#N!EU($VVcICN+8Aw}ffJ
zKF)VY#G0u>`$}GSZ)J&Dho8bil`;9<0Pr?9#|<*i`H-2~?K-+IQ&8n?gr1kX4>$jF
zvUPMuPv=yroV&QJT1BeEUV0W|Q>BVi>YY4@;5_7RIIQ7TtHLP(dk;y(;y(&Tl6Xjh
zdXuIOjgic8lh$hj5^Fi4hOuGtHjD$)8O83Bm1&?e{_6<fZi|i?`H1n?S<g0m6u<;N
zjr%-IZc?Vz68LhP<q_#)9)-`w5*}^8<4vKYQ&+_`xJ2=gN-10tMw5p3uB>lw+~Apr
zDYv54p1#sY4k}?tjy&>nHb{u4x_t#d;*bXY!>kogw#s^x7VF#wyJqb4i*}hy?-mOl
zI-44^ByHL<y2%DDuUy-1(KFDH0kHc{3sM<LRTGt~(y!Ds=mhuB(2b}*tK>@KlHe<p
zzR@bz_jwcH<Xg}6+d?B;;fTU&+_=WXTfH8RXR2p`2WOCR{12SnE4rj*8pEoXzw`^l
zG>GB<8>!#bi|7~jgv?4<L6tHlHv2S;!%Qijqgyace!O-q;{YZev^NDnK;@x%69t7L
zpkaRaozGk9d%jwzxNRRo$9Cv1%tly~eRhxiP@Xm(6G{5UHFI-Gb_b-iDfP=mr=|uu
z=-2YKnCyEs8Q`v{aIsnNH}<JpE(a;CfAI9n6WE8>D#rgHakNZx+U#|%T}xL3_`eg(
zk!UG3jzkwEq)_m;qYePxGufOdznxA9FR%1YBInP#a@Y2;uxI{`PDm_vUeZim2*lla
zy?47Ah#J@>A9n@(m440-y!u7&-|_cy)=rG7&@(M_uS&E|CJOlR>Hnjr!i~wMs-hto
z)q*c=8eVOen%h_&Fxu=!ISn)o@x*f4MExccUx}ES0GKYxT!GY`<|3#;d9M1%NwA_i
zrGH<z0OTb52F$Bx>&?u#0B)g#@oMG9c{ITp%*`qB^UOxaGn;+qqrdv)DJg7b3BPY|
z7dpzp^ij62tiR=xt&WO7o)G`-Y^z$2gF5$p`6T^(K<}6TC+ggnP=1ny>~?dXK@@+C
zZgIjD;IHZX8L7?XErd$uA%)4^A~MY1$A|F#E*a7kmK8V4>tO9ZI*V&I1XSn~tvmQ3
zm+e%0mntzkR=cjvh~cs_ar;iLKU}N?<(!|VO4Bot26*j|`i?z}2u$>O0;e+1A2(AX
zp<h=kUx&V>*GJ!SJ)Zfk@b8lR&l13&ITAg@fJiRZiP1ToJHRpLV9yw4YEgZ6Cdk;G
zTJ^T1v)h@7Y7K&mhcV6zNHvoJ9!%Kp^b}KFZkjyiKG6OGV*<oCA=1P?pM{K@uM&l6
zVMW5Cn>-R9rM7f4^Kw)4kHvYoQj4@Cp+ET^YV>NAAogaN+tdZz<$qvT$4jHl?rFq|
z0rl4eh6H^D^DW(wX;GUW7KltrSXa+L2TCG$KE1luO&P_F%WEGbrerBeU9J4aipbO=
z%ck9GNjpf1<kI#m^fa>kBk&i&?X#;UP9=!5f{m6)3U$GC8^lrMWNT?$%XE{Nly`|`
ztWO(e&=IzFcVDjH#;rA@Pk}7_5YJi|0K;hym6}v-61;)!hO6~L%AZt9Ow2hP_Ytv+
zjfl~16EcYet+WxuR!QHCwTTph+k-+0Tx*pA<IE_VeHF><dNa&4)KU~%ugxcJ(Uq^@
zcIfk-0)N_Uo(Ru}ZsoMHK%?;(BbXzHveK*aY9xYC_jA~HNUvk&2lYsGgh~8QfFyuD
z-5y2Loax(+3%WITohj5;$Xfe$lthK^*+~S%wY(3*y45_0gu8WPZ5YTE&luafHR;`T
zs0ZK2(9ktY8Xm@+1qf+4GFVEG2VLhMp^P$zAtr)Mycp*N9r!=5zHW^cYW^ml+Qp~N
zULZ+xpyTA^T>}s@_Z0XzAjw@xzzsx{cnHpRkW3gmklb|)JC`fbY8aWO>N?S?pCM=f
zd>s&yH4qumDiE1<bWkgk$$kj7EtA8825<V8tLpB6r52sC;C)?@-_?de_jZe&CK$@R
zy}K7%Hal)~in~QZp)eUVTRN-xuX_neBu_%zPP4?U0W%bX$x!J!5_1@K0At-zoOX=S
z{r7QcM$hLL>9@J>o(+stYl2{5%#Q4e(<;u|>qpOLEkoTa#K(H>!wT<4fAk#)C>9zh
z^`8LT!(GZ&%~)S&@j(7s>#S-LPl^f=eDSJ^6860PzBeP*v6-`s%Q~Jj*kdjXq`q^k
z^5dYpDI*`!m&`#dZj<R)fH#2=jf?FW;_px$hSBBBElI|IWzaN7;-3!?|Hd;ZkR(pa
zc|D;Vh6G}L+XX{@cOPp0wMzW($ifWWY=a~kLlXv%{R``Nywvzgh!sSJR;NJ~i-WbO
z@;CdK-fByab~HO3jbHS%qtDLLP1`N|{mDn+tm$j@UN=82UrDb-9FQfkW7I!MF#uOV
zcOe@}%=#Zr2%T)Bb^9N7?hiuhJeVw9cg++<<vA#x9P;|l(+vWRf}x1HHARa-jlMaW
zO{4i-;{x+krEjN}f?nmOl(j^qNN%OXv@Z7!(FhkT7k=$R$nnsK9CP8&Dit#5`Q+hi
zSV{p570v||4Ulg3c15Yw-hJo7l<2uz-Ygdqks-r?)ua1i+bpGQs;8M8M$F)>-aIRp
zLr`ecd_#xz$c^h^<08H`u9>Z^<2LW#oL-k52v$FI#I><hx6tf5>*$D0gjRVb;a|pW
zdrj&@u??;@bW~Av+${!)N$iBXv{HTCiq_aJ-X(7;fCktJR9jS(QB{$sNIjjnjN4%t
zqc5;v?qmqdRw4KP9@ru5yh%24l{;nsH$J~6*;RANlND&^toMMcZ2@ee8Fz38RcBCt
zn%vt1x4RIyjsR{GEmZPJ52Gr{RPdkY8sy!fiO{&;QxlsW3h_);roc3%YTFi&c?}w{
z_=2OgfE)#Zo(%jnK@e(GRZ~uX)#~K=A&ApNH?p{^U#)-}(Yb^G>O9iLcP|Y~{hh*N
zfsD(I<E82br!B=Ots>b0OHx~+NU)ml(+V<BEix!`AgH+4Q3NEGhH`i-Vpxk&?^76Q
zlW(TEp97zK*iX~+20eBZU2FvPdzZl#oW>{sFpG5H-<a7bxfLhKe}Nx*{R`2cCW+y1
zN4hx`!Bz4f&Pe8&qH8FU_zS)5#t+=N1n&~#)DgWYRiZakla~Do{aSse7XM6S2v|Fx
z!MuG^p^7}gunsC=9HO99*QST`Pc!`wlr(5xk`H|2uFC_=RiUi>l15VQoVR=fl`La(
zfUL`O>E5J!(+|TW*t+&uoKl*5JyTR`GNRKgpvC4Sz#<;GPqvqJ3*ca(AN==VIE-*p
z1qU^e9`Y4YQeyMxpJ|a9QPk=QX6nN!6uhiy8iFTtHKl2T8Od<3_gcFgwuR#uTl=}R
zX}nvx7KYaK)O&diH#b8ExGMin;qH(W;8o{6$p<wlL_S5)U>z={FXX4?E{xc{rsQ&$
z;zGtqwk8)p2!+P25Ep5NTBuC+Y+$#+mqV;{<uxjow%Mr8cMmG^{Lwq>fg$~C*HG~z
zHq*U;Y}hv2@=l=ng=<beL8=U8quw!+;VIvFJZp{D^tX|9Of~QyD{ib=sOEcnfLG+3
zD5pVNqqqR;hFO2J>^utR^kF<kiB9|1rT|gVq&yU4nBM*5?4vdd(M^`>oXvU@hsM(3
zXAR=Oc)%Lip1o7sDM<U-+CTu<=>(PJM^Tt&6b}L%777rF)B^G;Z5I=2{{<_&D|nJ5
znd?g(EP#%J2t2MIi76LfMR4REK<sJl3f&?O{F5XMCA1VJXRhVzjs4rxs4;0ofoPd%
zIo@zDF^u6wupp2PXjfgF*rfh(NjR9$0(F)mK5b7&RBR~=+uL(%gSHobH{{?rVd)h5
zLR6ef7qb)T8Od8F&LqwRAy#G>_Vth}isrvp9!F^!%oCKO`KJD4KUN|_P0{1r9upE~
zfe6Sj;6YIm@s&4z;|~L1X1_G{uyn#Mp>4a+kW|5=2&;1n6IO%4%uJNeTTE|A3~LW#
zT0SCjDSfj3#dHzxVzzJmCKL<u7;JUmM#X28D$FfJI@5psG@<ox>Cnyb9Q=lvPL=2F
zfrAjIF*fxH=`7sbL?i$>98ED@zl(%@+&cJkJ9u6e6*7oz74^<iYxEHR^Ho;#c&;&R
zYAho1N=c(+RUl-)QJgR)2w+hll0QWK69Pb_e(^3?W6FGBy=0Y4W2(si=c;@*(5;AK
zXCrvLn@HvX$9g{=0w)M{j`Ji7F48+K7X_|H;s1-qC>{yGi4F%mB8iXs2)&2A9Oa~L
z{v7c$AK8Rw&2>)zzsi)&?32RaPIIkuGpV*VB>400MHV&V4KbTtm(>nHb0q1$Y_kvp
ztX?!Q88(nq9pux>m7Yv{kHqDlWtew(LF@I;S^z$G&#mv-^S`OwopFKHqwVZ>>Hn0B
z6W`;tB|sJ1bjkoGe-q2DOy-wI@*zfqkBiDs6}$0mqwQ0D4c1IvAHuxCc7s`#9B$!C
z&VRyI{AYlKcUTJlh0Zz<PtbW)a3cdSU3d!_q0Db77@(~|&_lVvHOfB1SV;Ay17_{K
zm?`OVRMQ!{fGlR$<K!ET{s$BTWE^q}2)lU&K$Q6u3;Z&Cby$XKD!0j$87E(3=BY-#
zWPwgrBt(mNrwh&>+Tq{lOW2t?7WiopAunvgTZ+6W<EUXuq?mlR_>2{|+w;nnA?l_a
z>##b{HZ}ZgHP&j{nrK%Rs332sLA_#N13V%@p>Ecg@K(<Ba#ODf_Ee1_FEol80~5T8
zfDR!eEx`&)#lZbIU>B3J*gM;HU>@SXVQmM(yue0*rM>=P&{+G1hWSzMUB9gL)hanx
zK=M1vGLN0pDU1jNgflxDv_XWx;ir+9*s8PCO+3QxUcGCHV7)e%w5hj<r}IXHVqr^1
z{4Ir<F9j5_*=Vu!Lx8sMJXm;1<YOEt0J%<H*mVo)03*!q&ySKN-;dG4p8c^0PG3R+
z5#+XbgWKz<d0^|{|KU|DX!T<0YEP$x$%bP7JmxGzB}>*xkpl|moE7W02>Iq1o6;W|
zQa`#&OAt>Rxl8%g7xq|Za9P0I#55s=Sr*DqkuogKwaI38ootgreb3caWCVa*uzG!!
zM`M3e2pcB5zf=Z<suoX}|EXFZRG*>-{z-gxCI>D*evvWC9+oNfgYx0z97>UMM0g#e
zY;phXH~-kterP+%c@Fi-Rq$q$CelL`ZPRe-8Pm)@cJh6>pl>)7^*_|~p~RmX{7c*b
z+#d7a{^^SIV_)c#0INy|0E-lTGY`(`e!5jZT?xZc%RHL(%sSW}8$1HuK`yW<JjZ?~
z)f|up1v=6ou{KS!RJhplf}E^fsey}9rt0S;&N4-~i3x5hb4<r&%EC@M-m@(Y?glRm
z#NgORUVuzG5*Jh2f=CisF(YK_yMFYsNl{@5_fKNPJ4@L*J#Z5QaKo{*O3@=-U;Mf$
zExq=UH-G}*yvH?Jdn6bVpa6qo2fc{=8jud6<{I2-ipQ&<GCn!eIIJw=eM>T%#7+&U
z&hL?$%0tsaxo=PKP}=%>N$g*?L34;=6XWK#Ng0sK%X=WO6R|(G>>v{;75#m}3Bd~c
z-cg++^S8T&B&oCjF#9PF7YVL;Zq0)sF5Xj?kx1RLn(~HrOLVx&H+ld;mK*n0aXu&x
zx<GVl_eC2i%hVh^OjU=IOVq$OJG$)TP+sCb5$N63VfS70%j?<2#BKjsP7y~c9E62F
zdVFE{31=)3x3TmncRVRFGPQ#O+ZJc(;==u@RtVKxJu5I8(7%p<_zc6NZ9|o)rri1u
zAN8NtQ;(O10rSM`<<^$QaC5$;+)af8dMHvg1-3a%S#5L#sx_Y;9(v3joKNo40QIXQ
z>Y3#@62lkcd2<id=~w0aS+?CQWeztwWHVYGEUqF$`}nm%vbsXMxb(?emdNV!mD}os
zUql;u7?g$|;Qge-gP`LNH91ny0=jUjxa}RM=z+MXC(D-c=%+hQKlJo;aomr@QzD#5
zj|JK$37s0d81jQWfmHmcL&|MXg8noF3M2QZjK@KAu{~@haiPg4<H_zTa!uoT?khrs
zfI9`zN;O}Li-%MQl90m!t+N+GL{l$=?g1%R&dQ(y+<pCcF+;nmwlXj)8x49s>GBAF
zA|SYH;GJ?xFP^Z+3k?|JjM5poA^fV>4|7r!GNNZaS9f{lv7bf}SdDpabYkR(d;!z~
z(I~rbGIdP+iyf8!6J#f~8|V|MFXa`^Tr*nBT->UOiZ{A03S2`+8uc4n3j8T;6$gJX
zIpF^T<(fY!zQ8ZWf+plD#M6a@mpD~4l*!VCFcq?*YGypg&cyH^TWy+edmeBC?Lulf
zsq0W(^ag@7%Z~*+k&ZSSTE$xY)m>sO1WB8bV^z7Wjt<jLky%kr4t=q*^-EJ;q~G=i
z|F_xAiwb*TA#9Bs8zu+XanpxlK0WDmN?LRJNId260N4N>`W*iqP(#y*USCrqB^HYZ
zI{<~3(gI8rZ1kmq!-0zmaGHSH{mdhkPc?dWscETa{=|K&7YeEsiuGCih@w@$N`8AC
zC>2U#CS6ct&h%R(Ij{XeHaEleMcLizjNuzXJm$4=$^p*Nz$pdHr#}alr^O;e9VMdJ
zbpi<@P3(gr_4^Yc%#@}L-~F1KnbGRqnKy3{slxld_V3#}piZGX4Q?wBlSQj;fkg=+
zI)7hBi-+3aNPl=-aXY|c%M!$4pz(^g^`<SRs7#$Or<@96(?ZeQO*C4V(~;u!{rx)k
zvo>xJVQ_y_|0)N7riaXMj^iSo0)r;Hza!7LZe@OStfnk0H>YFI)5?u%%r#2#%eBee
zxRH-MV5z>hJLp>Q&-eZo#Ey+B?_5_!PO>`#SOSf*q)YEb-CL57euc{dG5MbLc|-)p
zO1JNOiJC9i1tI3kK%0|ZC0i!Ce_P|DZ88_xWkzb4(ug!5z4l>Q0*?q!a~}sq^}3;#
zgt81OcvT`%)t^Ogt}Z8U9`(Q<yRq3Q2J<)+^IW2xS~$tP7A5nl9kf1xjLWSq4pGVK
zF;m;jAN~Yk`zJo#O~lYson7n39bLeHuye5etz_&K-;QymTC9NmSgLBMrapu6=u1rT
z5v1IcD>WYA*C+X`J4`_w6ci3rBPrQ$+KP%-Unn&GS7C1!X|=%Kz?-!ixSm(wh%rj*
zRSyTm?=23A=~wfqp0T2LT%U%OVGUhC?hB;b%-}F|ykjx|S<ti=!%Tu`Fueqf$P@@_
zok?PAuG<pzxU~U(Bzr?Twc=<UJt<Yfc;Kg@Rwo3Ah9a~&ZYQwIE&~2>^Ai&7>5cfD
z0;vMg;L*gK)d{6uUvTq}tc3oG5dow7<&N<lhG;%)0t5nk$h@yjnXFyTfaK6p1J5xD
zj~;D=Tt4^R*<tK6tJR(M$bS4z-JO5qIMLOif4_X`>Aj(=kLKY1_PAYaD|F!jcC}oO
zy-x@5PwV+(ed$5~oFkD&zxS)ryUu0A6KFFl1-Fz^3CQpRBjh{Yt6b<yLt0JR$?Ub~
zZ{Oca5+FRfeIn;A-ip8rb-v;nWs5!c8`rgEdR>5Rk1eP3k!=P~?Z(5l$5Qw=FW*$X
zWc%TNN%1>uxvQp;UXVHnT`IV<vk|Sx#!CPdOo>=Nm1+(7nQ7_fhSur)*wG`Jf++Vh
z%BG6d_7Cg5Gffp6wmt{~WY<ZXGx2XNbOgw2DLyNXYDz(b6e<QAb%t!+m70BB&i}f7
zl?Ez*cAsu5GU@e1?5YRz0xUC44x`v{Z*9Dbj04f3r>t4HcK)%D!D9%x0yUX}(*y3Z
zGmWG=F>Q9EQl*eGl~OmXqy0YuKE{c9j+3y6y8kQTi0j39*$an?%=qfwV02^WD7Xcq
zMEN<3K`OW>6Oej{dRG&s5O^5OpOc}JWC44jXRL-hi5dvoZ_-~AWWkk;4Qy|+V7~^w
z&)Iz0IhL6RNdhhh9#jND`&ALp+)$DE6w**95>{)}0hMxkAaghtsa&2NoMXIUwI%%2
zPP_5X={}(*+2<;vby}BY+0L?d!Sqim)m`Svs~ZNLgl$SFAceh$$$-OoL|0KR-azD=
z7blLI83N7o+5=?4eZ-(g)x~-vf5%r!y{0D>#<fgf5E-_d*y2TUEl3SeM(jtvZX2$y
zbc{X|)H~y20*k-KTW+8htZ%urC`U2sjVx1Kshs`iOf!1AqQZVf&i*b9g-5)aSt)mR
z3sB9>Q-I<Yqnd3PS$-KGo}r?$G6j^Le}3{-q1}bjfkp`o{3G?fIV@uQh8Wjst{_`)
zrz7wr(M5XZ%HD}(z+?qvb!C$7cD7ne3jQHWhc8d1U9u`T88Zgqd3K5v2Ow(mjPXt@
z6Z!%@NC0nfRpwG1kpX8z(6Nc&qjqqStIt9DN<;U6Np^M$9z9Y>Bb4AL(GjLKqLY_s
z&&gq-Zw1YO{{oKg&eW7DoNK>hkUlPDszZNO>*d_lG4(cZB&!DS0^%sE7sR_9DUcM}
zETl`E1Q)zD7s5by`7cs~&Ilc=;U0^mhu+M-MtWRG0(mYOLp7m$Cu}@fhevPNHhz@4
zZL(od)^37Av-gP(Exelq<Gl{Ng?Z6^p5`+eGZBzsu=e-{Zc!|cA&i*tRfo!8N9cv!
zR-8tuTz>qnA7t0JDb+;_4jq00ja_;{j^jcGS`Zv~*lFP^n^w+1v{a<PL1f)cB?j*r
zC;n|tYTc-9@%9e;8Es_aQEKgVglg_z4M{;&H-*%|vQgb@Hw$es2BV#)<4@)Md#{ak
zlyTG!=MHuYM$lWxI~d*IR2|5;YXU`vbUFQ69ViZ1CJqom?q>@gzzp{n-9A8wHc&l9
zomQkZ05Ourx~23=iL)b@u$wA-`!gouuTo7mdz<tp*K_gC;~j<!miK6BS)qK`*l4Mf
z|7(-#V!e|TBo+18Kef^P*85--qGub?9dT9>r_<uc1YbSc)S6(oY;wg>F;yYdfFCc{
ziU~<12{b@`q|ENiw10-tBIMH0r2MD%)0b}<V8fS4^o+_kniJ$qmy3_4F<pLY5fkFl
zleIv9vk0fK>V|&u0_#g2!#`Mgd{{cV9fO`j`ox2inexQ=c8Dg2YOMqTO^bniob`*P
zs==cy+6+So1KYC1-_glPikliVAYQ4@9+BEJp$}l-XW>T->=8bq#ao$bB^RGHRJ;&>
zQJxqWN>bXkNm}Kz-&wb%QnILv9#hV+XB$%^$M*J-Gkiax>zzGfdtY0qTdw)sf5N=H
zR!+@nGu&mpI<O{m5FS`)vZ&^XxHTP<QmbyWSUY3f?5&K=W9JjpH()y~&xM(m`$0w*
zTL;kjMf{gi^d{TwAqXfswP5)DI@DF8UECnE`rFD^y=Apa-?$C-J=3cECsE=3%NMQJ
z|F%iVFl`v)9f;^mS@La4q;%=ej?2}Y-18Nz72iYB$cAgJTP3e6;U}k?DyU1bZuO=g
zPS~A)=Z~L=GXVpuA?HCcp0Z<D#n;aF)f3PWEdwg$OPy9@vT=rDB1p&b5+QN;zsZWY
zKBI}m=VbB1Ih2Yft}V06vr9$YJVz-8eK8k(xIcov#C7rX=z$|@gP1JCg>_d8GVkXo
z5qUvTa91xNGYvD1v%bZ<y>=a8);67?aKga*b3P{T+U=e%AMIPOSqy2!(qsH#fWQKp
ztmPDLI-#Dip(-e4@+g^EyZoG=;WgDO`=dB4YBwYia`webiII<w1c5>ncu4?<ag5Ag
z0<^>Xg{3WC{|m>>-bB6F>g3vU^A-3#BqYscGEm|F#5fXVpBG7LmLhjnn}xYJ=s0)o
zp%xM8KCRAv)+>|^F}=p@O)wt7Vp6&0iqqi<8gAe|(ccc#!X~Zq6uSf(N`(OB%=K#A
zv9^`YE{(fayC3bqH;h&j+&O82PzXoKvB!V*D0aZDESn=~AuXm86lPxBU+jx}fsSKx
zFcgFlG=bX|imntYEB4(V+n{+b;v{J))F6AV1oh`Ya$@}n9=^XP2A>B&aK<$cb-4CX
zLIg|fvzY5kQ{l?MSQsN{wj7PFB+LBxCalK`BVn7AC6ks;4AcxaCM|jCbJ=iLj38)U
z<cFtohMHy1uu9Ti&f>-vi-2nn$SFE+e8rF%?%e!;r&BQS;uku1xalok5mjh`lw;%g
zW%uCAu3m$X{l-f<pBw@d{Mm5s0$77JU(cmJefo6Vr{8uwyMepJ&%a7^c&!*sDXA>>
zGfQ2J%!972boU2CI*=dXIc40dUArB0IMhE3)F}C|uPm!-h;64Z5~cLgHHZ?<L{+1D
z{0GOWotoO@3*4CFU1W?puF_sLGyGpFj<pi!_=VbL%w{v75CIUQ6f=XyUC#fTl*qo?
za)rmlyiTe0)4?ZkMoKi_D*kkm$Ps~Hz|Wyf4UJ}oJ-N)HLYJI^BtrhnPqGK5Q@9p{
zH7${^^us!;Q&usA*<BJHl4>=|oz*H*Kbp6Ungflb$;>)~QL{QF?Sl6!NAr)sce6t`
zHsFU^!)qh~f)V?A@kqBIMj%EJN@;j^3&55jKNTU8(MA+66rWC&Lez|?T&8~R_09@$
z>Ey<u|BbE{dtb04F~at!*O3#5Mer9(AEk#s>BH36LS^tM1z+^>G9ne}V`7oz{a5-S
zaz~vh@Y3c6`M0lY@D?8Vu9fW({&%4&(o1k>XE4ADIEGMM*&01OYl8r9+ch>Nxf=^E
zjLjbO-)<bKsnBJ@sdGG+R>~Ff2w0l4*3BEPL0+MG!LO2`7GC|*LA(Z;jJ-Hyqm@lz
zGFQ4%KZxU~^KO!s$nAHrs*7V-_9@J=6(2>IQKCYU&rH^SncTG)sGS#=iCe}1uU?M_
zToS;;F&=T=u3#U2P=HL6ls$?S$stjl?yng}cUCi0W6Se|VNf54H@F9_6j(R)_Qu?B
zyS!RK1OIN2Z(88*N~#(NuL~}azi`gjr!2F-_s>~c&!#<dn&fdGrIia~5{PJq2!y40
z`8Cn&XbjcVkNR1OWNeyP=;&+Sc)J*~v%&xx$N0O331TR)4d{RE0}@BFgO*g@9pnC1
zIM?02sT=en(l*MmahDtXe&8t0MAblf$6*JT-Z#;|=*RB%nULvk3V{r}Kwu#F*}`Ig
zg%A<UQf21^{vPNlP*QVHKJ|(Jd{c6`s}-0S$jrKtGb`tRvnVTA<esN@vE0Wh;9dym
zxLDg7R(pMr?bN=oFyf<DlbI5=ss=9&>N~(%djM8V;IYaEc;;D+d_7>Jel9eZ-!*1G
z6d&{QFI;3;uExf^ay5>X7FT?Y<XiGJXn&s=<W3BYmnJs=eA`#`YA5*q^P(X0@AB@k
z45eKD;=|>H*QR{9Rpv&|Vw<%d@dW%c(NQ5Rk36CHhh>O>gYx>pZ|3|jcyWj0GG=V*
zY#`;GGaN9NbzA72w$n0H#B0q*wB^(8oF@Y_C_b18SzYwZlEY+qW4tt%Q!f(7x@=wa
z2SO+<#9}eBRAY4GYzt3iYLw|jPw#Y!ElgzfI5fkfmg)KCa09+2Fgty!M*)D_Q0paT
z@*%63%MN!xEj8aGI^Y+qx30&Dn_r3s;A$~E3(jReBH7GQc6fvhP*)})H40#_z?<x=
zS-{WWTpDVLa<C^XskHUn?KPoHzil}!m2?OITz#H8;4FqHbV_H=r}0LYfrhfMAbf`Q
zI*q4{PcGk2sI-nLruqekTLZi&I`v94+Cq@sIcm6hp+cBlp(l#UEJLR!wTq>#pY-G(
zi<$i<v5oA=^=z{Lu3%`2+^nm2ah$Ib2Hax$$e-=4101UPldA)FQQggj>!3SXZ>CpM
z@j~1%2K_ALCyUXx)Lhr=&@R30QW}Gp0~2x?02iDNP~LD-IODydMgX;l4H7|?l%U8&
zrWt_b&$PFaNu02h>h6x>l}+Te2Ir<pc%fdUaOdxzzs$*bbq-!mrI>}R*&zo0sxyxl
zQw?ljK+;V4;Ad}Cbq9fV1?}U4Bd-{-a6PZ?jSvlo^w85R4Q=vz`X4M;t=6-y`wG-=
z56=vfc$nH7@5GsIO8~!I-J|oh;Y7ZhRbXTx%(wn>>Cygay?Mps_}Q9Px%%UBx{Qn{
zdgXH?gFvekLLrOka3MFjI72MSx1ZD4l^~XnuLbFX@IoI_G#xMg9~#G{m-$lfee@W{
zSZ6yApaG)eF~)ex<Eg|qHMLWhjfo&)yDc!v*zzKi#r?-iH^6tZxKo7mCbzOn0KNK<
z0E?27)%oA{wUr8X+~Q2lWN0*-qMz?J(tV^4(K4f{kz`kqtlSkJLPeQjoYo3{2&oeQ
zAG#fWr(W&gsE3%h*})BRq?p;HkfvhJ3bPmQYJf5;K}e;qJY{U}sAt|qPK}oKM+UxH
zkbA5$%lXID0QmpA8XTL|vi6s{JmSe;NUg?u4#BaGtj<)6y3ZsSONQh*F(5P0*IMC|
z(kcfn{;mGXq_h<Z|GhJ?KUMg%<6oP0uK!hwmz|c6l~D*jkdZnO2Z8=b_mC~<Z~x65
zm^12WiO)F&efO5j@FBC7hFZJC5gH>PJ1<L+yV?*tNoDy!gSQwEDDLo-RED!m%mj2m
zi{1@6Q1B{Sg(>6IA~g>cbBW|ci%|RG4Ld7v@i-&ZrCvec>APxxxuWuvB8r!%LRF@j
zdL^rzbP6+G47;n);<PGqNr~+@00X>c7D$XxLC=!^Wi|KTFgW}tdUb9EG!+0c2{zo(
zHKyKTI_oS7{#(=KQ{=UWLFq3(6+#UKk8|}P@Hm2ss~GRyhd+kI8G&Q;r4Y7CtWx7A
zx!#v2Xkvb!$2#J_yA2-LF$AFuj0Uwx_~Fr?pTtQjf*1*h$UrlSBrBW5z}rU*b<0f<
z*I<KN6V}oEn9C&$P;>Iw4F*7lEa`TP;ip{>kRF0KN9EZ1ZQl)OQl*Jj5tK=pMR73I
z-3`h7XR)#FNF_AndBQD4)I-!0Fm`=hSpFY3M(Gunzs=Upu;i<8#z_$hn`35?g@m_!
z;J{&qq5ZFaM$C*g01*s31kIrze-?GNW4e-s7{v`X>PL<w0ntVdV6u@A7D1Y(qq$l~
zl65nad&b(hA~R0;zx4S3uw#gK<>8g+OzlE+nu<QRwldGum(;2Vn#pkat-`Ndv(|+x
z?`}YAPRJ*{P&z@e&o!KLC3a&jxY^iBPVK#XWf<r^JN|_3khhC+b=J~gKGj{8xTabz
zmp2WQ20+01K)<owB>WNc!;17HiXH#iMOpBNpj5EOu?{1=ZS^RyT0>s^a*=eZ;BQ>3
z;dhgMbrB(dCGpH7`tH&-+Y9#pTV+xzVjarY`?0aG6mIxF-{LcMbrQ-bUFz0zLGd*u
z_1Lrtl8EIWJUfH}L9`)G9Z3~G!!q92?Bs7E1K7f|q*>OB9J1Pk$lzPLz4+FF92A||
zDms~M3bjRc>%(>FrKu|!Bmz}j&8tIHGpfBbhSBC30Q+y?>qXVPWTc#>@s>4G34$V4
z+air+4(=b)B0wy6mEq+1noaqYaQQb7_*-@Z(~y_kl)sqjNR$2=ruJbj>sy9TY30T#
z;ADezz%*63MM)GBeC)=bf#2N%dJ;FSRV>Nf2N0a^U~;0vtR+E@O$w7f(0)f~^-GAx
z$D+#Pm3IHQGhjr4Y}S%neEqxrxq2#`bZi|^oO0y`y3fSsX6?3$s~eQ5$@v7_xq!-w
z8pDFtA`cs$9Asl=^8K!(mcRr8&6Hvas2Kg)g>X<&H{<sxubRXwOz&I`+Ro2sl?j$0
zwMs0@H(~TL5#<<j>{eBlruR>qZcLx7`O$6Z*&2I)m8ML9%zX+)y<|aXfa=2Avzyjw
zNFYmH+h;-dp9UHxoN%^#`CBGuZ~qS5dgl8C-UZ(!=Ww^N{f0Yrk<<lw0r}aYM9@7r
z2eJ*1lP4`0kj1~f>YXMR*4@jT)9r}LL3P<_Zk>iY$$?v=<;Lxc=C`CX*IIHWCU!*G
z%rPO@twp~gHm+7;&z&qr9lgrIoGpO}+%*lJ>Yt3@iBe}{1{{l!EpZqOXWk+mo*1Ks
z#(||hI$Xc;^@b(i=P~5Ei%CjH=vakic`M8D1<nXWTho#?Z#AYl@riZS9<><WLH$^j
zN~`&^q($TWS;ef~2PXvtyqg`ylfv&4l%2<&p}ydh@@-?XMC;t3_BVy>UOzs%BI{2z
zZnYn_hoYJk#i+3eFI&Hya*gP2u(@8vEwO(R^4N3jQ^J6DPBw`_y~*Z!EeXmC7{39_
zHCw-dX53ElPiznC>wj$S)6!8jAxqij%|dVhytXM=Pzbcve$gX8`6Hy|4Izsj1+P=7
z*GxY7Z8Dq^{IMzc2KE1)FwgROB4XPt3w%J9icl2rJ7JUsT10WxTNZnbEF1W3thXPl
zk#%=o9W|^{i*HRcUJS&^q{_A&_GgVHiy^E!9zlKdZ+;65!GDUjpX`pz&JXyA$Q!1I
zu<*CDE&`!%L3seKRCz5BJYd4qKUyHT07Ug8w$rT;X3f#X=mZ}={T^Y){$p)Xs<zE;
zTPnT<!Lc;%4V9=P8RwH(LywEI#)?Yrw~9u{X149wwQxtbwz@GhyC`JwUmp6})5spm
zp+T6V?OT8QQ`P}?LYygLGY^5W7FIY4=V6+XOW6s?Ep8>{0_Ajxm|iXdmxR+@fZxVM
zw2@b5DEvs-X;rL<;xns<__!2<4L6{qUS7iseM1XY-Aa*v<Q;kIde{_oIM{C+a=8Uo
zdP6}8ui*K!a;4kZbj1;*_m&>Bke3lw;!YZ=H?!I?+lN%>Yqd@Z@LBWIn*|GQp&yg_
zY-%&<QuQjylk+Qbk0&1gdWT5k0kno|I&9si=C-$U_#!-%{Pe<&R?O8&im(S*2emgC
za#;y7g2eU4a2S#N4#9ij7iTQmYS$|gIrEfW|L6me2|7Z*Ow6{*DO8?RmU>LHGM9}j
z8DXrz^_&-Mw1q&Roa+a=3ZV}v55@oD))}w+d+g56{MI5azGPgiNDm$!1o%cbm(Fvd
z79}HUlhq~j9RJ>S^^}I)TFm^eGN9p!g4H!@%M_tG%4BPc5QRB};ulK*@mUfLESac0
zyWFsKnr**&QtEK~#AmmI9N$tiXKGWv@^ssd^y(KZj<;*opBMv%nC$%tkgz?R5W4lB
zjx~L8N`)Tv?+84y%2%!e1B|_FuRRpGLGIKSq(+m~Gd1A4Fe$z&efv5!M<#-fIV;&@
zhtPLOJRmcVok9r2g-G2fd<jti%YLrqwW|T$V_i~^!4!W;3E`!6ONP!3l*QY*7F&Ek
z)2bq`#?<68CMC%vJZi1kOy;q{q*&$nv==K>Gp43l7-$c^S>i6h1N6S|81=^YwC|Xb
z&FNa1l!oic=k4n#u&-Uy*NIT?v$iEpI-2Crm!+@P*NXIiaXU0<I=&VNv)V&Gt0D~B
zTxifip1)M22ghV;a@XqOb+~(xpwq-@xa%vy6dUU})1!MB<Q22oGBWaPO_&C2jmy;-
zH*eF7Q%+8>5E;m;0T61GcKis_UOl6<cq}!s5}G&q)ua|&@9dLHFRN^<lb9VS;Sv83
zj`D^IE|I%23a><hxQ=ZOjdgK_c{)W%P!Rt{`J0lHG)D+Er2sQ6IHnxmCwli$WIe^p
zNt^DdOgj-5emG@@-FCwd5Sk@{3%9fWR<k+UklRz6fk*x=0+_NtLS6V<d-U<!be(eB
z!I9a?L-fiVkq-H^@E+E5Q6x2`I@&97Uy_4S#D%LH<LkQL+AX(CiQ7gkI1y;4VK_a+
z^*;xxFmZn(S`L*kVC7$J580vJpy2)xnhwWUs5))rSR1VMk)}!F1K7sE2%ZaKGn+@c
zu}=~%c78cHK-{uRlcwpwv1~tYvB0&B9{X97EM@HD%WH2&U`VZsPP&+tdNn+C(~ijb
zweSXioKo#5D3=428c6|0_U<};9A|++)IZlEiQD#u{y(YaEd<+O={DoUxs0=?d#(Wa
zv(Hy{##*g$dU{wjuBFYtS-pMEF-QRp1~nrdd89h^Qx60i&GGvJDCKp=71Tw!Y^k!E
z2=oYWluDxy!Um}2Q|EUD-e*CerDq*1__LDKxdII7@^<P~x^wN!h}{FZgSIFl{cMoE
z?eg3)k53lXk8FAIJ`Wli`=8LFM1Ev#lJS)HjQEN8>3u}lA%1J+zl>HXypu9UlF~a_
z|DVueqwG<v&qZA;kLb!c++Z7SJ*Gq`vxIE01eZw(a8f|(UV-uolOQ#Xr!+R6!bLTi
zTe&nBc7AK;yhxp?Mo|0}Hiq~kv0UfyhQjHx@WwJC;u>L0Z*eO{#AQ5(i|)RKkx~Bd
zNVM(IU>hG2ZpNMyt%z+v5)om=l6dtzx{vlIzplO{sKj3nWw74^qhwsv$-hfJC^&-H
z-2!Y>R?@o|*4Kn7L>B1913iNL2KgPKn7;ZUHB0eRB~(q}v?$m|Y2n7u$`@40R@L3l
zzQdyLli2EB0uskt1*gx`98p~-(}6ey9XR?(wfsq)Ozftl{PUFv`!|mfnMfbJ?9pD)
zvMgbGn)QAI4;SRupZ{RIoYDdi7yi_`0|<*FdKpa*=~eF?c(vQn4}R`0cdMR*`xc)z
z_@_v*`H1W-)W!?x49&2hg<>j_gKds*LI}NjL6#%rc}c@4HBZ3`jQ7AYb)mrX^_JPJ
zhUn>Y2{f=F$WQ%}G=#DCBSiq0Ic4)6q##%y1oH;@DVW|;O&#u`U<IM-EWlts&<*rY
zNvVHR*GZ&vwf2C?E0|i~!lifi`}%MYm;BQ0qsityJbj*m5@;BwSp(|%uM+xeMb{so
z2dIFUfa)Y41X+OY>moR<CrM1kMKhp3>?ihSy&JV_BGHIS?9*x7rAb>*HRdD`d4;kw
z!YBA5w*)+h2hyvRCjcV3yDXGfEehX>k%7b`;28Ac<W3|lQ_08*z!$o}wMl(~XWX$2
z7aF>nLu$SJJhh<I8Kdf0yg{IkNd3K%t5a1$9h@UV;==EjQkT<{>I-?Vs9n=&6V
zKNglKl*!h%?EKWiv?+$ggsO=An*rwtFL{$C*DsUL+F;946abjv?k`~Nv8!MCOBC0N
zl#m(rRT*`Q%3mgnz09AK)!x+<+h+j6Kl{8qjYVw!u#{XLvSuI|+~Y#$zcLJczC8a0
zJ=bmUgKH6pU(Y`<xW>JdaW?{?Tqk57Ez1*DGcqsW;OW?%o4b$!p>#6oQl-eBe0Bq^
z3Y43eSm5MncLM(Qo}H1m-&n^b7BaJTaM;TD-3-bKmU-s+Y+(DD`#y-f?WEW)iSHe%
zc9OpAjcZS&-aTr@?83@JqsA(X7H~3=oh=z14ac>96ey^@Uxn+NX9BymIB+(OrY4&q
zx|s}gn$Q#F_<GPBMi?yr6}p*d-!`Lj16TMEgTDpFga)KibI^~Rt#!967q+HKHnz2a
zPc=++cv+VZ@SN{+r*ly-NzkLDPfEb)eIhBhPCX`NEsRt@*W!Fvt;f3EjeXQ4N*o$p
zAR9doN#_z$lW~x3^IFAyk>#^@_loHF^m7P36<QI?EkXap(G=?Qb_wfe{w>C4Bi+(u
zdoQ*t5dz@#V_dt8=#294HS6yfuC3;1eY`I+zyIa-98>_?YUIDOeW!bqVT03L_uX=9
zMAnYJ&oqHe(S5;IV(7PICs4z+JY`~NAd{_=?(4(q@;RlHO$rO2Q`%uVxCrS7x`GzD
zC<&s5x_y~$4K7f1x}ya5=SueEkTuhDBwa##<^}xRJIskt7`#tM1AlIfeIRKdKrk^4
zm`N7QBtFz519Sv0;-D|VCGxJH32EH+SE>bu2POxh^A_$67u;}}#Jgcad5Z7rO$*!Z
zMvY>e*u$5N@p+Xx-LrrJe<_O>W)%l+vtuz%QJbA>Ma0m#lyU^Et|zgS)OPY&zvAE4
zpO`Mj<RoPoBKroj=Xs}lW0M+DwiNIQv$!R+P$3Bfy1n8RVdL?HMnSz`I%8Y>;xs$?
z*+2P9%XdK2!~UydiC6>bBBzGdCq&FUpNn<B3mLnM4Q&zMJ7*Mlsh4SDC|f&;ro6jn
z^>sJ#cH4J)dpx!jKI#JHM$RZ%TAsuBNdOesTf30+OweCr0&UtHT?JS&x@aO(S~VcI
zy)LL)HDs`&DY(I2>K^1D@1L;(Ur7<s7kk!BkX<tTei5OQWXKy#*c#t?-~6Sg3&hJV
z1LDs?2F0mYV@QnwDFXb3@j`Ti`s9o>Lr=uTD}%>bYPBEv*RT)y(7@jtZa|)rPyoEN
z9fY}_*|{E>*wxJ>+P4$`FNVDaw9fyBv3CrPyzAP&!-;KYVtZoS_QbZWp4fIWv2EM7
zZQD-Xoac34cRf|_^WmwZs%wAff9<O7uHI{}^;_%UYP3sjQMHXsIgp>FC6E2dt;66}
z=MVNf1A=tXmX8dUYXzjn)e=Cq6J{k^n!4AM0T^tT*YcfR2Ccg&hh02js})-4mdH`9
z>U?D$JDhwP^+kgeZqOUq))W4FKuhQ-WYyMFv-X|a&g)t&P7b?JFQX$Sl0A8-^emne
zq1-&YX=pM``X$W?9ic%+^=Yx9u6^Zkb^G{z1ZFQb%!+NC0bA%ic}RX8wyp{KTjswX
z-$P_vl3F->NIKeWyvh9d39l0onK2Xh^9mG9y#zycv_>{BloTDFC2mKcZwE(W$1kWj
zu0l%O8=I|xMhjEL##C^#(UH7~bmWPSBS+%zX^c2kAZKD})N|?W-P{w{j%Q?bu-o@c
z19u{_#pwWbbRR-=?+qUlv7)IDe}m29WEv9?LV$3e40F*b71Kye<Y=0|(H|rBxE=JG
z>RL2x?EIn9!B^ZnRq7QHfpBB#<18rWylHW~=unCKow%%S-V}lIk1&0{Sx@{}^2E+e
zA?v>U%wS`~*T?d!y!Np%W;DpXmE0`6IW|Qs8B^Y9l?)YynqZ~V`f7UTEb~v_g*GYc
zI)Ji`Du`GnUfKMhV_)Z*ti@KhaV0zI!x4JmZcXguduHt+x>S6@VkubQPxAJ3p;I-n
zHl@@rD@!I!au}4g+RvCuZF}VTUv5owR#JnV;p>q>))w|;SygWwp1h><Ks6FN!@Z6$
z1tzVpal7c|e!)xLHhz`L-jBTmwEbyaeE{#;Dy{-lDTRjSpk)pXOtwPdsmPjAYgPE<
zf!^td!Z2C;=7IMWHtLf&-o>zmc#x;}C@F%E-t$U7b}InGZ&?Zl7xcnH>T*>7N|EjQ
zqMZ_!GIFWJ7{SS2&eafcHAfbrO)Hnv7Xw=Q>5<$>li-1`)*kXo2R51gyS1`8L%@N-
zo`-$yF>-S%U&EXxSL1D>+pTxw4ymuZ`7aTa-vihl3^(_&GS<ecLFAu>#WgpK@W|E%
zn&oiy5}!OuGE9QzgECz4LU0F59&#Lxobm{fBLM@B>da1lDQvbs$C*-xG4saOqSF5w
zZRO)^qLecZWZX5yBYg~}M&gG`)B?;bX3gBq=CbxuWg6?to6d>+!x~J{wG2$BR^ND!
zJU6ZSdvvOf^;`Z{sR?tA<etiv7HnFQ@DRyy7;?*2u|3Jf^!GbRp3j$?H1ljZ-=0$m
z)?x#_O^-lxT`3RCe%U734et|JkY$<jDP38D!D3Ae+qzZZXlIT8)DL!oN(3;N4ZiJ+
zEut3{YEvBsa(cDd(ocy}R<YzsG)pK3;x_Z{Li+vo7#HLwmq?VGXBVXO=zaI`Q8uq0
zEX`XN%F^*}wJKD4{xM!cZ{u?&yRgtk$*PM0EpJ^=Wy>m2VA3#;PI-`qva;18Zv338
zcDO-;tBWcp*&fKcJmX)RCk4Rxgh1;q9<@^5a{k<(AjQJ>lzG^R<^2epvO~S^Zfn99
zQ%K$+S8Yy-RT&~6Og!~0){4H*T8RA!d58F6Y;=V46dDK%%1&Q^u~MDn9vy%cf{;~#
zQscv^JQ6Lm0!qRt>+b!=&b0$eG1hu+px&QzoZ=uR$b5=-k}v-ePXy>AVNuhNi=oV%
z-QsZlOdc>Pk54;ClsBW!swNR)WJmQUpmCRn^=UR6g7h`GHl+WN+^_TYfEd<;Ad4Ur
z-ZLPQw8Z4{1I06`rfr>@az1fD7M-GAoDR1N(&3%G7&4AF5EM;-!0xG}VsOfL%#A{F
z_L)|uS)VmPU?dYgUj;y$fM!NgXCwgf?axZ8p8iQSSy9)g^OiKSbf<Dq50)vvo+&TA
z0%ov6BmAjAQQa6<K2g8VXdMAWHVAijy>Nnkx-ejCUr4pkC~uglRrKk*JeB7Z#1Y?7
zVzP|xSgW{e!cMOW!2wZfoPZJ(N7sZdibpKv-09sb|IXEfgal|Z@0}(C`CSw^Xs!SR
z!WJ#@n$79QzL(Y0Bumg>+cqI^)Y>M6G`DkCt#Rd21yP)nz1BI)$>c*cI#4G2^calV
z4qI3=C;HH|v<7u_+s?^OFg7au=(LtW4Zcw2A=&YOvzs^ZdYnG{o6E9G+T&f7)FA58
zNW1^17<86NRw&@YVLKd0%9;q3Y#L7#@qupgje2=y6%s|d2dtICx|{*!n9wzakdUGC
zcG2$b5?z>`_oswjU~4Etw)(g-WGdUXDWE~Q6w3K@B~1paVzFv@z9dG&eC@AoIr<HO
zz3WKQO?6W??8o;<M++>q5HblLaZ5**y0zrN|2h)kOag#Jy}FP=S`a;q#s$5@oQBr&
z41Ga&<ew<WKOY?_w8&trg0-{yXkoTt8k0OLDirARlSmQ;Nm&s${lqMb@i<^ttPq*#
zxC;L~^o~_(y`mP_PJH?8#I*?Kh3j!0)Nw*@FDC2YG>81pHDSSCv%%q2dTtg5ZrNOp
zk0Bw<4nQbh@>j0B4mm(k8hIO*;x90d^6fV4v3?t(-4Ep;=Iq!kkeL3Ew<QSpAZTQK
z{N4mYNHb*g$AJVw3Nnnz&<(!pi0gA^u*^^g?Y!D`;@1!lt~{J~e(ZUa*Ac#BJ~TKg
znjmOfdzz@%Y3pZ>lnw>SPP+X8&JIuD(<L>zQ2<GCU8o6}K|b+rW=A>#C0bZw8nPv=
zS^xfZpJc7LO3@g+)<#W#|6!c=&T_pDD*a@~o19z=M0BNX0^O7{ZPc#TlYBDoi;k_%
z(~Nb~9q0q#3?bo);saD$a$sEN<M4NpM1gbm{pv%ZU+2P`ncW1u;X|T8oR0>w%<0!O
zCBS^?OVMNE><ZeTH%5d_O|oOEB(coiyLQ4VZX3V!iO#l*dIK0mUX?lMbOZgAvSaO%
zY_dhxETUg*U-v@6WBUizfM$-Ou9nx7L?}r-@Z{DpSTJphXMFIK6+0!8wHMW>1eAC*
zxoclU)m2*J4+#Bz!>TG1M(9kbl^2=jLqO!nse6w6ZaRJTkwncCKYY7H?)ff*k3#<0
z-UqhMfQR1CAFZs5=Cu~@Pu(l&V(pUVy(C$eRPfqgnfh~|mZUJ9TpnWThFST4g-ZIc
z=9h&cKY-&Qemw?<yF?%s*YG`ZNX0E%cHs+heDs{`>?F+z<x=YFZ?Wa9PpYj|cmeXc
z<=W>N86ll8q`p%YH#;aq36}HVBML+Pj>?BV^j5CGr7U4Dg7w+4^*{sk?3Y9X#O#dS
z64l%9A|32>%x(2&VOjejF^+!LO5PF}wc-TWtL{HXu1`VxN=MCGG(#yTFiqJLAzgPN
zc9x2#L&4pJf3ZrRR(C9$m46#2KLV<dg8WUGVr}xD-U&xm%YvdZ2|}Ypxy9trPJFd%
z3)3u=kn&3p5;0qtj@o-x=WxuD!(Jp+%xt>_4oO@*)(4RNq3NUzPYN?^R2^s7QFo#K
zwslasVgakrDxa0{h^PM(%fioEe66*pwk1EaL74rrGA6C8Lm*4FIKgn51SnQrVir&~
zmW<`hS>hlVX4H>1<jUvKBisMlZE<!fe0H>GlCbu*SS?QUkGx7>Kpiy42eI=!Z6!st
zd!b{m>5%}}4{5D7d2*;ZF_>6EIa*AP++d`bdd2-|?*?8uoYP9A(a$YLWr0$(<4u(y
z;77Ec^s%jeBH_U2RssWk2G}FEqknE-4Agb8ixz=W&qKFwWX*N&eY-LeOYv}Wr&Bsr
zzyyXw2xKvAw$=(Z!&=DaLMDx~Nr(`;G=E|!6`t@qD&Fd7Os8TmQwMa?@>;yJkm4NC
zhU*t7SY!^a4wC<1T*uE1?`0zMX{XOO(o8%z!?Nq-ja$NLbi+HL1)xg0=WsUDJtXTC
zNMjQDMGCAshcR#dpnv7cl<xAe<EzG$2BZG)U{&GDL17MMjJdIfFnK|fXNSLCl<jr6
zG$^`Qo(}4<&3egnqQ=*!(H4H!GKYw0H!(>^W&Er=LqZzzoT(+NETK8u<E&Q^a;4^F
zMPBBlWsx=9&<eFM1qgnr^A6~36oV%VJPJx{wK*1KzpBWUH2DaPM#}=Gk3Q%($Qt1S
z*;qdsmat$G=qwb-P3*jd@gs(3+bT9F&t%RTxj)p;Lp@xT>0omU4InVX>gG&nj5??-
z?VoMZ?LxysmZ9Oy^7<|Gv8M%9uZe4%rdWfnPA(3COTGLZ0VKB9hvcg58=EtSLb3$H
z#{Ej@ng5>dh;+nUTk1K4MqVTClpVuC7r}hUCWIib3#4OnArEsol0LYj_=KVMqi3R&
ziG6=+KNj!&hb={Qv{(j|Ua5~JziP#8P*YJxsO-SG!Z1~0T!F0QTCKIf)Vuokx}7=B
z+jNDW%Ip9x*G+f^dePZpfedT}IuCJ9GUCE%wFm7fcAcwjiXORG5648w2m$X#rw9Q*
zx7lU0#oZ;%s>~VTL2()sZ&<F~iWpM%1284@6zjf2Y1zGm#}G%sdH=Jiiod1NkxqwC
z3BQ`ufq$RM9kS*79CVad%uA{qVlPyNg!Hu8k-h)|9N4fMhk`g>yf&s*?kFWp5woWC
zYjmDUQY$<2-S&o|@O7S$ESmrYr*#;gW;I&xc$e`!iIa(WKm{mPw)9X1;ecp*g95H|
zo=y^{iIX_nf{hBsWnyG|F8vxjr$6=hJ(FbTeixLD3BE$**5&Xvs#1wT{62r#arBTJ
z9f<*t;=KcS`zr7zmu(p6m$X8wRub}Ha0)L`P-_;D2+Hn=jo5{kM0{6BXjq`s_W6FJ
zpszUJ1yA0&`g;SAds#T|JXF8<I7bt8&31}KO0=4&^9YJ_O%BUe3p#|%KX~5kjQyb-
zcY{LJfd&PI{T}_41{Tlex*7qcbH?OJSC{~+%JFsd44Jbd@PU@6|0C!)!&~j>RA}|`
zq3&>6INAK<U*efBkn3DZU-EWZyLjjS*m?y$;FA2}Pwse~Xp)U*a{J#y=3LKANMB62
zz`!}KuP+Yr{&Si#YZ>_B@WB<FQGYl$ow{K2s63u}Ey1clMw&?fb(f;i6E;1o{p$O4
zsoX;JS>dUqRW6O}u2>SxQKqB~>TmCn3smyDSzSZmWP&*|+$z;ad?7E?u;2Gb<|Ux^
zPN09mZ)JD-ph$88p#T7i+um<LfYMKc$<f}x734GCO=kWKpYk8SKEb9x>fZkLf~+KL
z_S=T`D?>0M2UF7z@WZ*FOCYgvss-70vGQg9AOxCDVYAnj#w(IBn3Co(9X~%Ae<Twi
z)pDR2i7~RJT&HMJoH!D(K-mfV!@MY}MC&<5!KTrg7&6gVO$Lydjhx}ZKdg{5kBrSO
zrU|V)EKG53aw0IE(G8MyxX0=^<QdP6<;<B9+ov=w0mME?THAx&8WNE*#1f1+4rxeo
zB{^()ix<e3C4pJHxu3(|0kd!z)0kYLn^q6>2#>}J4`Vl;<Ijloll&WKkoEArK*UF7
zXiFdpz4!+L^jzVnl&K`!e7w%kVbg5Y*$UKs#Om72w=ymJ+p>Hu((_YXuZF$HI`rTF
z>fM=c4Hn95l|-q#GrsA#WPWfvvoQoE?v9+8cWSvTrCLf=-@T1|G>t%P+OkFBA^c4%
z)!&r))hBC;_=Od|?8`Uc5_xlvq@ac_H_d~ELnd$z@MlfLN+2U^gzTm7-|d)D-l(nN
zB`NzXZ(@bU;Bz%zUsxODL%bnL5>M;j!y(MkDPxfo99x;5E?#Vriko7)MT3{T;iR~1
zg_6bZE0QaI5f9><dEjHqE-&uS$jK3{9*I)t!7S$e#WJycnf&Ybf<Ye#vSa)3Ez&>^
z(k8<mpk`~dDy(TpM=a+qu-CjKHb7A9do*L^(#W#s-Rg(No3_!0Mnt?rj*i)7G0-Jj
z0L7n(bPwNf(jREt-jnOge<hT)8&t~Ohiv=q4bu}Kg`Q=xT;!4Ek;&lc({5t+GN1(b
za-CE1RRZyZAUL=qUuu2IOM@8dwa-ge{2b4U0SW3UcH~4@tRwwyL>zEiD74ZE@qU@=
z{^`QLWJHJ6$gR->7~N>yiBH?3Qt(~Hp9=IqH2={RXc7ky%(uq$>GS2g4OWOxD^u>R
zX{dqyVEt&S38w^Ne)hDl>4amlji+!kWj^3y08=gDcnaXV?V`EJ@i(glrr!X70&oI;
zOYUUK2-FMI7ob{qp0+ULd=C})t4uwhp57!Zjf!!ik`~bNG3|w|hoCoQAV6Bt+`@2d
zd)&A^3%|oB{w6!it4PaL6bq3XZcqtgmOZMudM*Gtl&l3l4KZy>{Tf0%-N^Ya;VTN*
z&TzUT+#=k%{k`P5M`}7%L}Lr6W35%krCQVi?P;mC_h8{i;Ex6q#HT@bV_KN6QL3Q3
zpFs+l5D1Lc4AgiaGh;Ag=x>4`nj&6EvvA8Kre~ee5QUay^NYDMEG(5QQAQD`SvM&!
zU#hNNq;MK?)D8s&=dSKZKC?Y1dk|Y1p|GpyM|#aCXC_(JTh^v<{yG3iHq(%v;+S}S
z1e|7EjnuODOzw6YOzx5!OwLSCXN62Yy<3BFd<b6*6=dcr_0_r$DfIi(#YafLkEcaX
z(mW=!apu|Dx25`xw`-fU7uZ<|e6Z+NE*aK>GeRngf2{`?+_AG5&SXk|gr5Q@vMv5i
zJ^`~GD1*;1Y40a1Y*+=vy2|N}9B>>5#IUcHyS6QXQbD64k3g0f=HtL<VBN%aL0Xda
z<Z&<+3X>DtWs<mNdL%b(C0=HCk>e}F9L!N6I1=^w_xl(1!xpxw_YMzL{+N?9E=a37
z+)IPT4@NJqe_fZ(Fyn4lPv53%e*T#?hjMwmeWIMs+?Dkr#OV$Y4U#RV=|%tLWSE%7
z>cn?cg>@etTK&S1KE2%g3Zba%6NpD4%2MtoX92a1w5@;}*ya`s^X7+De>~z&#!yrl
z0kO}5Zk&qn!;rqKx?GIG<3QeYyvit}J+$E-%HzaXypVNw;2@9YtM_Jne{0uRo<g}Y
zUv9Si=?dgmETc(shC2u<I~!=hU8uKj+%MlB*<ji0zYzu|<*!y<ut3hD5z4Wu5RVcA
zXT4A+dRk%qX@deD>w2En{AS{PM3Yb4LC63D@IbA&rBhahe%84Y86E0<FX+w@Bx4or
zq7b#IJ$HX};`Hjf%|@*1T~3(bMbG#nlH#M>+Gf(aHq0cFjzIFztm;Q48FejvXI4kd
zTEa=ArU8Odv|CfhDCN=B9D=e(RzkX)ym2!0>@fw-eWZ&V_lX)H!?#~|T?s-H6Da_4
z+&4TJy_yu`iQe$Z&0m&5vr#7w!^Rif^U8|{53?f74YrEV*@_0m2~%B1%*gv7^?GFP
zi!w&4;48)>DXzdaK}RNTE*h-x1j*@)2oB8c;csnNF{0ePEW2mx$;|m!Q}?WyA?9U5
z<)M~~e&!kSk%|wr5KzzuiaCnNkS~DQ1QHbj-dfH(C{=xO+-W^qI3$i^y|_k=hJ)?c
z;DzoIp`(Ou8V?rf2YQd-T0H6aNQLd=KY1u)eo|M*ZbEaD$9a^oKacGfJ}H3SGMA?U
zj%D}o;<<~1qSUF8U6_&`Z&ED^ZeS5p#QlHrw+V@H*$YIDEi&q*J=<^0R6~P<aOuql
zC&zkzr$^_uT%SX%(Gv=PS%D{Y9jR&VRoS<ym!X!*`0C+WPPy6K6so6*vZ&5nIgM)b
z?fv|&hUE&Px9oRvD(fmJ2xdwSa*7Fua*)1?<4v+y(T$y`DaWbMcD$1{D{f0(oJw{?
z0q{frCTE+Zrko<qxQ?qoxmXdq<X=$8;&W=B-~sRz#R3F<js8TT_Yg})tIXA6=qzK(
z;1>d`Bs?*Do(MhejSK3^^Qr|<pk&*t92Y;?JQbfJ#z)v(gPNy+eeDyhb9~;<=~5-F
zU6isH*K()6VO}eqw^H%)ZJebEV~H`={2>E7L&wIBFA=pe^?+hNS0{)Y;H7G7ZcRSr
z7y!^^Yih72Z+tY0m2-6MzMG&Ud5RlqKB;VMcfq?syS3UbPRbS76`w@9Kn$bCWP5sc
zUa}sx>tLj7XqqQTE(Gz%r7<aLB+M>nVZSjO0Zh)PvMB9k^i}ET0t0x*f#yu-IdAEz
z2Vp!#$)3{I1YOKh?tAB_#k>fysvL1@p#UcHNY$aKLe6Ytg@T#0O2fIfIRw|7)3fwo
zDDY(5*eID*3z}%YhmG!e^Q2ZpZDm+bN-K0niK1NiRCuS#$%`GuS_HkNr$X2jO7+ar
zEJT-vL6+$0JG&wh`^MHoDwxQ9%my~S2-`T@Z_PVzZd|ahC8%&6{w3>Bvw?w*x&XA$
z0X;7n57(h<mkK9EGmxOGfSpT&P=SF>%uQa}gi>%{2CsZgCAcX^*|#*>eu*Bz*d_v8
zx|$B~K^7;u!yn$Kkv=Z2=p^ye)$q-`4mZZFGBaR%3hLVZ7Jh2#Nr8yZ(%^F`E5*N)
zrH>vf=76FhxpUE|a3otc5zTdHNC1!H^Nzb|7|;#Bw;|t~{bOn#elD3AM9kN4?nB_b
zPq4TG@3`(*=O`xu84Sew$`u#E+IVglwVZ3JY&#feZAvUP!3Z>{Qz}ad2yO!cVSGI)
zr6Ktj;?*9$Ixg-L&|6E`zBIa&Uu{jAEnB5tOgiu*QGrlNXu(@^-o_UA=Ku!__m9?1
z9LHL&Z^avJpHXQ|pa++w%$Qn~t{q%g13dcKxu}u@tlj+TW*3#w3=YIw0-mfi-z_RZ
zd}_5mzD)4?5yK%x@5O+wmeqj*Uw%;|UgpHif1;=wxT1y+jOa2O#&AXDF!(1hQ9b<f
z1L@q;(h38(46bxJjr-bfJH%%XYka#2yuJ}?+*syOTl*O-iVly3)a7qOOZUXJ0YM)6
z_mHdtN*9>c7q7fTO9zSdfiq=DN&TPUnANm=Ppd_oQ_sA^x8OzKVRvpPs2m;glPb0E
zPAQ(nF*>7lQk$VztUbbV#xo8t6?Z&RXxicQI#dYiNtT%;MXx*B08nNGow92kODXep
z9UYg}<n1x>^RaV{OzJdmD0`iNg~=GAh^nFOkRR-SyTy=7v=EU0L=!1uAua<GgPY{j
zV?xM6k4xx+fU{E)C8md>(w;O7+e(l{RF!bt|02{CqaP(Lr-tJCgN{^kPp;t`Or+TI
z7ZU;=`lR%QjrmU~5kT>!6&U;-C3dtyMo2F&M7h%X9t1}uf3}MQfg3e;biogLJGWcw
zK1}aVs2?XAaG+GT6eG%8*?>vy%?%1R4jFYd-fgG}u>+r{UI9NZXBWSW_VH|OgT1XH
zriPtlDPGi>TW#M+b$2FCpy)xp3Ysj4*BEvM7+7UArgDH_0Dzp5gN*X{D(%}dE#Q>K
zknqMJj5dA03HK*rq&>#+sClVz5&1Gc?_rSo-0Z$_*n@(NQNSdwal&GFfnXaEV!=40
zjnp8IPJiah$5U?>AgoMXBSF$sQ9hjhyS6BpM&T8FE1|*>S`70`TwqpFFAgxJ98VJp
z!e>wIxW4^c7J$S~@Bqi$S-KaHrrY^-xP-&N#f5f~T4AxcvXTNli_TWPNX4dUPs+HP
z&5*P2joZ}8-awBdRa?G>y!HF0+F?KfIZOu2$tUVIcm0d=kMU>S27d;s#e)U)F#n4#
zNA;bv0<98L5G9e%>i3`tvu?|_P$;%`UC>_RPr91(Y(OZC><~rArZa+bYY0&|{VyTV
z>bfO-eu6e2d#vCG(Mgb*2207!v*`d%@B5CqE7#WKh%J9x=+mDCZ&k8at#8^V@W{eD
zA@vX5C%%1Rx;wr8#c>Xej540nZ^56zKApGG+Iemb{FY%VIksa)NK$%-ersIi#qEmp
ze-jb|>jCVtxt@$k&QeBFQLpO%l=e(wtg}7+t~*4Rg}AYQhI1pKSegma!0;hjG9JIY
zzfZg#IF^AXz#H5Oxtf2qkx%^uO@^J?8|xEI33POoO(4oY^d>@T<|P^#F+d*W63Nq|
zQBFsiSbC;_VPs+P|EI}_cWYSfc60TzX7!T+K&ouA{!f>Yylvp9T2y!v&L3CXs&AsR
zGs0QKOB3}X46GP~0);{;V6^`b&7alxTOv6!TCu|vL(-^HF`8l}%9aluPt0Vire-EH
z!#h4v<82}{R~gortDVf|0&vmVaP%Q+1d=)_*qX|_|33r<@PB7u{+|#S%nTT%h>ppE
zMBbd<+9N-M*l!3X{qJl9gmXv(>RIVUW(3`JhGvn^=X#fUx_~ZH&Q>}Z7$Ov+gre{*
z*y4>Vc7s7D0ACHGfxZ5%0hEpq^;-#qL|1+@T1+e;3?K|1@iw5Ccz=f5^y=5!aq5BU
zf&7B-{DU+LG${v_=P<&1-&))JOMb+F#30Wq+d>fp^pSQG|09izLd=q|5H2ZVHQ+#?
zf5i(KxSqrKNAQneTrOfF{D0{_z<=>R^E_5oO7lcWIR}x*W%hGPV5NVMn?pr8)c+xF
zX#a<}xq3Vr_lBSWjh2+`H<$$r0soKIT!TS?H}m;`82=Gg)6+7c_;njl?LW$YXbY39
zSw=5uUBSA5M*Q)w2EmxO%Ig6*M3IgfX*RIzKmOGqm{PO9iPT<4G}KUGA!9-GkB;9!
z)laveK(IjPdrn(8R`g6BSV}=0uw6mF3(=s4G7|)1*#;~Y6~8mm!1fk{pYN38gn5KX
z1#SqQ#eLyW2|Zz6?Fj_&Xg%>Ig;5#AYeap>{dYh{|7h0$6TsC1{qM$)ytpXAQvbar
zHPW<E+9|>SofNm4Ei)@k$<(|sa&A9c{D;0NXC~0y{fk<i&-h>ZMs#WHo4!f-rf*h_
zvgFQ9(ls0uyC#pdMZJdLk)dQWos)lk+7pQZA&>c6@oDNB6#im2SZ3*X5(fa0xal{`
z(@YS(cmr1h6o&?<S-2bAkb6`0X#N+$i7Ju(Uj!$($@f194)EfC5FB5_{~|a)qM-f{
z1V{Pqe2g{_?`({eY&AKFaUV45L}{VZGj~a&V?4upeS+rJ*tmWJu(8Z=wcOqI?FIzh
zvYXtn;`;7T*@QKw-Gc&95c294=`{|M8MF6jybStfdH;xv6^YlBt<mW28cpl$CZ^3z
z4mSjLSS&&mFJXqVvnTP<Dl$!abZ>w5Q8vg|_PCef#RRBX)?`O-{2oN0TQ2=8QP4o>
zyYJ-3i5#-#9t?LqV1e8U#fYyIeO5VxP+ExPbYPC!sD?CT7B(%`h7fu&>S*eB@lUIp
zkxeWkoJphY0ycHlMpN`JUgz;nn<)Gn6oIFJ<4sKc|L{1T-#ku@JaxUr=x%O|z4_l%
z7}xKP)?O$csDn^)nW%Pon{#S>Nrir*NX!8t5{VSOL;z6`<1GQzYT%m!LmC?`r`$&`
zN@3^$hsdyqh;v><$tin<CWm*Sx%nygA<IJTT<SR8+bDUUX2&YDLCaSn=$at9Xi?t0
zv5s9Lj=>sjt3#8w5L>mXy32Au<$wOwDF)6;b;3((DV!TY12;*Bj1top6yNr*N8m%z
zery3az2`eGSRpdbY|QovQ32@>HcuXZu%TNos4EN&i&1IL>lL1`wI5H}wm4<J8L$O&
zs$!o7rPGS7>4(T!6-~2|PeQ7`i8UR$6ZlFWN1}`cV|K4JrME%l3xk|YjQMJ4wG~{X
zy5#T@)&(?LSo<WEsepFtBL^5VAju9pMIS(9Hy)D1KMbV84=jCf^zx|qL3}S2_qe@G
zLa!>MN5p8+VIsNye(W^Fv0jA!+^V%Hnl5{vPm=<F^#g`PD<0LACb!&-s>a_h;#hPl
zaijnE9oQ+NC|MmUYm1O!VWUF7`%4KhMX56dsV~8$7tI#s*?W{*lq)B`#b{WN0fJIG
z46smMO$8%quuJwB8AExsYKhMMvHF>ggFeKsQauW(ym^9GgCs8E{)hbFR``+z?{#ZD
zHXXW;R2H2}oibFJYDFlT&#(EH`&g`Mle&=L1*}G?S{O6v`Gz|!yIz@W7xQmROp>6^
zSS7apx};3vk#!Ami0Q))2I@81BWl-AzB{%J@taP%!?^IsaiLOt0#6nR1tG+Wu_}rT
zhx}QmUY-eLL^iCa!!F=3J~Mjql<X4<g7s_S-1*E!>cHecnLQImO$!xbs>!9;E|nTy
z<fA1~>EXYG35>5IpmLT1mSe|P?o=&#JQRcg5E*n)qc6*=`{Yi4@ALz!l~oKq^Ebt>
zxrJm@xpCR+>(*#>vPa1UQ~LKA!SWZbb@HoQuF;nyMGt?txc;SknAkRG1Nokm552~}
zMqev*1%|uR4a;fEb1S9agh)YmZh}92Y4?i3WKv~qp;P)!XD0pPZB5b$>28kOLxqVZ
zfX#udhu8+V{{*IWl1B%G1mlV>e+_AQJZsgwVxzRbKg+Fg#=5VpVw-%nJ!1p{MUFs&
zPpZ&>rC%Z_+CIh%WzGO%{~sLKA`VHC*=98r*Mg#J(Q`D_k=52rnrxh}KP?BvNIKN!
z=QuQ245Mgr6Q1i-mUq03n_Tr^jOip`OIGwA$CTBBglq@2$%?7|Wa`EbxUv5H&qOW*
zu;tXa#(->2_LKMI)46#|`hsR1P?G0}!`o7k*5nd3(Zxc^YrI9t6ZQ^Q`Jfwabzd4`
zXefB)F$%Utkf*RJ`BbILc?tSu%^!++!MF(fZz4Aj-ODp%*N&3`Wk^YldtDCDgMY(2
z`!kBLSav}-s=YYL+IoaJnM!4p*SCM2$LN~>{78W_UZjNnAJ9NG>Xn3AOg3OwbJkQg
zoV-9X*Mc&FQ}j@;FLh2Hx<ktD@SfqtbKO*y5~GT9V~s%Znki*am&@O_E#~N`j6vjp
z8tnk&)_&<w-P@QjH>EPJNC9xR{9VjWF>>-J82&Q7`kjIz7U%pW8R<Fs<L{vQuDNFv
zp0B{Aefip@VVkRIUR!h?Vqk^!PsX@w!>3C?S0X9Y9kL$H2vsZ|qv205<;CQy(NRu&
z%nwT5W4tY`!z6_~n$b9BZinu=7nL&8qWlMs1}bgDM`VW%`h=2l>%V}^j4l?-=}-2y
z&K9yR%b1^2f%IJ|r^kl0w|xTw@cLLe60FD}DE+Bu1072T_qAEvZYa0QYM^fx!wYJl
zj)#;~VBOAr9lr!(vs6n4o!BMP!q6=Ha#6wOaT4C<8T{w16Eg32Y?aGo8j>OurTq>K
zU0J|1H@cLDcBK!|xsL!G4yoeT=gPFU!w(!Z)`_Uoe9CA^y;^+%nF9iM0lQ5t7Z}Vy
zuNKr;OmpR7DWpmcVa+Ule6?`fzPX7QoraRl`?OSjBa~VP?V+n0y7TC-ji1A|@ydRJ
zgE@wG)q3)~u|VKBC$^Y=y=55vp*vC3xJ8}E$=BiUf)W{;h^Bz&=rGP<wAzg0Qw?N?
z>#($lj(lSB5+;<FVpeFJ;%HL6cvg}c+q#m)g!xu?eG2fuf7Yz*(q#(GdaYSk9&x^x
zK8-6**cLM3<WOlSV>~>3%@d+=DGZg?te6AChBmsaQ`?hJC){cgtW63AjH~@-%!VG~
zn(PeUD@V(80H<DkDx71~%;8!(=ltVB(jkGFpTW-`l+j=z!*lweHkk~!3G;iuDI!Ov
z;|qgsTNH~a?2!K`=~iU#iU&(LNh;|$7}#1!*3`_5fKKmtJ-24h?&c-yiM?B`g`p-o
zmeyGLIbqr&^}SR;jd2APCweGJIr(dj?}tqObl)kh1H8BpK*imXeegyL1B^D{@xCY1
z_F%=s$L$?(u@#_1TqU)0TC{@7PILd5Y`|E_x93V6qju=g^|EQ8uNqkjuj3nj;H$K)
zpUS_6cuAmUsM6H+;j?u7u*=iVXk5j9>2Q{@q07mcY;cJ;8AjnscA%pBD%5VE=d;Yc
zQwa6q04!-P<tib*%yN)snS^#FiHIF;rJ89w4JW!hJQ$?bgTs6KDGwL{vli1L5Zy^C
zRWuE<ZPlXjPH?ZRao{%f?y0(sH1--&DwmSk`tKcA&k0oSJUnPmE6~D^-_-ZexYhW0
zxQfJz<$B-8QQjgJrFTdK?~EAJ*Ac?SA3g5Z0{Fz{Z$1MdlC5(p2C>!3?I+M>)c4Yy
zpCBx?nAP@LOHKyk$CBEL)%VjlPR1{jM*rd}k|ER;rHcu>%`F;#fV>2z^Eiss+?q>S
z;}7TO@1PQ^QJ8t})5uTH$|L85M;{tejUcW0UMl62*YY_IINZk6yzn0SgElDqKBizr
z1vFG-bP7i2=A}R#aJr!839F1MlToPPm!cJ4Vj+~n>gcVU9@>M3L|hwPRA(C961%dV
zD1MOdOAun;XxNJ&EUAR4VZRa+vcC4)x+AxJ^!y4cFJl;Pu4<Sh%s|>mnwR7}WQMYq
z4i>s|r;?dyL1bZ$ZpSq<{wkc9X0#w32hinz2f5_Y^QB$p_)p1X+|v`)aJR^kd-W(v
z;F>^gmAUJBKKt23v)Koi><EJ|HZ9~tICk62Ni75L7-?Aw8uVEL<}TXfwwO$>;9YM*
zFD26M5h+o2vNbldE7u~yw59|paB&0nZr$cm8#wn#?TSq)YCjbLxpJGOaTgXBfc#i9
z*X$H_HxGi?_U3)?XguoLIpG+_6J<H3W_l-In{jGWQb92*D)hlu-+&^iu=dG%%VvI7
z>;_e0tf;WWa_wvH5Q(;N>811By7b)5<!e+utdCL>AowF8CQD$SM5iBnnA|;HDQCsQ
zG!qpB_AT-@yxJ5jjx9`%A=WlE-xk(OJcga{cg`tQQlRU{UK`AZXH3B`hi44B$;jVZ
zL;SnK{Ju52FzP<B{^@(^_R@yV`yt0baESfSQ$ozig{eYTj^PzE^vya|@5(OE@s!R7
zH(${!t;VNLa}~>Pcn{gLQDTV9S-rmr?O{y!GHx=#SSfNHTiR*?hE|(=I%O0SDUImH
zIvbV394V2oIqmv-n0(Et(deXLZ2mV}``4K!LrD)Wu=P3qs|-JMDfS%dSL6H4h?{1D
zB($%>G<XiHH!(bhWN}3SVF*!sF{Tkhw@AVJ-Jzk3@FRsNcM<lGgFi8Y_XPrG+$Sp_
zZRsmp*t$V2>~<CaYEIQMKIc_8zU1b|E77VNZf#dGJ=9{I%QEyx+pbyJ$XIZF)ICq1
z;Rd1dW4qZOf__Y<3VVHg5&{Tg(D@Tx<C-e?F&s!#u{|5|Ho#0;RaUiJ_^4Nu2a>Y+
z;<QMCFr#zT<Tc`UW<g}$bamv>dBj;}P8~-re&!N~yhNdZ*@mpq&AV?aDp6B8J#G?o
zBu>mU(XFOG93UbImy*$=xf5zw5XMqr!;KE%v%M;F<~|7~d*MY>$Z;QxRr1rd<K9sg
z0M}$<hiR))i(E10L5k%8-GcP5fxrErqqd0z)K*<A)9<db^09}Ai{OpD(|}vZt=79s
zM}rU5`UYD7E{j}!ywP!p<M-H)#=2<dx}2i){Yf{yPjtzwmF)xtH}Ll(<kMc@T%ola
z*fqZ<>Au~o8uBV(AI(SNC%2cmc@(H$_#c=_Zc5yTNh1va)sa8F-d`|yq7(|F57(}Q
z@glStfmI@i#*sg?*=ZZzq&+(GbjaA=WtN)j^!O0~qV$+7M_D&wI~d|U2CmAuk+a{R
zMs)iKc~Id6Qt*bof!)?f*)lkpcv?))F~(sS77DKOxEN&0qxQ-|$w;;7aO-R!kB}l5
zn;7s*u8vbGfpcfNzg}IE)lN;eMtFWmf~xVM@e0#NKMxpDg=MuQ-M-m8G8VY3bAfxW
z3mpzXmQgx}1s|!=(H*6G$7{GCDEY7R&Y9_p)1CR*yT@eya47sE+DKie$bWv};}5q|
zcJnE1=}Lc7+QzD!LM!;jXxmv*p~g4#FX3<0gMuSk+V*BkbvkDQ2<go^&rv>f+`nHV
zdl1RcbwL@((6?}$CEkRO`Yh;Ky;5cX-T+O8yth)y__1kgq(;U<+_@h1L);SflL;IS
z)l{5T2Dw)IuIkc7GnPGPZ>A9_>Wy?gA|_U??s-H19NO;B@RN^5ss2HN0In*#w_a_`
zyiUsbH_e7s_DGeTyeK(|p(4vEU>P@T?S}Oh3*dSO$V&s|r6o6uj>?S2M%zsQI(56+
zc)Dr{-ZTCO*&S>9W2Ovx)Y0HIuBw6`5_q(e_1n=OPQEIhQ@3}CJlA?9jz3oo$A9oH
z);E_uI%)xS+4HA8+%V=Ee`TH#-+w2`FVo1)5NI^zerC&)#CC1twT&CKiqQy}`oO!l
z?_J(xYV;dtLU6skryJb;U2Zo8{0`7G2;LlD3YyD<f`=s|Pw7@cBucItImk#;v>tR6
zgqi$3e)lNHgIhgG6U|XOakyvYh9tcThL-d(P;=qa?|Q2i(W*1{>sT<k`sjQ2ar$5{
z(E&Px<rmLBfTfsLMnu)*klG*0ufrTouKXk)TQ1XVjf7PnkrMxS2-|8v=1z}wK|%fR
zqOd<UZWL)%?!tgw%muUtg}tpjzP@~1T-<Y*4KTTfH!V1e&$_~A#_e!C5T$CN)@#7m
zNM;4V9+HA}5~y>O4`-T7)ayMs_;sL<JQ)58?qhz(7IfzR+GcmM1r9rb+Ioy<U&-mp
zy|1r>N3F{g^fc+>XWcWv^Shl$KDx{4^3e|qC&d#@#QeMm-{N!51t%rCVf`loN6}E$
z!9my8qHe}>Df|P)j&w_p8XR24P~f`g2SOJoZdC9<)#khPVjqSOMA{B`Ru~X9Y5!^X
z7fVUEQ{dEdRyG3=!QYn=@0J60R<_9cqaa7#58B;>d5YFTM($q#X%rve%23DQ=3fVS
zyISQ`28tZXnQc0D2Xbd!)NiyR9L0an8_y&?n+jPPHWmdcTGo{7O(N?$S5?=(eB~TB
z30k^U=g(my5e*wy*-}A;T&<QaJ&iZ;&sRQpn_hREZby_8|1<@#y!k|9%0H{Y+D{dM
z44!hnqGe_A*iU!^Naeb|Z8phSo4eR<f>KhGNB*eMsd41LJ+EBan;j3=4H$7hWB|V(
zfRf08v>oZOf8K-Dj(3z>5RO5FX_i0%uY0~cPuq9MjpHLWZ^uuFk#uJA+Ht;%^@(ja
z8uvB#N2e^#NAi3{Ve&<=H}8LiK9PYL+&0ihb_Sum+Lk{97Ne7=*Yncv;RTe-H26v+
zd2Cy)SGR|r)1)~RPCGxMIK=AgHc7}4dRq|ky{GQ8_HV1Cb`>>3xky?^iiNn-*H`j?
z)yvFf5}_C_IJ@aa0fUzBDG$>y%Z1yDlsGKUB5skKDfuP-lC6W!NSmc02*YA66P7lV
zC*l)ui5i9n;KAd)nL}K<j0v4?+@)jaPEFkfTAa4oC~6_Xke$LER8i804Dje}H8xK2
zc3~T2rJgmkE+UN_Htl8Zrn&kx)BhzfPx)z}Kv;0wI}e$Qp8({xZBMJVD5GMnhJ5;-
z62dU`AeL-;i~_^=$-5u|!#B1w8r+N|rQ?n=p|8dQIGKWyY#e7$urqD~A*;bL{{-87
zy0O3ASsR@o-@exIetp$wYi^kYlFD9t`=miSc+?EeAJXd6?+EonaZmhgUUBY3{i>n(
zg5q+Lf{_dOpu$WVx_pkFtNiE=sWqRdomJ3Gbc^PlOq_&16vMCl0-{_0>AtOiQXhX6
zKgw<Z*ww@0L<@At_3mV3;NF5rJ2%!POO~BlGPf%O89OIFi~|lBdjAbQUAS3|pR#8U
z4w)rhqQQ?`r$T0J5lfc^n=S{P1%n~U&aTZnL!nHe+$k^M@)g72A0EIwc@ysk(1+~@
zh95v)vF90@=J|{QD3EBW`=uLi;BH}9daBj{q$gidsDBVAznBX1=l1b%qr$qcIb(jD
zpoi@ZAdAa`n@S>1c0{yJ*@g2h!Z?s<x#)@xj+-K9n)JYh8V5q_N#3`x63(~az@;3#
z^g)YLB6;uA0y4`$Ki|+_#FcwLaj`y;u<*~*Z=WGO`?vIoc}*i1kRG0=A3SFp#8jgJ
zoKtoNKM+RGJBMf-hG-72Yr1eN7;DL(GH{8ct;`=Hs>w?|=^Yh_v}Kbm5mHQHw_UBG
z!gw&#q{o*%m>V<&iWgpe`fTMH@?Z+LkXi|w8)RCuowMgPY0aH{|Ba2-`D_1;jriFI
zAnuY$`QX(Uk|BP?{lxU0%jstaoIK6%v_^i(Za)g}6wd(Q7ceeb8#!(<ff)!+m?>7a
zX^1J<0+?sCE8l3M#rzBtA;NG4wRjcyIPx5yYUrIvz{0N;bYo?Fz?Vi`li;#;q9<wH
ztFCzUEv#Ec6Brs?3O2ktT8Dx7^mJ%Z-ucNv=DT}-PvAyMYFpFLJpc6ph+PZVQHLQ^
z|0VH8yYAARs;q~9`R0fl9wdGfkKvPVzaV<00v*YAOUtb5m&T}3MSiTGyv~)*CXK8d
zdfB-`6sfD~M1G+53?nL@0`bK+{Jfv>*=k&U3%nyn;fu)lGgpOikzO|zZt39R{E}MV
zwQ1ILWXi^Nw0f~%{Y>TofLt3SdKt^<sf5RKp{!u{Q|8pyF&yBquZWWMUWLY8UhxG_
zmuBf7%Ttl7m(yXA&iR;meLwKZ2=VzThKtLxdf~y18*3fO0BzgsWk(jTEYqGDMtIV+
z4)Vi_kRz3T>=I?bcMrKuhFY@T;v(C<7LKSkbr$we44x|Kxn&9zFiBzlU@aF}IU4q%
z{K9|57KQIUO5m|jTUz|vFMGb%Rg1jK8dO?9VyW4mf0`e;y3T>jv!Eh|tGJkp$OE`s
zSHf&&mwRV++Ks#Jv;htheZ<yR`ID^mfNzc*p|E>($8DAzqY~~oRA5Ija7bT4oX`jI
z59s-xlzR@~RzqVR5RTp-@EicH^)V4N`-+y@ZPL&PU|JYq+7p(8s%&3I1K1%Wc6njR
z1O3v4PC3I|1@~T^k^QYZWjpNL=3MEU33sV$20*xk`o)zKscjF@c#f(?k;q4sh@7xP
z@Y3oz7eRM;)kq9Dx@uIXlW7s(75TN@?S+4M$B-5L!rgKL==I8+7hbqKoC-<>9Y50N
zrJs9g*Au#DB@B7tCh!<P8VSey=gs87Q9b{8EZKvXzOIx}l*m;$FDC<ylbN6(IVEf0
z{k$vy^sbIcDfC6-1*!kt>ae8$o#xQgcyvmTZP*$JNvWRqHRU^J!X-Up5T*)`=TE6A
zJ<|yg3PW#_o+7{@<7g=8wnP!5Xb$4R6RfBLVG!Y<zJNLU{-TH?&2^L#))MwdP178H
z$Bj3WYJsFMh@>*}+e<Z{LmK$Z2IJqw-x>tJNQ}#k<`^<1W%Uz#^>6F-`U%keC2(gy
z2R$((*X8r(Y<sT%DMhj1A!ga%hup%AI+q0?NL2^vEd#CO6*SE#gO6PTPQ2Z{8MXZE
zQPZ`d%m=lmyN@wBM)Kve+u|%j#jiiTh!Un#2Wt_#gE*6zN<hl@A+(#lf1X~$Ka0BL
z-^lwActaWa)u|A-v|l``I{X+uMaZaZ)uc*9Ww}2t!hc0uN0ReI#@2G&VSU=skYEC6
z@WGCLN*W4KWZ;{%iv%RZOyQv8RbLo{mzXCe(e4(4uoD`EF)@f#XY%D&4kds%j}`8C
z1Oh+Yztzgh)OBYD&7UTEy6n@rFhV7kdZZ~Q8X?C6!Q>W=RRlP7H%?YH7~HI}Jq$%7
zuTY}8vTAGcOT}<Wb+eP%*8La;XVwBto3U6Hxm}wsLW0QZRxTTZUM^R|PGytjOm3KU
z(?o~N$+A7&TfV_rclpJA5Ut!^_d@Je^N9uZk=?%xp8uKj5*!}l(aQmUr8IWf;(eAQ
zuk&Y1bw1j_5Y5zY4G#966#up@;cSSPRWz+96VR6o?IQ`|<>lwHjY$~pMn?mvaphG0
z4GqorD-n;18eZR3Q61{iTiDwwgPZSMwfe;w4m*F(h4ou7$}(-1QJQh*q^4(@?rc6?
zmPjONbBaI7d>4HPeRW~s0F9%KP^(V5U`EpDC`QZt>+RuO@=Q30;6$2VaZvw|-J|7L
z(^cl+{!HZU@ch>AuXg;D>CzIw6eA1hDt+shgGhaNSBAE>j@KfcjpBT-J*`zIFAhcG
zV<H`Q2zC6$N}ph{N7wJIio8^SUG!O5P2*uJ)oqPIDF0#??23J?tPa_&rqo82xuFdd
z^xH>z==g6c%hL5BCK@7>2666mC#)Rp_#V=WhmbM+6d@TVy-KO8l(8+q6Fyem^N|j%
zu9D3INKRi0_Zi&<K-F+l6-TqI6>{^$6|B97Hh>N1+F%r1iEacIul>}xv7vBY^CC%D
zo&9n6^dPc}0gymyjqCYo7T7TGuHXQ=!2)0b)+h{>q5&klgJl<Z`WM6_=N_bXM8x~2
zb~rP6&M25YUJ#i)GSmT{Twdf@>)42O%a}Y<UPlB>5G>C&k?xEv&a&ND246X|NR=!6
z0_T-zYV3TBcH@_+?5QpCHj<)#6&9*C$>1QO-4#%fOKL+Mx%jVSM#&PnlclhajD+8-
zvI7#qEx}8Wmgib?OL=CnQsZ*nOpScqau{+Ltcj$_q7y5dp-chm9)KOX_t3s$xtktH
z#=GVCV27Y@p$E?(`}KPmGw8(3yEG{lAEq%miFL0A8|~kw<i1J&U9&F=`aRV+A%h?<
z%DC>r$ZfLo>HP?104cVCYN9RnaUdH*_3nJUtQ`**24*v^&fjtEK<$Np+1;lAQi*Z6
z3B@-GF0f9~2`G6*h1sd?3?P}#WXJ8T!pu^6bpS#%-}ac#D{1JI0wI;s<d153+_~&v
zpWXI7$-L<99lDEI@wOlJJ*eUO@$_EG#(tSC?C}_K*23gG*7Ig;k@iUJ-)YIFrrUsA
zv#g!^br=%rLs(5mgnsz3NFg?7c|Tnrc@qp{<)67x3>YWb{gpb#n+LK1)rJ70@3%Cf
zd#v}oCb`tEXic(E!iRLP6>|YX<5J5frlnJXtYGqvP_k@acxerv^P#FzA)de&t(C1U
zduU!ZTdBJL(=iiiPQQml92jX7i__dZ;2sQek@m$@Ph+%hOR0>!*OwBVz6e6Ybw3tA
zkL-)rAK+QxJy-ZQ$PW;NRRbtIHyhQy;hxw@L?`(05yZtOH=RllC^{}88C})iS-nj&
z4X>}{^$u*6x_vo&fa91E57*Mtd6w8)6bbffY3FU@cuODb+7*4v*QQ;v9Ld9X6#s0r
zURfjCO~nX$AsqmRQbBx_@pLRr)Ra-k-wOl_2ZYv6p-3s+A`%Ou@4coFoU6(HI^mL;
z{NWZ_#wnd%z^2#z@IYDzX=}1GK#I?i>?UnUNJ|2P;ogh4ylB>qE0m?RTS8K=&YY+U
zXKS$1C>&9*2X2b3RiRL_7%!<dC=oi{N-60SH|ksmUTmSgfGrs(KQQiPXC*py`~;(g
z0eIuVu)CqeX|>X@VrtT44yqQ$Wu>9W*>QIV{Vede@<7pF(UC71b{eH0CMK>Nf~zv+
zMG3!Q`y9Ycmrs`GrKM{MsbZ~CJgADhc`Y(7c9vILW6EGbnQ0Nyp*=D!Nw(x%kq<+>
zE|zYu-=-}@f#=)!!96raeha;IRbi=q5AX%gDq9>Nk3^-~FYhzwv}~Fj+0s1HZ8$^2
zQaoWX+f4Ru<iSW=XPUNL>CV$;d&Y^vbOYi2RsZ5OpI}0Z>HSRMFv;aGNsSSXyo8%#
zB8Tzy#Sky1ozm^GrQb#CON!8>aONIpx?o=Vpc&_d&Mbl~mV;RLo~-|YvGii93g{Ac
zdHD(-a9+*KEn_dcHf_darb*Ec&7c^swgS!M7_P4Wje(KOmhbIuKexLM<r2uRK=-v*
zq;?xfY@)`>lzO`zzJj*@7j7?ucB)d`-&1QP!K<%DdpF`&$qa8=Da&f2@SQKC<!;vC
zJ0<jOM5pliPhnU2vQi9}cer%!Q^4AdrLj)sXe(6568TyjiF%!frV2|%XJD(<#G-D`
z`>+7G-Jcoy7~PPaq>5N`bm_d`_=;CP@qhK+VdI8`bDUsnGr<$WhS?`_o!o+mGU>sR
zFVCm1N`#!{>aXJZSr1DT{G8hB9jv+C!*;a}QhzH?`q@`0W50cyGI!&R5CB+y!sya8
zx^6=sj#Ir0zIQOmrt)A_9?7ydmpx{XNLsGrVuez^@N2Q|AIh<W2l*G{Tjhzhdx{EH
z=sj&AJ?xtj<}&|0Xr<7FWq3UQGaRn^L;2lD#S3ZBHue9b>K&siX~S*NPSQcgPCB-2
zr(@ez$F@_kZQJbFwr!(>j_u^;+vD6b_8IHv`=?f|HL9M8XHJD{1@b+*%e+MTf<JC1
zT&8N8uZK7P@DHR;4_*1tvg%z<MNff1F*iwG$aQ3J(Ul&1KanHFcCxSAE&RHF_Hy_l
zpJ_w+``1OFC`%U)O_NLlybcmnKVuf>?G+PZWr@wvOjp<p+bW-#)+y9q=OSXiG1#(t
zB74jIH@mT7v0Iz-H8nnEQJg_utuD#zWi&sTp9b9pEEzYrOT4T>5_o19ql7^3fP^gG
z)x_8!iH3yEj)BHBJztWUG`NilUzr|am`z(bdP%%nP4svv$<_dOwJ8D&lGfE+F{gnh
z#?i4$*Z+AR+LkmB-KS%dsq--UxZIG|Ut3q|0~1FNTR?R@dZHQMDi(9UK?t7zuDd@q
za$6X3cKVOdsdHP8Ar8RfP|Mbd9sXbWeZoIKr^*CPG-U>U2YIA2jYBz@#7u&wx;>Fa
zD#98`0rJ8-F))s3Sq@=sUk3;cJG1qce?IeQ;X`3<z)Z51k&H$_v7(CRH7Pxra@s-S
zH%~7DI(*+LXzug=&YH}mjZV45G*4VDjJZBNtgUmfZiLry8(!em<)E(fzv1nhqEA2D
z=6cup23Irh!=KmOWNJ@4&U@b$9o<<wQhlCha#kV>+)FG>#o_z@h9qv$H5tAR<p!&{
z$U~1Q>y)-&pK-L>ntWe_rgYNq!N)7?t)K;!BdP<Olk?<Pj4(+2zieN91w1b8P2fwk
zbDNir9FOg#Wx%=W+$HiL`DBOkV<hB1#=T8!I!3q%q?IYqkg}ek>_5IasKQTXt0a*L
z&;2~10re`mKYfrMz@E|U=@DA+?nSGg_zDp89;Gi%z=(B*Xg&=fx(E7wG56*Ob`iG*
zq}L+t#*$-urPB+0szwU)gl*|&Z?1YEnQ|f2Zd_ulLxC=bwMsM42F^R|Mho0oIm0v>
zvO5_gd6!bOpXU?5$iNkl@M`rwbD|X_LCZUOooL{>wFC06mJ*vJf+y~l-1vO4w~Q9<
z)I4EN{W$Sl>rN7OyEGAu%38Il*k?ql&)seX|8(tEQAGumd%<}9aOmxOjtgOp^Z|KM
z6NGbKc?InEU-mfC_B;PY6d6bV#&Nb0_L<{ua&HLs{z+Pg5^DxFP{2cy6D9YE4JVTz
zHcaah0uBFKk5kud=I!aq&SG6tSK6k52eo|G90vo1Q5i@ie<`hsjw8i_t-*pl*E&Sa
zCT^UW&ALomQl;O8s4dOdd{-?ib2#?<nBBTE2>~EW8ksplEv;g|OC<?43`F>zT*jzy
z<suqu)N|Ezs$)Ns*6*RmaZC<hz&{AtY2<kTCuXobO_0yWb4PE}PTYTnHstD9jO5*B
zVv$JV?_A%;ccTB2%I<YTCD$j_?cKHbOaW*4J7(h~l*YUwe!OXJ&Rzf`5(cMv57Ig$
zBpVQ-eO&5(_Xxf+72sCLU_7s^UP7&H=U3B4o`_-5MxG7C;Q-f6#^fDR34P>Q-v|E*
z@zdm=KJqEZ>em*6A6gxUp8B!3RmLMy?Y#?Fr*9<zC-pC+tX)6uUV6nK{P|1`U%jn{
zska%@kUg~ozIEcuWx~c(cbJf9xDDIw1$==2An17cFXEzwbCqHeRQ{3P`!^j;VYacL
zCin%_R}#Y1c#Zbm2KOz;Jn>R<tw>C|TC12)$(=kS`50UX^30w90Zsn5*zu1*O$R#n
z!tVezf8YM@xHeC(5uM<6ddf7A@m9`#$-bT1cl0R;**{7?NDpqSHivxn-NcO%I|)#`
ziknj`k18zhir*#fhrXF9)T~ZUx^CrYI=}Prrb9U9w?0~G%+o=&Uzs^zo+I8oIvGZ%
zjB@U%`1O)5c~s~}r<ORrSW1}`OcaNtF*B>1-Yd2eqo?XT8VgU{Y!xM4|BH8;v5h!p
z^5DR@k4C*C+jf<RQ?x>s03Owp={rzlqC5(TjD`_O3o~0L(k5Tjx4po7*ixuJN5`zg
zk%j4z-eWRin3k#WhIapSOGk~=PyDz*O3#DWVb1-w9MZJtXkhtw#s$tSB>uu=EHm?f
zG>V1Rq^qnhC9^Ch&KRpYg&Z!2{r%%bQ@XM#0ryTOV+@iLdKHtfmqG$>q6Dan{M^Gv
zpzMJx3F>jrno6pi#%QSrTK!|`q(1O37MYk@yY+PI&mES9^bDGcz|2muXh|q4=5jv{
zukzNam*nI~4INi-|E&;LG*5X5gefb!n9Qmg`-&B3X`~%Ib2QX(GRwBwV~w??ad!*o
zO6y);=Q<K2b$nP<_v0J+F$43=FYkRVryIo?!Q5G2gIfR(lJQl*b6k`H#2uwRiqCy=
zQ{V{x^8rFX^mG58tv>uHkK008i_)S!1ZfXZ@CmSj<}JevgpSC_&p1@FiIXj9?M%8N
zduBtD>H{#0d)?{ZfAILF{;)LQ+Ab(^mcCc8S>o2_#}r#3<v`vNLjq{@$kyt3pQL$p
za|Wm1#lsL+B}#x8)L)Ke)QL$}1UAZW*%WaSW>pBn_?+-KQ$#IE{NimBo`GO#9by@7
z!P%+;0P;f8&;Z4E*(|R57wq!mLvtlw5nBgA5pw38@88`OWbDHgGro(UJdoNZ|2+<+
zpr-lCeqQEcIw#RnhybvUlYbve%tx_rWK`5>`lXwsy>sT9AWlH>(e|Rp`Xg6=$aJaD
zbWQgTG3LQ!l!($<SsQQ#IU?i#yj~`pFg0ZS80t<n>hBFr{K_4f9J0V}sA1EJ3L33C
zadl9m;n|c@0(E5Geq&zQ67Fw&Cv@g5Mm<zrB)iV`m4_#4tN<#$><lgfE|UAq0^?u6
zA9<CNguUDHtrPqFpKX<+L;mKQ&OF`7<Tm~2q{eH_^1+QVkNu`i*6mUCpkJQx&2ycj
z%L~FaJdSRE_ZCrAg`6J0B?fz)dh2sox<;!O#d*YS%A|=-RX%jA0(eU{FHNAs=jMwI
z?&Mp5C-**F0<f^Ji1MsnE9}Hjb`YBRA7M}wl4efp++95RN70Ts?b)+kLZ_Md4$=-D
zE+TXl4HDM%e<Rs`;uR1_9gpD+`#V>shCD=w>O)Y%fr-q->ZenwW2D;&sx5annO9kr
zv4+!Ry{YaB3TXdqV1}GpUiGghR^vPnEnTJJQu5@H%;A2l-Y86fP3L%?^7p3)i&)hq
zF1T9|yMp=H)HI(G4+HN^LHP|(=h(skWA=G#^*Jb;@S}PW<{r5$NxV^*%GgJP5~)#p
z7lT=9;gq_~rD|*DfO@89vAa}TFy_Qqt`L5yi~?pSMU3byZaA|~JtJjAU04-x(+^Wj
zKU7Rapu-^+fQuMUGrJ>-Mj;@E@n}QGdA*K>Js10`pa$!wnR7h@S*@a=I!NFvhlC{k
z06#R-#|@VYAWKsvpJA8lHa%ejBpe?#6`_B7PimdBa4)86PmF}Umf%IJoBkn4TW}hp
z{TCN)%3z72tp0#8-1LgHJ*o#V{_~(+&V~6~c{C%&XZ+?cQnJ88BkB=+x_Py|hNMQm
zUvYbl<~!EyUb+NdR<yl7rEzIiqG+vRd7hwp1#$WlZ=<?#ZBesv%bySdVKo5~P2y#K
zR#wGanVR*3_vLnWN}#H1yD~g%7vcU@`jO>RLFo<sY7pM<=L+FwLDmq^>%;5TVtT5F
zV|b;v*YSN^S&Z8hjH8|AjqQDJH$f=mG1%wdRkNR!ha0>7(Z&+cmTN=v7%Ml|c?5^>
z_@GVq@qdK0|D~gG{->kiL7l<1;O+5SH~YP-kK^9-I;Ot6>wdi2zAuK={F9GCo=V2{
zTATr9I^+5qPd~3J`gh&b-~LT*oA7!uI_3V<y()wHqbOL}^(2=7Ev@zx`S}AH?5Z~{
zXCp-mUhVNQ@?pbns-#<_BNIzxnI*U+*xX_M<bCt%qUQ1>wl`##d{<@<r%oW{+Q4B*
z_1X5MT8!5q1a=%H#-5MQwR2}I=K)ZA&<Y0}<T+QZmy;fBWXvtvs{%9~o6F25(-qg%
z)-7K7ugvLii7{59)3frBZhSQ}GS2++xSK2=JL_x^9?pEWXEZoAz3Nm*cX$ae!wGa@
zDc-1qTaHk`jhu1c!>92_l&jq3Q9xC!&z6Hk_iR1k_f_dL`|YydG#p#`h&G3(U19*a
z@3`(wsUam@%)Z@TRtFM2FF26{Icf4U2w08sGuUGS#F!ROCwcoGIhgFTcD<%qF$!MX
z-aDvIF~KoC*5knjVcQH<W!Rp(St3sh6;W8jCAI4o>!qnx??xL7-ma`kFvhIh%#E
zb!3dEoae*7@$w?4<HHub!grS40SF-FjsN;1P+YF-;XbUUC<q>(+V)Ov?@`WzJnA1M
zTW-{yu>W&h0QA)!KVr4|8RQ4|&aD=4^`3~YzQaC<{r3fj)<4j=Dgmv0msaC@H<Dzd
zOH?-`>{lzb6JkA^X@9_$<r`jUP=R|}Td4P@(x}Hp1Y3VTVJ_d#(6jRJx^jSpQ|q6N
zf98nrkH=Z(kw{FtkDrfHI}omGdXvkkjMFTpH3rX-K7N|hLEL7GdYFna*lhZqVeP$M
zpC8e4>E8q*f}Zwe!+7j-ljqY*c<JI{H0`_AJ9q*<*-4hko16R6=>7%U#4a5~>+!5^
zvRG!?>(EMKgx#a6P}Y|i>@WaIY7Aqzj?fpSFt3J{?f#I@$<}k~@%>zmPk(WgC&7y9
z>6OUOyLMX~JB>*p!(fzqcl{q&CNg@3ZI$V4m31R~7_7l0G3iEx(mHwtzh4T}q^dyL
zfX4|i6&7oRvb3f%RI}~jG0kQl>znnBhJJvJb?NyQ*2Ro{EF7Fw;mQGnD=0|&Wr>4c
zzi6lJHvbUC^$~i49b&$OPts(0obUW9=8N&cRQF9>R0g1y3*d!=sI=Gmf4I8LeZGF+
z@<bB&W?B&mxJZZ`4veu;4LHq6`nJ_J*npYcHc9GkEI1U^q4^)aBHw@}ac4%e-<;ZM
zuo7lRdG^G`ALooVW~O}gZ*YLEwziU0IT$shV%t}#H#Hi%i<EOTt6J@8GGd8pQ|ngO
zl7yfwhm(H$u^c{&J^2lLPqP-M&CoNlX5*V=hiXSDl?xWEHu4sU6?8IWe+DL-1Tiw1
zb0@XQed^cje2mNeKYq~Ah(Y>@!5gX$wDApUoL?ug11hf_D?sJfB^&;LS3CbZ0H=0N
z3IeTmPHr~X5B7iQd(P5t@Q6~+^C9lE2M|1+=ICym5U?$KCo+ZJtFuc4u9~UT<#VN>
zAKnfV0)ZfW?Phwq+BrJgW*Fzd!H-)gkD{=g>)%x^NAi;Qk>r-7dEi&SX`Q0C1%vFO
z^(LW!P<InF+lBP9ZGp|l$YI{QZW+9u0klzJ1GTWX6?y&z3zfgZUG>Ju+Rd6A|KU<|
zcRs`jUgTQT9)8ZOJYlqWW;7ywy7q~{i@7d%zXSm3jwfKNIm<uK-?0o&7WbV_TB5c^
zl*`(_%b?q3&clAVwpsu$TipuUs~T~z$rzZP0jiyu<a)sC5-uWXIuS<(eIv%i%EpO?
z=iVpf0n|}l#K;J(XaC!klB@0`1e_$~8XkXQS_Y)Dbk_lj!P1^~_9xd>;(Exm(l}~l
zg~vNTB&>&nW~&(3mQnWHf!+%#?af%YAJT9-h~q-}%>8%GuHd3Isp*SCuzD3Dtw4}c
z(r3oVsq}11iroa+GLe*xgSR}ua}qYqd5lkv+wLm^jlN2Xomn6LMu-5=sGf@QN9vOq
zwQyfHXWHQJkx}D1CSm!D++c8Feu#o%mi*-}p^${xz4qm^j_37E$lt74Qkd<s9r7AF
zAo_40>W0zkf<Bh{i}jl(^OqY24Pa;Bw?K3FrSss4?tsLDygJD!?JfFC#QaW%9=+BW
z&lsx0tur&+So723Zt}1k-jJ1cPsMC@vC8xKI*~8~`8tUk*D7_Bf~vSsYZMjX5)Ji4
z<BdluzxSx8QxZwy{q8zPd({gkAY?aanD~xCjkWu|k^oYFyn(X&!72m@6XnSCm`>en
zLi~_}po)3UP5t}RTZkM-$iib9JDkgAS%dDM`)V~YZ3R79sC}1Kb;}J%=6$h_D#aef
z9+NtA7D$9+F!It*H=ii~39p`}GWkoj@>j^?d^sp*Ca`QQm-t9T^pg9fHU6S(Hh3zD
z*1tbyaBPx5U-;ttsy#}9zK4ua70MMmmUoYrcc%p{3%8nY9I=<bLxl_O$S6rOBRmhw
z>vw+(U8yHV{wqGe5&sz<sUHC*`5(|;lQ;N~$HR4=@XOw2?y>b5Ps8=-V88P$1)-m*
z=H41uAa3vZ$Hd+s+nkU3m)Y1Y>LVNlw82D1$uccz;MrKHzq1v1DM!<<jrnip?WIBr
zfh#JjO)oL4CNDUEN)n%W>ZhXPI3TTXsyrD*`|KP8EL9dI6&Fb)o^0_)GR2DmG>`|r
zkLd@AON7j-7m|pQnZ_2Dtj|eNVbR$~`57oeg0{M{NS)9o{%Vcv5G;zn7$N2rmHE)<
zZzum^o#!cQI)OkPj}89BIw{+jj$MjMbNKyBWx`<!I^&T+0o3nCt!GLNHyk*Tf47AE
z8IlubT1T}8c23#AjanyW18Q-G94@B<(Y{WZzG8*uI3F(Y#|aNKu7rnahsJpO;hqgz
zr&QCE=2cx>tLP}vRwkXjm!yye8d3uNJ4F1YbI8Em0k4DunNke7KJ~exr1Sllbzh|h
zJY6vja8InHA#*K0IP_*P`T~f6y<G6<;n|2TWOmcFgF<SUlfiL(C`2jRtSHxkjE90!
zb8r<=&H@+f>3Ycga@N$c)|d@!*GHwC>$AM4Uul{5qs)#{fxPbU_xQfPF&_DbY`7|E
z!W(0hzy<^aB~r%z0S^{k0zPw2f?m~4ZCa`Dnm1K*Za+HKZer4qJH(B>$G&vineVP-
z+Zn-9Y%Ord7XBtZ^j8EdcjQRxKo73w@((MkvmRRftI0&I@-goMPa-rLO`RB3Y&5Et
z*<gjCLgtJ>VqF_m`S`G?BHrQ`^ZsFq&@_7sEHCC-_RJhw)Jn^y%I=OZeo#&w3YVbQ
z(%$~WDxn;3c?vl-c5vk1t6hH@hltdqtM%;Mabr?0tidQZ4^F7!Trv#KsOzamCWEsc
zR_f{xFYoWs91+?m6imd(???00E({=b6?2t+%%iAMe~E*y3Hl3t)1S<q+2kBXhtDwz
zc+EL&!7l77lO)+%Hku2$NaKd~_6C9nB!+aW6~j-^O5uKG=1Ejuf+%Z&C~Ak{25Lf=
zzBgoa=+{D;Sc9vlqJ4D|j2+YKlfFiVB;F1csINpjL<OCzuh0clYpT)K6towUk#&SQ
zUm#!eEVY`JOl^K%fDbOz8y|$VqdijqvYCepH2EX3YiV@(Jyg(Xg1C*L_h=eoKTfHC
z2g03FlkD<W){JocYOSt`#Yk=iD=!vb?YA;NIP2Z`9*_uWv!T{ZIkc-R{<AH8;l4U`
z(Fp$3JJ0oOz~HXriQodZC-uSmoI<o(<0qGBHD;@%D5+?5JOxsucD<4P-yl#OfTJ~;
z)Bys+rl5Bt1`}sOhC$rD<?$X))lMi;`l>o@ZCx#+O4Jl-X#Vm*Hn3zGCjQ(0Gg$N+
z7DHBWTI>J7NtgVyGuS`}#4?>P^)jv=rcQ~-`vTuR*gA=pOP{pPQG6{0*BZ{DU#qJ`
zK_N-Bh;$l9IQ6tX;^FV${VH%D<udyrmdy~8+e{VF%{o26eSIk*dY|b$qce?%E|~T9
z|D#3iil{}7WwW>#G%4YbfY}iwPU|c^%eY=t2LC0+N6R%>YH;?1v~K;wEqSnxYV@aX
z<fK2*k?OSq;Tm*A&6WeHv;{qR_v#=I<r832Pv^J~qfdQeq5)mnz~dVe)vdiyGq2ub
zv&F++K@5)h)NY-FCGE}X#qu!;*7|Ux$+O@IS|m(U#w-CF2(Irv(O_Ti;1uldU(K{P
z-#1wtC{`#|+>xwG2MAwy`-Nt1&(ksA`+ELB`{1g~M45F|*_zyLy}ZCp(`#05+lgre
z3;#lB`CPAlDFB2a?#yZ6#-ndihZm0AIXN6PP^qhCHKpVFAs)a`x=zlLH@rYRU^#>J
zZLhekKSge{>MpQcF*4`0^7B62@IN3_ahAtt^-Wjx<=QIihpz&MHhwQ<x#IIZ5O3xq
zDF>QU3kI@JU(vxSjedF--1jS@XmEf@$^2RPT_Gi7Q3teGt6LyH;z*Wd*DjF!p!F>x
zqkPIj9>E?GSjT`=IBnH$*?5Hk%O(oucy7wW+|v)O7a^3*LZ#C@NN~y1X|$Q5qc2>Y
z3?-D?aHJ?|8;XS#!ZSax&Zs!$j(YfkzALNG$IGuC_&W7H?BB~H)qqMxqjF;^%NjLZ
zRniBWTPr|=0&(Gh`24y0ie5Yv_j|KjQD}PgkI{yx513;Hicct>q1n8p{-Ln+G_oLF
zK$Gz(@hH#3^|r0CL>;MCyz!qXsj+As2oi(DBt1TAY7{V6*bN<zjP}}w^>5b|VD}~!
zi=_+lyhgdver;xcTx~7up0mr~hk~^zk3?z>t%(44lIppe6tp!qDqN!ij|`6Aq(@b4
zck~*QNE;iSR(S0Ki>95Zadj^#y;-u0mm(zdB6qO$@mrJ#{yQ{$LxRR3SDQeZNsUNW
z87p^sosuq?W!$1h(gCa;M#{0X)Pe<z2c0<xeJ7FH>9BS)eYMt;eV5HvQBRvpbtXpE
z%O&7g#fe@exrOZ>vypcASZnCr?SMLNAZO7p2g&+_598)J2i~Sercr^Hi!b?~jBQq{
z{=iv>$7ej7zFvwEmQ0kE&+#LAI_^pkQP?{-O=$t~WUer)jAiN1a3O2M=ujDzc}%J$
zn9CG>`nU3KM5f{NX4oyoA|urrx<$=RHD-XinPAvitJP*xw|*_G5^mxF*;s!V1a&z-
z89Ge3U6D7vV7mg`nLb>|_!sLi(&_oFB&*5RU>*&j`>lP`Bd#n#myUmD$a4sd)F1WT
z>?a}|I1n9d`iWYhE>mge(?i~8VPq*+t_hLL`|pYik;UN+;&BW$__X1WwkQK0Sw8?H
zG;jrm>h0cKvy2~RpND&UQBe8@F87#71el<alMLOoD%kaXnG=lsV4N%DgG^&65rdc{
z+H`lQY(rw5s<&2TL%Kb?I;g>GslNLjb|`e^QNquHKH#_C4QN8qy|RnT;8zHyUCUNG
zg4Tx8$QX4opI-bLrn@UA4KXI7JcNLY$;Qo$r|iLctF0F(u;2#MRsWu=zG;SCttbgs
zQMUm+**Qr!fy^CYp%8}$>Up#Q6aVuP>ZXG&F%*-2P|@lbjuCwm`>W@}h>h+p4^ooC
z?1gGkygcE3Om+M?f;`K8lpArC;Rb#7Zpp}#r)W(!gNZriL4yu^ay~PoNkRbq^%wMX
zFX>;6`tgG~EX9usyRv_$tF?O^SCm0WuiBGtg~>u8Z|pA0_8rXoic^UoyuA2G;xp7s
z?@rQiA!clx`WWSqfsDvq72=SjIpgUdEd=|8XlBb==fUFC42@Zl0;p3_#$h6iy4f}g
z#!vCI_TM#}QkNRj@Nnl19$JCaN?Bg4cPU+s;f-0u)T`5wG94m_PE)qWIRB=1jK&ET
zdnwU@Wyzw#0TW&&MFuWvn&FVI;;)6zDcu56lbXFi4k;Qgq*e@7rhqyF8CLYDwrgnS
zpWxG>b#~zX%6>%=;9B(~kv}JR?2!9rEoLC;Ac$tq>d;4Y=_toZ0s?@JC}Q=IovML2
zi_~3Bu`AsjpOH-%wec#aN22gk(29VIEnQGhEP41+#zJ&lAC(JpNXhdjdXpI>r&Fke
zdo@+>-duflkEr_aE0$<9qf(CI#!MX)$hdl$Xdd6W?IEOO9me0ZXOdB9c5nI#>r#oh
zveH@s{X%C5eDX%_^Z|gzJ`-m*yL54&5uV1zg}6bdP|{RI)uV$!gKHSTq-iU?bS6xU
z^~^Huk8)t)vSKf1l4F-7HJbH3kX^^eQMSAk!Lh$liaAMn^imJ@VH0k1*Mv&`AqRuG
zaWMg&n?>n%ACKIZ^ibr)_$CSe6LHyJMmi)cp!hhG=L-5=w*?q&W5kvw4e#&NPqo=a
z)1Nghm+k$0->xj1zKvUlZ)+H^$fje>#bDZflBG+yf-epWxJGWx;1X8JMQSKMm7g$T
zlzN6TC*i+5Z<00QKV1{NS1cq9-Ox^cUK9QaTKNc98N<=@{0rjdeloWk3n4D)GGjmp
z9T#l&>Q8L~Lm%+Nzi*MqU^b|K!v>;f5qY+Xy$C($Ayty7FQ&aDYg395xprLRYht*@
zx)C$jEL<#NTa}k5N?GF(lwEV36asl$kX8ELQmg~dKh-Et@FP&u3}}?!;W_*PIWCwW
z)6x)&M_hdR(x3L_eQrI#UA%S?S`x+drwkyY9k|^0C<4fld69mh=SXx#YE6vEq{<7i
z^^+!2G+CFYgO>conV&?Pb{0tHz)m=n$izyB+CP>{zwPM+(Nb}0n=z4aa<BRz2IusT
zutjI3u_93+`K^ndSUNmQV?L)eNp@FLHiYMUHOlCy58X;%v!URP4oZ`g!yic_27SK;
z15*w#wgM`8@Vwr$4?%l#t^9otsi*pMW1%|@lkA@RfqK%iH?onvaFa}}wHj`@96>~W
zA3ssXGykMuT097=>5g@9lh29{FuZrpEOO;(SSzQ;?I-oSwmf;JJcZCi@ITbAGxF&Z
zUtY2wfwhWtHMMs)@sCqtJ<%|1pheZEFIYFT2LS$7XD4BSp#3a;l)iV7G0h+DXrCkZ
zNI&41v+P{vFLiVD9%@EbcWz)#wzS8aF^-m(F>cOxmbzOZTY8CnJA2?Frxfr@Upixn
z5<^d6S5Bu8s11u|Kq<Z&yMLhGd^UD_b&Bt+-ThN(U6PIpz1qh?+A87@K5I@-eZjDL
zZ~(X_k)@Z&X$fKBKt?}_Xp%~37SfH+6h!4Mr)13SyvWWv03wTBqfTY|WDY|jGTbns
zJfsX0v;^bXF$?>dIC+}%$y0;UisVpbqQwC+EHTB#O&b^zV^UBsbBk5bJ#*;gb_eAX
z@houSIy**v;Xc<J&BXI6e~w?F>_li}<G|cTI40S=5{P-ox)|vkO&nK(`=df^;tQX!
z^KrlFYzTtjaivm?GCIrwb1jy<x{eO5kOT}Oq0sJdE3A9UKqBGYwGbcQm(y49_7I1)
z)IH^@apMTVFW{ev#-_4VV{BWQxcMF2;pHtzM)pSQ^V@UBr@OWIPU{x<g;b-c9I&;-
zF`h-*`1BjUnp53Mom#C>GVU$&c-EC#ruF<{E8Op8v2%#IceE6>ZpKS1Nd-O1x@6Df
z$p|r}lHy-hurf+l2N%n6w!&EqubxOp=wOlgke3*GlxYElhx<hH6&HE%3_^!)>L4Q3
zoiCA)2YV;vG>tHi8{ZWnVa-*G25^b%sOICY@)iN2#`xyuqb<@4f}At&{PUt##g0m6
zT1~ycs~lY)yII4AA%z#=?WPWEj|rCT$f2)1ZJngWt@puPWh}zi&ykw0UlkQN*m8}i
zL=Ah~B%%@lHAXpZ|M9Q=huRA=ofDE8ZxsV!{x#{Wm{z(;Q05x9EGc@}4`70)xdYDI
zPgpNk#awk7T?=C96?AualPB8q>+$yjdJJ>6fPtvS>pc-*4dO|}5YcCxwI>DVG5KmU
zGHZ!GV~MWO&j|irJLgMJ9V`Y@a>n3<BpAJg*e1MK-c<Be?8??2+E@-#dc7ouqp?Fe
zMaKqfoVmx~ove?fscK+Q3z$=V1L30>`v!f(zWMx2oJP?kROD}%V&EfXP|5L6NJGV`
z^H&aeI3o?h<kg>w9;Pdi(j_WWI6ridjhOo1)Py)j1}0jyPs8|0{GT*&0fN_yCFnEA
zzo(FUcKTSW1Zrl@MsDGQ=}s0!mH0CDNmn#U?^@P12RjqV&D>6jkpUK(;|Ew2c+;mH
z-f>0Fc#S@Bv#6lKk>av%-x&fEO?T{YgM(&x;W*-_rK@{-;U5LJy8Lsv895O5gZyeB
z>$id^i**XnM*`5rcV&nWZ@vY_nWHs8!wSP=1@Cn~e!~|`3#;4{!XAuHcA)&1xfpX`
zo#pJoEi%3&UozBrD-U2YR-T)U8)7S)B}LX22@f5{_EqQ91j2`l+?=3j)DjYBpUC8C
z#ZXnbFgtYpev*Y|Z5uCR&d@COec8Cg0O#X!;$a*w-wT|t2@9D{usYgO!s!ekMSw^F
zbJ&vOfxwYywk!Xa%OX{C)!|9mI^Uxp^>4As8Pg;|{KMn`wjQY0K%_iIhtAdvvc)<h
z<Yu>&He#E9KfyhC3-{veZsWQHASRvOBcvgOFFsxPTF!vD*X4-Q`bP#nnR>Ve6aA)7
zyg?Vx`ghZ&^V9_YOIa`GuAe#RAHA$(=gbI!KfKS329f=EcbLwSZYBpw{t`4(i`^F~
zgkB6*Rj$iYSpn({c|QYdERcqTw-xd?PEfm4f`ts{#Vjaatg<((xZX<aAY)~qa7}~)
zBR4<VA%FEWl=b{7#sAt?IyNWc+;qVB`W;qnb_a)CSE!RzAFZ|op$G|4Nl;IHxPn*L
zP^7S_m&Q+LeIqfVt`8&}@$@sVW{h#&4^R&D${=%1B>^QKUPMhJ1EL0<%lNODS50TF
z&bVfQIKvL8dhVvT2cyYgpPLzo4)u-e;PEM9)(ty}&fU7tf#-D!%Ft_iV?vaecF_o7
zQ_E#Dz10S3S<kCyhzAHa;wQZ|wL(C`<3{84a4-FBsjDSUqJ8_eB$S-VE`u1XS-{;b
zymrSc5>Q}Hc_+=bjw^gzze5`H>wU-F*miF>yL18|3vR8)mfOis88MnkAHI70y05)9
zH0rl5Cd7sbNleg58u<ElNUWte?0xxza4_(_huLw_|8)|@;kgTU=vJ|`Z&DOS%%gpk
z+M|3MQ<IA4QWUyTdqA_q^*p9`_-b22uvwq-1t=9PW_Q&Nv7B?uP(BAS%M8YvCK!*3
zcAmcHa)dZSTB5~QPgbi0b_y^GF?vR&4nide<=`fgeSL~$#8EIvxM%{x{8`$wr1CWX
zk-4&io8`oo$%GtS1gSUIqdPW7oByNCCqWfL?Jtd@{l|`1FijB5CZugJE1-O&*5P#s
z#Gs6K%%8M~@AUbDE5qDMecalC<3*#`HU4oL{!}4@dK0vNhx(y)(5m-v6_gS;(t6%&
zfQL>`<;{qRXt@&{UT|hbx5(B~XlZ1i;V^cG5Rsg>ztc;sf;-QZim%OHY-C|h)xP(D
zslQzAHNSm6^+eh6i4ZVI`6>i1sZ?47ywomK;?rYoUk&<J$3G*k#a7?o9p-dEwo#d8
zc-!acCH95KT|sf2Rcr`DdA}_@t6Xh)ZZQ?<%l@h^c+G`8K`kmL$Eu!*1>ECBT%O!|
z<TZ^^`v7WWtFmcScVngS4WX}ltX_)L*6-}{jjr!Cg>pZ?KAmLzb={jIA7vUS{pCZD
zuESQ`MrS8R=o&ASX$Lcw=`^^FjVe)@wMHAjcz~kT4Aq~MsIbk=0i_}N%Q$!5CS9`L
zL#0{-J_&Z)ems0bYSpBs8}<0JlNxtuuL5Q&MY(#LtO!Y`dy@%Mho5jUoB=lM?j<wj
z2dkK!1T&6@+^!~r1Pe=vD;F(5?Xudgr|Punl`rJ->S0H0J0>N}PBgEZ@9M_j-?__e
zBl)%8fasbxOJ}VK=RtNXQ5$(PU}!;0iM2H8DWh@+Ue)q53q!+qp_JWYb0k{Ro%WZc
zyHCFsNAKt@(I1hMF^ID9Ht@B&392$TPcJ*FI*E6tf_Y&lEj46r*xL@kc++Rn3_el#
z>d(FB_;|8;)M|$uhe0+9IC3Wg4yvF3JiiI_gWTasYFo9m)x~;C$r&?z_V>(#p@~B4
zqCa4Lt|p*{g8>~6f$$!Ot?X?z)(MfOmkCJzrkC4kxk_f7*WtI1ab~tIRAR$%rlkV{
zp5h4&h3N}hV?U{SDM8!-F9J4CkR){YHu4a5Y1Fg+Y3M}+&*Cv|iF$n2qlY-1knlaN
z;f}1eJG<S6t@Duvn_bxL{o!HBR*x2^%osg7gm_*I1h+kU3+Vx7DSBDdaTlGpz%~z8
znKs%75AL5YUDmm)(axSFE9oBNpjc`mAo>n)VTRYdR#WY<FSpP@u4(4{bMkBJfRJ3}
z`kBFp)1k}XpeoL}%@P>MMrOYiR&2c^L5Xgu?=99>ROJP{RPdxz0VXIq%oGZ44Nz*w
z9n@mIwra=Iu?7hB1kn_1wyJ$vsckYyhX^atd(ze)?rTtd7ePD}F^L-j_)0aOy+eD7
znbHG#@aR^qyGOym4N84+Dt5Z;0a3n+b<@-oB9gs3r<|d%f0&F)LhT?^$yvoS{69CL
z#7L8s!?idJT1KIMqHo|0=hPrEVaLCFB9cQ*eDM`UAQSCQ*KNCzF*&L%*_-U=>0u|P
z4jbq={0gD8O|WLs#qFw5BM$gUz$(L{H=S<zh%)}w!(yreH37pEi2WZ4{#dn)KNA8&
zA<Jgoq;1UOeimnwO?e<iSu$d?kD}}yLS{1Mnxjv-{FR@F8kQrzeUnyY%4ejhibo3J
zt;He@`L|!Gc_~3u7W2H?W@T@0itVoD?wX0+X_2Nc(~+|N->Y(O<G56qY$|Gtnzf-L
zE7Gwb=np#ZF2#h6yJ=k#*WYQX5IOQb_aQ^}-oy>(X>NA+0fsP4X*%Dv&+%7Ak~3Oq
zD$>h(Wo|dGOXix#;VV*$NYx(7tCaqb0Hj|h^9G{<%7mGefnxeO_6Sik5bZ*#v`loE
zWIQ^IZO;>ICRk&6NZyP270E({A*&r<d)SGt*pU&)ia6QmY$9Fm{z?O+DQHjp3p+=K
z%c?oS5-OxMq@E<t5qY)>_CO-%BmB@o&$;Q+4Y<p2G4VUKiLarb^lWD8)MI}Esz*Fm
zN468SeH&Sm0~lAwt-8R}cO5DF9h>M}r$Z|9ZU(0LzxxX9s%1_2s0+FXYg_sLYf~^G
zx}Jbz+S_l!Dpo27rM382n*gFiZ}i|bvx5t122<%VS>rckSDx8*cmZyuyb1FqE6f^2
z-x<FTRuxbh8l(tW=y?7^Gas-I;|UYwey3dT#aM9ga!~KeY>EgH8X4z13`2RBeZ2*y
zJo}`+<Utb)wxB6p(+cb9;OZk4+HR9^>sJ696PqmFwzDXAr2C+}<H5dQju$0l0#kzj
zGQrD6J`7?$H5+G*!ybmUe<lmJYb25BrN~=kjfYtd;{3xj4YmBBV|aw;vWdO}1v6|^
zx@<GP0+jLb?mKEMcYIhLCy#F!4aqBp*``Uf3w1;BN}GDDdfo=fMq5u+L~}dI`X<mO
z9$Vu*g&anym$B8B%T*7vR+@`RZ?Cx9%F(G>L>Rw`(H?{Uj6eB=pN~Ej^%dFjTQ4G`
zrfV><N7oKp2{aXXg%@BX^`-P%EHq^%-9bBZrhi?wCd<&*{gH{L|2gU)1_2xBdQ75E
zkoov)MT=3Qytzm`Z~gU{!?rIuKNg^G9*xGGPT0JI>9_vc1UC|=<$4?Y-42#j^-O<b
z_;a%Z@~@qKD&&FfnA?qGrp=w?hDAILds6{a^DnPV!8^Ifi<#Y38Xan^#E8*KrHOmt
zs)U|oV}3`TjEVC~4s+ZSY~!awXOd30IS)C+jH=2cQ!t@PX2R}WTyc^k)9*k>#_IVk
z^-a_>=XSDN>t|#?ix}HEaWN;Gd8%DE0Qq3ge$!|WkWJK2s8-m<lXHZsZ~FvqTH{K&
zes-TCIHRwKjW&XSc>;rT@`$BagPbM=zkLl4eJk}QB>ZT`lxt7ec${tPaIJOl^PA$P
zAblrEms*94&m0ajkCxl)vI*b@O0#aeH+Me=Ls^D%Hy4r1{^cn$Kx{7`Txc3yaR@#d
zpImI@7AgFfR2log+8Vol=CFJ7g5iGk`kbU~NaRV2FmmtysCILI^Y88DtqwpFM{uV9
z0==!to`XGcNRM)VKNiFb-G+xnUu|O<8w1OVnzaV3U3)c&XEfqGdw^f2&Dp}?3-M0M
zNuck)wf)pvFn8Rarn0X$v^v;Bwb}gFNx)Yu7BeQ-T%0!9j>Bf%`76y_|L!UiG|@i%
z)+m(2QFNQfTyMbP??kqY8$3{9A(85nUe!IsWO>&(T>FA0{ZYPxuD87^9D8EP>);o$
z3o_V^;>Jdap~^JK4?w_>PXrcxOY{yta@d9pE+5UXOZ!EXlafIpTH@#cQ%=}3@8rWL
zm&AOJh2fl2Gd;AIm&}~gYh71=p?RTL=5C=mNvDQfI&{FwLE%X@_NsGy@y)jFl;@G!
zvbD>iR1QRa@tMo3|DA1@(*Z~;4~&Td_s#uf-3Gbd&JKZgFW_WR?({D`@AD(WLb}9|
z>;AmSq5Wh*Jnyuz(4*idsgq}xt!Mj;!Lu&k_#@NcmPg5=xkH}T@Bj5D0Igu&PKsWc
zAsi1%%SaS$DB{N82`I_ZkL@XrU`%k2av!AaI<~wG2T49>w>Q5DhY6>=bsrS>sLxhI
zg#c<t-3QMTYImJdYGW18Mhef9)INr~&L71;>n?FS&-KQ0{tgU<<Tql(Bu^!<>`}r;
zn4V}6Kc=mVFSe-?o@H|irM3=EoUeDrotxM=VykzBcp0EyW~owFoJD0mtp7Kah$n>2
z`+%JiL!kNkB77&?T~lUV1!x?{y3&dB0Y3SA?#L*Wm1&_t1t?}LltD#;axJ(l)q+{0
z32B(DFT#Xti!^XbSID%IVOxMTgaw==+VWo08Uz^zBOn%P)1GiwcMVKDFxe?17Am&m
zQ~B!4hM1ItI<AjH(e6B>doh>kd-w-t5Wl}!4NT02?KDdkRrHGm2|%*n`PT6Oe2ca5
zb#2tBXuxP3=o|b7+UNQ(akP?_e5Gm-p=^Z+Z=Cdu`4Ju5U{8NnsZqn)S1nA3NH_!o
zR_|=B-GsXD8F~*QU1QkBkT_JRt$jeRNDu$%(A)mE)lvXCW=*lADfI_xv@lsHXHxH8
zcrr&jnHP6&n$eAr*~L$m|F~s9ChVp%C}Ua#Yg4j$?QEj5wD~&>2AC+j7kQ0MYBO}-
zNH9PB9;B=8z}!{{M_{1^A12|#b!fD>iAT0EPJVqYKOt8>T-AiGmEe1+W44POeK#`2
zeH%&RoJr>+Bn7vWj>jo6yu6pToBp#fpDKxjrad1}p={F&8M=9ko0ben{_r%m;8+t^
z+U--f*QLa<wEpo|o+A(jmh2FvkxDU*PmYJRe8qUu?wGd``r&TMg=Q$z{|kbUaiX2z
z+Z!fCs7uI)Oa3m@U}N~B&p5;<SJ@|Qk7MD-#`E-WUh?Vn`VhA~OOP^HWad2oa~5=6
zy}(3a@i!x#Js!m_7VHC1ORX4YW~^zVr}HjC7XG&<>s|1cjp^HE#G$pkGTBW%Q}(gb
zkkv_OB#omq)gAdssk9G!)g_x7GXqs=_(!DM$3)XV)YJmYb*GCd%z4H}mPJQX=;>Fp
zJ$cH)yi#E+NGC2M0xwJ1KxgyVZF35$_?W`6jdbvY%ZR7CsM18BLS<U=&t;E%t^wjS
zL!jBJD!N%V|3u94gmSS+D-`2DT$$N`Fv2F}@QV&BZT%^(d&waDQ~^IEaeW+_cirsN
z+WiVuLIF%k6ifEv;D_Zz^i+&zaTPLNaL2LCNWuF+u?*t!btkq^+&#oY2W6>=O^S~K
zd8mV>kJ7hM>9j|H=Mp`mp%Ua85sbRE?A8Epbo(MJ=Z6vB?Js`S@WBMcj0zw=Cw4N8
zmQ-Vh_ibFneKav>Ub<a{r1~j5kdFyN68m~${KqU#{;IscgB&hjbDH-8i=CG|kzHM3
zJpHiYe5vp-kybSLB|3id_mS1kZ3S$_#r$x%qeH&D!RdIwXt*6L5o+R_`ilH|;B21-
zGvV&;qTEzphm%lj^Xb&@LWHoLksZ;?IM0sf2eX+pq{KP?SjSc5VNWw8uBESUh%aI3
zE0!@}!s4irbbVl|+_73;EbWzKin=}VJp~9h;~6j`T%HxYNXS7HG-Y~zZefB)r2`M8
zWKh<@f%*i3aM*v4_fa96^dS<Qn*W*z%$xR4V>sB5Wj1i<pHH=yHKL-)xZ#!x%6l`U
zgWV6J8H8yw<OZ{VEfWdROY&UvtKEOYP`I3|`;A6fVqO<{+!j^AHIen>M<6OON8e_o
zT7!1w_@c)>NAl6ik=yWD(1KYw&#@rji7?SQq(KTeNLxNdbAN8U2XYLV-W(ME7iWoN
zInPgmhbjpg%n*1>q``NRX9qczdA_KuhTa6lG-hz)hgmE{^RldL#7Em&P!0-Nf|^A;
zbTyb&-(OAbdT10*NM$+MC$0*^@yBlXp(lu+pL6&5=E_>&{$MX=EQfT@8yh{9e?F7r
z*KZ&+0+#mKcSfORk;n0!i}83@EF090UNFMkXO^Ff#IELqMQa^zm?)6Zi#&<}ZxM?d
zP1GRJuE2uR?c01OvUj?i_D1mEn0Xk;^&R01*ADBD25U)lCCs6bFvV-fBO)?M@F8kr
zxnIP~#-E(sED}_9yCp2v;6woxZsb(ke-HuV9(5<!Ls?LS$o$)q5Da)pkc~w6_*WOJ
zZMgMAqW>XIYDV0*BN)ZMsFnXjHnR6|%NDOMMyh&E@3#Zz@hI5b;WU#cBXC8Gu#u2#
zR>AmMolF6P$$>kP8RcL*HA2z`%2(<e$*sX3MF!pbwS20-Za%u$dWmyr!(s@GK<Ozf
zEhmM^<O$JfWP>E(e(jTlpx=1EMU3(p6dOH{IIjXj^xQtWG&KSkb~^MuU?RuR=j~zH
zP4QnkIp>p!7qi>C*{9!N9(0_UMU$rD5sjuc(G02&4zrN0n?fAQV}%7TLq_+dorJ>D
z^#kqad)Z|#?I$L|f$KPb(070aR@v!CR8Adcb-<WhPzL4~Zo;EZSCI90K^`6?Oi(L_
zXI+=$jwv&&w7I6(z*$@*v|(Er-S82j)G|VUzgdd5q^|ajhTa~vM5{JK3|(bXN4zDG
zfMVbzF6Xkc4pLjRFEsh_=OFfX#ps_nnCf?7s+viXLG1icX=JW(&=mkDTeomJx+rlt
ze2^oq+)=nl$r{f&h>3q1{O^=YrwP4MQx|>{0m^a6k7@$bVeWTHo)@*1+^R5tDAHq{
zZ~>4*3D=3Mz;aWJJ`kTUfmy}PTO#ki*!UT)An|K**T6d#$N1Aw6UvEP6!fAJ{7&B0
z-1!5%c4-qT2Rl8ECnW$C(vd2nhyj0gS0Dru7thIUX4joG%ZC-7iJlt04l2QE(AnkY
zpmInuhXeI@JS|MtgRr4^WK=XPsAD<7HoBnj?<q$lK`6R8SNa2#u&wjx?>*^_t}Qu4
z#rb;JP$Q1B?p=REhJ2*!0{RB|GMo4hg{QDuO((?WnFm>|&)5M&gx-1U-7MBAMnk2B
zx@ffycUmWeuZjY=$RI+Z$9c;wx^i7&lMh0q{BT@1*YUroN3mO5=3gUDj5fZ#Z*CNH
z_;nwG7xm~mNM0UN+MLavW3BjykPks*81dwcDhrB$D0FkP63qv<$iq$PoCQ0voJ7c_
z-R7Chw>|uLjb%WCfa8{@v;3)~{Eg2_!as92jPB^&(tXBOFZRM%j_gL?j?-6F2a=R2
z9jm+`4T4$CNRtS16t23C@u+;8tXO0!3h9~V@vp$E=0zNH6VXgcbPej48Tv!AOIJ5b
z??3fORzFYGHEAq_@LPU5T-la2lyKs@7uD;CD$`)21n}@Pxf)yZOQUi1h7s5}TAW%;
zpS24dUbIjv-!f;pYk63>{DbSLc;!uhZ2e%qo3ngL^W%N}@}r4%d;atYUC7bNvIT5=
zQ6FZAr@^IW*gMvC=<;0P^Gjv<i0JKF@S1>o{aXR{^!+In<+8?&_aGHS*{O3P5$>`(
zTXZu}1aPe28qxa9nizS$3p0b*?V&*)CUH%qaODcJP1=oJxr@^bmmOMYO7gU*f1N@p
zf_Uw0Ou{m`HEBU!;HlZxr)>rBOwJ#}J8%6A>|Yp+1Tr~LQ3Vv}@j^mTd#vR)S6azi
zW#A%}*x6W}iIx8JlwmvCLA61>kN>*333sCJ2E01`228}dK6H|wXQKF8IX)eH;FPdD
zj&#I5_c#K)OZwz^PS-HGk84Qm0&wpXy1P?Wff)D~TidAUyv?l+yRnh?%2|$6yK0)c
zWhq+7)%7pO$K=zT`ovNP|8&7Sk>fw<Vjsjj%O<56nbf+`E`I^msyq`4_{P;D63=iI
zfG?ks&s{Cgd&vsddJ%1n;S?|LfE&SUJdVU6n@m-+A*OIjLxiAh!#ATA@gsq@r8rCy
zeaOB&u$r}in<0aaK04S=>8LTT{gtbhrQc7O)hYg7Hv776*S$2JEWmriV&OERq!5D+
z;ulI+zremuI9QZ?BVrDB*)J3V_s_C#!1q*?()tEY8cd9_iz$@(M6pQz(~9J81Z|sR
z_9czc=%Yh3WPWb8i6`}kB*`&NZH>CFx;7W*%Pkm>Npr{F;R6;bk4+IMw~B8)*jnC@
zHB__mXDGS9iGK~xlg_0ivNVNj05ozM%1L(<cT#c6>l*NiNS%HFw&13r%!<DXz;GZW
zhEPbWO<HuS{)XL<mp&`vY@{URryDehgBNR7h>Jl#{q}YJov|}nytjf|)hA4+$sW}|
zyt5gK&`%33f`?p{4l#mGdfKEN!%{rj474&wV8*7F4=H#=&<U53IKy#lf~{0$ZPTHD
z{R)R;h>1etLD(Kem>x9v&$I&=qi$?gp$dNOrv1Op9VZuMbG*rv5mqm>R!#os%6KUL
z3pPYYdhjJ{hrb=i3qX$LWUZOX&DMzL!4yk(4BF{yQh_?Wry{lwSLyZ0w|MgZZQS|b
zYWKhHeT7_2UJo<pZbsSEPVfDxKS6Z~P;%MPp#J!zR=|e^NfNrCHNN>p7G4w?#^itr
zxhsMbN((c_+^Dg@S=i)a8yjcNsJ7;UDpV8!-_Q-u6q4Qyy(L+-x7cPiN|<RMwq+Xj
zQTB3KtQasFYzBj&O1SP}(EnzUc6cm9Kkbtre%@d$6Xrzf<l^`%iqQDqafO2%z&+!f
z4J^$lW@9d|G@}eDe-Vp*5Yx+%=a6shQ)lfFtFFET5A5VgFF%w_5sIDRsZP%-=P$AZ
zu5mVUg(i1Xu$@~~%y0cOMt^v_V4{Y;z%GbiUHX>NpM@;h!}$2R8p=h9Dm(vv)X4f|
z^gRRtX?OZ{u~YGSFr(bf)H(J6OwEGWM-4rvB8hJlrDvAD4^QV7u--?Wm_d){W!hDp
z>$EZ~Ub2+}{d@)yk2@R*3H?G^-nmADbQvfXe>|n#W6(n0%{K8psM(efD&iM-0tJ@O
z<T7-6_^sDb9u6b%T;CHuKeRqAb_zVg3lJgd)w~vjl{9|!R&IV8ZD&aXCN+ayp0_1G
zWAs?j%!2V(6x}-Fy>z{ubI6?4={|oxvZK0zq;QT$aCiUKIGhJl32y2>+?Ctq58m37
zT_!V4ukP3Ly$o{}9Cl6YL=&<HN9Kdhz3X%7xZE&)cfBvAl^9-}8nzMo=Tj*~`)u^u
zvh}hOf1s*Uj<3>|bOG!F%|9PX*m@*qJy9l)?2Eb@QjKI9rPpydjTdq7(+lE?6#>pw
zLsb$4E$u0t>9FqDPrZ-u5IVH0>ZT(2l^2|^<niXer3hE&p-87X{IDB4)XaLYPopN9
zF&n(zj3JS%V+)MWhCDHqjBhEO1n!3&p4Ym=I@u%|e3_pw19e%z^a2jbQ!p)?C-KMi
z7;!<r_X6;p4x6k4<CC}zsm<<KclQ#(_RM?yWBL;OJJC};byRime!6|}6S6<`9r4WK
znyTHR4J%jNkxHNj`{IIz^geAEiF0AFWfFVb419}z+$kA{J5Mm_CS<t}c}h3gKvh1n
ztDtfg75Z8GRz@Mf%jmob8~2lv@;?Kbc`g$YgGyRz=#L1y3L;iZP?=omGmVRs*TPid
zZeN=k&?a4TH515a)n8k$ftVwv^@1xp`1ap-TPJk!rcNa775_lP_``^%U+YsqKO=%~
zcF?cWAVRqBs5pL=r=W1A{7|v~|Hyg=;9i>OTQoVbZQHhO`^2{G{BmO3PEKswc5-6d
zw(;`af8AI2-B-J6rmAb#^h{4p_g;Ig=}n0%a=ERelgD`=fvX6c0vO7}r2y(6wiY?)
z?X2rcwn#Ezwdshn&}|ER5)Fp{%$DX;9tff)nNP(u*$S%2l(%q4O5_KTEG&IB6)=gh
zi)NNJNw@v+vSkT>$1P|4r@U3oR^o+?M-H!4v8X2AOg>jm(|0O5l_WC-iWX*?y`=&o
ze<u~@BI+d`Z?}=m2B6HM7nxz4J^QZ-JpN0$dHy<`q!(#t5BP(PXZNSI-9u8}8Gs|y
zy-WyLTq~+PL`T~=<otadQ3=Ng2Kn21+3R{th$Kkdg1p4~LJUR({TD7UKZwT~o|&UX
z|0b5IY-!q{c&D|e<2EK6iWPZ4hGh>c$`uaD-2ESy4Gw^K;W6+jysz8%1yOf}TU6%A
ztGFk-8(zx%gO9;ZC0@^pAB@>0(y=3FOW8PH3gMD<nCpWM^GVyrX6h!o&Too8lryyB
z*L1eOmvRX=11Keu%&`t-Ge!FP<^zuuE4dNzPmFkuaGXZf=9vk@A$4=Md(zARzBf09
zUl#+Klm~#6L^u1?<ah&S?4J0@{)lFZQTINZ8G({o%?WF%jqq%52Mcha=sZ7^Qi=#Q
z@R&fJU=ML#7y?2@GTWF5*d-;b9LXFCPwi9xS`OduE=F%-k+p3k1f1zG)bGzP!IdKM
z6=LK;4?}||h&lf6diL$@!M&8h(BOjF9+K_9Q{MmxW*nb!h4Q$2q;v&w%6QL_P-I8^
z6PqY-;GSH@*llv&OVne+HtWT6cJ1tl6YC<*K_-e6x9=T9=fTtgJemr|%8o1IhT0$f
zkP)OjhwLDm|JYU-JCE@OKiHQX?R`AaCizvTTPCWFG!%^WR0(Rsx{_vu4Fca5idc&r
zGZVlTjx}qFLde_n<btR8{KZwwoeh8f4-1?>kY@#3vXyJuBKmfQye)P0n|o-WW`<=^
z8hi>|43o95#1FkiKAFMiZoffwBfFE{Y8}!~&rkgkV*WC8$kggNh*NHLt?p=8*kV1g
zgzwY+O8i_?vt-stb1)_S*{Zo&y1MsuUjhh>{ZJVS>Cd+GBcNG&rje+_ed3LptafbQ
zC|gbaks1woerxdGcuD>7vZjH}+LO&`EuV#RNu1JQZ*?JV?+aMTOyA_0cbsIRQjOS2
zXt2mawG!JV*M7*!w7c}P`p5jbIfpAvYBvs-Sej?uKPg`W5nEoBVcn^SAf7fjIs`z5
zFeH#P__YVs#&{jR{tGz|WYp^mlJG;H*!qGZQU}4#q<8yniBM~)iV#}v@I6y{O9{dD
z(61pJiz^bW;^LZDdC`!1ydaBlz_8uFf>}+&CUdPYZ(gE&J=h4ztY;mi-0CmJ-Gks}
zV@ky><omY1ikP`$?LhywgjOw{OeUaL-%)xJQ$Qrc;PZjfN^rryz^eCd5Y}8y5#|c7
z;sdpt0P}?#p}?v5HR?1?E90y}@(D5WS9*vEJ;7~EBe$$4V(h+HHx+b5Na}PiKcz~f
zLPm8Wk2E#KRn6ohRk?_2sHb9RyHjX{Lr?nPdA%>)NxMrvbgY6Xaqa8nbP(VZcpCne
z8DZAfA^5Z{91Jfa<VUYub|_TVW#6%7JKNY!Wv2evIBFAF{Lj^ffa+Ff<*wu3#ay{(
zQcuLuA@k!s>OxSyw|eTqbV>wci;xkkHr!8oMUJS5w{Yv(pianT9~!<M;_8@{HDq=B
zTy0vbdW4X1mv=seDsy0A9uLrDbWbWwUBdh-_pgL@he8C!395MCmI`@_N~onJ{bKEh
z$>*hfD-|sly${ONuZ%dyPL{$cCsBJ!NGS_lm{)9|XulH<4LX*QePTbyKsW3}Y<a8o
z(hS3J8FB+(z*LTe@1cHf95~qH`5vk)me^|O;SWK!l<ZghB?1LY#teY4`#BmrcY<&_
z9FVI(Tix#t5~;Hkq-M+MbIac~4CePPWG1GkIeTB4spQF8+we#DJe<55H)Eb=Xs4R9
zE9E3%7<!`2=gPAHMIb1$vCPqWFLD8PseAbT8O7Yb_z9z)=YWZT3vA9P+?hvbyQ`Q_
z&qdYUHMwe>z*g<T0>VQgIL%MEc)iXCTk2$QesqV!igkLB8-{-Dt~r!!qrVLk_Vt+I
zdidK7u-{<*iEjXwKb=X24oxCBok&e(7zQo+nbcsz+R;~|r9qQN6C{0Z)(Eq2@TI;q
zMfH$$O79ZG>{RlE@n(pLm$qV>1{=Z{G6f>T!p6A^UWPxU0BZyK^n*7y%iV(TDsk6z
zA~~SDyEUKYiy`m;Pu%T)y9|xJ>CdN09Vp1$vHMz@j$qL;A=?B?0#bueaM`3zH9Jzo
zE|_<Rj6Y{3$IwGSN8xG_+Qjz~DbmFEp2jZN#CN@s^9`xec%rDVGZq4dA!Hp|Fr^#h
zABnIaF#KTvVn-pCF=P}GJFHz4(a(#xfkeV5S_4}$cqI0<B;u%4&Xu?=7rVg>YW)~C
zMoc$6w5CDi-CpbogXWXUx1awvav+JF){~TY$c}^SH9?ddMP)mGgBNYX^u>Av8Q)r<
zNj`Vs5nY#m&z9Ccvd|?+3{)1v>@7L6>*PI6HBAZ-s2DruzE?rqy+`<X0I{N9(5YF}
zF{HL$d|TC2!KGH{b4uNmc3&`m&pAJmz~l`^hlGGeo1s#fx)m*n$s7+P%{3}@K^N>5
z<6J%@FbK}DYaK=_(1_YUO4(EhEUb6OM*ngMSCpVjJ)+QkB^VJgD&=wBd8|_`iOL{g
z`{oB&d<}udhn*3q%%vr4#Q=5xi957?>3FZVIx9}DBy8v-h(|lfF7WGVnWR#I$<psG
zhdt?F7C7KJpZledEG6;P#TTcr5*xd~?c={66Cp%7ALdg^|NTgyIFuhTQ616I(rBm&
zfX>|%Lu+xTj=3hr(kuBT=GrJRbpPNVD7p-Qm4^1~z_@DeOQx_r;mu17!KOXOj>2+r
zG<_!}9H{<gUgvsRyqA4ZN>fT>p`9$Z#8N~}RGg)=U>XhYfH+MP#e)A7PmJNZ#g=Uk
zdjB(llE`sS)l+a){Q(|%=1g%Wlw|Y$GH_JX)A`0}s2-T0TW9Cadi2P-`{dY;<~0Yv
zMHwW~z~OZw<H)>>2twx>P2AADpsP0(A!b&pWJZi*TK>>Y4A%Z<N;|saLmY9{tkc$#
zG!|IA=5Qi|r4p+3<ntW(jHQ7VxS?|7E2Q=Bd{US^#j|9-PIvE6JT9K`Bi3l`MoA@o
z&3i5jg4yg$I#ZCwc?)fO-PzMtinAQ>GeN?rx?%svX^I?*yB-LXI5elYyBj+4Q;&bD
zC-u^C5Dd(4fvkJL_;OzBMBxwidhhoqBa=G19KCZD-c0tZ^=MpeC;eo8`%8j4Hl1EE
z<ArOptKTK_^6la+6`#1=6*mW*MZ3X|Y^&uW>7E8zLu#om{TSaxzs^eyl>h@~Js--Z
z{f03Ulb%As?b}&%Y&gv8vMkKoxA9cG>vV(ZZP_2iwctN0=3eaSW_Bl-_`DA%bcH`^
z4sy^8n=Vat>v=@2F3|UtC2q30k391q{OXn0%*#`2H}gVOXUNGW@LPqiG<KPoeJx*u
zNYw}?g!<dCZ`L@Uax3l!cYt>-`(TGn=x9|<+G)-@1M)c@KvTv-r9C+AqSZYm$WE9J
zfg$2R@9n;`Cu}fr6xAV-FKx$bDoQ^+{BfwkYbA~u31-EIj2-&O#Bvg0&)<yVp<i%A
zhQ;gNquF5*79x#Pv&})QF?9I5;{H{6SmpeOyEM<>au7$%R9?uKk$^C9(S7Xp*kc0Z
zq+D&Yz%_*1C$~;I<0&x>#qZ#o5NJf;Hk4!g!p#!*FX2qrFm};(?rD8QOQ~ssJ!r?K
zW*m%~;erj$1>>-mrki4#W|q9+I4zeKwYCzxNb2oZh7jCNLg++1UQoOMJG^};ii!Oy
z^PLs$Uod}~)D^LXmjLy1!WKl0ddF$O{%B8Qh^jo`$vw+}L&}taoa2s9HV+gLta$~9
z{~d|c01<*<`s`wMvf!11L@3;~BXp8aoVyW9_!lf?64C#%#qSP4XM=5vCig;03>$-_
zJ4s6fMT3+WSNbOUZ~7I>0<I6pVDM*E^@be6;{x0Q%627>8Px1v<igII$`{eyyOA7~
z!tR$O7zf4jrUozy;QlpRHon^Fb~<Uhog-R@H7&Qh3jJIK(FC$SI*>Mr2|kZjtQ`|6
zI4r_#R5%aX+a!!;aPeHHrsdL>%%6Hs(QL|>D@YrE7hHE<el*dj54igf-zvVn^ndBa
zIRTK{yPP-lLpwkU7gt$_Ag}))vABjc%ZAvyrrcTcdZrL?&dDYlH0-=2lPM}0A4-DX
z|1A#}VH<vY4EEt^eYEdZhGRXDuEPE)5>sPE=;RyJC->y_7*#x6^(MG@9gO-Hn!W6H
zw)|2Hun$}@1;qv}W}#`OG&O55kI@c#kI&h+#kNMday4vQ2F~YjRF%D=b`sTq&AJ|w
z!feDtvMfRBG4{Rub@xadJ%Sq@=Jr@RFtF7^BAgdaq4FItr6*rv-yhWx@GJ2M_*C=H
zSZQ3%zbWsQWfFUB7=gA`oa)Hp;T2pfl4$rYfY)K-vhekH!!z5oCCO^{w=_aJ4~AJ-
z01$|LjtUsNiE^^X5!dupNAg&q$qAyZ$)L0mY=`^z^YUf_<1eVOyjcD{*+t9_;39?Z
zY4==y-$h(o*c|9{yiwCDqrN<&*a`L)mwDun3yJ~HN8DBziqhBqhTL_RB&4qL8`Y#H
z0L;E8rE9$jwQ$`C_B=j@-S;xLrfp0cLBmPcD~9X)mmLB<TE4-O{&R@NSMB?5_$#Bz
zVCPDl7Di}4`?uaC$G6kWn+<H5&7s*wol(zyMJ`%PC>@RkrMau4W`p}`i-3zcJ(@Y&
z@pjE5wQIjx;g_piKs)|z@hN{AN0HhIV9qhOt9VrNH_J@u{=vd1O+05!Im4p(U-}2y
zcXM^LFK~x8M%$7gqg$`d+%(qn&ZZmi$Qrcu*sS|=JqV+a<L#GVEH^QqExa~7zg9GW
zbForEN*DuTW2CL#$hL$kyxORku4RsAo-&9MXk-D9<2ts>IqAyZoanDwqWv`+fZGXP
zBBrC^AR3CJri+m!0t-VoG4GiZNNq`|7N*+eA!$^($pb!9`C1G-BWWBv>5V<xZk>MG
zJOLyoBM@c$dDxbK3uLgO-TPNIXwK(r^4b+NfHK~k;G@-{tCQ&V0OjcdSfy@%kf(jX
zFF&IShsIAIJv%ING8iissZ?1MzzS86NY}=NhQxJyQb=k?>%PzhRIz|hrxP?>%=R~>
zUuHNs_w(s#Gwz|LIz%Hm;ScJA65$`z-6%ein4f^sFtdZB1FUY15ys#_+?DHix`<%K
z8N7=AwHir<YBzf<-TbVF!0oQ>YI~q<=*X41_@fW*5Ef-2EssBmb-ASokYj(orAHUT
zyZ+&^ET4)^3p_rE7sRy7m$MBbGU`)XVgATc{S+{^d1mGqjHUI6c#x7`4o}b0zQxTl
z1gJU1*r#S7Uw!FOr#R}GJ~_H(eMNlF?|<w71&%5$YDPtuI$oW6ygpZT`)l%2A6M++
zw?gCei%EV^ev&&GK?4~F@TejP%_w*dmyc;0hGco0;eb__jW24oV)s<@<{yWx(L}Hd
zJzH;L5w?R*kIVZ;jYAy!sdEl0CPr+aDEpsD+bM~HBbb7jg2_RTLybeNC7~AUi$7aC
zvtWw-G(|u~Kxrj)7VAr2!~AcTA2%s%2b;zy2_b-xkd*0}X_S(pya~KAD7OI<RE^!O
z9R^v=XnB$t{jTc$cxsr;T-!$QhM6wIayw?m<;Lu4ke16`2YV4nq}BEiM)q<hgk1A7
znGNkj`@ddhJA_a?<J(+(_U0n=focvfABVQ0hg;Zxt^z_uysUgY9IotLXECopG5V_p
z?UVo<e)MdO84m)bt%rv8@=MgZUl5h@>Xgcq%AR`-<%zFp9J6XXuFUM70u|8r?1NN<
zj#o`4Km}96z#^H+#0C}(iiJ+Y=4$nfxw*H%k^j!q^yqeKSH?K|f`{cKlVI~4ifSv6
zcY0O%p@dI`evE7Q>xIe(mEgy6DZ&u|5%51YT^EnAy(<PK{A1jtcmMhQc#eL(9Ewr$
z|7Qv@K8h~?89XxH3R@W^4<*mg=(JfjM+Cx;%_Iiq+~J3oF5ok8oo2nV6{|#6i?V!b
zdB#d4!A_$s4UNIgs+XPC+Y}WSHJI*#I;yq7mW}tYpg+_BXbFmtevFC3k~Y2ePc_Hz
zl9jStPi5EKE^$CQtg1<H7#uZ|s!{2nVt77>de%zPV+QboN$HqkBA=9vVq`wL2wih5
z^@7n|B%IlZi3q5StWtbFv3LDS9<m5oh%{_6rkC+hxSVE9i;!GM1}xiZ;vjevrWMnI
z(R-;d9XX(9B@aCj8lD#<)CzWSW&RMH9=)h*2mXIt6cTV=C_G8vFV|8A`~>WSaJS*&
z#{H&Wosakva7UCBE+5*nqTn2Asx$sZ_tCtmX=6@bjR3YWkrmu1%4f(gKy;p&*A~>Z
zY4QHb<;pQon`c+cRDH-yV@zh6oV@iKW|;#HtT<m@j;7YnIgo2bTk?rzP-l>3g5`3Z
z*o5`!1Yn)ZSB$52NdA6xAc!z=FX3_@YLNN)JYP8qre{7K<>hbQs??(0hBhx;;*oWX
z8ElGo@U;GFRB!%bXy3*a4iwn?eF3JF%tD}KMHtbmly%bl7t=-B{G(k(Eo$lt)%OUX
zDC78W>u>R0ge{UvAJ;1>N~4kNUT5c>r4ZuNPTTL@_lW{GmK*3rbxk>|$6C7?|Ks
zXW@^THt`*MYP%ZHTiq|w%$T86_7QEH54-YO#=%B>*Zu{g_|#iMdYvuITW)h+pLx9T
zrlFNaGaup${bw}EG7q~z{==k~*xeUEEZ8w$zIY?lF^HjRGX@ZMm19+XZ!o0ulrpEM
z4!$pk!DQaSC1_KVzW|ATdLux0ayZ6rdNgQ@u9&wPhITj-^bdJ4)=n4@48d~POyheV
zv&^q%8sxK`i;r@U7?58R8K&xe+60)YqE_O$DyY*`2mdXEyd{$!;H1n|7CeAPB<DwV
z)*U_TW+O&~I<Fe8eS6w^mMCi7>84fD^t5<wi>x`nhN75hAf@rYi%Mg&`@?PO?+;R+
zA6V4}i&8~e6}-MPzdl2v1gD4XcYrOhyG1&dv7RzU!5G_m3%+$u&^HZ&Z#lTX^p0AI
zYO@gI6ukR!H9oE7H{oJ)KQaM*PX}FY`7jb37pV+A%;FJ_ARQW)eQN4If5xI-Egbfy
zbE}Z*2f<U}kh(hAM@*$h)X=nKkt#ASjt+rY3Z{Qo|El=p*%EwQu<LO~Gs_FOmQ<&W
zUt&HqPE+om&wqAKv^Vt%hhpfArB&PrkhR+DpJy7#%-pscUAOLMl_3PM4g9K2JurI#
z2y}O0>0NI6jQIiAGs7>A%5MXmNVlUfy?OY0zCX2C-#vkNem@O_wk+FoZngBlkHKN-
zeP3Qq*)Ly*F1K)5?IsmIwm$fOtYPSQ#HlM!GaakxKYlU28@ts#5)c@_Syfe~S4Vtc
z*2FHtirltTJw<%f);#^-f3_Bn#i=U;^7E7d`2&_K<5L!5^(JxL#0{l_O{}Evad!jT
zH;g9MAAq&KRJ>qz&4Z9Mz4Dwxs}J15EqcO8p1N7|Im5(_wmXz0W0g3P1Vic2s$n~Q
zT4@tzFf2y?8H|;>VmdxTZ#WFbLa)#am*&tH6!{EKu5-|bh613s92X$h8K-jzrqHQ9
zpi-D@pz)EZyBMQU!(%5RKVmispMCr@NRh#&yD^{cOl#ekX?1*u+#Ku>i|lp2XaS3D
z4hOJB_<d=RggnNw)elU;FsE(!Q3Sbkw`cDdK{IYVVJP-XIq51QC=3G{1zIJdpETya
zfHnWeiU{R|0dXgyyW|na5|@dn<=G&M?l9A93}gLf@dKL?zx9_&DtkDKEmtT*C_^B^
zcdgF|MvG+VL)iCQ&zb`{{cvxQ0|ch&?x*vf+4>=xX?I%RI1d0hu{Q`3P%o|R_c8zd
zKw`AxZx>RM_#-~o#zZ-4g>K|?^PpYcRDY|VYp56oQ2dZAFf|i;*=f$zQ?I-Gj!jAw
zR@seKuJv{~e<QMg3M+qzIV9GK>miMQWm6YX{-XTlAWFC~a(`gd!?AqrDy|X83u_}i
z*ZofbrFz`xX?517`Y&}9b~cjAn2=aoSZ)*cCVz7Pa9)crjqDhLXiM^HKcmqZi}sZI
zG4oIVyQ`ATC(ckB<<?TrCvz@1f{PA$>x#rJz{xUy=m%ah%`A;WwI(goEXUKPwCBw)
zvez?XMPTg}74Wp40;-AHLrS|E2>m(iC7x>!tS~l3R2jvn)8H^}8MxhKuL$0b&&?ve
ztxk=fbzs{=DXL0tFulTeO|g61@*AL<P7d4ww$c#ji`|h7DelMiJxq56*qkO*4oFOm
z9buwicD)5L!W!u+dn}a|+4S$1;w=!8JZQex2I#E8x+tLLPh%fna$vlFRf75Kg~Ve0
z{1Q-;nW;@x$NiI;bsovCXlStTE%5*G7I<#>gH0KSPm;7oN$&)h{m=T9Pre@o1*rYt
zb4YY)b4hQ=P`Rfe*h7_fzZ*tzb~INMYZ!vAeMXyNoP-#+=N?|;Sg<8qu<gdyrgB?3
zZ8pnpymi5e=#>f+9Fyh3H{nsmF>GR{MMKl5!X;VVDTlfx#8IdI8%mfR>hD#^c?~pE
znJ*oY#W;dCQ;A1r2gE4BfQZiR0WgE(6M6tuN-Zpsx$(d*FKmS(ID#VXv~VPC;q-u{
z>nYnOVqFk*MB&$!1FcEHpUE6!^QILM`;2^hc6ngX5xud%(h_bAvVi)BJZ=HCdnkH?
zibY-JVjxrmrfS(o!!iP5>|yY;vX+(Bk-!hYi%$vRBnXgK3z!g4k3nHS0EQK#@mLL2
zjcXS%?<Y_-`W2*Nr|6tx{C;zL<o%+F|J8wA`Z>WO2t8SBjHFY5JaJL$+_7oZqD4`<
z8bAm<Yty0gm{gIDZGLIi0xbD-T|B<RDhpFRyN`wOKugGUfs_Mgl}fA|p+ghj_4}pM
zmA@6Vq?s4;q0xUuQM>OE3y{sxL0#yDbtID{>&UcBIZScpzMyxGo3}*#T{QC*8K!?}
z75%}GuH|B<dYmzk1Z3~0_<D~u5B4<@C=4$>aVd+}nr7P;!~^Q-#fB3wx8(P);5iCX
z-(tiXNM!k3_nI)=k#`uLJu_QR&yCS`i9Qw&^=C&Aw9D5==P5S}5P-v@l&OdF{kmMt
z{fuf|XBYXE=9PqfvphA!emj#0+rweKA5z+r{g#i9Kxa4EwSZqMSs>Bru?a`L;MMZ(
z_qjeC!*=&|4@qiI*ZXD2hpmU_gdYFKlfB)xUJp^=x8YGd+(XxM&HHue!PI&WT+OR?
zw~7XHPf5Zxxwe!I6#&;qY$)pdzRk^9#}~Ieu}O`Ob$f@2_YPceYqs(f-MUiZ%qUPM
zq>Dze?Y^FSdnKRDtEm=k|J2aW%6nnzNzoQ$PD4c|mvR_nhef+pLf0uf(tDxwr$Q#7
z<|<AOkKj;pPD2=2w5cO{@E1-d_#5TPUw<Rx8NAz944NkRUI5u8-zQujWHo}i6<QC-
z3y3DoyX{<nfzbMBbRLK1E4mKYZ+S-d$bUcze2`#E|IjQm3($Vf*%x2*4!}d)?Z(Hs
zWkyn35)cr%Cj1a7lVhQ{DU-o8xU}g!y*B-Q7&1q390eXAGDl9v5sLYL2B>~BV{~bG
zm~UG4pECs00OmITterF#qMh1K!l6@c|6j2^Nx+|}%?6+rf1`&JUx8T*{wf0~cpc;U
zjN9p%Nkcjey<1<loq!NiwG%}HKFV2`Xm-}>fK;4(Y7sTJ@p4*`a%JIT0vNv{q#g=|
zf8>Xyh8&Y+{7bN%*H~ys)CIE^HGpM%I3AI{G{(G4>y-W<!tV#PtHlLC?_4768Nvwq
z+$SZr?HqVbb|T#C8ScSp)ZT8uyurm`J|%sGFrTDOS;NC^x-nJ@eWDt~iNx+aRby}3
zp|Bs><(k^NVn++@SMJsv{OfP&gUyG5$l{%8i~9E>D7Bf4Y}@QrIb9wP19+axaZ)DO
zE9Bplkkgk!b)`p39#rZa&?)s~+cS{dVl|hwZnf8WHGUeoFgh3DM|XN_D3nF)W(=TW
zAI}%4Lc2KTu6gEYY@>UESyf()(66WjBg;^bAD(1fVH;5V_}oq#z)wfc$%G0+MYxmw
z^*1tyOHzuPwe%czVarq67a;0nl5Jsl$*vXc=at_JehyR;)G)$g)BqJ_+2r(kHzZN)
zc7b8fYSr1wJQC?WF6nhE?%w?|H0<hZ25_8YfXVDmo9!4%5*Y`V9r(IiZ0!E6|FiMI
zF`!KP9_*X#2AHYx0=FE_l=mePdl#ho-qLxhUblpTUEgcfbl&+82Eh4uHRkb9zTxo}
z<7lr9%VC<4VhXuQ3ki+@n&?pt_HQcg>3qt&L^xnJ+g^Ws<UY1UOCD?qWo}beE;k~D
z-<IsNGHB_giQVm?hcy)WIMWVABv+}R4l&yE)+VUh-A5|9!i<>(N@c0~Siu8Pf;CeI
z;PYH9jyr#@^!Nh24)~W9OqWvvIE>Yfw_npw+EC(l7La}I8SM9ut`8NFnO&&%S`5|*
zKnLQPm<(>=XDoFPQ5a{uZ5unp2pP`5V6qXe6c~-;BXsQVDrSi`ln*G$PRPpzpH1Qa
z9RBj5M=owV<RZ$;CLbrhag5w)6%#6)h;Ug@&sB3B1|MaS2h6mTT-+C0drx=hCiBVo
zadUz<5%g67I7K!Uj&n0|jfK$}XFqh1Re_${9e#kqP*qeiUQ}H4J;l1pjD5FF^R%cM
z8ult22~NA@o%5fsrgqi;!7MA5cySy`5@Y6b$l)-Ym#%KN5jBKh16~jqVeIrgXNQD%
za7oW&#Q;)ki~?QetoScb@Xa8FMzyV(h(P)0N<a!34L<j4ou!s%Gd`zF&N%W3w8Y8Q
zM&>`fNkX-z+mb>vBj+4m!3tWN3g!1gzv>ajNU!1;Cn`0WP)(p(Dab@1P3Kf+Acxk8
zjj5T?VH!&>n7I*}um0P^TQj6~>;&0FkT(LW6#%Mk-=K#XCy5K#=a>wqf(uxr5uzpz
z!^Q9WgttE>du0!cA>6XhgkrDE6<fq(eBQguxT9cc`ac^+#@(IMOxFBN#%xE+Or`yK
zXD4HUQ03}^8kqOzZ6j)QlQqZF<Z1ddJ~{NxElfQzudXg&4QNXLXGaFFfXBJl|GHtF
z!8z9WbcxcmWcfe(p*XC;9|v=4tsA>L55F!S%_&Vv{O7JBmi5GtQQ3$*2d1n2N4lMH
zl#bi2oY)7+(7*m*5z55a;L-AUO($;yuh9CntWA!Rz+g|7K8#`6)oEnQh^3uy`O~Oc
zw)cLD`JEadqK2oFL$u?aoZs^E0{c{ak}txgS2d8zr#)829)O0<5cSD?)9I=}K!Oqa
zS;(`ND;kOz*Ik*lZ}s`Ry^k}WnWfqPR^le$#aLux!iOrE_VCI|M;L}Tvb?EC)R6c9
zpu5gQmYC|f_#o+5Z2Rw$;L#J#FAMRUkvB=f!FvHJPmK-NcTgd4Mlmga>ZHf{+?y~p
zs8HTgjs8*Dr<~gjS40bVDm;$_Y8`Nu*~PG~jy^F<D%=WX=lAtu^;U@MA!5|D?Rt2$
ztsh@)U!btvIXi$dfrr(>q<IQJ`CoiZ3EVl;;#9u&ej|I^upBwn(_+w0eRU|g$0GTP
zr-%YB+)r4h^4+pIfO9jO>A6F!WwsP`I{t#&-&AoG<i2E|avt$9iSalvH#NR;C@&5h
z-$qe=e<eW9$cLmUcrgHn@gUZ|lQE|p&Zd=^eaQni!NOug8oPxp3X3HA=n7$B<EMv{
zMK`g_0{IDbf%UJcF77xa`XbfQ#7=n*E&~9_0M01IGsN>mLNljHE^t{2n~~X&d^nK{
zgLTHJ9AT&@2uBtLz3}kONfy@i{~fjdzd@^G=5+=CkS+g-1HMrc2uW*K;5tv+s__`y
zV6lPcI!}>e!r^R4SvjAJvB=Mr{zWHiz$1HrWom;vV#A8I{|JNtE(t6r7lwM+d|kGt
zn#5w^)p|i^BB2aP1^`4ne$OJJ_yLN7#Wq839&TDfK??Ja*Y=m;YzBne6bLz9Xq7_7
zHh7^dRyGBNjKTgq&U{WjsVom>V%4<fu2X%T65IK8o<0w>oH#>;KLx=TrFN&K_Ph1{
zQuva?%)yB?DuPjfUBj%tUAm3FUZ>>AI*>gW&?rAmT2rot!?M+G6+t=>>yS+4e9-qm
z3)8{|X73o{eLYXjr*WhXWGr0X(5OefRcA&@k`QoMI7}mm6@SAGROwoQFdUCN=}9TO
zoy#QkeekU@+nwEEIJ=DlQzx<(U|>^ybU00TwY2hS5e_!Me-FB;>Ul`$sB1sC%H`08
z>T)wT)|r<Xie>(A+stOF*+|Om>CyS}ohQxBr@^r2;V9@OKYg*kYLrw+A)~yoy63+u
zz@^k5I7BI<kXm#TIK-p(7!$iJ`r6Mjgyl@>^UA&JJ&yF9x)*Y%CS^ydP|i)Xt=J7i
zxTUC95Cky#)v3`><Rk-aRI_S%X+z!Os3E?mo4zFPmayNH9%6H8BSbPx$h1|N=?l90
znF%2M&4Duv9-DjHL&kn2Z(wexMyOpVxt~*E(O;&}aba^P762XgxRF${n_!bBRHt|X
zPNpwBwwagLqe8A<f(w4V(3$!(u6O{_7cv8NQ$WwepZ7zEDjG_a0bl5mtdX^hCwV*s
zD>^CLDK|U?@8<)`pK;x@$;CZ<NvV7k>V3l14|~IW^i)D!_P`Y8Lt;^MgR{k6V;zLA
zmmBy^{oiof!uM`Yv`(xzn8N@Sgs#-|j)rHEdzb(06u}`iDOj<1(ntWIumS5)p;S8u
zfINF=K7BJzZufb?t?Dvnbx%Q+IZZ7PPKez_c`xFXLoO*5ER<q)lJ4tGS63$6VoMh3
zJ3P-IpJo7Qfa})xvf;<I^}mp}v_$pgV)X(q$N8L74EPk{MWd-<C3Vth1CQ+JXvcFN
z>`dyOSYr$42Ri4Wq?D$Vj>RhHw6w|w<gkJ{&r&K<Du%02z4Nd)HgmIk)VaNaRmR1@
zS&nN!&71|sSN^>Mj1+>3P7b(^eX;DkH(4B4@+oTrwc^nT`2)N8kU<*3Cba7A%u@w|
zhJKz~rHihHR?re-_M}6A6Ko_+*6PWx`ASu$^4LJV-bvzu*%0GxUIIlP3fjp5gq<B^
zLbZfb)$}m;vZP-RNQd7cMs7ReAz?ua&=LnX^zy7T>1Qs?-#$wFolu3ULC?oiA)*Md
zSE0eI4+OajN<7VQH~&5O!dT1&Tv4|Fg;3~Xy|R#0fO5#vw+Kh%{NifkSH)Bv#aA^D
z`~wVTlnR1|sc6Dw2usPB*XzFr7?V3=oQQ^}3gUkMlsl`X$HL4C<!t$bWCId_=AM>B
zX{%ho8qtl5&8M~)jc-ZW_eE$$0!cra<NfQItr{d%Lrj7#o0hh{Vx<XQPUIDL(aGY~
zsq`N9f&H!gzt46BT~g0dGD#KAUK2x6o#Ld)pocr@Z^@XFNl&{!9yb?2so%_%QVv2=
zMYN&4>cIYYOFK$fU7fCOA1}0i|5u=nLoatXmf%sq&IQ-U+j2H_4=jbRyJg8(|C$Y&
z%HNgqmi8*~6wX4~gsQ3I;d1Ub4>|fUGfe2uV_}vv6Ie|0CV8#H0IChv<ugVOUbm)x
zWy3`Hql(=~^=KWPxb;jxtO#Oq4wPbfw!_5g0y=Ux)G_d^6Mq}#<F+s6(wMR~GvNia
zBzf$4X(+E>2JiHS>ZZCG%s?r9G82qf3EN$wn*cy8CBjf14QUyD>Q-X40xyH4A21V=
z-41pBxRzRVuR7SVdhrl}cunFYMu8X%PGuQ`oxOwwB<&$N4(9~8lrRfD&Rs~In$a=A
zA$#O<hrB&dw)+#lR6PR;!AK3c{i}J}Kz9U{3-91TRyM=)Xqjc62qsg06c5k(0^?3F
z{POqP3X{5t_$tnKL5dV>rbQ4P+I&OXYzFy84H|PELWY=r0)&h^Wtm5-+L?8^=hX&T
z><K&5YVK0t^zcuznpFme$Hoe~0UO&eHvKl}jjzepRvO9}y<C%BTUcpf_~lqJ_wh`Y
z&pxZjc)^SAS*S>+);p5)`GSne>eQ={c<$jb*Lvwr$!lajfi>{izSW91tH7LzYD}re
zbJfahFi;bdX7k9ZqrG`tZ2UL+1DHvr&PuTJiQOD7{BIt>8$zqY;bI;U#}To_Tk{6Y
zZfxxHrg)dSeNxQzk`+f#2Y9)(0FJolT@*2wkjYrOx+&6TYdQOu&z9Ao?XqXwxX5a;
z<EVrpLWb$R>7-kY4-Cz^uJn7dlV3&p)hmcT1kRDFE$EzyUP9Ej&zIZNqafaet<-xU
zra)+&%+ovIiHeIP6utCuCgbk}uv4ENeJ`!FlXU$=fk!3?7dF9UfdM*L17(_6JQVF@
zf#kGKniy9Xtu&({GD&_KC6`4Cb)#uAryQ~=I@4n6AuZFt(D3dO(l;l~&w+Nfp$!9)
z&AlFZ3Uh8N*EhMAPP_(FrB!D8_=6n2(C^tS%M2ZWq9PZYQ{O+G+3Ru;HeLErx}CXs
znK?T~f1BT-s96lI9pjAowDQ_l9ESLvck_h~UT2~cCDWW!WdsYH2(a`RVWFxXVZ!Lk
z9eOJ%HjYNWei<PX8u-7^-oeJj|I|I)Aylg9fD5^ViJUgROIcM(F3><Oo1#u`y+Fr&
z*wb<cB!kq>^_ud^(yNl&BcKfZ8`%#0H2EYr!O!am@o@NM^wc(yzu|_^e<WdgmJTF8
z%EDaN2;s=R(A6!Ni?=pLVCF{#wk4Hjkf`&^4aN2uUY9$VC;b!=g??Jktjq#g#SSc`
zum{U#<1DCwd<Ke|7PlOxN=+P>qWzK*JbP{xKq_HR5lgLg7r>g-KX8`>yx-R*H?DD~
zLt)$=m0Z&h&uulK;nH({4_PDxM;3uAFWd6(3ceZ$(fItET4%B3!X0<bpwMykLZ?kF
z-jXY?&Hlbv_i<c<VFBk6okpZuQnT&uqk)Az4STIki`ss?exb=O<I|5iX67SHW%K<O
z;1Nto&W-NLa?~?fJl#^jB-62J_vhh21FOfI84^~k2FASO!{DxZ7${Or)Ioma$fiKE
z0<#ANJFI5N?$<xstvK6y%y|Qw&@XxNFZiUpW=PH7ldl9%+nZNDZi`y76;}HS&bx{g
z+|}v?*6}o*sBQlXt;8Q&T^?TXOw9n;x?^dF7gMbQKb2mK`u(7XO>RXcODR6vySlHg
zHBf4?^5l^ZsT(Nj{dLIF#ur*i;EQgdIZ%gObGN&SutH9+h;IXtCJ7t|F!5{ZN*Lfb
zmHYcIK~vyJJh{}|EQvj%_8Rga{=SexZd&eNPxfYk$~!2vFF&N8qmCM%+$wUZ@1i^0
zKTP{U4Y!JM-@HJPwPaKJR_8220W3Qh=$?yCyGyTRL^`t~bNOx?X#Cs;6`p~wo;Pc}
zQ#a8-un-++cFA_ClAZ-ll@GjJoYhmo(LtyH-bv4-*Fs;;gOTru&+gA~&nj0M8`m8O
z&bb#~ELS~aN>mM!H(CyQTXjmP(_gA$o<YRp6n~Z^gHAc*W0A$+O2mFpqpsGT4C#(Z
z8vGCQEq?YRY5lgs@v#5FioWB`hGk*nBe}50i_X<f&-boR2SsVf<+L_37xK8H`>BAx
z4SLe+hCtVYvZFUYsBH%6+prps>{V|kN4fiArGvNOS})5kfU#e*ebg0$=`gFL)sB;#
z4^lh@zAFTkr`U!@Wwmx|p6Rf_<x`n{Sg(Q2211DxFZZ)TxDG=!pV5E;viHQ00aDSy
z_Bo)ncm4|NCw8!Wwrk+ywXcB?dIJm_&ogor8uf|)1%IE>-}sxFSyRaIzSCc(ap`xC
z>_2JmF&9fu$qN`^#%Pd=@&*D&$^2sRXwceO^YXLo(k>}XgE=<Z&rY$)u0xr9iyY_e
z#I#3)b>Py)PS{IAxVJE{ci7`Eo@^YvHa4q)S9TjUDj=L08xP8%#hnb~_5mD9n52B^
zkv=-MOqeS85t)Q~#a1)8wW92|gJu>f=*yk!IhU95y~yWMKW99C?)991>qfG;&E~DE
zmEnNb7|6sUbls8{vaX%DMiZzqk0o&v?`W*vkF9j}EQiTq-CM_4f~D+<-TBrDHOzbA
z4q9dGw3~3g%W`ixKGLH4R{%;xv?I|LBWZL4Vo!#^;IuA=g1cJp$jOBm=q-#~Ecz@P
zR?pz)*y0aqhVRj(u2!92_G`*!vJw_%t|$LvB^70CPU3Tmrz>=|fl;g1^1<}a)b1-C
zlVcXVmJbifOBD>`evdt%3)i1QE9u-<)~B8Nk-ew`Fz$3fzcQ-~0D7R}%}*i;&XO5?
zp%LwhOqVFN@j9Rvyu8k7**NH1ijneu%f&7RN|zlk#7{O3RPkt17u{(Dd@C(PP{Q=<
zuh~$h7apNY66|I0wob>_mb`DwK?FX=<DL^JPS4!zK6(1Y{BA8q>Ne}qUP9&6@5Ac9
zd*2VhymS}4Xo=k30Mc#q8d#(_WM0|@|Ev#^Mk6rX-l%iCL(Z9uN*Ew1@itE@7Hwk1
zvt3|)EjhqD#S*-qaKFAs2f!eG3RaJm6cH_y>-F;kcU>+N6A~c}LK6ZC(x_jB-kQH}
zr_^iLrSW$RTaqohZXTwdQJe#S-8Br?wlp{u3+XDK#7d>X06F!!nQoW<dnqUr<QPQy
zqXn_FXYS@cr)J8#*3cf5yj?bEwv4DY%Z?hFgD|B;<N`V#$52WGXnDH{FG74x>(@a`
z_u$*{bvYcHKRg`DAt(2PpPiH4!$xSE8Be29l}*DDc{fg+l$?&ow9$XrYdBADIcnEC
zS0)^ub}kiw*&1I};`qd9rz5}1ap5Vje>x-Z(C5}Ofxt#WU);tQHz~o}PeOY0=56Y-
zrtmvI$zt4&)r~@rYn*|9#Vk)h3!=5zc)Fr{@#TN#9o2V_Gx<aY7HkeitX!FwU`2>9
zOaM+P&V4@|;tz#5(fX=o&zR^sQIvEPpaBR+t}AELd-aeAcQaR10_93a%={9gn<9K(
zXl|lvT791eDPno*G7f?bR4CiL5`n>2-gu)!*~yP26~MSL@;}Vl26Oekca-5M=uDN`
zGFISk$(&L*k0YBYzUVw-KVGHgxO&KifqN29!SjY9Vf9I9&JGGxp6r6y0kwF(PymED
zE6`|T%A#NS&z`p|n5h$a#5`WVm+{)WFWy&Hi%b-(c}Zxz^OY}jrZEa!xMZ%pb=g^<
z$JQp4J{h*~sj^t9rIPHnso9zK<x{t5bD9XV1^Q~ZcgNR+lyL!j>5;-GgqWYzg1sZ|
zAVtEg$TXim<kW{>rfM+m;&t*skO0nZaEPM!O#YHhRvj}}PQuMA(Q+?NX-eIcCKf>>
z#h0q!3%A+rC0VB>kSWuoB{V2s$&N5x%TV#q#MLK4qC5eGxiJ?T^*C2IXQnqNOegG)
z4x1_!K-Df!gt$W?%zhHr0Wy!uM&{Vt{QEz5bx5$KLm!W);_xe_AI$hUCIHBP7_CP<
zEQ&VHQMP@01NSTV>5{t=3E6Uyna|Hfh^f=ZgJW%P{l0b-BiI?C-;0H0mZ9*^<aCK%
zYgfc?1bsg+!hLn^Yla90BSI&e<j1qWN!D>d9f#jS_ayt?ARsB5q{#(;f8U4-*M99P
zst>${`Z^eWirnI^^}mJo5(A7sMWBrMD*>bN=#_vte56VM5HT0c5YzGR4<JayC)-LN
z%He@E6!Buc*B!brfv{wRxHbYB0uzzK!eoRS2ZN<*QcfTppGuB+G^|PvU>q-lE+`-^
z`SvZ)g+nGPbjzEjTXHDlA0A&xEg+lbu$P@ow$8aoHk;<Yb*Bt)4>;O?-+i?0D=a>O
zy6JLrTJ0h4(d~IB*y>XzgWb;V$2)CnDg}?V;Aa^|^X4fo?Yg_XJgHt;l*DtHqYF(@
zJ`paYaUmkjtagw){G-0r1qXHGOI;m1g)P#k>y$RX(VuWTKKI+CVPk<;+Uf5&3$ng&
z2A^r@%|%^Hbn$5IA)w1v75;=wE-UH8&-N8~n0CC(RLQUB)%M2jCFkSw^0H=c8rEM?
z!c3(cR`q(>t7v$h3Wwv)>$>{;DEG!$nxI)j$m@!q^n6b%_n#u39zT3>ekBuHBGXD*
zgp?8(dVB;IRnKpJr1hjC5fyq%;nL;}h(5WtQH*YKIQ9DJ62PCNnc2lS$PGawvS?8;
z;{|1$zbNrXA}DD8rBL6EUEWq}y<I(_GGOWlx3%$KZxK#06L$+$_<{4L)R{i*ygt=i
z7DtgolcegKUN?TLNf6X%#?@UF_U^&=GP0c)>e8q65B$X*G`e2&GtsINFM%V9Cri*E
zQvqkgx4Yib1hj}?Vp=C2pg+*LgK+fcSs>dVe+d^3pS^Q6uU|Lw@V#=e6C*Uwr9|ak
zWN+f1o^rJc9hi#P!p>WU=q0I|onCuSkO+LezP+DXq&N^Q*qSyoMxBlG6{?d6SU%rt
z5O&aCEGHXySa%8iX65tbe1ByVp71YR-YBDmTE^Q^0eI!hUE8?8zb9PJUu+NCV9tCv
z02-c*Wk1$%*|_N-A#|D!KN8>SU(?T%#brAo$SGrO57w;po2MIyNK>bIq7a;!epw>l
zw__)p$dW46GTjeVP6yd!>jH25b1~yKF`S2L`)eHiDU459hbKe8GaNBV>HXh6W8r2f
z(Y0a*0C0Hx9%>Uo0@lcGQiPN%Mde$EiiKs>`w$I!&rX(_v9Csa7k2)yV$HP`LzVfo
zwk1gK$*5J69{ylPp}Y*!2C{>{8|Apzge@~2T{$?*_?L1?&E`vs>0VoLGuRZ+iGfLt
z;#GbO^YCyIo*%obT*d3bU(VT4K1o}*jij5tfL9I!wy#N*wd&?0uS|J(RnfnC6&y~F
zBXtp_i*~KI@r{OZVZ&o_(Yeed`bVvUObpgKYB&IXdp2ZlP5dzAfBk3@G1&WM^VZ!_
zZLK2RXmIETKgF^-FxU@TG89t3e33G)iLifg-9h3ze;ClVAc(d*N8VjM`p5BoOG6C=
zK#SfUutgw>!5sxs&tN1uR>aj<j|fr*s^kAh)k&HerqHQzn`-{1Op{rWv4AgxILd;l
zKyL36>;EV2YiTypjdE;A%YnA#GmUXL2Ss7llt-9nN{i`pk-eIqmZQOC?@If*#Yb6+
z2G_02@^WpW5H3J;X;w&!q|;2!!1u$XdRvF=@Zd3y{{-K^fm+w}mTff%D{0M%4=<25
z?%UEW!T8^y)v+J2y){^6^`7&Q$^H6gqR#Y#S5@aov^M7|=M7=__z-Kk)~<Vcna+m)
zJY~mO-kT(m{z;lwE3fihL8IQAMS1JO_ez7L-ZHi+rgsAPWKksu04)q|$UUY(_ASzE
z0@NxkC|PmQ-ZE`-XsN0A7QQ&CQ`2Y{e{stwc1h&wT7Li3OBfGLZq5Fog$#*S{E$X-
z>S<fNb=>#-xW5-ZWc|RMTWJscSh&$6G?`9p&U|RF8yYj2Q6&|?aa`lmf3kNBkyR3@
zmml7>!I1z#)<^=xG}>W^YvF?EKe-c9CswPGT;Bs<9o|}l)mo93$zVQ5{rT#yI#unf
z*8k*UuRYbRau;Tr{ueC5nVMn4$BA%!ELt-FsQT?@oBYIo-vx*a1irqT+cjW@`}!2T
zcMwofnT?bXoOSDZMRiR$?K|xZUQ{?Cry=4#fQBed=4Gw}Jx1@hy`=rZ#+$H4T}%SB
z2$|P?lBnqVJP6~z{{x6NdN-`!WYh4F`@^JQq3tX~96f3l@m&OMoexXJ8(db66ZQ=p
zp!5=fF=I*pKw_%CkTXI@Yy{Mcg_gs6-LdT`K+m<7&mEs>X5XYXzt1`8XIw>|bpu!j
zC)!?CQ3N{n%q2ct2zLCoKii*ezf^qGKFGm`-uM3QY3~8=Y3|kS=_~GIs{kWgv}l0@
zV^`^8-AZ~^<j{-p>u{KhS#KsmRosaX3eVhWRDZtEXas*GpNoBsCzApWu@k$TbVv1C
z^-LIaB<eA$8CD~oDX*or0zZz!QUPg3XcnDF$wKBy?GNpPEqmp$y>;*;<6^798$;Ws
z;HfM+?D4&3#j+l<*6x(U<igH42z_uy-w@bDF3^Yw@`A>HACU8fl!SN-e1o{L=8O6h
ztHNNVrN2B%&8NhjA%S9(5s}Fvu;pd`2aM15({M=>lPg?@=v?^<0c-&;(jE<v#NJ%W
zHQ2%WnMf7v>X|LSWQw6dz#DIm{iG7?y7W61bFauIE@uuf>e*^l{*v(245neeBCh<t
zUOLA1+$Cb<z(oLl#<zL4{|e8QQiF?W7O-E$KPq~2ns<{R3r1F-b7j!h=#qtzQz;})
z_`Yr{M{PoWRR=AR(Jlat?>GUeQ}nm?fltg{$%D}NU7&EPqinjN2Fg>ZoG?g=gr7Qr
zz;V@S+p$EAFBzzqm<xL_Kw{)!YN4`GDjg)zhJjOvhw9NSoHsIbTs{(Rb<yBFhct#2
zgy6Zuot;!dJTrp-@rZX=$nKMMB=2Zve{ViN#b*oeg`Zt)>c9j<oF5unQRj^=or_#o
z54joFuLtQ4Z=TRL=X&?C8X;qW)d;N_v#@%Y5fy7NVe<07PgNTB&(g!duD~KPdi@+w
zbyi>*?Ws-e!95o3berDJ`{q}ZBOjtgmMKK2=iGmwikydCv?;w=A|?Y6G&#~|{fWRQ
z4PfJ(a5=U71|X1`yB`(pwRqJ+PlSo{k4A_Iqr(SMb(r13YRnBpI^guXfLWi%tOE`f
zohPQ`L4|*^1ytoTtZ7}jKcH#dY1dJ!X?-;_F-fFCq$N9)4t#yl3vs`cPW)yG0-gjE
z8tXG|oF*BiiR9d<Qcy}IW{o{|K`H^7&S5skEqKXc7NGuqXZswYIi6TP`|D>QGsf??
zRy_rZ-an<+HI}tW+stQnXIdJm4$&)VOx;ROSVFao1WbDT%@CyU#REQAUG*+A`13Y2
zcpAx1?mFi+AWkWXP8qPG7umP0B$?&%)t7|JXwBLIFHYz6kl&bV`%TD``kT0`<8$QZ
zv8>)bpcq5=IjQmFAlQHx*WX_`-+xQ`bugOxlI~J+_m_UzcyRWJ`drwF6>hIem1B?F
z1h#yxV03!-GXy8<nBc<UDEbNE<*lE%)kSnf!$%dnGUVKkZ{9qygxd=w@(TsTGl+L9
zi+5u_nh7C@cS#}M*$MDKW$hHWJZ!qq$6S7`s#PS%w57KBQ7r)qy<vz}UAT(BJ((L<
z^i=E3BLB!2ti0D-a7LW@{uD*Mu3S7VL@`8LETep_C%`uRV<}IZfhlD1kT^?Lo<LBj
zV0Sqzv(6V;YpL`D`sNG%1AV{9I%%<^nqRdZ(cn=nVb(j4%9a0=TpSnc^5g;C3@WdR
zjbhZFkVJa@%K^O2P0jQ!quwa?Mr!6Xn&jEsa9>eJdH}Nx9)!X$fJ9?g1Hq*ZBAk_N
z`cXVYIo)KVtk0!-7vQRfZq3L|XXG$2E!%{$mnYluZ-RP*2$Y4xC>X%n+xc}6+1E%F
z$#>J68B1`OkIQzHxGH;MrJGNOUS%!t12LkQ{cT10_yX*QYBOUdtm}s!h<ty>b>ist
zw0$N&2qYoPt<sHN4bn((jg=S(_HvF9iJiu9)x4?i$F34IL{@#j{0@F@xq*dKX^1DN
zSRDJ!0voIo_nEFW4K~odW{t2+yH9T4)hD||ms*e$2~#i=_InY#rmyp_Xa`Fv<w%xn
znA}e~+zb%sZMD9Zw07Zp{AUj9eTzvrB))Wk|7HH1MaS|atK!m(n_~$T>;*|UwH6Rp
z_|6B#m*?uN4nzSij9n*6<$;0yn}hM91dKiNHm8S?YclfB??Od*AJJGUzcS#!Kz-|8
z?XTlpJQ<Aa1i}FXhg<%a=qdh}|A(-5jIOhN_jbd^R)fa2ZQHiZ#=K+OZeu5nZQE(s
z*p1z2r_cXc@4MF6YwZ1Hex76AbKKW;9KRFRvaJCY1uS{=Y`|{ow22-^G?v_HN74k(
znRXHn9sc?op;y^=vp_6X`a`thN_z)>dq@{453ud9{BaRj^dR0f`;TUOSsl=#Y^Qm<
zsd}~CgzxXVmOo^icKB2+A4$)=5TT0F-Aez$=8?+>YmP_2*}MN2Yd(MC!Y5c)*!OI&
z$h80#^?0KCts#$3XeA@0A_arUvq3}n#$+@s0+gexKT!yJ9-q9ZX4|9bbpJ${+-awf
z;{E;)fkqBGlkzr`;AD(o{BHm{Mb?mcHd|VcKcG%xq>DtJMtCka5>`K-egC!j8;&dp
z?3rUcHSvBAu-o{xMgiSpd>X@A3)#Q<kHKA;@obs>lysob7raj>kuGG_FH|djMG#tT
z4GXE=XWHd@<CH$OaC@NuC)c{O>M{p{c3W1o{PC0kX9@@?xBgzkG|aAcH?jP9A*WHS
zjj2Q`Vf#*~g;U{HC<R3FDjC{$T@GFqb?ZhT$s%|Zn0QUH)A%!ZkZwb_8q}+6Q=4p9
zjp1BnC96|9Qzr=G@Wix2T|qu*TCw00I;#F7KV(!Lq`|NbBdO|IA$^xRv1;)4dtZ|Z
zwrp~@9S?^Kdmd)aOSRl_PyVX=m6)w0nTANYix^k-+uLMT+}}EA$fV57xSofP_N({I
z7q?g79PltHCpUULH*G6AT+08$o@X@qS;KZEIsbmF*(|Afp4mE@`?%>Yz_TE=897l-
zgR=@s*qcomD#!2R-PvV6u7e;|JPPV-Yk{192Ag8oAx9^ce#Oq#A$O*vIZtCtd0}U*
z)x;!icdsBtQG~sMUY@VOHqk3PS7w&@L43gfMZWwmoPwn!EYfP9${%bxm3OI%VzQt*
z-G2f%3Xh(RNHFTX1Ycb}>!)3m8ODpGp0vBcnbqivV(RTpK86gdHM%b0eEGiTJby}+
zS{*odI#^i4sNtF(e5W0=<XIY=GUJ)5kjoB9v8nSb1I*QgT=bN&AvRV8kQeFk>0?R>
zE-Qp^M9-3zp2v}{F+h?d^Zc72A+ePHY6|s6<NeypPk)Ii<wt+n6zn$jWmMqi^v&-K
z;ZvZ<;?=$)-)U9Qiz)<DGmTT7)aj?6CEhE%-8g2WAfAPggGxh>%(pi;sjFj@O$sS%
zX*-meGmp}9H)1O!8K_Qx>xCgAuie;6B_5$5pIAfG=tmdA0xF8;>`b99h5^Ou9s$-a
zZ=7X`Y=AaJd44Kfs^jRk)IsOrcZf1Yc|^OBWK<v$I@@sYH;Ac}!bz+isYIfMsA&<a
z#QG9?0iHz8lHz1I+fBhrHe6bS$449L(Yhf)#JFKAk#J~3$pHpn?;X5nFP3V5uk?=7
zv!(~lol$*pnu%L1B8)eRB{rl?X8WE<#8JfVjR@E}ut-r!d5As{z^y3_pg4y(c<Z+`
zCnbDqcz&vAnIF^og|hV~+sp}OtI`~=&pfeBd5ahZA0GH?!49iPd=!FmI@ez7Z<Qpv
zGy3*G+lkjG1{d(O9jHpJ<L-{J+mYNfz~gP7K#2>l!@7>n<bKkr*&|nRP&%S~En;#m
z48E%X$NP^5;vjBltLTU<W<sjHxk*9TuasYoTh30G>F`>zW!+y(hH;kLa3qezQ>5yG
zm<;e|GE2x+thVYTL-C;uAPAPq&X1YOyOSD$G}{3KWcS5TV#_*qB&lAwm*B(uRPl?i
zud}?6Bb5uU)zimhTsz@h^chngTomjuavPnKJhREb%Sec5#Z>6kP9)ibMLKd{JpxAE
zjOyT$^wNL#u}Z)s>YWJMHL0AOt7-rHr$O5NvY1o5=y3NjS38atWf1kR^G79mQa}cL
zDqn#6t`IIKJiXf*%h7W>P6`BehjcmnE62HoLaEgF6~5+@C{%I(=9sfuIeZK2_vaK-
zT*h>X@H4`K>XXRYkc6&geqW^g<9ku_`oLg1ob2B06}~e_Yzb&527+zRZFx-fk)KqN
zAw^L-@us{#KeYukY^df;r)3UGK3@W%4INb!dJD;PsWf)v)UeS*qdmjywbO=#IBb+0
znFQppbmVo1#i@&jO-vSfUpB}S!~aN!lk&J-C|ie5T1$FkX|N3EMe6}dme@`-(kn>c
zjFl{tiCtwv*Om2Wb1`JF%PIMUr?q@}<sH~F;+!IkNUYmA+d%5db;64&td9W54#U+2
zgILq3uQHVTHhFboU^Xp0g4|~-4rHG@rl9~MMJp<G^{fhO?OYvFZoIVv@5d|*R=wel
zB;HqfN~u*T?%(3&<2n_EaBszyfV0B|`L4E9rJ5$;?qKdyO=NL=_wUj438Kmo+|s&(
zVJUOxC_Z1>)IRgs9S~l>P5@}L395<NBwc;XTYDtu3w*}?ljr(=mN$DkAK;!d%-=eD
z{3|f)*DD9NKOT<R5y?I0^U)Gi^`lwhn&&C_uhyF{jyxgu(Ca_ULPlc|NU5Y!+$si9
zx<W4n33K={khA>xYz(b%?a&UQ^R|z2W4523TW({}wrM7r_gJ3<<<fyQ{oQSv9(CTP
z(`V%i6kciHnQo|Vw)5N{d7G*B+<IjQneS)~JcLbC38tj$=%U-YiBk8l4p=x8%yQ)w
zaW;wd4p7O0*tJ-#CEKw&8ojt(%_dx}vm+YiX4S}v<;Y%%6{vU?h)d8hCGA}mg#6z@
zAA7e!>w+@SsD6RPkz@lTFVMlw<rq??d0dgWod=ZgYPG+82NmR<jX)uX(shRex|YpC
zSzx|?9Z$QMJUuAoQCgU73&mYtZ;=S4)4{HaLGTjnLgw^v-m(bd-J|U3!2yvXuRz=K
zi|vccV+@dO3GFOW)S;S4u&jnZ=W8ba@d{}V4h4l!=&xJmR~HOa*N{d&?B6Bey7S%D
z@jKR!j9I(te=}dX4)!7Q@YqtkM%3X5)IaN%`ve8Qj)$%(=i%G+X3W&v4<6=)2ItZE
z-o&xl7;qZo<{Yeu0y}qHVpcX&=Pij#gcgIp1;Hi<0n9PX;ct43(y%sm8LO}OenT~f
za33916i$yK9Kis@vZFW?IZ1pWBy8>x4vAxlri3qq6cyVI-C9E{576ZW(7f6E;PrT5
z-mDJ8_aGz#3HcJxd*dYBx3l(HQ4sSL1^0eLQ0_-H?#eLW#NB5_c0Ao;tciN|<|Oc#
zmYxb&%)C5Vip&p@q|B;~DT$Eg{HA6*I{ivzP_-DG-!2M-F$Ja^hdeuIBtCWZ_iS+_
zG0Qfr3UPxsy#@Npg;fe%_czQ^N3oUkjIt*YK5nE2n3`3N7#a(kM80$1VOBQI;){|=
zH_-Fyr8^I8)se8T%$6o+LKbh6pO(wFqy>s1X9`D1*N}joq_KL763YG5jqEY5Ql3m3
z@cETvKBECx{{H<(16FtiQjO5<GZZn357u`!v1<j#_fy2Y>64ukL3I>Xsjgj7C<GA=
zwT<S9bWzN>!q}V7-bS{+n@Gz;jO+2NsOBAiHdfU~YaAw~q8JvNERnDJp$@fzQJJvH
z>v;|Sv2^K3>-plmy`=E<$cK={Co48O6jnH^v>*%z$Z@#Zd8nVmAg+WQ84CX5!?Sqd
zBIgUfc5m*#0HJDX+PRs5_36;rI6UTahwjoacN!>xXgd_v@K|dZdbzP`#%1lty7=Qn
zZl@h`x5+=hlfnzQ+c{zNXIK*LHaZ|2i+5`-#PepTn_-m}^<o35_g%)+>Xw(~Uzx&+
zU%U5C05$Z{ZvYp7r^CyZjeVQMP6XJR@#>Frm_-sXfa<ppi{Qd}YWt?0^WI*$od$6t
zm$oKC${l};$yJ4t?A>JTbt44xRfs`dk~ANGOG=dP<0P0sUZ9o5o%y7wMwPH-7PwXt
zD?y6s7y+&VhtDL67!Xb03-^XeRVmi(`VM^nJf645;CwT_wErqI9P&#`i~CKDy~wyg
z%Up7{lJjc|#=uq}yvpqFLKHGrnoLWu$%JWW#_uo|BSYOtR}^2@lCD*;ItbAMYV(b9
zE)T}M-6p@kKid}#_)^2?J%HU7zNysUGNdNWDu<BKN|26$Q59t~MC|u`hj>SmraG^(
z1Csmpg<h_=2o!R17d|1yEY-y59A9lfyCXCP&RDbd<$=*N>=c8^{d|Gi&7MyhKju)_
zQ15o$^mWMuZOHvIw^coVu_+bwtH;};@G~V_NY;IwDK{3lIiXK_Ue;bt$BZB$>P9L*
zJ9eQari>?!?B~@_;bBRld`c3Zq~EQ5ynz;;fsyTJf>hTWZ;+%Ze*>SNLzc4diG9Qk
z^7v1VhL(f}YI!(nmWJqt;Z(mF*sFGjz{adDAQH1n&F;yzn#l52Y3qWTwi=&F2v!!m
zp0hAmhRuKDdpU3&zL+6N?2j84O)6GwoEA02X<Sg4#GOj`Nuk0$m97$9B20iTtOKc;
z3eC+q=N|)uR|XBc)-C5_J~i8dQ>_Oe@jy5wutGMl-l~I@g-=RU-fgr+u-Q0F{wScL
zlM*?E{Cs*qj16<q8F~YQf~AYyZJ=5!vWz7YodmH;l9IAaySTdT@*^S%%*QZBS*!KZ
z$OJ=2EVyQ~Hml=v9D1~2#83O0)eRsrl5IdwWPC+r@;jYEePM(>PA85cu9?K2B|`qX
zIq@#SU!vMn)O#t~q9_#b-mGfuH?QrFyHM@XAyqvt2jRu)>-OQ$qqEBh&_M$5Yk8V<
zGK=x-ZP~CN4-M^=Nkg_l81{C7ef0Sg0jE}v&oUahhuY9Ya>X*Qk$M7qZ3%GPT2-pv
z$v+g_FbrSW5H;B(;$vIMSX^06S=<%=B8ekS-z!T%q)a}Tl;UJ&EMA)p-_iyKMstJ2
ziNyPzSWc~yb?l4~av%6=%>VeK2R+Uh7Ur8ghdH*_E5^pzYbc@a;05fMyq8n&`RB5q
z?H2*^ScaoJ&XZdRn!eZ)e_8>MaA~f}FR1m6)gpS=`pFDOynzlm!^83qsax&uYG8bX
zuE6Q-d3)itbhI)jsr_rZ_ttF@gY#O@kL<5a_Jw@2&+Yf^ZjJYG>;2c(L{(P9w{-o?
zY5tsy^-;uM+covA!xwj$3pbchH?k6X=u%0zY)z<ezAF}`(1t?y<V6A_WOy<DVv!hv
z7!&bky)bo`&d-oU2*Ab0%RWm5czW!`sKQ2}FEHqH9uS3gIQwKIa@*HRL7|-6e&9=r
z0!@&51d*Ba1aEaOAytr$HC`1DyXKP`8eY=zgfrRVZ{p*rQ-vXGpe3z+{;A{H#l4vV
zo>n>hv4v7M9O4I-1vxM!OT>%@!@9RV&>G>a@_Ahp8z!bsR15{G%r#ZE5oMDn8*Q$7
zo;u9%@zX>v@sYdXY1_#~&A>{0{dS(MM@+|~;;r=Z4DlJ=5`w6w=ddgE+alv#UZgp;
zpnav4AXjI{eoV<n@VI3MPhY~3>tQe$E$CpHh+BHl@6qBiJ_H2A=u<VN<*2XE7r5pD
z<)<s?yCjw(j$%m;^a1wBTJpq3`+O+0!WFu<qH43Gk*>^Wy>(o7DK4e%=!f!{Z%u01
z=2#VlF5HK;Dj8m(L%f_Blp1fBP>FuwmmKrZP8(CMJR+iJC-*6)Dz>ZTGsLj`MJH%t
zq0~!Q@{iPA6oB{uL@Z}TQ;2j(r-=}OasIk$%@0|^t<MJdc^Uqc7pAnXDyf2JcC7?O
zy;aXqTcl9461Nuc29iGN8L;%E{GvP+Oa|X38Bl82JT&TQX3k_GMpEC<f+^M4-8Qkn
z?xEOiUDQHC#n6be4ZDNUrV+WhaM_$KMJ?dr;%+X@bO4`27)m|+43<%;;UTZoEaS$Z
zUn}cw=2g&;c_^bRc)K)2drqLVyr9@cDpg+RFi~G<+~E>3ZBChEAO{FAyf33Kl1zdF
zs*nya$EtF`kcQg02rc@j4RQx<`J0_&WgRTM+s#^KJbr__Mj^B~wtSvxz$>8m4pB)!
zUIm-0PYWm&DxqbXWd#v6pnLku@eU(NKp#{}5hp!GU#EPlTkySmLybMYId~=;?xwtV
ziRaA}Xdo5(I^!i?_v}RnLWzRjZr1I&B>YDuQc6OuL$N8v!ji;L0RyfyG8DuLJG{zc
zmE95LEQ35B+wu=F95m8U7AFWD4#cU8R7tPfI4F>caMmieSrR>Z`l|<>pM;SVrr2Kh
z%Ntn-f+m*_Uz-lnv`{qgj2+1!9Fn>rvI?DHc?&{OYDN5*<TpDm4^MLwO+i*CO%jZ1
zuGniVLeSYQPpBXq2R&<~kiL~-@Bc#PVcCgVgEv`|V~@MO>Hqh-Yf}QbI_Hp+dJFyM
z0r>9f{!SSnu0V`omt9XnG2wbw3SAf*zU+2(BV4Oi?)214taTP6>Xa}(6^XxDS(zVm
z=W^gZ=2b5_qaL|Y-mT+?r;%>*ejF-d@wR@58G#ytLfJULRt2N4s8R9^XGsC?`*Q^f
zW7T9M`}L#S+ol_vG6%G+xC<1nON)^0Cy@LTT`n_A=;;k5K4vB-f70N1F|hI&`az_l
z7=ghWqaBf0DX`&UaufZ`f%o_zxVVeqm^_J#c<74ccgAp*+Ei}sIdAF;NVY`A-mE)I
zX;8m;h6{;Vtj*Jqy#~_uMe;YG1-|o$sd*UMRVn}>IA<EaPy7I05DgR!o7zwO6p-tP
z=L<0oNHE^ca(Fx#N4Gjp5dK(XilE1R-^f}52{U)8S?f-+4`(!Zebp)2QGWS)m9-8x
zh3b`HP1KW<I>Aj>RzQJqlbRs2oj#i3?J_l4fY!D@z!6<SSfAquw%nrrvCz`$6<v`Y
zo!VIS%Sgf2CmZ`2Y3ujk;?Uy>X<+R@2d7?$+Zl#djBJ}te$0#f##e*{N?8DxwM|dX
z0A!<QZO{Wkm(^Q1T3{=j3)Mf2N%}2aa_QSuToMRBGsv1&dWTiYkY*K%LS4p~X{3N(
z-mGVB2Kku8@*#`$S}>>4PgYO>G%{uJpnS|xE=z3`Z<ktlNc?x0#G-J-4e-mt$eSf}
zvTQK<kDd)%+s1PjZn@j@!b-S_4KKaPMy7JbRq*g#w;O)1Udi(<!{wtuobNHN<>qeQ
zo|pk{x77FfxK=%Otg~-Y`U_mQT-U+gm^-T8FN~RH4}I;}3%y3|WRb>_VY5tM=^p3|
zO&xN1cqClfTbBkodB%0|g#kZzojisM)5L?R&56DTlQ;?Z?7cvb0YL~9@aU7`A7+D*
z_SgL-n&>!1>>`P}{~_9NX~(}jTU<ZfGQgSBJu%j9qqVs4V!QK;Ox{+aO{srZate{D
zIfe&{-==SY%aIiH((uP;r?=x?!@LUT+!-8>DUKX}hFW{tU0`7`inPrgjXTW*ewG!f
z#CT<M6m1W6D)&WpM|YakBM`4%*V{2LTZVCcZL)LY<3o=)dX1V_AhsZ_AWiLdZX<+b
z<}mYmc%5p`y#cQ#Jf#>7ahxhZo*9|{cYGMLyu(nAf444Z0L4cW6<4$?OBINiGwxdP
zmwJZ~R%X`(;JJ7zjxn%VTGPW8dBlBRLYAkG^5|R3_L9Xn(Fvw#V&g7SII-b|m;Eba
zr9|770MfM=Ssj+Fd-e*m*uJ0z(P&~yL?7dc)(ng9&$m@ncaq1q!Ff((+BrrqN~*qg
zyn)4}t~oEgZtGfrF2t}*>-f<+L>(y%`+cpstZzd;pgDbLx_LdhZ?W@oB)Ep$nXvvk
zKuY9@#5(-M#QB~b9a-PpH)Msj;;b^xJ?Xi@to2g3-EV6^l>(VJfxO1!(&NHbq~Yy*
zAH5s=4ffkNNq+9_GHt7cvAnJ9@970E^-BB;ci&hn-;ycKlV{b+6~9u{J{A7vjgS_j
z@Fzw*1PX=KkcA_m1!!6}3`aItV?(p9EW67_vJ?zz<w)mZw`6kfq2Fgii_a4HeHpl>
zTci>GwCRp%*s$ekI&b4~ijBGH^p*U4gT$X${ps;(6*`lIktvT!%@pu%7sM<|7l6pj
z-r@}BMTWFEBbZ9cPi9!kMsSnl%@7EtD7)1?1V&L`i-(2VR~$5W2b_ijc3as6S^E8O
z6;tIbdJyl6+l_P{=HM(7>0j&mFQ6QG#3(4trq1AljyCH!IIykQnYF;P4Qxf(jN|}D
zSKpVs(Om^-ZMS}9!!3NFL8Sl5!R|4#kAbzJ3<%MxYO}~F+HM2|e1KR@W>iJ{AVwbH
zOIE06rHEYai8)jhZLOYt>KdIgWf-p@$sl<%+}WI5qjsKFwfL8DF=crJKe7sxk#6+(
z&C4+9s%&>iUVJuul9kl#TGMUSNAMka$Lb&UsF1bMHKC8?iWMH4c*oa4FyRn%MV7Ji
zYUM2baG$1@hD*mc4gsCvy*8#R6EJ*D_ZUX6L-{m7TNKh#{R_CW^_n<eTKyhXrVxz`
zHSD%UVLemBm4tJmGZCkXv!@W{p}FGhrNSv)A5y6iwg;I<EF!<#`<n#dy02zuDdInR
z9<G5MX*(@fdG!`sb24PC{dm&vFrQ0+o>7_M)AQZYiOi$L3vd(<;ss6P7<&3disl$!
z53{~cH@{#|3>{00bnw<Z6W65$uQu4Cq?+;q*80i(%Va3@G^|Ra=6AtCRd0)x^AzwU
zft$rp!iDQehN{GwKg#ITb7|sLnor6oh0;ot>Vv7ve~_No4?<eSMqtbBM6MiZF**FX
zR7_QxZ)<Z>?f@-$f;bFp&#i$TbjY+O%1s`1^qlyIG<l}O%D{<5=6(E7g;;%)kHwoM
zdtalmPJc~T@qUlj0t@n~`+*M8pXTrfy5Q54T@Ki;tL@Z?0k6=dw9~_Z{4nS={jVvy
zZu|69->6$`Lafir9^$umm)JG^*e3`?&1^XnQE%bp0cjtro?874q|%Bj#zy)HL&&Vl
ziowaA7m2_&OYuBI+70VOA5h;*rqwwu@9O)zV#5pbU5U89(8S^rm)L}p&gu}eQx};B
zShb9{MLvkpOG4zeE^mjlEc$A>Z?`yp+<r};JwUZpNZ%nIy3dz?*R{29$E!ie)kWkU
zeK$HP_YwqNht%7%k&d^hLG5!rD2Yge;|^E3v7fP<wI(#AOLS;0oW&~6^EYJ1;UHzB
ze$(2?ASNcg_@hv*MRsArUOKtCLd~by*3sx3PW-+(jm1l+nelCF*BMXQdKAEVv7p-r
zP{_d&RC>6S{f#Rio^>Heb3yQQ8J78UR^#ma;M;*vTO?H5)c_|7?i=Y<Ci!J2CRXmv
z9Wkfq-gUe|^g;c2u!Z%{kRpqAm-)*oAfSy$7%@!?=~M+M>gRvb{k)Vco%dNI>=aqY
zLo=hp+e7ijD0ew{KdlPR=QNK#?u$T#PwPf;ki?HiKbnbO+#hExoKZBB`UZ#J3vO#d
z#AE`Bww$10vW5aqY-Y!)O<nx2q{I>X)e0(>!Yf1FCM)Czz6xTK@n#8Iab~3WX_C6c
zkTRt2)G|+BTfg>CQK#iLJ0q*PhMJ|Ye<M;!kn$PxF_KZC@do$S<#&1-$R8A457dRP
zwPK!~xBd|!Odr)kD_u-$JhiZEjt?#g%YFkKZBz0w%Fy62x<E1nw2Q>W*Z)TDgFKRY
zEzwW2nKw0AardetSUEG#o3Z^wwDulMyw=%>u##Q0YjF(lx_T+&^0I1ov)dPe%X9aM
zdZ60vE333vNnrdMx7!zy#nj_5oF0bMAcpuksnWc7vF=XLD|hjRYsq(4^l_eVb|(R!
zIs+c!W!UH!HPfsM?ZafR3+IN_`;%}mgkUM=>4{Jo{+?q4knwMekN1x5;VYPuC{NmD
zIVFA6$2^U#?m?_WYO@j}1CTiVGR~I6%I2YwAYLJR7wuHx4yV{y1QDtUr%47T8|q!e
z%zN^`GQ38os4c;%93hoAydBqUcmW^2{4xX$&1b?KkB~cNHCqUH&h4>Cnu?(O&Ge)P
zn?Y=B%qQ$iqR{;vKZ0a5hV#7gxqfp?`2&_<te&E`=OTVwG=A-#=BN~NnrX-v3g|~u
zB6ACDWVhOeb*CKgHhQZU&n3A%3oM%_7NJxf;~=B^{0_2QIOnR=Q9~>%W1vqK)(`KI
z2pQGCH3>c#i6__*bXEE0_`YerlMgg;OTT`aBEr<X_E{dahXIbKZ+>ViZ3DGycA|NJ
z{Z#XIrOhwL4Z7diX&QOafHpE=?hQ;r+Pzk*ys`f{I(-Cjk=|!nb5Ol?OG!FaF;ZuY
z9p96n7qte#az%1^*VY)p3HUiPRvu|qVlk6iqJ>><{jYtn;liS^DJrKiKz7LPOR%k6
z=o>pGH(It?=abHuc(~Y!Blyj{L?&6pA~9g(*iJ-DDq+Gw2>-@t3w2JrY{$cRM9Vl!
z;5;Zwb1T$8&LoXWgYL6sDnuE~XB4Pm)?p-{bF#+)cULZ%1uW0suB)q7=o6R-ubSYh
zuMV285%IIDw>YKuOf6thQA>%8lBFM$oETr5?q}`~pl32~oxnm@P6%xE?D`vTsMXF<
zHV2_6zmG<E*7bv_@J3RxemkF-xuRco>%=(E1y0LOaQ^WR{5b#d5AHTDD!elA2C{#$
z6+EJV3|;i4G^&vZ;hiQ{m8K=vBkagDLPYMkEuF)w+WAg0+m(0f0Q~s2Zdb1dJ!Mc{
zaWm{h#n*J3!`l>6=k}H#c;$aB1j5sQSqPZ~NluM2W2z2U%>z{AL+4aZt1=B|tZ^BH
zTJ2=ag`Fq9qe;mv%#YVFBvS|m0-#uGF6nzl1m;*er@ZFK3Wn(=POjPsyN>B+-%VJ!
zebJfK?bH4Zq<FVFNoCXX<xnyktHP`-(*p5uJxu;f(Lfk}e*FU9UE=kN=AZ;>z@<@y
ze1~XJT%tG=L#hs46Wc&ZgQz>`@Z)upHcfh0U2=o20F0GfD}?~{KN<q~*F*fHA!2g>
z)ezACM??Iq`OpxMl5nGgpNxV-`?-kDxYKn!4jFyG4?s4v6-Fils!mUyRaUuhzp@t&
zgmx_J|J4zF=xO>K$ou<WeiA)u^e==_XPWtE(VP-=eb@*J_mTf5BY=Nwgy~WLV=MPy
zvJ=h=B<0_<hng%;cKxg&$?0(`pbzB(;Agr}c97NbW&&X8+QS-9m}ieb<rU{eFDYJk
z)WPxKU=D83QiqG$tMG<?v_e{tSoSY}?Jf|+N{}BFUSM0AOn;Ef6+6KmZL+Txt)R;+
z*!Gs+BOS#kGy-tsy=GDbj7h-#K%w7Naq@y`&YXGJ`S16nsa2(Wom>t>ciCDR2Bm*l
zfjX{+msqol2Yfkp>NKIFb+sqHRdNGxdDTnpboM_V99?*|%e&(4IK|+6u^1@OD%kjx
zl!@*`fEQRSH#8M-;r3+D;dg2b%o}!X?2kRcC$nt6`2pxZZ5+4PHYrPd+1ED3#AY|N
zBQJp6gRG?GckYpzf4s<+CqcI;wpX2+Wcy#CbpFe$+27E$EX(POqPLE?-Hhz3--VR)
z!3hV7Z*V`U57CuYicM?$Bl20Lxws#Dw`j+bA&-d$_fZCPykU=Gz9x+Qa~JXt0eaBf
z1XuteYL#7H+O{vk0xBmvCX<Wxf-5v+7(2g%4(<;`GJW$z#$KVx7n}WJ&8k|g-mV-i
z8et(zQPke4M`>tJT@W}0LGfeNid^qL<KeLl;lhK}w%-l+;~UiXTiBZ#E|WORe+=#O
z*0rk#3@fw16NRo`_O&JsoN-6dZ{3=jby9)$;9-8n=4F4luqaTuaV9S?-C)1;25Jwl
z+C6OvIm=W##LyY#Jy7;5JO(PF^m_6L*#nk8zCL5hz>3Hb%<!$}&1>SW0lGK0B>p6U
zwgDTLB_GL(Z)Ylb8g%4F-6pRM(JR$Pb3VB~<=Pla1Imw*NJv#N5s)1^oTeQsxC=l`
z-X$~K{QL2E@h^Yd9gqgga?YxrgTg|3C{6Z;LrmXq%;YuR;3|*NJ@Ls#&h-jc4BK`+
zW3!L7VKE)bSUGs0nm%jO@jfW>Mb$sb8B>A-Rd>HIs8bRl^V22_!&7S-9~GEneNWcj
zw+`sEWIPBCa#%GALAI2}bck`w!bbpz3{c8ZW?djER<S8UHL{j&D<iffe=cxriJ~&j
zH*d`ho)1|Tskd55EHcvJP=&}5qK*Dxd80DsG3sl?C=HB&i76xArHNj?6PHieW{M<R
z9cc)sYA!9yAYX1^(V9tqTRD<)9ga-CTZ#VctGHOpRfa3sFHXoPZt^{@_e=$V>M#r_
z?~dm%3YvQbQ4_v!aN)$+=9sc$6eDkg2rXNaY`U6O!cF#1Ed{d`AB$fuQ}hjR*u%<J
zMR@R?@{V1^606m-V=Q6lJJ>iQ4+bY-SFcqAL*asBJB!REH3-yrjJy&vrvHVh!D4BA
zWkXCtrypp1fC_I%9Cuu=VKM-?z&kY8>VVaU?wy)6$3}nk_QCzM1l_CXnw6n$l{G({
zKeUON?HjEN_B`pBKf{6RIM~tzU725n0Q<v^MVVG4#G!h-&cCO9Uv+0L?S2eT3O%bz
z@|mpLA`?QdMhk5EH#FaF&Pwp3XwC)>SF=pXL2AaB4;BF8V{#{tzZn52p>{eeNt{i{
zBeC(rn+i@1M&zy1e>zHCeT|kd?E#)H94m7mmhjMz>JyOcrYxBEy6=hI0#*joe$m-%
z2zE<z85pX9bW6BjU<??`jilKq%}p}RlX3cvueR0d;9~*)@AuAbO!`y&U+pBE9*>Q!
zA>Tp}o@5#8#)cpz_)7q>_cP)*;^BR&ov&)v3<<|R<RUN1JkT3{KHd&H`(a4GLMtU5
z*vVxtapa$jH1wjLlrzZReX|i0HKRsDkaz}!Mu=Wd3QN)AXx<9?h?J$Qc|6@<Ohxe-
zPA;8{5-g3b-dr$Z_1#>1YuO%shEynn<3-IJwTJd=@s~nEFO`5%3R-qj%;if4zx%cw
z!Tf^tpYGQcZMorrz8M=S>Rn4+5vZ8bEjx&@Z@y&1p28T{FCm>2suOvnQ<vqAC4QD?
z)B=HtL;~4zg3+aL-dXkCGZZ%AZ3k<$?g7y~p<Ldt%O{{lf2<NeYYu0B3-z$wfIche
zujw2ANk|&cnivkS>g^_*f|mrTac?>DIJIE^-v~87E9+SJRM3y$<j*uVE5N@x({SZs
z=8{6K_<&K?@0Egc>)M)Jvt%tS<+4f85-$Q|Fdh!B3NU7Pc*4FdHH{xjCCjgma@FiY
zjA1|1e&$iY_U0JJ3=vG7yf9g@TRFhDsIJ`w^nVf1Z=X(j_1=?CdNE^vRupUMwAlPO
zj80uXwHOBus2}!@1W=XrmSzz?`ep?dw1DjLi+<HUB;-~?e$}}@ASa1wpf!Sb#Eqtp
zG;TawbGH5yRCc;ccQNn{x#4$cQIFuMvA4)}0*nD8<5f|BkzABqH2U~Cn$&;58yLh8
zqJ5JSp<T)&S8%NlB8$*OjePnV1?_na5`&(#PR|y_Q6*k0i)pejTOl_N(_uWSLU|z5
zXbd~zpi$PLiFqji%ff5TD_%#(Wl}Z+5p1mN4?+PMTK##DM567*5R7<azz3Eyv5Tdj
z1|pwiKc;Qnx65^+6gTli3Iy!58ZZ*OSU<j2h?4Y~EH}B7vs)`Y{F=%u#$stEHAn2n
zzBGybALuq>{||IalJmaVb&5fwvC)E0f0-Wc?K$6kKdPzjD3|-AVHn7_qZUfM=8Q(u
z(!E>JR-2C5_2+V};SfrU+=iThky6(d2EPtJd)D809Cz0TJP`)q99e-lzfQF}VNijP
zNQH8K*Ey2dqzx@MUZ*UV_6yJjua7ks3APA1)}&SbZp__TZ~HI?diw3jS_Ikp=^#yA
z%F>q#7UNi^K)uqZ3|hH}61~D$#`ZPL2Uq2No!As|u_@1k6Eh08C!4CSOy|$?)Yu=|
zlDOf>XkJHS!^l8DMu#R9{C^>C+|mDqxH-Z8KZ)D^|DU*@?Dw*HT;5EF&dJ^szEv>-
z11A<?xL>d(%yxxuxt35iAucE9t@`9_Ri)*cwlU30;?C5Q!8wM~Bz#4_i@|wrytj1;
z$@cu4%U8ulAmW5pm%q`Q_?)^zpVc#MW26mpQ`IolXi{PsQ=Q_QF$SN8uIw_6ec<zV
ziuNTrc>2u4S?S=3y*meb(RK(xAcC?5jA|VJ?0BUYFLTE}WAgOJAcxT~t))xXVARg^
z$)Rf(CA;ja7Ip%d&N6(xB=$1O=PplsAokVNWV1}$g2Yy7S+F)dK^Cskirdc^Ou0<M
z)e1Z=#0B0@yjx<dH{q)T{%8w@XtpV;Ec4olX&jTU2Svz0M0qp@HtH@g*#?0re8-1r
zU8_o!hDTc8%QY}>z39y4KB=XH-I<Wf8QxZzXGP;NZ{-=+jU2DJQpTs!I+*MjRp$PD
zD|d2F(E8a&AfHc>Q#elvEfV2iJkVDfDIk7CYmKfkDd4-@#sG5QK|RqMiZ48QF+T^b
zZJg44bgq;W%m5kkDQY<2G2Md9ZHGq`a*k0JmShh>i)&3{87moHyQ|&;LFKbsa+9u`
zAiYr$cXgIrE?eBh66_YBo(8^yVYY4^o90h1sBj{1*LE>fM*_NvP)i=FI+4+iwj<8R
zq>-Cod%6&vk*~oiT*n-A(p$M{j|n?zj{NqjK>4<zvvxVHCEy1vy$`FDP&D?66x`YD
zs&gCe;*s|ySvI9+_Q~nw%N&@Wmc%y}+d#2a?*!7I1fCw0t(~}^x5sT2yT2p+&j5a;
z2q(?DQl>Gk`8O?$Sgkrj{+07a5oBN3m&6v3xo(_I{BuO6H886DjOjxud%4f(POHie
z!1;zF;-Y{sIsMFM{Y(*_KkMuV>?;pdZukeu=n=MDZ24UN{4b;9RtUNQFSX@iaGA~+
ztZVR0UWx}pm9}t&@;hS&#ggDN#IH?d^}fE7Fr)#eu%D@RnV7BIE}bM`5jhLR#o~pn
z*`{(PYpIj%zWU!IqMxS77>=kYdvyIy0Wul%P>ok~9G8n=%t31DL_%{Q@l>aRxf%wN
zk5D7}(O9I&jE=&~_#JOm1a#Y-NHTCo{q`-1FlO~yn`IbQR5%EFIumTs{&;<ofM#%Q
zwv}Oo@kf@muqUf8;Au&js4}%DbEbKMuiU#RD#(I#7RkUJThEeJKVKalkiFe$1Ked=
zOgWQV$vXTlY-x>}M!dUIS~h<mafVgV(yYx)#FA)pILrM(tmd*K-pyoHm6K-4iyy)`
z=HKpBFllv@YjNY#8f|tzR*fWs&lz6hGKwrG4drJ!d~5@;D|>SBUOWEPl||Z;Z;ba0
zBdgUvCcb69+?rvOZr-e38l#(d1r!dpIGycl&fHT<d~N?*G+=f>*#osi>6!ue&s{3b
zGdYng87w<;{mQZ#I@LoCZVVb%V%V1C97kL0qgLw{e)XD_Mnp}nKm**m;J4&v+YK>|
z-5%VITtr@qwx-YqDZF<~y98Cz)0^JQRgm6L8$_84`Q4iwQ!_5H@K=D<4ZLp)so!l&
zKYpVLJV+n;Z`FV#-6f>&3=ZgLvIq=Ve#bgO^uAoi-U?>(7N)4kh6p~nX*;pTlRGBz
zLcS)WB0bxtf^tRtL0A)R&T!+CC~lH}p<UF6F1J#*K094l<&$nzJA4MJVORr0d1Q)s
z>zBgsFADM#^#C?z`i_z;(e}JJi^y=h;o6;gi}zcVUh|To+gudf{j6l?FJB~=j~sGx
zSIhX(c<#`rZJifF0TK3eocrrvYS%WhtvVGXrI!W?oFaK#^7Q%b;sB#_-!1(es_@()
z=a+Aq3uJ_nRw${ez`-h)FWKliJw0t$c@?geg2-Dn@LH``^-A*agH)eVjW(niaa<Q{
z$7c$ag6_`x+}N{>NIhyUP@A_KsuLS#JzwAf3TpseomDU|lSpsjp8|q^>IXt{6JXRB
zMN%OtM*61Du!31%I^A#593IemT-IH`p76xI<3-#UH`1z!M5Y|_-#(Jd#-BaTh`6M-
z0VVp%xdauDlEfdtFbl&<af4mebjdVFc`*33Vz4RHkvgX35;eG35tVltAM~th@(lcc
z!o1YO@wvSLFhxnUvzzZogxaH@ox#ZxG9t_F)yT!+WWTe1OK-27XZG+zDEbH8i$BMJ
z99r+U+bEsK=(^81W<bO>vAoeK?Mlqjsh;f@aZAo@wEP+duOC_9KF<6^98gWw6~pA}
zUqA7=ySm48lp6dWy#MkS@74Yb??GB?i*fgFuTwd4*NK9fnOd@cdF=RNuz+<siOS=v
zKz!MM)3lCC>6n*1a<Nc|mr?4XKT_~p)d5}x5v>@eN5p4BK_PF6b@fMveOK)<CSX$_
z3r?Bk2Si(uv9t6@6ZwlQ*)s`QTw$^WITO>nTBQXLby9>$2-FJpN7DG`4mIh_@={!X
z=7yA)cM~Q)4dJ=fJa<wYqgq-f4w@W<n@hw8ao})moBp0@0~;rdgnxu2<mHT+nFUh3
ztV6Vu__7$;HE!hS3{cjTn4#mgHv0GZ(UP-Am8(JCp{Ak@?o&#t(8+2&>XvzsK_bOO
zMKbY*3Iq<)HySllG)B}1#T}{ED5EOWPqY+#9kMSzVXNUnGEL8*Y%hII-WaDnxC!~H
z%I<+qktbAV8z*Xrkn%HwZBxuN8UrBsLXG(&IU`><2|~a(A0iISfYgwtIy>x9kv)6O
zF!+1qX35s%mgZuffG;mrqYNFk(Ko*xb9;dwWC4MCli%`TP?@F#l<5&hZaCFc@lqCg
zDfLfAUDxW!dz%>L3*4VY>li*)8{C{7*CJ)2TKb>+V!a5IxyKC(8MLK%X<Y(pCm#_i
zRL(GWkF9tSdu@uDo?4#KtdGi%49hY0@;FPUVFdJ0=(qTK6XSBc(+RiZocT{d4Ye$=
z0gvz=ygfXZ=ZA3kNXW+x(`5Mm3+Nkmj+gQNEnMOpBj-`s7~@vjzwhSGHUi)0w9iQ6
zoZ0XR6YEWLOFimrJEWM0$*BSLHQ}N@{rr5SV5Ltt_G;bq|CWAi1^zAlSP?q2vkI6i
zmmGkWJ2-j3uknkGf2=IO(v|cHcT=xW^!ZnBeAPIPVL>76^>^w>mr0v`2=Ra2ez4`<
zmq3;0<vH53U6{yYGdKRoIH?WWhqsm|Wb8A(b<Xj*sQ?=Gk&8IHSDXCH>cN3C;iMGt
zS1HKXEg+b={?T(~>U<BugRv>?A(E4gt{LhSz5glsKr5N+S%Pwg+h3|vfHK*@i_cy-
zWSBN)|F$NnP}E$HON0~fUs?mm{Eybi?*3P6h;obnH?09&;k|SGm)GDP^&L}w*f%4^
zn?Ia<kcp)TC<)q|eaUP5ofNdhlDP5n8*0ZpmGMbO$Zg53YfPp`o-NJvjCfBjHKrm4
z0%&-%<#Ob(`%K-KY~}%k<)3_dBK&9$ClKw1N5m?(Q_)@8|F_csk~>^aJ+}<{*8Eex
zhna}9dZ?*#bAkPMZ2Cs@M4Vk-YNXbIezRY<u$A0HPB!s9FeHedyN#jh<9hko-A??t
z*JZnZh#DGFHfIbI^A8DKD!ovivUY^RTn_-A5Jr&mMGZp7QbkJ+t(=fFC?Ys-d9-H)
zyM(NLEy>s5joaA=(4()7%6=!rz#XZe%qL3D5P&P1l(1Lb|F)R;J`v)Jt`C+m5DqJR
z(60IO<u$4RvtRTC75N8U2sW{2>_)%aBEB0EXhQ!3SsvkHpH)~kNOd@Z{ZPScF|h-`
z@8HMwL2ixI7p}h%FzpI~o#m=WA#<BFhE+6Cgn#$nC>QuomaFRcPnIjJ(;<D{&DYhK
zg4J4z_a82-9LQnL_<db?@#nt#Ftz)#A7xzH0z_}o)VQ=Q2Z?nc+dz3J622@>szja`
z#2|tK7BL}G>Jc?MKA-``33Zz&-Cy;6$}xg8;yNIT7%L_lw7^IZF@G9#w%g3Xo*9_4
zHLd&9f*_?_ijn!TZ>Y8S7^%2AuDJsZ5T2>SO$7r-Ncqc-BbP-1MAEM-B!hL&q7$<f
z2RkSBPGU1*o3`PD+zWeU_HI9rQ5=G^)a#|a780N=3zzrY^2ng&r~r{-+rQ|@ms$gX
z2=<f)c8&X{Se7yZRR{YQ8-ahgaR4%yUfJKD@PixShyLZperu|&qx30w`HzxU@V)b!
zs=uabGyRcCi?vcg{rn$j95MZXM&$nvjjpsZSM{~_Q{Q6*GLy35YBMs&;e@#)UK}1v
z?l6h?7k<F=I(OhhjkN>s1>J4e<p8^s?|`W1Kn1*5(;>OK5HPIaYQjY4bGO}WHH{Q0
zyClnRsMyki%B#!ASpdrRse@^XgD}HCC|hoe-XWv-jakgm)RXa{JH+pH(zuj*C?>yr
z4{L;t@zj!C)I&WzQmjuaUhz1`tX%_9^L;`CH&Ah$qq{*`_*6!zRMG&a5is7D2}#~X
zjZ~{8YkCvuim?LhQR)J%*A#j2q6QHRr_fgx=sud8%zRVlhc#<Rkg`9g=Wop|Ma~LQ
z?7bGLOmI~YIycI;hHWPm{9NX&-=}KD+Z}9saHc_`G)?Ux{6<UrwdZm_i@&clz*(oH
z$Ky-Z`?pR9BYC{mw^!;f=0H*)ZZwKRhr5rAfBM~IrNz!Lc{95Xy%wt`c7Udo$8Vr+
z^Yz|J%3lz*F2>ZgqB)TX6YTW%>CW?k%lT0XADK_-7oT_BF1^CKKL3V~kJZW)h3ynd
z2Vs<4Zn$6PAIT#}hdYOWaX7Txs5cZ7lorPm{k&-;h2Nc|?<E&@c0l;q=Ela6G$-}@
zQlJ8-rKdA1Y`cBpx*gK8ezQ9^!A=$K?q2It2N50Y<8?D5RJ_y9RJ^%ibMO9y==B_H
zv>^j5QE-ysw-~68Hj`0L<OfI7usG4J07LD9U!>2}&7$Yv4jb5^m?l?CHQzqn<tsRV
z4T}dY{m404P9Sl%wE{kcRiDj}#Nh@B<q&z5k4#$RqXt_bWpkL>1D*Na0hBZ{QojpT
z)X5EAvgx8WE3m|bwD8pGIvXY=d;=5AwUP6MdZW?c=c3W(`<)7C>RF{g<dMuwzL*a6
zm#PvL!Zt$94;g9o&YW#vH*@z3E#D({i8MF6NBBj=Z*j$My#w(z^bGU@dKAdJ8&i^q
z7biF#Vb`q;Y~0cY;aBUIT$FhiOnl7Dvn49F+aV^r{lCI81MQqHUqR8F42d=ZdP!(L
z2f^p->Cwtc_>y<es-*0}cK#_G9)0=MvW7<uKk8GX2_|Zaphrvo`G~CVV^&Gn%N3OC
zG@4Op6PDKlZ5_anJ-{ZhxjV6v;8>Ocw^pm_8xSOCC}P$`5&9~l!_4y-8^1D<{>!Po
zp+>UPi99A_uJCFc9R<>v(1i`ZNjsNll(CGu&0sRV>k{tI{b7#n{5p>7T2MoutJO%k
z#1HOxyQ$iO{)MNAh&=gzp-nlO>zTJyu~Eb#wZfl{Rnh<*1nw#~)EQl$Be8pLB=f}=
z>h=hyBJRfo-p@XJDC7t#a5VO6+y#g$oA6cA2pc_57S{qN>gDWiVc(UeQGG>Uy)fwH
zxY^d2(u#s-6)~Fyv?2lCqhL$Jr}`TZc4)0sD%CkLy3x*$F~u)p75S2P_&4g>bo!U|
zVg-4tQy9QiNq839Q~A_oPu~g0AE85Cw|G#qv_(7B<<HZ6hW<JjTYP~c@ATXaUtxYT
zA2D2P2XEeF_Tjk%o&74UBCW+i`MPP;(2Z_SEqDG?hgl?5HCDR>4`%AQ*GxVmc@esQ
z0(uS$F*iZrRnkuV!bK^2=zL)<k+^nB0_U`_w8{(Ea9wKZYGe*KC|K2rUGGBZhkrUW
zkkfI$m6)cfEyK>wpyB*nfL(ll)-M0sW89_G#yRxlHWX@p?maBN2uIDhe@<qg*eM)b
zlwOkWVa(b}WXaLk&w%i^C7bYw?8dW7#M{v5F9NQJKQu2UEYR#mD9}-a=)cQSy^Dw<
zjK?~GM|PB^)*;=-fd{Ca-)*fzIUA%aPwzsJ^7G$niLlJpZSOY<VQ@{k4iae^gdhg;
zI}L@0a#<#+II8>6>WRG?L|{PMzY=4*OxRH{UUBY2+8ihZS6z6nFZb9BXb#^d%1v8Z
zpN&M#GivlyBt9_MwSx7!m(u)DQ)bWwMZuc@%y9{F9chh&JhxC%#0I7|)5q*HBad!S
zbs2>Av)8V)n{C6){WtKqNNbmJBItMrF)>U}#l>WBf|#+?M4xrr4UA1DNxKuC*HJ}>
z8H8xtLAkoacP4;#-2e$KgIptLKPUpq3)VCwj3i>MCK~b;&_gJfqV;usJ@auhC89Kd
zElthzh){!&vXWsW|C`y^m3YBC8s2pBu0UygAeZx9?}=YE&v(cpF;%}ps<I-OFL^GD
zjJXlk8I`PxXyi&lh_4GZ6qE8_Uc*PayNM*twc_IWdm6Q>BDMFblDa3d3G&{pQ&JDo
zX`qgs^V`0>`%I`Bl!iWVP_%Wje*er51c+ldo^If)L9~y+b%`V#>Y<K|*S?BxNy4W*
zURII1u9fAi%8?05(tS*N{!#101K*46_EVcA4b{;1yNg%<Oh9U{JLOWR4A^OFLY?7;
zxA0O}UsG>4Bveaw6&u}UJRd$j;lLi4aZ(5pCKnvMV=Sol<b`ohO`8W5n0Sym;Ee30
z&iNQ?1KOHKj7;crGJ)b)^zwSYBaNL`bA9P>O+h9^bi#x^Wm4J7R~jcVq4B#O&qhBz
z3PKJscxy-4F~)%*j04MnzGn9EWPIAv!@J-^@V2x*Rj%u8JUPIYkr(2eDxX0o3E;-$
zp+{cj7Yhjg4ErMKBRD3o|E=2(kT-4$AvRu4_;fcDA4^H}OPe4SdAm^`l7YV9J}pDa
zVUb61&3c#v)v#KarjRnzNj*e1*)&rmA3jCx6BG4_fz2*V3_UBiH%{mJ?}c{_@Yj^m
z`^cFy;bx05qUO`k83JLzu~eWXDkucav^sl#Oz8X(YO1O3@+f2p0MucCIDL=XCqrZ`
zOl&By?P2gpw{?fuK(r|5DWaR)?O&e~EnZ|*C933%_8hUf8>Ma@bDPSnDU$OtV1P
zu~RjbYWv0D7{;|*Szl-k3^t`^-g|skDAmeG&RS$XYi}lz@sk9^yt;WW8&awP`eIEc
zC`hxZXSL;A#gL-EbT_DhY^e!Yl#HyTs9Uktz_&K%%q#uIu^$nmILHB{YO}g}^xL$1
z!&UR~%k%=hkjo2dx3c<RoYX%O%fTnb?qKlJopNO^p(1~r{j}NTH6(uC8nHq+S+#M=
zdEw|Jxj#=(+I%+D&~34=@%Ar=maQh%Zkz(2j_jcytEB$wJp`8v?8i`#rD<emVI}NT
zg#<(1;4{1UrUwLtqen?P`K=tRSyOa`MD~w{m{VsvDVoOIC?tm~MSeDmL}D;bCKjhM
zPHq)14p+j5Vo2DQbbEMp79y>O#*8v_n(|PGaEmZxa_%)wT}CB3N5aOxQ&Kr+W|gxv
z@vtnmk~=RB?eZA}0-R7LH+uuZ4-PpqFNQI8n(Lqs55pX~*bwRy3R=kXpSZ}SRD%xF
zag-2py}dXm8d};~h+LQo+B6H7w8*3imp;CcWYr<RJ%#cqapPN;E_#`3f;Lv!T_K9S
zpYo{!WWt0`$r4$Djzxl%*cO(>sOzuv1GAr8EFSp<Qqf-lDb*_&6%qolA4IYUm>J}<
z2)zwH={PXbecE?!HM9O<YjiMhT_n<dn!l=d{5eFt)1qiy7`Ne%m*XT0Up_Bu5$~r}
zgvCicsm(UGKHunnG#GZL{(;YmT4v_pN0l&nx^YEemXLB51%crZyS39_QQ)AH^W84a
z<6K*{VP4q)wvvUAo3^6aIxuGJ+zfaNc$*`LK<r`AK7K#;1#esA)PTU~46c3^)6lf7
zl5k(<k$U1EQOzV1L|UWsKO~JD)=wmvd3P{n-0AH@kb*AIDSSpa+a5?l^589?_Xpe~
z@AC@6iwY4}rNfF;wk?!%SPq_u5PQinCvg~P2hnbTKB>jAw67RX?}xTX<Pp@&%}{#s
zDXfcBcApR5x<OnY^0rlmj9|A_MzS+xIw*mk6jR_t4PrhtmDBh=iVm|pvK5s-6S@So
zTOpJ<IQ>8|42q=p1#aa=fqC6NZ%iK?{g&aXZH!6dr)G}zwNo|jUx?gaB;~G!PZBg@
zK6x$zM~ND85gpcUCwnRZ6Z>xE)9&kcq2Czhi3X6!nLV|_O59z-)_43|%2pAp(_|d`
zv5BXPlDbud-$SO`x@FEU-X8jj#PR+g!oE4Sm$2)$wr$(CZ5yXLwe9|@Q=Hn?scqY~
zZQE|YzVH3(yZ4WqNoJBsCX-B_dG^|Cuf2BDu#5>=%!($j5S#1V2@6?ns^dX>*j`S(
zphq_6O$twCV0Yx*^9DSRLrK5`bxBH$*qX|r%IowTAaFsq%(Ij0cLU_6V+^r3d0!U?
z{5(m-PU(1|mEmx&>1hv^=TI1Dm_&D91a#BQK{-!9C~W!m31pV@qLu)V@mV8>d%GF#
zBCFM2;%WE0NAcn&^13PI;=V5b`e?9xE`HSzaq)6W0fg)(8dNy1<LSxC6+$I_DYHcr
zmS#LMpiyoX95F7j-cN-CxHr;y@iB03UnaAb!=d93B|^^n+cyy7`n$Y&^Qf@`fuu@*
zN%9vcE(gZY1*|6hklo=@ZUl{;T<;jIiZE$RZVh`K@m2N}+ug~Vv1!%C-CTpKBqt`)
zU1s1!fIppU@uow?b3g15446ZJ|FIa9mgL?tfU!e}?|IVRgXJYydb(Yjuy)>;x0^`4
zC1CWyC;ZMC%1QQQ`fYDoH+-&9=1URzW!!xqTG@M&=s%kXvjEMq+8`yahm+nHmC`zG
z)2n9cFRjd4dW+vzZy$XlXc9$CEPr}QA}9XcM#_@jf5gZfBHj_cA<N&GOlB$=Fi*_`
z$o1peT?<>gUW%}D;&JT{%knas1ulg4L2;;q))jLY7z7W2Q23r+OyE(#Pj8QL+bZ4U
z6j^q1u(Xp^<6+qA;G}RyE{|eg_bYz4dk#-+<U2BI@b;Y;njfQdGV9Z`={VIY9Seu@
zwMG%cuYD?qa)W^-v&-Z5>k@L5geAfO1XQg26T`0BCt^LW@bjWjPHb4N^ct;0>_ou}
z^Z6Ip7MaFgEK3<E8pE-@|0>zB6)unfTPn%K)Hx6FbEO%Se!P`GtlMAMRnX8R(H4Rs
z!xb)*nC(Q<C`0JCE+eRt_pX=H+%nACPsK6sYWR6ui(T}2wO>^qXJ~S%412%=Qh*ia
z$YRErtTl9zvsCMA_l2-q%B`@FY%QzhK+=X0;rNU^NG?bF;Hal;3?$!82$6aS(o_%l
z48=HxVderx*6}z<+J#s(P*<o>HFZetC^c3jVCRKp$9)zkJp!+CPLGXDI{Dh@eA*Jx
z9C!rqDaHYC-3kS$2q$_^bI=%oFW*944_LxA(4{WqWCvO_knzz#C9Wq>P=iNH3I5gX
zubUi*c0~J~!3j^QYP`*w>8T<xDaenfs}gpsLUetFpuWx?7Ty?yW8G4?-6(Q(HcrY0
z&vZsw_YR$+6}13o7w=BV`?rsG5S!vNMm6Ebf~t+xhL1HnDSQt~^VTbX3B4-E-&3g%
zmm42T6wa;dkdtP%$aoT7x^Z^L-Rw*!*T+K`eZ3*~O4+P|{fJQ^<J`YoX1myJDqnuy
zj)jpg#(O%U*J|0d;+k_|V=qrm9?rLiDEfN&IgC}+KZ#XuQ^%{cu%=Y&$Kl@D_~G=}
zp?t{Los<Z>^a}r4#CSI#{`XV_ug1<YbBd{f=$G|v_26*U_YO|nfOVlvfy04q5>4QV
zCw$;2+XzncA-eWiV<NlGM+;<U>R6kp((WCwoRM$9Nts5h$v^MN1#i%PSyvWe4!?n9
zt6c6TTh|cVyRLiueel<fh-b~7cj#3Y5|?=Q>HQeZyN<<ks?Hg}=^9DMlmy%-eeaJK
z7!wq<^cli%1_y<1EYy|)ajm){6;T|dRE4v>tZYvC(!K61lKy!Q0?tdh>j=x?c_57M
zau#F*llxuspoqHRtvL8(hEk#8_}Sj8Q1oxAW$%1Hj01x3vR1owus>JWCope}cZ_uM
z?>%usep>C`Z=b$^1Cb>^!Q#6eeVD)hK&~PC=WXl@ImN45AqKFE@wPFu?-q-7p)P)r
zmWjstaQ~Zc@{0#3&m3a1sSMp+9^*ble7tJpU3=~NJsjTkjIbTm#!<OVWEt;n?8tQz
zz3T7HomOwXFTEFGXJIhpJ761NA7C)kndK%p!oeCJl0>5e>K=T6PAs^zg<C=p#Gq+8
zp#Mm`v<6FAIAqJi^VD)H$qqXXlIrwm)#D6?W0tgZe15$0>o2N31+M@bLk<L=W#iR?
z4`5K;VC0?*C*V+lLd~u5U4yZahrG2d{SLgo*7P4{<!Y2wywPz#*t3GbyvCbnLs?le
zzD6^=&NS%;5F`2pcAB7mB)VTx+By$LjC|HKOn8G%(?T&^oZwV;v>t3DYVI7B=$9rv
zZ`}&Y7|JtVOuWejYAe&Har3Dv3VHdjz65R&&AOwVT3*OLI%GIErAhw|$XlC9Ehu2b
z<LlQb{kw7Bc95<nj_NEfCMVx$z4gw%$TV?{#RInm_`!hc4L$X1`RlcP-3PD{6Fd%t
zj#+`eb2KNC!K0m~kp{SzCr!NlX2^v`BBqy>h`4HG@qe+6zeeutNT>Jz{F?GiR(ipo
z(N3I^aHczMhTIf@Ofj1&HOzJu6Hk+k9nSsLuwt39DdEe7RTC}Kta@wqt4>gWGWt$!
zow>Ldu&^RJG}f|uQLnakHZ@HL5rmaQl4M(-j1G=Z0oQDE&Z$)fS~!o3u)7K>XFJk4
zBQVpuzL{9*9qw0OqxPqehHmkKUfpd4oE4gDp6b?(n%AYpnWSfHKyen8C3tN*ONonZ
zXi4?>g6W*ANL~zyVXOg}8k(XOWezEU`fr~T0Dd|Hj?nfQve#v*$s>M0rqte$;CP2Q
zH2AWf@*4h5t>2UR*y?3%13CZJe^9p(s=LZ`!(P20EhLY@cCB!cUt|Ch&A4SMLkO^1
zTCH}<+f)-Zz~*H;o6jW5J#Spi@6LYJa3IVG2es!ivGi6WEMAal=uvUajR(WK6p%9o
z5ITG0T!o;}Jq3^6bv>VK^XYKI&|UX;lmQ|jjFl%4>OY*jxp+vn0e^o;_!~7JS33J;
zeW1D7>P@yx?B#Ww>Q0NBbEPJ+EiZqLa8^jDNKmM=?^5lO{;NuVFaww1X2oCuYTFne
zTemFvQfC!1f+<x0&ckxhffYRcQp^1rP*!Q}K2Yo+js{dT6I-ITC3@TeTRIzb2-|yV
zeEmKd9jbcn{$V6=YDnO>qC>Rm+vOY?0y`ivAOk~RussMWhucwhbP`h*6nCGyJ~D%e
zRl1-XOzYtcQg^|LcFW6{kP(g=s+N)cT`(E9L28|WYa{8Z!SN}O6)op0fp|s=ka(UU
z9_7t?!s%9fxK}{?W3(TzYrEe1e}SE5oV75!ZO1@i7a1??0}De?YYI3FCd~r_=e8lV
zB+Uk?m@q%%gmR*91klRWbE(eFQ@UR!$4I&0_QgW?4A8l@7}2n~@-<dKRV5IeiJUNX
zTDdS4{D9msKqOyJ<oW>WTSsf-_JfCDB^OCZhWb6k9O1~eQ~fYWySqS}WLiq{Yl-4U
z6Taa>9}aW?!_$`TQ?}8Y_q0Ze?0%~am|q2_1-b;@zZi><tddQs9H$v<&bD1c6Tsv@
ztL)po+n%&d<?<~H+m5!1CGtnJ$!RxR=Eksv=$yMz04E(3$ipCLlM9pwWhvKQxRWN`
z2^gf+;J3U!pXQ9g!t0&n1Bc_BmGotA5r=dbDzU;#=wKHY^<Oq^SfpnkG=7H~cM%%V
zJ!hj2`@u3%exQO<*puV$>!eGhX1*Fd!=b)rSo4ocyuhS`4dNS#j>^x#N6#EG$lofE
z>cmGu0OsmG)ilJRGmQ0Vl`anftapx0ZQ;lihXYD3l|g}sM`{imy(Z23sJoJXe{|}P
ze!YwD!B73L3hw*u`~0-FUiS%3mAQ`|Trk0NlZu$j=hj!dEQ%Oa{QCyWvRG~yQ6IdC
zjEvnIx!>O2zCjQ$wJ%)ybFm<WAnC-J>rJJS0JJT~biMTQ=Pm}z^!bBecbbm*nykzn
zwfqqm7SOCnS<T!Rg~VzoLwtNKVWnQ6G|V0u4V^g;)h<scjo$TpJ}&mryCW&r&I3op
zffkNJIprqe!-E>~m^#Jn&iJkSEQLVHUME9vs^m`lu)aHQ?qA{qPSSs{N!#OSd)=>I
z0bOi)=S4d_-fq*irMasxC$T?5wxVE&kjvD*Tw>t|bxT*C%}r66S)reJ7_g&1hKnS2
zaU?$z&d81-9uUf2PX&V}9w5#L6O6N;WYqCw=-NOq_Gq*^vD60SMO7sHOnm<b^sGPP
z9fni_1&z7LZ{!ymbDg167|BrOBC>cSz?@k;@@9F+RhnSgQysUu#M)<A+%rXG7_(aR
z?Y?!pnB_#BJZBZ&aJuDyHMC~LikzfOKLRA!RoxyBOeFv4qoTJcm)j0mME^%)8gLQ@
z0=nD0ci~0frbkHgXfFt0U_#(x2mR$L7tk)$-wr{Q##y}mm@1N9K77p#p8e(sz%YU|
zFFgFk>qBie)oF2v1DSPE)LaOkRbv7JDK~__;Lt>Z9QX;2{$yMpUWw7U7`?&G?J|+N
zaN>f-4QF!^iOtJUwBliiN$#A46`Ri8TYd1Q6qi`L<5wVdyA+BoB?iAhL@)@)3#X`|
zw%aqj2FzpWcBX6wsTi3b=$Eo2M%JziH!2UkKI~_>bN&pu*<*^i%SI{9m0f^ylFE0y
z&j-uP_uL{6>$<T_X`FvQ^#JxG!~Ugkz_q%%Acq{-VG;OIozmk0iUwei=3`M1lm3fN
zjx6G7?b(VvCic}=0_<C6&Bh#01W#$h>}FdFYEBWlFQKVmz2FV=5FF7`XgQp1cFJ+q
z-|LT0GyPq`_>~!zJsBzK!)#GpjI-ZP)7FAnqC7NdmQ$#~`CPQ0dcVh1vTnnX)F2nj
zR#N&e-h^S_(paA4N6B-_@^AtjSVJ)MAzO7x``i{ZTU2}HmQd5f%N;Rcc7*Fu-UERi
zm!8q?53dF`m`+-1)xqY|=53R@&ApE@XUoQ3>Cw(CK~vk$zn32Xd4Fs2pl!{e>^Xz5
z9|S1W{LQSdavpT$(b761W5p^Co<0Q%N$~E`j-kUM-n(w<pa`&Y%Xmzz$%Ih<WiM%u
z)L^u5b4|NgLQ)BLdq=|)=XGFkC-{cS@}e%bf0p5xuzHfm7Ph@Qyz0ctJNO2ST=bIw
zGbJo#wtav;0Y}UNn*Au{w!ZNF!VQQObD8T!eknjfphHUkNoq0Re&CC_%sxuPDM~+B
zkXeiGnmu`_iw8UmZ<*@G3yAI%5YrNz^q4i*75=g&3=#KTPz%W)sDavIFzk(12eMsO
zzJa^L&=*p#Up#qqoBVYo6-5>&??sYmI@+tKKmz-HiM!D**3JDp=sEqtp>1X7x}$f$
zscFLo;av@(UXF~5Ba|<o;7D*^Hm+|9(U`1v%D(9gz7}x5jSC&4i;+XN5C8;DM5J}z
zA&w|G2BcH;57f$j!7{R~os`2<)tl`={7bs({smYVg&j+>6u|ro-YeyV6PA4>Hlv{u
z_ca9-+N#u`1=Gt{+NxqZ&%3uBtrj!*X&Mv>CDR4Ps9OXGII=*luBVTBgdb<Ch-4oO
zf)`h$KnyToCsjq}NfV7M6s8`JB+a#B&e=9FeHGfW>RD2c!a|Oyv^Ld688!%`ghnrv
zGxCLGC0UXcwRVbHDXJEu#0w&`tHh%A|3p6bv0gAeFuh<Alp3yr*rxXMxwZ4|lKaD$
zs<&IOg8&hcZ}D~akPV4U0C`>kScjem^tO~gi3a*15Sj?`p(ZkN66(LO+$&Q_<VVq~
zp!HxH6d-3{3bi;O@eWP0|6Ceq0!7o_%SyV7!$qz^m!(TcfSE82?O|u2zcA&_spu^i
z3KoaLgGFxKU|_D}fyASQYXvIbXL*mYfrc=AT~fyBTeFrv2G9T{5nMB8XVTrfN_vC>
z5^3pT$HiTpuO<DzLr)QX0=Rsxhty*0`b)taf%1G{ckl2a0&hMe&fi1DQ=aj@XSCQG
z$dRJEg;F9Il+#Fsi0AS^&?K3qL~xfRK#4utfj~%wR;)(gZsWu{5p`7Rc@g@BZt(vG
z3`xcFUiCoa5#IqK0)Uo30a_j}VI$i)GXZvk6U?VAqatRnoq|5Wj$IUSE!jx$g$vy+
z<RL-L?VDzLqjKHcVsH`p58Of`xq(nwL~rd%!;HhgymX3&A>F>$u&8w3LQq_%zpv-F
z5rafuEc&$F1DGtRv*4!@L5LlA!2{i?VW52A>P_*X{GI?ndTv;me}C#USZZbe;tQt<
zGF5O^{bQlsA!kSihG0u?vIQfc-CfA=vi>FizR<!`OV$1xoS)};@SNv4t9Qeh`4B^B
zLVX-~dR+B#41kPib?$?5ii@o&#kh2gZ<v;DDvwVCi^K*%ERwD7Prn5i1a<dz8Vjw=
z_D)PD!$AVJ-MCm41YO@R9eGth6GPNw7VgEcIlYt%LM}pfF70ZIy`?Wgd}vGhoo9cy
zuG+C_OLnG4B;P_{hPUD63TOj5m)%_DaYc<AqzITE=2+83YS4ht7|(D%&%Uxmde9j1
z>A)g^z|onwfdX%o?}37YMSX&`s4|&pK(s+H80Y{4rKZ4Om{8E4@he8Bd#QE)39yv2
zG^+l4mmFbA{34HJ*jfw^(LM1I<g)~#{vs#%z!79MNI}+M9V)>o4gp|PTRNyjXv&W0
zaaek!K*^-lC5sktg#kd~XjC1KMv!R3O5&1W;3;Kc+)6fT%0S3pf^suLi#u2Q^^++j
zyO03htUj=5tesUiLe5?$368WClZne(PIx)gB?j}5sQ<jcSU-0;#NrR^05(-H>_9eb
z13~~yGP7|O^mk|}ev`BWR5Ej2{}vmkOln)EHv4di1XAt}9Gd+-Ah^6g_RiZg=#=v@
z9^+y%xl@-lyMy9KuJ(<L%`ha>Q7oV+0FW*av}xq<?P}I0@CmK!;IE1~=bzHPVg%1G
z<cD1Q6?0%G8@G~od|2P*FH<;s`mp)BV&f2XK5A;upm*?hG0(9I>2DyE;Z)L&HDD;@
zZ;R0yZTVj?^Rz=7y~{X5+IpL)Y9*9N)8ty^dc8C=OFH%nZn{5s8t32;AF{g2!GIlJ
zw!b^z7$+$$fZ+Zshl?9do}=!1)b@U_aoC-DI#E=_HMBftH(~lZTygVjX6FPSI1%&<
zU%->sdE!@Ca~=t`i93+TSE+xq?J*1^#Y$g7^%)!@U4>shVPvLdn{`=@h!vP{@QajW
zmUJ@Dv7y<is=g}zre24SLnhMd1t9FgK`w*Vy!Fl(R<s$ParGW~HfM2VM6HwbG0{=R
zLD2}>@WKv{U$k@`ug#8e{Aym(#b06R1w+A{j|07wT6Utn?`nVtOz>&cG=!u2?l52X
zR>if``;PV#8*YevmM6RFQYz?8se}Zh6B3q})>_#uX?h&^2ICi~HF&wFD*!ER@2XfB
z%Y6`~(=Hrh-z}ecOBv<TOZC+LsMW4g_r|TwZmY1;RreR~^Z?dV_OXOJ-A0MG;=!hq
zefsuM+2iG!GJWVqsiHu#4qJ1p(y>^|_GYS1`|SDi`DJqRR9=jfyKFG&<tL`~wad>i
zY}QznNPU@?f2UTCoEe+s9{`PGpO7qG(CIK-yvMi#v?SUSa<Ey!s7%Pn4M9ZKdGq>0
zBc73tZ_Xt*iER7Se@YwPbhj(X&<<SlT+&PObhGbtv)%09t85((HG^w?m*FvH-O|a%
zy751AijyrBNV^g~Le$seZ1v$Ag(;HXvgoiqTghmxPGBP`+AcR@y8-I|sv*ubbC#?9
zeGf8HRL<q}Ef9-bfkg@gtvGo=Z_sz(`DXE3<TfC+?W7;#Qp{gQiG76DP%A|7zi5ds
z(>~@dn|CpD12@>SP+ALHo^c*xfEi?<l5>Nvb`74V!b^P{k&=wa56%)JT6y|M)G5bV
zFYj&>3K<ApY;EG20{|9}HchUF^`*Y`!bpr4taW0D(e-@1{Edx}e43Z(tpl)kTMG?G
zWuktKX2n~7C!^*e;`)o7)eT|=Z4rQdw3<>5MxptyXokN(o4*z-t8q%B|3Xn2+^Wm!
zlYT+W@_uJ@@aR?%Ms#@HH`5bzpW)2Fb;lOoc&I;nJ}!3RvIC4-KfE^-OKyM|oS!h=
zok!~?t`W7eR=(09yZ?OQc>MJiHHdcY_wNR(j<eoo~>7iRub3Pbg2!@EW6uKhiK
zGWzJ%&CG?}bl#t!(OI53@~QD$psU$d=~Ye%U${$nV;T6F1Ye^PHLLRMN!R#<;ap{%
zY?+~2T$FaN>I~pnWWQjf)Rrr~GK#n|Tdw(sc<@~i<8h_}JZ+`BhOV8X`j(Dm68!|k
zSRAZ%{dT&9>tf0g0o=VRc98piEtrAYhvRyf>cX5|w1npu4=JXY=svx*-!hKcoTA7{
zXrR+{51ru*#yDb}K>5$0S^$M-L-$j6=dLgK_#a?E3k<-J*`|4N-;e6eIQo6+t&3~?
z?(WXnh7vv2-MABR+Zl`kz#RTpb51e3`Q(4;Ixp7`mN1B}^9bA<*!`o@m~lXKpQgH7
zigLW`k$(3NLDYH=>H4r;e&6?RVJ*PcYW!gs)4~r5EwRk}Ck}WE^IZy*EsjI@t`pJ9
zDOvo7;{ZxX$XcVc!0*EA1#tv>uP%YYloY5%w_+S;li3TS^l{!du#|I?H<HRQE>GT%
zOQ}b_O<~Urb)fon&umm?;rz7ZQRyas3Sj`8Yo6}ib5VxcY$<#dBz=Q8FViQaZ9mK#
zck*xTGdPgle;caiC(n(AXMALN(Xb&D+__F-ih#rfCuI(TmB!P<h@5&;$B@U5z`;>j
zpXSx9vQ!z+p;%M7L?bwHcYRAl@4V2sz-uTM)|e}}1qU8uFRL_XFkvs;e-F`RzO00`
z?CcZQY$D(9FzXE{A8J+&8J}k5?q^qlsJM*kvaUDb?ScwwcK{F36W6K7x5eW1oTTv#
zYe3$n!ITAnTF;UA`J6s2ub;SD`L&d1skGz+ESg>Mf_Hz+7`9mXQNEc=1^3dcJV#@N
zh*br+c{X$|l}$y@Tek!^&kmtH!z`aWR?-^I<I&_w+zsg(lK0QouITp)55Gt$KT?de
z^9buov^WDXiPMR+?j4F%nx)Tr=G?O15&)!fr=MGPz986BDvm&bjYV3YyU9Sqn#}cV
z;`%q+9~qcM9Uw#XfAY1aakS1h>t}Go*@C9gIqWXXx~Ul1M80h~b<pgJI-N}mu$Rr~
zKB4K9TjX!#eSAi1d`6H9ei(JBW{|u-Yb(ex1I#2O7*+$1r}g%dCC^Qk6?l>UX@KoT
z^L0)d3Z6uwj00rb0%R2RtT8K(@xJbAmss=Pd0hc55WP)A4E0?Of@gorFRMNNz@g=B
z>tl}~oC)O89v(fN-3{tJ!Ob~ZFI-Um{!G<7gNH5r`U(9Ih{|V*QT9lx)$Zw84i+`_
z=(+R=BWjaMwj|Qyg=zU)xGiepg@DzlKbx4zo{E6zn#!cTjPIq4M}hkY74iR;B65{G
zGF!VG6y*a5+oy8c^IX%#-OIu(7?tpIMT90VHc^`GCg|ga1Yz)g?8Hym5kW$e1}$c6
z5Jbw?g07W1cl7@mkjMq<rGf#vJ6lvFPRjW{GCriGW6=dD`aRje)`H#uE7pyZ4a7CH
z=PLn?<-s^2Zs=le{~vWiF9jIxc_6$K9ZqKBhO{IUmow#t|02?uY~SgOhx|k4&A*Vd
zAGrd#9MyLs;OmV$Jc=PIoj%QORsRTbJ8$BdL7ds(;<Mt_r*b+403q=r{U@h-<9l{f
zO`Lsk>5VBUVmZqK>p3Ie@VqiJf4gIwmd)^Zm%RG&P0V#+4b96u2|p{_GieIvktSXH
z$Z8a`pM*ELe*Aw43SvPW3^-(xXh8hddS+fc{r{^a7}4aZ_F}hfvWZcbLIREB)NF>S
z2l&8Ant=~aA8m~3vB(|Hp^9DDi7<w^yB&2={SaB52Pk^|M|@4=^wxTO2I+6c*DWx-
zF$NG(l3)<CTPo<2i-#z6VT7D`s<rxlfv~y2<tJ$!VA43lqDKk@>c|cB58=O4(rW)k
zwPt~X0IE4&KfOi^C}|O&<99`y|Ewgh1H@=H&18+sB!!jcl{tn$6ZtPO3<UadhEall
zivvXq;8jFD$6nMtgQ_$k3gmiMl+xnb|26(d*KzOf!h#Fka~2nuI(cIP#YN5;|L%`?
z$qbtgf6=p>pxR4&4Jhm1CBPx}w9U*@abS$arulVv7lm`FQ9XrM?xx@6%NUOT7w}8B
z@ZZR0n^Wc~u%1S#$9X>jl0a@8OUviO>E4|x0#R?i$p8-7-*pIxibOI3$XEH521gEh
z_nELMd2>soVS3hP>7NIjdPvN!R*Y04QbiSpu0^|W{*kA}%*1=iZ?&EwV;|61Csw~t
z20K2oO>6;_MP08^K2hvC`885)0VPe&e=t#Pc1JZ`Lm)9Ux@Qr+Le)3S+`Z)y6MA~Y
zyV)$BB{TN#D-20%b}ZX`xBg2;>pjHnSq0HWps5nw`bfKhS4oD%t^VF<wOY;<mhr6O
zSre64h0l-7Nj@`H$+-C4-fRIbb10!IL(iEEQMh9t&Ap%f@#9CL=JyFuf4k&#<J3S;
z8cWJ5;_dslGxiUt9oWh;+K88J4c~MFohqyj|L1px#yR&3vG$n>x(+5E?2|H#N*f%h
zEd>HY3SL;+aiEu`xA@zsWWuKN7~N1cA{!V4H!o{%&^yw4V8h3B)7W6=ul%=eP4B0;
zG}*wjQ6#MS$YjHx-e(BF^$6{AvjzsUaR$+~I6@YyoYDWYsL%Lp1BTQB*0<0w)&>R%
z(=?ezHlL(A45_LUsOfVu4>~*8+p?%SNAu>vEB-z<(-V<)uatd8{MdY<lqZ{6Tln|r
z2<)!jutJGgtxRoZIHrU^k;=P_Gr;CjTEDekdfG00c83>wz#<<oo0nk3f|^e!h8i=f
z2sxc%Ca;Rm2(fec+Zca<YctLJZ$GN_*y3Uet&1b}cBf#Z-I17jX&VB}9GfvZdeo|U
z##S>lmHFTLHZf;G?%x&|r;Z2iv=W%hC&v-z=Mm$IHKq1`Dl_OR41m^_Hz(-%cGiZ>
zXtH91xJllF`2zz0cJcoFLC$KE3B!07#n<wlK|~DicQ1S0`tn$Y&D+R2jXIb(6uZ>Y
z{DoVlgVOKP<+VSt?#xN9<GIN<pews?AhYU~M8jgwOk@*1H!4|ju~CGzg*fNRQ3t-q
zyhE^et@-vm$Wn%Lm#?IdLP*GxAptXIu8F^bxQW7KGFZC-b4MrS=<y5qcJNggv4$`x
zW$e0>w|&C@r*L>aON2Fzvp?&;>FFP$@uj8Z^lfE6;O@jdbKM|;|8yYqvkY!$>J*q<
z+oH_!g<l6D#LxE9(gx<f=BHiPDS80iOKg@%Nxs(UY$ailyi9G%ELNNeWBKM81LJ(b
z#PJ^CfCP_M$|=Q3a-yN`0nvDw-z3`NHfnRXzLw)8-xLPEZGRrt9^W4&qyv@)I+b4x
zN#3Nt#D!A~W~8vkn{(`E81IgC!ADa_1#P3d^vo7GfBGJM);&i)aQR8JY2pj*w?EP-
zWW`HpCqEyr(io9f6v8`C$y=idHVE8x?))|20nk!l>{5olslCtu)m=>_rlQSB)aj6z
zR4fM95Xp3`k6&%_tnjEnGMW*5rk(oFnosD~_?}N}F1uKij867QaXR@UcY)3`=a)TN
z32K^THKEb%B*Z%6HWamK(Jg6s9|vKZWNry7^200gE~LBC@M>saJC`DMG;HMC-3(_L
zfSvbubfI^+N*g}m=jo9)(jRB~7LVi9qPQMRlyxPW{+dM0AE($o%9RifVLi`fp#=`j
zj*>$}+3^m}n~uPpzG#Sru?*R-PX}<j=(Qak&ujeee81{b84$seO^fFDM<$`4nB#9J
zvl-247WV4!OJu`sJ<Eu-JFRMf1G5(#U@~d;?7lVB8>4B%eXQN*AIZ5)@>h*0ah#1#
zk*|VfV@;b?4R;Nc(CTN@d-L*He|gB&qV9gfkajWTIH=1}Yu=WCbVW+K#6N|wP#>ld
z`()6lxlI}-K$mnb5nK_S(H*^MB&t?^b^!LlV~e(CVpQZRi*isi=Azi+QYbtP00{W4
zA@VEOrM*cD8={}S#%E$Ug63#NRa!-t00;lJz-dGQK0F$4rPQav0$!icESD&A6ofm0
zo6*;|^5t+*<#(7XbJrnhNNF)R`;)er5m!`~YA%0CavApujkxOM&orB>N~VZZ^vz~P
zQJJgsn+d0g0N=~BTnsIb{lWVoz-M?-cxKN6aG%LO+YwOR<u&<Zx|70zV+V!A>#4nC
z7nf9Ix_-6s{H*UffiL8IG14EQYfk~1OOSf7p*aQOsP^z*YWn>kGD9YNWkrGUGk;7n
zJAChHejU}xJa$EEr*z*+-w0$F<gkS&BNIy}RDzEubkhAE&@~k^0_)%N1gPT54xex6
zZC*A0+ew=BL_*Py2D(XRKS1+GLSY=7a!ZJO-eRPl$GM6cc7>o01A;cdg85rX3f5f(
zeqldA%>gf}zqSDH)}HTxkd2jzCNW878;}6Jn^x*SavAW;`$$FW=#OR_$qi|40)dNI
zd)D&7^w0U3`~<H}4!`^mpkPk)_;!I`65V9g_GqTBa45vSp9sV!K$OAfA2MFJyLw8$
z3iEVIiH=)Ls+ymmHB#N?{47;<e$?0HJkg#MrRW}g_1`ik!h>KopG@P>>DH>BSF0Pk
z{exGJ2|$@N^U#&Yxm2g=9G^hW6tVr_c8+Ff9L=K1WOK8;sS2=(Y_Ni^43sJ&`cK6Z
zG3%b8;L8K(B-6QV<=wA-KM(#0ACS78sZR}~3rp5M?cUbp5U_$dHT>fU;fIOHWP4MV
zo-bI2ATbqWM=EMySfFB$T(u(~QB}8{x02H<Wk;car((O-8tF#w&P!%%SA)N(z54L}
zK<%^1>ZL@;+y)3;I{d|2u>4Y(r;NqCnjnBjr{#8a9JhyhF0dk*esxOnZ~eq&IOFDz
z&_Hy{lN+2YQ*EsoN|V;n-w*4qAT9wE^7&?C-GR*g+DV;>G(+z|bx<{8^@fx?+5lpR
zb+dekn9rUb8v)355%NZz%-xzu=L!|!Z}HKUB|8oyK25+cOW7?zAo<k*vsfvoH|UWr
z%7B~xjnr?S1oBOz3T78y73T6-EgEk*I4gk$xf+U7T}{coVyp$cBg!!0Np4P!f!rfF
zOWrz|){nrC+6hOO!wr9dy<fjWT}8RW>Ez?G)wB<z(~>uD=a$=pkFsf)6?tBrde@kY
z^RWj=+WP?iTb?#paH@HQ#%!rGkZnMwJ-8cca?9u9yOK+y;=ttdLi}$heuqKkg*9E6
zPvJ}H;@J4&7)U2|bK_Xan*t6s8y?kH29?~gv&G+9?=W^Xk3<=hK8_xFHC_*ivW^`3
z^pIbdq>*P*05<G%`5%*veFl>PfQ3{#UZ}!UibM*3EG5OjZbm3?SeS*)Ld=ZV8i~nt
zI_#sD*}Zjy(4oFt>}!W!mTE7G5e1I~&_k8=2L`p7_wtu}4Q(z-F)6scySGf+Pfeh{
zDskYYg5N(lTObTV>gI_IXmo~lL?NgUA(joCDEH7m*WC7ME(ht>Xbg1|P_|3C>2^`c
z@9<#Z$G25~VUq(g<&1c$DSp2ntan9?usz_nLHH*)d|!5Y+mg8sYBh?XDsgu*rTjaz
zYm8GcZV_7_?hx1X$C~&)p>MOU!>5Yi7)UF$c=ja1DT>Ia{NsM_?dUC1Vqk+9d{DUe
z>)?;pnkw8up@>=FOH&Fu;GU7Wa8=w#cc6Kj0eL>k{(3}KgKyA8Hox){9SqUAjC8an
zye3tpAs0S^+SX~8WVRP#T0a`xlVCH7!NYsXBU?K)tkKTp=RVc`RK$uH(}KG767Q$k
z2~`Y|_wRB{d1bNqA88!X9;oA5qx}GbFF%VGu&^O=gi%DWl>Z`N4}pt-F?YCAfpU1T
z0~%d<wXb{G#{(dy(|UlBz|lwXyRN@alvyfU3yWn*gy}46R0atHtw|Kv26A_*#_FMS
z;{Z16oRU54Aikf2jo0~8s|Iww_Wii5ODLTi`l$*2(64<Pi#m%*;wKD6UQyv&4oC~*
zLPC7I`zq^hcn<>bP(QThH8kSGsR?jDWv`;IpU*09!l`@LBS3>^eh6gz@zDmpMR<pp
z6FGta*Z^1|e^vu8l78p3>8XyOR949N3~}43X*^QcrhHcCB^7WH9*AwB*HE9rmSYx>
z4D2jo=Wt9Z@#h2U275z_6C8;jdaGetLD4q!1@?%vsR$7A+WA8S&2MY2g#49yM=OKD
zu`P9f``mr~ZEXAwR@9fJMVY<91q^$CutgFns7<f9fck>Etw8Z3#8<n6eoF~nLe_yq
z%4H;Q)rEWn2`S^wN_H;-2{twv=5@3*UobVA;*<x6n6JBT{89DCH8>9#A|4Y+uI@;=
zl)QtgMFJ+}ToeoFJA#eU4=yD)gTqsqTu{B^TVeW{1^vWO(oI~A`Xl9$c0BevzxVi2
zyZM`zQeN9w5Sk!IByf4R2s~T;(ROh|QnPa1<W?laYog%z?)mPn5!;@Wnidb|JQz?V
zKSR?@x7IUEWDbjAyxxGPXw-$I6U}Jx<s9y*r2+joz(pchYx8vhn5@SEAF0ZT3kjL~
zKEzx)n-bmi3NIAEkc35G>pM`*3eXBf$Y(OSKq^g85KhCBA$(@1HqLbth$0La!VF&l
z@VJBqNL_gFB1lI-@W5%!91VV-NqXPk7rwBX`%-mB2BVITS+(wZLx^yR-lg0_Sm~_b
z3jp?+=3|#<r#Y3iDdFUW_z9D%tnY?yBfzRzfr_7uVGoB4x78F|9o^_yB;!_GT=(~v
zMC^t7({r?mZ!K1=jd*|E%Ihp7nJkw@z1vKLP#AN@0yUV(-+-N3!-G{rCD^>$;9_=7
zmB;Ke>fGk9hvm=(*_)5@5^pD^RbD96BY+jlM)}UD@vN-geUHgvofo{Pw3nY6KP6A0
z$Q@<lW<vN?hF8Ms`1Tt`U_rDqXTjE{x=^Q(-soiih8=u1Y+OwL#$}_)LOG;CSK_(r
zAY5%Vmdb<bi+xBHtb0SulbfOl^+)pFvK<=%HwyGvE30!)#AftYlb~)@>rL|EK44AS
zTEe00Wkq*+*UjNv`TNgZXeqYV)9NdTcIL7sACb{NOmm*Jw%Zsb=rgUq<QpKUWWnQi
z-5-mb7DRRd23~{LND%_e_E#ZLH@4I-=o{#=J<rq3ht{u*CS%~mRGq31ilkngL92o4
zE*UN0U)5Jv_}9tBrUjQi&v?ZtHvqZYaOl>#2kt0#PPrI?FlyA$s#N8O^~lr>#42Iy
zk^KDI+2&j1SVg|o$_tJXTR6OHxseE`%iP7PD{%v+1V{(3W?#MPMUm$|+nQmanE<R9
zCPdSh6w)EoV6v`-UC>Aoy~T58Qj5@K8E^?8Ol}a>Skg_Y!dCcs<}Por0f4%=5AxZM
z)k%dwiC792=5sNDE93^(?Z7jW&Dx1#*6R_LyFi%Cx_y{*U>g~gEDMB<)F(<o*PuK9
zFz%d?&%+z9)l|rmg1QF^uhJQV-?}C6&wk=Q*dFl=ef;{m4Y;mW;6F=+NQBbr+Er%Y
z40a}F>;J}(gm?eY**n=(1k_)5JA)3v5e}$PSw)3E@^_?P^tp&H2fHhI1jKShB=B+{
zdQSEGE6>Lq`8-kI?N8{s&|hUcj0YO{o9wZ#)Qk`4LgQ)nH0Alv+xz84_gQ{kp9$CN
zeW~!jZIq?x=D3f)UrtXlw!h~*k6(Un-dzb7s#|{IUSXUNz5>x&0r1=T{yCbu8?l6L
z+~94Wg4bP4o<6qp{yRlW(59xfk*{dMe-)WEc>T+}OSPl#JAD%Nx*9j!R|X#mjQ2EP
zxcHh?%6Kp#M*%AEDsmEg)!%9YGdH8WGcxLN;VJq;`zw}DW{tme8M^qG+ioju<I2m=
zVJwq40F-%kFX~|b6wrw!>0#*to!Nx%!T3s=%<pVW$w#_PxTS^v_j~-h2+R1*c_b<?
zG6BVb?o9+2YSF`M(W9w|YT=;W;3=luYdT|B3t@#a&xeLE>3nR_XFix=HL-kdqndE>
zK+S@W4;zDsUW-JU;hW&N(^8-Xf|*W&U)^=Z1%Eu<))sY~8c@04d8TZ0eP;ukAa-$n
z+h-p`Q0SwiT1%T$in}aai8XPy2gYRXR19%BV;C5NUS%rR_lif|h^6V1>QEI9PvV1%
zBv4RAYepy>O*G9g^63vTqmq>|f#7lTaB+ni<~-0^#O02k6IA+FOR~Zo|GZEIIxcEt
zim1+3=PqOx0)Q3fIN$-ne1(`98c!33S!IrD6wNeEG||d$NPsAl_zH5goQ0cs4D~<<
z!cS*P&+p(|E7*AC;g=_%$*VGHhZ^!AjqA&V5Tiq~u4CNe0YGZ^2OdN!gD#j>c@So=
zeY7$jMJH=3gPuR63R04^i5F;e(l$wJ`+ZHGSZW_v3Rpw>q-{2J*Tm`JfJ-07L!K+g
z@WT=EQl1sl-5yM$9zyZfU8XYiU#jo4D;h{Dl9)sJ+|~%wxM1k+#e`_{BbltfpN#hi
z+Wd$eh#$C!iDN3EJ5c}{Umx<uVm8NOz#7Apaz{h5LtR`jZCOz82!qK|X1>GT1+IO;
zm+cAO0N_D#{EBTSRoKG`PA{rhD8$J71z<xP5n+$BZTQfvcXo7h^!lShyzatLBHtFN
z72Sf82wtBC+10WEFzs^0oD)pY2|)RdL4FhoM2&g;Q^FW03k-Z6@LoaHYskZdV5j|C
z=o;H7c~F0Qm*N(3L42QhpXnC=TmTIwK+<*_9bi!#MY-ptuxu;}6vBk>V?Hu+I*A=f
z+;1p#_6f~P7~PJD<Y>bZv(1|Dz-LYs#|jz|`YIKE#+i?(DkpcmvXz&Ps@4LE%p5Ur
zlUf5UH|INsgAAC$T7A*q=~&2D=Oytq?>({G*)WgPD3B^q`h6EnpU+H0{_Yz~L4d5e
z4G_*B8JH|$CpT-Tt}^%f70%1+Ys4Md#w#c(wxx_(v+q5mU=CSuVNPu-*ng@*X+lI5
ze_wBX7lk_kA^YxW%z6G8w3&vmKuzg3?wG?zl|$y4FI09oCh*FQro49luq+b(A?CZw
z$EwSv5LiKhjDQNt=rv+1aPVzxLcx(X4zN*8oydG;CEe;B(cUBL->^8LVdpL8Jid&u
zhRZTdn~}DQHB#O_9#SjVpj3#F8(xh7TS{2VYw-BBL#Q@Bsj;FUXtL%jWm~#l%vUl;
zgr!q$mYx<aqA-R)X&TAe5RL2;r6C&+;W>h3bWjr<25~IgqZkI?XBzR<aWm3l3xG>y
zl;bB?#_3ChO&vCBf34~|@QKf7m&k_9E*`!5^GzJ7z5*l^dG+`+W^cC?dPx}y)|0n>
zqf!_HL&AFM$VqGD_<G;@w<W&znu{kW(X|$7T4uNrh?q!I9Fw-`EK%}2BxIx56oo7!
z5MpPXexSbnP_{J~nnjjfj|LFq0N~&F`{VwmYvq)dER>?|Xm+e}n2H;K)SSOHm7-!Z
zR!yJih^!y774pb4S~6tf`487k_%2-DY@N)OLZO@N!I1ggYVh+fhq<nX^2#=a+cSp*
ze>rnkaB9h}8mr&<R7y>8%p^lGnc?U6eRsFo0{-%OKm9QkY=>0xHt`X9p@0JI+7LZs
z#%q~MzZy#THqUW~CtOSHGSxfcv_?4d@U0J(En5y(%-rd#t4t>}Nm!N~^C_DdOXaL{
zQA!gb{@wVn?N)OLHRv9{^p0n+WU7^_23lV&C?l{S$Tu=o>pHHqez&M-5AhnkOMfTc
zw&9>frEDe-HHIz7`bD>)M*xQzRaFroF<*YyWe@+}TvYYWVB%+p@^iS$+A?QW#H0$g
z2~kE7OOyt0neS<e-nNsES~>op8M|3`oS0^Dd{}fvr{~hwbVj|xd#aCp)7KWCEu(T0
ztK`iL7guSLZXcHxG)$evLD}mpG>#QA9S8_;9?-6%GXy-(Xq)Df0{|379nmd@W*8@q
zYTC9RC(ST5bjVv0?*+J#j;4Rb1eLH2*FOr;djN6JJ*4FzTnhpHnE`oLC}bNU#9sM|
zzV@6|$3LwGNYRc#XW2XjQ}dv}*;7G>bNO=J-h6p^d#yQZ{bZaI1~TbQ4_d5_3eQCr
z-kcGlja=I2xAyN<IzTO5Y6G^($_EM_uC76cZe<e<Z5iWK2tEd{;2Y5{s^#|_-}jeo
zS_$6JmMceJH6h+!dgq8ep(a^*!TLNdxk#l(W#+)!!8**}9~=llEzPrGE<9>JF`)^A
zbAC@4E)A+f9C~M83TgWCDtS9wk8ht^P3CsLoRDgO*-Sa}3fN?)l*p6H9#Wc5VHhE^
z^m?wim74P~WHUH>)M%`nwfH-L$Y(V6BL5v6nDjVT)zzjF^-)3(OJ2R-;;LsxIdtZ)
z^q8-69--imhTyMfn8<F0AFn5=fx$nm<vr}}P07=G3_Ey{jw%E7w{+_Gj>v!r;A`|L
zTdB^yN^33F1c0M29tp}X_$ZW#gDX83(f?@`k4$`>`{M;Sxn2x{ggsbnws(JA^N>E(
z^sQ|uo|uc<SZ4g^@MQy5MU|7*X)e<0e4udjN>-sdC5JmZ?N3%AY2D?KU#mWD&H&yd
zo<YeEI#$gYaeO~HdU&|ldx(mCnXAyGSzHt%i4gW;2MDeS8JCK^v)l;n=Em8*vYleo
z8U`~k!Wv2r@Jn-v_3$7w-LKIK`;+61BCs((rM_l8{os$2tPbXvF>g;R?bp*vAH-4V
z`*rnprPNI&C*c=rA&Y)UXn_frP<)nL0hfTQx|;cri~AqDL#O9Vh6cj`D;MoZ$jiah
z!-*#~fL{B>MlOavI}&Q6#bL?UAy+4FB?EMUV+#Ii-SM=>&HgbwdB-x-KsAGy_&6h4
z=9xrG^#Y^z2I8G44ptBf9sOv%WXhV^Xr`_r54#4IWxY;$W<|M*JwEB5;92RR5=^T?
z13=}+S9@2aI$5g^@v|49KBl+zqH|GRC%VrkfV19ke2Hrvrmp3_2wex>g8g@&)riY5
zdZJC-Z(R{De6^<x?_o!_>`w~kx$_4pR-EZBrWtla(PiW%$J~RsJ8%@-gUL9h7~v+~
ztZ@YH_E_v=KR|e`c(K~Uev8|jO!b>K-VLidxDG0dfko`np`)?oj~jk@(%mmtea|U&
zKqMQnpR3z!TSp5|dz+m>!&e;n7)50BzX;C;s71~eUt+Zw#i!_W$P#(ig(AP5mRG4i
zUuNHCgj1hE&aEWa)|({KT@^}XNR2eYE|MXl*(NV;^-eYG*+KC<t=X^MfCWEo>KkYu
zOFLBVc5w9iVtrY=Jt_Hwmy2xAh4cn-z;)dYFozExpARW2O(<&!^sKFP=lY#hx{05&
z{DHn=NJjXnfp{`*MDCF?60abSu*rsmM<ileH#*s9g;HwTlx@;$M|0j_{59ZJynNQ!
zO%Urq61e$8BA?R#L4?8W+vE8}i?Z8S6nyF2<Iz#Hs3NXFDkOR|w?Z&J$(f-Xpz)i=
zKc*&+z3m$CEe_lym#lQm)z0zPQv5;AT+cUab<#2R(<HRTi<vv2i9Q}Kp@#XiFO8Z+
z!M3RQwk4U~DaxxLzMj^}&&}1<%@M}N^HVj$Me2R;xPIzz(DQxID`?C~T;$QWPz@U~
z$GU>9naZXsj^sHibUf)9wciB`z|-r%9sdkE8WG1F8CeObJ^si))Pn^vv=zhx7um@r
zpD`<oh1K7)f`-q`RD_RgP~F!kMgj54G*aBvOJv}6@>jS^{Z0_#wQkadkl+B;vdg<>
zSwKr!m5DBKa|0H%wqIlwT|=yF?!`@g*zg9282RTT?go2(&LVNVc~Y_g+{%+WHDvAK
zjkuCQ(UR`f8Gdl|p@rQx(|Wjdx-q4~{Vur<Er<a*G(Gx#z!ETH!~L)G$O(s8D|51t
z&e*irG>s8CU~jnCjQ<=Oz4B$9CeUhV!L^^2ziH-BrwMtXD9+s^fi5Lhu<PH&I&>YK
z7D5x61NOB9>WJ%hf1K=ZXd`qCY9$Rt4Tu;7?TK&_SZRn}g3h`1Bj91d6ecW%NCmmk
z^%2c?_Q)VO?_jsZnqeejDPgd$H=1pcEw;uJ1d*>&F5*l_<*~NrG8J7<xmIkOAJ^yH
zLgtsT!7T|jpUThIK+TeJW9bB|PgK9xptS>(8sw9COQIw=xBwmK$4*8-)4E1&K=j$h
zgYt7`S)j~xXQG6A8^5OyHY6aM!z<e5WkWoy$7~zLi$e~A(qn3XPj_vA5p8G(25dct
zX0ALfNgC_CS$^$yZ<<xC@!R~z81C?XCjJx1nZ!G}VsHKx_$0j1mf1{_LrY94@#%o2
z7_!(<F|Ow66(CM~_UGkRx^ly+s+97w3>GXkhY>1C6yykULd@H^eX@TM7eNOdgdV$2
z^FDu=N<`op6bFe0YM(~{ah;YYz}Rcf6axQM$?y!RAG-~M)8b}nE8_jJQ$t=AxK?DG
z2$|2y?*U8vA4w8LsTpq;-?G4w=bvn+bb8AV3wz7;2EYM)bASrYuqGwN6F)1+u&f?$
z*n`Zi|3BBe<ax{{pcm;?lKeA4;*gR_3A6|kttrhcY(gz^2i9~)i=jO16gLDw5;iBV
z+9gNR9psegsIjlvG(;YYj99}>>@PwdqJ|qbEYY|Dw4&VEW;<Un)DQ_m9A9db<yE$~
zmj7fp0GY%|2HsIB<k=HZFW(dYnUi@#IP1Eb8Wqw)MjG^8xkK}z20j~&sg5~N&Y)h1
zRVVR+-naEd=DmxL21U0Y=&%fZ;CeQPc=t~u_{_AEvePnAzWS8gvyAdFc#`$nK+dai
zJ)ii#g?LJ?qMf8xyPGVV)7T;50xjHa{%KVn&;mrGdawb(>hYi4>BO8l7r%o?Av24x
zA!rssg4bgZL|ySNk2o?J6{n9j8E?GINDraQbF$&W=|)@Sw5&`nr8~FSA<Y8^f@P_9
z13@b%03}<ab$Y(0(ZE_`xsk`^Qj{}GdD)o!EM;LucA+j4(#+27aOJO<sTOlHGb%Ko
z{G{keP<hrEUEd20Rb$ptoWV|&xa{g!^VmF5Pm3}1;cW)ks52A7M5~<MU4^!J-L>H`
z{>Ncd&?{2JY9<XWf3TfA1tO@KvRMN<maYkQ8IDNA8WP_uW5VRfea&2wAm9B?0^MB~
zN{hc5Sw-MO{V1jZ6&(5AJR)$qC^ixR{1^;LOsbP^F)0r_NRJUx3(ieAiC-?4F6KSi
zgE78)HAk)VzIe9PTZGhAn~tIHfLwjuBQ=q#@q(IH!;bJQa}oC*AZYqF#oTK)+-4<4
zF_-8=BH;YzEn$<<i1E=oG=mI_2}K&b_&Gq$;2UlrLI^}(xE|cU))X?R!=eZvG(fm;
z5HwRS8x@u^p~Pa5i^qe>@^I)=wRx)NHnIv1O@9!stw+`>F_b9sF@cxkx>k8UJ|#Hv
zeDtVjOcOn=GkN~REI0P3X!`*2NKS9*reHs{j=6mBQs*nqz)&syo>E_aUOr^s)YE<d
z+zE;^`had%YofjISV&4Y#Cs0duu{COrgm83tU6ZemR^23)K?Xrw<Q%~f<z%j$=OWA
zlqs6?PlT5c*PxH84@zvlBC2cDuC)n^6vt11Ao}e%+q(p`yVKIqZLo+7)feF4@8{$;
zc9`$fi8p!hhV+l(X9EQ``z1$l%Cmq4JMNdhq6Q8n8mOH#IXmp}&wlc!lTJm-XF<jh
z4duEonvl!x7r2RFy1^3P(hHXU73VwWyTcV>rg!Zd8!J&*PAU@*ceu?54@ONl<It0U
zU)9ni)hO304m;_H0)06e{h1eWZu?Szy&O2%@RMMn(CI(ZDssxs#nN@&GCUtkOUSI>
zW}A%5W0Q-mAgVL7K4d_E7{vihM#CDPs>PL7WaHs&?!t*RZnwzU*l^;@%)6bZUzx|z
zm)j6imm}u96z9=zLMo1ZwIA!*`<q_P6z$(W2Gr{AI}S%6o^1=BzmjQpP6bP$Id)ng
zmJ729arhaR44R(?&nvBOXC(rTAh&E&-Zom$|A(t{3=ZUrqI_)IwylY6Pm+mkXJT}0
zb7E^^+nksa+jcVX=D)jDi|S8Z{i^!gefOR7&hJv`YndbauUJcCU;*6O)2GUS59l#&
z>Al<Cd+;8y2TSWD(IK_mWlKn<4n;@EuT3o4u8pr*jZjTDOJyw(uu$vNSS`2#yN*Uf
z<804m1gZNX9O!GLqFQ08LZH;K%DKAhKCc@7r!TeLrp^JxGlu))M>j<Lmk|iZ0_1wb
z=hwziJrwKy<xMFoc7Q66*wP{t?Y+m&Ounh{3wd-f4GQ|FbwL<fsp`g>&i70s@&1wo
z5{Prci-wjU5IMQO;#c2$98JN&hjLp;t(&xqHIl^VO|+Oz8O{#8Y#j8sW?x6p=bIDX
zRaB$fa9u1AgQRQqA5OrUO+Q9Pk6wJ2oCB?OAJvE_6#iZ&8~~|@meiDNn_l>?_@~Va
zG0?7=m+g~#RCHYGwq?u8b@|3(L#en!e+$-KlpD+x8bZJY_*OUIn=popmLg6c%&0dB
z(~ts`H*1KmPFs-B0!RXc6_UgbleG6J%04wnf_1P@pn}s^^a56Yfr*~K9kap-_tr@f
zE2gu*T{?a{RtAKuX)r-0E-fLuTy4!aL_gc|k;lIiOf#TmrgrX_DrFpTJ^nQp-Bf2R
zAY4RF5G*f2UdRtdC}zy>j(y@1NH*g%c{Xg<9gWs*d+QIRz6gK(Yc^1^P`+}^Re(I{
z8Mj}!fm8CZWlTyL_}aDox1v;4198|YIW9BK>$&H?^&Lp-uSwZbfgw`_0d>$zju0Nj
zK`~Z0wZyi!tm@+%zCXlbEkkN*dT1WF%i5Wdc-NPBk2Tf<gM9b~CITih+V8DJ?m4bN
z$vhaA$a=e3`vFUus1QzLSO#|`X_DpiukZeNpi0ipoYDdhMJk*;q7avsoL&hJ@tK;K
zzwP{*IUl&WO(W|gP=ASt;HB{e^Zm3*?4V7Y;TAq6btm=Rl+VQ#@(%491mfLBb)<8-
z=PfcaCApBzqc-C8N0p>=mu8NbFAJ~B+Z%=kq^V5Fi>7Ukli>NSqg1>mkes#XZZ>H{
zVAYt*d07md91C}1-}<Jr1X^RU_XpITW~HJnih<ln8ozmkTsX<<_%8<2)lp&FsyT;l
z^SWceXR7Y#;mfm>sXj*af4?4i->H+%iI*M~|KD%r(z@VyUadmmJrcWjK;Nd<C9<Qk
z_#NaVwy?iK7%AO$p7cxg+l+hEz94@mc9?P=m7s3nY*bl7naV8R2qE+t3LXrA2#XjL
zORq<rlfcsk6?~pTU-kH>ncUR9WxbNP6pJ)!(xcUP+IO=_%XqCjtu_ytyO#~5bZI>7
zdk0_5qu8ay<%Z<t!;kn}?ad&DXsT5LcpK}kFwDTxocT%kyGm&m+Za^tqUp60wc$L@
zdX%}NYzB-+bv}xe6&TJrI=e<db%WJkbr3<JKT+3`MDf5D&ym%#$f(`w-*f^p2FEQg
z%e2do?h-7R#V<JtSNLX0MNrYr)g99(!<w(T97r(Mx3xCiZ5g2_((YBqf<^08p(az8
zd_lW_%@Clg`Q`+U@3hSukTgB18zO12S_r@+hQ}MzDv3qEiou%r8Z`iLjf*N;a=)T+
z*9w#-G8LwM5g1t3lEAct<#<WCO)c8w;k^d`{WbQH*3+Em_HW}8Eh3*|kN7E+hf*IR
z^q+Q1#kz@o8PoeOIxWahs@4MSyn1h(=Nm(_N?5q-pchr$|5SE|Ug8bVxF+_|$%s&M
zWd>cmd-;iyX@c5+0p&{J9uO^$B~j~nqT8#xe(`}G?DnzgQ0n5}3dVbA4p9zp#oHv_
z^T^LN4q5NMT{Mu#&3^OTJI_H)m+bi{fLj7D$`qWh49<Yy{Ik$$KBAwo0tpF;Mu$)|
zoq57OAXmn!Y$88tbI1W@pBQoqcG7?+DXJG>E#l6RCR-fW0-&(4CqdD4`YFILeJ%@T
zR0Lcu#krrYf{G{*F)ea!)!9A?3_`8ihyNYDmy!Q``9RQ7yj`Z{V@j)N(t5Rc#Lp`D
z;a1*Q(P2<DXkfg#P+sooQWHoWC2{342_E--{4{2PQg5UUU-jyeE>EANdjHVKUh{+O
zzCvhXL;ss(mx<T)fm~a6>7nN%=9-ju2Q}k!1#I;k^@BrSTXuoAJ6p;vWZL$`=4kiD
zlS9W8dF5CZ;I=FG`_;h5r(1ZtIM?$Cv+du`ba_*;IAYja%FlAArKhZf-tH3^@jIfM
zWn%J6)-?Bea4NvMgIdZVr=^8X^rVTArr$Eh!~nG$mMpuaQA1WWn;9Y)`(bqrv)o~Z
z#BkFK<K#{$JEPtG<S!0BdAj6HT=;!$24Z=X38&lJ1we__lHS9=GWZs)i^gLb$MZDN
zaGb;}2)R(7bNDTPr0m(NAVW$dsQ%wTsr21!%h|!92qR#B->_xL@$!`6LA)P7cBFXk
zXjjwS@>*OOisKy?d>DAB;ohl_Ox;!+f$Y`7N5q1MP~Ji0+UeP&fhhH_DL-Qm<@>{x
z{?X!Z81V^i)paK1Jxx!#C>BX%<RiPa!%0U5=%To+#H6J9f7l9JV^lvyx)l^ch6j?6
zjC$Yv{TToT*jY+xY8~3XfZ9(v)r{285(5e%bq&Q_mTYM67xrGm7nj(zVgKhgFb`Bg
zcxe{H!F{pa`~Ccpmzj5AFE$`Ep)TBD@M`WKq=G<Ezz8wylZ=Du50F}bb(+-kW?azL
zh2EZ+)mid|{O~noSL-ASAHm{STbF0CQoK3<(*qDL^M}0ScY0FRReF^x(o~`Eg6Tgv
zs`}5W8G7ONa&~R=QLpZSew3rCTkNzF?2NC_0XJVy81E6<Lq6vFU(Stkly4Y^-bj5`
z8m0VmnQGObWJUv{JX&4N2z<|9m+Dj-Xv`Qm6|}6&S`(i}U=WRwOdrz0)d{SV{Rfaz
zhR<aBw1bSYN$>S4i<UO^g`y{7WU5mKX;U?*Lg&(=7pZIm(`h(BnbSc$p%eqZSoQN|
z?s`oe*<SagkxA0iu|r2b9D|q1o*Z)2!1#^IR%R3PYs%TuNSP+=FJ)L`&{O=h1>BM3
zPm1H8ZxZBx@FEw*)wGoOD0wPHc?19i5C1vXVl}GbdH*NujJZi9*dev{p`;Oc3`Lr(
zdYUW-!F^B&S|<B$U$JQNVfJvR9irCSQpA1&qi&y7sc(e{bAsRBNn=4Rf<4d%i`PJL
zrd=7W*NyZc5mOoP7#5tNRg=@?TsbsQc5#DwTCUQuL57twwyIadof|dTENcLWb5@$?
zHv{?YPE`pTNRb?g_`6;(f0J}{F#Zk@=6nXJGHM*(2<4?IhO}4kSEv-Q_jZ?{P~Khs
zD4wS*e<**adqLFwbVy!!Nur;a!tvyz4c1xXR9LHPbe2iv{YfZ_3g`_lFxTg+w8#q$
za@8YtDEpQbBr^@Q5Wn2znDl`47{Ao-uJaNRl(+D!v{+vsva^E@lm#i*jFod@F6q3)
z_HA~0wqSzWlCD7esfMyMzB-af%x_-@9Wa#Z2#)C~HIPt5`?zpJX{Z68tu1IG1k-P_
zOE<oo2Uy5CdF*M{UP@)Q(9!=y-h%(#;F2%gW)F{u87F04kx?X)8UbGS0UV<wFzdrp
zQv(BHe~_W57;&c@!&ZEwUW2ls{YKoA&!68$!tvd65l8NRN#qNy9@Pe%SWqSldguC<
zV#CBG`<666<Nj6m=|1`h*M2j&g@PasWS&KT!Qkq87@S0SjH>%3TFShA0=>`*e()Fm
z^$@}R1BaEE<>2c4@)B4)dtF?Oec}zY4jv@}Ob;!dYp0#x2acHFiGF#McS~*btq-%6
zk+x_NJQ6tAnmseIdmyjUgZD<}2~aDmW2b=G3&IM&RPKAc6P`k@*ZvYpO}$3w;eZL%
zfMW?|(xYSfd!DtK)$#=pQa&f4NXDd!E|Lve{_p6E+7ygk55RPj2q^s)KZIw4r;C`L
zQf~JR1-B9QS&k6CkYjlYzC>T=%44eF!V@!xe5QQVk(#lCJ}&XL<qBubBjN<R>FD;4
zZT&$iE9yev1jAT=ttgI0CPPYlYW$24+dq8osc&3da9Rul#`ILq@R6Oh4>udKYgC&e
z5#H#%{00^f3J3?)eC=X`NO%87!bK(F*+0YyoC_-Yz+jMZ$B<ql0-<8!tQu`g2BA-A
z$dT{GEM<m49e-Cw2t8JzPWj|{hp)YYyQfvlmD?|SH<(oH5Gh;2mR&*GuKOJk=}~e4
z5&UE)g-{~B+~t8~3&~Cdp^Mf%6R(25OatI3H^C=(p*z)i;Rme-uC$s(Jb`7BToc&X
zA93$zFk(%s)%qmxZfj%oz!Z51FzaM<O1qw0Nc;aINoq}0HG!;#AR`DhwuV#vXdEp5
z3IVDq48H|qqGV+4gun<wY-tUIovMwph(O!yp34?Z7_;zGHeht0Etl|LiLd%rHUKHs
zObJv0Xnd`*7a34+Je407C90FMo|#DSi%<I}ds6Uvh#(q_pkff%AsSrAqsStHlW=1Q
z{*#mC)c+2wXeI)9r(mk$mZ8%|+xQr9OQc&bprbP!%^yPxI8cL~NH#@Tifi-GTEZo;
z)^1tUn?%~zOE%S7Q)=NT4#F0a>wz|g11||Lo4I6`wrvvJKr~4Snl{MZlL~T8Rv|p}
zF1mzVN()W0Wz9d}INc4MWf{6Lj8{H1NMk!qfyhi5JDec}1W0e)d+o8);D2ycY*KE(
zAxisiVRpL59JE@e*@oxhT`-P<!)<78s!0f8Pff$qMjt>N%d4ZdaRm<wB7lFQJ|X|$
zUeguWAo7LB<vg{a60l`Wt>{5I!HncznI3-G@smcPNRUP*>)XI!S51e&(%C9_ng?e3
zg!p{82>nwzd9Hs^r!6u}vJqF|Q|7SWV`Gj@21z8;UtrZegI^CJLN<nBEn<N{3`k6p
zsAr<@gkcT(mEJ#56rmn?4-0sg%ivOUa*_pc33^$FlZ6z>0422S$Ff;jNR#<ZK^`g|
zu&)FPM#bn{f{B0FzGL1MzkPl%voegddT$E`Lie#l#Z8dN2T|M7aMU>k$0>%%YY7YX
zLm3qsf;a7{<vk-IdJV_#3~!N0P;~I4IL>`f`u%L^%G#c#twxjNt_*CsBPf!_R?X?}
zc8-Lf@jk5wH}K=|6bg64KwP3`z$jpbb4=p2#Av79<vCZ**g~VF@3~?K<Wou!-N|hJ
z1nZ%>%kHjh#OpV}!YS@A(xo$?jo*QxQwBSQ+(BDfH*nd{b&P%7mKq=#n89=tOqTvu
z$VAX4f9UXu!z=pj$PNe_ZIal1Zk`D5uMs6$xZ^2u7b}7Nx_;3u#11c!So1qK%(Pxe
zxyL}cXUj|g=f^ob%saB@$y1lV>9kLkFb*eF5qCYJmJeFD3N}tME&@*DtGKhZAAStM
zAF*eoBB0z3M#t<nWrrunbFHB@lNrO*`iJ?};R&>aiTXNoPd2d5D3h#)Efcx2t2yY~
z98Zhtc_8t8J$Z0bWdsX4fat`8=P#sa=cx0>(vwy%<zEbsPTDYMEDpwtP5>hrM%HQA
zE5b!POI2<(laPhMx&{@LIk=HT94Ozi!^$!{ST;C{xLtZ9Y86)}bvbUj-!DCkEaDWl
z%Y@;cG%P#+q6Kh$X<Spvy!2dYVQ~}jIPdsvj?&9%5l`VLurA-J!a2k<M70+#%S0@f
z+$$5}4X#-O)zvG>74h8tvy8*bKmyMkMbxp@T^Q7sJ!wI(n7HL3YxyqDGN&Nn@CJkG
zw?~ZVNrg?XmOQ6X>jJeRYK>t{Jh|vcE^^RQ<Ay^CYk?Mo0;&8XXqp+@!GLglG)WlD
z;K(l$n&}6>`ZmEXgcDl~dP-Z$7it_8vaHaj^{|WvTSAha)#zZhi=k3M<b@bC84W}P
zf@*NuhoMB=-bWGvPk4Loh6H~%lAEpVO8dT?0{dVh)G5_}QKia_!%Fy_tkRqb*$6X7
z)*q8^1vv5lUizNPZIS)R`n+lT-sQQfU(zz&o*y8nPM4Qo4-W$aA+C;rR&6DA5fJ`s
zo`{+oyXvfAJnGBiYvZS`$~=3#XlXb*bGTTWvb(30MH^HI&-`l3IiV(6b-w1p1RC22
zgUW~wLKqw4q35vIPpaPo;uYkWd6xgnK?P_vT{pu2(XgU2JuA=ubOx(}DWshQQ~%4`
z8DV(x<nUdRm-z9MM|}hFHbv5`M%wlUZ+G23zYNQQFmc_g^!c4y^^e_MKLf;;c^~)Y
z1n%4d3_rwXrvH-<{+y<$b2}t^_)u=<mi=lQ4)2ew%69D1Gl|7l%Qw@wHcxT`+Eic%
z*zLQ%F5im4!z-T@Nx+WB)}RblP<%yFpAC~1ry))LTR1(ktMJDBXsNxQFD*t;jTh$9
z1+xVqH%PgAJK1WQL+vM$Qb7`rB8~5yWpK@TiVmwO1sc}6&cn2adDp<ix{KlTmE#)T
zb3|&!?PWZ?67;|zg^OF0;w~+<zy`jia*v^)vB!Jpn<NrrgH7@BaIe4~v#h3HjGWl9
z4X%WNBJZD+cLi^Ej>tcUI~qg|)Jp1aR=SPj<a#OeYU?rqVAa^n;xXU;CDcMU^L%Go
z92#l+)y%_ANLHi1BaR^%wsy2CCAlR30-DHz$21ipuvL=7kCY}wz2Cla^8n1;n)A=d
zjEDb$oZa<v#T;`qU9O`<b^8YzyUwIFAz#s4tMxzw0*Xpg^ot5;4kQ*rPkRBSKPoo}
z#7s1##}h?T8z1|!I^h!R`T7w2N=iVf@_$r)S3Jd22t=OavJ%fg!%CLQzdrkta{%T<
z-R}6vTkHi0VtzQx^(Jb)37}aSpM)bZpY`&cV>pcA#lRkvP8)%C;0w173bWny*7;lh
zY&@4}vZr(C+>d264$S^JMp7xy#)ze1e?@3hfNxWY(GyGR*L)h!9;M_Da+w4KLNzF8
zvT{C)7%PbqFkXI&gCnb=Y<2p#r}3AIy!x&6Z?S4-4O>MF;cdUIG5}Sh6F-$alitv3
z@~ppxe@pe^W9RREr~G8z!~g8fSdTKzgz+w!2;N=(xU}h*Z#$1~fUh5HUE&#BgkMN+
z4RLgAiLQ?M!*?mK!ey1>#7EPvosF)va?|eP>?ScaU0prJq!<uT16U^^SsA?f{B7&x
zaWK$znDN(2kk=Q3p#$C{-30_qUm`A`V>9C#FC88o9(UG-c&L{$Xe1beJ$B6?mnVAb
zWmPoJf9%ZIYaqF;xuB?0FR?jPAoNkdgeV!h4E#-Hseo+C1P!SoSzy8u?64rJj8;Y*
z9(?KpH)(ElIFJtuz-QFU0?1KNVEjR%w~Zhi1Zs#Lg8tx1umJvP^&&<f%x53qP}eN}
zVW9ks;2x}mJ98q+(OX?|6ee%>QTNQlR~ob!EPiQVB~jVL0+OIGz5fWmgf`GUcbp!b
z>54e-C-WEzoL;$=JQn{nS7<PB51ifskepP|&eq5GlRO&9>rKie*1#4CI2>}13Jf@$
zJ~`0&p{Y2cLI4E9_6LNL0#S8?K8R>51o0ORTv5M2;zIaLa7HAME7sygt;~rZ4WU&i
z6uAIO1-&5j%cP87>GOshvu~<jvbKVbNiWZ~n*H&gLwRL2%X`K`Z7ivfbvPi`7z)`f
zi>6K2OuaTcX?PCaX4%A!HlRN$KPpdeO0`0T)&|=Kz_wXIu2bMfq*E%Z#HU}kB$gMs
zu(;7H{Jn?;Sw&>K4LO?eCEKjx;XNJrv)1dxK)iyoa;;$dFGa#vY=yiwpBW`Bmm4MK
z4=$$TN8dwM^-w5e!I%8e<>O4(PBhg!B!ghTkKG+%j~Pv(WUC|fii=}N;Re^xnHJ{P
zyv|$>5be}wdgj;4qmHej7Py;@l_b=SIb%}FQt4TOVC}P<yE2Q%N%$UPX2FZFfB$YC
z7do*Q%40I0ND3wG*|=9;nGCZ=PdtUC@D=}|+N7Drfuq8ZM+^NlPn{kW=Lfx&6;R59
zl)#8A(+n?zb#$(MV#m;nn&<pVRFj>9qm(=VV}ZCWV81y{^T*yiIEv-*iu{_|-D^p6
z7QR)qd^i^=4I+0YphojcX6ak+emTCc>*lVI<?G?<<~l==^hqS`HdICn5dW{QG(4C!
zCRZ|7Bcb}-_0RUgt)sKYn?H`imF<EZ%O?eDyZ?~rMoj+Q5-pM~1jt@%e{*y36QSF`
z6}OW~Yt2yxGN#~%5t1HTJvS(*vkyHx;^oZMHuzhq(_md04aYD1AJVa_3nQLJ^bt+;
zN=O5Ct(=8a{s1uE424Y<Jr1}gc?pS_<i(Mavdz*_l5lvm&ij@{U)fo^qJ<f0Pp4kX
zqM|wJ^8AEE#yRsjK-$w_2T_~Rd7s$0$<X|h{2yed)G3_5E5<KI6?sS13NEG7(VKSX
zH+Ik2UnRB?dSi2N=k84u<6xBd#P)xxv{*#Et7a}Nzx<`V9xUFTwstBhyEzK%UGbwz
ziK)z`>;{AIUTc@JYRWY3r7ef*KBs_Enl(RN0knrZuc=ctV3+B^&_fQshIU1Oevy5t
zir(qs`^yR3cxDc`<=C^Q3Q<pE9(>~rS3CJ?Q=twkQ;r@UjNiBFkRO;d#z#_dDz+8V
zk=5gwvV2FJ7TO|2EySX^D<$_@bTTgn4O;UHbQE8=!K@+J!ua)r)%z$DM&C!@6JN8#
z63?goJ9~yYf!r94(5s_32_7>?5wd4?!ge7)j*gmCMOdp5A=I{L!l4GL`xj*UPw0fb
zhdanyp`t6+rLhhU6j6fu)3(n$+LAz!KhHbH^|dot#tkOg17tIUTyI9_G&nQ<Fgsvp
zh9O>$MwsWpox2PVV8<@a;@%<?Kg|cVJh<APX9;>Vfa>1Cs;LMaU+E^ZM`d}km(oY(
z$vwBT^Umj(My4J;x6$c~zfJ#KTsO{j2NYGLK2c<UdpdQJD~4+7a!jvj?%$suJwp26
z<7&yM@f0ViB0I@TF;5GtMRL9W9Qt{$sx|!cei$_N#8O_$!3V=1L|*C)%6}0u|4%N=
zliVW>6_8QR`#bC{WO8$@>HTHX$yy4d^rWvhhd5dADmO))#O?QnMnqd6#zzTsw^w<j
z=X;d{L$v3+g*=@vCdc8!K<9_}=odRRXWbuGPFhP*VH=u~p)O=XN>)T(%bNC+)u<e?
ze9k1yq@N8FW$S0Z^Bx=)dg={`0&kd2X6*v}HsP<@L59b_t=P+VumQDA!oB<mhem-l
z;G4vTWKh^?2qWMGxE{9fkfevkjq0Gw4{8q0nW-JWx}<d;!P3%1M!-pc15GUA+q^&<
zx%(i<Lk^7rWKw9K`#g=@sa6ZgZDA%8Cx;p*x5CJAd1cp5Sy$5_rJMvW*yp%P83Ah{
z+Kn1jf@k&)O^Vzh02>Xbj-cl5H>~O9lH@t*Z$fl@dz6rTr^+wot+n-wiN{=>xl7cG
zZp#b-_}Tk)SQ${>%whl~_v!4Fetzcdf@FTJ&N;n2yCoA{{%*bx+<0JlM3>HltRfu;
z?|PhB$EpljAHp<7J3l7~3ME)WP%4ua5kaK?C6@;&fE!lT9g=eEg{|S(Z|cwb?;NtC
z``ot*Qtd`R+|v-vI17*OxTX$tR?468V0D=)8!twe*IsaJ#VY`ND;OFRFoJ}0y?b$U
zmK98CBHG-dwYW=%oUJU7K}Zb>|Cc~sD9b|m;~9-xt=SZ&R7*z{`PC`tc)Xu);^8|@
z<xR1sEj_;ua80dcoYrK2LZ11_YAkCd9s+Y<TpZupDbCJs-=4UeM@1Y==a3}sr*kmq
zdDjlOJi_}h?g;4@yKeH^Cyk@4hMH*=i83*FzX$D4XKDhPE!%SyEy;?u8SIWH-}!~r
zstfP!2ufRym!OgP76kI{Zcq!ZwkB?IXICKm+K-1UQynFTnQ*w%2_1v9J(G|ZTO`m3
zH5*QwZGV#BpXj2`rOkX!Z?>>d$5Nbqq~HNwPY~<*W<<?(0xiB^LFZI|W4dsr&n4Xr
zq)6dr>`1LBc1A8&KK<zA(3D_JxRFyvBR06Oz$)l^d@RhI(@%QDAwO~$Qw_{Nldm)!
zd{i>sV=@m5u$RgzCfy1(x~LXAme-=-o-3RGP1?=1k~2PBYE(7h_f#x>^1qtj3NZjg
zRCcVn5smT@wG(1W7(uJm;c7_;OEBs8Nj1N;Z{O7o$)1gYo~fVW5-VcRM7}SR84$OI
zV7@Qu0+P9Y(Fpe^?!q0D|LzT&XavI6u3h}PPPCNzC~wE9`?-t1JzmY08jOfG@6O*X
zfm=7AiPZRQ6+gaqzIZk1QKnU$V!HxUxu3f4dX!a}Leh61;~}eLr(0KTCR^R4EmfH>
zzEy}Pt932_bPuc*+Q0YL<9(%wh2-$*_uE^w-{7-LQU=7n?@e{d$-jgRfIY+T^)yqV
zuImSPB@h0^`m^s75WWTE&>r1pTYBT)mFndE*?pnfqb;;hjBa4$c5qXlAEW`<baCnT
zi)UEp(f}@Emk$v1X~_(3XPgQCHPTUVBK@Pp=>f-=m6dp0eJ<)`eAF@9npQ(QG*e>R
z+!{~SH=+@k5A7(aun<c_3P13IAz{JPI}_|mEV?jPTpA_>Ef#rAt%9O`@NZpiF20||
zuwfT(3knf~Q1=I=C!K~$tX2RuZkH}qj)GZIKQq_fAyinEy;mjk;rEf3ByXyyqyxdi
zghX}n;$X?K3V2C!$^=PS*qG#>30N^09IAt3v(h5uW~<V)g$cK(fwR&wMG2~3i}Z>V
ziW6=_Hxn%j!c(q>#rxQT30b7Te(zk5?lHl{38-_+l^{MUV=97ksR9x;T8d;nSIhdZ
zE`{Hz1NK0vZ+2sE7#*tl%yyXAA8Y%pMR}m`D<#1>cHj{3D=#G}M}wgG^Eewuj?<$P
z0w%TGggY3yP5P)R*0e@7@apB^hG82``(AE*)3$^*Qq77(D=spNiTg?_tWa{6e{$;W
z4@fdpXeX#y3&%nu1p!#cM-BrV@zChUN7Lu5FV7~XdAU8{t9DNDW3Ar|t?G8Ht<_R>
z^kBGjC8qr{wYJO#zMr@7EjnNl`p}-FsbYrP3${NWAJ)Va7Y#zRKjTA-?RQhWo`huf
zc=wz1EPkb|39eS01DKuP{_<{ar+P=F>mAl!GF~*mhs_f21_9%{X5*b>xTC?J()pO=
zZ#&OJ(lUEEvN^?!ohu)eP@qxPXQv-gtG03UsJ~H~lyzdXd7e^Cn)nFMTH749HluK$
zeB@9h@LPU#b~^lBjmb1$4^@Ee9}F@&lQ>>WJ=F>(LYxV{Yv`0XRs|*K()?-sI1(WP
zf)aTk1_1_U^a=<IU%8bd<1d|9Z3fHIjbc9(nB%6ZMY4k)$f<>(>okJ8mK{d;x$Xrp
zO8kIowoSSM=P%B;wVQ+^X+q?1o4h<LR+HD8XBLls7oBAQt$LHBuGQzxB>xn9L28Jp
zR1je5RY3V-0E>yYuu&W+4w={+`E0r^HQEnhB2Oa^A_IgT;HM&Iq`$BS7wh5Y1AXRh
z=@)LKk3xJrHoUY<UTn@t!8uS$cnq!dZI;%%RbqRyvL|@Ic}h1)aVd*`+%&wzS?g7C
zcB|LqGyheHdaM4s`Zl81v1-kop}D`CtIPX6^8ws%wx{sNd%<Qx#qgCi`X*w`?^edF
zIjM-Sr)Hob-g#(`6+WGyj#37TjS`mH(~(pH3oS|Mn?587NRm=LC<*Hy>OToeh~C0#
zi3ySTacqie0&J5FBvMU0{f2xtDQswKv1WH60sV5&+0(3b;cSFvUJTyHhvY7u7^%28
z8pM3M#tEQfV9AgbPH`}ytuH;s&}Dz<5j7eQmHB}@58)dWYNDUh()pIXUgF7+4ok^$
z%lg+5fz6<+P&CLhT+7I7y@Xmyk>l?d7-P`U6nQ+tPS;eIPQSRF<uM>>z8Ie8Z8Wur
z^v!_?kbU|vU<~AuvYik0ED@96ClZWIty;bTrWV}Gy^`3ly~C>Sz4`SF9;BW;=t>F~
zNbbPwFdVkJT&eY2y-Sptf^LgP<HK5u0WY@1=wTI;^V$ea7ibVCv=lA%5jC*U<<Xl{
z&|w(;6q5r2GntcxW>Gjw9P?;#%XoLj?KkbQ+&jAPNx*6ZRh3^Tu**c*npC<Z0A>me
z()Tx@Tx^(&VfE&^L!im#)fZ+gWoqhiuoaMaxXwY8=-_PBvEiO{OWFFc>*w8KyM7+W
z0-)QYD;8HRoLybLedn2C)}jhMqprp8ut0C^{qsCC`AqXXJKUIlusFB?eR+&0Cnc9i
zTG(JMWx|_Hca|VwP}gRKE3K~x3KiXE&NwX+%lSr8AkVP=c}DnkEn2YX0m&`gI>2h~
zSSyBn!*GOP=FXjLn-+66Md~!R|Fu{8J%Vyk|Gy%C2S3DwU#Z}pwefjGHxYa##x6DU
zYi7RpW^H-DZBEy7kQh<ggl=mmZG9W_-7zw~z!?%ZlgHwbNM<GGP7>>8y_|}JMCWGN
zO>dH|{(QH2s1p=Q#5+b*GNLs;Ujdpm8>dF^MXl*z$9ywU`#=b-aIRZqe4$GW#~S;Y
zSt{w)770TH&Z>_Taa+CF+s4hfG=aS1a0}EI%M-nXVT9tq#Wm7J?w+sgo9CFBN;?Qk
z^SNQdq&1lMo6HY!IF`m-9MsQldDr%2(fg?dd`8wEr@8yhs~;0B1)F8x*8pomuHyF}
z63oi_bBeziPPtDBtu`AYUgC!JUN6rxd8=!sX?-m!B>s7s9OcPa?@;Eq8m)ODJ!CrC
zaqGO-56g$?hAQ*16GElkGMc-k545*5PBRfm0^!JGD%1@RaCGvam38sP*;eyEC-$RF
zF?hG%*oCzU#d|-DVS+cnM1W`wB==I~RvH4#{%m=0QLZ-r-iZpUk|eKuRoXu-3*mJ&
zhvS1KV<P@+u!UF6632-9Yq9iFLj$Ke!Gl<nq3PR?iowB^Sq<TYOJ6y6l#-36#^r16
z!b10ShePhW`E|gN@orUpX2GAt%jB)~fa37RmBFxwS@PYjM>Eqr8SuzDhn1b$u_WO{
z2njYLa)JjGwrtIrn@ccv(LrG?Rf2}wvnC$&g|7Y;csOe(A@SWzXwxRcFFtretIZ_%
zP=drE4;~=r)!CM~xj9Y`#d<qRq&q3MG|tA~P!j5%Yz~n3RO^(yZvQ|gPb2*K!(stC
z@fJRHNARBEsL%7b9vF#>lPLZP!fauWl9uw<xeE(K=|drg6@CP=gUeNS3(2cVt3zWa
zPE)2Y$FZf?^H^DPiKun)XlWnZ@DUHKZ9O%%=EASMrwi;pS<S2G*{}absh!(Hn)r~~
zg7+z;())`qafJ7`2&|6v<4_;|1etbW_(=WkUh0%}7x8Br4$%GwMm^DmUVmXBtf_$l
zz99V<)=vLw%Ns0@Q*e8cC*YeMr0UbNZ|WR$rsA+4c{T5S?aFK_OpJTNV*bL-{!ebk
z-PRe!Wg1Fv5=Ccqbf{f(rz0nNo<ck+4qFj17`lHT3_5LE$TIhWmd)Iai;#4hlN9_X
ztl?hD{jSPR8Q|7QYjbe4)KvDH@Uaxl(C=I5C=?c|iDrqNm(@L6lWk}Zk#c4V+w%=i
zNSSU4e_mGBEIF;+pg$C?HUSt+w@CYz26O>A5uw;`mzKY$Q`Se)(&~!iVF$Ma6o*WT
zBF2w|pb^`uU?i*!#hJl_3yB9a1nuSYGY7G1!mz6=&%lur1HLhxJ=as(N{a#Xlx_Mx
zJ!G;~zcgG%Dp(abi16t7<ltjl-dDIuFLG=N69q+A6oLX8G}1g~qE=7DsHZY4cWb_J
znBiU@3-9%~nyo0OP$(+@Vko3UutQ|Dn(hLXR1F5zA7)SzM}p@YKHC6%EgN|9MYTKl
z<4@ZsYG69QWUV2KG=K02vmkQ@I<l8*&@RvciEwdTM@Gs1C;Xhm(OnMsB)x=N-C!5>
z_{k~N!!LNu&%*qZQe;xh6U3jT@Qvng0bOa^zUp~2M4+G^ez*Z7{=JN}Be!`4q&&$H
z**&Ouf*irw_nb<RTO)1@g%11?QQla`QdxOSA^-^kB-j#Vk_ejE2K_D$?BFm2{tN_}
zI`P5PeogzcW`P;LIkS@BEs%uTMN5bUuOkt)PV)nu4~wiz4cm`$sA-@;m43<Q;XOcN
z7qzxh??#=B=3Fyta=(-$ahf6LO&gOS@e-kt;!!MX@36Tk)4XHVaue1_HZPm~^Tzw(
z1<>#Eu8MF9d@>nIY<)bEte9%qe)*7N?_LVI|Dv#0Hvb`xIkTA^{wE6gdyDhb+Gei&
z<uRjY{x?_6LYsL{nbMPN?o_0F*9byd%!NxUDYo065Q9L$Wyl*vqKx5iq;E<1&N4V0
zvCVJQnh6+T!~yXjB>(jNa0vI^8)%`Kl>qtQJYu_HDR)JsnS8<N<a~4v?$pice%U`b
zoR(+2OqmcU(Rlfc%0>E-{OT=YR)`4R8|3#S2w2qr;lDvXgiGoTE9ieD3fvCK1OIxR
z2akrfLQZP`jy6c;czkkuh@pf6-_sD^2aOu&J_d=LS@EFRc-9gJ$ZlYSEne-uH2?)?
zofni2;^|koR@6I5!`U?KRK%YZSMP_f)BA_cf}^)>UL#%hDc-A#J>`?^ztRq#c7oNC
zk&fc*&sk-~BN^b(@T?Rclql1?e)`U8^b3!@90z>3{8U{aT*-@B7kW!ls2)v(@^cjr
z{h5MsB2%X-K(w|t&ywy-9ibNXI02x=5ss$oAGF&p(3EfZv~`7*XD-Ds$CM}krASFR
zI@J7{%UK#)UN!!AGL4u1BDP1BbPScE&pu=6Giz&R-(dl&o_q9Dm4|Ykop_V|1LW9C
zou*<@4tBp!3k<4RbwayvvReKaxQW>%E7q>z{%TOas&brl#fqiAf*=p!Jpcw>Gb-c*
zZ6}eJD}BiDk3PWnF;()#y2-=2Y^u)C(TTjHC#ChFR7spW0;J&h`;t+x8v+&8sC&m?
z2o1*<C|!rCaa#AI`w7DCRM1y+4a-hP7(@f|Kd~An7}tSYeM_v%MyEXja~CF#m4Emb
zjP4F;p*ad{IwNJTVWq7!&^OqGQ7g?4#^UteoV^k^_@L!<j_bPTykxLojjp;Rf(7cb
z^a_b3Z%|{yK5xWY%;p<|^d74T>d>M8|F!a)&j5(&{vznp-1Y1ko~oPcFX~d-{O_@E
zw=xK$GSq!kDhaxgmZ@xX7iQ|{j7-rgb_7@vfE37Z*=934PN6KjEA=cj$csK?Mz3<_
zK5yEx^;{AWnWC=aqQk<s+0+8JdFyib7Vb$ttF7=H3?peY{pm2uCp4TIooiZ=5HN)w
z_S*Z^8eJL^c^TmQ?X&6?RTPcXBIKT$l5C-rNt79G;dBf<VkEX;FkZ}Zex&*G0qkUy
zKqa@>Px+ZT9<h?-c%~v9e)on;H=Yxba1ik~gmkp2_&!`Vv^ic#c2nhmOAkz$x&~Aj
zSP)S%i&5M6V>VIpso=;A(mjJgMVfH*fFyx=%43%<DweVQ-eZ-?hOD{?n(05$7u6mW
z`e{QKc6SN9x;3}hIrg!Y339qIcoR=OKzn-yZ*&i1HKl(L%V|0YQvWqKm<ULpQtZQP
zL5C`d0{fe$BJI0M9pbo}`Kx>}5q+84R#EUU$Xv-OF1%G#D6GPG6~e|$2=ba)whD1x
zxHE<)IEt{I=2fPVST$}ro(9Rp&FK1&0j&DV{VR#HoQ1_l6ay9mNiRfiLCDoApfXaq
z69m3>_)Tp;X^WFpDRiNk_vSg|W-_@Ax%%k3V#UIb@T2T3?s3!?QP!0TRcMx+QAyVG
z1G?wv&-icRZJ-^3`1mB0^%M8<@F3z_Z4O4z1FqCL^(Xhv$^LFsQSqY4yiJp>Wu#0y
z7Aqq&4j$Pp>&NS}qc9A4mf~9&P--)Hs<0tauQxu9F2>EP@vD@U##(v7R%ro-$`_>B
z@Az9|1^Ya`8RA(4`wjmVmFp4KFjCIl(NP5_19fw-kV5=uQots$v;vD(ufB?HSx`oD
zn&I2?Q;hzuBMkNG0;{bPo4*0vw_Mk4t*&uq%f9@8nxU(t=}=-CKgNI|z;p(wrD^p3
z`DjTfHR#b=>==`*?kAR8obnG^v1#6y2+PiA>0U2$aVdVjW(*dZE=EITFPmP&KhXXD
zpq%HR{8JNQ;)5hGlrZ!K9K^!c-fbp-U~k7I{|!BP4P?Cy#AlFOn~tF3)sC@&wWag#
zC~+D)LVA%!5dc=vck4Y(fGsJACoMG}?G63Mm!L8F9`-qzunFmnh|0)Vz>O}kPE0s+
zv>DS?2A;b1u^8oC<(Qxq$<)qCTj9z_Y;!EbyCj|+M#iLSlZ3o4`d+xIIz4D=JJ;0u
z7JU6L>EO*je`WI26wEEm>gPTi8{!rLu5y|(l1Ib1{KZgB4c;mgfT8R}b*&$MJ*)R2
zy!&rpG=v#%MoO~Xt7e4YW0CA!QQ<R6V8d`FOY{<l9DKTAqai4$drD||XM%(O`u-9=
z?VRZw49~u>yB^a6@>s|mYWq*iU=Xm!N-cPJ$7ZvbNE1C|f%kORBYMOlP4YU*21B>b
zJWGk#)hyuYcDwf+V7JdO*2PL_#w~hy&AD`1Q+<nK5&H$na^{9<*V{-88opgbnGqKr
zGqivrHRfY%*&(-HGCSf6;K=;$Fou#P8rzn)hsHwd#aq_3Z(*2q|7GXcb?!L<DsMVW
zml|X`%H}DoXkVjLHIA6hDMERp(BXviYL4o!&|%`aw|0&T^qE%L;X`^gwD0PKt~cq#
z9AcJN*fsstI4hdPc#l?-d--MTWN#DW&HV~~gg4o|YOPW8XL0f3L7@X;?Uu3&zrplc
z<vvw!liTKESm6%IcAj+nP`c@d5-|UYlMknJorjCUF<Wnynw4A8Q}NL|B-t<qdhMMe
z7sKBu-lO~ncqdFF!T!WR3+_=AXSd}LbT{)Usl(+euu9v!@~Kr$Ryilaz(HFH3GYi~
zp~3vGs>7n>JwI!SQP6xugbd0EkF=w&HV7JowV(GbenMiN6L~*kc*J}^hv`NPba1O)
zR-6ezj6nCVZWaDyCuk!Q3&ka1D|rH0YJg9q6T$}?WCecJp6)6bqQziJocyhb56n!~
zGPH9CxSZDaz?}p}*k0d|BTUe1)iDM~zxjh$Dq;?i5Ptk2UeP6T3gkgO+-|rWj#3kc
zFEB%8WIZGZgzQ5j)#jMhhn)?y{1J(jnK7Zvr~V3RF+H@h&ogmO#D+5_3ubzKjzrcc
z1x5vIqQLxDkm32jUF_dVE$1;W{~nkHd}3eP;-NJBIe?sPN<egydO|+^mE~*?R!;y!
z#oK54t^LFdw};e5jc1dHXO;}{h<eXh)p@94<xPhP`TYsOx&x)|PrQLls6n8l1$!y_
zPHoQCdi3W<t}dA@7HVg`tSCq0Y&V+nwFEFm6p7ZRQcHD8{XtT?|KCn{O=7{|&^yaN
zxBuDm>T@TRq<$qW_x{@DwWM(-!{ZYS=AsG49upQfgR}NP$8s7Wi~}Cv{V6&bNh5~y
zO7STX_UNEy7v*zO7kHyjPfUbu<C0gdmrA!Ni(zT*fSm`+G!#G;v(jJ)?1RP#0Z_A;
zwcF~-2<DFhzhvC{Pww0Z*%8CFZ|;o|Fge!TCGLhok5bEPEjBU{O_fAEXHVc}2z8U8
zH$Z<6<eknj(SOsvq0+1WYEQ8HFNQZq7yKuK44n@dYN?;VLF!`C)np&|XUmLeuJSIe
zo@;d#KkBCk0wC|`<v)^Y@>B{hfa;Z~@r&x<@)b^NX52+Pj-*h&UjhUEDLfh*SO)gv
zWeW;=!OT?HEq?}V^ouQRWqi2hQona{y;G@@8T|v<lZFZQjU?xtz!zaRsLU0!oS=D3
zu!!q#yjq}v^thS9J&|<xbe-=w&3NpvhP}ZK6Z%Z+=}rDN2)zrAvW@)v9tcQy)qW!)
zp=38)z*qI=`UvA$u|TIz5Zk;#F}^L5wqlS7U3d>5gAgGTPK0pB{h$47xAWf_J!XQf
zi>O%s{AuG4AHy=d`;eZOC<O&`en;d8q>Y+&$>$tu<5SWemAx&tg)Xw53~JLdcRM-m
zF5?`CG`TGoKU&J3rJ6lJ8U%*bJP8DaN*E7x0~PwuNZ*0X%|G#Y>zw_DE0{#+v^hEF
z^@1kbglVQztCPl`H<!K*SDo0^&_PF5XE_71y9jH+Lf=v!Or*1k-HBZJ{Yp3AlX)?i
z&~Ywu*^uqzzR_i8<zU4s2Kiz|qh<%cqrA1m1KDIVuX7z;3C}aYl~8T|MHl})p*3<n
zFWPG<e-CeoMMCxE>bfF{z&PpdfVj_c_N#g;;%{N?TG&2G(u{;1%8i20nvm%d)_Idy
zTl}X+<u)fLT^h_G!AbzL?=JIimC28;Ty6Nz7?q`pcD@)6hTT#pSHhRUh*5*ptl@$!
ziZmnr%o;&*eZ6SFKrhY!H1FS5V^o;M!U9z4{5IP8zgl%Z0me{=+Um=`g95LgmHHS=
zM(7vpOjNvV;g1Lv0pwF}ZWoZcJM(*U3BY#x))x$FQtjQ_Xh)au8oy8H#=vP81AfTl
z<M)6;iW2<}bw{e(8460-`u)RqS(8z0`Pd-*#=|{KX;VF*V;_ki=px@jR1#XVn$gmA
zI3@5T=Vp??BgfaxaoZZ4tEk#|Wn%xz=|-@K=yUOe7~&fb2a;?^|4m=~Db?q^2I6wi
zcfHJ-V4*^#xC7wR;`4MtbFnYsz~6gBbESQl(6}1>lNoR|k0Xxic{y=lCw6c}{EN_Y
zT?J$L9mWc9J=;<ktJT{P_@JiY_gBgqn#|U9COEU|wOm4c@_E!5tYKH0W&Qc8%8l?3
z{yKxl!gkr{t(@m~6bU$O-s9>+pyo&ON5@Z9MvT^``#LxEGGBfM^@X^_`GK(!S&=FN
zEF`7KCgTgCYpW@Ea%@A1nD(kL4kjWd+e76gg7GH6E2rSc0C>gHJSUbt_11(jX>aPb
z_f|C;GR6l+oX2(zSk9trBi2?f_)W%|aSoS%nn3sR_DkOazOe;WCTLHaE=}UpJa)yn
zjF?moCD&ft5yB2NAbiVU`Gm$F+aoaU@oISMwC2=BZ;lAtxt%$gEA1oTyMK1=fank~
zUIGbBQ+|XSB~njqnSG=fTczf^i@`;9nQ{)?xc~8f6k%~$0`-xoHPkY`PW&|J<Q@{4
z?|D5xi;^bQ9!W&^<>c6CzN)~iNGXh}UwGx3zbR5~1zG&~K#Zy4h+XIYlLo!H53P!x
z6k-3-JoujC>SQ6#j^o-}dZ|-ntrKcA$-oXIs=;9osouPUB3Om|iCpgCVfY@qU}p#@
z&}bnHOo;iq1r&>>qkqqVl!VR=^l|k#wFwDzb?EnXbV0a07=tc;@Q_jVg8jlLr56R=
zE0MmZ7=kI*1*0rW!9eWsD?^>LgNTwk2BJuFLQElc6)vl1)kDMr*~_nZJU?ca5+(sm
z9)n^(&lrv^svv0mrih(zXZ~8^>T<4b+q6NQN@fW(c(kww;su0pp*Jx%@d<X1xi+6t
z<9_m4g<V9Pue@_xy2ZnwgHo2yU-Uns&*Ap%M4X&Zq%6`uPr9&iP)<<iZ+jH>fT97I
zebPg(qP%nI+;dC44L~LH;#)F_12yoo6|>t+S%8s1qr&))s<KeYquq~Lo&FsJuxs0o
zn|W!92&*E*GBM7$Pr@Y^o~tlA>rq{7IBTk*SegW7G6}h&mG49;v2pYTNb+Q?l)KEY
zB)_L9QokQsjuPkvW#rB-r($U|hv%5JRWT|Wx9M;tsc2AQUa0CQp<qiUP?Q4C6p=Wi
zWUwk&`3$n942e_cVo6G1WD=P^WM&_3nKfF1e7sn@PWn?*-Q@eWjxs*5V?UA`bOOA8
z5lYrkM6EoHa!fn-PatafljK{DG({(jz@XekbhqXopPo8`4e!Psqs1^MoX(j3=sc(u
zl6u=<`iy=IdVEIoiu?7KCu9M57+Sp7q$!%^X8ecH<9c=g@r!7c?Nl&6b+y}pvqjEq
zg2^FsY_F1SA26VHdoR%K4rhSQ^c`_P+B1kG=c>Pd{PA-26qBk=-Lkm4^;`24E`7@t
zjX%cid{$zozPtR6*>a>B<SdT>%4{iTw0iv2ylSJ>Y7>=p{I}bd25~ijfAQG#GUEla
zCEgbrsNmKIMUBbH6Pu}5TuR>m!*~a(2ujljQ+E1UM<lb!x;U@;=-5y8r_TPzXfW&p
z@gj4}QpNIdLO4w}(V|=e=y7A)LE&}aIPsLB5O|W|JcWKnucEQ*=C`GW+B^(Z#*)?f
z#jxf+81ItnyUBWxLZ?K4KJaN#BGtNBR;8eF5>AhI|A!pUKHYpLsb<yb#nFU!72m>6
z*#LSNu(0E>xT6HFS7h;8)}sOxekgW+NV>+H;w?UG{<g8k@iR)9Cq~3Iz_7k(v9aSN
z`nLhs?x)pOmXa9s`K~&1QZPB^7IUj=h4!%w?R+q_qfBX8Ma2v(Dn15liBOA2tjJw!
zI2ru1X?M!+6sN~}4AQ+-wQN#jJ;_s>!A7%D^t1Lh8VC%FUMKdf{R~)GjC!r8ki4S<
zjqg2vxFk)O5;S6O`*FJ#eW1_K7t+gzU{`sq>M;-7Q=3u1Q~q6cXg--Hk{cdU9Tv?g
zv`iFpUSx2;)7={I-yNE@5^(nNC_rUtsLEUb>H8S1i_{80A(H?tf@m;nE=1?59dyZT
zC!+5}`<@)*c(m{xVdJ4*?9>WeqA)bUY?_O+L)QJW=x9OPsCZGe&BK#oR#vY%mq|^s
z>+~TzW8C*$`|U$de|RiOp6e<0l)Dk1OMHKF88O=7lg}1_MxbG6A*4W{iN=qKz9Cg?
zbV@AJpqa2OMG!UGKt}kv=na3Uu1YLuY>&kLx=Hfw_UC}Ww=*Y_E@OO+(=cpTVjt&9
zM^YP#>B2arp^_L1PqxN(vQvz)G2U?*I<|w+QYBN;wUmV;0}R88Xhl9+jo%{=G?1LH
ziv0x}XR9T^1x-oQkZ)M?P?sTdxCVqEVEf_|YZ{@i`!ib7uE}1zXg0dabZrW4iJ8hG
zP-%Tgo*8zUN>52uVz6}z%el&CabIRRDU{=KG;Mf8uh0|H_RBR*#8S!Z;nwC>_%+Zd
zVb(^K|KgILy-zQ5(%I??sV*=+6^eJ8`8@f?@k9*hjtl=W$y0B`i0xubqek)3Ad#jo
z$QaksufkG7T4dGkZw+g0P4Hb4y}bwj)!!~NSbb<XrUKpGjAy)!df^r^PLwVd<A{_v
zzV(!Y!~T4<Ibp!QsRS`^bWCxQGXnTYmXLX}vNbn{jg0<MAB4>eFODI@!I4XR>+fBz
z_(Ek6b_EBpB19|9qJo>yk{HVqE4QVFII%`Y6{~3C>9RNfS}{X9+nf{ZaKw2h6%0ph
zaCe52oM0)sz1w}fKxNN845J=7m674gFj+kLq2<~>eA}O=sLS{V=f{RNtaYsr+YW{8
z-hIOi4+CbL4QCr?usJT4viiAZzM~2cU9dOs8oQD6U|`IOZkx3~9@V75(~SXIMdA-e
zO=)mK)HF+&zEkjUDGlLG%1+8&m5@HDV!KcE^^|ZrCK*!v{VpsaBEP&LS3)N6@ooa0
zKhBqpD*PH9t$WG)Q$PH!mpS&u@ushVVebM_C;>465pr3}120Da9KW5wX~h#<B@@tv
z1q%e~2rEo#jbb6}AS@hmb>XhU=p9`Mf^1eo-23C4B~h@QPXtoGiQV`jMxJyrJck@t
z8^<Du3DdO5p891|C`oCPt-~3pr&NTzm>J<0SSO*nk-`;St+Db-)~=_@Ku@+W6zbUc
zwZS?EAjo3pPqjgDmBGJDvBk;XF#s07wG{A`!>oydZ2UGT8nMy8<N2*+DK$Xi^WNec
zy(yXD%C%W<)->L{-l<u)Ib-Nawm5L)tfhI?sosuYQ)ga_PTH#&I&jcl>A<6BXTtdw
z2}_L}$jV?fUNmFh@xh|ks7=ns^R|cLxw>T45l*)Mu)y|eDro6|Qi$5LcLrEuH>=h6
z(L`zVl11IIQ`p;k=<P_`dyYT;2UFh^9as}|JF#tB6Wg{uu`}_+)`_i&&53PHY}+;`
zwr{@wzjv*>`lTMu>4#pux_0fV+BKS+JN+sC>6QjBP>S^(<5_0k{OM-`1ZAxV&ob9Y
z3f9x&7(`HotX{=t?%~)U4pIDhyXK1p=PKXd)0fbx6$VoOtWMEj!*hBk@1BSBZpT8F
zaDSQ8@!qvkfs%WgfKwOAEO*900RJApNEuV5HH@Zc;EMfVTPq-;Lz%Ta+jg_Jw^=I;
z-))f_L<WoAiH7^6!@qGaa(ATE+vW8CCc&cTOJzGz{f3}+Z=j|%IaGOhTaJu}_*?SF
zxk_}>xv<87g^o~3U%h6z$DVpZ2?pL&UX<(U`b;P;s_Uog+LA;>05e7|Y)r<wl@7^|
z>a%EY*>BG;Ql4Om)00#rsNPGZ{Ay@zH2hQ^Y~nsM6w$@y+&{zJHxd0?{ZicUfh#ku
z8<p()IYw@A=IralBU%hF+tDNX!qj!To_81L5^MVSyl&K@M$9p|wQfz)kuY+f0m-<)
z_x{^6${lQjBE*&j5NVAZtaSGcmRArk|5ALI!*%Q5;dJ^Ee9*v$#`0Nw3oK5+(~b2c
z)9}vO46ODd0ecyif=aK1FqeBa*tP9evwG_ok@z=6L5Rp<<8^VAZj9n9SmkoB{XID*
zcHujzumZUdl+LO-*D1Tho&F<OXvp=M3DoF;V{Los%0PxDut;k;UeK$2?oLZjyJ$+m
z^-qz)aQu8(2v1WiN}ARILtkzy+oou%8DWf8yin|*Co|8D>rgCB`-$bJ%5Fyy11?ux
zMw{TTePZN&sNT5Kb%dt2`tL3-rv~%iQ>cb!l>swZ&a`r*2g$E_r=V{iA3j}2KkUu?
z&K!ePWdJoHppdpcSRf3e|G$Km{oCw6)0%}DWg1LG^=CxcUrMoJ7qR%|S4okUC<(O;
z%yeUihas$Me3lp%xlwU?5=u)8uV$KB`~`ZwMl4RCrWd=_e}?sjyEFKBZw;ha{n>hV
zdi#1Ww?RFQ|7JHOoi(S<E^nAz7Us1Z*GyIHfZ0m|?ua<PU!_}@Vh<~V7%!lLhZ<ZH
ztVw;n4)@M6<^hp6|HQIevir1VE89}gW5#+%R!U5~Zs;s=^AlSGa;F4@4Q~2#3Cyli
zL^y61&OQV7SZO&}6Jibu)9%`1mYKA_qyB#EC}w1e3H&iaEAS;qLsFa2Byvgnj6UO~
z>N9)^3}4dRi?oB_)rCsRUUsU`2b!73PblO_6rMVCaN&?g;{<WTL%otAAuo~_hrEeg
zeQ27(4CWj>b5i$<2<vH-S!tXbbHFq>o*x^Vm@{#Ju>D?)X?Xsl)i;_H?8LHjRpZ@W
zTCqzhg<<VHUUgQd%gs|M)4*WH#O1afDyA9&^b+?rXByAKL_a+w#3xvPX#~?1g5Ec$
zsaNSp_lO7x1#SFCKzZoSL@3>6s0+4r+-&0}{G%zZ=B*oWo)s)H5jp*baoM59Q${$F
zh}J-`aaR{ffb*vBJcFowZL|35&+$`r<D`W(apa)MMZud)?%gc=p%l>NX{W0$)<FlZ
z8ri%%53-fcd<}E&*HqQKL^mvGVssjk>0vX6MV|L>-X(SZ@m>xX?%X!GOc{yKa2fSR
z5RAPr|Cx;CjpZTiQoZyY#gB-<j63=N<Nd&7IHz=acd;INW&*Oqk4V1gqbvjNdBS7d
z9Z$%j+~Z_!H-o2=G9f_pD!0cN*<m<w&3H6ln}-i$@u4lXt^rh)5Q8+W4*Oqym%@@Y
zbNv|pVn?Bm>kiY4X!|>?IRh`zYkEl3hZ)z*1@{Y62AQm8XZhbIVFT#T8x$w$soIdB
zctOeeJLg!>7qUG&Zk}^&UqQ?#2@&=Q$F^yPfqJQwd*=j~a1P)`K)WORwQTt6HI!5^
zcX+vMxa3C+!>G#T8YPsZ9QjtJ1IoXQ{jpmzKlT}{(?dI1)5?lSUb6I_5-`wKyWeFK
zSsyET>c2BpTBwsFu&Cw8qkMx;eF(?u*S)k#)#i6OTHFe7#;=!Ib`mS%>HxHwzzh>V
zJvw!<4LnjW*<m0=l-Qvpn=GY2HTx_qkXivR!e=PL1*2Y6Abc|s+6vqSy`-=5O8r6l
z`uItu=)3B1h85ER3MsVAw^xr6*_bo?Q6xtu$C`C37+uJp!|nbH(O^#7+?kjR&&@K3
z=2D(RljNdR;#Emu){VchPz)4l*Q{0oRJcSI#TTQE^;!X8Kwn9LR<f3i3}w9}Wxu=D
z)-!W@Ug;}4(|P+C;VC1@=lLw<vTvZKfHtggIsm6$<gT#f$$|j_5px@$B1{JFkS&o3
zMvr_HcH6M$ZK{~p9z`q}SB$(Lvq>Z;dy6XL1n!CJZ+qQk)XUsmsHcpw4Q;Qrp}2AX
zl_xRc`wi4+eus7j_k{J|tj)%_9KeLCN03t?8=ix9<Y_?YC=NRBjI>B4n^lc=eyI?>
z^abYu<u4R+>!`Y^;Cka=(qlr25K?5GHb6dxzH0!}GntA&U>*oi5eXyfpM!Q7Q{urK
z&&6$tT?uduTI2@hj}~I2rV6tWbLIIRO>ql-&IX9QYBBRhP#0<N>Am(;CW?yjkn#sh
z0cnyie3KMV>@>uCq<a#~RcvB{P=jh;EM*7CVj4;A0aU{n8wKJ({OODqAw3<PVs_GY
zD;_#i9US(Aqg;)}k-3aCqj^m^qT@MU>kr%RSUAi+P-iy4pvR-oUb7V&saO0w(c=7T
zRC9D$X3=fMyiCTTc0&T$;#Y6ojI#s<K7Ho?p+VPHR$u#X6Nx7MXffeG$FR7Fm5|Xi
zAjlxm&T?E~>x3S0zh56S{I6Kqw!YXD!-7SQd-uJj(KaIN;}?H5JcB0LT^ZqFDDvw?
zseec)du(CZ0%kEG<1_3x|3$6PpIdP^Kr9~_o$rrzqS67!6GzcqUog#Z4ww|gF<Gvs
zUwt>jBqx1;<R9V9X3v^9{BQz)Fd7EFt*eKo3l?n=9N&)RnECH2muhk+hV~!TB19Xp
zhYk(QMQc@MXhxem%re0Lj6Zxf;WLk0C^~>I75?~m)Il^tA2|7<FFTe@5ZTh7fgYNj
zfF~HQj<z5ky7~1AS-J#wFbTe~kiS8Odu8kOQn6NTAfjSvE&l2A?hY>3dRMRGn#$j8
z0!WG(9MX=ibHZv8*Zlm_;tolg!Z>Pl%}`_FwfyA$t$!F)1bpeQK<$uH=<+0UUS0R1
z?39z)z5TVrfoa``ihg3>*cTa?1~!_@%kKLq2!w-%-+#l1E-Ws*V>PaHdQp)tV#~@E
zy0y+2?qYVxVhed-!eZDCL+8+sIak%;hwDw1bdT<9i-n1Mg62$RV~Hq8Di{z-rt2L@
zKi4?6me#Ul_`TkZ&p10cvVlzPCk@WP?e@0xeco^_b6y9a&mM=9)Om;?0A%*LLq9I(
ze{tp(W;>@GgkSKk!WqB{JedT2mt}sT7!oIhwj}M@4^Nd~rpn`w6^!ZI(6E-$HDh9)
z{FvZ?O((nQ4#dMzzWqHFcCH!gxr7uQ@=P4JyzT85M5yvXnX<z6#5b~<JS3-hwEG#%
zHEuQ!CrZ@%33*ODYTDw-1i;{}Zbp7))#o)l`gVzw;fr3C2rDCk|0wE$U_+a!$df0q
z8}2&j5awZPf1JBY4$Ucsq|TvV8am0vc<!6wmtes`CPa32S<WHbFoN4FpfD)DiEv3m
zv=Vsk$7iwnZmThkuAg9lB!JpX%dCS~0=?j3Hcs!xQ;mI?`Ve4Q4<wH?x{qqk8?E<Z
zJD#P{ld?yv{nJv=bO8}%#_wPS1^jhP{N+BJ#9DjIYNEzp=5aMZE@<mVW!n*5-obss
z#PXxtW0?x|N#r9nNllD(s><DR(f1*YH5So=N-t7#K9Gu6okk|M(pJL3<jxD0;wY<0
z{9Xix*}}<gj|!RGfulKVMm6MATvR6wQ~H%<lwmZ9Mt?>1D)V-#K{*v-C2@4GHeUI8
zJ?CEsB){fvhM}Y2rOf0-828Z{a+OvCm;1&`jTI+wS3;m0Lh@<gToV(h$kxmh6pTT6
z<tm&6Of!GT1eBouLEDv7V*F+6IfR?0H~QIzm+Tq))2U6$8z4L?97UY1uodwgZNf*f
zlfqfJ*|*Gf@-n3J0GAjt0~`A1+z|X`*d$r!WG1R0jW@Aie(c>hY1VP0s5^(A*sGk4
zo;Wu;vs-sCD55)RA|&pLrj+Y37}+hf3XMo9_)U;0;W3~lvuMlb98Xih;MP<|uhx75
z+l7SB1xebW7T{0N4Y%_Er9zjH6oH4%PK_rkmINpoC5ZKjc?pYtM2wy=KMVF}$)^(-
zYQOyo3Pyk<RIzf$oxj%}|Hl&QVrQA7#})N~OeAM?OsiHAQLfy_=qH^f?7%T;*z&RR
zhY9yWfgn9oa6sbYHLGsB#ogUgSG%0f{_ex?^F`U)8Ylx7t<jn_q<N5vxbIIw))j<X
zACer5cA8YSyd?~s-?{P{P^BYGj+FcvE(Y$d)>ev;3-raE1~EE<yEqaf*_@g>UGkR(
z%o-vU`nL~$BmTv0T^uNpS0N!ok)^`2)UFlt!T1qWoisw6H$5#Z3WY%M44?Iz;Jb>T
zDrMIk8c_4D5B-nHob-i?s{&~S9XeL%o0!sdxKrVM5UD*a)rZn5h2H82?L2tXgK*Kv
z<<dDq>2$5A$hj&1@iKS07W?sXUMX3Y&0&68fiKa%<7pdhIB1XqFS)SfcbT_3NPG-^
zWD8iDad{%`T<20pHNGlo^1;$;F#}!AuZlynZXnHaCi7X#prk7~bG7R?>D?&mSQfTg
z*KZYA%XF66XXKwT6XR7?(eS5odv}~F3bZ;Q*Ya$I(}~&(i2=s<hPtKW4&e@V+72>P
zfw2Gb?uYfcF%nUcu={bZEr$_)7Y|t3tYC4u5=u7}^Hy;#BswlPWBp>B#qm%jx`8B1
z!3W~LoFyZg_}<@+@fG!C+0c{}mI=&L&8dh-<1WX2jI1d!h7zz1#lwM2k@@4Qn|gBo
zHtLjEPWBNuG^q*AG`<n+D@BNpne7~uw_AR)>hO3)tR!Q##;t?7gH@CmA%`@uu396g
z7}jpD^n)w7rU5<Hk4}S>JmnOJ*MZyA0wB)Hu+JgRd>r24nf?vT_R8TSxTIG(xLPt8
z>wD6RvN~xGw7JcFw(&ILTIBj>K5mZQ!GV!*$MQ5f3E<*cv*=cNb84LJw@%`KEyX)V
zqKDZI*?M^TiAt(?$7?vX#?8<kPBG|Z?;jYki#PD-8+ln`BaM5K!f`^-F`6XS1~lw?
zQ=Y*<#bY`7l#gABHma4(D~D8*_XSSGn_zqkMe4tA8?E2H!owM-ES|<LfCH0#q078|
zeM7lo^>X)7eX9|#Zr*6qkX~#E7;prO$>8BehB2;y93lL0d7uxTPll5`8J;4Vtc6eE
zn>%V)02L$0t_#?0>(admH@%Y+0CMl=#i?#LgTNw2(IZFik{bQ@fkG%Xo-lN)=Fgv(
zIxTY&@lBsHjA>b5tYSMcS=<k9E!=yzlnRRuW?Me2+}xnJ7142F<oN9=pC2yG<Qb7d
z*s}TMrTktj5`G&CiW_ZZZLlnA%C1_>SCcan&>K<H4CNtL4y`zk%C-_uTJ``Z7;oRN
zN4joyG%sOD=PLP*C8V9u(qC5)^$EO}lH3IAr<0!brpNL1?j2%kuQnR4+lKeT>IjWQ
zA(l63I_)9;FJDq~>V?6pa{Rx{GPJj{yAB1H^S4uc-@!29ygHQdPiOrnka!<ca^Asc
zf%n4Uu;DNPvQ5DhluM+ZuvE>V{ukJ!ZJ7-2PixA?!Oe5m(VDjBX{Bds{zM*_g%n08
z?dVuP&eEf!spkpu4P3s-^<5;54B_DBmtJ`G%W4B8-?d8H7hvEsLtLc8UqS{6SMAMZ
zKD~rJmRL{}3is4|{+Z?YP2GS?_IyeJAf{XbAKe;|Qsu!Vl7uAUGq&Ny6bzG?kM9eK
zuoO{1w>!y23Ke@is=nT#?)Bfrio)K>tGo6J%1<mvq@_972_|-E6yfGhIct&&31^nj
z2LJPynqgj9GQR-RBUh`8eJ#$)4KouhdaO=6{j9Xvh*$($esA6tVX6h~QM|+ifM1jx
zG8eB`wC;5?VsOhf3wFuVQ`%VC4H#4>1HzKQezG*s#KPCBf^!BWVrLVnKtJ|UynCT&
zQOA!4dXW<T$hIlFj;j6E7w{fZyUj1>f^1RAuK1m%G{_mW(mTIdv?I-=QUx+JY^lGv
zTf;CaALc%^@<z<_=_SS*<ni-2P_AgV1({>6IPmAJa%;QQvsU&a%G}Xa!f`Cxiu7K%
zS+BBbO*y~oEe830`}Y$CO1?qWqS77felga5+hen0%2Efz^YrPP;+g#AUIGXbyinu3
zWn=sfDBCwJClFQ0vEq~s|LVt4bu(yG5W;~^!(HI|q|C|b`QwzVwZ-=nh;bI!xa;if
zXm9mcd-Z&0_h!{G@v|bfw#A;Ps1DTOc}m{mENIcf<9o-?4dDn|Cz6tL*!cVG+!1(e
z0*zI7eAdsVo9)`bw^6fP)HR8IEmP<a5N00nqzGru@5@YKV?k2{Plpe7-|5pF=oB{o
zEe8jZb&4>0UCI`EQ`A)n$R^-`00$gPY^}zkNOzS4ZDyOzeVNy@h-7v-wVu6L$B-Yp
zpQn)e5RSX|QnoWgsnD|C7rAHXH`+h34j$jd|G~EYgHD_bLS)ae^tOe?9?tfFZyqv5
z_$0S$q+_70okHHA5anTUw@A3fEh{lG$+mMo6?cd+f*BkMvwkxH^y+nNA92HLP$7F*
z%nLU^n>Kyi|H>ZoqQ1$LYFASfNFus~L&8J-LB86RF@$+0Di_WWYAd9oUUe8la#5r*
zH;rTLVn~q@@BLJzzA1-MRqZlATsp09ZXHr5XE0$)c(GE46X@AQ@9>NmTM>@m<3wk3
zONHBn#-RCBrKnT`j@Y4!CwN{}^IjJEF$}N0R*KzKiJRbGLl2P9&0}S>6MUk_dPfUZ
zoOgs9kFlf~`O0dhp8Duk#_EU}GNBHELalvztp0uMPG^B6lUfA+&*UcxxQS?@Y<f0u
zAI&BUjOLqhJG<TLV(bTRLq@v|-X;>4S`w;Vq9!WZ$M+Rr$ZBBW@6wwF#sEg_c0}CM
z#{GQ{a5IM6)z&B@rH)XmLlh3GrV5^d=-ToNM2E}~qK5c0FPYkou!P1!p@Wc)#5Iam
zHaIff4i59X2k64<oehTPF)b<rJ#*nEMXGJ}`^iMcSUKkkOMOH$g?4eFHr%8Xf~s4G
zwOL?^XebeI$NY;|?US;z#oWQHQ29hJvlr#i$yW@7w)G2I-GoQJ6=7Q74R0!#{#~;9
zK{7!e%SBN>Skm>2|GvI?A%M*jD_In8jRw+*X~JyWB474HV2*x@$>d^|1T<6?GlJ%+
zGEY`a(eaK$G8u(ug};2D8^GC#$FC~<1K&%*t1An@)bsRGaO<8u%LU0f4inBzKqtKk
zXBJw0LuXh)(WzPxY9SA`ZcKy(2?Wv9Jao2k_s2tPQ3@!+q_UOE{@EKJZv4k=Ub-i(
zrRPZpogAI~j`#t>GH)4Yga4#w^+#mPv2=`nzdA_2oZ~pDts}8nyh)SmjD%0WN*=Y3
z45=M}_ETJ62Get>oO3i&_J?O=zDQy$&62GY0g`e4sM5JVIQ9+~m*JZTo=NXL-mm>&
zi)FraEwI?waJ<<Mn{mctx|tVPyS88J`kkCTq&ruE$5Pv5wkHWbe}5;g-On+!+mg$!
zrPD!)T_C}K=lI?kr^+@-vu{54dj3-)oe2UJm?~~Nn^E!oI}iZ5-g;rxUTNl=?-i)?
zD>uU|d$PEVeLm!2afhr@rqNp&N{mL4%z!$Cu<JvO+WSNI;|8M25JQ^5V0Z_?W8fXq
zS4g(ktt5-hlL|&YNRnYV9S;A+$BfH+RdyIQKkM)4QApu=ke~M~++hoD0sEBn;z0Ej
zK!0n}@|<R#dU3WyHYmkxYgn&i?dEeBC~PG0dy*~rg9%lHT`I!V7)jj6p@dc=lU#`j
z<Ef!4iY02w;Ij>s_KkN_C@V3ea&XUL6auXAEyh&OE`GJh23}Rf-P4(+am9ZL4iv5D
z+E*ThVfXL1wE2GVP!rFkrOWD(0}P{d;3o}RAEIG3(GeYiE7&XxBJFK};Y_1-K1pVP
zID=rkTMCh&zJAA!`_|hg+0q}k`=NvT*K)Ux@(lP!dy(0f_A9@)Rjl=E;?mqqbH&@d
z)AzZTIUb#7B4BXsF=RFUq%4|<799r3(8|apw~}<rI!V#01IqR~tAE?*`)jf=kbe#F
zSPLH+rbmC(^K~QcI;EduU+krHJtYY?$k9&QUQ&_NiZ+pC=pVki4VL0{2uPgWDN{BI
z-!HRMO~4uO#^)mYD6xa#Sg1N+Lk?3kNgV;=>+p*xbF@vt?X(I9mCEMw;$hqjO|oSo
zrV5mEuYPuK2&!*uD}rxO42Jta=+4F}>!lW%OYG2w4kG<G`^QCNk{KJrsGEI8@~JZc
z$fAojW6a2~g~Naqw^q<+q;SuiEdm@>$@Z=XI1Wp&NvsWpj>WGz5}g4S&gGT1GLqHc
zO-_2kZwoaaGLX-ig1B*6%1&*2$B#88cY<dLCw8Ql$OF(2@Bu@0klz*nzn-fW!zZH?
z1?=^b#d{R+;8<@LZwGD5<heGjIM5`#%*@663`J!WyJ0_D;(I9)-^Y_Z1j|P$S^B&V
zV;e7@6%yc%hf_|u_A^v*v(dJS@sm@O3QUzti&cn20KCXPyUot3$X<r(UV(sd&y>tL
z{*OHq_1_4O{g3Wk;{px<gwKW{ojCjNswoQ7WQA=ePSPSXEq$W>GX)>Wb(Q*Pv;tJ7
zJLCC7H^;BZYGXLrLf2A#*k&)JZyc0aq@0Lc9(d1@Be9fRw9pE-Frf|`33KbeDSBjF
zB}q@2lv0&aQPC-9JfI3;z7`FLC2D|JC@!ZmpT{v_p;9G#z$q}rW!*)|nt~|(nayQD
z)F$e4Z?opC46h9Th}!3b%YmBj#1O);*YZ{t$mQ;h`iH<bo#(6IIstSC=?C35@#hou
z8z0as*e=}-_1x&zTg@Vjj0EeI6KJ)d-3Xg87Ryd}nKFk{Wi}zl+QpUbQ@mO_@jO*5
ziuC;xrmKCnIuC%I?=xXCU4ZfH8*R4<%ls`)X!a79;(n71?&iF`GAR-pP$N|%MaV;#
zB5s+q^?q*CI$9>=OP2V4&R6(aHZmtsxrV%YK5QzEz}`BnAz~#(VS=Dl%BH{~&OfmX
z6b}3VNTrh8?vbo+YUkNR-&3vN`RM#SwDl|_=FOD7ngQ~@?KnQ9NYN1!1QNTntEQo=
zzb73hdI^Hsc7pfA8E5)uAsubc9WLUde_R$QG%oG*ox&jg#YMl>xirhG=Ct)@C(!AL
zP>4+oTGduKoAxsNY{IJZTN8a=aY`8mYmSQOL!5k;uR=6*N{wiq-+ClhYqgl?fq{s+
zB2TOY@dZ3Vs5V=ZFgIP`I_-RH>o<?pP;Mc{$sVEL$!uf>1w-Idup!W6?GRl9dTo|i
ztBEi-6s%fRts4#qVrpXY#@GU-*&qCI76L@`v@Nn3!K>xHY~Y=pXnr7wf8;~UGLOcD
zKqH`_Q-p*-^Lb@Q3Ok{lZmfzP{EOv3NKezr6$S)`@CM~dY%MS1KA(Jwbh3?9e8DiI
zQE@5QWGoV=sYgr`{|7H6%W7bV*Fw9?nRo}sJNga;b}7-smzpRB!gL(-<=*;Dg7F|7
z^Wme5Zc;87*;!u*alUVtIbuSSm0N^9prbU1-XyqYx>{I#*05buW`_!C6%4>pV-maq
zTJp>vS<UETPC<vz6Da9q7KdSDnCWC}k`^)3_<7tSPR@o-Y#I5?A|^qD(9tNkWNZ>p
z91=0iodbO+t7Hw57O~qaWR&-t?2RGwd3{JPsZ9iB0GffAY+-s~-O5%oADhaTL`yDw
zG&*i6!99CA>B1=<{L-8pk6AX<*Z+W|ic-h}^!e9bKhKpdJYODkZ!e0S4-_gU?0Ro@
zcXSgz`!ZmT%-(Vs--wtAFvlX4C-Txo*`b!s7pYX|uq}%=iz#0={A{pZHZ0<%iSvY2
z=7~?7`oz709;9GLrwQ};MZAI@Kaqs3M3!4Ol`^amU$NCEueN=F241fSNMM!VSz7e3
z6SN9i1Mj1S+w?{5hSs`y1MmNwm$}f<D!Oa{^Ib?0wL6qNN;)Z<IE}Qsg^P{q+1f2c
zj;DP&?@M=qE>NHmT?Vg?Q^tZ^W>cl*Bj2#Hz!!+5>Z4@&w&O0IC;n_L&eCA+lVH_G
z>x=RlQ<PEk9*79Tp(+WmV^jSsfjpp+_h5IAwL)GXF6$w8hA0yrFs{%gvFg0(6QU)a
zYqF`eZ&CK6#3f}Dvj`)`C2jE>D+D{o_7l+n(SvZtm{u41U*n;kzscm9nmo&E6bDBx
zrJz>WBydoI*#ym%qQ})wGcp1bh8|U$yg@3oQ7S_UBPbcLnsCL6i`VBuv8=jg9I!nU
z-wV?RCx#Sf1nq#nBb|=)EBV`tOEi>ffp~8g6U;gAsBA{Hp;E8qUpWrhWa;FKsiXP5
zQTAS?Wn&7`upk+Z<LgPxG>=k_a`clqZR44*d~LsoV$VZ`+o$8vs`x$~49Mn^rx!Z@
zkRe_E0hR>x-+Bz)|FiTh`%6eqpsVvo&bmP`_4WTb(PSOL+wY^`Gk9dFq>TCaB9*?r
zN;L)%Z-cd^ih2$aAFA+9N|_!_S0gj2IFNsB)3vcBG+pKEDfn2?YLxkg;b5W|JP(N}
zPP$&}6dE_HvI9)AD2(KjMBm>1g$E=iZb*TG7Bw?)<w3W#U*fGAFdi7-Sx4ho>GCB9
zC_Bi^AS&=IH+@&O9!lTCST68FtQy9siV2o+tw(U#WEWIuTek-SzJn>in8#xE|AfPy
z_O7~pS)dw}xhUu4VK?|nj>lm>%wBbF)*g8)tZPht@Hm~6lblZt;;Vf5lZ|S!wP)be
zn_%{Dirdf1n~2YrJCJM5znfLcpo4*3rFaUJNF=+Z7cC#mzhtvFwh|N)=o!)IGdK!b
zvu_)#{|;``s`htA?7}<E(<xKq_u6qmmv#*>we3m~HLnZG@I5MbmJQv;j%=9t--kJK
zs?H7N7sl#H6YZntaNM4PO|{JN%d`VJeH2AT%K7zVxVO72XfI0rLN~il7Axd@RP{*p
zqWCj<j`t_$YoOOQpZo1sAWL%{X|%pM%)4C|+3TlniFY)OZ%C>iDa*{;mPfOW9j+?p
z*9f}&0SsMaF}Xct*RqIS;CdpQE4esP5Kg+$%KDrW?K@V<*++%K5I00OSiOLLEd1D2
z$!A54Ea=;xbsP3JWt*C}Nd}uZqpH`E9jr#rRiVcK&kkb64(7ckPy%>Tum8nvNz+#C
z7ty-jZKvzHon0?;s0;R$XPh~k_f@u1NAC;rXE	>Dzr(SDiu(_@7)hbyIYywjA*
z+RYd(S7R1(RpbEq+Ua+IaKCN19E!u+=E#}V+)uyUTg7W1bE^-Yu&B-sCurN?iie|*
z)L0{UY~K9h$JU=WNgpn9hK_<5+jtlkJs&x}ht9`9udgnqW6V+;tj7r=D9a}3ab}>s
zMv$o_?rcB&d5=o^>fw!9rol#=4}s8Kn39L}63nZ`JJ=cPF7^m;OvhD@l0S_sP(NiK
ztyGa2rA6>`m!iZ&OBg&I_{Ab<-++*0TdaS^x3~(yd}iCY5R9(-RBrU11b@>Se!)GM
zNzmc=AG)H2vOk7TQ<kAfDinObKuomek+#E<+F&Gg_q|hbWA6eJ4!RM}0K6sdsKgDC
z^-H%KohMGA9Dxbe^somM-YcpNv%*<q3lN9xO0RCjHxUj^8a4Ds;{-XbTr4jxr=v;A
z=N@C}K^bwpfa(jBkk6+T#%`FI&w#4f(QQHgg|Hztt}BV7H8G4*j@00v8qF7aoqyQR
z#5Wac+I=U;#RM$47fU#BY3^^wNj*-NrdP>G2dnF<&4C03dA7!Vl>|gS9`Cv<24QHq
z{JCl*i^Q_x`v#fCg#MiC8jifSM)Gb913%eSX}KfBWmLI*vE^#KBb}SG@{|jQ5pm<p
zMrDRs(^T+--g&*tbZybo=^ZE^xi!Rvu5iW|)Z72@tFuB@j!k{OU+t@i6=gNu^`0cY
z7YPFZ9tHRIp=bT(Md7LmmAtL5;kOAlzbExSa?4!Ovy;IOD^<>hePWQ~cr-8?fBLgf
zABe82-FF3%Kd;avrFk(vp9CQl73$%WirUcDj5;mmaVxE4m`wOjRpns2_PF*Jo?h1N
zJwQI_*^5p4>5jG+n}U1x-P|bOpj;iBB*_MVO5{WEy_dq|vuHS}XR675%6oHh3(>x;
z^cSBHPmai8%r0dz2DINf_ak@BxzXOfW;k2sJ3m!E(rz~ClXQ|KinEpaXxUK-e!qfc
z#ul|DTzWkkhsfFFY~M+5M$X@_y*uf#C7_URFuw8SAaX`+g+!mr3=#ytX$HS}TdnE?
z*^22zVQHiYGu!F)pRqc91-jsJ9a9EQBhbKjO8^?q!Bte9<)7*Gv8i`Uops7Y5=GfH
zOH4o9mV+vr79sHjHrKn8N@lJW$#`d&2mn`0+9DvnT(FXZ;J95izN9$qSadY-g6BF`
z&(tDHVtz+M<njq5YCo+MXJu0Ck;enrSO|Hju$yfky8I0NsUcqXOo*hEyA)4vpIn8D
zJ`R<?UZVbRA78#8QJAgN?1_<;V+d^92l3w)A&CnY?Ra`H7ukt)=w##|KZ}Ru3$`&r
z`wQ?nf3ZowFT+Y`2n1rN?}>l{RlDbvP8&_$U#@smBQ+$DTJ;AlpDJ^Op{)SP3HF}O
z!%bo8e7=~i2DeZBm6iNCT_<13i)Ev1WzjV5a5|nA>oR$(#?;OcnoiaV*Y-%wmSu8E
zX}rwCw14tWSxzC2Yt}y-_)be-2}wNW(gSh^zDT$^4eKP^GZ7xbDCoIrkQ8>3k$ex^
zdv;|n%4wA-8VD-VuNk#JC)|Ow!CbUnlgjmPCAqno-!!Ig>ainWg_O055XSS&FO5s)
z5?(B6)24L_QTwkFd0!zj$OZAR3%oO-QU&N?LQ{pbn_O5L8cDv@IIJjjJl`(AMkgZ5
zpYAlA)}tRwH~Mlxyr4gXEu?1YmumeE$w*P#4`i4zOGXK)damg7wypY#|66;HMzL<<
z=>(_&nm7$v#WLQzd7;TT8A(PKkid8xj6EZ>=mbV@LMHoS(M@0hFSS)Ky-18l?cVeL
z&pa2){G&@nz(jzscsG|v*CLPcw)78e!z_poZfK2-=8iSBm0qa}rsU`tspI+K-&+@p
znQGSO65R4;3SsDWH8QASZ`K?O%dhfo+22B%jV}aqI8vn0!SMmqhcJiDF`SncYDjE6
zY*pBGu3t@F=g!COgXfQW&-4YaZ_+Bg?WhS+(Y&V-+yerCuW!6QBAjx^B+(QJysUWh
zt)abS<v0&Dydw59etsfzzS+6NGSN%Hr!<3Z+eO5-;yznMZc$LafOk#ji5*qa^>(qc
z=_)ou=Kss-!ps7|D^~s}5~SQ+j!ZvpvOD%|I!D(IU^$z^2a2<Pcszaogd?|`sl8y%
z7sCVfI;8gv*s>r)Bw#gusM>mPV%+H?<F~QpQx~(v{Gwe(n|L#Bur9V-hK#3MJj^*J
z|K7c-fR~SxX&#_H4_sp3V_$q{5U))wzB&tuy?>@}UC;zpXyX~{?QAr%G-;ivbt)wg
zglvk$5V#NB`hqZ6KE&(!KPljs)Mc5OzrVO<He-X#T>@Ky!z^lr3WYQB<>p#K`n|U}
zSYpRz6h#bQXjAk_<BgQcsqrf*^vH1e^wXc8zp?=SY^Y7mao`e*JFtMz)Z@XW3dK<~
zjyK^QB^1zkOcYPQ$X~H}P>DVz&n#MVd*S)fC+--y4tF4hSKuwzN}rlfc0TQF_)fTZ
zSxO?K+0|@xX-qFL{=zTaIJSFZQjysH+Nq5j?v?eR�EQYgXK`qB)IMuMSrvB7$6-
z%em6A!Kva0(eZi(hIuFOvj?WoqbBz;*LpBOe+5A93DAe(hJo`eOuU3Y=?>V;I_?e7
zAEE9E*yKmd(46)vQ5B!Ha<xvo=+ObAko$#P6J(8Ha$1PdN4q*qu8o$xt4|Kj-E<VM
zie4K*5I5QV-Brn+Ke4F~qAWkPt&bNH1{EhR)D6C7oA2Pj(eFB@XGZuVlNWM?4DoOm
z#1QCz?2w5gp1eUy`u)wIAH!keIN>Hu=*50mQnqpuVmFYdQF(nI$K!skH;8_Q3fSwf
z9C9Nvvqi9(bA6J%-t40lh#Je_PImu3nvdAQt+<icv6feCm{pH_OqJt>NW1LCxFi_s
z;&rZQx2db&@%!6;AkW4q**UFwQE<P|;TC|Il;NJ4p|ml(Zp7ejols_Cu#tPV%nLVq
zariP<SG5*gdX;Q}uA_H{+cDS^-HUjg)!35|h2~A<Kb5$yI*`aSAyFJiSC=HNVAW__
z4(lf#rJTG+hE`>7Fq^hU(0T_Vo!o=~a@^G)eGB&o4Kbv(5zkNyEfJZ-i}&!{tQA=A
z(cywRo{nRpg<{CEGtyhT4nTYAyq=GAE&MSA{SqMp?J8b(04bfYkl`sfQQXHrWA{UL
zVsKAX3~<oc7yI;BItR73*QQT`5R$YH*jnm9<dD8txNZLb{qsMe2YGL0ilL0mUys}K
z_8oipm#&5S0^Gr(2f(waA591Yc0Oh3r(R`}_d;d%;610bFUK<e?ThWoBzf1{kKgt*
z?!$>%ZHICJhL(9!>qd9HEBb{Rsil6$$p|U1FK=}}J`lU3k6oCrgF#xLZiolB9{^Q^
zTm4lj)4w3<0@VeraXkT4GOyWpR!J%a`l4vH3u5434=h%TGr*UZ3>E9Mu|NR^K`d($
zBB5CUtU!N>CO(pYfe52QeQROa+X7*DV{^)yx~eHJ^%rSid<uROk<qU2)bFGbWQsUx
zgyri<zIqlxaS0v0xN<Hbl|?$mnc9TfryQ>{c6ZHzQ-tP1hZ*8)t(Yf}&BkYV^~pI&
zHcuUs$3y-t3OvzhX2um2fd1hk(KfNI(p+X}j@B<-JRMQI55z^?fpD+Gb87^LMa(c3
zQ60MYn;ToR770=(L2I-d(f>E+22y@PKFFbNwDH9fZ5Ds-fVjXo^&Br58XAEpzcIAJ
z4hyb-8ay9H46TLH!2-0$s41`5R`)wf6KS{t1PE4VDDZ}?dr%VbyehciqHMPipVpCQ
zk3O<R{@hAT&F*p=;gzKH&7({sjXlXy!zbSh5~)8G_6Jj&eQiMFgC=Wux-zc2H$_SE
zlMOYDPC1_VL|D2qM=`mK-i4z*1k^qXzHZMlNX4Ao-S&6E?IE@JwA^V%Lwp)&rjqSp
zC*FEJ4?vGCV}g*pnd<lwaw*evtr<K+nGn_5BoX-cL9o8R+6-MgOXD*3J(0pRuQYHX
zppWKlo+nTAJZ8$xi%#I~64Y1U!pc)(Q3hk@w3llM7dL9=sAKAY(JvDB!wDWaIDy!g
zupnqee)?!M<zTJc(o5+(t}K`R7gV_Hp(YHb1R~hp<mmH$#B0$qqkYy3q<uUXO38`A
zpF?N&|MchGQz{W*P1@Hl=}t~(@bT1y2=>4^wl@gS<lNSu!%1I(*X=Qn3YTRvD-jqf
zcko?CmKs+gM{Qak{y6R+utwh={)4G^F>f@uH}7AI@JM!5cXmH##wq_R33)r-tZw1_
zHAM{%oB+64zB`~feLHWQ!w;7l3hp|ZnL_ZL%iq%>htmi10!Bi)ib(PIRZu9lN8z^J
zK@Hk}EFqu=`_RS(Rz>ZI-cjV*U(UVv_k%EU%Kl_evU}-}K^ljMj#Md49=9)wgJb!&
z$t58ji)a#MmTaa;=+(o=<(8YFhKGAGKYs}Z{04jsw{af!xPs0@s@8tZua&w=%ZKej
zHI)W08L-<c3_&uP*cJN+NuoF>qGTasX*iWi2fHePlM?a>C+2=B<^0v9UGk^q7O*;Y
z{{xjb@&w)$(rsD1e$n`ixCZ}k2rY&~^u-&R+V_46*U265cH$lG9*QP$J};SnC^)<?
zcdkII)q}c3D`hDu_>O|ZZlJnU9chf5Jq0J%Z1^T-4%G9yDc=N!w5<TaA*>=}Bf^ad
z)`VxZvs$-=t)H&N5l+K6Hqs4+JPUhU%k-%00`=D9Vk9^x+X({1P7NJHn^fwc>8#5_
zK-0#nnOlIU-4k+RJSPjaTretn*n`r2>ke>;!`u$G<{G1rGZ_@QUKUzJMZyyy_87zu
z{F5ETixB86^QJ)$FMTXMSu&Bk<vv_hhc%k3E3|(M^SK*Rx{d1L{5YA-<D?Nr@rg;M
zbxHpslZ%pNRsTY~l6EE9J6y@%il3biZ8F~oV=f}X3i2(j7Al%IEh2K*`onlzXaN;}
zR|yE*o6t;);9|4-$yA8)iD&Fs%CV1!)>yHk^t7EPIIbukr&`P+>=*L!TSV=m$O-Wv
zi3$+~svJwj#^~%)h@T(S6Wm!dTnzuVTI+$Z(0#3c+3@AUYo^$RXpwLNopNG+!cWL1
ztSUFhhzy^i*b*!l$tr0`*t&}n05S>TD@w{Q8zgw(nu)LuJHi{+9UdLTG=+|F6nPYR
z_nK|Dw+`A~GfM-#^mGe-X!&Oy;U<B!&w)dDp?<Tnd8k>94J6WUs(H)8YB2JV7vUHX
zqkl&(pf1|j4}q}!A@5gj0Y+UL4m~wQUz@)vZ!?p0p{Bfvbm}y*f-FM;4xu4rW5a%}
zK4@-q$%zA-4H_`Yg`D7!Ju~@#!iT~v4L8pA3B1e@Ob+V<g`v)2SCcCKMZC!}#5bD3
z+_1_C6MQLlx(lPlQ1W^4F{OkJo@>1Y(^#j?gkP#6NRSAU`^_lj=gEP^rbhmhR8?0h
z<>bX+_7s^Y3$TxUiUBVN;!MTy8T3(+olDoTx;+bKEt=6lls>}AQ7j}bLd1|Sc}>6_
zlJJ&*6-<oipHyu}*iNN(?_E{Dnq6<sZGUTya|`OweB0FX9Bq{7aPXh4Z@@#{u|9o#
z7$0*Y8EryDBu}TQI;IKcEzgeRXKp%6ax7M{R-fJf%b(J<qeQF?NSbRic3eKxt0%K>
zN?=oeYv-O9<Lr+?Iwwk|?50L#__H%rZ~2sf{NqzR&)p&-!H8Y;sPSBBrhbL7-Z%u1
z2?clh&e3QQP<JWXch8m(LsM^2EhQa!;*BVV?e66<gSuh&4<zzSzGsuYFObLiNtkG^
zN%Y=sS7*~h>BN8>@VWG*cZ3W&;yAAE<X>_(mlnsi7dlEWeFP~t^<iO;MxhEwhxJEw
z4h4^qQ5!XIRB}(D@R@wv#8#dwH2hU`hS1o0SVKm5OYQ_o^uD*B>)#`{*TG`39w=$@
z8lX6qq|IBnV<t`al$6h$Y!2_MQ<b=fhfa6VO?;s=J4D6|tUX`<(YEGP`4yUlOtYy?
zi}$>Qzh@)7@*0i8>>k!0H8VM3)=Fo<8EVnP@CI)CS-@?{lI>gN9ZnnQ4?4jZkxOb_
z;)SJS^c9=^SISaJy2v<KZj;He){z+B5bmLn!F?{y(Vk3BF;2uhvfNU@aBWL-ofaMY
zc~JfKz}5zg4OPqi3f-F8u}cdX*er0Wmg>}cKsvmxQ#{KK<HP%Ez_>1YN-h5#+w`P|
zmL>QI<cWq)?&a-l?T{F3;7(sPLWgPAOXu(~oNOiJU6xqyhjg6annW@lPhvwxx;_u*
z1XComix0I8h3zhNgz?{E3No|*isf=14_o29o$LoHJxETwV~92IR8xoC=rwisDMvay
z5xbZ{-CZW_*{#jewW|1bKCtADi;H{}L(>;Rn?IRD;NfqvKHg0UQpMgYJWs3EDw7ou
z36a-898!kP(&n#Mta_H84*acOV^J}A{cQ5VGyEDNz}9L@3LGAt{`!W1_7%k~;EWtR
zC~W{EaHrCZRbR*GHP*Y?sh)mXB`N#dIG*;#)wtP70W@s-p8AkqG7e8%^3VR-Xb^a?
zm3q9{3vI3h?%J%j-Lgf^L9!q&-@5n=V!yHr$}P63-tCdx0p`((fhX-!j)3+t=CCYH
z8^kfDee?E}Zmz8_M3$%;w9J#nAoXQ&7H~`Ja18qF@$Vf|RWU)}9OtmZu(wSS6(-hK
zl5_**gz|HytYPT>9+duAd{^ngSg3D=t4vA&_M`q0r86|NRm9`Iny<Cjs&b#xsBfFD
z&P}4VSN!85i(^y|dSdT$v#qnO-Lrtjrh4SHy!n$D1<k9z2vyO|3q+H=l_`0n4Y=Tk
ztdTz3crE87ZOPiJv?ZOohciF~-MLy1C3dcEu>GaixcRa7>8VH4`=vQclz#9NW<Q@S
zjp8)*Kr$6kMA<Z1@DBumr}J?^saPoCf5uFnqEE(i6^Pq0#WpD2C>Ix&;WbBHz!WU;
zMF?pIe!4u<7Hw^8=<_BzIK~Q~gZVJ%UA;fw{$SU{*N;>s&|&>p<)KR)hU{&j*SvXe
zyIimP8~#MiO*h@m+-HfkJq>H=0SinjUZ#i$;-%yMcY{|4ZErREM*EN-@v-0b7SQ$Y
zP2nr8{p0nWh@WKHdN&G0Fud|7_7xf9fO<qQUKkQAu8F^U(V8G^5EM8-eTsPd1U)BA
z-ODFHIIpq-<@^tguf%^Yai(89lXGEP=;Ye{)c+bg#^6e5{h+&rpf1v?(QO98@UxBp
zJGgSBuK(dAiC(imV3|+@jD(5*)a~2b-Ji`$*Me^bzDL8~Qiya+5~%=^9X2>w^K3h#
zl-*#A?mg036kTM#)outtGld$=n(i6|+IR4KK?MZHyOrr>|BBb+_ksk^O7#%?b%cPW
z>YkH%0ux$f;c&`xW@E^y552}=m&=V;L5^mYkXWQxyRy9Z$pxif;-rX(5r%c5x~PhR
zqdC5=l|#axxbN2CG^@o@=D$0@@W2z?*~hKff_5B?NxEK-Uw_>Kameb<>G8#fCkkjU
zlDQy5f2KSAX#IcfM+Yrh-sYdc9l~y>KIm<Hq2lg)d=s9Q1p*{PLh65iYBt&fdIBDO
zspDbpZO?bKE^<PSl&LhxKak7|*;x*XONyXb#2qCGrm$;}!g^4}9qn)v>V-T1EJXgP
zSb!5}as9WVR{u55$^~X>4P#_|{`F1Lz2!YGgS8CWl^Zd6Q*aEVKTj&vR)Ch4OdN~{
z3aS^4)ufw=Ta+)JgkeVqR!*hiMB5#8{Dsr+E+VlZf;2wi*H7N=g4v9R-Q;ToSWK!%
z5$=+Lv_kHBhYw%g>L35hRq5*$o1Y)Yu@g+qGMwFM`T!F(eA#`>kXyObBhZ=>|J;@0
zH$bbay|&CSqg6`hED8j4XF`Af7ckwOKhHNT?MaO7`_&$;dD;g+SgJ1N>8vq-3}B9T
zx;ed<Vl4@l%~ZO{md0C?muVG@EQa=Fm7J@OVRfM5TeXsZ<T}~PF2eKM4_i5CCNZ8?
zs+3j90==58*#hh*-5eOA5e`ITGA0$&e+BFJjMTX&UmdJlw&JeiuItR&VF*aqNY;Le
z5j=*}{cKhfEi%{IRFu}dij@`0C5WFSc<^teV@y4`Oo9RpRZf0xmAj(B2)7rLzsx>q
zN}cE6YGGpWG)>EwwjVPVaXFJB&AUM(F1yHr0p=L_)<eTrsea}5?=}|{973|t@a{=~
zLxq-zqR)^B=u=HNma2eOYDST%(G^{IF3sbrgdz$ao2<H)e8_wM9-$^sq}Fltt<zJZ
zkVlbl;OU>eNg!1Q_tPu*!m6ll>49V@hp{B>qszCcvob=jf;IAfuQ?`%q(u&_*zm@%
zfFbA*B?b@bEf>dkjT;gC<rMXQw0EcT52y3D8)_$EMDjypb*`(#T|#_Aef4^8==P)D
zRyVW}A#ctlWH{d)u)0}w*hoY$O8n=!#|tufBrW+y3DPn7Mt=#6E{q-gz;Nfx$PFTi
zCfm3%)5}!zqAJyol6u0tU!Ey$UR`JF1=!xz+-`z%%&Kl%-`Z{{@EQ}@_}0XXOnLVg
z?K`~vkE7VTfMy%>Uj4gL%g<$1F<h)d75X%k`MfYgKQan+yQrI8W4hO%Lqz8UEAMrq
zOoH<4Mc$jWT6u`NtIWto3R(#0sg%Ax(-dY}m}HpX3Hn+VY6?CNxKL<yi_^z^fK*(!
z!+=p<czFk}JNRMqucDF+>RxUduW=^oAn&7HHRsv6zW6ryOk4yR9G7g{Qb&|bz3XcE
zSnT|Bk)^tjk;mi7`NYAOS8Qqb6lKuP%6MEV`*zPHt%CtF+I7}gHC$b-A0N8`>H&xg
zxN6&Z5g8=t>Xr~D+YhJ-ah@dsRtUZ9^S{sKoU0Su{-v>xzxt<MvToeKC$y!@%JW>o
zuYE_CE>wDYV=?f};w)u<uq6`|*N=DfZ#T}-i_`{z&@qLS^Ep*X<#&v<+O82^+ZRgy
zcl6QlQ55FxZKb!N%kr*}$ZCpIB#ZJV2;xCrBj`|qovJy#?@4j};+h`@z@!rVd%-h<
zw8xKq+2gJl(NAr>&MBv;CS~!B09!r0?cLI79bwNlo`yT>@0W%2Cv=em+l`lr8QZO|
z1+T2N!2@N__WJM8Be`bg+rPF;ww*bJ)(vZlfZDdXpg?(*6gZyZg&$}S^qB?icF2_*
z=V_c-9_f-)s$QbBoj;!d#sB4wx;&I?1@Nr`arzxrWL>7de68J~=B}{Xh=5X<F#|7T
z`qw38Y27(kfp78m8TzOZby2;=H|c$zA&gGGTqH_*kdF%HSlwpBp;886l3@Pben%ty
zgjsCpV4!my@ZtUw0<xUnmXe&IZf8acA{2%i?+Fo&E7Z6e7-z&*@*Z-TNfEz&`4=6`
zSgt6zOS&y|x59mi$ee??;(uJ$Xg73uV_C0PO8TTz$iGbf%-#FVJaJ0WV~0_(#LXIM
z#NY{H`5g0WmYut}plgOusZM)&rNE%(U%0r`bEzG@^LdSr+Q+7)))N=$qq&K(M=UHN
z>5snqSES2oz>Z<dU_h*eWYNwvObNGkq~wZMSGUGd!fXS+NyuKlI1e{LGMYYw8;AfY
zQaw%f`sTNiO&C08Uy^gMIHkFQ-R$M97%(t0Rpnlw$uGzJB~F1YKF+yGojio*e=H!L
z`G<)oXeRHUDWoFN&8k!sVjk|^r?E41foMt2&-p2AWP*q_{!Xf%-F@>Vn=xS|^%PTD
za5CU^lZIBK{J@nAhZlpK3DKk&mG7Nyx%6PdN-vS;evU1e#i)wuBvrdFwLC9r&aoil
zRBFw%G~x%hlk3fjf_bZ+Z4z<j;mw<jjO`)M7yEzn#s2SO<)|&A*_XLg|3k#?PQKcW
z2J>kzD#F^PkeG(7UZH~BPdT6pV^)hg9|7!POg`^P2ayzGgyFD8d@2Y9ERz(ik@JHi
zQMF!0eD#uNR%eE(vAZ#+ORG+dVwUn3D_>aB85squY8gpqg?>M{oRiIdoU#Aui|rZ9
z_$73YQA!PJ6O|O?3c6-W!D`x02YoY@R+~(;{_GL=^Mm8M@rW`+ogARW8sX((a|S+~
z8l@;-@GW|}gHDr=M+_Df@Gh9fYg2Ao@rF{FjEBhKy#>v-Wj`jKh`0V`Qd5X$XPA7p
z!vXcY<OyjfII6eVpKZSdu53cz%htC7%=(y@P8X&XgIh?THe_0z)2$x;f`WHGa#%UX
z^kubj`?*@w$cx;nQ)eew(h{y_$AIGwDO>bkx9~B_z+8-G*JATgge8la0m10GM~E|p
z4n3AUmgtyk%NdHfVjvwpwj%_QqnjLacVa(HW`cvoEb5Pgap`lfM|1ATlW34Eb8WEr
zaIa>9>PWNt6<RQ3@&5v=-;h5{z=aVP{#b-(YiN$PtSw$g!(zffoq_3F0Y6}_><wxt
zrCiY>*o9f<g4l&M(f*JRYNCBvWn<Kf;3Tt5af9i^$KA&A$YKFsUk)~!$WJS<3lHTw
z)Qm!b%4i;tZOqn=O1iR!stc$%qThZ0KU}>7a3$XtH5z;3iEZ1qF-azv*c038*fu6M
zC!W~0ZQJIT-~X%ks@|(p)ph$;*QvgzyYAk5t##JU9`q+@s?ANcbnMWE%uBzP%tyi|
zte757_dUYFXVM;UBv0r1+cH~Mt5yooQhcJ0sq<UGDucZaeU&A6cw6?qwSicaLFb?&
zBh}GY`aM=BRj%WabS5Fq|F0sb&os;-u8}$i28q-vnTn7?a-I&wGAH2cXEoCS8sPNc
zsv6)?_cacn!R%9#5zGrhM&IgBSi;$-LfH@sYM^j<sx`pB#SvBb0m=5~BMU+iFTbr!
zLro^Ge)O>XK2DA{nW)ubp>q_V0}&Q?42IB0YkS{=$oLM{ZLU(+gI?MN_F8$T@5fJ6
zUEQT?^W7C)o+&1W4nIQuzdL}1wzTq3gW1~d&<9ODqd<r6GA@FJ@2Uoolm627Bfxk4
zOi`i(5iKwvih|S!ru_QBm_i{8KvN?v)tf=iu;w6=orHy8Wm^igc#@rlZ8EA0POCv6
z$aOUOK!>Mt{QM%3Fzk)xL0qI0)7y+_B&GF7pP(Qk0s^jQaY>-eyZ3ku3{(0lFkbMw
zJPS{x`LU(dVXubd<Fn!2lF3C|Gm6wXvu~On{=<rNwYdNWZ@1beTjs?>8!NKT*^)&C
zS`e;jtZR{>RJf+QQt72S=k{EVMZ5>*5rrG5&a<<=z`kyxA{k|j8<5ykjYgdC3#>al
zyyRUP`rz}f(qp6PsZl52Yvo7$*&i)t^`n&`Aw-ijp!vAc=aif6`f=1?+0|YfI8~Ut
zIE@M!nBxE%$k>0t)PAPzjp+I;XZR7==ZGR2qZ%`vKo#5Q0r)XbkmGEsSQ9?=OhWZk
zB)+#txjjo_w;0NICeC{SObolGKcBf7bq%^OLQ48o8~orq2t)`vBOf+a#AtX#=t@7l
z>^KunYeX+^$t#s7Sw!NiueqxA;>9%#^4Y_<cmTRg%9&hVb4h@7oiS9D@&O;e<Etrn
ze8F4WWU<h7b)0PHA!Trqo_%XX-wb;QubbD3yzo!0L7#zwg{9W{v$VmiXGXQo-oH(W
zvdgAppa;{*wD<o#j|9)$owo(fZxIK!*zi|h9r}vXPOma#VVY}WV@#6|#s;3fcYj1h
z0LrCmvgl6;)le7?GIRnl;7@y!u+Ue-A=2B&I+4j!aD)^}e4!>r)P7^~27x|o_nzCI
z7NiFlm4gkghe7`ZVDPM3X+LI?FFM0!`QH$7vk*DM7LO<)Vng?sXH~qAg0CS}9*u0u
z^FOsp4n~msfB~AZv660@vGw)~Rh}Gx-re;tk(5~w(lmw`Gec@c_9^#()0YzMG>QL)
zG3z+q);&e9tw7Yel>!tT=5Z*87OkmO?@LSkB|Hq>hp8$?>ZtsEZ9q~(OyT;!5`+<w
zGx9>-$bn-hWuZXNr%6FkRf1;r6c3hId7&)4q(&qz65o?!p3Yn3xtl6w7br~pak^fW
zmPz>#x^~^dsg4p!0JjMOrc5^Wn=j%N7jp3E_v$-^Dtsvn#Z_%?{7&2*8wbx~(XR)x
zu>z>;@6_hZ^nTk*ALKl`=Xi_eE*6vx8@-~Dr(*WP)zUYP!WwkstvEW2x8bG5e-~)h
z7lNE%DrR%>)1!ZzPdIy2bO3q|IW^!<oK#RmU>I2sbk>7Vk%^b(r=#Z{r4LWYQq+4F
zTA1a7iZ~4lb;%SxipK~Tr#WNnF%$eZb;s`ypUbZp2YqqI>z&HG8|%KU2;X{q&tW|S
zLe_S<)|)U_d4C(3_7IEuIcrKql3CToeEO8jpep(?(Zia=;kV9OUjufk-p>;iV}$vp
z%48jNt`}zZPk}d{19wV$!wH=>f)@+U%{nq{-UO2z7|`{`m;vQ5q<oIW-bwAhIH&7D
z&La(CZN%8<0t#ArZeaICDjW>#<uGQfE1X&dLJ-V3C%~bKXVfS?yw~{8RQavhW$Qwa
zdv&F?QAW9<f)l!(TmjDGKkhPzX$}cpQInAeGG@qllR{%gT{f?XCo;z`^_kqDv7ZU-
zCywC;IN!2xf?f7LkrcVi^P-V+mm!JDir77W<JWAR4WC^x_u5I51W@7$_h#KD4Eei0
z%H$PpK+pUH%$?>96W>Prd|af~-sP-ioE+P+98CmaM^ZH&>j2YMDgRk%zq@<P^=bM9
zKKZa$i#W=aSqen*3)aCOMxW}TW;e&kgYzBJ0tS~RjPcad8&1VSY;}rqqb4yao*X`F
z+!E7O$&<44%c6ASu=&)Vv9;|lT`KOlkDw2p$83#t>{u`lf8$gpRu^kn3@xs^u_TkM
z_k=&aP|18#p@FJvH-j(f7wX?05)n)Cj@DEcRsEL3c)5rxvPeFnAxC<P3T_7aeUD`8
zn-js{ABpOqa&T7uGN+%LCXDLAeB5o?ky4i<eyY0K+siW`Vpa`KnzjUd`gNjVCt$_j
z@!$1OGB>)9*T7|BmgsY%(4_^C<09s?x8YkBG~BfNdI6VX2!RVLAGOJm^V^3r;uk-I
zyIn%%M)l4k=<xD0PtFEwnsXU6X+v7+rx1t$vPRj#y?^m(E;{-@R!em-*|r|c7a^gT
z@zOoqUV3s%KNRfEEYeB7g%pi2v3)NodTwgfnXKW_Y$do#j2HZ1qW833FrsQ&-i-8`
z)m8baYY$M<Kfh?1ZoKIp1_mv+px|TVGCf8^7A=PatDwic-@4O7q8e=ap|l4-Kp=*<
zTO@ZA#XtLJx?TCs3ZY~%e1^ycWPsAxm$E-LJx1byvrC}51<rNRiOw*jOw7UkT)2h`
zE{0KS;~($gbT*v*kVd!Bum9$9G>${El*^d`>j&hYPr(n+b@_Q-^04>_eeBf6oIfH`
ztK%jq(`He(RokTEf!$(Zj&(0Uqddpo#?SBt6v)23=!2*wXzzcdniZAfvT;s+hTe(#
zMOvRy8^*%0RdY~H$o349QqO~_<lI}Nr=_G$(;=(Su8;~&F=FTMZ-dqOUHHHp*Jf^I
z?Ev5Rac$Dxcze{Zl3YnvBX#<1)(%3PNQkAajz)I=-RL=oV8Aq3k5Y(QC~G;I1Llfr
zZCiv=BL`}U>8Y{l>1mmf#|?7pQ4-?R<Q_F7riM?uNTQtiC$*pp)xK1@Fe+OSX)iE8
zbEa1|W)s_^ppb;+4Y^!^%T7;2-ByMq1wb*^AWJM|8KNdy+Kf3{SFYXQl8T3aJg2|o
zBJv;nU(HenUHKt|NJqyEJ_)xn6YF7RP}CxZx)|w?&DU^Aq{=N~D&DuHK2jvtkReIY
z3^-a<gjG#Wb=TscTeap=4nx(X8y(vRB<k0>8=nAJP$clTv%C8BA>9V~Cby}Vcfing
z2dvG#hOE0$3rinan)^|^%FSm&Sc7SR>rq0@IKy@hR7XWY3yLriBzhEwm2L1`xlYk~
z<+JRY`E1b&=XE!}^6a@sA%x*xspV;Yp8R^4wf`(t2A&NgNlQp?-~kHq-e11(@n2}J
ze7}yfo8YGi5@?bf>QHAZXwb)~^MFp*f~7i!9gshw=@fltUif`aIGfEX0Z<W5i#i!J
z<N)>#ecF2=L4k=XH~*#}Cu%b)RH=zKuKB}1UiCVi2Zi>@RADm_6<p^2H#gh+vXSLy
z_tK&#Jd&=}A9UNFVQsZPtLTvULdO`GH=pezg3#`tul9MIxblk1n_El{ZU8c;{f)nm
zo6>AE|FmaVE5fj>a!`tkn;9Agp*+VVaqo_WIQzLEJ`_R_tStL<k(02x4=EUU;V$_j
zXf>Y4L0QU>wFDRhT<jAnjPtJXX$rL&l4#s)E-~byJQ`qM*F2I9?3>A=3w~N8mx-`l
zxaaVyeS1EH^77PH@FuGKy#l;S*nIwu&<Jj6hNmbiHAPz_5shFQNH);m#iEtTgiZWW
zq~XzT1b3-4kV+!G@>f?~;cwmQ?d>{;<bZl-f4v5%1hr~vU@-Vta914XQL5hp+Jko4
z8x90=fuV+(6S4A7eQb_DW$K>ah=?!6hPsr1RcM#az(#M)s>QwHFE5nieL}H;<W`g$
zY7rl+*w^>Hh>4{s9S3P1q-??brcDxlq`2GAWifZ!Djcd&GB3?iRmHkhd(XwLGTFg?
zzk$ocE^9J$RY`j2pT5`B#qNb1QI<VaZ>|G`1y?ag1Mu=WPqP?&uaFg3%S+T?<29lA
ziCuVF*>DIM<b@oAUrmm^&>63UA2sc@<5BZESw9(b2=O2fsA}_c(b<xtxm0tB&Lm-8
ze|ju+FVRV_&}eH>_g|~TrL^nDCxktZ*XxeYE1V47N)n$G_h(siD_r9a1}pXI)lWZB
z76cg;Kq_7NNLS%zU;ZTXB_;FCt&GZ!X!(VA1SKxRWhhh4K3E9w<!ndciNOseI;TpA
zovPN2?<_6I|AZYRhN(h$FXdWT`z?5rK(oZM5e0%duvW?sE(i691g8)(*@QV}|1N~X
zsW%hE)gRmrbA90Q_#))=M1!`zy_mTyINY@$k$6X!(Qa*(<_(3^i9^Z2eY=Q%f@)dq
zfXnk(#`>NU$N2yV8?mqqchwHB#;oAJZeR*LCwcP+=#kF$b6#Ek@vyzgI)Hym!?w?F
z(<!gFEQTGPBzub3q+fLrg`ZGxaGxm-7MRgdU?eCtYh*WAny~5i#N4Drq{NdT4`z@$
zN~-C+rztr!mq{+q<u-5ojtP&6)$Oh~(K+8RutTqye5whMvK)-VV?o}e^j*?~s=wQ2
zRe9e!-Euk_kZq89bXNeZBYnGxVcH}2sb^Y=p>k4D)~Di^DiU^71V-pI@nMo1XnkoB
zqQr_=fsV>FV=ewqwF9KpThQw+YeBC0$MpY>-(~Yxn~%L(M0H3xC)Vb&*kMsi)eJQj
zICqXjRS1D2{&6Sw6k-N2v8K$cXm=5rF)^_$KKaykLlr}Z!k*7tW3zVK?n5_RISRpG
zv<x`a?t<vzhOBx@Bbh8S#SAX^%|ShUtd+CJuReMhD(uPQ_2~!7AGkyCi+pF92RF(v
zuZd5PMx)csQxllO@3(!_)rE2D)+5d&vpM`mBmzL8JF5J3^j6x+npHuB=ON*5o+VFk
zV7v%?V<G{FI~Yrt-tq2JRJMGuBhh5PFEwTQ(Uj~=(f<3Ie(1;L?~*O5w}pwM8F%(l
zA=Pk<%p7XrK>wo(6z6-C*=YeK$63p$#f8DaUK9>*o}w(16V``Xw8nL>K&Tg0;hd9|
z#7Urq)XRo80k_Cfj`3-uxPuA%7h}C1nX8}ERUg&-yRLQmE|YMdz@$kOWwHX?UY{}a
z(Wh9dc*hS^QZi=Gzcwt)heu=mZ0iZG5qN?`Yh?|S69%KXg$(kGi{uZKU3aZYP-WNN
zhG_{}Liv$4sR=pQa+Hi_LRCEL`_N}%0gOPHnc=#{Pn?DV(`Ry$e2nn%E}4~Yi^E`b
zWQ=QWWz2EAyssn#`|q>4Mo}?`Y(+f<%t5x3Tc(ZZgJ0!|eTkWEpH8nhs}eH$zxI-G
zpYBklpEjnj(h?N1CT4c*4joPSn0nv^)wyK*cc#%z`B~=ToAuL6Ta`Z`4LY^-46Ok0
zL?|mKL4GeK-bbvpCGexK1$E&(ES!~4PN<3(7HnTtl&<J((=in7CDmrq19@;zznL#8
zKQ2mfymO=)vw$0|8PmI`XTDTLR~n?q;|Pk_s_oNy-oYxs&=2&O96LU}Mzt7h5D%Of
zeQ{Ui>M+}x#lleO<7TAf!qnbq%?672R)<Wc7*w2pxaD4ruvy9}xk;0^d7aW6rp)wF
zw<o{cLm(*?fRYp<?d6O!CoCzSyidn3i3P$0Lxnq$EYbdR+x|`@H>ZFQ@cvuOy8Y0-
z#<--D={bnV6i2Lz728Vj43pWJiiQ7dT)6gQJY*(;QcMtv30hd3OVRRL^a}8SIfGY{
zJkE};rg;<5I3FIq4g@AXowB?F&)Q{+P&n6#vPy^fFx7o3D}HKia>Vd7!Q_5MFh4Th
z5~i}R8QD56xwkp2=(Y{yWPHflg|$k3?Id2un7<}Kt2<w9k?k*3)+HDoCA4}O6#M61
zsL)FewGSDU%#8+N#@eGa`JV*04<1f8A9lBaa<W?PJ)QGqlNzly@Z+(EUa-jm_1*Ta
z)6bU*ZZ!MePnPQtk|O>01_r387po9~nBcQedW3d-ZIh@;`;XPBfdWC4wB#TyJbk}a
zp-}GaiwTpECCYoo5y-{S&|yaZap2|jhMHgLG3jM<r(<ICi0&|%4A?h0&_asMW<{J~
zNuY&PrhiYP9rq2~-ucG~&gd1$ujys&Li0gP-=^}$vQGXtJrs6QFvxuGQY4-*wNr|#
z;1N_*?r4Bu%pq<J)v>SLhmz#GqgwgmB=I_u?g=|NwlS@@Ka(XGzT#NA{z9K8<hVQ=
zr^-@1-Y+!NC5F^_f54(OXD?C1;le}9B<)v5;mfk~cjUTo^K9S#5^cMo6fXfNFDpx{
zWs*F2%#u)3`K`V5(7hBl5TKG4%+6v|w#dK7)%@F#5G&rvZ&|l-X9dO?l0Ax$-R7Jv
zSGW|-@?=q7I@y85XA<h~vZ^e3>Bq4r+2p%pg`<R*!Vjb`hybo)Ms`%^kjcs5g9!Cs
zy=8`|>h`Hz(D3H=AHspWy4m*|w{n5whu~uuQxVJ5lj*(Im6Se|Lxm4goX<h2<0&RT
z-taS?5~`i!{MQ9*qmUaJM7D!^XcO6ssM-opbz#5GkDbY8EZtMKiY%n2vju7LgAAX$
zo+_e-a%_zP0PYpBM@DqBM!S{(VVISKdy#(Ul!WD2jA__`jLhmAlkV7W%(>!hyIbTW
z5eeubrdIBo4A`6*$y#?0-%`{bd_4p@217iZBhz9YyeeX_Bk3I>nrmz?R&id%NkGmg
zC=$P`Q{r&#h2hZVCr#lAIqC<1C{0*66KY_MQ4Zr_0QJzdQ89usy4%)UruZ0c+L`TF
z#?6y%DWJ&MJ33JMZeiWE1he1Ye$J=+HRvmF1GkRKlz;)plp-aGQ!URwEAB;;curxA
zGs0P-eCHd&%Y=`kx^qI|I{5LbaWVw%U_K5y%<%*(BV!Fwiswj@XA4&D!|~$>O8Cbt
z^UKvWpymxXyx=i{j4SY}ySJkxqu<YF3Z5lDQjtKO%L{`tA0@(UL+hU4cowr=GR$=8
z=c!^CdlAL?XB0Hie~HB$*z=tosZ-_Sh?iT*W9t*{Wh+t{j51wE!(_SWMfQ91bymy8
z>6qAb(a$4!g{1Z{>sxkn{^<@Ycloo{kfMGdfMRUoz&~_#(@^P>!f3OTelg?Q7Ftxo
z&wn4!LL_clj;SX9gpRG5gOWLE4j5zO=8Gfx`2McmnWbvx57zYe7r6}qS>Hy!lZ|$q
z1l@Srw3vq4k~UL%a~bRLUmyj-nrT=iMHYJP9Q5(bd<%GVhqLf)_strlc&*ScXl~p9
zGX8d9Gncy>!2WbBTJ`DV#NyL9ioIsezNY1ht;=t368M-Z61BO9%knhY?2qX9BS>MW
z$(Df6xS;v!%Ei`OzMVxO>QfOA_RNLnQT#asOM%Iaf{7I(_4{;eXs#v_Gsa)%;H{RP
zg?}a^zQ&p#+o`W)!Q1o_Td67`yuj!hu*m*}EWqHctTdsNZ8c4~@!XYLg+$Vzj_1hh
zAzbX{(vW>`lpJ}={kr!4c*1^!doJ&0wQU<Y6Z+Jo{*71mdvwXJndqOx=XbFnX1{>f
zj29LAiss2xRq~DRTx<%rCoGfx<P)h6cin79Yi0b*<U9s0M9{aY8E}qDiP^`E!0#L7
z-`CN8zp*{~-_KavZMAbCRX=^q1NHnGpT5i5;B`_V=j%P{%+zReCkgTN<PIVgly3++
z@p@`%ol88otqJwRpm~IJFcr_j<O#$7nLKxAi`bLUBMoL?(1sW><}Qw9+X+|Z741_I
ze^bn(YHwPXsF+VuYi9TnP{WuJ4OC#RFs>|EPJ8)V;^;nwIW;1M@L4h@@?A^Ze>_M5
zBw*f|1BwOt&TxB!xQ;@ppu?X#DT3{MJgm<fBX8GiP32p$Yi|@(pEp~$v+Z%dXVGT_
zoUPyl4sOi5mfaUad6Hv!a2CeH?6o9pwf;ER%Q`kLl2MU#aDT-Au@-C60_v8r7q|Q&
z9Q6TpAMl+e8G1`~XB{>c44Gd{2i@X+A@`Z{B97;MVsuIHZ94O{UD4<)8TYlL`+`(@
zT?%JpJfPNAGNxO9zg_o8<IsAx_CbQ(Iy!dlZ~k0P5O6VZ?(GBWhW5{~+pmp*`!8<#
zu8aG7o&2j_1F`kW&sPcopebNN&~o0w`g`n`DT>d=JMEMXH^)3+*P+u^Nmws=mwxf|
zEuuYySO*u!*zA}0&J^+A!}-9J3qCEhe8w`0!2G{yvn2xb%1V$Utwy)TzB7#4s{{38
z4fEh3Vqo8QEfeNH7&F*MG>uYSN0pGyEJRCFXsd%PCt2Le4KJPk0;h)!vpAn*ycsCg
z*NY5_^|~?C%bFlluA+@Zk)o3-6FRS`>u7#0)K&6K#r;m)mpV);1EfP!6dEOmN|U0q
zx8VpIvVr+|Zp)CfmUA8J=E4NYe?lOIr3Qw~%72^74jPoxGOwOcNh(F>;jMcf`?UUC
z28m-Yy4ajs)ZVxx2Y!}?6E^n#s!Y<fz6wB2L9iDDrRI4V&l7>voukCY9t!>ItIYDp
zO|?`4?+e4Wo20GdsZgKR3s9;ctV8fv<LXP}sT!QiW6ZbXJl2Yd78ectrI&GD!igi(
zz7ZJ6vlR#2`oZlfGiY$c-fCicb7`C#@A&Q#H2GSYGt+q>0q!(53zf;R%D-nkO%V%u
zQ3^rCZ~9Ldos2bT(7K9deyxcK3uwEEYue{ryWq##cNcAh!rzR#S}<%1xk@Rr{^oyh
zqP$v2d3d!yd<l3kuROV5>v{=r-ReVH+WdLsKGJNdU{H5zsR6`B!e|iDW5GX^ig@Nn
zV+!6$K8M5k0GHV<+|>`?e_Y=ui<C)a(=^dtl^8Z`n#S>QB~r6;reOYKN+rOp`~KV>
z9|0e58W2-NBT@~mbJ91X`uSlutGc>pnSU=BrFf##;ybz&->^cbBPvriycqAywTC$)
zmbAFNa*|8On@ZQ}b5ws}Ys2fh*W&paR8Md*KRC(x2s~!b4_+8uk-+hwJ55326D|`4
z<LuiFweR&<i>DY2{usVKs2FbzFCJSCy%UZ-Ankb&T2Zv;Olrv^dCNFM|Mbe4hRE(O
zz@BraR)$g@fHUFZk3tD<=QNp1w8$q05g&ocNBHytQ64}QhWG@nf#ktyot0!pHE>4j
z>J}!w0I(vPvJALiPc0gh7)!66$YaSKusj4}Lq*;<2U7lKKMs{XnQNG4{$bzE4Hbp7
zN~Wc#l08$jD;SOo2kg53QJNirtnh8~pK}GODvR~HD*i9MPGB7z=yBuj?8>|<i|(Hx
z1j1{~Iz|SASCe<?2=!+$f)PNsTuTm=8|DBEobN~qQ1fk;Bxdd@4KK&P8X$hBs3ST@
zAQrxBb1VAXTo&nscSh_e_*_K2n`I>PkEqHPdilwv_z3u(gqiO`?lf`iG8AuXNAwtn
zSX)(sfe3W5Y5<kGbMRu%nQ#4ez)PGH&IkMmMjL$KyW%hLOhHKaH>g>ETL*Aq2uh&j
zO_M<{@hkH~oQRD<eGVJb2vr}bwC`hie$pf<FTO8rUNwIw^IGn2Rc{ubR$>qf9gZaI
z<-u&fu1ssoZ}>O;_IH01-Ya}{L}lOAXVrWg2hKA<d1n%%N~rqxK5FNF%}oa!(D)%>
zvAVNaxaHsJIu!22M};{aXnPaY%L!0o_fr-$6#t3cqEWx3=&^Av7#-Klj~^M26br+=
z!Z)8agp+f!iif8@u96-tRx1`qNu~1H{kSAIrg)iTHv5Y<Bomf0fDbO!u?mJw6Sql!
zVrrj*YkBKH){nrTxfnI+ieiAq{lkAuxp9C`_Cf-#9<FaHJKShIsw6KqxD{A1VkKmL
zuem0BUVe(u&V`oFf9m3_$Lrsmm?@NbenD5}&xfg$hFbD;pj!o-Mft&b`kN-_SHEXJ
zWbX8B<3d(!W!K|jY+gJ=sz#hZO)at12mUff6GC69HiMJ+si2&fxI?>fiS49)x5?E0
z*&6df)0?4amw^XGoSGvdRx>a-<Sl`PK>CQ){PhWpY8b0|lea7i@)VwO3~5YnZ{%6O
z1XpyOx{fEfg?Wx1h2QAEQ>>E>4+L{<I^R(<<>0=r?|Bt>@ngPgf4M&#N8yK+>3n<p
z6vvGMdb+qECVIQLpaz(kvciu-z)_RyFA3iUB^<o$-@4w!gR=l6Jl!Xu6^q;L+2V|e
zV%G$SA|K!Un9^)^80PzgZ{!WcKRZxK2m2-;<o!>1$V=QScN4Al@TmiRrw{0GiPNlf
z<f`!Szlng;(me*{_#rfn#D^ek>LztJW%=-<S``j6+YYF4G`C2UQK=H!%+h|3wI~ig
zXwtEGDE(g)0{(9pI?~laEl`PFO>7)vtRG>#An8<Y)EU%?coVX9$Rt@XRMJoOLBAi!
zO;Ru6ueC`HMjsf4UQupdbAjl-Uv5uqWMPP3ojaYqMJ*7!R{s9S%ItdejS`{msD+E$
z%4+I>oihSk?UE`|bP5qMovI;xWLQSo5E8R+e+1wfF#elvjWIaJ8igED=}_st9_!oN
zL%}V)Ij_j8#w9~au~-Fo^GW&T<B{9w4;gspm{ZkH4QsI+2ll&P?a@A%a7}O>$B|p1
zo{q-OCoDHZiU#H-<xfK#$p=GRiSS4a6{VBpS4|1^=1G{_#Et4a6^UUo^(3T9>ODg#
z>H<{_p(#kU_x^GZb_+04A{>1?2)>b2s*4}<JP3I5nf6z9Z0vZ$l&|`b*v!1OP|7!;
zq1J^&rK}ZMe!i$`lS4ZEb+#t37;zOcG-nBk{of3zp}sm07h=7xe6K#nmSwI+2mkLt
zS|0ypV1isvXScF)gd0eA8*4Bhgy6>q4D)*1TcSoVS7U*LnMp2DgG0*_!<)+31S7@8
zBXL$BfFKg7L4k;%(1Hu)1SWxMU?Y2)_~PBxbs=}pLkqaI4MIo5>z%CYLuX0ex9NaO
zg31o!FhT}qAw}v~C4K`fMCUGia<rO30bs4JEym{6ssHJy3WHzBIEDvbG&RQo2M=Fz
z`Ee$PJL`%xZvr7-c#KJJJz1#l|I%-s^#J<MEaf$tFnRPLX$IitV>rW%j10noMFPxZ
zkT4|IH(%`^E*ywt<;q=5n9VHY%Mt(V1}9Wly@<gv7corGe-Pn3v}L1;8fK?joU_Rf
z$0LD5iYgl}OiXYIDU@LSc5h(Cic*y^y$?|(yoa*Ji9(7X2qErUJQ(&b?-4X|zY6V|
zKSJG!lqo6Z<D@Si_kZE5-#TYie^Bdsgq^0&D9K3=&}sAFXpJGcyt@Tm-e;GrUs$KC
zIdwrWiS!REixtJH*#`6pw-V{xUm|hpG<?I#O2ZiSi@R<;fT<$nseA=8RuP+<N%d?l
zE^K(RxoFovE8-cHct4wtH(WYOGCl3~+>%_#NUbwLCd|WVeU3DDwTWH<R)CdJ{p0Gl
zT+YNCBw7*fW-=o(fd)NP>161!vqipF@*I5yI;uZmQ(AF`7@k2^6NY=9m?sO>XYs-%
zT1dwaE!Lq>o@kB&L{DJdyKv4p+mn-({?|i|9ab=;@;sLM=Q`T$Y+brXk#8^FZg1U_
zU8J9oV17PL1FM@MQG<RBAP{>bRu__h!oGe<9`)kBUk}C@Q|(kQ9Ulw`Fu8P73vu3)
z{3SGkL2uuxxUAMPVvzZx{h>I@*qLD$D}g#gB^?VsxIzh1Y6(msTM?caPNt^{$J%h0
zydmN$Q!i@LFNsanAw<AC+G`go*Zri2lx!U|U<@hbD1-mp^0l?iM;fh&&ST=j5R>WC
z`6H%iJlq)mC?*ogU`%0Lgq1d2Xv^S$_BB-S*2u8J9wM~G!$!v>hkqJ(bpRL~C$q)H
zu$1TUEcRyEz!&5BK184p{I;EM`71cUD%7e86IXY7>jcbu*XoEVe6$iLwEY=Pkn*nh
zf8|pm<y&_+)Y<+)jQ`^;<=c1$R$GdeC;1bV5yUkSqwTMI)E#N49-1iS*Oo-6b&9k=
zG$+uHf$=sH!&butaMq1x(gv+1>8|f9+87+l>T&=9ZhEgRnqO(mH|PONN)Kf&C<Xsf
z(d}2}2E&+2%~|<RM0etz8R$XN9)G`4{Wp)(eFuAT+ZNenN{7Te)yC6bH+9#Tu+U!K
zv0AlH?#~H}+agY!Bp>rAe|5^LheJs`sh?|R`-*68&JA02Dph5d+jD9=<dx7F(`x&1
zKF5IA#h@co4T`9XVrX3gg$};8^@JiBoPzZx+t^c^-3S?*ztX!|UTU~!>_PQ5f`VSh
zYv)QS=hyo86*FaQ`ZDXgNsYGVjb)+c^GYbfRRXc$ohjj>ljN^`HSAq<!5(cHEK{W*
zWHTFj-(#+-u6QJF>O?27hM^%BGx`kPFh_t8bh>)ABYw-q1xIS&Ks{bBzV<4=ZNM38
zxpzp{%=i*I-jg;#@bn-%MS^6%Mk(;mUA5&;##PBLE8}Y^b*Z)^!`8FErG<4oAebjt
zrQL~%%yvQx{+?IJ7Eel&GuOLuB1rl$4xMQ7lI(q09(_W(IQ7UL49k?r2z&VML@@vk
zBDqqQSP8BlJJ=>Y#hgoX6&q+MtT6r-UG_JMOv>W}W<fy$ilLW9Fw%#Hk9{U3_)VXm
z_PWg;njtwY9D^JI)mU5}?K}ql3;*VSo3`MYeJ%0>7<n~)oTej7UgHLD7#&baWzh_m
z)SF!9y<L;4o*mikIHph9wrSpO2nF!+?nPo-{jif~4v3P$O1eVCKBG3RjtL@21+j0P
zJ5$Mc=(G@BsW1K7Tx++9OZ&qi1{ZaM`s82J3~XxNBoU3$0#wlvQWSF0(JtS#laroR
z>36y5Vo~_q!+3A7Zgl>lqkNUSTp!e;qfRY3{qPTD3LUX<pKxjqxoN`SaRs30YnGO~
zuKzA{X7j>~cJ6NTtI;=jB57mk9p%GWr%p%f4ITQGO_QW@&m(Ps_xQe-H@+eAA4@fR
z_J(@PanoUmox_%dKq393$JlI-+b=OD^+pb(ij3Y)5Q*pV&1DtE9A7}0PbB<HnPUla
zT_j4~Y^)NN$q}aP_yYupM<J9p){Cuyf*zgE#DlRS`pc*f55w~hpV<gj7V3dl)aS>r
zR%p$yU=5a7MEbxi*#}q%Sturl@GBYkzXC&Xgr)|~Uy4vDMWniKbRA*B+dJQS&yd+R
zn6y{Rv}?a}L~7X*_3ZO>>Wi~fumC9N(8?x4P?AY}w!gz$l+8H+m9(qKU|X`p4ee%!
z5ssR?xd~;HT_a~HX4OnOZTld|m@Aiyn}c|DxMQu8+O>_KLB&dmVn<rk;A~q!sdew;
zP$I;5XMw?@BQO^_(H~sl|1SKnCLALF${og-eLp&0PsEi-kSlj@I3^)eFHIDjPLa-7
zwP{EWb1xhmP-Qd&R}G8o*y~%}vt~zQ3WDb!dgj%T3EH*q$NUYCb<mKBk9?$`&92t2
z|9koX#W=Ru{+0!uGbb3@Vn`1R+M<1y8BYrj2Yyu8sgoJ;op!AvQEJn`m}&Txp()z-
zIA2@0lKupp<z6IrT|;G7`TPubLFlK}wEb}t*OZ$nAi;ped<y|n&c>p+G0m2=;#?Yz
z<B|}3KM9R+OlA)#nKCm#cZ3X^=8FDd(YZ{S_m|iSnY8I|%g(ITj)x(cBkKbFvmd{}
z!fJ=j#L7*6CYhdI<Q1@J{sPBaS|r^6-)p$jrBHTsZiGngGsYYekMA&m2S?RJ0teac
z?wJa_gG%_OzPopyA*}?UcpS7Z6LS4=XoG{fRZIN`Cc5x-pw@}uwuOWypY!XIMq;%|
zE$@oNHC<!c4aVk-tkxg+#%e=!#xd4pFhK9zRAN1voDL^h27*;kV;zN^WDlQfhoE87
zCQGn!s*v)Q!mH*OSAI52w7eTTt68a9-Uo=nu(-%SOhThOLlj&?srS*k-(;WX;o~sj
zioFgD-1=6>!<Wppju&sd!ORxE7RYu_Un>>Kk}JX6DS4<!<<C;6m?XQlYDu*@69@nl
zoQq-_mA7MS<LgQ}E6wMjXuc5)9MNl+XTB%Wa@-`}LsaaKnvJu$s`tr*<qV~`JpkPH
zsuat64%8kBk*i5)L}HkCkOi()Wv~xYSCDtUg6w=WjHeS_gWZ4Bv+S*D{g2vsOJi|-
zpnpE$Hj7HXds@w>+d!4pgz7SHHL-SkYEli8C+u}6(N)X&G#s;}rNBST_{1~2ZM0Su
z=)ZXrMh6Al`dqu(?Nzw6+Vv|+9ROW1orE#_TrrA(dlTCBDU><q1@2zU>Qi3)s^v@*
zW%m4lPFaB8&csIi!ueqcV<s=c`Gsq!kT_GYOI&zwDc65593Iw+xFA#T6oVn6L=^Uz
z?a}{FqA*Xg+#oY#QfrfVdMJ;iN+hxR0onPXyG!IM4dDIWs^6MH|JIzkDA2jxkeuxn
zVaMFPU0*4xefg`dAxrkTE>SZf7cW*0@yM=UNbIwoQ0hXS7+4q=L{s?kmg=^(yV_S_
z<P|Irod`6mC@Ly>{L+7s8vp9usKq8*HQKNv){b28%(b4m?!6bWH71$JY&`4da-}Ko
zGPbu_0)FEC7Y(8#`<rV#NKE1$^eng%#qR$g+NVx+hYHJ;oIB{4gCV1tr*x8AB-)f4
zd^B{0zCX@q_J~<8>hh1a(yA+-BL7@G$WjuHMawF6xxql68pB86o@&uNt%OMIYV2Eo
zf2JV*2b#=W{u907j~Qk^8Iu071qdI;Ed#4DP<8nor^6h#qqHZY0S|_;2bZvI6p~xz
zdB#;E77bGgoDxzR-l4$g7u}FBcWlW(04-D=yhD*jY{0d{76moNo-uH(1?OPQDpgG+
zX=Uje0V|7LO+@)@bYR)<|A(#Q|306u%OSa^A-RbW(Ri?!Lt*KLq@9*uzwgP_07TI6
z4v4*`lqz&X!(Va0ACGQmdiqL-YXR~WYQR5eJA<Z8qlTEXjY|S`TinekfOCO5mwACg
zY8msJWg=oG#j0{seao!LQB{)i`3oMhlYdggZn5E}=4e4zk<2qil9tg02KOL{N=ug2
z7}vc6WwZd>eYdpVpVVZnKc8&OKxf)Qd2@UL6w;0=oO6^JBbXtm6LdlFq9e3v{x+#q
zo<#Gq87Qi$t$0Rz5JqNAZt*YO|ANl5yWjah_f=PJJS6p9p;1C!?jd*x##BlcZ2qmq
z=^c9fMk<D9<33lvZDeKR<+n_8ry`XA*cMj$3aktMPw02kyNpLB*sUo5{AJv;3l}d)
zZ)}#T;jiqJNn^?7-8n^5kEhMGb;_!Peq(pp_(_!jyZSZ`u97s7nUHmRGE&N41j~{~
zHc^&}nWJdm6Ney~lOzW%?oqm=3Rr6Qx&4;6byXcxygKKJ2R2d-_tx%8pLwe~u14os
zo$|tvDD|JY?R(*9c%GmD6ojgeA&yD@h^@NiyfQwr_h|r3Q0iN>;pl}`$6N*p8f6(p
z4otq?XG#D`)pD_ChUUl~ZSYPlsVFl#v<&)kn`bIL=1;6h@Icjr0Q(XqBt`NPCM5$?
zF#|+v>o@JW@f{(~1(_p#(qIoXWVI4bq#RD>JV=r}NL;q%9-VFA_18?kFw<D46qx$r
znTyJCn~pgfk?NnvW7+8`e9nJrnDD=mP{`Z2#5`B5xoQ+uTb^uAZvVQpYORcZn)=V9
zUD{WYO&{ywt`A_|>z;$Dxx1e5H+SC8x!!JPIoU|(I|H{!NnR%$9Uoq^B-ZzYob*ZW
z&VL&(=3OXw4vXo5@o4pMgF~-^9XlFC{**H`$pM~2qmW^`5{JK;9oiD`5;0635HucS
zur(yArapq|w-kk|emy@nK_7^8N{|@P<Dsd$m%5m{sjEotE5fkF$T11tk`gz+H4es(
zG&?LP$AJ<3@WuSxs{Y$o=tvm^X+ia8ul;Eg|9PeTc&*$TAgJy5{6$_FVxxY(2CdY+
z_Ac02#+Q7qnQ)!Dg7Q-Dq9a%-#I*AkW-gcUK;goWT?7*&d-_$53K+u0O*VsY)D@fE
zod4S*4lhD5YGX*e14SAgYf--8+zUWv_{>Fan||Hi<Gkfeh{#*Uy`hcFs0R(`Zc99$
zedH9m+3Rrz3Jb~%8TR80I(F9;KfUzN{8hSc_E1CtnrQY%_s`%`X4Fu!Ld-5$<Lk!m
zD)v|G&An)Ymi3krkzQ7P`)1@d&E8GV1*^(K<qx_zfBUDr%TVzRg6GaX*F>`qMzj(B
z{_6t)=Ofe~9`P^H8UFF+L4=_LhsVb*aUI9er*v3A6RlkAh3;<AP*iF>QeUo7(MMAr
zX_a_ArtBr!V_LirjYa@?Ue+6|6_wvqDqVa8h-otZrg5>o8;Tw`J+*n*Q&Uh#0f|rN
zjz@E%$kTAWZHvX}N8_vFO@sX>D{uINe0fc52dFpH;0~nCl;2mG5yHs#NHK4d)_Uqv
zx?Bu!oCEX0_2w$)8BaN$^Bt;}Qil&&C7+&-u7E3X&#V8kfus!w8q4h`eecvj^wkp(
zfd>jZkvJ~&lf*eFI$Znds+&I2n86gUvY*mYhSDzrC#P4?Ue?H+iX8lxzJ{~_w-{71
zc%N6`pd1X>b?<A;oR`l6!~AJaeLw{FEfPRKk+tI_Q)pVM38%oKU<kv{wY9p-hL(mN
z)H06@zVc^6{=my_Tw!881On$_9C47EOqmTTS(%xNtRun&ZxQG4t)M6OnS8rFcraa@
z`hI_eu(Rz&3PS3s+pYW1ed}Z~&G+4#-wD@Qt>N~>{hV*Ojt4y|Q2insV)|A2Vh<=1
z#W{Z%e;>C#<h)Ih_~$tL%=WppBi@kvie?+<;V}4{_g0j@`3i{8+nGevWsNlKGgzFY
zEl!eoKzgMqvEd=gE-aGeK*fBH1wD_V5Rf`=|Ldw@nLIh6wI%hNwQfN+Ntsk2+tAO+
z_cAcytCwUo<ek|?f%EmXUl{)3E+r6!L%6DD%aUg$ODK4Ju?z26rB!r$q#1)2;&V1z
zJRB=JLZA_R%T;|Q`C6l>=i^0<nH2MuEo+M$6R{4%+eOOv6w=cDBlUY>4EySJMe&zQ
zXC;r8t}<tk3OU<g17S9s<Rw3g)^w|D59~pSG)cB%sr%UWzm^T8*OFKPn+_n`@vZzq
zKyXCz_V$@`5bP9X_@cc2MHf$34*y^52T?9Z$BakSM_zMc(`1V%!g@jAX0pCoTN5Yh
zTu1^Q=6GK{4BA50zG4%aBwL8nR|+<9drF~;KlAo2y1W@p(kokFOOKO$as{4TF)@83
zgZfkctDSU{Rhw=5%<(UI{r~`TMgt62q=@GN(o>h>3spjnmc%*RdMtqWj}Q1>5Gc<~
z_vzIYD;Zp1ACFs`>RV6BM#pwk^*=pFT52|R{H5`QWrLaxXE*cX4D2ATiM9YC_KR(=
zukT;%khRMo&m9Rk)RU2&cINcvuyWjq{(fH+WwBw2cJCFd-|Aod<0bOg)xy;gavc<j
zzvvZm8s|Q2{pX7~Ki4WsgC9B7X-wf960e`BQ!Z|1=n(Ix-7{l>mT-86G-^Hk|3ks}
zZo?*GZKc8C`fn^EWn%p8BIR`58y#O?mp@dQqRNw<n-H{V`G>z@4}!<8d(K>l?CAy8
z<y^z!sRat0$Z0)U=hpo!Zqf29UZKj)Ec4$o3?WMj))_0BlP3W|<4&ZMsizDGztDLX
zus?!tU=N7=kMcdKUR+nt5@>Go;yh(&MzOzQ4JK0N*jDMN4!bIXXOu`~>PZ}%X|}hC
zDy7CH!7>>6YRU_vvd3JQpTg7=h(f_lDyJQ05D35`up0(SE6n?%6loB}(f0&SLvfYC
z1V}sdMsycgz2dB;2f<I-J#j4{oV5Jp&}J$~B&QB0E*vtFk>oXyf)$l%4Ne48fG7hq
zo*u)Z1n)J(qD)ZPH26b@A;qK+?3XB7N)nkc{Ecy7xb|GgdQ<WCXQ?{c3{Tiu2ruTC
z*l*xuBU9ijQsOZup@5rV3B3|r2ZPrCH@md~$Tx7aSI8-{IeWVYe_=Kzt6*U^Zf63h
zX+uAV05dET?nIg%qV%;ZLuC3@1@xycJF@x~S?guHZjE2b1y}<kBr(+Z=YcgWkx2$M
z*uL|V?5qRBUDa3v!_m^4ps-oOyCebUK%?xqK!f>bIY)tfhQ($dG&W#BgbeOiJ4As1
zLDJUR8{yYw<r(^;>sK)yGx1n8UYlJwa~H%gf{}0BU}Cgk4;Ik?NbVz(X9z}sr~%e7
zc&`-Jae_*@fw(qZtUeI}DLHf<gfe2FU%InL;?)k;EgN%;XE&aeZDcd&AooKzFlbV!
zrh+Sl#RMCmC}DJhU#vH{6U>0zee=9^cgjWz%JuHSxF)>9+B!~E^^;~_1`_#Gob0Kr
z(4u@{<jiN`S9}`2xZa{~5-qOwv22sU<`w=|lGjU|k_(5F(4`6<{LyDN;cW^7`I#su
zQ-CWjuE%fU_IN)(O)d+4yXSQeq^_}uQ#XuWt+9|n2*Z%db4pL1Flx!Q<^yF^=W3AC
zAz>Y*C{e_AMu<|`Eprj`*OVOizs>^47XQh~$eqqlZg8Pxpjry8{ifigTQwyl*4-xp
zR%W$%_#)f&PkVnq?8cCIL_&(g(j^A0ENWM3qxT8Fim!Umrz`qkD;;hFx4*?c++tmk
zAC0<ogrg<*;&x*oT)ksw$4W~=W=Kf-I<Aodl1Do+q!`BoRpczc_AL(8)4?J?Z*GIe
zP9}UJXOlZQ=47<Uk{d)`EqQg(BfE>E3J7xub4Vl7n9spVad7`=fYF>Z?>g1tVPfA@
zPfz{Ay}3nkW1`Hj)EN(fFhuE8<fs$Th{VlXNS~MpZPe|CaUT!w1<%RnMDQCb8)Zz+
zu{9C*fSUuR8z=%SKUWfXJgSQr@ip7wPM+lR1B#-L!9W&2!qnVudJTyi_X314YEi<y
zH$(*kw)n>?iWJ1TElBAtq8?(YCHDnAEFJVA2MBWdG+(w@<Zn{Ih5>TnOYP6jHZqFW
ze~%U`e&QP66ghPB0&BLK>pC}@En)8$8!U5wH@q}J8y1D}^9{3{%3#c_o|5Dd`rLf+
zP8aI`!v|d-_ZssFyn3038x-<|cr=TJeD(?l@$?_mwc6FS9xXERD>*a;6E4mf_a0VQ
z^}=r@RR04~08hnq8uuonOH;);M!m~V>6NHgsOAffj*{|Xh+wdX|M*bvE98A1qN$o(
zRk|s!&+lP6w!4N#<q-FxJah;vn>}uD!E><?GjOH!;?q8TOe-i`(FSa#TgLHT#Bvnr
zcSaRfqp&h~w5tS)7VC|R{?MRrKK}w7N{iSa#)uFB!vJW<S@U(5i!TIzUg0#moHw|m
z`DQ~YQL|l_esFdigj>KQdXP$nZb~2BBr0iIKz2z$)KQCu3sXFssRQFH*`!nNzNpGb
za@?;dB!Lvl0Z|-uSW+Lm{PXvg@4sYI+L)%=`Sp1_qI$7EWypaXzkCU)Esc$myT9Jp
zlmgHU&W*xnW{!iAbx0<rj8e#8mi<@0;oW{-|BP4swOXzP<vz{t&_;!8nPu^UL2Agi
zidCIEd!#h7USSB|Eb1A_su`2-Wt!?Z7DZT#$6tk?yjj(!9=m4>QLM}pNV?xrR6_Rn
zU#5qJD0fq-y7a-?|37~?Kp*txq0d?#G|Q^g!(t@CkU>h@JUF$3&5n3$$Ee>0_{ldX
z!YOaAj=Mn0t#p}8q#xmEc38`MqWbu^8tC4!q=KO?JlZ6ox=l+>vZfR%4d>@4YPFpf
z;_<T~E4CU249Rp@`ke-eQdDto-hqn--Bx#W5xyeNfnU3&5}a<N0UU5RUMor76E{j$
zc{kU~>Pr1VEhkc=Ww|0AtNbQuy|uGB1Kj;_&_%kMHa}=6mn8{%B$YMWG}rWI#8oNm
z(vkW?7ol=ChY}Z|-$^v_UEP7bh9TDCPD{Fl5%pT&g&QtgOJ+ptts^jpeuXcFA?I_n
zzF^1E{9?))S8L0I0OIPB0WR0v<z;ny%jKCZGszM-6pfX2hqUX!<_g|-i<TTCl`4&#
z?jr5eipVUbT8m?*Xa=ZSrsOcPST#CFYA<Td45K&Hex4p!nRv!NN~D<;JnynQj5d0y
zu*&g$Tk9(0yasLuuNJ%*rNpm!s782b^SzTRyn_!Kf7+P>9j8RfLOtldmE)u_Fn^Ko
z+*sq;F@Cz(T6d_m@LO7UX#e|~Myb_!zQcDT19>&MP>KoK%sN_x2$I-WVSrf$go9~w
zKXGc3>&n#lmNk5-%ga3AaNMErBFQ9oL}=0gvqyLm6t25~&*?kw!fzZc#jVA8k5|4d
z5<fQH1pojHIZY~db(x@F!E?`^<fF{a7%xEYN@c}SV%<pcveShfyhjWA>UG~M>RD^3
zYoPkPi^kY$tshW!=Amwj`Ns=;y~jf&?thwlYRAmoeJ;0R6g9?Ky{miZA0{u`;|aRi
zb|-%`4K9OF7UF14l9q>U{Z_UVFBuPhw($12x*iW$jehx@oz{1hwmPE0v+`VC{t@tw
z_83lk%PlaBp2F>|wVM19PM$byw2<REndzLgSE0n+5(;f(QA##%>TdfOni&6dY9?8^
zpY->Pa)IEkwaAImdcIs!Y4-ZvS=VA^o|B@(41RF$xhkAChU{+iJHW*EU=|vqoj<r-
zck>%ylxyqpZ57|{(;Vnh75wbU8napaTy9)GwhPh9dLu{KFW6f+WW##q7~ClRpdDr%
zrt8n_I8Ptp=>Hr>@(eyqrN5~4w`85fwp*ydKj^sg&=$ZF*c<~RJ*$)64l2|hw61k#
z6x-G^h|AR5e7B$1r>t<`vm*No<LMNnseuI8thY01W}P?fP+r&wDBfKP%d@E@q%IJM
zi>T9u$L=We1wQoW@_SfX@zkroPgTLKQwKiC^@-@^G<cj}(G>;Rn%nw~{X5jEmv_@X
zl+3N;;dN3&;c>*_S*&c%m8s>DQ7(@kn^<a8(KL+55w86*!bY|3&K&r)Hqwe6Ah`vg
zxFzBf1m({mPNXxN?C}^z`3ka?X!x=VI9yKK)cMhpk;b3+nRtxV^Ib0;N#sAAwB+{+
z%a2~HDDQ>mLhoxf>msIQ|51jFBvr`L`#i>GyqRhupD~K5Bh)_lck?F|w^DNX7w6Hh
z8|~%#J;kO!&E6BMmNeSU9j(BR;JQ2@6+tj%j$5P@H|{;r{A}StJU%wooJ?r(&n#Cb
z2b(kvzPXO~#}zTAzowk=mR&qCOaIX)9be6PeQ#A3CdbiZf9ueU$D$7^aFWii5<7o@
zKf7~Y&65@VK@rzPtCs-1T&uQ7Aht_QE<8>wL(ntCHLe}=804Ti-EJU+;4T6nAR>c<
znu9Wf{BKwxGnZtL*#;x{Wxxsch4I>R`n)LT;i`H=6sT+STdyjrN%teQ_zUC1m$1qD
zkAOQY6NF?Dvyj6Z`*NA11>vZ+{YgfE*JVk=jBazWTAGnG%UXX#`DO?DHM+Un&9}fh
zxgfQLWNo@2sl@0H@2~)d)L}xLpHKRJHeF)GH0425crnVzXdskuN+5qg{=kAxxJ3)l
z&fgehZuHMKWo8&_T*b5LE<vXxBvFBY!#RL636aBffpifO>t7AXpc8@Q{xEcb3%+N;
zD|*s?9`NC)%>GqwXYTx_gUow6c0t145E^-9gzCXkZZLsp**XrSvVfWi<>4@`K6e`G
zjqQLCnt>^Cti3PDv2seqqXW9X-wzfi2{Y7P&(2;>s(jp?aerOk{(T9V9M@{z3IW@u
zy_oQ+f;bHizZ0WIul5=2(Dr%is;auw$fw?S@pRv4yMAPnX7+v|%37QI>toHZ%pls2
zzPQPtf);C!q!a?cUF3A=K*7p*Fw*U}TnLFdiO7w3Q2<9&J&waDf<54$xgUQH4m!v@
zNZ$fI?LjChi%IGt<7eRq=4X<A#5v22-t7f<9yhnD7?x!%;`l*_Ww+gp0AF49(ge*{
zO(g!6pUlwM!>EH_A7cC;vd%d=vZh`8v2AN&n-km4gcI9#I<{@ww#|u+$;7t#WuE7p
z^PaQ5Rcr08-hXsguU*ymUBBz9s!PwwGsg9k?$E>HjOf;cQmqs)Nx%oub@CsFxXW<<
z8^Rra9=Qk4b;U%Q=rEboRnQv#tA$&@DC9U@nC{{M?}L#?%vpJLDrjUrix?A+bn>62
zX+&zzFcW<y<voNKrIF1qdO1P07x=A8U$iQw+b-GqCwchUX!%N}I`=LfI3$(ybDA7b
zsHEdNx5xz4v|_*iGr$0}N^Sgg(IXxrz3?Z8t&P#XI-UO|9BmVBy%9Dqy{le{*?{Dn
zo6Gz4ylLA5K{oq<5>A$B<S(Z1%GvJ~fNtyHLmW(wWx@a*e-HJ2bGJ-kx$u$}?pINN
zNZI?0Vi|>ifZo#z#LrdDN34UL+NDxmpe6;Rj?>k5At=?oFu?fOAb*@PhM$dpKnfQ>
z`zxHOrKOuo16i-o``g>i#m&o!>#6kn!O6|Z@+{^^hvME%Q~1-(dGZ_l%`zV3XVP|U
z$*87Si<SY!S^mc#7Y)^FjT*cQ_JnK7GL>C9xm0r|vf3=Y#b*28`!_LS9aGvGdCcvV
z6iUv!ejPh)kAT58vhGI<9<d=R$@$HvGOR&qiB43VG=`(`$@;I3%H7WFSPgtMuH6ga
zWNeaph2FueJChVvEhCZzw}aqhOd{wW^{a9{$=yfY@@$d#Hg3}NCM24qbMj~$Pg-jh
z4vWX#3_>Q8&<sMvOk2x!1};@!>FeU`F$0GMgLMB2NCXj(Nsln;&13^5myqAkNH-A#
zv5wq(hhD>CDUZ6_^J5=wZKc8%g|N+wVnT?^jY4j!kNU$2_ZTXD!163Cj$Hb~TTc4D
znzso#uvq~EVS4y2>5M>iXOzw4#%?d2lfT|&FWe=G>3MPpdvHy)og}W~k6H##=er~+
z4txkM;P)bCNEEH+HzbQqQ5qtoI~?LG;;SKoCxn>5pZG*qQ?OpZ05b(*w%#+e6?s}(
z8>t$N#cp8WEVBWvLe;brwI1sa4D+UTKx#S4-nk0tZHq)6@R$XVO&q;mfJuu)McbEN
zKSZabRGb>5RT#=mwT*DnA`zEc_}H&heM;~N7}3a@5&rj=+}ORz%Tm@*cJoi{ysD=G
zCq1o<36t^J`<OYOVk(g9{9sbF^cJjQo9Gv>LdQW0fyiBI_k}NQ`g$y?6c2ez;hHr0
z2o_Zm3ZWE~rA@<Rw_icL0LtCWX{mLo94BjJ9PaC`9Dzku*e$5ebf|M07;h|qD=*bg
zF$0?%ynwwyE`~q{6RWBX<8#jbi3$4aTsbC9mI^tQhTL3ZXUe{1`9{90@v+>Xm&G<+
z+?T`p0^>+%*|#Ysa#qXqK&HHFJ7wl@IG4fH8BP&@ro7TyYq<Hm$ciR6#bHrw*92p2
z?wV*`0&GtJTJCL}NG?zm2&(}A!~g_lhEgDb$}RuzjoRJep=b~%JQ6`*9Q5%h{dQ*t
zd40fO;lJHVWW)>-K9Q{DH(Qsp1z+Fhj@bT@&}J$b`s8jm2J;VTkdK*MA&cJTaX%^i
zkt2H(9_tGzjZvZ<mnrriKeV$07rs&y`!fPZ&y(&LD2@YA)E<Cy@A1#GG*nH1VxcrW
zSYS(A`?Jy7Qo0!GA&kI=sN8Zyxj~-+W!YkVr&A?%yk=B!&G<J%SA7U7<&?Dj<mxL7
zyEsuvZ8AP0G^DwGxU(WjZ*BPgo>=C7e$po7^hT)9zFgx#Kp<c)Pw%`;#j^tw^LkQZ
zvEHay{LPD=Hw~aR#`ZZ8AYrl55%iVx1ST!xKV;u5&MFZI%(#VlNy<y;Esh~&%H%CK
zuJkhVnz*jq^6a`Bkj|eV!0%6f=VV>peeX?*dS-vBg+nG4RXM?;n|4`ZHO4!nV>h-K
zac_`y=OHVQjUuAIv#Fy?;0YO`vYIQY702|%obrBZcL#LSZzJL-tRm>0DJbvV&0>BQ
zRIjoORlt_gSdUyj**KpIW^^fM1u{&s5z;moHp5NauJ<mhRB*l)JE_oEcq=!&D=s&`
zlAZDa4bs`T1o^_CLVs{7B_DSWnoit|imWD(^kYpR`{-l5c93K@oe^%>C&BZ*qT21t
zY_IauoC9tGML9Lme&e;?(5PLxfk%sDcefL#oOZlI8e=hctyR(v{DzVl#ozic`hFEW
z5OsMH;Sku#dlk6gCJFARwEhr?&(JJcC3hz?v5UAtMe8W?bz-l=9C-8*WXf|bf6eO{
z8e`j+A|u1qm%)<aI%LR~#T!ffx=@#+Z36G{*bk^X5X?szm`z%Oku^^bmBh~cWA`_!
zp@Bj@Ta(wu&KFhpowF=$#49BsS3%xKY|B5DrouOt+u4+rCL!+Xou<-v2Z}!+Vg`u^
zi8qndd=4lyGG+^*pLlTi!xbkvBZWiKIblkrbO6a$etO+(4pw1{iCaOwLXSToBV^QF
z3b;c-*`Lj8Yjb>MTU5Qb=V>2rF`&m=f0|qCEUY{(HAZ2SCR=&Tgx8K!b_H{bd_Zoq
zcrNF?%HX4}r&^1Q7Dy}={-7pemPOK#d2}(XIV^Ec<LL9?!xs&G08vqNTjUL|Qgf@y
zjF`3-2ZJ8irKd?H`$^55hj)SSF7-Q}1~BXXwallVoMwRQ?qy}4?r`p2%(@3>x=KZM
zM?H}rY`W-uep89NJsS)<bUbt%YIp(sJIwG+rUeNRgNXoA;t>?vAZ8udUY7-3Hq$XK
z>?mD_rF0y1unr#H`DOX5n!g~g4|z_O6?PRyOXqv+p>1-%I*@#DFH{;Xk;2e}F~Foi
zMO^E}^fhg+_iHP9c{S2BfU$&ZqGJYoRG?6;!Q<%IBmQgmglDj6bF_1Z=~+U%deFne
zgz2H@LO^pOJ#4w*pDM}EZZ)E7DMbkTQI6Y0YDEij3U?_(nV(QG*}F1|?>KuwN)o-b
z+hl)36Sx&iNJU18U}ZRGi}`difT1a@Hmh=NP&Q}*s6m;r67$n{CP$-eh7{HY7f|J;
z^|`_*XUtW3Zk|i)bZ77~JY8sYFV`a%AVPz$5ElrhYriEuWl2{_+d_GX7_!)RVy1z=
zZzD%m(ev+g<Eb48gf4AY7U8a{W6|kR;H`%1?TXH~x>-wYUPT$y37{C_qKNT1c+f@)
zX<7^#^={F0cI!)|L0xs%l3i@JE2&J%QBwcO;_(h(%!@UdxO#h2PoBoVZep_rO}jQw
zCORZt#`i`PosmH{Ns`#(96};gQ6vdd)Hpf<k7pyb9E@j6l+bgoZEviSn?{jD1IZKd
zM}d|irHrGkhc_CI0LTK9p|BE?)j*B05{{j-4&uZ%ZVV*j5;~|?<7drkji#g{LO%%y
z!2Av=_5N0Docz6kV}p{KLjA7Quo<R>q1r=fMFTVQ2X4pyH(LA53YLGPWv(F)xbjp3
z6(MH7Gh0vX_zU<mma2$!q?D|v$qyrsV%3xN3%f<rl_LN#N9~w3H^5Y=f&?d#{#*P_
z%?*2bu9kHotZk>v1|w4Q9fp_Kf{zmHMEXk1^>-8gJhpe{?ma1<(`yFj0S`Ui8ikBQ
zJw_W>w;T4iLaeo^D@4J&uzanXVwOWv2Ty;Cw$&z6uSaR{n*e9q=DuLoCSud^U_5<p
zA5xqTVp{+^S{@EN7_LXqN1zF*{0?CoYO?l1$W3C)PR>m>9B&hV6wRF;cVC!n0pTXi
zBBBCSt{vISAlx&V!f%W9jb=O!#oyq<miY&)-oD8JpnDBY(e>(6ltf^N+md5A>fGEV
zu=v1ZY>Lv0UiL+l`|`rog6`qeBa$pL)NfGq;1OW9FKEMp3{xmB=25XbLc~ZeY!CCK
z=qv)nPhMeeSnC&QUbzWUw_K{7iJFt0Mydk$&%-bbq*E!lE?2fzIj)(Xzah%dCP3B}
zF3gsTOv;e+zgMz69QT7WoVDk*kw;m+Uf?-4^!c<yEH-_V)@k6VGm7Z+!uU!3=BTF*
z$X?xF=}V<b@18@%a%aqEq@4R1U6yw4MN=;Mo1RA>{UO?Zq%O<AiH$rKHg*E4O-5I)
zMA$F7Tdj4GVnW<NfJ%U>7B3gAN>M899fd+FtN~s|JW;4R$)TK&X({#lu0~u?t;x>p
z2ILH@nWo6R7?4$Pc@%%9O4)3~vwW^KV5T{=Im&CfRKshS<%!Kq)62(Nb8(ok_;{KA
z!SXc<g3CJMmD0Pq8F|ANe>U(+mvoC)SEJ67Pu;{+?$Ipq$fB(o;1&xp3q1{;F$KR0
z+XGxB>p{Mz(x;%Dmt77EcjGfHZP#;XL{AB`W?k@G;l0)I+xo)LcX-a`v--R(zykqg
z-=1;3(;gm&Q>y&FItPR5Tzl~8rd<V{nHhkYz-{2qu<QL$5&}AilcY{8jgvm^;{zi7
zkI${!>UY+q$TF`~p3UO-dXo%xgn$tda%_kdgq3VCMTey)QmK(RLx`fi-+juVI}{K+
z<GF!exSA{rsuiA11*tY<M~Zp*fNyUpN_OzZaK{W<+qtSCEay#B?&t*f=eS9BufZmD
zV0~F)b`)e-Q#MobDHvm&FPReU>U_Z8Wr9b+q-0PzT<~ARGy=A|o=(>+5ft8u_iP*x
z!yuEN4}Be2ya`BKADCI37-ZXv|MHykV(wNV#I7Gq%Dll&(QGM#E`vsZ64_(@g;6G-
z=z7vsl0$rCa%A!(lHNs>X>lTfr_~`JjE*+uTqwYG^+ZE$8N4C~=-SoEMA@GZ-Ys@`
z%bQC9$C--IpItlD)U2@SP01sg4(Dhd;9Z*2hIAc9>E2TyJHLEQ^cUQi!v(Atp3X?d
z*;gxM9ieuFF3BbH>00LjgXDt)Gq*-ZyA~o#N2+m+<tcvCJxy34aiD!<g`4(0wTYH8
zp5Sqaa;;#M9}+oW_&f(GRw&=Oh&MinM}kkts^;yq!w`xS8;UXuWCaxyG|Aqc;lk`$
zHLY7Jr3|p5nk8v+eY|P;F(PNpu%e1*%~;#AAB$enXUzs|^;k6l-DkYaI91%PteS}w
z>5;3Nrmsw!ZCd6A{h_3O&?BAjjQUHVbM_a>$x3q;Usn{5`3<_j_Av!<@uyju6TZm7
z4LT@>-Tnff=ce#BEvO4zxQjxtg2xRyXxzQSUX5OjMsyzefy}!^B5h@>>abQ4Gv+;_
z6lgl55HE~pD_TG!an2NuVw^Bxd%vI@>^A4H&=@0-xNl69FCyJY2C-lal82KDAGbnF
z)z+D5T^A2*c<yipllIm)y$n1<iuMBk2D?3fsHTdx2ftbPMehi?u<J&)9sOqDp;P9=
zKaq2!pjwZ^3NrcgR?Bboz1J|Amv6hjVHLEc)zD2FaF5==4;j3lrgPEFy%9gIcjJlE
zvJ!t9^%c;|+7OWa^M61hL_fGb?sEhoJUpd&xzrmsYHr*-2s;?+_x=;IF*Z5x!i8Hs
z3Yv8{iwL6MNTVuym)2H95gul3$OGjkqvQw6u;r%W3*ez`mO|4Vxe^h;IT;4rSFvEp
z1_1skq}5QqIt}`m_I2?wzR$cLfxP_$`Y~x@f`5adOyfcJUTtbyEu|ioVtrw>)oVeN
z66<aul?O8(4JSGVjCnKMG=C8@oje$M!K=mk=;<t~8|aoV$MNkjBAqAZ{xEXapOUbS
zr?xuIzn`E&i*NvEfe}eDHY;NV+wdFE{l}wId!@a#?PHXlPd95*=cjI670D@FWwPm-
z+8DD(=V){E>ZYyigX(wtGrvXpJDG}wmQ9k1^U#tr!W*7G)Lr+pF%B1Z=jG1lAMc<%
zJvLkQ6{kEdF5b>RdOBga?LVdl2c~xvcW<v|t!KtqydL&jCK`UPXuM1L^voIoWS`yN
ztKEOTY{oN?ce^fiwpnCD>UOq$^jGr%tXOSrG~SEd-#W;)<dL=9JXp12q=zHO8SYkB
zecoU$ro--1_{6zU4=3Z`*0)$W!sqe1p6b@wUaE=GtfAjiQ<K^~NUgN-&X$m%b>os=
z;J#Jzu1`+y411TZ##OBb*)@d&-hlIrd<GK%&~Ne@yi`(2%5vs+pgfkAdXXi<>9nSK
z*ZoQ>r#AAD$_L!lvxTk=hO(l)E?q<6zHI4fq!K-B=?%w&N+L=-;jXvG`fV;wVhj=`
z2N|FrGj##yS0`jT%Iq`|=c(y32P@ZO(2mbJK~izf)*ZU_7ZFx9+FKd`(^WDn4P`C0
z^jW5=c+&^B*eP6Qz@`0&6}pYC_I+V(!$5P|3wFoAOZ4VW^#qu&AGu%x2GK+EAM}2Z
z*VX$G`JBs>@=hPu=en_KhLw=hQ~7$EX!Uq(|2>UiPpSm{nHBD20Q%3odoxb4iCwuo
zo>o^GztHFFI+~fTMVC*&)fxXwzv;@kKL85lp7!!h*Hp9kr2|T+{G-KZ44XitLp{-t
zTW`PDwptk9PFZDiq}ugcF1V}PYr=IeE&@k)1w*wi@pt}8wucjiICpB4tB~;U(#`We
zst%B<v$nm^nyZ>cTT3I3LOF!dOq3x~^G_!Qps_HYsG(XK_aHd{3g;Jes<vI(UyQ%8
z8?Jg-9xkLcby_^473(oPdKlP$-e0Uao7i7`zQ=sf+O!33H(#3evCOU@d{#$X^D>g#
zdjuOJmHhtv*zwZCV2ch!Q3X5ig5PSJwa=hj24J(b;s0$x*`qj^MM`@+r}_G|doZI?
zLUh2{#M=GYM=b<^j3-)-{3FbX7&;S~m3->CZk1O~OW(HW$u_iBho8SE(Yg9}6~lc7
z&7U`M-9%l=rZd0&vZ4J-l5VdKGC;r^+ejaP%q*C?K6&;$S)EhLz7E!Wi6*3D%1B#+
zx}r*6Zf}yCtQMau?IN3Hw|iorFv=yP?G4g3lK#|w8QmR_N0aR-pQ*LuS<`H4<SUtO
zBlhBN0qF_3kHeK#qW0}8-4RFL3%p1k#TSF$T<WT1f-OO68?T<Wn2(j3>fK$~py=8!
z7;6p9kv%mn9!vD%1#hv^L}QDo`F+=6tgJR_o_Ps>1Oujz;R3=FwkUV!&*LPD0i-$w
z2i=R>%i<33Ajg@fEL+t?Jw!!U*g&sp6H)s4Ctpgd`q*RRW&uCb(1bpDwQtp@D1L$N
zX2EDlA40#3>jAX^qRaDQPf)G?m46>Z0YylQ7r~N*ect%Qw!a}2L*Ft($IBRez@mZB
zB%<MK&KpH`Sm-g@I4To!->N^>r4t;AwIe%t8TSZa+pJ#20MNsXP-6`ez^BxXUUIcK
z=oeIP?w(W!UjB?IK<w{my;K?T9q@cWB<OM%mCSb>z2tO}H@B6r{vnVD6?gfA?+Koj
z<P8b32ijbGuZhF0TdazaXB~`$V)Zzc0}QJKsHZ;1Y5g?0K@28~dO3n>vGh@#C1G8%
zERP$&Yt}MVY@sFHn1EBr4^D{)4y8~{M!7sWmTw6@h71QCoX9UM6b}Oc0w9I(3)B$}
zm?W(m2qqHA{u*w~c#rgly7mtxJls-3B=+E8>%WG#I!S^<2u&Z8=Pgs4DOW~X;m!P>
zdk|N{Kfxh(zP|D{xMB8XoKD6eKlG_cM#BJ{jXew`|FJW1((znrGEhH%+<Eu+w^!tz
zz`R#}_p{)`b<Cb6q^mo<6m^-_`jUto4+nDYTX!AUSS>iO_3vPVlg#Ihvj^Z~S_1ge
z%8)V`r2TTX5{sS^j6i7OLS_UPZ;YSP?lzB!3TUqk$ClvE?}<nGdxSpPk^ycxcAWqk
zaH&`pve0Bppk>qpFCRguH<|8$slo5GrNY&V#CWz`$%!($c5-oS*pz0Im#T@FO5)4h
z?=&&T-yUffyx7foHFEi#if2y~GOmhdf+x)a+TEzwgcezNc`60OTEF>~ykU$-2s|)V
ze7ON&0f8qZNCf2(Tq&P}c4|jJ;Sm5zdJun7RqeoxvAmVTC5$}y=f&W$xqS1y`p|I_
z3&~Se4CTi{h2Tlb0Tq4wVIAr)uKg6Xm2jQ11%{nAn;d0@95_E@W{%+Bj?p>_)Dw9N
zhj18JJ@~C{&mdP|e~UR`dTq`Mc`#(}ivAhnLsVw0wGrFV@R@fJya1O10BD6r&qu*f
zuys0Wb61@@L<kYi+jP|^b0b^G{l?^dGLD67SCp|?Z;8>*tKwG~-tso$)ivZ6n7eWy
z&86ewPqfQ1q3{#5JTi7Yru6D&zut1K+(?cqQJB>M0{nJ^@U7-D3$C)Hhy07>3Go{6
z*|KV38KJ-z7`c|F!CN(8fRYPgx_KrBXe_dcK1FRp>zN2ctMCsanylbB7E&O2py|Ei
zwC)ALgU!R%26!0Ci@6<b66oov9D+E&<m3?Zyy2=}C#lE}i-J>G_t+#r5Y6KM49Di9
zCYpBte@3Ev^7DX|{X+mxb=@9cF#)q`ONqxut@Z&g0EfgNlmoeE#B*=*Jl*>GFyF4y
zU+`>pWpvh6hNS0kV>~EoTn99J6}AN&O`I^W<nALAP`L#^#QzA!H@w)9p4m%r+aI^Q
zwf&5v)-B|VjYOmtrL-(u**aOvIUAV7>v7X$$(rUWk7#Hs&j39@Acqnk!NN1AuO#Ku
z@NFs+K)V(a>L+bouY!dFXPyuoXaW2N{3eL^0i>D%mw{f+uF+WaUDY}t%-lg!rgh$$
zU9?mNxkd)u5PVA^HT-(VYd^Qr2d|>fDck#IxdJ^ZT#iirEb?S@c4$k~F~@^*fgZ;p
z&oyF~kgYF^uZJaDmcvTUsNVQ?=(E)FLFRe~;37VRoJOspUSWy2COEPV>o^5Dah6Ia
zYyg^uS&gk<{<<%w+*aJ;86ait`VE;}(iZ6|1Ook3)VDU!&2H&CNeNV0P|#n^Qb2u1
z){e04(X3^)Hz{sMBl=#EMrcsD0M}SxCs!Wy6e6bg+Z83xcM!0&zE+BzxJt{`t*B7I
z^v0ia`C(97&W)$)ZcfZhk$gOjLWh@lNRb;gxSiX4?cbaHlmiFi>^YOL;y-zs3uIK&
zC}mZz3P_gr7}Grf%ZrDIgbu`|Ln06e6xbwjBR#y_C&WZc_0}c$YIeskr(Br*ICrl*
zJ+!b_I<Jy%RHZ#F?jwz@vH7Cf;nZ?~!-2CsDrgPCsZ+{Y{H5j`+5?YHhQxVIS2D(#
zW)H40#(FcJHECw!5rOr;<r)RLG3&V&u@Sy83!MtYI)Pj&-+KC@2ipn~nC}g#N)_5%
zY=fZ^&mO3m>{WW4KW`j<Ba2LMpSW9LcyrEi4+NC*Qi9L}v@s6SnlkR76Ci?<&pzVi
z24)5NRn|jUis9B1Pe;MwfNZ{zGj6k^E;bu~Nt)jx)`!cjIroc+ID6T9n`Lr=RMySL
zU`wOETSV-Jq7AXX<XU&GKrDy}ItW+`Sf>d9Q#K=`V(H=PWjtNwdMR)fa{M9bzi`uE
zm-z>g`>s;e=A<4cR{{iEq}@&44StKr(UXb4fibkbS0A6Bz1hk-@>;4y(=Vvyw`5dA
zqw;@cbpN1roM-hD&ftxpv&Ocbqcybx(zbpqc_E931pbJXbL`D(FE3j@#%nFejYYcp
zcjnA|Y@}*>)YNQqElV8P%tOfxoUY8}xj`64^dcw;YT|T2H+75*mMB%87hQeIc>EOd
zo`|whvSPgB*a~z<$MePpRXt@`#UNxLh)_G;l+{AvnX@lzQc`ZL`*u$<Ev3*?NFmY(
zCX{IKCv`u6KmWNx?FD<_K-XpjtZJ&*SY-NS3Yy=Td(~Wmd6M>R&v&TD;Vvz+jI8Kx
zXAj<)G~G_XZ~aSXP}a?4py}j=C#s3~;y5BkQeiub`x|Q$&5t*mj)F@o8jE^v)rp=Z
z5n9fgR1w-Q*(o5rEx3Z4qtx{MIrKw#A^0hW*uM}siCb)<A&U*4uAK^lN7G^xZLMkX
z7rEJU{&xp!Ml*lOc4%_-GF%@hT@HQb74)oYGzCNeu<T`6F-<Mmw`gaSm3q1@B!lK%
zb8Vh4Wz#aMi#z|QlM?5O0-Ktok+h6nXZ<3j@->%xe#hhoROxokLT8{U@a4)xEhCXU
zgb{ap<z6PGrxF<Uw{dW#rj_6CmJ)QYkgnSM#JZ_FC@egVK58CQs2G3D6ulg`$7|I4
zcF{R?35=gkg7FTWoikxdEE-Ct$vn5ASFh~Mi_W1fH*@!k4pKp4NI9DL>`G!WU2g*>
zNOJ5DgDGzG<6c|9(1W==+XiKLD1L#YxPrrS?i=;}7gr|{*%LHsj4`}T;f@5u1tju&
zHLlnlJ5LoE21|Cfn@eWGxEt*+{*cFNY*(wd=+B+ND++=J?PcwCDxVuCOeLqO;BwMv
zPJx{c!wXewNByJYGtu`CK&QG>CC(liE<cu$;U>8iElP=%D~o|13z+hvLupPw{0E+A
z7$;`_bgRQH0I(o5sQ7zNqpkL$b2Vj3bI-QFd}+uJy_X$jqbWy`U<3fS0x@*wADaOy
z7S7A#ZBH96B9V;WzZ@kRJ4Ga@<9{w@81CrO{1QWy9f_^#nKOlJVsy3~a(~(`JXBmq
z<_Y3S^F$w`*UL$N)HzY%)5!XRY(sBD_d=b35cEO0_p$?%VXzQEl{f|BkQAur4*qO%
z5yX{hJgcS?Eyr@Pw`&CSR9uk5LfrrFiFJ0528`j1Avu3S9=FgdL}|9A*L0EJ{bqem
zrG&%BxIhBsDSH(oG&uRtMcnBr&x3c$BiytV{waG|KZSE0%=vlB^smR#5N0YJ43&<4
z%l;4aR)L+Qd^qb^+XE`q`~?}gn}s2DFM1YHwkrVCJaUW?QEB+ov7NI9XJ`=eA=f$A
zZBa5Z6EJy6X|hR)o;t&aUlBj6Zk|4dw!o+6bMpt2N#?kl`g#EKNp&Rbe1n?)!*2ei
z!l#Rm_v47_?kR}uM>A**jHsM&-16o+kZ3@D7nj$>s&?Ld1&Nn1_E_*`Dfu@q;kX8Z
z?peTPy}ErUzMc&Wmn(NxzpPbx!Tv{th$@8*_12Cs!bDR6i${cl(k<~{to-K{iA{69
zaq?@L7_2q3C+kPxI;r;OVMtXThUXLnFCSG?*^yuYf;ij7szk5cQ+R(jR4(w7L~v+f
zUtvcHT5K;KeKgZ?PFJ1m#O+Ah;yhhE5|FKu&~@C%rr$jN_|L?i2rUKyY7Yj5OZWLw
zhVwUkx`+~^NOLMo`=V4*t-fi#pip3DOCi(Zu+1|i!CkpA1m0Ch<a_2r<ROzXH2h&o
zUD;lH(7=#OPAO1AooCMoo(p0JAZV>K(D5K-|Md^W&<s4>C(_^)8D)I>@#HlCaUE4<
zqR5JOw?h{;k+5=I2?}>^A-IzY_a^Z8$<O&Iq;pjv@MOg*)a(NSH@%8qXW-tZb@2ef
z_CIKnE1eDQYr||0wBIna^4^s4Ni-j2H)As)pE<LqpWj}z*`tC^(jYB2Oo}!2_5vU8
zz;07LtlUvB<qC%K0i7@3R(cnI%?*7gNuiW>&`sU&eVqQmL%b0228e^2u8Xz`w_mhM
zpHhN`Am9$KHG-ybK`(qf@i9&Pd6Vkt?tqtvgqilcE(Nk}YA`020Q<Nq`oeiNnn3We
z9B|Hd^pj6syek3dwr!;*+Gt<ZE3uO}L~+@gV#88}{#V#<Wq^D=5yszyLJ3staa24Y
zsaP}qH?uhv)ZGFrxgY~m!dGGeb%J*jt<Ej@9U3Gy$Mj(F{iNOG2CI07qVFZBslwO(
z<(I5ZTHqy~GJMSA3%kB2{G$lh!9c-}lk~1IUbH>3Mmwjo9MFNAN&1oGe3{uhbOo@l
zvbQJ>-uMV`GdJP-9c~d`HX)}T)MdC~Z&iY|xr3hf#4A85Kx&Od<qU%rpZ(@b%yNIy
z8@pN$rzfE|I)%g<OgwRhsW-ZkMvINnNlMI?Q@xtB=6<C(V96Y)llW23L&vN#L|gzr
z3f0Hh(`TpRDL@-|v~7}riXxKej!cYnz=0HPis23@A2>}$@1I7AMKMbB#wNx(;J}J5
z!UzpQ!r!2+?|%z7+Z1rC(_Ob^9R=_A&3Y!#xv5OkL17>Dx{q%?2rzazfMW<l(V<&@
zKt!A);%=uu$*28*HkXgabqHO6Xk^#dOrf)#5i2dRxJkkLc*KMI<bkg3tJG#meMEVW
z#(@H$zW-8J$xc5i1D;NnrCMCy0VPY#WNS;M*q16w`&BtQb!EhK4?Ha6_d*(`ZJHcy
zZ~M6|1GSg!dR|hO2S3i!$_>A+0U~xk<Hr_qo_oMKtUbn9s43C3<bCf|q`6e@6cN2`
zcMs0qh>UwfQANGd9cP8l1oa^7SE^qchV%gNG4|D}x>$@hz+nn&2u#e9rotlTR9jBn
za*@&+VB=GpuDGK7ZKhiPvi^L(N%$kG2I}IAu|oZ2tgEpv#{p=}HQieeDue{&!bT`$
z=SPjkBR&RSRz-#n3)iKMl<;&0V%G#uiSM<O{E&&Jg!_X2sES7D41v3pH`S6X_yGE}
zWM<{|Y6w+Sk6B<buy1Rn$ax+?C+OurXw+)NvXGpD3S7ZY-^bg;ck{KY3usBc9{1i?
z4y71)h~yO+a%G0Ce-KkyA#k|ASx|86h9)t>YYhD5o|@6^JiT$|U>GWc=Wa)JSYvdI
zKVaO?d~O`=kMZoZ##!3?-LP$*3(&%ztGXU@QZ8rvKuQOn@<=ifCFiBP(<E?$ZNI19
z#L3C5AGM5|p(VP6Gh1q(7z(2n>h0W1iJ&NlvR5W%)61|YP6)(|G$eaBsLE@O_u0}p
zwr2}3m3m@8At#2<<$iwQhM}eZ28kyj(Xs2JPxYhW+sJaN%@~eThef%hGa&2yP=cOZ
z8>Gz;FApmCUZ^z*^hX5C?Fn}_KSP_QmxEAij4*xr{G>;}pw5-~m3dufyT4f5t+@sG
z{;$vc?(_HYG@r->W6KWX#nR~YZ7QnoFVEYaC}b7}<+BDiC|TK9#0<FesK=<~2RroG
zBo%Ec?;hcqC|d8u1%ZpT8-iZz=NpoJIN_dSGrFy29uTySMiQzmy&0BaCj^fBweEu(
zX$cR`k!@Uv?YjvN6*O{d%SOLAgQpl8;ove5kFvB<S`)ucaYCo?Meg4M-97-`7CT^m
zDdQb37IeeLB)GTNQH?wmviq&~S44?C0r`bvm}i7#hryOvv<l|PO9r)^D85{nhH%Q{
zg<ZxO3H8+lgd_tysT5m?r}q4y4?Ok>nB*9jNguy#g52Rq*zyFySKz3V1B4ns{Gy}B
z#UU*orZUh05(&nuiXXAJq!#{M1C;ncd7OT+$ralD(5RUAh%K@*mVT>50Ck$zpWIc(
z;TJ&8q+gVNMF0gQi<G+)c2~?qm3ybft}V&k>9A<A3Gy6@fpjjmhoww0xNP9bivZsV
ze)-F_q*dzuAL=Lt*Q5UWFQ(V10w60DU>^K8r}z0koL+3ufDZJu!<b*g$@`Hen(H)b
z&Br;V9X5yeGniX3-H&+fJpC%H?K+H8WqctICBENk%s7mOR7*OK1_O^GFUdMGC|<u$
zdM>W*p$b&d;|2!HIe|Lshxje`>;h^ua&%m_$}ymm`yx;V_}IeR_VLK<iI|EgYcG*^
ztncbGAYq$SR*sz>!Nu~<CA@HWtmK$oW{5H0Xbh?rKhn$li{b+fdk@^ZwbeS?JHE&1
zcy)7kbaZwbXo)UitJ6(4R>d3Nxy4!IH+1|?KdW%tSplukIZ9J2v+WR7TSt~(<==Dl
zrvMD!t|Zu=-JDdre|aMzU3@k>jejivw`CH&5%Vo1EF@>BVEFJ)!Fn#Q{?plnZHj46
zUBgS#n@_=5`pxZ&&TZ#i2br#n*JY)<<KlP3Za2sGpfBs|irLmi{fq4FldE)R4oUOX
zomn$M%A||j&!vWI(>LZv@ou%?ZvfqpE(1rg{@{0pj05={P``b+Wt-)r4u7L&vO7Zq
z5ytOS-;Y7$xzB~CqR%wx19|lJ5uwO%-8Xta6BiTmRk-#O#rSI*njwbOZ<dTAXrnNb
z`#)$vnU@W<?qoiWYM}a-oGikT7#W_|89c0#ei!o?_^=CVA8i`MmW6B&djhJ*!e<y_
z4g|I5@vrg4je0F#4h9^~k0O7@%1v=VzU3K0t{-nn$@qgc)UQOBjTuGb=VUUbd}@@U
zY5e2f=^lL;H;yxWM{I`Eda%|o7aM7m%3j?SWVaoC>l(ptP;K?LUrgx#g|KdNVtR(~
zrEf%Z(4zuDq=U(7`mSgJf`354o|!Ha@_TnzRNTC5?vIdnbn6jT*YJ<Ea~VjQK*gJ4
zUlz#(3mTrPf2g_zEfxI1qKn}Py4c=g0w*~b8L1X{R9k-^X!tyW&Czdx{UDZDXz08J
zWSOU(T#`VQz5IS5Gj&5cH&BE)@l6V-f3aTx<`5myI`KUWX&U|szz=7XNB(kuDT2kB
zLzo`Jq}Qj!w`YDfy;w)#1ur}Ur8VTHlAh-P!?beFCe>kSlhw>f_U|4{P2IzHbZRPr
z{c2tOSW8XZkKDq}8ajFe-}Gty{<V%Y$-AwSyXT|dr{Xn=k7Md()ppA9{u|-x=R=0O
z(ZR0k$Eb?L_vh&jz(4GLlKemH9kuxXWA7cOJYra7lDdKq#^)X`0Cn?tZ&3==PZ5tO
zuwPJXF$ZreXL7|=wE;?R93m^#IXSxH7E6~)DKAq=b{_={;|z(K54|XE2ZBJ$|HlBD
zyI(d6_{RbI2dZO;zVKiBr#*4mSw25k9M54=uS=?FgXgcvSNaqH>zeh!4dDh;d*z7E
zvfFmG+Xb(^P1vRNR(2eQ2{C1r7GWv=8GfSW9vX39Ix&74L)!;-XfgsX*}O&Ao7qCT
z&E|wfo1ItOYx~RD16-Llt++$?k!3evuV5_e=`VsmQ=X1zQSp%YLGq$GQ(q>vKvLtQ
zyps%XOj=^!&5=ukw2i+4eMYLr1AH+V2ySj75D07<+b&3?;qGh%WOS5f#$8lsVi`3m
zuOeXoqvQ}p^6BHSjQXLy_|$$v`SCP5txeqo()2Y+N(q%jDM2FyJFw_rMIJDTb9d2}
z0I4yS!Rut)1eUIZsyI$IOG>GZk7RDZUp%q!J1~NSZoHWFRy!0Ou42k3mF}01-3VK9
zW7!e$JP8Z$(r)lR^W`j3o>phr6Yq^u=Q`W^&a^1<AFzOxf<Gu2&%B3f^i~T^p4vJR
zx0{yscY>T(+gTPX4sO#QeFgf<T2zuw2+#N<7gXo)W#=$5IK(&a<D0C0Vu6JD8RW$0
zp`9-I^geZvxh^V9Uu0gE{eP2rsMH1mhRl5<VTXt!k?L}unmi>I)4GH?*~~kEWA+T{
z82am!x(qNZkVO8Ji3~6@K#Ph)XtYMCofoKd9NE|UaFZKl`Epf>Q#6q!WQ29a2=*mH
zCBlLSYb!=On?~Dt6>0bcSUMG4Bv`fhSN#b)PL}p+-ZM1j$_LMDAHlK?%B=l}m$DB5
zwznv*>7$tBmHS7Mjzc=b?eZY%Khqu?_HF42oOlkbFW>{|?Z?d1fIq<_>w>m(Dxp&;
zMZNO2qe>><AOZQBdffH6dXAzi@LaKHg{im(0+e3ZZMWYZ$R>IlQ#=&BR|W-uu(;vy
zgS%&I3AbQYRot=Y^;qDrY##$*M>Q&AzLKH>@&ww$zdz|tmCjAYZ2?b;ckpY6@CVqA
zkj;S$<>psA)FE}irjrM?VtD0Pui)g{I<Z6eq#4FU@Im6<tgMM2@bPkHy}&*Brjd!a
zU}zdUpBZ>J5ko9=x8Cg?B5}eo)%=fhjBI>%36a)Yf=$FR7!hVpB5F~#4c0|RVI{Po
z2(!S05U(=uUm}iF&aC{sGx|qnJ_5JG+1dh(377;xU5G~jZ;9mF_bmvRg|YCH=Y
zd}>WXy$8nEjkI^8!vOQVYv3qE&NGqSfP9=NL^Cg6hEyYg6JO}+uO>{NdZsc7U%rp;
zmu^ZG`5;fH8YdRSrtE|whEdtSU<y^sPA3IZ_<j`$Z2K|n6-x1em3s`HNbkxp3f&R;
z2JXY#QUFk`Ywug`u)kf*ys3Im_Z?(#cY{@pfDK{>FIS4&LGMTAH3`<scS|(KZq-8-
zgpP-0!lzRZ|4yp@q{lXkNHW)YqzLkVlZwAq=rOnLFDvwy6S`@ur2)7$SL-T<BUL>7
zHskDD6c5vn6SYN$)9qvOY8e{^z7>!Zc)-{(SrKDM5e>38-H}rQwuuFS{x@_Mw*b*C
z-)aG0SG_NguYL~J%7Tm0IHyR%CHz;4kjc(Ij_@<u@{Z<N&-LA@)|w`ZzmhsVK3^P}
z)TU+OeTbciD0V7hZzsus)UZCnv6JEgZ}Yw?Q`?Cr(sLFl412eF!?4aDL9QeM@L>GU
z0EYW9BV2qEZ4`r>JkcpG9B6Y*zVql7li8-zK=+}IhAhVz%8P?*ZWkuG_n4{GR;$<+
zv&rK5N9=y*D4-T|<af|TOF`q9)YsrG?-H~4tEuVD{ngYi8leAbYToSsXlkg=*Acux
zO)$@y!)NfifX+am-;@IGVKz)zdncA$mUR8dLhPh1>QZK<8$j@y94ZzEO(nKoq_-Ex
z#*ph)<#Ju53-J}@#5J(soE$at`tiy>x7<^iwA?eHeFoMymjA{6xgr0H{oj+rU<7*1
zLzoG4g&}+$z7%QZASB;9<@iPa%aCiz0N|inTM9}M_N0>k;D4S!&e_!W|Be5PqyB^c
z3%!p0V9`A_@J{cL=6H5jod$vx$k&aIVa~buK~BF`>~ywmDk}aYcR*h{SqpHP0fiwp
z!Od;^JF}cjCbn;?gs2Mp^^XP+bl)OACuPl1C4r+4aE*^+-A06`UJ{o7)7RIhaT}UR
zQV3D@kNsL)f-Qq6tO0|^ix=!M&NId(1716<{s`ibELI&+W2`RBqAK&LVm3<35<Mi7
z@b{~wVe%z-Om9J>+um!IhhdFo2hi?;U(R1iA$H3pgR9f{b9(@{;;*0kFVwfz3?2dK
zSa&%DmR(bWe<ex|-go=n{h6M`A!e7dJN;V=RH8qF=uQdA$gZ5gvrIxB*Pq!rri-ro
z9cLdJY8}2jo}2M(M=r3_<upY~Cl_aLmnXL&h<^=EpB!&@HDTBlGss;|dw0g@&Y`8f
zsLvOvDRZQPPFtIp@}3U+(B6583jkzTjlRjYKR}S>9iZ>@Z9V>ooAK<Y1J&b8x}uar
zr%5D6L6?8hLy~SGih`Aw6ONm-Tp1#hO?j$k(mOJyr&QR0$JE-#md<Kx6C4zL7vc=Q
zy^Or~*T~jorBOA529|OOWhNMiFB0d2$V=35JY7iCNfxcx<+k(8JX5(WgaD$8R#XdD
zEU83CVG36ynkhROm8`xhb<WWr8Qen%HZe=(_rhehl+mB@U$9y`!zh?Q5lITd;bnN&
z(}Lu?H+1OT8Q*oJn$e!em`SrcvEVX~NR1B!y)1_K_ULq6PR+RR-eD}aXxTS@ai8oX
zWX_x0)o_*s{=p{6j<+{9fa&MxG>5p9`53$D-XYX1*n-<W=$*b#^zAKYKEsXUISZS^
zmOi&~#@a7gF<sLT78MLN4E`~*WN3;YX!HyxC8QdYop(QA(r09Lse$KFo*&fkA&dDB
zXG?U%oIRl`76meK;?8NUylh%J!8t<WcS~{DMU|^Kbw%GHJ!BOo077(@ss#DVU`i#F
z&ZKA(l*QM+Ewp)8Kp3~B^o}UurR{M1LD5H5-(bJ$Vv%HMoOrtOR6=S*c|eVYSN#%{
zu?j?aql7|Ui_k%aeBZl|LqVd>8j{S&0il-3>``aZ(fS`bT{Zf0c#_TUG&qkNZAhoa
zKmm28`8*(ulPE0!OxKl*mbw~Vro7>LSu9wF(fNn1me2Px(m!HE&`4xu%%@g^axKqx
zG^d$wmE)DF*jxr-ld&zfD?SxtuD~mnb8Kvxsg8&J^|qC1&bMQo4csgfURT|rh4zn&
z`*!L~y0}1IaRek_vG2VO4;jhx11cHN5{LzoXl;ChX+!h?-pExIt)#eb$r6a}ufgXW
z2q|;IXL#d`4=Y?oIi#@-h-MX{O492&c~!~0N6c#$-k2K~tf%7M4ylk49KV2WNAL59
zMJvYuDa!2W83M@@9jb9=ppW_{QN;kweka^tUB*w|3kAncE-cNmBk3h7R~}U!Yt=%1
zZ8T3X-lHdg*N}{|Z+;IH7t&K0Tx%i8&a7E7;96ZxCoOU@xFb6m^qnzkKRO>jdhw!Y
z*SUVuy21c8%Pa=m?@=|0^VICH5dL1>T@T;hmPb8Y?!n1_1BLPTe^#8yUdtsmQ)h+0
zg(gyob+w8qekgBE*GEtWRaTBID;rq%^A}Y{62vqElmfx$oI`49d&478_X!?2&GR1U
zB<0kUO4vWh+uyAsnA#QFdTCK(^J7JTdXFt!wpeDla3G#Y$6pjNJMtT-9>}APr`#k?
z49g$v&esextJF-^?trLrd37wZV<>+>I7N$(329|e7YX%w+$6qey+O2K4or?Xj=V#{
zFQ-fcFtxU7%suTxnOOo8*(Yf=Pl(D>)gQ7Pbv>F#K0TVH3g%$zp{girvG5q3sUMcd
z&sW$fLWXH1a(SS~Nm9EUeb~y&K#Ocz*0|3l?Zw?)V2`mW%LIh27EaAPw#wBB-gG-Z
zKiph=?ni13-<DTC7(na7nl}WjKu|=6$U}w!mtzwsg9NLz5%r?ryID-#Y%+?njKeOi
z{o5aJ2Cech7H^9Yr(&p%e1yNFt8fQ0L-;k*<uS!ttiDfIWi(EyhXN`8!Nmo3Fl}0X
zxaM^J%~9O)bA7}jPG&~Elh65+#dG&+)M@x)#KnYa3>ab!soty%JnasKOGmE?<h>jQ
z5P^hj19l8u3Pf!-;Cu9R6LCEdOmEByrVgwQOy56Zl=f4VJ`JN(wE%5s22Cb{^F}4M
zkspJ|qFh{;ux<O2;0RQ;73+nCns83-N2M_ln(m7gXZ|?<;QUzBx5Aici(Od$ms-$Z
zpkSc3E^;`eq3@L6(FkB_AVC!-BZ|Po0q!sQ(FQ-4^7B(e46>CfXqK3fEb5$AnoKB8
z?^Xp1BOGP6LF%(@IY+@#fLMT7e0RjZ(<!{oAUdM)*JGd$A)@goYU<P_`zd%G2YemL
zNegc9vQmp7o~8$_2kiukXPTCKJ>zEf{lq=2QGgfm{(EpS(0!+ssOaB%k=hTySDP0M
zj%yvK<d(KUnv5QU2mXE`0X(JefQ%h9Bj^@&FI?`E-wtFrA*IhA5%u2}%BJ=&5R>;b
z2Z@`$rVcZBSEo;hz|x1(i*f|PiNGSboA>pd3%4ldejWrT?r^feC+@4^;T}Kb?mJ^|
ze|CmiBDH8a)b;u)#GL}hluuUNwsqvd5sfi>9qx%%*&HZ_A(>kYmo!x`cUL0yc3wX~
zUwrwFk6F6uzDB{W`;bqzj1);lMasly26HwHGZD7PK$tx~vs8*V>O{BWg>+xjK|%b_
zN6!4x?M2Lczs_pj4e6jg;a&7K1nWZ@`bKZb^@;uZ>-ntP3Ppe{vfc%{bT%Ga*P<}(
zSgK_$t)>RXAbdMl{I=8A7MK*<JwP^D+aI&P+aF?9a>)y+U}MSh$-%5cnuXUj`ti%5
z@o}g#x5n#)&?@=lZ1;b4x<aHGa#L!U$V#Bg<o$G?!b|XD>tE-QD&*NnM;s3LC4AVl
z<Zhii0wfaX#`p+ed}MsY$?ya$+rX-AJHb#C#NflR4x7?_bue*5uU`T5_q3X~MTo&V
z=6l0&9mKBlFc+<L#aL!Aus{)ro%+|$eNlZ)GfwI*+wc<z6VcvSX9IT3FDoH^d@@V<
z280l=j8awA>JtK`Lvw$Ouz55}rV*5I3R%<ps@xxe5(xlOG2Fq!&Tx&0M56&+hcwh@
zyl1>8OQl)`Hc{=)!}<ANFHQe;%Wwv7y8{VaX}>*LmfpC|JU<|E!hg#>_r**>08lZ$
z?=3odF}@blN_d|dT(dzfdO}_4-+JW#bZkzYVYwyK*S>6@_3f?LIU+!je#V^HiV#py
z$Xr2mo;^M&9T?P&-sRVU5}t^jG3{*swk4~s^ly`@7EZ#O-~xA~3Guk+hLwoi=i%%~
zZOw>%$3p%o;$6`^cVBPKA-BCmgUm<%ZA3s7mer6EEp@Yj!&*3W+xbw0%7dx@5CLY>
zo&mrQG>>FCp=6=%y0;@!Ds`ORcmG7N-f~Ya5|QQ9{-oPd#uolDhq=>CpD&x|g5Rk1
z=e&2Hr}xQ+sYxXHffrvlevFLinU9s}Zc|EDIbqJ8af#3snq`R_45iy`j`p##rD|7<
z<tE?A_Zio!1jj#XPL?K{Ud)%3zb2UpI~f4Q`dVgVrHTrx5;V#%)OhA$3WkeqzoOFE
z(m%|nehw%?tw6hwAX3_)ipb-PGD%WY73}^@$&`fA4~XS2(7dxAd+NiQrg^wEQo3O$
z8Q=e$VnZC|jUpX@cgOPVRm<)UukX12=(1=0Cr$;RDZH}pmpW@#EbedTJlfyI0?Jh@
zvMhSigJwguYvG2%8%g7bP_^C5z!DH?_$o+#>GQ4c0(iBG6se9ILWZTuuHKO^lP7SP
zSrDO`^o~~K$j$mG29YZZb+^w~D3~}p0~(trpBHObqGWyqk<w3l^baAk*;@9^c?FPp
zn8&lOwPwi@rXiD#Nfh68inR0!1EghKq_vvS>cVf@g`ZOHR)sfBe|3_p<EbW##g{@I
zU`{0HFv@GA$bXABpC%U!GdJ5vY^<zCKr)SRTaaMyo%+$&l@9_`LPL2X*cOX<PgJHf
zyb$0JTV9$ejjp0f?R~EOOful%e3iU!vbT7~C@Axc#>3yy%J<1xNhw}s0;upvjLR>r
z1)3mdBy=qCU%*X|UM+}LyDNg)Vqk{n2fVd5FvR)%iXHnv)SYq<8t@&ML3HMXOG5yK
zgiym5bMuzd_RF9fKG`d&vn^IBh$FgDg%Q6(_g3)R1yqi0Fy|xetXB(50<dxt$4D`?
zq-m-}RL<i;%S<X3xy%pr0nf(ZVxpoVT&cT7HV{CPD5Lic!CX**O7zT*MhERi2l0cw
zlB+{Ex#!k^iD`k-TtB@8XG6zm+B{?ABW96hTMqC>Jao*eeoh=IRjEdp5&Ed$s(vF?
zHy#d=OT5x*NCD?fVK>oZjr+sjHi*CiDBwyZykMCZ0|frTrG|*W08gkBk!Xz{m3PNe
z^lSbF?d>J4_8h&2N-&M+Yf@@jO#2Y!Fsem=)S2=^+v^Q*xe;#;$PvlWjA>LAi+-po
zsG5)9qlS-;CMe6K4i)T{0#y`%fMYTK4M^X?(K-bKkl7*GMDPpv(NdLxbo-^t>}J5a
zMT8grNHdC2M}}$xLXdwFM#lL|Syw4xeJeT&fWg}xkpSN_<rkL9@YO~!)UKAHef`zO
zKo+8XuX*37V|?~4Ihd2I2F%kII~8W?**8*}<;!IkV^hx43)KsCg^DSrHwIPdG`0uY
zNh&x#EbscOMvU#y?6XwyzCGY3^dW*Tg5)d6Qdg2~=m<LjB=xHXQRdB&f6_c)fPV;6
zH2(X4e|`3L5~DxD{}TNItQ2BL3$5k*0vrkSReY1ha<%zb6U{S~s8m_Euy?fvSViJF
zo7pTjSI5aM#-UT77=m9pYld~Z<`&HiejV<v*Et)eI@*EZ*>fp<<u2xGt8Up^959ni
z;|+TN?0$+sxC@~b)oDhhtnSuUSV0yidrXkv@XBhb4uF?OI)ID}6s6KS5B%O(2aP~k
z3~D@Jf*uA*wWty<MQ27)U#3<-U$15ng9iJ%_g+6TCPUw-sP=emy#26G|Fq@~0N7s}
ztPRRRH5pRS1~{K&l#a1b>goqnRXsvgs5A-S$Palr&{&8K$5mWFE3c}s3H__0npe(>
zM?<Yw>YE2rT>rgQhD56jhRQ{DvdvL{m1Al0Yhm%0Mpl|JpPTL=8)~bGf1}TCI`lqJ
zoSGz`3z@B_BdMZfw5D}keBV53ibzQy$MCKdw`*)n8t4uTQFM6nPnhU(sB)6h@KOLR
zVmZmGl&Z`{%o?NxVWYCOyJ?HW)JI7c&!Oq>j|2vIyJK+|#FIWM`uCYV%@`Hw2dxrL
z^pvY!kRKx+Ug>IL2L_cF30$s~7RR}+_VzydaKXcUdG@bXF`3s3)hh=D&SrRT22Rt{
z9%J2AKPw29*Rh|Q;V?+ZYp~!t4hH}eefz4mEAOU!eiA<Y`1plcj*c4^Y?k6PZs9om
zuEp|2LFu~d%MY(fpJ88negPWbs)s4Oth{JCMa;j~tV}0J9%pbhOk8kmxMF+;TZ*ii
zng&$#JzuSTh`$<eVtqWk|3Ajw0l2bm>-&vu+qT)U)#=!F$98tijyqPzwr$($*tXGk
z_jAtseeXGSZ`G|-t?bHPS#!@h*Bbxv8*|LDqjuqwX75+s-e+$)pT<+5ce%q2?&Pkk
zJ-XEqy6^#Mi86y8_V??=%-+^P_x%hPjn>qNu6t*;i}zVA7S=ZSL=I;CX?O;V{qKCW
z0|t{B_$GEAR+GPNPG-}mC;6=}yRLld8P-3Dz}VD5wPc-r=lJ+8T?-9$|LP97$sh}o
z`7h=EmgBkwH12;1-IRIf^@JO!peEBnR1Scuk^yr7XQP$1UQ&g-%>(hy*t>qAgi8mQ
z<}PYvPAkWbUV)5A3jw}~dv}v%o65xUxbO%-0Z+jeml;i*|5{0PLT5|$W%MvWES50>
zZEqdXz2i|aV?&ahJi&r=cHG*GBV+0)2|7jd#bW?rAiVZ$cj@p?vzNggo_T<%IXS?_
z4;pywr=^L>pI3}~R$Q)$hf*eAup{oG56maFhNe0#`i7eK;;s0*rrT??f3Ng#5)P>%
zAY&I@HX7@02u()^WQ1g}5C6$?W{xBe*tzRIwGCC!-;&?keQG=Lc6eQSb5)#<@hFhG
zY3nXF|Jk+Cj{EixDjkSjiS<4LTAHEpe5Av7KKQo)1~-1ZPYvBYdR4~n9<$2Ol7cPM
z-Y24E{z03e4z?FX>Cf)t2Len9_t+itC5stOLw@^OficKWcj+Zq47!{0liQXV2D$nx
zYJfkoaJ-2ECiMUpr@7I7b}T1{1=qAoGPpmDPPBpgk?Ht}jY<n7p<45R`9d*?v+sCs
zhmpyFvD<rsKDvG~_hP|<yM(F8o*CDrr?w-|*|y{L@*mdOU!oqXyJ+tc#f)abz`U5f
zpWc^5m%FKk$>f7t!N6;DL-adG?7f+>DtdP%G_f^PD9?-)Z8cLZ;p^!b!F%6nAi10q
zC&ee(7xRYGl^;2$ueCA(U_T^c!bp_~AD%oN7Wmx^lJYgQlZ=Bx#xvA#UA3q~3DHgk
z@a;RNTzHdcCSvRny<eLWY23h&qkGqp3V72z_=80j`_^eN@FR?KZo7|$Rsyzyw<fmi
z?Zh~gr4lH!=QMX=qM;wRF?$08RD|}wes~nisVjbBpdTF^L+=UzxZi~jEhG=8V<U8K
z{9x%E`2}AKqy_1;Ci-l7_o<ZmJx}9M-m|brBKN#a)@#k3$E{_*T8-DhSG+MfG$cr5
z(bBD5n#NT_pLo<s?5Mmd-U=H$IxAXp!d9g#ZEgcniKFPd*qIWu@&yQJeRI;h-*cM%
zLp&xUw^|&~_XMy3_dACsY`mTmNi0g_VIl}9d*Dc&tbrb8&(~Zckac&2VsM*fN*>c7
zCC!Wv4mSA1P3R#*sNw|cn+<JVI@kp3F^wz;K_&$2SXbN6T90PKBYz_UhBz^{_JtZe
zO<BqT|NN7CiRJDJ*E^L+cPrwogAIS|iYQhGr0$V2;OB;PqiM?&<&xAMANE6tHH~8y
zJcpnnlf3RnSIs&8e>jV~c_|ER;qsWN6B5iY8@EUsAj}pX!_Aa?&?ZJ&5~zjxUpES=
z9TWVF^t|xUNmu7{i<?kbS!9S4@9${pPggNfTeH6uBK`1rr}zMqHx(J*4&eoiCMODD
z8tZ@vgWFpDTH3&oiY|MKa^Qxl%=%!w8lqX7GIc333KH1?udhRb=#h1W9W~yQ#3TG|
zV)s5jBw%;U6ohDYNDx5py#>QeVqN%f4<@?>KA2^JK+_mPYQ8z9sbx($Grx6fwebc9
zf_q9kgn69TE&GFB_#7;w#P1w{wh9OP6!r`WQdU%s6WSLsjT1_?*=yBb<$+v`c%X#X
z|E+0`KBE6y3vt?5hVY=cqD^QA$^0zTDKhGv7YMDbkz=?BxGpPY)ZkAB3f0sEOc4&>
z!%~bi_K`sB6nuIHsOa(Nq|{J2p}4z)AxIiEk-U5hXG%YVPAUx5nPfOXOBKO#l6S?o
zhc+h;dKS99y*8)NZ_IicQz&x2I*ZA6>Cja_9Ml+Jp<B>;_12SfnGIcNlr~Xn*ooP0
zn2K9y6=j6ZlRRe>*_@aI4jBzett1DDi#^d3?HG>xC+Q8nu<WH`PqfzLr}LPN8q%=9
z=?!CY8(D@yunu{oU`^ZrJuKP*@k2!yEB#AZmpbmLT%~6BV7jcfXizzxofttpEZ4F4
z6F8~;%8IPlir626J>3=R*tPKWE088EIBP-!%;5$s*NV2iz(YsS3x?he(Q5*5*ueh)
zhYjIftO#8K=CzM(zPCq6IMUa2yNsy{=Q%6@AYHPM;w6x~1_-spc}@v)Pk^*<V@N3Y
zdoy;sh8q$183vRaND!t)H5`p*@~bzx%(S=gbhAK-?QPGr_%?M4u14AlYKIN=GD|+a
zSEQWhpEaI3c4mcLB-W#_R(Y_TXCpI1r)YE<J>@_Sq@}?vf06bUyLT*KtU!&$a9MnA
zW%s3HTzFptbTw^Nzbkeee*efYbKn%*G1=4Ep_KhM<!?=-T}WO-U-Cx*`Z>ui>g#(K
zw|zWp|K$6!5+DHHJ|%8K#w=fL_&+^b(%%lo!Ef*fs$VFdH98FFm4o8(-_?ISNp^qS
z)HN?Lqnle*>+xINv(<^@ej!!i^ra4MiLvbmfLGb@sO*)Sr5aEtKbLztn@lir)3Z<y
z<#Xm_0Mo8~w_eP(n%)E;EGxhpbayHn%QlBWtLv^`-LCdayw-l!2P9G-zUio9-a4%+
z54?tm4&9p4#vR-;t>&tLUk0tHv58wvWw(9mypwRH2J&rMcP+KME%-l7wR@3O$<g2f
zfN{NRp0uqGx61iT^MNY{Q=ig_-?5rWFY~5_r}A)RPOI0K(jGyy73uemyI?A7<c$Bg
z!G(mVr499{71WBQ=wwy30_kmm6NFF1_fH#GU0(=VYb?xRXl+xq8WwghTAg>G#{R=+
zyRXtq`&W#w5`ba255E?4uUbboCAT1Y1d|}Z<DA~o#r9ekEsCIG<SK$wlnKl+J~8*;
zXl9c;G81_2_V|O?NZyNWbrm(&@JxcTxw>^t8AtV|`%BUMNr@@Rurq&A_GbbL_SyCa
z$Jg=3*cd&PeB^!CfsV)}gP9+>G9MD;Ci(65A~d;<xd243D)?)C2~3AGu;$G|=Zoks
zw-S1_)84V-^K0Y=V1>3=ClBe1NHLOmQ*kktFJlHAr(3(Rp^=xE1jvip{Fvqzf=Y#8
z@Yg%5el^8gk@A#33j+ueUH<WcM$-yVxL`z+1g%8qYDxbO5__qgJY7YD1H4W}P63l&
zBe+7AzNC@a22%+7bpKc@4l!7edF-B4FL$1s82MvPYU&alIdL~U08Wox1twD_O@t*_
z#n+8v;;TV1F_N!duXgaom3_p)Wu0+`;9+1gfnXq+LUL*uf*gP3S_Cr2UghhO6WIEJ
z^=pT}D%>yoV`BMjsg9D<JA3>WaA^R<C(QezJ_y7oUtUZPaez>1j}VK>dA^@57y5xM
z0c3Q|C|1@7Y?(s9M8x>5dDIOU20&6QM6JisK0QGGbxNO3;{;0~h4QIMlTU!b#3kWM
zO4EMT;2g7Yr%|-O{;L%W9w{g{^$RzLXs8A_C_9+B-)^3elNGDeRqOy!T{=MA(C?@w
z+067eD?ztuiwV!@$xe*XU<m&ZVL#sYFPAvUU<x2`mpGEb$p*s7bGj;a@K8hRMm|e+
z4QKZ(NNGw*P4z6%WOC6;ZlD7tslW2*EhxaDJ13N<M>g<3rn(ulHz#L770aUM=h>aU
zz*agZe}P>J3&jWo9e0;wc(Mb8Rp+dEV|9veVsN>mbZG5HPosA8+n4kI-W^RcH-TCT
zrz}@4#v}nVXl_{0gsFyDvw4)Y>uHa&h&;e1Ko5r;h9Ilpq4MMw>HEWiq?1RLON-Nd
zjpHVr$t@^FW|G{7UB7@{cf!*86Gq!GTGcQKOawmsSGrCl&OA8qVgms0!WA37HTJ<<
z_!!s$<zZG*V4<_%#QamMd44~UNEn_f6fua9m>1wf&C^s*hi1DiC@pAKmqN81gNGo`
z2lQ+zo{asM^9a%>F8tb;tlVKei&MrXKEl<~gcOFp3Kr48v;H2K^6`0H)MU~P<4hD(
zm5-C{PV%?@RIc>0fdCYq^?uhC#x2}hW2v_+?Oz-NWAGGV>!x@!`3%y!4_K?OK`oBv
zu;|%bM@IYjQKYeMJ;Y-G$;tl3ODFBP#63zSFI*fY;+g{{TwFbc@=C<9r9Trl;FgdD
z`t`aTFv4#fQ(<Y4Cs9Rh{h0e&gnQUF*wNM>O}nyYT&<d>HVt@N;xHkw+YqJ+_!ekO
zgG!erHth;ZltOE?g}`s>aMO_+8YR^zY;o?~%`P|12-O84yS^Ra7V683*_z#k!f5ak
zB`Lrq46q1~QE70otjrp$cQ&5Nf=`L~4~sMKfdr>JQC{(qU{XP2Kp%X~vG=Y&r9`g_
zkSaP3g2H+0FUjumYRA0V!iWC+QXaIBv5jU6m#AdHL8R&bfyVoVef1BQPUVygJNciC
zkhs}JdbVj|bmV9J=;_C2gK2duRj>rr_%91q?Ij8SdMGEq?<c~+qnPrvU65PGc<pk@
zA{on}<#)Ut?2SNS{0+~nPt-ML0(#tD(gm4GlE9AfU!~-!zc$T>V=w-A2c-Z5KAPW|
zLBIfnN>bagJvpk|#HqB8O5I!;5w~11_h-RdDD{~w!FEkJ-qpLYCM7j+Ja-E0<}zEq
zJ7cN&zpI|%dDRWhofzTf-+2n9ddb~yri*Yk=Ztu`isE<H@byOPuTV0P01Sk?{Cz~M
z&z-MO2klzZ#Yf!F8Q~6b1xuidX;Q6~(G>p*a-@nQEHZ<GYsGM}m6<8i+#|K~rRL(C
z$kp!ddmj!$bdcEI<0=vFUa4N~q}16A&-37McFqGAU+10|KXNg(1*!4ZEQHXFyPbpW
z+cNI+&wBrBDwQ2===Ljs!|O3ov-ac~Do1wS&~(;`idF96@?+zN-vMBkQ{!xlA*lQY
zmKc8vj`_w7^qkw-JY~ba<%;1GaXGwUdJ0M-=x(R+BldC7f#v<;4SNuqF8`$d;<@nb
zzw}tg%>eg*p{~(u)#@+(cDFgev6*QQ*tf?PSx+6HCO?)8H|>CGaAlU>EV4~hqszT7
zcC@k{F6AsuWa7jw`Pgl0IY0RxWtgpAKB2_b!i^}WTaO{m^my?9m|lx^%OV;gWiOkv
z%dWaJzJ+y4(gR^-qAod>#S?T?6|nWXomm`bc&JdaZQ|3)3Qb}RG3){Gi`~QE7Qz1z
zX#5}^XnG2yvAS??boBC)jG@kTok{>^=Fh~5vTVQY^g5(?Jr{L$+MC3UWb~Oc2Xf1G
z-0<!`k=i*jTcNei*-Z4{)taiEHh_t5pCu=2)9fS(XH4pkB6_&r$k)CHnt;`ArwRvS
z4pUss={yq;8Obe7a6<z2u!u|uBkqQdq)jK3pNx~eM7ap5QqV-;E8^iQlu0Z>QB|Gg
zP{AX%fmidW#>h~R@-EW!>tzi9?p>oYohJd?+Jj3&BhA=@+m(*GmBRWig@}WQ6RjIj
z)7>`M1imi~v8wFDjThkQV8oK?JzkIVcP(^pC^Lo=TB8Ru2S6W4-poHQLhRoA%gWM6
zhwVbcUZV(0)GNAsc;oxtxu#ZeGm~~)V707-G{`bbm@FiNQeIRn%^izaQ2f}w?>cei
zZ8#O-<;Qe?)v;nF!aCq8w?8Xeau~)vPc)35L(u*z7|C5CQR0;Fa)jSm{lJG>7IZUn
z=RjAT1YeHS2%yGiWTv|(q6I1gg;}AbevK_UT-QfC1_u<MF*xO4B3edzN%8K7<zqDT
zHKEwWm~o?d&xy+J=d%m#!|)LdXj=0AM_D6FWh17IDlZY`+t|szCX*Ezigs)BX<COt
zPZIkrBaZP9q{EOMr~Oft&Ty;fT-J-2AiQ?zh2&NYfcU*v_*}CX9kH15+%fXRBrrs_
zRaZ61vczgS@<cO4W;QKMVu3H2Nc>H3XUJ|g@sng3ixe4^<Qa?K4&o<e1E5yV+vvzQ
zMm3V+C+kA7R~a77KodF=!0WUQq0m<=x>2;v>!n7_IaNJP>w(?jYOflTcE43KGp(=K
zNphhD(08yiPGDpH@Tzm~&pr4J8BS+ZBNo9Vh?7-ZH@1J#X1Ng&uB}`#mzP}7<<J-k
z*GN&~vv~>I2+YEX^T<nv@KDmA0mgk-4dHOh=Wyxw$dFIrq}fUgquT<3Z7*S_m(W|5
z+?LQw2SB-oaH>h}rq$EsSM}iOXKD!3MV$f^+oS+Z#OEi?WW~@VH4;oZBchgR<3!{s
zfBQEM>|g77Ivx3{2ot==L2s0R6P^~@qZx4gT5bN0A5;AJ-{E6g7iB8X8Iwv3kPehn
zgG{vhAvPCNK*$9z(>`=<44y8|6^>|$=~}$K^ANrjUx!_XqFr=OExpRA3h=ZMk_&)J
z``gH(2-QE0Tv)*IqH*N_8%g`85r+8Dzl~6NN6M><B}q}yL*kj%S(p7#g-&DGPJ;et
zihTPEcZFrsfNN$oZdB+mz;Wk0tbd@p)k0Jwda1y$jGL}1WkVy{ulLs=>KL)GGL?}w
zFa1dv2bgB#P8+fK2$^BasU*$$+v#awr(gblo*fave?E^7)?2;scjOO|ErvQR)B1GN
zdRNnWys|HKTC_7H8G0&a1^grCpw(YI;wP0y;^Rh)?bYAWf}2LnF|}YA>sQJDzfLZD
zKIQ>f%Zqu|utuq3l~z!d+JKi<gwn;Ts>5+en=Py2wjwEO_0S{h7u2n!LP{r`B1)TS
z>5LcX5XJAmYUi3Itcxd<8yyEzXaLW50-moGcPnG}3#W6*s*)c1ZcPb16}_ip)%L-+
zMB0dHeS*n|s8x`oAlwz}%Bd|%kUQXa2f$2$Vy(y5A?smGO?uthv2*KqnbOlbM*Ck`
z7M6~V=YHw(SwHs$trx5>R1il$UYzKipwGm_;ClVthbHh8LOPv=j(@jnjYnl<g?q7J
z`c;o&B<B0Ja}&mg;0AsUxtOoeHVEhNd`Z3riupEQmv6bJ2R?x(_N*1<IsyO$yUBXd
zhSC52!ntzs_^}Lk^$0;cvexYLYtKdf$MD7m@m>50&(g%6$E1h7{E+kTpIjfy4hDUX
z(gTy3`kuw4Jy+7<vvh(6RII*M5N&sO(6AgKKMDj)b{AcTB6My|m>yWw%h62Ds;Q*Y
z1LlV<J&>!4lf7MX=-n`(>>_~4I_iU?^R0_L(`LDPx+wKYpN<jKl_`URH;}`SwJ}j~
z+lrvBX)Zo|CiKNpCkG^xXO#aihw+*nwT_-2F!Uy`wK)tVIhgZKv#(=bo$lVfhLRHU
z`sqdq0VHK{+l`HFN6c{kpyKb2kZGCxU|`FXAbK4Z>mMz63-U0dWdt|_P^v2q*utRL
z@D5sQB+~b?*mOM`lYhPw;NdIhiuF&2C1AJc+uvuL5`EF9o{$@i{?^DsQl5s+PGZ0b
zhIWF19UlLcG%F9xaV^n_C$uKgb04T!WJ|L1ye9R9!WoKV@%%!8^W@hG_$f&R_IO*#
z?<}k^1#Z=^FLWDMTmXDunJ&6vWkb6zI!$6cWf-WAJ3g<8UvsYfh&Fu#ZO#|ovI(mA
zi|UC)GHe_e@&8ObNH^!z;nLX%B3)(vT!6+m%p}$;=A^6n=I-vtrQB|h($B|~@abiG
zxyjuvgZEFlxyRVRcm`bF^wc{~GTP<O+bjSoU^PqMFyRk?7_GpuQ^!gdyRwoK8|?2&
zPY8&CO4O3oMZ<F5_Yw`Wd~ztLV3uWZZbDW+7ObPyD+cJuk^&rqp@*%+_|heOsSo-1
zc#5%Hqg4RjRrE{wc-wj!yj-cz@oS*BHJ9JIV6F|D5xeKQJokbtIu(wr8Db%zp=5U*
z>zM=i185NNUCX^8(Vco5v?*zJF)-8=oLJt<BH+Y*ZfBYIrf#gCLM6rLa38q*-63$5
zL2kj90Szq%<NU<{jofk^TmMH0RL6?s#wxSYy5wLyabAyaT<aU+yuye}ti9Yg=P_b~
z-7?>Won*t5XqyCm|GO0%67dBYq;jjxyQ!ey2jDw{SmV;%R|0ybZ-s)YOUK-=7i}2q
z+M0oT?wOGA3-(z=+%bdPS>9)}r;$&a#Hso5X8glZoSWW@Ci%sN0-myjE0^tQFD?eS
z>q+Z318Bo|AD|m_-)c%km1;5e$GK){`F!inO`H4mPnJX}y7B~p<+KjIL%Z2nm<@2k
zTm$ZK_j0RqrklMYcRjjj#Tw#onqNDvnRpi3dohSPy(YgB$&xT<()1+CdOa)jt6g^L
zS(=9v;sLyjE0LmkJy4~F4F)*kW{y*XG;UtYJp?#}LmAFz)4fQzj*z(&2uQ*nMH#LX
z+tTS)j+Sdbpdtq4GwL7J-s;?4Q3M4<hyYEg;Yloh0RC>AcY}Pu?5FpYFnk3JviXa{
zHNXp7122G5SBJ$UipdgozSn$7<;3sP;$#!kLA6Z%yK#qx!o(BW(Zv9L<zcQ8IyXcl
zuVXp!Q&9jRv+wKuML#}O4JtwmSG0V-Tpaa$>k-xl2qxQ&>i~y{n#9^|Ug=0~1Ylj^
zD@C!@)CwxU%8>RKVJV1sdYhKqz-&+oc&Xnt^QQhD@&uGf3CaBuCnilsF_xBGl<e#g
zUBR{*auo{Kakzpt6q^0fPotR^*)0W=r+GSQqAxm<yy?U;(-*c*AJ3DStF7GjN4}CW
z+ydyo)FtOF)?v9=Q}mysP-9A)K>_^X@L$iUH;8bA{Xu0i)&t=|8MDP{lo$zrA|a*F
zIC*^AT(Gzy)NRDf1vq{_E$Z-&b?dI*S3HEa31W)RnZ<`gIz>Sy1i<L}zk(GCP<VK`
zE!cVY2kh*r4ts3GN81l(8QXKKdDmVTv2=;NJG4Ci`qSUr=xa;N0Ql8WCjw{^aO4|)
z@Do#Z2zKG<hQ>17`$cW-;a<uYSN^jSNob1oF;4GJ>}|o=*P+f}52-6(a`8>#c-y3p
z%jMd0+#gBMw@1A@2-%N_(5DWapNJjL>Wf1~ya;s87;?2MG_lso+{Id4b>V=k?wp#C
z28kW@_q{a(+MxR}6pDK?_7ed74$av>CjbAqZHHNl|F~^~zsNJHIJNE*^2zZDR~u?}
z=i)Euwhy+OcYfJtW*O+^iDbZqhHzN6o4)X|N+HuX2P7Q9@-@zuBC0-0hA`lH{fskK
zE}}FkR?!g9K2yW&wXWp;G=2uW&Yt{QC(`#MrTal;l^>A}@P-r?ofeRpm~kc^xnoC?
z8>mXw%-AMn_c4^I$ve`|Jz}al5`0n_Y&%K26?iffZbA^A?yQD6ZlxBHzyD2-4o?9d
zko~qz_{_tZN02P=BrSF4Z3xQcb?Mnp{{!r`HHcC)2m7^kc6u_z_W0;A8T8}#tADhg
z-Log;J<7*#fVXqOd`WY9ntI}XHt$`>1yJyo_7d-2!>|u9x83(g&}=2-VjgzG_e}ep
z_vVN1%B|by&zgh`9;<AhF#iN6^CY?2<^a0d2r?gjly^V8F9pg~$KCF#9_|66(5<L>
ztc$Z@c&w%RXSSyzFs4F~*xh3ZeK0EsxwS0_*HXg(eVp7-C0ylENk3&mM?%P%vbW`*
zVUmO>^Fa7hT2mh_zf-@F3#NMA2`HyPo8duHvT5LiU;t}#OlpypE|l<ENYO#PuU17)
zNEyH}V_T5S54Ex2KiAQM^hbZcl2nr(7bZ+t(8{@Qbvoto>)AIZ+sh5Gi2=t<DB_Gv
z3v(0$fDM)iK$_2+b$<B-VW|{L8pLS00)Sb>@Kh_bn!q@FIOH6DpHzx&CVm(e=-rRF
znsX4gWaf8)?!SDwc5?7sgnxW?1@-hiHxcPa*Z3S<TEmclDt|wCcHa=|P{^kkJnl{<
z-Yu%R(d^4KZ(_ArSud@>US$kzGF{a;A++)W#As)+>S=)!^ExXq52$6$By--!=dZg2
z`FYOhfV%0S{d|iqG5P5c4Do^69-gqIUSWED@2Nt4_fb2OJ5|^@K<vx`tp(re*@oat
z!YX4=Euj7;f?Y63ArOu0FXPokW|%+(YiyjL-#Z;G;}zhaS`a4VH4-$Ku*Y#x<s%LN
zFq9-F^SvRw1)a=h3t1L2!~x5q2C2)wvgIbBRsM9y-zWF@7Uu7<j<BkeC9B~-9#66E
z)tN~j+51~q7$_J~qcNE^WYFWuxmqPTV(5XA)#474Zl3!zo@f>9Z{pTJ9PbTc^Z%@^
zwLJ_(IL2oGU~T?{Yru>CzL9a;1xOwN_~e8I$ZmTvaXO|n@71x*Hf8(a&{f^6r0~R_
z0dw>GeOs#JSmAEX@T<;llR=kVPr$jV&H{5ZgpnQfG~U05px^lrIu<?DH|AE17PYJj
zeZP06tr*#$Lwt3Fb0|yL*NlShYgs6lqHfe3B9u;$qNsJ5)g)9oFWe7P-CyMbfaFHS
z3URGzPDJZxalR}65w!PkUAiv!U{2-k;IsB{rLujVZqzE;>CFlIq>@-b_{JuV>g<_V
zq?ceMef~anU=(Zih<EWBNl86-()3Dw$`{WVa-V-B+SEo#tJTLc^q^s#7GtT4KO&vM
zO1z!F*wlkhEg?A>UC;)F?tG~WXt@dGthW<Pb62-8Peb0s&O(+0Ym0301fi(Ua6dEk
zR5<#&!Km<qFY%G$hQTSB%rWj~Rtybn{RnCJJ*y6pb<3H9uv#JWdzqbb$N`y~2jz+B
zv#M-eZ|}FPA31Mt-v|*_Oo}XUYKnQmSq!H&AP`o_MdT4yRH(3|p}@c&0fFw@dkf)C
zKXGf8d8?lAodU@V@_7YoK=6lGH2d*q;S<jf@a!69Ri8%I+Jci_o~MYmBIW9we+W>(
z=ZXUoiY7__nWr8yn%_AtUOr`{kS}$JA2OS4zl3CnYwAxFPupdd;w|uZj&_cc<j&#t
zeilV!WqwjPjcz_&`~U+GlvVJxu7uz488v;6%8l0i!G15UH})Gjn$hdqv1nIrkR<#_
zB45Zc+1V4@2?&YV2e_=T4C7C(d(k;a*58A$Nh-Prn`^d>;UX~LuDs8-JjeJt@^``N
zm`Li%^W1WSWYJ5?$n8jm5&Lr8MOQ)9=+5^J-wdC)TaDLqYfS(EN7d#VCkrObw#p|{
z+jH7S{HRq_f1RBB^&CGQ&f}&q4So4R1YQ@la(z@u4bHZ*#U7iSs#BJ{{Rz{q%B2<Z
zt^1w?Ly7W6(P=%ya@Wl;8P|FC2c61Ze|cPgb!!9EnePgLsg{lE|BQF#QZN6t<9Bgh
z0-ehEn8W8DFn~>i_Fbq34rId(lw9y5QNvAMbPD|X%+MJl_?pK9D86mS^&Q+X>6*vW
zp>(m5Jtuo5m5iV*mTE%fF!j*`GV>UH&(%mg<yvGi!Py|{@L1^4P;u;$#<E-SvMsP8
zE3nQD{U<D=Bi-4W;2$4XUY;{t4ZzuplnZmG=vb6W8X)!$<JY1r8@kg3{-6Hr?i`N!
z-ktbW1k!t^$1KiIIDHzuYmWJ@OS#dWG&asA0pIJ}G-c9E!*zLfcVVX}hzpHLwz0@r
zUw&yB!ECP(DcIVAbfoc}ScJ$n9DT<iHk?*jaUNZVVv{KxXcBQNFz(5@-~xL-h98cz
zFtN%(^Z~3cT<=IWy62aSm7jdpap2=r$p}LsEbAgH7ewt~;R+xZu)L_k`wr}@tbrI8
zQGElGt^3xuuTqi;M;A6)Ebfo@H;d+P5)CsCH^hcAA=`vdVKCNVP69Zic3oCj9##Lw
zcaE|LU8shTv4_f2rA9`;vXjiK`imUshy~+5pq9z<^Xu}G@%uHCb;wP>y<EUee<p#e
z3qX8Ukhb~`J2yC{ictza>eef?b{0aoZ^bep)yT$$WO|y1hsMfwP!>v&xYYqSGR2;F
zaZag@*tr4WuQ{$E2b$vnum3T}DAqu8Ou7U#$Kjkna~$ppG{>Y%0sc4uV}-$C1>G&z
z|ITYtwdvqwGc{x1&ul_qseoFLQONS8a_GOeoF+yXruoN8@8H~<^*JCQ4Ey6367>f0
zVGOLB-#(m{9E0h`P^>P|pXe$fgnEM09ppyFfa<D|6R%bVof6g^49Ga8+TaFKWg{=i
z0#G~cmal!G5isw7#+v)$=ynu7^89c_yGi<}4`U>?uh#&Fs|I9QR8orMd?b1+<H+iI
zR1UMm%Gg?~I^BH6bvleM;qAqwwBJHkq~#&MW~V|c>cvKAW!~Pcj(I2O)Q`JXngjm3
zjcX4>0SkNra>fWn5?Yu-9XTfWr;YmxAUL)Q!*7KGJdKNp%Lv?nD@<Yg@9oeMA))Y;
zQn=cqzyM(iJlxwGuhJC{>oJJW4CpbmM1}DAw;X`efKqDF>@Z>db2qdDbMYTMpN0DG
z00@qV@URUIi15IQC5Z66GY*L3Uon<<@sf*}EO+s+bo&FXHGyyFNmT+4IT>(MPf2X&
z2iR-n+8yVA+!jG_PojXpa!<w$AaGAY+`j(^ivKIT`@4!2B@>u>^R4IoV?OvTsx9r*
z;H2Es^~J~(mzS$Dv5S5)kXDZFu|o(w8v<Qq+m}RW0|OfZkV4#-m#`4PR}FatW_HIS
zL=`qfnag%)<~#m;0ISmaX+!}QfwenmF%>up;HVf{!uh7La6pQwY)h6(UH?LYym1gi
zR>WA{n(2q8(<pavj`m{0kSZgOQDtB3tz<qbm6(8@i$t~0BC{-=5&6@M1Ku-?N*`zM
zy!H62nx<6cEYZBtbt$Dk;q^v_lz~(#w2x&AdQ-dn)R*8Ia0pqw(!ep)TzFuzJ*r>5
zRpq+2@uef10vFp?>8^Nmg_cA&kwWs{wD<}&PtTs&((}`;y=fMhG^W3kT9ttiLh`Y6
z;@BUB^xu?7x#>Nda&ET|-1wo*BL5iynueiDmOY&n=(F0i@dEZJtMLq|4SE;f@W2b;
zUs)wk_Rmoa3G4C`+#(>LO!i@i^1TY6{fGPx^ao134d2_jQc0V<Or!_Y@H6eQ!<WVf
zH9gH&B)1TRBQthhK6hwenAlzMaVCwR9OqhgvWhZKxm$iR9U3Rlai5LZvoKKg($NHS
z-zQ~U#)W{-bBqQC1FX_1So=99g|l;%$baZD)^hcVnLqd3HhDA)rUn7OfnMn830BTi
zQD?%JU5{Hp#cz!aCyo)E#dcN$LSBbLL|$G;^&itdj}`Q9B~5Eu`KzP}f0gt_0H~zY
z37Y-wa*F}Fg%kXFDrvD-=Qp+o=8@{?nRkkS+=+kdXUY2ayrB-phIDBYd~JmeGvV?G
zLBsN`?_fgH*&*oM%iaH-jdPfNi#f=ppd!eT%RoVfwCa{cd;W_3<s?<4RW(1{veVI^
zw6bZ71^i(<=~_z{8_!P4akas%)73HL&+p2Jk&hdv>vadf%$B(-iC5&$UVU}N_^3b2
z5xia~4G|YZvjLdLJjydYtWhD)gIsdzfd=k4+5@VXeb%V(p&H8wLjJfe>;3f1kpY+#
zwve;@MWQ^xo2-$kx$X>cM;1RHP5u|30^gl+#$M)6emlY0xR-y~WMv`If%@e>p#cH7
zkBKLt@*4gb0fb25E6r;ppG-U@r14GGe-B=#NyiSRB&e@UFWOEA+iSv1$^Z~#<bb}Q
zuM7}mx+wkunGLD`%P|x=jPP`$|34uH@c&LR^G7NAqz8kg6T^S7sOP?}5XKE?@n<%U
zkIopnIl&!?hNZO+BXz+o#o$-gBiT-f?WE&_t7a??%8pTp{Y`>ZbfEGH9?Kz6^UoUB
zmG|aoO#DD4mhbdiTR0E_mgA0?z?38nQHs-b1|&<<aekj#jgJHcG?kOx&8GB?58{qA
z)G|##b=ko6IOMOjEB$m?bOC6kBO?4HgT{rDtqrK#J$Ol_mOxdSNK{VOaypbj{GY`&
z>~mD3G7PKH{pfVrL(rT8L$g|Ki629w+C$94-SdquIBuav7Gs_U=62xyQ>rN7fc~ix
z;-1K*s_Rk)zZ;2$h1`X}s;9nQ-rLQyilfgzKTw2ml%<f_5lZ{x_|7@(>zj-zgNkEy
zTM!1vpY9LmHuqwAwL>dHet(e+6^z44YfdC8|M&$zXKb1cG=n2w3{Y9vVj);r*{ZjX
zhDJJsg?~)+k9^thH6C@`_0kZ`1E69p$jD{+7QOcVm9~3%ENLLniA+=Yrp>TNWXrTD
z-J*Ue5fkpPEdtiV%AxN$kWR4e)Sto}G!!QJrA%Bl2qX1XGd{>{e1_r1g+V0{xUz($
z7q+vSw=YvOb&g)ix>S_Ka`>}1uNE{5WbHo5ZdQJh`u1CHkH0UG0vrMWCnPhhIIbD7
zo^2Y49vE~?{~J9j{)HY)f6*i5FM2Tj3q4p@IsF>@R=v7nc;@rgViNT-KJq0EHa_=O
zKIVVP6oVAg$;32zko=JA^m0+=rwWdeHa^L#<x^Oe8(oGfEv1`o0)^4@c|RE{2kbc+
zoNkW%UUY1&-aXK`ak8wiTHE;*)tzA0AC?4L>|*uJygK2sTJshx%dQ-dSB1TzEG_g-
z`9+P(%K!Yeyisp>%q~?+DjI@#n>&d$wU*sD=%tiiC~d0KAsluOlXrp~A)2US9ugb<
z4C4rQKEhR%GX3v<$>{sazxyQt|FdBdhn!yWZkGH=F#kn0sq4DD&e3W)6gs&eamSf-
z6bicDj?M4#s@{&}DJCu3=rZdNq31V0aB`nwfRp?5HBFNQ7iOXpels9eg?voO{8I0z
zWz&8HHYV;LFV}{E!UWM?oLjaoh1}gr!c4HMez<ve?0~=ofIKQnGfP@v<t_PFCirS#
z;M`%dh;qJ5i*(tDg~c@^9HY!ZpWE;SzuG0Z?ypKddp>&}ptbk$6j~Wm7mJ1K);;<1
z=YxdNm49X8GVaj-S&64&NshILAfS}O_9RB$1s}{bcO$@{?N2aYE{3hxT`moMPGe07
zbR@s$V$1`grLj94h$&SS9M!z7Jr>E%Ne~wP73Q&Fl|k25<I$9{T_)&7bL{Zj=z=z&
zK0MwLwp(XzZ5z+6m1w5bQ>SpBWb1c|&L|Un^|Dq~eyiK+pZI#!0$eu#I2q;(vh(&h
zfc!EMCkEbpI;fQ`P<L!y6t*hpJ82{y0ZlFhFiGKaYaCdEKu+5MmkY$4oPrZ@515?t
zBtBYDm|n1Vg92wW14+;Fl@P9Hhc(Vy)%s_sIryQkdWK6;4;*aVB>B?!z~O6V&nNKp
zjWA}U@flF@Y-XgRB(<4r#K<d~5iAQ9p=~(k`ljiBg;^GCWRPb`b}mjdq*DOBcEC58
zCZe3oWuXq5YLISnsboufZKn!sDFS-@Z2TO-Yy-XZEdAz>mpXUb_UdQqhJn%w>$mqi
z;P#yg3*C>(BVxC%dpKI=uV#Xu%{*ytb)_Y8m*2pqe$Ng>-(Fa>^5Lo$Ihh|`vIbWN
zHttEDRfyr=eOoyk0$Vh9txaGXMFF%r3=E=Y$_UC13imLED&LCK_o4Iv^U2oMKs?x8
z=M>L=T23$02=-(b*Bm7@Bbp}2M%V_OOAq=Bt=mFe&tk`rP7loq0z*EfJO6}yl#(>&
zKg_JSST+;<f#on-aqMNV`x)81h@_1kX;csg&Xwv`jLIng#S!O9fD-Nave4E1{{RVs
ze?UT>q%cgo!IVFiQX#Z`(Ngtn*t3U_?yjj!YO(2kql)rGM*^XG;;&(C`}M-FXA%Y4
zg99=FA%_4`6oeTD+A~Y4=n^J6dnr(jj*@E=Rkb0G9s#6?-bo-uY?}iq!oitj8Zgb%
zM*SZcK|McbRP{d?(Le^o2+9A05ja*rj9{GuV#E&)AV&Of0b&H}91tVOfxrGnb@D@O
zH3W@m5dI13$hU?54eH>CDv^}pTtkb|9FGT+V&I5^a<@TOQ`y;h0GR{uA89hGWKGbe
zG~vyY=O>FslaM^*GIf<2bFFzeKjxuS$HXuR30uk}1rK$g7x_)cBNF#nR@rzIz7SU=
zk$F{??BGM=&>coB;n=TDAE-#v)Mb!cuaafKtB{z;!U2p{dcbty)6V>B$hrOOiQIar
znbm%+MF12m@-n7qOVKBox5WrY6H$$UBylbFmn3dV{);5sbN@F<#OD4biQ#YmCJBU)
zzaW7?`4=R_{({8tw|{{I{J%j$j|K=5dNlV}qJKfcmgaj<`1&^o06v_;xx+NnzwQTH
zns1F{YKYCbD4?fz*MQKYKRCN&7-*I|WEfs0J7hnzc&Y{^XcY8j1|(?2FIbsGje;qe
zM00~MnM6l*M=G-P<5aA6_f`)ETFxXqRRY`fVf=&J^`(tN+x5Xb{f1`6WvuDd**soH
zLdL&!_~84gwgAp3|1yNoUxx7dKMb)P1r$#`W*|dUVFMXLkNGb{cmWwgkGbUx(z8a>
zsp0ryOlsmQsOy}vMNYfEp;yBZa;Oc)U-9gXLwsd38i5M6>4{d3-4-E)+uf1G&ZepB
zo((}({gczlY-EQov-D2S>C_Dipi?jMAU=)b@@T`fmGi^kWM{Ss10$0|Qzz+x5?)Q;
zc)8M11EC3o@PGx054q1+ZX%VHJWSCrJLqkI<IXCW4lxFG5C*L&;Da$wlS4SP=x4(o
zWu1~&G?yj~SuP6vAkbnt(m{ZUqpp$_c@=-uVR(+Lb1y2`s%W+*>%H*<(9$mALB}V+
zU@&SKg^#dN@CP<&GKa_R!K^{(Yxay^{fc;HVb0OIx_PZMZeIjenvxLDS&lPP;9L|G
zF2O>@9p1khGkxXPI`8yILZFk0TLak378;y1`?t#&z|2*1TVtXUt|_{R^1oXjNsU>G
zyS5+RJa6b4)U*e@|4jT2U^1G;_7|rOEDb+G#}@G{v~71nJxRLUR-+h=Z>wz)cdOYe
z<rNX?#wJ>QxEsyxH7eBH&grdhhdKnw=EzGiuwg{h1SiK<CDp$+Gs}X}{qjQgbCHuQ
z&0neSffN@sW*!cwEXfS<GYOhUM(E)ABcB^**V0Y%S7lmX*OzxKfP6+J<Ak4GSG1o#
zJ3TLck;XR!4TR?ji&lT*O9(guooSyTnQ2-IQ3Tgy1^1r5fO)6hz~Ff&5#!K#Cq1vP
ztcb*8j8phelqNnq;f4>24SLM=Jc&$)PktFPDfd2W17~V=MHZzwicx4N)ezf0+ES<v
zf{ZY|11}Aw5s*vZfPg!zjp|>qa#BAN(_b1cv`Y;lXn6YeES;DqdWZ6aN_RFW&p?<|
zsLHl)lV3~9kzZ;-+p^$+3#OljN$#QTt#hooDDACeiCpK0Xf{1|Rp<U$pLR_5cPo>a
z551S)*pg<02;H1LF?xY9^26J<)*j=QEli||0XSU=9cfrbz$0SYhWojUV<~v7dvPsw
z9p-P@Wrrh<{7Jy@5|fp7J5B+2N=_Bi+d6{y;MeK}LmY)-Y_UK3Dw5N>`Lze|I$d*S
zMLl`(@s_T$vLnmcVFF&i8ss=GTUFko1|qDO_$$jvHL0s<lLZV#xHwOq%WtLEsnoIu
zxz~(|q0_x*0Ea~F81pDkMk!?C6E$+Pg-00GDWa`|o_k%UKbQ$nD^?0q^PU8CQ;c_A
zHMQpJ7u;#{AO`3`R|=15z}#p9QRL5+<rNreh+xi@UAhAWnFy6)e(FVf5GY)R@UE;B
z(r|bDrZI%h94D*zF^t4>?w=^JRs;UE29$rh4xG!a2aubi7(iqU{kz?)K#z8@yw*IC
zj498Vebj}m(cnt_r#nbDIwE7QI)}x)0atZKdmEL3kO6Y{*@b?wtM~j{RXz3BMcnGK
z*D8Z8BzvZI0k-%2BVeQ;Tzrg~-@f$EVWB(X37v=;?mMxjP<W%86&nXnp*Qjom=qgu
z9fkmt0kA~(V296j*yRvC2omzWchB@^AIb>N_P%4<P0BZ3_Aw;GBV%e6V%%3R00nYU
zUXsVvz2Q$uQcP+M4mfX|kZ|WIR5&$0r>Ps*B#7I3JuEFEUtW(k9;kG#!;r!N=4=R-
zw%LhxCHa%0`Hw*>E{77WbOWNW^qp_)I0<mRC4e;PmTh_VOe?Bx5lDnGK+)=f+R^Lj
z;jQ_KGxN1W9j@hA9-e!XKHNTUHDhg=U-*r@<NR~wBY0J3Uw}7{1GB53DqC>?_|u#)
z6w>eUz8-;>-6+E{UugK2O?Yd(UT@}1;?z*!x;?b$0KyS;+kK`ns<3#+L8i-=V4Nds
z0wBf4=%TOj+8$Orv?b#AmmO%qIG#xdYJsm$QC^}^^9bw=pgfbnVtAS6FrqAu={o^i
z_{F$w5W5^=ndad~hEq-qp<8N1wNkhe76!AP(>)R#0%<A4HWmWEzR8dXtxUn?P}x`v
zx~aS5Z<Bjq5Q6O6vL8EmvTcLveid)1JOqrJNz7(#gnug}rn7G=lGhCQi!wI0t`!Af
zc*=y6-bf!N2EFjTWv1M0(T?V8uwrEh9Yav(5F_#iCs>sWx8g)r(k$BoijrsuJcWF)
zil`f$6LTdybDt{T9lOW8A*thXoRlwQi7%Dp(qP8uZCj)iYCxY#%%E+4=ZE?U0wBG@
z#oYT_4}>bI&%#X<$}V_Zn%9^$UQaHBkUE82GhC3iRmt;AD2cB3l<9ur58{#NK9gcG
z;mOU%M(`x$w15Yu95sE?MXXkiJL?#{>}h<ia38yLJLY~>&-{Z5r=syOJx0REa*M*{
zY-c2W%!+f`gs8ZTU*J}}X<x8R0<e`SK#u?=vTw6XI60l)wYyW`Mo<i*0&_r^>7q+l
z_fjXcIj&K5k2{*%!gz~lAK`TVG{yOGlo6m22+tJrRjxI2=a_>aD;aHKlbw&Ta)H=q
zoa3^6?YeJvBa`X(dNnr>PZY&pb0`t$JW4Uoo++2?2;0+Y8bh|N%Z3sNh;SZNMu)F}
z20Kb<H4q_AiJ_t9bRJ!6RPIJoa0nirg)m4BiVhJ}?#G{Dtl1n7(4mWKUXarB-IjAg
z{~<s|MTSX8fXkSJykDHDddIBu=znVReeOo%AihbV>`Rorytlnn{nErQhcUXamGzxE
zP;dS^aJMg=guZU?bnxrp08|i3kW&6zLir*#ul=Z^X7Jyv*B5zvAV{2L2z=dnsCp2+
zjF7v{A$N9MO^spP$~_WZ3ivQ@f8={k8`^t99HJz%et_Wq{t1Dt1>jf1VhOGYi}JPq
zm(4Vdk}(Qj?{(C%3?Bzm?`UTCCN!0LN@w#DjI2vS_&vKH>;WK@cL>;;@lJ{smimT(
zEf$bu+Qx}3-cv*Km}pZ{k$T<G_4fU3vTaJK*jJVBi9(0_{N4TNi^Hmw9^zoGFc&9y
z6!zm(ASA4}g^2NE_NJCl-u8rP9b%)$R_jSfK3{^dn6205n?GWs3^tUx!t~tMH=q)g
zLl}xx=EJOfg9nJR)aG(=*7kUj)Z{LGz`eG(i1V^>E$ljS^kN?xnut~YzIMv@pG;?p
zs!<~!KWnO?RAI8}PnPe6vKgH~T3Jbm@Dl{y!;EQH$?<N5s?|g33Xv@{6O$D4qYHkW
z=t`TQ?_W6@AarIk6#%Op73ZhvJ|8{;?^Vs-SxXlRSAh1ZY#vPQ4!W~;KVSZi*7L`a
z>)7mhR0nc2+V1m?!p{tiB2F$!*IrCY!%0H<-#yzQK?k{+k4RWzJan2BYNs1qHWR)^
z_)(B>i-E(VG$EvMjy<D$Z1r6~Wx>FjaeQcy(>-B9RUhJWEuea8jjn`H*+jjwWA?$<
z?gY*~2LQ`ZN^oj)RfxOjbg=st?NK0}_Yy<E#gXp@c+b0|JNfvN9r%M4K71Pnndaco
zlwiM%+C?IEC&Hc49KVuTY61^aGV;`QJk+RY+ELDE%+-9$8r&(|{n_3!QOAXZ0-4D$
zNDYQ#ZbZ2)M0TEE*X0(Lky*tU?C`o5{dyWj06YWeVIHDS^gILX(Y=UBH}d3kz&{~r
zvLW0cG_CZi4MCIKk~F7wCGzRsg@<gHF0&JK!GTZ!4`{$wr*qDc2i&MP$<A}WZ^A=^
z()x89Fz}GL-ISKNyHrkh9vsiYyEH3V_dFZSr?8AMfs-{5+ccU$Cj^55=!7WM0HFQs
z2_VnV!t)IsF&MIgc*koI%``j*dXZ@O-Xf>Gix%@vMmydO$zMQ2WWio~C7qr*jL3Pp
z67umwh<fvng-Pr)Xlut!)R4EG$ntDC4VdU7FSm7q=^nm84ui-q_DH#v|B6)BPBTR#
zpwbEbWnJ#iUKpOIwJL<KD>`Un3c%YtZ|ESZvmm!y1Ivmk5Duk4`b-J7r3*iU<RX?q
zXhk<N9r#m}%ST+DT%R;wp+Ogtpjk#4al8P3AW9E9>xc`T`Gtx76AIZvb1&ETJA&!d
zy#?^g->aV}Z-2dZRYIbshq&mv{?c0MXAsx5K`Z=Ln72Y9n42K%*F?`1F9k3_(ziha
zSHO=_C;+OZF}h2BntMJiPipRQ4@TDF0koVz48Fys+;Kvk>koK0Ipe6lbVP6w9jJZN
zxFXH}wl)pSecYHWD(}Aa*OaQ576ZOToChRgL#P@9n2sCh%lGF8;rN$&SZ=@gz*Q}R
z)wM0{0o9)!ZJ$SrT<h8>2L8`>om&#wHvN4uDOo9b&)Kpx$Uc8Yo4>15><rtLOKiG1
z+N@ZjzH<-yX=m9oLtE$P1T%NCr!5;6JJ0watR04;r44OF<j=7S@!L&+hvizp-!#Mk
z>sY<^q@8L7`8k5Ufy#S{f9tY$>ky}gCeAJP01OhM+VfIxG%ISm>|&+eBPPWLj7$cp
z?|?^$9k|c!xpbFMH}I$m_M`cW9S}<N<ld4%pofhSKX*EobjXn)B@O4^dLSJ#q#O}s
zs=`68H!Hb3VrD%dBuULDdcO1uZh^xA{gNQxu4}74L4)%<32r!nK&jaPVIriY6m*Fb
zK-kzgIOaSZK~uIMc4Sfw=)4W(q6o{~z6pyE{^6)xjH9}wN-(A}0~Zg_7*XNj*P&@#
z_&V#;ab-Nbh;<Am@ELX3|6iQ>FPb@zdGX$6T!P=-14?v36|sEb7NmVG!&n8%=0e|D
z^7{il-~;W;gx+E4TRqEy@!QM%cMZ9NE)w1k)!?BPVl=*v0pS@)GqR?+;P6K|wzopT
z@UqGG1y$FK<>u>4`CE50wh!N2_(|H`-+9M{+GXzi<4sllEcTGuM(@Pfw>yGxiO9Tt
zDZ-yUsCXvd%R}v8YYl1kIm?9Be28R&cIw|`0N8u03wS@&nhlRVU*+@&;m)v6yo!Hc
zP_X?ZA&YmD;5~Y<BTVn$VYkooJ(K=Xgy$%M6-1#pGMlE;!c5wbVGbQIlo)Hh)bLLA
zJN-bmHPa7|84P}^<4?isWV3^>9H_FJG3j$!o8iITJLv}}VoQR8He@2)y{aB&mYzLw
zB0zOHyz~p{m<It{D%1~`OUM3TXjfs;NKez!9oKZTmwmbBC$5u8d`GAf<3C?F3()sM
zV(yA=*sHhLz47l?w5EYo_#k~3ZXcf^J{Jy#L5Uyz%Kq$#NODP0h$<SNAd-*21T{m3
z2{?S#2y&G+&^Ic}tbo(#ucN}r`7SI%cLAO(n5gYzJCJ)N)GR=YOh+I|LZ1uJh-+%j
zH4L=qu+l3<(WGDCB3>WNE6H9n&#C!x2|pX}$jvzN+Vt)`P+;qN#||w-o0ZG4jI%sl
z45><9!?UckJO=%oh7NXw`TTIci@oX0(hv-R6D$8wx~Z>v&Ikl`Fv@UF*MqIXCI$pV
zAN`)+l+__~7(p^}CI9kNF^0FBNaaHz27ZBaDzjgG$)q4fo^0%33J%c($=_363{21u
ziH`g8-jeM%L253I!ID&?w;d;tPr_W9DkeUN%`1L@-N;3O2A?F_x+trQ0vF<zi+AA>
z=qz%X%Nm<qgHdT83qxBwu%l~3+y`izvBez*(hdbdK~`h>Wwx+B2M+IMv~lH*3pLdC
z<c>s3Vi5l=o8js#qItZSD@v4)fe4mo>~CJN5U2dDwG8`t{5Q4y%{K(K^#X~VkuBAm
zO_prx$QK_I`l3m}(imL1K~wV00uMhIZVt&s<@{BKg)ycaa04Yvxsh+7g#d#9yUnf^
zQ>aB<B!|xBGgw#$gDFaoLIg}!S_y_n*(CF8(MAEx+J<PdWN^&d-CIxiEFElR=32|~
zPkuqtOabpH!bAr%q|c>UbV3322c!O%N_Cp#^(m3fE*zB4SlZb0h2qqo;lxepV7hm)
zXd$B&bT;rxuInV{+p>OmhX8hTZP5qpMUMN?(^v$SFMXf6>JlYtjmEX#P}pL8)<^Lg
zPZzmtx+F4;R-8tX#<_uH#QzZB63p89=^PypCi@!~*GiQYmVGG}%1QxIf!)Ux>bq>I
zR@DOBDmxY0kNtaNxMs}xjKn%Yc^a<r4|9H4vzTyMfcwdt*~*7QH~^tv_W$ARoq}uY
zqHgWj*s*PM$F}Wc$F`lU?AW$#XUDc}+qRRF_nh-potyvWuTiz;-Kv^3XU)<2)B0!~
z&eo!*BFM622{^D-@_Vcu_yQQtg;m5lsY=`-d7K<Fpc)H?V1wugJP&B(GV%9gb{hS=
zMJGnEX<R>y6lufM!P(RO@#I_D;=o|jv8Xu_d1{_QMNQXa@bO6@+7jG#;I(3buK|U@
zY}-dY%Rn@CQ+kk;)wg!7@Z^m!U<3T>zLH<QLtB|ki9m&myf@4z|7_2R#E3a-_2Lq}
z{<&k%`W=-|>(?vo$j<y_%GTlyxn}MmBNjf5H#r*I+j_Ovjq>W{;#d|8F7<Lzgz>;N
z+(cWnYN?ofcv)migl8VfI%C-Fjt6U>-36Vm&#pT?t2!2odb;5%Xn9N-Kyp|wP2k`E
z*2){Ju?}ZeL5>YeK*mGv>Ax=3*O)w2#KV&*ay;Zx#22z3(?eEl&1F5TWgS@}Le|qg
z?Wm8eu3laMH7KiUHA^Rtx%<-eE~YzWCQf6^aoUzC<(&_<(`)fME?2#oENQl(_bV&6
zVh6o5siF-~_b$wWfLlNh6mp14G;@mA_1CH+-fM<gb7%oLy6VKNvTep3;Nc>TC$tCY
z`|`?T#()5u!i+sIKQ1XDBtgaNaU#Xh4r^`V2w~Irt*iqEbf}wAP4*U^y1|G<tRnx;
z2YG;*eBm9+&OCp=wN=~f*egq_0c9C8_CTj2!+yhGsar^Z1OUj`@E572KH85QuA<<%
zzkpyCg51zI6~wqZxKm_bQH%nx-g+rdkWf@Y5{%@sci`z<M%HI2Tcc%&c0~61(Kmuj
zRc@l7luX`j+N6wMy2etzcN2@O`k<1JnGeP{8-DwEah|x0E~_0+FO_VeHycX(wWH}=
z3U#kXh`|`Mngi$oJ)yyo5VFr9mFANNji#UL6;n5NZ)d@wjjdVh2G>{jhxRY6ZtN}h
zU7^JYvRR46(QZOvI?i*gCjZ<ewL7-Dwe-4tx5%ygxxq#_ye6Z@#YO|>@wC({)ZZK$
z$%h43#;Rrbc_+2<Hlr&TNOvD7F^QixTEOI<qL=2CtpOOYQS^kK&PEma{O%mP;GA6S
z&bB8T`GmwQQ<t|6<yfxJQ54wntl3n#kWm!4l}AG;v`QEh9Yb0w2Tbn9+@~8>;rIM4
zFV=ownPd9-9x;slF&>TTv?)C)-@;w&r%!9mIedR~eWcV<>-=D&8Q;qWC|Q3i7@)#H
zlx@3($^zDtJd1!vsC1pD<hs*`3d&?8^cG89xH#~<X^$&j_sZ5Ep_cjI>V&#OeUUzi
z-v;}&h$Nu?Fv=9GG%prOsSJ=FGD(^7>NG0-{Uji4nUXV$R5*W7%v&pC&bs4mDJjmA
zQNgQr<jt1TyQC0y1JEqX(JdAeu)E}e1TpSvzyWBu6N1z?&cj}j47~@Y^u5wPEgNXi
zFa6_z2(<SjhyHx^<PtAvdc?;c$M4y8OAJL03+xT5)LCn-*nh&px4;e}Tj@HX4C3=1
ziiZ~;p461(#{@h%Fr2wmsL(%E_)#{#m;**<_A_9jW4e84zFvUU>*PihuQ*OU2_tdh
zqX9USMgY`0;UdNJ-&;pXxu|~eAPHZ;m0wxoTPfXtj)^P+2)Ky|2_9jcbQVLBMyRS7
z2x@7A&tjgGZA~lEU0k1fz$}C@HLlgx{)oTmE2Ju9(?+Z3uqTZ{j6$bj8ja0+XIJUK
z#0&H~%S=s60*}4dVAt-`skN|%KdM#X8~}1Z^dwiwj8v?phL}VHOJpgO1Mu8!*qL_m
zuZ3$qQwXJhJM^BWZt2_e5~ipJYNG2|2IKWiZHcMiCIdTgh7^T?)}E|AIga2E3enq*
z{A9Xkj919xUhfj{G81-I0CW(lYhJ?Rz^d$3nqjV`5DT~8{JHT(k`f!DET&jGw*imW
zZ?AI<Uy-X7t&OHl)DK3(6oC|s_ge_ljnvy#XgEpD7MyaiCw!4XcF;NqfE}jfa+Kah
zqptk@Y{ZF0%I^YZe3XbT1?Ab1K&@oV7a#e{oW!$>ch@BbJq^5EXLoXWFNyaYUZ!C<
z9p=>Gl#nP!*yF9t2dUKb=P(gM0Dy2bjNRz1GhdMk%_<AQ8j*>aHGqWrMV0a&>`y_t
z&fC&?=&PIZQ&Yl5sik@!W?iOD0cTE=Zz~{U(};U(q0)(iEWjo~Cs7HR)SoioG(!I@
zDi|B&aFJn;tNctdO-&W*KJ}1EL?|Oq0RMizkC-VsH|It-@_jUKChr{}#I4D|{7cDH
zMOdZZX7R?~^D*xMLtQvRQpc840JoFZ0mZ*tR(4;vAi=1QfYzf&@#-?}+2bn^z#-Es
z-kMb7zCDI2$itD`tS}kk&K7T25l=beI=FlJWWIXR4@9`od)t*B1@FEn)Tk^y&?W-*
zXKyl|X~(5EGND+bL*o@NqUqPH0Li@lBZGFhTpo|V#y5=axbrTCk10i^gOa6`q`@JJ
za&30?88xk<Ht^Oj!juU{v(b#|rK5$<QeZoEe+LX%j_>=86L7TU@g0n8hgu#!rWGG{
z3?1#;NL_+}7b?&$tn`v11IHDBfVKEpAMb$#0YRu-N6hr@wr&F0Y|WwY*-S(1-u75M
zc8sZV70E5aiSq|i^7oT0<P>#bGQG-FtVqdr!;(@3;v)0`!@`229evwS*Tn=CrrBRg
zbn9AB_9=;knp#Gb7%)}hjVFn%{0@Hyq!skwTxii3z0`W$IS9{god{RJkb_o<Tfy}4
zR~?9Jw5S+wqv`=8s1h!|0XBR6T-8g@WTS_u{+#ojxDQ<Y=e^&tHC4L=k!YLaa#kzn
zo|D|?9)2Y~sKU}(0rJfq$KlR3UhD^Fo)VtlZ5>!!nPFY%d7wBTzO^s)Ms;}>V++WT
z23xr#nt>_){hoWi2%i}oYqs=kyV>;M?0B^xM4EP7OG^M|c*thy&hQW>GH6rL{HCEh
ziLowO4@KQ*|F-wFIb%Oq`X{tNN;?qa%&K%OL33x=N3IUGKLoJ*w2p{Lzy5nMe_+Qg
zGc<J>FjDf@8@h6sfih2Je8QVw!bw-Nr6P8i@a3Q&$u0e=W@fLWxldTy(hlxsTQ3V1
z6Jc+p1E4;^E-lK=3H06Qce9(j?qC(=u46nFd&6a55g3&G1bE=_)AJ?w&<M}HoZGhW
zpQ=CBth<k6!8BH#3iSUBPm{pqpfO?)2Mie)^IHq1R!^%mq8`I)hR(1_m@+~>RU5B$
zSDY4Ss<_u%GBh8Y%V^%OdKs_XYo5l3Qw+PD+W>P-boj@R=Qgc0?l)=J$e>eJ_9fvK
zoPyIsZ|8DZow~%zS$pWL<80SIDXn;$b-$*m8je8{EaEu$_y)#a0%G6ZAp}!!Pp!{C
zGlsrZE=#yyjeXbAfbV+JiPnAUSYw4@!r(dWOxGl8uAY*UluV2(7{X=4J`1%(lwG9U
z2LT+gm!g*)&ek)i<M;|bGO}D><vepul1XN_QGL$b_a%-SoJgCg(_|)@2E!Wp8iuqf
zsjinQD;+*P^=MwE&_`v6nQk=)k{)>n*)bii6rYSu^#L<NZl_&I9GzO7nut1Iz{ke;
zHdhcsu-wPid}y5IVa*(HSoNebs7Y9mMu3iMUm-VZWYz~C+Mw9bh_@ft;!JodORSD>
z0fZeJ_U^8;s!g<qsQ&xMEbG>=i;#+V#z_@UvNm4q8KIgW8dfb?;GH)A8k1;ptax1z
z=nW?FwVEokLp^^mGwdHXB(&FdNCIg#y8Bxze8ucuu;C9>v(!SE3`uSk<I#ax0{DUa
zy-^T%kPv7(AYa12Bd+jN8Y_ahI$kB2AXICMh*pLeaNMk1bKb$dI!ZH3NVc%-))y8$
zTux8zzOT98AJ=E0nNC%VQp0KS-p6Lo4&`&ThK1b?^Pn}OsA2~B#F-oh*+y2F)c3CS
zNfLi0kzH=_YX@u<OA^k+>eh-51B|T|B2I!Q;pUtzQ`tDl!s+O}+;!L731Qe(QIg66
zK|QnKa!ixmcJuC|e1({xx}^=aCqY)Pw_mAe(879vsM>+&Oa)6D&nMONG3vf)Qaj`b
zIwJZVUz1LzEMu?{{`3>Tq1YGe(b?MHIu?qisnch3lH5DN(nvS8{nDtP24vv6_|3W!
zgyrDo4lzgj0+kF+>3o2%z648UVf*{Wek(nG6BpkqUqFId+nF_$YnqwG?-raTSD9nu
zd5!JEZ7oJ3aY9C@hSJ16*?J@Dz2G-PT0JDunk79dX<IV=!gi3l)ct3@jzc?T0KJx>
z5w^Pz&kh2?I0|QPv_p-x4OrNVp`#Z(UzDor{G9Fmb}$g$6yKwe_Hy<M@LD+SfTs|a
zx#{Wo_-yq?4P1Z~dO=R=4%$kf6NDAs={)pf9wJyz0%{;^X?di}yf?TNR@%uU36&K7
zjXrj(?{PU;4_A`y7fT{I)6;?np6U^A+1ecdw9mwA0E@=8V%_`Y2goDWh{b50Cbf)G
zOq~b%p6eq@1S*@<Joz(>tl<Zq?rr1pW$m-Z_^J_oKjkIcbWKxQOU3iadz1c^sZC2n
zY{9hFequXCeGuJser}Yg)UX{ya*R=N(Btt;;&wj^O#l^HTDazj-)%87*+<TW(Y?4j
zwby+3)tm)N3f%sV2Kd+Pz7>&@&d|&*Ja<*szd2d^_}Z{i(i$ybK02%6pP$QRr!toa
zE_z<rvR=B>9yT!&tsIWg*Y&#Q%znoWqcIMYxklpXvq-Y+({=e=EdI8f)85%TIVK0L
zatQaPeH|wQ>1OtNYr3jiuGwq6t|Q>5>j81y9(?~b*QEU}5AZ&pi1)IB3~4E9-0zip
z*c4^az)hi^aU1^N#h9<%k#(0bQ(+jp)!9&OwzXOBI1w_g7UFB#XRFYySSfwjWAav<
zVua5k>W!8!mji<xY&r0fO7x1l<G(XNM)%1p&X0p`q}^ic@3Q?qNQj^Q^kLzqo}X%_
zUE#<M{`E&j8K9*t4Mk5-%6GsjjOEpC@Qm5hO1M<yS+}_@S;+(@fo{tu9Te8^IM`^G
zH!dp-+GCivWG4jU!W;ixmdce-$14Hh=6oyhb@uOeV<p^Ty4Y!A_3xz0#PXondy?rH
zP;pNle@D009qwEV%Ip{Qo9E%3az;Gd9pyFjt5va*Ho%K4TG|x&HNlXYBqQBJIENo0
z=MDmSI}jax@cXQnNZI*O%^g_+b8B;be2ax8yddm^xUEb!j3|QGAkHiGEQwmyVCs2q
z{b9&=vGkcwc#+LQl9tFzdC;rhJ)82VfdjNwzT9X?>6YPc@9VyxJ&&n88GF(+PtE0(
z7L~*}8qoc2^{1-WH;!ieH=9pm4%|~lAA5A%3}Cp+Ud+23ao{!j5b@PRhX!@r#th2R
zeLI;2HH4C!>}Oy+!>2+_3@=eRJ;ooo0xpq4g#VxcCCRqyp?0&{C7V_5>(P3#2Iari
z)=8sne7Mbem7YT!`@5E`Hvz6ba@}taF)?gafQ;|oTl3hZg&9^rjC-y|(ho$VOIQ5H
z>?MEh!(AeqK1FDM^n6)*A2MuMvujFF@Cyyy%y0IjP;}+TB_88gB7sjGXN;xhxY;P)
zKpe!Xv_OE^tXuSWi-;Bx%gBrvJ)nejY(JH;;Y_qGrL`>mz~<UnH$#`bGi_nI?Pgm5
zfVp)5dBeU|_f=&T{g&K6xTC6foLPxdUSPN&b=OMKa}8cm>{n}0uxQNh#aqA^r?St{
zbG=zKTS|;xfK0OEK^YG1sesj;o#5wTCZA86)MYBc5y`qCv%Izh7NCjqwRknNvUl-#
zJSEr6oTVm>x(_h$B+t#PIUZF{=qeooOb|09{xv;8XDZ`mamhakOecSLHrK1dsc?aL
zRPs#GskzIgDBYXHsadj1=4CT|vO`>(@EpTSVCiEok}GaH*-w>gRL{qG8?G(nz|H?8
z%>PVg7{l>y`<!dX&IjHGHIG5-s>OhNBO+tRas6=hYIH0C8L_Qtor**1Uc{RO*ypBW
zWH*!=8nf70@wh`G5TkcGJfGpx&gkT|!mz2^5BR*AB21zP_(xL3+@M`nSq-u1%3U2K
zVZp`WDPGYDXRNAJ27H%wXl=BGJ>J26r_r_bY@#d$oH++h(@Hspn_!Ws?rI;el6=A6
z<3`@c?enyH+uKFHq%21$HYk<>n)k{DBZdL%XXeFB5Sdw=P2X3&tmJN)QP8q=k}|2$
z$1W<kD@^=0)#%jPiG1370#j3KV-fk5LarMDEjMXS{f>QnEJsR2zPV%>)-91(SZ@Ju
zsZQTV%tF@58D->n+p>|jGDaOYfF7xnCo-nf(S)Z7a+Js0QSvr}`&DFsoYoR+nkaE_
zmnBZFLgi1NMDANEGe<Y``Z8PD9c`B)tqz9sl$~<P#4mZu=7IYMm;d&|C1^#P{5zG!
zkyM#fQw}y4YbaL>6_31XJQIMK)T5LyS9-rqOTs2PV9pf%{S<vD7c)R8MzEO%rElea
zhNF{z0=HrV7W3i{ohe}4SzGbd9mWs~<;+J!nGWA{YG3+U*N!Ma2=(59P18m_eH=C=
zS*SwEc33qS3DQaa#f8ERn7-F&SdO6{&gS!~2|pUho6v4h<DAZ*U6;mXAKN)Lm+T=8
zTK;=CeYjcJ5Pc^@`-HLoQQv%j092);LwdvfF}OLPB1_Sg!4aS}Xo^(v4?Xw!9&(yI
z1l`_rB|DYFCi&X7eFAg}Nbw0E8e0EODU&!GPG&dw_5S2IbMC2q-$W^@0kUyvVuA6t
zsmrQFSRm;13M$5Wz+z6MNWI4DVWai^RL;`T(bY!u+53nw*NXRa#ixf5lH=tTz<nJt
zDh0aM)E^P*-2{l*Rt{hKyMg}u`=q<5!F?}VOGm*jn4ke6Hty=P;_>kO*rxBHr$e|*
z<3i;aN5W)^d!hsK<4dF9`qqwq9w^~x1mUwtrZ~OD-my(&Llk3l4ekK}@)p5PRc72E
zT5kom(X&WTE~eJ5kimJNTU$+)=HI#l{A%cNX8H|S;Q_!_u5Wk&G^rEb#<f$dNL?as
ze}(w8U9|y%C-Z63FXUE9ZNI3N^e=`;(Gr`u#HD-^I6TWb0UdIeRB*~uxU`*Wv$g0v
zsv_H^8gom~!$pWpnQg^ehk5qR+ATa^?krW~7?q{NI>U-@{biYptLNJY2S@vKOyuZk
zgTbfQ5C^~o@HUj1RfR0Jb*ji3*|9ty5cdomI%0hQf4nF>!Efl;qyD^>&~yn|g<f7n
z{-HD^YnrFcfYz2PQtRaUw#bSvYci*5efY14hD4Zv=5U_fdpF!-MgKHNF&XFPLKV9R
zf|N(xtV9`dnrT=$ZGY69uTdq#_hHOIo}JXFCk>!Hp=oVMbVQphas;aAzDfQ*d|Q2-
zh226(#i)xZpXt8G1BT?jC-YgcAI1_zF%?w$r^#!;iw`a68<C#RJ-uhFhisyv)39|R
zB%syXWolFR{b|1B2SJR&u()dUA(`?gxf4sktNi@T{`>_{_|iu-yRxA1-XtpA?qEe3
zpr?ZBrKeiFL636TbE-)g(;VT>KI_%Bzadu;`3w_zdnxp|dm$V&`}^j@tpy6Gdb~w_
z2uiGNaS(&sDJMiwJ$(x=fWAWz#)$Md>OI~bOltvw^PJ(Ott6}06|8Henyvk}J@Sr`
zRi;7nI02?gAU2r1(w&ad$oB8|=NRZs0QJ!S@FWFoi2r!f*nQkxYnWk7_8p1nyyL^N
z%IxrKU7f4H_x}K;RKS^Zu<rT2?Dfmbdh=ckfj!!JAnsxG^GWA$3lVwje%SxRl%y!Z
z8H;+fUw+1L84~ZRu@wG2>yF9#6&v$YXjGpG3jrd4w!iOk*2WoX6hCKu$0n^+8>Xi4
zQweRLK+iX&xSR0M9e$va?tf58{&y05JPI?@P~x(WSCb=^1PJ&%6DL0fIOTP>Wl!|?
zwJGEx<&t(|Pqx&zx{_-KCR#?kgACHH#ZN1MKBVrttW3-B`;7k=*@%*?jy+c~JEjgI
zVB`?ZAbBt)-2k$-q1oXi#6&=6%E01?iMFjtw#TdgZ1fQR)zQPr$>T9>pZB9c9HP+e
zF=o8?aq)+QGYZn+E_QCV&w-vcckM(&Dv}*q8>QdNT)Ta?`*uA<@Ebi@+E%4i!F;aV
zLhnLP@W?-eO{{PwT{nhDjjdC9H3v-rJTPImkz4%UdyHsRDf}yjz3uLJx7p6D%hiXB
z3}-PcsneH_?CNUP`KcCHpD!!*x@x%uPJbrAJ;grpU)PmEF%;_%dwIyT`hg5<lM>?)
zr#eU`eD+0-i?NaIr!emOW2I#IfRgJl?(X@5qtKuBqM^_$p7E3G9H_MgQ0UYG4yYt2
z_G!E$YA;6VR4<bV$&U<C!vwwr;!y^@Gi6I{yVY5#iwvvg|6US_1)JK7fOr4rCu!Zd
zw?j66mLw{q{pWyp%$Ns_s_TX(dwrpA`Wql=gW%p;DE`BRR$x#eS1E={8ZGX&ULgNj
z#gzXfDUtDxTO;;f6=@Qr0HXLS(>Vmch916rheY2aCum(ATL5H7cwQV{2Z-0!{WG-x
zk*UswA2J2s`u~qivCRCxWU5NX?SIJBDgXx{j-v{@C`X5Mj*R3S7LykaTmx&G$vU{V
zxhq~kF{mQO_B-oc%gna7_5188^+S&G-GWnUmho~|VOOI0EtUj8Ql7MImY%GgUE|du
z>mikRvky=8R#f<bjGM?7)d>+GpGx2eW3W1JD8o0}$*>FB3(H8STKO!2ptlpMD6j}{
zn?n^93A-K`g{8~HPT1VEJwVo_Uc{D^ba!Mg7C0bYiM!RmoN*+mBma)NvCy%{TMH<+
zS-}B8Kt?w!O|}7sFlqixv;)MY%a=?htppcDl$R8Oo>fn&sDpN6*pDHMhw>hVJZ*rD
z2+11y!=O002w|M&N6pP}MFptlrVaoiX$EeKcZy?1s9$)a5H3-Us*KRC{$zqkl48T4
zT)|7vi#;w=pX1ZKt+<Y%V*nfYZ~Ye+2JR|@n)Aj90btu_+QdY{CWstc4=0pU-Dlg8
z#mCr(FBcD%4o(z!QXzF<_S|a-i*B8h$2o3zkE9M1W@c8|G_Cy2wCB4KvIc;cb$u{_
zqCP3nTyVy@lDkY@jbBSSf|9^q>m2+6A{=!KSS6hoHiB)+8g_FEYB5VAbw{Mr!jcBM
z<gA7W%a2$|cm7cID6~nSp5DWGzP^Z<N$sKjwKA95LJcN7^Na#UtLFSgLe(*Q-iqs@
z|M%msAb1K8z#+By$R;PhHWI)M!f6y_Gd?VUgk@62!okzLaR3F^J6PU;!Br2h>v#m;
z8_|i2-E}%%JGx%*K<gV$JWeF4<%w=GCNVU}Lg`p%cZ$U!>nb%pRDO!a(+S>^ab{&$
zgr?#no0%!0>H}{LoGR{mO)WBdQ6)ew<INgGDd|KS*01~gBP8HzAOYVECbE!}_{N8*
zm<E4CNMEkB@PFn$jqhEuzp*>~8@vkziyQ?82bb~yo2iuexo>>QUgA)hbhr+(nA9Lm
zbg;@u`!CSGo%aYrh}Y*tLH$8c!X2`VsQn~bxH|=j6+RYl$UsuaI?gZFJ`kHSYQ!GZ
zP1{PI(472{IhHszd;pM*phV>zP`J*8wg-@0f;u(g&RavXy9RS48^+%ODdbY(SPss=
z!fHEbCXPZM+2zTU{g7A{TKXY~4l0cTPNuUejfF`Yi>!Yu1trPN60Kp$TQUazjN9!5
z&0sl?oP>({LXO)ZGZN?d`yaC7Q%G$(u8wL3NWqKQL><PV(gM1}hK*aoL4-4Z9#!Q(
zhyNZ`G%kBvkY^_t&OL{leK0oA|3JoU)E~&`knjT;sqpS;DG`$o3i)=tTd#v062*vn
z^?Op}#Q)=rbwA%=C#T+!rIb_)_$h3BpXKiJW9jafhJpn6RXdA7!c-=?t_M23TN_(_
z0HD(J0%5wM1&vHcj(Oh{^Ww)u2~#g|7U)B90?s=V<SXnsH8e(0N_gU3SX>(vYXFsV
z?n`RarNq#h#REqdf#e|5;o<ug(b*Gv6wPH;(mZjYQIdmNp5+=fYb}H&S(2mjeig*4
z4!_yf5Y?V0DnI<-`!qsif(SR+=EXR{4Y;?J(c|kvy)RP!65oa#*#FS)q;msD<Wm+l
z?D6n;Rbir|I9cUbs8u$rIX=O#7i_S;ztXK+n8ky={pmuZXG}v6+us*P?ZgXQ*k*@n
zJCHEbzTRg2DzW6>9HqF&xKD1Z{0~J=PB^PoIyXf<Ktgxk1~Jz!Vhk1!_$tu>^2@l$
zI##kpZ}TsEDu&2@VhO`+f0<ThjpW=2<0{fRy1c*MGhM6Mmy2!j>92<KHVjNF6THND
zk;#w2l07G$O@&Z>r>rpP=asl^pT+A|ZB@D*BMN3q6X`IDfA&pLB``mE`R|897=2{N
zE6^@lcA`sx$Vlp|J!^UtzQY0ol9LAl-F}Kq@FCpW$G@VjI`qloAgq|iHZ4Py=;c{j
zI)CThPc@^EW!}`cR40rLb1_4q&ea0iyEsYX%$2n9CRx!Jz;y`v<64`kR*l!2Q0!d|
zxAmLXX&13?uC5se(tbil5u7$On$`U=-*X9}1XC%gxuI~3x|KKaKP{U9TV*M|dDli%
zfi%mdKdqisd<PwzuI)CyZh|4gr(fJK1&MZ?*rGeR^vwz(xa679t<69m7uIi{X7w!9
zCK)nkj$d4z0hAz>ORaR=#XYMcF3i7WGBj$?U-3*pOoTLKOv$cS1XJ+b({w7aCoe?{
zW+_!FZa$9Q1FSVgjpjK3Fc|ijwSn!^hy!8Rq5JeqmJ{9!C-C9%?|}~)W^ZoE?mTy(
z$X7nnS4@+j823S})2?oqsat~L@IWq{L0I!#kZM3HdjS@BiO@3Vv@B@zTn71nE$|}I
zS<vX(;t~rfh-CKlvCaO3NOjMhJ@gu6r@qsxj16kx#C9>@Lz$riI-NRnu7X89_;VI!
zQFa8c+sQV1iQK&V<n95JDhbhHEG2frq(^7AJgDSYVUwdzHW+)CzC)mS<a)fUcYVL3
zeVK_faUrktd^~Vl72NI--4KhJTBU8-&2SShW+B4`7vf+1VRL^VXzjF0#CcU1>8QOb
zg=q2bw}MXW9a<y+tzrg_{bV2bsn@D92N1WHj`jVbY+c7p++^ixT(Q>EEVRQz(qLQl
zo+nmn(JI+`D7fi<?l_EFgC*7d6bypC1}Ab@UfdnuKaJAI42LYm0`Wh}v|a*_Oh3*3
zG!YGJ|GU|r{-xG#d*|$oydMUR;eBryk-f0iRM}Ru3IL!ko>~|lV=83J0?JJ*ymS)Y
zvDcY`Kg^s*(2+$s{%2=-3%Ol!2G(4i^HH4aYb|l&A8;M0Q9e{+6m(Frt5E+O$bM9f
z2Z=z9EOGG#mnl)uEe?qJ;$!O-`S0TM<qM9rTs15#52v1xr+z+4n20HABwT_6!8Z<^
z^{I>pPz3OT{g&+~MLGI-iKNfpz+uVYF=sgs;}lx;LkAcJi%oc`$wquiW+^yMODw@3
z*NKWlz~;Q(^Snn`J38DNuPdKlJS@~F%{JT$O+rq-q!C~x%y*kE?19!+&0BWrRI!LV
zzM->t)B4gT^TQatmYxw*sgKtW%cukvdaGCi@ORb-Z1xM7vn*=}T!g<*;1`~lZk*3q
zV$X*<LfkJwHYcyOWy1YyX`%EP5u5tXgym-cWs=E=+WNBc5q+-Hpo8+IJC}#`EEgAd
zIS9%W=+XU#@q5@g(T@4r33fI~BE!Xwz7>+ToixddFJHGCYZWDLP^@gW{A)$^p(pqc
zK;pjJsq5^*SP^h7QP+9OGG%!-L)I=m)BEf2|C+uBC&pceWHmG>o<h+=$_9D6p`?ez
zdfHk1+Bu$?6x<C!Iw!57jRmakdX*cDerj7cN=6>P-h1JYULCzyNH6gll-c{2j;A~6
zn>G6RJrAU>E!{nQTn4k986Z2S05_gD{ASyaB>2SlS+`2Pon>Fr_%re&BQvZfmWr+e
zoNv|A;70E4-4f|PqFoB{7w7xd{&L%*(oHejsIZ~^d%7A^^M9g$Z}<Am?bG`d127bM
z!xPB2tA%@4XH<Oov!A4uKWpdM^?5O3Ec7!-)t$-2yNMqkcc^IEt5*V)I!Z)Zmov4)
zZc{ZLg3`l}HYGGJ&Y$GFL~lR5<;kpB&o{#*-v5pfh($q5kUbnnVmu5$(s>u<ri(sj
z(P8F!OKZ}Shs0LuwlTnnmFZePqpXVKCb@0;WAjJuNH|-Kb+trW3-e;JQgv#SrQPFu
zyCBoCHyN%qoMw6t)-M3WH|ZuFVKS3BK|)?+;b(ScK6-}(JJ6uG&|IL@q<Xpdbk9ux
z<U$u#ZLs1gBy+^5V@)2<ZOq3JH%({MTu>I8ldBHZ1lH$p8A+Wy$MUr8@?*zvYr;Zm
z9yEI$!Xbo)jq;0r_Em&zwCo<5_lYm>gUuZ93w4nktiN|V7W4ojy`NstGaa|H?R<t>
zeXZ=eWbH20qdc!!RBRH^VXtsn30m1BkHZZ4A`*z+3%N{btL}2fBbcUQu7Zb3wJcdP
zg0v<N8w%(Ml6~W9Iyk~I<G0>wXURriiMn6Hf|vg_xJM$H#{Jj){eKRM)p@|IX+;Zl
zVq8I|+3PI>$e<q|HyuN0*P1}u)XP~#Wzdr>lbaMLwQ?;zz0tMM;wv5=oHcndfYz>|
z9vSDaWvCrdhDjfCpf105)QmbuPgKtv>-%|q!7*Rv8D7lY3{%&?3uEthH0sa{H@*qD
z>Y=@U2Vk;zjjh|M*tE(~^`JuIgn6RpWiN({qVp31wq2X#t2`UwGweu&Sz16~lTl$V
zg;X_0+<$DDZNG%$14=G==LebUZylq%JY={3n6Zp*-p&6(#{J@^wZ*8^n}*O)l4mH)
z6{@#(iox3G*Q!zP*Ehs|o3Hjz7q!R^$U9T}b+x&+(Dc8+z(ca6!M`a7<0b37LOCb^
z;<m*kcHi~OZV$OS1v(|V)tizW;#m#z+7PWSZ<2Xy7O6ViNUkMX7C-tu#V2WmzA@Xz
zs7(e97esD$?gQ}D6~yU4V)b6*d_#`(h&(a*!fayq7kKz_YNcUi>!511QaTjk9|<Wj
zv|g9-*ZXpb3-DhsOcKF}gf&?KsLCtAeWq0otQ2-HS(NQT9?Mum*_uKosBGD<1t`lp
zK^yGpU*pYp+b2+!!nz^;wPG3(@K$DTkah-ssbVrE179yr*14yD-Q}RO>9`Cv5L{yr
zDC%P&m#<7aNlw|pS>-&>(iM#IP>O%dddw3~0n6o_Eg3_0-eyhOl!5adOp;B2@Uu83
zW2IIeHA36(g6W3pkAy9jJ>!T`5v-Bo%9RTw+I86vL-k{&W@AI9YcA#hEAEMS1R^T7
z2EPMqJzfEflJ?(3ij^wqgYc)5&Li@1rHJ`^6A5TCN*w;8iyrL`;<^(sQwn#m4M(BY
z%dwSee@{{=mU$iTZ*Gkc!Z|zuPTq^ejGTnx*{WVW)$wK}sQ9fs=)DZsIE+hfE$@*6
zJ6V{WRMCK#AU^>~JO#q$-AMCb;(*&Q1;J7gA(J<IEghFOYcNeN8T_!<q;g^_O@}}G
zrAP?UXQhEgwDeZWVr1ejGF%ZIKeeLh^Sd0%xr>GTVwtFrA2pX^nn4fXkHj)~oHz+4
z^A?%X^6b^rFJ36+q8MqP3JIObW9Fhhd}OgWCKx5<CU@@KYy-JjHJDRW*kHI1I4xf<
z^H*@@mH`t#AC^xZ)d!wj8SgX!G5lY(BLcpAJ~7;hsm>#nZ}bv2CrA)@@geoatv|W{
zHHzbge0h~9QD;70v1Z)>!(RDG)+zzux?vDQYK*B6f*5Uo(H#PFPB+2G-oMDNWp)5U
zQRXIr^I8UJ&0zj<Q#O)~!VwZL>$wh;$k-LcEs)3@)d)Em1aVbP!HC(BT2r6-MT0Q2
z!BhW3?Vnj&c=FLW?c8cE*{>UkojwXSW(IwtI!+?&p$RrrPW$TstX~tf4)hZr{ogL_
z4*49?S;>tjCy^)n1|g1QV_VFT_A#>iuGt#_#z@sRX@bKjqJ_IF^}Q1TRz8;1Rt*NW
zBZjQdsW@}CTm1j*6xs;!xAbYlK<(Mp4LQ4T9vxeCb9&v16sXjNy)+^1j$d3EM0dHP
z?xr>g9Cm>2uJjNA%a^A8I09`#avN~0TR|W)fOW~gY_eAghIyXY>gxa8jWfH1{*kEk
zh|=8npxNig{z;%~hljKyzw}Lh|7WZ8it}wKFxH4~OK;$Lj%U$m#~(Ky@_w{GeE$9(
zGdcWJ@VtxQ2U*o!MM534M8G<yoXb=gh|ge{DZJ;$3aIg6e*17h>$ihII@!wiTNeaA
zej>^FBND5Ulps_^8L#;6`Zww#>}RFTjP+?P!XU6Z%B85{gNp#@An;w$wuy0?qf9?{
z&(7YC&L;JT4*uE)mLN<B`WGZL_+WF&A-vXNh|TJ@@OcT14|d6~nRgJd{>hu>MOF$_
ztzKBilcegSPI+b#=L)YK8;p>Vrj>W5Zlx9DD`cHXV)Mki8KbR{q&c0NYC%K_>ES|T
z%EoAItYyCv4hIDs1waq>PI$1?{Ylg2=d!s;dlDidtgUhobKd+!GD3?Z?h}^Ee%5gw
zdL}0UH`3S}lxVWn(unUM#NMShrl~1Qrh_-p#$r@fpC^4|j(@c6vJ!df3rbp&Iy5|&
z$W7AHHz<ts>{t|j8c*3<uGwQ&i#9H_hL{Fcwq;jn`pf{-;!#s99*vI8hx`I>5Blku
z<5ZL70teSkj5X~hs?8!+@SR75b5`|oN}jLg`%v|H+RZwWnOjl&eST3n?DuZ~j>lHz
zEUHDzVxX=sP%diE-MPD#%I4?Lao%iDe{LiNjaejV-bf<Z@o^#H*!?tYt!i+27HW*u
zqFAawjBh=)kp)l2TCp+eeDI0I{O-t(8MmB%pUXALe2JMv(|iOh;xOy_<vLqy7g34;
zAvi8TuKIeU<h;H8JcB0p`LjlLj(jo$px10?r~jhrujDSg@z!<Ib0w{}yAdD@{(~_|
zJ@TIsKM_B@j|f_Z4w4wE(kjC{=rTxt^-Oq)hrw!gY0_k?r1NRvlFW8uj(3qck`7zb
zo~XmMr@O3U2`?MEQ@OPd<jXH4{$?&rCVK{y4e~YT{Kqi2CiH+CxB^dSD7IY^p5D`0
zhoN+sSz{Ow+IuN8b3_92UV%u3p2bNsM@#`cfe;<nfnr4}mAbm&RYa)5EPnakKLFBm
zw{++761yGGS(%t=-!^k+JmQ}kX4f6ARy8OBC~r!U?_etlF#Xy^bYLw8GYa!O1R!m+
zn1Rg^<!8;4w}i1EUmgjHa_;|qiYq(1^Wrx0d&0uHN3y6H@P0_}C9vtAS;Cv}vu?Sz
zb{FjYP<R8eHfI`;Y=FpTZry_b`(|Q}iWUpe1i>(6%_Wwl_jv9{YvBy<MYFg<xvq#r
zdI_@jL|Ftu-Hyp!;X&qX{KFvwkc%-5>cRx`GdmKNpW1x03-*?6B1B=^kC`7m=-4$n
zE*z1E$nw6t-yt$c_9jgH#0pb7@I!SFXj5v;?<k-$0X8F*D_WJmvuD+_AxFr=%?^Gv
zX09kK`8QUibxs*c#l&l}j3CCuOO~2lzv(jE*K$fmR@PX9PdFR6a5X{zNaJTNbXe~o
zVQ+puxNF+9-SX+`w1|DzRwDVX_UW*VU3wfQZ!Pmlr>RA}2qWoo^_AaBt3*zmql|D9
zMDH&2*JMb39nQ5|%0igsD91-*Grl!?Qe}46_@P$lA}6Pp=$p;iK!FFFns>!~iD<S~
zZMSqMVeDbHq83)-s*PiSAXDYSxyQ))Fw1gcb1qK|=>hs$Wg*9J(^g%3u%Lm5c->iY
zP!gxQj-)<42NIsB04I@N(H2MjS`3b^chJACw0{;S<}nuQtEw7qf|nt=xHGa0HjPuz
zgwL#vZGuI`Wi0NEVJL^)nHT8mAQRo|Cr)f;ZNmVN#0Un)<bv}6a6eet%Y!WZ$>W3K
zCV#`VjaHd56<$dxNE^%XV&`dF{!%;NomLsHX%jz-dFheSGMNDjnfvGScCS(_Tb<^N
zl;`Hl@1GgVxO;I;kjS1Hd*-$ys`_aDxT&H&s!Xi1_?kRhz>A%u+2wiUQGXKl%GN+G
zY(G@!90c~u)@(BY<+~P^^+Vv<<wPEF9fq-P-H5UU6AZ&o9OcPcQCyJ4n3<fs!qOW4
ze`FTT&BgRojp1on6~u}(#XR;X8M)wBfmZeTSM`})5j=oA^!Yt{vK&I-{NQSH;Phu_
zQC~hk<lkLHqVRQ;Tr!@H%bVMWS_)=6tPeM*@9T@PrvFI;GSm9%Oe3;cVTOnI9)sW%
zn5L?!l|Rth9Q+v&85>xqRThe-VDU8pSzu4E&H=S~n-*jbYtYRZsL1P5EuuuX^cOPh
zpIdC*1072@H|P0s$6mw&ee6S$ztb~w0k_~6%Nsg;BTvuNwqjvJqg$nEs@iW9wKoGQ
z0;FECE;xSx`6gY@Qf`~3-xP-hp06q#Ct#_+)dD%*NrtG2$;JhROq9atpo(#6m(cne
z54Kf=zbrf59jSGHf>$2BbSN<_>;7q@d6NOffEw08DwHO+@^E|WZy4a|j;e55W5S4|
zcvY;$$X~Qso-7sE@I=#ja?Gzh*O{fp%v?R6rj!^0x|}(dUsLp^nlYR*VR>%ZfA>|~
z3zw3t{=;QAwTPcQWb*0|DOBk2%4eFcnhd1>ZSmZ&QR0#?w*7+9yUk9(U0*~UBNNC8
z?CCeB8$gz^89cN(rgW@reT7!^tV+z`&A<C%wusMDjCfx$l_Q6!`7Gh8H-VZ}1X71o
zSJe6Aglpdd_91F2SI!cwl?@S9BFxL>pq;h#pLnmicVnJn1z1rwZ^XOmGSa-njU(iF
zb^}DR{EJ87lj+utBEGZN0&={_Kwrb+5oAzYLMUsy2XF(+DkEJ`V9SdfcO~p&TpFS+
zPQF6x!ohi<N;g0VY(qDxrAf`HLq(ix4Fr<`8hi27mV`uP<yX*?^rw+FAZaHu#Z=5t
zvT0!=h7~ifd+BhCUp=nI*%!?>yZ(u0_L9kYymbU=I-G8S7TFt!D|)0!p|YqrtllW!
z+)dUqSliC~X>cpwtLOBMWgzeDDs<IoJgV^&%l&Z%YxPsE3F*0lnVY7TLDx9Sf7%QH
z{pkE<f7T}J)T!GsjDtWnf5BJ#W_s}^H5F%ba`M`$xJyBw{iGhYy^|ndzoM@52q&a1
ztBjwol4yTqu_oG>#D?6Pgmp!xQYBnt!)hZh{Or@#@W1yo_(@}$(4{eV&Bx|}>h*l7
ze90gO3SkmI1PgfR!O0=YDH!w>{`9Q<0cZUsW-p54CiqJhFp0K|u&o`)JCF)781w9e
zt1OckQ-!N&^Gk``1aa4Uq;9D2Z#hS}@m*4*vXvrpkt}m7<$l6ZaiH=vV@SQJBwS4_
z#FynE?}5~f=>Te_t#nEc1vn}oeQW!%hYq-HSNFN|U3Q9$pV8`YjU2s5IO31V6QWlw
z?Mo2v;y+v#`z!g@%svLLhQ|$Va(Rl`-4S0GB`<*jW$n>psQaMR2Eq`+kQQ8!OyHj1
zL#|u<=+R0ks<a-gbGkpD14EG2G3}cE=sc56EWN9&I6ZyzzuSu3<6YYT6UHFR0$iCT
z5HY9lm}2-<i>O|F`{*ib+rt3PizRF0O!2W_p=lgRcvw3>2@uCnd;DO`z|4R9!Djka
zEWjp$PI28Ec*NNfWxL`DR6oVVq3)Hyw|@m!x5F*5fs4Z*cq;ID-3s_jphm!&3pl1C
z{R5Sdt%lP#dgaj;JRxrY#Hqm-$b<G?nzr9tc{wGh*GV)N7Myom5fG#oZhkEHyR@$o
z(z%|0)=ah+Z}UAwf4KU5)3mmdMVVGS-NgOb*>2?yAkf*MQ&C|0aGiY*g0?x<7P1qS
z@Hv|xFN~b4D4$OZ_4iYaWmEE%#X1L<&Xv$wn5|F8-?hikm^K&#$nHgv;IYYw9NXZp
z^jcPxJ=%i|t!8ouFLp3lK_zR@CdZ6bVI{U>4Cd`QH{Rv<{n@q{5%jPeuyJ~foPg!$
zrzn?u$=NMq!#fIhV;u__aQ?pnrmqq~`ie!qL-^8cq@I(18(1(vF`urcFATX#dvlqK
z`Zx6++7FBXI1RY#hny)KG?kSzHsd115t1SYGb<CtBHWvv88tMdLUo=(LbP#(0GB6)
zjC6oe$v5oZ8&Yt}p-5yN0E6&D+FT4BDjqUd-`#U8fl_FGJk0-P(CEf;Pr6RC$858&
zqDYb|qn*)eIJjn=rSAiWD=3A>mkF){<OQ7A`nin)WIb#Wt)hy5oCrlj(J?rabrro-
zv(z|QW#243eX>N<aQrzK(h64UsBz7J_&Uo9dt?;61Xz1qMnJn+YT`edSD~(+hR<x1
z2wE`LmtYQzpw(D3KVGsHC%*X_obL|nq?j03oI0iWOc%0DNaO(T(9AA8^HYATw{v3?
zIwo*H#{pf|XQ<aO;&CA(01e@ET9WL2xqi$NQh$$e2iQ2EEeXDsOpO_alymot6q`v(
z3m<8lmV1kgTyD&pl$47FIt{6dgK_A|-1@k6$lm-Lb`&fMK<Yaddi_8({l4)vo(LyC
zR+4rn6Cp{B>xed8EWmiJJkXF&>I0q-&FTtBG8^{no2rT3`Hh0sVk-zyRZc^m<iS+P
zhwr+=(tz_Cqa|`zLD;F{{F1IM4~&MRs|{!>Q+@)jp;x72E_~`aK^9%w-9jLH7z{p=
z`NN@t>(Aa#AQW7VAaNdK0K8d7;O|LpmHV_)p2uCW`&hkYB*^C_e%Wquz9p4_wsQ!O
z`kSLegmdlxq(<K4yJN&o^6%H}Eykl2J*Y>4uI0SPQD;JzBHjf@bbMBB_<^jh%#F3<
zbd}qIFxWdJt-pTX5p|C}?)#m|uCp*S9VVv~I^Ayaf?N45!mo&Z@rI*)R=GfK8ZGa!
z<4W>S&)wvoQ3~dtMyPqG*=xc?>Xfj6hRol{I%W@Y0$76$SMK=}NEAA|if<lWI`12h
zlUxLH(3uJYL?Aq$v@U~aMsSOYGW3+H-WeIug>d@{6jR}-YeX<%6;J953AdFL_@5Z)
z##u16j<S%7n&u0;FIA)EZD2lgvY@6?i_if#g`g@DTgIamPo!>LE$ofIcS$V)aFXOQ
zw{=yPORH(bF&A}`o4*PnL56Ea<Fd<Vot|qAN!LM-R&km4M_bc;=wrkb>_Pia?`NEE
zR#h$oOU4UICA3fk2$6&O@nIl;!BC?7VpwlNsN1o`Atsp0R-yS8L0f0+befp4KS4Yu
z1)X&Ia5fp`oXN~|Kpov4cqQ5bC{3_%&K}9IM>k!W==-YVqw28&$KCprHt!yaU6xz?
zWj2CsgDes$w}7kHA8Qpr^1Fcw^<z(1rpw@?N<L3OCAz?;K$uK9==nR$#M)MU*waw_
z!FP*get;(+V+|X#dqP(p_OF{KLW1X%rrBC18PvBI<7$HVWORqRj}>khVBs828dP*U
zIPka$iTxJ!8NWD(2`Zx|y^pllrO5M;s9z^yg{w+kr6q$*7yWzcFgN1<MAw<j1Qj?z
zUMHXS^nwoe_YKdl-~Do6mkRF%Pum?&!-<fA-=+caL0XNfnKD2<xrHm1HgZF)4`=}}
zvrHMFhAW^?%~4YTgQgN^fckFKvdU(?$)7yVJ0683;8X05$M9m`f`NN>g`2!B#5=Gv
zpNXJ#z0Q~ew9pQ8yhE>YdaGW}k`C*{X;wq_o`L+bGN&|`_V=dfn6X9aU3>pxLJ~HN
z$M~3J%}<Pdp*`03BgCY^ypgjq;SX!ey1~7AQ=SU>x$qDN3udS)fbk-_T(4w<D#(J!
z8L^97a+`zo>{a=FbLCo&wL0e^vS(El=H_Y*_G9+3ZlkOgb&LGL)*Mr$o$%_1tJ!$*
z<dsT0oWzG)Udn*S(e0K(v=l{2Nl<ya2;@1^tKAj9wBy#<pwj)q?AioA=6Q^dC4cn?
zT1rpW=#`h%9o8%!fa7f`Co^yB<DV3j^bz@f*Y0~<2D;^x=!UJ5m-QKJh8Nx<Oi^+|
z4Eim}0xSIejV7B5ITu3KFT)cScWn#7ZDUP}2XNm3CHO3fCxNCs#fJGJ593s}%c(Uh
z{H}Bz`_>ajq?SlX1&-=;fj_k9Fd34$TNOSuOVKfGiSAuh03T&I_&e)AoEoJV(JhQR
zPQvP$%okS~JL907E&}uhPqeL{+u(+O76!ABL<{Pnqwe1!%A(+8SwP0&juVB76gMVx
z4xL{J1TUtw;oI@;94;%MG9*%iL;~e9Zy2--4Ot3EQ$<>Vy}8-NcX0YC+JS(N;eJ7P
zy9ItjkaALB0j?&De}Q;G{RK{HzhK}t0}u7%6#>ctN?L}m@q`zw4I!*OjhF<o|3Qx=
zbNq`B-MMcG(x+%Aw8lJ0_iR%~5Y7X#TcoQR9B)IhkPZ(2!P0m#2QH<Wjw<_nMe#|)
zR#I<tub0;pOo8Hyf;wwu;&Yz!E!=wn)V~8P8GWmO)K8gJ01Flf@5|<V`>2AdTOaka
zfzu3EKW}0{N^Jjj(j<MH^HanJI!MiicKrZ<rm|NcUtQxI#v%;42Z?r(w4<2AOsf2T
z(&U%Bgp1cGuy>snCl%~Mt?|@2in{Ir4-ybL?G2rF;`oqmU9=_A;7m?wH3*q%mc~#l
zvS<@P4Qor$MzuWUk#PyZw@-7ua{vGqm5;|5?px7lS8Z2)r4ZP?%*@3}r>+HJ0>QGk
zeYSR<XtE;({tGgCB)6ASBd?aD3|}3lW5(8(L8unF18=dD@vN(ucf1Du@872l?}Z%l
zwr>xCg8P0Q5e{dJ5bh?&aJ*rmLRv=F0-OoJu_0LA*XrXsg%h}U&#$YkWk!h~MGlHI
zw0T((B0rh&-n<T6?W0%58extrBGk7@U_cnxRV%6CR?9kfBBPJ}+G!nTlJV<LE&N_k
zy8>gLzEqLEe<3ldj3IO}k41S$;Git{KGyVtE6I^=Y^L}!i|xfm_b_ZT<hA4WtD7uf
z@A}t3PI&M8w(#Fj5T+!MS;;!#xR!O5xkx@4;OPWxJJOva(ef$D0t~)~z(sz=pZk?6
zs=b?nR@8weuI2;nP3^fu;HG^|@=90G;^D+HL7t#u5E~7Or%5LfN!E$&i&T=X^-JBj
zWUdH*4b?Z{gs=3DBb%itheYAYe4qtH_#vK}{VPqLKcSYdc6r+WsdV~R6{eeaD`OKe
z61^#3l@bKY`15ww9S#Qua2`PKGI|)9Ansz~<bK2B)g#;~L))-5EPvrv98p8lxEFR3
zh_789eG0#uLf-h90Id%T!%Vu76Y<haGYZFhjaJS!=PNFkri?$%cyH}c;JyK_iYQNv
z`0VbKuJsUquxfyJ^`9@h%kowsp=r2YJP5la*p$XKQw2Y)OvwTP_6Ia%av^Afs@K?0
zaKrDV{>j77$IHq84}>ZJ7FBS$Pwq2>L~2JfV`dkXCVcX%7-=7FYI55uhYz4LNSqE3
z1oH&24A`NLFQNxCHq-8)tCWC^s`An&$6$lpQs>lzM=fQm$};KRu{B<h&6ig!D&?KB
zP9hmag{w@UIhkN$4d})xMXRiw$H588G<RyWm10i?uF6DDG%aS7cOsnDlcVO)^Bc=T
zV*T}AW-n|S8lhRJ?%UD)A;Pkc^d<o2GZtm;|5LK6tJMkKG|NSZqs3Vkn?2A|m!qL3
znTzOFmv$t6@hYpH5*~1?sP??y(&v96LmreCDM%Lujm*PH>Qpr-0g>Q^Lr#LTV$A6L
zM4&??Fr{X4v<k@*M~Yr~(IS&QEkG0uY-y@(CG7J%KpRJeC^@*Y+E5n(=+0yo(n?tr
z&8~Q8uIo}7+^drI9IA9s%&IVS8)#3>JO~ker>0h+EegrXHB}l)O*cb?f44VOMp~jT
zR$`v7XY9`}fML(<2}YRZ4bJmWv7Jq&GW~O(*;^Hi5NCD#GeH&}c7LKc4D~Y?*{u2D
zon*3B+RSVUcKR*w`@pLhz*9JPePMH3sS&F65?P9>VLOolL}Udwr~NNTkE81Qf01++
zKy7th7cTCS;#MG(Qruc7Zb6EdLUAeXP>S2d-JRm@?(QW(ad&rjhd=N4Pi8WAl8Mc^
zIcM!>?YmCtpXnHwgeZ%jAn_M-;F1iuFh!JW9|^|!m8>y=(Q@X-CG%op*Q5}>zA0R1
znKSxE==&CPT$lzpbs}pyr%73{GJCTzXDJRZ0Sr2sG=J`!r2Zf(YPun8(<Cil!bAiF
zN(}k3K&2KF*JrssZyA(ilRfOcQGnA0(K)tT>Pcwh5CbTME^l+FSlG6qFs<R^v2-8P
zy>M8B%6lQRn?T+C9D>_}D30JfseJ1sSi`qodFE@3iwEt^%iyQKIpCI|T$-=Z9IDCW
z15bZ`W+<wY2sh^7CR?GAIk{?;6CwemLT&Q3J=&X0`Zv@I)a1HBWK}_xUj#dqe_{Hb
z+7dsR$Y;J7^gDZhT+n~W<zp8kU-16Bgd!b$q|z?>9yg5GIQC<;9JN>4FU$nln_c?>
zk+4(rYIxXP7Z33&0g@~6>aqvvPj}Tf;Ggc?CwJDQkHZ)!eh3q$GnD1!<u*XM<D0j{
z7psErClc?QHGJ}HOGB;RelLj<*RKfTNTRJuu$St?83Ice`0E$Z<6@d<N)}+oZ~1Zy
zEP=pB&*|b^clL4|k($4#nk82_Wo6_jQIwwP^{tBDZ^G@9d?4pztT*4wmEqeIbmsQ%
zqby&u&N7LY_Ch^U_h*IvZb)U^u`@-@rVa{NJDDf^@2JrTdx=vFm$~cqg!w4^G0Mjm
zj&qOQ*AtGPEzyb_i4u?`{4kmy)+?0LAMx&y1^wOC+_(XezuI0E?13V?buw(S9l_i+
zRP~*|1vv*+3Xc*P@}Dn$QdaVCTEm-t++g@FQKv3MtNp1p$;>95H-Tm?-%Dy8tLt|3
z0E_7YH+=Dmtl?jTHgln-WdcS47|CJzM{D<BVJPi!4X2%<3I>oKkoWRQ_KV}v=*hhE
zznE?V!e1(b<^rma9E^0QJZbASmqh1X?WDE46^ekiUzAGZaXFVkBbz6KAMD>vc1P^-
zXI~O)(eCUW6Sj?+h-4T6?g<k6d%4a$27JmglH0~T9bdMgh&H<sVQ#^%cIFhxW6z*K
zO7~OadG{G&q8)%q$|12K%YS@5YOvYCai={hYy>ERM_D}*T{+qf`kakE7lW=DNU!Pm
z@|BAwv;vWqHTBd@V;(E4JMSYb6m4Uf362Gl{~OVf#|&T7`sB?P_J_H#WXARnOd;1f
zlM=A(Y{dG{@@kF$gM4N9IP@;UA)4P(!mLPo-m*%_0>NY5F~;%PpcB#hi$CV`6N>fm
zB0ABZ<+536)K7O>#xE`O^6HeD=6K(l=ILS9DoF^8?aUt~vi4Qz_8FCu4@k`?i%zL1
zqmkmVgLvMxcvl~<`W-9s6mcm?!xq4AdJ{tZt)OZgBqQN?o8XL^{2p`sO>ggVbJnbV
z&r{$ID0w*Qu~0r;L@=}bFUnxXKv+W``5iaIwS<<nHaVm0$0W-B9I-<T7O@G57;+L)
z?LD5jb3-GgW!?=n`GnMrSf7uhly+W^vvx=hO8n(2Vw<~N3XV)5)9;IMVfqVm-^OhX
z*#c)B5%9|0I@pQ~Lk#)a8qaKHYnI!sVh_oH?)NfOpL)nEIVY@N4%eHPEt&p(2?h{0
zl&~+q;=`9&On(;vwq~y>?(IFCoIHLbUA4c~eO<_Pxja3*IUX+98TNoVH{a<fPSJ6W
z+OG&k1m=_#N4O+n=v-gNUdKjFUc4=2b?ztr3=~5b^)@0(&0J1XEvSdME&5`mCnZCH
z6#iYUFr(GCQ1mZv$YiF`R;QV&uqLt~C#Lv1gUc*cdw8FX!W#qq2_Kg{^Oa3j-8ccH
ze+RhvVj1qUato&x6XZ0u17rN{UyhJ-Y)=Nh7AZS?430uQ_|y4<>Km)VCxO;ZwM@lp
zWZ5~jIhWnrHd1Mtq#epQOec06Gty`SY$uf~8#c<+1)GD0_39-epqNeHi~}sbYrXp%
ziJl#sJ_h6c%=DXT`a0OQ5y@l-=EYI6_}0JCYEDAQd*C@Ezq~zyqfO8cn~}FQOE@dE
zi{`h#ktDYz$DKcl3{-AdEn!d?l)S#67o|ZbJ$XSz2*3Wrh~EDubp7;MMi1#0C|y9P
z3>4e+FR5|*&tj5)d+2W;L?FibrotTq!{60=p7juqia0)e%gt#MQCH`EJMrrlPxaCz
za=KNUnnZ6i4cy=VEGAP0!w-h}SR8n9tW@08u$z@n$`_h(>vzidsVJIm1IoYs71G#+
zA2B!Jv?p9>cXD*Gu1(1Ti~(vgFj;wD0MQg<U>0j<0HIjwCkrZuSjLn4T_K75@$o%2
znYHdPWaZoe+EdR)cy3AdM)0u5hDSU*d`Fa=jgF_+#^W#KFQE{*`XjFCmZ{w3QM-&8
zIR;=;*B&odD^qr9L*0Q2*=#MjU;&jqLXR=fCo_yAN>071D0u)x2Mi}_GhZOm7g!$i
zCov)+faxD;-sV9`=y$@VHYgAJe)kf?WxIH^Zpmjr<#=1RK5TEP8)6uS3g}@}NOs4W
zkA<}tLrEEsx4b;o-V|7%Zcv_gAhLoJFT_)ypM{kfD<VIK^Nn{as}D~a$1~)~UA)^F
z$T^w!I)N(}D<D7b0PtCD6x1az&><_zDTK<v0_#w>KpnlRO#hNJ?%)zCxnOovG{bM-
zUmJ0CT`NWTSz>lM-$~m3{Y}yaKCbi$)9zYc-1xnnDu&PUT;}bFl~})1Yvt|eB>bDq
z#eozs7LMoc0zy{jP2jCKG<<f5`PJ1j%|s~{^G>wFYQ;)P3a}h%75jf@<Ja8(oDEw#
zn{T@z0wAh&vK|rk_@tXYqDtLlIBU~C{MF8ZoCR)@gIzUm=-hIQe^`yjbavx?ha2z8
za^Sol=U8>iYpMSpxZNKLikY}D=psVYwjex2FD+bQkFhGwZq{H;PYqFJ#k0~OCmQtE
zV7lh;0bqg5zl51?-k`^GB$@=`U`o%d$II6B1Z3G-|FltgJP+0mMBI^FbSBrlvd!l>
z{Lju%thl}!`p&M)o65Mnk$8FB_qGoy{Vs)cz3xGh^nY$f`22P#%;HND&;cuPS!b%(
z-_J!etGF2Y%z?hWNoos#aJ68y{A85T<^|HZm*ZA!+B0nV=RQX6_dFGU2Th302*fNC
zaf&w;30wN%RLzRIGU~W5t;G{)dc;`mRiRpaQb~Jjn!WF!eZ|Knwd65{)p7o=C7I_J
zq~6&8ePv~V-Ya8CZ@yZf{_eC$;I2uPBP!$LzHp>P!%W&U?GCbKC#w}jJSl;tFD5{i
zkowq^H5WRG?DCzN0%QANdy35g;-e@<Jjj5DE#b7RQxW!1_7_b_h#~vrM1j8qwbV!~
zirB$TNnv_T?~lcwJ}#-!AsH4mH(@5+V47jQ$nt}+#|5)DNPJaDoQ7hx|Gbg9PzMu5
z@MAfKavqH(`t4_vq~0Ms25deiNB08GBZp~0JX>fN_T=DDo(>ly%krN?@**uR>Spmz
z(O*!HmuYH>g^Cgu#tRV<8w#248+CHNCHh`Z>ene@Q)mS3U@$c)(`$1<ig&+Ijm?BM
z4!Z||51W7Xy#H7boyFJJO&aNZ*tD9u$V4yGj9iDh#P_M${51YGjC<eP_S+z!MQhCH
zo3au;Z=={@E1`z)v#fH`s7AR3-!!-1e~VNQQlCt1D4*pU{rv;4#|JCp*K-L!nSa{r
zbJUVlsoOdHR#F09Ux-`lcywLuDPRpgNfld`XmQ&6tamJ74#SG=)5$FQ7Kt2_K8&%O
z)gcE86OQuSn_9QkVzT3fZq%FrXli!%BWi>x_8^x^EYCKcYy14xJf$SdI(e3y5Zl-^
z@nTLDcLoX7nXz5VidK;Z!<2td=i|LaU;E$*tgyv2K_;=$+|Tae(xn~Ee%`|7zGl7q
zaN<6JfbVBQ7bAZBb#j@s8h}|^4k@R}7g5slr^+>Zx<|36^GuHkdFFr~h|k|=DBJJX
z$$5TsHFRN*Q><s+r2&?$KlkBgYwMx;c)Y|i=a0`USMNX>>=S#+!iqBM6K*wO(U;sZ
zp32vhS#VYv4XMYs)OV>rOY5we4?bkkD4W#OikfJPoC+7&JMOePABCxuk#(-LKD{VC
zg}lZ1iGr%uP>=2%#R5?Ik%`D=AO)_wK2MrRuDc4nuLhW93_^eF>a$_9VjIsMU`q#}
zPC?nMq=Wz7NJW2!Z&JI_vpgSdj#GzYb0SA(su(crkKc6l5F6kUy~dNN^7u0JuAAT8
za8f9_DJr6`MSLdGSK~XF)kt~s!j-*FYn}$Xj^m%sOW(Lcw-Ruu|J79d3*n!0n^<x6
zn5t|;s>HdF=D2xIe#SJqzurc_RjDO*JCv)+_?xDi`NQq!qc!t>@bH$QDXdTiefnnS
zU5SmC^WBb4#xo``2M1AK*aODF_x+uoZaSi9Xy)sSio|F#Uq7u+tiG;iJUwR^w|FcD
zL@5qtLU;{Q8aII_aw&gmv;BQ5-uFnHZ)gjK-Ebv^u8(nY*SdEDo#2C><V0Ul5iF=>
zM72Ks<hllX<-tuC+y)d9MhE|0Qo-hFTFhIEtSZsosW^wW{q*Ep#PGjz>SW2lQ;#h|
zuy87#(|RgUF20C?4|_cds=0hIFy&8^BH?9|4HLHY>bwNt6`80=9W~4{OXuM1kCG=k
z4I)2Mu~iTmv;Us$368r<KQSa;SMKhP$;>h}2`WLt`WDWzJWnu`<d`ZXYQF3m!x`gf
z7uN1!8nm5IKtTH4?iZJZOl@Bo1CeTx24Q%_y3DcWCU)dmerS0p_6U+m|J3=KI5Zgk
z+cGDTbC&`@0Co5vxyJ7Um*fP}ztuU_Ji2>M_^A=Eoas7{77_c)>$1qjNwvj1#_{sf
z>al&^YpLs9Z`9L4th(o0lX*q88XR+m6dIC34XU5M7ExN7T5o9vNhsU!KjAebi`Sab
zH+2Fl?GGZiE{|Got|zUEIZu-vg@!ydxQm{`fnQF5JmZ+`JP)Ul3AqJ+5<7c$K6TtX
ziOxxPL>~mivB)v>PNDSqY%2Fxc-1gA<cQ2}?98P|zm-rI{YY7rhRfzEnLx}!97i~k
ztn-u<O&@A<%yp`!4ebeR#PM;$MI<Ujt18Nyal+W_w?1Q3V!oCch9YqyT<mS`fmRbd
z9G99vna6LKs$@MqvEUBZXc5UxStOf{Fi=Hp{cSVkLmzcs*5Z0uePcWOzJdAE<Jggy
zUDkYHw&1rKR_S8&x^|C2B+UXBngssi(uL$eYnxKH@TSWU{_6e@4sskeb~zATiJchB
z&>uBQ6+aL+<_y0v&x|{F7p`0(NRo!2<eHEIJ+#=4Z91kuG}x(m0`he1_s60d4}Lyb
zFd_t0OgX4l=}Oifh8LUR^^Gx$<?!6-qskTYsK1(}R3uiyLFe6|^MrYj872cXC+-k`
z^RJmBbVLY?@6ssVVQ;_E;jE(-m5=fJKCjyZpy62q-;YJa>bHA<h!KU*JZ=M_%=T!2
z8cbfUHk^Xk)SLUS-52d!H$xCcM+r`qC7xf$&vMns$Ypo^?1y|xXSnn(PW7U=lUikg
z^4`zu%mGNc{3U<%Teqyt=kLpXmA^)Z@N~bZj&Ka2=p@?Z#`rIMAaG?19A^I7h8`5n
z2Ww{{M|{3{F{TX)j{CX6GkVie@=d`Mc%V+-H9GzX;Y_7j{C0;xkQS~M%gK<q{UfiL
zWaNv$PD4&ttB2XXZQmsmBWg`pUH)pb8X>|j46!$Cf4bM$!A%DI--r<sY0T4iem`Xo
zzH5;Q9^X8Tp|&GuQIvvuQhq$LDEVwue~92r^vyVwrAFm>;7(%kpLmR%<{)N&8;~xs
zocxp1gKUS8tVAp+SAa9~18uuD9jrUbB|RM*&iD6+N$D@@lli-Dnl5&d^uKc2^4a55
z+DrzU^$UNr=a4+RkLJcbA4UaJKg3XCNxYJiBzUOl^Vz9=(SH@NEKFD$-(#H#VlY=w
z@jUm`RD(d$kV$&I(a#eOY-qfxu>t<=IATk;VTZtk?A;U&?eiOR#oVU7PdM+1d!T-L
z3-bpPnQBa*Wa@^gHgnzM@Hxip++%fqHbm}E2s+fcZdW_}>D7SA^|i=LquL0C68X;H
zqB`m}ztIuS>TJd3a#y}8P<e97-xg<~-cax2o3)r^-LOYbhcIKA3nfYCVnCB)M@heO
zq-k?`wBYSSH86fUpGt`#knshVP11$EOZzABXnF7Y5ArD2%La`W7cTWYPVxRloD0k1
zdN~Kd#$b9ZaFyKHlnpMs-9M?<A44Q3Q?8BF)$^2MDZ*3qh2I;s&Xe~W?Aa9!_kJ3Z
zqJMu|E5lncNgec=2i~JR?w2$oJe+r@_gZdRFdqS_>ukl3ruT)BOe{t(>&45KH>^_j
zzg(#Yf2=6rL41xsbSX(B_(`*DIA5LG%ucGU)A8!e9!EH<7uH9Rp(e^mA&~%sf-w@+
zX5Y}7Kd}*YFfQ70p`b}HKxXof#2ka?!AHh7`1k4bI7M*|rEhNr!U_o}1rD2?eL|vp
zvvsTBsne+#p<%7;8{Z~ci%ZJC{Hl;rx4xq_prs{3)SM}o`NSK*I+;T~`*t@s$>=bB
zlFUGJy0nH<)o6=CNc<smb$12eaH_2p>H71WzG>9WR%f#J<cXaAduQJN!>qeognGUV
z5h-i+me(A@H28x`7~Re86F=p0Ww4Af-98G-nJ-JI{a>zq(y6BXdA5!&9_;s8zm!RA
zO>3-m#@at+X%=TXmP8M?iX7@2z|s%#5X;+I&~Po81>2Bhc;Ctt-BJ<)2g0X5DUGe_
zd5tEMhbv59X8cQSDF%s+4WCt%KFwFB*4L<C<P^V3;nPp<hJSv1jy9f4KH{b5;KKQX
z_?;B@QwfwA<y~G4=Wd^YNvXLu(btcLxUBqzPw8I`Q)-LrX$8iL7pheVLP7o;Sbbzf
zHj5~t48F@7aMbUoh&v^!0U24AB`k{GRN+)gw@vo?ec0B=RMy&JnM&n}+_5BSZyiB_
zocRSuE@EU$O&@B$E39o^dkk;>*f#v$_cYhd`76p41|xjU-bZCT+@L5Wor=9jrY0Pa
zU8K5*&2mhn5KK#mC;?x8NZV2)jFF?3`D!l}h7tkcGBEfs)nIzt3s9#}NtqAiMBnhz
zrrNcxg%;Jef0fig=4DyHys=GBjjldh{~zR@7vkU?$(Dk&&mn1RmE$+&_%(l_&Q=g^
zPm0zi#i*%7H6mYO{6TJa&Tb|%!7ffS-KVDdN<La4or)xzTt!cV6{a@!$;I$oFR&5+
zK`bM343-P>m6s4OkE}JPJpIGNky~vIa-&*oEy1r1&)0r;;V8ZAlJj2u@^J2QtGRzs
zeAL6?fUmyP$*6|!nKr4j%UanmbO##q%c06g02>G>Td^WBxPr=-D-z!HdU7q5UlNj3
z^`A?(x9KHSqPBcNU{b#d{)73xW(eG2yQ+1pwETlWINBW0sBgjhf9umS@@wGVSk40R
zKu)69$28TALFX$x)jag0aNC)!7HqHaaNPJa(i-VS<x{0tM<LFLiWKFSza2r?0g8TC
z<YBdMncHrdiYID|l-}F?fw|ndaPi51{c@B98%(ro_6u$8)!Xn#EJHv@UC%sb6!G$Y
z+)PVPfR*PP*z|*69gOfC{Of#M=KOHBk1)V=ss)uCJ6I<7Rqeob%W0BM1|7pcXTsG8
z`!uGkKqY;NDZmoz$DfeGpFMqhVMm~U+9@TkSh|L*q)rogH1ML=LhXu!Tc~?t_3nlH
z{-fo1j-kEPNq(CL_dd4{zbnV3B^cEW(ULeG0Y>H@91H!XlsO(@H|Gg;2>%`PbxPgL
z377u9!H?o*n+VC6Cd~b3Q+G3bJxp*v)!I^`{rPLl{e8X1swLZvs#{d_x+<?2y*-yC
z$JnW_;T&UQr%uLmmdV<1h*=n2!M3!86QjxxWrVG@;noeN4t4(KQk%jX*}~rqQpo7Q
z$>^^-W%^kIAIT@t-_`;wZGGyaZq=1ssMb^1i&NMoJ7-5tY)<~uAI{RHf3l>WX?DR|
zCq#zU%@~`qD4QgF`tl`nDf|5b{;xju@%&B8?_^Or)U?<x)9Jk^F}n_%6iqs()I7*P
z>&*gPbRLG8T?%sAMfX@VYNi?eV?$PT034*gk*p6(<<IT8T|P21&Ob<euRVVWCy`Il
zAHO}x^Q1#CSDMHQ;je#s;42iGl4WqxMU!=)sXm@cZ3~3+$tdl9W4b`g$Q2u|9s1eR
zT*anL{Yx<_g)83fxQM3+*|(1M<c~7y=S<G|fTq+f*Dc<!uf#1s+lk<4e(kjh^Z@*y
zZi3QSzNH*Pa{^1=lyCT{Sa@6K>g4FCnvDMsW#J_eA|91<0uX;Kk`;AYvMiFF%!6t>
zezzL4fY4m-t2i`kuf?Xfyggx<llgI`mWP>Ws8<hlC<dk!VqX&({#bGpoXoSeSV)Um
z?lXkwV{O+8%cv%1O)O0lknRuzV&*u`#T51(RsH!RYMGb3)BL7ejyMa(qxHm&+udVz
zs-{;x1()BRzp6dx7!K5!;{?f)zlqKIskCR87rT*yM3S%;N?XIF?9<VZ7CLe8c4c|i
znzYW*SBkrVJ9plCwH~cSbS>?W(U?2Vdf#lY?*WdJg?*EwQ?Ki}D{P1Y2tqRWyPa+k
z=_M@b60v?^YmX8g_4jG8`&*F~D#bSO6XGSKQ&j6=QFWjDag0F7UuVn&rtny)uUfKp
zvR{|-vp5_sND2yCx~oiU^HB#>A@tQUQQhw1|5TTLe>dx4z`G<6&uLP$m2+#~;N##=
zRYv(|+^Zy2yT58PsTvs?4=76c&t17?UHW<BFI=Tm?hh4gZ^EwWMnDu=$6V71(f87C
z^1-acK#8qF5MXzV*xGRcg=4mdjBi;gWocPG12+*y4I;!hjMMU|%~$MP*d?FB%?rN9
zKOhOihxf9XPi8+qA0cw7?L+#6^*0kBF7pwTUypjHkrf}ybRxBc0G-VVZ)dw0l>Pqh
zjg3$KY_%M@D29{0La*a32}i!`;h`_3t6k(@An^D5Hs7orsFEg#Z9^KzeS=N5etg_4
zZV>sMeEsyvg!7hiur$9S+i5?4Tqi2+v6ht1e#jIS_s)OvCV~+waZvx=-x$sx=#q>`
zQ+{DFojGjY-H@y#QeeRUC76WAgEgd^_cZ&%(@H>HaID+L(T#;!tE#E+a>UtLC?Z+b
zi~4V&#?a3t0tUq73<nOsm>G_)DKrl>U*0mXRD_scb1?i{f@E@KNay~1)wG1<{SM?Q
zN4t~FVeU$c25qq&547~u%CS?sW`1(LHYG=*Phd64=Oj6XTFVaB2!2Ge(Y;?P9{l}5
zar=BHeV@V@HB)+iPb#uwv-F=WT(94y7NA6-{g2vtUWR$b2+XB~^kvAujBJ^3ZtN@N
zU5mV4ER6s~$HSY%staWNL`?(~+y;qnNaw3)t$(j=bV}a5k!%oVM>b;vc+I3jo;p7e
zonN<sBS-y1TJ?%<?(YW?qqi?Pz>6h{LF5(IOA^^>pOtQ;qrq6#!&zeULcY<l6;5ay
zj6o3e*2z8If(|bCdIiLt+@PmUPqO>-X<-rfvT`WKsma!;1@1n)tl%R!hB`oK439Fa
z9tE*>()aEGA$q;wHbfS1{eA5l>pmruVaA3bW9aKI<a(aW5{*!QL5wwW$Byi|gQRI0
zhV<=#9ou=sz<4!uEIS-&W$6ev^xf|V1_)@6A@&?`Qc|45bLae0(5Dl9{29gb3^VIH
z6<+${zf3_YJL+Z*86Ps+g#KB9V6TI(DRtg*>=w7$0GQQzI<^+b5Dyux47Goc2d@48
z;-i~W<I!H6H>}gT5-a5o9X%R$$`28a3`gC27pOtE^yl{{T?FD-j&<yG1pIVyLN%@5
zgh9Fr0$<+i-HxCwa~~%PBYib|SePNK*?WGx-)G3`xanTFAE*dC&G$IS7-3b=O`*FG
ztv^kqUwB|dD-G8L25=Ye;;~ImS#!U7^W%mR!ZEI?3PKBmy`FDEi%f;<RXx!*EUF>c
z5JS&8;hSwD-UMMn!*x%G$3HyLn-;1*V+|20WYCM{wRNKVbn0jf^%C9DDB9=igGVj#
z%oM3DU$>LlETtb?qE~m^q>wRSN|g$p)F0fh>Rh+#x4#8|DUX|6)=Bsm!d@x_%<iPC
zsiL^uCawg?!HNS+qC8u4+?7EYn9}s%ii$kKtTtx&hWXf^uB}kGQ7J?_N`-b@EC5}(
zO+uF<6-5C_07`{R=5Fpic~OLQVXT0pf)$PMYmV-~-m5dNi7M*ebK#BKt{H2pzQU+2
zU#OSStv!&0_^P1ydUc@k(PxY_?*S5x!9H#BJ}6cQ!Pq(&KWKyB!0YozndBd+MZPF-
ztXI*&IIF_xAP;2odS5SsXQ-dwE*B8x@3w<-v&YGuCUx1FcKsnKk_Eoz1)_&6ZXYbR
zb0W8R3o+NE)s1ITdiv)EO1h-Fcp%Xi>3nr5asxCBeJL&yL^J3Kp1y;E9y{{7UR;I{
zZo4&gVCR}*VrH8|&w|V0b+NoiX7RQaiDLq483W1ZN^H`!Et^A~K5;~2yAY<z=T*nd
z=6x&skTwceL`deJ$Ba(SJu*8hFfLh~6FN_@lYQmH%K0TJE?6;*<zihI!9V#%-W}r-
z0l;`oL(#|9jq3!#rEJJxwDu_MMgSNYN8D8R&Ut}11IsrSSB&-=p4%v=SByA@s2v)5
z*Phqn@RO1>xA41rfyvY$XsKs3CKAW`48AvMScSlqJ*n(3!u!mHa|@J%^!9Z3-AYvP
z>UN!d&08FX&Gz=S(gz8oQ*AzM{f1RFAc}A56^HAH8?bHnz}w+#ATWpu68>wPhs|yS
zg36WEx{oSqHu+$+aQyp*0blop5jd+C&-`oollC$jtx3)X#hgwx85c{J{go8;pE)-C
zqy4n34fWNcxBXl}i4wozG=c{cqiZpLuq<;F>`!e@^mIJUXb4J3AaieM?=;Z>40^t@
zh_33z>^6U}UkK5vu$*=rv_&~Z6!5b;G<tkB*@btrHd$EY!>Lf9=J=U$qU1qY4L$o)
z4bC{#!GQmjOYyrt_zt_h-okqlP#sGmOjcB-*{C}f_gs?;79Q>{U-jZ9s<HxCmUFNC
z(VP+DAFuZys)Tws&RG-bue7iNYUgf-(62$A!L|Kf-(4odC_rPOg!6hd$G^Hb7h*S9
z+?0;YCgUihce6&BDu@~E^hVQ^Gkq0$50QdC71HyJZ;2u@Tp6i|D66*MJI)Ls@CV0=
z5D0HwQ{;e$Ebx)NHN@1<ugWkACxSlh>F^l|Gims>i_EW)1Peb2BVHH-1>JE-(eUgW
zc+y?zwTREZ3sizKGiIpX*=Yop(}JK86v=I8^+xv-uQgQqAK+E+3dX}dNedNgC-Igr
z3b+fEc|C;xorEEt7w1gAz3F!DbmiT=WIzF<9F21p4X*uz2ZeFcySePhblAW9vN_rs
zwc5UqBG5?0M+Tu1a+=%)9ByG-BG)d<svo5gZ4FxwtdT4$tuZf&e&%S9twgrD7dake
z#xM`vs^}yEj3AF~9tMI)QIty@-1rFsW+c^f4Gx8D?a`D77ouB|H)y@l)~8VvEgs$2
z?ay#F2NBqGeH1voVxlB)?Ok_IxrEVJHxZ~yvS1zB_o#*;6(b0MGxBfI;r%pxx8}=-
z2;X$2i#jE+V&s&Vx3}P>tL}Hh`!Bi*Bs&%%hO4YJ@O6j+``LA)!icRn(K{_H5Kqs(
z+obnc9d!B0@Mi6QoBEj$+)l25a5<{}#Y&OZ*%>2^2Tt~&QErJIZ&l9YZz`g%qoa$|
z0(K=*d&6~`(>bodLEm8@;(7WvT#ebb2WJ&r8>=hSysp#z*dHo^6igQ`a<ZVp`cvvi
zNXOMfL{$3UQB5;;m*L%U-?6Xkgr3N8b8desPQ`JX+81(7?q#K(&=@(|tmrT9h=0{q
zIRktTr0$ODTds`FyIO71L8ykAt}N4wD&XN;%wYcZ4hSLu_7YG{5}=xJM@mO6{246F
zBDr-Jl2VSha%*sfd*SD_py$~J-Sx<)S`h^`t*q=f5!rkX>g)x9n%N<(s~<1Au}7XS
zzj1*I-ygaNj}%JQynjk(UmqFV+@+}n54j-;jBdfkx;pZ05WbCSu&k@~<@GNakRl`A
zyx@<Vl=r|RDQEP0*`AKw?K?Zx3!*2TkK5-z3ZOTjWF7O17ayq`60PW+LwE>s>59_6
z1S^3#mK&ns6OV>QAYWtcKA*(q%uVtJl@5K!klE$uMQ~)KYOq!#FbLISOdM~u{)U^u
zhfi|l`4^)=hGfX8@rncEx$bL280;J{;%eN5v;uT2)os%8&}$05wg5}`sQnyE5qJ9z
zRf^n~mg%R5ds;At<=mfvg+khUREbYvK5{yBxKDg#+vnPyV6vGMa<mK(E{W~MbkZyG
z7Rt=`vdCL9ah(hT5Ov9S`H0BSV0%g#jE$OY@X+j=5b_MSM1qXge#%}a-%$Gs8X69d
zC7@cehs?tlYes^C;wX(*cu3Ch9yJ!4-yjfC5qJD*@W~rr2n~{ZUmX0d8FVH?*V_j<
zYm$SK1bewk+OSI@L(k^bK+S2}Sv`}W^Me=xzm6w#w0Y*cnV(wlBx?kIUhx1gGoX_q
zZ+Nd(fU1577NmPefuon_zZWl5p*dg+OxAO_zt%%jo(`<GFI|K|g;~dPi{lmJxEALY
zwFB#cuLbRT%Y(dUJPGgM!DpqKfpZcTV@OQZ&$YEHh9%vu7;uPS!Clf`)yfdeF9Qxy
zlMDM~)}1prZ5h$-1X7U!CU+JFuKpiDh0Jex+MSvtAL&|;=Q|V4em{oEozjZ{PniWr
zCa|PSURf{>B-E8xkdPVvn(s9m%hMn7r8}cipyTS*l)C$sre#p*Tk@UWDO4-)xANcW
z(?-H%z0)gEi`v9dlXpH|9+}}t1xd1e6^e`(zADp)n~%egOt;L#H8`B6*pWz7d%Wwr
zqO;j8lP<nvgb{Ag>gq?ku96dg#@k`BKXQdjZs6^_V8#67KhMEPRCQ4g>{SCA*pJkF
zPK9fjmttF$HSFRh<`oSR#%Ym1#JstNA~~Kx5pKGnLaVHJy{b!y-YviI2NT04IyETs
z+_nXyiR3T72|ieFyT(TenrE44({s>m&~XQoH#SIG*k(l3ucuE&_!a{FQLhN6wPK%(
zp*Z}(<CKjG+k?Wj4)%&$rv}1=np-)qCQJ|37j|o8|1fWtk|>y3*a2L$cbDF>gGrRV
zC4aTns3XxXcfqH}4gwIJN;i{-PVN*m`%p-|sst{NZ2;8jN+23MdCM9kj1-MMN5m0b
zqou2`Vls@on#@lB77AQXFNshE@{=9Mko?gT{&*>n6EYCH(h0`SLb=PMct{gJBfCOM
zmf|_J9r_ndJ;Ngz=nN&Ne4apy|0fV2an~lWi;@t9YWQVG%XT(Y_~k0_+*j9iE&VYM
zXCtShR!{qN0*UWDk<-us$bDuKCCo)ViqU=PM0wgq`j)D@q6b_)Z{goLoG~stIR#-p
zAaCzZ>-z}DJ*a*}xMoDT!06EE+I*ePsOZGIZ@o~%p}=&W!P=P&kzGxL4Kvezg~;bL
z=<a8BalPg|RN2nM4xP@Bmusg^T{3(lyz{gJ@T61P?O3gs30uNx=2$T5paD3Oh?!&k
z>U9{Semhq02M~s8x)=JXl8SQv<z-kP_~b902d=NXoeySzES+<?g6A5<^#UcC545@r
zLOdF$v#F>V%Mex91wl;&wtb~GzjGP>5T`P}oq~_{r#jm?!*pTzI2EGxg&wlU-0O!)
zg`}^bP~s^X2=%&rFcW&1`Sd7(Sp4uvGlZV*@Ood)0YoRRh^{JTU0fm1O>m+^{-uxa
zkzp#Lfn3fiyf5r<7ieR%8Anq+e;5OJ%3h(LoP}%-qEy`Rj1X54({7)|F}8<mWejKU
zN$o*<pgkKnIa-inH^IFK|2Z=Te_Mm`Di4$B;ad2-H*9hL)!xvhr5la%x^P(X;49pE
z5MQtjoMtrm3YJ1qnhqLRb*(5#-P^^~b9+cGBH&C~nb6PfR)2N2du}6kFC-&a%Yjx!
zSVUK*_$V9UR_{gP*3$Z&?t`>eo2EPBGA~pSPfA&ld*R_|6{4}GKj=2Mw+Nsr+j#O~
zuib7TkFE_(I%Nm*SNe9yjy?=69$E!yLHPiHcHD^(?ZCYKLPYtNajY!cx54Z4ok~&_
z%>*n7ZyM!bM$E#ez3YNv5R5XSlZK4z0Q>ePW}!?KQR5wO3z%8qs(YNuDEGtIb^WVP
ztvju#_PnqKAWc>36ah}0EgIu>bTD&*k)zX5-9?mkeh|l*sYF?guSC&5QQkS|R~c|2
zqCGq@ez-bXjHkcSyr1<$8nN1=#Am@pcdnyS1F6Kg4ml00<T;a28S%#xx)q3u$a-O#
zu`EqKK8ew>fezuSRh&0f%0d=OQI0E!438cGUUrf}_MS8vrfh4vcb4M-h0v9|P^q?k
zki9(NBWT=p48uj9$oj(<1pEfC<p8IU>tMcbeZwO-YQA7anaZ!%kX$i&8zMz+Lb&a!
zb_?W4qamjN6UCq4WYp=*(|T&hmK@P^ROjgyg^D)%3t#tv6Ur^1UY_v*Q>E?*+f~6M
z5>Z_owAZwlBerD{iStUCg!x|d*M-1G)*6$GKY3sR@NYNAiUlf|USGQO1$g5G5>=HB
z=>k3@?0f!=K$T4*yzdMU*!L=Px4WQt0pm-)_5Sk^s~Z%WhS`?;9EW(jM32NT01DDf
z=g8DQ#!pJZMRaAvE}{zd8mUJu`v$7Gm{A~Gvq-<W{B0nvCtn4{NcTfMnw(^T{FA@Z
zegmg17M(HMi2;ApwV$aq4vZG+RLSK|x{?`mFOX!+rX87lSSYCRlJhk7-5X5qu?U6{
zfu8jHF8UAf6f7pzjS4okU*3AVJ?7)Mezox#hyb&(enK@2utpX~i$<S$7^XP!Pborw
z*}-8zwr;o02^E$&PC#jvvYyu&m-1`Iu<_q#0pstsT%%NlYYLsA1J;a+3Z9z2)&q;k
zz1cRJQTOe~oq_~#k+1a_0$fC+X^VW%?@F&Qrjga!JVYTojtB7V)xqmc*I1rs)Au+Q
zCSJUaQh~BnsD`$u&RGci8}}dJ<7vRPqQ(I_C5#CEFQnHQEBC^L>mx``18HQXwTv|3
zFO-FB5h6d^Z^aR70TM_tU=5q5L_?oipSeVj;Dak9utJymk024J1>p}Wr(gX>qI7M*
z)BTN*OcuGVYZW{&TbG*QV=-FvrotT6$}IW$Mv&J@6?FITr|+FcJ(ZxRfYNdM!ndt+
z#|o&ZoyQNu-ms5qHpjbVdwV)(AUROb81i`?h~~;&kD?YhY)g$6UG@u#eD{i{hC%`U
z16H*7GEpdKg5PW6Aug<`rek~V&<kGs8iaZ3fp|VXK=tBC=;E$Y4L#h{2M?_lh>cK}
zu9W&8JAI9)suvEZ2*ZQ3${geN<xb>;{L4-yA4$_HzRia;kAC|bCE;vs7u8k=nhw^n
zu-Qs-Ktu$L$xzN!2EdtC9NBd^f|Eb<CE}(gHHl#<onVTS41TJ2V0ca>$#gaWh@rCr
zYtSfPpiW^VX-KPJBs&;u?^WaL*d#%(YF&`Bvk+LvUjY#xM?9dR5t4uEM+k@Y{LSv-
zApyL`4J(nIAe}4p_s*1TL8CDdO2v`Qq4ujd!~kc#hrki)Ho7zKXBZx!(#wng(?&qi
z+usMtgjd)?d6ev(Z3*HR^Z_Bt=PFUZa7hp8@x*a?Jl8=-P0An%k<!6hV2T|`h#_ps
zU%O(ZFb)e*Ud@B)X_Y)w=Rh#)Qf(e<<9H4`>JO-?;4TBai*R)R>_G&_ei&=bcuhJm
zc{>#3t`|#muYM-8d(Iw5r6&-4x3^Wu0soXzQq*}+#*y&MJvI}-*`7~%l<_C?f*VyW
z?ek=r`tRE&E|7-J=bob*&_$c4$$+poXM4aEM${&a7kErp$lR)ZLe%g<nj<=+LJ$u&
zV%vpf&M;ka!pIAul4q%Qw`2>d5CU+U1TQ$Z_U<)sqS5JGj+bY#ki8#kZ{QNdjwzQ4
zR0M@uk<=X_?^b0<5&3g?sos(!@(1(k1~E-FnP$k35l1nqJjsegZ#oG0+_lb|^4y~n
zMDJYR``l^3NAEK-!(JE*+X}Cz!#^RiB9*f2QxLTF!s~eky)Rm$=?c@_2XK}unRhx0
z`V_=gx6r2J1>ODan&rG*!wM$L{!;d?B1=d^#`dQEM`wlw76USbbm<RR(2;S%qpuUS
z4a<4)*uSN{F4FNUf=AW(!{{t!ko%fPEju+so+jK0C8z=owu+E~e5;;aGTWrwtY6&Q
zI-p`5CZ4vF^)4R$^0*a20KcoWAR$Hg*neZA-UsSpeRXry5VZMF-Q{}F<QvJ-vv_z4
z({_Tn1kdG(RoL8Cql>#ylB~mwsF6`qz8xQ8iC0ALMeN0Ax#sZOCfb%6O^?H)>a$0m
z+%uw~o6}N|kEAStNY)E0EU@RfU9HF4Ft@JN@5d!?Nm2^$&H)~p=wmkz@$y+BvhhlU
zQCCNe^Gn}nJa3>wRD$1{YSQEc;@PgDwsk&#vB&LY*(WtL`hvO9;tu)9OVah}l`tDJ
z*)bW-I_IAAzL6aqh1enKiRRugu}6Zx$B7RA4#0f6J{0aP2@7%8LkqWide6-xJ3V_E
z^hsCqJ$!RB7>LUtqke;8DEi*H7Ukq&To~lCr+;>h0><L694DGn(Fn0M2<j9lV3jP9
z=0<q_T-ZR~oY;v+S>rQJKZ(RN%|5_QMjZO?GS8t54{c>s%Qj*}>Jh#OS174SBvn8E
z#6emD75w;_mvhfBSh-Qb=t2ZyoN#0zFJ63lf(n1z4KVfG@`du-2K=jbH%cXx@+b2c
zEJL2qQ_6$-9c^0*HTs%7c0MMB>mP#G+D6ByQ6|=PlG0`z3Y~D9bWpTY5U=nI9y@t>
zQ40Nnm>Sb|{m^HqY;Q)AyU>#w4|n<XI)+`SVPRl(>$gj+U7t{*iAHol70?TXHO{67
zdCllUpd>?Ve_9nIG%EZ(e8F6W?CL9xC;>tT`t=r=ME`Tg0#)K=2_Ed=!S``jRisD;
z@j*enyrQ16BVn!f>GPn}>%v)EE|~Rhv7=ru<3%5wM<?Lhc6rE_X!4|=feXjCRj@dJ
znz72%KB1K>DmP;zsAC|!mDeCZ<bFtJ^hFFvpK>Q^gN~L21-C{y5uUdhYz|;C{7lD8
zO9js|21FEK&iz(!p?aVtBenYuw*4Ux4Sl1OIkQ<Lq#H~x%99|O(pzy?U=1r}I|ZFJ
zF5>gRV}bnXI_}#QsP=wAisJov%~-;EG2T#=XeJPpMaW~EckfRA(Pgb-;4ZL^(-Y7O
z<TrR6d886jvq9MT8zJe)&u>8?btb<n)?qKGcPkj{e#K2&xhYO%_mB_IUe5iNM%YuY
zs9i1Reh0n|Wb-1DyEV+3I%|X6H%1i|>wM|2xfA|gDC)39s9zIrZqR1f_HResL`n%(
zG0Dn<aQaOdrD5#KkNw+P^lNIHLr)+e3F>XyyEstdY4*1e>i0(iq|45HaY<SV#(Am3
zakmG3VP4%^hoKiGn~o@T4-{Ed!u=>0XPOuYB2%1`EYyDg@59r;Z3I?)=;=Q_%An5S
zs}cD2-=^N3&FO4aChq(GH3vgzU<}=IT%_j<oTtC6LbxUO7l9ex$)qA{&wzWLsLHGA
z2Z&F{g=qX@fdPIy+>C^KoiFpvG>Uso1xqD-yEK7?Zrc?WY2kQir=fK;+BKhU3-0wD
z@BE-{hi!(ZfvEcxnX>Q8F#PzY0`TuN#)fN-cJ&A*bkQc$A@71BJugKCo%@oUsix2>
z!BQB0rC#yxzPPDAwC0yWfdJ{ycosv@59@Y76Y*1iKLdV;4y~IJBB(e^JI`04Mc<1s
zic*<^nIh*5wiy#AHiX5HKvm8*u~Z{7T?_GoxTL!lD>`~DaR#maqL|eg3ch=7)gckt
z@(ub|JakAJG2J{Xir=K?P_EOapCaYk;^{C(jH|BqtE><MVS+jvXfpPB7m<e-EQk=Y
z0mhjdwD4K^iyDvnI$JNeGutIER7gH=CKkUaPh>ruhLOs`#~m~^6zS!fn?`r^H;r`;
zr3E(aIFcwXHe>7&hHL+{YldQcQ&dN}Dm%Qm{3xy`7z_1YsL*Dex;##xx~bLl=G|B>
z<kD|IJHL~wkrgKdAU#ftQDt*!=AMrIDg>O*L2J4>e<qm=yj~U#cQt}7gL+P|f4BF)
z%M_=D|07LPE4S7(jCfxvK=v$D)8o>%4c~p5jMw}ESsQuleIN{4!1&$uTw@mT2kLKm
z{}t;&MxIpX=Lgie1UfmWO2(D1H9u(2SYY4V#A5<uydqW)7)*uw$i93f`!W3$<wBi~
zuFzqYu5*-^RwSbrrgPiA^FnqYddES+OBZn{cCH2gk*`6{V6Rbo(}Te;@lS6-Uw$f-
zMY-U5Yuhmt#=`)YQ1$AX)hx!3PBVg$!%026OjOy$pr(jgvPBeuk^-wBrXVi|M)>KJ
z0=l5UR6&3`!u}z02K(=(g3y?*-4;sE%8WN+i8($r1kwAu<(eRcGNqk_ujvX&ru=(&
z1{Rtr8@*%Mq#kNfhhA=iz4gX%L0CbFY2AFb16d18B*L_iM6|p}A)B+|b~;<IUa^f?
zbze_gnaEt&O2Gb=sEMqMsa0CjG9G}m<$rRcNeJ{RS~uZh5qUr8fS`riX$fb9`jxCv
z*f2?kdySiQ`<>4|dP(|za*^YzYKe;q{v-T&(4tPEYP*$4{^1RJ`rR?Bp^Np$kFFfT
zZr}azF;1MReBJu}`3}v4yYbQ*i+8RG5IFO;4#^?Hr(8io_mH*xG<qD}KyY=dU7u{;
z9YDvM<AwT?0nbngv;V4K9JxO0b@R|10(0JW72Zepn{10GRInvPBGl(WucYrq+^3^a
zm9P78Ic>iGn@SngA64|OEa48i4ch@wxTDq;)S^chRXF9v-IV^Rg`Q#QN0CMyUG9O`
zR6><k*?D`S&ra#+k|cS4j|hqp_x3jekdRp-yrOn&UBkP74XCCtV&0Z^x$8D0u2(-h
zHFyrfj7Hm)FCShV`Q~`&K>})2FtLuB#Pgmnh1GfKL6`tasO;)3%Z2N47?mI$wcS2v
zoAaZFVQ2b~#{_4l;0x6Yb~;ah9ema$FAmfuR4b$IkGd0b^Z~OCXbc{0g9g}xA<OcA
ziqhWKT!|H4sPKoGiq~OM*Mo`R$Ws)2IA(_EO~!agxhg1ymPi+lWui@hH|Kn1m6(=t
zm4L=>za{?YG929ZNDS^Tc^*~@RP`rJ7Ifng``vq5sw7ki)1-eta?GrgeM)j+h5zFT
z?w3@}w??YyV}$=+FJ6Bt1>iTWtfm{;!?uD+UinSnEt?D7q#1$L+IJj`WxneBFxFTY
zjH$~=3xBx_d8bRnT<y_kg6qnQ_leyEs|ueAPaXixVSMf#^?GuzRWW=N3|9Yr8Bz|8
z3MX)J|0iPTrfWQye_po^9%Co%-8F*fWX$5v3o?RXr}8)p!mSxlhVz_`&iUQM3c>M5
zyc}Fdfwz*ha$sk$vQc2CiIj?>;hhFOXb;xPXr}F?NEzj@yk~_o#tzf7+&WkPe4c-7
z)AX<;PtLBRoR1qsqgZb9>gp0v(SVBU{v*Z-?{tyC7x(09xMxu*|8N0@A+&3AnHr<&
zK1k;ysu>KE0wg<?q2#(UY$^Xp?*PNrJBKl8gi~$ASB+bhGcdx8JR>VaRfA{Eqz;`y
z1&qKy_#<*wFoii+mw&%Pl>lG8oy97^8{UR5?jU$f8uO75Rr$tGr|>kMh&r@HvIQKK
zko(M>;tcl8ir8ANyuxiKA`w)iVvl6^Xrl%)W>Fl_fTvyU)l-Y&|K4oJ6-uG+(+
z0_siHme}7P1S827culyAdmIdF&wh=BAj+Z)_M_mJj8TOvJh<T!DqP6}-mT~IQ)b^f
zY8VOiz%AIb5_!sO*knXEne7g+86!)z0pn`#d)%=gMN>L6X(u3f_!7z6fhsrMR`o!)
zp6-C7<xRBvbq%V0NsztnJ>vxeUzMakI7-az=Z~|nho#CI(aa^lxOdEws;4A7m`Rrm
zgs8tljeq1c0PQ$Aos1UBFhGMZVfe(m))K#E@efiA^YB0EI$S9{9g}aQoVg8ljw;GJ
z#PzEhhql_bhgVY}7xf)73i^88f<h>-k^uzn$uXlHg~V?}sarWZmE;n>nAW(uAli#d
z7Z}oF!B_X}_JRgrPMIdIaHZ!^^B}Ee5(O0n@PkUD1gQzx9u6%oA%eO*l!)@%<B6vX
zDF%j>&tkIutF_dEzKADHQK8<;Ur2lL#2);ikt<7H5e>LjXKi54UXrzz0UJMVfLo}t
zqTg?Y2vpr@OuB|J0(o~KTXCv_`3to%*;xLJ(3qt?TXE}B#TuNcMQzu%AkMF#OYmWW
zs+eyG$MLKbOaF7tOQ*#o0cLV}W9oAX2E0N6?W*=ic-~r0zmR=SN{C_hGfKOrg5Kj$
zRG@pXtW+b=?UnPoi*inpjIBcf@S~>YEZEs{4#U;xL$K9~HNc8w?7N$Mjxd7DpW1OH
zntqcgSfR!SG(A`cC1Gq$%Zel4g?Gg0alc3-U|gtrwhzi}dGpEiT0nvo>qQaet1|cn
zoRJ+K7@VVVstGQ9)~%mD7pvoDtxhuZ9(k5^2|j;LN@am}UHN{)w#5#3+7Yj{I1Jjn
zN+7o!aAzjK-t4Z+rZsBuIsEW&bJxa4DN*W1fF3;LbT~>!dzS}pPglt8Z1#LB+lzni
zUY7u4<zO8%2SF0!g%IVl-i)O+78ed>p-}!ac}kb>XP6f#-{?VVZ3^w}uO@<d-bf_U
z^k=VA$Sxm1LYf??YJir)v7V{`!HzJ}d0Qfwaw#0#ULw2oVBSBreO{M2UXcEiwjP_i
z40VoALDTj|Fr33lfe&{Xa7Cy%-e>HAncNJ>=}8QG18=OXL4$)rn+O)qBwI}(YbGCG
zFaODo3S$$e^3{l1o+ZbPx|$jLqJFd-QItNnd7K_U%vemx1Z3RWn>v&4j2C{}oY<mt
zxQO0d43iC<H?M<<E{MUnVlFZPu}{oM)%-Rj?JbI1S#%nNFjdQ}Xm-xE*N;2U?yHr1
zZ~jPyd+(wdIOx~#OtpGfk71&Z>-@V-nCbX9eC0+;?^j#am?WZI95`yyt)Jx#6M4@v
zlpOjHBCK@}P~hXaZjW^>L>p|3`m)lh@#%trM;ZLcSAASq<kap+qUTyLun#A6*wZ?n
z2+NUxUbu>0hAOF3A!Rq~mfJo6YFQWC*y?_)I6<F0Np>)UTlM+@lB@XtnnQF~bUcK&
zI<;Tpw+Pb3J)_p|2D1yk|1|c6<~Ar2rLySeVFK&3480}o5^g6Jv$wX0-cL8PfaioW
zZXpLMyp5Y;S`Rnx+}5hIJFEqz&$>gMO6PhQnLCr8<V&C6pF2KPA8fT~^xJ)Dtm?0f
z&z*_t`~Gz`6E_rmB2RIpO?Cm$r9ldg67U*pDSzgK&G!B@i;q`$t?7twP&zs8o_|V-
zBCrI+qg1oX5QgZ^mUz8W#XZATl~e4fDa26Ag$eUwijABSC4!A&GJW3?mC8+3UwUJ3
z9D%xNOX<rP`aY}$A^`!%B^<M^*+Lm=Tgu*LqlM71I9GcpX4Dxgh4uzc1q<YEeNp?w
zJ!dBJ-njs)%8U852jB+H34w8W2)y}$hY_4R48HFk(2*;MIF)|=kXJ(+#$Ne9?SYV$
zI_I(+k^_db<e7cu-Y?WtIH|!Gmzi$NsLw(e2hZH}q```P3xCt<B)!W+%lL?2GT=!6
zf+-Ex9ZmC`IWT%RZ%g&WPw##C)L_+5o&15VL27NQ`56B_3c)uzR_NE%)~sUSbqnEL
zDC$&ontRtf0i#ZBU=X1x^}8UHC8Ii%Qw6b^SUj_5dkPb#7kvohp@^@$(D?KJqv*`T
zp?cpio-vFqjP*;|21!Cu$v&6~jgU2Dsgx~C*%^+p4l0I1g&`!n?E4bcQ1-GL6S6aQ
zW6SUS{&Kl4*WjFYdEWc}+|Mhs`HKV5n@^bj?;YPi9#!{%j409aTSmc)sW-YQne2)K
zC_dVv)O;4bwATo+Kk34ZZ~CL|!jU4-m7GVtQG|c0(C`MO4M@$_TY%s2$NQua3GOb+
zqsg7NC15DJIfQ#{=)IyD+vFt0uSMV*BZ{Qha1iCZ(vw9eSMlKVeui8qY}!Qh2>8LF
zmSSVKfhjd)0k)<fDpuRW5aAys9085BCd2{vvzNuzOz<Z0$daQsyla-n?i@Y`lY81P
zFwly)+6a%%*R336;MZAqd(kW3P3N>dmV4RWh${u%4>J(SIL<S}K2Ffvtv*6@$T;HL
z9CZobgM+lNCpgkV$vgY*uL~H|yIKM&y0!?^`X0d#9v`xOqQ!?99bqJooa^+m?FDn{
zamc^NBDPLRb61$4?ed@Hb&-Ro#JMjE-T3;)FJ#4?gFFeq(l!KC@jN-2pSJ3vk*zJA
z!8PdU5oy%N{Ve;LoWF?g3(4_K<Xa+I2w{tS33*#^gX`e8B8$jB$Cio1Nw<WupNQ0R
zFOlglRya!ye|O4H=>L%5_DVHbz|hN${QdgB<OV#At!a~Z4b|y3Fw6I*eYxt+QHgrX
z_no>7W`{?<L7V$ab(#DDHU9z{9Xrnq&7waf=-JP>S7MqZ`1QS(b6%BCt8Tusq^rQn
zeIl(q3KCx?9>?AOv!yavmvPhKEW)RmKd>3C{IIY^<{fEgzlJm$LI*CS-UtnUpi}M7
z)9KW1Jfj4F)$s6(EsGQQjQje0h%w|}HHls+8c*%H@-Rl&4)NPkH1?WGzLGF2^jH;S
zTZlJJNLf^g4AWEfhR;o35fw;8ENwCbo_1Gq>+t>*;o;_u2EAR|S~;AJB04QGm^q#D
zS}!TztjS0(_mI?4cXo9N_?S>1hrGP=in4vTjsQI0P*y;UO(-DF&S`ab1T&*9tah5E
z^f3z;X!Cv0+Z)bia!iS1o7SgrY5p*C3=-*<)R;D5LGLJA`A>@JvB3K`xO?L2#koTG
z<1knyBEi{}T-S^QQIW!jqQ8$a<u(J)Avoc6(Df*vO6JR5Cfz57KeHp66FYXeuby^l
z0>13$$OZU5_Y-cQ#&_<x%9p@THv0H4M@Qvl0%#MwK+n-=Uxe^WA@{|nEp(7=r6N}L
zbN<-gW6iOn^Qh>3_xm&AMX9ul>7+dkF~W%DfIZ~US#D}UCjMzZBN}Nv4-qofErl2N
z--k61H7sRxLn742MZNFl9;r1xQ;i2W#U`w%f?X@YSoX1KjfbH5pVBMoaOZDJbBy0>
z&Njx&bSfpFugQH2s?db0p$bTC1xz1ae0D&aB+_a{XT%Q>9L}|9OBFTW!j?jV+#?qk
z@Z{k1@<R!e=bsi2QTg=4Bu#|bg39d@4U@}*SSlOguS_k<E>Sh_x~8)69I&}b)Wn7K
zXAjkaB3t>+UvUn^dq2JD7W#cjhr22N!scx~40@6T3;QbgaNV--#VH8|t?g2rVaV^b
zRm-Ydiy5KJ0XA;M4nua@M2mJ2#2=2~2X_X7_vPsXOicIg-Xy^Muy^`(y?@<rGrH@&
z`|RVhbNWO1LPVUBD{j3Q0WcJYTQlA(MNrBL9}gT!!>-L(3r}ZT%xwpD0ua?)ZBH}!
zpeUL`E|m@AXze}>Gac0EpGCKsnyC7`=f#z3A%5)doLj32U^EFqf6#+52Oqi9sy+T}
zhU7*6RL0+_)+1!|DK)KOj!RuPbBr)_z1o<r1$A3Uey-&Gfon*hCAN*(!{z8=0STvP
zi4h5SCg>|ASOV>5v?-u~NxH5jAi9i|SDErMag#mC2%7si86UN8s<@a_3xk5N&wY-9
zIng3(c~NQWnA?Vg^+z+aj3yLE-?Rd`=akgFa|e<T@`<nltxn}mE1%f3I|oOAdJs+Y
z)|j}@h)8Wl4KQtv&@k0^41+c6lEOH%(@;kJiZ*i$?#=mq3w$r-xY>qfIf$FmkRex)
za#{aor?M#5e7=7$D9YtNg5GD#P6-tk{IxozE)p!{nxkq?njnN(#kaa69qp$(^D8uY
zx}KwtrfsJC^m|5Bk$W>r_ts9GB3TH(;nBob*E}=<7%h{j5{6PB*}*qn<HPcwb<!zO
z@K+Wdyto#!J}242T}s;eZHMSrcYSfbtybgt{y|DA^0L$1lH$p?>veyCZm5fL_s8D@
z%DYk+XmQP5;f;<sN@a6mt711SbjuC;MerAs@_!26qMna_!I}b1#I#=BE};XOex0qW
z^cjl)Zrl|7X?L#SKWuM-UiS9TgC|pIV&PAebfS(E1=#0Ie{LMyU%vY2Op(K_MID)>
zwvKC$z8T46eUGQv$s&Y~u?U*WTujEZgL<Y^+Vav>ZXY>*0k0aTcMh?AJ8kYahFCv`
zzZD<-Pmv$b^iz$(n9eYt5^sDyB!z!&x435{4i8WS%|5;34HQdtzcC4O$Njrt(wv)S
zd57L)(S09Q6(%{`4~(;1*~K^_pY*Xv%v#>0Sl#;*fe*u%g$c<p;zy=ESnT#nZ?>yS
zL&gu|V2Ej%(3abOxbUiFJfe0LCc6jasbRzQTsxM0RupwRm)Xi@4ex(kqa&D)Y^#X|
z6@dR=#@g-fsS!_m2>HVsv1F?`+P`u{S)U;Pfhgd^n|3jn+0m}VWa;OW108%LtpxI>
z;L&f2;tpDG-`*EF8K}S2zWnPfdDvVtvAO6@U5`aDk^_!o>-w!x4p?xsgxWajeEhon
zdZK~;7x*jQI;t8=JpaxFqE577|EF2cfT@45(1i70$QWKat7_Aj;@Awk=@sXvW2_s$
z0&y|=Pz5u;Q0Ed>pE#cEO3@ja&Tpl3TOa&;T55Ahn~E+dJ8&{Cy4TVVoK6Yj57$*g
z1eta^&o{!x#Sj}$)wSkH>%3ji9>Y#c-4o_?VoTf1^hub{*a-u%pSfS$650Tq*g_6_
znYpokQjGnK3B??(!%}k7cRbRem7A!nvUSPC9xs#K<V$Eink>3)<r$X&BiBKj(X-s6
zj{Q04CKr6;Ryy1EIrNRym1O&kTY6>m_2Q7A8EKfuf~lSLinef1gcUyDVtV#gu9n<{
zi$GIJ{&kYU%?b>~y>{)$G8VuZGkXPnP#j@(R3K`Iws;-FG-eiR_!$+njIdw!GEAb2
zZ;=iHePBnv+_-=f@1c5}0m&oiQ>nEhjhzkg`s_|FJ$0vxZj~Zs>i#vE71fY=+@>6|
zg;`0FK0TBc`taH4EoC;S`@?T(j>#}1*(yd^<eibBi(4Dl7LVn(v;mLKA#vIR5m~z9
z!OXUigGsqV6c%71u&WeXoJBy408xdV=$S@aycz{F^e=2;_07xm{dP<<62-ol;)!ij
z2%3duQj{b~%%MZ!l_9bA14@L+KyGy0^o~}@@O$xqcyjv5PeuY`?b!Ql%RCYtwOcau
zd#ixIUaW&#%U>K|5MRr;ai*pE!0rt~sOn`cd);+3<w=mQ(wBy6hO2XyUqmdb)J1Jk
zqW*u^=&t>qI1T$3v;3Ca(r$XX{BhUwPh$=RXNE_H(7FCaUsSnpGGIwdkiQVXYOltr
zdOSm)vchi<eiQCdiD1CG*b8!c9&`Qe551RtoX6quLS72kXvDrfF1Oc6WWA*AX3R3D
z6&h9H)>_e~c{tDsCBW1s5XSo3znSWP*<-4vLGWVDO@0mvs7KVu%{WQK@zn9R7#_$W
z$m1EfhQfHqGT*qir)2`U1>?NDIF{6Nb=z;!Q)#cb5DsGTmqmoQ$Ep~Az47i2$fxR0
z9lh~X3<Bi+Q5wWPC7tLHQHXMWYCt2(F-Y^EX-ThKMPVWAHAS6uyCyp-Uv4N^5Y2aD
z64%{loI`yXq|<2U4<93I^WhGjAUM`m7){4eaS6@w*UIu@+b#!z0f39)eZ#f;;yr*8
zN|fV7c&gW;;pcy%(<Y8z4r2e%ZVtWp9DHy=0r=zoy{>6DG`5ySZqxiZ;`_*FyqcZP
z+pB1zjYLui2(B2=(k;g}8kIhtJ+kgx*-UMF)!Iq?mi8*4x+m>@YX&-cUp)4*CO2}9
z7D`j(Yj})!+#H8(F?sY&5NS<T2@O~da+a!r<X<Ic`yx|nB~ZfD$J_^4-i;UPG`%=I
zATrpHzoc)2OTf;X^KE&NNmzp`VM(1ewT)VDzO$pAkPy(DoAY<t2;qLhMbtc$va`q=
z-5B~<JAk&MK7KFFK%M1^Z5gT4+p@sv;iig!&ELe-0HYM9pLl+&7OA%+T~!pH=5eMP
zw#vvcINk&rG!lo^7<%}Ti;{_XOp5~o2!Gv@pA^D9Dt*51Lva7)k+Qc<?LDh)XDS8c
za7Sj)olEfk1CuD1bE!0Z&z+4efh5G&U}yBZS^y<Ve;-;wLElK4^%h5IP<PWzW}EY*
zDw;-&=V-2P-sZhNvVTXvmQk#E`czr%H%0<(?;9H1cfG08()H*0%GyG7TmoQme`ZAj
zm9_o%r*r^t+yuJ)`}1ah<hp{Uc3kLvJ{R|3PPY%=nZwfru6air#0Sb8H&1p2;?{|I
z;_arU-)my4PZ1*bO6w&KtCFx-t=u71+5V*m^D1o`J-f8a^aTF(%W!JfyLtv<P9XE*
zNWG+H>JAs!MZl%ypROYfoa*pT8~kG}Nv6<CD~?a4f+o}lx|Ri=`oQZJo_lKSi9wR-
z(Hi~IxAn*vS;%%Jie=J3(twsHdA`30g`KBfhO79-8q~R9CbN`ZGgaz68Q^L&UHRHx
zsqb(kWMB2?SXlN~h;?p9LF&udl?b#(kY@GKc>%(o%2%MO^g2%i&`|MXz84f#K}2VP
zcgo+ga9$c%sEEB*?kC9=QdW*>gUFop$L)<c?Kz+0v!`B$5J*<D(2TNgeNNWklhb%|
z2fqI8T~flUSTmyMfz*ymLZ2>|*pi9&4p?`0>B3Pgq?LF7hm>}-_^mW_(2-3yuZ#2>
zyW+^mN{fbtOTeAS++CuwZc*63n6{yNE3E$_#<MDpxXKIjMh^5+na1*~wXSC}{jvkm
zj^tG?Vhq6StX<s!vIUTTo#U#LyVbKci}ItrvdHh@80<eBi4kmF?{?_qb~Q|HcqZI?
zd@xdInkzKHu-QJsZ$gyLN`plTGPYN_#1i>inxW|_P;$+_M4{|6=9GRnEVZr6^>5z3
zso1#GAeKE@FUIEMQr=<X-0)@xbzp=xG%C3BgcIjtgn+3jl<1agdlos;GBewVqd_o%
z={%bttKT}in+}BaN;|Ye(uAa!QXK6P1A)Hf?1`X#$1~>DMByCjf@o3*V`q~*PGfD?
z{Bb3azuFiaiv+L0qm(TxAKr%$Zn<UXOQjO%k4hH!cM9zwwBg`bFVFZk!&S*Y`FDt>
zh%IDs$;cV0x)Df_XBl^Kfn7r1yENMm=Pk@;lK#b9(*w7}onQkzO*jfc*rOx2H|u<T
zuLpCx9Y)p`;Y{u~=Me6BCHRkt!Q<_-_qTyFxTQG+u?+c6{6^x4PJ<?vBXf<wzSt}t
zLh%<6PyzeoCpL>G!!~N-a;dm+33!}&0(K*qebifAXJj3Nm7P2<sU_T<XmwUKx9wL~
zpei3N`8vz5>K?I<foTOD7HPHW7dHUDTNz&Sw`@ai;YqlA?VmUT96i26?yXE1zyQ5G
zp=sD?drqr^lA#OxKjtU@QEZ5oB#4-7P&9w<<O}AV&T~n7D>r6g<3k2?A9$4vSd0?6
z^!~n87|mDuNxyOnv<q*bMxBB}<QAK7Z1QhB!{}OKE?lyO5N6)mZ{+o8-xY>$2=|#R
zr7lB{hHiT5=!wFwk)$yYasjkh6L?wq#lVJemT#+=WAX_w#Kg4oj#D`?UU7o{wDJq|
z$SI1y_e<OrDU-;i5=6@kGB?8M4_9lEqb^d-?JK>&U5rY-{mO*=nM_#*lCE{ZKxH1|
z0T$DsfKAxDeqt2!qT1Zy)*Ge+4vPb%U76sYYve-jd$mEwzYqj|&9fiV!1|-*(v#8@
zXKEGO;PYxoXzuSzb5n8xljXq^k(JHk;j1rx*Ys)e2Z}BDR3}*E>z4Cd#8~YhrWQ)#
zjcdA=lMO?rTfi-y!fow;K`{{qNdJoJ6>?b5)1xc#{?3#_Yw1cxD$LzlW%6xA3|Zd5
zU*4!$XKE%c7&1#FF*&-e0k6fD8%<&7yz>931{$tjz2eXAl|yY>0?Ed5#hsURP!C0T
zf0@5ki%b{^TtF5|xz`VEYpI79A<cbN%v|Y*7z(p>G;Pvj$|#*5=h{>se8zrZ{p=3S
z=7__nKb5L3UZGVse!_t9+^6oIdVdl>m2?mz8vk*OZBteEH@*UJT1RlR2kaYVrLDvo
z!!9UL+(o)Lff^$<(JnOBUmjk;ZWI}+z3sexs2;@J6!m`ZHgwD1FkCJLA%fa6iY5yn
z;0RkOr+!_dQe*ub|FeYG$^46RQXM=3@K43ufnu4yw{cDHO!c=lu1EO5v;D30Z8Uco
zP7ck|<Zc+TQh*+A+{GHE#fON*?B~`y8BkP1*&=UtaZXO^ow-fkW<<Y{ukgEgS(EH9
zfBStT;GfLXjlvGB5JGc{-tj@|*>ZXahShIU>>Fe1*PcDoC}@k*U~ce;MJhSK7az+i
zwA0h(*a9Pr)<eT{nG6iUEzuM_Q_aSZzEQ_S4>GyLfc$^b@nr36=L1vLxekqL?XRzx
zI$va}wCNt3rS)rwSUHrdlxY2HAL5%Mskt9nKPAu$b`J?B&KVgcb{+dH-WFGAV-HpN
zshUm|ilm#P=ZJpls%d^Z7ni^Jl79W0l1E*{H0{9d!S<cAg$njNF-<OEU$;J<y}PO;
z^eHt^k_I@ShcHFS^rV4WV?L5i7AujoV@pCi+=wL29+mz~4f?Nd!ikRTQbHLEUD;3<
zMP<qFVr{}-^wpZz@so$!$e|N-&HHQcrUp-`gzX|lm}j=7Cc!GVy#N?k*x2K^yeSV&
z)U(f8ekIS-U)R#FWrBSWVYJ8!Qwy^fRi(uP%*W>Ts}3W;v$n3oT)PgcKApme2ay3-
zZms%r8ISZE(%m*f3zyTiqMq%@>)5-()0QM48G)z^|N0^uYUJh16I+|KbPgQ#+pHHj
zs$Q0%Hn^~P5!FX;mKx2#y2yKli!^x9oqb~AAe_ha_&v5-TX@qF7NFI7`|fI4r~{x1
zoIRYuCtlyJX8&C;(({4;t!Pn8jhM`;8;4#?S#u(KcNZGSUaP}c>`ZO(9FWY~CzA!d
z!jGWD<q?rjr|?bB2Mqd;k2DciVlAP)1*A=ZdYQ@RA&5|mO1<5XP*ijRGFHv*+<mdf
zo`3JpoHQf6+~N%Ex0|)21x64<x<Cs$g*xRETdR$((F8w4{8>~|Q=eh(z{E#kC7ECI
z!FNk0hQ6$md9C*>K9O>blH(jD2Z?QDmkHfKj}CvTH)kN$Fo%iuKz{SCuq`NUEXO&D
z<nTyW`3LQ$M3Oe@zazdot(CmL$KuxCZeRTn|IdR*dP5yiUi5(v*WyG2N(&HGItVLw
z#KaF20oI$&RP6hrjQTGywpPNzfYtG#f$@X-hm1z$U?RSR$)M5Je-d__c^t6j_jRy&
z#5j)iR}21HSq$C)ppUz3?|ykJY;LHUQj`N0)gp+^ap#1Zc|{3Nmgv3(?!cz}XJ0d$
zr!GRwHBxE!ZQ=Mw_@{t<2?C<dFTdF$NProJuna70mz<+)c#7TqeIRu9Psj>FNFp%F
z@DE_bQbv`k@@PFvBi_S0n&%#@`oPoXWg!}#gprANGLmWEpEFOslhzUjo$3{d;XB#r
znKfnjhG3kJr%>E^>x|<=wbyT04BUCG$T~6=THbn5RcGNuBj8yqwMTU8Af4j-9>R%}
z%aOT`--|U%lq)!!xLF|k|GJv4-aGjd1qj;OtjA!#h3V}EUIdv0jYjCT<6rhctS8&m
z7Ax$)_H?pNO!AwKjj=C?iQp?1DJFmNM%T3e_&-c(+uJ(4;hseeq@*vWed%h*kfH25
zKRVh^9-=FNU>j3!XEL4#wug_pCEKh>d7=WM)6cE45?q<FdC}k-FPb<CMuU1CLgR<S
zlpIg_uxyY>oZi;;^N6qCwb2gNVN3=_JXRX^29Y}hlZ}0$pwT~n3DROnnDu5xk^V{3
zJrRvX_SwT11tqo)k0Z<-1rM$#ew}MRLU3QBa`CJCMzVF8a{&eeP0*c)BvEnz^WvQ3
z^^NS;zyA2ixwPF0{&5el*0R4XNeRBRM4w>LH4uCjt+1AHpHeDcAO7;Lg4v6>r-@>D
z$W#=8Ju}MWg7dk&r^LTtC{P2z5jxH~Sn+F7rgoj_s(tcvSkypi;rC!?c{-bqzhK8H
zh${a}jwz*sm(~FCFct;wuGh`S&xlyMa7zTb^uxStcLg)Ebcs?v%+BoNW?&T*W?_1`
zwMj9jpVGYUGaaQ3VgLlq&7+cNen<0{XUW<x6B4@P6ZuNq*Sxo@&LmcfLYh(Xbd8_q
zXWaDV>DCJ7nL|(CJ>Yr+mWd|YbP3+FbnAyfUllk%0gJp#S@xDLE<5TlEM2vl=(Oug
zjSd<Yap)rC{}RqiI~gDFS|%s!;&Cq8SJ(F%ut&dbW;mcK{k9OPy4%59ZXu2SwCm1S
z!~1NYWTVDjL&OeLrQOcDMo%?Yaw|wrjIqq@2VL=0ig<uC+5v52N9l4si1HraTBy8Z
zAp-{_Op+BCsk-wVA>)zFL)|bFt6MlK)Hk-zjH`gc?C}Pax&$(<M>v|4g>KQRgxgY<
z=u{fW?cg7x6B+8lLBF~-VK-K$#c}1g&fNAGThaGr#mDkEp}|?GO{FfW@V>e^J_H}O
zke|O+S&SCmP{#LD_2p^jgI;CD@1|2e0&9V^r{6J1R?bZUJ8*b&ST!fEu3Zm%Gvr2p
zQexU?PgtwGWa&wVB{!pqJs`(y=}Vy7FJMX^A`26v>Vk}_rh~Qv*%2K^9$dIWNy3Oz
zRXmnrD<Fpa*UmHte#KG!B3Yas=*8qwaV}&8s~W5L(?mV-wrjtyBS3Kp2ozO%*vbFU
zR6MlY{}zkrSI3hu^Q<Yxpx=7G{X=+^6IZT2hjuAF4;NBU*qP<RVI>HCPQ@j65uFh+
zJXqSbx34!aKcFI;pPqG43z*e3Bj#Vl`Xrj?uy0WDVX}&s3^?G*E&RulSz51#k#R2>
zw>3~VXy<PQhrCHA!K~lm0ixhD=26z!LUv?#SZ%-ltYC9Q%~h4T4m*f~<Lzm1m5=`a
zIniNwP;%p!q8f$!9lYT|teL&Oc!h&y*F%Wr;Oo1Tk)R78f(3aK?lun@d_K=`Ag0(F
z%F@8pwPF<xPE6^n-}wKFMm#Sw7@qW$TIt*MvDlJ_UxV<%kUdN1+xhF7VLltleYXVe
zR?6Z>f3jDPBE~GuY-bu*@kpmgjHs#0?T~Px>8|nSv)_(;t{p+-7g=}}hAl=#{Vf8-
zY}}$a97%Bus}x>-XFnO?=~!?ioIvC@$M#kjQ=309UIe)CPF*T_zB2u+C`X-?#Y?am
zNkn2nRkyC{XKahqPix-gPWjCcOOznZ3@w^UX&3_mLVEv`LXH3%%lBFFQ|uSY8P(hG
z$UNe9#Q!|MF;&{K3D`}iy04vSo<w~pljD)jf>VvPOnKX-?q$WVAvcLacKG`B4P5>+
z-HLmDKx7;ZBid5_h_0(v{P7)p7>5aTl_F%1hf#S%I|8IW(Q~^v5Z8Z&xEC||@bD{V
z*~GQI<;MMY+S|S#%BP=rwcMLA{`pE_jQYECBu+T~&RSE5(SPv5<T@cMN@|LpQ)ye1
zKY<}-MGl6IsBL8}HVka<kkEFs%|>i;k*-bvN!{haE0KDs_%g7IVOE~I{$pC+F8#_Q
zrMS<lHQ7N9`)9^*jZ*!x{(Xu3<*Hv3{Q+c$CeCMzJFC!OF2eaB_S_cU=);Ro$%k7-
z-h*_iD|1Z#CtU)F)qLovSU$2<iohn^Hlf)`EH)K%ri*F#2s&3{#*W&;bJ;;i0u1=D
zmEXj<r1A#gn$P#?lS*~C{MxUxcFRK_%1-#DTej}zhFf?G8Vv9(Ioyp4yKxb+zvMEz
z`>w-9zVqX&)mt=TrkL(6YkAVjkc@K#!pBhcVA{|=u_xq^vFLC|N3}&WONRJwKy^)|
z6O+jqlE6-26n-Ifq5Nv*&8%?@v?&lUN_QY?Jd0QjF@XXz$(Me~t0iGW_cdI?<th5C
z9t%=JpGM(y2Yb2v9}q^p6x|?&og45#t#!2ndg8Ka67|TJIlz!;(#{2RhwIImaxcyy
zs-|5TT3eZD+(Z?z!CkYojrvSJ%4;>*1C9R>JoLfVu%7iLc=3Kdzjdt!7VuXiz&u_k
zq3~NxVmS7rXtqs;F3X!QQ~UGp&d;0^|7G1=F@8MJ0u6Opb$H)QxI70k#^h-;QR{yn
z&u*?M>6~&B`usoHdg4m=Q^P_*^0hMf$Hx%k8=>Qyw)a6T+SQ|skCu1u(RR0Ugi(&w
z3q-#TS^}51e_J!m$&K|g(42ZfE)HDV$%dOrwWcY(69wTjYw~lJv0=8SerZ8XEyk3e
zA-GGIHQ!$NH`d~X)h*xsdggaA!uZeWSGM7p%K~*hElj52=vUOFx|Y{%gGcK(ABviZ
zZ%C#g#J99_G0@B9@^{rRP}i)M^fvjuzcY>Z<BdZ$Qeo*7jz=>Hpo%eS*#*70&r_Il
z{As9`#QOW2$h>bTZ&0@#q#=#We(%^m1Y7$bJdQq8HViiFCZK?6wfOVoF!yQyKBy`Z
z{x;ItLo{BOsb|IDk++UUG{!yE!)_hxbAvQn+RCs~AoVlT&I|6h8kM?^2SPw>LI}R8
zWee4<wR5z_<tY!~C8*wYovl$A?ByLUYCb0&b8u`eh~qLQk@hx~2_qw))>Jn}GKm<h
zBv|K+IU!z#E(vwAJJNy_32WkmsR7b8I)>zlOD3Cu@S(bZtC#=9_TVv;LlB<n7w@WZ
zasoEK_{>q(mtSv$F0$T*ey~6!N>%A3gmRVX?P<hdfF^q@$AD%_#{-WSpwfwpMOJz0
zHhZ+HfTwF)`N5`YnZc8_%L)!9XsnObT!8vrXoC^e264uSQz{8;azPkb=vYrn+W#3F
zmM3Krl5h^5HYrhoy^%=5+42spG<w+1?4sZemAQLgLgNDiHH%5nM^Y(^+mL@yjS^(c
z9Rjet%?QetmgLV*sPTNtcKiIe^dSs(xbtj%5V4TS;0oUonauar>$Fq2xkgR_dlF9C
z3uS?muC+oX>{)#1$=|m0u>7AhcD(+L<=&-zDr4H|8kg33`BHC3=p8Dl&yVLV*|O4<
z-HZ6~e28HVX%E!EzS|rluAsNMZb}v)b?(IJ?FfTm>UX>X8d6|bOV9{j-nQp{k)3Fv
zdXhQbY-#BTmZ=!*8aShZJ-fpbF0KCx@4L@~!#*d>$gH5U-1bv5J~0ZoiS(a;PV!PB
zhbsZMQP`di$rfQ-$1oo5&6aEM#r0}$HnLG@Y{I8lp8kX9=`E~LOti0ZU|J=gu2Bhd
zc9UurN!FIrK!n8hrSZ8`sRV{wByZC#5iuLPi>b8HMm>mC9ue%Vgb+zjdY>v*z{m?Q
zlql9htzTZ_C!Ss~3AtXUx1CABG)diqUWz~oCz`)iI6>J3cuP9k=SvNSuqTr?l&u=|
z+V@od`CZwhR&{=N>WxR9MwGl<EXtSPyl#si_DzgR$|v$jihvt*+u3`!cfs-ZCwOTR
zDlanI3ccKIqS3GNLpB9w(mZW4#{d*%87ti)^!0rn%VW4Hna69dDOJ%Pad&(n=TYO%
z+$<|dDm;!ua_|!V<b<SxruE^Dun+tbXxxf&u-w@J1#ijm(!mV^jr<A0(If&V{zhzt
zgBddLA0WG&B~r&K-F`Z(fpQ&K;`3i6%pA4G`fNP=&2*xICjGGBHUJ!WaV=ShbHyF+
z*qtM;!$fS!{sFvdHqrPny_k%9iY~R)(DEG=a-ryN6!(4BUz2NvU!#SzdG)~7L~QU4
zWxcUa<sQDY`^zDw^T$&cXDTOAd~3ABFg|0O-OV6)+9!#v@UJGSUo~8fs%BNkSP?B{
z*`yUW6F=5=-K~B19AJr9!q9Nl+Q8xSWTyCzX8)Ucu6_>dV(>JeEd)v``jhMz{#~x(
z30Xbi5|QT9*T_dn9GwS8T5dfw7ieTuN?e;=M2IYw%SsWsv9V?5JO!e&HXfXPu7P2a
z16$Bo3GEdN#He?P#KwM&3DUY|5Zf{;+4B4z^vOb8f3qwA8O}DlS_kZ$WVu(dSa~4c
zXq5PE{H{`I*PUEuO``5aW*bFZB2HD6IQRyA2${+TsFE!|Ug^<;LAeEyRbf$Z;vcBR
zOo{)jBP%KQml-BZS(drW;`&s$VjXhg!|Z-@n`y*39;zpq@3?gMr*maD4{;kzjH*j%
zg1KXGls7H>b#le4i53OnR-NFV&huG0DD6FO`~F+TV4jcg7~&Y_Gf)$q7Et!x^5FhV
zktSMWJVzFn+5ol~TdtblHb$ugNCltUF`Qt`D655GYY{$tK^s;Zh_3w%o-Rtn>g#NI
z$nMMFNcO|ty0yMA<aC+(Lx_Op!<hg6;^~{R0P#=_ao2>`V$bvYAQ<c%%;yvr$ex3m
zRy54Tc__2|EyW|vLKh43r2@n-x=`|7q5#XODeT^H&KD>z;jiL2+Aheo;C{JI(2gw~
zie=P0V%{k>vCm@%ruIb9(f)_{AhI>{&#w7G@HeP%kAQTC+pRRD#8z0e0=G{VcWf3g
z6-T>$h5)hdG2oqfrzlH|y0{fS;`QWHJ-7Z&=yM#?(NFHxZ*Q|jHZvQH{sem@yND+H
z>ES#cr-pF=DI7DJmD(Ehq{y3xPx})<EBjyj(;=q`kicpp?FkTr_s(ptjY;c1Ssn}*
z8kDvcgS?HGxz=w+rCmDxCG&I?Ces9Xd6sQR2892I<-bRu%Z%bx3fQ(N<kW}b_A7Hc
zzEJf!CU-Fw*Ez>}H`MFD2;0_oNLvjCpQzdfMnu{Q1g)(B-X5Z$bZCCV`Fatyg=5I>
zx>o7ByC>Up&Is7&prwVHOM=}Pa;IlK9a33Kyn&p!_i~1{NVebAB3Lx8JO~J1+!`#1
z2@z#Dx93dEiYabE9+rhJ{?upQkZI7_@b*G;>HQEYq~7ylp3GOhMk<}EJr*I+oi4Ip
zQMR1}<vzbF*+x05ZSTQ|QTwZY8@Fj5H4uClGDfjGpQ~M9Qqyv!u#Pf=Udq#QZGUo>
z6gl)<==b=r=vG#rQy$zaK;$Tz_m3=zYd_y{MEfX9uI{?h9Rhfy@C&5wfseSB<HL`U
zC_2^8mea{k?Zt&cjq;zG!8bzJ{)Ov4+3y<>qjF!Ol^Oj<+2zEU0uRGfS33ERV^e8U
z!6rK)KT(%ZXv1jXA%xJwAYJRK=@H$L1Ykl+WSW4UW8gTH##sPWj-(j%DDs}P@l+_Y
zBWH$n`$t|nm%!y{+Lr>y_pxD`b5*eC1!!l>U8Wn~md@=cO-XRR4ABmk3V;|e^C@$h
zV{`WptNC!Ir>HbOS+KW<5P_q(KuSwjoxr9#r!TZh)D^s%XyAF&CGV)DKE4?W9-H_!
zUPBe`LW`&JuYvuTjIvpBLDzi6alkaWm<u-^m;YpXO*>cn{7C`*%zs9bn3;lexviId
zHn!Dw>Fmc%A6)7U6+{PS)@W_%3-+tJ%g?E+tJ7S);NACv55v-<bZb8X-eo)wPoY$I
z#WNuF@DS|uVNcfEBr5s#uabAPOV{6P1VP}fDxT5GfGa}4@Z|l2%oRxoFlJ+K0G-j4
z?|IM)oniI-hum^8!g=aIR}X7>NZMFQ+heN$)-eUoz;?R;nVs|St}Ne{-LRHzi%CzF
z_Ko@RtmYxcImhNThP5Sk5$EfkWfZFbRrA57lL!M%IGE4k%&O4x6Q-@A+GdZV7%bZJ
z<A(vwU4DAUSJ=-Tf>;8A1Cc0p@RdRNo&D{W##~Z#@@^iiYJ&QTK~j4qMU{5YqL$5f
z^n-OPvgH^VIV*FuEpD(_lfXRu-SWodqJ7@GnX(%smLpe~{I1l|YmGHJ`!J@_cKe6|
z=+T~n=wQ~3rN&30VFskm<MC+M;j$3p?G0cT3V&Z_$~QB@YQ}+W)>=CqQJwToGHdzK
zF3FjCaJW$I`_}%qQ;OY;y8yhr5bf|}pZO~7Sc$OeS;=j(;}@&9d6gNAm#Blz-yu3B
z6^8hJq+>`LSTRS8=?cqj)<+}0y7?h`zH6Y`CW5pJ$i#ry{hMGn8n720^(#COIFa?%
zjuPC%U|W7j+F7<l4ma=tAQ74RFg4&k*jg@lJ(91fR|+y^g>HK<1@R3636@H*HwSM}
zY%M7Few|)K^yB_4+HDEW)%zZ!3RBHf13YP9fbfwT`P<E_3{+;m0#W~m;Y@!BlCz3e
z(S;+nD2+t1&uy<kgH!S|BLMz@KKUQqWoO<e=cuwIHEeo2Fl3>#2K(wQlVg1>(#K$K
zd;D9hLJ!5=AHNB5d-f7T4ZF(=Y_Z5YbCTo;v8}^^6L<Won+YmzUm4G4|3_Zo4>JmB
zePL<yU;4*;P4b}JZ>1|Wzyl_GR#Ft2hhB@5YF{V!X>z^=H!^_f=c0OT3z-;YmXyGI
ze@5m$>SNh^F01m{ysUm|!C0_3=>8*PFPClG&aCVheh|Qdjp;0u3v}%<mrPDjX&9{e
zPf1b#=S*3xkEI{|W3G^Ua611yto{+gSf)(oLU`5Bs~z&q)qgjFA4W7vVD24Ey`SCC
zz!GM+-*`s&P?Z676%DefJOnH=-|d5%W;qKAsY$921daC2aYH$pA!a+2g|L32hnTRu
z9*>;vN_IOVo#OUE-CA)`Y%nNf2gDea>!fOzf+z7$K^iaOtrViH_PWtJ#kRildncN&
z#Mj=->2FJjUo?qMxiFadatcD{N0|-t4}UawU^^pl8GiUZUJdk^L2$OAXyIAjh|Sew
zRc%+{*LAqskGT<k`EXl0$-x^BOeFLQr*0(%_nGQ`cq|os!imiAZP{+t`CY0`<)p9~
z(mmb5tloRi6wo?<maY8+#qmxjZlG6%OW%(V&(u}OqmR(vu62_~3#%p@KpVJYpN>vQ
zGxP(QCgIEH2<{_3J+H+qWVzZ4Q*dm4kGij%{mZ5Lef6<3v}p*k-H8s*cq<E0j6hwY
zNt@tY=nx4)n>+$prV>^6C10<-irM~18cjm{_HZIj5d{67?EUPWs1H;;a$oq_bNTB?
zOZa)=QEz27ReAT@A(hS12=XhrR?fN`g{nM&fWzmxo9zo(kIx1j`(}@>AY%kZC#<Zm
zH8Hpc2am*$PKlukr~VvpHCDdStF2rE9Sa(oGTEZD{71kstrp~9tCW>g<rKYfy}1a9
z>D-&6o3{v1rxZECUQ@U#i<HSy{Tq)FFT9%H-dj?|5%uT&?0ljgqr7uzWxGf5_jN$#
zdOWM&K05(2m^WEobO&2qDMSrB*B>08&s|PRtFU~y3A4aQwhF28;FdfQLfGpFKkNhY
zCHDZ8s}p0^2wuYraMjsuu13V&+=wPVx`O)%mjTH@)f<wKlWA0iZH}LQg#WvDQ!i1D
z{VIC#7UttwXH%fw!cXe8;Y=tdC!qf8De}U-<Gk}5;;~`G4`z7t_%@z446;@3*MZ4g
z*vq?~=N^l8^;&U?O-D1OPyBAreMVp>>%&z|E<>!kDi0m}=j);1bzM^rY~#cJTAxU8
zWx?w7!a;{W{oQp+B|Gb?1H61^y<d&yb~W6vG=llzt?OTYst~@v@{Bh~>;{tX6#oL3
z(G2^KW{=C;3e8jz0v~d(>r8i9{U|Sj$<g(cy1KP{7ttSk-Z>p}tEzu!<PXOw+inlb
zO)R)<_a9kB4r^_HUFY}edQ?W6&ldI~7d{CV!iPPUfb=ymq}{ueM7a#7ds)Pgq$hh>
zb-^sMc#1FCR$m+fay`QRCxA_4C=GG0w>|}u-Y(mI|6okPtS1iH{E^8ZoL@RiruP1t
zAqRp(eWhushu6u8J{Ed&!Me*YB!?7sQClXaB|j+9yo8S=?NU9!lC5%u;Nlb;cNwgo
z8(iegNYihpw5SW1Ro7X}Mwz5)qkit#^~}^TsiH=gedD)N!R2oofEc;*)X@_T6yK<(
zx?{kEx>;hgx*Fx#`h)Rp7pH`?s}y08jkH}oWmbf!f6yc+{fbMN+M)o+WC^MdsELi}
zK(W`=<r`y0B)$IaKu+rBOBiF6EcN#GDmE0)IhvRC|6bz4$L?Dr%a(lyGltK?Ck=fa
zwx6*-463b}Y^nJUXnE?wD`;rZ+4<p|p?x+I5Ijleau5K^|2fM_`zov{a`=<+<;o_^
zut(}<`FJMf!B<VO@aZk)+%f&e=<}7j^XJu8(iVdeJB_@!C36I~VWmw6y9B}iR3Ai5
zHO90Y7rU67l8u-}gNcLfO0?Ioho*?~b9uHC0>#_I*bboj_bp3X)qrrKN_*q!F9!J1
zpYx069g>N#*_ti}h&OtEgIB>U+OvE??z1IzL+DhnD-p9=>*e-2*5F>OctC%qU9{_&
zSW2)U?ph<<0r8Pk0b&<V8;0(Qep8q^Q}zcx)EnL!xh^G#A4#Zd=Mwp~dp8n$4&m?@
zANEQDVgc;xy?(EBCB_1Zh!pSWj0~LU8FW&a3wptp@dKU3Q9`1-<E)2h-^aK0UZT~Y
zDdVn?spooYQTe8-&1(&7hPK`+uranF0ukH5SPvhP-DoiK7gdmWxKB<W(((U*5nXZa
z_PL;)LSS-)^_~eaL<^=%D*9|NOfLH!W%M%b0#6XIs@=Ee7sV9S{ign~;_{Oj9o>3R
z|7dR?Xvg4V4C3E*`4??-U^^9^HWGU8uXTfgbOCxr(_A;cxKRUlWm+cW#Q0;cX1{-m
z4CSj;={Pu+*d3eYsOcv)sr!CTXaZBs{1&(~lv3Jfn-DvlJY8AnveC6nbwo$$`*(m$
z;Hb9@!Qeh~kY@a<OW7r`$gtc?Bs40ZLo8n<gAH1~$_w>w7;cxV2FCRZHt#NOpfG1n
zdoKn2&I>;cdc)|~b?J)x1{Xf3Y*~xLsye*Mj(1+DcMUQ6a>6`-_5H`?<(qwe_7HH*
zXj&7&+j(e;UaB~W?v55hbkg~M@&l^>NhXd}>f{u$a}-Ai@r7Kt&~N5{-fNgzr=*r)
z_NRE7`Qw<2cw(vprZ!TBG*+6=TOi8e=Rt{?Qh7`n_vz-;H~vwHsaZ(H8>52UPTTf>
zu?bKR4k1s4s=VCaT)K?T65hDUK~RNnfNczy)#XObKO!7GMMNBSg+;{j=Kva`%GXV^
z9KYjmwl1CRd#DXXI4C7L-UJMby#_OxrmE0<&KD7h=fofbamnXNv}7(vw+xh(hgAF1
zS&Va(W7_gFW>l8p9_N~>@)&m+%HjDpNlJ`*pqAfI*E#HckWVB+hE&ijW|&L}_gnDw
z=OYp=y`9J%_C;4pOrgMq5j%)oxTy&KQrzh0@V|}<Z&g>gY-)LMV?(K<<PgI5J(1#b
zoLQ|ks~QG~@1-~FdV_6R`MUgIS(sJ55y={u4s@#t9c{~S8M5t{Y8Sqi&)Kz8tW)hB
zo*JJRgBV<WvFB^!-Eq?CZaJ+N75$2Y80c;1l-`*)shxcX+R|eF_32NoBu(5CUXUJ*
zfL(tYJ&V{yoPy)!J~u-VnZv#XXH%6L8ASlFdE}Ge*lTsJ)U4?&rn%c)#lX{*kb^U(
zd8ni{0^G%K{`c1BQR^N_znB+)O+Ong{H?4EIk&<diCaiXUbErHEx`~~Ga;FNp>IeP
zzAe5|RUL{K1#(;E`=__KRLgET_7`6IQF?)E%r9eQ>o=}=>-u;-9MNBN%GJ1RlsgWD
z{_s6_tv#;n9N8$#t_v|%^V|*}qS>y+*y+CJeDu{~<BIZzO6zLxqm`7cQPrE1!%WDy
z44gl6dV9NnBYR*J#ogUlV!WZ1$#Ds9HZgQ@@aZ{Q$9{8IT80ehXv(0{ge1AZjU76y
z#a45A$kiyPD=|0}`+bjec`VrvbqA0bSJy?csINd%R#=n=?GmdZReQR6i{LoJuc@P%
zrUb?<YfzCPMUegM{4z6`=8npYvs5f4Mul9_TqR@^b+uXuL5@Ht82YKbMqB=KqYmD&
zvBeGwCioK1zhrM}Krm}oJ~&0i0!K@z(@_xB_?S`D_yL;UjQLQbjx_@QtMRvox<V~-
zOVbhL%vTagt9BlUv2`{n6Qp-thsRp2>>tRX)(hj23ks`jyl^~`T>iq{GoB|XpSf^1
zPgFV4^X}m}0-MUy*Iu}4aoMw(T6TW!2a<l#b8!my{}^H_+>*gchh94u$H3?Nv_tTX
z-DB$7J;jnl759Wn5*;X1rAJh1)H^uIwQvTfO7Lr~7_cjRwA|`lXYq>7*!E^l&+iIW
zt&suuTm5Soo4L_}Tuj?`a(J!ctQ_50G<AG0J?`4gt0s#ZHF<=IsdChffC`hJesfy6
z?fUO?*V`su{aCO4Z1SZ#=STEHN|tJ-Ud=3P>Zp|&2w*Hn^#i;Hs-X=YBa8tmW4P5j
z2Vs-&Y+u{(jP1%&<LKIum~qHbO#&^tbM0aRsm5E-vM_VcPm;sogMk5oWl-VwgW&xU
z-(M5c;tzEnjR?`)T8XQLw%26_*NU&eKh5&vK-`DbCw29KYbI0Rb}s8+#v%KH%N~94
z{;Cq`F1&pFIp9Tk(=9e;+lY3^tkbJrA*UJ$s0zv?tzqxM7b|5ULwL*KY=#xtz<ycp
z+vO47g?{{TzO31QmU7vH%niB!U$?D>bU9awyQa3W4{t-)q!&=ZVmjXzB>f~za>oMw
z5E2B1->LYgH(k7Ob}9&RFTh>k_v(cntRVeIuDT_VW%%&#@){c+v#K_OY12)2`>0y-
zW1YD&uJcof=GW`0I;UK!D|*nG2?TvE@GSQkFGsT*IxD_{DX>3Yg4j$(wdGqwesb_F
z`4Q_ei=mKTUU-A@M9e+EX*{M4F#huSQE3rs@$^DP+w6Vk8+96#HIEmd$W$S*7<U1p
z4ghGz(f=f2iF$}E84Q(?8Y<menhVFVoSJ)VYI1(M0WL|xaeDMd76l2mzG4b$Ay2|W
zjW5k*l})&XlNhOvyUf=fGG9%E{q_$S{;io6W2T9`G%d`zq24|gl81uW`id%Ff>3w3
zf?p1=L_$rpVoH*Dd>A;KjjpSJILYW6z!w3Zsk+XGryr(;{k;S^s_GuC`m5Faay|@s
zBZCAfYh=3j`npiRj$<&APY*q9+H<A$;S`|*$zo=V4`|FG=$&zI=8I$c1yQyta#j5T
zF2B@A*He*|ZtTy`pl_sy^>j>LB|IGYfj7e!-<V!4xS5KaYh>t-bjcntk72e03ivh*
zJLpc3&|V^6PKMIOBrR6w&*7G}UtyOw`Cv3-@W@$HC&r&I$q|~Hl4KkBJT?Qg)$Yy6
zs{2_A63zxU{yk&iocsV3R&MG3V3jcW0c)}^F-wd^_({GaJ+gGgvz6D{xO-*R9QFql
zT`8k*ZNyjKwhLdsn3~miW?&1DJJ-<?zOFJy!nb*>dVsN_X%tG7btY^o*-xg2;dpM!
zlJsQpVo?Ji+FBU=$*UgvP-IHyG!cWmu!oc~<)FLl2=C_(zF5Wn<&hECS^c740z4O!
z&~FK5H6r4#fn`5s-{)p!P~6UlNzDSXB=el}eRWO0J}CT?<Qp;iZD0qbK>H9ZO?j}a
z@Wesr^%J8??j=9XMvUmV6y(rXeyDeu9~YWYG~!@6Yzw)&YR4ULrV-4-t0(+OY(J!h
zL7$N0+3;eTncz(D8rBTeQKR`amhXSz&hl^2qmw-(yn+&A*=G!lUhFlUC$RJV;8voY
z5a*C(K_-y~ty1cMSBe^Q;o?YgPR4*-(kgh`V(^KRvfL2ChgodT!78F2132~l*C5kj
zR8UolVK~@({lp(<Yb0sZr8D$F9ydk%1>LOIo#&Z!$?&s0OE~-Xijg|ynQ(6e@nIbc
zmtWlze<~+P=zMcOWS<M0zk-d~_MyQP9>+}o-?AN62DBp$w&7fknhD13LpLR)SADN&
z$I!koNd%)EQeTuK#f2Y4q&h^>*Zjo48rI77RYyF&!=|rMw|q~`;>tr9M-~%p`j*6z
zQzw0(pE$#(oBuI4?8IH$xS6m6MzeC6;h<rpt;?R(@oEAgq{krgj}$YkwYhg6PC#1P
zBGd;U0Jx7RI3~SRXtPx8j#hLtORJHCY>yB7u&)lO)+K0|uCZJtL{BSf@xQPA?+l4f
z)h&oaFqdgjm3ww^NU3q%tA`U=R1YncMUkY4Cc10eKGEOTk0hV3da$g#QVH+tJkuC@
zYjI3m+X|2H(R?;GWj<^mpti^zQmfdJPW7DtRyw?YoL#MdtBMMdv}udbPz-E+f@7wn
zhV}YS6iz5OlH1r4=@#}bb3flkiJm&}rHx#8E}Cx#1@l;R%i~W<DE@DEksqUapWp-J
z7PJ5xb`=Y7K-`t<5<04vjYZc%s9GHLkFV}!fEPT*HEVxp<#RAeBbCdzqa9p38wwte
z?G<v-^)1EGskO`Sfz!qzq`1x@YVz030LFsRlfmI`-Wg39e`~Qo9N_c!tlbTG!w|$&
zTrRsvj<-n|L3v}%5JZV(BTZX94qt2PhG0Cw><Jo?)j@6(P|L7hw6aVC9j;WFPDTm3
zh*aZF#+#C}^ao-#9|i8)oX=udz3CE#Z#$9Tc&o134t4{FexAx;K{5Herj~Rv{s*+j
zRK&xiYcP>7Udg_X)zHz%wiR!V8gEk=_l>YzzRBUXP{;IcdEoOMKvv6r^_h?b>Dd+F
zfzWjJtG~1d{uVshz9Bn7s~|Vats)jasho}b8}ucNR~|Ba#LT)(GYXS#N5B+B4PS6d
zs()^;4?+CY<PCI?$A^_-8&}&V3<2SpFkOYt6jg*nEK!vOt{0<{lPq*ETME`l&1Le5
zYUx~uNe_xbBq76px?X1kxl&)M->?tFnF@b`XfVu&6q^do9j6Pn^M`mmhbQXG<Ql5(
zW0nmowq?;-51mvwzz0=ch^lbs#q&(KU5xrM_H3ajZtRWL1lo!EZ6dP|AhXK9Jo;HH
z3YmI{%364%SLyZ_-Qe!zE_DRU)Uw_axLmO&081Tx`_rU0eX8r0#Q4GH(T!VSsy$y{
zeOa53jc&=G04=rtp@)qSMP+nS2IxdDvfTdMd^2^_4nnvs(}HCG=Ev*Wu(zg8?^bmq
zfR!BB_~?~hh2n-B&it<su=A2L@Tsgru!uXyVawy!d;d0DCKhU#0xD`aGD}q?CQ{&5
zHM*vAZG~GrpW!ZhR`P@H?mEz*_INhcCsoumxd0V^+BMEaqkr0zlU!G~PK1}M2QedK
zL3Cy<^9o|1biJ-~_&h2ugHCK~^n_9V^X^^vqC`mS8UwFYO}?EYFtJ-TkqXU!k(wLk
zWUP{|*>c%NYTE7)C#!>%tpV-GUCGqZkTFynnEF}^jX)%h)JGI~dBFBZPavoZY8hGL
zKSCMX-5Up8V9apSV6LVA7s-!;f6CFFh|(Sei_3l9e&C;ILBG$<vIrAHrY2&Zf<S;7
zOI(<`6eL*)4ZO<WFV~Tfl|?OnkxEMBX*78>eM@JL%w8*+fzK$AFj+jA{Jf4@t$U$Y
z>3zkIM0wm8X#w5>a`>~nGHG8({EVfoUMoAp;c}b-cxR)P->r2?(O?~1ue|CnHkBwQ
z=Yed#{;%3=HNGGn{>9ISG7Z<SIkko!TG}S9ISF~j14z-SS8?|hR=KY<&{^2*7>mnp
z^;aLzD_mtHf!LVG$8*6;_{$B%|5lGd7PYHfd=LSXus67pXP1jxeOVAO$4wW=Sy_9M
zgP=3C?z%D&-1KO(=Bqq}dto1Z;Z)sQ?DuT~SAU`j%v}02f~SJCLBC?*9kQy4(0bJc
z)uX4W8YD&@Ab6GS4Q7>=E-Yve&j%8p2v)-$pZE22flvEtNaeE%sbRLnFOj<`Un&%P
z{NsA}o=#t3T&>S_=P<hVq|r2&RQ`yXIvTpg|BzbGqc3G7J@$JIab^3i<7nnuhz+vM
zJ~_AC->bW;RFMl;&x-(i7(_lD$2bVhc{mlvvNhTRJ%(R#!*5xB{g}++)dS$=iKZ?T
z*8UZ+)L%7g6e{5iY@;F`_pj5Q#ZK7=I`Owjy%x%t0DG)gMkt^EUs2x~)l|@K4Iw0<
zNG~GNLQ}ecAOs;u5Rj%+0jZ)Cfgnie9I7-aK{2Az6h%Zjf)pu=7(wYsi4+B-8wfo>
za?g9;?^`#2CTnHRS!ZUQ%sjK7z4wC@X@~!9OJPR^?eTo5n<+~X>F0?92MnPoM%JFG
z<FoqEP>tT}cPD}Ck82VfVKyNU=LqGdv@am0lrY*$^I9B0#Lv6&0O)p=dp2l|m7c%V
zAgO_bTfDvz?;N7Uv;JHQ@4I+674#Q$aEe2Ge>2&n9jWx(xIXwIhIJnnoaAt;<wEUb
zS^}7^`O<fqy9}3~NCerO-u)Bg(lBg)EI&(;ykOyaD+BkNUz5@o2EL@V!NA>Q!!!3P
z+DWjcjtVHp@m42t*gZ=T;S|Wjjpl%9I`86%CS-bn{x7$ef@c$|aYGIA{-NKvw_Ugg
z3a%*O6l_Wo{$5q}$(FW6p>7iXfD<&w)LNlE(79z@VmFa<14Or8fnd%lbh4uM;IbOL
zq!xi0k&6&`dnd5ZR&(u?51yI3f8`=^KaEo45U~^^X{G<q23RZj-g2KIX5p0Tb>g0M
z$NMDTG23*%4bicaXzL2qgWrnAHl4O#Jr*ZY(PCcsfKacG;K|LLzdqwBtScYeDg4Va
ze&xM$Fob+}^cvf$xF(^r2POjqD8==+Xc1rVI0Ocyk57DRxh4qUeeIo76k3`C8?)Wl
z)!&l%%W$~$fT5Ii+mO^_5H7mmU1Om0;fv4S?Qun|0cQ-F@*XO3V0g+o<P?`8*5}iK
zqzC9X;Wyy!v^yK@ZPl}^rc&%Y&nWyd7oq9`^{}o`ug-TW9*-QZOa|_h<8gN8v&H^9
zHHR$zK2aMO)MVcjd?3KEgOg;y;ELuP?7Pgr9N2RMvd4QrU#KziUQx^T$&Lu}WN={V
zlg}fPHb2)3jMGv(3RZoF`K6WuS|f4QtC4+Nu1ig5o03)a1lgykuExk1blZ;3S{fJL
zB-Bx-;w{fwng|}D4XmvXGGdF-8QcFxNmCL^jJ@wy+3(9bO0IXG(zn}T4bMl;e-<p}
zw<>;p-LvCV^l7tW9-rMVQG9^C580@pE}A(DJF5@qd=F@c&M(bz?mJ_aSW8?#=UWlN
zK-lzlx<bpdEMnY#7J`^<Wc7Vo+HzZbrr^4}7f2o0_?5#1tj-miL?ZL-jOE177L0T}
zjTpK=eDIAi=z9Ts(;<?t3P`I%F{ENzbg!4_uN~v1Ik3;D&7kkjT*S&v%Llx1)8>TO
zUG*D~!j%CI8A|M;4YW2VIf26~GHl5W3E|&F@+-U~rGAbfMU3H+h43dFx3e<>Ow+iD
zf+~;$gKj|Nf?-$XhF}?Ufs^;_+$r}CmaxN!fjzT<TqE{-{^u@1L%X2UDcp6*dj`Xy
z7#QZJ(XT*QdaTaL2$n$RvRIl7WRQTD37fdAo$Y!)2;|wF5v4#bvlOLgK+c~g#O_EO
zvdu4@xw<M}D}GzOgr$?U?8m<fX-X{SDaY%gbQ=IfBuWxw!<t!E{hOKZE-n|(wHPy`
z15|H&Tsq%Iu~~ZuEicCy9S=`cXIhB(&Tz2`^XA{p%TLy&bKH}R8_-bSZi{|x+(ttm
zB4621!#i6|lz6oXTHqv?9e(~*@_?82b5%9->Ib*A>pAL${KGE>7klNGPhGoIGU!5l
zRsdM&W)GfAuS1okdJd!*b__i`9~<?oHYi1^IsPeH?@7of2UYodh10##V1PH~=A~#f
zs@dBarG`I|bHwrHJ$Z4rv$`;oSU;AD7(pmJLWK^z1<N>fm+>$<`+XZvD(AreQ`ec*
zcXbC<^`GMX#KC!Ysu}-y52JnET3)yQ5`f&^n~~RwU`tiavzHU+wSQsez)d`MWQhO~
zH2i3;Mi*cw!`Urz<gscjujgj+PUrWH>t9G019nC7-Fq8tc|0{0pLU0^EMd!SQfd@d
zjumRx`T~(~B);6LFCNTLUAeRUSa7o696H^<{?=V@a9zMN$tQZ;nL`QGR=2qWp5pMJ
zsm9R2$0@P8%^w?EW>TcC+d4J8{&=aOwDF_CeY3wJu&xKVC&mI@rnn15VrwVGje@_j
z*ZO|--<#%>Adgva%<ou;s?G%j$+wD`NKx;wdA1@xYAxGV^VB7*WIlTi?-#c|KWHAD
zkoyGHRfmg93&Xox;##4#Xk1<qC}=?2jVrOWhPU_;vwxi*w=Pp=v^;*Dx`nhcYPyqf
zH0_K6^<Y$$xH=lKpl+;h^U-2pRJ@lWB%Bc*Jt!+RPFm!62OF9#;jf7lNxk}djfJ;9
zgSV)6M|Uv6;%-0_E7z0Hpy!7^c8|k}le@+%F{R@<-+yFU6!Ei|j#VHg0HVP~(HEV_
zsD5k3)ZcIN$V-+@L?@#*M1Mw08f5N4>~`=j%$B7c4_xq&Hj=g6zP6=95y&?fm4Vz{
zv=-Svu*fJ%e(osQnly>tEtr%rT^zZ&rx7mk_r1tsyod>RDsCznF(2aM(XYf;*09~)
zwacAuhGHULAo~|)1IF*=2QfunS%@yy0Bc#)r@J@K?3SH{OuR>aZs@=i*)QD9!EWT8
zTnkG@d~fFR!O=+mR<}%hNhLkHd6?Sho|Jp2w`Xeol;02=`qxeU*c^JNCW36FCb@m^
zah{VJ!%fjwW>a&(5GULR`2*Ym<9Y(P-*5`VJtdm1QK_bYz%R@E-ADfTTZO7$QC^Yp
zhP4%1tz3u8VVno^$bsC;iFPR?awJd&>W8`xRk(}ySp~iLnousNla=B!C^ZWy(zTi3
z$V{$0+_tOli&C%rc#&`IS4zQfxMUpU$OVAiKR2*qGDQw_+c2n45;<#IARH1@VO#_5
z<5QJDqCniT2MF4)2~6xRM@41eIGT-<tS@){{AL*_YO!7P!+*vyu(x5*P&{~(hg26G
zukrR6lVFLvJR;qIb%s(3`b7G?suE_!ntkcxyPu?St@DYq?p{9h$=wFT_9ecF2JFT!
znay1i2RB9AFjK_6y{q9mhywd*D?Jwx;sL%jG$>iEu_L)t5Gm>;C#l-ClM5_bal29;
z;Np$u1Vb;2HNyDMWW;x80f!>3rH<;kld5ke93^tlk*v2!*Ah+9c1bI1YY$A(tL!@<
z+=LvMu!pJwmiIV+m8JNbJ9|W4!N@Fu0F!L@DnniB$cg<2S-5Shl5v42#QI1819o_Y
z5g!Wi2*l~^^Po7b=4)P-$aa7ISBX?zZ|z&<sM*rCG!yN6_oS#aiO35EQxV|-YL}hT
zzOak1feE(Ya|Zu5QyXva5N|G4KEKZU{ONepCFox+(KR0lN^BeNYIQUY$@?xI#q*=h
zn)}F%DLnhvP0K1Cq1{$a?k9jTE_EH?jT5r{PaJuMinLm)I)KPaE&m?yg6(VMB&S}H
zO}c$pj)7gwSJ5vwaqo2DV$9b`y6N@nhKK!R0iWdBEMJRAuAmmqdB>zIzB_jaT4-=Z
zc;2E14ocxs?RzB<txhDVj}>L*+c9Z=ew^llzkS&Hzoky*W1InqmmN;!<rS%T!#wBO
z=*F=SHcRyu5Z44bA^F6Ho1?P5MRhZlyS)W*QlmdRQwt1A?zF7;#45BHJZwmTSE${Q
z?{G@tTuV9)53>(?Uc@|)5Q)WbjQ44U54dmn+dxVZ2<SA;iwPK2sYUivX$%!_*@t-z
zwf7_Ia2MYtL;)Szt;c7<L^kNhv<8B#^{cq706s9hbWeM2+!(xKDDd0?Im;`jG#*12
z^l*(s?E40Yu~XS?zNFYxKS*1xzLe%vU6)3h%_4}y0o%=ztz7g_ZK7M~C;qt8?08r~
zH_v@%FR5BL>BrTbW8vwj+5B#(fNAQNe;wkGy1ss3-G{MX@lQm$a1_q)t%yznhSFTZ
zF8UxxbmP&Z`VM*9AgGXQP=kE2p|<xv5+~AoRyyH<^2oKLN0Vz0;!J11%kj%#oFS)Z
z(i;t{H}|%{6GBgEb3i|dKO8(aSVz}x!IZ&(b!R}?(esnqn;3(&YS<G2aY)`3q)=bs
zsWiY@Uis3b{|S+j>nE7{l1?><M7BfyKn|$>&%lnGb}#E>(a%Y6rh-gF^!=Y986q#D
z&*uo1b~8=7ne*cU%_}-tO$O?0JSVkck?>Y15=YOs4zAx4S7yA^cx548o~Xhy&x5iy
zfkv36D>#lFj0?>!W`-r~*yT7+DN$nU-vPG<IFx!TeFY|Feco9q{*Lrb3O<BcrQ9Q$
zy$R_ooA~F=D=vgg{1MXfcu`$!&9}-F)UQWD)7uX`y<*clbjD24nQ5^&d5~yu0F!TH
zUaFCSgd8D7S|6I95^ODG__%JF&hdg@@zVMeW0-T(7g%#GVhO8=Ymt3GjC~OI7${PE
zp3)f1U|0n2YWiqCn>rLF#Qj7G=g0v9Wa!f`7q09*EdJx@NfL4?<%c^cf>ujc9YJMG
zy;bK6zgv@_rHlr#(T=G?t}J!qB9Bhs=uVt<-6)ehZTD(TRWs@dRod#e=L&~*g$txK
z*htSRlx^NS6kg!H2g-n_yojN|vgm95G&G@cI01fK0biPyutIPBD~|s$B3(XvTtv2m
z+F`S%D{u419qb@gLcnE1s}$);04wcH&&5uS&p~`9tQk*OJ<hvmZag2mSnrjyc`>v_
z^y+*KsHbAo%RwxqUVOM%zsM~5+T{%C8gx6~)Gdjx7SJ~T$&)9Z8wtGn{=tRG(|DKl
zp(BQvO!>eXd3bjoOw?Upn!E4ZFI9^+2;b&eYg7a+iUs08p3p$*)kM$HIrvoBBL=5T
z-I$4x;zbtjz8qOT6!?n8Dz6Xrz9#2!Ek_IlMZamUMI2}p%!Aq-6P7=>k#1hgPUx)|
zt47rQG@a}WIf!gG@X*91Jyw+d$Zrgh!q@Wi3iNeW+UW=qe=XIJICnGxqwZLry?iio
z!r=vSHcPC}Aj<s%1Fh?>k^OiVoH%`2xAXmpcC-vfy9CDth{LEPhWPW8MCYWC`=x?L
z+}U}R4C4tI<SbcaFaZ=_w#D8}{=#5VbRzSEe&gwHpkm4jNFXL4-hr;s*{4%q1WI;x
zeNd}SINC4v_cf*9)%L-Nuw33_>}C4BNYDcN9!jwI#vWd<Yw@x!?&IIVvJ-=pYpUx>
zkewZ5LP1EJs2sAllBV<WlHMcJj>dDo5bX3Dk=i$xqdRg=hLA#%y_ykC7eM54O8~<a
zcY1MIY$FS>(31$zFo<hZ3HeFtck70f`*xR3TG~)m_J1nJkxe`a-__l0e?T8*Vgx)t
zg4TjgNET1x$n@*}d*c}owXIwrwADO#n5<Wv=sPfPHYhRF@i=dR(Y50pN09yP6WQ|T
zUx{($xg{wXoj=`sE#@eQ>4ixzd;tHrbo_TS3mpvzTrZZtkg{CCHT#{1F*4wqs|KFk
zC3IOf@Ts#$l_odQ*TH=vzqv0>RHrk&xiMYp!a0n1P$-MeppAZJrF!xGo3EJhN4X8o
z4*@CJj$k>6kfwZOWaSIvdK7mWI=>uY*lXY2n12?2^VU*=Qti6-SY2Aeg<&~v)pPbH
zZb_BE(9#e7C1zx;I9tppbg~<)7=4a=W2I6XO~o2f`*0%aE$gC02d=o_$QJi8C0B^{
zm$p|C7&p2f8B86SQYPd#GX?IGH~E83G>R38`L+M{oMe3Mb_1d@EP`jlDnYB?l4BjI
z1&Be4$eD;HCbP5Q5`m?I!(N_ho%!ADGyt>K5`<eRDqd3u^=hRz_m(b%zIhm}x~vPz
zkJWTF-<3{D8$9k*I^p-)*{v_1Slz;rj+X5=6`%f4&IFJ}jv$dOVD{!8wd4Jil{fNH
zOVEvE?Xfu`IVcL+KCVcwpZT?E(;KYwZE#NZIh$-)Hp&PjhH+e5<8|XnAtt^7j@B=G
z^uBdE@~C<%DGzUxwMcQy{~%5R5oVH!(9fN3D#;ta>OHAu(QCP3l;TA@nd_s-tLq^>
z_b+sp6%;s0r!R~!PGm0L`FjM=graGsbSjOsSJig3Jfj<ZJa(`#N0@1)AMkFC9+LGw
zM(;eESzX+-Cbu8ch6}%v#M%K0dFJT%=Tc3;vh8sBuxjHOHU?HW(;l(+FDhc^v;9WU
zOLBYk@$S~woxhyQ^q9Rv`$F3O>df(irf!~I^zmO#`WzNmP1e+l3Ex_7kEU)6e=S^g
z)%{4_@7HUK*;+N<-wr<c_qY)Y98r|9Gr;k|VqqKo-)d=0)Xs7h03^pyH+SeqdzQ-m
z%VS@+2-x<c<sXNddk06o<zn>xoif_*Va-M0_^28P+golC`$s;ajegC-y2dOV>QQHo
zcc(~twB>Yq1#K^#M&2p61Y!;f$@G&k?!S|g3MYE$t<>Msc`=6s;D}ncvEv2oMQ*P4
z|Ebb5bp;;B90LwCYTeOlf6rK9?qQwY@y^^cOY$^re-XShmfosI|C}63);(JIu}Y`O
z)9EqWTLxjrzg_jB!egiu_n0vH=J2!rLi+kTr)$iRp3Bmr9ys#L*yB-4a+~*N^!@?7
z3OMM*(vC@*^v#9P_UJ9@I$?(Pck63o)l55;6ZqKa+U;>fA(35<cRpLX(td|3x3nK^
ziqSu<+Pk_e&nYWY#B`5p8EJ)g24A%w{H_pK4F#g<$H}3tGh~2z&@lb+_-LU;Q_M9w
zVrSuVY1LYLt4iU$VZEce-8|Y3WedeU7gdqHcYwV>e?SlYrd=_#G1GNpMYD{0@FPV_
zIEHtK!-{LF?$VK7yWem>wk>MVk$PlFw`cfo)$7LoU$7%VX1CMTyMG7++c0yW)AzFq
z`+ai4$ArPD_s?$oIR4?YY8+S{ce5Y6%VXdNC!H0!N!CryZ=SRG^Dw(j%r?12Xwdwi
zDF0Tq%QbY*l`&tE8d%eFj^}F1rv{IEcIT=&ICi$INAE|=zd#IzhotrtjSQ5zTu7)=
z!1kE;u|>I%FZ5*@#tBoc{yYc>b^1Dt3zie6oj6nGug!as{hUSQrTEj2xZM|IZVgul
zPp3Q?*y{?(f1HNW622w*?Tli|^M}(>Pq<g4`^mfIE1x8l25(o<?=-EfzHughQICR@
z$y1EpZ+w<t5&M55X5ED$7?}qsa<8Ks1@~YwjKGxdoC?=vGTR}?<-JTPlzBoR-?)-9
zD~saDt6Stmb@Q&tTD*+;)^2Y8_ynyZUfDC+i{m9eZNbS|x@tDT|5yA5izLg1sgtW3
zY?SNMCBZpRo4z-1?^cV$5uRs(8+^_D&6|PbX97Bo2ov=t@p~=ePF;3aPi+hM^2^`@
z&jC?BQ57mQ>Iwf`TDb;Vt^0O8%;s|Oi+mM-BT22Z8KSxCefA$ayjl52W>v20{@(tM
z+}DnB8jz)DzyD&uHuJ*Z!VtDIb2bL#2W*LE90yZdT`q1LnOCgdeicXj<Q*p<pUr$x
zdG1Q&)$y=!;cMwAv&zuvOPnXJX_Brm-}nmL39E675uL{}+vdE(+LyP~7P)R!c6}~i
zue?T4=#BLLCG=o#S{o{<*1dSP<x036C6DM8U!(UD7n?S$kQE(g@`t_au7?5Tkt<ye
zYrK6+;g&<sZRx_a<<Y^tGMd6W%M5xyaW2$>a>J}Ht6N|E@U>o>!2t&Y1D(#qaMhI1
zM1ipr$`D}r^}z#}D>L&pdVrDAc%BZ%wm_R^ZJpE>>*X)m9!u^{OkMu{Pt9aMd75O8
z6itj}H3-{J;GSAWWo-}hnsoj!Ev=KJ7QJ&=TzE%RQy<dkSsObrI9_<$Q^%F$lh^<o
zJ76)0`8?B}P|?T{ZF5pJN5wr~%;VnJ1Lp7!i$tMUCvG{eqPSV87wi1XrKTMZeiZ|$
zelC3S_m#QL6Mp1AgKk~hq@JYP*VokU35u?Q<>NWd<1Ersjw0!$1?M$z9WnyG!bz33
z$x%&?F(uy<-Q{;r?zf(a%aG7<n1Q<R+rEcrK*moe3aVO39(dRa^fFf`#`pL6NE?<Y
zZDFrO2>S`w9d<_iG-WU~jjgORI9LO0bv7Z_S$~(S&i^rZ|Aa~SCS~c%G+QX5RpD*t
zGN!|VHAC1Jj!oh~tYyTa_>SK+UfNO0+N7qaIu<hTq~E{TUnhoio9POP;d5c(-Kz0}
z8oYB(DHxNB7%JG{dVS01oj~hp)-xXzPIt`4{V@%_7vCK?Gc@%_@B8DjiO$d7!1<bo
z+Ni?K<dtrm#gk!Dws&Tg&CGt`@6OTPhMByv+b%iZM0Lu-#<e7TbbAf2DeAov#!I;y
zDQKww@zLdF4812@btt*yTKDzujjhlZGc{i~CANaj%P$E(Uim&~Wv2Xq4*UP_0V70_
zfsX;kfM8%?kYLcz&pMmKNN|C`+gMo)%9xum%najZ1Q{l{C8WZF5XuC<1nEmBv@pT-
zAz$zW+JD-@B7!~?Y#Cn?62X?~b}uu5_Jdh~Fb##vLGZ%_R%Y-j%uj*^GuW%>9{~@x
zOm8U61oRHGA^`#ed(ngmN-*$OO(fwi*fJr`zz9WhFabg@49+Fh+P#ByDQ93X(u6W_
zgU|VYpGqYdLnt=LQ#qKx#R3;5oPxtoK~R$f4LBSHu~{Vq!{NG+a~p&@I9%oQ|E{;-
zwYH253@88Z9^f}k3V{xXV<4eh1VaQ|`oHTG0gHh1Lx63<eFXd>L~W1o1p(J(Dmj1=
i^p9Zt1Ys69*MI*s1Qiyr<BKDL5epm%>-o>zGW-uvw8Fgr
diff --git a/fun_gg_boxplot.docx b/fun_gg_boxplot.docx
index 4dbf5c6a4f6b3a2f1669a16ad2927746e5bc8d7a..407d7d505dc8b0b8a5472652e77933be5ebc2849 100644
GIT binary patch
delta 88364
zcmV(}K+wOjw+6zu2CzX13fzP5bazhy0L|~SO9{FJf1DIJvN*^QUObWW2PonC5YsRg
z4{#l^5_k&;3BRd*`jKB#`$4^fy7msKD6f>n(V9ut5fZ}|MrK|)9Lk*J*4PHKqzi*(
zmDdlRhHO=DS)G2qI6-&9816LODY+B2uT4I~R|Q2$-t35NYd}D;Tr-(R-=9xkB=e}<
z>Wtdge}+@Px4{7{c|<13UJ6$_xn#%cHD9RFxyBGfo{wE;Wp*VDD7HXhjorWs#sVAt
z=Lb~=z4N>4wgJXYV5pCy6EJ>r;~1^MBQufiF9&zM%S}*>I^EtHDH%}YFGo<QL#G1@
z`EhgtMZ_5FN_3trC3=JN``eva=|)N<A4e-df5;4^t^tGr1Og(-a`=+`RfCCYVPcGd
zr9@Ud4;cmonmZZ_1fN^RcFHiQVj0_$<<W7D>|5YA#cSnV)iEr4d)FOY3^sK7WvhF4
zjqJDO37!wm`}q!f>IwMdYKn7hU?3P)T;Ub&5)=`<WF#nJ*XP8(j6TvF-YcVzs;-aB
ze?CUV-3!6Pu5U7o3?MRq;E-}w3%E(1H-G>@#IF4y_c)D|_<EoyAN>WXBWCl!L5dd}
zcDhwQAbj^N2?2N6?RCqzqj=S>6eW!KFgisD8aJ@Lg~O~rSjbIcm4Qaq+~9MIJ;fol
zd>RTO;J#POwDd9P0CbFb9CP0Q!~kR`lO7Nde}mIXX71I5{Hqz0&oE|188d0!VN5;)
zjSMtYhEwm$HWJSV{n4ne8}Z8*LHb;!m2h_qg%Cx0w}_d`v!3Vcv3y3xL1DEcfy3bS
zd(js`j2RoIMamb-`l_g=M;6AeO`+HHGh2<BESHy-1z$PKC1y|3I@zbzB(OWz-O=TX
ze~bwJI_uh->C*0OE>!b1JzJ3P&$~37yidNDD|PX?kcr(Gjey$oQy>g@P8vK`sx}0S
zE)0qciarxXs>@6?<ICq8p9VzL0MW5%F%wX}&UfAm3(eeD3f?PPGNw_sfdoQg8oJPi
z;^^HM)0^aT0ijuM66ydROil4upuA?He;{P>BK(rASd+qCHbN_!TGQ#=<FFYgNqd;G
z7oFZkyR#wY_v`)q<tNGP>JyY012G+M7ay1JefLJ9H(x!A*Xw?Fa8b&1eXz94|Ix3G
zkO{^YsJH>3bcFE@4IU5=ODfdHJahv<MUnKqyzjMdOS9RT6>Dhk?0fZ7$(}V4f8H{i
z>qq1qc<$Jm(|A@4<jc<a=<=p4K>mW1gRSNo$EzRp4Za5Q0qE5PzTU98YW%<Ff}=Vc
zU+RmxM}ODDF!+``T+e=8pz|v-di(bhkbKa>p6rI?-<sqejpFw#a9GHkpXrochm1bp
zU3+xKt3vDauHPBleyiBPmms^qe+xUH=}XJ2GSJu|WBvqx(UL4zk)~gNk3pp;5&l{&
zA~-zW3o8wU6!|NGy1o38j9RX|B{tz9No0I9XSO?WZ5q2gh?*o~Y)EEtJdcKFXX+vx
ziu1_Ah*><Go~Z<sC^`Kao)KBI*!N0+BxBRRAd-F^y?^)i-H#uP1UW6bf4}F;h}hz7
zs2hl{P2P=&8xV)awEeFpvFj`GAD9Z-_%~U)o=3)vgv{rU%n^ST;)juW$ZZxy;^wBf
zB%TpUAW{tvs|y(<F7x;@KLPr9Ty^eRHy3?_tDU$~AE%VK`kVN~EFx3@BoGdi&Hecz
z7NGIm2L=d>XN=%%R;=GsfBs3mK_-j91_eg=guE}Z&~7{Z>%qltxRmdU?CE<~=nQON
zzw9*sAb7pl+z8o(g)Fey*_3hrlOlRq%nhS<zArPpA|E-K(MKi}2xt?JfmqnsU)x|2
z?P5ucMoiae((X0cArtn2pxK1lv2aIh?k)Tvu>=$$QPLAC*iCvVe=C&#Fryjdd^~f(
z2EqFh)c`h!&>qu;&w|(@_u|Rv(;^nbZPSo1=0c?u0i-o*pP%oM8s+=KR|4>?TH!8X
zK0X&9IZ)&QL1;uKZWzUcc?=KKU5i;wIF47go8%$jOE5y?8I9$^4)GV_7V1SD>{#V6
z3j`tP#Lq=;yg>K4e_;_&XYxlYB!JTIrzys+(3vpC!fE-aAB93#Phd<jY|}vSGK}&C
zjRDQeJi3zlgn{G`vIgZTS<YBM79eTJ+?f!*h=nh57(_cgjQt?WUv5BDerfoMe1Hn@
z1W*yF3%ppK8!!t>!<z`Y_1`kX9`{?N@<ZJBySM#sm3^N3e-IMqYTftuLYB3X?{WAG
ziQjdnGf(_KPEM?tCnW4952iifbk!S8vaa|CD`TSzKm!96!VK_zsrmm#9gnRbY=NRi
zoJm|;@hR*FE}Ytl_)eC%;OzGWi!kPX)*f`*?aTHcji(Dq^m-Av@oPSL4ZkBFjs^%k
z4<1M7%yNT?e<kFTWzWRJT>Ig5##5G0I>WcL$S>kygpYq?cAUmmsBN=(%$z0}FCfwq
zfX59X6cCWY&RIml#HllDD(oYuFB}f#H8DJ_E|0)V)>~GmNosCj&sbDzh`;)Ntx3ce
zU9f)y-b&>nH;#zqU0-y2*E=&F^Jh*(N|uiyo|Cw7e|g=$yLb_y_Ei>hs|<pOF+U<P
zn*0tA7>SpzE&R@a{3>t;;ctgMiX%e&V{np^u}}dIg4zDYmrU@<Az`l|UB2&f6>zcQ
z#aQ@Phq6CPOWQ-)AEo{Hdp7C??1T+kAvR06#}a1xB5I=XjSK|)MAQu;;;$lMq`Q1^
zM0&#Ye|dX!w>GbHG)@j?wMp*sDr%GOoj?DL;=Dx+aNX|pu1DH1fZPsMt|r))MQa;f
zwy#H}w2(3K&cV!s+^8m4jRz#J<UL);1O!qMgg2yhSQ44^mcw@K&W}?<Y`-F-_dovf
z_q%r=O@50B`@E9rEyF5p9KnjnB&5Es5t=<he>;<z1g@Lf58{(T+Vw65H@A}ibXer?
z3mU|3Ok?-4Vtd1n5ITxv8d!x01};ve=(JjIs>p0T0(LAg91lsn<cjYiU_9It<>HkL
zT}!l}tBw6}Nk)s@F6JJjgfl0Rose{~&dZ|^?Prd$ii}l6RzqUXXh?0?QjOvem9E@P
ze?@~Siy*8#SiPoCt`Gz2ct%$T0)?S#%e{k0;P74UK$=P%33TEH1`i#~)!^&IP=?}0
z?d7hKD?p<Fy)#t^9x-JB3nAyEh@mw!@~W2|gn9<SkR;{cW|g2JmR%zBHY>G^c&fa&
zjx0SSj|>yW!06bJOQ~qsEts-4%GzZ}e;*QoStGO2w6tC-RuL=I!`~q<PjEkT1yRFo
z%TvmJi@cp=QU)_buTK^s-dQ1Ok2wJ0hHa)htwFS>IhqaJEeep_Lbs8*_;HyCn)n%Q
zB>5%|47s%=H!6FXh<ljwohDh%Tzdxc+!sub$=-(oKZ@ifq;3UGH(`D=DhRa2f3Zar
z>fW`wz1uQUNEN_Vibg}>;}ho3<5lWXDTm+MQ*84R)f}WQL^o@~yQ#GNt^EGR7gu~3
z8Sn!5`gX+~MHFsd_WIY~s=ehGQMkAyMJPS7R_Zt|7PzWd2ld1!bXL;wRgTMw)RVim
z#{Fpxm>rcix%HhH-~J?gdkaN^e}xJ4L*l_?Y8BU$6<gr`iBo9+a2*faG9^jK5MKd=
z1c9v}|4ddElRErEG9?tC(=bNHOhW|nIhYSV10`c~@z?Zhr-9zb)(ew2$v}=&sfN%M
zVf`9g0Zq(188bPoR+bTv58MYQHhrxkk#~97z3vXmNYIKIwJK#E$cwn(f8b^s%9MaI
zL2&|edHW*BnvWf3@P-&t;>=-@9lCSSjYc(|q-unUu3YdNOT4zW<^VnrJ1NZi(Nsj0
znxpp!D9na<4x5+21nzXsv2xY$`3vxQ5$Nx_P(`c{=ry%5uGV%v3fI)4cO(-@nTrmG
zD!_zrw7B7s?O^7^{CnQ%e|B2yQgXwp8e5{TGEO0a$1M(5X|@j)@Zfc@@bVqGbf+0I
zD;BpX=FJBki^!)~w}g}u#)WFmsDP-PSGkEk{1T~@KXZ&!y69r3I#Z?n5(20QubqCc
zb5Tlbzus=-vNw3*%2;GnC_oF0UtMVCdB-LrmXL6T8&_4KkW53we=X5ucZ!D@t8f3f
zr*d;0gJj$>laYkYIJ(ghC#fU!;@7Ks2NcITAf^If(nrfrO4Wre_jZSn39P#lYxYh~
z)R3mQK-s%W)dGmca@(aF`)7IzU6t9A+ch*ow=zQaD^J9#g85YMW@U7eUcW7^RRj15
zabvQikxcb2&g#@me~4!)mc+tjfx@~xyQ1ll8fu{7rZc)D{hVNaNIs7(Il1{MO1Q+f
zTobMiTOZ@}j%co@QWdi08&s+qjcVMO7us_Qgw%lQGe{T1ddH-5Cvk%)X4KhP>V_Gl
z9v3Q9{7J{qYbsoLkaafZi<o#Um_o8Grl3K7Z4BAn>2D?0f3=#@17mwuFl6P^boyyd
zJr>8T-cZevL38=^spylJH<!Tm>E1ny%EYK#Nmf}#O{kBkP_adI4l1Oxy8>ELvna)4
z6TBvU!tBYKJy{KfS(-qRON^?JyaJtEDUziVQ<|yPk*0{uW*bB34h<C9(qTULR*(U&
zD4R8MB-x=je;?9p5+lW7!tU|ifXOWUBRIH(%v#A(Px0<h<z=dHzu(IrVdr!Py^FzU
zL#^!}p_#wjjL<AUC%(ciww|!_1IL1b+x;z@<V9-G1(-)0cNT|SOIuCh9^hZyvc_fF
zDK<mcB+#F|_&;B#32T#l{#ry-xmgp9r-FMsaPPLAe+9;gdPETtx|iKvyR}=SX!DVG
zmZ58kqbrO28oDtfwHgpM;0dRP?DvIR&_JZ@t1?wm<xz`eu95;#HxUlQka&Cw1+^Br
z2me78{{8^E4(1feQ54yrK#Hlv-@uytY{Ni!E`&&B{csT{-Q4zD=a*&tLKX8pu!cLW
zbBhGof56BYd(?htKtz1Wz}19tph!Bth&^~GXd5W+6@Mdgbep=jh^(3ghM?tU$YJ5t
zBKlail*vrp%X4tv7t)3`g&nRbovtQn#0NPGtw$juVe_h<%2lA3Box-oi_2+%7Ow2~
zwMqpq%+E^%w(9G5Kl<sz=g4e%dP6uQtsF8^e?akmVg9l~%!?vs0}NxKeQ7*f<KeJH
z*KT)51GKq}@um+Xieef|*NJDvBRNW&9JMD{E|Z2TrI>Sly-9r%a-Xz78n+w8ExRPY
zmT#F?d_oX<5H?fYXXPX>p<L2v_mNU*aM!;d>`dFu%c~RFQPu_2#YskrsJblXSBNAY
zf3cqAkQ76u<<H%V+z1hANxrKYEwo9iK&L7();bGV1|yR$kJX{!Rb5X;l5I94Hm4ZF
zQ!eFK2`B4K&%!84dFiRP!*<oc!M<0v(c*ZO>tDRN-V!bjQgb0L5{nf@thx4m)(~~r
z#0?JRDd?sNkeU#w!Nr!JP`j6LXcW9=f9{;}-*lx$J>Y9d?RP-S9w-C%SyvK}x=)|>
z^R+G{*sCeauHuCPS-=#(NcI^<`j9b;mrN#<z_1iqkmz9OJR)uKI-_*J97<y7cry%Z
z`v;^I#6mee`OvUjm`P7SWtCW@^6%9iX}tl#_PKZop_h3mXLiBU#(WVtG{hN9e>8-D
zkS94WZV|Y88YVGmH%X^Sx=q4jTiBdwslHh7*DpiJTbm|mR4T#uiWEj#su)%&-!h6~
zDAfp<=<?!J4(Zx*tW}g`e>IMFF_99!qYQu~!*NjS7rOo%yQf-mN|EQREf<09q?Y7)
z99ZaZA&g$cyh3wnW1Q2${&MHNe^%$Vj1?uH8zsmrCQq<x1P(N&8N?Fge(!;3f_^02
z8DP+uR5Ip}Ay$P*Rj9WsB?dLqTLod3oEkeW-135<mi<ACZ7pnuv3->oh9<(MC>;+E
zn{>SL$e*IjyN%I#7|WHr;V@^?4T`8(-e{|M!@XmT^sH()+TYPEH3qt5f19i`J-Etb
zot2d`heD{jYA#FYDWB1xc!EG#dxq?RSiqGN0EBAG^(hA+nuUwtp<0?t*cs5=H<SZx
zJvV@s64QW((7Uj>8;p_sM)DiUe?ao@OfAkW=FB+bC+m!(yp7yHj6+`bHRZi@K0#Gk
zNVHp+_qi!{%WKT(oaMJie`h51eFRLo<;;y94pfWlMY27sb_3$)=N=jjQ*8~Slu5v2
zdzR^)^A7&K?s@YAc^j+tIgT66J-UJoitH6lbpS@Bg#u7>9^j}RI-iI9$@PT|BinvL
zEb_AdAL6oL;yNs_WsmAmyk4@w;WNjjf!s5}r>?JXBf_4j9gAnRf35xndBUFNA^GQm
z3O7V3j`spGjMU%$^L>Myk?8j@t_O`KS-SCz$bMhz!DlFsBUyLhQ%6<?!G;I6p2+_)
zw3v|_JK&E?(co~n2-{Y8bl!yFC2H6v-X-`7#}*0a&z%PDM0{D{CaA?Z?qtG3SpE?u
z#D?Ea!lI+zd`68me^(yaShT$&4FnR0G!V-MPF{&C7SP-zJru*Wc#O~K?+ex>AQ_n*
zyN^N|Hmo8_k1pHSqcwZWPo)m(LVog*ixvC)h;^2b`j4m#K)FRt*tTEuD0xF#Mrxm)
zl>`CZjFB{Q9E|Zw`A`J%&Taqxu9Q(GXpE?-c-|*-&`hO;e`>V6&B%pu=nK>Nbu^>%
zs`c-4z)dLB1)X-Qqt7G<bSl7SnDNm4L%gVPPz3#^JP?1W6zTFlo$}(=qOW;qvWjMe
zRX#2>4uzJ4nvzmx)hiRmm@CG~lqsh&mT5f*+6+mE;Sv>hNy3~KxsyH5Y5}hi$bnDJ
zG@8%>tro?Oe>kpY%t4DZ0Ghgnu-_L#{DG5X_4mbR^yH|50l!xmaZ;bhr%$=hll`^?
zddEQv(q)j!Q69w*R*NA#By^<7w>7+)=hY+HV3yXyIMr>wt!Pn33djN#6_u8GP+(VO
z1T9%-5MdNu6#%BP6&|J~9AMf~rKrB6*vN?N%149Jf1c*Ubh`aSH6OQUv_gV9KSe5B
zcu740G*MTB`jY<Ya7ejM>rQ*@iTZOP>LUnybJ!j8`0`-=@?96U)5}PEMk9FI0?a@#
z+0Y|a;(c@Mp|GLzVZrYr8O$8omy`@|Cu(|^(B|0UUbVM3I5;^R)FSG$nw(yNP{7#+
zGx{hce<PetMJS{pq_f)H%+}L0anru(wYmd?8@+I|&l&s&gkzR2v<B5uYwlhb#i{g(
z;yh+$T9dKfZZt`~(}4I-y<6N%LNiLPO&&$i>d;b_d2GxbY6`Q~W^*XEr*X38IgcWD
z?0Rmz0{Tk1kI1gAbNTTxbMbch9>xT0==D|ke{*;_+$eO3!-a(jS!Ys>aL6`OS_6hI
zfn5y~wP|1rS%Gm`G}LJzZ=h-7nX=vsTT7JJ3b)A3z8BY0He@EAS4qG_Ym7?K$zgMt
zs>&tDV-z4)1%{cU>%(r89p$?**PhPjo-4<e6@MdT%jxos@-5ooJHf^YtbXAK1P2j=
zf4p)huy`RmrGVY~a7BEub0A}BKhssfgV%rqP04)KEP(Ddwf7P_C9jMQsmRcCJ)Dbv
zcW~cH51QENj*nA5Eba4fQfv$f$B7%%ORb?Dv+Rq6!>HZv-CgfSi!L8zfBKr1%gKs_
z`eYGecEM*0_|h)=DrCZ@E|{{eCDE|He?OAyxm)NbRDt6X?d8yzwo`kqXw~PPZl|?b
z@mF4qTIA!*h4L(9MX_4B(v1na)k1JO!krS2O~G;p2N`0q6(VdCY%A(GF7B_X<bcd_
zu~^Nj7GbMHzTFE<c2C!{vvae`92Qg8iwGU_h3xoP^g%W(<+9^>N#&e(1z{nxf5r>f
zi-m+r_MYQqVM$RH0FaF>h>x(1g9DwP%B+X7TcnD}cF3@=d6G(?{7ukO8@bDy*zUGR
z?XEW1)|F~|bDxCqu#Sl-loruTrN8JxFJQQ&>Drq0Na0M()=j4?;|W?UkwQuR+@o%Q
zQJU%yZ=Y9Fpe@Tzt5gkCo=S%tf3UJs>bXh$7x&#!?>0SXqMAECP6^4K31C_n47Ygb
zurL$M%%q8K#4{laBG;-JZ%w1=@HDpG4el0NLwWR+{HyI@{d$yjesY@kz(hvPR{9n8
zyvoZF$L)FPhRxI$+vN21R&T<vG-OFO^g1s_xjWRh#a)?NT5KyZX0nqie?BnRN_A^T
z`I*|q)Eg9N7wxhFi=;)n)fu%nL-OTKvI2lh-mqYks90!}cNrOdziZbrBLQMJ_87+U
z4r_JIvVMwVky=?OSI@UaG?ON1ix`2T2ZOzNqil`O=d^N%Hh{sZMXE<E28N+qu)kWz
z42XQ56eLE$PHe*>zlLtie@LwcCf_<X9<tvTt}q0@iAC6f@o^@LYPUtGK8`M6lR=2s
zbynF<uC6UBGb3>!1)l?2eyJ{50+a~FDdYYpagiD*H(3O-SQG3MN&<L9$nb}P80SY>
z$8@RG<YQj_I(q-^?Ykd~1&?M}7=i%`tyXZqG+Z2L7m$|%TCVu_e+3PMRZX#au2QtL
zIR*=av@ebjRe_3f$2TH%Ilg=)Csr1%`m)>Uw))yE5DkJV(eVnDL^=gXMZ6kh%xntk
z1^VQ=0dz6ZlqWH9Zc&&pR$2&NH7g|urL96;G@2=k#6Jb<+{gzbm_`w|UCbiEzB*Lc
zKoM4$iU){`FgXs&fAv(4b^+NHaebJXOI6{{{phC;(Bd8h5QJ|-8aUkdR?2p&y6jEL
znxM&0qDVpfRtiI4^k#aPSiiY31nJWBI*j;WladB`U~Hb|oK?$)%9~3W3nRD0v*ULb
zqzFt7R>qH%Nq6d3Tlhb?Qy{`;H*oGvEoM(~JNe|t69K;Me@?IS<+RDSTig*oswK<<
zD;N)<ZrVi@!qki@g97H<cKCW=hqxoP41+NoSx+>M(k5#w%X59%+0r6+NeMKjs$m(X
zko_<&T_^?~q`?^RXw{GH2t|KF^n!wM>N60L1!Y$!@F4&svI*ZBF~VpNAc~=wX(<F~
z1ey4d$*Q6%e+sw;P#X>&vhhhk=_=;3??U8L&ttOm1g9&({3EekXeFD3b3rJAA&XY%
zbJeif;4e_DC2uGwP0=C94OY-dRC*<-)N+mIqd5z~%Q-^cl<yPl%LV&v;i*}i<bW8|
zYqx7xN82?nNxr9T5oLGgx|Mjd`5^D<xp}FQ*7DOIe_-mBG=!4bc!v}k>@^+MXT9q|
z?|NM}kx^awGg5WQL--J*y1u{a+QfYcUo>JBo2Rqg5Wa+8UHSCYl8e$X)8Ol4ygOpD
zL|EC4zy}ujB@KdN8&c(}Vvr;)norMysVR+9aZ{3}0!b%el%jvL?kbmFCp!bhuAg}j
zw5AIee=@f~D28wWE_6Pr<@1z=ae^wNNs8iya?JhK9Cg6uKFI=W)ohG<8fn`e!>tb)
zufC2oXc)U{CZd{dd?2_1Ed3=jbq%sk2U!L4xQdj?Sk_nV(JAt5)l^I3u<NT=_LVJ2
zWAqOea?DnhQx_B!(*nz^?kUciZ{PL?*PFZef90jCgl&t!lx;OuKmiznZMZd6DSB6q
zb_bGcbR&<)lIa!-h7b^j9#tT1!6=MF?vc96H&4>d8I2bqGrHTjdpQlQ25fD$ft6df
zs;tss6BqXrR7s8Ov7oZM5){_gvsPqHgUkiN?bs$+&S=bF(62b;s)9@|mKrru1(JQE
ze|7m)2Ip!USv6Bk10HIw3%m+mVUba?o;2GfBG@OL+AvW7OU0v@Q3sYmGpjC=e46c-
zNYWK8XwdI<uS;1Gx;LX%KRWD9vZV*w?&!9nge~VGw^;;*c;gimIenT=Kh3Gf;+WNq
z_BpYb#FORG@h~T1lS=&%I-&{q;I1|3e~TwshN23J*9{DJ75St3XctL|&Ry%~LZ>8*
zqzFD=0Ke{D&?ae<yp9@wh;0y#;HlfY?%-l@G*9I(N1jrL3{RESqAOyB`}4th|GbO>
z;>n_5rd3K24pXBOeb2;DhWti?t3{TELrUOnu3ndg%I(Z3TkC=5xE#<b=gv*1f8c(0
z_E$2OU4{Le<jo^7BPovg0+MoUkJ}G0FOY;>7Hkhm3olq_3^vK7Je1_5HA(wI9$C!A
z`oU{i286}7CEj={3q`U&b<}UX-ujW(`+mh3V5en(tv<s5Q+2>c8DLlgvb%w(LDeas
z%Jqk&li%6d>amICK`B}gT}5_Ee{axhU)+~*#TNMo2B3Tjc{hcN=ycAnyQ2~=;u*1i
zwU#_CTfJ6yRK^8fsll9}1)<BmKwIHJovyHqTi4o9>OeObxfo4IwR?rB!l@9|K=M_c
zZmF0zbkmA%PkIt8H|UN}s!k=@xUNF1m79iEb?R+s?K9&Won&}cI81yxe<aMsI`&$?
zYfVxUhdMP<PkpBir#dr55VyB+k|6w-OhOQS2^+y<WbUT0e|<IIN%``*Ya*L^%xQq<
z55L`ReVrd~w~qLAagqA@qq-WFkFv)#Fw`zq9q<@>gGBDwV`-sDrsBDXR%#=@gtWqG
zjFqC)y4}_kJ=-R67DzBkf5wYxCe$uB5@&HdkA`PwQ*qZ~Y}wpDgY-<{8b|C*U3un4
zk>LHai;MOVdx+PaZu=IyHW(#UStS*__5-#HqVsvkpIjd*)Yh$m92P;3<RscxBpWU-
z`!}~YWt0sb9kasec?#*e&@BnJ9_Zp(_Mo!rIORj*L4aXlkX17he^n}?mE@`D3bbHt
z<0f14kae?&*>DKyYG|fr`g0e?kTnCMzWeYhHcY^~a3|8Af~;QYO2yD}4#>FHRR_g<
zc{Nl)+OD3qtFjjaxu8sbyjrJ)>?N!8jn|%PhQ>uitwL__8pC?RS6>6Gi1Vs~u7nxa
z{e~*$C<~uhPw*x6e=_W8&LbK?^Hj_uOrmquB9m+&t0$%8hG(5wQ{l0+TmRF>|8`J|
ziNAE4Y5%ou#-WipR=Dgp#I9e8n>wwQuFdH-hA;1?oz@U74pf&Hod*KdRUyxIXD&U+
zgJZ|!*j^MEZs87g6)06`))y__*tjnlL{uX?s*XnJqUPEQe-C@yz{HV6kz1Bk49D`y
zly<A?RQsQfp-p*dWA*i=4w&nL^}l{;a;G!6>}+(@t^5|KE|@KjA{G$3gcydM+lbjb
zaMl~R+ncuX3L37IH#%nVlCdD;pyt|yrZ$@#y%7sgVz8am;W%!R5D4PNn9}o0#Mav?
zA&VDbAe>z2e*{Iq#)UnjKf%?1%D?HOj724RClLKmhcbi4#MzB^7cY9RQpJO<lV6>W
zO)5jlvF`@9w{VyR{mC*zd2hMEQD#mN%#DHTNb854K;Lps43bU=Nd-wUd~UI)IHZ;@
zM4TI>&64IJhrxzWK3PwsX$$3DyOAx06ClI;iM&4|fBqth!PTOUBjp>SRQ5qidpygg
zolUrCyO{3qz}9xVRnD4!e^K<_{r>sgO(}B<i#L-?`?+Ix?zz~EXW#ql8dKWj%fyeT
z<EV}d^{>vq$!p15|7!ocLH^H<n;O7=m;}P-z+5}3RE^|60UKg0ELrYv$a&XcJ~Xn}
zfwl1hf8$58IUU?;z+=pYjM)&yG`3)5(J)ouK66~*L@CdVj+0)pTa$k*awqFVaF3ua
zTpV>zG~>&l(IhniN8yN3IdVctu7jyDz3%nRWf}Po9K&iUr;tq{a}a$ABs5S936AqR
z1dUB6JR>DCQo@3SfsF53$jEy2A-$DD_=;9^e{<6lSk;HUfmqcNSXxpi=|Glk=9^Qp
zx@NXr{ZK@Yez$Yp>+6H9Osp-HcMxv_>&9q7qXi`@_Dl-Fu7cuxp@eF8JKdXGT{PnE
z&oB}1rfYS^o<bp*M)28WyYF{z`!6zDI5(91{$9wAR<hQ^s~5lPPG_F@f1I3HG0)@p
ze@j_WUN{b>)Z+Xqw;yOk_bH3+jIR<GNohDw1AA|-IxMn#-5T6WqH1Bk2o9dJ&}KnQ
zH@g|CO_{<0>T0>V5IQAul}Px8^wDXHNIJnuh1ArDc<5Ualm36~B5z@$yROn3LsJk&
zJS@gl`|zCrP7Y?oc>Zz}<CzBJMvRxye*)h`cnRT*2tJu^@v>2T`N^_-QCtcf(;41$
zhBx+n8PWTUp{@ynGMg0ttqz$cVI`5fKbY2frJ=<z?S4oXG50BQinqcJ7%`Wr(P{Q+
znmWCmZ!OGt!KciDRA<wwPY23>_f~yIPgmYygK1JlTP8)apfbB92R9h52j}Pae_d@B
zrD#`gH)MWyI$*2bX;f=#=wSMKX(=Zwzl9Sk<YB$5DN(K(C_vJ-&(o03XVnoIS3(-f
zAy%Z0+iyl^p4eVh?8cr@jr#_dA>QRh1{ubhl1Q))*;IH{g)FO#5w)G9U|8WaSwRhd
zgi3=<r0;|Hr#kfmdB<Togm_I1e;+~+V`F5s+G1>yF|+9+g8f!;h<ea$Wktk8ScrYa
zxq4)iuqT_%q1IbGOufO<b-;_xJhO=NG|P!j8bE1kKkfz(M5-WeC1lE++3rNhPn!zr
z*CjksRAe9q2h%12Il>!ym+Ac*O+g{T{CT`eENdh`kjWNoEm+qgAA!qNe-kOe^^=81
zme8{o8&H7tg=SX->XbbRv&>tqcLtH@O>w<=)FZ?3*ee;}ls^|izCy53CS-I}<qi3)
zO!(}&d#svgp4>W>Vnvy935J;+3tZK?bb~|8=2VwJ+_ka3UDbN#o&uLx<iF8IY^(et
zlz;Ndz!3t@VUMnz+y@P?f9IYHon1+%De&!0(qZuMFkD0KXflgg{zyyxh|ruN=l;f3
zMe=em>RxtB*-5WiDIF*zm}}4E)(;ng%%f7sA$4U{phbQSU19!U7M6gO8wgqGK7vKG
z1x-|`(Ly{CG-)&4IKb!oG=PLeL@uBqWWO)4(@!22Ns8OLDQyO$e;a5<wvn%ilKc_6
z>Le^C#dlLcdd_66s^S}PoRUz|4k#KieVVQWh3o(b!=!G++X-5GkQY>?HvU&ts|Hk*
z1BMp1Sv~ou=RU9oQt^KD(+5RHN^FltkudgAR|@eXg_OKmb_T<YHpf|ebA<`hrPWot
zzm{w|qN`;OWO+kse}7e)WjI8#eNr^7H)^RiL7P1#c9zIS+MkwDJ`G(b+h;ALOCcMA
z^%&7HwI8M-Uj*3PO4tt&g2z_7c$Pn@M8;jQq(s=fc_XDvg(L;3!<Qq`DvH@UAqdY0
z@ezAez3V~mdQT<E4lc@HF1bh^GF(&wE)tS=#|{;R8bUPEe>tj5pNA|p`zMUX3jr5W
zcX%u;ab!B2jO4?g6M?t5iOIa6rq|UU$Cc}SYj|}!=NATE-_<1F6s9F)483zfuZEbu
zgK<sC(AeuSiiKK}d4aXR)?`A@<~0*Ha3csEz(vCDk{Koq*)b>?Mwqsa6jZetbdMab
zNG&zqY8JVBf1>^zysp|`6S*-Q+MI{Zr-|pz>kX1QE{(nMQrLM$KAe{Nc`KtnTM9^6
zV=;v$rEqqQEJR98U6dKdmzZnWnVK*~b2|Mb>^!(xAP)0fvHw*AVxX?lOe#-yNka%r
z833LZ0RJWgoQyH#V*`i^n=Vk?MQ&gxeR2~SC^wnpf3_3ARfshkmIU`<ATwm_$&F;|
zT`0#2sJAwul7|?$eGRyk5?RWG7)k*LWA%hk;^W^Kz{Evo6HooV;4#DfNB|&!%0nMj
zSl$6)VaP);p5e*>_9$hXJYwmis4&eVo}NkMUSv~nc!k_cFbxgsRS}1}PAYQ@<I4mI
z$iY!ie_iK1tlL__1yhfY#c38ix*-pwu*Ek?m3}IX|Ln#8`I^*dE%C^gJbVCt$leSF
zh7AAwC1Vmyr<muI1pOL+fsI|*8;c&uOgQNvz?0$=xpg0@{3-I0i_hU<eLYk?V03Wr
z^NS6<QNJ@9l(Jdnl*RSDu~berQ1sLFGYcabf3YgiWBrK0zY89>Ix&>+#T$P)tXWJr
zv`U7I9Cwllh`yKePf%uKrtw_Z*gg~DJ<2!sGvGMQFpNZuC3WH+;K=<Iso)bW&KL1~
z5fe`sx497}YT(X3Sr6Z+-|5`;Z+4?fmk*a3M#V}^vsDRu3O726oBmdTeyLqvw9xDG
ze?j+7^Xf$EE$IqpSh-WlxmzKT)CWk0Rw$c&^TNiOzm>em<wVPL3M?6iULeI&_^N;*
zhcBS~Jxx}s4wOI;%vm_$q0b6)`KmxhecS$gfKhge6O~$ifA{g1_wO)e4>)?vJif%V
z3EaGMSWI2-h<MXQZ*be`?xt&UFk|@Te=j$NU#U=%+IX8VFX6jRV>;HhtJ+nO`CR^w
zzb8>dnf){(k~k0ooQ8Z{u|_Ee+?qn&-tP`BO4-n67x45ZQ`X=-!c{3PX?8C!8=05k
znskpHj7&M8aSC0V@B+bvm{E`2|9BATZpmn%pmf$x#ggwUS&k-TSS8DGBPjT@f6B9#
z!pXINNE7f0?BJ=`-*!+yT~57B_!;lWjMi4E7ji3e9`Z*QHrin)UFhW0$u<BqF>8d1
zH4oh&hfVl*21OF6P;{pDtl1N@F_Ys5vc_#Qxr6vDis%%20t@^FSow=T0s0swY{H&=
z6stapA+w>R4s1hJuZhqX{|3v^e<B|;Tk%^zP#2JnK%{!Ee8zaCsp2l^-FlVFBiKrL
zfrJETyC58kxsVB%Yy){*>n_cy6v6Put_*y`{*~81o;$YY6zZ+Fk^?u2tLZr6-Xa#N
zgRbjH(=t>cLmnjWLHZxsHVrFDM=#2Rc@M7;Lg}d^K_g5e&)FR3_I;4pf9~Rg3=&@g
ziSO_J`ODk)cemu{(arxI{do7mh@#U1fKp+H$$m5FU$#aYWN)9h&u`WN^$gkbm!Blr
z)yF9(*+bX4gK6uB34h&d_qHm<_Y*#UITBtSGK4RK>A`&163>x6vjBMg`@8e=_H`Lw
z--VtTkRCh%OJ?61N=}60e|S|kQQtIGJuzh*373DPRsZ*)y0O8(cQ7b5%<q4Bck}V>
zKi+*H_wWDl_rh(V^0uJ*__sS^gBz>d3?tZ1ORz!CojqgrL*3}KQzSk2=bze+A^yn{
zzhj7hlEj}*KmE3dVnh5BCH`DEG;z!f{ZEnpxKH{icI}7A0N@k>e*i`O!tNK}+em<u
z1ORMM8_J&|<*_%VA^j<m9;4W$AiW7=zaZ&lrnO=C)1o?sX2Y^nbD}0}Y~)EBM4TET
zAXzv+2+|EEJkv&F6C%U!j}g<E@Fg+wCx^yVc8_xdH+HE<;E3!nsLCo3G$C~H=7H=_
zXi#!uD2ZYkOOzy-e}E&<c+%cmr;BuL^;<XRC0x|LuJt~TU|1XEb&DH8GD1Zpre=N`
zFI-P}Qphs@`d8=Q|1N=LZTR3t`5<IIe`Jmo@}(7whiU|t_*W#mM2=y2l)8vG3!J6)
zZdAk}7q{mlm6BPq<(|C_Ci~vJ<fPie-42UY^{E?>6@dale>tzy=H9{&B2vdeWx_nd
zsdLMOl7`Pun9rkO%FxK?1|tR|h!I)4oVAHy+7Ad72PggT&~XDHUP_Qg+PUcV4KiNN
zr{b}Y8U_WYjRK##fh9!3FpC0+#)xAE4yOkP=D0K%68Wc*3?~eQXfd~h=NLV(a7#GE
zd=6eEoD;U@e+oO~jG@#D42zWo{U+cLdy0kO>BcUL%3ygK{JbQ7++ea7KVu%p+&2(1
z5GsUFMg&>RnX)1|MBoB}B<b8d3H%tmoUq8*jhDRKk&rQnIZ?#ytECi1FqptOEM!)N
zWF587x0)Q3mo<bf@AF`94|6@^<Upkyf7f~7tsv$pe_U4!&u%6T82&!REny3~!+QDF
zB^&(Qy>1QeCCy5<hSio1gC&t*;g2`_Z)wlbDFdfe%W<%t1cdH>TjRbpVj?3mjZO4Q
zIOGwtg$ZFnd6t#-+TIpUcolRC#5%4uOXnXDjF0AQ%}lL;2d_0|_kH6JaPaEN^7u4I
zILDn#e;D>HorJ=#ss+4e<b{zJoAN^3yI?|C$Uk?;2@6#uVNW)jV`{h&7e-vXn7A+*
zDc?QhGN%FaWS+_+4P6|u1LeRGbncTo0dNj4%V#&qLyh7IDBZhOw|84c0iZ@OYRdmt
zL-KBfP2piW3lJ2>nA<kmgwu6aa)&JD(~!cLe|JtEyH?j~8NNTUPz-$n69BmQ={0?F
z<>-r*{43OM7(f_BK>h@CA|y&SCq`K0mxw_sGz@FyFqeYQ#rw$M-bz`JQYWZejZXMZ
zI$>+rM|R3b5l#L-dtbwxHj=gbS9-|I7QVx1_%OEPsklmj9D82}=VF}MtvIhsMs2K(
ze~?&7*zw%m|9-n$5+IO-z!G$eGc^?(A+Wm7b3Xfg93D51-q(e}t6z~8H{L@~0>%!V
z6LM4u2=%RMd5;NBeP#&eX1Y(RH#>-!iBp5##%#0@D$_9|?AOFT<1K}V2Oed^nIuVt
zVs)R02h@+Bj5XE89*uxH98m}T8YrC8e^au=gP8#oKb+p56?#pLBp1^e_ems$P#OnX
zcVV*m++<!#P{$f!SEpeT)ikYFs)|*I-8fQBeLOL|a(LEUf!xkM#w4wG)6!(zNDX~b
zKRl_`Q)}q(?VzDK0$I|eI;y)DS%*I^Dd6bY<xDdUw_WIw^vN^bSKFq^%l73Re~&~9
zrZ%0P(qARjwrOuoL8*AQ<Q;JMjN%%b^HzWI5^Bvxvm+3nvo1lZrm>V27N}t};}e}t
ze~p4H{EW`?^lH4V2c%)zL*Gx)2@FU4LmiD5rv-^a+Oq>u`6>pKDV?<>jGi<a$5nBb
zETJ}eIyObtwGtgqe_?_UI1ro#e<D?7ee!U-!ZQ^Gk%0&gM+hP%J3Zo&BzdIaLWo05
zikYe8f!Op)A950thMI*PcH+fc!NQX>eLyhpBRgosz0D&#<sb5=&sUqC)HYA*%|;c-
z9!Yk8bGRgXM`_)D{J5BAvKxLcT*GZ{2_Kht<zQEGFxJYo1M~;79-JEbf1Ba{@f+*2
zEIr!l2$%c%2%`^ZP_|9W9+hj=g9@7dihf1)gJi#<b#mC@GP&ohP5W!s;+|bdk`hAM
zcw8QoOCL)WRJte;;$U^~M6MHRC6TGoI%>DY$-zp-oWGbbDY;$Ynd1^ubLhy<xjJwy
z+ac6inu#C|GS8?inMxMwe}M<W<!35OzSd4_)o;F>>B{6SDr{<G%}oCMfnLe4C3<Ix
zyzqM9Qt?v)A;Sspl_)(KVWfefI*w%78u^nSyy)g1iJV!h*5z7@M?%r|5e6P&Q<G>3
zO3~DUPaB_XmSItX5w05A7@LfubORoB!w8Z%{Yx*p=2HDpm74w)fA;n43T0P%g}BR1
z$IO4u&`m5k`d{9i!Dm%t{G^7iu$CR3zzhDO-^=GawHApia9oXwg88??52ab|bV;QL
zLmhml_p__M_yFO-jqr$<58ir|YV&=q!OMDR_KZ0Uu3_`HkF7-_O>rg_icn$EK4(#;
zlknlS^BW1F)JrY)fB9O*YHY631|qF?k!Y^2QA^R+8^IOnK{sjP_y^Ri`j`KuyDr0U
zRjLQt!q-a;^hXku<z}-kQp$s%<cwYoB~#d1pjRH4KnmLxbJqhaAcZB`ZdQ*Dd9+01
zw}U!VWN~$IIX8N3uf$j!&8X&e5ct3v1+bSvH;B1A7WzcYf4!rd17o6L=`U<sb0#VI
z7HcfgzOt$`QthEmjn1l6r2ryQgLR0Ed3)BEYWioUmjd%O_%rMcADxGDH`3CmB)U<m
zogW<^HF+FJb;edfbWQScwEz4<Ly=NFSt)Q!rF~Ee#I00A+#Y)QjBO$iRO-R{W`RpM
zI$*~d<v&+<e+~MgL?sJ*A<f@&vVuhq<>Pj9F|g=g<dAQD!`6ASUi>{tu{dFnT`(=^
z6Du6r5K~|LNLCgCYX+oJ5yM1Cpz>*w{oG4`dhJvMg;PCe_a#v_G7p(wcg$a6iWPDA
z@X0kU;!z_Oy*}`Sq*AeLn=@Y&_!2_ZsQmD47}`WbfAXr3X0pQ2AEdF$)WTA&>+h*^
zHx`T+$lVZ_pCNZ+^~s2MB03>=Luf+P3rZ-N5~@WBMa%IEK&VDR33W*c_4hELlE?(v
zP`d--M%hsO+Io-;wJ2mnRoiD&h=3O6MCQe2q0!o-uoX3@sSgU3ys%czns~}PIcKdx
zbG;@Qe{$DA)Nf1V2zKO<Ve*<_l-e{eS&&6T)|!asW16hB24t-XEqEYn4YJnw%UVm`
zcfs;LAO&yb^9#*a6vF$If+raJK?+`OChl^&-b$S*7<l$A$4#`9N+x>O$Z-o_fFw6a
za@#R{_b$ne-{kY%B)R#J4K6~Go51P_lH67wf1!wHVeDy=o3Mn6N^(mgQ)IdkziO(2
zHcB|+cR`T}M|+=DBpCax%x<}$T={Xcgcg3WyH-MrcoH=(p(V>ambqk<`81Z`w%~9o
zc)bujkIG>gsXt<KSg2@OnZY6w8B^p;2qYy_tVTTHxoLQ{RIFI4SfSwzbCCq2Q#QFs
zfBfp~wQ`Y)AxT!#Q3RrGFBeH@{w<c)fCQz@6O;rabzFiH`BAG1lcVH?l&Z()UYwpY
zIa*R5EV9>y!p%Hk0{>9V1tv^Bj$S(CkU<U^7$Lc@W*a$V5vfK=cn~11Y~{8>;#``<
zIlg7|KBSc;FHSU7137Y2@IsE92)qc5f8e;m+-`E@V0xpsTmI8TZVrcFd50b+=77!F
zT?(|PA->3T4s9R*bl&)JDM<q=v*Q#$-*K!uF0PhUM`8&eBh?)|NyJAo%J|1vCAP0-
zxlWQqKRl_`J3KCTBZ<y0@nVfzAj0;Dz|`fz!Pl>~Z{I4YR6Zz$GZrJI<x!{Ff2uCE
zNjs{abY{V@IT0GZorzHY68mP(Ri$^L<NwtvdF*C2*x2-29V=V?UE2aOGFb;jhH4L)
zymfqm%YH9ETJ}ga$j=G&<eXPW3WRf+z8I@EHr+CPuYN%OOXinccO`r3#*2k_-tHjl
z5fKf<M#z)xn~ca_gmqA~Rnj;*f05rW@LB8dsO?o-Eb>|S9mZ$=rM3A?Xr5A9$MQ*2
zIG&m~^jsOcE+K`_S|S)RIdf35pb(JMoR~d8w)Mys7)rm@zU^>r+tpoE_E6O|ueNBx
z1eH(@BBF?9IZ$b(c}fV8@1GEDXId^A+t#zLVGTXJvd>fLmcq1zD9pOre~#xU3?@7O
zwnz*BKxhEi_}>%V#&akbJmJB2vhjd~&;}PgP)Beu^#^_1c2Hg$0wR~Fuf2UMPWWR%
z&^kXnQ50T4ut|}SIFC*seLt{9<NLanY!m`fw)W`J*6c#PJV={8l{PW;Zj_OS!~1_t
z9Jd5e0jR8o%2+jsBa8tpe}I-%XrYep{eV&o-~w=2370YL3jjC)oUFhJi=+Th04S?~
zVp|ha1E6FE6k#yok>42*Q?*oCoBAFfoXCw<J2I^YAR#0qLO56wd$ser(tIzD*kd6$
z-E^^iPwTLkaw0?}OnL@|6OjZh&hLo`oQ3ZdxgRaNtk;lk(j(F7e^J=M|Kk7!qG@G;
zQJ_DD4?UM2UAS{11Diwlsr*O~wX`&GR-v!x>L(moDrjVBtn5D!HKl_C^x*@lCNGp@
zeVM(eJd}^xN@ERCS+5_R>|9iCC?Yv|B1%FRkH!}5Ul^{~N^BPDla`}ZD81fw4_$Uq
zsWQ!OVs#M}RJt=$f2?KOoVVrzqXSAk<^SGZ_LNRyEq?yQ%G&p;Aic)QpE#u3h7vQb
z2g7Z*-M_r<qRSif>GJ00vip&OxPPIb9#XnrE_>JAPfEAHMP<G&=cOr8+Y43jMyr`4
zk~*$6E(dzx&b%`}Ec{2W5iXso4klz{Zab(TS}G$2-2f7qf4W_-oi~~bK=Ci;vO!rk
zb9hHhI^t8Z?K?oOg&r`Xy*cQwv`2!2pRuhPXoMX{9b$*P@4jGclFwz%L*RLGY(D>&
z19C&S3CO)Ya{D{CN0l~EYyoKS(BNALWcw{!bC@ykZ0Q<$n9#btX-PUAb;Ksap@C&g
zMy5ln^fYQvf4EP&GldZxRqU1S2*4wG;0daAO8zV!KkF!XNo!E)%js;J=?nUP@PZD;
z34Yb7AzvEW)?`e7`7C#ptLc^5_kU94F8w(*`SGx}@Vf53{=vRq8e7inwQdHl*&qHP
zz0x1vKeNB_djDu!BU(%!JmIVVllJ$&_nr~pzn8i+e@TLFDke{qTDvjp-LVQM;oAWx
zoZ)0z5<j-}k!ruLVQzKK>y7jCD4K`g*%O-cN?!D$yHCip78)F^rf;iLZMJ7T?P2--
z(P8H(a^Q0K_C2wOID0Zm(v|1)N5}O>J+F4sZ+djqeC723e*Q5>y5kVQmf4im1N0vh
zTzH*-f5EJ~6+OcECB@Q%^Z1M@j?))9jd|Pdx3>u{zN!6KZ2UK;%@V!sSiTK^;*RB~
z2i?j1C`eDIbf>|K2l~5SdcKrDEnd4Wx%csw9%J7il@5tOa<_xVmV++jFAC~jqmFVe
z-(L067y0T|Vek5x4d}YNVxGA;)J}C$UN|YAe<_v<AnkW_xSszuZK(@ho=5DtGv5|F
zycUZ2GFM0`PGpM&hEVK9U@&CAj#-TPEdi#;69<2KO9WPmf8dtBRx@(X#8^eFnxg%q
zS@hZW-PmNDWX;U)sGoNLe}Mm5@MmYffPcXMp5mWDA8XR$Xb!R206h|GYnscZ;fIF_
ze+A>Im!<UZitKxGaWbSA?_J9>z-|J@iDR5I91IA<KPclkbTZAN*)*74@xviMk8O)M
z`=bLyFDz)8L3ws5xSw93H^;H$puF@>@KFMFc41*UPI?>{E#WQ`_b`8;W_SAHV`SRh
ziyf1q4=ho$MlfZljzf!j>1&B)Zk8&@e^jLbS&*vJ;YxHq=0c~Zr5)j})Q~oD@kaQ4
z2%o$hpg#~MpDAMuJdq(!2vTgC$P>vJ7p`LeRWzQR6GzW@;IhQupy|NiNww91i|IA5
zLj<{6W5=q%=fLOt%;zk%si?dT{GAJ|IN<Zm^2`Wm+#WVHY-8=SI>ap-tKZqDe^{<1
z`U&ZdWKImj%lMNR$|TDsA2@*CSfB~k(FFV!R4rj!VDV2!0)n6i_Js=bebVEL+B!~v
z(KXbnb-C8!Weq7T@(1gOWZ)q&Q<w-7Rm*_E1O`(|g9+Glr#2M0(7@+lS%GEckvpAb
zC33XW>k9&Q71&i=%3Eea)#7ZZe>nuFRpAi_@^M-{(Ly_^x8I-hv{O779>56sGD3Yl
zMtD?xMgY(Q=tTrQF~2+o5M=t%dOwO^^fz@h{=-Ynt1rIEgH0gvG=goTVngFEu~7mQ
zj-EUeE#EGBS!>CCI&y8utesu2eN($$vnGV2CDoDqwMeRAST9)f^_O5Ye?8k96)XTP
z0(G&WlJMR9;G*zKVjJ}c!i7dw_8Q|?bS^7E-z-XAkZ>g8!pb}Grc>;QfkkQM`4${6
zcDzbbj`GoQvmx`eh{Um3N5{3!`&`HQW@36K&C*)coC;#%J0{)|eRzN1pf{LsWm2DR
zZ~CaEAQ_#@?LOfc@_+k6f9{bNN)LTef@_!EzVdN;WuXb!R@a&Bup{45qmA4gowi_l
zzyo#SV5DKyaFP08J%>G;ebm_NLrW-)y}AUKDs&Elwg7|tV31h1&Em5h&>zCHxzV48
z6d5>^DKvQ@!41d@<lTV0dsNs(Q<5Kb!`SunuYXh&pU7lbLw%qce^9u;O)%|{yWeyM
z5XaDzF<yMVQ*<Eh7rhzVww-j$j*X73j?uA=ifyZ7r(-+mq+{E*ZT$27X3bp8tTh*P
zRaaGO)!F+z=RLce$9>9kYi*)S2IE0!{pRpV{x!e}!I2~r79y<oClq`JCpzcz;2P=>
zGaQMbBNv%3RN+2%UV-`>ykDf=TWMcx!H(%0oO0!rg++*S3@Feo(dE(#sW8j2{4Ix#
z0BgZ(X_L}@zbWM1to`OgeFdP!wpX&4DvoRhIu19$Fr^Y`6bMS3Ab&qb(DbqjqsvFc
zv4;*40o6Vgu9%V|WgiVrL9sKJloA>qXP2^tW%<X%9t6Tk!-wC`85Cv&{tFs34jb=g
z3dx9A5%ioyoyiKZ40)X7Dbhxm4MvX$CqsA-Q#Nd#WD3D(R7vp{R4qV=kDMWFy$GXy
zh7w;$_pS$@Reg()p?e1N!)tyoAQK0mDioda0R7Ii2g2<zNA9c#f}_}Wu;2l@wzj41
zzpi_<s`zUgB=}uOZ%BAIdTe<wmnAgDl`&|qM5Ny1@6ZMRh)qtu&mz3$=N8jX8N5Pb
z@X<gs=y8GS<wME2z!ZSVn~zD@Mnx*Dye&0XR}~Y<5`T^}F7+K=+z?*NOBc~CM<0Bi
zfk4h76pk6CQVC*$d9Iew7b6RqbPV(XZRv-Qp1esD%(jKbqFy9LQ(oQ-Qcr{E#YEAm
zOb^aXpuGJJN-)=>OH}2c{qhRmj(T!{@UBLW1r_&u>IE7($0N`r6k564hjoj7*i^ax
zhlsE#pBb}jy)VfPmYGrtVn`=Aj1oMO9_?_O5_-1?$(?|VMs15u*`&WB?_L}V*}p<o
zTtcU8cwYBM9P5BpgR4vmjrTW4TaM{${ixW%&0RJ(`NB#C|7scs8upB)Eg2ZUxp|lM
zwS;>H>OgJxRS!TEa%dP`$x>L?&1oWx!vf;rk6|{lpPl(uU@tL9SrqAg&kc_7A5bW#
zJ`ypk)@TOLHGSp;A@HZ>NJ**B(R~(UQK8|V+rnl6V{6dqt-ed&+EvF}8fTM5E8M40
zvyQ@+;$d^ig(UrAq)N#1krvg7OV5k2jtK}CV=~~i3m1U$=43726`pTsT|U(fVcoO@
z@mzVgTs{UxE=znnC;qd1jF*{R+&kh7aD}*1fggeDV%JY45yVyUx}~273Z;KU))JV0
znAn}uUvSeGSKczUJ;NKFI*h0qAlUG6ju<Bw7tNNj6n*+MI~F2eVO-7W-OMcuTRXW4
zREA)79(e$3TUVr`>cl*&56^mo){oLIs4OaRyDR<CjwlXx;g2T@;|{v>uG4+=p4i&=
zm+kgX5o|x=ROsDo3!?;MO)dVyO2%ffx9&|qfP*4HFr&=SfqzNyL*mT2g5*e&6||HD
z_K4tK%uLrCg?`?{J*HuErtf9ltHWaCL_>&CFcZ+5^QY^0v{N7t%<>(*vtkNVaQ}Nz
ziez5Zo3t=7Q#Z}e7vC)@k7kY`oAqUGk}nl+Wn)Up4q$vex;aRSP^NkwMOXQJgb+Of
zE2NQ{reVqz;TcsNL>k2rDuV1jb*qVuG$rQ!&j(6e|MoEV)qIN8D>w}FCmfHON?S$q
z0i0XclXP@lOvyE=Kk_grEK59In0Z$3S3C;RCll&i!UesB+QM;ak>jlLprhuX!Dsc=
z<V2Bt2bwH%QC$gK!U%7=UIS{S1)4M77stw%cW)KbNFcwhxuf|BZHOeg{iZ|rt^Wk1
zLSAZ4LjfyxAFhr@oqt+;&-+Quzu+K#0EP5yVMOlK9h7f@+Ilgzs9x^#=`z`;BMW{M
zA!LLCQF_jgu+geFvA5qnYY@yInVDj|Xp|RRrZYO=%FGT}Jr4V0;xMlizsKXtd_*Eb
zI~m0K<w6Pb>uh(yo#e|Yg7p5***fs+K*?FA^x&1*ecI#Z@aSu1yLm}<r@HiI2YiOz
zho+yWua$1Y`drW;<$6qDc&!<7PKFfqX)bMk!8cyQ<>G+8XuwQ~`ek2&@CVXB9JxU_
zg}v@UI)T|Xhsq(YYQFwKau^*raeFB9iGYDE^CNLaKI_DzL}_E<ixyUdCod&+Mwb1p
zZto<d7q$lC;op%N*UaqC<3o|r1_bvIq_zKJar?L1%^{gR*Sj>1E6r<iF6W;!Wvaqg
zhOtPhq*_*(QGLfJM*K{8Jd%1~4EZkc!M`Q)=!j~$@|u*?nj-BWRPuxOdzH%ulWXq2
zLfJFqg*1J9Y)mF%dP%6>i4tb*^&6ipdrE?**6*9|0bJsS?AXOWo7Mx0D}jU+$7U}%
zUtPCX);xlFgoa_H_ulWPYYcZDQU#O?YA{tVTTg<4F3jl)dQso%$V2CMMFFkeN|Dnz
z%(zUIZ#!3Ekw8nrV&t?6Mc~cZjILxnArJH`6iZ(1*Q5oKMy^`51M93E6cr-6>9j*=
z?I#T@_}2?yyD>v~8Fye=2E1Sv?60pq4!hy%2Ij;Iwl{@aO|<)M=;K}H-g6o)k3kM9
zz`sr;7t@#*RS7&XrYjDC6f<R|f+2VP7PB~oIs9^1NjhrqwAq9Zp)0KcW}!FONR9CB
zrTl7yvefx>1SXf5PJe|TtSRl|euvKKUXtjbG^L|w{s?q5nd`jB2QJR%F<-oWoR+)a
z@*%+0x8As%@_U{y7t}biKH-8Wxj)B+m&45udqMLwO`vYTJPwC4y;YNT18r@IdEE`t
z@z@fvbKF+jJCDh>6yfi9OalBGXv!P>cDcBHSB()t2krjx`|idei7PpdO`sTb_knBv
zTi#8hW8+R=XKF>H2I}jRdxouv;M9iDc10{zxkV42actz>!mfKTi0nfU%!jlLVJG68
z6i1z;!qzY*{C*t6henCS<9^wS-|=qMBYDnF2D0o7@4+@!ixrxJA;g=KGc_SELLB|p
z8Au(!@y*bdpVS;HYsI1fon(oA$JueXFA%UKxC_PrnzGOX%mnVYEx>F+7@kMRk{>EB
z6KJ?l%DD~1wKSM>OJT67XfoWXkw!#uswxpP7<Gt<V*255H+Wf<p2dsuVusGZR2)=k
zTtkIoy=la)S&-DPaNFN?UuB?kVTBIYzp)4r=i2$)p=zB~#dcZ8x%xVL&+*AbH(Z?~
zSbR_ltyIMW>SLBj1EdgqP+0RDGCE+j<=!O1&+c}C7sLNf*77JosHtMgvTM(kcY(!8
z<U@zt!{$gIFyjCBcZb<THM#_@0Bzrjt}r<NlN{9}0NDXL%OQv@{D;<}`=<41WS8rG
z%}U%p4MK?BCjIEjCWbr_Te2z8`77+^_RO4(NV+Kna@l3uJOQV_m5XEe3w=V656LU&
zYyY;aM`Q^blG}zMJ4GNH{ygStWbr2QndW%3ShUi7T;`he9ZC2Hry|lv=+B0H3yObG
zK<^em1gEVGV=u(16|xr`n6@tcD>=LG?89E2LK4ay$32anh-jq^3V{F^?)jhoJDDi6
z*gv_(`gi`|W4qGBf-lQR8Bv-b1XZN!9~?#n?Z23yCRbYckGPmE^yv+!-`OOQJ<3L@
z`B8JPs@IS~Hj~+$$y=kJNf{U#Y>U(~ha%G<h8p)d3NMMlWop+?pD{bc$t(rgK^ubq
zg$k0>TxvN14{GTJ>Oc+2|I~n58Fu4x&w`ZLvlRbI!Aef`QM$`tDQ0hSaq4pSJ&khQ
zD5aBe62vu~#i{Hz-?>tF+8g?h*`5dpr?%=AR@fiWBsu@AI1sV_3mL^wPA2Q6OB?nZ
zMikpoGRl1y-<XQy?^_!!lykgaW;kd5L0T#6D|I0ny7DYw1N%8exR_o~h0WZ#E^dQR
ztVx&WHq>jJA6(icTopWL{$nxjHORscV%O}{A=G)yzKmtoM2(CAg7SEdGr!_^gaBqW
z5$d*%RL*wvFIGs<&GpnP7lrdG^T?d>h3-R=>u1}e!kyNIejJAz<Eo?n@zney6NY^e
zWbU?glI0<w-736(Dxmz~;4$0H)5!)KkdC9ZtpNLD2o6s(Brk>#OOE6eQ4*$~^c)vD
z50Nn{)d=rf>tylxm7gRxzN;OQyBLm)P7uq#fIgF1?Pvz%aB<$tRn?FyD@CDy@<?^L
z?t6^d)G5{~r7bo*4L4y?)shOvYiZSNP6vKa&Sl~N`3%{i2Ke~+4fpeI>PC!J^e8=9
z)2g;8l~U2n|1H?`YGUWJSe=SY0-R0*-SWXx&HG0)7mHo4LINo<e<K!Jpi5%EHD@|j
z{6hzY2~+KhQ;b7P*g5V?`n70v7@hStX^lJ+KCvLzFSZTJF9Y<1rBKwWxcpDXf&Y(=
zi!vO<EBE+cK<i|AEuO~NyLPI;IEJQ$(a1%j#lFw)S{Bf01ls6!)@DK(y^UjddtgXj
z#)%Rc_>C4l9A7qz6?1n3s<qc%gLtwMX$;4Yv29)oVr8PCL7VLh@BO{q)Md#M_cdkg
z5j9+RzG(mg?cd?LZ+)#m!<pBsUf*Ol|6*aE$v0(!pflK7T0_x2GqDuUSCNAm_FJDn
zD=~)!B;TXwy8uN(AA1Ep%<8B>#~$Ig_4?N@pK$D}{k{#H5h_Dhef(+bSN57AG&!PT
zSqpO|bquByfs$q}l@*MYHJVr!{YvvFSBpEzy#T(-Kif+Oc=<A-Gy6fm;bw{8OtqJ|
z!U-GI0CDZx(s<ejrN~}u9R>}|IH~7fJG?{Qpy3JVG+|>M;<AT|P^7xYir%R|sE1{>
zAW%gv3^`)i=D~L623Jg#h#@v3I7Sc}nN~w-$zfM+Qb{gP3#dezLF`Q&q@UWcDoM<!
zeSl0)7(JGWvmMOuCSCsHF=a8(??0xl-J}?fera;hD5v3_j<9`e)MtjZsq0}L_9euC
z)MLc*56LkVWdjIZ*~QODbsTF>(ZDLD*fC=HjS4&`H`01*ZTL^|YUhdc(v_sRzzgYX
z<jB1F^jQD+r4qRCaq5jq91SLI(;dAv{Q$gENL}gs$-MQnMt^^Dno_BKfoxV{8?ql*
zS($%lx&oYUVOPK3i*y#m`gJvk2Y>e|%BY4uvOD?p1`Vc*0f!BXa`m!N!ydb`@oT2F
zN4oI4IQV>FsM%)*w{5wj1hdSeXggl{?n6zPA-+}4JgM!n=KF-eKi1w~Cv&Q^@+v^d
zOd-iWQtlE4MfUy;l7Bs-leIeKDK$fKF=lM=l#qqU>1gur;945_$G#5P_;4ze={rel
z?v^uI2{vgx;bOXF^1Vo>yXL%C=yUWIr&2Rp>IbGId{gZ~&ps)g6Mfwc*=mnrx2}$i
z?@;<BUYN-hk{>ps>lZG>@9z)AvDyG!ZCOOE8!d%y1bV|?+RuE=*kztnC#_*=*iM!9
zToEnmgQmcS^r&aM*_5}&s2g0jKJTqIhx|3=LTHhsEJ5Zy7bs~q!Yu-{+ypQ>!?ZQM
zpPjfgjRV)Oi9>It2>xW2A+6m!S^SJew}0dBNOX;w@NclI3b>m0Vri!e%H)B@e>gA(
zhD^0raNHmSC<G|jM984<evkpdUr^_+6yJFSp$BwJX!|PP>H>WW^lEF@|0g6L2$476
z)Ckk?^e31s2o;!YS_&`7Z%}ubt6aQYYRoGT_p58>zbDF6NIizY&nW8k=NOw~UQ)y!
zFoRkU_$#5nXe2_so7l)`;zfXWgux<(zP3O~<jqZrD=-Be?~8XNyJlQP5+7Y@3qeaE
z!r^EvZkd2@F(X5NjYTPMK^1F<IYVzlh|f1CyuboX|IsC|+%c`+4<Lejg9x1VK?f9o
zWJmXc=dsn8A6#D<EaRQifU`m>Zs+ZUiZyM3u!G=fsuNdQLok3~;Q(0wkAHy|0?jqZ
zwXduV0$Q5rqE#~$LH$5`40_yabQkeJIYICvmvRwX|1&4{&d9ki{$K=3J$(Uj_Mpgq
zp&-a8nIJ+Ce(nQJ709Zkj9}b~+%?bX27(|&V0=f6rEooe5#Z6JV1)kh(IN2l7gO;l
zEWnc(HMIh-3CL%Of<w@uNd_pm^x`2hKarT}K@brTMRh{q$o;^A1ESZ%5hZa!LU3^k
zx9zPpQM#aH$Bh3Ly=Q`=`@sglq9DIQ2iMO;%Ix(lEy!{4!IAOrH_uq((&TXLd*FnD
z@;~!vO8$|MB2tH&rGcb=iSRE(?*l;vB{PahIXeXC=qin$3)2y0jpQV<DaVkuL7gx+
zyAZcmG}0<)C{Q(wJdh9QsH!KhXTkn3X82{wMUMgV=~hP_B@#3K@y!|<|CU-X{(7n0
zYM8W}@g7ujlbM!~li#it=f^~iB|JwQd%=}1a}wfuto<3Kmi+5an)D-u3FnOv&6H5p
ziq{rE&cth~eMk?BN{ODoJX4wOqgc#BTlz)xyFnYC*^-{46%Th3Zgaw5Pk#qd0(VKQ
zmgMG9GD^)|2v-wNNkpMMa-eB{2h>9ox;-JVRssW4N$Ub#LbT75WCS*IvdtzPLl}ev
zBo3M+0HY8uCRhOdO;fCeG+)5C{@MxCM?D;nHZ~Cb{88jj98sKIs1-=;M|3@fuKh#e
zXqtHRDfFU6zLnx$Ri$1+_9|)VI^7oiQs`w!EY?VXm?3B}`0%%;hA<#@+wRXE`2=3P
zPi9w5C*_8EKDFiUxY`gmej%}t_Jk5Oo2_~xjPmHO>(OwA`kbP1rVOPzt}(YOn{Ww?
zG(+oU?TSy@J;v>APw>{Ih3@zoLhiW|s57XL)eU{Ch$%oGr`+y@C?X*=(#yC^MFerA
zM*8ttbjVCAKOqc%o@kQ(lFM#eOTy69K2oXeJF6^tqHe+Nev8jrd^<B|QR8k9Jpe`Z
zl>;>2KamaT<mYBYd2C6dnk;eaostDyd~@F3Awc<i8HN<JqvTCTiM?OI{Ui5$!i{GA
zHV<y$VXPw<#DV~)qaf`q|IBf>5H&h=iWh+?lsfc&cjd($wd5eEp4ym2UPYgDOMc?i
zrS6_UfVVQ7Ofe3@8fl$cK6z{nd=O_rk7N{YcB)iG;HlMbFAM`|KGCzWfZ-RAw(z_i
z&f3mxcOGoZl;1B^?b7eQ1%ff`_dv@ymVnlpHUQ2k=6hSXVJ-H`BSI#Ko-_DA>kQJs
zUPEQ3$gTwGiJiXT`!(dl#eea0eWsfQLoNAJR;A6$t;zSej`;3T+RsGuzH3>Cm5e7D
z%lB<Fg2t*%ColS$T<GxO7MM5Hxu7SU>*{KIpA-N2oah)>HHo_0YcUUTqW2i@?c(xY
zGQ(p!yb+~N>idp2I0#RPS(<bJV=vcbo6Ut$R8Q?ajvByPiC9X~%{j`{#i$av<>yV-
zga3>lbiXHmSoan1*2(sn<h4F}F}MtU_$@4V=s08nWV$VIrtDK=30NE;{-$&_bY6T0
zy~MqI+~9)lq(E$z?b_+OG1QnuhM&W~839{^{jU*whQJBbL);a5N}=>E!gqRHUvUW&
zL?L&o5L&@tu)^ba`pXwy*37HtyF{1$8GyEQVYed|^9!lzwaFFF+O1~&QMa!|?%`)H
z3yw7bzI9iv#GjnkUI3lQ+kD|Ws^ggAleTd6Cf%8v8P@dFrj>!RY$YriF5<kyQRdf}
zL7TIcHJ_GkFW`UZH;%qo1|#seT&&KW-#Tp=)7xg~MSGKPStHCLh-&AVMd+Qc>03U7
z#OdAqs1EJisU$UtxMLF8*2B3q$F#a`5YrHgaNC}=SL!(+RF5*kJWhr~Il*~VvAH7>
z$uQ5IkJmD_x?*9U_s>4Bmt+F$+<xYsN=CvKE)ILBox^S6#~jnFb!pzFeKS&M+CM5@
z_j+DE#NDR!+mlD@uDEWRqgfXP;dl7f$dl=^82c(MoL$(gI-7HLZd3cJ)<c<`-&SNy
zF40#3=GVdi6h01M@OD1;1TW7>7R?X)lct}j<E&sb^Vt#Wfl#p$EQM4v#BTNY7aN^#
z7)ETQlx01#8B<zkE8O^3OBUKq8qg^`S-PI!u{&7g>>KKxC)BiA$I(CYK2!`22RF7i
zEz@ha!hE^rd10G{wYG4=d0vS@qLV*_)U3LnBVi7pXwrNbl0C{PlSpz3zK9xF3|1`f
zQqjgyX21rJoHbPb4gr@QLR)w8@(&$CK;9d7769qqsBpU>I09Qy*&OA8XwF@-mOmmY
zS>q11Zji_%b31e>bfN7H&?lO`?s1~-_g*HP3`axU44c`O%Daq39@!pMq1RP>$Z{Za
zfVu#97vGDdz%Na-IoKH{sV5c&Hu)1aoGtrN6F7BxKw18jas_jos836O%QI#^QAM1C
z9seZHUNX4~B3S=j!_$FTNQT}$l{R5wl8?R4oED1Zor=w4<YYLMlN(QhSlVf2>6F{q
zYcwJE&CeKN5Ar5LMvsZpR&MQ@f*ZELr=%1(=pdd#M9!&KEzVe}AK{5R=arDm<FA(-
zWjuAU44IdIpIP<jdHa+3VkBl=Q8BXcO)=h3%UJN}Qq~&l&TanEd5T0rbaW=pbO&SB
zzc&jN`zB9iUy*Ym6!}AX2jNZs{-xQjYYCx~tBD}B3s#esRsD@1f06swm4kN8>X0dr
z2YU6wk@C<H2`MCQTdJAS%;mPNfnl2m-Op4gDo)!Kr>Xgn2=I}p%0ROx+Er1T@bJK2
z<g%uYVqkG#s)}IWAX?JOVBaTAYB=f~=Pn3ZWE&Jd8G0ywEBdy%K~w7HaVvN?*=Q?B
zc0<Xvn1k_&`*)K1aj0`7dgfK6>unM^)q{>#s27~=R?JsMn|`)wmUAj{JOTxaJCxlB
z?j`yy!(|+OY?%ERV7f2Yi<e9aUF_7$+$XFV4}XKi(3jtD1XOC<sAbL@yawfZQB4x#
zCwZi5F0Uz;NB70RRPf#=2Jfnq<26pmcQP^zc~VdXSRzxC%SNr%?k5npS>gcILj%Tv
ziYnj&Om#Sq<Rp3_xMIQ&=nzpa`sgem({Y{R&|;zwX-}u(qFA1C4!EdV?h;`UAa}d_
zL}KC~Y^MfL`D7bi!m9n?5MttD(Gl)uU}6#v>Pb#Nh2YHevz^U-f{R^J`6U10h0cEm
zAtfe}MZI#N#Nv*P0hPHCg?$93Sx9_+`}LW=bE1=bOAe7`tDMo`)(Gh_#LdBX)LgSM
zcR8V#q6mWDh`1+twTZg3W3Isy5dB^b%L<jsd+3<jAsEUe_Gm@60lB^<|G>+O&c0E)
za0$zYe*}FDL}T|=G&BTr^{D!MG}1IKo8@u>T~6EiM<cn`JY_J5g%<*VEAybMh)1!3
znBDY6xwY_$hm~7E9|_L(ZR}&wPI0tcx#KA36uku=Ap}6tk?LZj_uh~u*S#o&qelr@
zvzkZ3Kuw%ka}X9)*RWA~I%4|ROQgtrlly*T=;`V*dx^T5l#%P0<Naq*#a2Myjkgrw
z=bgE^Xcay*X(xN6yE_EZq-6x;2Wj+&)?;N3CklZFr^_~%BJYIP)bp#`>0LJ00STg{
zHTT`szqft%Gq`;pFW(*YB@X$QUnM@r;uwa3RUJehD_`;<q`Qyf%8cddkNU<B+*h3V
zg$6sO>ECT<3Ba*qb$fO&@pm$l(d^HQf4)<<`m(mWMm4cLH;Dkbd~mP8SFy!8OB^=H
zxqXxjjlLx~XRTRxGWN$%I_emPuX>RmZa2W}2v1<<)FP2U+?*IX)d#z={oR4e(x}Q3
zzh=Q$V}B6Y(ZEP1Hc6?v5L+I-;uWKaYa{pl#NE0h<x_a<;NQJ{!)`gh2`J1CUCyTK
zaD5$$DdaPYOfNuF%PbL8UAeJsQiw>*9c-Y6gl}Vda5+9ab;T<i@{+og$}p<P_?F5g
zU$u8OOmn}HWmqd@gj_k_XDOQEf3Y^huT5JS@4yo6ndruyQl7{Bf|cW>;#%lJFWm43
z<uQmj!d@;PD640}uch^LHn{zA5O__{XJ+~6Ew}NDnG@(;!kAt{(12jFq3XUN_cj^q
z7K%NP^qd=-$P6GG67*WUYiPK4BfFSBkA71kCHp+6EBJ%cI>wY2RD_tuhMQ0F&*S?O
zGRv!^+bDpCMB84GqyaO30YRlp)X>9&H@G*>&*d|?x9(2zXgl;){JedMLd8prR}%xJ
z;C}Vn`VFw87#h6aJ&{fKisY}J-+Otbp$M|u{6X=Q*+-1VS(ZW-(7y3~2X$d`X@OXU
zuWhPdYg^c{J%jDkf%YU(q9cq!TBKc_K%aDNFu~NckefG375^Hl|D``9s?|o(SLyBQ
zF$sTI&i0YdlSA;_o99$ykUk`Hz9;$P@nYgx&<$|wPN7p}!uVyzS6>+aTG(bS#Wzaf
zW<!=Rg#5p;(~HaF4%*G&ZJ(#?&rK|Gb1F?nO3c^aaMb=Hl-^&4;}>Dh+G&9x(<}B-
z)!iwf*6ba#m(j^*wt?B2^~}4W+{MrCa#@3meHb)FwWcpME3cC+AL$84+YEnU+IN8f
zsxYW4X#MEbh>!+D_7^^lx8hBS6bDc48h-iCUj&o4$UYn5=~5AuiRd`e+>95^WV&uX
zCZ=fx#+b&I`>dajn%uil>QnsA-bhAyPH9I=lslvR3SyA7ElhWeSnCoZ7WD0EyI3p%
zz~5@#Z~Bbtvc#PyS}-*$4(R+O)AtjAo&tziG(MRt8u5FYbmZsT4}x(@x>&RN4W2!~
zBLMaIOo_|o5Z3BWh}}ztsKZO12))D00ZD9<sXdho+Zkq)PzltY2<|WstYvMh^Tx*6
zX4(sLL&?Xu&K`%1fcMX8(Vt%yTkP8ANP$*MZ|BBVuWjUg&>K;EL}7U~H2!@+FcSmX
z{;)--ZbAfB1C5Ln#<XF48C^Kp5B1ojK5gE*V8Wz7P?>$CQJZ=3mT0$e<B<0tLTie%
zYA%Up3xa4-Eq*Y<aW$x5IN|TGwR#w3US2Mc>`biVTovqNzjuOpX7*A>mwBQG_un@v
z!T!6U#Xeo~(9iBX_>H&fJPJTu|1CFZ%(SdaU2Prr5jN&k^vO6-*xUxIGvWD-`$oLh
zb~DP`{HAAWqH0_rlHr{RJ>rq6={utu3V%6ubvmi+M%IDW#CpT=5}T0RwIW-Ci_KU5
znw%e=(ANc)YC83VyH?K&!g87@Lel?qt7T>!mAZ|V=fh`gjYQQY5A6JPp_flUE&(g%
zOlCH=9QY2Db0SXhn92&M5mhvaQvdzPy$M?X<REwU`7v*Uu|?@^V!-$@m@(>g8*C+r
z{;Q1q(A+f@T#=@$Fg^|EI!-YI7Dn5;iG%<I!HWd@fGmWe08Vz64Bbcjnw5=MGl_Y_
zJ(uyK?s|)-%OE|#6nLQ;Wrx)@NXH`78%;URB{0ogS%`E}3}@GYPT2;LeVDZ@G9t5X
zWbFktSOQN6RX^3!s`OIW5Xq@QaWl!T6ntvr|G#L|3v5{k>ROL=u)gyx_9otxsz&9v
z;GyQW1M5iNodscy>Hj1dNwR|bmVOdq^{svuBTl5P4^S?h8b#;i$M5S-?dX++qFlm#
z9xhB9Ikky#(tV6ijQDdowtjtNR*~NHqhNjF;@A1<7;-K=-7|Xbp}B}cWSvg4&t0qA
zdYiJqbSx0>mCEYjZ>FzptFOQAdopq0=4B0-&s=6qWFiT;i*{+J=PdJ4Bd3hCExXQE
z8|sIP0Kl(ps{=~;yGQRXWs`h3uaTV2(XKk?cf8?sB##QW)p61y5X)gVaPLye%>{YU
zZyr>Qds|w8=7=sr`3cP`HtNXauaGd3!|&y%g+=p<tFWao1p!kzrH-L3z8I)_JpoUc
ziSfsZ@I6k_&WT4pV91MgzY#%|b72RTK?A$h0n^x&p9{3TO3`OcJkB{I2<Vw5wkFc<
zzZ^2Lw+jgU6$J)&`#a1*-LL=>d6Yy%{Zt)?(A1Ux<`7AIsTmI7gdvZ69M^-*`u8i%
zc9;@qSbxd)!Xt&iJh-<R=Q`i#ls(zs%4cbY*~_KKDZxcu2=JD?22{@zyd+Wv$b13;
z>UpYioLQ@4x@RZ?kI-y^8Yx>pKSf0@+tH&h_jz*uc*wWyF?&Vv3m^1})KE+$fT%o>
z2#g<MZ#kLhW=(L4C)X)xj)%dDW(cyW5*j@Ah!lEs>FOqgkFx%W3fY&*d2A9H(<`O9
zml*T_DWz9riHC=?21RB=DljF$#F-Ijeuc=~`+NvqBWD^4?V`C`8Gxl`?(pM>lq7(M
zrT!9TDehsW?LuyKuMq0ww{+-Pusze&_1$o=aoOA5wjucE;-350zFlpT(T^+MT)gIB
z^?=zgbXWKvh<m>SRcqcAc-f07IEPgN4Vd9{RN*Fj>bUNG8*GK|&c{RP273AxxGUml
z5U|MLOgG`~r<7<36|sHYMn-iUi*GeM&+>jUS^UJ(yR(5+OS984{D;maxKKUX5@#1y
zUEw@@dVyUeSt0PU*tM%84Cp;DO(oxmx3+yLl+XI<R2{q{_&ekC$!+kO@=L1CbHvjS
zp(%PbP7r8!2Fcj<bj-OdARtSqdkJ|PUt;T&)SWQ^qgy7_!@3qtT#u*%f|U9-I1gS*
zDZ<|S&0HT!DFA^>cUHPaKa0=_kgT13A#Fgc+|XFfOT4_h4lY_PO*Be&)Q3#5Aa6<Q
z>2DbcxFGs`=v;449f!8IP=Wg#wQo<AfFhz8<1RZW(hvo7I1M-80sXZiVpQ9$UfEPW
zZdv0aGR2>G_$W4f|5+n}H{|mdKACEvK;I&Ng`0+5IJSvHL}3Z31ofOUp{`nb1d|(A
zwR~SOl_H0@kzd~o6?xlDES|wG-4GEU0zat_su3A^Qb!Wr@u_po9YE7!A?W)3`Eyk9
z<0bR&u(78jmo6b-ttq;PV$9DHso?>rRv#w)%iO;%ml1Nd5Fo>kWw_~`A0a~<9$&eB
zH8V-R)g$FerM{t<fr74bxVn>a)vIhS=eh4$J&QjsB?eYb3T#gpv!UShx(NxMhKG8&
z)j94Cl)+FWN43+#wy%+>f5oI?T$%-uPDB`!BDT8=W!D4rIZ!5AskXD1HL|<OBYBAX
z3f`%~_xs4~!^6>-8(A12{n)8bz1M`@`QNp#iiR|v)>LehY<2REH@lUrB2q}>h=&7m
zc+teGKO?<eD%K#@Nq1Aku;;$p5+pLMUk4~-kf4BdsZT}J!3hl2dvdUxtc;FqMJ;pc
zAsYVCBA)^*+-6L`7PRl1c8a|1$-pd9rypEmr6MO6f2M%Uwtw70-6tt5>L(%5ZPLKK
zq<a`?CR(40vl45X&pcbf7wkW!(wSR0dSGGj9@+aMwjersJgGgK+sU&<uuJ1hHoh2=
z<8#A%V~nEyR>^9U;Ke@QH9PEpN!sJ6OSULJ|1<|Kt<q*C>KZ>~kkgj~=t8&qViK)+
z)WCw|FroB&hR5U_OH5EW?SyW~c!;{)9=kUMDzT(Uu1SBC>SR?`MuC#f`E)7d+!7p+
zU{DVPS<wy}E*Y^9qEXlPFJ?VKGF>Cqo;{oVQqUhVDa4mw53{)`IsuUHB{P?hGdi2B
zwbX$Y?t!Ev+Z;WE21o*Q?$E|20i$^lRqPr*?Dy9q7~)SyVaJfxzT8TodDZf4jSc7p
zLFiH{UzT>Krl=@DrQhkL*=#2f*bAe3P~lFMskv--P74ktEU~=rgFcQ3(|58xpHXAL
zx+Aj*amWo#*r_C;$U}A1b;L>h-Q&5@;|`Rxe#LMbsG*-EIb)dJ<`bMs<$w%jQ!uu&
zTgO~?+K$?s^rv<yKyE%CQ$=g4l_xiTlql~vtnev>fS^`bxoQS-$KvO|BdGYw1Tn~Q
z`6j>{b!v(m^DUZrcG-(y^3C6)<EUPYr6xrLq38t^)5Y=nl=5qRkDePNHnE<-&IJq=
zrA94lL3iW-EiO!XujQ?MxB4i&^XQI)=;KEIC!uo#pJlT1wJXUKRL~g*gsQF%8T4{m
zMDDFWfd|FcrWS2^&+Hx1d&{ArefBIEBIhQ6m#_Lfx2<>XNfG$q%&wduc&-oF{Is#=
z^h}ugWt^UTgsec^USJd0f~uhM0tfVR;yUO-Q&AhA*VH5}UcG>E^@d}#%E&^6ulRdv
zwuKgX8U1<bDe^ixp^}tgSYI7zy-vL_gkwVbeyQK)_UAsuR~~CJOyb+$J^EoVk6ma6
z>7+5~BqpMP!j5vk%X`A>!=l@lXAhLo7s1}PZBbg{voLH^ttC_@Q^_}P$4@{=zpJv)
zLQuaT3d~K4pcTeK@Zk_63r5~x+t)V@fmnp&o3*zzKTz_=XePYGn;foY$dfCdu`DaO
z!t@}(L?*TWC@`}4#{&_?`M7dKwK3LDz07x*>jZoK_m#IG@68S{Mvs7+Pbh^Wk}<SH
zbaSIkt$&>@I^7jnBY*enSe}8zqS(6Lqb6yhhMe#8BgxTlOnU*&^}<jGgf$C0hr`Zr
z1vIVvI!Vfz(B%W+*$eo}8Zed}^?3;mEMww9jB;L8xC={JlM5P;uM^oHW!iK(kYs_x
z)g?JY=<DJ;`eq7TY*h@V>W=+53Fr4E)ppHLIDef64)_o8=;GD-r+a{qHqIYy>VF9W
z=U&uDvXD-6VV$e)D21ybFf{Xi?Vr8BiWSjVy5)9@(P%A)zeFi=(h6yiWRDh?vAKOu
zy|3$QcYN-kQ{~Y+Y+8>Q8aOmJEm{-fe5k7xBfxhZI4o)}PnuU20l}Pp!qn~02N$V5
z?zKuj@n~h72?RGNs0{%acYO{l%VtV>$;d|7MjQV{%<s*j^3a6q{vm8+59D>c$Q-=b
zhkGzNMLNGos^l8A+-Fc9i7+<Q=iqhuL7g8;2<E0bV;Y=s&sBIl5y%hQJD%(|oC6{h
zx9rp7Y-ZjQB5ih{3y9lTha^!oB@Ru=gE@KWZz?C`G15pcai9UU^%0K@6Ih{y4nk_K
zD>}rJ%2|IMk493`lBw8+!>-FvaL>NUuV;Zi8ELg7feW*mnbOw6E29O#)&&NUhL5=9
zl80nOg3(lLxYV8?{pu_bC|@}R_0tL6lgAor7Z=SuMwk+vf;>a1gEM^D5r0Guwn;FB
zbSauWMjA@K(V_%CD<9fG2$z(Yl7%C?di~E$FrS$!docr-9M>KpD_>6h5ZXw7*7>cf
z^nTLNW8#3I&UVG{I*5cEIG6Klf6u$d7#4@3cu_1bZ|u1ktv6Fs!<XDqgSinC5*g(r
zzOBlGJSvM!{(RmlEmi16Ef-;gAJ`sgJre^0SEP!je<1?e>l0kNku#?a-&04KJW-=A
z&w&KicSk{JH@{Wyy>7Iih<#Kh1EM0|;<=#m$^wx>xf1b5S-aGdM|g!!Wm|Rhk>@Wa
z@8AHc0*zyZX^^$~TfU*a3<<C-K&`#YdC|>GNVnBm>9n1mJU9Qzb@XVoNRLz3mPx*}
z3-<dHoE1=5gN4!y1AEEd1yxiZnf^@q(faqZ?(;iVSx*<-rRo|N+;reN&N4M`ukDR>
zqDT#W6+_E|L${Rj-uypwJ-8pWHru8OgliAq{a7~EOilz#4@c7D2%fzphp{YN5g5~R
zX?<p@>1~qd+84Q+!6Dz&W={IV*VKMCDo}q2(p3S2Z7@Y-OY8_8JB{`Me5g=*p=%Zh
z9n4oUGh6O=Kei5F9$FZ#(EQfYX;)Dh6(eLfialTS9Pl*bI_NXWP&}FoKB+#2t_S={
z4MaEK7-LjuXWKd{ylCa4w(l$ou|K;7xO(N0J)84<2=iI>_8@e!BE)xuG4f}*(x2m{
zym5iABd_gkvu{IFg)&Xn$;K%ZvunNmMgIJmr<7$%h_`s&DV*p*+L&}#ghM(hCBiGi
zBJJ5DWrKB!g~bcWIxm|EDs_~sS*T3h6=lwdTs2V>2CGLurNh7vt(kSxieAVOV8oy$
ziJ+-Z=6XydZb>H#|FW_CyF|jSG-fLl1!RFn%3?PA5&fvW<os@q-V$DK5^byOE)6v$
z49hvOD?J_qhl&&6#L0_w?ki_>9p;q0yum=SOn2pKKI)`}$BG(L>M1I{*W~8o>tVt^
zPdd6YhfX-dMJ$_B<5XI^fvgPI{h+HkrTX<YRGvPy(~rCdO144Lt{JuF#g|K4b0D{2
z75I)>4fkB%BIS6V#F_G^$jz|Y#Hn<_sn5WQ{E$>>TSnL=?<Ov-G%mp0bx5>~cteeD
z9P`Fkv7C5=UU=_bpHaZWI0$ktlWfQs?B{+aPdR?U_=0+i9Fo?MEL($@Pw}kcAVK}H
z6P635^DzwAR=(FZYdL25p8CCb3&3-wA7h{YmydClxPNL$s@Y|Mam3yK7IZjP7TnJ)
z$3rmep546Qo8jBhRMzCN#*ix4#W}RQTj}!4J5rYz^DTS1c~Q@W-_RKOB1X+O*bBu0
zIv;MPiZanDNx4bn@#EPs8YJSo?XTaOuy~iM%Uo9MFg?sBvOF)UJh?-a%>ZD1Z}1Jh
zO4}RSv0C5lQs~j*4r|Qa!n5s)#*LJa;ar1Vlh1K(W``Og5fJs%M=)H4QBuU`n!(`N
z99bq!_rqrDP0Tz;%rEC2(T<w_1+eB<AoIX7NoTK`@df{}zlIq}t1_ph%hvc@UczJw
z89kV_JDw)ph<Oh#oL#am<p;FbX=V`?nia}#QF{QzO+2#tXtoZmuBMNwYdwl2FsKe8
zjRQ^Pzr{H%uE~4~#VVS`l{!rVWGbF{n@b;EPD^EU`vc1?p}#M(`%<n;xU;y$)PMLr
zLwQQ0k4Hb^?#L-qt3f=uWS%Eb!fuap4VlGsSLb0=@w01~hx0jiGy$h1j`-tuXC%#a
zXKeSd64qoHJ+FjsN9chot~L!<dP{3xRagXDxDzL`XbkZ^-5J8nQ?9A=JZ@WevFEeh
z%TUrzboQxiR+1BO+5Pt(;z{4E9pAC~-fbcd@@9=MH4Fa7iKP}r*Ps3z%arxmXw=4m
z0{nBmYNh-8K%dc#JivrgmcJh3$8keBMF!bs>Nu{bsK1>gDWXg60<<N#rPt~D!?_DJ
z`F!0v6)_t;H_p9nb;*NW+i!^qCuz@hCZ8FV>h_%Dwp$-H%5Kl%3G8Uz2f|A)b;Hyd
zxg_6KH2|+DGjs>nq1e(q$$#L0Pp}L`%Idj5#if4#2&0}A8ps`yPb5!B5hh4XBk*Tg
z_x?2rH#<&tHM=Oi|LQryLwNtt<r#AF4VN`EB-`TFYVL*-&#`XkhehH*C0w+F{2Z)(
zf}RJ&2NZ@VS5%^$fOr_T5_4j~rR}v5&f?z;%OcCp84L@#%mSTb3dI&pf?dD8(#{|F
zl*QUHnG2V5vw))y4;YNa_=5sY9ji^}@`ycL#O)MmDDtXT(MIwJWg}^V>iHZ7B?0J>
zu>y6<?;Ok6*jemHLD69q;T@Y>4O}Abg{)Y2_9s1V((zTKRs*%qm7V$9K9i>!57!*`
z6RL8-iOB7Bu6m<><3F&e{Zdr~t<agW{y47d#V}px^Z+lz)xwRGPXUxpyjKpZlnL#%
zx@xt3yAGj6<{P@uQ&=e~z4^W+Pcb9zT$s@>F{@A%5w#J^V_lkHCoU~$WVEAalZCC6
zsTZ$lNRm~g>VF(y{Fza(s3Y*3T(p(iMV@-{o!mX2Jt_Pp4>q$Z8b6XHewJQiTAc`;
z$*KP-<^&qXTlXg{u{F<{oO7Mr=;NVd77sBx8pouAY`F01tXo=m3Ez7DyermKeu{w?
zLq~`F0C6-=gKA)Ac3bwHI)l6eam&Q&fj>yo8`PN*NsmsS{m&4gwe^>Ld(h_t!xh6{
zz2ls^GoR$>`%1J(huR9R8xC#0j-66NIr5y30`iDBT+(}<$p;Rump=IwkldXw=rzf^
zpY-||a#IyGxStybLu^`y`?gvuu!oG_r|u<hs;O<hZ@+H2Da9(ux*iHj_xIByCprie
zPKxJ-Cb0%F_GmUJ@6B9%t;q_||05H${g-j~Y~Wu?zkB_uXUf>JSnYLGZKsqs@Hz#`
z14uh$Gom>%FJ6cLA)`}#$1Ym*fAAb2lfQwD14W$G<9~PFb%fa|$w1uRi4*ALWEmPO
z)Mv~7TC2-~ybl?#3Ls%+jUvGl`#1%|7lWOQDddb$&YsUR6zg;63t_R&tC77X@o!FZ
z7%s*7>s9S4ZZu{Sz87)Ev{_^NM*HQsfw(~kGB%Uc6QUtTO#~o}LTm3$GmAAX9mz_%
z!wTt9#Fw<==dccGGiEvjB-^((pWmw808`v1)qox2{#T+n1-hR})_0L#b0-$Cgy}&;
zSqL59(Db2ppey#zhnk{zLFk+2rGKhO{v&Wi31Kyq1f_ySEJlHvjg)s0fzqI}09;qk
zFR+L1=2m68s~?8{vt-fg`mL=>$=NF;GETV<Qj%IfYlN&`eHzi_w6z@k2W~2hBUxQn
z-Eyn^b-B(jKwuV6c<;#cwa+<^ouLZ{!dz$7KUDF9V-XfoX$SbA2yY+>71KhU`wUQ}
zv2xB|^`0iS@Jdz+iW1>1VV&Xu&;@4SxPUVe&^==~4Y+JnkkW6wX^~QT3|EpjWFEU@
zmnf`W-Xe|0i$8Wy2qN*&ia$1XEt#9XgQ)e>SPG@mvVbWJ(3l;ofQJ0wKBhGLs7W|l
z;v>A|70L20n2?KkOA8)i_n0n>5P~#QW@01=e3>uM@4htpnHhc3Y7RFDa6aGa$1JFI
zkn>5r?AVuUsZVFc-nGV3wqUyYz24`%%>4xy=!HEQ!+(pIO9<0;4`8gSeACOl(*mIp
z)=zgeKMp}GIYF?D>g4jCI#@x$JM}#Q$!T3rFv`kc`4Qnr&*_uK>=bG?#%z0=p{5JN
zc=FTmXz>05)*%bWaXVuGa50}htn|gsmxW}53BUe_1e@OvKIc5_<z3H7wcF9~&d^=+
ztgq-qJqzc`%t)SiqI(6H7Rma*iDvSDu%X8wYSrNu!1L1TNFNY7+tea|A=I>uPjW4W
zSOgAmMXifp>;uBq;d$dYel1S^W*yAG5<4R2{(^BthuGlx)OMf(uM7PFAQLT%l6S1D
z6s<<JPZ|ZhL$C&R%uzqaesjlw!#Z&C@i7L1!wx&w8iTu4%1CL;8HkE22Nm}gug~^^
z9RE*F#$C>#U)42<hW6p@a!-f~4b=69RQpC8!ma5raHSRhZe7Pb)qV1Z&k{B~&V)d^
zw~>?kTscd!HUjOwZRkhbmyU&jJUAA~--wP0_&wNsNiz4*c^Qte0^mj&X-}SydbNfF
z8Ob-mbIqMT`p6P%-CW|BFV>n?+4(XxO^R2M55hE&4`!YB#<+rN#)mu7jTdL#eWwtq
z=L$UiG!JACb6|1{r~XOF@Nt4#Iu7*g{jrWG!fF}K1O@)J=<c<2Y!R~uEnMOJA(hEn
zy6ZhFHqo9-T)*?o(9Fp;U8h4s!QXK#QaDw<RrS#LnRhS4`Ks2ysG$(QAW62EwO|pS
zy0|==Fdp;PB6P?l7Fw2rhkM*4&_G=*Y=A&BN?`f5_?R9dk8qp5M5UUmqu`|OK0zlw
z_MGJ#%neBSL^|G;x^{0cRA6GknJQ{=Q*yp7$`olI|JhSE=k08AkI&N`{9lg;1!E`W
zWIM4mw1RFvGfWDtL*22bK;lzFlNK!Z)jpH>Ozf@1h#WDwHB6jXc7DY=BWz$;AkAAa
zowDTgJDat!&L+J%?;P`wC=5){KV)`uQ-9_3sgQE;i!NamE;}VPud~yQySWI6c$5uk
zoD~hSzc8itG&_DoSBuVa__kXgD4pZmDiYwUdg{}1%wan4dFk_1qy)H*1J5jHE4Ssd
z!(5UzgOH8z>-VD%+kPjbgr2KthoNI*SW?-@WK#gOqK5;Id$jqi-cT1jjo;y5ILZ>x
zo)a-;dYC4}7r{#{lC(fy@2$b3kt?b0)oF`-;BX*6^fYWBT}x&_UE^R_7ry4{RW;Mz
z(W%Z0?&z~>ZY+*DV>{j5CO4zbn;J-;|FKv(Lls!{n&JD`_IA8Q*^ABsQL186@D-*#
z$dbdf5zZJ8tmyo(*%U>pGB94OHXjGd%Q<7jy5Yo3|JN>l1hocZ(|_KMvlh-B?jnjZ
z!e=L?ISs<iZsyV{TCi&esyE+`Kjq(jubkMwi5!(a+arE_5R^{$ZOsYf?XeNK(TBw=
zJBp{lvRs%<%1|^npjKV~Y8V}bGi2JUFK+I9x63p}#w|r!+uh+Om>5YKff3D$o=Nxr
z-<EOaag|w9Mjqk=oHeG)qu63>pi=D-r#XaC*O4&`#oyk{GCp`r;2bJ0TqFetvFp3>
zjyOu=nO{bjq?w(`-vs$=E?2xiA^oYTPx5#082ZE92OjJv%zve;grGEl^2EiPqQR5(
zQ~V+>16<ZH*NQk@i+inJk$9T(O!}f?ByE=>xwg)b74rW|{MQU<twG0<92*#b#Fb?}
z-!ZS;k{wAv#m2=vg#0nWaKkPPn_#?`j#FV1yqi8A{cFPc`<1Ngk*BtvE{(N?zl<Y(
zJ{G>hfLP{)i_M(~dlxW)9wRA-FzcHe1r0ku3HQ74h;b%Fv^*Dd9K=y&b`bhpzL{Fo
z=EQ(F7i5@1NRlW3qJY`mM2B@4Nsg&HU;2B?-DKi12J5&&%9kl`vji=~b|M~T@#!~V
zr85riz&#jmLmErG#<u!pRvvBL*Pc4G#(kp0S+LtMT!V)tg8*=m=)pS=<KeVWI7b9w
zFox^y=e|&}-PF?vB*+Ejmiku)!YNK%o0g7^jiE}Z+HM~(l-mvEPq3^F)7QMPYB2Z|
znHjEvXR7$>mjS;b8hFwjG<b&9f2n=wk=SWP!?Uwg`buPTM9M~m^G_GgrgFqSLU@`Y
zva(0#uxFUQHv(9(-r<}fr@5(31ufs2?cWpsd5pVftWVRA5sS~Pe9YSNjpIQ{CbA&L
z{0`ap4Vbk@x)vwl-K+VnXy4wwYx!IkH>;1iJu8_qJ{QA~7vqPt_iR(#Zu<IUWxA>O
zK$!{$ZdA67N^6lw20@#>I`xOTojD?m?P#odC1X_4c>w+k*A<nYHlqj1)^wBo0n$o8
zgDBvs0@cbs<=CqN4qBQLu~pKf^eC$1_KJ{d%sTcY6=GGwGVV~Zj*GlR)O+YUqaq1h
zLy9{xdPKOI=h<3}8>*bYuY<NK3se+3c+J|*C;P9E+&9f!R`@Kyj|_#au>Cyu0AnNk
z@&NzpW&sLn5$w{jHdPWZW9*|mSG+_#SXciPndB9ZI6a<&uVg<Ae=$GI3lwN>+G-oS
zpEBN{p>?PEN@OOg0V=XM?2fA%)NjZ(4^IkUcktaKt?)D01$t4Q=!I+&oj2m0jPhv6
zPTQ7^l~9|8*WZ}bXCx0Gc;g+Uq_?&E2RNIhxd7i-=nM7*fJt}P?q&GloTK1z$MK`!
zj5D5hZKP3#-<=3TA`wT<h&_DTWX*FwtU$Vp7A(A{f;rfz(81MHG^oI#L4NUmlAmp#
zlQ(Vnvd3RUSz2m%cMhWrt?}NyidT<yPMFu*?Jk3+tx<S2i(@saDphtb+A@j6#<=p3
z1fXD(%YicpZb7cK$Aj3~8uL@j4}c@FbG<T)@Az64H^f2?`Jnk5FvY!K_I5uDH|RNf
z;K41|`Ml<Av<y1FVVxddaUM5av;}&|Okpz-jhxsUFV@p6)-}$$$=dNaLgap#rYjeP
zH4X1mhgCiB-A9=$=KaM*SxX!&DLPUE39zMG&8_XHj0IuDmmxn8ew++cSHH={91FbY
zPNj8Zh9ZvgC}Q)$E$FLrtp;QB%)5%2U#T5wN7;m&B}iXj=Ple*NOs)k=WX~|RQQRm
zMT1-M*w-xdB?Afa1JAs-mt^46&=yRDG&nzPVcNROScO>86XI|SO+qvjWKSvr0pSX#
z*2y>uFdM4I3*w2AyU~plOlL@B=?@T%tNATdnwOg;e|9si%BSenu|cKpiC}f`SjT5}
zB+7GeQX9WsCB}1L`;EHQ>E>n3DcxQRbG;ZQk+ZI-dY0K_Gd(=S*z~^eC$$GnFe)h~
z1=uM;_-0KXvsIIyvN$dEN^tum07g@wxM06I8{M4RIPW5?Zc@JamGQodfz*n`rV*z6
zV(+P!Qeqz$xOi4WCg>9MlghZx^h8dZsLyhvK>3-h5Fs8n$RGqU(S&mhYK^}yj4}8Z
z26FwlRr+R(V<oE}1vJ9&_{+w_d<ZmQnhy@X#J3X^3q%>b)<J%cFnR(D!2L+=h%BH+
z+S`&NX^Z){oZ5aZxnXa7$(`LL|Nr6Y9D_3nxGx{uwl%SBPi)(^^~6plb|$uMPHfw@
z&CR>-*6zQyx~ltYS9MqSz2|rCxkwD4r|0=FRpfG}Lbt*uvO@!`23lm5jT7dg+#fP4
zyb8{pFVTAE>zKXE<u)%;Vn95Rnr{OpU=0sb-G(WCiLsuEPH5=sKR0-*T(1JlsjE4K
z+I+!<&K@AAb(HuH`F2MK2s|a<=Tjj5YZ>ZRGwt$EFYTGcScny5sz<_kOWcomJ$B<D
z+GYc!k*zu<oL|`Gek{v_6ECm8YHv4)|D&&$6n@e#gJ(B|D1-B`k{lC9)4`Thp!y#a
zYEU_imx#CztX{^!*H?{%ZTWihwg8`$U0g7%PbX5;{S?!kkUZxSK!)9+V8_(4@Q8As
zp_vt8Nrt|&`vlq$+{hvNUkIr?37yUTh);Xwnb;)tT@zZ<>4MLagp23TM~xa=Rv=r!
zTWr*1Wo=+b+Q3*xgH?9>M%D#d8&7-fUFuUrxv(fCg)2fTN}uqU!n9gbl(dh}`+2f}
z`(y(}B6tnyG@lp;a5h@by$rttp5s>UDK<9?RhOyP_v<wLY)kGN!A3Ah_(xosw&kGM
zne|7-F045yGLA1$ydyYX<?2tJNQCC<@+UIHhB-E0+fJ)M3o(%+V$8Y4%2jwd(~JU2
zQ%q}?jC}~t2AFV3l|uLP9^S%qnj3X2FB*_`<NJ+Zj`WNIK6L<DM}%x+h|ZD-hhN7r
zJok&Q;)b?Fk3yJ)_LJG(fd#9KFxjcf8V<26Y&ow8n$6;yuN|0o6u!VhXoCk}2O~oc
zQ4&BkT3Wb`RdF<rh($k9^~+MY#ftMG0JZZVJBCZsp>=*z+}StKqTx4RD}a)3xeHDY
z&%gfD9=UzMwCQ=nT7bMR>AvOLZqP{A(jgF?pf@5$s^&S&05zw^hXPMb&|nc3ZS1a`
z6N&$8j7(yxctgf*e^SrUJ)zPNSA;|r*9(PFK;VTmF)z|&GZ+lwSHOUL?2C7<JDB+6
zqn&wfBeyRO>5;>P+|gq&SQ^5~CiDQyNvC<IGXw&Fw9flTGNOl!moPaxI5p~KHmQ9d
zN-2TiNZSiY#-|W7C+Zp0arY3z_zbwA8Y?@S@jCFZem9duusyetjGV3yjvf2BtFm8N
zub1(|mY-$mXI7ioW4lM>Qan;3T9Ixjh~GUX1aOENb@J^sR~&Z*8N{3sJ^Y~ck78*4
zBK8Z=NJDoDD?P?!Hsl+r5PLjiRtazG+%3dddIgc2%w<|KszVt6UuIa>%VG*HoV0xM
z&E{7T4yWqvUS2x6Z&X@J{;}7_812g`(YHN#tCM+(p#B<pFe1h&5n86Nan~{CS)`}6
zo9=#Q*+vi7UF9wyw-L!TXvJLSAF7WH;4$L(i{qG%e(d;6iw`ryql2yJM?*fK;pnCs
z;<X<78!nsroQ>WD0n}qMrD|0yDDr@BmbSW`uKdt^=50cspv%5i0ClXm0ib(cedh_b
z^+$e0rK3Quk#O1|HMjwA5S&WrlY@sY0Oh{-cPUqUS{w}d!}k#mU62_JbLS_2C4!+(
zs7YD8=d3y>Zo|nfz}wHa-%N?M<I?RJc7eaa>$uxpuWCBCQiZjFbdu(<LCUIpjd;vp
z#(L2X(kck9-C@65>ddxiTiw_Fpe(y8zy=R}VPtZzyk2E$E~wbjcE*4B;2LUxz{BRR
z;psM`;psd;5IU}^1Qa(q)CNBr&_Mu$<$S-Y=^oH&YD=(uWWgNz=ck$J+uptu4e0!$
zqR4OkoBW@0fOan<TQJc+l5wwHzkcPOS_a85e)y-0`;$afmY$-5D2bn{zLj=HWiMQ`
zID7ZiIz!sn*6)50FLMtSHf`_~9H!riA&{Hd)~oI5_Oh0+48jXeoM8!IMqPm&*2V2B
zSO)sVUcm2xA2sQw*7-4%FK{4r+JL65eW`;`O$2=Gq1<(aIoH-zUDjnUe(44zBCeh7
z)v~O79`uwWdivb)n9{FoS+3OM#B4bZr{@qlooT+m$2m_xSi<mnWD8kuaY5=8K~reW
z-2O3Qj<Gc7_z)H>$QlRGNCDqd7iIkkq$14^KnsV^7>Ub*swE?Q3(y#$7A2;Ez+k3|
zF(BQCM_#(e-!dw#Ir1sA-(w%MH!RlB!A?AfA{wf%pMb(iNxoExQsm<qOkama)_JLC
zCi8O|0MLDn(!bJ`lt$CS4tiOAi&r6P<t(Zfwx3|`mr)y+MP~!fHW9P|p*A^{Q{yI|
z*u^6I28h6C(~Q8$mUl*;hJgp0N}?L#q7pLvW(=?45Tr~of#`AAwsG5HN{Up2g$G`D
zfNFua{b#WG<=Zg373yrP3WJJ=_)`d*WPYQ);rRKZSDUj<-3l1AlK8%4W+JJM$+^uK
zwT7}METGo;9XP=1*$%I(57!w3-P<{B6vCew^=j1;eeID8HfE3dHTaH_!9DDN70~(2
z^b|kCF-oRp*_9CNJt%eL1XycQo%_SUVWGl`DRr0tRitdaXK4YO)7qN`rPGf_>Y09#
z$k*f}!OQsAJMVF<#J^sfpOuZ|q?R{hcz3GmqmVT^(|-Z=teqPgQNo`$>*sRuI<vX{
zjT$_uEZiyE%q=5go`?lInK_Gk-zj=ka0}y)6`NQ>)m>Y`H+2M;wOW7Wg6sp7-y52d
zySL`z(jv;!Gykrs^SGj0tSA;HbjlNHr^o#oduzFbppuHq%+zaHQma8tiw)WTGjFJ3
zgTcEXXWRfFiWQL4r~$K!6eJ-MW-ibf6y7Ktp^;<;r350E*bm8!y{~uBTSEQ~i@`9~
zub`B^QcCV~{;0H1xnJ}C-b2t-X4kN9Y2T#c70H#642>Svh>4P{$BD<i-+j%mI_T}p
zL5po2Jx@$KX6qszC+kO!JCelLE!n-sS<a8zaHatWW<eT76H$Rc>Vd&d-&L5Xg~oRO
zaac^*cHryOTto-$P~rO7z;BCEAU22E5U5(XvZ<Ieiex@T=AVUknb}sUC5(#?T@0RB
zs;?G+2$Zr-8!h@`2!jBxf+JQxw;=zr0w=R?^mX=8{IX_@_^dpz%RzJbbNCv7y7iLf
z%n1M(6Lgq$PE54LZv6O%YyW~SRPNmfWm=19=e%2W$0IA)23NqlQS}($`4ND$V#F~q
z+IZ}OaCNv6OH3=IS7%2$4D~93vf{vJ>>V2aguV%ES<&1N=a-uHkLR=hMxrSvfdlzF
z(MZ@%%A@e>>O+x>xSfx-U0Y3KV}o(@<kS(s9iX4f@mFrr+&7pSPqkG)q(-`|q&<>X
z{sMB!ha%~N>D48n`EO15Kh^uuPf*|2=0GVv<SKD&@sf4^dALixoIXy7mPS;r-4o8f
zX>ihcj2KA9rmTw}x|^L|X-DAMkf`=wo~Q;q+09C|dwn{zKE8Ig7wD-)at6f&XvC1d
zN?hnaBJEYRP(I7v46=1oVY=?;^x7k=qc<Ca)nQ?xP$xZ*Bl$CWVkZ*`Yt^B;dUakv
z>6?;h)2B%xfhc*_0LEZBx@E~=t-t$WYxE%wBokS{S!1;yL{Ge>Qf!DKvMd4Q0)#CQ
zgOhFgH5&q~r9tfEnASc^HwM)Q;6pWb4fYd3_hKY%fv_!cpkkQ=_Ata8nS!-BLaUUL
zS|I;4Y49iMHQ#k;<D`nX(F|L&44zH^;fEDjKM*xwf<^3NBtGc_(WFe`*@J49#G#;x
z51~mJ(6O6RCuE#+*Q3juJ{Fc)`*-qO%ZuS4bw#isGh|r~R6TE`Qx25@pj#teQ}X!j
z_v=HDY`n}r5QSGd^OW|FL}Q!WR}L3b=~iy+_~+;TLgVL%Ss1yg4;k&t?=Wg=WBOzu
z-A~C6mxcx&wu=UA5a3bPxieNRAh&G?30T_XFA_q?7VS>i{Tq->DK+Iqgwrw(+akFh
z-FH-!>JQ*Kwd@+Xky)`Xz~TwQ(%nYBGgziR;|a>4bIT&pvek7$LUEN|Ilg$__;{w9
zyTWgvFM`KFMx9#A9Wg80{nex53-`}p5>WcdK@?1fTLKUe7=VZK3}h_FVuMLn#2^YM
z&VnTWBphXF&`MUFo!?6?B4B1ku7d?|CigLXEy5PgMW|WY$83WL;Iqr#ekh`A9X>D0
z{}*TI2)ner<<(ZRh3TM`rd{hYHJF^r@%$nV^m&(|Q!DvKz6Jtw@*{+YMpEaOVX%vP
zTNuLP-tJhAJ7GjDw9lQoa3ry#6m2Re59NRvC~HqHG1Ht+j}+@YVP*Qhnl*C6tJ->?
zLPuR%J&rcbDG=uYJd1qPMjvgxXWA_*f^o<(jWC`XN}4qxGW8(6e{eOhrJQ-Fb<48S
z`<*6ZqrhsjK;4qA5qV+hFY3x*ULyOVe4zXYe<1&C$*qHt@y2FvGmxN3jowidf`DN%
z%Ben>C?$ps4bil#IE@O^`Rk?C)HJSTRrfBJ6ETO}HF`<FRSAN*(1OzR1agqBes9<8
zGh7vCA!3ap?oJkZ-N<B714H<_3E24cPh~3`|0D|bjrOQs|5kyL-^1Q+ZEl`CR3<15
zeLG<?FJ!^YY{HC3CeDgzK|sd#Lxi?&9RNQv2y0fhP!{Zm^jF>_!S^W{9d!jEOn_$a
z?<$k*(0xE(Bo&*ybc}ccm?X$s9Qg|VK3TdV{Ov)C2C{4sf9^uzca8opzBl2^_-X1+
z%)dm*XIK%P5xM~+7sA|;kcK~DJM$hZ{$cRY=gGaY$dX_;wYx84bjzsXzzteHIW;aY
zgf6>^^0So2`Lf6D>|UC1t(*a#G(&tc(shI@jp%^1(Y>RCszFH{TTa&F0_X|G{4?a@
zd}UY8yykxF*Z`dP&nRP6jLv4b91q#k{45uvwtuQa&d%ZlF31!YyNp{wrd9tG+M{Uq
zwN2hc;Rft3n4~5~^LcZz8D0ndtC9DA773o6b9%)9epJ^jN+e+@oyW$cqsDw0SUh3W
z5?=rXzR7XQ;Ckm$Ho6^$o&6e~Y2lXIn{U&%V*o~=7xUpm8dtW}AB+lJk~Xkjyr(Iq
z@k%yu61t-_L@LoV;}oLd=v{dul3M?#q5ua1LG{`45{Ve3Q`6iuxi=Q7Z|$8$9>K1g
zimVPISLGx&A)|G~>y;RV%prNhXxqV9+f6_|DZxZN^LEqIq5aTVR)%S3hud`%XSA($
zqMhYPfeX&LJHqlB&&C#=CPc0L_KOh2OB=08tnR&m4^><uk>3neCBjB<M?^fmIw;nu
zQ6Wu>{Sb*Rul5-K!KL1!a#VXhY;;VBvy}<0ZRYvUD07e$UbRR=f@yM)So<nP4G};u
z2#JJNQ*)=gF;t1M1tD%R2c(}4E_3W>NBuP(cWYq4PD-88Z+iS2s}DPU@o7iKoaobm
zdgU+*K+|?BGm9dSAP_Mp&wd8tK1FmUqn+?#Vm=v+01|TguE7LS<>FIT2G&K@<6>X#
zVjEca+0Sova>_UrLBRJJ_x`2lFcMI>Ze7a9DLDBPNwW>3Y5SClQG`nrwz1I&Y>6-T
z@3nM5yaMn;)O3C+ZTngLvAddG#mOoT=Y7B-=-(HerXX<K8)iOk*6t-wFF-<i8b3qL
z1y5CeH_Epf4l&NBzA`n)hr#cm!t;-yHVFl1A&XC>Mk(@jE?QU+s;F-C!~wu05{#7B
zENiDd2oiX~BO2CaHUO|=s^dDRx{19=%%Kpy@~4dhus1JWWd6675;nSlSC**pl!a0a
zbdx<!AjZFp+a1K4Vv0`U3{RdiN;!rJ)WkIh53!V~f2v~kty#r4(7tAK50+0}*qOng
zZZ)j@xtG0tmC~10+Zo~xIP#4G42GrG3aVMKozSGT?1*fxOXW=ZJo{~bZ<5^Z($U&o
z{kEvIiUN`eUaSGrkEDXCnWF-^rO!XWw<<yA0jdLyT~~%aN+oyETXPdwVqVv8rY-UF
zTjQ;PY0Z+i*{S35{;1#`yZ5WC`jq{v1Wb7GT@$4c65~|_7(dzo7(OIfW(UtE!p+We
z1Pu!;b4c0u@?WuMzffZ3+0Fhz@`Hx7DCtuxShSg&DPwtQ-M{wOE7(LgQD{Ve2?xm%
zG-eH*#S%BhiYdB!HDFCig0cy;9ssp5m#D)#EFH<7g)+C9x=*BD@xPPEQI8B8T&4Ih
z$?~F6RY`8--IenOfWPy%IfDt5`RNBBrzWG=bw#wC5wTFoCdn9<Nw1zZ6=GOdO`wfZ
zO|InEi?z;Rp(*dLyb*zf8I5Bw{|m$>X|(xm>5~Ep@%O9#r}{KEN5gOCOvmqxYur66
zmzTVBs~vDMTD0+bLws0wg}kWhcB=AAx`MKjC@lFiC(Ks`_^bB&SGcuaZsh`$0E}oG
zvWi_{lO8rybYL5;MtCW{QW>!lZ;F&xeSk0c`r2jbLG!ft(k}cV;@fZe!|rep^VT^S
z9QIj>ZNU|~(LhL>d<g<CI>g;F@&w$`Y*i=uAy{psqlH@VBx)(H+5Y-cSpPV~|0rzX
z$wy;n>Ewz4vfha1u$>}nHGQ^oMWhVjzEsES?h;&k25V$4_N1M~G0QGJ<dBFK22Qg3
z)7B#I8of?Q6qQn@Uofo+8Z5x$^k;B07cCLD+KiDw`tat81gS|IhGln^V?vzIiy*JQ
ztBqW2OIG(Q+2GNAbtFLPLR#mY=IOl@gH&pjm^mT<I`e?uGqBQ@N|9)k$23W*X<<1@
zJ7B*Fx%i}RSRY;s_%&+0WVZI8FRkqcGMzW*hVR*Dd8KL6L`umOsN@mjdP%s^@h>qQ
zy)fcHzD$l^-i&p52(U+f6w*z)wMp}s-*=v6(QACfm-HiKLcHYsC%>3jDV$Onuma&a
zvi?K?C>bduCTf2s*G<=wFiR@u@fo-OSte+@Hi9J~>nU~ea=WT0HYZ}a3T?wteq+hd
z<jQxgpSIvwP5<_FZMS<GmyKle8cU6ZavdopV2nzFCPZtHOs8Nli{}*SByOVQ8Myn8
zYYAurH)BZN*Y2EDaMRc_lE>GIKasbP4(=X+{WIRm!YsR`zBoe=+o@QhdG1m&DFI}(
zR6Ovz{b!ubE{I>u*^U_0BojCk@$Ci@*UZl1X<i5NjAg9A=(xL$3=Zkj5nntFYXLHs
z!)1Wrf9QSRNDz~qJ7)X=m8kQmcxPL~aKC67QcSsKSu(Lfq{siTYX5K~X~A#Q{tv^G
zx#Rr0JaL%^pT7So_x*C+FVhAHL3C-WGDrGx=PgCTo6S-{=!4+?>O+V=#qB;tF^n>n
z_Wmz?VLhFB_z(C<YBNKY6|M4o)6?z#)fEw5_r<gjc*B=yjNXfmLi$#n*#pb(ZFnD9
zSP&0DI-X3g5Ry!Se{y?uHwfaeo0^V+@}RvbcM*)1(&NV9!XfW*$j(muHTsbjcgG8b
zlw#=MSNz#Em@q2MJ_>aN27HhG{0$pY6P4^Eznx&%PQul{Up5ebNU^H|a-TR%bHnPh
zu)(d#0cK3f?tHZXg}`0cnh%tk#8enBE+Pupi$ec1kMEID&4PN3B)a1yaojhGmo+~t
zrY+jAvMu5)$Kb~rgKSFYWd21rwu+D8L_YkuhZ#wFi%R-Mz_ufO*;@MAt@Kt{T#hd&
zL=<#or0~dL?BKdCV3_?NXL_4_hmBPB^A#%GvUT_Nr4lp)MQiW+yVi5+h67ZnaQ+2w
zoWDW{rV<h#k`BLRm(`uLVq*4s7%gL3dXUUwZ$;nU3P><k)aw@flD#^9+h_eQou#rz
z2<vM5_DtMwxK1n4*SJjLIl0?^eA=+0ouqJ4_V}&=|GT~5bM$V!y0$^#&LC`%lGiW0
zGB+P*HaJsr*Dm;{^2uwH04{Z`&MF#U3Jo!TdkSsYxCr(lzufW(%&L1h4xa!2l+5+p
zxDoRv{*$LUJ&R6%&HDH%0R=C(-HInlPiJ8&)w^Vn68SLh24APuHyK@nw>oRntcr5*
zv3Af$rOU@v+=+Mmhm)!lySwBy<=?OwlmZ}$9rI-r-(F6kmz)%q0NWAikF5#KhVt?l
zQpH6O$5hkIKc4F7LEBz+cLF&K(oG!h_CFZ-Dao1MdUiW^P34CIGm;-?-in>NQVG!a
zoJ@iX5?cOG0PIx8mjE|WurQsZmH)!{PI&zj0jxsSxn}l%J;T!pno6B^+k@aTZ2<F1
zRe$jco@@(uW&E!NaZ;mGfyXv>a+b{>#mMx3n}$oKZp7Z`iNt@8MgS4Gm}PW2RGM}&
z9@cnvLNdkaGuBH-U#0a58141Y2k2kBK}c21|E$}h*l4IBXRM0l4MH%k574AO@RcRU
zYqlZ$b(<ROKDHTmBayct(ksjakeTT0g^isE+ydka_vp=E6ux|tPhagOnpq##bkc(?
z{CgG-_;9Mz!-}cxUp2#{#YL*I@fW$IAN#d#b?oGD7|r}v+5FBpU=d-decfzd!vl6F
z6vc5WhsN*6g7N&6#qW(Q{ZTSy^PckT>NkErS@xxzqa{=>G6HH7ksK-rxQKvU`DrP6
zdsFs+a7EGvM2XV$F`9b%WW9m$eXSdqTC1r(>?0Q~6@$2`QAx95{@oAw5n&v*{<6P8
z5k#9|5b)PNc{z0coIP*1HF4)X9cQ&BV)kw(C)}f<f4p&b<SnVILUr@R(FF2wqC;6f
zVCk;NQa>v~|LalmH|4w^AVd9R_b+07Dqlm^)&ev<C~GING^PQb8c6u%yu6KoCB-$^
z?UlG`eGT8$ysNkF7l_9Y?m~<Vd)8e9_j#0RbrUz*?`6N4<7rgs5<~7s6Sg2;3jtS)
z_4%0k`y3cgzueg^CwWQ?>?7w3mhi`H0fp*%8;GD)8`TNcidHND;EZQYV{AkH2o=(l
zZg)h4dAHzpN5@^f+Q9~~8-JX*R={<J*TOkgj-62?5IcdC87YleXbH*~ej?5!qmv-o
zmHxfE9K%Vt0Q_3<MVRKGbUN&>Vv45mDOgo~a2Raw?en5IlLof29se<3<zQ9`zWvZu
z{WLtUp=2?V<<r^+;78jQYByk}C}#4_`*s+HkL`zdqqbSUAKgSHq$;!i2gz?X{L9v*
z8ELbVRN+Y@m%@T%NV{q2jIpxkE%%dB@$J*Ds|aoZf(Wu)>N9d-{XrIRyQZ~>!(jeC
za^`WcRxX?C&VP`a{`A7>JIZ0tqd-5*A84!50zc023A3ONu$O3ofxXfz0RvYe&~R`s
z%^nKmp>5@HI*wdiG7OU_wxs3vfkF{r+PelO6#n!t((k@G+ij0v`t_FefVz}9chS7Q
z$pDl+a@$>Iju^fZd)=+v0v!Zj1Ki)rjP;<^q)k9Sv;M3~si)xkEZqbN&ZW)Rlp@xN
z>f(;#;vUxtxX`C`{s*X1DNmk~x_bI8^sMohkQ2CkiC4@lqa=2yj6X!#6siyoSFbiB
zbJXL4EYEd!Ze~Q==yplUl4_r^gn8kYnj>`^P<2il+X0<)9g##E8W)HQMIRFnQYod7
zy=c+_ER3fStDDqxaXW7L&AlUJ{HTQ2Et{H!+bUlVz`TMw$3N4gHAsC}A*=HtWkZ_6
z;GS0-82fDO$zDxCi<y@0eUa^+L!DVTTDF^h;EHEqc|>F3BX?SNga5j{47Tk1Jfa=8
zxh-7$Bso+zx@0wbU&cr=S~L(PSwQ%eFBh0jfwpuj)UE2jFoQ0#j@BQ_lys1)#*-wa
z0AODFshT0`H})kvY>kIrY;4%K;+=({c?(DOu7s<7?MUS9!zobJt&h>|!P)mcs^|!!
z-!7OXn0@fWcJtYNeP)1$8(gq9<bJ>cQm2<zdF~0Nhy!prcifAUitE8rb`xqskQF?j
zO2R>P7`U9Pd5GE(&7FuLs$yiWbYH^}VAFisQ*h@7W`sz)I@0)H5Ap1>4aWgj&n9=3
zvTB-LXtM~ehw5M(OJV#H9zb3N@NOw&^t7eW1u`6P5jN^?l|Y&K<dOL>2uOS2*(xN=
zm`IW-hPAwbo!vN<kDZun9nGK1`WHd(!5(fNRO&E8C)fg1#44iH?gWDc7Z%0@*otp}
zb^|43x_##4&?-+9qq#xH6&x0yRqOy%QTwC)m?X#isV>=hc&!>UWSbx{@8=0PO=0)}
zk%w5}o5#{oU)((2T0enk66!qoI(hBmo$deV4G+>cJ<U)fG5d?~qGIq!whZ0^Ci?GN
zyDt$XWyxQi`t}26*p6wy%^qF>nLu<S3F!Zou6JuM{*QEBFOpG_^#4oOrlh4JWoOhy
zJIUwdN?ReRQ^cl{T?wFYnq=w=ePnc@9HSe&&na92M?B-<|DO4Oj81f2YS^Wzz{{at
znK!yc^E>)3wKU7!-OO@)Gv-Ci9`nlY0IQPf{&t>7mFQP!4SD7iSma`_=jKfyK9Ekg
z`SYc<GglDfU?eUUd~vaZiGP+ZqEn&=6QYk&JH=3BAR4J2U^}!^;mA~F&KrDck}YoC
z5%t+jt9pD=_p;a<C7xGQX}znzzO11rjyLE*((rJSWX_X5wOsLeTcaL<+tY#J6_^1Y
z$KN*bc%g6uyLV=5h}rVM`NYZh5{JPjPqmU-=kClU@+?7pKVzm+a7N+R*Y_se;D7u$
zm)W!lvl&(R??;R2za5z8V)Ep!-=7r#jfs?1Vt~KZ@!q8u$d^{WF;9|P#YucZ962!L
zgc>2ILvV=?L!`Pv_N#Gmmi)5>k??4^&PK*j{P9K>1uGN|?g=75(k!MoMz}Kz$hk^9
z!f#v1zVDBCr<QJ4{T!2VPkaZr=K}Wl+0>C#=V82vW<j3IrMKr{_pi2evSyh0^j|u2
z`hXGE@OUAlN&DY_7oD|*-45=UtgY}ujw+`&8!Z}qDh{o>NKC+?Y$nmidCSDL!i!G0
z9n`y@BEcn_M8?hBZ_v9jfCGA1b|l`xm+2ICwpzz479hdxM-q+AXr`m)T6ObZWMEYi
zWJ}$)!XiCADw@ybr3yB!Xk?wfUMb2iN&vpH+&K*NlHJbG`=28vQRA7oMVc!@M|jK@
z0Se?5MJy-dYxq;)I~5qfomd-pNF-^fLTty#evz=KP%#+@*r&z&GkO8NwW+Tz5g#Ki
zM4G~jkE1w2FLA>}C{v`n<4_fUVq68gEL+(KG3|>;iw8j7(#3l}fF=%23G)KP-2q}R
zy}{c^koa`AHzQkLk&`y=WZ#WnDm)o=r12XX;lMmY{ucY2ss(Ka-smF;PEhYH5-P@b
zg>rGScyI|&g%nYs)+*|}7dI|Pm@;Ig0%3E=>Fg}3<y{}~f6e~5xF+aMYJqj~`F9rP
zS(Sov2kXu8lRzanhU;bdTGSB>mH=uWYy%PO`a4qtf#&1i;;wv;(5t+bcg5YUP);n&
z3w5Z~|IBZ(L5#%{1>Fn^z5;^+V$wW3{1OTKIw<sWO~;y?M9vufERP^~GR2=ZF#IN^
z-B^Nw8F(-O@$#u07weaA;B6m1&a-OO8`GJW=kl13JxOM}6-@i(1j->qhJXrRN2YAT
z!6j$0DF%D~dVz1HYEOye6G+TCd)3C6M#qqm^yQ$_EhENLM@ppcT*=rCImT)l$G+LG
z*{_@1P$1SR^gHwSc3(eW&?cP6=?jZ<On{Ewc3ftK*YTS&eg^FrMKPmMmM>EXvhhH=
zs{;h-P&&kQUU~xJhN^mf6X09vrIj`_TwI!Y{ULl)R(T6P!=lXLb})U3I97J-!()n}
zCL_|DzJ6Dob%z<eDbQZ5!?_51c7svJ%wwEtgfwxSA#VBtdrHu06^L}V0HOa27#SyR
zjJ7TW*h^>M)5f-)q5t1fu~H+uSf;N#S#6|rv8M03RVBjK%q5EdGC*0~<W#cd)(Lp3
z1`E^S#DHFY^_z4-i-WrVK2E+j5iyO>k^^TISOrTh-Gs4~u#DR^%VG0)cPv_#HV@5o
zc!)3H>-2uAA_R-o%Edfl#k4YS0^x(wHc|Q+Xa?<LbC1`OVW7|0!JjqnQQ-$qCg5
znS?&5fb(60XrZfl5MaFj_j%|zJ1V`Z^U1HF_nwsf*R)?sc)$e^y&m(3NZd~PURU+B
z<P}edk?PH}toXvkKl15N<7Ka-7oigJMS9O=DZL-6X4fUq5#3dF6OMyeT<4y*qqPyB
zDpwKHLKi<}rcC&fMJVERn@)5Y{!2GqHy#=I{p)dr89NW51Yk*)p1@ESdv7C|Qaknc
z4P)7Dmv(!ve@1ATwObkXh|B@nxMS*XRlBOU`M!RVUw5CKM%@8f&|lPBH}o_K)E+iX
z8<ko)bHC#c1z&MUvV>W^5lLncBle=al*&AFYz&wdNvi)G4A6)4jXT?OGgd4Y261>r
zb)lBN`C3o_--)n(pVEaCSO+)UZ38(K?vg)Tio`=seqIaZHK-4WTd~n%j6+7Hg%bE<
z1va9l2DK+MChmzo)2z9~2Z^a+wtwhg;9GKWwU--dpOSZ=-q5}4#7N~E+;rTuOUXlc
zogR(WRp{t?e3cd523ZodK82U|r-bNlE8!WUNX3=_8F<o{N6vaBqH!2)MmE%AnTb|S
z3EWUYyu?|QTr_Ms(opx`;dLjNi~WJdQ+tZB%y{K#;kfLPWW51sCEMiM%}se5>SxJE
zx5VS0MId^-S*CMBY!sbx%D%||P%N^2Is;AMh%>tFFlmsSECYx`<k|In)x#D1`=g##
zR^>1NT_S#6H0_%+tMyrSrd;OylDqND4VoMGv3l)hD+sFIojvTm1%=paC=_IP3M&+n
z>P>Ca{kigmq>OVF_1U&zE_7oy8w8$dNSk;btM1KaIQ6^Sqqr`DD;z@$TPp#{xzDIP
zF_$u-_L~hj0SpW`%O{$fWr@X31*~>S6s(>AYaKnGuL5WI$*mXnl)|e#?rum64B`fD
zbqRJA=tC&`I}v*A>h}KiK{ooD>ZEh|)NBfu<c{_aC)w&G0s3zf#z}n|7|%&i{6_Wq
ziIRld-X7FWvl>=eEUv(fJ))ScnfHF|tn_AMu(M1xy$a#s@-ZI!If}^8hpw{<7y1<d
z8$68Sbr8LUi%<b9H%s)TTE$kEi%7}8EjjDqku5jSqmWRmwU>5F8e%g!=kW?Zp^O}K
zC+Abm1p@`hQ{)>Si=k2R9FmnN$+V83YpbU{lFGv!+GQNoy!NC4?Up&kT!&`PHQ<h*
zCm1ot6DuwUjT|b8<e^GJ?*5FN97Y2`jBOTLU@m7=tgQ$Z<!AES>}-TWBKT0~Q_gL@
zM~VZG!T1u}`sn$YMWOr$pgo}@aDeBCJR5UB!y%ua@YcXVoho7J7=dda9Q*8HbG~57
zH=59dgrH+pg;0X{>eEXw95%LfR@jODfPjt}TzDmwS?tC*hsQAB&r-tzhbc5*abKz`
zLE$u9ke0Y_4*69huqzP3SFfrIx#DcX%V30zF>iFDzdwd&kb8NLS_lR@$tz%qqJYDH
zSEv<P-C)#+xss#~KWenORdso6(d=JxT43cHI1SQ!+_{@{Na)K9u0VZav!J!EQs_mQ
z7LTSwy{T>BvR2g|c*0>VIKL@C&+S>TgW7xdsfdZLs^UzqI$#nGaXrs4`Lc!87x~0b
z@^6Xd=SN_)P5{QQqHblPo~-!0q2Cq~Qf*E;a>0HdEA0=V9fC8-v&EV;L}BJ#oB8(o
zE(e$vj+#tFgPNP-RlIxzJ&C$DYsWXzpXHz-{&-?K(!lNbIK}&JyjfxZa-T(H9Q?EL
z{aXF&`o)S2p-_tV3?>5pULWkR2%NC(MUJXn9ep*laO;1+4FaG(85F4YJPQmqj`I~6
z2^tM#X68u}40H(BK>`(B5)RR4o7kxHnIdcm=`&-x*nZC$m{BK^%|3QATi(Im&iMod
zJQksMNv2u${k&+Ko9H8eZPtBz{)1MfXd+Y~69SjuozkWPW@=_0kb66=LE;-zAU{@Z
zy#jzvM<RF0R-@W>)x>l#UdR&&6=MieT3<DSnA5!?an*QK)HuVAyM@j|O--`(UENoS
zaK`2NTZz2Xr<7%a*SUBbN%q@EDEQjCy=_4byw)wQ%fUKb56>{**t+GD-2Z%j@DLuT
z+V`?Q%EwY)CvNHxp6gif-9_<R+7e^y*3H8%QHGWKfQ8#`RzuVnStzCSqJw6m%R-Gt
zkeF-@u>-V2A;xqzga&*HYG#Lwrz;YvGtJJgbY~FjIo7btZP{L>=@|%}$!LR0$Q{fe
zyKGo};Y2a&iN6d0>Sd?+i2IFE^C$#@*1J@-70+U}4PmZty!X4<bT{zzm<8;n(jeQ<
z{E5M<7z{FN-QWu&j~uyi=8>kFvy)~cUR@7^Z8}Vi(8h{$2i~kQ9vjJ_>b+FU3NxXs
zNBIOR0L1z7K1IR+;%p97sUFBnIpjkZKh}>jFfA#QBF+QQrD;m4j)*fcQ7c_MI1#;0
zlm_D-0`mIa`<kr+BY^drhvHOGl~O+AUeZRKl-o6e+jnoUb!sY}EO$lA1&!>0M5!is
zA!@ckM!6VjmIds(x);~>JGYLM-KQ1&V04t<J~UPRa&BQm$XFHY#dC<-f??;C*t}bS
z*rFhuNQep0nP3SWxwDDw!6Z%rf>z8_|M%1&_yP&aKH%&uh;ZL{&ax4U@ZSqoAT6?D
z_-&o+)ccLuz(k;#avwH2^FHB}cOa-^=7Xfk{FV;vz6|-?RpQ8Jf13;c&qL%TM3!L9
zTtQvW6Il$PIC&c&r(P@;h2`!4*2<HFOAKQxmf!=xTZ!h_zdeK#*KF3Eddj-uPxPiq
z7b}7q(`!O74C$BzBmSHMw)@k@Wd;YsrBkqJm*7Y$jSafcHE7t}S?L4W-pMZdxan@Z
zB0}pqi`ez5&*SoDuyx7#&St6$(jfxWPB#L|&VnQCf>8eXfb>j)l0U8F4_tI0V9wXC
zr~`O|u&dk4B5;4AfK(OBHy%cb8XP|1w`%I}lmCqKyQKWHj#z3%I`2eHcF~4~4^%HS
zO&8f(+Q!xBoZcnm*}C)2<zGRYfF=HO=AO&%jv-Vcx5D7QfB>^qC<%dxh$x@?k0wWL
zeA2T`gR6!;sYQ_zr~8u#o#xsBO#4cvt{z~y4CXq9*U*zo9H1SB&!=ARKZ#1QEM8s|
zc+`|tLmEgcgv86n%Vr{uJQ4nDc6$N1@NX-e1y_wFzPeUPe1Z<QLYSO_kAi)t4v;0l
zW&^Y~^*}tQB~}hyED7Gm3P$TefSj;lIz`lf09=y;t|D8x07@1WF&M`=s85n3+X`TZ
zYzJXLvvpSxVMo8vPSGAJz=h<J4BprlQ1qB&KCE;M86zs9$;DMB`u6cJ$AQW_l^M?T
z8!y;g>V879m+j1j(;R6ZnA95c5Xh9wtE1v<0inQ*%)i0zg%D|>pse$TSyL{qs0N8}
zz6sLoj_c2DJu&W)ZLn-6n_$E^qdH*3SP-*JA&WV{WAe>TX&``{vv#a(7kH8|PRSH&
zA(A`yhSEc@Vix}|j#EX9{AH<^j~CAAF$}<oh(xzGxoyt7)C%G>gRebrL@%o++A4GA
zXUi0X>-?Ez=$S&MMx(V~$x`q^2ZpJ7xS2%}<bB(KJ$3uy?oEOq#MI1%<tHE_Q?$LT
zy9ab&)MRMgFP^$=5mah|jnn;|wQ}yj>i8w0{`;X8Ycg@knE<;18$rT2jPhbN93(Fh
zoe-Q|kG>as07@dTTs@Yz%iqJn!lT!>i2x^Q^A=hL^ptzeU4c87cg(MmbsyKej9c{b
zG5kFS@{Cu}DlWh@R-zG-N)&KO(y^gB?55!kp9PgA@%hra#*1>pi}6y8c`4>{f@6lo
z**OCR{ZVVe_Y$bMMaYZOx){QXwS&=u(n74c-PoJZNGdBYnh^6^6FlZ0MN&=Dd(Qrz
z!u_!3grQsLnAsM`f|+gCRnd~%(ZT8YV<=X(O)B^qJ8-5#SvQ24?gdn@(`^+&zV&SA
z_^O#Kg0jbc|5cI0VKaGRp1VCK>S|}LS|gt{!2Fa`?$hq4x-)7Zpn!tN9K_*iG1};I
zt~dnmAnCN4ea`JZvSL=GGds5skyWoEIgJT`j3aIWGK>#&ixVQIQW}(HaeuT_DMko@
zwl8!1v+^%(H3q5<lnFqYr{Z?vV3Gn6l+c%CB~Nm^q78v~=?yJaA(d&yN-91QgRJXp
z!`Tgq;kXm$;=0(}6cA6ylpK_DPfVal77xYjew$OMM1?X&Fil{zn#8xiYWP(CEZSpr
zzsd6~)*VNKQkAoOk-)U{*tvF&4kO^ymZQ>giuiP8Zs>=1?gx-}rW^tSqg(m5{Nvr*
z$G&q}O4K}S?#^=p&qp?d8$$uDxEH-u%GF7jKCySS)O^iClaR-a6JYAe-l1Of)OGGW
zCeU!+ovcNM`edi{AYMJePvwA_{0grH?vMQ5JNYM}>NB+p{fpj~m<Yo`ZA2%)wtvcG
zP0is`EdQ)PA{GE)L;1l5x>&#sqM=Gt(U>TX5lm}AhcR$YcI}<9*;FcW9ahd{W(8u*
zE_1`Bg}JfD<vwK??@(dFwCq1$nN;CgbD_()7sv{cjZiwfA4bmb$AB>US1);PKDgG@
zw?8mL_M1!n_{L6SnR{ZHgHB%1u;{dz6XBzi+C)9&R~o?Ykg}eZRkc|y4(rgV!NeN9
ztmf29L*y)@@mdK*-p+gB!d^z=4dMh;L=T=t{YCL_!tVpV=6PcdRjJ|I7WhSTH|_P4
z>G=V_ybjd)h@6Vl`9U~}$<TCDc_d}S?|SPvojqO@!Zf>skn-kYc{6d0&uI<g9@u%%
zuGn=yJtaU^=dWcNO3IFMpRKXB@rJUvciRa$1;ce6L(o7`Us*w4It`6mr9~JWJ>69p
zALvE8*M%xN(D2U4oJc~^0VRf6HAE^Vd!Q+p?sAPASBpw8)ZRmhfSG)b+17m6?h*UC
zGs$pof2)N2`esznWhV)bf4>>mlgc+G=pBCa|8GD!e^9)&g<O28NFtMt9=uo^I&Nv@
zulU&0TFBFfj<F=KD#ReWdb3?A#Iw|p=@Ae_`MjTX;Wrlks~zu)0%R(wIvm+N1?$Rb
zyp%Z=_>RHp^GW6I?psJbaSW6Xj#vE0+YNoBJ8XNq#Jh$%zqh=wTsZXmj^T3=jo!u`
zkNW`fm=#}mF<OV>&Y=c-9d>m?h}b~~(S{W`0xB+u{aZq&n8>ET@(s#xU{ZmEe=bG?
zAkimAWorR@y>cwMDAoGX4KJqBiqS<@k}lpP?R-h)s&y)|v9{pLe~PR|0-q)WYgJ^~
z32ylUl$xyeimU=Cjmor`y~=!OrHhr=ssY4?Wot_y(9(+-W4N@cJY*qs#oojib$Jvm
zqKVR_;dG^;hRE8&7rgjtNrG@VQ7i1Z$6#5Czs9PAbu8&2h5pvDhhbu}B)B$<SNsg2
z^@GWCx^paFZBVIazkyrou%kyn&e5o=r>Ylj63c|0{fVJZV)B5|zN$bjk^3b<83QQL
zecad@RO2dK%N{a4Zuj;5F%khL7cV9M^j|xV(ufULW#qHQ_w|vux9+$u60b8!bmt91
zJytlfKMS;`+*i>L^>FerbjqFwd1QI%=ctSuglfe4G8=pRtO01YUrWud8A^RqE!li!
zd?)eg<G6sBx-RVR_Q?*S>^0u<>;d!>;^utl+T$%xD=$+9Qo@IZSEr^a&X?=!&{S0N
z%V@_~6{+%UO{OHcdNbY2us}&vfp|0$P$?3+{N?TqdsHkotl-90T33UVIej{Q%Fc$)
zRnJ?aD_rZsV~M41!tmAuLkSkxmApCXR1&bH7XeB44GOG+%Po1lKfv6B4Fh;&|K_4!
zmJd;c{YRqgD#bTRFljH?+k~NJSy`Xvz)KB{B3GsAU173TMo&HV>21Sv%Sx&_6lxAL
z-dKW8hTh<J4H@M9oA{BSmUbee<rsTw@`IScQ@qKmpD;L}HR&WSyMqAz0R@SNi}c`p
z5{;5JbeiCooYXHq=FK?rXA{5$W^PH7G1!Tp#UoD>-cJX#bLfBf0dKK<sA#=fVf0TR
z0;rz$=J~|i3LJ!h7De3oF9ddTs`G@R9YP6tpxjlT3!XC+z440V?x7q>gaTu|KY9<z
zmtHY;u>6I9#W5sz`;xf~RK888-QU#G+qQm2rk8-TBh+cMMvI&qW*orXL#Znv@g)*{
zLqYl>N_oLBN}D0WByyrbhR7?b^I6a+0-EP~-aY#t;#TvCufy2ejru_Gyi+TI3EK+Q
zeoJYuepsygH60lcsybItL~sKpzl>iVgAs_(ME}~*9Hn_0z*F%@Z69_<cXNE>If%A!
zW`~n?2~)M32a|*;%GCh|%!al2J{GeAkZoxb)P?P|h>*&R1O68<o~*+$`z`k7tmndR
zS4NC=EDxhN*lMd~Znv?5wTJCPR|Qrg%ml0xA=pSWk%<u|=2kJDVortTN|~EGan)j~
z#nM{ETn8}Z407OZIr;%KtQo`hvABvBC|{x=tfN$*I}x44f)Ici?A*I2+t*Gnl6)2F
zTfcdcM^C@ky(%{L-9b8;=%Bt#@gg63lRMmJ(4+y#Uhhzfd6*e!dyekFgNo$yMaio}
z<dA1*ry7;EWnq3KVCmbZ+9Wm){73TKx(VR|Nted}Vd*4v_CtLq*FLE(HNNdrx#bWP
z<-O4IdBNbdX#zlN=P1w`+HZMWi<y_kF+HuL(hH69Viv;`dYuHr7HpuOBo>(6I1IcE
zGRD;LJ2__O#Rlh&u{RBzy0&{5{^-k3eP966ThH7BK#V0@9R;kmKpmmV8e=y5q<Z9&
zn}l>cm`0*$EmI8**<Obtnl96Ax}odHj(1mx<&B{;6$C(^UZosGslsLbcw|PnArQ^h
zl?vdZA(rF1DOC_Gy(K`GK1&!RM|FRkl|`kw?#D;*3{%(r5ogf+4o3ujcg}e&_&6(R
z&bSI;py_47AY^h?WoN0!iUyxsson?h1?Q0&5_jP_q}NoXW|(Y!2i^Mra@RN1=h^6Q
zm(HOM#Rh;#&}K(Lf;g<gc)X34^533Cx8B|@cAYW@QR-DjNl@QjzBa&+WAPxha4_&m
z5^>U9oKlf#*$Uw*?V-X|$HC_G(-k4-D1ioJu1y1R?)}=C{9SP?UKj*WbjQ}Ib^w;f
zlxKG7PatxY8SNI8q2S#Smy>KQU~Eer!jC<bp94(romf3<S4(2J#;#qPp+2Haam$#h
zdwK=WS!?z+cVg!UU;SOiBg5XoUM0`CmSoLGhKKyU5jyZVIViq9Ae-|hu~9z=zZhIW
zaGXMC$E&48rpcT^%w6T2Dk#a19Mu{48=_Hyplt9M=f*O7UWI&-Q6}uU2^xREp2nQ|
z(g2VPY({-aaKHIFRlErMr)XfJi0>B|(s_GV)kc-VIp%M-n0u$9{*KAX#Le~nv)N2b
ztHFS{Y=5G^N-O8)e}+^fqp_8w&%IDrU&HE9MnD@p*4GCnam81W2w+1sYa`SH*d1{8
z6Loc))UhBodHG?U&>h^l+?fQccIC-3-x>gKeNrrrHuGf>EP-Dt*HoA^@X<F03?H%E
z63q17(;*Yq&pa}4uX`%yIp7y!_-miBSZIW$rYygcNUSK@CZD-K<sMg%hP2J<@UN*V
zSZAw}MqXElMO^z!X5!;b`A=M+vT5JpqU}l)qwe_QC$ir4$HkBla@Xb1#0auhDj)$#
z4zg{;s_^>PcyP4b3iuge8^zd=IcTGxCM72smta`$z0AV}Mmzke^O(QSox?Wa5$RDx
z=rl!Yx|M?49<VKp>g`?D7it5z;n!)_flR12x?B4@90iWAc)=z<^A>n$9t=cjxp-+`
zfi{X&(s##R4||YQkyJ7ArCdX)br=D}^>}FY6Fk+Jk9k`7$X3TqOY2(9;c-sK+n|vK
zw2uEaoOpi8Rb&g6biI(paZNzwMHv9O@TJ`Al@h4BVrf-p%7Zxkwu#E7ca)LZlvw*5
zs$^fJf1J1tlIt2uB&h2>U)D9FLfGJm(gRhfVIcnw@^HEFK<h=ip{xuLYU>3^w0Ryy
zjkvU;tWPd2y?OR{_C7e}-4rN)L;vgcs|>N6hdj9Qv-7to1(z~Hs0=`LZ~16gNlDe4
z<^gus(rLWl^S9uzwA_}hI)=YCf)G%_sVOfTVx@>C*_pYN>=n*?q~^n&pM_(@b+nhP
zC5sW{-#W_FiiPge@rEmFVWb0~wBlo`efbM(<M*TK97ia*3-5ys9XZyCsr)njb49-0
zh#AIwASEP~%*)<KuiKPOV|C;3P3bB)O;_%>zYU`P=bR+RDCC+YM|X$b`S<Bql%cb0
zRl*(xyLKmG&dzvGl}*N86fn#$Q_WUrf)a_|pjQH$(;Z+h9#JS2S%EtM{DSJ`@-Ox(
zqkm&g3f9lI61Hv}O78H|x_I#v>UViSA2zx+dWHy;{oJIr2<P;?ml}eX;Fod{s%bz<
zc_^eru#$?Nztt=Q&fHoYj0$n?*+?L#dzk5lbgtt;S+sSA7GHfF18xMcURdU)tyR-@
zm49Nu;IbrkUDB-{4_I9RH_}HDzeSllci=5yPh{B;?0q6l-UgX-Y2`-H@d##d$CI7D
zdI!b%%cI}eKjCX8b<HbTm22A<%iRqMcrE#k<1MenpB0<#7_;$#DZzJy*~%___0<hM
zgsQV>(B{CFQhtBx&cDdfg-w&1u*aBjQrynO?^27XW|#Ag{C#5pG|AXKMbkPDs2P+E
zakPF#)@DM+1qhaOmI9SQZ*2%CV?F!?+(c~Dc68<}$bi}f0f&P=@xIUEo%0zfUVGnk
zR$JCRBkUfc*H%J~{k=u@ikU2hNWdRrzeFCf@DGT%ApOph`F&o_9)cEjy6N+j|EHYn
zn#1rUL&Pev+Y})WP_+7BxG~*J@Wn{{S?74&hQES=C*4HhlcRGft$!*4Y9OB_2bROu
zkJI$ATFt(AXP$*3qcN0gs{8_)8;1_zEGrzJZIYJ#LR?G0_Mh@4_kBFFg<wA}#<25*
zN$sIM5d=|l)YDx>z>+E+eREgd<aXXx(h8){C1d(D!gw0siOvH~5nB}5&}8vhu-T?q
z&HV^F*J?vu$9A+u4<4B8<m&1vwuTQ<T~Rj+&N72LmGy2lg2*&e2DUL6+uAAWpxM?|
z=%b*8R8;MrQNai~3*4eJI1M-O_2T0zGWC;rLY<xIRzQG2GY;LIKvTrOR5LnBZgcf#
z+^l0!E#n^GjcPI3c3@k-uO_RQ@kLpRdj*@+Y#e*nE&Q_hVwE-diCBH;L+XDTr-P{J
zk+~FR20SkgB42JQC;;>n@r^6fRK}5cnfYXTTyb0Ew`l3kKhX`PZ=hsh!>H@J^C3rl
zF}}>@0m6Hf{Y>*wcRMd~7PJ4>rVy-0Hj%$5eGm<3Ji}ae_T?#1C#3i*-ZWQ!S*q<_
zzp8I;jwHTQIw*oaqx39v45~onW-x!fpybM2*j$A`^&sga%{bjZua>w&b8_Ct2wdDe
z4s+genBl~aUlLgB2W@>F3>Mh5bX;6m@JK-f<-y-u{FC3+FW|hG+5y?csM-p>C$#1!
za+eJd8yE~)l+y55-st#6d(PI_(ebDiX~Y+`6JA?bZCV10*+GTLrLC1^J3i2hz1;4u
ziT8j8liuztnd$t}Q6gcsc{ji(2@TfhSR>MHfLMRB1D(<LvhtR!6s8A7yfOwtU7IQN
z$}#zpnF`j{{nmi)Aa6Zf54tP>IZOW3o-qaJ<VACkrq_<b+^7eS0|8AV^Y>q82Oxvh
zS1!Y}K0S)){Y3VF=C6-=*-BDed5F(O=X*xcnJTmPOf>%~ua^zcv$mJtD*jzFnz@va
zCy^VuJREPpRJwiPYzTf@&oHRMP^@LuUX*CGekNAf8KcNWlQbN4vOE+dV&1$bgqa0U
z(&NEc#s=?OJ<F1oA~4aFB^h`?(|9VviXiF^#gZMnFYtJRwOtim&6B3Pt>w0+jgC;e
z)g0BiO5q*Godgxb@(x=nTv;jv6Xk+uFpK0|R$4L%oXdbIS`ZY)XOW$CQjue;d-WF`
z_c2fC9pgHw4QUp<)$&}K=}-EniQ5MFl<cnZuuNt=(iq2pdYbLB7lBSN!LV%4vn;#Y
zLp}Q4q2q9O`RN&eNLCZ;GQZl`2XLFqLCno1wWBD8LiZz4@i0wt{$E6WV{j*Nv-QTd
zZQHhO+qU_SZDV5_8(SOOHaE7Do9DT2y|?;9|7vPxK1|hA<D5Qim&%!1fE^xNA;E3D
zQ|?VOE@XgI_>aK_%41+(u4%sB``rCDBkz15+At6q$z^`b4Qk%}$yYsTt`yKW=0>dX
ztMrg=8u7?3W?ZYFYbxnH0c0W?g7~QB@^%zHcd67a02evZ$PnHquoyX`3ST+~Qty%`
zd2Bsb9TwSa6+oMbxvgx6QYr!OEC{vK=5~O99|z{`i)#-8r)Ah>MdAgEQwDt8oQeh1
zM`N~QAlYK5JliQAYasAjCIkR1qt1Y7PRLvw3v1uI`(b;vBO9d9g<kg&Tdi!r-%=;%
z^9m(YCHPefFVPDz%sNe_tMyAO|J+f%ujTqImYGI9s2VaqbO9&*LYYw0%-{Q90BAMn
z+6e3mbC&DHvhRqQ_Aj49MI@@^XO^!LJ^k)Za6ko);25ng$UkYI5mEqjqdC$rX(4hN
zB&Cz=evdr%2<bQz`ECAY<vb;CA!}QjOO*NX+#t>*yW3{_zO|&<AWl15=gr&4ok0)p
z!Sz<fjhOR^KQnvNcMM3PCcu2fSR?;Cz@C3Z0L1HW-4V10*z=QLXSrj9D={;Up6Y2K
zvUxC;y9Mph^kzTOZa)ARU##$vB{yb_e(0pw!hQD@e@nHnS__mT6)`p9v{aa5sgfer
zo|<b@Rr!sGI79Xj3Q80KhdVcEw11{DUg{#0{t#2Q#WOBIOjn7Xj@LVHgR{@G4SmRz
z|B+-}74RMUVjU>Uqq%lCKjoNHBARe}$MQ$kXLsnSK0Y{_0|!9!0Xenx1-6J@!}|U)
z@^CccO0j1Psy5?xec3SsZ`yL**ST2cMk*Q_Blkp$;yv#4`DU+z(_FC_XAK2shbFnB
zsW-U8T6%={pKLD0Q>@R&uCt&@m=hT3BSh=>rdkJx2SjBpvL=nr*&kdN{pA9CRDrGY
z(<D+7glJJ13IGtU2Vu&KY~F|wbVNc%gVi3{aU7AksEK%$n{LG-%3oL_Y-MMPfbXiK
zJFnNai@dA5d;KW%?TWER1dRoCynWtAftDip@)eJSwiyj9%q@1hj~up3y%L4hm5h2n
zOr9o)M9|qhYOy;*exFOeuDXuvAtMD&5u4TtH>Ie?WPpc`C>ZvKAPVa}Pcu2`HJb>C
z#x$x4u)JCvugIUWM+$o5$oyAG(jE6}%2hIFvJ1+L({)7cAb3h*C~Bc-=3+mB^Wvdd
zvyzxY(BLd#Ex9DALeN{nb+jQSV=2o85U1Be&Mc1$$MbA%5nYO<K4pn`vZ>RNH?OW5
zs5Y1uR{))epjS;5ipxb8eg*i_wM1nJ{G|?N`Hi=-H`TzCj3tj%(T2s1NmTB6m<jZ5
z0#yLJl3UJm&_}b~Lwha6_xuSx9+&7j%7Wo>=>4cB`+HxrTt}sx3f#4w6Sot}a_;Ie
z;S#kgZZDG#jQ(9R>u+KsN)~LMkNcfHkT~iKM}W8Msk`*`;%fmNhV|ToGQh=q(rmt~
z7-r{6hC5FKQZtjb$Z;bjld*<8mvG4Jypn0mC($8Y<i#876~K#K`*wZc`Mt56J$xoV
z;fPFUW2kdKTwYUO;dfvt9K{}<mBQ#_5U`nK9fkZ8Zy7R;jvHRKs(mT4200YL@iP`#
z0Qv7d)2zUQq#_H2x~y;bBS**%(}~@ahX^~<Wcqy-G}g^lL2J%QkOP;%-wajS2UP+;
z8B17dek0BwV0x};=hSUkv%0?!{lIhcZ5Nx18#y+NSpljsJt@!=L0b)YLO+Xj=3gD}
z6GM)Qb}`$4J#@&X80LrcaHr!CUvA0m0qCV_*N$fAci5R#I2tt=u1<*2AS9zKP!=fS
z8}n)y%QH*^m^Q;GQy&7v3|v{CfvBUBkq6EQPBL8FHJvHs;lP|!XXJIj6IjOs@~TNf
zXQAtaUr8Kx;+i}axq0cyNFc-yeP|kRJF00Ae~7>rwFU&JPnN>&)KR}~gOn~30rl+e
zn?8xPVkeUorTSIsqP$07;6u<pP(@1Z-l{g;Yg6DHvPN<+_)^%;;<oegxCxr|Y}URO
z>aBc6<~=%1BeqUcp2ca$FUv2~mXwIYOu;xWc7a$6dT8L~|4LtDXQUptUJR4(Z|6F!
zH)hXLSM^+){FiC44WeTi8Ze@Uf1=6F(+ngW59<YT_`I~%F~uZhmV!b?Cno9cECq6g
z)ELE~mjvOebNf8!gj(Q*+Pv9BqbweHF;1xn>^fL@I~!3YQ=n{GGgrHXs0Q_tH3`;J
zE)WU#-4DK2JMxF<l8O4<DcCt>NJ;3p_VrynP}DZ#Rkql(fP!BC-D$}Vz#KoczSR<0
z8zf>ID5cl?cJqo*;v$0mD1UMn=3D(;k?>umt0yckCGo1*YxM^Pu*|?uvvMf&eoWH-
zT8^mUknR7h-5!X)1`GBAVL!Tw(9?NQ3%PXI!QEIuODf^1`&o5+CR|}aBR2n1&ich+
zwe0ZkOW{aBk7`A_tRP_mu<XirDL?VvRqu3O_IJ|dQhtTgnG}jYGN<<0eSY=K+asq|
zNg!QWV5dn%{_#Ps{U7&O3wWz!j&re7HF@<q)S?Uxk12yIS_vP1;81-pqZIrl$bIDw
zWzdHl%a?e^Ys(zN-QC7O)G)XA<FaUi$lX6O&U3^cwsSXez;eM9fZNrN&8f4BO(Bbd
z2ab@eG11Z^x%$sRn*9oz^a3<_d5Bv#;qAd)2*~lfz(2qkLecXH3=fQqs4%HNMp8hb
z(zh?)qkK#453}=puq=1X=1Dp;^&@d&T8tKa9GJHy$K~(Vhgr0)u0P>Q?3);^s7_7M
z-D|Bk5NTU8&C$0nz}|rBSNH2BWQZp*thOi}Ba&j_yO#k)?3D!<I9(xuH0iU;L6f%6
zhBnsb-Uia9a3Ih4Wt#+@#e!Ltc-vUMC<XoCZAbUDDW8!*wp>uK{(Pm-Cd=?H2si=b
z3nF3hAT0k-MbN&j-NJ5gE6*H9qtK0zdD|2iK{4>I&&vG<K#4~ybXq!)@OHP66p<7{
zGz~O})W|{xSSTrg_)IW?woeBUa$FJBOzs&&T!}F%NGYX#DLfLQh!W=p&9%y+5{(Re
z>JZE$CP$mO2LByzrD(vw=kIb7d?K(0p@}=+cpEOkYcW>L(OL`~zYRO_C(&)2VKo0n
z)9-S4XNX8pz!(I$>7j3HDh~ozsRO`lJn6D<%noh)Ef00{`^DQ?;rw|^M|WAvap{Oz
z)mb1T&IDfOVXa|tUa=X|b&q*rSa`+sxF)(~Ms+;56x+V-??F4CX@{f(i+5GnS#0;`
zN!kOuf+mb5;u)MoAA^90Z5Tx(bs3hP-)2$CxAFuifB|B1OwNY`s^9EPb^y<wl<6+Q
zMH-8oiSj^xlKz~3{)(dxD{f`TaecJe0z0Mq%whYMP6|J*Jr220*<*9{Bp+%7qVlNI
z-=*032h?vih3hU@`+Ea>$%?Ylg7oT>h1hfLwH{vVOiGS>sKEv5%tB%waS-b=)fJ$I
zlA9C>@C?EwHEa^(Se>nZkKO3QZ}iWAi)*Yv3p%LAAmgR{fe4DScyIh>h?cHc>3LQz
zf5Z<LUZPbo>=fluhqYhbLL@;UP)x;s2~XawuSl7srK>jIm0J{*#2FPewNF83+(s#a
z7ei@YZ+=t27fPbc7$8a9FmlX2-z@&u0RMF_*2D-BKTX{zwHf8SRjML!H%-Iqtgy@S
zFcI;}Bf=XIi>xzMjNb4u!#=-zeOls?ZDvo)9T}1>SLr=}@(85xASAsNnhB-C7*0n?
z&R;J7Ai@a9A#WiCm>I(fT`<0-s6&YhE^aS~(&W}ya`Bwm;h<r``?M>sw*l=8cVPHP
z8DU6^`gMvgi4&h8nEgNKB*mly)ZVWD4?2nEx{JMHJHJ4??)Cr`xY#0L?bP~Ot)H+}
zBJ_~b{ave#{alspw~hPL?7hSxQ;1(AG|5H+JNskcbOn*NOp_QVC7w2BPaynL0h&;g
zt^%RvvFbqx@O>l<>YpB?0OmR8U)EXqOAamc{6--xU)qQ08L6)-;qf-ruXMGKH2~9d
z%=_5;>)LJXnS;%d;Wj83V=p?Q%%Y7o3AT)mtP&QLFdt&3+^l7gE1GLkEN(XC*#82o
z<izi$6FE*&-^R496tQ+GgKA<};w{K!zW7=O5#b3hm{+4(%fQ^j0TXYeJIBygz(2*m
z0OqCzUT(l2uXMSbzrxLKcgKe>qJ4hd0Q%cU3q0jpJpK6T2?{i#$I3)2mc_d}wqBKq
zxsJ-KbY2XpX*<VXT-k&kjUaztt1Ya4sgv#qnsAbEC0G+r3vanvl!p*xAj4<+zbg&R
zHDRnZT|Y|Omw{{e0Q}}`L9~R1nFP;I#u!L(3A9O9O=BUZ3R!bzFfha-7BOX`q7i*k
z+n%I}VF@%{L4j083eW|&5t4_k9b>JV=TV(R40_kUlZhkko1Z+cZzi9+-SIQb%}E(O
zcDhgt8hf<)xWtarKbV1P`|lKL*TCpMl2wr*zBbmKiuTdn0A>U)!2N5nO+bz(Z+M5@
z|MW4A|LO+%>Ydg*25xZDYivU<5D>PFSsf<0G`{LZwC-ApJ=nzQ?&vo25J;-DD`9g;
zqpWjo`fl2AM2Wfsk2eYMomSn_)G_2et-Rdq{qU;f(Ab~Cg|+i|;K~(y-4}}Pz?839
z(_O7<mKCzl0Aiw|hdyoB{}7R`$D_hu)AgLiB--L11<;&pDe=P0-NF<A%UF-7KTV5H
zNqz$sY3}c66=Wa3Ba=g7QXTRw&><c)V-W5fWW3k;p0<GQRsmP~x|GrYoh@eagb?{4
z=nBCbzluUp56(;~)&+TrE1YdwvvmQoN++oOm|f0fKuUU6tDh9X!)ezBUT0r}T6S_w
z2x{E!4~i~QH$$cH%$sz0{m=1%=T=QG@FQiFi2n$2yGm21wlu{?q7xe1Hr0HO?MSD|
zbMBAm_H7~y$wCWTdF;LSw3k0_@8j<xxC%Yu9GPAyJTFY1t~K7KcM8U{ep3&Att=5c
z|1DDhfEFoig;48SAy}a;g!C|CLZF<zyS4dSxRjohEp@jUF_bLA_<fNopBDq<_oz|J
zE~AKh|2yxb?Ep&iyinP~I#Z!aR3b{<$^a@;`Lv4p^`7@v=tm}v45wrY*KgVJj6L<a
z(IK5*x5%p{lwcgG=)}wZ@s>2v`)Cb#6RhB<fMKNya0+zNq~MdP4s%gtAl>K4d|htw
zIB!Gl9C?VCqO_hlDO7hsmptE<ML8H|?>u}qh$LJ-*9u2hucIEe_s!ec)_k<==V3qg
z{e?ZrQ?W5w#eDeRrL0u_LJX#&0oEpQHV7T-pWrU6hpB_J2`POow(}fYdkH8)VbUbw
z0JJdT+-{<C2pN<aNtAGLl%EPf=n>iBkwf*{PP@cHzvTO8*zg9Va%^*0b8fLA0@d()
z>G*bxI0orP(L98|{`ByHv*l%LBRgOt<6rpEeD(n<*OGcR<p(XX1h-`z=#wI;gfKs9
z0|!uRA`<_Q89{mwjbkyV>^%cGr}B$C;83#Ax!CIh&9PZzijhvM!dB0m;5Y$U@WieS
zAy_oui^vyL*F<n$6+vu1|0)|x3gw_+A80PL#miOHE!96x|3njL%+QKKjVkap3#3F6
z-^c<Ir@?SAP?>ZGiQUS6mrVeW5=04BRI)3KLNqKAd#9*aU&t%!6;ZDxg<psY*!;jf
zN)EcV8<`g`foOiWGt4yEj^yE{sYYN~ot7-qCspy#Ec+@vZ}ZgAT^&G0Mt*_iX5Q%e
zH5v~_|9&56#J3^TzoE8~INW`}6JfroITS`q%H=vh%zh{snGc&*X;gvjtU#+KVyZPt
zU#|*xYjrY7(V)yGnpcgJw2|TnAgUxR>j$NQr6jR^`h^2a8I7tq3`&y{lwlZphiDXh
zcN1U~45bc11(qyVFAbr?NMhx%8iXYk4(5P6@2t5C{s+7|7frnRCQ_HP>Zl&^5V3Jd
zF-LL*KZQ{T?%r%c3~N#yg<J=&-J+wIP^|@<cw(EiHF#8A#8+AizG{{PVDDx@e68{H
zq?usKHiQpkTz7ZOP5c%S#IN&QF9AB<#(oH0*&cRX#k!(9@871ySF1iwe8oA7%4`pk
znp~8emGGIoGwOBK$i5qM+%QUs5T!IM%NcW}7*C6;McJrmS}-b{ZYY%CBV6@n36`Qo
zG2F^)3~MY4IvmgyYjQdQSV}%1$>k;qm1-HY4BmC$!PlUk3X9;;;k^_TX8illep??@
zx8<QjM{U~0puAHm;*B9oO1kmJ(kn>%uMicQ4#iuvB1X_Dv0wz452&fgFGIQFzWV+w
zNhEec^Y~vWL4}e+7f1pB`zdD+joyS2qraX`pOFhrajF8I!2u4mZ%+#HX;B@q=`fD|
zV%WAsY1VQPCet$Hx67S|KLi!qO$6{~gi5g6yIO0^QW3$&V>D_i=*=3#dqe(vKF#&r
zIij#vRZ@hesyI98^`ZQx5swv?HSInAaJ^HT<=;VpsK1@Uaun6R7FF>ST|BP~^n1#<
zsTbPz@}v`CmjH7Sfh?H3iWPnc{a_{u5Ot0kZl}8~&KjhKcnYf~bS}JE{`o2<yS&G*
z*#mg*ONmxCapu_l5^pP4KIRUB_X^rGULlmAe<}mpy1CgR6kiA$meO9%i8F^8jvvsy
zM*a6BIZu@SK)uf}Mat<-v)3=qk&M}pX;UMo3Jxni;{b5OC*Ea;86w{DRXm|=4>Ixc
ztR7i|fdT{~c1E~HK`yh=L$8%_)OtV`fGBxhYpqt}#$Wm+n-xQB<9~Tt|F3EBG*gVR
zRcJJP9V~F9-Zv&FXPtX%r(25ekfL8^?NG6--)GG&jquHmhpVzFx@`<(Nka^B=wAuL
zJ_9az`qGg{(%Q{!uA<+=8%m$=Qus2l;?v~r^{?q>k`BLV8EJlSU-n{~^s|4XVQuDd
znJJ2Dp9oLaag)+BlQNT)G4-Ty3ukZXFE*<rmz_ce{sezUdX%aj;1a%L>^2x(LJ5wn
zaQ=?^*p`jhSEj@rIN(&yT(q>zeQQDD7Xr4_+`M-dE^5W_yaiNanS$|oXw6}{wO7Ih
z!;8imjzpr<UNp5zbZaz19<2H@J^5@538LAF$lTDOI(=k;)M4L*HtC#p#DNg=m0yhJ
zly008JzL}8E(J`+rOKUQrl-?Me<<>vRy2VSAG9CAdi0W;cNb6*Ux9A}@lO%cEm{^x
zZu!>WMSBv~36ZsMWgJ02@bQ~kN)!Ys8Xpj+7rhp>$!qS*z1gZNZO`&jB$vB>ylU{j
zXIaonk@qyH%+3{*CovO2hbj-#e{^~WHmRM!uXZ^*!J%Kt)`$E6W7ohd{uwU6H{F#N
zzHhcCpNMO|0gs0sU?1|<tFW3672N(fr$I*dsvGoOZPR={W&02J;X$GoI=pnJ_Q55B
zI#RqkNNiWzMv-Lw0I(pQ{B^zt8e5B|cD&!$SO%1uy?oEw5-fkk?^iU#{Yov?J}jVR
zAF(52!$yc3>PUYi!~AFU%+f1T5&lh44Mu|1;ie<N#}~CRezbk)mZuEbr5XvN!MZcw
z)+*f()>-_?|9!JXt1%|<@H~@SJ^MR|3tw61I>g42L=CxM9C@XLf}w>ud5O#ORo0j<
zvnRt3+=D!MZ92l;b&3#A#08hqa+0*l$PGUOSilrI{;IwR!0U6N^gDYP9up=*-{G$9
z{7Cq@O_Q6mdJ=N$ZLfc1P&;K7PT&i=VB&yPo|#}=T5^}{0S40iOv^k{_~EP7y7P7k
z#VuNELL~Hog<p@b;fR2@NLG3j5p)%~qI-(#Qq}qz)L|nJ$t2Imnn(6oN|D^wIjaoZ
zuQ;p-*GnD-aA<pxRPZz~m^wPzas|f?x|Kg-d3=YGCWn>hl)VF}ZQq;Gm7NIdWpYAa
zy%~fl%bmWqdrY3;z~$LoAU<S^msWMynN%n{wBzz}?);7SH1Lp4TqKfk=*?j7vahqy
zKO%lxBzh3LZT;skJTz{!XxQkI)5(pY?V`avq0i9*P(ZSQOHN@rg93C*bUicO3sC}z
z5yDm1Yj6moJ3R@5-#nV>d0VQ1fT=@Ms$C5Oes1Y69fo0Egs$;d9VEqdsfeRY{k_s@
zuGXKd`QF=vBYfjAt<%ZL$zMp@E@@0;KnJ2Qv&*x3GVhr96!(Yb%zR?DS)VdYU6Ggk
z^osThQ2xvVZ-1L7OkZ+<yAxQdTY+=-S8~wJ|Fu?t-w>(yTCZir`p>74Mn+U>q*>hG
zktR78kVV1r9c+Uo5SV1Qs5?5$U;0eANnaJLw$0e6kY(5S$17~>77#pkY2CfuVV%?Z
z%?hZ~Z_>FmC}S9a@sBg#`3y~2ok+)e0e{H?%xu@))Ws^;g5IdwgfFpFc!CR~!W9XA
zl<v{Q@K5qKJ6C9NZBD1KXBT)m3%e#eS{|7<GBn$#b3P=l|BPZRIyLc!{+4-;R8H=0
zb!rb1$Bon!9NMpyIuO4%IV#5#J5ZR4w~(l7)ba6=(8gOzgh{_SE#uUjRj|mBEfF&W
zB$`i0GpnX?9+Kr9bl8u?LC%Ng2+*u-?x1JCg50JO{H16PA}F<KCoUrvd$^xo^cjvd
zh1QW*%@s&~&%QlABzXmD;_hbb{(e530eodr71vEifmtCGT6Q3Ap7Co%^_4kV)w2eg
zTn77fD{+;JQtwLBSUhKIZv;4#KMV2!*k>oxX){!F0>qdkzC<{_gDF(G7chg!PHx03
z69Ht)d{sk#2CHU@{0q)MBN~*zv4UXs6BrgugNkyQ|3H%f9{qoY+zihX0q7hwQR|K|
z$&EkCr9v7{zrxj`T&3X0PM*1q3i%rpd&(u7dB<+7%-I3IwRpTCA2})rgO=U_8V1uP
zsv2Fwea00Oc{p&;n5f?DDBF=1R1V?vx5J`ZQjJ;kKl++Bk6NugT}LK&qAB<;mGUa*
zZE0}LS^3S$){Fg(jM7l=D@h)q-J3KP?bKd_@g{64R$(}u9+M-Gqa1fClngx$cO}i<
z4PLIOq1!h(-8F_WFd;CcTfmn9iP1H!`(2fyhMXyrqOU8DuHRS<!&jEPD+!8kdn1_h
zf4Z*9%sFQXL-qE}Ha^TLu|ddenLwr6u+}G_dE`vjV<h$=w;=`Mx`+hgFmZTAax5aI
zOZDyywG|MnwwhnWdS`bPLEUWZ`<q~^@*=_-3Dg(uqGn=a`PIK1BeCrOTFL(64K2+z
zAh<=jrbBnF>ONApFE#ArtE7!;wV+rMzli8F$Caw_s04%33+%TaU0q5wi#jftOv)vQ
z+S7&adh4n%G{{ntot00zjrf#sxX`##D_CN^tD>-DtABGZcIpM!4pb7zlpL3hR!;5l
zj8@(S2Rf0Rj9HbSx79}gL<G>|rI$cnxlmKZhn|H?8iI5Zi*@EaS%piGxKe3im##-k
zBN0A7Clqz&=>)NOFxUGOB}-)`zrIS`UJX2I+)>2L?h2h7GOF1H(vs`G+Xi_1^a6yD
zLyI{mA43n<UOi9~?~)++z(y<i(brEjni)o`ky)y&+H-)MFT$1qTy(fTUY|J5)ssIu
zh)Zxaf8KCm{xRNFb$vutcNCDR=c$EyA!MrQn}a<=-cjFKA3uD&{j;fZa*C{`rtq_8
zUS4HejpxqXP~&{MY4g;Lw$;S#3C88=Z<S5uRu_C}@}8C(VLpyt`g9a#Mnp-Eq*EN3
zBfOJGJBng^s!dn{2n6vtis+2uSBA)l9+bF$J$?{B>gZ-sKK<xcfmfElmlWpKDhu1W
z837czI6o+v_jX|6syh-GS<@bjA1&#ggGQuz8ry7E*jr0u(}p#jY?MPa3l=cE8r@b(
z$eA7>C~pfyXoln1-6`kv2KcQm;)?Y3elGyO%5%=Ss<>1D1}8FNhUwq07fi$aF~;$i
zHkpFyRC%2&JsWj)a#i$~MXv&Z3;r6RNT(fkQJ;M{SQ=rQqAE?39%QEtz-1$mQAtSo
z=HMtnOBHa>(^yrn+lziZNr{U3al}h^RG{c%$L<`d&57=DK&*KMU*Lv)-fG>Rs%QbN
zo_Al!02u-Umc<vKqdn0ycUfD7JDHp4ipN`aCt9*oF8Fe)gM6QBmgL9oEbd52Dqs5y
z3biAGK#fckpo9vkg(!a&rBwIKz@hDegB>Y9`fx)c)ofZEZw!@lCrdq94*f<w!UKE#
zE2AI+Ecq9A6ln~u{n8?k2nu;YXud17s$2~!x2@|7Ku5=0P;7sGG+i79PPJC;R3~pc
znv9y-^t<O|zy$ROYR9K*@T_{{R^?i0uppNJRCo8H*Q`Y-<-S;T^0T+y{m)%6q}-u0
zEVkUCCGie=)2wJNZQQIVeZTCiXzi|gA%1Uuq^|g}V4NVzD}M5?|K3Hf@N*3qyV&&4
zFrfSA!R)LAZ}Pi^4K(a&lZ3Z@w-B*=a0<Tu^dtH1yc8R4vdBs;27+`NXvflmSn**P
z1@hmS7P+i-1zS1}w(h!H%zt)V01UUCzTYR0vMA#{YmgL1=*9yXZx(Who*P>0g7vY5
zNQp2M&!<4h*N|qSz(xmVtVy&Au8Xh`fTZ<L+Yz60FH=qWxz#~Mt!P&_3sXymb7>6F
zW$V5HG&y^*)`Yg~#nKZnH`2xopJw|u!8XB2`UDL;t9bNNt1|YC=5U&nniEB*f&z1E
zKkj1kg$}`q#>}6^BqG$Q-|?y_kxU9fVW-TnPIO4egG=K}Y=8I5oEx??)<O>+0a-T(
zkZxo7U&S6<K*8^hqw;9jt2FAAX+eky(7**kz39NWzrZMSNG@D*+6i~{W=usaj-`xx
zpJmpmWn{7~f%q45rL3)$sHBV0%-I*GHsz>V^4nv02G{XtJVaSw_u(lu^M}&*5vNK+
zzSx#wne*A{5?mHM_huekH=H8P0LuH;2#Z6P=5|;G7w!uUw%P8w4ilO7Lf^o@&^Y@>
zXNqYNSnCTyKKLcN)5!#AX{Bi;AR-BQhNyJ{_onAh>(hKa?sst?=y|Zu@uoB_n4?aq
z9L4<s?)cfV`H1azy2H>VRSr4_N^u|Ma1y-<8goj8P-06gY6<k$fT|W8fI;x$9)@xB
z=!8@Z8Qb+zWS4fq^0Clk`ErD<E?Uoh6eIbWS~Mh{F!4iVsWRTw3av!e!Ti3juw6n_
zw5p_6DSFLa)4E1%wEH7XMa~Kij2eMM8uDD_|IdIFj7y*u3;7h70a;Xtr{x$0`XSzP
z*<}rJ`Y8#rJLQ0!bl=XNE@BI~`AUv;jT-I8s<+NnE1?TjgUa;7f4Jn&mtEV|E!dB1
z?NY4AlgBxQ;ZxE@O3&B%{%ttQ*s?fX`R3{CP4C&uWfts9_%Awyv^7f08W`WK->eGe
zUl1!9rOb1T?P<vG85lP;JYX)iKkH|w*j+B*+(*9QN?!n~=a0*3=Uty=uRf23Th`y)
zen|R;^Y*sPaxK?i!qOK<GLkA2eJj3C{-FRn`vTJHW%w!!<;VbDBb^E*LW*LoPwzEN
zvj?M=%2b)D@jds_rZK^255%*&><|$eh|97os<|sTXn^(BRk6rY0ylgx3G0b-9wpnF
z4^BF#<trdJQTqS=8s7n)n*HnzZl2|hL?!r0yuLCSJRMn8Xx{8RBH{7Wk%Yy%{BI$U
ziNA<$vAnBTX~Z)w*kpTH_~;OegCm57WLtBuJ@mWF86|}gr3~HAhf85I8^4k<V)w<$
zN%>Ok8rkod@9t<B41NT_k@b(28b>EaCwu^76L!7Udem)P!+lSI#f%z2=Be^})alvL
zXKLxj4Qx<G=u3lx&Yyxy-Jq>Wu{BQKElEw3VN40%P_4+D(uxHVRJ2v7xhW$Oq4*{B
zyvmxS7kqH8CxPo={um|9(np_3<K;L-s^S%0FwlknaWP{%UQ8W|zaC3Xuo9FH0j6au
z&WGVBmKPd!7WVgDuAy?xS1_Wc^Yn99H4;&d!kn*;&brx1eHGktH3!-)gYIut(dXiK
zi2abVT7<+RbG)JVBB&+Y3Vzi&0KQ&Dp;qlmQ7@t)KKi5i!W$YHlF|k2heYi9tr)q_
zI;JUi*7T>$BnQnW*>MEv-`(5(0MRXyDMn9@q-Q6|6IFVAbQ=liNvnNV*>FN_LT9hn
ztx(Z-RhpkkM>In1O6>W+i81M*k~l>rrQKGzON8O<dNFCaaZUa?aP?YQX&YFA+qAzM
zG%ET-!Z4fR)rAHik~hVOX3s+NB%vbn%=ck}X%&4-)isuBYzYGrW)4S>0I#j!dpK$b
zNvaIkht;<uUY+t8n4~}3sSI$^!N?iD3|sxXzm_cxz{nygnd?K5@gd2_NTo2v=fF5i
z^RrR+5eYo$;-$Re%R8_m4Z#uC8Y!u09?UnnG3#l-l?kF39lU<lxYZt^e`$E=J;nN~
zpk71~!?>Zli9taO6qm_K1FS_-?WJ1}4NS)!pycggYL>T=vaQ7py+ajewHBi$hf2iy
zvlz0jZuF@O{nD52v+GbLX*PRl>ci@Tbf#oC2B4{<!d2c@G%Z^5E%I$bDY<2KJ!=!n
zKcOmrZ-mF3-o}2uu(w~6f&@2*!?>bv>+aYC>4cx$jT8cW$l*ov#I$DdO%Z{ZH4C34
zK!5FYdOw5d8WqaYzYV0Y3U>bm?N=KL+M@GPk05G(U)(@m%@LIP$d+ecSbzhe7QO%X
zas%OD{0opw`|um^kzumC_FajT9pbr!i*UUL5f&NAOPf&DzdDApKulhu#~JLgDkXj0
zcmjOTPv>^YCoR`J-c(a|1`e%5GXoA|pnq20+`sK`q>wr|#8gjyNrr`_2;<XBmsuj(
zq}}_Rjg89F{WgEI_=szz-)b{^9J;9E25CZ!5nFX0&0Yau9@o=lwbs}aVpMxNU#g<E
z(pT+etJgtJFR!~w3qf`xDIUs|?9{fph%6RWAYEMcD65e{bPhe-#%ss3<7BpMWyJ7C
zbFnrtE>3yhz|VHfo5bIHVetZ<enhz=LJR;hBUy91%+y0Ln2Ma%HpbOz>m6g5L6P|8
zh<&Hdj)Mh|Q&IfLtoKT^e2M8%1q)KmaJ)kAQO+=(m^e7Ue!A?1euy5w+9bv){kNLH
z6Vc!6Z?g%drh}FmT=X)+{r7h4V?f&>+r-P6|Li0r!%FT4+-gVeNxAA&F*X{lpRMX`
z=a{FXBTHF+z+^X`Z@z;Up-tX{OaTOm+=6kb0x%3_L=TL(8N}{-V;AZMzuI<8*i5l^
zxX&%R4zZLE9K&8OYk7vnl<z7-$CgtpkkTxW$`L;}fG55co1;xJu@Mm!;zFmT88=h9
zGXFxpa6|z@dy;uezb92Wg|$2pio$gRot^&mAZh3RPw5v*Innj>DIB{bqu2D=nAu|T
z3ajRZqqW)0*KC&}B9V-FgJP9DK5J^QF_sVsi6t-vJ(QuduWQW4q<l-}DNC%!DLf_7
zeqHj*lQ{TEZo%W)$RE+?ZXlq+WZ>5|fW&e^p>YeN@~`l;3}bsu1>~XNLf7qPdh)~&
z`c!HNDP)1B((=fXoXz)<^Vwtf4mt$3WW3?<MOP{J^-1pzCBb7HT@T$UGQBM<tSs2(
z$JB7RTI~Nb0=dI|Np{@=(M_EhUaIW{@YRMyD##C3uJ+>R-3nm^07p!@F_RBe((HQK
zhapilG;iiw^p=$Y;4=zymF1ow1t|zy#J`GuYAM2U8%!eVLUBPZ0p*awc$S|h`w;Sa
z5#8Tbk`0AV)Q61!K_9&ZVZoUP1d;)f2$k8C&ZMnxTCmSkph|<VsrZ3w@elBZ+C>hl
zA!kBK0YO^s2Huev0CRtoS^rHoI^$gF<NWXIqr=V`P*JkmAH)2vz0cZiY6|Zob_IBs
zR>ffU(jc<5i8UI|G39qqG08^aVwainJ+!sEtHboM1Vn4#`|vdaf^-~+C9JNLXo-^f
z#26W726&B(-@f+|k*<<R1p~Kpk{DX2NxAzg<#)w;M~MNO0817jXOi|-p)12?O@+j%
zS8n>#x`;48;X!1Zv6zD5<SE}?9-eRi6>AQnBW2&8G@9;M<do@!kd_88>32>ncCAw5
zs>+|{blas>QMn9c<+$gxXV88VGwAelP+FvHxG^57`A@va*l>GQBb=HS7;3F=MwIg`
zsL5M)ssQA`0f>BlZAX40_{vS!p)xfTN?%wC$+@QE>)e6dP6RzP!Mhb?K)zi)!$a{#
z91x-o5#iv@;!sQ>v2ByBKn)y=3$#s%2+eWmC-AZVO3bXLx*Sf{RC;#yXDG#z-Fs?L
zC>QeKNx~VNRgtmR@-`!izJBTpQaXVIjp5$W7zgz7{p(U4sCOAkwzUQwPcStbr5T(K
zMTF`num8p97sI1qjF0IN=PDIO|Kqshi4VmGDgUEUtnO^G$yI1{t7XCBbaHuLwVn>f
zbQTS=2x5Tbhd{DMhRYw8F5PkSci*Ui{HWb>;DDOpvsJuX&8|uwG{#0j@@9A7VBk6%
zFr?vowm@{Nv?MEeQ{H(qm-Yr%!cp@)Qk7J=sQV-a$Jrg(eI}5B7)D;m=@klP6ag9t
zDhnyX3A#(DA-rl&Z@7-NzcV6GnnPr*ACK+YO481B2|Jc!?|hy7?>#rWKsZFfkQ%4q
zKn@MHblE&Pcm7NHgjE6dO%je%>nBhH>X^{hz(V89e&Vz_kUh1JS>Szup=F#({RMsV
zQT<?ne7g=0?lBh$qA(5~=JE$@Q@tEfAXWi;BZDZkC3U5;rM*PAQX;Q%u~h+jKq)vW
zN0JZC`AN-YuInzowqAGH0zHrca?{?GcJ4xiE3FAHt5h2LTPT^6qeUM4zY#M4mW5_;
zMGEq!=??2#=i93qxkQ-B2-$N=&R_1~wuuNZy_xHplSHF-(+fs=qLytxpmigHhwh{m
zqawK=M!+L2;by|#a;NwdLT~Prd|Wg!qgW+t-7cu}p@bdh8<}81q^p?A`j^u`SDVkh
zkTK_OMlq1A#V2UkIvDHh#pG-N+g%7Fs*TjHrR7iKhf3bmAa8Y-!^{Kt@Y-+xdNjX#
zGz*i0f|TTi=K$hLjuV^d-h47&I~1IT_nNrf;@?YwpE4O=dzFpI1TAbui}o#{QfC=n
z<vB{%F$P!`)j3IJ-QD$=uZ`-4t;m<)$Msa-*GHf%KUmKfyTsQ~#W4e52297r4d!9x
zY$Ze>F|Xyeq=6)MAp>kyh9KUwdeN)HlnPz$-)Xt0F4`u|O7TAL292yZn!ns*g&fv#
zBwowId3VKj$%iQ0GhNkTSzU%ZM|9}1P?7T5d{WQfU{sX52m3iwVjFyi=JI=Vb4(Gj
zD=riqzqrkXpEkj%Fwg}+iM<PRKIxsc9HkhlzC(ny_7SP}5pA@cC;ut=YW9JUYy98X
zv7I)(QB<Az^N5Ib2N9?axmWRh=pA6#<c7c9ucRt1@To3`h=^F4AiC^=0au+dW0J4&
zM+u=R)9W4V!Yvf?5efSt2uu)DYk}BPprckyEUW{)ZL6RYVk{PLxV`;$nY|qJ7d}B)
zCqDrg3_&_In6xC;c>0JVb7LE6T)oFc<%TjV;e8+8vTt;h&Tg$>czXGHndLRosBC6?
z6e+fw{T{PE1eU!mWbaQ)jFRKRypI-mw6FpPY{dS7gceja9cFG)ece9Z9InK5_FCd`
zThl3?+50E(Z`)76zZJh#m*LWLD+m@00C9<K?!$Y+j@s8pyF$%gPWn~qQP4@qM4oak
zZ1WuCe6M_|g3V$(N@>~!OPk=UP`hh31;fRH4F3FEM&F%72(?eZQP?2Y9XZbYsIx0Q
zElBT)LFc*l7XK9gnVv`ARcl?ZznQ}eL7>{pj@SO_cj-0&O@zY=i2tO2YVMZc5HKid
z#W3dj&0%PwMD>WW9~oB`c*U#OdGqT7Ca|wGb?j5`aI4QG&uNq1r(Bg&_hVJj$2`7y
za$CTw5UU8t5FGyya1zPp4#&P0#4;-`7*j7|l;oK@@wrtSb#-_&FR-xU3rh!ptKGVH
zhS*o%mEi@H(|yASDDUzi2vmKt^WyrhG}nYL?L3L|-;;fL*FNEWoB^D<?&8idK9KrB
zy@a>q`!#sG96Tb-S?EHn!(_l=`9s^q95f<r5^MkR1@nTr-u+t$$l`@b0YsWJF1i`K
zu+U$kOR8|N4H`1{&ET1xHZ@Zuo8n22@v&7?(7pkhHKZriTc{b@iF{Y(K&36Swy;tP
zQ;8IGGNmj<3@`W?(`ah3A|^m`*zO_Ti+;*$Bwxf^Z<xi!zu5NhbK$2vvHU@Vj8Bl5
zWiDlfh_4_hfmXh}wYc~*VS(Nltt_omIOMB9hhvQdxW<{oFs<4_s7DkINUEF~)+9*y
zf{y@z2X6pd?sN1u-b7>PhwF*wNV`0{BOQV}$rdeRVup}ux%s$140EAJOpPW4?s_<<
z`%VI!Q@E46tv~!rF2&4l`Ohv8FT*+h*qnTiNCa^Np-eGSE}q&&>2T(x!)Jz?iErh0
ziH8SGT$ja1g@H^Tk#k<teP0Hb%9|Md`V9ctTGVlL=i0iYOiAuRj6}}@tQwxGo8r<M
z;d4pDvoEzOo?B#4C441!`M>J2f2ZaMYB`^vBeJd;ki+i{KV@@N24$Rf9jW&9mMBl=
zzuxWUiSwN_n*mFsbvo<Wi*#i@PKDjte7msEkhQ6N@83Ok4t6(G<v`uz$eQ_E{HXvT
z3Gu&>XhnRAF|(2pqC!7s64e+})fixR5i3*@R{Se?B28b5y3xzED6cue$bVS7N|qK`
z{VmcMF|>ZoDhjS|7<J~|s?GN-YZGYtR?4Pc;Sbz*49dBtas7;<a778o<w^1oYT#+E
zieTo)E*h#De6fmaB)l)OUTCM7Jb)&cW=E{FO(3^Nc|sNcAJDI`(-CL>QjCcp#lhwp
z3)<evYQ}wpxn*n}AiQ=}06hU2*9|eU-=;9@`jtV=)fE_v+x{;i+tlfRs@g>uEUPmz
zSE4ln#Z@ckE$Xh%YZCol^kb*15qHNy{%e7%&c#f**EnrQ01w)AU@u@Z9-wzu+zG;T
z8GA4K$IDeS3l(S;WsQp9L!sz5yHmctxcO^ot8%uaD%iZgs-9*JDb`(4iLu3DdI2NE
zybieoZTZJ(;1+{wK7@OGoLjWCxC~O<DOW_d@YesfD@H#*Kc^?2(a7RKayGs*Md|2$
zAMf5;Z>k)HG>3ZF{?!1ieUQ~p0GpyI={MlU+?4iMK4p@wZc?gH7wU}~m7q`T>jgEa
z?y|j;01XfEuqWXwtC^r<;6NGg_-nu1WT2oPL`ny6=>JQ2e@^T8lsZ7M&`V{rQKcW6
zXgk2Nz_|HqX6c;2_J4{#GQ%BNe2~+@4JvTqpa8%|QOPf$VY>@$L{MyrAl`(+b?u{k
zO>xW%aB!W;qU2waBN8K!#W}yMjB+CtIr>{oJZP9fA&D80!H1!(ntx?k2sA&LdFj&?
zT3;(2g(`$35WfCaAB}zL46b6WDJd5A7xU+Plp#uqLxP74monCr)bkZ+P$L$9f+k6x
zME(T`Ak8F`(jiB6QSXlYID>K2`|}_Z=v%;^gP-y{&V07B_R7`v-+X6*c^n%_w#f-G
zO3y)VR9;&dUhL=INq!VTS=MvW*gZ0~kH?4%u`S-w3I7}}T_+@I(*-f3J1p#vt7MO_
zl$x7F9st^N<gUo(Mi~g$6?BxHj9@X6#|OxM)Q@126me3mxw1aX=Vr7#9)`+$UmrPt
z6CNTupL&Xpkla-X6W%mR&dr4&3Q4FedKVU1NsvfLpm1l+A6g~_Y-UI`dH9Q&-fqp|
zqJ$%YpX!Sf+CfVGIKwVr%=5PN;1i!cHE~L!lv&Goai2qt)EBfrp-(9A6Uf1&fL|b4
z=N_P8Vt-HO65_*rQQNbT+B29M0jv_q_`pzd-jFePARs@kL@pC@wzv$sBEcZhfT;#%
zFThF<GA^|*@W-{+2-!=CM}TSyA`DK~2l#_NkmT!b!7}_*1qqDp&yud<F!^%tXoY`g
zMJu@sNe{sXU{$VG<7w-E_IO#RMIBj7Ngo4!4kiyV#)Ep4aLAD;tj02db-ZA6QYiQ%
zsa1xDAH4-KV`_{aW0J;c*jX7%GPuGH*Ma#9j{~j5KGHkRE>=NcT!GU5Ste<=duq>j
zy2I``Fc+z06)@^PK4s^sp5pJ1GrfHZY)9SJrT5eRJtBs!GA3Ae#}8nplc{!ivdZ@W
z7xj42YvA3cD_wH3pmkN$r-Y%2H9pFx-z)t6rQdDYy537Iy|6H&f_2(}MG;fQrzphW
z!rN8W0z&<opm(_P!UEsa1K%?US)G`>>}1B7ZrpODe>T5b>Ev;U?oO%rfvZ4wWeT!#
zxB24+HgF4xFNJ@l07zzw1($9&JoWW-dkdmZoR0FfWd7RMhZePH;*N#ExcW<g8~z5V
z`5l7?nU56Kf*|m;We{P7<S#1@XOtH*s7}n8nI9OBJmgro$8JP3Vhek+u0KDo)UdW|
zEaatlk!)e>Rc^l8eeg|^nG!;MP_hvLe3-d5M>AF9tDd<w3RrE|^6H|*@_<Bx*o;D1
zC|30Jy(Z>J2&hqB?JHGqvmO7Ek$n@z<({{lTB2L%-?Fg_C^37+){2AIgp6YN#$Isl
z@XPX>@4O1wf)Jtl_9<z*bACL~YG)4q4dU)xF1v>+DF1TXEgIL#SxI=J&*7Kvxh+QW
zr{xL-0Tz`C8UQTfv&=;<7%wP#!AFd%an9@N=ICM0z35sI+a{=MdK@iV7aMh8T;k*&
zP#z%~`tz22WA+PA5V2Dv{IkH3VH9IT+%upN8WiJYsW2}YV%ezt7B7^#CPf}M@a1F-
z*YWzLYX5HAp<$~lLwbh>f?@Fazf%zAMV_F)n}#?)0T*Ri0k`Zi%#a2dt8Q*zA&^lE
z7hj%|MlZ{&ZiI+xMe^(ttQJ{A`w6tIFfdN#4Gf?q)Ea~u!3`qXrkp>zqLF$lqsdSE
z@8ju*P1RHnv$|b^r(@?U9V>D>ciDbam~S_wzB2rXw-#sY68^{7m?0aetAD`m{q=k=
zoT}m80p8sLjH}<BU$Ge0nCRcaq5a7NuKj|m;S5w`CMAkgr~)jJ+_N2srM3$NkCJNu
z{7lhhIePc@545BgJg?#>7_Y((II2L9OC<}98vTpbbeIy}u{7u8oLwmkT&PH}9v}7z
ze>I<`huZYKjEQ@_AU_O*U}vHFbNC(8x3s@506aZw1hEKS6$>{N7a~Vw@Ju=d0OSC+
z^iYB8Y5kd_Uj|*{^4X2Y8>80M#uM(yMR)E|sD8ggR1Nuz#xWDS3`BnYQgIrcAdy<1
zxhkmfOqQ<wufE^M3obKq{=yd<W>+p+8b(Sbi}YX=TpA%=yK}!GDbPpyCg}uo%Ruiz
z0s6aS!!WjF5@};x0~4~I7|Ms$NYAz2*0~h;rE5c8KS@*c>^V8ssD51iGn`-`FOzHX
zF_aq*7}g>qtik4*e5L#lPoc3#`K-_0tjae*g*euqz^{B;eaqQmN^08i**zG9znvn2
z<H2JMjjjU&O2dZEw0}|<Db*tw^AL8P046kt8YrGp2>c*ok3p7|MltQAOvh~%WRp;x
zQA_|CPOmq?eY!wj!pJz>_~cZ^RP&>9^|iF(P2O!Y%H$|!1}#`ya!7onM0`<OW=MRe
zM0{P(p4PQbg`T~<mMW+0rX-m7&G1c*Sn?mr!H;clL4E}InF^Q+xYF6$v^4+-CEYfA
zo0>EcpL-*4F{#>d{u%xQvzC^^8TCl<PRykw{o8WKu?O;phOra)4^8nr*0OwYUlYit
zvFhHm2We!BT7b~AnJC$*?jt!85HFp?N#zuYNofw+hI%)?bRaq1tiR=Yg9L9zJ4e&B
z2>IKsTgO{O1S<tai>^msE|xYxubU!7??q{D87UzGE&v&;hZHV=@l`n1Wadt_O-nq%
z5&2pBS^m@n%K0+eoKUQms4{5GJH8%pmbFO|kN#3>W%#YU#Obc3q9W2+20qJj7ilJ^
zptr@gOOLwE@;Jasb2!v$4951c`9m5*;E$O1DQNu~O9s<ZMO!_Bdm9Cqv!xf+Kzk_I
zszK8}`ZQy#WpLGxwvS42<8HIwsNv8R$qHo#F@u?nQ-h4?U(CttUW*2FuPzl)JsF>@
zkRr3zYj0b+M-;S<iIKJ%yZ3WM#(4B2V&L=r%4zL?593@VdB3S(yl|%VEZwe%jQDo)
z5Lk}2PiuVs*$V7#xm*I6q2I7-d*mDr(|yBb$2@tFnw}N^0rIf!P{sn_AMuGR=1V2P
zNc??Ep_w<bKGfbHZmXSRK1xslbktf^&yBfisa4Bb5;(Cj81cznZ$?w+><1W@MoAz#
z)u(2_bK!`ww<bHEX3LDoTC20)67*v#$P^9EnzxFQuG4kSq$2?7VUcK24cH9wlq}J{
zv>%a|09`&!(I`Cp^%a9&<cMvm_jpeI7(0E?(;!V(%MoUxaSUNXHfIOR<KXWX`-INW
zqk}&_jdb`+Sag?5fuj}acfb;o@{PInJoNDi_?Glet{#&#A#^*F@RbtH;-;BpH+X6T
zZp)i>q*zLVM+*Soou0CRzb#P)rFRo7&6d-)&2`uqm;W@BdFnXOvCF%(q(!TS|Neai
zvuc(iSMWCZ1HyZ9wwft0xS38I<;T-&nNY>B)S|r@l{;x$K*)94R6GBtu%}&z+fEKF
zRfSS?NV+uiJ~QU)N#}9@pH7jVn7@ObAJJv>{f^W@Z(SzfqlX(IH$aeD-%m>_ziL2N
z$YgS5`3P-{J@cX0dg+_+@t&rI{RHDtz0-TCZ4zzB{;9+pYQlM$aA*3w#=J@?PY*uQ
zu`_gaVbHu}uKe9YF?<t^3U3AbwMC#b;=?L;@&lpP?88vGMaI#O?1`xxBFMAx53f}V
zRHW-g+&2zc%v#l>_j#q;_SeG0rrQF>9+-eWS~W4*)14W34^YqXpC73JoAgBlAa+36
z@+EROzp@U?XaaZKav6N*HNTQ|<1;(G#;h8ZALK}tRU8$Jb%F=7&9IGWiZB)=F~qw$
zUGytFTajCtze-1J@^z!irMvy%9h`HGY~b;n(!Hfl$ts6KxPR;70KQ2@T+7Q<Sh)kU
zngHDDO{R4MHFMAKkiRBRC0jI0ha8|t)0+J?m#y<P^x811TGiOq{9eUy#{2u?au)md
zcG%)F&%f~Ys>ZjufrC#M`>|4W9-1_19d6VfHby(;naK2@eJk4uvt6eKS4U1amME=d
z1O$G3arQEVAJQL|rAX^@@?<#a#92*GJ!0U1yVo@~o4|*6*qzf9l7nIQ0T$rtzZAga
zqG+y_!KMckKDlrAd`ctG)ilz2TNiemWP{uFdsmFL$dJ{qm=9R^DJez%9|EAhc(L;#
z7hS@i2AQyRm{x`4M7u2<S4bgyK5>?G3&mG8qRvBs^eTGGtT?28=t_$0nZgRysOBQ6
z&1Zda4<8xAHAfS%$Nk0+K;of>ZP>s*{A+Ii5K`&+qm{ceo5FDE*Gfj1O_@21?)$GD
zT84$}+kX|u_AXw|2yh7}1V$dX6Jftm_JGN5X{ryMqp0fuv2Lr3toLS+OF@30W*y^M
zMk+P$!rZphS@zt<d&;0KDoVHRp->gVT34;P!z1^VX&0nUBTjx4fKl$9Jg$Tch;h-G
z#>fNfS{1?uIomWl+DckYjB$TugtZPXF@QA7#8T15*@-686rRj7CbuajIO#`h_%Y~O
zZm0-Ooc}F!Q#1tVaD`RuI|%32&H8fMnHpih<f6rQmAEbPzM2zu7EnZar9Cs2FjXDr
z3@Dk7;*+i+ug$pypyW@Hqztk7&TBLYk$OL@p{4>XrS`}YL8dBDUimN5i;mhU!Y9^z
z@-yYBUj|gqvGQc@8rP2*zU)|~b~coU^TR|Bw<_(v?oAemNeFZ}3&%%C5Z&oVNK7oy
zzHuMBq2sa)Q$r6D`@H<E=Z^{U%OkV|DIVZP@ey~j#>r6DfDMx&xOs}-`7NABFwFl0
zqd;809>hUA2eEE72#6;2R-LrZugFE?>iVKhs_cs{E*iC~G>NVmk(E{~kb3p1N}3nd
zZ;h2xJmlu)H)MTmPGQ2`1$P$@j~h*0!DJFSrD!6_BzV5tw%Ee+-LBn_Xx>sLKB*k?
z!_CAl<)@a?n@%<Z7Eyn&+|D`UGS5?w+{zBQI1W<M1p&<%p@-tc7323-Tqq~1rK2Mm
z{Zgll?8~YUsYW2(WSjVuJdoa1=-JR$#-kLJu_ReutccgS1?RzY;U!mCL_(}iyh|d=
z<wcq?F0AHDVuGSxZU5T1_$-_Q+>t9Xr<jiYLcrV0P)D--?wni66$Tt^Jv<}K3#8>j
zP?_}*8_F0&@7bq=`f*uC^wwZASrR*|ngmjq66JJJB8s;-Ycal=)Ef*+3h~B_b_5Ks
z^FoE)F+U=5Hr6_>D0dfrny_DI+A(+w`FFQn6amfwfBq`pLed8UWRC!H8ewk{XI$Sy
zpFA864l6u+5^Z(~$%~$44>h-zjkU>@Z=rnSliqa)(0!>h(Cq9`dtUZ}sTJ&@XYR_e
z&pkuyvyYh~1Tlngx7!tblMCb><chEppX=F*M7ok!3SGpc8&vCpzvti7wz#KfWLuDy
zVr%9Df4hzxw?_gvn&UEs*#IPzyeZ#^mkL>Me7s7?iWe3LSroC!u~yPL?mn^B;s{7|
z7$!^|Ma598zmTU^@EEHyh#oH0d3a~Xrc5vsjhOVw6I~E613BtkoavSck9pFZI0I{J
zbcs#9_e3_k%|5%K3}y+0nQDeVm4#XI<i;Qxf0T~P2EHVkqI~N$nALmJE4o9*j;OaV
zNuTC7EjBO*0t)+mo@EWC6@pP2b8kI=(4KV}606b6j$s-Ipi}~=91Wl{0%%_W6pZuf
z`2Ld!u`<a_j6e*A0qqrJco-1I`@DV;!$A+Kwl-cei&#B6uExmXS{%wu`VDnC8V5(U
ze}gR*oZiY*Ii*_d+Ijo3b$Qij*Zv>}aT8P%!dTE@y8>>D1AgsxrbuZAtdf9L5>S_3
zNDxI7s31#V_4B>5jL%j+h_8NL_BR}E!tu(QgO|;AXvfV?i=XdVd2``{bf7%4$S2@&
z6lw0aDtlzhXVM}A?UC(m)PZJ|3qZ>5f3Zh2UY?!<+PhI8-?SIlBRkfRl6$T5!1v6+
z6AM*^!^3j@KpixzN@%9gbTo%oCa1OyaR6syml40K18eG_#E7Eb>;cUb&xEUl+uaIN
zgRDo2yWZN{h=RLbbot;)RQX`?e<VwO0Q<=ulLe!@a(DJEQF!4tKU+oUhFh|4e{=8k
zE;U`9oiqy2b)Nl=HHlY;lq4epYGF;{C9{->R%x$E{L1pfQ_&>h)Yp{aF-CoUj7b53
zvHbANwuj}1CE6QyW}}uC<BJLiqhTEl>u6Zt;`Seql#z=g*>Hi*XjoUxu>L07)H?zO
z;y~e9-d7PiUqNHV<grltSmZUWf06a~g7r(K$^nua$&HeJOOn0|eG7GGWN0r){x_2!
zR}EJUR}BU9h6MDz^(-383IR>BdJnD@cpqxdMv)gCWuJ71=80z5RAah2yagcy+(*2E
zN%P}z<?wjwv=b{cb8#_jN{!N4Sww}P8ke+_c#09P^veo<+v;)YtjdYBf0E<~Kh7cG
z;Ux|NDLA-oXJl^_642BK!hJ#{p2O>LvC=q8L2S>AtdWcwu|9RliE#R4*$>1rjTc@o
ziN^;r`7t{i2?st~&q;aULqwnTtiZ~vDsWD{7F83VLuKRwxulc~TzBMr{`u#=?heNH
z1twL02Gi#wgkk<H%1@Mkf7TsmOr4+0hYMQ>t0JQ!b%D2^B9^Fw3F`uiogzYm-iVaZ
z97pe)z9U>XyZe1F&}j0pBYeC(Xl&-<Po$K~dlgyApFcYt&7gPen9{MSOYer7Io6D<
zMjggruHJzvWf|B|{y%%);+!_J?ESCIlGHhHmtY<SKep^D3^;LpLkZ5cowJ)ta$N%q
zpc5f+C1IPBZ~yl@J(56@B_ZB3Vp3aPtOuT+{`F&edb%@El5j-mud%XIEZ0k1`qtoB
zt7Jg5x8)iE9RYtotd576-T+eB=iIVBZvv-K6hV<5u0W_xorjB7ad89gYnVXW9_?Z$
zP*4{sd^)8^dsrSVMYuk(R7GRQ_~8|2p$8mivVl4i%Whc0#KJdy;OW8WWU_5K{bFuv
zf6BGc?i9@%wCRP9YdPRo<`+{P=F~YF;qjO<)N8w`m41J;`^RKoTxG|K6GMHN={)ij
z=2+)hLO{e#C~B;m3hQCw`FrlQ$DZCBkgVIis{Q5TM^7^=FnsKD9+}F2z}aA{9I7}A
z@H^YA?Ya-L&mleq44(Aq6({@Yd8n8_iBb3oSihj&h^A+9zlyWQ+o8{xR*ybZDpcbi
z8{rrF*o}YI|2uE1czV+u;xP=JzJq>4?s-uA_&P-MbHCbkTja|n&ui|CFMf?a0hevp
zflUmIr8s}UbJq^^d;W~Wy2XD5SmUM2=Evgq;2J#aeb+MdpUL0$5a#(3m(R(*gFXE>
zb1r8l+6Gk~5vyBp4YBj>w(r>^ShdiQ7^LuuE$x3g*!-T(@&#BX=IqtyAz=l1>@NxA
zT}rinNTz%%N_~u}dfh*c<Vo8-M}Qoo?%OKjp*yGq#-4$(*uTsCm9bsB%vbus?e(3x
zx`pd;zFx2avjo^70xaAam%uo9V1z8bBq|Ob6`&*FKryv2)Y51@E_#?tRP1gU1OWKA
zu0VeT$U(sV`7f8jd)xk${7yN)?Yf+-N#d;So#$E}0;P&ppk4wGtq|ot$F7hOS72}}
zp)!*UMns$Kf0!0jZ46%?vvsi(*LSuR+){tP{(7V<<Hwu!wH;9Y>#=?N(D=YE<k5@m
z+Qhzk|K+p)Rk&cUz(2K`H}y9CY=4dWtGj<Dk647eYjmK<P7e;>B0>L!%At7{KC#;^
zfVPs4BGrgAFg^oAdJ5L0c{wEWvZIvZl`P9&Gehy0t@oo8Fih}iqf{%O@hEtXG25f{
z*gQUFwl|I?S7((*X-n06*JH__A8zjdpTJDH+?cZ?9UbZD9Gs(LPR4E?lJs_@w<CYO
zoefH#v+9@L4nHihy&Z9xBE22y?MQEDWp78(`qJAuOmGBp?(H)Itjka_NfFbu=MzZB
zmkY=Daps`RU0~<h{g=;+onaYg+AV5i@GrgYa2~zNJd+-B`Yhkd_!H0xS}Gn`AiV+M
z!A+LFk@SrYz&DbiLdg{A8%f_t`bK}8`bM7UAveB}3?c3gAw0CPl<eMh#CSw7^d?W-
z?n<tPxPrhsp24>2Dvfm|*}^CM(EX=Yz3n-7AK~4(g@ei!`q?d+6O!G(zzY<~se>aT
zbPk<sM0@xCD#%fgW=6cA&g$i?Tvodtx`j{uRvQ=-lk}h>;H+4&r{d5MKM{YVSa2K4
z7CD7~(FbCXMYWG}6mJ@QM~B#dqC>ivesD`5*?*j1=Z{K#xv~us&}62%V9{x%Dsi;O
zc#^MRQBT1mU1QD}7v%sXIQ9#U_-s;eo4(FHruTQZUq16ui%8;TzqoO|-&b1YTOMVY
zH=Oyg1)%r~>>mhF_`}3ySh#=N$7~P27qM9s$V<H^^&Y3Cwj?U{iwZtVZAnz@%pUSs
zYD)%icE^7nN^MDg^R(#=YazMOeORcF8{MMqa<0Q}emZHw&X#PX$#dl3$=h8Mb|9B)
zpqa}>a+ULa*YF8hsa3bHK)%~<CSl`3+3SzjeR)p0S4%@I4e|PhnBjj=T4EtAvC_|a
z4XtBPAY-wV`vqp&b9q^NZC9Q3e8JzY!4q>f<0=R*W5PY%<5=@HCp{GX1CPdkpsdS~
z9(gLT=c#qKOAR<ru*EgI-JKY47rOsFQ?;u<o0%N?|0wJmpevvA{{7S}C|Y07=Duaj
zsR>_~`N6*APxni5xnF-Jn|-qo|KH$dq{{zVvy(J73*8|$sHG)_PkN+3CjBw#kI6l<
z(01!w>wHny*OzZV>aaFN?+rX2B0335ci#R~9}|t<?NKSE(&*fzpJo%7M<A$ur0M=I
zfEP&*O^L-`e<n<u>l$)iM06^tzl`27UQsST5sh^lrdr;%^0t2+L7mCK``W~WY%``}
zP-*{KK^_X(B~q!$z)!l;qh`Smi*p8I1{-V)CUCwTP})=aed==3U;2hVLM7CvV1=z#
z6J;#}YgEV3yWJ^WrdiPa*S-Zng+%+VZWyF%sZ;tr=2*p3JFkD9xNo-%wh;+e@6h`%
zHz*538(P-bJU@Rr8YtFq(xLWjbYv;LDfNj=a^yHnM~Z1uLwQs_u9kBJ)YchMp6|qH
z-0Mq>i)8VZ#q={aG+zwCbK!TX;8M-}a5u*eA^{8>L^uJcOqSZ(IXd#xb$NeR|NP;m
zftsy<-85S0%YCbfZriAi{=8{7Z~weU_h0WHTAyqi`(1x~xU0ACKiBVC?Z#JhdyO7G
zw9v<!cIzJ9(Er-?#>36O>JRigZrb*}*85hwi5_mz=eyRwTI~nVH#b|K={>l1_2oUv
z-nUx9IzY>+>OIoA5`8(!`CD}Ruhw1T_EU>;mU5e-#NlKfV1w%ITgve}7fD%acd|Kj
zeS1f-aQ}Z$Z#P;we?rn8=w3OsMk>)1D^X@a{hCtgL-;MLm<4U3s#$md9K4xc&ASiU
zzhu5^-?fbQfnc<5FPM?>2A=b7*5Yibax6K^4`)l&#&x*N`tC5!I+r-*?B=5ZCuf!F
zS%pK+M#czdyDlt-w=gibRRY%xcNk=(Kfwma%PfBw=A8N0bht`P)HCq2wlv0lq^AM+
zn^0+7LIC%i&@1j+^m`%7WakmQyu8dSLqJ%I;sKT_CI?_ho#(=>7TkQ{2iU?jvrefK
z)(m%EsmAV&3mdeGhP#&Xge`mf&el#VA}_{gK14Mn3i5c!RZpF}#q;J1^FM4O<8Fu`
zy4HVYCt3aaLZ>>xcvehSz|`JHgtsixL%2gd8Xi{HqC-c{nw&K`YjW1oHfzr!nM5ev
z=-NVQ(+-RGkPl)w%kmywccae4Li*1<>X0FDg<aF_bP@1=Bt8Ve#Y%Wk`dRcFa<`m7
zFBCv8<dPIFb{gp><-7lJQFfA2y!*IT{*HfSzPQYHcZ_{pc7O0O#Kl=e!NGGr=@I8q
zqNo5GDz)QErNKQ7?Ju9MT6g!cq_j>;m9uM}WT1I-|M_G6Ypa<84xswoUenlCVnAeG
zmOv99sCE`k1n-Pc{HfZ!bK(+`q*5^Lb7y<jF@%Mo#bQPtx=S=~>h1ds`v@QQt%-j&
zlW-ZF`m18S#N~!of?bJ#vo8H|eivmcq=d%7_N-nj6_0r$QYNq-Fv6Ao-0ZQVsFps?
zBx;gXES;T}159sSpHynaP^SAH=RebZgR$rG$?<h<TLl7EVS3zgA+`}=YrHruR%@27
zYr*^*?g+i~XA%WyxN23euYzieKN^3qKj1GWBVI+e1Y|^^Z{QK}_c8KN!t1a>qEj&Y
zdGr-splLx>M&HUm;s`2U7aPrp+`ugowCZYs_4cC{&a9i)tU?JCJb}6tDPUUc0?rw(
zp9aT+*a{7sEn9ShhPVfX1iA)+^i_;ldn=|SSROSiA7qi2Yz~<^vWC>K_UV7uuzCVe
z#70Ak26YI!T^2x(wYTL~>7;tf(X?Df>OpZ>dpU131^o~y6Dg~DIx$(%h91ExdRDBV
z!!^%|3C3tBM&kCYa!W<TuKc*;VK}sN1#}BlSS*saDKD@}N`P8LO9E0q*ZD*4+|d<F
z;_~Q8!Oq&;?wSX`_jJl2sP}*FBjzZ<l1<>%+49y1Cd3(qAD?_L$^DV&z`8tu*lMd*
zuW}Yyijp*b1{tmd-`X+s1uJ12L8W@7e8trs0QEjb*lN5o(R&^uRPS8S+?fGi(^Y*C
zKj-{Ot%DfRppr*DWuRDe$x)^s4Sig+46LYBr;8@C!b$0coZ&6cFsy%~^RO19+K_dp
z5H3mS(k=iTHEL=0H;eGMuwO2X{w9Z*ZlljHz%K#HqG4EiBhnj@-U!Fuh;P70VL_&=
zR=Hja4>HMOcl>a2Gvz*_@C!88tJQLoC%<$hHNqd~W}6mqli#?mxz$a08Md<<78!Ks
zb-w)$^^!Wz;e&HWlvaP)E9yLsm<Bs^aq<zDvd9k^5gIGLomZn=t2Rz}k|5i11jo$L
z02l2Aij0;(UV+qKLg+6yvB1+43zOLF0%r@CBN`s#hI@8hu2fn)4Og()WjSLU>(m8#
zl7eW@mj%X_KU};z?8VX8{*qI<RkTLu5DF!9(*;83B){^6l&^m&!iq;2Ifu`siaSI_
zuR_-$l<>}uG;kyS+&X{yebKpfL8g6`bL;#uV|H%cJ+>abI|2mom-P^ch97Nvb41{U
zHs=x*uAxokKhUF^2^)M~FR&u{%{#x{TJ@QZin~8EN!VSvuAH^lQS~6Z{dX9<ol7ZX
zH#eMBv1#2Atw(<h())}D<jNrUCwnV`=XA1F5b~hq*&;<7?OMZ>6C+{;xH_k_?80D&
zj^WTr;Re_d>>vJnVnS+3l;gB3Vpcs0%is$%yk(T^n~a_9zeqt$nIJx;Mi3b#ixHgi
zzNey8B43Hwff7$U9r0$5<VVrLmSU&J^<s^CDnRny{)K-gCh4Ie^;>7h&Xpd~ETxN8
zC^<0Cm}kY}huCD*;r}Ycm%nTkE5}=^53$;j=0EFq?VI*{!SHu^4n+VoLRqBuk#5lS
zh0{8Z#iPv`&!C|ZJDdIX1)0QRJYtU*&^3J;RIAdbJhc>E%hM+aa$AT~26unzPsHe}
z`ZMb!{=a_{V#0Ig(Mh>hIXyl*dLClyT~B|<e5H3->%C*j8ZJE#iS}-Ue^OQX4!?s0
zf_D=WezCI&4ql(HZXNfi*H<;>41p}O*K|^OyIIe*Jj%=h$SR%&!mfKkhb;KFJ!{Vn
z6rxvU|5!V<WQG5vtH%NJ*J?Dxq+ijn6-X&NTvLB=NP%sChZ(zmjH(;Cg&nSW#4~tZ
zaH@bgQ+88GG{P@X7Ka09%RQT|*)fV@fUc$Vrcw2A_QLbZaglBKNuF<>=kvk>1w~UF
zXLeQqt6=mhQs&dl|C#UVm3%l>D+Pav8TD6W=o{~@?*zibg3%{X6960z6vdwLlID>1
zq5^;C0E@RTg{kjkv**T(fR3;=gaVzW?FU;r>JaJ;s{}qmC^Ce2aGfU_Q={NS!~a>U
z(X2s6WYnR9Vl`_6g2P@_>=LO1UGLd(OJ2IMOY6Hu>nqutX7)1Ozy!VI=tVpj7-Rrt
zk>MDWUxR$hOfx_7r~_v;e!Uck_T;=oy)%E6Xaj4QWo3hN;G&gszf{GvVoOF>t}54O
zTW#mG4I?rz^vT#Qm@&Khx!!Jl6z(uih=B}J_AQSXSHG2h{K(C`@mKsVOMVN$@1$N*
z90(%sl>s2=55%^ZvP?nb;x$A!n_;sX%(M;Q=D1udHTcf}$}mjC(88+{$WV@F<{W>H
zm+~3U;y6}sRIls7>^4E9As&kkWjt}uhMj1j%cB77=4ym#5vD)>&ZFxuZR&4~6fI8`
zOTid9ol@k&WBF-^SrWuEwQY#UKh1r6eTnU2Q=il^G8tI5QlR)nK|&+7i<Yunw8%w^
zT(n4jZ+(3d3y|XR*>SM(RVsX4hcbVTkdPK%cLPf-2|Ee9%`UO<i(F!@JMAz_f-bSj
zt}%29M8CdI@$}d?^ihE=wD1cLdE*ixyF|#VMR?g+g!^&fUy;)b2al6fpC+8cq2iu4
z>Cm~)&dx%8vky$o&ql?W|8nEximN*2;XaJB!b(61@9YzlEtkwG6U`^1M)ZFNwEvz(
zF&)`R0ggOG9e{v~zy<342k}-;jyEoeDyKkDq;&8ABv(4(m~nPiZXMSHfGQVjrP7wl
z3fNSwdh8bj9cqM|b@Li`E!~JMaPAPt9~~={OBIY!vE^$eU-rfq2z!a9($ZPV<tTBG
z;0)r$C$oz(NWdpd?IXxoRkMGyvNl=jLufruG^z6TC6CUiQMPZyDVB2O<g9td6M+~`
zSC5FHbVcif;SJz`S0vWF3)fK-AWr|qcK8k`HJ_&23Mrda<sMnc%}OKrPzL9Tx+*qJ
zl!3-ar1#iU%&tK!l0g~g-2`Fd<9e;ps)Q+j|DBYd;^qk5cJf6RyFGtIo!z9^z3jN)
zKR={}qy47+(7J2aKcdEMvlZXc!KRWUUbXJ-Ztu`d8)fpSKhe5SY|QBgoB6JZ&#}wv
zFZVaMZFGA*ugm5OzM^`&=?#vddZTfB*Su-JM-R8&T5i6#*m=mno9x=nJ!;=RP$b;l
zwHglv8xkMt_wZ#`E%blsws~`X(`p7MmZJ&V!j^9Jy3%UR5>=!k-mA7UEq&~EfTkk;
zUAqYUgKHT#g^p@*rU@b+Q~Hp{lywUWJ!3<IECG7Q{To!d)SJzl2P%f&HK+FD{6%<y
z3J~i!J+S^jrPmQfZe}#5vixP{9$J`WfcA4|YS>&v*I5NIs5gJA^()R?Q@Siu3&<_5
zlVJ3IHP9%RuIU<jD<o~!MZLH{iiVgx<N}B1aj%C=@<a@*q7gAo%4{g`X+0AHt#&r9
zA!X7FM<THDk@w@u_31nx)5Bi>ox=)4*`)KX4&uv5$FZ9q$}XWKp6!e}D=*J_8*<>Y
znHMlp?t$CcX(@mABjoe>H^U_OCHMsaznzyj7Xdd@(Q}0*D!QPbWhk--=#|1)H&;$m
zl;{@#{g-bJ4d>oWTpQ>3Wi6ab4>knn{N{5yypK-9eTkEZZGwL_!t$PooRfbvzaVTj
zs5Ck$Hah8bxuT9C2Nogo3y$`gUwHch@JDZ)=!-cqSoD8oh@S{+n_@ldNbmfcbSagJ
z#x9xE*xBt=SSiDoM9ptFpU6cWaRl;C&QTX@P$d!~X!A4=Y!(#FA_G77nFgl_@Z1fk
z_uPE%_*t=rh!}=$<QWjUx<N5F);0P^y3jGn4$i$GO!XRj*)3S<!e+6W4QSLwK2M06
z^>9RH7wv!4kiI^G<c)sVn#Y%a5R^84Xj%5iv;Zs-ZL(eOQcn3cMwt><eWzpI?B3TV
z9WH*z6}PpZ$X=6-6p8CKDgF4;SIx@l)mebGXLMP0b-ZsX{4F7f_U`>vGUfLh^5mZ{
zO+3fJzSJ@@{<x$=8vj(p##&P-X$xvRi6LDv4z+)juMJVL>RG+?3e;r|{R5B2e|Vfh
zX{n?n?TQMv5RleNJPDK5ieGD$ZlQ`kgej8di&!$+wm7}KlU|0@fV-z6b`wB)8Pdy;
zGA3ngfzu&e=?B@g+L^dz+~@^JfJt-)xoh@skP1M>vZWI^Mfoaq68IVYibWB4)i|Dn
zOsIcYu+TjgQ*<I$Zp~X(kKsf8UoG_Tp@r`2pIRtM+*laGsY_|^+XHjC$g9~2V;i?0
zzkF)b|52sVevm~gwrW+kIN;^i9=w;Ykd+`D*^-qYcHzv+?5ORD!19$K*5YKjb}Bi!
z3r+^KSG8C>Y4X%v{1R+Sf<S>Y@DT9%?=XMxd8m|XWXC>>u;(D0I7F26h_wf_z6_;T
zfpf&X1XlO7S!>q9t!}GwU1OR|GHvhL<Mo-mI5XTR9@BLMPimxxEnMguctrdg<4a=F
zA<T<O=iw+!3T%u8!en4GR&rHbu6{)q@~q}oM)_?P>)NTX$P&QY6Bakx2_>Ssx~hL0
zx%D)m|4z!8>(Xy)6%U=9$g9%lSebwxiBi)ro#k5Xw0g?ZbY^=>zn`5JRArR2L;OX4
z=XaHyGyc5p%=A^mXueRKM|6ma7T9#;eO?oRLu!7h`FF1QIi1jOf$<pxjwc*!l~ATO
zA^^>?xwQ=ej!Y+Vw!u%U^=6qTMe%>wS2V10Jg<;I0TxzH?4~C+L7tS1c=5)yKv8~=
zAh7L?dgZiiZ*Pe1XgZ{XK+k_C<<N8K=P_Z^i_2iEadd%RwNDP^vwxG$Rp&nK;${9x
z_2lf>&wQ@;^fp*$(;=M`j51{ur;7qS^f8u`sgCwE-ym@2p#GVqQ@9Ql^Cy2X^6a8*
z*xg$_-CB(CxTy%ic!+KF%v-84&K*83-*{`sIy={%y<p!KBJ4ap5zv3xS=ZV;Qi(RO
zhV+(Z=+E94;<JEHTXZVw0nj)XIhr5ng}GSu3&7|E?y%qZtRlyUVygtv^YNwk<Jip+
zaJxg&{aN_x=q79yRDHk-ICy_cTz}0`o7UB}Ct_F#a@eQVkU|V{-M`WwmTqVEqpVvk
z^eb4n{zO%g-beo-hQ9e@HKJuj3sN(ERXa!MOao*PgA8?YEMiKYr<>=D-#x_~t9UAT
zoo-%_iP6`M5$Pf8S?Q8<)K?7CVwJq7h`u!;Cfn7%D6;y{n&+wtd)<HeStV&*Q>O?c
zlk!L<_CpA(s{TyB8@_J%?+N{4to6|od|Db<hk9~`lMw;8Vv0d?okIt@tS}-N4&*t9
z6%!nuxo%dAXACw${Q2)N{5hAR@F(Nk8(?NsKPh&QmuMC-3I^#XNhIfPV~FQ$mXXry
zsj*pEywB&>#akZC+!}u;|4w++QEF>83Vt{zwe?=c_Ni5GOI@8#UCj~G5}d(wkmg>p
z4NR+CtJde`1U;Pe-?2Bi=2%U2V89`rlVL{c?)boX9J~3U4A!u(j<>{-Rjb#^9@WDu
zS@FB>Vz4I3qI^lE61Hm|uT)agfmJG%-67ZZYE}K&<I>tkKsbLelM>Mq(JLZ)s0@aF
zNCyvI?FZ7*^OKNXe#w$Bk|bRWQPHbn)4C&Cj~Jx)84t*nLGVxZ)<HTSc2cTzL8QJ;
znG3=Xu%#m+fxWyVvKYAKa`mKKSqHeKa{0J*07+G;0B#(+IRdWhX$<0M3GCJIN((bA
z;`1c7Dyz~DsZM{B`<1Xgm!m;=XXLk8E&ti3a_n(fvH1$An5Y=ClogYC9w;6bA6fEj
z0!JW@kM2e32&Bdl5JYTejsS!m5*wQ^33nxdDYS*pI@)is*n}QRSW0Z1eP1ayB8ZJ3
z{<H7w-M2V`1aYG0dLSL%^rfwj-g%FA$MWf_RBcs*eT{!a=93`lF0OTOql0OH-eRj3
z#~k;fv}&obYJ$k3->Ugs!_=!&N@I$&sk4WeKoeDJ<>tv;r+So*{|^5N*=s2sj(GTn
zbt3F3RMfPt|Fl*T4%*O64_iB%vn3AcoD4Idaf8~2qfx$%)r=?AdVNb3Q&t=9SAM5*
zrPgaO%Z-0Zwf3sUm{52Qp@w(9y#0!PU(joYKNfgHcmHMkYhX*z5h@k^6_(fc2RzhE
ze_(t50F>@o&L;?UzXKONto7eXxvLjCDO~Zps~ZOCT4+Ri3f7QrAXf6iA}{`8D_r3;
z!n8q68l0d;OnSh%yxQKj-dW9%D9nuL1N7Sk%4mOjCdr|$rMT!3J^oTnQ^LiCoe?9q
z@)orMuGDIcW3CEJIL+&9&|Xl`(*Kt;c~s6**C+QsuPPpKH1rj-CwS#Vf?|?V-}8pS
z1e@`)dP9i#3Ds`Y%gt?uuU8v*65Yqoa90ZE*qABRE9I*OPg8IE=?yW<0;Uz~yn`l>
zddh!5vCswhSUp1@7cB!TD%I!0Nv&v8e<5dh%QFmJ4)RX?f+9TAPtsCr{lMFTRj~}b
zZDp8|uzj7_!TX3IS{yHH{KY&EG2*#?X7izxTTev}SRA`Kl0>DS#o9FM&oj+II1TJ!
zk{EEcSUYK6uY+sPtVmq*j_WvfbA)Rf(5!zU#_LQ*KicyN(&pChmBBU0JDUSJe<1^G
zn#w@Sj<IQxqNC{sgMx-f#9foqU%UE5bFmV*_&X})@@Oy^tGZP_KaZ7;QklvJPt*^H
zK6*?b&>Rgp%njrnWd1-}5<}@x7ydg*)P>wE-2~I1^P^xui&%tQ#M<nDHzsgwS15n3
zBx#H-l>!?DM<^WO@t95@6{(rU$HaWJ`=W}rUlI<?mk@oy`<OE?nSeZrI{VWF+S+r}
zj}wu|eC4=tMk3wIKUOYP+#xDT&pr>IP!8_F@;lv^xx8<1U@hUsZsQUH`xK#*!n-(`
z%6|y856l&Y&q;hp34)Y?fb}+kJ?MXqin~99eorU{8l`H*)7{)#=Gu2ht|Ru!S<HIu
z?$rD4G|UaEC(uibh?NynsaC2rLQk9>uwrs1g?ISct)SzVFIV+i{j?d*bpBv&BO1th
z%ZQ7jOei)blFfSYs#!}eo4&`HY&w^uZ1TZ$)|@@9vbl}U^M{glz+b2%7ukRGZu=wN
zWa?LZrbMMxtQ<EvqT=$;`d#~`{a&!izdT2pjx<7gNbe)v=n*5EgS0!uLS`oFkS?B>
z1a<Tmd;93^3o?m?tRb-<U#vKq)#FmN9#D+YOKAV&GQWchYr$C0(mG5hkE5HluLS|m
z;o#Y@^C;sM@EEIj`YPxy3MYROB+koh*rWm{vtf5<X?9YeS^`%BchkUqmBy{=&r4--
zH*2NJabq1=JKqomYuQPTCs@a^n<H2!%;KhCPm;tfF)lIAAIA5Pp3eJ}EX-LKMt!|l
z2@=D%t(PjtMb=wU>Lsa{)}#zF=Fx?tfxs9i*=mybmiXpRImhRDLyCViOF4#Y-ON||
z!O72jRmCMqnx2#ZDFK2|wn_q`gtvsZAe60;TDk3ruS7Rj=oTa58=s%&l*WA`hNX0|
z>IQ=Is9317AL#datALyo0x1MS3c)PrEe`KqsIv<@*`1b=Vu|*hP?pOjOGD#blo#Zt
z38a+vm98fBlMgXzT5x~GuzM?%xrtw^2|#orURah}K_dffDRdTO+L#3XL8Xl?bw>A4
zfXiS0x!sYCWuZ7aYybQ%ofPdey~<!ne|@*l#d9a|tkTibA-dOy_U`@F-Z8PU(KAf|
z>(B)0cA@aNx;{Q_B|W{wwK}lpbeZ(=RgvM$nHW?6DxB@ZSw?@GllQaB0y`U~=%)S9
zx@*@zqQ-5rg<5xaw|D5KjWT)EpJ-i6(KRQTDf3;6GWLQplTm!rZrrx-<J_3majkL8
z%?ZmSEWLkdwHseiRa{iAyqbvC!qw|at2K+H9JUtT;((WR&rC}nyXliu#eCN;0{`Gz
zX4ZZOa+cyAb7FtVidoPms+xVqhL)pm&^zwmpvtY@Y~DQF+_t}KPTXq#BD~<HYYwbG
zP>H^LpMh_HuQ_-A3j5mVbFhG|dZSvu;_BRP9(A#*7SIK{u7A95B^kY6wi@Nqb)~u;
zk~X*~MQ>mSLD3L{Rt{M@>JWrGDuIvCn%N{z#K0;V5z~Lf0|G%k>zNRZzPEqp@37Ve
z+jkl^wKiB08KtXoqZJz7S{@mt%CS%eX(9KW$>D6cfCkgLBU+Cbr1zO>h%1BOpX@CO
zgmvP}`vJmQ;qX<$M~%jDwH2(+b5^Xht{B^b{P#b8ko<QhhfDseWB?EtVU22n5z!J<
zvPnAtR#tzl`vJhp!v#Jo{RL&i`5JlxSH=c(!5r$(;3>1!fJh{IXo<+G_(j(MBS<cy
zj{lGv@!77ud$B|MI=vi4HIu;V$!X;zT!uM>DW#~;S2U~^;5bC8*csNOc|U5`S(-_^
zyC)=U`n2s}qV$NF?JFvr4PuHgy7WBE*Xh;qRhfUS$i<yRH;H}$xx6bptd56pn$Zqs
zEvQMyt$!IA!lJ6YYbj6IA}qB`(G*L;s^g4upbDi2x-+CFh&ad<Cs*v=K*V8%RcC0}
zDH;Ta7j{ZMYcg0TXHCu;2Wpz<!0;5AU91w1jz-xo)->H4V(8J|V71|SYWduTL&pS{
z><53hoO<jO0#Jwon`6?okYb{)J{qe^mwN6!gefBc4~k|@6iAaVpnF36Rh}dOL<B%o
zLOgjHtC%j{Aam|~Q5(C`Nwv168sqDsduUl>^Ze*&pjg96r$8ZnWGTJr6ERG3<b+R-
z6w{m#^Qdxmd@2E*T0padZF&b`V=!ShPVax)cJFce*rg*^w23}3v$+LWzi!khSGmTz
z4eP=T=mp&v92ulf45D?(kxlfR<igMzsS^091-?lL!m6n+0CwxNSS}X@QR0SaO?rxc
z)EyFRnL|P>^Qfchoud(i)Quyk5clp$F(=fi`46*U@5+TzVche#+Drm;$HX0%OU-|z
z{lX;r_ka=rI#p^tY)p}*bbp#Fa{BoAxY7viVd&pUm7HE<WWMojXN!1|DDy{H?5^>A
zuaA9r>`WB3mnHO@vSUn*d=|>QE@-^vOb3T7Wjmpf`6=SD(jh2(gST|<-_6bC=~`P=
z@z{i#cNz3M3LG0rG!@+TVacO^3JHJ82ytARGv#`<baKrjXMRUY|5dr7K%a&Hh0WvY
zakUsUnP-)9vCJcdGp!l2(jzZp2y0!SZnl_5MJk1=GE(TMFOaDlR(4vTztgL*XuPnO
zist#b9VsJ^ny!)MrIcSR*k^`k^TjG<JoZ&(oSo8Xpc7zEhkp-O6j8oRqJe+>oZ{EC
z#GlvVmj^4@G=6yyM~kz9;BVL~ShTIuvgUe^WyIc~p3b%=Sgk7v^1ZynFN%pkw@>Yl
zI{mwv#lir2`T{H0H+x*0j5>5wdLNCjqS;GF_A=Z|nx>SZ;MmFlgNboQBZ?S1hjGEy
z1%-gW(I!tS4$80{x8QL8mq~y2<`+e8RXo)v*4r`Gi26uZ<9K9#`&Rmqpc#~mHFBj)
zwh?~cjbd5LluiqF=(GUFuQ1k41x%$sfGRp8&Jk1du%k^(w)f|c7Ne}LsZ+Wj7LH+p
zvgxv*c{XFImqu6AsQu;RN9wKt8>Y}y&-Xx2>9ajd3-t|sgfg1Ga(RE16scGrCsOXO
zDSe2_P+0tSVY5FnRwbm4V#g&cy5umtYY|qoS09uSEK2ByQjr7D@A(ATA%Wg$bH-9M
zZD&7=&4L4z>`2k<NM#;rxL`n}jzw{E=p8EM&V7&KmxxE<IGNO>hXktiN<>zR_0oJ3
zKq``u$>EUaOp!P$gj#=6x3(CzEJNu`s9Wr8RNCd>r*At@Eym2{a=*f^+X$Lys$yD;
zzhn7oGlz~U4gm!H-6>!IICw!Z3C_)FaPS<C6Xb4Ho1B`8-pfznss-Hm%AZ(V{Um$^
z5mndMUmcfI^sh>Ee6io#_Q$WNU5ESeDvNW&9Wbd2)URM@UZa0dZlo_N3+3*_nOGLg
zvQX}>W?B}?-A18Yxh~;qWGtAmqU}ggKvb}e)^*q&=lOc0q|_CBWWQ%U35<E9%{z6v
zMq_Gkd}<xY0VUeN8ro5eZ|FB<FLv3lxgYfY{Mkh(1^L~1gKY-qKmn>0Duv?A+S)E=
zVTAv!8;>Uat6hIM8$MR)WcN^K>VGW46XlQKaKseQO&|4f*IIPEd8<jgYcX<jSQeFu
zK6DDo#&F6O@RomuT^@@nR?(8NAKyYIaUGN^nSyOl4mrz9&eU2&6zKKItf4JrOG&&+
zymE}!u`a>7X|RgI`~}&1`{BQ65xAB2Gr5ArTKk!vcn*K0vKfyoFVczXh^iS`G)knw
zugbza<gAL#D1DGFMg@+qN<}QaMxKa>2-iM&mp&;INM`5W+7wFJl(H$BY|fW?sT54l
z(LpJlf`x<E-6;*Lkc8g2RVNHG(x0HM4eDakZgJC@4B)h4UwI+RSB{Wq*)Li)uNybr
z>x&iUr<#8#J0LnNfHNi;zzNw*9!*1*S^P~?%IzR!12nvmoY-ZIIgeiGjC-?FUJ3@)
zc;s8@2fb%AB>Xm&FF084JGbQA&7mW7TI6C2SZeVd+5EVc>2HvI=#83xu>H@v@e^8R
zI89lj<Ois&SF7cwKz;l6s0-He*YrMf!5m`=uGoM6Wq04l3J%C$kE+45Lmv$6&~7aM
zT7r^>3Nmp%m}BxKCw4=G<M;-gtf|KbbDZMv+hIPKb65IchZ%Q~XF58*c}#cc<dVre
zGstcrxtsM|%cIP0RQXy&@K;A~EplXawZoD@wz!3Wch)&80%s0H6FX6b%|21~4?G(G
z;go+qypUOQhw1uI3EYj;$#o0VAZ5q?i?FGuXt--BPpspu6VyXkH3;raQFlnXKiPF?
zhb%>MX=Nu9S!E)t&@4<vB5NMmhqjTw@P+$>s}b|JH?etYJY@^)(gggoB{lwbx|Eho
z>Rtm89`A!gtjeTrAq7$<bxYY4SvF--x0HX*Wu#LOv~MpXcmFUFu7s($FbY?iuM>h+
z1V#O{R6UWP2nU5TIii39@?u`?4X=z0sV>r^F62>x{qtZ&VU%*`<$39<+-RK%!}Hx`
z>8>Sop@$FSg^ycC__`R(kKYKug^*rZf=>H9<5q-e7j(v>b`Pb6IVcMwjAwTMMkarF
zzCZx3ygG)OMk|DV=bsZ1SqSmxRb;!vlq&ncE<2yPW7SLmXVxVo#e(GFo}J;BCo7+w
z*`y4n?(7V|{9g6AVv5-9%-jYKX>66<D-So)Cwzlo%@?JyfS_8V)GRkSBq)*kO^)ib
zVx@J(z$j5IKC1h=5v;0$o)v3opHqL!(|d}EdOGz!+^RU|h|oWg9b>~JbdmI8&3hSn
zLk+Ag<*PGq^#7$Ibl{vNSg`cy$CpP6P)pG!`os+GtG}9L2Z0<)T9B-%m#koq(SsB@
z-z0BD<hF{2m4^#_-gGX~YfP=0m9$AfrEW3EEug}6xO?Wu0grRsL;~aBfiZtlG~IZd
z!DDktN)jYOnj;l=sF%jPHK%va+#}2VVtT_oKer65nUB;h9APV)QF?^h8f!{@GWP*`
zedyjV75}I3lFQACOZdLFoZwrd96gxHt=3iCytwVtX<Z8!VT>(D@s$;=n0X}6SfM_D
z)_!oo{uN&nkn|)$p^IJT%<+HYGB@pq)?K^)5jAd`E!4WZyS+m<ZA3++Khe5$X&ndZ
z)^V+IEYf)j&GA<=-!(?>O}la1zQ4JDXtf((Q7v6O50~sr4*d}u`l()5TCHt`epZud
zOCP%n56Zdk+C|_WT+7T_gJX*^IP$STui|HHXmnwcp?Bs&f$GiX&BK4qZTq|SXZ@~y
z(|$jH5ngZ?I#7|LL|?v#FhB;r=G^%!>}y}QtM<-xy-}@SUGX&2c|_HxT0l<3BX;zq
zYU#SMt**#x*?_;}EwMf8e%Wr6OV^d^c8J+zx#Krrw-gOAk;oxSM;(H2M<wtP_j<@A
z)P`XdjfiRD0Ws-)>zRKLjlSob6ma?A4;_?)y!j9*zm5=j8Do~%)y)?3sF;hC3C_ci
zk%HhfG-0<8?pyRPIzL|&6RFBbv3S@OkZZ#Iu4HT=P&qJl!vf!58Rf){$-hDEI@SL@
zg?c&A#E1Ix4uPYIX`!xeP{PML1t7HtJ#1n3c4uOtDg8{1hFgCqY+|n5_D2dxdIUt*
zJ0Km-x|v>Pfq6~WWZ?C{Tm}}`ns0!z#8F#evY-K?eniFq|I#F|+?fTwB9{rGx)$Vr
zCklOqk_T$^)s~er2UL}UR_#)4%5*V7V?(E|_z3BJG<|2gArPiSJ;lW1F~M|xsjuSh
zNGl`MBk;CKfn9$<55tL>ce}Mw-CRIlOycTn`Yrl10*B7_abPE#1L2hWuR9Y?x!-<R
z{9b(6McQMTa<Np(1xvMl>|(<rrh>IB*cP^z3MM{nM%-hIXA|HUQ)-tjJhYGkT~T`m
zenuL3cFT>vLGjb%n&)}pWymsn$0EVY8h!X<sxpQ$EP8*xp8;oMh#{uI_O`JxMXc^F
z(kB+>1obae*2s0B=$b!`O`3*7G7TXcl+qSRTd;yH=u_>JZZoh*!HP+4T<_0>CIhBs
zBP&;_ht}3Z(qgCGVoL{2Fk`*uE)ATon%Y5iG{PEH(NJU&WxK2dCv4i=?OCW!wrm|e
ze@X^=*4KXxa@ORmNyQYRm{c+V-89!WQrWC6HoFDD%ENV{%AS^qHW&)eO3xbRWOVtU
zpb(KQV<V!kXjt_$0UHb+2gG``?`G{F!eeKmsBm7^2OG}>Fm_i@PAezjfZ<R)s8eRs
zUH}TC2osXCETNypizv&S+<rG5UR64UF{W$#MLmDR_@X&M!oN$|KEqm@=OucBs5m;d
ze6iN9RE{gP2qAPYdbGCxaTy2ZN<(cr4m_ABJz{42imJL<|9RBaRo$5Xiq-KD&&~v-
zE<!wSrpiL0il$f!wSGyZt5H6w*6WF-%m28PF0Q1Dy2k<En0W(kyK5GxNmzMQJUJjw
zL_>c{U#KmS1TFy9rrC+6DyH?SS+yV>bG`DOVvbYBoE*!>0D0QKqD@~VYVjAJQE&(9
zu1dwz@^t_Y&PD!6zMQ*U>_ay9`|JvJJ4WR3NmqK=)Gbt~_i0Vhz`)}nb*-q^ihHoj
z%!_x$h0qHk?8_qFdr+E5v*!X`hDhe8nE8KkO~e5Pfie_X=fu2?j(jGEKDOSs+D&xV
z`tL8T`-fH&W!txH)c*4EBZ9MU1603z|K(Gw{eW)oQWv$d4WT*IpXX%Ww;s^L?Q362
zR4nAaYgb<$&<#-5`gD&TKGZ3JA6n@1$J+<=p?;6rw=N3o^xD>^&ktYSOZT7ajTV1u
z+_mZej;mIq{^h=fZm-!_%%CEuHFXidc4lS}5RC1Y-?b#grKOS=+rs-bd3bS^x!#`@
zS+Rd%by{YnTf%Cl>;*Gi>_es$Gcp%$b<?m%Hp|q<mFMs9QXgfO(dK1d&YGMxsc=FR
z4xHfR;~a`~D|Y5q^c9WeXgRA7b25L1$b$LFDSMZ7FZ@ZD6qFZc_H~~Jwh)QPq$r*H
z(<X8(DMH0#6Vj(L=ywDi5Rxoat}5kHeZEWK!I%HeLE_7q0uJ#tF!WBr#81S>VfP1D
zV~yB)u>}L7U{W(4l4ex;*_b?){jguAFzwJE>Qh6dgGYo^MI)vszinQ#Lhpaij{GW`
zz4Ky_uKt9e#FSGz*1mG^=n(`ptD?n8D9sS<>5--S6YJ3dKqe1`Q3~`<mW%3B5(hiL
zfzJ`1MdoLYF$IGHslg#|kq|jFh;(6T<i!ZPAVDHQlFi^=k0OKqlyH#F=^P%`vre6g
z7bsh;k@H!{&XEqg1bK9WoC<#lV<@r;^;hXla8JRV9UBVt3S2ePm!hWe2ZWiZUay`t
zi|3r`T(j7LJ@3y~K|UXjUd@PL-EWgbETM(XEH#6w=3MR<#E0y9a`gAV=@UNz=kx_f
z=%@}K65J!6<6uk-3z|N^LKXYMqodoL&efDrh&d7z5tlh+H$P_bKn#CSmJa_+n_aN<
zpE{@hvbov$-&Jc!OH~@9N`-5i?ThzBXw!c(7{%|Rtg)A}SToUo|D0>5d4xi0ClR&N
ztUcA5+Ns{CUbiX%?R0uwtDdo(&!Bd?s+P_wGre_?+R2$5{$LWNouW{`i2WQ2f?a(l
zYY=N<Ku0#VLF1e)wl9Ac;;5}PD&^B))+@i%4*Ux>v03NaQj0$O%_uuLZeE{=LRsX&
zJ$I|f1}0U@Fa`8`hD19&unNUe8?0gHssm(+2TFD}Se?#`@Y^SYOhRJqM8G4k`ByzH
zS1Uyx`Dcrlkes_L*lrx^5wmM3WANjAgn(lLPcB|$M!RvsEi!+7^UA*)xAJfuCug(u
z1!Hh<U9C14V-w;a4J97<V|lo?o@LM_DCH2xcMc)GINLOeSl%fc7)r18Vj1}S96lDX
zVNg+tj64X)yR4fzoG5!&8m9CZrnTDUNF&6#5enLygEBV}YI~;Z^3I7oyS7puX<c2t
zDCZIL%*=~!wlsfx!g1E5ay%sE7}rlI!2w<wg%8Z^f*LRT!G{da$GZnBD!l7bsdhXk
z^Pu*^SH-zAXJSyX0{nU3kzMRjaIoFKvJ0}w?3C0UT-z!rS03)#>ob{=v@f2orHD8=
zt(G4290+Q{Qr6Zu>MZ#$82?{ofBzqQ*V>!54u=1V$NGO^NLvoqmR2efg|H#5bb?O%
z&`p&AmqtqyP7=1Fo&EP?I}YHGKnOP*JZW3v5U}6pc`v@bR^St<N&<c=&^OK|tiYES
z{528>)AX@z{I~n*?pqtwy0aX@Nz#!jmdI@)4v~&j^+LttPU>E$g8Q^Opd?Y{W+~*9
zrq7^PQd56YQ&LmLc}vFG9($v)C&D&W1-lv)R_u|2J#3pD*Xd)s?B>gSa)h67%*m!-
zwXP<lBxTRyAOb%&fZ1gtYv?`mBqH08a^OF}p7AtdkF$T+H?sdYYVB1>TU-Q`xz)Sf
z+DE+$nw`7b;ACmg{MxSHx9-~AD?o)t$6>Hq9<G1Qn-L?za=Zs)%lLPULSSR6WU}v%
z0^1|PEsW6;=(zPNRO`tD|KhJ24m>e`K$oSg8B?qXDpX9NVv58u#pwnOvbr4RTDV9-
zxfEp%)^{l(zp=#OD3_vKN{mZUUD0FmD&Y=`axBUy%p8jXoRg#YfTyl59ou;FV*t8Y
zCzO9r?hO|vBnl8N3T6Gg5yb-j$w_YAWg-xe1vX8x<WiimtTo^5V}DS=D6H=S=3#G^
z>o*12U{49Rh{#$RdNsV*;B1vW7qCf;q(Ru35wK}n2}NlDilfaSmo3{InYm)&`H?}&
zo0(91YV~sM=0+gMsyJ+hrl|!vq8Dl}P*r~u7;xT9uMFv=v+CQ3;RVr~v|G&sPk>%W
z5xL9-X<SrL5@#2Dl2)39LQ%+5H2Gjke1XTz-(Ca5GBgFko+`Z54eFJA@lwdcj_EmF
z%c_1^Di(8f0jnNEdpTk7`|pC%e6{CF5~`@tr`4fKq;LeJs~T)PDY%WN(7_dI&T)S`
z1P~BwfB&n$`6AAj3eMM<$iSa5Wix|USN_FgZ)})@S!Uprygt9l=k(c;7pFd#E;X+i
zma_2rYydqNlW1Mxu#+}Wkw1LM|2SX+HkmGtJ!8c|FZM4)Pu_hp9M=PsTfw%gzU|--
zk8Eq;0?($OGAzS0pb0Rj-URgQS7Luel?_eJt~nbji!#hs8!79vKkJhPZ?1z4M}6CN
z26UFmO*GkGs-4&8vh&U<8vi7Lq6v}EUc4YB9WDDX$ZwT;igfp^-HNMeweR&#yY?B>
z?;3iN2~mNt(mS2I4rsNbrg*<w0wR{y&cs;t{i3r=LoXpFzU^PTK{w4(ML&PI7Nwm5
z1+{i#{Z$jfE7rW*2De|k_uxhct%lydZ`EsjMoTf;yKXx*-@JdSb;11|Y258yqt$Hb
zjd0rps&nc*f$_bUa<i!G>#U$uLDg2EbUk}CU2I0yUR{U&a2fw_@Woco$9yb0pI|04
z%_Eq>F+r`-Xp#K}->q-8PP>2A{y2Mama5X<hmprGcQb6;@pN%mRN}#q_RF`XDx&th
z^3)6YX0g1RCT(|w<eDt20F+u;G6>Q`0QO8oF5zGRT=WY$(3~A|-$UMSgvi86TDwvz
z<QvCv#)$v#R!2pA;Rse;f^*6f4K{)I7!H?{1zsApvExttP6iv5@%w-8bD`bg)n4)9
zFx^hdh_EXZug*(nOXEDb!eRS$evnnzan?>Iy1%&wnI0TG`ZHh|erE3=W?MyTdyLi;
z+@}OL^$yI#ah}426MS_xf)+9#N6>*U?!!a=hq4AIMGv10j31`qhIU~d1b-F8?+<^v
z=;fNOec{g*nrj1TL9KsSx-9N=ATDhSG{Z9DYA(3+qvNyCu34pm$E||VByc@Q&Y@kV
zl7-81Q9s{hi?%b-N!y}XwrTSxk0?Do5<QhfKi31@gs#^?)&MzZ@C6Rhjf3D%`c@1b
z{ND6HbL`)m)x)~W7bvzIGh0@3!sQv^Dn2MLUnMKZF)#Xyvnzj#sZ+#jQ+1-bsBc>X
z%`^HxCH3u94VRA*QP0KhuM+!y^?_lL%_{4ttYdWqEo5T7e8S}(S+s<PiWeuviz<AL
zvCM}-zEUhpg*@Lme{w!FI}#FNMJr<M$E)Vx_ejcM8eW~Tdm(Jdftkdlzr}pPlKGZb
z&4fK<660b#O&5Q<LeFynYZa90jNaG_);pcOp7lt--z6uc@g@GEx1h}JLld11k@x6-
z?Yr#XRVlc4DzMb}k{Z9A=G0vx53EzJhh+y0{I3hF@k<Za;2q7F0c;2KpP>W$9x01h
zD67o(X)4{wb`6>_Z$rHsTQnqC@?Kc~6ysFDu!^6PP{6?M6_b>I?ZVth6fjH{3|CMx
zA23W31PqF1N8Y!cB=&}@NEPP|XYUEH=uZus9N2=XK~=?q3{5tS&CP7DDBFlRJx(bF
zOJpX*$&ZBixdG)!w?0kIsRgwTa&!hs4Sk2${`&bd@ABqQA{bRl7mb3DQNhBA+OqIy
z#c-G2DCG2ArC;BFaJVZV4_L_Pwln!}RO5xMN0Nv-hBzd--q3-gXEL0~C*NJ|ikK$z
z5(pK}`6vE~@=rq9zJ>Ln0@l$ZDG-+HpJV}m6s|wI>T~e`#bp)`3>~|tx$qa73kU@0
z{-~`@`_t3}GfwB_T%|$Hdl*^%i9%K=WPP1~=P#FBnt`W(Tqrckb$Lv2g0}D=Nfzt9
zP0~W~8dLS2jo6_3Sg?ug{m22&@naLwN0JjNHga?HfXRu@30S$P!^flmC#F3R2cg&|
zST=yJYYeRyWTifpV2d4{qdT%n+;CWgj4lU;Te4BI4achymb~BjNsw~yJbzKlEkMr~
z&UbvdCno}by|5_1K*Iukd%%x4CQ<QqwONV<i3{lkB(c^Bk|ZQB;u|Wh+z2a)D2Z%C
zBCA55Qx(%LBE;hmlZ*P^m#KSMB`ezeVV6|nH~zswp_HQy!q?>rzNWuF8s0+kjh;Qi
z<)!=u{VrqpOyVha+p|5-ewp2~Xp7xFBRD|#cQ0~1bduu9w!OLA;n<_M?C1KnNkqW<
zj^s16cwqNGItGp83=4fRygsgF@tnT&Ptb09*|P@|dVyaVlMZI}f0kbX6}Q}20XYT=
zIAvpse-i)z9bUH|UjbhQe_W?K2gFxuOMoOxrOK#aBDH`<t1>tR8XoCnj5b_{7;QSY
z+XhZjqEdxO#!Wbi!=3qhjH;f>n=w$>oYw}*lJ5@&xygGhbU8(wYJqbHTA62Ypxt8O
z6nH&}jEt93_-=cdSPbLgdUM*muM-{w)f|6MFg|e}AOz%*t4Y{Yf4)D-Sk2#)T;-9g
zNst51QN;Cef*9eM1t5`UaZI*q!s`M(%B8BPtobe!#F`VtC(Bn_eC68Tlpx&J14+Wy
zQ&sO1L=Wyg!RysL)-(+=$~8qZ*oX-dd7Nb$#FtBA^YB$E(v_L$w_y^Gs?7-C5JT+Z
z*H9i<<vZ7j5Z%FLf5+m`YifV2SmD`LAddS;KqmWR5%F`&1SCejQt_I{&Q?O;`B4jh
z)KVQ{t*L|4wm_UV>mB!Dfuurx3Ts2++R{4lT))ARL>@C)udU+y<8{ju1Qv*|$9p@W
zI;g;Pq7Awi{vP-m+`qQL`9x2a4PFf{iNTLXb)t<L4&ej|e>_WVqm0CNF5;X}O#x)a
zw`c>;64>aB;%6a-+W2g+B|Pf^Kd39t9ax#pWhaHAxLjw1u6(5#5h~wPSQ#E2+HB3G
zmiQaQ@@$dKE=~BBedT_l&7nFz2J7VcQCp<C2@)*#t3a~n@wSDJrny`<4Q0g<gvgwN
zW<B>uZF7!if5U-9UJ<Y@cs)Y~w`Pzj&3z=>rBAeSH!1P=K*RH_2aqyZ>ClbmukApb
z!)Y7VO?F{`Ix0JE2{-AU0&$({10aF>us|C3Af0ju#yR3cqf|Ss6wj^z>4{2})1tT^
z@3hlg5)tnweiodlz$M;^_Pj>v1d%Y^QwRTmpGF{_e`noc0G;K#fKX3y3XIBJsvKe_
z%Xb0(7SA6zggLyb=wQsaZFJ71Mr|V;vCebeE<p2f3Ie=njwF@19(Bn?#aEha<hga1
zRpXUAAY<^^vhKWc=duJI9ReBLVz`Rg_<E{LvTP1onP-?>g-+yDSC`Ucj5oYe45aef
zq3a3@f8vPGj!SP3HFMg4@am--rgu);@SW#-y@INsW4zXZd-z`a#C-~nA^gk%aXdTe
z5zX-!&<orHda{6Y?)Q2L!gE4Pf|lp(Jk$=)#{)6EqUNm`T&ldand?#SRODSC9_o_E
zW<V;h2Ll;AM)zK&OwW(U)@jbSFtevVxXdNdm+WT&2!99Z;ob5aBxK6+%(jPj%O&1t
zFWk=g93uDmd_h1pLF~x<Jykbe8S+&WkC}ipuKj&A!gDpgTF31i981T&0N+$md_7HP
zu1A3k&*iPz!SntezNPS;YkqLu>$@(zr^I&=#aG&ARbLO!vT~}st7-C{pm;98kI*><
z0op24t$&P=R(Q=dERGQMgo0L)?`hC*94OU6o5uA{(9Up68w56(N>vay$w2~1JWCCv
zbA25ICzbAYFurqM2dgsA?g#4vuLB0_JCCyh+-P2_24W`^IGwrgAK<2O?E++C>w&<L
zo**&tU2q}s$S8Obu9btg$#XgI;5L`4U;?idgn!^Y_jf`R;gv2Rj{E*0n#lc8AeHwt
zhKTtlV?yMoK0w+g98GFc;}6L=o^cLYYsDo#i{X{xkWJ#=MkqQ@PDzm%IBi7+&tL@D
zpTPI0YQ;5dsBV(=fIgbk+CnqJJEB6fhG&mMbBR}ZLi57&Z;O8A`Kr*}q^~CQFeWE!
zEPqYG6Qp&17KZ8*9t>Bx>_jD#3{zAFMuQ_br_^FV+zt2AqAXd+3DO&X4<gse7^5fE
z7r0`9cQZue3$LU^3t{9G*wFY1hGQ3aloOq8yzdEtW)~_43&MNz5VB?Nb;iYP(%lA9
zcttwGt>P9iLW^0dcVhIBM>#PrCzT>eCV$--F&W{WUCgRxK3g%ut0F)K&$`EA3-3^g
z1<^A-!Lh)aaN6o3+8oh-GFvsNaSE!lJc|>X>O3xq@L?yXv6;&Ama$oxoP~y1g74{4
zh2u!q)o^+1Hdih?!(?8$i^Dq}%ft~B&sPB{JUUEDTi$67bJ}^gV_Gf=(3K_y?SD_t
zPl7q+wL>5|atgF@eCLw3nfLD`?L6<E0W!FkmLQ1SBTgcc=*}fEiuYgu5iY$zCS7i6
zX++^GT~Zs9F(83@0dQ*aWIe`JuE~>KIz%Q&^nB->){T#GVkg{jUh}MB@{3PQkpL3-
zZU<j^#WO*$dF2jB5x$-zoon*6Fn^BfHz0-APgAlpxlT(hg(FTT+)1uZc%9tf^@Rjq
zgZI#;tSqnpq^t{%>l3_No{dP^YpyL*QJ&x20+J?Yq3Avv;}q}6O;u~dDX7|TI|pR&
zm?>31>E29DC7uOM%{HDBN?prDM!Rl2Hwecv@K`2wW4TYBx}D^>Qg@yAQ-7o$CVAyF
zbmm!)G~lXo2^m)NdqxRfG_O&n5h}0$WTgX-Br=>%o@qqTW;=aP7FBq}l0_?i6Agi0
z;5CUXR(QOSja1&XmyOYr_ZXXaRyP~pxxUUIfS<-}ZRFXn><j`ejyU0WI5M;{@8-@y
z0>AZ_gB6}B0n)fd&%sAt6@ST5JKnViWOBD6Bc|q>FE0)cr+RUCs0W#oE<Ded6C$w5
zdz|7q)tsgBOmNQL^HUFGQmf7dP7>!RVm98Zo2wRXF@RWZe{vOxd_C1=uHS$RUSC)c
zo{h-OTAts@!&k1a%hH2K<pAbKcsy6&z47-HjfjQMrg69^M~l|NJAW~;36otY+G*ZB
z10?gf1W4yuwjyTY?*IurYgm>GRg*DRb4^jibzYk)%k7TIY@u*(qZs%HoI40yc?KQs
z*5HviUNY&;D^^dsxq(<7wUwo5oJud;kj$g^;&kG*(Be$scCI)VdAwcRQm>$h*O{N-
zf|#1;ErE2d$&0s{SATtr4~F|YK)m5{fFP9jew6@8oQ(0FcTJS2W-=Oe;JIiZnP)*u
zG?OmG60H)>ooMsqoJQMtuOX1ZBjgf&<9F^#GL_%)Eb!xbZm?t(dFNuux=nOQ_L+N~
z1wJ|VLrSr3a<Xt3Ca19}$1ChW8qeyMW{2iyuH!X)JzX_kH-Ce}H@QV$I@I-KHUy}g
zqus^8`PSW-NY&7R$1-Ke<kVMXGp~F@JSLsYKrGK_R#BDTNUWlV;qMW6X;coXIK#8Y
zKr)Z2t9ZwIzp5-P0~9eh3w(@me!?AV2iFHnv7A@Bs#Bid?x;>rZUR&fXO;87>&`80
z^=49IsX^<^SAQC8;F1VrOehF0ai1K>;3v2SU-<n6Ad`Ec6+Suda{y9#d|IOz?yc8o
z+HB0Ml3%0CJj+&NDBK3uggA?8iUr}i3c;M?^@W<k#A4K?4#=s_A@LsCT9%I^P^4PL
z^I||Ox9C6|zpq)VZ`?Z9wp`+yAcJQBYr_HJylyUV-+!gHka^XrcAI!*s1DtEPe~op
zd9J1+Uf@~`hDmWhzRB#HBTbxepASgk)~p$?`S}52xb!ybEbo|Z@KgDXg9b&xeLf(4
zf;iQ=|JIzId_SZ)tNGV1fJ}C~Ij6V{Uar^k`!dZ3&HE2pfF7m0-NLMr(xP?z9&(G`
z@vM7GdVgXuT0-2&`Bv=U5pt_bTO&t8;hL{4OQo8k^;xyAbz`~S0TTHA(>BELn_F$z
z!ez88J?Y}@S~Qo@uASnQAs~@kvo2D3TmqzXpS;7{;NMy3Mr&T-=tdO3&Dxh53*7}E
zj#s|137u!Mx^a%@0)PbWzxCy6>VyJsc|N3D1b_CNZ`M?vAMMs!E>#_IFz*@%(s@+f
zm#hDL7o5(#SGPL}T;jWPldiDt)#T1y_j+*esRujxvPchZ(kIlT5}w=WQBS;@p0x1W
zYdzV>t7APm&wE#T^1`!H9g!Z-;rFGdp37*qmiKjbM0(t!_bQcJvtE7S_i}m@#q$)s
zxqryNf6%)U9?$g-!^An-ed1A0AG7|qv3v@HQZSShp8XmKX?f*gEPcuRJ$h!*hxU!%
z?m*b6`CZ=uhH`lu%a;iGY-YJe!VzD&7cdZQbB>PX8VfzA!<oi=sfR}}!KwDrcxM7!
z&M`xR@a!=V&$|ppfCtHUAvodp&0x+Q*MDLo`owJ(kj``7Bgy33Gb0(pD*_|I<KjD)
z+(h>dXbK(&1DSl~0Ep%NJ0k^n<rJyj@*B#dY0Z6QAeC3XN0Y?+ghn%gN8$rcEU$Qu
zZZgmNkM1WoIe|1DXN?{PkD10;?*DVdDsD^QI%#g1#_;YryYl84I(_}s?t1yo@_&CG
zFMlncpMAA^9#L1j+v8^S>bvDj%x{)o+#Y`@oBaj9UB5dV*IyzxkGB`__3LNgv|DYr
z<G8xJ08bmYd-~jOTfckyd;8t$`2O>}FTXg_r}yi3fBJc@<rBi|<G&pbj}Oo2$K&eZ
z`LnW@pXirHgx&RKfBtr}yF5QV-hbX)M(tO}PhUWf`};rtczoj1XZC!&K7Uy6o<7g}
z?ds{9wokuZ?_dAr=9$@d+vCmh>Du*otA~f@&)(j?*Ke=3oA)2idiiMW8UMRq9e;TG
z&+U78!Q=^(XYA=etGi`o`1Q*lpP$L)*%!YvmuKelGkbYvKR<JqXYTVee}8%AmuDY7
zJ*<!0&HjhCS3m!`Jb!oCZVw;V_y7JG|IMRkCY+X!Hn->RHAB1(&ktTc;Cy=Z<MsNd
zGq`rYIbXdxJ#6lGtDmq(>N&Od^6TyD)8X;_i{CET^5chJ&TzjvuP&bf{>9)Y5&sn}
z%Xg&jHW0^~PrKXCpNIaJXGFrco74Gb{jfT&&WGa#``0H-yX*VI-8c8km#m+k|F?>M
z$&Q5A*=Y0H_BdWgQ`_qfPo*N}iAMVWyqvVlciq4KUjde10Ts8sask^I1Uqs1k(Zo{
z0Z@O0Fc60Cg}#H#JsaJ%t}ey}tHNGO3wzDRr~#9aB$k%acW<Pp?Xjofn_=ernezOh
z=jdHpr_eA+b}>ZK2<(*^8N9uAB!|d(ZhAhWkqivNxq5skdw1%&=gy;*7cX@b8Y<LF
z!{F;sKQ%>@w4_M8BT2I)CwY?A<d_$S=QdbtyL#V#Llo2|oN){$@0Sy$PE1mBx5Gu6
zFdWco?gQG9;xH&78x*US#?yTqXH={~udjNBYRmJx{UQf<k6X0Ws{dMU>NYDJEnlWj
zpjAo#s5ZWPMfC%h0bc<W1t*l5`BS&ti~%PBe@nwK6ouaf|3k@po1}GPgth~fiLjwy
zMe!}9cU!PYLlW2R-`jRux8Xp@OLETra_&vi=~h+22CS2{&Jc|V3P6jxlzNq+_hJ@K
zP~be*C9i4?8QOtE)BN@(6^4o0!mF(fSTDf^ElE2jOorCp8-}qHYf#)ptwXosvbKu*
ze|BqE*l_X9SAb(e##n*pCHEY6z|agtbU>D3K$;(0^?*`fs6c`C4%3L@lh=b)?rK4=
zoH>>BI|Em(M{#g(og9o!(?m_uTig5;e=g>4eXF6=-K+$lJS_zi7Pzl%{=}h5@!3H)
zyox)&Xy?3?P~Pu)PY^7g=ku2$#eem#OT@xPcJGu=CZzWa2kD{h{S8nCZCti*?x=hu
z50AwR<uQp#NbW+K6f|a|gc0)9g}bOf0o5VGe?;1@aX~2?O)iN?K;9?5{4)73mR|uC
zw_uL}Qw0S&V4vUXx44u6=K+6$G9wo|b$}=f3Jyp-yHG_rE0*(>-~$ocqY6qR^-<=_
zd{z{KCn534`^IR6)MWY9=-{HOF?;BpXPt!LX^EiQ>_NW-gJ(GvJf-4`vE+izpu~X5
zVo-;m{4vV;vNt+%vDd1BEj2^ds)-=@?|`@m)eU;5OlR_h>l0{nA&Y<JCyjc>vhONF
z*TM}usbG{vHk1vd$BmB|WN9#v_Q-pfmZjJ%WZoALxGwBT?}6~OJ366(_)yc`@E>OL
z)NJozu$tIOon6lOLv`y~uPim0kBD0QBR=}l-F1ud*2C*yGW}<R$Gum&o6u^tSbR+P
zAW4cnMWLd>j$;`E;+`We5t{A@`qv(yy43ug=r-$Vc)c<&CR6(QRl3Az7sh8Le*u@l
zmjT)e+=K6QcTWKT&F`0`m;p}#*_ZX00Vo?dWn+qe6951mUH||Q000000000000000
z01}rtnE^flz?XNK0XzhMgaH7TjD!Imm(`g8DglL-ngJynJ8}Av;Q#;tRRI717ytkO
z0000000000005hpK$-z71t*l5`BRs0ngJC7+Lwl!0X_j_m&KX^Kmlx*2Acst0o|8I
un*k>qI$)pQ>j3}&?E?S+5C8xG0000000000006g_h?@Z;2H2MY000207$F+~
delta 88329
zcmV(^K-Is(w+6Ad2CzX13cSr1+iXt(0MhKUO9{FJe;9D&z>x+C3^EKdt{e`PK5};`
z2ZzkU%IykHL$<27tWH0lpRYS%40jstl-vnBxMoZLtAe5=Z+67aEg+y+u9@VU@6V?%
zl6=!{bw=%L!>Qlf-~g69BAH?@g)5z3v19d`FVyH<V~8Qo$F8$7fq4UpEl^lvH?V@S
zz()W1e?gT&@BHq%ZGf>880zEb1dQL@I7Vym$Yh-R%fVgmauXDzPPey4N(L19%Mldn
z(CL6eejJ@Z5itgV`OdSYL~9^;f8Ai>gbS}G;p_YNN8v<$Gd)&O3?G@Y(>3rh@OcJ&
zk|csJNe?|`Gv&DmDwp+g7(ob@F%NSP5kJEDe*?6D{1DSH7Wcc3SRA|<L_yxvKK;nA
zsr{f{OI>@1^p;mjGV1+l_AnO<6WRPc-ZIv{=5F8^W51tWXLrglI1rD^phJ%r23kK^
z9vyFp%?)sy>bml->KK;2z3UDx1}V!*qF%PTch|^%TYmca;Jlx2@u!}EPp+mo*9HcH
ze__QHUf~{25y4AFf+BW(PGr*PBhBHxGWw|M`pE1uRouN0JnZ@=|H}X(0|*WYfwh2}
z?0y3X07UHC4{`_ENQtipit^E4kUC;E4;-X;v0<kfh$VdYEeQd4+3j`9xZiu#t`sGV
z_%J#}2pTuAy@kW9KUm03+M9t!*4*H8L5n@bA+>xO3L@aXSIhJqGUxzwjCmY$-vGn_
zWG6tPn8udlM)L#aUcb{jzmSvNN@niWg#4=+lg}__MHw^cEn<@;5FY|%Dw9SK9)E$&
z7s~posHR62#;#4F8TK=Kp_wd~mzD)zIm;zxPtrQsr`9B}JJ;ROWl0M;J}c9u-Pv5I
z=4*PkAm5*NX*hYGd@on(;&UMryD=I8wdbcm81S4lc&t=y2pC-$6d4qKCW=%KplHUI
z&o@2|h^hgiW6@$pw0xcKycZUlxqq(|yjQejOeb#x353KnbfHPe(Yr6EbIj)gLbJ{>
z)B!x0n&PiOdCf#Y$l^u#C0ns3g}ZEoRy4Jy)49iCGftBBFl8?~y^D5dL(K2j`}xaH
zlG)WKC@}_NI^HfmF5mm^jYMz0dKRzO{qEqRl<E3lX_x<_UmYP6j4x1e1Ajp22;&<X
zJRlyHRH%)4=mvm_BI$d1-)rBNX0tOZ*3jPB_v)vTJ!>MoWj5E3$T{%bu{EdhtQg3b
zo%7M<O<92a1t|wx%{7i!Kk6HN4desRs|kF)VRO~^f6oO+bvC}#7j=*Ru7_dpEqAz{
z{klNsS7h||?<FAlpoOhy4S&hMHOW01#qU|*u#h=F(<!?S8GXRJ_UMdPh1Tg^zcaY~
zR<VIEL3V)`c0kjYmRDt<u|vlE3I3ubS*{{Yzy2PBN>3vEwOT}Qc)S-@8VV`$R|0i=
z`6U^(TzN}u!b6hC_-M{-cjDSKc6ksrNyONY%;I<+4bRThML4qPk$;5|vv@c?Qwb<h
za{4tqBeG_(@09>a#-@KkB>g&i|L*O(A3qoga$0nM&zBLg#oJIf5MP_T8xc1k4vlI1
zUrl1ySK>b~v$gSWvUEL<j2Q`;&mWm1{wl-|BlD2kEQ-X<O>s#)Ba}d-8Xi^`GDuwJ
z@nwEK_wl&u+_i2l`hNyjJ8`8xPAPHqH}Q#CM5q8rARH*0`}0LCK;yX&3=kI27{ST8
zSih(IlX`<p7J&^4jPMD0Uu2=(cKX+Yi`{T3-xt}__pZ<x*uZ|-Y5qa*da=0?vIz@W
zV6(F+<NhZ_^s<;6M(uoGW_U$Dax$ZjOehf0CLRN^u(7|k!G9v!#gZ6}n6A;J-D|Q#
zChP-2vkA3h;f~ncTlhg@2`ECMq$gCcoAgpvDF0!06v+8_=7J4^_a&+UY!0D4rVF11
zu|@91lhda~EQZ^rAz#deN+|+JYt%kJ-y=23_l2(n;90f8UBY~PE<kdi$OD4Vh)mot
ziV5==9;mw(vwxa!9ItFQ$wR=GV1&jq8q0$n;xEK4)QdRSvC3f<2tv?_pNrghf$np|
zBB0LXk5)(krQc6ej9sBKVT^^-@=-qug|MDsnPS+cf#78rv<w;pnwNQWCG`md$suG7
z%2TqOv4AW<(vZ0`A$$=FU*s@|c6u25L6pDTfT;Y^@P8Hg02SZ~pdwNic(FV;U>1~y
zHxYE}zo|KY@<6cwz3+E#``;@2Joh0a&egi_?}aQ=!#0okLgIJb>C6-VkCPKC<_QVA
z$%APRI9>HdldLQL!OGa^0?@!fg)jqrUuyopQO9E|2wR}25oZ$DR(uNkfeWX0BEFL)
zE;##r!G9u*xu3NM-FEx3JxJr}LK3}R1aADAPhP|C$cLi=0?&iT(K)l+U}6dRWZ5(E
zFxP%~o$-|Alg{uhE%J+a7~$jJm>s9F6>8gT9y6y&#tVqF1mJN)2n7VBuyYoXFmdY4
znhN^}>I;WMc})xttIH$slJ%C=X_A^7*fSQ@8h_%izF%t+@kJNxAAz@0xyX$pVtLmW
z-QM-ijK}<$6OoeTV~FP@Zd_is?=D_MsC|{i+$w`0V$6?7j3&Rs14iPdYYV?KAioNn
zLHOHYkK%|B{}`O4WGqy`gJ8D5@g);{a!A-KNSE)sTm@XLcrh0K)uHT<($e-&_D5+y
z{(qj0dI39OgI0*m67I2tnZAgcXnZ3B!9EdngNXR6NEqoZUmTI1aDCn$-L1{*9F3EM
zS#6TLyo%c7d*{!8qd0F716;Rzz3Y)S3?R2dm8%K1WzpJ3m+k9ODJ^7-ymK(~AUCQB
zR^tK5D|t^BG68{91mO*79hO8Uz2&f7yMOcJln~pm$mso#zx@5~-A9w(V!}SJWO~c6
zN*hP8A~Ff7uWN*6kI>F!CV}gw_JjDOkaoR`!OgAYKOGkN`+^3s8`Ic*tk~Z0BZQ73
znFdxNf`N-uDLSpzn<_F}kANKu497zfFS+8o2pA9dM7ek+L)Q{5=xSqsT$0fuw||Sd
z2PxsqNn|G^U99u+C`9|2W2_=$6_M4D*fSbZ8@5!VI7Fo@cT>?|$|49W4_2?~lPko4
zI-b$hfk0vC+H&t85;%O9JCLRlM*^L=fx$xub2a!nF_fXWQG2;-<O<LzK<`Wyf=5hQ
zz(UA5DPm|1jlAk*2ce!pFeFJixPMtCXozK(2))fpEhC;P@2w+C56L6LgfTEWHsn$&
z8g>h&tc|jE8PbPDVAjZNG%c-{idDo4_3(Gd%M;wsTtU=u+wzpM-y&}(nUui{(d&~%
zh<8><+G7qtxM7>=PHPbDX^v(CcZ&ifx6o~5E`D4lf+l`O8%e&214C{t$$yQ?UMAum
zrhKPKmNVC$fjsvG(_^ytp}>zKc?qdoLDNl`-;4?ZZE<W7g}QgGZtu2?6jBASm7>v5
z`1pkR^LUlIRLbGE_7vN^L^TJg3(?J*@NOzCe=EPg@x>J%Mh3h9zP?>?M-heFm%aY=
zw`y<sMHDVBNfAm<td%;Bi+=^KD%L?g@d=%kbbOWLvLf~5uB~x@S_5WBrA=;qXU4Zb
z3E$pAkzip${g8MtnOep5WW^S^f8tac09?ldw@gVAGQ?K^Awgg($Ul>n#iS1ZkW2{$
z=roLxG1Cx%d=BP=&p^r8T>LdX+i9TpvGu~_O)`)pRjMI$MOeSaR)0Vf^G?Q04y%=A
z1mpwv!HG>@t4QQsUUsj$gEA7dVn(e>nFsPBE;zWEhB76fOi-M_T;9G2vgTul8N4Be
zlsI!(WQXn?bfZy?C#f2tqAM5t#uBfstvP@X#7+vcel!(PrRL~80t&Msp2OxPFo8Rr
zbF5r7eEtG_UIhBPE`L-J>jQdCZH%k6U5~;wwdftmL{jFW!=VZ=Asj7kcw{@6`7r;U
zce<U{x|H0os>YV+tBg~K;Bkw?RhsQX1w42iEWCV2F5PK{%!<V=ih1)v$0G76)-55W
zgmIyoGb$h|=T&Z^55Ghz<<A@=l`gv2sm@eszk~oP!fU7B>wjF7(%P@L8@cQap13j=
z85Ii90^?T~nt9%_$%rK+T;ax5RVXCW5OGU1*`4BH#_HQY?y1~d#~>MZ%w!~CGmdU_
z#7XK1z4-O2-T}q24v47$nDo)|lTvkI%e~zpWCH8%#G1X66E&nME>QNaQndhLvD|j)
z#{QX}LRV$B<bQSzjnJ)((EZ92v8rG`)w@|4out=qOKa5tenQ-sENLWDy^FItH51~Q
ziY2iyS)i~k&#q{Cq=p)3xao}UNIxf-ACk{wOHOWniV`ldE!Tvr!`8<*y(60IsZ@n*
z`39A$Mxz=x=7sj00wFb^`V7*=u--B0+)3OZiWzlwmVdfo2C2t|3Kf6SG4z@W7anAt
zjrk%b9t)<BY>O#qkY5`^c6a()iFK`}^uXAj6%1MVG@X8$Q;)?lt2b10WYAnbeJc9o
z<;^8<eY$tgqB1cmSCUnhQ4{JTDpYJyor4PL?5=>8)GSJ|*aWXhpD=r}W=~c_VU{LP
z<PxJQB!90!Cs&GO>BN*~s&%9(BD2}X5V}JHMYeR9&%G67z$?mTjT}jKD9(p8o5V<Q
zn6P_%H()Xg{|F8)A+uJp)Kk1WRC$>y-0%1DN7y-?LGNNP+E8ozM`-3RHzPF5&xx<F
zi>)W@{J^oG;C6q@CV7z>bOGkk#+}6>*V0x~xPJ%uSGTNjnRbfJ5H<<)XD|NG*J;Aq
zB%i+)5mj#1MB}O8-VWTmZD)aTq8?F%gzjax*KX|=DcXGGon`2n;^@jEzlLtiNUa8h
z4S2%oA^Uyd7Bmnk`>ITpRC&~5nX9Bg)J=rLFeDzILP4!X?!kXhg}*<5u7f#6auh{2
zD1VS*D)Be4=04jn5S|MmQdvJ-#7Q@|{nq(q8NX1)d=ISQPV3wvK{haQ#vZjF8W0g*
zGH^9v94L~GFJceg3EBqAd&S>~9NnhwEh4LCfgxzQ8FE;7wTM2JEoCxO_wpQ^_l2}!
zO<{*?N~fzy8u3BSLhDh8NZ7opr*akOC4UKpb@SqK8lZ(M`+coa!3*>A5`nGy`rVIy
z`tUh2Tb|w!4oNGAj1*A3Uzopa5c8sl*#N^>XkQx7)_6E<(Y4#%(Ex2OW4!4DiK3Xs
z(skll@kox+CP(cFmdm7}N-5@CUvE;Mgxn`BkjCvsamy~rujO0j6`v4<9)!(Q_kUSA
z$xA4gG}?WnR2tm%?*}{6cJuP;M0S*Q0d;Ybks_)ti}@8IiASs_IV8nUY58;aA~!-r
zT9WT-Mhk7yD$uD)jJ3`JmchuR%VTwDcvaVvkz||Ah|MX+@RUpWRl>=7)3Y#2QeJwh
z?XX=naIo){ZL~OE<@y(IuD67XgMZXqh>OHxMG<SReV;W%9X4@;LwO3iX#%7sL~3xc
z<tNncWgHp>ubDfi{5M^xQ4jbUQu`gyvIok*eb$u(r0&zF{d}zp3HEBrva5KZKo&5?
zFOq$Rkv?S1;w6&_B`_>S79=_tI*&-3yv`^cFo%*DI^GP!+WrA)1+h?0Pk%l%EEi_d
z6Hr+t7ODJuwMSZSK(Ku-UP9<)9?F?r@U$^s1P%>x1``e8ALL2Si(3S)o`y+G+D+1F
zl5Uf**cLWtTB<J={PoKa^46va8kI`$y&{E?mMVr-%D0T77)mt)Cc3;hl|#C=9BUON
z*<X#LT}-5e?<fNx$#5Lh`hSJ4|HkgAmYh=LIcv*BU^}TLc^(HAI$Q{&7csBUT-q4t
zbg;kNd9T&EEn`K==SB%Ki^&tL8i50iX$G+bx!-#rnxG#EcLo?VCY6jiWQbKEQWfg$
zN{K<u^j1NbC8x%Y3%9&rsAYf9Vp|KFVQgO|hM|eDDN4t~!zLZCJb&`1DD!S(bRNcX
z<!(64nRJ69Dwa3eD&BDKSR*~F8jkjNG)s+vF4-pQOb@OyS!ZRX%%KpfuA0jdddg=s
zD4rlt)}A4IAQo`t1OTDha(&7Hh-Tp;c&L`<5_SeO_YLI$Th9%krNlJgA@nXR?gnEd
zzmfb#@*j}=J5!5ui+?#Y&iKhX<0x+<_YdQcmwionFP%?NRTdKM7Uq3!irw-Wb2?}F
z?a>)YeIEf+ZaH(KhXd8(dXa3;s@;J2`MHNi!&F<tC}k4x*q&v2=e&b|uY2D7K;Fix
zeU9TsbC0fIgCcuHQyqX2X`ukroCi3nhtB6Ae{y|c!^pOu5Pyrj?Ei<jESR_s3vAh=
zIux&$Y;gF@acLm;Oz^4eE8K{%Cu+yyS#7I-L7uRuc}V`bpu!CiisQY23?uco|9syd
zXC(SPjO#(8NtSLrBeLJudhi*_<4D$B_|%b=L9pS0ttaxo3@v8l#t!)7QZzUmF2c4I
z9-TK~c!?UeiGOzqzQVCZ!ufNjfjbdjR=5djagIBgun?AiL<zCsx0A5ws5hTcW6hOE
zHWqEKNCSbyAq~W`fs<F_iUl+`Ne{(vEgs`@`ul=42}nj}$L^z$h7GHT(xc1v^=QrB
z@>8jUx{#lI<YL7>KVqFFr2Zo+15j>J6SnQwJWAe>mVc4jr)MQWKsRF~jT{GKyiz_C
zfxL6uzrQPGlnELmYAT-h$s9COX`vb|Z!>aX9Qwj^ejUx|ylVaX9B>l~bwQ`y>gY4c
z0i6o)8D>0m{}3-K927ynDG$V7Dn+`yPp7=NwdiXenyjK3VU>>yjYFX&p{AsiS@p_<
zG3JVKGJj>tsf=Y>4}vyB5@NVS#a)sxr$z2$&$C*<YXoxOlQWGbbU>>`u_KPF8FSDg
z4S=SuA?){s5P#q#S^a(S89h0wV8HJcMx4~=@#$0U^JKp*f!=Y@f^-?Aa+F6ggw<jQ
z4+$M<@@);T=6UsqHkhULFiv%wZ!21qkpi+nMSn%5WgZmRRT)7`))_<?MOOuYsceOZ
zX$c3Iwp1yq?<h7hBD?a@ptPs?Fr98cQO(Eg8Lg0@&QFmF7hY0N08P}@puVKPIvi52
z)4J0hd!qhai24Y^-W+zvJia_wzkJt)?esFzp3w-Nwg58_Og8k0m3ZGAdnjz^d|2?i
zNPh-1hxR2U!`q3P-X*j-wzyaA?F|l24hOY}`m83WS0EH{w!w@(O34UkQxOVj2<fbL
zH?#HhOx(0@dadrj;6^Xp>~jYH0pXaX3#~!5)SA23MR6*9qBxIPnbu^iw;N4T?=&F(
zQ|}h{lF*EjYm-M2v^unuWgZ)Ihnm8ywSU<hitTBfta;9($Q`?$8?S)AQtl(NYwKKo
ze9T<DUA~7g0ULUKRsI}a4mS#2;&5SMLe`m7BOJ2Ll-7WuOJG;SL~R<_LRMg077cY8
z$Qx*yc&4oP!qyVywZbiOv+u>Vlnt4Q=T#E$&>EvsbaL1nrmAwu@fZcjRe@pV=zsdK
z8)ZlNZp^i(^SS5Bv1P^INZE3_e4~7ecKA-PaRRGf_yNH|#2~NS2`pa7PAOowK3owW
z>>S8g+Rt<q@ZdGzKvObbH4C7-P3^sePRT2yLn<=#To31>-yPg{(t{>;y5r-N4@>(z
zoD>^F!g1mT^-^nS$1M9I;V^2qdw+M=yV0V{2ic##rsZ<7BB4H6gqU6M*#f?_i@pk(
zu&E2CtZPX$tnZJcdhQnb302^@M0+_jrtQ?8D_Zq=r`u_5R{WJ0qZau%bD=y7Sy8N3
zu5@F9ZnY4cj&P^MV^gr)!9j*tY=sEh1lx)_j*I(iDmfstTr5_zszuo9kbiIY0+Zd-
z_3Z52tTKni)b%1l$9y3>J{EnD4NJM~cwSOD=UqWq$gJ_g^<p8RlD+46Sy)n31ps7Y
z3*sYe<KRH&r!woI>=vmavK=z)Yo4SMD1Q^Q)JE>|CbqlnQM;=Rwsoc2-rOf)Jgj44
z3Z+H#Qt2<c&<hwYX}Y#%J%3U-6SH;G>B@M57E7d1Qa|^o8(@^CI>g)O)f8yUvePP6
z1C^)JAqTANlzMIw|HXZG)Voa&nyBWEk5fW&X9AcO2E#2LIxNftGc#$T8}UrYg2=UM
z##_^9Iy{Z7cZ0iy)=(ZjCI4!BSic@+ou8cMJus0`vz2~@J+Ja|#D8&nUb<m3^~E+h
zeZAG2Ff0vOk`2Aii&5?lwQX@%rj{1lN{pH8q>2yBwNl;MQGTX&G4%!o+C{spz#?hU
zZgocO&5(R~ldJ&Xk~b{aBq|mf<y}Tb-|yPB%t(OPjXj33yu(^uv#g)uSfo}K%GL92
z5zV9t+9F1v=)qub-hU`t<MTPK+@TF%uxgR&5sQIg=oaj+)-eMjpC<*0QLq!+u*k2W
z8#7X?fyuXyjfd>_g)0ofZ(<R4V0@g(qS|c{s*j@!*klkQcAZtWldEgX%FIYyNWtfT
zmS3t%mH;I}amu*=NnE4`%1su5EY<}3gpvRr5i<OtAjbJo)_*ZwDmD3-SHF(lzkB=c
z$6~>w85V|MfI_Pk+%F9m2igVXrGS<z{(V6MVO3MCo~slsZH~bLA?=GJL{*@o-0_V_
zU5+nb$%&OktG?`Zx~;x83q*sUN_4yeC6P`6QW38P88e%LdVxN<ZU9|OH04Q5oLdwo
zjFlFGSItTZLVsziP#2A6$|CVkfjT$x!3d^N#BCR|NU*OC6*f?W6{g|=;v!6rgK|C9
zqg_CDMO+_d=2BI-b3gj&1GKmY0R-XOkOmI-y_K?^sxEtzvL<LUlqgaVzm>ue7`>Su
zCf09m3_-dyy$&Nj*rcRE9vGXaIcL?fq4MTZ#=^)g@qg_2odqcZlY^D<BW2Q^`qdWx
z5AGC*@YxNVdsB<qQ`}BI`SC=6Z@bg$d^v6M?G|^0k7^0CzzW7gsGD{Xg)lW^%AkNb
zw;jG7*dgvnEyG|8N7fUKqqND|%JN)acDA(0T~Y##scKk;DP%v4OBafP2Wc=yJX-Z*
zJ3`T)5P!X(V4V64L}WqP)d_qE0EukEw?>RG8U%=9C}vs;0UALjK4h}0sEPuv0n~<r
zhirTjP`Zk_?7I;8)bp4uJ;CWpF#kv_7h1_C;am`kV925s`dl?^Huwt^Ysni5N>g+Q
za)T8#5|v&FDz#kW`Do5U@N$ljH|6^T`*OiPTYq?J7AH9%2KCzQ+SSo^jZ2d6X<J0u
zow;r$-fTX|dwOnOs-(62^aq%FB@LluHr^qH27671^;z$F(7RrjO=MJ8{)|*z@(@16
zsIKp?x;AlN!WWHL#pda3H-s<YS64oLwdA5S%ryA=81IglED=_ABk+Mmeo2F%*oIWO
zs(%<HNsH#wvtVjU<5b*~q^UsC2^gj5->kdJrPs;MK(Xs*9t5rF!iCH&5Q-sOfD4^Z
zYWX~+VVt1KXp*9Mp&WC+HAfwAxlgjdS~VM^o<`cX$8hUI#;dPm4I0L-nu(~U8y^U6
z084+#OkIPl(?M3jJgy>TGM4pKdvuCCTYojxk~r-8>Xm(E3(^?<gM}QkRprzLMa8th
zGOK%vv*z2ky}|Y7E`E9GDq-6qFlAef6;J?%U>j~tRf^t~quqhz8r{g_v1Gc1f*}Ni
zp+^-+TQCaakb9)A^39WUb4KGu$c*kb?p{tqs{vbEZD8e=ttzW@*u=&C1XWTadw(pb
z?5+fb_4TY3S<@hML2x^^NtQDjGZ^$M4!NozlZ&NBjZ}eT-)LQamBG2%Mpn%f(}0JX
z>jJNWS6F0}tS8NOi3s*dr#4I!z*6xjX4HXY(9EifB%fycC6aVS3mWu$-Rn|Tgzn9#
z)sGH)lWggMwmZ75C}GQa$ZZxuA%EU@1w~GurqfSz>ajRxb)$VwEGF?}d2~F?iP)r4
zKZK5G0zSBF4f^6qmZ7ME;&lVVT}A$=KH5c6qI1`}xzH&IBPoK<7r?K(7qm&*B(I~!
zA7UGXBY5iet~<CG9L-bt%aNzlA;VK;wdjgi;r@JZ-ajv+fOxVfm}!+#gnz@-=tSQ$
zF_a;{k>F~PrQwhgc$=%&WubCAGs@O_pgArFw92`2(<!*0o&A-}WmjQ8CwcQo%t(r3
zzJR10+vD~F%nKwTmj&BH(!vYY8G}u7DGwz%X-(3;kVh7Cv3~GcmH}b0ZHYIY%0iLs
zPaX9eueW~W^}b&*2H0sCV1KL6Fu+tD@KFXB)`0A8AZk!`3aE1ZA?f6IcD8zKVtG)C
z7DQK(ozffh+86g_T(L#|fdMF=Lf%c`B08P(>+Yz8i+Dz?U#%sN%T}+|9hGr`S86aP
zXhG<5FVI#vP^T*_<JPq{lseE2MlMDZQte)0s&FbqHIRH&r&}uK4S(IVqT7?61j`M&
z<CCgWNj9#l5NqY8p;euF8(RC!_(mrgo)r!gpAHFgv5vhK@LH49#Gy`&)KlMS!>P_p
z5yb6noFoYUC6f?DU&2Q47@4~%>|bBacT&E5?wZJ^9&;Mt`NMCwTVLnL+pQyhU0kGo
z{-~~o<)iFz4Ggu5ReuLOhTb5NJN8&wXp*UTE~1s%h%X_na2jK!D79|4HAT<1Nt^`|
zjFRzUnhCYbjl@|T&!gel*;L%M7+W^?&mcWhxW*AXQ&*n3Q6zZ(?Bb$*#2(^xr`x{8
zt_?;>RaQyGuKj@Rg6Mo6@+a4a3bl1>AcsZJBRPro70HIn%YXjO?M)eFgGa}#aC)9X
zx-N7}f~^O-c$Ph=tU6Bl(0CAFSQup0j6{`+XeD_nx&kd&+qlWrJY?N0Vm2H?x*D3P
znf}~`F=WkvsP8_!iVYL+F5HRqry#3Wx>7N;oC7khb=5&JUtSGWkhZI*?W*hrK`tnh
zAFtMFA$!RxeShP%r<$R05mBp<8@$G_p77Pzz$)Urs-P=j#&y4;iaE-{C)N{uNxckv
zn)8ST&^#6M2$Seswa6qJ$m&Tcx#3x7)>L>b?biRa@xL9^V&X5|X4-$Pn{jAljukHZ
z4YBK&;-*fkrE7D#jp56?X{R+riv!i=MdyJ)bydi--G7-&5AxvHF*&vu1%_L=gIxtm
z6`J)$i#ImzO9m0u$d0O`5xS_k_QJy+H!yJ|QRJ3o6~nQ-GNs+BI@SKCV`x)e+E{&k
zsRQP^VEwOOn%wCOE;}0?bt}IGstaa|qlg8BE+K|t=Qd(C51jP|?)Ijwyn==+<&BP6
zyksoMIDe?QHleA_CP#0?0+bkRCv`ZEn<NB+xG|>m{1UPCwo1t2MHmPt7dk-^uyJ9}
z=udF<pYm_|C}UAc-U&oM)S=9vF>!X|-NlREt5oq|>*QDGW0T5Ia_qZ-?JXQ;L4UH$
zP~KZEaFm%-1ao8HI@0=KC(yT?6N97^LQ+9e41b?n>?sbZ<qHw#25GaTdB|a~A(T(n
z6KUE)dDm`aOW_2_@O~oikBGmBVsN#n<4F02D3yJX(jL#UX=f8I+AgL$Jg~LhZk4m<
z-(M8HcfWsrcT>uo!s5;3(thsPoqH}e<JtHAy2g|?`7-h2={Tw*L;b7sZ}M95*1y{S
zZhw&fv*V@)upcIY@HsHojw)3n`A@)x7z<05`x|oJb(jy0EOuaRyukR8Y)%Ka8t@pi
zA!9Z~F^w%4Su{))xX&C{I8n+oqvNEP?AGKTi`>aN5!@rF3l~S-6V3QCXf#Ppz)?72
zRF0fblIvh<Os{)=b6H0I1IMsh$|+=1$bTF}UjhjY)Ix&eybeKQ(+ST=iHwx6AYmZm
zyB0FCUVTV!<q*E272VwQ1XlH7Zy;8+1eTW6Nji{aoB8IHtge}DS3eZdqu=eE_xk!^
zD-&x=<sHP^z`8M7&}c!4ianD;u&barUnrs4-A?!BRu_%9`!h_$yXjh;v8PZ7rhgH9
zHrej`-P`_)j26xfCBMHHvZIx(_3-M&@4C~OC;lHNCsxe!IQ~*rloyVJDYZDi%Iyal
z(S6FIJL9XwMN%5h)4<-Fs}76oUbhDKlBinPFM@;TEVNk=)6H&%YE!0gfVx_)E`(0W
zTqP2|A$@e(B9cyUQXw@pA|Cpd#DAp!AG^p~nCPyn^v2K>gb@#man(M2CxDZK88M!}
z9L0F10l5+5WwgLI5ne(#BZ5z+TfA%(Uw*ReUKE!C$8?4_o#Bl=Uq<vkW2kF_pv)%4
zf2%{LNmxnb?hmH5UTJ7COuHY_Ma+GQoZ_vp14hhcYIK@?nx;-~=UWRiUVrc@b0F2(
zwCdA=^54BxpV8Bmci3Q>RMD16ku0dpZppz7hU>xk`F&TLMJd|V+YOoDoetQlcN*2&
z8akN1URuh@%5UMs3VB%XYD$!=1`3e0?ejFG^I3I7#+8tUa)=eF<Mx};nJ2ba6}zz~
zRO7zEWr%lqkwJ#BrX&)qLw`0EUR5E>Dq}=#Cn*?KI89bi!ylp2AQS2PApWUN{XpJv
zm<}Oc6T^oP#Ml^Ft+p7OWXx>3h+w}}9HJgHTUinD5Ef!zajqWOB<#s%bEx$e4^wZj
zbRF=bGtVsIJk4^VlLk<l+K;=z1Cc6-TM3ymXSO>L^3$e*`gIA96n_;Nh{3_MNkES9
zhTdg*|3*_#h%kR1uM*1|$q!_*1zQW&wa7={a@9miaQ$TAktOu(#Re2$eWBSEfjVVR
z!YuPv>zzR)dQ)649`(p@JoZWkIOWepkgpJIlnEIfRe3``D-%Au?jEb=nJ2eSrC3p>
zT!LX{#{yS%F5Tb|vwu0&B@lOQtZ!Gfp1G&MB^LQ_v=Q4XzX;`@yfScvfOFWRYbW<X
z1MIoyLT6XfX$pLMlXMt7JPg;6JDSX5mOs)`KO!_|$hp69Rgt_LjJlWIQg+g7R!Rp7
z3Fg`}x%I<^AoHjca!6fS6=;!PLsytTn1v-^<px3)x{qKHZGS-%Rcf>lPXtZcOg9ej
z`92LGArX-aXb9Qw3+(ihheeX&wr)zB!RQ8>k!|E_q9lKWt~v>eN%7qjke)MHtE%`0
z9H%6dv;&GpOrNGJK_NQ;!Z4{D@pgjN9^?g8sg3_t)v5s%<$$4uZB|eI>A4TAfmFO7
z{q#YRkrLaZQGX<iebkjg{74}sZ<d|GFr&?J*4|uU!gOhM)$Xq)n~vye*#lYLklJ6B
zW*H8VY@ZZO>y28fP0(gfiJc{~k@lx$lutv~$@W<b=~BpsU_C~3Oznqh$QJ=Nw-WXP
zgy6B&E}rF2Dv@zlEGZE-Z{A2LQz1z~>hR@Aw2ET3PJal(^Fe&X9#!vp(7WDKNwR~B
z@|R04l7|czm4J(c<lV7DMWKcejdYGG)8`>e&Hf3a@j}3b)EyoROB|UFCnNdr=S1Kw
zZelVosOfd}$8qI)-x^+>&iRFb*LO9^H-%{l8AI<}(5oS)?_gY0GBoyjjAEhIWL{wH
zuQi#_vwwNb#0}gCLI-e>u)Ac2NkeuFN`?`pts@0hZ3f*V$174xjklUb?w+VW2d}I4
z*F<g%hc@S-^J(I_^Lm41j!R>2ycBkxkq@V(e%{Kc&z1rb)>urTNhzFNBMXsIQx|21
z@g?S3cBUpw(VR{{2|EvN7Kp<<SL}b)fEcK&G=G!IlU>pf!cqo+rv<>j2>~Z#4Efjq
zqQa&N6nBvu*h!z<L<Y)DCb{hda1~<Bh9$v$7|0A6dvYV$dKb#E0_v?zsN^9AZeIg#
zr9_r8A%;@G!B{;Zl=%2J1~75a*~C-7FL=ywKN0{4pz_d16_$5ESQzpUjAytqfIUhX
zCx4Gv`Y0+)^N6Qs61f-I6dYb5_YzD)!+KT3p{|q49K-lBK>~7c6jawa59_v8aKY5$
zV{w`Vk8a2VDQxjgQl+0t<3D@xf4(L)T1!0gB@Z8fAF?-tfg!^`f615x(<$aTB|*Q&
zUtnVw_Qs+IG80ZZ2=Ju%L~h+jDu0T6<bUFGcvxQ#RSy^)-241u18>ytj0UA_Ryk#H
zJ#Q?PlMNL8bp6c2NJgv*^jJS4@b7}htxgOjeDTI#4r>+@4y}?QBgdU&0;2Eb{1cSf
zm}xv0Hnz`%c#rap{R}uxGYlgUV@aL32RL$nMJo72i}OW1U&O={#%*qdi5j@GPk+|K
zH|lpfxBZ*lsM6)brG`<lQqycz!k)s7j^d`jRiIyLmlrMc`h3v6)4V#7dP};(8CLF8
za_&}0B=rH3p%uzz-@LG~=5Hl0ayij5odQe7p%+N;6uv58$l(hpe@~N@sskku1alTn
zc<8gjT)rxhQQx*dA7GT7;zXsE-+$kI{N?>SOxXjD9y5<GF>L}j?;I9W*E=HKbkQ5!
zcDlRinjFj+e)-Fd;a4h@q&D6r%uD#L)0mF6?W%TFWImVw<L^lnQD#4lh$Ie#0H+}z
zSFBOW0k@`5xA(h)i&8eU*#$hk$&@uXk8o8=OPbxw%SPs9xF+3W2P0DsXn&kS*CxC`
zFd=5tWA{HEM7mot8Yn28^;5Cr`%0Fh2^m(&a@+_C{;cw>rEqfXAJPPT0y}sr_O~4r
zP?u9L6Mn`!GNZLs>V@3OoQM3;g^hODNf$afb+QcrP0SjhV$DN0$YB%yok5WVDiocm
zJ!|&FY|P~Nfvj=cOzt2)i+>_Ig`U6ye*sqh;!l7+h6$UnCm+SCk7CGdD5(S6P}OT9
z^u@oya<s^Y%vSu?57Y&uBM_;cE1xl5X{xvjdbeKX@(8w4ULYX>+AavkVlHF?Cfh(B
z*Sbq{Dn&57u`2`Luz%(CkLQl9IfZ)bt>nOs;%YjMxVMOf>Y(d7(torJRmhMB$$OCg
zhqg__O48AbGGX4sD}+#b>PXNClgM*6$GLqUB(}TwAcMr0K;rwmfBy3J{oO73d35uC
zM?c<uFrw(R0H9RZVY1&0`j@TI2HD%^?em*;Ks`hD{N*P}cJ*<}N%qin?qJ&bVZvYc
z+P$qx@%@C)Uyg)Vhkp#=%V2siAGXAEWX~)BUjP2?{Jec##@BbDX9lDPPr#Ddw}z4v
zp*UWZP1H9{RZmPAN5bXbXx0CHsBUcV?;Q+E4fFe7-ranB`;T`Y$o>0&{Jn5nsJtzx
zKK|{F*x<%0H^T_F(-Lfub7#+({ZKbL?G#DR{rRW1V~BsU#DDJ?;-4h(r_)crEuz>E
z|3ryD7Y<DvGeiGVq(APHeu`cDAu<3sMF2oizp(qo_cjvXBmn>$)Q0k>NO|l{X-I#H
zq{k?BDM)X^*e^(WnQ3hp{<Nr0q1mu3)tsma8yk7j1`(%*2uK#r4}x@q3D30A*o4UN
z`(wm(CVWYZ{C~-zF_qop+`x@p>Jd01I}EC_3It6EUA%c9`x6?JoES=?n8p$%2`1nO
zG@i8g*6AW$Tm9C}c?lP_uWP-}BN)~OdEMehkc?0fiK&^N#tYXIo)ogozy8(v_rFVE
zSsOlhQ9cNn&mWm%g?woR<DnXXCH@u3E|Fsx9;GfK&VK@Dsl6K&amdB(`ADT?mTb9a
zZ-dFcH!nG<_HehuVpV<W24qE`Kv2%>w7IwNgNW2|P?<1~aO&JLp`_vS6Xx@1m@+i-
zxxt9R2x3ImE@y2bnDzrg#lcBGJapVZh?f$ik#;V+eS?gb^Qm|&q=rGkX`{fWZeR(K
zFwCL=qJJ^sn1RFT!GSq04TePiX(Yo5Lm^tsE#Wyv4=mgg4l$pDR|)5Yt+~PuIb$fb
z0>ffuLB9z&#GYbdc)GF6qB2;X20t%}A2*on#m|_>G4~CG41@|Hlo3G|bEd3F4iUIO
zAW1qmPXa#%FDEQ=cH<>4cO+yCVonq>`)VnL5q}IOa1INZRUuhN?encB2jyiAVaxkG
z*xSQg&p0_yDaYS+9(XH=c?#Fn!n2!+1BSm(aZA{O?yz3Ib;$-lcduK6dr7mBtzosL
z!(d4ySoq`3{#)8}bjrXf)p8uHCjp_m-`2QqjhM*DOk)$h5)OI9Y+*uJP@ZL_y|%Z7
z6MtR>odU6rYt7R62L$7zIa@PRE8xLvjoE$Q_yZigy0Sbz%@NLVCliK!ODCc5t7-wS
z8F^vk#iqOv_b!+a7V^&>a>7CtN!XLk=9n68#Dx(TFD5QbM#^^&xy)(6JejBRNJAG#
z>_9ni1fBcjP5_+4%ktSx@=&9A0!sI;)qm~XmQetx5saGh|J9JZ8(~v;n9c$Og)!!~
zjW*$Qot4}ni}^I9Fy@_;$F9}2T88gWEEGeZzyts;etJ!xTsit;CI1Sw8wL<Y5s*K@
zoCt}M&502f`6XhI3Jt?r`Tu9{YnanUvUdMU51HA5JB)@8V>_OTs|3ig_jPbC#($aJ
ziu1Yzw6QiqVkKe6b9evy?QThcKoSB=&@Il?RBVL6>ORl;?DIkAQkWC+JZ^CB)w3X7
zo#1^Gbi$r=!pg9Z#VH?gG&wqH9KWv#gIBL2Eo!`npahH@Iw$0)5)kS;)$%SAoO#UP
z&&_n6RCjg|F%xG7y@lCmAylMeMt|6^v3<sy3K0)H%7$}El5)lB9uZ5_kDpG4s)Jn`
z0d+W{4*JzsIA>>MiKV##6g`~Yn-zLPjU)%t8TWA{g-{v?T6bx(`NCvgN>JMzV@IQ5
z64fxwSE`Cto83584Q(>j-Ew%=T!GxqKE@=id(+Bf+)xdDT01(e))H&z;D7C)p*aFs
z+@w0HyO&vqKQ77P=-JgwGY+?1=#ljCGu_wQrt!=6<Q<Ph^QShQozY(<)v{=BO+l%6
zzTzEl_>AHjn~PR|{1U2-dZR56pR+DO>Tt+XR#>2h&5VzAHoY|pvhZ^{&(p2(Hv4FZ
zX%BrbMaMTB@eg%8S)LXo4u5ISOQP~s1Sk_aYjGGot=CT~;w)K0E%J11imYSCI-cIb
z1R=05ICDg*^5*2>c7^9E@*@Kw9*z)1ig$X%BT4*7!=(_1m=rS-$pf+Jl|1AmCJl!s
zw%Lgna|Lrx&g22XtdH!V5%(^S?391VpFUr2ds5pttu^WuAbTj;y?@Q&lI(4zdH3<-
za+=A0_&s+Gx49*JRNj@19m&RcSge-NAIR+6RrELA`Qta%XIXl*)G;phv@u2>(4c4;
zrZp~BE2T1;{fd4?wNkv_&^$eAbD7+8-lF|At5MG`BuNROEIcU=iiMAbGAdja2yw7F
zcp}#cwc^NBZyvW=;(z2|#beG}Opuh^uJGJ;h^aZUW&1)MIHqM2YOTyf5C@s(RF=#n
z3-!Q#;qo(;C10y&)yg+d&U9q*78N#CvSucK{y?wf*8;t>Kwh{#aH;qyfRN#Y_X?Dr
zj4;wbS8ZD|ts(i7AKd8XAF-TSt<>adlSe|)_7OTBVPh!K5`UDusRfTV9@$LYqy!^e
zHL@@^7)9v@JZie`Cvke0UbLY@^+!c&cvslhvn!Nc=@sHGGaa+|IbAcbWNUx9clw`I
z^~sYOxWZa?xB}1ri(W6E@6=i(vcPtRR1_?}6?~|n>CBc?=<Dj>JH4M>^~47V4{n4<
zw0v;aqf{F2tABM~)<d&rjFEp08^3*QE)!{nGbvYu3X1jxi!zyn53ZfxNC>4~VzJNH
zGS*{roi-3@waY|vv>~+=J-y*yk@}iJ3&%g8M#a1QFU@fnhAUFd*A|{$s-r*Rpe#2U
zHIY&t1SMzqYABh4)&jlKX96i`S1erjt$+lUXsb~<K7ZoT5{=#t>QIrz)x_o8@U^`X
zV{tsEn%hC(18Wq(Ui#f27VcQ;6S44)W(@S{5KDhy%Um!?$+wsjiT0IMrLk&_G-`BK
zqzVNPks7Q+WW?L^zErb6GrQ!Qul}E5Z+PfDTDXyxM#a&MQtjgS<ha4(K&o@L3ZiS`
zkE6Zk7k?Uxl<M(Hfm<paf>I!Er5fP&z{_WB5rLpm57sja9Kz8)J60+GIhv!>7X>O=
z*b8a#p3^lfdL*B;8q0x2?;?kM>lwE8llkK9NrJ@*gY2AXL62C$(1w`$;zzu)5Lh!H
zm5LZ9LVT4^lk68>deduXA}E;ZIlnKCvY~m%{C~P*-Vzh6h{K1EuW=EN8j<MrfF~rC
z@@3ncd!oRT5bBW156_07O*AC83TZAYbgeH<RHhbIYF&R%oV&4Pyg=@T!2Ark8|zO-
z#1qjmxf?<gs+Lni$&^q{N+?>6p94bGb4sWyN~pgF2^B{s$cEY<5Vy*P;@8%LY^Y@+
zBY&#eKBqziv@j<!FER^_)*c0|s0B@3%2o2hMmcNZDew54wKC21nqtUZ15v*#kt5iV
zLyF03f>CPQyktQZ4Owd<nvY4c*6NV8CbZyztTo75<1cG1e%}Sl`+yX@wa+g!Ur`7j
zQVO16><1}$nVGn&>3VB*reNSXv>Z3lQhzF$=-nX4EqDQv+#tzq&+t9CBsYGO&-at$
z=0VoK2uW@Nt0PEqTYrQio`rFsNp8XtDlEw@j!dEHM*OO>JKPh~EW;CLA4nR*_)r
zw>G=wl5*w8?Gjq}#qLH4E#gVksDzd@^H>&=Q5Mr!{M-D)DgX6S@H{MsWvu>)%zt5_
zqGfFci%4WlkTW5WluWQ1@r38L;nhOEVyR+<hEvQ%5{ypi<RbB_vp32`%7-LbPe&1m
zw!2&;q4_soS_2Z4wogzJjMPyHO5{hiB212w8&ayAT)1(1&g5u8?VDt;bA_8l!ubB7
zm<x=Vd>p=X$RUFqGB83iU(GIZ$bUjojpFbiKw8<_ZH2_SB#Co;%jiQ$D~n&8XsQNs
z<YwT7961qq5gNg9gSq|W$iehRcelK!3EdnH!SW70PK^PZv%3;#PeOc==^WTT{^`8+
z<5H6PRA$F1UcTc*wH;h3s<y-uK*p*wc#?>Zq!;mziArqWq3PIh68-43T7PTvxZJHI
zI={q=G;V<iyCVWqm!;Czuhnng%BWB*6@nRyq0;iWU1?TU+N2%VPTTWf*n$WR-p)j*
zcZq#7=c>{@(eeIjg*<kn;%{v7t&Wtf-mdKc8L6y;B15%COy1g_z-7M|A5Ckl>g4B?
zdUE!wE&0N^NMB4;3mZ<6zJFIMk^kcPCDUC=pSsav;hnoX$b3XZ1F;eEWc#LLvKK)e
z6mFH&k5A?IOMKQmI&Qhu7R!7Ve24LwcWGlj6Pl-#=81gT5RRv&Ha%CyjzdV{v6cu%
zOvW4(FDL{gH78~dkZm)v`G(SO)o*)T+irCil|58-EvhYAFhM1hjem$JqFD}9S`A$#
zgvj?#iMBIL2Td&VSsUUZ>c6tjQ|Xq#w1g<krrM6@C=4b$|F%dB06=H}Soq&l&B6;P
z7(C&@ce3$-gun*pJWxk)F!KjJ+jdag7y=@fsBgS|YfkthLD0N7I#m>2K(I}bkT{P{
zApI~f$CHQJFy1HxqJQk{(Zj9TrFwagHU}zgBI?~JBacQ8|C-uP0iXg<Sr3(osuM>T
z16lwr>(D|S--iLE7{CSKvKB5A+7|$D0ytTN6BbDUpa4+T0mU+>#t?v#8c>A6giC&V
zKupy_d1LCkd~hn)o2}5a9)N_9kO<&lMeJ2CYD(k1IAV{5;D0p3!PWz<!(PdW5S1{=
z85B-LVzfBFCn9hbeo*9oG_9gmMVdj6L}y1~rT@nP3PjV&0;52C3?6!}Ji2h_L^?J`
z&QtM`AZlf4;;cej)74KnvXs%-9I~>1U(^&zCG_C~s>CmpV||&vs63L7TS|QcQCX`U
zpYB~$ZYd%;d4D2GLI;m0ChcDstk{Ze7V43fqg5!~?oAh6bx@%^%Wh(I5oJ`kHxjI6
z*POTE0>cAJUFHAYU3HaqY%PBF#7f)uDkH7R%AYu-+m;eDst3ber`5Z<>7c7y^y%vM
z_Nw!dg1C37pe|B6U#_}0oli=qw?k#VDd#0AQ9B4#@P9_DnIn?gjyWm%df?u;H$E)=
zN3Rhsov98cWFu}Xl@TqKk^F7|u}s~nRWIs|C7^g0Gufatn>oCrCLQr9+4hx?W1>Dr
zv^NL+mG(%m@iVqm9gVSVt0Qca_njAv4f46vc?djDj?L%)azJhfHvzeKM{aND4ye)w
ziY)*QE`J(43xRCEX$@^=3_P3K5cQ|DZf{nSPDdTF$#7_3>eI1d(<;3oH7J}X&7Q#s
zjw%jHclh9uT=4kSIt6bQkC%1izoa#&^yO@}&Flqz-+w^|<M_X7)Q~TXEOR=czkHTE
z%JuBZ{QEyCa)<t$nf!QEU3y(}U;kj=FHKB){(o9C{MYOc|Bznk5BHz>-*|m^w9GLr
zruU!l)%!{N`#-qP2=G5hT^c7rw-uA8O0`v=_wHDSli=-u6V7llD~X?2+E}$-H!!!_
z7q$AuMHtP4?;Hrtc_lA;(VQn_m`e=~*3-Ayt~6S6o_4YP{`jbU96E40c>96aL!3Ps
zC4cGK^ZDbGTD_K4JLxw)x@x@gdH}!pm}AYh31EwC%4!My2l*FX7hf>zZcUFceo3+N
z;5<HKg5&h1PGj!2d+lxfi*IT_78(D|XtP9bJC<$3pQvN`*+F+aKMK;*DcNc8@`3)o
zm!2)<&+^x<OYVNWqsQ2{NTov}klbmbiGOLMOZkg}Iyb1TT*!CVJ@iGszEjw{US<Qj
z>8zP&E)KO5os{QJ%BPB@97y{;9j<4;O*`s>m*o+A=FGRl4zIalzSI>`f)m+1fguok
z5f}{VuVWr#eoKHU^u)oN-V%b9{2#a_uhk6QGci^XsitWEXcm3;d^a{3Ctfr2JAdlu
z9l#&pzY+Y|nJ?fU@PDBAXVAw(X?Zk<*ld6viM2J$WgFste@el4=4L7NUy=1dE>1`E
z;)7$FI@nFXI8lsqj{N~)_y=Vin@*-#Hk$_XD_%I{=ZR$!XMbEm^um&s85HN2{QKz@
zdUF&@4vH)9_#Y)uXBU>H<0Qv%-hUGAQgIKn2Won!FCIo_-M!c`Df++?HLC<ux@z0B
zsF%K$SmtJ-j0{y8kOiq49j-*@V=i@iTG<iqN)2ff7k7l;hv3Oe3H^aE`AiXG;E5D@
zLXcw9Se}T-xNsHwucGnnf;f7{1BWI4`b`H0PpYL3989md9U{opDmzvMK7R*3KV&{<
zsZDw1b>QzzV8sESZ<l9AK;v|=F~k-gepW}gX<_v{`xMKyL_Z<TmW-*cyBU8HLz!e+
z<O3z>jRl%u6HUNxLDdSj1s4BwBp?WSU|*;(KO{ZAtgWL27+piPQj@DqUe=JnBJZ0&
zBpr{4nZiVvs9FXLCNP*18h=c{rhB!az=Z}r2g?d9E0^5KEGv<too-(cu&cnX;!@r+
z6RH|zLoFaUsR|D{kdM;pu@>5Ct@Zwbr=8-#@Bl{0mJw>3F~XzjF#>=dKrbTbiTUL!
z06)`@*85TXqQ9x*$scZNUTyhJ9&7@UrxENL6<ZpAv5gX_aCGINXn*;3;mg`c?$eQL
zBWCUFd+poW^+R(?I9gI|$y<x0>bm)Yho1ftjHc%Yqk;vXg|9BQR1%(>?_U&NN$jE?
zLAcP!+FoP)iq2I9=$S?F3lfe*Tv&Oh?sSShF|a7DJl}%j<&IZz%27T(Y1Cz&77;r(
z>-ePFexK<$-&9P`xPMt%>zY$ROnlG8TcHo{5A62_6Ru3^)7@<kH5DYI3%S)J97FzZ
zUCLeZLg}I}ihu2@(^Edqt}HbH+vz&f9d_h7YP6A?t<e@teLPU7Hbz6N>JC!-=5x@q
zIYf=UIkbe**y~GhsY2%<XbUjN4hFGx+bllI0sSF7n;HFiNPm%mGnrhI7ZTinyg=S9
z$a_GAT{I>6QPcH(KmX=OMbU{&x;fGYst$$wy9Co7x%*9L08tE08RJD8I!|RZvCPTr
zIDEZRcxFu(EgIXlZQJhHwr%T;ZQHhO+v$$cvE6ZU@_%Pv?TdZAZtAIeRt?NCXQ2>*
zCwPOFWT*p60hcTVr_^C<HnsLU=E$>^nb=wdljmCQB|gq<u#?tf3J@XH><%~i!!Kj=
z1-`=CegY`ij+7cJ-VT3(>wvR?Tf&r4k7=?CDFJDKYMew?8|1OQ^R|`Tfa(8rYUeJ5
zlYaG4q{E`vKMJ~nJBx383zp{r{k?--bZ-gzwgp_M81UPCVaiP<qVP;eCZfHWl&rwi
z9A~6GBx4^FM+6A>Fb|#tS0tnnxFr~90v6uO5)$A9RT1EhbdlZ~xDsxT^fB62fDOio
z2scaA1oP7Yfn*8MctAzP7Q7Z43pq>VZUaW|5fP!1s$UO2_s0uCj^STOX|T2Ppj?a~
z`V#SJ4^S>93s4T%IWi9;5FC{r>qQTceI0;)+@9glZlcvOV91|{@vz`d;@H)BAwyJ>
z7ZV_IuTr<pXKwloeAG6-(02oF>wlY0M-IOfA9y6(9Bgi|av4o}E<AlAkcU~^MolKB
zwoR)zR}BN%3U7@&G5r-<)KFT(Ul-QRSQluG5m(kB3XT==n+c>G>wFDf5N0kM=@`fd
z`jVW8k&;Cx<hG?4-r#|(t*~&3V4y|vdJa%>5jBscY(UC&7QR30m23FGtl8EU(cxBV
zxbTARm<1#MUCKQoHQPhKp!9H9G5IkPYG=bvGcGyHBn$4~!F;3-A`6WI?5sg-JUMV5
zHR9#@;~JEL5`;rZqjkWbV=r1;x+D&Y?$M|vE-F_!lP{x`q}<`u<^CmyF^JaHlmm2}
zVD2B>J-^K4qFP;`?$OHP!Niu~cc6d}x%ugFwwv<LPZVJ6xbG+fJKu|?Xe0gE&SNE!
z!wKZK$v6Ss+r)h_yoD5^F_U`IA|Q=mA22k>3Yj2jPZUpocacD`1p0QrgE|9b{%kN5
zRUa+igls})I1wu+Vs~kJ6?Ut87r@yP9g!yp6(5<K1=tnr6lwu^(p7A2f8!0bHF$;O
z>LB^t$>dkGYIbWHi-Tl$fp}0mVG70_M=0t2nCn*=Vhvk7&bThWW77DM-*UQSPlR|N
z8K~6)U?Gc&;9K=>XtfdcY3BnPT`fZA-YgQY#%~ngLcGQ8OPtkT*p}Zt5dagD7)t5w
zRJ7FM%EMlE+kf*r?h!Z-B#%cbkB3p5!{cPxT5~x%ZyX+hpy~24aoH~a#(l;;PZbiG
z6?{ETe2WSf`8mzAerE-5Q;8n($zZX2pbD@XKAfJf6_lgh2}^3$<3&e{D$nJ8*fB}|
z41on@NE8s5po%@Q9q*B!0+4s$P%hJ?C&t#J`A*oBzBo_zjn(G-6S8gZvt2zm?#IZp
ziW9}&2L@<*Yoqk=34y72Rd}qEh5SIXNzrBq<+-N>_v?8pay@<TBR$$tr)+mSJ_tCx
z;62*uQ0;>0Ixno=WTlbLJY<xIzuHxSD}+rbA`&0&31B&*)5~3>1DF7zr+w7z4hqbr
z``Pjymz?5`K9BX@8J!7w$L1lvxwEl~klIH-Egq?h+B>hN#v10ngeT=!<Q|>{AKI=~
zVLnN4$VmSo|GEXXf-oLoARZU<i&^Q5JF7o;jIjNPp;wbMvE4j}+C20-0XGTuwda3#
z$WU_`I4NWjhSBc01CRj)_H<I*P^eLSn16pKxVHA+gad5Y`~-Vi4MsChUXN1REn&c9
zN@!Uk(zZ}{K)!_O8Fg4A`FpNq$*0{)ZG=!n6A=g{7_%Ni#%Ny0{qdjWws!<)9u5ei
zQrvReDe6V4GDBr|z#B<T!TD0;O~aDDjkm|})<}segcavC0G<^=JPxR6LXH-c9o$5W
zBNqRD?X)Qia`XR^-PyZ}>DeLAoAkkwGf*aQc6IV~-r_oT%oP(xWjY%|z@0ARVOhYO
z^26N_!S4c@kPP@j2YEy)IN=qVCxQWf;~&KPE1)%)6R>$^Y!X4w_fZq)xIA?3@vl-e
z3JSV1fXEf`63|OPj?^{JA1kZ~OIktfiYS|_>gXb*AF&1G71*<xz{wfNZAX{Y6*_>I
z(f!KoF?!GsFPXP8ur`bJ%g6M}E|759Oogu!ZIxI_wXy`K{*g}%*9ZSBN@L6f`%~hJ
z-l@>16N(+*S9(fhx`e7=xfDNNb@L{jcj3N#r3d7>1CTl~P8r)Zvn*QwSOK%{=}*9k
zC*#yxGv|pvm|eh>1-s0!Z7<xc{)@hIxszDXo`*kEG0qxd!!-QMFxSmK{f&=o5ha@j
zO!8`TM*)B@dz_4kntVB-baXhOlLn|>;4y_i#88dq;`_Tt7<c&3*cE-UfUCb{LCI8M
zZm>{T7GOzN`P&j&x|!Cj9sJ#22u!$mn}-p}12bwyNT28?r-wSkpH)}(jUb3y+uwKR
zwLWGB+HKMD#p8c4<g^K6iqC#y(W*=}=<8;9E<SU3RI{=Ou}cmLgFWyV`k=fw*q%&f
zg}-0Kfu8DoVNE>R9K5H<tFmyZ^fgrIj)2$-u&4??6H^Tgz6>^&=bFf?2SaFTPRPo}
zH-#PIfHih8Q;dt@;ly9*?Cz7ucd)Pb)-XT=48Y`km&v1XeQ$SGg*_MqHh`V<V^i_o
zUGi}ep+*0<a1gJ9_$U_Ng{vzQTtr!uiJz67Y(LpMbVG5ifcPO~5)skg)6nm|0N~?x
z5i~^y?sWX)cezSL^;7VgS;5wB_JGu#{&5ve!O5E=%GeK2Uiy+dvSf#fpf`<hrQ@K_
zC3p!)=q%tH^D%`&<`55KJ8NqOu^z&$+=HNqaE7w(DZdR77b}`V+=F<X9P(E<Q_W(X
zE)%DH4W)J7sM!(>$6u3HW|P_#0M3m$<REwE;8+`NKWMm*#Yo2bOQlVWJFO(J<>*}g
z<~jkV0m<MRRyy#+aYJwkYx-OqOoXa(Y%1YfRji{*>GCk)jYMT))1|#tA&E=iR#o^V
zW7RJxf)ha?Tn*yTKprm4jTEwtRW;tI^$Zq*f1{ImpiI<oBG_Ir_Uw<u0*;LC=>TvG
zkf)n?eqtEB*~N_bC3<-Ic${&lr1YM=qFEl$32hD~N6lHmPEf${apJA*Nohhgw1g2#
z0@s@bJ}iwq&s5L@k&(t#7Pj0gFT2kV!(Mo&uiM7?1LObqf5+ZNF}((=0O{C?sW85X
z$B*F!2tss%%y9}~jg-;S0PI>$PWQPVRVb$)F~LDN2vSXNY@*2%uqK-I+D6<zT%NcP
zM7F1aue<GAB%O|Qa)u>+n)X3SB=qL>v2CVGz#1bJW)C5YabSxjeJ(rLn0WHN6~X70
z`_&sQTkHzFQ_EpY%R$3?fXVaEK-y-+)@?}wNN6p85f1>bn^GrXA{hUpEL8JGX#O77
zO2ot7Gukt%^Yk>e5;fdYgb=XK!(-1RO^yf*ZtsS>)c>0qh}fhlg^z&BJ(WocG!^o~
z#Z3~hw)<*IHI;=aSw6Mp%|*qX+!L#Onits!a|?E|38P-Jr(P18TWJ=64^0P0mqUvI
zIKCAwU9>F9Mu5szG*Yo!!L_46K!MKQyCSj`S_^1^hQXD*5{<y6!$sm4g28RTK*$bM
zfp>(#G=Z_OQu10y6=$q6lR}kY#fU!fn4`jciYj75NLFf#UG(;$K==kzi2Kj7J%R3;
zF(c6n6_u<jxsYJT3__i=FjYb~S^^wM3tx(1btP~ckKpcNF;^lV|Cwbfhy>eGZ9!%q
z7fW3iBu>pp5S>lBMVxiN1h;X9UhdXg`E3|gg5n)+%SWsoagKYKA?3%>O6;7+L?2_g
zWh-!<7)Q;(Zk$yk9THkVmb)(D%8s82YUHDBq0dn=d8N5Fl%Oa;@OB#(at*+dsPvw>
zs;*l)-*)|u6qxRnX*&L=CUZ?~*Y&nj3R2es#Fi7UY`06*Ug5GVIwe?V@6&Ga3?a;f
zN^0{#Z3L|)0jlz6HFCBH+O&5NDv+y$|AeSzF!U}-p_u$(y@i_K7Rq9nU<NoIO4RCs
zsZ7%`!&av{@tC;L^2Bd*3JKs34^y!=DTHwtzvXo?l4K8q-cBF>n;`M->GW3&?tz>A
zT#BVox;7gLy|WYh!WLb0vY`K&-49TAP#AYtvzMT#(QM&9kA;_hpSFr<tq9od!@3=q
zxLAB#$U^z)#4=m~u5cj_NzbTiEMvVL-`1R~i8-_RkOKmwR^6aqsR6{$c>+?A>b2fV
z^Aq2)&qTSWwzERAXlbWP-h`yuNh9*B;i9Go7Pgf<I3}Cwl4}$<yo^}5hb}%N#u(J#
zc&Gjiqn#eTaZnb2Hadws0wN=p*r@N;4EQToXq`uQ$DVMz?TE%UWe3wGogNX^4}l>%
zG%a&K{UuGKKHnwLF97}RvxBZNlS%&H72JqoDI2a*FTo}PvUVTuNYCY@(mu|ED_7x?
z%iawHVK1-kCsnHNxnDRPfqH3Op<3=&Vy(7+5JX)67C$X$*xB6*)ev*4<@jq*zg@rj
z{ZN8OrfH^Ufqw{<ppri7_(kw}Y`AAx!YQD_OkVD3wnv05#Q>5|K9-2728nBLWN`*%
zyZ>GE=24^YLGum;2hU|3?*dne{@b+P%K6MnOk&*)+036ZrgyNFH;S$l+V;oU%A-5P
zvhF+Ma9RKPKzgnTy#|P35~*00PvlHgv#76ZY~}<#xq*Y-EzDT%)C&<swQL+nmNv-y
zR|!-Xw`2e&f{YNhG;Z?mta#QGCgkBM*@(^)i&%T;+%#QoyZPk>ARgL<J9jC%Q@F+|
zl@ccQ--##M=M9D^Yo>ulNnbovFk8NeAn-Ro9T%7zbRTk;#4=5d2S^~bzsynM`AJA2
z=dY=nBh3VCM2)MT3^J9ZI6+Hj8|3JGb@i9z^3wsfZY+xi(cXLGG5b@ef9phj(~#Y#
zMo4`1br`cgIZ7$iE;V#m2<S;<(sHu5mil7Y4xn{#o<z29s}u~hA?oM!E5~f0KjrWh
zzJvmzb@xdiK<pjh-BWb=vE;gtxS-X_Eemkin7->PugGXJL5d%lT5PsP?5`T7mRZfS
ze&#>qC6dh&y8wxoao^IVKSSl{Wtcoj^D$BvF;!Phg+fRpjU!ePOvj2lxnmuh1=(T-
zMwIdY5IAuko;;a*78cb>98!gx;{MniK}*J_UQu>8-bRngEt+QSP92F2k(>K-clTMj
z+q=Wn8Ri$B&FThUqI(CKYgc>k(;JU={uO{~gj{NTuME*XN>0I%{Oif6x9ohU?u2iQ
z3{K8ExY-mBa(il$xs2J5?REc0c`9=Q@geYHEYevfAqY&#`D-9y6`O3<HNZtGKpLb1
z%*ffp@6)_M^@8$7LG+<m87!(ot7hle@n1*EoQ&jBup#!e|3rRw-QZS|U=dVA84o}$
z-GnG0G<al)b`Qu=$WchwlMjRr2m=b^?HKV)Y59x{HUM-WaA^O3dOrYiAUGV-9sL%_
z^Y&JVwx3Mr9hM0=<CfVdV-Cm_VsCIfxJn^UbpUj4aCjK2nl_UQD16wB`TuD65H=v#
zfKe2>0njH-9!JXwL54PeUBEL{GZPzrBz`Q)-+s}o{q`KJA_0PmmgNwQ4<SfX5FR0G
z#R0AjBzX=XglNZvw=808KcP{MO}rUg{}%u5Qy8c;_Uef7g?_eid@Gn+;1^p2LZ~<p
z&o&6ekhy_8tr5}Pmavv$xCu0&(C8>alyz_X1t3d9Ex(o~<OU#3LxAQ&OrQY1@fo9S
zCJ+n|IB2MeJlV%^qpDSnb_fg*91sMk#I7rz3)~uzEuBd)$6;Dn7Jp!|50E$z+Fk?*
zL0U)%s3#ELi)X+Z7|;;p2x}XjU4e)j^3Rq9&Ypt41K;i)2)hWW0bv|#CJ2cD(O@VQ
z9w2l42i!m;$JqcMs@DO6*H2$ZDQJkqMs?GQl_Q|$pYc@-OigSROdwQ>RGMp<+Kwr}
zh8e-aII$F6(EBz6BP0}f8~xZndry=Bbm!=)$i>)GA%-AJH|rQwOq8VJPdS?wcyFbW
z7aoYUkenavCT1P-VpPf$tL#`*k1fTs>V||&L}G6F3jnAqTH;Goz&#rvTQWV=8oER#
zVA8m|+C-}H3+;>&L2;r_Lo}>2{y67jrXM&Tg6`N}nYh%*?FyUowkE^J^78P@jgOTf
z>S<>Oh_~6en|wu)D*xkQVVSm@x|o<?%Vf;&D-X!B+e}QPjLq`(ro+Pm=`>vr8r&pS
z`Yqhrb_Mu9J|hK5@qa#}>I>5(d^y-}-0||+W0V+eX|=kLvF4z5RxGb{&tN4O4kfaP
zo=;>GbOuF;;{>Wg%2i?1Y1i8TzMB8?3Q}n3N?LQM5~SOgs7E2O7PBnyQA9vUKoj6d
z!Z6Bin_~IU{<0)nNJ@nLS}&bJ0@Z+#lEucl&(fs<L~#`joUCBN0D`*(R81M_(?x>l
zmvD1l_eLT&H51FY%suLc1J+rRoxsP$NP<yM2KvxHp(nXg8p1we(cOVz6nBtt%fi?C
zsfou#8-;`JM{PzV(c7^z(09aG<v(x~Omq#}d2gw0a!+5sRtQ&tQ(rNiNe0+WcR=jv
z7>KUhy(XC)&Gj*Dgc=<lf|<CYs>uiwGfh-cwbQ}HrXBJDm-gowDPj|(!v#8&M@##X
znKweo$qB~bA)91*5VqPf{-t^8p32+$VO|=yQj4>+H|ubjF(;=|KKM6~JMT1OYcEOz
zbQ2O|26!hjrQ{+|QRzH3+6&;^)xA&}4J!oIN6LeAFL{@lqMH`~m+q%aZnxNVK8(iG
z)9{!cOsaTbk6_dk(=7d>{K>HgY{$rxn}@$DSCw--nv0$Cr;-rugtLJI?BT6comhKv
zyKW_i78q1wVsgmmRAq`mz%|t14ytt=5+-qFe0GADM5@ZADsPk8)&wXq`j0Gr{4Hw5
z?BI__A?^H)d=TiTI0Cv++`%Lhb!3c#xQ6l?yR2PIaD4(5`Jw7d;Gn|gB}=%lG?0Vi
zzbFQNnrq?h^Q1a_CiVxSWm+Q<BVui|@VsJG+3I!6L-6sW5Jj;3F}^BMTiqIu|GvhT
zwxs6FKbGS-9_XqYF#kt`zO%eV>Eih9S_RVWP+^L`t5YZj(n1aOe}wD)>1j4!`aOo+
zhgBQ=<;ez(kTF%YKqQHD0RFYhNl7E+U*IjfB$RY3M?p-_23`G2mlS*u%4<kG=RGIX
z#pKmLwVU%Y7l#jO4_=Mj_y<G=Z6yE1W%^p+A~(oT)|+}jLw&4(cF|s<4*WW7gHzM;
zuL}x{Jb?+mTJ5KzJS%E(9s!>=bW}seC>usH?qiVqs7r)`e8oHT&m!#pf&yBgJfXT@
zh<O8f8V@|2PlDUqO(GUvi<B9iVBihgyCac!i+HxrdRJ^~*M{AM9bpPtm)&9}Olw>m
zYkVfLnu3UcU`oE1xf(Z;!;He)))@W%)b*Pgx}w5<jop$AEmSd<^D4V#&iuOpW_o8c
zSrzjMa1)-fG!7ySxKt{LVWmh@C!aN@<mvB?fTz|@FXsxR11PzX2a|v|FXK!xtr*U;
z>hxCSiq$v;;V4ro@qw$&P3xiB4kUvFlOWfLWcPZ2cAq(Micyw0lXQ;5zDECOD27&*
zlNh^Wp-1cTo=Eh*NU(4Y#KwNfg<J~UHX0FQf``R*O?HJr$+{wG{k}as0`r5M`>pwN
z56+mK%gO4+x-T_AX`wv3F33CG8FF<<1%F?cm2(W0%dC6D%5iE>_hBGafcGaye~z;c
z{^u_s7><MKYv5(G`tBrOHw4EArAsw}-*Z(|qdoeF?vl4a8=OwG0dBW#dEZ?15s8`r
z6Zwxh;fy|$4=Z}skJv&4h=Tb<=PA9DKm3lDIr~8H68W#3h%EC=nvFm;!JHgU3tn$E
zT;y`m)vcuMs8PMy)}9n^8%e^urko1E0pyAr5`TwBZgKt!#9O~SMGP$EuRjjRE)hzz
zq=5=*Sgv!&K_DZRc^tR`Cm><vOW<661^V!=zn*NahqYAj<2*sy3i55{F7!!pKEP|*
z$FU3B>K#wyC<TM{@01+}ozF#iFaz*WO0)UWN<)%{<_QW=K`_qOI0~=3%tel113WEv
zAJs|Vdy=iT7aHjDx%6-W-%{SomB^A(8}`8D#+phFFo=~p;*_mmaf1~q?26@V-|?Qt
z%Y&#*>=skUw`qw{a33NjJj;!N`fPd@1X<sD_J<L(p~Qe7iJ&l{D?gQN&?|Th7jtW1
zJd>Z}(@?$dYIJfrmo5AHg>}UX5NAIb4zTyU%_-$uUa3=?3$)#3<#E2s<j+!9s#apH
z@jg!7VE{R{m|aLCHRjb*>#`4wa$U~x6bATMHx~34`<DkpFymw%BV|I!GvQ;+U^@{L
za;YJhn)to!<&w<}eydgcy&p%;4)~xhyPswUr1F`JP7`~_20;*(gTKcDbdPa>>J3P(
z=DRh4R87O8QoVd&G5y5{F}#I7ba_-${>ns#iZi~BgNt-37G?Uv@)JkjX*rhOi!i7t
zuB0dd>rJ&+R8rxus0kQM)0f6(eV2Fz75)uB5@t_L^Yy4MxL&>bUJ(4#^BysgzP014
z++zVHkiSL#Ow%s?4EuTy5OnCW6BC&IlD$!emiZ-y>5+>un>tkh2_=-r>V*mB<4Lof
z!)_k#bq$-A$_Z+t8HSfJ$2|-OX(Y}t!OP^9j1v9XDqX5&rTTG&8gyOB2QXDU*0VKp
z(W_jE-yo^E-jczH(~h-V;SqG{ZXkL{%z&z5&x&Jawx=u;^|!17)b*3VD!7Ww65o2|
zjdoxHJF(6X!8m6S8LoWeL<2{J;nDFB;W!VB?9$;ht=L3OM@%fSil~AW5Yj&r(ST&E
zgRvr^J3Ye}kuVSu(U4fHj<BN=u{bvr<y?R$Grc#J@LtfeRd*k9)2ItptpGu-0~v({
z)Y~);c)7wUCgDp1I^yGi@0v-!%oiIwvG#dIFz$1VB_{*Gn~VcH^!GYr^8@d4GlfE5
zuoEzcino^Whl(Tj@(ylKw<Hev#9&%aIx$X!b}lET{VYKv^6~FDq3m2!<DJ{((DWJ*
zJ1|oU?t)ohQ5&XVKl{^%acFb<GGQmi&=@7Ti~2$#c-<EOOn3=XnHf_iIlxTG_~u)~
z+t?y2`wbyqth|?cc8MT*!7lOP$WK~Pf)R5>mIWw7PE5#jgecV2-)LNeP$LwqZ4rih
zGdt?b>fSKCDn*nU`;Zo<l`Bgn!35?O8tuIdjdGRHG|v;`U3y3V9^}eny*VF9>7Vm-
zT!q_cUF$&rfVj+(!voit`_iq~ik2JrHw!@zhpazSl${={Fam`90wxx=hr*Mw?RoS)
zTyT_`%$K$G^v|ZQi>ccVk8^O7Nj>e`s|)QIoI#gD&-vF?nfO=yIuH4Y@yOP4KL1d*
zE2|M+(;1%24zynaNk!T&M{o46A$wm&4>=eh-l6CKg4};qu`ZjdbHfG&D&hWhdIEe^
z`zIC{KsWhWF{F?}=dybCs%}zyWYh<<sh~}g!Hrk`e_NX9W^NJ}MMVY;bu#A|1lFQ&
zLgs=5<WX1H?5ab#zV8Gob^mustvtn>%^l&0t-d~@ByXKIDMkQM-TkOuI3F#nQd%Nr
zF#s(FMaD$zW#gP*USi`u5EV-OsA5O?_-iK!4!&>L^nWR*$tt*`A%NGxq4Zo{jPlJa
zHiitD`=T~?)ruMR=^Xv`aOx5SHfD1Lr&qX1B@R^a%`JDknWT36U&(2QM4zpKnWA&H
zi9?Cs2dSas{&;l`%vkfxw&w2he_moC04At0a>%i>hT^wV=J&LML6Yttu#b-ZC6ck1
zJ(G+<AI(0Z3NqOM%D)8L+f#(-B=-*W8&g?{(;q0(ni~P!$s{l+Ipx9MNR0s3pIq2}
zLesmnSE%BZzTwc3Fl<h-lkCon%ks1APA#PKP56vM7U36yQPV>p@P*}V2;}Z0V60v;
z$XA)SawfI2do9{pt*#eQROI$wuvkz#_Xn|2wzU;Whs~@Bfzf9S{T1_IF|1q;I{{gH
zH8&@rMv=g;k?4sevId*iJM<twIxTgHpypx4R%AONTY1}wj7{*P^if2BLTBpMq5J%>
z#>4VXdNZ}_mT`v{yzV~W=H}`f-~#xzDU^K@i8CK#mBHaws2UC%X_j>!Cn1UaQ2ei;
zRR4~do*<Zq|C{vu&7F!zWzd<pZtX^v3mT1T|J~I?=Hlhh9`Vy_wOi~t)Zk_^%RGEt
zMfCj{jrG7h`x+6~`Q}*748(;|G9jR<>4%A^rj{zJLCB}P$gcpX>j9*WrCeHa2)dPj
zE!h_3%Ks7K$IBIq;a4{|_q?jTW5(^Vn1x$8W_p&u*i9Brv};Iv{Px?GF*b+iQRlDv
zEmuyNyt<3Ym+0#5ZXk51a}6uY(U^Yt8O_j{Xg)>le1YQk@$&jsrjh*#e>;pYjTk-Q
zjIqdlpl(>3T@6HFmIaUjiAUj+zNH#0Mq4d56*!y^cVWY<=EI5AevkM*f|e+Gkepq(
zq~t4sEIP}+-`jS%&)D0Jb|Lb1TJ7xWEfIaWj)Vz#P~<{HA2@=zNp&c;3E!!{jXTFp
z4h=@yUo+aKHxriuEnnm@IW!m^b0Ey_{)0~~aO+OyAq*EP&J4g~@HI-LQjotm`m5Sf
zAg&}*6oC*etNIH~yUDu-Pfe$g{<6Sybr%o8=mQ~EEjT0GmZ1U`=x-QKTopE3^H-wb
z2(TDbP)IG(fYha2gj=yxZuf;vfmfbYpI)iaG1~lWB1OCI+?<sz!;RYw+W$Vx6-5%X
zmWS1?%P+dq0ep&V=6t;gd?t^1I?aL~B`tSoKRvi9dY9$aoPOKPB{D|2y_+9yE(?hA
zcDHfSH{JHjQVH~Eo|h;}JXYlOn1?ZWxujWiIxe@F%mcB%iq1|fh1n<#+k>nCS^KUa
zb<_lnP4DIH0!OUppn(vNWzVw^n)JY=BV8Tfn7}gK0bEfq>va5+e7Vq>Rx?8~6<L3O
zskB-K%rO|>SN{F6*LaZsL8ek(e$`VKUg#9r7XvZwlY+wD3I5jI$z|02%oLKr-xO@_
zwha*ZB99#ziHHsv(iK`TZYJLcGUNj4Y(Qc*#a1)SI?g<8j!N%*9ViM}90PT8(ujrp
za|b?+2B2H!)GBk@1DR&Kxz%o!g&bwtTwyV|ROVi>#n0%1XX2KvPK}BzSe*ic4A=)U
zh0v^pG;1Z`^K$VU64^==Z~NL?{a5f`%H)*tCoTd2ZASNBj4|M|F|6RRb|V+-TV^lU
zc|>hR=!)u6t|V;b)*kKD@S<2d&d(<gKwARXrip~==QZqaVD0oC!V9oa4bDx-Q;4hT
z-#F(Sl*Qv|nEE+f#!I&A%+~1PT)Dd|*chDn1L-LaW&`zdp?a&QZ1-Fg&TnTQx86tk
ze?)KfCQZV!eJ>gRr)Mk#`ZvwqeS-b}*zUA^59-{nPW_!Z2KQ=E{yp24?GOJh4M2Ml
zl`&|havJJCd1xg;Ra;a0y3~J4DLwlmb=@+lv)wHZK@F}b(?_@`O8@#pMy)4RD#;<|
zarkU1Kv7bAuB3~d(KZ-`xTFcx^(g6U!#3oF#F5ZLzQj^DN1lo!96>Qg>NDctv$cWh
zWQwrGZcsgcvi%Ow33zf$@`zX;fL0oxpOB*jbX=8bdu-x^s%gfWfE%gYkZAN~4HV%u
z6xcmCLHQUcY)A78s8dje?Mv526mQ4M@#tb@k!j~j3hpUW&v<>LVBcqzTE8DHYf=Le
z0kMKA7|O&jv`69Gc6r%S>Aps+#4mFtaU5#w(;qoAu)|troxa7&5l*)R;Ph0cFN$LB
zHLpmr7)bOuJp+wCe1n1lb1QC(vDbWR`RxrDM9pi5R|slD4j?p^AR9R?#5pBi5EaVv
zJ0t${37QMaBPo?T+!XeqYdC#*;VFyAtvME!&3Q<ZdBo{;>r-gr?_R;x&;$Pk)!^6w
zhcUD>*%%tC=qQQ9_7m^{{4e3NM)uv#aKzvxEbu5^_@_;tD{vqDlA7zUrWDuus=VmX
zFL!9l*T#*XI_u#)Uj{l+{l+ze!l9*k7Umyu`w33qx?RFtRZ*RPkh{RE9keYdIsaw}
zlm6$#JuXA4uE)NgCh|tLm5YHxNpCL5?MLiIk3l{VVlsS0RS5w;@K6PhYeb5eG^41{
z_-R+fC4YD>!^c-jdiSB~Nn_W#=)3+>8C}oiCQ234Zre;0<*|Q_M!FOAE{cK1S;C)O
zM(K2Qzejt%@%EUno^!n-o1<uFzvx5ti?Yo!Gt5JXH@*9(6PeD+$2#qM_PfN*DGq&R
zC{TZP;e>->{7pcoZw7bAHS96z{MIoOfISkqLp?CStb<U{9J34vj{(?=7^aa{W`3U6
z=%*0rd&{c2rqroZhV9u;wsm>yA>|`hWpCv65~b>3W`<MEB!Pup_8?XuA=>zG8>%P4
z<4;kASQ#8>1#e{q9Q#)(zat*suHZaEckCAkUcxT~6eWP;J;G1x$A~eYR@q7&2OaPp
zQs6@~e-o9I{cRM>mhcdgP9rxXS-&LK+F)QAL_qHnlh=SQOUfffR%(!(LhYzrp*mf_
z@T-r@zO!$Jg&^oAf<$YFWf+G8ii0W`-WnfYk1am^@YbjE>x-?$n9Gy*yF0Du<B20Q
zdEE8Lz5&3a$tgKOG3ITHQGfr{p*~F&%{hCZmKtz6<tc_ikbM|@kt$A=gu9yj9J@es
z5-j3Op>QZ&LXV(Zymyv#S)^>E&T<~uzD>HUB)HSE`h!(?e6VOe{<Nt*rltVAY8+1&
za^Q$!vs$?kN3U>HHc6Q%50@b%i&2h6BwkOUJU}yP{?BCl6#M2ICiu_PsY3cA1=sY@
z*&{@@$;lY>og6fv5zO?LuRHwyA|Bma@(GQXEj9a8d6~kq-F_vTs5IgP!n>e+UX;-K
z?`U7Y@~yXR;^H(h%oV=M`~=4BhaeO*A|#N$w8g9@7=ejqe|F^ajoGQanDsV&c%vFG
zQow?i$C4+A`cJ;*Z<((n35Zqt;%~Whspxr5{4}8D-q!~Vw^RjAgA@*$9U7RoOfO?k
zDy`#a(?>0frBNH1;-i;znv-A7UTJ8&r;h&c9rVs#FKVwR4)Uxp?7uF?S2vXS@ws4q
z(Wa4eQFFS&crj~}CEiI8Qx4giQmxDASpi67vkQFGM>udcY-KWkZPQ-TQ2TGm`$XFw
zzSrMzj$0CKulkon&D#p5j9wZJ$aLgWfs9l_2QeL4T#~n_u|sBY6MCfL#&5PiZ`u{A
z!BM1qCs58YENo~^grMB=Z<bH^ECjGOmN8kd&%#tVjkA3#-*!bm7b*S)o7dC-L=?~&
z;+V-xH_I`x?10C?=8S4<<2756R>A4!#lQWjhsOJL6LO5MnJj4)*;A`dQlA5#=7Fms
z4Q8%$=t&QD(e84-YBOIB3Au}+KGEh%Q7+#Z@Xhy)#w)cyZ-u)^{{G3)>2^hr2Je;1
zHo`kCEb*j;`ehk{myS1PO4*Rd$pqN3s{0wQt+^V)LyjN1-a|FcvqTBhL>Vc=0H1B*
zTfga?&2v|JqcZ5>-}Q@BJ=NN*_PbJ*yOv`<nOJD#JPZFf(cD3prQmQ%J_@nSa%}!#
zkS3%0BG!CsW*%dXBDjLvCm2L3?*n-$32{hzK_$$gJfXEhnmj4{10;6#1317E3vH=M
z+jhXy^sbeijnIRteJ)Fo+HZ%EY>;VAv~e-*1B5b@EBf^jcHffmaF=+MgGIAJZiB>&
z^9#tx^wNx)E1!+~4aOijWJIvRZ4I>IFyP$XpwHfeksC#laIB?`Cs@DTNk;=WJ2sDu
zZ9~(X{B3x7{MvGp$bMuwr4Jy|ASaRYCvb97lb4R}h`pz8K+cgwybdK<xX8VLAnn@d
zEDx(%!10Wrt7|e5MdrE9$^N6%Cu1l&c<w#@8sDb3G(UNiwMZ$N=~cQ(Xy*l328G-O
z<tRFwnYx~4&&N;l!|SZa-=AQl=}V!({zY*LlJ~E;7KZzX%sR4Q{uV%RA>D!2YGeKn
zRq+5miX?-`M*Jr?IAsv3W=B3D#c;Ty94xMZ^87A!b%yI->fHzsx~84j3K<%UBkRmg
zqRb_7=C{3~v!q_gUp~z#e=jn`)Bl(Yiu90eY4mdl67O=>2gu|RUG@*6@<26@ypCk9
zd8j)vUjM$KDQ|M@w3P*XnVuEfGIiM{L0*;5MLC-l55aa4ZQCXUdO_B-zId_Z4pB+l
z&tsIX90yi86_r1Qt)K_t+}Ku-*}=IW7s@Q>*@UrkoHReJe)ct&1FF)dECMI?#xKq&
z@5MY4U((l?XJcq(GE?!G!%TfX&n`A=fx~Wd8olN_!=z2oo!9^kg>|z@b*W6_M6G?P
zEakyG8X&o}e!muIfx*?Ow7Gr01(c|xHt<gwk)_q$iv&nh<X{%iBTkyE&S!Gx%z7_x
z>vRR`VbtROblSC(Hr>5yX;E_~#`fCMA&-aU-g%wX(U`TSAPI@H`GaFNo(3q|df#f7
zdh6FkKkfmoS=Imwr#<sKwX4|9<02yNXPoUCjoDva#So-PFl{91XZPZA-j5r7I)nJv
zyF|Nr%`Rq}w7H{GoBnF9Y0S!RFGZdjg%9ATx@r}gf5KaJJs2v8IJz9^-BSV~nYt4)
z<*s8l7$s$L|1nU)%_c65ydq<1NgmkR`O}_aS^+DK;vNeSyR7-kAKg4^B#Eb(n)~nf
zcr#jg1724Xxq0cNEQ_%(jTnSav<jyMP@g6S!;w%zjFy)4Wl&0pVbCR!kyMd0{%J(v
zaS_l|m4{wU7pV44MSRsiZ=k*3#GT3D_x6hm<~*Q}NUp%VAvGcyeV&V-qDI-KTEco~
zu3ezcrM~KbNkDZkgAk<qij3Jpk%OZlua@|ijE&<cQ5)W~pKx^_4^c3KR65<>Kh;~o
zDQR#Bp^=vQ1NoeULhfAa_%wM-K5^$n;7Ea#Dr@?N_xl<xwbijiPxK%@<b;K$+3_A5
zOJMJF<0F6o7tOhfV@OrPbdXbrGXwA9ZUKj~6`M97SiSvW_NC~Nr*`h>K3X5_z~?=`
zM7Doxeh`0;gYdN;tkBqN1U5smGM}2g(E6eR$!ghZk$-vf=<IVyr9LHlHN^SrPUM%a
z?!XH0m^L`piAOo=m3`H4{)VL(-DkPLpEA=joiz7)oky2<tv~Q@y=zY}VKvvsQ2#oF
zPQC*GGpHUxftW=BeJ7qkDXGqn{YT1mxBZv>apE)$^+UYq&#@wphMW@rCgmHl`8G?C
zE~98+>-@GDlhC}}?n2Uslx=l9*Hs|fdl3v{I5{)l5YIcEDaa(hc9fdLv+#puC@iJ(
zA1|bJ$e*sCX0L>R0jVu-j>sNr8duAa-|<m^i$(^KN+=F^(0Wh$93q8~z(2$uIG~Jh
zJxOd_1fM7!+(G;dusvWz9_CZ-A+RYV$?cVSeHu9s7-Tn7Wl&-}SEv1vg3diohEQt>
z?LpB8s#7iYjFb9Ns3u%~IA^2Z_49HK%fUN#mImP!F__$g80E%^o=Bmmk8<YzWGDsz
zh`$a~b=PcyENqoY^g2fw<dbY3OiosY^T*!5tW}45Ao5FR!3xzwVSFL)(~hhX+?y8a
z%AL;}>Ch@G-Hy_}+7DA~BjGGYVCSr9^oHT-idok<yfsk%4shebVVYNdL4krS4knHd
zM)I&TWiI+ouvXs6%HDMki(jn4oh=XoRO}+p;BxLUNxe!>9prA$5C|iFXO-WmsiB49
zup{?qAZTt?z2(1c<-xU5n*un8-Xdt|*OX||IegiRdF<yleu1M1n3ywV_6VXFK>Y3J
zxw)Ze>XIu;*tjA;RK30-EHBgJCgyWnv2C{(n{!<40#x`-&NxNXyRKffesymGu&J*8
z<;#y&i+s^BB<b>&#**8d>8;VO<5fLvF|BP(cuA>tC@bn!@R3o_l<Z~V-z%Pnd!$Oc
zg!T}ooQFS4BYAzMM=R`L83DhRLEUEvX}p@posU^Ix1!r8i(;@W!(Hj;pRuhlkK4ZG
zi0eV?@dN>Noa%SPSclSds`()U>=8cFNpmWg;-?=Wnb}yD>~dS9S@idOh3-k0feO|w
zb`XvEXV)!dumpM7PF&v5Syp3x-hp@Ut=s(jgBg)$_+rnss2g~RSRN+W!EI%NcvBjs
z3FoG*&ygOLmLEi2-q={9h9Ui4Ov1I`2&+<;-m6}sKOaqHzn;(p@`S4ZEdxy7Aa;(b
z9qqqO(&n;{qQDNhFW2>oPIRoAHCKXz^a}ONc*8!}SguG!gVR%4LiLbB%#fdI2ZmyF
zXI!+niyrARGxP4(d7pa0y6hSYVouLR<%eXFPTH^)h-r2Cf>_Efw4$O;RskIDqtW|K
zua7$3t`Y4f0OB(?4qR)10tWmPTkz9W^0_bAE#XRg*o5r?oI_fDZNE(qCUj{4Fylfh
zcUnrr8D*Uw`F!#jTAI~GT5bG9ivB5QTmKE7dzEyjllzQOdG|?!*#|WOB|_pVieX=%
z{*su>fnS(s^2#rCVE=9NZqr{PFZb|GnFlvERG~JAv#VMpi1_t@E1rlg308jIiCVgM
zS-&9U><Efme@MZ%k%BfoT>2hNj%Qhxq6Kk9#xY}R7f}EOeqg44vYi93Dc2wN+wW`r
zwZ+W^HvjX-L^)9oeM+z3x7$Ztg)a2h53;dg&YIm{GhzK*+8B|Hw0GrMX3wrBVzqat
z65W4nRQ)~;cbhx_RGIK9^K_%iJhe2DWfRY4E@KOddz(s9qWBe0fg1rEJKpVnKHHO0
zt+wrx<1-)#61|%jm7m%6<%{RKN&3$+`Hw3X*A`vZz6NTO^?8*K;s<el<L`NDYo)HJ
zr3Ut^`4VTxg`N^Ql^fVUPq%T;E6@NfX@8xnZrgJ6jMmHmfGe7o4<ky<kjBX<Bn@Rb
z2{)RDT$`aj9^I5*`|_OSC3}18^9sFX!e$B!EjB)IT74tKa_H`rH%gqUMND!~-9U0m
zGzbFwfkzkM4@{F4kpGIK%^aR~>-1oawKkM#S8CU@iEA$(pQ`gfqcLENa}jY{-z-T)
zR;C@4J@GyPT67KKfJC&LKT*SKWOW?cSaO2?^0`jz5C7Az^eA=yWfwuF#>*}aEeZ6t
zwj618QoeC;@HqN8uXwMT;JNvs76uu|enLtB--jtL_0T5Dpt;)L=K1u&ge_p3w@sPX
z6?q>2VC?xkTZ46nWiv$1oJ1>mJxHNK6sFfs9pvy1m_mT<Y2{Pitr%24+$o1c(UR^#
zU!~o;R+rqM_y$7~Ah8~q((2TXtCWFYFT%oE+#n87Ol{KU!Ia+LlKVXv4CXD6$=*TU
z@}=(zJkd66;}k12PhlbwazFGeCta;xnfEiL2;o++!EC<LZ}V|wRk>6N)7p15iwDUk
z8NIp^0B`4F-`TtYj?qnzf4a3TeHeJ~#0~C1?}B2q85>52%Wy3ZDad3am|}Y&xEOFI
zT)dAGu!UU)Ob<7s*G9<Z1I$0LT?T#|#9j91F0Dn;+~m0BhY&@DFx**1e`3Z9K`f4O
z^un`!uDi@Oj2-)y0ecqO-tXIekh-FSb?LJMOe!Sh^e7y^A{xE;{sHEbK?!o-VSZ0A
z_WfakJtbd3m-oMFG}5uZXY#y02fx=8bp1^HxcHSV|03|Bw_<{py#1<x!hAP1Y?_03
z=7M}mT>2C7XTqpO>uh;frB9ugZk&k!c{9;ZK$GZv%JZXNBNO)e>4M<xLPxdYPQX<N
zAg`QL9vy-=<K|0J9RZ{IEq>-!(3{s7k?ad%Dm2`pp&+>3ffw{rX)fBqRiaQY2gkyC
z_9v!%Jy*a&xYww;nkXU$jx+)y@w;mPLTRX(v@*6R#lp=JYmtc{-Uv>IicW=3YLDJr
zm!&H7cBfh|5z9q~q?6=*n$s2=5H)}gKsjU%oPb*|?SgEA#TX_e`GvtPh*1t_UN)?a
za+5XOn_vLtpmDE0VJA)@JSZD&Z$P<vvkQCJ9(k87!;KL}Gzmt8T^@&o)9S4)B1u8a
zLJ8^+J(3>yGHC6^+eCkWFc4+uru=6`i8&lEq)>KEDd3k#Fm=e_W06X3!eFX2AZy#b
z`yHHx|EX>1rrPg+l<ZgWs)oJ2YH2w~#1dWw|B{kACf&k@zrmGQvbwr%L8EW=Sy9aP
zKl}<CqCaxoKlAol-w=8y($%}=c(#EnnTxhtQUzBf_DO@E%%Yg%g~NS?A<>SBx0y7>
zl144w`!M@hJH{&iS)LsRX%FcFOas_s2up~%6@gwehEjtpz<{i$5spccFkwFud!%+c
zCU`?;2=W!L)}E5Mz{ZQhLMr>8J8)uZ5roj0qi`0?q+<q>nWMDcR0H-=<h>&^1#8H7
zI26X;<B%wKD_oLE1<i?B<n-ApNfZM!SE6CZ3;dr>|J-zI{xL81zRny1G|GK_q!+uP
z-$2DD_HpG{r>U`>kosMlK--7n@A>}_xYpYM1;-$s2Z(->R<a_Dyn<*O+I~zbKMWx0
z1&y+8tuCYCN}gaGlN)%#mQIe4v95fc0So)~60CCbnWUnvC^<rknXN)>)|ky6b9Bt0
zXm1UyH>X}75j@H;-HtMV)3(-=SM_cvmC7KTh_P>@Xb{yM5WC(z!ER0LBp0Kd;54J{
z7v{>=WJ?f!%=APF59Wt~7PSh|_kwx+_e_Y1XqxTV>5%-CdXg8Up3Ws`W(7#u^S2bn
zG^yw;kZKw)POKA>j<F>x1ktr_0X8Gt2H6csYlDznu~D{!0i6S2q>sIkD1i0)3AuNM
z<5!&~-Ph_Vf-}e(F5Cf%3uWAakVp<ZV*K=;kO+(3<=TK=HA<2?>Y8FATA@{g)w3&;
z(A$4-;-dFH*oU45>HIo`ThRqcoso*kvfALROOg#EE`gH5<k>^ocL`AV>k4_*+hPQq
zcRe9R$hVy+3%IJwzXs!(>&81XM3B%ni<ab(RM?3joG^bKL6GSREf{K&iSickc2vJF
zn2vY`I$h8CZkRB=)Wau>;cdU-fS;$p$UOTO%uVt;_>~Qx`vIQV)`i94c<qTTkI;1(
z`l*6|D2;phl@f%C+Lbv8C0;HdD~Fkhn^vbJ9Q^v;93aHZfXU6k_yP%E$j%306QNAS
z;m6cXor&>e%GsMo@w?nYt33*AB&<XCY`JTNPi;p{!?HJN)*rogdR3+5DRF|m+)exN
z$n9T~X)~!H!@?{4qS1M|ghW>@;#H&-5*FwbOZZ00p=4Ph>aZ_G8#Maq@8ulyyk==+
z7lCpEVgQ2jZV?`T)pkASv~_5x2(~JQd?fsLt8%3}4{hhlx_mvozA4q8yZ<yE@DT1&
zPBtq?)4!2SSLcbr-_VS&s8WP<Fy=<dkXMva`K+C7#Yk_{__fVl+Vz&?+`*jUS)y(m
z)7~?b%(<C$Gi>C7xt`wf_sNY9QJ=?*bCGF)T~#T$d89UQtABV$cAT-&4|=)riuyG7
z=ssu6v0u~Vx3xKdgNp^%*j<KfFV*)jT(ohB)Ey0JcsFsp`5le9s?q|z2Lpe&?teY!
zuJ^OZmrlZ0A|9Vjeq04_&fz<4Bd^4)&EQEV!BCI8(5m{n2>J)>EtpL=ATWEcP9$J}
zf7j(CuIt<sNb!H?B9Tg;rs#In=Fm?QSM}>RgS~P(QCa+1G?i<h)}U>5Gi*uPck-$q
z864_Y=K^$h|E+5(g|_HC*4rnytj-<nMOA#eRkutQT=bRYH*ET{)~)18YlAMGKg<gZ
zF%)XcY1Iy`?ao`*12&%%f@#;*o-VWi%~e&gq>J?+O6&iVi`9Zg_WPq#-@eUo)K9OT
z!#{G7lKgn9ap;k=0!9@l_IJVRl`m@_Rj+)tOXtU5XB6Jo@!n4b<uZfolRX5-?D!9j
zAPE|8vZ&CkC)QGm73?)=HIHShMBRk4rR(Z$zFq^*^Hk^NEX8YvouG$kSSdg)%Ws-n
zR)YVhWPJSOfJ;k44g3y>DR#)S!fAH0S>+$AErd<afu$59%E{C=C2m&e1~x59AeRWP
zpUdz>7Nh4ap!l1(j+@y~l6(@I3sFsccSPQY`Ux_s!4l_15X%$em_(Z}2uM9YZ1p{B
z^nQMmID^*=gDcvmA=A+IQ*TftlYT#zuB;4J*Rx8tuP^-fzbk%r$&!=vDs5w8G~Y=W
z0#&yAMT?dzxyf`yTw>H~Fo`9aPfjuD5X*~V%vxvP)7*tLi!~R4FB!ipFAWnDMn}6K
zX)CN^bS#;9vCKsew-+fEQ2!8epoBWyLP&KW9OM)!VnN>(-LjuRV=2TMu(k4dC)~Y!
zHK~}*vj#~e!~(sL7)~@u1xK)+Cfg*sJY8*>guw9mV9FCJ`;c7vL(1#4fzC#^B%Ngm
znKNUiG)Wx7yqfDkA51?+ak4V4NjGn=$yqq#-qL1Ix$IbK!a`91!|)M*K=hg*AQ+%?
zNr*#bNHn-B`X*yO=%V6HQHiOn4KE5nQe1qoX;@iZz)(_kI=G|Hahob!;5_bOYWil-
zWeh5{*8U9~r|fE+6K0Xv$(wJf$v2^Er}kq`=%^on$jMgcCtA#spqdyfw46Pf%^vUv
z>T3nV$sJI}nri_7wWFj4MX~!lr=?d^_S5vb-=<Fbta;`hk1!9Ci7jp4Ew~EJVS|gN
zvcV?hg&rsQ>(@oO)TI)B=ml-R`TPLt20drB=ns1Ss+lr<)I$+xVfze@T$3N|281Wa
z+p77(S_=gp)z{9-7!%0*f$Lv*cl-F=c%aQM=+FA(q1Dm={Q|cS>kGhUQzu&X6l<fQ
zvMMI=WYDCc+BpC<wz`13?&@$9jUowil9tS?60`=J@pW<e0Ii6^FGSq?EH44QL7MS^
zBs{l}x}Kt8Vb<;m?i#C>Hdo$5s4ium#%gP)MaRvofJ3Uc_VL|1|0#%trRblmhELvr
z?D!7{XuJflO-3$&(=;=nM+~fidspO+l}H5X6_hHMx)YRW#BtL(bBGfm8;0_Z0LxF=
zZ0e+=7z`9%c3GxCVW;k)p+dxN{ac0p7R~1WN(TBIs&~E)dK|65EZzY-i&3igRjQ9p
z9vRl{(6*}{^w9L3j!u0{<Q|ka#Zy-9z-VlowOazj?j8j4#y;)B{&~svZ}Rh*J?-tt
z`KIcYJ49%Au2n|NgA80GlSsyrGiJkX*XRCgndA^9L~=_bTZm<qle>eoZ;4mC?Dpw0
zFWWhbV1Cc*SdgHKqI}QAE?y;O$Del*mk#H)fUt|-OA%#nx4>aB%Tatmj_Os6LoT(c
zdBF{Unv`8G2hk+H7NyI90OsOiHB7B20GY_f@6jT2K=KmL7!58IjA|fai)+OOe6|cV
z?lgH3$g4OAaOP^d3pI1<m>*j6nz~YZ4sgT$%62&zHgzOJzNJreXr5_@yXkF$+~K9r
zKrI%36v?j{t!|XhPl+(;_s>;TPXsbM_?rp{&Yt41q;r}i4VVsFitt+ad?-jw<+B8T
zD0sgyi^7W;gf!7VlQkGSZK&C<9*EJW>M?0`ueoO!@fc*2EMuQVpng>;)%BW3u+yME
z`xe8H9y|ZMrANAS3M@%>`kMRbo)CHz-jIQWkx0i4x^}37OPC2EGlihe++Rgaeme&k
zE}ZLLvyw#$W&f&u4}YNkY-%?H*BcZ~5*$GHW%mS&>f>`Kiq%r5^*`Wva7vjg3AhXr
z=XuACKx+j_;@s#{dawX#T&qu&Ze7xv+!mldC76ByrR;%r{0Fyeu1~NSlm26W{vwbz
z5|aXt9N>}s%X!H(lSN;ksx%epzJCp92AZNuVc&XoJ-9NnbJI&e&pdzbV&32i6|Ebd
zK`%-X$jM7TqQ*PiW9O}zP{21zpUkk$^iED2Z^(MCQ0;@F7Bz(+&M+D#%Y<(R_LyfX
zk|D+!9kzYYK5j48t%f6-1QvB}?rZmJ-4}*5Cmah?islj14o;G|c2Xet3nK}@A@D};
ziXgH@$=8M_=aRKhQS-W#(sVqe{@nCI9GMn+>#7j3iB!>2>_O5>W@d)Zz<~b$$U4X1
zPNM$H$F`kJY}*stwr%?t+nLz5ZQB#uww>8L&u;DhtG4<@SNH3#dvAZvIp0G=_kf*%
zAb`q_xQNHVU%2xA41U{8joYJy+!t$<!q0@=PutySpH{q6WV~I3Cpzo~@k`_L5A70&
zoO;SbFrBwth-_X`%G=?0K#va~%D`*f15qX1JR^VWf@P}?^SFCV11ZM;;buOoH9>!@
zg@l8P1h*B)Hl~uSAP#A(i<yD|LHx8V!{ap%5x{sWA>5dIEZ0UBPA<n~Jsu*4p|dd-
zcfmL?ly@Gv9gn;dm`dK>`|m<A^StHiMLsSOmvj$=kYQ-Q%Q@O-4mB3QmIS{^>7j{N
z{x<#^MHdsmmIPbN;3cFHfQ5J9us4=10lUG=oO@m3rPvthLnms>&6fLwth@cQs7Z+}
z$WtupDKcZXurs})tYTqe$S!dBpz01_jHW#IDE*s5v$V`Nl`HtWzcSh`nQo0SFX1q|
z?fX!Mz}-58oX;-OVG#}p^=hb?aRv40FU_{mOK@Qox-3z5km4@rXkF+9#huIh4_tBH
zh=KNC-4%g2qwze`IJEfFHR|aWdrS6UA|%H^B%u*H@Rt6}aZD}xD;*VT%A6O160OH;
znHb0nt%NR-gwN=NP%{?ADv(j$vkT-Vef8RvZ8h9JjPM@(-X7p2utHe+CLVJ?h?gkb
z{?GGZ#>cfsQ3F@PcW!u8+xaAC*MdD-=;R0uBkMqB+Pp_NrCQPb`#PjcQeO~Bl!*(N
ztAS~AAGxnGWf{Vj=1}TO*s|<tx%qFX<)X`8?$UeShO~G3UCZp4Skt%QS-oHGdag<?
zb@s^3Or67f?E>J{dF{Q*ZC80^%p=3^mB^{ql}iwIZaZk$bh!hh9tHtJU{UVCz>x|f
z%D62l3(CNcWZ|%QzRIkN@u>FgGwhF}+`ghULQjN79?^#q6rxBYUBEyPfaorX<eT6W
zKS0s9djp%YS}u1aoC~ip$&>FFa0#fFJIEaz_YL+<Rsdj7;xdOzX`calR-BAv?}*rk
zwfLGd@Xuk?TM7XXD&g7i$^O?*D5!ln&8x49M$DXIs?(s|=CuSWo}Q9UQc{{RNOr`N
zq4M?*CL<gmb2j>E)<73(`{hQVGhrXuSS9K%Z&ufMPycO*_`Nr$iK^(MKQEg6fT?@A
zF5-^0vk)L~B_;hnkmwYh?wCvb=b*hF+gf-d%N_}?q7&G{RBp|Zc}4vHXoj@6GMWKM
zM{Msun+7$p@X~fRwber#7PS?0uLnJxG2x8jt-B-didn~?Y7a4&6Y@-g;W_5dZ~-U1
z@>DgC<7_Xj8rUHFi>)}+RwKFk3>oUYJ$1lH>}pJ@n2y=7=O&+Y7|0`ny4ar`DQG6O
zEM3+7?G{56;nE4O7})edAiYLYYq$A4Vo!P2s7kvit2gW>{)bfY8{Es;;D)nH{gsYN
zAA*3F161av>oU}<QI=J*y-Px_y(7pS3h`mYf$X+Yw$tTC<lkTbeLi9Ec({IW&N6`5
zzO;kF)t~v=jtdIX)-2uK9R1w7bhK%^&aE$y=Gm(4_M2?=3l>Tm^yy292dTHJ#Ec5J
zaAvGWt>%q@^&fF&v#mDKOZ)6abqC+m%G{a=R~($psj;nsHo5UJ&r%DQb(iUl2c#a7
zAk$pqokK?B{gF@s#N^_v52Emwwm9IVCJba`)5mRDm&jf-bF`T&BYLk8^2+@7#hpiy
zs8%2{g0jZ^zsG^7HZzL3!%X7nw(C^d=Pnpj5sgE~qOf5SGAXMPvL#TZ(W7LJ67Gpi
z1xweaE&=*SDBGG_d{5E^-hn-qG=aT6%-UhZQ#1NowH*AOmcwQs`QPDYfQ%`y=K;qR
zvD;HN!N}<{M7>d?#%y$k-uhF-cBOaP;bk?R)F2v(KCW$*+K&(yyIO0jn@q({{J!{k
zwbJdYr^WZYFABx?FT3vITa}E8#o8>Hb><+<ti6ZhwKmR~rqR(e=^oFG09u@naJ_w)
zvy5nZf&283W*5A#U%!yx0i-FW0k7!D@CeNd;TA;T1;7~0q!b{u<6(Y87!5FrW03(N
za8e|zV;;i6Pd;I-Sm%{oglCy;G56YOR2b?bhuk3H^;NVifuW|RJWEC@vvQ8dZbD#~
zJymcKxp;T_QGf<%zey@;l1ZWZob_JCi=bumC-idq_hFBU2@PsvK&e;j;BxM-4T{PK
zW=%lo*&OGpu>U)YjG$>omxf*%k!LF!atiW1a#GxyG=D+AgiN6UxQQ6<p?iV~niLaN
zd(LO>Mcnr*U;m@)XFhI=6vYVjDz&dk2e5W=T-vMSVUs6+rkDFl1<0_ai9<15_+Pt5
z7x%&?n+iS%pleo6fN%>pr#!xX?7MW-H}_=0AjxC0_0ri22D5hzTuznS;I(DFhZxaw
z-=kScv3}aUWV9`+n_(#DfNJpJz&g~5Py4PT(pkkrvXB9~Na@PgUqo$g+RrK#Zh>l;
zrrUTUzhW*0@1nxbZRU}pQayJ88ryLRoG#`tpC#4%ewvj=Kq)Oe1IKcKg23l<hcc-;
zD<u)#^2||m>{06!os;7BAm!_^dGktdu{sUtlhc<KhgiIIEypjPDsi5wl+uM=?cK%i
zD%;@NR#sB80;}Va(&r^PU2q&Xv~toL)G;-3lgNf|+fLv}l*8kamFniD3$PO7z1IU5
zEc8rp#O8%n0B!g|9O^PnKqhfKRCxRprFuOQ2h}}H;#42B;P|}9UdcfhZC0wQ$lt)>
zs0PMAD&*`|&^un;i?8Fa7vJBt;<gnVmmcfdbjdkJv1X=1!@sUVM2ayH!e-iTI^|XG
zwm0FWM{o{YprV{`vyn}b^v5BXP2(66YuRHfW{0Y{0T_F5BFw``$-qOkLSkl|Xe>5D
zBU|Mv?amu^kSjDi2Sl6`p@iCl90}t=)xR>u(Y18pm$0YiO1cZmxQ%KxwJVbk8xokj
znY`9jpUe6Xt)?3_TysO>1qI#ugD$>oMIyC;$8l|U@bv%u!;vu(plRQr0N*n9^xF;Q
z;6K5e0qmrW)?hQRIM<N0nrLEIodB^_a{A02ug@)?^kmqZfT`f_Sq0%wFkpq{h5^-p
zo4~>0>a_{Z*y)bVv#69<ml9_&)hz<Why|5+yrmuGO~$pZPkh!>TBYAMl*ytDMOHus
z`BSi5g<X#!s6_GrROG_#5TmYB&@nhTWSG6Y09kUon5QwN$`6`^`7q+>*Xny0O6TP@
z#IvX$0W7)DWB?id?4s%untg{QAE$qU`aRWoO7bF=ilOsoEwik`oh#;yvHf&cB66JE
za<|U?V@;w5z%q0cA8mr>=H!)`;@g)+OaIjq|ELN!GQdl{dUxDCH?(xL8RVpda|;3B
zs)rLlNSo{4pd8TG|9F|d7-ZtBPIWTI=DdnmMWx?^tjI)<FF~~@jTtj{;3n+{uG^`$
ze7v7Z<{pu1JEV-ygC@Sy_DgFuxUWyHt8)S}v3ik#lJ+TQD%W50q{BIqEz^ePo{)j?
z1xFMM#mv(IE{6UXYT-I~C3Y@R%t8Q%*&z-5{2hf5f~m3U-y2kT6VWa}daK|K4<MOb
z;FXCeu95<4w0Tgsnw&cHKehy)<$O1sy~ZN{*?~EBE@X|V0bvFiD6cxfW$2?=&Oo}v
z&~Yh~y+|^q)Xaw^$mqvxb?MWGcBG_d#&<s2IPk2*PDr*S#|&!!izcj8>Og>5lv{WU
znqHzli~!QO6Lnr5*z8|3?#TmW+nS_*=5lZq9AyVifPb89wgL@nhLYHX(6D}g^CKtH
zq{#%;1yBUGE-&TaKWst;1({=-xMMK<w>UQEL8Oa&$42|FFzgpNyZTBJmsFaCz0PH|
z%NOidzJr5l0RBy?CzYs9E(rhtTSrhcZ|f;u|1l<9hsfKeE!$Z0dS_9QWo4>WnBwXE
zeaRNS3W|TPcsCIorVWlqg4TvdOFPwvu0Z~nzqE7x5V%fPn7{u3;Wpl5z=7-&t){(E
zJy<LRTcSM6AoMvAyLnY6A^+(hA#2OR^;B*L>c6`W1$d(waSW?ExD9}Coc0NauO;mD
zomcrOqlw05;8Sy}-gV^*coxPn25q*p<I&MHUk|ciZx=ZxZK7X9i$4hW4+zgMh_t8(
ziJ2$RS{)GM^;mDn5+!b~apB&AFOCthd{HUBqX1rARBLRWns{evC{N`ZG1KqCu9g#4
z#oZPeFzVFub*g4nkpNieQRS*S`EC$2*{xIOi;ah*g?wF>+otrFtn};s$kl=w3+a37
zSIvt(J{WLU@-0{LRj#QT&{S5DFf_jq$F$2S{Vj-zfD9-SIk@hbtFLm>`wUnx92$bf
zcbUc`oa+V(7GXIskV|+-m#x+wJ6)g2%w}AdbQ$oyknze|U;&9A<%60?&5JC~Lk26D
z4)%|H!ZZkHgVu`>oaUmmPR?hSu>0)T{r4X~CtKwrKIDD@(XZsH?~35l+L-J0HO*p2
z>NzDLUY0c%q>N~|bU4Te<Y~|hNXh9RMaZ3Z^9O^F!<(BUoq+sH1XwhV0=OlkCeFiz
z3@{9pXf#Xr03LP3Qqv1cg$bm<iT?a1&~Brgl4dHxyzOLaBFg26lx%1JF4U#sdKA15
z8)0rjC?txx#fa^V)bpS^<74-KuZ9R;u@bQF8VV11kvS-Xp#byRaP~M&J_RfGRblL$
z&}0ul>Uu0HKaQtaEVS3e=wX1;%NcA(IwCG8T|%<k0>I50+PFNenG(ft5vJcPKv<&7
zyn@S5Rq*A?svSa&^TbU4?Q0~1*U}AA=qYzlQe<mfJ0#z2W+g!C219#uM!y(nUNHW-
zCZ2r7Nasfet;h9_MsaSSM6?)@_G8Mk9(pC>OM2&y)jQbvp{_5`!AKzLe00TJ)szb$
zVRJ*4fLuw)wRY1x_F7l-kd2s)Om=bX^48=ZwFjV+7m}AF!^LMNSGry}Y?UfO56l?W
z?L5V7EgK*a#dQn{gM6GuHpXQ9nv@1bo9Amulm$7b+H`)VNFw5eO<9uk4L9)*!{Z4-
zf#H+-gkE9`%~Upi^?8uTjbyRpP8rKsyWUtE(0UpU*;qB@UhVsqS<hW+hJ91B<xPiR
zys=!QjdM?#J@TGC*!+yp>MorEe7V5k7Z2=551Veh!iz-+b4VIa$O=;l%CTo>R5ZB+
zB-Vj-HgT0%FM&Cm(tzmMss6NBP;&}mY`m9+gDt&r+;MO^n1>Q}F;``xd1R1a*D_86
z2wx!zhLTppXk)0(N1Lh!HF+xacYqx%ar)_==LtioF(7I)t<p%8hU~)W-AJ8x#+#xr
z=B~e7%pcdk;;5ODK?6b(gqoRYJ05I<J|K_IRCp?)j2J)^5i#S$Vga*m<|`$K@V54K
zWwdam2QKLN*;X4LR*MJacfLk=zwO%r#AnYqR)`1+EQYAdtwL&<U&TM;-^YG&aM%iJ
zg)KPnQ`gU*@BAog>XD1BgzJTw-YzW=rBuX--JxN(ob!zl;#sftQ|(t64wA=|V8LGv
z?_d`FM~gjaW$TW6X=Rb#*lA@#qYN;R1_dzbhW&7oheeShmODCBY_4+^G`2Z_(19Yz
zhVzRj4068sFug%xjmzUpeX&9s(%MG*@SHL1fuNkz2Mq%;4j%r6|FRTAW)AUl(iFbZ
zU>YEwlLqmG*{9GOqBv8{kTD(M2-EwBdvHP9SeKxHr(<>Q<TZerHDY}Y`vz}-1vFLd
z@nrJ)ol39ANo%JmBL%Jf9w6IuXJ7x$keoIiT?_s_@`&zL;icuTIm=G>&I_Pb!p8$P
zD$^SUr*fSb2-Wb3razS!W}uQaQm`wE(lcz`atJ~2(*NLDCGgD@Iy<#xE3SF!IhAg*
zJU5qpwi1N?1QplAW`^J&b=-Xyp|v$2sZ^y8^5fqHv5H}#-q}tA7r-5h&s2-T@R3yb
z@o{?hZuTkWFK%62r!2)cv{;$;BT2CS0PwnH-I4`cCacRi9RKwzPtL1_J2*C)EjWJx
z5XBr;j3I02JeHVoxi>#n3>g_n<}S`1(8ih~tvDC8lR4|akInS62!=iHbN(cG-_WrI
zS^&FrTMAv3j84970if64BY%z8FJG~vZZKq89IS1>U(G!RCxuF;tZ}a7=0#UNtVP2j
z$}GjwQf7lp&k8ZT$l1al-j~1H5HeL~5DG=7t%Q?%6cA9_d--$vT*DR(qmwg(WFWO|
z>o`YM>cpq5&(dJS&HV=ba@zv-uB_FmBrN6$Tt_}TD{NAb2UM1-D#7e$U{cbs4a9*c
z+JPnGT2`w=fD{_g!(tFyg{fGDqs0{`=GYkM_qVw5P`6(_<hXhWzXkdIRc>b}=)<mQ
z3JjHfoaaDtlWy1r&OA*B)0-6bu!=Yob23lfOu8Gu2<Bq50WOYIl547?JqIB&N&63B
zGb(Ox@G_GGpg}tnT>bJm%UDvM_CgCTQMfhT?7TS}*_hN4iG?TgXm--7RTnYbzk)>=
z_u;lX+pff9NCZQ-lKE+5CS8{4Ylz|oW&N-<`oxId-&G&f{4+{M=B`n}Tj_)}<J&Cg
zr_XLH7t_4X>t+sY>`(<IXcoWBeWOWUC&?6veldIkP?*#ts^bPer@KHZ6yYOBnP`kh
zP1N=$Y*Z#Kz5yXb)Cy{W5jTyw&g)%%p@mNM8@L2KekCnFj2%LwunUphr_rJa-!}k|
zQ&$Ybe@tt#69h6gWrKn3x#LQ<n6yVf<xM_%)P-*H5}mNjl1TCqh+a}~F%g?bu%SeQ
z^d|>^edsAkViuccbfa|kqhYEG7s#m>hYS-nd^%v$;8Zk*IJvE*qAOA{-8qiIaRG=0
zm{a8&xA&i%i?fn`c1`A&Q<70k&V4b75I%DS<P@>#0OT;OQptFfCg}{4y~NcxLOt()
zTuVSNp+9SG?#4Tm!iz@6zFcm`EGeQT>>$8rWE1Q>S|;g5m8~g^#D39onS0Cf0ogyT
znaVM-X2>LIBTs%lOM47x&CGu<i2t)cG#xkNi*+U7J^H?EwcVZ`8dL<J6NXF<@dSJd
zk6$15KX{)&@NX7|wsd)-i=jubu#OidVVKD%Q%$+YDYMYNzxO25%VU6Hl(~Pwi~y{T
z$94;wGGtUfea8QZ`=qu!&X@J$f8VGjM;GMEl{Fg$X*Nalr}l|S)qxtnhuU|DtnRP=
z-TtrmLMO}3+Wns$-?T<k1Xa=c$EcWUFJ(_`R0V)*Ci0m*)Btw~84B;YIKBy-)m!Z%
zKC=Lj4tY8sTOpyC26gLx<FD?<WIi~PfDpiTQR5>OCZ@)U#)^$I=$N0FNHGhL7x5y3
z1e;=I<yr#l>rL(xWtjuHe)su|eEk#JyD%c#S!gv$tC5Jk^`Le%X_w?o?)yBvgW-iI
zWNVdMhVNeupU&c83k;H@t)c8AGljD(nnwa8yc!K3xIpHVTucq|fFZYS&U-pAL!P*}
z%wx>cKEKRmEkqH@8jEN~ZKeaFoZP}fbjBUKUqkjKK8M14#o=F-c<8NqYX9(;ms>$9
zAdTg9VyS#hW9e)?%WF_{D{J}``#^|TnX`hFyKiji5yHf;0fpGrmOk5eWJ3T^Ep`9^
zn@yddzog<5?of=nWR=w$wW6a5e4VVK+4xXO;q8W7-SZAL{-iR*1Cl=4zd5G`tXL;9
z2Mz770o#YKmpsR3m})#n2pyju-F&S$(v8qtXxsm+15P;Da^Jo+TG(61x1x~JOv@Zl
zTOL`9G?^YNyXfagD!cVsAqR_{s&D|tnt}Z;K3+mNwrs%P3s1HHAX)V{rvS?TM>5@b
zN$)kQ5j(n@FmV_RQfy2v5dRPezgB!nZf7A$CHoZjoh{_g*yQT8@GE1c`CV#ZpHx@{
zG+YaIt9bi9Lp=CF@_AJjXZjRBtNaZ-9*^rgymqM=@6E|F<cXC8XyM+Y^l~^aS5;UY
zNT@UmZknzg8*i%)@3!hyea@atDqT%)ZT!f<d`DT6T0bf<yDCl$hRpbznUA6u{zTj)
zgO@Y#oVeEi6o7TC;z&UsD1QXt6iWW7?@@2zb5L~X+LpEd>ltn|NkOaKaCZb&23%uX
zqvy?^%TcUvse}nM&xdQg%lp>Ij?1*t;&^NR*EC$t;*-z=8?LD2MC=cloleexXN}Du
z^JYzO^Be9EGx`#xkca&C2e#Bo@2+%a2mC#Mv?L9Jez~-UmZ>C}%^OU!F-U>&)K`WM
zquUUd`Xt-ScH$uYRHS@Ds#5^Kp|UXC29h-9z2uh7)vmDeQ27DI8@*qP*0kPeYNLRj
zZo3tTdUdKW#Yn0gTCgD}A%<vhcNIFLoF-b<+wpX`3ZwZeulV7Zw2ao5JFhk_5CVMW
z_GLRzL*jO#hr9VoXZ40q{*|+A^c?wS@7()1SNvNnRf|7Ix(CuWD*gpP^4<r2{`sXu
ztSjd18$lvxLWVj{6}_@|SkMy`-&wbTtD%AUmucjxwU`$>F(gqsaLZdy7%s|XH-Xy)
zh6MgNiMWfw?f0ewc<ivlUBj1uXP(9hpUSZv4||Q2=KIveo3*sL2-eLJ!vM<Fi5g;m
zi@u>gN%pb~q1Y}jJ?a23CPj8{IPAAHQmQHAt^@58QM{5`9oPiH_#=IHSk;V1o8sr|
z`bk{1xQTCRGC0^o0_EI6xRNBnm--aUyc;E3S0jl2Z6<ttdl>p_nI^}lIZcGPE|0(d
z>{597St^u6XwlTVxey^5`iA!bXXJZ=_@|OKv)_PS8=YmE`ltUY_dS+9h504<ZRoeA
zH0OQ1%&P@2=Sp@04c<nO{iqYvPh~nQ;(AtL68u!WLFi#P9GICzJllW<C<|#ic`c-I
z_GC|OMM&11MZY&%&m0Wb`9r?>AJb(G?%$QPdwU=bpT5nCaB3rInX{gV)=VdsqnY)c
zm(HN_Scz038QtuFe?NJcc>1*H2um6LQ-Mv#phAbBY?zH_AE%cgNU8Jn+km(=M=9KF
zIuVW=s6TudWzZSncWTwm+;A7Q{1*Qt|M>N4+f#uy1CIq%DD@IQvHLFQceSWBk4&!d
zH*gcQw^*T`W+T3tmhkb*>EBOpApDuEk2AnswjI2e?FD%PtZd9z4+M6uQ2h@01BtPT
zZ&g~qivW8ayV-8k?7F&_V2KljNEkRIAN}gx7p5R!aF)u@<<(|qBooO`@fG}3!rFP|
z_5v$N;nZb+z6MI%cF=jVXggpqXzkBf0|(Kqc9jwa`}pj;6oIL-^QCn4`}YinOtmTU
z4cJyT*cLu1fZ=0heDhYHCfTaEF|pHEX^t=D1X9kwf=xal{kXi)u|j@N1sjBFFch=u
z*o>KgN76#Kjg|F2MWd@(31@<HsvOQWQZ-A8-v7ef4&HN8iFQnpUNjacThd_)0qk6I
z4i}m1El?;2D<=DIWBGL`^{3C~Z<G53J)ao#3|+Tbz|C641jQ~n_Vw-(6B-3WH`z-f
z^eXSHnqX*0!&fHypA<OhNk7-=&*|i7B;%x7*!mBICpKmbHi1)T%+}dXyK|xQe_jRG
zfwuGo$X>;Us0WsAC!9$dXe0@FB1drweux%=uxPLqt@*mv|69z6GNOeg5?YgBny}6Z
zKLVU$TLF^E6LK2>3wJo;A!iyIjL!t6!fL#r<2{Qb%iefn1cWp4l{cv&3<Pqw04J5K
zzEpZ73I&R{A8ofxmXMIO?~WE9qIIDW85pcKCxx2weH=;0zXE}@2Tv8h-PQam_%Kgv
zQ2zp1XUWmuDz>^_3U3&HY;vEvf%TkI4S<(v`+if_g&(jAD$UMJ_m$btx5Y6CA7m4g
z)M@IbNkOIU8n6PUlUX>9#cyOFd9kxiM<t_|DTyhF#grYN#(13o=JF2#RX~G?{418@
zER39mEa^-L=TnH){XL1Kk=3rDjP=B!05%7He~pM5&joglPLOIk{!dL72*lu?At2iQ
z&?a!JPh94cZzd+?++-fcXBc9xUeRUoTE7y8DCEZ^$xI~GIgJp9Q*EYv)j#~%1Syvu
zIGbz41!45^_P3ySW+n#%^<i~>Ne)uI?3%gux=tF))66|*EqFh~fojx<BrKdG&5x+@
zmbb?K15`;_vPb9EgUDt2O)3D}7w}Ck2-{#H`oGaNC+)}oLD!e@$JGBHx?Y1iSE%Ta
zqVhQMiumKcSL`f-tyFsmB#beR?#d_`wI|EWG23lA6Zba5ynoyMAET3Wtt!O1X<u{x
zHo1@b8S5)1{^dk-ZEdtN;-h9c-1g%lpl7|jvhF&T7~R+pWH~ANNGP~`yW5sSFb=48
z>#?J`#VflH>zBxkOqe1v*MkENegX?J7mH#yN*fhO1kf@W0WfD2ixHUAIi7nQ`rqno
z`vMztnHKf9zdp(kFP1qxFyszyk^gW4$GKXgfXKrj$QHOwd)0Iy5wDGW{q4*I4TGi#
zx8DS`NyC1DI<|GCKMS8I`&&pHeJXPO>fobITz=1<I$M}BsI!vmMhU{&|M=idm+LPi
zoN=9A7N>wqp7m)cpA^W7b}=qn`1s~qB)l4CX@E<9Dnq0=h~OvolwL_9s}=>A$<Ixf
z3XMam8FnxYl6K!otn6yDlbU46H-#C24v%E5t`^7_VXafPKyPLrBK7^1!R}gxV08s6
zTa1nGaTnR~;}iGF$nLzAWhwGS^!$9E)fO*}B%XNxH5;^Mu+3)m`M&4lr!AedCK?tc
zkot(p3~ywDB=Vr0X#TXB5|96rGM>u75zptkaOR*}r`@q?*Q$lm1P0M|9%h!WNWd_n
z>Oj~`uI)AgR=S#NSi|=TsTCF6y@7k3_v?3sMQh`@zNdcr-H&^Jn6@_YP{3-JO3A+f
zl5B!nk^PcyM6g|P^`7+CqIF#=DGSgOLH%K=pEPq41)F%IGvFa`f3^>}KNhpZaEs>(
zgVQ2X8Miuvdv9<HV<CR65Cvj@XypxtIu=ow_6*M59{>|7ED;v#IOA+mCt|!b<<}(^
zV8)k7QE>Hb7$M?6WC{;{ka%StI5#%NSFp#in;sq3J`1O+_xCG7W)uK&_FNOU$eY(i
z=4agF82T+Pjmz86+G_-0*}xwcs1{g?HM^7`<-joNi)A8O_i$gbpnuZ?ZwSI0_`X73
zO5-g@H9>_8A_XZQ-|xd&LPzA`&3*$%g+8CncM&O>mPxi^02Gz3k?3Zds5fE&+anS&
zSeD_O`y=G1HpM{&n&1^+niFc-K+RFn`ex<@Wis9r6ZN?i@fQKSb3H{Xw_V%jvopZB
z*RiQmCzcMH++zZqNXLkLm=yf_`t(W2337H0#cylIF)h-asyE}gqxCb|`)5iL_*z5s
z9}>6V4ErW$hV>5@j^IDsI{kjr3+(7LCDXh-n0>r2h<-5?Yra}dwN;F&{E}F;($|qW
z5wCyJNqCsdjJpKP_WrTnRwR4x4|&u^w<WRF`Q>!{WYE!yCF8XTJtRPxNYaWLeIcde
z&;)1_@ag&m6yqH7xz=NstKTPBE%wF2v;ICzn7YDocwDjF{;d*59@P**6^&)CFJm}_
z<!FYV)gSIy((i3iatP|al>8GJM&qZSBrljxoMPuQa$f-aaStNbqSo<oEO8AtPGT0w
zd6=v$+0Tixc}0hBllfh}t37|MV=2bU0hfxs%?xutc6bka$l@K!60gJFC-GVqR0j|Y
zhA3f{wlVZ4Sn(jx&b*Ggt!pkvtCpQN+rfvf+E27p+iAh64B~L<1lxUDMAzj0Yulyc
z|3EVWj>ip1n^b=O2i3eB6M4&fh<bYndMv&kBjyH#9;!-?4R<YGA%{<)^WxdYY>X;R
zDY{ngD4*w-^V!gJ6V?ffy)kncrhaX7M=&~$B|&<#*aX`8@n%`<r>7mpSAg7m6$uo)
z38o($KDkRd!;b;wT5Ef^+IG@?=XqjeMpN6pLwgJO440i~%q24-;_;74g}X;0^(gVE
zy<kcFkjvLV|MX2lX8Uj$Z`@mL&f)T#vy^O>-fm7@@mSB)zL}*rwx{7O7zwhx)TM00
za5qR^usX1nE^1s`gycUNjo0%bm*6pyQa4jK5$Wpz@HNPYl?__rNR}K*R~_);CX!V(
zmi7RKn{_!STiY02=GbCx76ZAUbp}>z8qF*1mbTYkF^hBU>u}X*xyJKOM7ngrkLQQ%
zWifEks}|CCxCB&k=Yb<knlPCYqy5ulEz8KPPc=hB2Ix>=ey#~=p>|lh*H+f1bvPa@
zQgO>FL#^C03?agYd`3d*<|B}-tr3niR6z2~nX#NoczsqbLG$?q*jG?1i7|3?Jr?zu
zV%R+;t|B%TEjR1h7U>QPyam+T;gR8nag@*?-Ra0O8(owy8E2s1u%6BQ*g_S4>K2BT
z<gYlL?p1bG=}G$?l|Q<7Q$=dM@lS2e@zT83gOmFb3C-lAN}X?6D^$ouqBU5Vkw5EC
zkydpP!T^yx)QRMb)bt6;z^8x_?Yn4m-ELZ=$DcyDFlzGs&=_K92cqFK_wf{4TgsPY
zuVPQ1F(*K|KPIAix{E>#bnSwQ{!ks576oA~5w>u6N$tikOt6lYfy7=?jQS2bF<Rbj
z0iRn-lITsm{!OGEhf51>skY`!Isj40qiD(o<E5)GwFbLIc!}Vq7XFsfGPETa8X6L<
z1^Tb5?R6vLDe4(i^kapM`Nl8ZC}!MNaRPD?4oJP{?CSJTn>M+72z;dH7^Y?pHv_WM
zuOT^Oo}@kPmnyN{N!U*|FE#h`qVi437~EoL*=?Lu%-nycI>L<{-`Qm4-hi^$8{iDl
zvCHM91b9^;u7S+Y_{o&Ynp-!=n8`|-qxS{VQfS?y8v0(HB$}hRNdat`hm}Yny$3<a
zt7O`TaueR$8?f3n8d+4aSfW-=i9(LY9tQ~Xk}Ir$_mbq5in;oV`<cw9h=O0fweQr~
zux_$JLW>`HusFJNr-3tY#(>VH3U`LBxwE>~g>|Ne)-8X}{QL~|9=mm^@brbv`wIQ|
zWAl(r9WM+vti|76V4nyXjt%prP|w6D$Je`^8QpDA)?A#E&0?x$^kjF-)h=k1S#+^(
zeb;&2Ldb9(S_pW{6cdWYOc!CXb;l<Za2m$A6d^>EGX=&QNF$PfVt}8E3-O{!Kz;s~
zc^7RC86MDTgX^^0eFrzr8S1~>n!_5RyI4=j6L2Pstg=~gUaU>@$fBoq;TezLq1?VN
zk7cfe2axIEqju~|ql)3)y#z@6fh2ZK@ma9#;!!dINYBMG$e*hxavHmZPqa_6nnF))
zN|mH4+{8%I5_K*j{D2gL+C1T0b!+>f3hqV%EM{mK3+Bc;hXc9#ndUaBcp;Ety}ah}
zQ+NXpB)cJWjfOP1N(q}$qXs&=rDmt*H3za1qDsGhm?8edjxDA9U;K|FahFAOb6Tl?
zetB0c%VA{MXlCv<t5tdgm~hzy$fu`n@y*pt=(YJ;%1B>SbO2JV4Oj*T-7GXrKj>uj
z!8!;NPb{?rdWlHX^F*b}Z_vgV%!|4lQPvZbYVcA~^9co7=)3pmkz7ohESF)%_B8EV
zPd6~}-@vtW)Zio>)mjy4<l{zfOfWN_xjvP8&jo)Gi6gHl`P+qsp0(j4nkz!<y9SGe
zb)0ivu5{WwRRGN5^T&J4qQejw^+yZ~M)usFWUV<@QqjTp*Bwwah=l&ZqD6XaQ)0ce
zTcAZl(qW-KHieU7t&Fz#8=z&Ic#68%#zb1k5O0J{ni<>6CbFibPmoHw@YTv>d<u23
z?id*OT8c8Dm}xn5E?K#|gWq#Zzh)wG)u54#fdQpM;sbthT3wYzM^44|b!DnKLiu3(
z6HceBoD!~F7tNG&R4=*HJh1!{&+i6;k@*E?R#z#Uu<4^BadE%De-HbLkAwMiVSSY0
zQ}wSr-t42}`z%SRS1F@tyF;-?qLjbaFyQ%Fd*jqRNbM5>|IKakX6Ei~-4+17`|jfC
z3F>dH-wC)S)?dfOEOPb~p5dC~*IoRllo8_Ig`cxQiW-CP79Fe2iZQPt5>H0`aSh|n
zkg)`#02T2(c-=>@Vyx-1Cl$Z~!pb==S4#}WP?AYV?ZGtCZLCqd^@OW>#V05ljm04~
zhYze_etw6}=CxSREr0Hxqpk8a(mQPJ8xI6t=L)EFKa5^$?O$6-^GLMRY%}<Ng9GNJ
z(J0+X|BA|?>IXhyPwN%6kOZT1`HG}LxPfs$OimS@e#l>f%)~+P2*S2DfdJFF;IBwo
z4>38vLE{$7>(lw@Izmnh=y>%1L+ih-@(Z9IR=5*EM1E8PO{!<Fwl=mD1i`{`nP|zx
zasXh5DjCqt7yRaT@S{i`Kot8ME9s@OG@f+Ki;_7XE~i@{m+Qe))96eJP1%Bh4-VZP
zzDPs)Wb{b8@P|t15w5?J=2mP&z~V9-p3n~1)2U%LbMHvCla<YVP6HjnFQ*=69oEBd
zEd53vYMqi;LIJMUL@SWk!yQ5|Iw1lm#1ar)WzLQDiwA5#JFnw?cl0&Q1&3~Al#xG_
zNLmEV$j2t>fxiozsWER&#X&4&>NV^kXIF5OoDXHKiDPAeQ#qpPgVd?lfj)Z{6O(jS
zd<c1({CUXjRKYb5x`W>_LaLc+9whFzSB(;7@X#>kOi?}vXZ|FUpU-H_+TF%WPZ^*K
z$-;P?Xt~&zCY{nB#e|8y7sQ0Ax!cNk%&hFlw4Z1S5ych+b%CW}556oljUN+bFwGC`
zT_*?IrB{$oD_0(qdckc%=^~~s8OPmpVNvosPnFx<pn}`5U?XpcF}KhDK`?@^g_7x`
zW^lNP_8EQFAhgoXYCjZe(L(hPH3z_#d5l&E5cFbgBYt$yfS7%Bt5v{8Kg4|_LbkOm
z6ts(Osd*gV`t$;UapU|t8r%9wP=?43{vb3~_>NyTn=Lw5j`}=_-y9UkL6!J3+k6o9
z^md|j$;nxdUTvll{>XMQBGg_$Vt9fFuRn!Hxyojp?RPfw=p}=N#O|`8_$(lq*5Dy{
zyg9+DJkOKv<RY9%8VO7uULHPO5sbx1>XE}0sGNcxRBOgsYczS;vWPSlVKE;Sjc^f{
z0aI|D5cdV>lB|9Hq`FX96xkFsGb0e=OMZOxvXKm#$np0w{GipzBBjv65NQ6$rUBz$
z+0qQ{;7s7v$kxxFVhz~lng~G6o;(7Wt|_psoq;9a(OUiD+i#M21vNQ2>ID9Ow-k62
z+r@CgYXXpa>3?}yP#j{o@n^7yJo?7gMY#mABet(AxSI#du%UOWaK9nJo-C*tx@1yN
zP0nbBMqRD~(tT!(yKKfnJ#h^e%-|LFnWxb2vF1dn(Mn~A^cs6I(HsE=&<IsdRvv;6
zkVPuk5X}V$l%LQ9b5<^4rJ<Wv2g*DZ1i3k4@9hJ*c;OP5Hl{U?1Qi*-JIvzjC>t<O
z7z)?PUB2&we{VH;W#n@&l4D$D?3A<pdaDs#LodMAI_vkLzsroYbLH$JjSpa9_0|IH
zlPB0!+}r>;GXHpQ^@{*fR!{yY&hRr^{P2~m?dx1UMVA76v|<lMPg&w4Hlc&_t9#Ji
z>;;0SCd1+ZF{!W)5KIFJdFRO_@(e@-csK+O`qg0&rtUr<3xXZ7&-i~3DB~Ik>84*t
z^ePY(z4`Y24g$F4|KOMyq#qz(0mCSEj?ud>)#ESgf>MB-C<*}hYuMw!c;mo(E=0YN
z@jF7XArKf?g+K&oH4%FZ{=7iOLD;$eLWF1=t{$?LSmvOsBfXMXKus_){-()i-Yp2P
zgu3mR?kPs-ZpjkGwA?hKKb#UR)xJ5mJ-Vis+s;+lQNca3_{kmMMGm>84Y1T*xJ9|2
z1$;AbsOO?<KKTP;4tgn)lf>ZDyQ5t`xg~1pWh>nzSyn}R7gTOj7$dt@tHPiI`_1Y{
zW~#B=Z#Jzw`)i<TwI2D(YTGkpRHD|n*A*62D<!;)cY#eL=ma&2^6*U*pdwWq7p8Q5
zagxi04F<HWHc6ft&D=<YE(NC~DU<QJ@zRU?^Gq0v*OLHaT+b=}Ka<1)|5p4h!JZCB
zb}|-O!NZNc2@u8OB-PbzwzV}l6qh;5JMsw^N0THCoYMWIC`XG7ZWwJ0(_lB9^>9n;
zKIfaSUH5sJ@l9Yjj1;*zVe2%BcIB&m_6iw@)48TtvV9Ng{?JS#9PQXy*o<HT`VIRG
zSU(_esDlT1@KA@_KCW#ccnKp$+=&%R3jN_M<gh}x5hrhKYj3{#n2Izhixo4_%$dGc
zE%&{7&$^$z?65D!7!U5wMC(qVX+(tC6gBk=QUf*^;&*r`G_n3?Y8}!mtqm0po`ue=
zZjf2WlHP)@14t-)KTRkE-hc*73qF;_0;w%cP+tLx62O1KYDIzHagTregp26T=RWeS
zqSG__(%?~jVwb^N-Da|#HApw9HK*8iS!s@}wrIRDquYpLewTt;J-!-_Pm-)c8pAk1
zmstd)JO~){Ntg0xUovxW+MaKjP~)ke?lR3crf-UOXQ4M<MhKMXzeUi($|hH%lZv%(
z(yjrtM9-}|2$thJ$gev83?*$4uzlbzB=i8k_r<IK#4P;&l$xM;#$7$G$|nDH{InH*
zO5a*$e}7=D*FU`;a@jvQKXPFTl58|ITSFaES?#^VAx-6s8-gss^~|rRK1)hZ5dFup
zym~Y6C|G;YlADS~VL$OSF+N^Ho!jbkeQyJxKJ>$FL{(X3S6}rfQq)IP(1*oP>r!b3
zNmW;G0nrz7mF9P~o&+MU-7npTM6O?yZdwABn!@#O46LQb@WIES=?8OoQzU9Sm1V6t
z4P?02uHr`7C&<evDyzC0>TBLa-gZElVlTFONr2UvCYtT7m^C=W$xtRepEH3=RTl;*
z+6_ZcTT2xgb6@`E{;FXv*1H(Y%dFCD`y2Fby2sMcH>g~uh(f@NWAJ?K_k;+&M7AP(
zGIRQjMm{@!VkM4ec+6USzLm`}#BewX?VHs(`}SgU3*`~x;UVFHrJmSI0I@Z#e(MQ(
z8kF|!wCQ=fFkB%ZUZ~bSpY<#4wMH9IbtgPR#AUo%F@lJ`6;$W0uqG;^8A+*85jsHT
zkMwiHss|$J(5P_B<E&eND+|6@`Ah4!riAu0&lY7X=f^TxY^6%8nq;iq|Mc^!A<wsw
z=;tv;+($l7r8>i-Tx&N%qar0*kD72Yv0Np_T0*0urCD%5sX62xLV5`SAZZwNrWY=1
zQ#MXFe~3bD1Y2RKF_ePDBOiuNh7>eH+yZ~f1#GSo*+BD4J!A5Bym{61-tcg=Np|%-
z^-l_ou^)A2Yr2_-V?wpm?>A1WtmuBg<HXX^5xS*2)I$EdaUrZJT+Tqsht*KI^2EI4
zfrUC>t9!jFLb+S1<665NK!4}QNu=-W%JaAI_Cx1hYTmK3gi_YTp>7PH-b2@Q%xx;c
z-hxT!%PMR73$f<Nr!wk}L2ee-e&L;npmckMVu?BJSBi1&6c#}_N9I-Lv*}6YU9sPi
zr91yr50w5bd}qJ|Pk9}K{>d)%ZJPUb?d|9Y1z-ht^yy*oDPmU=5cI3yg3N5O=^S%q
zg1QoML9I~Baw*omvDjowZ;E#TZb0#88f-PB)FNbdDZ;Ix?;p$63c1k5mL*{2j<0rK
zb90bW)iO8OayN$2=t8OL;T+5WuswNBlrFB?Mfr@$ctNwBJOdigQ_C;jw~*I?UwY<B
zv6D~AhKmDrV=CE70QpAo2V5kF>JU|&N?LNfIH|y3W$UGF>U6h?s7R-O1RHX^(-JB5
zd+7trSC-(C!&N&U{kwWf65Wuru#Lx8o}un6fHCPD#p;|}NJAoeGH#QT8gcM$upzk_
zs5dW`Vd$v6M)AlbC8;<W4wG>{><{o$%PTD44nnOjIXdwRKn1r4R~|0_bluf+QR}&4
z^;VwPpBDC$?>$E=z#lRycs{H2cj(m-p7Wo~LCc7|ByUFiQM|xOEOsT2_9O{nv}js`
z;o9)tR7;3G)DOh<w}9B1Cv4O}O5EDa6EZ8G8b%p7zr2pOAr^4Db&9U&P+mT(EpZ8+
z;K@r1v-aR~z-hgR-Kw<n5NX=6!uL?tyFmk}=(gKw*R(OzHFhJQ-I%wvDo@d(V;#|X
z!z$TfJu&ANWTcB7GYLq_W~)yaSR*>u3`Wm!Ul?S}fnId;?~F~qskjp^pNG?{x&9%1
zc&pf>qv^`%QHm{7aeTx%dYTND4J4t~<LqyG@<y0)0R7}(KN=hpQKt~WwBy0+)du!-
zM|^fCmdqubZ-ZE<s#}F_7YUM0drkdE*?N35B)qdB1X$C)3GrIGma$#}-dXyh83$*P
zwPNyFVu~d~yU=))a=_M!`Yue&@m=PD7+OZKKmw2~^8}Cwu{~5gunR<4m#;Q|ZB964
zvQ*a&;8DRhN599zI!4yR5f=5p;I>4ODi=2WbL0z%><;)gCx5&N#ARS3rk3c<>d1>N
z(c4pG&o^+BM)kTm9#N!Esf&-+6kbo<7utfhagGv2*Eeqt#WY0rEmcRiPN`OP*1coF
zHGk+&TmCsKQvP!zB)vVeKfCDP1qd8g{_2KQKnfQ%PfEp^Oj<E)s>!M?1P~g1iV$ZJ
zI3x%3)KNoM(v;5A<z`I-e~OsRbypywvG+gqKEmMdTGlW9d6|=BFyPD6WZ{bpQ5T91
z8mHdau`%Yo3B~H?5_Hia+*D}e$&#%Ho7#3<d5^j1Z`hk+e`&Le#eHbi8LjU2HOLo4
zfU)e&k^YQSxH1ASH5!stm-ujEhl%6(kgl)G!q7ygLzu|+p6bdEoC(qAx|c_(Z6aXX
z$x^{^(TxWW!7L3NG?TNYI7uc!-s{j>@-aX(q?FW<V1U#)w!SDPPk;Gmz-utErM|j?
z&{}u>w=~i)v=3SOWIq5%rv)U(kGVn~(8Wz+*X84C-!ZitfqAi?Am!uzT_prPf-rP7
zEscl*B?s%x4HKcJJs0}t6)cG6aG=yd+H$CL5oj;unFT1;1IEMouT>ZF*{=SgE<`Gg
zPT$|*B)IH4QVD$J1_tCLXvDSzRFrE<nd@T5Fyr?mR%gVH^{*6~MA7WR=gu_|KzDE>
zg7TVb4nh8-mRcR1eJEvq=Ls96B!q{UTf{LBN{ofb;BS=HUyi=c_4DkG3#H%4tu~Ir
zY<m@>?WR$<aVcn$DbZ%(au=IqNQv@8#I**Az*h<4R(I_XK3XTNs*-GQ3wyq{LXdYG
z6X{YsSQh^2Qk|nd?mdl`PXoqO048Tkc^KfK4m*1qcbdg_(L~*YE}imQJLd<Jm*+OG
zN3#qqM<NpnqVQ6*=l8WjCUg|z(8a$U+ahk?hE-tqeYD%p?2ZqiitNFofrc11`>J{H
zStG9|>greMp}#re;sSlaH+5%p;S{W2QO8TZtNiNz%HVCdSnWsW2`uqU0EVSNNqARb
zdk<a`qN8q{e?6>!A)pHPa4xNx1A8lmx%HAj3`1J2$)FU6XGq#Tbj$TB^RWltrD)oW
z@=Q?6y3mk0@w<&FWZzP`n3$v^a^nk8O!9*kZdfiBbjg#loP4i1DF%a{wxEh8KvF!P
z1C!|}+exSnuJAyDMj@z#0givQT8{Knh~5LOQFWN{0FUt3P2H7Ywap$qivE4yG;|*f
ziW*XoUY@J0MKh%Lg~&*|#NA?Nv&oYie4bzj!iH&OsJXS<oNebp1a<~cKFz@JVIfN+
zAWHKCwNj~?wK@5{F^HgxB!`qKW#_|S%7oKGhR{C5UWodZsfh=#13asrT2kPMi#Fe0
z{E4xoHyK+u<suiXDCR5~_@PY@ScFcAv4C{rj=C@_#8vS_Q!b8^fOJqcj7p{Pl9XN&
z+<KWRWt*k>oImrFYwwE0X>Z?MQ?VmKUuFtcfEF*L$Nv!#;PVoIGlG3WTN^0Q-zwVW
zyb&|w*9^HeHaGii1MCSHyLBi(PZt3Lb+v@nhU<^O?VY)}L^xIfiD{tLyF*)a|7cr|
zidP=w`m!-F>AVq%)@8NU-IXsq2RqaH5tBzMttyyepbMcmU3gIH6fC->W=C9DL8rtv
zaZzX`N#f&L-%3)9g&0=|gwN?@qQLA%Ly`sZ=G4L%!B;#B0E^jkoB{QoIyT5l4r_(p
z;~ZDR`Ecz@@k(TIvUSqxwkMG|yqo_d@#mbQDt2C5`<4zm#!5BzKBmku+a`6CI_`~D
zb#yC>+9IYgZbm6y@9k+aOJ0rqg`_kw>;gy6kZ%ooAho$Y{I?O1K~xvwd_Y}QKU^ba
zD_83p@D|s70UL^$TQQ5egG!j<MdQgnlt2R7n3?KngAsKzP*sB+u<|?@3f{jvm5tEJ
z{Z-0_pvFLyRP+*6wemW)=(4cNM|@(&`X6d#qL@~^5BsJtQXO6S^>v6o5h1>#A0M>P
z&0bYj=W#(#6ghQEwD`DW@jnwijZhNc7(9g6LA;b>0E4)>#OZ$ZaFkJ~PT*nUk0VaS
zSpp9A331egJ+gp7Yv&bgnrXBuyXJHJRZ4lSSx=Mo&Us&zYF-(0@jgl5HU(K~@0}G@
zEWLP|6POUzfM!!AzzS=R^5otFRJv>-cB~|?i%Hvrf@-N1VngY-G+k0AAIT(MJ-SB4
z9ehnMfcWwl*o0_~ih)|tDv+a9fega;o<N_N<C6Y{f^ETn!vJXCpKsQWb<`_PWATH4
zx7J$gvRl~06P)%UpuyZHSkIWHBETf9Ue+6o8OOkY_+tz}sr<L2LiX3Np%+JP_h}*Z
z_=hZ3$8r36k$t+bg(bT$)(cBrn1GA~z-H4MU<cU(KB`m=y+fh?@mKwQ4rmp%6t&+x
zrcR6w(DL(Xe`1${p{6idtf74aQJRGG>m(_dP+*%;_(4^P!;4>3aG!oR>{pL+niyj-
zctxXkUl;WWt2u4sFUjLbnTWc-qH2D+;3j4WGG-q;d+o0}>kZ2VL>QOv+t6wKvv9Wq
z{Hc`t3S_0vl(LCwZ<P)(DAv){)k|;-^S`jNap{M93}rd#!MPWPbD;)yuP32rK-N*S
zqo?|5`W7j_$Udf?`t8z3r+&u*S}(w_zZ35q(v&H6QnFPRHaOuFSbHc{9`|JJ<P4qp
zanP`4*PML(xecb}Oz)X_*`})er{o_r;5_sLr1&bWu&WlykF5vC-1%RSHP;T50XON2
zFvfPtQ@&<jQ$k;o74{sWzwiD(2xZ&zX|s<K-?cX@pK^WI^=!mOh5^hpG#p$Rb$rg=
zB$$tTSA-m(IM0(W7=X>^W4zZvn;$(|J|=Xd#S;n_N%i~ilg`e}=~7rksbX~#z>4R5
zImhOCWeqzRnbYb{0kUb)ckT@cEgXOKk*8TD8?M~85)_6naT8^#@wRD+^cCjai&4hE
z{HkH#-L}0XOMbMRs7ileo#$78LG@ER)tN;%G!#e<Vx8rTe&X${&ePHLZw483yFicl
zmaTXnb7aRmgJu+rU9=ZFm`N{qz}2-~cS?cAEHNiBZMpT@*}yRUgy>8PhWQ3Flfy{!
zwJruEUx-kN9pT?&yzlF?1kF}nM?@tMVB5`$c!t#=T5gX(Ge-Uvzv9JwOd+V227j1K
zGlg$imhWS--!--0HsG5J>-Cp?PqV*WBD!lzSQrq&Hxnn9_rY6jfQ$No0OZg_yDiiD
zU;(rj%^)?tU-6hQW_$jpEe(3Oi4i<GimZhe+eXrBXwwXhSA`We%K0nn8%eE}f35FJ
znoUlT%l4lfNj2ckIX-kWes^3-Gc84xr)1cj6|J^-Aye5KtIWV0-4VYx-|fz&S9{EX
zm;PCa9d;G_?a;z~l9&jo2~bahV&Vf&@ud<Y4zK+smURC~ne!#2&93NnswB-@HG>0d
zT%hWs#+=$oHhU-bAc#D&ozF(q!bSzGC^s0DPCV26hb=v?sYHm14M}lSM!5+uSy|SG
zpJ<+WC!Ms;34z-tpH87mjexz0@yKCW%uXQDo<>iLEShV%0X)#x1#q*W5J;LXnqg<5
zRnhet?D>}-Gt0*tptlP;UP-d$!ftz;oBenKN_iTd89^l!k_&~dhkc}@P4>T-ddJ|(
zp6G2m$t0QBwr$(C&53PePn=9_dt%$RZFAye!inGf{?GfMR@LcUr%vq;z3cQ|>%Las
ztzcF<Y;1BCAG69?zWvL2Rx`NtWkRn%yVeu6imh_HU7zpxPx}ptR-V{kmaDUVU}HLe
zZn9}3wWD+PDEd!ouSixgFWf?BK!eJm;gILB_jW%(%FD)wuRh44$qJV!Oza0%+JFI3
z8TfH|7z(L~=Igj=eRBzVkQnwOp)5sci`qu;X?zf#k_fx~*1w=IlK=w3uni$m3@nFr
zDT0CFvY%ddW)dJbGHXwm3G``b_tnY=D~Nwk@_Q4OU_-IP;jBo5HW;}Cfy`IV#9h>y
zAzFa6A}O==(q_THUu2dhuBT#P?mB=u+66`vLkQ!9Ckwsy(#N}S@>enuEtuur{V;h4
z8k9zQ?slsqNb|0bR)8R+D}n%yV;h{T(~Pb)34}@yg<(=8?7t@>qccb)7TA@a?JB*@
z5rLK3!ORh*!Q@~vc!{kwK;kfSgmaWMA5}{^8%!(}eDzUQLAuxd4P=8nzDIQ{j>^)F
zg#8plr$EnJD&x#Q9}gPe_|3N5DELdk10k~I+_k1v^b^6n%$dJ4LI!_r+WPhTy8Lz+
zXz)m~!zU#$(c4Rd*Nj59uw~m=JsiP3Ykx|{TfRq8n$*Mo<vvZv2RyqaIg_buF#6=l
z#mGt8%Vy=9&ZG-7We_-%ll;S>C{Vpbz~%-*V@k-JZyL@_P<--Xxgk^Ob4ZPlHIrmv
zW4HfsATSYBLolZ+zK3%x?Eu8xIz$Ty8XRq|ta=vI$flidlXMFNTy^{x2fO=I!!Uv(
zE-e6PlQ`{^z$ddG@XWd;&4nwx<g(?{an<lG{gU^F;lT<YTAA1+k$)mM$62qZdt*_;
zj;5(lM?_Q?VxCiD<%Tob%Ikk+a|Xya1{?M#<w!9?;DW{&lpe`fy;}k072wO`8ZUGs
zU;A@QafcM2c-qOvMh8gd`lUfe@S(^B7XCP;fsv4z+X6bC5p5>n*mG%f*7%w<4`R%s
z;h{;J(7T`3ZN3Hq46h{}Z=I<HAs*KC7XwR;>S0{57qOS6-_9QJ$*%6yAthSmVF3$y
zAI($LM|KKZ!eN9<VKUJB5}A0uv>5{*#OsPHUZ+*m>Dl#5XZ@sti^DAp<p4;*S@5Yh
zg(~PVA7E;>H<@)P?Wbxo>CEhf)q6yw7Kx<**jURJFN}MH9`skARy#*<O5V3$Qh_qF
z1#nh7BRtg1uG-Y3+<ke)N*al#2~@&-;qDMlQCcYL_d<FC+%d=ayqhn@(?~1Q?6r$?
zCeo}gb^ZhnHegqQ^!gJ>N(Sd=gLu3(lI6k=ufP|05Pijfr0L8_GWb>dj<T+>e)5R@
zmR9hmeJ>4+%M>P0UQPc`7ZG5G>sNi=*K^+{D~T}oGRT(Zwa3SjG9}<>msoARpf=#6
zo29TB+(g;TE!6~rKjJXC9GLn>245nDH6AsL@cHKP?`I(lZx#W5d-go9o4@5mhz?uV
zBCwN1;F8Y9c-r6cnlSwb9ZZ8_f@+h)j6$v=f)dJUt=(Gg*5GjC2&vmZ;)5_YwBqBH
zYx}eLq1dT?fh9|90%vu(li8Zq)_R94Ev{fHpMrdH8<VKzcSb?*n*D>Ud72lUKPT-@
z#PJP$QowhL2e^~&8rJAurm@*ZD$MBa5x}W!Fw@19k@b0?#U%pM@$8q5U4~I}7FnQU
zNAKrE`SQOrMJtXjWAf6A;mV(~F%5b}oT_U!e}!v>8rc|lIUn5r;6zanp%*e1_17DD
z*h(suc;>qbZ2FQJKT~l*R`+gcNZ&=WIbnu1kb?VnMr4LqE%EBDwRcJ!pAD;@Ljd}C
zgZsTuw{rb!!E?>7OCb*~QB1=bf5lioV@G9U%mE?@7M1<hP0>x27=?3&qOp7s1)27-
zieBf?wp!F=Qzh?wj_{sC`Ap<xh4Mdtj!XxCLh(BesSj`_ANARWgSw^M#EyG4<#(co
z0G&o?K4f2~p*|V>wwdF0DKA?oM+cA<y#he%`VIOm)v5`WbR66p`ruWy76K&`LbH%F
z?@7OtVc16F?4&Q*DW;`gV_Y)p>ah}(6>)d7cf;Vo1UpUcLi+0(4ZCWFfR?vg`5(Bb
z_;CuX9{Tz=R<CoodXu}a;8o>5{2E^QaUgpUQc#!8*7?pVGJpWzF__<|k^($Kv;I0>
zKs{!Fgyy=B05?~;h4$J;hHBQUW}9|aC-O;lU8|>44Ff$5Jw*y~?=!ZS54V;hN;I=m
zxWb<M@)lZ`3hlD!4Uggb;%n!BZKZso#=ST5@W`Fi;@WIJx|NBLam@u#sSQpdB3JzN
z+%hLg>)c!H3MzJybd6M2Y5)P>57E=jIEk)>woYQg^DjO6lSVT&PIX>cV9{ayqCfW`
za%P<UdA|Eu%3gqGaA~7>r%S^=Pm)6v^dkV5<tgZzp@UY0y{$UNmI`(ZnLzW?+P7Wl
zY(*xXO@Jo)Q$FJ%v;9-`Sy6MmY2J$Ju%2CCv18#@fYEmI#fte(59nB6kHM1;RQyqL
z-j&Vf-oCA0VzW3?v^c|gnWW0^r&`^vP+!$YuTp|%g;xPx;STmNKaQ{lt1oU9D@nwB
z!6=s?`r6xh#`%i~ScvM0x9_&QkL-MFt}UpGG3a7PK8)-1hMwjbl%CbCfy#5UY(UUA
zl;6Cvo=Q9(Q}`wo0oanQZ;_o}v=(i+h0HewLem)QGfKU+^6mY6{)l}syd5}sf9{{|
zM~O#@*a;!avs@H3RP0Kus+bVme7PyJxXtKNF((HO>s^o|1L{Fy?f&RFaRO^3n5l%L
zZiW8HuyY#crM}%21|ZP(mom?Sxf_}=LJt1CV8l64K^V%gfE;+ziU0PBbm6OZjBw;h
z7|PiDCRf$+cC!Yg$3J)Q7Ny-7XWlF2i8a@?a%6iJs=0^>$4>ieUTwwI8S}*5ywpZ(
zq^Bt6e?R-a!wG;xoiYU>I+f>f>29$7+r3n{kG4hfjl;Hg>HA&=@^Aa$&3#cpBfmj~
zNQ$qO4jOz66o@AeAd+5POAv}9%ah_kB7?ob01sqA6W3mN4^c*oDlkGbvwtw&@1q)r
z^o7_uOScw}5N7r2!#62Ok*gT-A9*OfSuv<<`Yrmq7kxBMKbD1IL{h*C(!`6yBxoih
zS_+WjV^}Si;h?81584za5xxxo#pd3zJDL`aIo&W!2kK2bES+^jT0STtZUwt@HWyy}
zo-k6~F!bEo;F32Hix1KUQ2gBKn%LCnA@y4)8|!A-)4k3ME?JZ7OUgoV?f-h(Bc?OV
zZ$xdE8+)5Le72EyO`$G>;fQ$;Am2>P;AiedRE}FoXc9hIp7t!V^i6=eI4^9=i#TfY
zBszxS0@6`@`&lZX(s1Lgs0|bCv5l`&HlYXX%vl^?M{PoyIlN?g_{t`PnN=4CmoDOS
zK5|?3{|&HyuHL~b)1yKWrPmg9uykkU6Xl&6Y_b#k$e{?n_Q`o9F2jp}{{F0v5U@EJ
zj=k@BpvTH+v=m%oN|HnL9U6;%7M_Fc;>$Oz76_{(G!H7YwGP?kIv0mV{0|Y7QXu;7
zxJWheMo&z;y?(hLvVD*&iQ^`rhSX{J;}fIw^?_?6|3qr&d~};nJ0e%F;yYWjEHhc9
zva@dhGw&ox0<{8L=6(8%2%?%Fywe;+=9`qi=jUVpztNpBI)MD8>Hb&(8Wt(t$4iy*
zF;+W@jd{!W;^M8jq26ZG&3TrFu8mo+9Ctb~BS-66;0V^A=A}5ZRa$eQe+=duiLE@0
zJ10?*BH*DQ^k%&T?1e##@b$;SO%ji84gm==`BGFqd_Li3s*h0?VOjD)U_qd@DQ6uX
zSCPH}lHkS0;7u#EX#R&^E>px3H(zuT`c7AbtMv;{N<-qe4zp+8vqie&`-u=WUB>Hd
z+6h!(ntE6v@ta-uRHO{QpPA=!j(mx|uM%Smr0k=s=3>Gx7n)hG0YwjABr74_tbQXb
z?yCY!gMoGj8KRV5=~F}$JgmiRe-37gVe96IR-}H`)+D?I#R2>o>R_XOb(D*KZ%O7&
zO?h6*b!m)h4Q*}4sFpu=wA<fX9~5W#z=(&7x1}Y+g~OYgOFshxkkb*SsQcM9+~?)C
zaLPk@jC2y?J6#T^yo&tmHE`4*xqcbZ9!wsjk0k)HmrTLmpgQ21_b;siw)W};##-7$
zGK9X0e?-V9Z6J(t*LK@X)uK+xF1wSPTL9V7@H$u4_gy<o%{OeN@a|SxT&m03bRDj#
z(|786(Ydj@{FQu0Ji!GQPiUHAH2XrxN#I-?=O_9Y=PcH=6vFR}@CGH$Eo_>-@5RBQ
z#XJ6JPHdOr9G5;_2^(d7O9qB)8$ud!PH?j?UbnE&BS07iG>7(qa08Y6pSvM(WD<7B
z#C?)M1M=#w1gL@U82G}yi%z8waZdxk>u~f6b*k9JH)mIB9zTCe73fyIZ}IZo`PF29
zFHXmXp4{SO6fdo;(NI)~e<tsd`)O1AmT2Q10Q*UnJQ(oV?7msrDz3lwy<5o9ozxCw
zv(5jzR}Ij<65_?INo0u0L#MaP5i^m@*?<2X%HA1=`!LJaA9J6cO9%NqgLH*c^K$z8
zyqbN|aP)jI2LdPI(|R;XeD+V4OGU&N>$maqiIR7A0pEXLja8~4(eL?j%YPKU(JX##
z6)5!$Smd1|*Vn0dNHiBmMn--nxa!~$8|<jxH~s+3QuXGLuT0(*L$YXP!t%AV_ZD*9
zB>2UhUB(y^W>%sSra$+S1`Q0KvD%=<QlvRS|0;y6{!}Otlk9F}WH7qEM9iL`Eu!Al
zXo-R*%TLS`p+WLBET@{oS1^Z}YFvnT4ds+ge^iXH)(hk!<(eitDmu0^od3?xbL~FD
z;Ajgh9g~p~hmf{@QS{k!m+=MuShD919=vx2Uv*6)L2~zSMeg7yn&p%V>ZA3!>Oe@{
zB5hCkXAF9Sd%AcJ-^0n#Mmu=xgU%e69KsO?{y&zYUv>z8<{d;N35F3II1n#Cpyj-o
z*0L2-j+HVEIoAkfgZJ2_mR}ckQ5lG#xxf1F5zFPT?Jj%b%3!7C$vx@CjG+s@&Y6Ru
zni&G2bWWq^nvB!G5kO?kyb-H%ojq-Ik2*>EaX5J1#3T+^ewVc7)0zVi1IObi2q<O<
zl!+|nt?Fzne=&qQ!QH84`s9m3!kczT*T)My#cfAeqCrEP)XMuJAR-QTpUbPU<HZ1w
z6+aTH)R_1}f;5Ga1<-@iA_rHL5d7Iq${lvMg}|uo63M9mGD$>z8a#c2Hiszz559YP
z%c0{gr`+g&u3RYI$}Fg8rQb@Z<z*aaVX=|(*3%BwgR0$o`tabrj+ofU$ZP2{@8vzX
z4#nc~=SPi0^2040Mqm2o2U+I_^XGuU*LZ8Cu0h>*_$44y<1}6Kw5N2I@$IKIOAE6G
zn;RVo#46C_<y#MB0?wC&osUSz!_6N<@Bj2jvCC2XV%&Cr%q6bSv_t_T7&#h`kU<*2
zk2V$`hbVp_-)m<YjOesBQnU+$M~$%X76}@I){W1dP>N?o-~*v|sl+NLs{zpHrmlr?
z{|2A+!o3<GLbf!3A_QH*iep<6jb|%;HJ6wl^epuha?Q6Rz*p8U$}L>s)(~RRT9;Lq
zG3q`RqE3)p#qO7&TXoDs8}kU3&%pJZnjtJAlIgRgZk-hi|EpBUhlW;3g{VwGVylr5
zMg_qsDEa(**Jq=yQTbH((jUOpI!^Q0kC}!!AAoCSoiD>2Tf|>3>p9_J*v7zYW(o-g
z7YK@(Y-R+xH3^vb@HNt!aaX!=UvmHLbo~#Z0Gkt|ju=)NA-gslo?D58DyYc%3)yd?
z$}S;kW7Naasf+rqbKgl<s=?1uo`)WCkn4rhj9M}E!F(Hs%H;2b{05AZKqN8u!4wfu
zXyx;Xt>*FZ8Lb)xE(y!@nW9$C_hC#*t>ZQuNz5Df;f6KhVYK#Df>v0BDAW6rU`|_e
z(}8mb+p?=s&NqE0pVS!eU2|22Y*-2&(-6H%XP}l!t_mHk=bXAEs$W9NRY@73xewRf
zN*{eq?&8NF(}u5;{0Xd9bhgEW3l!wdEc0I=2e4Gbw*^{P@4eDYv8Of?+H1sCB-e-Y
zBCv%-Z3rD5p8TdT{XFD!&grOx=PLA~V;K+nk;IIOP{_MT3M-EBiw;ejs(zMS3+XUt
zwp>65)NDE8FMTXl5Eg0xGm*o=m%sYM;lJ>bX({6bJC=2#)-9lABw40`cz+s0kv9%W
zuC6_^1rs!tzUxXO`EC>Q%pRrSBhow}fsRinCr`ymNoHs{;)7%{gs2?~=|>*{=o%7;
z7AiH9Z7qdLFnstLDM(Qsl?sv%c;5e-N;a>wZluA29{!guF{+ngn%%BG+{o_t#KfWk
z`-86*e8YGEQa8Lv2X#N~E<w8Ufgvq*@vEmUsP8?<TDY^b0_UsfQ0npuk@*GYLA&}@
z(fcB%aCo|g8w;S1eY-3-$xPs2rn>yQ^&r=*(K=&{`I5n2VnhUCk!4J&mMqAw{(TW|
z+ov@3x-4d#&Hi7B3Yn-I+mBs1NJKU39Z|<VtiwKlc2ganr?qMm9a?Jy^-*gAFSeNQ
zWD$!i*6X9#am3%vbUmXmUE(JGqv<tg4R`h}aoL4{7^bh%n#jIJ0iFcyL(;a1{G(l}
z_-`b4mpDPaAs5QD*P?Czhl|_~)zk)fI=6eNTb+oMUt;8O%!}`kBYU?3iGN`W1WwQl
zz&1Pq^AQUj{wX7nUBvw-Cxly}*JAO#jzx*&>t0uoafNOhoaRCne<l@rG~x|XkQ|)<
z37utyOKE5MGxf^PfA{i7opisGj@xsV$us6SBg8w7J;tLRJI!4moDdstiPhoNwwh~;
z=lgD=Q#+6b_+JrjiU#Y<s@F1J_rC}taN5OZrf5D~XtvO~;xo}e_ACfHIX>P|3xB$9
z*3MH;T%B%0L9V9$8T8&k+Az<0htfbpN+&(w+WtZQf%7Q!YCoI0GjdVqut?+g?NHEU
zgJV)}Ptt2$g87zU<Fqc5rs&|`M(7!9N0r?@fDssG>7kyz{lwN`qa3!4sBSU|<c%Oi
zVh6=6H|^c*SvuQ3&lZpiXlPJto}&}{;W$zl%;2n!6v#_SXN&h=5vcw|5AsLyklk!i
z1_3Wc^Vf1k>(nFLt}+pQok?>>Dc=lixFd(&M29h0m)Qjsg#9#1g+*!M^8qUW^15#d
z`3o8@wZ-sDGx{w5#p9S3*bJNk(MDhR*dM)UA5>ag5kw(JHf3C^oASdKn*%+v#?9?V
zO!f-tQr-(cYQWiSx0%TC)kCn~9M<f{8GgFA?8-l#qiP|4-9Jx8`M7U6pX`r5;XB`h
zcjlZSOevnOLc88r`M@HaCLAvmejx9@0gv4w+)jnEuQje|YiOOuP?#HumWQ+eSo`jk
zzznzFiyl48ECDsAKkvaF(pyT*MC<0>S*&J4j3zid564=*G-EGVAHLFM&MJB{>-r|b
z^ivcgTIEiTPgP<1kHTl0$^n6e2HOt{7`X?Wi0IG}lDb^74Em3}j&TN6TEfLl)lgJe
z18xRFJOW84<9|-iZA#Q(8#DtzFb&p)>Ars9ahTzZxiI%$hjw)WaB`i?ubs~g;llU3
zc^i6VSE8IuB$>QgR?)!LjH1Hj<~DuMo5hd$74Axoq9Ggk@+3o;C-j(0X*Nw-W8jRB
z3BrFAjbO`I((7@xkc%@Kp9zDhYx~e?Wg_Itsm|3=D;>H0VIU|rtceQ9$Mt_h7fBtl
z$}$y5N{;K5J;6kpoNt&!2|0OpSaLt8pghB9NC|^EweW2h`Y|EsDUqHOM+{X<uHuvF
zvR<~h4ZYngNIER=wCJ98m6#{DddQ{-{~`r1$$guJF}k)&E_xXfMjIb*w}tBp)h--6
zH+e)!m&qb<z|r%m=Q04~wxy-O119@)wR>Th3jEnS8)uX|4qWc7CE}wt1j(h_-RZds
zV;j!5hwj|G7a^yN;u2v*V-E(K$Ac}!L9x;Mk`cqWT}$R;h_LwalF<X}PWxvDj%y}U
zL;-u#B9hJAatfn4lwkX!+qu~Qq6{(vl#2n-Gm6=om5#}09S^Yn-IuQ+U~1NoYtq7m
zTb_GMf@7SNps)T_1xa;UF6p39^{9SOp#Lso{pVrW9--x&-s50@|1Gd_oh&gdxCKFs
z)$6ZTy5N-fEO%u}rrGpFt1)%7x{4s><sI#9(TgCw^Fxs^L&XlxAy95ugM0W^u-z&I
zS})4~1GVGUxB-~ATzoRsObJhjv59`$)u7=2WK%T%_`bs84@SOG+7=h>CwDE}s;}|A
zqSM$kmu<t;;tsCk;1iyrrQ*T+yxeX3W8tmKd(OTI$nJ(>{KuW)_k^kPgVE4@7jxG_
zzy}EVwDtW#`m;obw=ze1h!v^85({FV6W-AH-uSpISr(WY-*Yc+T!WiRL@me+7NzA(
z^ec1w(x+k<m*H4`#}qzLAL_L$CS_c7MNgo!?Mn@Ukzjbp<8WpW(n=0d;Q-eDe7xSt
zTk9LV^Jk?odac2df<t|Q{ygWVH1uGRY%T^#`SpXujL�ZoKYdaRL`_+HJ=l&8w4y
z;Wckm0zeu18<YM7IyX9XL9R#fHmg0E>Z)_W?OXchP827l;TU4x(rL))dk-by=d(p|
zTsIFb#&7=cf>RO4l2=n|u$}&*zIpJNCd9`FiK9G}vY)h7Sz8X)cHhR+H@`is3b7j2
z8>zT~VGQ9tBW}GR<|v)YnLwpv4PfSre<W)?RRSn82sHw+FCo=Wq{9-}rceywqTO<d
z&@tZQlQdCpORwri1OZi&AKIIhRMrE`3gasD5)k@p-|#f3=gGOra(B*yBVGnYo(o8p
zZpj8p(%0Uc%r8%f7k7RKLTOwZ`f;bK=>LTQjV#LXwP&L;QaV|a^Q6ov>cnp!#6>kD
z9|Nr0mFl!Bmn^nkZc}2%P?fzW%Xs9g4s<vctpmSgP37+s{Z!-^hO*n}kIpq&N9Bj`
zf-y^SMcDQi+suTdL_4F3`JMlU$5WTjhK}cy(Yyyd9-9N18ITy_Twxmti7hOqf)t~N
z-8o{TUpjXFI-3vTbQJ#86BRiPBQ%kkxdqyDGxmgn(A*<Z?d~%3E#Z=ChmffD%q<yd
zFWBO(SP3FXJxSlVZbRR=e%Ri?x#bY^WV(+9S@8>3y07eGzKaBjA}?08K`gLT`jg;{
z`)et;Pz!K!yl6cSkXtvJ$-iI^tuHqtI7WFVAW!V*T#^qRw5(H^B#vsdp_t-<1_Q08
zG?I1hzd@i4LfgGo7N*jFM4#1<MCB8N92&rkeRfwH=p`#Dj7w)fg+7Sc9jYCw6;HFB
zRnb_p)jOXlUAx1#!&LawC1>Q4RB(h{6IF~OLJlWqQI;g@?Dmr&yle5%s3RW!QBuP}
z-hoXXhOrk<uopj=gG!b-RI6sw=mAJGDFp7H7)9-*TR_Y%j4Z#25~s5eKfk6f9|2oU
z>+)#HW0B)SKkC=tv=q8O4q)y;JV0Tj(Gs@v7BLen{_H4=jmr|;qY+iSXj$eOE_D+%
z$V^riAGt%0*5T^>X>`0edT<!6V{xIw(co;HK69Zf9T=^-y``-`3r*4T(*V$q`As#v
z($P0bM=M4v(iiT|D{88Shsf*8v+mo*v`kivfW@)R>i)X0k)Z<}(}DdP{KHk~ZnN^m
zI<WqdB^P_16lTN9t1#4rfV5UQ+W?*z$as%(D$U9|kC4s}<ck2@b&*FUay=3p#-_WP
zKbU1Z`}1OTPyAa<O5$6L=)i+(GKzHt>}$|?-=1VV^ryL{(Naioe%)`(3Q3zbOg@eE
z_#yp{<~DY_Mz-N!x`{M{FxlOsiB%m8>=F5bLJ^4gI?V2kb)pehD39_c9(WID`(Nw6
zM}I_dW`-=j!LTWt`x=l-!Fu1J^jLZvgKe9?A0axBxb^awbz4v23?QOG(IaFk)sy{E
zm;RHRA7T}!EXJJ}<)8z@X)BTR8;|_2o4pt{S;(1G&F?zv4&>(zN@V1REk2slVg*kt
zHvd#janz6<Jkvk$^)6VCBNju2ie?Z-HBa@75I>={aOG*~_SKC77B*on$7h@4Xf^^F
zH|$mFpIqx<Zns)AcmeW>+_A){(&KNrVogX8(EY<XD1l<?fl43})Y^WzxHP@+FylpM
zPp(KLI&CwPEfLCY6p0se;oQ{YyfC+ssX3wGalg3ZNE2|KH)g;@QOJseetW@cDp#QJ
zIeES7>UfGsZ!J$`OT)vf)N3CYWUt0iQqx#*`^|)mP>rK@KLMU$3)(FQCEJCeqWpqT
zt-Wi&gl!PjAL+8RSHR`c{OCJ!#%K{1TgK>|SU0_8egvO3UVemLV0vDJUSE|MzrQeQ
zThwF#UI4`{KLzOj{7OGZ2(=BIO=`?M+J`-3l#`wKLGG}mg_bRAgy>=D94hh{Ucoz%
zaU$DWka`7ZltEE%_=0_!XuEQvxY1FVIdWXAGkjJCq8%;Us~4^9X68MAJ`DCfo<C;J
z(<zet8juzG=>~(DA7(O$Zkih!1B`IRNl7pjuI9dxZ6l9|gZB?lSdwYyoRnfCkTt!o
z#XK$Fjn?WF)P?2OW1OB%kIot_Co@1*F8vIl&DaEf*%Lc)mdVe+o=KT9KUr@%1~>+w
z=o8iO%oEVhE-2VB*&u3C>CWWui-=AwzPL)s7rO<dn6W;L<Akcwd=i#XpqS)<!_S#v
z?dy<ChE*h2*!&t&Jo?ea+yFar&b~K-e3l^mE`8np4f1I(E}Mp}R<lxx4uY5n9b6>H
zpB^0G0)bKDkzD=5<09PKkuw%LJ(<|=f1TQ>o|3{b2j*MKpSZYKppqm-J84^**_Nqd
zE9`>p5!%F`a~*DuHHfcND-=OLM4TZH{ccl<Z7O7^L-eQkZ-4ITY4ZWf#P7i+g3^eM
zsdW~Swaa3YHIB>H?G)Cnzz+xz8pogUxzgGIfu)fs^s`@#H>0e-mR6Eh3<8R<d$@Wt
z*g$sCoH70L)o~y9iM~7AGH*uRv?<!0>TcBA#|0lpx)8DRQCk$cjOI@Ba3S8aGH#4N
zaZN#?80xQbi*h3U?N1Fm&Th!E0mea$_|!x!S?lczRIetHs;Qte<tpU0RvO<w=*Duu
zd<6zFPq5ets(cYoMvZn1+ep#SbMQJ5I%ZkitsK41kyTR-4(8>VrV4KjCq}uzHXV6^
z`v1v*atCG5^Td2|Oh4FEiDs1<MY^CK3OE%_@Vlsp)4P><8K2$#yu_z*eyck)*61{y
zX+C(=Ed;KXb*nLsEpf|TZ2(?fOSbH14K9h6lj&2u;)t2a5`|Yg!taipzuD3_JcXwi
znyj8Us}yIxyt!{$gtT-DD{B}&tUj!ACf^XNndQw2jGgJpADJ0<bv<B?SDES+GwhC6
zaW7&%In(x4t5$S1tL`3ikFS>^-7DVD@5Ma>q&w=y*w(sVVX5m<83AH>`fq}FYp=9#
z7A~;l#!c>8{E0H)htL}`(O}%fQ)ep^!!(764eC{T)*1*ctVQyWI-NmMz!n*v4!5PC
zn_`5!gVMiJw~j;vb<r~}xR}|j3BQ6x(`!4u{k$U|2^w*Rtj)sjQg7e(jUhG$m-q6=
zLzBG4Zf@v}uTHG1HGq?~3pjj%a<afU=jRht5;3sgL5q{B`ARI~`W2?<>H9{ZI0O<%
zWY#V3?j5g(f_@@sVX}a=D%?yK)A1KY1GZn7`3d)G&137M6QdIioxWd-j*Mrz<QNGf
zDdROcW6OKFTiNb0*0+uC3H^EyY3e*~WqS4u#hUuRhL*{KwE;Cm#EFM*>Q5+}G91mr
zr)x6v1z5vEAC#Lij<mud1Vy!#DxN9?gy`N${kKxa@wHEEv)SLRG5_f&jZ-B*$<h?s
zN2y>J-7-*x|2MEu0Dp!~L_ilaW6XqP;uEtCCgKSd^6CtmvfHNa=FvHp+8L4Jr8~qr
zS_r87q4y4ffiWjL36PRAwx$@HHH3}R8is1zF;Q?*CbNKWIPp8QX+-6uGros9_m77+
z5r{48GL)kbq^H&-*~tDHy67|on^{TgUUycGhn``!jXAw_3z=@?HCAdtDyK*9S7f(X
zwt?L(sqGQsY}GbTtx9}qrjm#SX8b_QfYGBP4@?x`QnkueO+e$PWtp`UrX-623Xxq@
zO2%WclUfkox+}AeH_!Zw8%M9Xm5#jyh*yK-ut5<dD$b8Z9xVuHLOEy55TO!Ge-d&s
z|BMJmxO$<#={ow_HMIdCA_9q|seg8Gf*sVNWmUQ?;~Itu|6KAI|A_OoY69%^Fw#2z
z1#P_m4==ife&|VL1*5$fGM+?fWQk<PSmKzwX@T}C;iA5KO?(tb`~_#$WP#XxrV|Cf
zD`zvE&W(Djv84RyRC<o~z_L41bpIJ{xveq2$Sc><{-RsdUPLFygG$Wdpe}<~Z>Qc$
z0OF~8$hq3+8+CjnZOgF(PSHh#EtF|lP*X7g2%B!p_5q)=@JD@(8Ha8aqAJJF>Rz0F
z7<+bBJ6PIMDguR5LzC>Sz?^_;^nw$*fP*%Xf=9|yXL~}Hf)UQggN?)PL^!x*0)|;_
zPZz%q2wVJ|3Bt&a%XBUzKXYN9IAQQnqlk4vWDwu$vn@==kO-Ex6Ij)GkOvG@a7|=V
zSofAgZQ{_yD^&+sBM&h0zYaWWa*CXArASj12Ls=?2SEf9x$m=pXXcUKhEG*e4(Pw-
z+yvVVjBu!^URp$&L3N3g#o}@@9gdLKwTVgFW-}1OLAqz_K1mtY$yU0G3-D;o8aeP7
z-JSEwHbGrG<GD0p5hh0R>#{5)c^HrWx-7C0mQDWGoa|KY-Va54W#`=Uo%SnvQ?O+%
zXGp_h3^<xAn06|FSyD%<<zjVfkWu-?N`<ESd}pb@jZq63gR<ckJv8~5gj587ynEC9
zDynouj#O#YwSsOc(IL!O7q2t#x})i=y)nZ*&GF*!pcM5}3qQx6KstZ_UyEDttaHjk
zQ6i9U;}Z4f>#Q9F-5IFaO@mzZjzFSd4rR=T1J08w2R=3+r=l<@_ydTweUEHcfecW|
zakxeASI#k+9@##;ef`r9`y4TOw?cwnxJtKt*1!|m*&pPz0;8^jmJnKY*Uz>1F!j`}
z<(h8k<tTK05SU^q_X2LeA$Oxxd!UjSkJ-ghcE7$W(9)cyq%>l(o+Lcg&5P6|=S!{t
zhD2t|xLyMELK@S7A@7B;dp_6&dc$va?Gd$6Z*3p*OYT6=Wq~EKH!9j)V>4xWD$=uM
z<cTEKiX`$x50BuB?MD`9Q;jT#h6g&+X=x-)6l_ktk+1GifYDr}-ZTD@E}q4noeo0d
zK7-2320e*izgR2;p->Xt&Yr`yjWfE<T1cEIrL1AAYu;NN&w0=H$|n*@nf#Hjk<DjG
z^L>aVP(orBTtN?|_we5dR!e%eE$f^G&h;$5lE`35BE*vf<Wo-m!`e!N@>3Tm=tLso
z^BO>WuB6PYi$(oYR923evyu{eUs##vS_d;xN+eY#IfyKRP)Bufa9-i+{lLxAg?B3@
zEMpAeK-9XOjMvVbe~XIf4!)YV`Z%HbDhf_Mbp2&~08$hF|3q{wg|ibGZnYB0Rh2~!
z+O^fUi`7vyaQ}=fy?LJ-wfqtQS3<2WtqWY*@>12m7D)plXYyj`s<EcWUmWft{TXo_
zN;sOhM?Lk}T8!;FkW}2g^ps*0`WBVhGB;b^IoQJ*R*>U&Mg&1Y7Yd;_BkV?)Ra-V_
zFlIOs40e4Av!0O!kx^&v5-pOB$_w6^cEBagqXG^S-st=+qU_9N%q<B3=VXe_=vOR0
z`9SsYu>bbba((f0ZmeCKNoo7Wf5jjfwaYn^GOSIvR2Wya4`t%$0zJ#L%rB6HOdEN=
z&0_K9m#vq_?S!cm6nnt)z(s8AM0C&@oWb}|p~|V$FnKy=7_FQ_x7VN;&+l-hy?>U!
zBk5eF7600(dCgbfN(tEmjp;;f@tYe&9<@4^C8Ni__?Y*aVgmyC#t}`X!%FjG=G{Bl
z*+7p?Q)Zkqb+?|uC&nH5J(e$WO0wXwzgv94)7L4KUt9WGJ+~{>)X#siwcfPe)aZE*
z@BMc{Wu2tuS$D8J-IHTDdD;F(6K8(<{V%O8y%?&2<#~ylo;6@GAUI0!)8ohIH^Ia^
zT5-9`QWRbtCDg5=k(4)~w|XM}Dp7qZioovIE>SSVeJ&^nr%}l7_ruaIV$mLBZh^IJ
zbBOkLN=o<9n5MDtLZ%cpG97=(G#0zEb|S4HRa?6*QLSgUV-AOz?iUkuQw+36_dED&
zjTX7VKntgzl5YWrSRTe?rVG5kWSKYtEM(f)x8t=g#9|_~&~zYjga^}!Y2SylN%a%*
zVTZE|rwgIl=9qqc`4*QXZ(@1GSZet$ugPtCy}p${nhDFj?1Mb?DvIq4NV!9Y#qJd)
zF!<Jez2~I87&2~J$7MgOlpWMJ%27cN)=^RT{+pjy3g9og;a4K<tH%EI)ft0tW0Ak;
zsr<R0r@Bf~#zc21XbEQ>=lgfm{#h|J${HqOxOk$tpU}ooVIX*iGEf^+5OTbtZ?FGG
z8d=4r&Xt3%+{rkLvx(al<gffx{B~Se)xvL}<;+OkJIez57uRj)Ro4yAvW_E!3Cln!
zBkRgO0rp~&j=b4AxQU!Gx?Xv+8ahV;B)BOjn3S!eW;}PUMQRV5umL^`a1qPGZvz>*
z0T|8HWK5)s5|1HLpz`ACFjJSMoHPlNzu4#(!DbhKeoc*o*?VX5Mk2zdXhp@7>pJ<a
zY#_&`{)7uoe2iIC7{%DUNKY*p!qvrs@OvvAZ~$k+l&lh%d66jrXbSd>AC5X+v>xWv
zH)xGud0^M&oj-Ry@UhUPM&S>W4=#5w3h`b9b!*=8*fvTd9oOVQ1mm)7dasl|@2xbf
zjX5cU7fU4{qh|@lyp_UB_JC8d#aCM@YzQi3+gAx@F=<z07F!A>t{VjGxe41PdOJ4&
z5;;hrKj@+H;HQ362@FMCoTK8t6<;X)eCtwt!^dvK)J8ox3^oM^EZ%y%pe0Gox1+Gt
zDrSvB{^*lE$>9q&DwhAI^l-DYOb;Pm(x4?Evwmt6b@K~cc`FcNHS-gDQ2*dfcZjym
zv>PQ1|LnkC%@vFrYw4ZYIUeodTY5VJ+Kg<f?E^k|-qUY5Zx=ZH*}i}tP6LSd=|Ox{
zGLK>^-B5uvK0bPEMWD}7hFX?|Psmtle@&S{6ZHUJ-9U<e7w1w(Qf}$7zghCzJ^I3x
zdEG3Zr5cT{&U}k!+z*kQZTbv3)2V|h47crmIFr&*ev2Q4Gx0v_bHfPwR6qHF@lZ@r
z9p_v~w?H;f4Uh2QCV~E@ZZK|-?~DHP_!$IblG^-ho$?dpa(|aDK0Kx5IfIQR$YNc)
zl|E)X;kGGTYFw9CW<Q!($;8A?P5vgmhYY|#Cbmgps89XmV0GyE*;S2{FCY?RPzwS@
zYNG>wExFQEI;yk+*23%h^tJK;n;W`4ehWIatDq{DuxKiIyTWD8aladtC)?gATsC+`
z*2X<-VF<)goeeiZ;jysWis*S4SJz&)y(t8lFYecHxeMx6%n81w#Pva@GWjm2FvHmE
zcr#8IDW9B%f-3vIQPBo8K@6y(JgRq+W*nUOvGUI2rDAjy+V&X!Osp~l-rEs<rdM70
zFO81(6@Mp9B>EHHc8uo=K9#X|-pzN_>)m8IKC6x==5yc(zpRYiTSfDD^e-m(L1Iz~
zQl;A89ftV%jEU+HqH29aE14nIZaQ*FZkvkYXva*4nC)IauUxfjv?w+xdavNuzFwZ!
z7_Q-dXd0n^a4X`y?<{rzKTas<d!P7Pxp_W<-dLlO9uwrAfNvZa$tr@xIVYYtC)3}W
zU=e$Jkf3aJzGPVOZFs&%?2*g*Fqx2np;-7Pi!W4n@t#PT!Hff9zAOB7-^x0h+A)l*
z01mJp!T2zVgB5yL*Y4`Bl#Hn#>D|XBmd`ehmpQ4tNG)K3`@(?Tvja#H*tOR`?MsJ&
zfj$rY4R5V?JBLG7!Tp*));)c>AC#%2*RB42gvfGnl#7>WzC#1AZef-j3nHcFt){)Z
zs%g6S?#Znfr#hRgqpCubH3IOZV+}&jv6hgc`cEN(2lMs;Y;LNoa1`I(HkCsabbVN)
zm-CZ0RR_HL3k(6DqCs1;irWO-{;(~vyx`)sY+-60?AylV^k5F~;q7lGZTlGkt$efE
ztUxWI-v&<XZe;k1ZeB6NvZG75fY^@JJu-Meon$XpHZOX}C}A3u+`)m<(Tyhu++s&-
zcAMQcQy&76V#@n2(gd49>k`UuDnYNvy#6*7+=v^H@=u`S0JQRNr{VpTR?5~<3GWE+
zv|D|EjaM!1sd+%eARLJ(yXvEE9CDY5?GVrItiWL1Yp88FZ&ZWHoy*r_^MaJb{{;UN
zApXkc9=J^|fW;4Ph7o^w(W*j=KQ5K9)6>ZCDSJdPw_xHlsk|ZxZE*!(br%<L*ELzz
z&+6T0p9+}LW-wlv+mO;Hd&jU6JS(uOxhX9Pt6KzZrVel4l_|L`lEakql%D04H{_HT
zX7a1KJ)+{V&KZ+nU61@%HvcH#vhGT-?D9apG%f%1USX&KbgQ8YL#$d#g&PcI-1hha
zIs%)qA1Np8A(syB?R;#46ouhW)KhxFE+5O_03H?^iBhb~fFHpUB4jZCUFZKGEBt|0
ziz`!zmH*e#1FCPI*GJrbMq@G(QbuhCBvyoX0j^oS-N^K`BHO!Qp~sW`yr?`ocg&#r
zMkbMDg7^1IJP?0N<^mT=Y>gF?B$VvRQ_52NJBWsfHMgUn{md$OoM5ldV!T3VWHleq
zFAbTfJwzz2@BmA{{dO6TvK6usXV3WSInP(uXGX~}Qr4%LH>FN~`2%&_9Z_9N!tsKi
z3uDADu=s97ZEO$T_Ha~Pa+53)SJ5~dm1cHR{+_R0JS%VFY*91m`?N`O8s+}yaq`hW
zf0x@_1+Q${#2562w~te$JXD15hyE8p0_7HdxRpZ|WyVzD!2EBh-kec|=Uo8js4?Q(
z0}IOpXIJvNSOHDkW8qo<8QEAIAw4=TX%Acocic9Mmx=I^B;tjCJ{W(kk87o_8jH>^
zN3j{<3-mLA|4}O>uCA{=47zKmyzovoKW50Ro%SkHS9(GWLty(!9{)=RfQ`gjp3ifN
zg3e%0qy=H8iB(?lf(qFJ8KahjE*0|HmN^8wqFR#|vyYB;R+dIgXGI?jHcY6m?A7=B
zE(;FnK19fiRmpMb|BXkXH5f~^Qh^W+wxf~qe;7t6yC72x9fP1qJ4+pmps(h)SVDx`
z3d~3-;F|yeJp>e!y~#=-(JU9rVF23F@_&hZzY!<X>JNf!a3LN69tomLa4Fb3x!`u_
zZu}12cw!->fOsY*4XFx<S-#h@CdP?7Yh-nwSQ<uOl3?($KaC>xjXU^yMOO6W=(n`D
z=VgX4`L0opLR@kv10rYpuMt%!%=MaB>82?lAOP7Io%A|Oep<ts{Gp=Z&-08TLzqL@
zrqlbpLszz-Z8x_P6V=phcy^ac>@_@I%=8=FC$+bTulLvLcfs%KZ%*Z20GpSAh{^Sp
zr*v>t;oBx<MxhyIKyZpa_;5@>K#IQD(y=l-b%wSG7ZIKUrJJgS2o@3^R7?c66UF3M
z5D-E5h-G4mMoFF-bG$u<y3m}{IV*kLrFTUbj3<JCAA8~{=C2L~Zh%3}lYvBXV);u)
z(UtAe^6<n`L{c>aPaY)>mo?&Xx+a{in>WK`AE_T?YXa&wJSg<n5TyxXg8$@v<IC7N
z4n-b}f^#`%8T(d_((|Kv*1z9a`$rr&U?k-79@AS}`i~Fib}lZK#DkRtlU3p|_SkY>
zlqg>sF`&-GC{G$OzmXahG~-^4_`eyYEp{WuW2%Dft3%=X8i-riN9o6);z3|fj90NS
zW@CT)gC|vF?vZ1ofT2)45kyQ|f;L+9h%{nlNU?v93f>e-Gako(C55FEsRc5MezfT%
zBzOsZ-jFd*lJ@DJ$}fzfx|_%aFmyvsk0lz4C7<mXxB2J64JX%gLq;5^>t<lh!{P`1
zwF$_LbMAFB>4L>$dpPS0;RT(je2}us^kcNl>I}VVU5lfcQq-#9ZkI=}X;7>s)$a5&
zVhG_(mCeQY;}k!v#xh&cSp&#Tpqyv=;hbcH&f)uO(#3j2TDfgv!{R4lj`zzHYH^p!
zwo;Y!Xfm!>J=Xl1zrVjzgY>o?hCZ-HRGXc^L%c4#{3#<~iuug`5%sOACg_?8%;xu3
zVSznelxEka+0NE%y@}rzf)9fxIHDTKyVl#p=Ml-9e!~$~D%F7mas)uJQ7gH<apY)X
zt~i|Rd+BjoF00hx+U}LoV@TG22|7FRpjdx|Xemv;gINaWUk=26A>AK3!Gwv*qN|j9
zwaHKl6qdUc;LKT;t@5km%(w$>&8dy9YPKA$T1o_HuZP&W1m)Q7*Z)4HD9H22Jx@6b
zK)lS|?`2$QbJfi`SO9l>4FX1)VSJGx(3br{Hfl72osLm>GNTGKj@naI{LK1KltcmK
zuoyO77q<yl$0r?JBdfJ8iIl=`7=il}J;RMz_POS{_tm^cPkl<`f;DKG2G#z2w3_B|
zF1-o}$<$e3qtJrgsiS`}B+(By&x;Y!W;!fQmzVwa0|H3Ig#jbW8J-dF6XfvlF3I2n
z>+G;<gsU$r=M07>WX$l$N#FR>CS@nRO15bPB=q>wo3Fd5O@`t-<U#TGlk(rJdWygW
zlJf6*nu$#Ej4?S#2TLO_ayaZKV_26=_j=z3DquHfEY+k&uX0V{Pae@f4y+2bw?6mU
zqQ(A9Ou;J21>V(AnYKffA@^WWJF2I1k%XNxwg?fa;dXF35Wowp7srmnkL41z@Dh&P
zjfoN1AA|DLj$jV0ncSr!;YMO680yY(yygvj4WARTQrJTXG9odd_`TkBp4v}!{Tmn|
zL<zT(9LBzr2&CA%99mQd2)D`y@ar;NuMCY4jI7t@0xPfIg5G=1cRce$jsFs=VewJR
zVZsIb{YyyT(|}19`OBJE`mGAqc$@=YCpPK*PrS70*5QxT2}9zDoiB5Po;b?#jW`sp
zSpqlrrL&<A?>43JL|bakHRv4I;rKYM-42U}iyG_Ll<kLGR2Q#q+1j0bV<iSC_pQB5
zxt)5?YM@Z|G9N15;;ho%ngLv)pQz#JlrXS16vc@OpU0L*kI+1_(v9aWTR<-5wV8i}
z*{8<BD_`@5b4Vlc5A7XRX3;mLRcn<AjJ!A+BUT!)ItzIQP7ou7yVJHm6oSN$mn=_h
zl@+z6D%PoiiCMHF1~--p!KjfYM#mULGpd1JU;yI{$^k@k0_nsd-kupnCoHw|a=4Es
zKl=<S;)3lyU!B~6_=#I0oGGK^>MKN^P!3kR^j&a!b|9QJRw(0@Eu}i;KE84tui8nM
z<4M_PtXg5znh>DG!N_K*f`*P^c5)LEw!cM8bkujKrOji!M`gmqouVy+g+?ouHWTLX
z3e=(BRzP?D{+=69;4RFy#5AmzfaS8eh;kgPKa|!(p2g=!^fyIFAWlp&W@JnzO{T$F
zt-@AT#h&1rB}rO5J(UiU0~I_rTndqd0~<WHRSJ=TZ)fA;gL3Caabu}fUS}*s@=Ek7
zQzDV{llUtx@xc9i1?maw3EMH6**CQSq~APOhB`WQ;je~b2vJF!@NP^GVC+S|;SYIb
zx}+4*5`O4DXWj&?p{MD8eW;wm*-)w&t%KS#mplI9O%U6l>CJm(DL`?p_D&23$w?`6
zSvrnqUYP!6S7CryB$S9@DadxYN=h)f`30bfQQ!aZ==+I@WhA2Ne*I%-;HrS?ZPb2-
zw^GY<D5*h6p$Is`1W2JY-~8b^3on{YI#MaF2%jn+%KJL-wnur^IFc14^*+-s$<-cL
z`Kx$ISl<P<TA<7tucNZ8EPqS&+X?pbV102(^%agICd_rZw{Zr#^S*9fAddglGHw5N
zL$TAYS>4aWu|S4$n2TFTKU1I~V`5Gtq=T}V1|sF9Ydz9-5>M+;?~n{1!2#!Ad31_$
zah{xTdLWBga)1frqa|gv2T8BN#n~L1Pu-mj3N(%uy#pKX;OxdBNrEnO$8M(BaEDPi
zY%Iag>5W}KQM`+Izkh02?%WuBs@IERgMyn|v2KN#MHlZEQVsYSZl^%QggcHse@q>5
zsz4b0Ft>jKYh(O37>&?3F_t{ht}v<cMrxJ5aKhZOz{zF`U`fD_uL6%0pdK=elvzIO
zs<co)DtB9+lOR19Iv}g@*Sc$;`UjztcG!zS>zyuiPb_Bgw(0?-vqy<xaNK{!%ibmp
zN>}1(@S&k(-&Np^)d2~AXG!MJzZF9lQ6SfnBU$qiX(D38{|F0vd|0UgFYp-uAdSH^
z$;1S0C&IvAuagVE4MCEM-o(v#-WBY|J+wD|{&$gM8!71~A?xEt<a}1t*^8WzSXr_W
zGeP?Mdxk9D7XBTK@dRhX|BycIV`^W?WXEjwCt7I<G^J-YV5jA}%(1R&Keg=bi|}Qz
z>Rwh1i+8*vFHWCFhd|Oeq@c`|J2^S|3+vb=j3s_SV2~Lv9{1xJrl+FtA3M{GIEQDl
zO;s381(dA1R8j~$RTcM@r4Cmq3c9MoM`$p}bqQ8`z7&N$?;2gLFUyvO@jJSjhElKl
zZyxjiu5BoQo|@?(N}{-5nuZ!EC*@3<vsiacZeHT`@)YkiIuC+U?{B}Vrk=&SmZ}Zd
z8C)RhGX2U7`s=gUM_cZ^E7q(REmB2EHtmTY?`yTK*v>xi7l}Q@Wk#ALx-Vm?Nq`JW
z)-M|YY{9goOB6lBh_Cjk0HT}=x-Tp!LW2Ka=M?$y#gccI+Y*n%r<M0r`)TA&AYl!R
zMm&noGt0LgkpCYBrtTqvasbNb?h&GdWmVXwQuvbRs$p9mxm4_HKDep0C*?^3!3Rrh
zBS{eKqr4I92ka~|`EcOzL4FM>L%v}dbM2}<)LX-29&2UpyqwO@kZl^pLU$*nUX4{M
z7nv=hJ-XJ$v2|*ay1pKwi=64@xsbO1<k+Rqu?<ar2v*`RWej0&6$8q2?3unZ8C!4t
z9!&BYG|W5=&(yURUBIKeCA@-zsP#Gi!@#Dx_PgcY-w*B<Q`L%WblD<$e3%`aG>)=A
zgR;VZ*_w}7Zd$c?xU$-@hv?43z7rynW-f#g0wHjX`5Uj&$D#>G_Zo&;;X=k8ogUE{
zg#P@Z&#lMdoVDA3;ec1&rmcTxV{=1U=f*lO5jn*GuTl~dn5`^|Rx@{I^jMAOrB{38
zg_wvfAmI-gc1%V^$iXeoNUHMxbpLIXjEy#5!Se(YwGb4hPq3E0mB;Dd=1!<%vdNEx
z8$m23PFc*U(G0E6kzM&P7It)z$Jw^l6?i*oe*jXqv>c+wPGP<?M*C0;4=-%IZ5fm%
zs-HG;15K(8*wubN9a2(Hr#v6jTw2=r+QJ}5?_z6tW6VUSBOC(~{-J9)bBiah^bGUZ
zVr4wD1Y7rYf3WDAP1Tky`w`=Fpv!aO*Ev#yZIqRIcJv3U<<mXsDjpnosmr=2v>vo}
zqXMSc*E4uhli$n=EY=2{*jK3IRw+0pI55;PDq+n8%ff88vI+!?(vQyOF7I6`(~e`v
zZQ$@(;J%@J42!x1+RpUne@TL#5bHzVzz!B0hu=Z7>^<#lChlpFgv`(WH_qaMp8^q6
zQ|C;F%TLnGvItC|r}AN7%4vG|>JD5}{!NbDaQ!wEk5pj<$y7-RiB@%R7Asf&Q&9~B
z#a(~J9LWbqDbcme{AW$lt3(M(N5zx3oIobrLNhy>v(*98%fB)`0p4|1sIjn=Xsf&1
z+o0`<+we4O2mwj|8xgbORP!G%q)r668xP)6l;<WXh~vCLwUT096m?_ZO~57dKIAE?
zf|735Ed(~=eqMHvRtpZeSmOT&tUy!0daF*_=U3#SadmysCRO%B7Z;7%RhmTCjL1qW
z7D&B%RVB@f>bJ(qDIRij^Bb~0Hm5M*?t;6EhsTYku3$2WoKiHAWD-2zZCh;N`EJ+l
zM>KCK6Q5KL`Qc__m-16f=}jk_0gEVqSZ?Q>ahd0-M{Z?@TpR}}>4Jb}jL<`I;)?Nm
zD=w51)zZ<CjDD$8M)qY@h*TqxZn8~$N*+k>D)emVE8|fL%2<*tFIL3s+=BDqx$u%J
zEFvLRC*CCy<?<rU7#CLaB{4x!ueN_}TznSJ0q)2ZnNv*1ej(uPWvC-res|7W<O%~0
zwjQ1l<^|GnA*jrHhz(^7qWA36LH)QaBYJDFnJkH&RZRjZOo?*3C=tb5oV6I=OzI5=
zC53onMmqwA*Lk7B?wB7DIU8#oSCqR8KTX(QXWB7%3;A!iY7_y^0e?P~Zz1Ue0kTH`
zIgPNlh%>J5p-&zT2Zt4&J&88EgycofvWJ>m%f{N|%C}Iy@k#Hx1L(fg8EAHPs68)x
z!PE-&&@*@C*!P~H_1V`<5rP;(xZCXtzR3mh4su1<iO=<HMIv3vD}^p%(haJ0!Qb<5
zYFpgXGqNp6OR+Wcfqz{`j@u&v9L;f=!fXH%O5T)j#7l)NI6ht_WW@^$ge;2K<X9_d
z9e1BtYjFf5It&x0j-p~H)?dg|D|n1m8AK14>O8!&V^b!WiAGHN<cTf_n1LL1F3xnz
zgvUH-PMm=?HoC;7-g_dO-Dcn2PzJLE!b~;8pUT24d2(YA4S!0<WdmOlO;Nt}8qDgw
z=@s1}V@K3mn50khn-&|G0|ABoKF_j-(h9+-jJda-KWNXo42jj~WydfL1W+mgRE`Ev
z83DAf01C$WbbS9wgjktmCPp9z!+`b*GCT|j<9%Mgh~c0IRa+Y`nMJG~9am#yaV-vI
zCjExG9F2pc+JC{83Qlijs+>}-cI~`<*}A-Hv}=EmgSZK*31KYguw4PS#R0!|J5!{z
z16E1EDha4dFC>T}3RI9Ku=@F4S;l88AH-KbFZ&w~H{p0?&B4oNJGA3wr^V0rth~8!
zK{`+#S>zM&IEpm)Ta`Vs<uhrKf%eGuHtImL$^{^0_kY+U8ZS@J0qxx=kZ;-x?2#R7
zNXfm{dEk3y;E9E*!r@`LexMGTRV6f2XgZq1E0a^(hB$z;vCD|x)qyp2P+~;UZ}xy@
zif6)A!tHK_sX^8w#a(ahZA8IcFS>kiC8~Td`9G2+KY;z@j>&@2UAa5^mMFY%o1d*B
zbi*y#w|}|!dY786&Q2Ny=sM3nV@=}KAtlL(fLd6Sc*!g!qE*^!62G$i@KiKOIQ2E9
zc#Kh>A7fHLU@Sj8v+ZH|VTtyJo!O|R#rUEE!f04W!#Wz)x48WWBxU5{NH$!cGaA-a
zGpxVKHua8xfjCfjmiJYJ&R5V_F?lSMJ{EaRYky?@y<q)Psd9kiMslO1-;$*7Lf=B&
z85!COlK;)*$5q2s!&O58y&(a8Z#|2~vO++Utloob1>T3+vr*(lN7*Ocp?RViHr1G}
z4sSsS0rwHFVAA|}Tsb_RI_<>D%v@Xyn^L26Ru)kqsKzDjB%Wf#EB&&9-?n;OI;(Ob
zt$!pr!jE$ZczB6}Knf0S+Zov#g#<M9fpDJ?iRbWoT&y(CQV`oSBWon1MyyX=aw42Q
zS@r|5Oyh;uOXBf?On%JHM#6#5)^k!G_z=-&Ju9&CstTM_uSL}a=ujECKrSgI1J@ln
zpMU<jue*cseSt~UpTYF`2w|8%i}DlYpMQ178B^!y^5Mc3!m7xqNL}FVr-&u$V8Xh9
zVyB4Epf@6AG{@2Vrtb(B&hCES3pAR1?Fb()4;q`f_!BAR@?J%j^5@S^M>FUhJEnAO
z>e9QRW{x!@t5JtBn5%c7N?8Urlnj(498&gmA@3B+)smFDH7M2w?bH9y-nTfXjYBMZ
z|0}a3bq?Gmn1{iSExQT>PF!Dtb8YABrjlIO00ZbmNL)$S=H%P|{Z5Z0kYq`S_l%g-
zIbN&>o}T{oV|seJwYTyb0UZH<e^?z4F}(q#vd_6?ecl94p(uhPJzRlMojMN}t>WSa
z+}ALHwmsU#PN1MJQuuUAkM^)UT8eOeVyTM8j`71Q&O#43&SV31CYIf>go%Z3`oPnJ
z(aB`nbo#~I)c%xfq1`E(H)zufAJ=lgugou|I?SnaG{WOCWvJJ7Q!D*{X!noFzPQSc
z6(@%JFw=SDDa^6XvxI<%n^4qPHx<^y#Pj#uYmYs>Hy~NJdsX|($B&+7R$%zp=R7i%
z|A4c>RykC07T|ZbS=)6VWS>KP3K%@;(<@H))ALX<e-fkc6R>_ky%9~%<bD-rjkiOe
zF|8hbs8p!NKQ_WI^syU%t^ar4R`K+vImBZaI(-NIhTQX@_VIOy=I4I3>$b?3OP<%<
z8DIPweF84qt^=DG7)x>ffak6q==b~?hjokp3b4jYmCcXE@4+>A*!!+!=s%Oc?IFzb
zB`%+neFuB`Z{}RiOtcNEJR(-N;2L7*+il;oN3d$4Au&kd6<gYWb+Gw8o#hL#Ow8G<
z&qKlr^4MPz$h(wk{g6!gR+Rb}Q}w!k9LbZmdyW7(M%}kn#6x#b35-1hW3hji`72|)
zcA2mAgWKynb9D>X<9xkf17-=ZLj+j3GcJK~@W2RJd`VOsJSsp(z=2|FVW_3idR+7{
zm#Em?G6(?hZ(V_Z2#|w-{qtWggZH-mDfyjpe%p09S(C(B+dI#-JOoM=tw6m59$F#F
zeU4orBd);URzhVa8H|WF+y5{vsM;95JZ9@+C$8^oE4Zcpe*N`GSH_Pw?Q1)r{MTdq
z_M!2CUC5&s+qH>(_5RCe|EqApUV(pVHE-%|`q};(_g8m+O&+lbch~4Zk)0kKzD0um
z3zb9jEPP_OTL5h(A4RGWX<&Q?hV&GyN%L|@=4D4I#Vc8szh;KwFI(?NDPWl3(?+RQ
zKI2jF9Amae>#=!!%xrHQORml;jnbB?_pZm1KR?{u|3875a=9^QM>;yv(K$Fr$DEAa
zJS6GuNN-1fdOI7GK4;Z0y&Zm7VtYH{GDUhj(%X^V&dT17qV=V>bC}==<lNh51X!1$
zVv-`JY0oE+jxQIE@8irtnY+Nwwfirh7dyi;&a_+9$lzak-QhfXm3byT<n&p-mGLK_
z6SP!3ut0hP!h@SEeIw}`9e{5nLxqwl(l?U6k@Ss!IQ5M@(L-*0BN;;69YT0$V=39a
z>xl7)VCYSrxZRap4RHm5bv%P@)m0knO0tDd_@Vnxt$N#Y?moi1a|;KREA+ElGAAUv
ze}NY$l2ZpqMCcqk*NFD+{Z){oAkB<;L7mmhS-GrsJ#-77`mHuFCMM}YMZj6HVo$}P
zA$}r%NU`8Hlr3@!|Dq4XAd6}r=P2GZ_>K;-|3rs$G5z3{K(hZh!OkC*`f_C(B%sMm
zb-|+3N>$=$kMSg5!J?jmN4mzGGcL*jNO0^I9P!zt;5L1odra@|ZohoyqZX0G&3<v?
zdcUu<%C|hqFmE{XV+%m>7uY`#pzw!@%dl{Nw~yH#d@o|ND3F(WPwG8ROKnM1>=zY$
zmfDi2*qJ@#vDB6f;Ovh7Je1m!{N`!X8`eT{qx-N>Avd~3+vQw`-TZXYgq<zfNR#Kt
z!IQVUChR~i*FZCui{vWj`>x>=vQn#VUx9qL-Auy9g|gQlulw?xbg!0%SQ_H>4Kc%i
zp|r$8SYoB0^%`2opg_iADfbJ^wCD1&_S&vG>-mDeU4tj)Y{pd(UdDuby2r8RZBBY9
z`Uf73|3F!nAwBX`V9!(QZkHNxo?we>cDp+<;4XCkd!}kve>O8Y^#4)VH$YcD=l%Ps
zSx~gTp3QyBm{SwJF7tzZ$)E0*<Z{1%N;dmuA^yL?%}AC1wPq)2Y!<phY*0%}44?E!
ze@yyg(jSw1WTEZWxz_ojuCFiOfYf1airyP|JVbO7l<vI!sXitez1yQwN~O`cNk7dd
zFpoe``$*IMUjQ$X9-0!1z5Yy?HrF-eyol&jQhyn}W4xkVej*y{HcYj=ZRKr$JAyis
zf%mnE3E5^$#h}vuwSqhpvP+~=lYyUfrAN(z9~S2f#0)mr7);=NJD{|u^!wE1q`&kH
zeS}J=Pr(XXttQG^2G*#Kp?AAex=gd6`>%ZqfC`EBUEMH9*HWkSd(5$lr*>ZdJaOM{
z8EhjGuHK>dUv5wqhBmaUv3Y)fbTm+`;iN<D+33hpdQ<8XndHcEn2r?Fq=xdSd|WN(
z3aG6!qCDS;(YV)_7#GRnEsN=AY-qk1g6G2TQo*H~`QdJk9Yg{cIEZipP?;>XwR3dj
zsq6CouKxMMO#?Ms|GH_k(3ksG6Wz8^9sPOJZr=WRkM6(TKeRsCHuk%J_Hb8k-+!*(
zwc3rZ==K^td}yJMH|^Fvx}pDT*BcKv|EfRG@3?8(_ge2;?IwD-MW63l|7x`#Jm1`G
zeWv%|+SQl$D0|;(3F`nYtE%@%=SuYDDCcj{?Y~-gjoVKx%2~>7iV}yDd4LV7w{I!O
z?_4Bhsolxu(Dm&d#lro6L%rQ-<@^ard!T#e)EcQoQ>;Xp1@&u6r4Qk^tYQ|liK=Gd
z0dVkUdNuDpX#bM=u6@@s-Uouwy1igV#v6FfyIG5~rOL77EI*trRU6mgGV8m;IO|;E
zl(U<U2ArH#s%I4rIU5-xob9@>7~aCb+*S!(Gu&a2k^Tf5951tfV3>2}U(?|#F;UOJ
z&)U)$_mQ3k;BP{uaR~w3b3(7UZ_)3CD3hH>@bdC9uM7cUEs6(Ns+b&rA$6V$w_0%X
zg&$xG*UUPlPFOSCd8Hb=H!f_@DjM!u$`iKi?K@jLt%$rBpZO5gkSNIGAy++h?iSCR
zFU<e2jf}e?g6LX*o1J9!>kFOg1mjsTSpidf9}(WNNDtu-^=NolU5gGKIcsv(<gCeA
zOWUkHi)0d^bfaqvrA<35-a|f!;VjF0blr_Q6AS4-^Qc3Hz!i2)x6?(y`;qt%1Q#pe
zLFs4FZ^+$p0=-ZGy^u>%xY%i=o0RYV$3@vmO7ZUFTKPMFlKJ8?-`z3xaoPRB#}F51
z5d{a&`J_jjM~R{WXsFbVE0qTKG_=2bx@z6s$CA=IEmh90d6I$V&Hd+(^{=gF3OIo3
zcY94^TZsXYd07HYe4yG{I1#)vLh+|+^UjG&NRmpyw9lRGS;r6-h8Bw%dFU?Dys5YE
zGwdUL*taHs+DyV_aO$s$^%9pGS_yU~0?xYh%lTcDt&kEL2ivoHtyDbbiAb5idcX)*
z`g60#j-p!nIFqPJR<U$;S`IM1b$wE)6+@Zsdz}AF_YKCL%O}UzwQUs$ScU0v$A#EN
zgst)7v{<cKx~>KDZ@44$(w|8bpy8@jy}k;nE&gbK!2W>0n2dN8*%FWug}#AD#NWrr
zLkX|L28m9=?B~%}bb+P?RT+IN|A-@~cwKBXBXR?`NYJXQ1=ib-S~#<AUb6}%Q1Arm
zQlx-su?sk7xPBTO4`M4cY_@FC4I1Je6cXqf1kzVAV(qP%l3;n%tbC9~Ua~o4>c|>W
z!`i2RU&HDNKoJ`aDH_xv=yq8EJ=WfqTcwlgDM!<C8L0=wVeRF-(G>JUq)eo&>gmK}
zMH_kqtLRy=h7Q*}Cngx9p%{tVv&t<M5xer^j)&pU&K1xtRAI45-ln|3Dk%YK6)g!!
z{aoh{xpPNXEQ!mbCj~oebGvIE{NB?kgP`7jyN{To1WPu7S7*yxBbX3p6n=d2y(ITX
zq66#l0Aj1HTD{6yWGPD0_!(rl5`1gN&=;(PZ3LC-mGTu=djQn?7-6gN%0%yZh)}(A
zL33vYd`(yNLHwNaC$$b@M1x8m^^}2P(IrQjel+xP(K4{2Qk^cE$O<Q=7jlNTJj1Ym
ziq6AYjA}#HokF-IrAxa2aMY-!+21U}-@<;mH2RwyV!Dk!zW~1kD2s++>5WKlM0z6}
zdn3L9AB6>(u3F`KEj-91i{0_V$<37eh{7+>T(4HkO`iPHmDC7-oSSW0#7%zVy5?3l
z;bqv)Zdhc{o!9yHJJd_+Jckd?9Z_0;Wv{67IAR*?(8b9|T*@LpWJGAJ_;y~6a;@4p
z;YosQ%MlziM+0267br4X0(k{ee+i+#+{6M;Pb^GgvkROpT#jgXj2rIRb-7Y$@ibh)
zW|!rRaja7p<VgymJzo|WTmEqI>aZ6_WBW@^<yO%eokJ*;&`lQzos;~^6H>l^rU)w@
zVdNY>mn!ZM6}<{whfu;hJJP_7_;c(0>GwtF)&-gNRnD#R$Bfyzb@$kM^zH}{z+cuw
zAR2zO?adK^7uuXlRJevVmH$AGY9?&(dA-1j<Tvm9c5BsVIx6n|%p_rV<+^g#Vn@}3
z?DpSb>~=1tkloyHR>h`uN3<S)F-Y$-9*`@8;GgWR2%gi)Rzb*vmS>9;ZM16*Q%;PC
z72xWe(y|MK9Xf_XCxsheN3ehR?}-ViB~gyku83LnC@h07(D0T~vTrhWw*MjpF=c}I
zlo~-~lq^PY%KM&*Qi*&eW(P_<?R3POJ(3?q2V07r9@mRC?x_IDd;1rEnwX@AhSYDJ
z9XnTgM6;AGR-xp;JY${}iyvZ>Rfqqp5MTbXRjeFusXoMNN1Fev-?eYr?*+r(<vA1q
z&<JIb-bcDY*B4IfJQj~OXFP+3M(k|%+ZSXKi}8p(UO?CMX;7_7pYqgFbS+PxAjoYY
zP8r<&sXq~;uj<dNllcFCPKXK5nMWt(TIKZk?C5!jt#>{B9rKmmVXgO$DQmd&JS5t?
z5&lV4<vaWi5(wT+O!&plCOCL~zPfeXqh4Rtm@@>j%wE$;<?UuY*YYSc3m~g_8VI}Y
z1s$^B-}bCMJ5Y#TmHlJw)RGncldc{I%wMa~5R-mI!&V@r>~Kwg!65~<{T*iP`Z20*
z<Q8_g<`K`}b-}3u=1kd5A<+oGKv^6Ppe^@owr0mDiUGQo(wj!r$Jq<dE5}8);U{^%
zd7jS;3ltPhah%y%0jz@2t4NtoGyiA4t5@>jSgjQNC1%uLk)dzAyS@_$4+}=0KurK}
zI8YRO#!H$*+KUQ*oC7T0z7(dulg*wRF9JHk)({GGnzkQo>8L}fH>?u)2%*Rj;=y&E
zXiSZQ6Ak}mtwys38Ie(k4vN*R4G0c<Rk2H?4s^X|$1Qp3#xAYz7Ok&jZ<^W5bORIg
zlA{;#U|^5|ltqSPP<{>aEi=vh$fFLN)%f*NAlj4j67|l1RH6;6VV0E*&Vh?o%KcIm
z(~2z_UAd}UpKZ0B(>9FAz|bdSw_wKX>gRg9^-;LPI3WfyNZGeMVqE=J`tc(-^TuEC
zyDa%F0Kb!ZNpT>EyjKQ*pg$1XV#+cFk&D+5-E4-<ZZOj}fScoTt<>N@11Q5V5km{F
zN+3fyo|$uhI9|$UJd5L4y-~fc2eaD*k%o9II+XFmJsWnSfi8~%u$!w9rbU?k_&blT
zzqF~pF;cWVRV)Q#<aA1r3y<Ze9cD=o&(yXd9{)7=?e!(Li%oq}$H-)0*-C-p7X=B8
z)Gk`ea?v6eEppK!`Mve^Ni0B$$7jdE##gEEbsfroI6^{NeBBK!u_Wvy>^8f^!Y^`(
zweGaTED5^AD!azeDG>eoKE=~x-_S<|w$Q>aJmif_gzOR_uNL8DXA$nlg?~j(FC08h
zQhl0m4u^_++N4A0K07-L_02voH9s2_Xa38LlPj+3n1}l?&I&65CA_mwP_|q$r%W`T
zj2h8@AJG1L7R7XABLz6}5On|oE&>;*_aDStIXT|AB&wVOL6OqI1CU(lh-1dtS-Evw
z4*;rMtd&YzDl1@9wd%275Ok;!Zr06f+_iKgw!pbV9Dj7IP%c$4M#Yw|m3-M7Um)xy
zno3J&DVL+fL4q@g7oW^7${+!sFtv{$V^z(6&dS<ksSly`Jkg}e+m}2#qej`j5vN$n
zm6NmP8BYXaI9)v=hSC+S4~92@1749>^DbORO@KK47u(@GpwxVtZY!j0R+W2XAvY_H
z<U<*pC+e!$G*Jc`ACcZ;Pcgd&u}B7GoOct1jgRZKMynF0{QY-Qeu|qTblb@nUF`OM
z5OsEwV)wG+g8%%G7LN9t_CxEgUH^z0x6M|3O9z`uj(F9&ySu$ZH*J*3qy9weLa{NY
zA8h8kCO*e5ufN>i+_ur}^}H^dFZhb;?WQ+4hU$&R?OpSx{T@BsdTY7)-eTt=18=fx
zH}|N0`#_O!ch_n>6l_R*sNchvUA54Er`zVu^-Zf8oLG(~Yzteu)$2;DHA_^Hig>Tu
z%Cz*c+X0%2_;>9h@DHwK+!Q*h#hE6Ed`#&>9#hsWEcA>G4YCC29rtfg<x+1pZyu-^
ze%GAZkMkGd1u8(S<MhD#1C?G!6uFtvn9A~(nR{qqk^$P!ovC4S5nX2$#Gu}PsMfDI
zb4}^8Of4X{xK4u6`_(|BT)L)f=&g{nSr_%<0x23|@{kK0p2xi&GRYG$u!=^+G%2&8
zz^C<02(;SSxQ3KTFC2-$%17RhE7zy<d`u5}{dW#43}utfyE=$3BOS+Xeki+yl6bZ=
z>a4sx>utz^&t_i0NVx}YXQ!oq+>emY=idyI;FsVR1pIbh;#>sWOhwNXmZ<20ewLxg
z9-vnWW8GXiO;Ms>0Q6tJIW(MmGjVO4<CnE?E<M-~ob#K{>F_=}4fiEZBDM+s)d<Ub
zB63dt(foq2*`U(sq}b@B*X4>jh8$Rg%r7|FXMW-B3&0<}aiTBg#9+~Xmmz*4tZj<*
ztRubiZ_=eyDjK_FQe$VgQ(>hHUlKLH;d~+&b;J?KJ2^*PtU;AXh@j2WJg`|%G>Z)U
z+-Dk`BEWMupx$%yz2j%a8X{sCx{+r<=;{W=+*sG>AL&BJBs)0wf-u!<>}9uLr3;(I
zYBr!z7x_FPYSzOMnO(GhQ$zat2$DDYVQU^={y|XM_@QOlBhvz~M6}6vy-PXe+Zbg^
zT=kufd9!<8n{>GNAy?eif+BlOGEyY2*QE60OJ6lBr&nhI)}GO2)z$I7sqnXiAlke4
zSILy$Z^)B>zBKV12m4aX$oS)u4r%;T5gThwp`<OS@g#<H#W>V|Qoc4s#j0oZ(koDx
zIrI-a8vo&O2BoEvlC&!-*g`;BEAb>uS}T67Rl0>L`Vgi_nlEC>XxrlS@=kggQUmUu
zir7s6>19YSL&}(xu?0?taHSt)(`skpmT{vOAOR-P8RV|nzd<Sh70Z@R;1uPn)JfoH
z^eYxc;8o*z5;CEGV!=Z9SWMB0Sh+QCSv`gi^?$X{!-p2SuYYQxBynS52&XQky>Ac9
z<sz?UCyZ^}e*E&OP5(ocO8Y?;t=Otn-Qs|kUwiOgzCu=laAZqXg4l&KFSDbzCj!e?
zf>?`_<=Uy_<SsZF&|cMI?WD<5ckxTGEeQez&cH*!=fA^$z~`Y-s*xT0EW)0HaN-bA
z(j(R$(E2izUIoq(^AcFy(`Kz%3%9zh%5{xtGRd^PYme7w^5V>Jqj*f$5j?4p9=33y
zZ{QK}Z;UUANrx~mCY^_)Fe$Jx76_Ap%~;7*ak=^xUC6VVTN&lIS*&ZP!XirmZ%<g<
zXeX42>guY0ZsgX}g#J4zXRb@XtyMg9aw4xvpJQbLdL&9s!*rHwwbSY;Pt%$0DgAzS
zT2Pfy&JOVx`JLZYa?bekx--*P4Ws!&aURhjDq3LEk@tB`1P-bBrRLwc=I3-m!v)4?
z5ICN2uvJ2t+K2!&$L7{H1UNFC$k_%zt=5}mo)pD@V_(s*%JIBH0tHxDIkB6b*aUe}
zGUCM>*8)ZPJ%YftH|mwsvc0__x})ik5&}K{os>h*rJu)yO)oBkt;W#>deuHTl+XT6
zI#-?hw2PPdC)Ja)V?Xn`-qYJ)olS>yPB6-pQJgLc@X*ItPNq8A(|m)#nS=UgmQLY1
zRLq}$#K^OYwqbW~^>k}7#^a_U1mhvL)iZCY#yEHQxP0TS9qa5|d-j5TTZpjp^h7}a
zWoKP$^GGGyz#7t9nxQ{?Ux?2FK5fyds0TpfT;ynepcm$1)h_^}6S%{E<Fkq!ABwFK
zM9;^U-j8E9N5Jh4N%v>rtD~E+Sy1%>E8yUNEph!dM{QbH*Pe)BA;@8$T0;sk$aVip
ze^|Pm*^jbrwa~9%;rbI*MS36ohZy?ikJX5l6)i~3^i}N~p)(DTK@2j~$+3tjd7f^b
zGk*6JbFAX2<aN4vJtjt9H%6p~tY@W5&QV`6Op8_Wo+A3zgqUns`=ZF|Lu;O^D(rQC
z=Vz6qbxoZjj7-WSmDmp<tg8Am{ciZW;lC&Ji?P;6Pw;7JU>)ko8BRt7+=?j%&2<hP
z=(56yU^tNH99B$lc;>oUEuJyh1o7v;!|>-^io&0ab8mo|QT?RYL0+O+#3&e~pCpl-
zyNw~9vsp$;ucyXlW$`|rTNiJ6Fmr2vocufCQAeq**(mtooYdBP8QZ5;y)AWhI(0Qi
zOiOSE(?Obh$u=;pa;;jQmlO1G(tpR^;F@DK)qw$rbWVmDsk`F?<8kcfhcZ~hzB=9#
zM^>#~D|=KAuVlsVx{JY@B#ZJTl}gyIdAw3dO$Sz~RCb44+pAUeXOBy39|7Tiz)VU+
zOGK}T=%F$g`XL=Wc(or$OV3Y2dif<w!bp;IF+@eLicRZ|Xgy+(-e)`@R|dg9*;@zc
zeAr2;(gl(FI%O^hKfsobhy?cXj>ux*mdn+Xa%COhmdfSh)&V3{r2@Ed?B)o#uBS1G
zqb0Cc!z(S!u!zr-*s82bKcqT;P3~91_FRqz;hmA+X0`lho651rVa4Vvq++6C$Wm5J
z=6RraSbSv3vk4r5I6k@;r6Z6UM?es<ojC#!c1Ub&!X(_41g6jyKI>?|#bOhBC}AnF
zarS+s*oYuDg80wAvv=R(2ol7Jp6h{hc+;1*K6>Xp-W|)Qt5UU94fZvE5}8keq`SD*
z!Ho{40eXwAS{!rSi_)s4#;OS-i+-!-a}86kPAQEk(x%QHVggN6sg;{2bDio@I{rKS
zCuFasbU5PS8`g=ir%+MTy8hEzNjPXjGd*nWY|fTAq;oRNfW{4KAC5-(HdZs9RO|IE
zRZLlJxL^65%9UEL!7Mj_D%IMn8e>A?IfNSC`SSKF`h7vK8U9${4c+~h?XQ6?K}V=m
z^jBD3-yiT$Fa3e-`2$e8XE~oB)cp=z^sv@{C*`hQ=%jGP@2+kbq-&uO=_y!4x`9~9
z3yZw?i>+{l(+JZBHED2y8Zqes=kjWM+j?g;L!vM<q7Tq-7bv5D>6s*lx|ZUiNA&nh
zHBAW@7j{OB+{#<j3b<0MHIBI|FyS<>vq5`7K}-Kz&g4-!PhFqf|GcVr#L>`K%%0$t
z6A6k*N`2281`}+?%jyjw;wMzQQ7<>Q8NOa^;7N2JKf_%qm}6t6RIik;8az$C?WZ@y
zEDM-ctn&_<JnAWb1I0oY;A8a+eO$B*tf*9<3n#UrP5p(O;VsWFbUDa7@e7LZOg~9W
zt@Q(M3s%K4@V1p<M#A=WVh8UdhG=oTtnnA~Jj96S`kBp#Qf@sJIbd<@=13Bieim!f
ztUu2*2jMiZhe=|<)ne_WdA$y<J+mTl%{#8+*v%2HZ9ubsh8V9i8U1L_CrF!HzgGs=
zAn$Ar<otyUuxTm-Ejz}hMT(B58w?5>9uap<PJiv{6V1g+;NtJ7l*^;RV65s^`TRUq
zI!a|KA3RY%Ao}Prfk1OK<S;jocaZr5X-N#FM_u^uBvBV~vvd<ogU*kF0WD$?auI8@
z1Kyaxv0b5mxRRtXwp0pi6da*&gvVn#eN?1o79SJy(e8^X+I~qmFkeFS1@B|dykr9M
zB<k!>7ieqGQ9n*ZBJ-8w#u<rpFaKD%RB?x>C_Vc;d_p<62g~ntU*__@!GX1e8@r85
z2<%gYP73ehWGeq5)IKm*7(OTQAteY>1_IXG1oohRJ1Xw}4EjBx7-*EL6;F3_Z<%Y~
z9l4I!D`zq5vAa|6yVEc?sGdMCF(Ou0Or=_>)(AaucEF0snH1jPYqx@qU%p(`YxUD+
zIMexqxs7Nb>n$TLiZY?tlt?z~#j9p5xor9#XR_&BlCsGM(^+%&w94i-I?o?U+5vx|
zj$CAa)4T1Dc$2AL@tG2pR<Uy2<cNyPKkIkxoA!IbCjasrX*$vf=^?$3bfZU%Y!1@y
z5DS@^s6)DVViMHRU+nFpw=c*f7P5xKetfawXjYF))p|fNMlYfLkIVcHF02J(Jxl8_
zoji_i*1i@5Jcom4!_K3OTfk$i;_0iPyC|H0NRT)$vtg48oXm#Zou%1Hfocg{3EWKs
z_f;CVsy{E4#oerxD#wj=VC{TE6s%<@Ii6r0$8L^boiK}=f;~wRx5T)_IDZ)5LwY*z
zQ?f8;T^RNCVkJln-?m<=92Z${MX8shURskf$e2eLjs^l_oMfv>;#=aIKjj>s=M5=;
z(k$f|vUM|G=?5o2^Hmj>Bx!n50;B{8LfI+_h!Wls-hxoJLTcr<Bfb*dT%lWxh;Mv;
zo>Ln4i5QmB#i|<!&ZA<X%6_2V=dA*AQV66F2q^@!oVPf<d!f!Q>|}RZMv5iccS2b%
zmn;p9cTrxDn<kJ_+E==o)K5Odq-nu_6~peWQ069nttJ4`iFjdIZUv1Du%*yhkZEHQ
z_y?6Xw$vHjLjf*-`R8^=HkO6r=&b$oyL3{t&-5yTA^r8;J{QlO#Is6APlxDUBig(7
zS9{0A#zxOH0jxt4q}zqU<LdhOw3YPq64&a$p3`O0$5%y$GiPE@0jO}c4`&&FZBE|L
zE(`2zn4+8ZL+h?x|A-p5%@%6i-QC`yn>NbiQGcR!Ek)OyWTwn_Ey~yn%1lP_O}la1
zzK?TbTF14<F*heHld$yuq1A4DMOAT8x$<fvS_@aNE3MWnl5*Huc#8vG);%*Vee9-B
zQWf)Ey9oS)YnfU59mrXVd(4S{B`an@o2Y8`85>%TzCrJ}e}gKwdb4@+aC6)Kt~qh5
z`HS#^o31&q{y-)A@_h!r0lwzk`77*eqtC$tw(5;){fet|yLr^bs#-u7=(_&#zLjM3
ze%Wf2OV^d^c1YUbq7=P>9Rx)~3|cv4>8L{x?x+MlLThG|JP`w{XhckZ6AuUk^{i(?
zH2U8Doxj6c8*JZc*wor!MP!t&%8gcNcx!oNlq$zU8Ki~WcP59k;Q|^=>yBtWVvycv
zsv)inf`78NBoNk#FYgBkYlXvC2_H2Y$JJJ_I?q|L(z;@73-aIp_(AgDnH(<puaW^k
zV1zZQ2}VRqP{}6k09aXnweAN1D-Rd=tn?R@4d-j<30xT)&;@g-KZB>tRs$lD=%FPd
ztKt`31B@WKh&uj5YQ$%|_U^?F>Fe}z6xB=ut0$+GlW-a45T=x(LSNCaT7csasbXhX
zlji-XU1w<~?e3nCu<6sbgNf23X11@Wa5ji3!sycTFkh!v$5&;4wjvjI65S;F1?2Ls
z@US`_!f8f3n6;oL9k>2vWC)9@@~)*kVT-WTGDTA?1*?uT%7H4B9_Y@Ho*?2NSDak2
zdjk=N6;_?0VW(&i9A4Nd`K-xcot!l}YaFO)o&&>EWOlJiJUSX>yI9k7Z-}8se}mPA
z=c(m$8x9>4T(Td3+;Zx%QwTsI3T%!^*FuVky839WDqZTi_YkIx06Zv~HBlf<zJTrt
z@mG1001y!XQ3>(nWvpVlc!SKj^F?j!N+;FYmTHWzhwh<ejm`6;qk&=#C!GR?^pU0X
zrccB$$&nL2IZ{k>Ld>Je+3~3abZP<34z}qXgpI+3**Lv_Z`-}c>0_6UT+t@_#LVUv
zVEwvLqg>@0>o%+lGoTlAV{l}UJ~4>aB}X>VbCL@~Yotoxrxy4oAqcCcz5v**(_*<?
z6hw&|rZwp)`cZdCuw@PjvCN~6s&|e?5K=dephDccC&iplr{+J*hP^8nN`-OH-)b`n
z&>a(ZTrM?#llBXf=-&fM0O(Yy^{_ETmeT!cuE^=*<Ks#rw1=U8CslHKjgk4rx1BBG
zMWW0fVX?c$^SwUy;juGO)Lxd*Z_17_HS$>~^SYq%mNOk3vXt$FM&_r8$4ZBw@D1M5
zxqmk|m#1rORmEcyYTjkg?<jC=B+*oG+lM8O{wXAXC?mviY0i}E)zZl|kDU1(Dg9UF
ziUNHa0u(lntH;%1&}5!f%EdB|6wb6}$V!jAj3KObfx6jZ9u=t+s>(>AqrO0<ZdloA
zf&Naf!lLoQUMiaB=XRuwJZic|mX}g~v0$GWp3N7ll=0YCm2q}Tr-4p@Jsti%Tv0^%
zGKmI%@^gw`(-MDPi(ekBVAJ^JK^!g43WC33uVB%(O3Rw-J(dxBgL*pKnqalAAjtRf
z4!<ZS0^L5fKkD@FY8DFv<mn5nT;J?*Z8GZ6QR#g&!ir`u9ofrpGijPqih^S+0}Lj{
z8I34n>>S1gTNe}p{zjWTsW>RZcHDx)`Cle~*_&S!y;bp4pIC3lSR?8qU5(?B`R!Zj
zM}lThGS<kIGTBD>eK(3_EmJxz*rC$`7{9_;Hx)3I{s5}zj5tS3&BKm1G1=aqKU$2k
zx~5L)f>=0)3CgC+g67$brCu6cQKR;kj~}VK25gu@Q$61UJ*Cg~FfG(K^byKv`pV^h
zRZ^s4eVj<SzozsdDnnuM+l9^k$XJz-I*J{au;`M*@UBHz(O!K}MzAQMA4)|IK)>e`
zWQPQLr_C8l(X^fYEH(=cP_iRMvm=#xq~U@AkvbN|&7pUwlsor5ieDlgh2vyWlO7VN
z)+-TNE!Io(O#rD#LMDepo-;+_s1Ry@N!{9F*s=_zGofy=vr%c6gP*?bK(!b%m&^SM
zyKW<BqN$2$E&h(>tIZrbsyGA?^mnI#0pQ>T#Uwa4r@_H<I8KneRc&%=Dta$Jg{u~D
z<12q+arKk%6+~2BUw?I6PSL+A(ecH8Z`&WgqIMnb$Ez&P4R^q#E>ORMp?QseLb;K?
zs4SGb6K7&sFv~)@yP9cPD0dr$a^<>&tC6u_#)`HhMFCO4Hd@zVcbw<zjgnGV@R9wV
z^&~Lnkv8wt=^Blxz457aAP1Ca18ZnUF}|VSkiFPtzvh0>`}1cPofPDE=MA<QoC5`@
zQm7P)Giz(Rn1vDkw{AR|^sjb*;cWO=rIX!5ovHt^2v3whg2NG0KsSBV$6agD@#d{2
z?XJbh&0$$oCi>7RC>z5mTfkfX8FqOrs#rx!#(sPYnZ$Kau4D?fK{@0sFF8|d5mBJm
zC$omOkS!(gD)GuOUdOrw>!!gf3iB6a>+OgCqDA0V-p}L;7HjQidg3{MkjiE}vb;zq
zsw1jqWYH**2EQr`^N_PDHly@Gx)>EWzA6>5^cs00A|hP-<X!rtOdy$^duvlDWmC$g
zXtFt9=A}|FJx2$nbP5&@UU#Q7tU?lc<5rz8$Vh*Jwl=7XO}oWSXEK1(ihbpUEMGZ7
zqGi8m*}QJtbgwT~n4fBYqU?a^umH}OWB?~*GkG)(S!VG!O)0m7lnv1EMsi}8G3Gpa
zp)>ByPI)O9RO69vr62U3&5-chRKDO~x$oSPb2o>M&}os2Enum|cVzS9TBg52_Mtaw
z{=xP?>&8!Lnc*~LjglXrwqC84n*#Oi+oLX6%U{#`%ms6dCAeaL`<LB)A1gQ@e?6)O
z&klVsutU4C{A&qH8Y;-d`CyL8mz>xQ5su><aI&TzAIx!z!*7TAV9s6XgB@nvMV{&C
z_~tR)p_5A{^UNT-f#h!1b1jcDyHVw95y4*_y|u`Z)zuD52HD~k0^V8YtO%Sr5KZhv
z6*l`s*+1}T{D)J2`tU+#%^jxeLnUxGQYY6fP=k~m`!B+#o}%Hdr982Yw@y$GVbvhG
zH$~kc>HcKbp&ham$)%N@Ok|abtU|Le6^X2QWFOi_{=ygT53WYc-`>RLsqvI8uuBv0
z&z98q+v!qTGO2qFM0mUp4zVhex`h--nba+1Q)JncN!?O^I+u}7LD0UvjNJXhNVpQF
z;=(9gX}(SfS`ifW(^B<Bf+8Fg&g6&!3doCjwKu#nGNiglkGhaY1@_N_6@^jCotNjO
zt8$}tCJfJam!-Rw(1jj8j2Av`8R6?<Fh71H02e}fWeGa%^Nd>&rd`k(kJ>$y7UrNV
zj4+<v0T`Ko;Q0apxbo^4Y8tH&`kjAHL}VevpI4FX4pXY^1H0^e>W)=20i0QvkQ57&
zhkJI0U!JUdc4m_@n7Xqw{PKI%<BBO_w=;7aJfyKzcCS3#NT2Wxf;C^1#sY$BjZ(AR
z<dC35>Nh#6&x)1S6$7J0wfLy+>qfAu3VK$op?yw&El=+$ChF<b`*5q`oFhX2M0Sh~
zlh8%di#6|M<P9~jwv?~VywU%biqL^`mSDlsqaR-$DL^eno9GiWxUc?dk{twcC}}~m
zre3mwJw^{w<b0F75s}*}8de@I@Ojg@NUt%qYF5%F0hPMNAh&=D*WvD&9|t_naT5uQ
zg9pZcNYQlTaR!ggB`Hae2x*Q~+@W3?^VXc+J#&vN_lxNb^ZeX0ux36|w{V25Y)0u3
zYHO@1^~u}^==Gs{zf}C6!b>hUD=y*t+H!(#jdJv0CbwEwar5G~Pp5S)T!b;U9K}~w
zv|{FwJY$9W{8{_K1^ZWgO+eC<1cfekoioROkIUS&A6j?q`bX5bZMIPB?(X&u-Lw%E
zk^V&M(xr7As9VRi#<57}DKy7l&3xAwy*KT~ZTtS_{-M=wd_}c%@jP6zGdc7}Z0M(Y
zU1_zp75Z6CrY(K!E<7mbzH1kOe{d}`YYmPq#^A`u0=<f#v7yn0Nrv8;3k9k-n>P=C
zH@EHY+Mo5i_D%c!{6%=dUFbkXjuL(O9>M?__?mO)uduIu-LBd@)AdHRes#suOy?0*
zpK1X)5s%o>m#U@f#<sd5uVn-Nj<>}2tovoVQ7&Cqs@oxEljV-zfZb9w#6%*8EFE<S
z!X1^sN8IZnlTaImRWu@|i3h}__pN7tLNxlGZ&JYJgFkdo4)W$hr2INU<YkOmVplg?
z%%fs1QYJVLLq-aM)6j(7Lbz|yzv%pYQB0&NBgNujS3s@__q&p@fk5TJ)C~)Ke`S;t
zHzxlEwd+*>_Y~^oKocM8&pQN;CZ>hDx<Lsa>lA?09`vw<-P@gsg{Jf~H5zVzp|FX$
za@!v%An6ehUGIQ&IO}G5nFZ!GU6X;=19KTzTx-4o$`VIyg~@^ji24y31N=*qz;b66
z_=;R6i0WF9|D7oG6-pkc(N|km&Kyux3R<;GwJFoZ1dR=yy5b|G_tEs7?S?>@67>`l
zkH-Yl`K7*!yCbcPP>;adCIxnX0X+;SX5Q`AMs;%meKCovv+1|!&j=hk+sA>OYz~A|
z?!WF#IOTr(VexzMVHatSWy-};DHkl&`mu`*i<k=5vS3@-UMiUQv>9=aEuKw)V@#=C
zw(!tG3Uo#78Tc7#<k>AZ`Ub^MlWU&mg_j}A>>Y~)FKhJSkEzNS%CP8v{eA|VjUk4Z
z2HV@l#uTx-yGWl{loQmyP+23_fud{vG&X4(4#_lxY*0#DAZ@`4wxCb7PrA*(A_XfZ
zxpBQe6PgT|nvJYnr5;*a4@rxic8e_?G{KDZn!7Y`zG`X*)zJuRR7FFPMU?Hb5}dGU
zZ?|WmI@z*y^!zCq=viNXH^^C&vnCZ&h+<O70Cdw_+el@zw%F_z04opIi7I<qCfZ;q
zJS#nGn3K`vgMvatwv3I4zM^5((*$fVcpMPx(Y~9tg9wkEiK4=JSs!dX6TsMAJvptM
zgad{{@t{tbO?v?-j3P`(&a#Al7B8YKb8`FLba++i6vmjY?HBcb4C9OD1PT8xW%~?k
zZJw9t4Wi=c*z(0%yHYu>)FOn?z39=}{>Nn;m@5sn={WFUqV$NF?JKJ4X8q?;S66jo
z{wr3;Lp(bZkh%!*yqPKsg({k2Db)HUm99qlq*|{hmM;I}Qo6X3F6tf!d}HPfxb3c4
zpeAAEQSsz}JP{3lDSe@~L=w0FSes@inyQ%At7g@LaLo0}dx|+u8FO+h8w2EN|B5z!
zm8ivEd`7_?sJki^Ps`T<JUAEmBl&Xfa<LED-0!n1)a@9N$0uFsWmC6Mq28x8MFRtm
zht##AUMudwE;BFQ6&FG;h_Ek<c<(`JCe5A;bQvO<pJL{J$2AcL7zD~tWStZ9HahZ|
z9QxRL-)c9}UF*NUwC*2TO_Xilwo&`b$Bzikz70_Q?){fft@Z=Dy-Qux$~J`NP=B72
zdEa_K54W#<B~h`E`>tJmc|bQnS?kk1diYSM1b%3t&mV6e(1-dxYTvpju+wW>pFTf)
zbuZn2t~XkLsBzb-130c)jry1S7P`G=UonG<pw`qy0Na_FK|nCJUw+q;6qlAtUTh2R
z*W}^FRpxqsR%FHgh1F@9m2L^EnX(tmaIp`WQq0I)xYbR=9@#8YA6K5g!%KaXSw@?e
zc{yux)}+D-Q8;jdlaF&K(yiE;ThUiEmZRmYKFrB~7$OViC#URP*1hm2T~bh9nAz8T
z9@s)8B9o$Y?oXS@v7`tUk4;FQ%Ans7bU;Y5RJp2@OZE9Kg$G~$I|qp`X9_sP*TB#_
z1rt9JABWu^T#Yqi>%|rfh=NJYcu1O2>1Si|RQAJunZmR~f2dCll@1;eQWcGup8U3X
z%?iDLKRfcPX!g#FLAv@Af)Z0s?O6ND!J|hI)U1jYC!sV$w5Law>QAgk2LPEo6h<k~
zJ6SHOPe~l?00%xtcovzTImQ$W3Zw>yz(qpj&>+%<rI8mS?1BV|1W7i7dp(K_`cuL|
zI;V4ZSkF3jDqf&$wMNco9Xm%l>=NYB4RR`fB#fcRD%4-4H^DsxcXn(j&?|7&NMDMY
z#vc%7qI$i0)-0ZLs&ma^2ll)_Uj_MmIC?cBf_1-560w99HnY?Ws+x1TUl1R%>&emI
z|E5p;1f0_s9HFB+d`NJQc#eZHF)V2M{0dd<3y+R&b2?X3Mj_@%P()njklp;4$pbNe
zKv_EcGi`Rk(tqll`pf2K=YLnNAuUyDj4BnbZMHAo6QNE2$zT+}i?YUE%3{q#|NV2W
zo#qh=shvdBPP6t@Yig%@qk7${1hmuXb**~Faz2CF>8e^ftIYJ)L24&wa`=Nuly-_j
z{UY{rC<u1-ovcBui2)ti*anSrw%ERZScs#x)~J+EgITZqQakW3)Wl|;Z%Zxu>^GzA
z<hXf#A_`@Z2lw2qA{&@gEyEPh?->&9@W3h*OKq@*ovRLzB_1f**<f`#FT!u13^ECc
zwG#o4z~*1|v|OzedE}ohUP5y2vS7P$s7K7Mp^U+g^AQ4$2|T%Yks0mA3Af09^vx^(
zZrsYlb)1~d))$Py!F9FTV2n+OgEW+Q;E(0u+Ip5jm!OnG9N#&F_~LBSC}Mf1Y+xw8
z){AA}^K<xEz=lCZB{K3LAn&qn=5V6yU1^xoW0=-zn<I@7=SC=KZw|`bM5yhVuFE?o
z^6c75d8Bo9^`e|d%ri4Dy4liy><PzNlgjaslw({!p#%qbWfVRzvkPjx><1q*I3MpG
ztf=s=OQqWJoXmsT3ttuI&YX!s#R~A}eMfe&N5R2%|H>}NCbLsgcW`a1pj>&lYp>5_
zM$*1`zLp~5<g{9P&~qTD2}@a9<EXRbzhL}-mHjQ553?!=^H*Z}+SP=ACG*8){t|_Q
zS^3yD{r~-k+dtd?$KJIyr;S75zoODSl$qAcH6fV^GXzYRnTB*Aoqb3<8379MtYfpb
zX<>Hz-&c}lK*k1RZW5>usV!rW&i9>*ZbzWjo#zlvl8#idL~avtjC7=`7b+fiQujg?
z+~?H+C5bAxOChH;eGa{UlA4m5lA1EkJ2KAp*c**K5w@u+*wvu0VviK;VcYDuP9NK4
zw^-(rBm9J8PB#6jbu}R+DSMU&5%{qI%q|mIL+_O*5!r^61OEZ`jHeNMoc))5Bm0k|
z)<K1|!$m-ud%fGOeb&36*?G7RPL>AEZ|(YH>!IDf0aR#o90sd@<?-6Q9Wf#-$44->
zjDN-`1U9BhCj0RuustE%!WgZ9j$5xnwVpiiFMg}xz%%m)bXm%VF~ypoLd6s+rbrx9
zoNv$|tIJ`og^LuFOHt-vbC(kG8!H@+aw*EC#JCjI6+I=d67H}l$D)kF+_5OYIXQ|C
zc<SoXv5gl$2B4dNbwc^%-f>|<q5$ElP}VOSQ7qt}oaNSCCISIjVACW^F2xzkTJzmL
z_6HS=!saeu5%y-eepiqU_LOjoh^(cdiQ&ZtXY1^_fK6f~4Z_ZhfKA&;C`toR9Bl@<
zY}w|>%nb|APYhDt%!JxgtCwqccLG6H#bGlvO)ba~y;OUDfvTFofb(X0ZAd4bRo_Jn
zFNofx-D)0q0yG&#<T4keaZy1@oL%rqT4@#vMIleo<bx^k1s*ehdkYN9&=d#<s_;rT
zs8{mEYatIirss66s`_=QSj^Q0ta=RX<%GfSzY9w9^`0w9sG>%n)`u#Q!V!?JYOwLF
z;5J@D2Un<nImg`)KtQbh{jdJ!i#T5@IA3ET1AoSp%?#e$_!rN;v0)D8nSoRC`tmNH
z)8|KCocmn5)VyX`%EIf50rX%@qIHGCUfMuK{`4vT<A@E|X1X}`j1>pH**_6IdH2O|
zTn|ug1>3Irwu3)BvaNv&Jez*Xunf<DCcvP2Q_!=2C&Y*<8=9J33pP|1WtguvQr72i
z)+Y<z-3A+u`nK&1=q!_)XtKXlyR0u{=Uq@V{z(Ew6C$C#ctJ`!TJ~X(-zxPK>Hb-}
z6Iat}KkA)!?F*<sH1s4Bq5@&1cRCLp&}v6b@qV`iL@ceHiLvU3MQ5LeUP4TK+rM#x
z?wX~4ihgu0N;?AzYVF46t0sh3tVOpC?!R>(!JQ6T4ZZ!?s@M38mSVJb({^gUdH=K4
z1&<G;arY06R<ory!fg|%&Z+YR#`j*z&7!VvvVu|tRXc&w_3Y7Xu^Cx=bsPG}W&Fdz
z7h63a^ReiBf|<-Lk6;GJ1hqz^MfMwfxBjYsb=s}==lP5CRF(cdj68n1n_=6IXN$w4
z5)X#7U%oX}5w+))r(Vc6i{<?^X}2RJw`5rbpw!BeL69B-uxBE22?qn<qMyit=KPTR
z9`b%6L?%ws+Lclv-#CpkM*P1!9To9~BUp6_&M8kc*aqHHI9yH^cxlwejz94`8EjO4
z#_zu`gm%YQd&P_6bUP^{!md!fI<K59jq~IRhuzotNmgOUSv#5N{_YlJdT{XU&wyq4
znZ2W!Z56E@Fj`Y^pA+2FJ1`H&c?uIw@zuo$TF87JK?lCMPfz(D${L&%J$y1Sewc<E
z+J$)({8bQtIQ;3Nms`5_g+E(pp$(*e1+`-7y13VYxU?<M49keCx!}@|j?Y57W|az_
zwhBg*!1W+Ghjx`p7Ou-h{c@iz+Ra2KZHs2vrp=!`qV)7c^i&f4+zfORx?Tra1LUB=
zS2#p>4uXHvw_@nv_ofG$WB<~u9@bsHK(XbN*|MG!F3$*8@j-d{Dp^5}dC^~go?Tf^
zog!YFs#DEHecKvnp3(nXQr}+HaQPS!^<3`$DzP6{9~c(dtg?>EI#x%}LMGPB7hLX<
zMN4R?cyU&|sKQB%Wj+q_m10>c<oV9|i}R`3k&qB;S`lkMCYpoaBPoMvcy+<<g|Hz<
zW)hSB4)X;|=36G3346#S#>IGlnJsjMp63|7u@|g&x|m$_NWb4DC#3NuRb6LPlU=hV
z^rrMCRXU*wNDoa>M4AFhk*3n6*H9mNZyF%df+9t!N|BNnklqmxc#&R1PeQ=Z#LM^I
zANRZG-&tqPUbAMO`7yKhthp>{emA^-m&KO;HE|EvPu#->O0*EcpMRdbxzS#{@zeGu
zL$z6Lq7xPU#SF#tNloG_T(t8r$yc(ox*yRInc4TGvY%|8Lvp+vk(48Eo68o{rEf4-
zLuIRh6ko7|W#5>axPt>R1<)T1dhR7m%UN!aASrG|&sk1GN_nq=1v_WV64n)1nnFk*
z5Xmi%DiVw`Jih;YSNTFGL6p09(;01%?b%((Tag_jBG0nAr+4RKKFY83S+NM_XN)zK
zs#<Jm0{X58aDEsc96~3~lm;%Vy06p!(P3NH(+b+?++yI(>UV~o0P4I|=h?ETq?`ig
z3ND%}@e|WR`ntS#dEVG~Z*`1Iav86~u(iSW3}GQOakLy4EPr)R#ol`?O`JUf?QXqh
zrLH<QsVC8<o#D4Vj)z9A%4bp~WGqbDoUllflamGjKc~yP+}vZhDu(=OC*Eypq)GkN
zy(!6Bj~)eZaD&=mpu<dt^~OdUq+CVk88k~c^>|Ut?L9R8Q)u0Nj~@}>e{_FGu18^x
zz9er64bZT0x2lG8tm|Uvt8IR+yflqsndD1~edNI+#jQP~m@v&<Z(98Db63Xb0I!vZ
zmx|P#`*5MFd3EvG+_focgGKCx@PpRiyms(r?mG+}hdO#51}H?^S!OT2dmfF}Sk~i#
zeLN}uCDjwtU3NtJ#&VZjpP88xFP-z$D=tSjTQ}(D>%jM8ttoE&Osj551NS1ZyTjkA
zVQik4pA9PUSG()pKN=jTT-o)bu%BGwzlObedVh>><<RspB@!+dE`n!8?$07z+DEWf
zjXei;OgsV?NG%IhHGTYxxy8<XpS|&Czx^ZTgasWU^bWbKA)n8Sd1LSUCB&m|*zwQQ
zL-p;?gj|)s?ajtYeCw!7)2C`U(}I7_2C>0+-+~yMiIyM<O4{nMQr-|25|RVkX0ScT
zo&pFjV<)>R8-mhkMNQRR8o{ycdR^GUwDE*jv>%Kx)3Y6H56y!Z>sn}*eBu=BGapZx
zeUhhi&-4GOEplkvA)p+R8Q)u;WG3x%piMlqxfWn4RY+PVhR?*C(k^+*vLGEGJRZgc
z#K6wjclm2rnqIL{GV)M}rXo$jAW@8~3OEgWQ}v&#Ux#YSmWDB-5Ck9*KD0?6cxnW-
zeJ9;&&X%LN6d#l17jxox`Y^T5rmuiG?pc6<{z+)%PDISs_Xy#Wulah}<vRCU%UvnC
z<#Yry%MD{s{PaEe_$2PN$+P@XGF<)Azy;!Ykz|YF7$mC-$EY!tsJHA{@`M@zHEC>;
zkP6yvUlZznM?h*d)C-8}?zaHj5vXS$QbXw6R|2go?^Y`Y`ut)~nsolQh5%=of5bR=
zKCpRaEr!@twtUksB(&S~Bpg6_g~W7_;*X);V&+tAEyj&H#LgwM%$TroC6(f>f%@-N
z-$mryRf1WBacK?d^2|%cNt1x~F(`{Gxa1mSwRxPYM?^E>*NMkxPDm`syFzl5?4LKI
zBEs<{El19A@59Tu#zmJ@|6%6ap7j_hlR4q;J{-;*kuf^Sjo63Dy;swu++A-qepcE>
z!z@KaXxC%DG{lRQw;m~7%jsr{Ll(w3YY^2u{WDy?M(ZLLnx&x{cPW8pt1O8Tjhl02
zCIQ~c5_qC^CV`$ZOWtbxM>?itUwgRIjeFWC=y{YHSlx?;cN;<KTeL$a5FJ*jeS^qS
z<VRhtx_bA6xq1!jST@Oa^240Tl9uL!#z`AyCZt#-d*R0pk;UZSsKqVv$I=UCRI*XU
z_XQ9xKb~(EZXlzgjC}#xv_YFn9zmg@V2~Z#7=;SC0_x6|LwyTHz-ZZhX@U8gQNoOj
zHo=#cG~6jeIH!d#a6q@mBHv>@#1H;t?TL!OOWbVz0g4t=PtHe1$~FI@C1v*NYZ%Hq
zU$@S4NwtYLUN^UZPdKJ2qju;@57pl7($cLajb#{kGbsB1ih7QFSjt@<Y&?f6!gzg)
zseJbmH64A9w>$5Spv}B_F%0sBOl~nX;2(_9&69=JM+F=#gwk{>Xc^S!V3r_1uaDJ`
z$)VIXYq)GGQB8TfkFn<Kyu5rnZ}c3r!FBtjTo7hB)&;~36jKj;fpF}vbosUZo~3Jc
z)=cczP`78~F=AAg2TZ2YXZ*t1de-;_(s!Zu={Qts-L&s3O5(WhGi(7krLJOoxJe{y
zrPW+f9a_HF70ZgA8)2w@9x)Ri9DO2Fa`N!J+IV+G>#1>Iu%gi-3v1@?AL=4A#&<CJ
zsc+OK@fP|(*XqF|K+Kix6hV277-h=a=w^c?-=}Y`&~b44WKQ7=3m44lRv6V*n;F#c
zG2Ql}T2sdaeg+f5!#~U525mp139jG{tIa7VfuK9llX+n$pi#~%UPRe>spmInm|2!6
zIG8hm7><=TWoQ8ePR8Kn|A|%<hXSY<{p&<^0VcZw($$tk{*vVRU2@Y9ju%O9z2!x9
zBX`J3Jv+xsMW38zrLE1bWi>suS#997o4YnEd|(z=LRimcO^%*?>WG8fi~2(o4pct1
z=}OoUZN0YjK|fIHwh$^&;`S@3sLE|Pv`6%|ECl{_%L0)@yI`dD{0_rQzxv*ZUUeO7
z4B+zoqvpzYpPV6MbNK<Gdrl6wRbu=fN>g?(_}EJgzIjzB@C{EC&@4)m3pLe`?qY-f
z)wzwvRX%8Byl)(RExmYo%W(ONVZc|su}B=u_9KrJHMab&LvE`#7e|ABF!zHM#dYfD
zlxH2(($1e2lgIV=hPqn{%~OQ)j?B6p<HCR-xvKnAj~?ne9}+G|EMKZLszR8(yZ?y1
zkMvciTUDy7zwY^f$5d&r#K+M?8zYZO!j@2HVo_nQGkDDSN)qbFe&YoVy29XIx{$3!
z*GA2pP3s;#EQC$wIu<I$qW1_QNl}X9yd~io;;l05hWV51=C)fQX+-<07*R~_KM3qb
zY5T3Sx~`p!ig+quTl9a0EMiF=ePpF|7+duUb%n=^3|_EeD@X)wLZc1OCMG7#J1yFT
zZ;8ItCOsxy{<%I7npR~46GH}s8D?#rj;hvh^RIOqx2&NyTfeV!6u%V&3F8*7qYrnS
zVjKu(JXwO_8v1H8>u%3^ro6Rj>wx%jM;E`M+!PwULc<oAO%YD5^?MaH4Vn@gYGh?)
zx1n#v!8aoYZH92auAi}aR$c;We?=i|G<s(|LaA3`@oquobb{HVrj?-U{mXJ@m5~$P
zdHD|^GkN)*5V7pG+~l;N1t=wBhSL>EJH39zkU;eT-?9TWZ2AKY!}>lqU~9O6&#mLh
z2D5zirgAJZ%GEB}P{`bFW8fhbpT46?5v2HPav<moAytV>5=nh}allFXtwWN*su~Bq
z$^8jMCd>Un3@fLEpZ8Qk`$rV%kZ7gP{_1bApXDOA=JD<HwsmxMS_Fz>$s<i<sU#t0
zq@}*8h}cZj@7EuKP-F~1|GJXyeZvt&dpjOJ<KtD_BL;afTPQ+SOxFQMRO#`rR5jSh
zTQ2;Dso?7`6)d7F6drxt?SVonSxC>nAo((XxejJN4XqrqX$NtkQa%e@#i^hP(hf_k
zv$2Pl6=w{BhBI-{+K#X_ti%l_x?-a>pT`;OHSxhuUa*KF9|4V%C_Tyto4zzzqYWPS
zhP#(Q%+Tr2QWoz-UsiACmn7|?NT1}MbHZlttSsJ_!!f7?GqxPktIgMZ7xG0pPPkZ9
zBLbYszChe14aih->{@Q->0x`eyCo4lc^&&UzA}eUWJ{Ty4pW_i)^`dy#j$6k+cq!`
zZ-u&jc7A$dFc1ap+91&p$Kt@CH!56f8rXF4$0LrncRKvXwdKdIR>z3)#*0{s6snJJ
z4ydMo#iMluV)gn(&U(EEavZ_!IabSkgn5Snu#$1$v}eiP`my<4DcnFSziCP!3>M#Y
zOEuj{)lQFU0{XQ1w)$-?f8M&4-R(+z8y63H;2UuQcwZ~la8rAOXJ3YLwE3V{w=&^6
zBv<0DXayQva2mmt)K@CWm9)V%0<Jt%k9%@->q3K6HhoZs%?0|K^TFohKsGEi(ITfz
zE=rKnB|1}X)THI*{_>fFL@wc>Mc^Qz;KtOpz6EgX^Hh}qbud5i)LT?B_eSGrc-ePK
znPlx=ATDN^nK%$l9~tus7r%6mM80*vY6V8nq!P&Kt58f?Sqx1tAL$(ipDMNuH+U!}
zbb9(2B{h63w`8oXVjNbrv;QbzD0Sa<C#rCVhZ&n)<8)dGx>xxGB1_rrmwLS$&z;+J
z)?4H6In;!tk%v-#R^kyD-Q6T5+Pc%_S=6`#&B+!CJ3oqu**|X<Vcfy16tz7LZ&`_J
z1;c3OiEbL=X%ws1NIpKlm|q~lz`J4e5WPjIm9r%2`iONNLpC9nWL9+UJ+eK(p_Si|
z4-}L3J2n^ir_3#H+<~;8PlW$K{e4HV^|)<LdKfGrZQbX6y?KjwtgX3JjQxX1cX}b<
z-E*DqW^>zZSwCUc+S6b4=9LoG#ZS(ym2-D_1)`TkJnz!C$4@0l+uTP^!0Z-k@A-4>
z|B|rdB4ut>+!f5}6VX;94ydE9T{^{iHEyzHc<JH`K1F=`^23&76}rJ)@FPUDiW>u!
zA$hQT`OLmH&sZFni@Do&&tdG;*=XYw8275#Fr8weA}sm{6zNSxv{sJ=f8QQf_d~)u
zM9}KW+qAoOMh9bobyP$|=x3OB^wF|t#2pRrN^3gbad*|WK3#kAG%OvZHVq@7g3H8Y
zA(B%}+jyQxdo@gJn)9m=dl$XZKIzap(<^_cI3mx>l^;V)#tTv}QYpSkt^jx&>@CB>
z4+~;^8%J`x0QXRzq&UT0{gv6e=&W1|zTOG=@X`kt_6XOd6W;4_>G?}@ybj{<<}cyu
zS!2$*LxMc`oG=jCxDsZ|Ql{#MzgnRi3LXk2XDM`-O%E3B*x*qRrXJ%X!=r{A$e~%`
z95F1BRrH(lzS2r1CiBi9U>kzbcw>lWS9l|b61>F>kwN<p7zJRX%9nj_Wesr9MOiMk
zE0{)E7{A_;Zuf-bPtRLWnuctI-PT8$xZY-g3TrNRAEscoL<y+BTakYDEDzRvr?;Nt
z-nyl2$Y|+RuYM)3g>B}#(PI$}lHQ2-lQZ`FwQyOXzZm@C%;9(79}6UgDX<Bt;q|l*
z{R~}n;e6V&k58<hwa3jJBfEV6i1-npVIceVNJ?YmRFHZ`6_O6u=!>jiIQn{wtY~ju
zs>#4=jh!E+3PpsrM-nTFG1i)ANX^SRJnjUS773~4LuiqQohC=R>HcHu{Tt)0j2t3N
zuI-OF{frE}0FEe*TH5~-GW*0dBb>+jb_t2c2ivi2OSK5}o%5r)<I7`vFGOBMv&Nzn
ze}OH~914YAKHA?8`}5l!8S;0f8a_6Rz_*`I;S($V&ZFTRhKO4sy@<<(wr5Wh?AtH!
zr+GGr2)w#~h_j|v1Ki~|K_NI~Q&WoocZm%7y)p@OowY3N0no5xLh%qGIC1|s&Nt#O
zwiv$z33cvtJ0125yX=(j-o;%kZ@`8sEVS(E0JYdx!~()p)nwOZWb+wKSSi9ROtyWo
zR+E=+to>v^Mp`TKWP_eqMc6r9S+u8*WX9W%*N2~;dXf2}P8*SbE_5(Jh-f%cwa9x~
zpP_1PCtStuGZ4|)_zkT!9}x?SYeyicGW#pr3!V2P6BZOMmcVBsA<T9zA4XoytwDK>
zJnPhX@=~(IZ*2c1_%X`zyZ=1{DE_S|n@9aXbDZRRogTdKnpDsupz9WW0yg$r?D#@c
zgQJ!8YPPA%k5{wrEM|O}?|M+padE%IY9=Du){IwSS!n*La}$O43>&dQxjJ9pM`h>K
zP6j#PE`O&;Ug<37k_W7MI@JDA&SVJW2r5bv)Wu9mfMZSXR)nt7qfh1LOpO^@h6i~#
zD9TQJhUw-Hzq;EWT4E}HZE8im`BknE{hf0SB_!Pcn2Slt>#5SUFid(9%^z)>rB|O?
zV5yIl-iSRk5QLX1b13LEY_@t9KjA;{P5%o0V?Xr7UTmLIb`wwczzn8UnS-?(zaWKw
z$p*1fB#IT(nl|sJfy6+7N#`xr_$8wY27@I%q!FmVQ`r5MD5Fu~_{nr5hpP>;yoUbr
z+gudrb$8#Q|G|vEcY(1|CH=HRQFTB<2}lRzYDe=4&J`+KzQRc{svvD?<<|d$ohN^N
z#zI0l+k)NB)2SnrM#lBQug6~Ujfabm(s=EoUwHjc2@sM2SH1NHGJ=w?tu1-bsE9fy
zu|Y-Oy0lOvl#7*UnBw@{B0Xk7<b97~NcVX&D48@vH7qw|lmZ5m&Gg$9ea1r>!-%~D
zi<UFT?~K)6!L&My3A4|X-NgfulGnveC@SVj2{|LhoY@kZ5{*NWc?5#<7#+=O<XlLM
z!zHBP8>_@+RkO4K`*eLoML6U}AOf9r(6_iiBIvq0vEW0sAru0YDi_bxH@xG=a$ldR
z=Sb;&TzJ#U=vk(AQznjIPEh*w{J1037(#L=gFl520HA+^3rFUIt~W<zg6>gJ5WxXm
zd(DH{APfj7lo#Sq4~!B|jbA2X9+NSDJesl*E^?yX{rh9g#Dj?JQ~6@H_+|YE;fH8B
z>zj&Jk&0?9TkgYrp%^Q%n^0@PEv2v@vfr1jj44b$oxG50eWxHU=TOi2^QJhE=7-)(
z&#>8Ra(6Ho$!c;Jq`p)aIM4}<U6h#2gRGp9->~fmN>24ZAinG&Q+_o0bQyy=onBva
zBHQ*|6QZ=bL0_tI$XD(pp+uKQq#&=Q3!4ZN2;a{zb`{Y48K8WTI15pn3*)R74Kj6s
zD_lNWXq%QaFxnI<VgH+yi4F)}C_x?L)ZQBYon7|D<{XGNT1&I@Pij1P^TOouN7v75
zmc3Ce!W9N_<?1n)O;jJRRJdYoBcV6!>V{JXveA(_GIGD-@m2%&B=8A1=oc+&gk#Rb
z1Vl2i2mksOWC9w6Q|5uhx&N=zLqftt^52IFuAc|e1l7Tl@<0%BuQ(71yeJOD2j9#C
zfk6RqrhJeH_kU7*<;v&@h=hdmzl}*!;E(e`lK-0%@aTLH8)yIy&j)FenP!6E(U~At
v_+I`Mjy0U40Hg&Pf?E`T#JK-u_6lmBjD&>o|G|~P^9n#hlzs54SCan$lm4Oh
diff --git a/fun_gg_scatter.docx b/fun_gg_scatter.docx
index 46b2e4f247e039a59a2a1931e3b85246329349bb..ebed9d4c1616780651a335790b4207bbdc88014a 100644
GIT binary patch
delta 98789
zcmV)pK%2j~`v%MV2Cx(f3VG|Jy`*UY0N3%89tkiIw&i)dcG5HJ?%w&3O1dBtlU@lT
ze{&DrlOS}d|FPDz8Z|_Gn>wyP{aE|<_2%TPhQbK@4)&-|KGs$wto?NHKmW&nJ)Aq#
zUe1XhA@&JBJbzf&A8WHHTAZJrhW3oiacIq5JD?$*M3znGr*txLZF2fR1Lw5WXg2uY
z76G+M7_#5H!u~xDYw~mUW4W(60e)aFf5Qhmr}hj75qaEvqxs~IbWaDTXWJi|?{1AQ
zh|g|LXfVeS`)4pco#Wtcxj13pUEs(ayPg}Z*e4sk_%rllZ5jCI@)IZPrNAraVkz>E
z_>G{<?}}HhB{=vmo(9BYYoLBOa~JC~oj><$?B*=~$bHc!?&n_o+{2>T`7wmJe-=>L
ze6Y;&B`D{fSb6amn~f3>flsX8D6@q0Z^kQ`W7psO@UxC4ak$Oy(>~DJ`9LouPy0ky
ze&RG-%{Qd+u$cZBlz$HBa<TcE`(t1G<ln8o1^ns>-;&sp#NhD97V~9>7mSqW_W7r&
zPXj#m*oqkJ5d$CM5I{gBwF|~gf5+5W!JiBCaL%}~bNjK@Xtdkiv))ZDzVurF7n`GA
z=cc{BcugjF=|ym--)eWS`|+LM;L@PkxM`m8Us(JWz#p#fum?OcVXt(1{OK9?r<~!#
zzfa+zf7|TxJ@!7<;6ZUM5bp;!)Q>`T8;7>bpmjx;flC7PnLO}s4gGNUf2tj(FYynK
z#ec6t{-;en8pOYP(;xIYJ$N%>=T5HR-p3lYZN`#%;_dK{_+hp`dF2Y$mAG~)SADwP
zJ^aG%S!%vL_L++h2s;jP2wPsXf;*A8FE+n}S!q-;m}M|)4`z8}N*H_@e<MK&^g3&_
zTUVKwl)9rCCY3olwr;l6fA|{lPkz44LKt&wFL!LiQTMFf$UL@AyM4vU;joWwhdp&~
z+MS!rw_r~%unXWelcoQCI{hBnIEuD7(9Y&sjrLiyr_bg<-#iZ&*k<bCA|N3N?n&(e
z{n?BDc7oW(ZLs9(`?>B5(m}J??i>Um=!Pp74vz3P`s=g$W*+4xfACkE-DK14F*20;
zGL)(hT0a`JHfUWvv`#w`3|bqst^`^$8-h&HNA&sIFTWTJhn;qA`kkBZ8pDNZJnA-v
zXPNH6^y5`T`waX}A5QnGbJ^9&faXYV<)M2D;)=l_7qe{-hY)ge*zX8xHr)KBxOo~e
zT9@tNKqo6%IK>T%e>Z6VG_=<T!>5GK8G_f~&+iNB`ccgwI_$K0-EOx>S%{v#V-Q_q
zL;$R}!iX#^#|1uLo88x*!3Cvs8}-{m4(LN?jg#Qy<~9doGS5tpT>w^KZ!yCRQyNyc
z1fSG<voEG4i0yBsAsPB8K_BwW^q~)jMpo$lK^Td&o=hORf9>}91aj8z4M%H?-%B9r
z#|?p04P`jhK$<6zO|uLFwMWgX%k7-bd<`W1_(8hb)RMlXP-4uTQZYYeD#x=iBi{*Y
zFhe(gWdbvj{wd?Vc?xf?hWEyEeZtv8o~os%-!&(bZEKTbO7Eap>Pv)!h<(m=*j{0s
zKAy7|zyC}xf3bZhS1F$zaSRqu3dGd=q%O)@qMF0L40G#Ub%u>W7IRDAG0aUP<`%6M
zhr!%FZnHOiAeFtNT1Y+DdA>U}Usp-rG32ZVIX_ySJ1>${X%#e$1)9^so)B`Cwrc3z
z&^urU&^sO*dN=fL=>1UWolXyz-fKS1&73jxZhVg=f9YLslwnLDa_u|IAD2&>G(?iv
z{#mDYc0J4@l2)rVXsshSMkwk+4<WIs?{KnsvT&5xpwD)HfbezO>ks69q-lH~;ivBy
z=~y+Cd9+;L_LdH@{vZKm<a0Cp<;&<5B#2S7s)9&hYq4{;$YT^>YmNg39-bi5czROm
z;qYBUe=dhhE`d9p8G3mdy<9hX!$cb~vHfnhd(+IMm-OQYl1Qc#4dkJE3b2YDbuI!&
zjFv67XOCw_z5@F<PUy<8|JAc1&(&b`r6N&A2Q*CJ0GPlt!qJ8a7$#ttK$V!lGh)=0
zVFLSO)P@NdZ&3A_zz@c#4Uw1_^<E;;i5^Uhe|l%HCPr;y)Q2b?vr%ScV$=qao~~xm
zg$cQt#M$?t7VQ#eSJs$D5uLvX!5mH!t%*@@9d-s~D7G?z?RhwTAi?O-P3!V1vma0T
z4o6XWsVXR1K^r<7Edm@y1lseI)|L%g91L1m0Zi6r$Wf}@n?a1jL5yg|uFR<Cn8sd*
ze~KSMqd1txUW^HrX%?&bHVqxW6dmhL1Q(GTaZ90QMlBemucGh*{I=F2%&_+on0__t
z42G4Sg;qsO_sEp^j>WX~hav=*Y)o?mXLS7a_S?u%GPkjN?OtcQ%XY8X><+UtW7_SG
znHiKunADjmvWpDjesG;=R)q9LsG#1qe{yNNWPAB;r*qTX%=k>;c>0bJVH$HliTcn;
z`$e2konC8nZ5lA^g&G$)h+NytbwTYKXp;9;wj*<L=V#sius_gelNUyJGyaCX=bEdg
zY3F2{1}+3pc<9xjJxA_VyU{<(MEleoj`s4>;i0|eh=3gvHZ+_t#R&rvEd$6Ue>A*s
zz0>P{@4DY#Bffxa{*biv<0fzLJyy0sWotTJcr;okR+RwvO~2bMKNBzg_<?{seK=>W
z&RMgi3M<IMfe4RxZBzBuTj)zKfx1RS$JeUiRFRy^SmOr4HIcg6pnbrM2`>WI2W&3F
z<i=HV)KYP}>V;<du%R|1PIfa8f84RPz|kzLq&*V>wmkMff&c*TYQjH09cl45+GkCj
zEdHbaw;vCB!*+AD9l2?A8tv=^hV<haa0hv!Zs~K~iw5#=6k9I9`2!k$`nv)k?>CwT
z<ZlA<Yz5#$p(1uO=v<FpOPhK5&2U30Z$DasN4Tw8D70}Dk$`XT!DX^Lf0_}rpke5a
zUC)hHD58kolfc6(<WRPqNkHdlxq$0!<hYXw2?*1>9UNhH56?-sb5OdRN&V!{)97yq
zYMw}9Ya#(9Jkk{bE->SZFu=MD@N|V8+5@~w5wErsCSX)4SO{wBX#kQ7$;U&s;0Sx(
zDz#+LJK!;+rBzZ(^Kcm5e=<(UxT5nLo|4Oe;5+^nsO%FOptVM@AhtVkZ72!%!($W@
za*k$Ev<S~nPvt{a<T@*cO(C&pFg=yYd0~9}1ilekvuN%qfMdq^dqd;z_u=j5PoIDO
zVt9+O0QIy0FK#u}PcG;~h-|hL?D`Dh(dQv@eXiR#|8b1$x|Yz$e@3mxt?+_IKw2?S
z;D}6*{>kiZ&qHH^0y1Z+nZy3d<S?_o18Ngyfcw*9VUlA60WJV(oe5+1m`5Md22Cb|
z?Oe56SFJ4u+dXS_bLt{ej~^t1ZI7B)mp3Y6%?I=as+VyvUCs%tsP0fv(R>VU8EZ@f
zrrOC1VC&1jZVI2le+jTv41L7DM}SAn=N9_HFzdgu<B=8PA#?Z<xD)GrVTA2FupcH2
z_bm3n>3^0a<Qn2>v)^tu2hILk`-^7eWa+!n37wq4?}(3MFQoi?@HxC7mg`TfW$0S=
zjD5_t?@j{3jdq*FUpnQhw9q&9F+|7zCU(S^>{IrqZIeYrf1G19UPi$7Kpkc4VbtXV
zQ7uD8QFUTXtz!iF*XQTrnW&g?T(`QHeSz|dp#U60UF-x7x8AT?$EfD|_KbwJBlg$%
zympM(%S-aB@2zBB$59DJVKne6Y2d@#pTGTm^!e*^Rq&fZt2MlS8NjO~A#CbV03-j^
zn;QniMtq<Zf5!-}0-iH90{+6?yujiY%)lReM5q{o;cl}=1$UF%yhC*`93zH(n91gl
ziLyzN%RNb<{hNO8=1Rf+Wb5Ttp~)vVlZKdNH*c`R7hwRfyI(=qLUSHC3%TzMf@WL;
z1JJ@hMA(5VUGaSnuh@O2NZkodJ_4^9`x}qRaGf3;f5{PDs35T`pxWy~Gau07c_G;h
zd)-z;MXMxxKVY~rv7&)2v1d5IHnhbJqkyeiQ_+$WBb1zo3x6i>z2)3TyomxZq{kEd
z=oU&I&9IK`kTGD_$5FmO)e?@oB2|}_e-c_w6S<*Oe2xJlee{UtOQr=eBYmR}83w2T
z0}PJ=e-t4w3E?ve1q7uZk{b;{#C=0e6cCRQn*&Nq`i9hQ*;^#CE%dvl<aiOd5#fo<
zoMRGf&roej==}Q(9BegkkHJ6JPPx=RK@ow^7`5$*YMUn5k*CY84nddGg@Oft(|{2l
zqz<|=I;bWatDp|LGCHWyK{eJvHGu5Bu800yfA#*R-EIswojr^LRyD}xDeXKVr&w{9
zu+?Qxgz-xF3E&(&?o@r*rUkT0`<yfQNi|AUG$yK40M`(dqWWXeHaa)`&P`XvZ5%Uw
z!{J32{xsWWhO!;VYU7$9mspf7fPHf2GOf@K2$-F5;MsA*g@;R2_~h#>={qHKmR76B
zfBSYCo#icbmOM09UAbl!IY(afaeD^}E;jC7>eD2}i7G~RElq61G2~WRa!WgsHrZP;
zn?K=ADw03f*R5f@Qi_4JY}MP*lQiw3z??3?Dt5MS*w=n0^~5lx;aplO3QAM&6TSHf
z9<07rk-fLErhGPgH|6YPBM#qrM{g2*f2oe(z4$`e$;2t%fZ&OZhjHcEPVYmFxDcxE
zm^E<%dnl_$<Zg32>SzeoV+4u%l)KFRtumNQ4N3Zu;!DXY+SDBr=}}A>;);T0dpHa&
zUa)bFWT>PFo!i6iXrNCOPQAI?t4x=*#}TV4fup4ZIGD<IPwa^D5~k8l^=ibVe{@-R
z#pt4~30=gdj8K9EmGH(`(K=d_D*of*g@l}txMAx0$)dp0v6&atGK|Z`#Acqkolm<q
zye>=M(f0IYpBKAX(z-w(Yx>^~LCBs&3LbpS$)V{?K=`(NaCu<c=9%78Ua~!d<~EBN
z9+Swmu_uIqVei_r_x)kN6mQvte*ldlt@y%4kJHsU*Jj-=Iwp8c<%C}?^a9vut$S#j
zr4xsb)j$0U#*MY4*8C_wmpD(>ANGamjZ&es%~%}U&QUaC&qRU3@s<%i6TW-*@Le%z
z)U~ws7*IOF;ED0l$^}QW%T}8GK_Vh-x?E%&jxHGf0kR8eP^w@~la0m&e{&Ez6)r1A
z#1ObEH0I$U*Jt9HKs*56Xu}(6sVw3EMX$a%NgZTHrsgCCbrHNlWf4*7<f7IZ9+C+-
zf~EvyI3df<<oiihf+~YL+k-~4$s1xD>QoMo^Ig7r(&e$_DsdNs*vb*|?)-=FICZBC
z9-p<j|G_vE-a?=h;%kZpe<@Ftpp{`H)2{_SXV>&7xONK-s{@H_T=P5;MS=0KaS99#
z*zE=Nd5~1_Q$FryhVKb`!i7mi435X>VTL1ePj>YlU3Q-pq^KH%vx6B5QrkU@KSA|4
zi`1u4iEirMHUmfBh&UtS6x$&<Alx{KPy^~?T-~SsiNEyHb3e5Pe+@W56HXwZcu0%z
z7*Su;*YehEJV>)n`z$&F$#N9jt%XKaCo2vrF$_db;A3E;w##E*JE>hor!i>tyO$kp
zHohB5T$_cA&B(#OFj7rR)O}XYyk-zAfs>kX*WJM1CU0J5V^s)#B98XphG}=wX=Z-M
zWw+ICuK}4*OVbLge^E=70`UQ;rR@T>G`ZQ<b$`&(XJ4+ad7l+mx&t)9IqJ9P-Vwk>
zH$o3M<l6Y!1I05K<IJU!j(o{;b-~y@%@rEWUjI@XkpDis{ru_k&lN&qn7NtqS})kQ
z(?%kK<eJS>n|oSf=>_bFZ<22}PE-U}Fr(#XG|1bR>5$fue~c#dK5%DP8{HD0>F&h&
z4Nu8sK=2)Z>ky(-pJW;juM*=^&g0qS$o0dBU?(RrmkBJ7GGWI4#AABT=w#_ZvDug|
zBjgc(8qH+onIs<zt$!=_Tr2`=lMrT|k5_22^uJH1-xt^;QAFxTXbF6Ui^Dm7jPs4e
zjBteK{I_NBe<sdFf?8!xyxRa$@8I)NU_(N7PNGB<pwVnxU5&Dsj7iwgC}CrrIbXzK
zB&>D8X-SeBZ(S@{Wfn)8n?4b{(|3#rRSNPIOyh{NWA|m#KC5a)eW`U(P*`Kg*pRUX
zWX#jzV~bV45sP9Vj$FTFIQuED2uOw;4LQCWV*d^je^H9no$^0jATn`-jNZ-qLM3~b
z3mS&**!A3Kb;NBHY4JNGv;%j{bQnk_DLRF-cE5kcZA&e1^I=B(sslB<S^(k2!dq1d
zmJzKyBt;TeLP1Z|id!gIJ)T!M4jwx80ERk?kU$V?(_3a5kkwY+(GnbY>+~_6wVL>{
zg(Pp6f4a)ONN&|Y*$RnG{j|5giukh2&$-E(1LN4$%JbwQ>;rQ~!$bx=zsZ>iAYAJ7
zK*@rY6se@9raFa7di6T`z=zEimsGd9-Lso(ZML*$T!NNh+ICv$kd;j*B^a?T0N_mW
z4p<kH#88S`NHFcjRZkyz!dnQGo1Ahr=njYae*`c}Oiq!o(u6xaX}XxQXn0Yg^+cwk
zwq?4QSZT)LC<yKkXjHoF`?bMd0e7c9V6TC4`F%lMKdPa6Xyb@+E|8uj@*)f-R0b2C
z;|<Ro<)kjI3rq*gnjyMPc_|kkj1Ab%qr}it5$-cH-S>$dxzsP(|LCmIAGK86>Tz|z
zf0m=`PXanGtas9KTONHNOv8^Tnz_>%{8dZ*%ii3huCTgDFpCPo1Bu8QxEo9CN^81W
zbBjt7SaD2AFF00}=0L~5lW4bxtuAmeW`QvaDrgoga5Sss_sE7V$QX;(kjpiDDg@9i
z3ofpa**5`2Gf0ZsokDya_1i;z!Z!hZf6&%I5X~uEU&QE*h%_RySVUfrdX3>GS6|B=
z3b`3ZWKk=VnxOSwpCvY{0OVX*dQjp=KWJ`ABs5+|$O-U+50`*u^M{VWz8b^UsG($7
z+CpClbcrQADawMaSdP1?e>9@^5W|-<8P%E5!2N?Q4SPI}Q^=lf+E=|x6*qN^e|(A_
zWlHZNp!cq`)gD752pL`YAPMqb668GzMEgF?$#Hc!<R~FgmK@WM8*)^Q9E*sq)9#F}
zm25REZA;g_)0NUPoXHBVQWTtoXw9d`cx&jGMn_(Pxn1>I{p-vo3F#zblbxy<AIt9K
zQ^edPs1@7|@wxH3-KytDus40jfAkc1C(u5}7o1t$pe`w<JN2YmeZ1~y>HXwUbdrW<
z49&bB&Fr?t+0aV5E__oeoUCLM6~m%$fSe(ns2)#MoXaZorxXk4i<N^c*{KwXC`b*m
znKOiyY!di;D&#VgPc$nH))Y`F-#X(uN!OmyRu7Y{=tYiPTYUuFU5N<cfA-cCTynCB
zs~9yM2i@<9w=!gB$WBu!2wHH7J*u~zi$}|IcUVW&zyn-TPw=b`0zBys-=BZ|nk^uX
zVdf3>nW4zcz;I(m@{Uq`W`M{8s>5U8cND%z{zJZKk6tPp+>q%z;lW4I3~pc9o*`Ey
zz{DLlQ{s~VN5%P)hOXY3f4>@nQW`!EnJ^K?huH|pwvk@YDABcN5F7MqBzrG<L<Ad(
zyck7?*1YbVjZ~cJmxsR|-A%Dl5L?TnAf?^=#)Q=a1*_#bY9H}H%4F$_(g!ZdL3c)?
zNleHiF}_F)<ioMx@yCsxF2_vz5X?zMWj3v#Ey!yKp~EZo<lsuCf6V`9=>{-rn|Biy
zlK{7;`>Y3YbFn4b4hLWcP@fMdmqU^hE8uUEM~K_XrEV`qu*+myVli`&zrOB79MFgT
z1f1`KYF@Y7?WYiM_1e#NE<25ZK2y?nsaip5O%;RWaUm^_zqs$OvPn+an<>%kbBlFo
zsjKd71*|a$8E<x(e{_i}02#T-(m)vKAg4+uOGBmP<4&enjBRzZtsGW7sRgXMd5R@k
zz-GebB~z|UlI@#;`w1;zUmfhGM#Lz6dl5F*eIt%sni>i&{C+2K-JI?#BeV|qoZEd$
zd3=tmhj{RCQ-rO-nLoW|$Ot<l><l|Dzwp_;Ib9>{sw3<Ye?3OHxQm5fC54@lYeuda
z&YQXLxx3Q^#2l{?Z2Af|@Ykam_9iC}c!kC!dLT@^Vdvir2tEVI$vuyBS5nRyqb@`E
zsR!9SIR`xfi^)#ij)t=+T7>7Pr_(7^`nIgdbyiHL42ea9>1jZCE{07`*KNwJSu{^;
zn`JQEL}7G^f5Mb&b)w$Vvx#$GliW_TdD&GE5ZSv)HyzbL6q+70hav-u<)3LNQ`}Hy
zi!I$;wlA+$Y)6@67kw?UnKVKSy8ZS*O@)ZaW}B`r;HM>^E^^Zl^#e&Z(#eK132M_5
zP>Cu?j&~t+<MC5TY&`dyNl|y|(_pLnYZ)nO%u5|af2)VX=$7~nGq{}J@RVEz1mE$u
zswZ@dI#N|gFvgW(Rp-SqgIb1+6GN>A_IP+oQ<j|My^<$q<q=m%=>bG#V)UNK=59}E
z-!|wDcP5j_Du?)DLe`|~uuKJx`o#T#QS)Mk_f&!uV1X_h-BSpl<(_=TsRGxQCQY4B
zB5r`Pf6UBY;AsK&**oCkU0D?Z@I|&469P(r%UivVQKl0hb#$-c0pN8gDOD)-lX;yv
zlQ4VT&YaEbR&UhE7Gm+mRB>hU@q=XTn(1vHkV7L25c)h51p#S_m_p~O-MZQy>aE!s
zuoEcUWAm)n9W?b3Vl%ub*1P3~(m;E=UGh5fe|S?jHx2f`c(YI1{zZTtcNzXvEUq;}
z#tKo}h3lhvVPBZRpwsKO_2wS=N|Wl~$<>fzk)Vf?RD@u<g;8`xrQb#84vrdiriL>3
z$Sz2(=SHg|(Mv>V6lJ0+!$G5e(^7G#dzw3FA3_nGWr|?az3QBev|-w7&tz8UL?a5b
ze~_dx$pZx~)fwEtZ@x=woCmKNd9HgF=5PIg>+h`J8}9T}y6UitJUX;(AEqBaNc&J8
zj(QkYXdi+}QSAO}x4UP(n`}%_kep&sl<FlSET`B_!YHGdS!f2zV&`7B(`sxx_tXVn
zZjQTtsF-8K7bFzx2e0fs7AGEKsA}^_e;D6BftGV2&ppm><E}d17Mff|@NRT^t;SFr
z`sAFnTHLdYEo>mDG6cEIe;W^lK?LOyp^)8H%WtGEay`dy6o)f+oEBS_{<AG+L%b@Z
z`fIr|+N0*x<+jz3AL>p&ZbIGaW6r*0+3s1Z+is{x#lV!tG;m13y?Mpk&<rY1IZr!K
z99r4i6yO6Mr9;u)gX`qaw%0SP&STO0VzuhBfZqc*BIpkiP|@|Da#Rvnr9MJcEOMT`
zom`VqBo==#@)%yiC1BY1p?d)&jN^;O3!t$fCK4JXUGPlgwFSB<Llqgya&`)z`xmDO
zh&!7U0jYwcdLgUOP+rS@>B{^BoQfHP8;CJwl>Kmc#7CiR;|BUD#;6K$AkReRuS*WJ
zmJ=B2_61V#CoD9KG0XMY&#*VaGVnr*ux*nC2r7Tq4<my4d}O>+1`R)oqxBCh0&0^G
z_}zE~Q{lc(r{5RYBTRPHOEjFzEamB5x6f`il>nxcC*yE*q1X>x=i#zG->C+V6epH6
zdi~Z_Hkv3%=iE~zoI871m9kq9j3?7`um-3xSKnH})D@eaV6GN{t2yWmhdX+(UJS9S
zl?H$M!6(B$-=AKq8@9cKuVKURa}8YX*!Tepw|CVUHU?Xmlb^Dme%z$&tB;6UblvBO
z9<3^toN~VQ1E=Z=G=Rq)kg&85DpvMThg|v}$Tcq=*E3|MQt(a_=pdlHYC(td!1u|s
z*5!=Wkk>YImjl_#TYdE`e$9yRg%YbZV4{E3tJ8dqptXX4tEg^Lx)Z9RAkmeWex!qG
zx+SUHX`|C<4>zSArh{q5;i!iO8v8^ZEzjK(#nSB9&rxeDDy3(N#6?J=*qc)+0HMbd
z{OB^3k1-}v-2SyW=nqDl;Q@+o_u^Ncsv#3~S~2#`xzZGJm6++hFglk>X*hy(21S4N
z-gSZMzQlUXko40;=HO__J8}|MGZaFtfy6n;zHYGwNbkF$M@_}AO}*vZM?P4#bW$w|
z%<Ys45lzYfPVMW|m|n*8ivOb)gJGr@9#@L5oM$8yYa1?M$~t+3e}*pIS5K)KoWKx~
zMp%&$P}TyuV1z@`)1Y++2~Wt~Z*hNUW)yO<yp&@^CT<W$Wm_#&O3Bp>YV}7bCP^s2
znV|$t;B_I;4m4gg31C02zhD1Vm-XfPj5K9?zvF-xy3LuO#mt&_v$0K=4H1usfVQix
z&cK)%4w3_%#w2<m&@XgNLwtL?M09G5En{pYwga0PUUypqCCfFL{Y9teaXx=v&8M{z
z`$$7dvpVEsRAXCi)li*lVVsJKM7!eft2(%1Elf*AC3{O;p{zRGQZo+|il?4rCS9wQ
zCpO@WV#*~?UChmoRvKD4d&5j$k`ABPQ_i;=+Sn_#Q>{UJ1S#Ag`I-$Q`ZU2<Bl?W!
zGotV9L|?OUShA1pw`#&~%Qb&^n%R)2)wphM7h(?@?a^gteKFaPC-wMYHXIj+qa3(B
z97eaqcSt~--|&=N1_a;nx3rFl=aHk@z{vA%N3MNm`Qvkh7mI*Cx^o=y{xxI5(?lH-
zx(s}IY0A26lSgCYy7N78WyRrmQFkq39iur0`+z}jr!qwspd2G8LEnELxKkGtKElCt
z$%}HOS_%KRmk~KXhgNi4e%FtnEu9^?)ZdAW=>n>~J<0XrdNPp#f#T)v#`zV*G+m|6
z1w=5QMFa#dWP=Exk&T4OOq5Y@BTRZ!iUwkl&R`Rg2%2ZcO+ey2A44AR#cgAn5uvb+
zYcP|5GyqSmepQlyY#M*CYYXZ-p(ecn;?isKxS%0s%8P8o8@rsV1t!D3-kg28a~)1y
z)MnL*HD%^hv+=Jc|8I+nJ@%J2v#|)~nP2UMwZ)qk@S0@w5!v+td#TlE?8$l441!0y
z`o4nj`+a!(`P1j0D|PbcIvhOH34Q9HBy>b0F^t4`9fenYSnPlCINI2UfKH|01&Oz=
zz4RhbcmtT!Af>2fSvr91qS>evJpR=(r~V$NZgn*|cTi>-ZI%eXX6C|gH90Z4@X5oe
zR4#n0)f%+g3U21&(F%!8eTPRZi}{Fd6-E$#-3_^l?c!i1?asKyWn+r&i^?=FTUVoj
zHs|n3G<K|p%UOTfvtKLQloypmj1lwT464I*9<f9a2Z)HHLAdh5Lf=BN3Bv{?S_VG)
zh`!!_8zC@GA;$ctVpw(C{ccyu0U-d0oyF@%4C)?mQm9z>+@Y`jap$-#m@PKL_aX;D
zY-OK(P_~2^hat7yxZU!WB(%_{8`+4!-?^FE@o8xdx7&a1-ZV|MXU@bwaX7jVLKV^o
zH9K<4^=)tI5bF;T&~wOMc9|k!p=+u|(xA4`5V1B9$dVvHji2&116#T7_s&||{YrJK
z1}L#X1!0x2Z+KyM<-wHY(SN3VHr*Hs<g<1jD2D#*fGzSE1=yN1<IVNgvw0T*UC;os
zWl=qLc{G0zkI5e;Cx|gB;_oBzez=}~gdInlDIL}8Z7#<4t{byV0y;;HV*m=bOV<5x
z;U9b3yA4L|PD@EA%A9c@cj38w?0BUBm_?j%cn^=mx}mCf0LnjF|3oK-p}q3r|Iz&C
z5&B<Q1aJiM1Xvc_0rn+oh>S90lo^IDfjgaTYcqeYL*_8uFN>RwmppM|Uvmd_r61pL
zwWQvU(zJ0P7SObDAijpAEha$C&OBlq(!yW?p2y_~n?}66vXVW3_yarxyF?)`oPul*
zpdkDT8$_$k{=xj_u*n^Y$NAn>bD&Wdj$^b055#6xWU%x|2qK(G>~(3Jo&bXo`GSUS
z<lcYF&|sx_2WJSjNA0@@m+ds55eEO!Ico4bGa9μpX#7!AZFm~aIVieuC~6IYfC
zwta#e`rz}S${Iu3uM}{}6mQU0RNs^UjV4>Zu<4QsyFgQjU6doblO3u%$s@beL7bNE
z50r0FoBlv%GR`X?{^}5ik}3#?%1$8Ffd+pxJ@JdwaMvH6h^h-p9m@DqT{dtyvrg69
z06+YS#cee>^@AH&s?F-3Hl5ViU^xd2rzKaTH8$&FgPG~1>Tagu(@Aw8kV+kTcCb&u
zV?U_TuL_=j(MVUL^DD9N`syI(w<@@HGptn=NF?X7R*X%(<giv_d*&QJoR(aT*4Tfd
zi|tu*GOG;XS91)jijb)kAo8Zn>9YR9k#br>r0o1S1xaJiusA&lWZzI!Z8icWuB#_c
zK-CgxPZ@WrARQSIV?>M*F$X1LiieMCvy*HYvv<;{;$SmlMPI{hpjkI&J=_Mx6lR8p
zs<_S6@KDJ;HHD)Y9;yx~%<xb>xt@Q1(-C!*I%asNvWA2d>6I8BS}~=}3=h?afXwhv
zwZUZA*PFAi$`23awZmq3==;LY3=e&$gEzxNjl_5>g;#x8?D067;h_qWj~O0nPW=T=
zy@JC-GlQ)8!$VDw)#PTshI(Trg_=pBD&eE1XKyBj?!d}S3Vmf_Fq1;xfPH_ONulK?
zg=VU=`HlXKI%^vJzXQiKlR^zNN-X?IlR_0w_}16;p0Hea330Um@VtSarlS9;V%B(#
z-kh<k+kM8;;BX_l(DZn7v=}qGwt5}#e|U7Qf>5gPoLVDOj7-t>cr_lbTs^+8&2Z%k
z9=({jY*@Kr<yAuK?AhT45O04D2u(_y^q!=YTKuW;@EH%Ep|XP$;bR&_v||3Cw3E+u
z$Q;H&8%#fJOs}-EipEP`m!XNiT0yR3RMyvB_>ZWpCikmAWi`29ui^14CNs^>JYu58
z!VoJxkK0CUM)U9vSnL7BAK;bp73l}87m)Otvjy>VjzZxvT0$J2BTat@?Q(;qM?%op
znOd`MPXsjp3mUqSdoOdJD+N4GHnvCYy9bx;G@ucNe1vn<;CE&;aQ|S-$KEj-h)Xcq
z8d5NiQS(e(SuWW233BLzk4R)|(ar(D>T3$g3o^wUbQC>8Y`SFpF(?PglO0f=<iT>k
z;xoEwU0$`f6rWzR+0B3U)ukWjijTa+zkCRak31ap5Gzmd32?}8E0ug5aV{xtpgDEO
zPmm0AGyLVt=$K&-yUo*K2)1WBu!B9*c$9_M#gx$b_XTCjkvZiz7oC9N0Dw)AAPg>3
z9}wC5k)@e?D!ViT5@S;L#EzKo!ye~Ts6KY5U`Ob=O^<QhTjGB^Bp}XjcuFn<g75fS
zc~EsG2r^1N3aTzejvFpKywao;nxW)Q)ui4`1IqB(rjB^>D3lT#acRV*5tl|>sv<7u
zI0{^QBQC=^_B{5pu^=HhTp1~%h7>vG;WrH+;6^CIcO*oSJKy6#(+w0%r<xL<1Te}3
zT#1Y<NRSloq85Jy3}q_-m&}I!xgm_5=LJ0Yp1KaBFGM`gU4-zDfbd`Og&!yOcmvoI
z9}lS~`!euxQwzpXeR~C!MPrZ5(SsY!*fVp2e*X5!LciXVAaEVxLkD(U882^09(M$-
z6fQ|DiDvA|jC#AWMM27rkvX;0Q{yN+U`L@~bL%QcpW%N8(<o%%E>YG9gghhebQYmg
z#;F{}ssw|StZo|M#f-h^kbBoAf>qYRN$vV!gngSAP~N*x0?F7JyzwVTk7nRDoik1y
z2_fKDOdc{g&}Fn(uE!L?a3Z=OesVUEg}#C?6XYJFMu?hdOl-^`4G@j*-XeHAL>Ptt
zS%S%sMht%oT|y4hjJ>{=tTW8VBa8_jH}U{;EJUzRv?h}n#ju5amn{b*>ZZ|x^lE?n
zDE!%mvl-47bGEf;KHmg|(}@XN{LORc?wSyo?iP1r#AC`{ePCE@96qnPaJ?Kp!g#2d
zt;+pA>YskTu~8{-q#bY}GY}aE;vm|JljzY^V(ovesus;i*PEPJDHW0!YQnjI{gq%B
zpf?#vy%VJ5kTk>I6xq;AJKA&w?5EAtoMQwUsPz2pYuu`?CYQkoQ(HOBvdntJc5~E&
z6(#nVUr3sM+!T_kkLOxs<?}T5(EVdi<U_Ng=aU7<#<M5+X5JoQhn=f->#C*A4kv7F
zI^cg7pjLeR_2!XiHYOYmA%d7Kzu=K6fw7@*q24;;wG;6W&b3QTJX9eVqB_L1=Ou>G
zj9T5B!DvfX$SFbb#iSQ6_4q-`3Uvp8e?XBNLDi5w!vVHIX$YD50S@3qY(YF^FyjC*
z!&=Ma95Ke9LR<oH_R6`4L6yTo6(J3+a7KS064ei1*gxg1pU`I-A)B%l@swN$AK5q}
z)0FKgtgtw;B_dTGOfp<;Qb&9$KY5JEqfHi&TE;#mS0JLJ$#2HiorOMu1jh_b7sPfa
z43BNhex)v!vHtbMR^Z}#h)wDT65l;KMqnh*7tu;SgSc#kXKc21wmyjA!N0`JNw$Br
z$<h<CHa=(yxGiV1A<z_gWNg?IVOm~o$A9sdJ<mSHj8nL=5>rNEMTG`zHP1Ej+O6Jb
z&@eQoE6s8Acygy??j#_dD3233E%|wj!`#V@>Qtz`0na{=k3dG}(dKp8)gtv3c*H{g
zmkWv}pV%D=;uBB$4x=s^TwYUDxbS~m2=k-XF~e_fvaVEykMTXEuFCz(A8E)RVeBEG
zbiVbJ2DZs@o-Ls`CgxZZB4YdCdLDPFK@RL1lmVGC1hvrbfy)>=Pfuew<ob;9x%cbz
zz$6}W<{d8C7lCIV$5qGT`*ZB$sSE%B=mYYfB_H3H#(_8&cQZgffsk9M%V~c^ER(bC
zdl+sM)05!U#|(cL?6X`~7DU2}?4_NUfp?iax!dX7G`m|odFx_=C#UZmW;+aV1iDbd
zno;{qnCIJB5=Vv34PmRqTZOJ7DW%mrj*G2sFuDU$Fe(s3LSxhuC~4JrC)q=P2+2TW
zl6?N`MSnXHP!ppFaD<z102qIHZWv{}v7Q^j9{?KNrd!7U*;THfbF&wR;<Lvhl{}fa
z*c7Ug2&<)UKzV|Ai}!>00tx6bLx3MnJiWFb#3lg`MUGx6)&FMDxgKq_{C(zj`f;QB
zD@DQ|n0*_$HhAsUruMqd0Ssq`C-MovB!*J;_@AD8M?_YH>_EUDN49^5!;l%oF`f$b
zO1vyOy8uSQypA>d(4eD!Hj@xD0W?*%F8~G~QSGl%0X{$HF#0Y5SU;+*m0s>gSVJ}m
zQNt|`i(5<s*Gcvy3Gf4B1U^~Ew8q%Jn+9|V`X}UqM<C|F?I!NhN7+Cp%ZV3rmdqq@
zjo1RJ!!e<76-H!^dAENb56Z)MzDtRqSijY7^v_C@O!{$7GV+ok85IPq<7Aj>3?=9T
zcN|bI4CfSTETk)9LU6QX=z{8^;FMDsS1sr8=Y-)~B-A_*z?&ipQ0Lwsy;`jvf83zg
zJMkYa@^gM4-hTe{`Da~~r*oe40WwVzG&l3Yh{-TvwblMwI~IQ_A6w5%T;C1Boe2*y
zC5xwqw2wp|lDJlDZ8b8e+C)~{)z+F?%|fF`JYRs`Ui)jzH_p#(N(1Nn#B&$*Bb2x<
zG`sNwQ|b(VI4tq=Nw@%Z1Q1ibA`Eja;qC^rIHuI(L(-eU4K}Hpt&KK1;WdWS>30p7
zO=x3KM0I60jF^9UT>>v)I|6T$D}_GiA{ag30HP32rfoyUx{&dIGch52Flpb3@DSo9
z7aE}p1=?ajZ7fNIuTNwT0MRqyV8fQ+JszS82zhivG4mGMnfln<D5CL4jMR7ubW;zr
z@)Ss^ot+EWfg$obL3I3Y0=VOHqn+RS&yq%j=b8Xk0;Yd?+}3ACjgLqWK;r-*fVeMQ
z_@hIz{uES}^HwQ|vUMg>B)?{SFGh5z7WW-O9MRilttauQt;p~ZRlyv8H<1aS+2CwJ
zaR#DC$Lu!o@7;j<LXPp3Xi#)g(1q|qDuZh&Eus`PT#MZ@4>Y&~3Qz8wCn<9u;9nA|
z;&jU6hOvK}_vkO8Ye3+JcWB0b8>9~dFHag02~E&oh({W~(sgSbmv{OWyej+q#j<$V
zZ4LVa6<P6=MD?A@Yz5aT;iu~{GvkTlD#IS@XHK;v|32+IQ}f)9{<2?I8lQNTRI`9Y
z%fL@~#c%lqA#ixg`-0hCbHZd(_>)17LqrZsh`)av91-y}p@_r+TXyq-$o)^LJ~`sO
zWa<Tvl~@~<4SNrG1v525W+f)IDh;@&{)YNmGmR(4OJ&T!;LTugv>_Bpcqw`j$qhW=
z1*UdDFX-hM@}1*kP4>)!1QQy}A>aNBA?Q!?Gm<!OM_<3)e&+e40O>LD=mQUy^C*r(
zBJ6*9B~0e)n|`-@Q<~GKA2*y{O-`=?flnhm-e(!89C^tXqUs-hUT<L`{FyB+fCCPT
zX+Xz&(q47T<lm#*Xh1tzSf)2jGE_-j$tdX|6qeUIj4B()^TyGlP=k{fFXZ4&1$jeC
zU3sF=Q8XYQoh45(PkSMq&{+Rf(h0YW{@;IO-fZ0YV>7=EfiphCwimXFd>nfTPZEEF
z+4$6Ud7iZl0gI5zx<@mF`Ixj_+cpLP4+{YSE38G}`YAMoe<u(|;3Slc$uJptBsPyI
z|Adkw2IU)QPQnmRVQecyHJ=xf^Pd3G3$rmG3>Zcbm>WZ7unk3<py0{!UykMM-=Kf~
zrR!97exS~<Hw5nKx#AnL=Xidg8X|^J*w1w71s}|i>O?IvStNOJi@tQf=_htIGjyRe
zU3l);T3{%4-${(xs%*#Okb29A@l?F2>s^-xGi~5bU@GM2aG`%>#i_X2p=u4ZG!ad|
zg;NTkS~+eCKcSwp-75Jdh~?=m0GoepE{#kJz_%cl+tL5|_UU$XjeZ+m{qNz=qb~{|
zfawAGh6o^~?0&E}dj0FcU<>T|%47O*4t9CzV8Cul3tqv$9k|ZuF$(ZA5w_9jj+z_h
zWe_&wa1;bPbjO%O;B^~CQvg?X`7X4Ffz-$eWX|R7L2)1y`^b`+%O&a{uls+!>wbTW
zIJ;-9ZacHbSo-n9>}MkmM?viVfI=I4L<QTysE+TgLu}@KfnF*6NKSws{N3>zz7_)+
zw<lHY;of?|Ym<Y2bpHP2TfMR)l`X$3>;}{T{lLt9W<Mz)r;U&AJ$!sO-#%Y`{q*bS
zFX-m>*T18yZ(m2>fBE$F>o0$!@3-IMoBOe&-|P*$8{Iuw`IowLAPSp$FfZ%hNu#g?
zo`NVW^Q(FD?C(5azP%F#7hNx51si*)N_<wL>4Mq9QA7;<4;B6}9rrt!jn77b=4I<@
zv~gV<6v#LnWl;d~1`Om65%Tto7|<Uk=y_YrXMo;7Stls-IEPW}jNN~n1nXtDdEIKN
zh_ryRJe~G?<l1*SWG5;?TgEy0rp`&It$A^%_gk&2#yT<0aMbq$Fox^^-YQ&_*-!Gg
z={_i{x4WE}YGzjgqca>9U)zj?*;K5${)FZ<dI1zE5rJ<u8<(B!oW}gLsr2Iq8H`mq
zZR$@j)JODTbbU8+vB!VM{HU*|<qh+L#-s7%b3pb0MkG*Qv@K1vF_L{FA1#kgVJzv2
z|7C0D6sq~Oc8nh3fjpU^I*eDFP1)32&i$Anh}lx`Uo?TTPi<ii`CSZnEr1gHv)O<~
zwD%<7^LBlK;>V2R!8BbM1M01c<p1Ehec4sAu1w4~YW?e;k2rsU-8LKl`jh-uMB`e>
z6fujt?=SgHip`sH^?=M7DmYd^A1r@-zNHGtyNKNlVO07U3ceYoY<NKkGkL{qYxd6P
z;gKa~3No(A$JRc?-!PbUxR@0d;?-;6Q?u1<Uv28Wq}MimF{yv1?&yY3Okz{O4*sh*
z{Xwt8Nh5+W3Riz{?_&+yHk0kW7<?~)fAYvQ-=4%*(IX?#6@yO+gT!x#PY9!d<N7!%
zBO_lO^Um6QUb4ZwN?;xg*E%;?X0Ytluq-cFw6a2G3iyr%F&Asyfb2Yph<qk6e+pVK
z@Zb!pQ{P*GOiq8mAE)f%6bdH}7ml_)?2ZOkRUlxW`X_(>()03mdl0AuHYP`hZQqFF
zP9}u6znKJd&fO#>!e)lG8P@iqtnHqAH_oM(BZH+$GlM+ZWC4L6!`TdHdo#{f5ERkc
zD6Mo*q+kwG9#8ty0{g@hM&>;ZT;5?B22zF4YQb60Ip>ctEi2*q5SNbfTf=gfSfknN
zU#fRhC8vL^sv8<PN1unhouHV63iGoOV>Mhy=A)D2{(sBGG{7*EjH3It)wUY@If8L}
zK9wUV?e@fvplAGl_P&L=Z5&JYuV9qfDfwQ9Em2RWE2+|hN_^L`b8V-0x4L^%v;<4s
zNTime6T9#1f4=}oNu*3tltfYhc)Dhsn4%p3-*<l=I1hkj4BzU{hIc;di}zQWX55<)
z#}o>|-Iym%ITAw9!$dtoY#%G?rRm1HORsCO=;5W-l8lEQ1wPkbLV<7pM@2XiUHluh
zUmgD%JolA<0Al>u<nw9;MJplEpTpU^k<NRqRUft@TVcR?UoDO(#Cfy2Vd>r0ZD{UH
z_s)NlCJx}*RktxaZW>i-wvzp|>i_r|-8DWfC-kmk_UY>4roaWHilPLPoAK8~`FCO5
zP&?padXZP0tQH%6$S0a07-6<=<O<v<L$RY*3cCi~-|41ly-@W-uc;+3E_BmaKFz(7
zGnN%6ud^RxMDkvGb;(PlR?7SOE^s0X-9RD0^(m(i<aa_GC@bN$HK`43^0d2Zvs+hr
z6zTOT!0+DClg~dIf9kD9YbcPMvwC}dGGP@4YzxZnJzF2NNxvl<?Xk63l_9KrD*UW3
zmUHZ!i3)1WillV?`Iooo`6*~XM#t{$a4lhPwDaXrR1nU0naxI{Q{!^sbwoFeOh#mJ
ziMd_u)VHKY)jn{Dg}Li)RMC(~A3YENor#6Lo;l{+%LUXPe^Hx3YdBKr;9h4oR}X_x
zo6zfUG6YwQeuK1s?|uGqmF7mOPP5S%>~S_k=ykxIGu($-|Lez(+u6?az;9hGauo1z
zWI}<!JJ?mo)^$>?pP^q7*~!+?Kj!YQZ&@pw!dls?-xugTvQ5igRO@6m=crEp-#9x*
zG$j!85!nVme=Y0uL%ghxtCe28a*irJwL(5?)oS4i!@!C$%3L&S-EEWNKjd%`&;6S&
zot`t&$F5~NWZJKe9}$NY(Jp~c57nbbB0E~~?C?Kg>`*uC_{iAtk+CC*8MS_0ZMJwM
zLD@OYH#(HOYfVS@;wxn*sEB^}`}f7~>bu6DzOugVf06lk$6LH-`={y^{TrR(Il7{P
zMn#uS`Z@GM<@=)Izd-*=Uw8?>K!tey3j2n>G}c`_4Zr9q{PnAzgjbU2(o@^Dj^_sO
zMwg^vU#Oi{AS3Mu{=A3$%_U}k=R>HB=epx4mOUZ=;CnBU`|FNSn_zZpj;Gk1D6~3#
zGefTfe<R4>=DU_p-Ojr%rT7$E)UX8^PM&xNjc-`Kc|?CEk?j{Xkr+Z}rZ_HMDiedQ
zRAu&~A{#aI`EsU*Fs>;p7LdPDL?;$88pdCgI*lCwcRTH#Ce9s_Y-TfkNvEdPcae9{
zxVYH%Ubt;@h@bES({w9D3C2Fe{_~YukWvpCe<lu2e0y@D%p;koonc#RYg<J1AM!Ut
z4;1%>31P*t=Efw^W8!uqv0g@^QqfvnKHH?8mASDHRX_-pl2)DHO&Zm@!Y;bNRM#H_
zrh4UHKTBn%lbmesnPj~xa1a^`N-xu*h&-=PPk4{&wH*+ts(PU|_=G5M7KBt#`Oq+^
ze`s<@!L(9US?KoEFsaAOKCzAIEb=2UCGI84O%Yw<#?|^cQrYHPSX5qHRMAnX++Tjy
zP}MmT*FLe6Po%>o2@8H*eKUM~fapL_p?C?n2sErDriTzZElNTULG_TbmQh~7i}GN(
zuC<7E9t?9lf0|(BFKk=LONQ=%xz(x-f0|szof6N3_lGSu+6SW*o;JS_k}V0LsiK?H
z_;PKQmgqJ610K&(inVFsvxY>Gso1#C{<6j|=~8k|G>{~cA8M^6f@W6|&aXQaWjaiJ
z{3KkR`Xa93&ROz@x{&M}PsIFpx;t#vKx<y~*a@&AtjVet!<kt>rGhn`gP}GWf6d0W
zuIC=CpT@+UC2EwO+R?V_p~M5wD+WlKXtS+q!|i+E_{C4R$sTT3E1!5cv0h?YpJf}9
zz_+&U(CaRfbDvtbJ!r8rV1z($cMZ8{@Ffdaty%9i8)9n`kB5p)TN?e0M56N4n)RVD
z%5yb4cShzrv*zK;i-c$Ay70@te=9WVJiRUaw)&n#mCEqZ4I8bRH1)=?qcvEkR$%J<
z9)RFaF~y8H9DMSDN#dcg%eo7R(w@Nu-gUjN3jB?YKB3;0uOI2bvWj=4)%u0G@t`bq
zcjmtpn)NTxN4m%2)7A1vBMMjE_{~RNH`1n_KF}se3+?Og+UJO-VXEYBe`o)x`i+1r
z>tE@y2GRAGe&7pqi*k(pncp$7g*Aitf2ls+KH%&@5Oc39IPkR}{tC!*j(j5y`NlT#
z4Ig>>6BPO;1HHw*w%OMX1D!qZw*Y$KHWFv7V~QGrzhBk=uK(V6-JVK6U@Z8(74>1m
z`>^qQGxTW^6KET$wz+dif5?UNjHj@<jt}9cVs;Twbp`j?@AW@@3laK2^*^BMVAYWw
z?mgcAJ!;(VVIR=nQ_&N<lUHq<^q&j<%SrSqd$%w5`HASpUjN&Dn|N0|VvlCDJ1f}?
z`AxmyB@?ZR5c-xmPU0lDDEoYU-K-pYumzp2->ED!YoQsN{|RT=e~itsnF{w^CjlGT
zyeyhrrkJ28bAeWKS<Hdol@i8^<Neg`#xC)4tK(|Dd5)SX8@mL8YFD?HuxW6)VA|FE
zB_2!_GN*d&(Eem`piAA7Bp8GqUQLEyK26J_V4R|pSww@^uAQT_oy-D31qM$Dm}EA1
z!ERgG4Bj3+xcpAyf0%aLBrHe$Mt>+!J7dyZBAfpNlgjwc=$FXnN$cUa+KEDO=60|9
zA=jkdek7L?h7LBVJJ=G{l?07m(4i+TL0N`KE~!gGO{jLmx$d~PgpAn6lP~%d_ie0y
zXFnxvBVrPE!@62;b%aqQGBz5<^@n;7lE4$I4e`TjZo!_zfAVNaXrSauDQ-7VCPCM3
zwTJ3<%bJ)<GB9$n5rkII(2O`7JhXxefD*KVLMvz%i%(a}AB`YfiDrJdKXm)@$RrW(
zw(Fx#3^h2E6!l$4R3Tqq*jWWUxD_PUG%XQ6(ElxID=4ky_q!TtcVH5LqfSQ~iL(!+
z7o~cwitIjAf2m-rY?>#N5Noztqn0>97r*;6X-3fI$$641O`E6JwkeTyWKF4}0+)RE
zGc?X}HXFGbXQSu1>ZVOp5#46DA&zaUDLd}47cT-GXk$(SN*&f4!wViUX(aLsny)#V
z4K!av^EEVI^EMY~zLv!tK=bvf@$qietk<CTnt!CAfA?Aln4tGsYzU$EdWEr+z1LDU
z^vY-K@H73^F)XvXq?NH%|DNf}IZ>tm`Btr+qY84)D(qxgxxj?(Y#%!4&Mtlu(4C#4
zvjDoYL+b(E+5Gbvy0f7>TWEEa=qNyUHpc{o?rfgPSklX2q1jg|p@U{$uA~ocONAC$
z9tRICf3gCgw5R&#UDuzaMb<A*T0K0fMYir0y~@)f%Wdt}hd1c)$s7~^&16@Z?G3)j
zIER*xf1D2<E`@Z3ve&%g^**ySSFNf$3cJ#2p}XUmGS=twDt=z-=0urTuJfKL_Or9!
zf3N@f=Q*lW&$e11BU#eAX!UE`Ij}}s?G7WUe-nOPG!@|$sQUkEt=6x#YAZ-G;FP~=
z1#l{pd3ZbgDdU;r6V1CM@zJsu`kegvL>+YJ#ZgwItkQ3mx{Xb@N<UN2$bU(m#HhPE
zMxm&M${z4sy<>ExU9c`1+fK)}-QkMUu{*YHXT>%;HaoU$+h)hMZ@#_nJ>!f!>h~IJ
zzO!7j>M8f`{pe=>5l>N3oOs^;u!`gP0Q%T^t^2jge}m<BTMdz}^~=#{o!3+{_wN!w
zugL5f!{P_tPFrbi7B5HhLq1m>uwW(|nunM5*PKx#+bTMx3clHg+&v5oii7I(z>(XT
zdorEga4MXz_fINPe@chKl%|LN)E!t)cLPCrb-rLo2r3OqD%`|m!6`*3Fc?`7ysxK3
zgMASAAvV5GIV@i;^tNbCt<3{7u-s6fAkFBX@cMWUlA$kM-hwVng3wsP2#7zfs!ya`
z?7Q+1+*sCEhXqIU=z%AK{r}cuKU9Rbh`xK<bw5Nkrb(AdN>Q4;jeg|Txj><z9O~zf
z^n{<>uLJpc?jIj92!%U0;m;QWri6~s^@C9=!URnQ=Dh_<_a=sq*mUVPQ|y$1T6$e&
z1Zbv=ENB_xzv}O|#taX5`%Qt=Xp3%Pm${pS7Wv^-HGQxC)LBIW!&U7Ck}}@uux$6{
zkZ`Z)_h#(1{de)gz>ty!{L%tV5{p)ZWLp0ID?A3sB%SJzBQl8zlIXfcs4U+S^djs2
zoSr{Q`lK~Nfj0bUiDX>NDBfs5)(N3qs?6tXciTj!s{myJXI-h6i>Og3=J=gx-}ZUX
z0wf6R#GL{3;YTh{`#|*Kmuyaar4jGKIOQY=(6G!voXUJ6hIY$^c!NuGn3KmJAD~>Q
z{={5?hluM=JN7MRUv_pHa2I!MsLZnnCP98;j7XvGc(Jw1-r>02FU-The+yZvM6d<A
zva*>PIJ_};*hqhz8fqxdVbYU13rku%<+&vz+-9W;Xnn+*Ju#xbf&`@UFjX5Tbp`*b
z9ZYYoU~^zI{rSVXv}sSKw&%(<x^u(UXpa1=y|c4bEM<Ar%rnb0K;!CSY8kVy0`e4J
zD)N-qx%^|hI;DAy8L>(j2rysJC@I)+-V7)H;nPXDp|_=7FwT4ba(j7s`$~p6|0e&V
zO>FJKE~S}kIj8aL4a_Ii$HgMuc7dM@-Y{Lv*KPkz__SqdjyEu6Wk7I3aN<epYt~Hk
z*Rp96jMcf7*-=<i4^254EnLie4`&^--ZPL@9RlSKI+Vt7)4F;b(8I2ouXrD6g%};r
z8<<SrZ6c3<pegxffQx>AEt=7sgMS-tK<Z}7ZB-={=t|j?D-?szryX$@K@u?>rwVO$
zsQ4Ar&5y1=`?fy*YIp8<%B}@HLw>uO%-0}QsYSHNMKuRajrv(gp0tDJvzeWr6T&M}
z+~_hhy!Gs<(rwNK4854dtsxEct$Cd!xDS*NC)rHXw02tvDZ4eFcfJc=-Y~FFh3|KO
zF(^mhTs$*7&M|-3;d!2vwx&;Z!_h^|9A+sV0YeIn{%5{l?>>G>bkpF8qoBXLoCbDB
zRJ)Cz*JDVg!^80Bc-a5Y>sVDg!oiHTs`R9~#dm?#I3q3*2>#;cD1<jq<Mwt#Co8f^
zlC}9~p`9?l=rC8lbDQ1SCop8K^y%T0@MEd{wX9bDV}iV`HHIHTm%TyRq^)rNHqB8A
z%tp`Hv0}!A_LzL+>FVS%n_&-k^G#4+6#3Z9z;6_S=NY=`fk#t+c(!DPB&S`Fg>!V;
zf!G@|yMs0vcx_@Er$k;z3J#@(fEtXD*Am%dJmO23QM1R=Tl6TY+P<zGE-xLb$7B-6
zvrHRg)E6}yax_Voqn#AbWMh=K&cyqOvC#*Z9(k1WlFQ+X`7<PL#2$h007n&_r#h*h
zLcmnGl^)I#`5QL3q=N1CIUd6`aSF*{#A$Ds0^s)v6#6oMogU4;{o~I1pXWBdns2WC
z%=*x`4!c=}90eJT(`7Y$o-U208hf<iGK>WEByZ1m8Qp{s<l|Fh?J9WEl{qHBdE~mi
zw)jQl>%?H?d;JLoA>rBk;-~Hwy=z9xjnsK=v7YzCF8kSFyqn49%peBaEKFdfj&=X^
z-|Q*S=QXGcpDJPhv|L_I@K@;+p|uWoa`tebX*p+9#XD-^aN7CwhuE%hDu#N9Yo=PU
zh#KC5=1!Ea?on+4RYp7Lu;?CttM+z&&zQuF*Fc$M7(Tk$LY~jsz1QI^<*f#fL~&Kr
zsJ3(97ODgjlgVG5EAjrPU3xpTnaW7dcD;Qdtg-?>PY?_M@{j{r)e@5&R7fTg4r(sc
zZz%-?%yVCVR91~#>828WcUk(7;b{AJJyXY!QL;w)t0kQ{Jk^TN&DRX}xL#*xV%vv7
z`p<<8jb4%A#=?p)8rsbzZT#q!ayMDd;EhVBsBTA(XwnFcZ4WO>UzV`6$+BW6i9!`1
zioF1^vt7am@d57FJrOxV|CSGOw5?YThqIG(h^LDbln<M5#*tfqe|#Aw7QI}@;4aJ<
zyT7ib7syQkN~|_UjCoutfxg^7+xO-G6H+_Yy4c7YXpacXwXUV5AJGeF1il1FRy$y(
zbtR%ek&uE^`McJh;XQ%Kf8<$3t0~LCHcL0=wS<~y2$AOprBxZg-n#BPrA<->S=(hC
zQy%Fue)KIgVf!D3wKcV=7oEA(k)hM5n;V(thuX-CWv3nQMI%bA_z4t5{40H5*7w)5
z+B-(LA}RdQ%ZQ}hC6ZnW#a;@U1qy!|=LX^MSi&TCJLO^-=dB8qmpY6s4bU{;v9c}1
z9_#EoBk*1{n*?5`gLp<-yo#Oud7=FR@v>3N*$pUtMF3k=*esaL%7O?*y#?X`VG<H<
zA*8LHLKZo2kM&t|w1a;7fj&oilwA)a0gFhef(}DnP^W_4VU3*i6fCbQi$Y^?L-PJX
zflvl}3--ItGJxO~auFgl8m$Zn8xhZ2O1a~Xq#$eWXhzPiZE1XRr_@TSAPd4M<-iE5
zf_^~6?>C}rd;L#cHjayqxFLs#E;oHA2@h1G3|2QcJ=Y!JX9>nE)Q*TEp<w!379=e4
z%+Xj;HqHd;$0L3#0C5oltlX-U@_Gre3|6kYilz)UY~Qb69Ur#lSwRsP_ZKoKBEb73
zU3e(BY^O$1SQWrhM`fq#Z!ReW+dJABJ7fP>wP-poD3@IzEQ|so|BYBd9K;0TZ!&BW
zs*sExgqjd;FAYCXV+<zMB2+10FBtR}Y!A!@r!YPMwg_U!wIfg|01bdx^nHpmj2)ft
z8U@ga>XnfL-<RMQYzxp&qf7}bSnM2QJbQ=0f(I}Tm#$g9I-egz_YS)OuA2K^5frL;
zje_GwGX-P)A;Gc|@Rd%qgThH?yf7CusjF2i+i-r1q_e$Ia8|9ZepLGPK<F{sAWhTH
zo_im}Y>pdZ@p!8Sq74XO+6lZ3InC30cuvp#Xg*bl)vll-cLnl0zDECCpd{k_RVvFU
zj&xWlE2p(KgzBT@&QymdOdSIm=>9X%-8+w7;_jzPhH)Bb;JNdEyui%(=0!&iHeDE@
zlU=P;5_7iqY)Olq(FgAv4a|GaLfg116J8c#99Ze|+Ulb7it$?3R2-evbp~fkcdHl|
z*(}UM*^qh#&;^)$y64*|0=Xp2;CdglG;!pSs1hOUdX2aLAs&-E!7Kt77TKmeLYdMQ
zEEoj$9OU^0Q9fcaZBMy{#DQS82OxPfI7`ou^ZBo;LrzFa;tb218I__ol#528*H-+8
zy%fM>oum+`6a~>&>4iTH8Z*J$s)hE0k)!#Z2R{v{{^Lk=pp2L&6P=rz>1zdS;KMNx
z8)*t$t-}-&11!r?npm1dp{J++#$zxN0zU`3*u0r$59SxQ0CHxmo4vPgcy7r49faDl
zg3C@No(e&xug*Mvj@hJ}$>~8U7l@I>5o{6U3oBbNH$bu1pdctez|feftdux&B)s2`
zl-U~C)ZU0C4>QvlZ;m50*b|N)`7LjlMbjd-!SdIO1%6tnpAZ`n^GBB4O{Qh~V1!0=
zJtB0*GluTwu|M6-gC%t$P9?Y0JJ~;l%V%x%O09qf>cIxzmV^hjvTDup0Pc!}W%Rch
zDU9Q1d$vgEmZTcOMwQ2y*L-BQEM<L|Lpm_k46Wuq<YbR#-3LBmH#Gd=o_RRplKN8{
z!$+XmcQ4$osZdC{Z3pIv+MC)7k)?r(oEtSjlr(_&t^_lElom7q-GiYctT&vyJaq-0
zG$Z+d#TX)sbH<`xOj+Zn4I0h4n8I_~ezw-?N-ipJ2a*^KF8;YROenHDR>s~;X&=bP
zDPD@fwo8DYmg(#Nhrd$BUTX!0pzw_eUxxzFL2yjWU*r=1kJ{r0h2NH<ocJYC0mGl#
zBaK6dH@jfXMV!X{W|RVV4%)e<huc@(x6;iaSJAuCCwmO3nW7XSoZl%QrOnDlaA2?*
zG3HQ_1dhaip3xHRzogpdXHqiS+sA>Qhl__FMNpxvJ$VQv>m$8+2#~~tt<?ClAV?<E
zhzs`Afjy_MGa|Tk^o@JF3L!Ga=$<2Y6lWM(kr%rTF(3_bN8Fa&DcL#UMT~~!<u<2o
zPND^hSEq6fv`r10HyyD!mtZ?|An($+{6a18Abh=v7UzNGy>Ed-gJE>4V2i+=k_>3%
z#hp3pgWWC2V^kD>Oa%@%;dutxMSPdyT=Ob-4L^A79;k~Z06u_5!Ir2a=H0wU%a*8#
zpcRls%D>TG+W3m#xBT`F<KAB_;+Z<1n_GHf5Nz%DXKWzuumUC~WnFX5CYEea*va>K
zw48wdDDvI^OfwX>!t~G0-5Bs(BG_98Cvi*zHv*Si)4NogdoyfYp|xi=PWhb1WIJ3a
z4eVS}uDax@OnnCizBLk^`^SA=`cHvZt-VymG+I<gGspu7M<YlT(##oZY1QzW$CO<1
zJomgXI;ql6#TS;h0%2+N&(aCsiq9ryN}$SAA0+&i14I$E7DynbT@8>sfVIC4^4pnU
zQ$*GUGrSwb_E9a&EM}8*DE(o8_&a0E@pkc3z09&J^JJ6|{lFzg?_ZG(gR)KwWMtNS
zSl)-apE*Or*O)b{^3h^I5=dg&Xs(;_d<T3~o*?F{a&zj$dQ4>8O6TA|h;{c{v7EiG
z=!dKGl3XXuXS5<DVm)SO;dUn3t}tAm5p+x@1mEoA^oqBOLy+%qlk~Ntdh;uzSf}8}
zsSdIte<oJj+t2636Ln&s)yi>9Dsfi;&wU&9+cf4%TF*HXl3Bmb6_a5-0DW8)f1e3I
ztlOPx2d&!OMY#N>p{I$3Qa%7+i7t<MD=%R1DpK)5DIk<69xf@@dd?yJnZ#>gp8P$S
zI6C=vSkF8YbHM#e3|qqcE=vC2B%FmS3@OdSGC-P%HT7qy+Wi8Ybn&^7aP}GTE@fiE
zZwlb1Tzg5Qa{sqS0h8E3;99-e-<51A@NP}(;%jAdJBA=7b$Ms<ilX}wMg<2aF)<Jm
zmOD}<Oo9SeR<A!yag}RE(?JO1d5L=;WjvU>sA8qbS>eZU1&i@K3z)_?)8HJ;7U<C+
zhIwP`BGTx5TX?H0c#=zcc<LJ-b-H;Vz5+UhNBFZ9zQb^rZCTtXM=Lx-PJHS^XUG};
zYm+|OWa^eF8E-u-i?EW>_0?bRjWCf1&e|;FOxM=GK3CQc7)hkHMz-qP1G%LWJ4QBZ
z!O^qvdT=S3{Gj&ch$9amgNz4`0O@a9P|=a_92^t7%OoDXllF;R_W4vyg!EnkHo)Le
z&m;t8oIxzSamEuABWTzZ);V1`bj0{apV;xK49prqUjH8{9CM6So=5q)I6*qv=y8WH
z{X*ad7IC95{ff_a{6P%(+kD*Il*_7G(XuQ&0|%s7q4m43cp&#&#@rDv>^WC0BaMQV
zu8rF0A#W1_*=MG5>eDqZk48W*N8mvHdUiZ_;FFKiD6!(u(#9GKjc?K6`E{0E$f+VB
z6SZ5ed|W<?wh+K9U}{1*2@U<%ewyu>5%#=UZ&E_xlT-j8ZiX1&-5Uu9f*D31MGdOg
zN5k+eUVv2wTRsht`dc2^sp2?hPf+>(W#cuR>dcrT6@eM|{kG#c)6%s<0l4V;lvFW7
z(v?od8$)v%>q@^xtanIKyqgs;S0dp;)$uXB@bM7TMD2w!<NRv#ce?g0q*4;sL<vKK
zFRZOk=MH6Zxj@c6@B`*tA|^XI>@6N=^Q%nv-=fVv7a4m?D=FBprP+HkWA#;I>;Zuq
zo0I#fOo(g2x_Q=&?KyI|0??K0C;>OuABKIT(in>KRd@$z{&&KD#XQDywTDjDrlXtM
zuour`T#m29a0@ogf<on!nMiZJx55?Xwr2;|Gcasl=_N~^-<B~GnQw?GR;p<d5uNT`
z_fJADyl{0&-E%T;a*j_^irzyFR;IH2P8hqPLvxWN`7qkJ-E4D_abm3&y%w9b)x||W
zU_T@E+YHLYX6(*&H12O1ItAG9z2wU~Y!AJ<Jt{+UyfPC-U{#0*r|-Sfwh3&i=~N;Y
zU_!e^ip!W<%U*6J-V-@;kA~guVbt&2k+LA&{>*LVPvq7n2lD<103%&=G3V7(>ot7A
zl76&NM(U)&CxmP&fFK2Oxo?wj$uaopS8%haN#Oj|z8xtFBa<qFK^`ZAFXA!6P>kTY
zM$!c#m9kC~ip)`?bSD6&YsN`V5(2{yKGO+eco@{qq`Vp$OQ+IgQY3YCbgho{n5sh^
z^@(4^67cV6fE4i%8t*D8E?CS31yfyq89eIXOvc0TzKXUWX1}SR*-=uUG!{kyltSKw
z_-k*PR!cwl!#(xb{Fn7GVs>QRP6Y|P<gHv@_#!_v((odU1tYqH<z4)|FB2%DK@X93
zf}RuX6q|9!QF*vdBzgyEJr~{Bj{{uCGlje!cgkU3fU5U&eFu7f_f#0#O%zho)TIAS
zXNiQ%YopUNAkV+_TI0}-Jl+kLWbfwf9RGuNj|@f@p!9oX3VHb+Ve+*Q$S!C;IK#!B
zI{Z5w$Cy~f_wmvg8nYo?IYtrtmW|t!(<kGwP#)yzoTW;5UsY{4Ld3QF#1BTf`1c+H
z^mUjp6bR|(8x2LAtM^hMUVH>`&;#<_VDK&|LY!##2TLQ`v~vEwq`?z2>eeDPX;cv$
z1r#G2I!+iB&eRWk2eF4vDntethS%ne%UY>=RvKtce%A1KQ#uFM6(mVD*Pj4J^(Rlp
zIFyRzrVohuF5^}gW000?GwMNc-sNX6y!wpz0>nGeG!J;-d9`R^YnU0uJ~Fw{pUmo1
z4ZPN};klqY7=6)jnY}yMd*Sxy;~8d0PQ|*x;-19pmB7SJ;&|^v9s3GALfQg0gD`@&
zK<uFTNfXPQXiv;4EAXk)8V!Rcpx8mVJB6fdl6Sl=kC0ndu35gY_H62*``T|`+!uD)
z10w_rGOIF2(ka(1FPX`j)-;ujq$?j<G+7u7Q|OEd%f8Gy)p*AHm6qEdB8MJL@^9dX
z-+vxIDVGQ<9iV$uTez!f8@=nW98S>wW7Jb{ar(*IHu*bdM5Ag4^tgqzH!OlqROcG?
zUQJYNs=bhW&lLHKTYWa_53?^>g9O&l_cUQFMz1q-4)~(G{m;DnZtnLyqAT)(?Rp|R
zvLc!(bi@eDg5^0z$Fdw=q^`4!v*u;J>Z${<>+#9W!;eF|AAsBcg0o_a?scV-H_NHI
z2aOAuSt#ck9Pq~md(%p94;(Y|L-a6d4fZ?0Z$<BEMlnyUI=1&7gln)4Q1^K*;N5>u
zm;chO1^AkFc4|%JR{l}g!gZ7@j^GD`hfWx{9?<KpvQV2V%+RDlSjbCy39lsz?ju-C
zgU@B_KSk7dO{_}?=rfnbwzY1JIUgTy!?&HYco2<bq{y_3Y-Lj2@E+i6G0P8vLt-Lq
zM9yKC>W|<-l<HG&$Ogbc`L{p-F%f8%bsUNCls1HWGr>EICr;12G_)!e`O&tn&x!)R
z)y#KiawpXfpUUZa$=WMQtiayBVTO75t^`)?v#-hJRO&oOH(b!d+*UTaR$%%tlca{R
zPZPT|o_ln}NA|NqVU5-p-27}T!2K%3Bj!(2<NaMbam9bM0aI=xqhtmteO9w{327^^
za=03*hb^YbHW_?#s)aT^@&6BQj5oolM+2s=DE-8Y!-Xas)(q=)f;9~m_4ob?f1Xgu
z0BfIERvr@)DGZ1Z{cVi!qhQMv^0U|y<q{PQuz>1JlfzD~qwn9!N`b+Ljmzs7VCH*p
z6Bl5%E4P-S5L*?Q1KoLchIczW`QQ`lC*?xh)+%-%yX~p2*VqnzY76KVKUV)N1#PU;
zPhvAfQyDAP)*tl%Nv;#_{a<EMjK+n#1s?7K5WaO0VoV*PDRzh@VK4>dMAY-Q5^!gL
zH7Wf3nw(Pixf-oS$4ARi7q*>=I8%YsSL73_6)RoeC&u^3B?5!<?VyJS-)|=~yG7!6
zH7&U}I1fi(HJy9D)|qB!Cz(4x3sgIQrz;a`tLI>{q%Y`cf{rg+(C<EZ*d7Xy6rt3)
z*8eG^yqy+mP-oL6*6xEyg>g9nq3sIFm(D=!4)U+T?Vx1c0F=*7&Bsz^dcTZi6EE1^
zgwS|x82;uJ{qfYUOvS1lz3+fu{HoV2&je8MA3qYC%^jk7RGYb@-V1vjw<c<L5!(}g
zLvK23$Lsf8aM*hxHi+MvZRc_*^(~Dwp6CleYYR9o1i(>e&P2K;%wDYmBb(!uT5bcZ
zK)QR9u{{E7$VBmGdJPd)=KE%bpgVI`?O}hsC{?u|WD^)~1qa~nK)Qop7uy}|Ki&3X
zx1bZUhUVh4LgTQ8_8B7q0D%E=qvZc71NvO{JFc0e7h&8ZM7w&>rqPxz8blz#XJOGD
z|HBKA!cTbx?s1o3;RYPZxFh!s%`|}j_t!l)H^ARZ@f8dc!6Og^6Ct==N}-n+B_%h2
z<giWp(n#bR%_r5#FC|LD#!BmYNbdsAw^QZ%A5hu)pVZib88ik*fB%zJ2v4@4Y|Slp
zQ>F5>;n#-PJF0+_m1WdJYbx}!->E>BwH-XDgg15*h=Horbw03VNr@zbPD>qYfQ@s)
zsiC2=D2EqP<3BnSS0kcdZp0dluqxNFTh!<=)M5>7bhUF@^`lNmvycnd5@ZFBP3x9I
zJ>u4IesMX|cu$z|jhX*rdO>RiH{KRr0{2pq!~8M0HM|<u&8hSjC~j#*;G1;<@rC*}
zsb2}-fTv#x{5IrWF4hs;ZL(Hq*0X#iesz;<?;yX>lE|s06cVx%RIIa(QT<gZhNGUe
zO=(tiJvJSC-{REsyzeajQzjO0l6c9}fL)*SylA5GVr(P~%b_~AzM!Y(F<nTkOU)<Q
za$d5D9F04vvaW-T4@X$d+yN3%=lk1*F~)h2N_rsSwWbWD7?lgSP0~wp1qnx0|8Es@
zk^|)=3)DbV(Wy->Yq^?eD!qG}%k|%1Q@*ERbm5+jyXH>3(btCvX=uHA_?KZl>h#wz
z4}Zqk&=4RU5*Fwe&v1YSfMvnjFemhv@JWF;T?NKL!(2%~d;lp!Bq_&6YtpZKh?La;
z5jdbH#=Uv?Ln)Z8g|F>N)iLrtvpuWKmb0a9o631WPHrJ$!#}a=!AOB$@t|q@G@Mis
zZ1uiyA&QEN3H<=41za1N>FhOY*R-j1we0VjZ4uvh@i>gs3!h@!oE>ld8&O#PSmCM<
zd;NbJF2)&vxrODy=3Hx&_e$8`!V{VbyoR<{+Ez<IptR8REyi|o$Dzh%Y)SD7q-g=)
zqow3lr4eyAC`ZsbG-?`~Y$`)!y3HKTdz5e1G3ua8QC|CCHkX`<%O$y6iW8eWR3PAK
zj80USb}A#5z_#peTqS{UHh4xahX0T_Pq8n9KRjN6K_kJ(TA(^vbE+G_WF(^n!ksH&
zQV4x1$bp~`kdze(=vXZO>AT0G?GkNDT4Hf@;5J4Ln(BB6j5fIZ`GYfr)DFSsTwt|W
zUlK!2dd>owH;Vyt^|Y)~+_={1`N5Ik^ojxQs?T#G@3VE9Sqt@_eeRfZTc+0K-{mxA
z6E>Vitp`_26LepKBt&XF1x^DK;QlFcdtE@0Dre10;3SdtXN;~(FBh=AlZ84XP88ep
z$J06csmNjm6LkOPCMTJ-5{;OIf@bs>wrMY}sCxPx*G%fK+nhEVTIqG>O(y_qSlSAW
zoV7p;)D|0LeuN*4G=f@_!X_bemmxx`Awktb8XoePlAGDzF_&Ksh)wv7Kn?r!DvXko
zssUvM6(#jCW1})gsK%`trdV1ua&A}ANpt*aaC67HF01I5ry?8vi5sxar`RMZBcA5e
z6B0Jd1@=lzvE}PCn&YxE8&{E|zsIL+JtHwqpL^GVz74^7TPk&qQ$W(tvF+_RIli87
z1G?u`du^&61Y3u;6)6H4VDYohE#ARS(NvWQYrGZ7N2rNR9B+R^bDC8BgS#+3Tc%}0
zjPOrItN`IQ*MMW|DldHMb=$xXQjdVa0BWrv8<UFm&ngqCtMpP;y0wdoNrtu#@ocJA
z2cl`k!>rO%9#7bi@3+65&wr1CGbvjiRIEq4hq-Rmh$rV)f)>KEK+gJ)Sbd!Qt&8F6
z@Ap10_Kw-hnTJXKxzCT=%g@VAB1!{hqlXxumKvc1#5|6iGwDzp&{6OH^`8#8A2naT
zDsm#^C7`!RH)bR(+bdD!84^e4HW7YMX=jn>b>v2sCF}DHNR>I6_4`Tg`qSYe(9m=S
z9_DpGs^2vQ1}iPcz$;Z_dd)v#L-g8kBX62oR@QpVP#p4nU|m#o7x87T&8nX7{%+rz
zr7<~lH5vmeg+LgZJ!HY6$7wr;i`aJ2RiB{eJ!fp@m@<s6wQ*k0_P%$~vQ&2p4JALh
zDw8jk{D0mklXbXi9O0GjMLS!TO%p%fG+{qpep&*ZUt6!FC*1LaRVyK|d<4{IXi<Ip
z!Vluv4tnhU$c8(&xA03{T&N*8NJvP3Ea&gk0vnaB7CajlSub^0M99294dDd%#|pSc
zNWgLSElsX3Q>sNvxVya<Zr)S!ID9qrS&Q@peq1&a9kd#xO>r&1eq%l9?XgoAB$gqG
zgX031Oaq9ri<5SZ*OoS39?e&`&+X9P_b+1G=46^kZ<!0p^_BHS?AhXy*Xw$;HI{1_
zUDj{#wf9(fsy;3NZc#T}d;dHemsP-~_O`b=p62b%wwEZ@KVv%quPYqB$Bg2a6ZSJ3
z%q;mPf}+(mze-e1Aj;-)B#MJa?CzaW^VNWmCrE9do-Qup@P!=}ga06h7dNRwKLqyU
zKija`=Yjh7ezhl^Q77G^LP!e<B_p<?=?kBwtBaf|UXn&iY&W^@gS*x(wYK*e0cBV9
zL_C!2nrlPC&3dlZoUOq@l*DNuUZlSu^qWOp+fE0m9pBJCgJ8G+%auF!ZogqtB!O9S
z*MA!yXHM49V9K4ZX<yKhr6ZTF?%U#9&%OgqjW+ey0X-7Mf4g>W*DItmD@*7vmi$#5
zNO!ou$*>eF%YU42T2d$c4AiaU?EpC)P#)GJsV!yeea&PIekoe+j9%AZOb^1+&`&oO
zL>>dH-iuM_uP(}e*hG?6NkELL<OZ&|zzXpi&g_LtSc{A!U+LZz*?cfl8s9x!_qlpF
zvk0{ZtdP$QCl<9fCQwM|F&;53dK||#_GA^zE?eCb%GP?t)VvrSS!lj6DVu5zdN^R*
z%~G}*(`k}0JRXfOo2_t~{~ncIue>fY6Fxq>a=tr6_z=;f2a55!ldQH)x&VheDKdc8
z`=i)C5pSGJ-Ss&U8j5l7z1rdKsMj&m24eCLbM9#VlJ<qX{zGlygv_tiRKv21ez0;L
z8kvs!wZFz8mp=#Ew3%gBbuQV%8QS+X-m>Ee#kjLzH`|p~(%5l(mvzSGo)NE%ov`v-
zGxJ}Sh-acC_}uIlJv6P^Uw}NUE`z$ndRYZPnN7Wuyzj|8%YU5bpNXy|m0opMI1R~f
z@+od-X1cw9tE-pRbGDbQFjaGYbs>2!FLON|e|i7+m>BIY=UAZwhnVtji*Plnn_j@s
zRcl0u!dD6PZH<Y$Bp>~5ExU2ZgHajK&blXwM`;*2LhCpJ6{Y*97bwS+E)y3B)Yr->
zfON(8i6FRbeM`jJ??5J+_x`aPi1>qQh(S!1?EvhVOhh+LO>^E^w5f61JPJ&eNnx6V
zMYWIp<y(5$6;Gq5#JFzr8?Miys6Puj!%5Hi2e<a@Ilr3+)b@PvZOdhlfdy(-Zk$KB
z_>kQY(7_UnNWkl1AbZhVK-H@&0y#f+TeO^6PfQkSHay8k1}ygfNOrbrDIfaGc3jqI
z*{Z%UC!tZ+BhK$w^PzO4FvwTH7uW_hE|;$35!4Dwxb?_)mAqdTMgbs7+tDRX!*{J%
zz<F%%p#^GEzSrXZS0>6nxac|nh85=YSwt*{$?O$X+Z@=?cA?1AiXs=zzh)=Vd_uif
znTAs@xq_fswTA=TC3_6aa+TL2GK4*l5+MXm77|a(=z)3p@rTwP_0~}}fFDyMLR3YG
zRv|<)gAn~{?KGYVW!CfAwaUF5XT^s=*F6?m5PI+}<(m(X+yS%qcj)~;DH>!Uh>|-T
zFrcD3AV=pr{%n)g2QdX<!(p>F`U!3c!a-WvI|WH~CK$?3pb((62Z=Ac)0YmvV-7~6
z8{gOXXGaEvMHhX)*HOm`bOXdzlsftYe^OJn^SVk8aiUGvjsq$l`gf?FS<*F8atPYz
z9*hWVMpMn25EOBLIszQ;AeIbNDuOU9(3F{Y6~X)WO|LzPB&7C=eXkP@Km>$N?NG`+
zVlsShq9P}hEq_ZH8apGyDOZ-!(19C|aqtCEv-fL0wWWR~hUaTS_Rc~;Fp0SK4M_pF
z?}FEPonvUY1+xb#e1*9&8ul4&M|#E&!ZpV_OcxxN(^#Q+{&2$)`Dj3>5)mo`7}^sa
zHn<y$j`u85Q{8pVp_Y<u(!wtX!1{5z4UfS!t8(p-&NWUp=La`vv6b_4w*9|A<N5`w
z1VD&1Oo#A-L8ZgBn7+1!Uv!GsymF`7U(MF?A>q}Zb*I^D&~Csgsugt5zO<9T(F@pi
z$XT2BMxq5%8AxZ4&`HIRN_+<Hr#zKNRg90+hwzZ_@AMV!r;vExLu;p!lZWxPY6rk4
zCYT+STY#EkBkmY10}chxCK#NzmI1RFxm~(CqqX)lNl$r~)g!*LESJ^Xw1<UmfWuZ9
zv<=AQ_UqBsCybf8q84~sR`yMczn!)RHxa1D-0bXGAEY5;_qP;*-NL~9iNHGIQ^rz$
zrk4|Tyy)H#?DmAT-DO-2#?wZ{&Gt#Zem-&Y-Ir;70zU0nq<vgMpoCMu!z_M=vfIDh
zGF!PlYq~7CkNnNELl^HGmih3_J`{MVt9<?DWfUr(+69}GS_0hIQC*1UnZb$Vk40x?
z_idZo?npq=K!_l!kRec{+#`3uk&_k6mKl*EEo(ErjI)LQSALp<Fy!Do&3gTgT@TCN
zftY?JMKs{&VtuzXjD#ni9^TUF#mEx+uJ!v?WGmSos@3ZU2f7g!UGZ~6NeqbsWGR|#
z@_+rI#%%uoS3xw8Fg&UD!MMutcjG4CTHBWmb}X>3I%Pm=H;oxUA>51}(QLLUk<9)!
zxh6)%0ohj#4*6UF(#dDV1oiynS7aje?4l2ngYa7~rwJRAQ8Q~u%(I*H{ilMzAZ^We
z6)NA4$o!YI|Mi&==qEKw03WkEa;JDX5!n^~l$l|xL;bGLF&uxQF%w{;&BCh0j?j|;
zA|5CR(BAH9!bW_pH0`XlP+JuX{Qf<-fCWm6xJ`G<w3v^=`T?@*kdJURIBpAN|EJay
z`58fP$n{wd)aOb~m$V(z4}>2N049FK@8cqJ7lC#*AUT3j6oK04q)(Ucw_b9qssVb_
zQ)rV2lk3?(q-U+omwt507G6<vN-p0{0XgtG@NXmF8pw!6{g)GO;{MsOvhaMl{*Zf%
zipsTnPQ<uZUF7s4C^KU<vDC)X!PjuJpjCWsw?&UK&+)=ukHk#nr$U!iEbmc?J9n>o
zi=8>DcB_aj4Xt7v-L@4kjSlI2BmA-XC}8Dw>~yuwBnbS7DzkCE%W(Btd_#Ub2dd^k
z0NK72q!epfO7QX!wK^}`NO{DO+g}6szgmCbD>oiw6p0nDl7E#6E-jYlwmF%Jd=K>=
zjo7W<Pn(p^?Ip3@q9?ns=>?mQZ`r$h06m;OLSO7U*U!e!Vl#*k)aol<iLT7Oh;5|G
zI_|zOAFZYnGLq_cI9i6^Ew9--ft(=CeD;Z(Gr!7fD8Yfs1A|q+q1MUakOOw?Y`Chz
zEK0lKFVdcxVNH%eL*u$>9T54Rx2ZemKwnazo(jplz>4ljp1~uc`qZu#@W>n<bLhm>
z9(=T8_m7h63BFt2%PTed&p1|D3wkGym^#<Jn%XJXEvg0MY^t_d<#W26fuXjI?M&bC
z`7PJQsi4Bu=jUVH*Mlxz{u^H515{2x_aVVLTnz=h2gGgxK;aMs4+KsazaQlhH2*0F
zFROMd)N{M0^4!zy%#Kc89(zT-hZAi_4*wE#UE;oZ(Iaos#%XM{tvIY-FlP;Y#ckIh
zP?t6MR7x{9ZkOlxfZt~xkm*nN>yaLWCH_0~^a<EcjpCs8t&f?Vbdxc#7v(<Qy6$PI
z(E;8zvbMP@!7ZS0CohoAxpfV5HV|dCJTv|84rwNB#y1)fZ$sP<2yAkhxsk!$xYn8y
zI_wP<!5+o}l~lnu-$P_RSBCb}3&yGbH@L6^|0U=akI?pt_>PndVBTHVHOeSmny?S%
zsMji4!V_402Y6{O3H&djQ7?bVE&nwFQV*2TKuN+_@@Q>*q%?T1HpgDX_>MV)Pq28t
zTnO`z(uY78!evyxJ?j0F$7>Hg@36gOnY6BmJ^!98h){mR62BP>lz^~@Ps=zF{9SI+
zlUTA}3v6{eupXh{z>itsDKfB4#J_Na%9!?3OKT`XnRLh@Lgw^22F1prPAjb~GfHgV
zS}rOe4`VmS5u-FHiiu^ev;qmeuDq)QraTC8`v<_>heDJLcqUfUglizSZR_?tKp7U<
zE0+2b*G*$e!1Z7@Jt|tiz*DCW&$0mSqZXD2S3W$BDKcpUWP)N@Xe5av4jCg6As+cN
z&8*A{RwgkR%*CUmd4G(os&&4$tc)y>vSMzwP_Y=QRkwJ~Wrdw3xeC|GC}h0=y)hy;
z^c|_h75Q2xeeE1)=MP>*L=wFuK}WFN%aixfUNHx}S23&7!CM9Rj|H12|G&LdmG-D9
z@Pz177;i^0<;;HPLaySjplYk$n)>TFk%<njctHDWSGH_XJCsd5;xkm9SFks0;kJFM
zE_ys3V<Z0Iu=gb@{PZ`bxYbx)n>OZ8b8h2X0D&T`1O8`ms)s}pmlsm4jLP3SkQBh$
zCqC-2*YS6M>3%XL^-`dG=la%;Zo8P<Q?oIt^RJZUr!|QSHIV(BvU;m-bCyi%Vdw@5
zE8u)U(vVz&_;&;Cgo#8H>Xwi`*Ra1x1w2^0*H3Sh709|0(sqz?1piI25#r-hVf_^F
zOLu>rlfE{i@ppe4>{d2>xLwkIXNE$%zAmp|H>d3n$r>^2;Fly&xYCqF^XDC!;RQX>
za^dPlgJE->wdBtQqd3`bnj|y1dlK{ko?tjW$QR%gS~=`-8>fNUnwv|()~A{Y1Vg%|
zu!jnwf^4x;5bHjv$kaBj6kO(XPFc<`#}7~1!2|{Lcbk5|VX$9dG}FVUlh-8T!_5LD
zlF6wRKa1TZ0qDs$!gSwI0c*MaCcwisM{!s2-2W*jf8$q$NtUtwdc1Rbtj2)?v&sCv
z2wr=L%$uzL75(-qsGl>5B<{FC7r!iDaI7X7d`Z;Jd$)LtU8HU#BusUCcn?E_#b0o%
z8uZEGpj5*k<Y@r3n7Oa-8OtFa8{^d{y|lRclJjO=rc@^0l08~b(i?sxIfJ^^=A}0!
z))7ujdT$81^S7y&?d=IkKXP&sSfBf-*QM6g-y_DJzMK3X9Gfi{9>0Wvxo;uN2Qc8K
z8JP|B;8{K#rWpYY1b&&B<m(uw3w>lP7iV|&=pykd<@6witZ_z!G`Q4|-}eP>6avYN
z8#dbWp6HL$zBXCdl>r{-)ST!KX0CL<f*PV?&{N3yRpGuEn>LoomRxEY;jF{@mr@8<
z?3<$HgPxzr0~M)76@mJrQZc?8wk^}?7q4kgU4=(^Y^eLMg-N#Uosc^_*%pz93S5+J
zf8_<No;-&a8n51)-y;+oR_6n;3Y7Xhx3h-ayhAsSN8dOyxJU~H**1U603(Wxdtraq
z-J7KbKuQ|A7blK8owhcBel+chsh4HSPB>vYai-Rf3YrBD<pAl2a`dJsvpp!Y9RA2w
z@LI*kGks6u8oT39vRbf)w~|(mjL@B}5PiWSud+G-g8VcuDz?tucOUtjdoDf;lQ@$*
zAwhPrZjqLayAZezcP^i93}3t3w3nNMBt%*=v>dO2e#0SeIn$pKlIq64po;0}g;`Qt
zp?RsK)PvHhuYr;>Z%tU*ix~N;gMpJu4XF|lC?XLja)yMkcakqh@y!2fGo*aDpox#*
zJf5I^1U9}fyai79x~v^@HOjqSZZ+*!^^)2l@~u*>StqE&|D!j8L~TB?wi&WET6*2e
zQFT_(m7ER4CyVx?(%D_t?wDP01^Y-%R)-Gvnd}W<nN+S$N|mXwJ+~-*I*RA`hirTQ
zjB(DKc~<5E$?bH;lTxpcFx!;9f&QzC3MRQJibjAk4qa$w!&gkQ$Y=W`=a4R<ker+n
z55~X2-fwXkBgyde`+ccWUgwu$6>HX_HvS-Sc&GM|==^(JWgnC6Gnhk{IhR2_8XL;O
zjqNY+>fbj4yP{PGvuH3$nYDfoiYjU`YDnoVB?vshK15kvlSfSlt)8q$6m4<o9Y^A=
zgx*WVaz;s|FVA73i$BMu`t;?*_G|z5(xj_29F>Y94+h$^Q-0tsO^{nqISU=)m1wEt
zne$qsF4t}El`lacK`xSt0o|PK(LYhH*ydkAQ^+;1m(w|85V4U{c8fy5u~zn6kDtj;
zxwX~WsjeU6&H)$ysY9U*)kH(EgyEXWPtFIB+Ne@(6MvM9FVQf&L1&R#W(cT;Lt2F1
z-=nsa51aUYnRM0$FrJ|>A`#5`A(=q6j@w}b58q5Cr4Vgzu|%_M{348RV&|v91XgG%
z5F-Us!&`SH76>-uT|R15YCZJn*r=S1nZk5{+diG;_hi9MJQjm!Q1&m<+WdwL8Bh}6
zf3da=$$`7(+{j>yVpNQlEAU*B6U&ciYMMh1Tg?EEF)&**<XJ!*t9tzgT&#AJ(RrH!
zO260T!`wn+VuhbY42TC~W5L$!0=29#OhO?M$E9<vhLF?$MyIr$jIIZ4y!oxnb%RTj
z#a+c;eZ1)UAvP<U=3mysz#V#9MF*VQDkSM_Y?M9!q!2!_DaIgw)C+upbv(JS=p1!}
zvFeyRI3O=-Ht;8LCok6Vv{FV!882N^#X@khNrJ|Hyp%90b(T1d(~b>E0-}h|4eA&A
zDujZiLMWbw8j0ucFC*`-CO|eEmJtm(|9=<SUPXi@<Zp<A6K4?&&_@6x+ZH+W0_DUH
z?BQV!%x#YZ=)3<kIYPv{5$;*ts2SbhPv{B74!(n|*u>VDKkyj*2K@Ubb}<&r`<O{i
zl)+N?6Waq&Bu^k^5FXR&a}UDRlthwiZM67@Gg%83MRw>Eol}*(sFi;o$Hu8%(SS<E
z;v#3UUw0=L?p}93CFV~YGwW@LG+D%t)Xk<S2YG`#-pj|#>?^oq<UL<H;b7I`w@iWV
zpqI^+2WBo;x3Ee{wDeMW!!SAQ@W<<b%Eqm@Dk42gJRIO(MLj-np>uH`lO^&)+DQwJ
zV>O#8;mArWGX}AntayJ{n;IP}yCTVsOIL4vjM{4czg!v7=*U(ulEw<X36V>mqpRT%
z4VOpCN&a{Noy9^fsWiziKb2O=nfu+19%dZET4J#qq6bh3w;QBtZxb=|=t^i3RSZ`+
z^LmoR^ap?`F=(X$seh%VYGwCV+`1Vx@c{P({84Bm7E6`OZNKh!$1ay0Y=w%wYk+VQ
z<*y)&>2&pn<vaZwMJEB20{DHCsu(`wACuK0sK17`I#MjBgImZ3*o#WX#CrUh9W$%B
zsTZW)u1sjtzK<6%J}L#L-}lb;Ef$-VBgKToe}KG;W%N6TYsICD)#Z&cXEBL;J{~Lz
zqk>FO{3kbmES#+-j<I5FgF$UtPurOx+W$Xe$kT=it`Ka)50x9C?}i|h=Nc{tGDRAL
z9x}F;-b+*RVmW4{!XidcoVLL6I%d|n#b@)fMH#Oru<FON@B45*^<F6$Mc(w74GafB
zI$Ojvyh{o*{-8H})RBr(C;2JWVlZ4ZD<!q=3E^ggdiiQ}aIq0C5F;+<FYB`3jGV=w
zQ@njP6m&%2FHZyrr7&YKy5+H11<zW*tt<EI#RiR}aP@+W4XQ$7gq4fGCzlIy`YY<k
zUA$xrLtkF3W0b7NRpi_dE%<?i3B(B21k?y@cR#}&|J@^Jb+9YZhsO*p1zl2YO@8t_
z^0CZe3RST?DHeFC*gPXpUhcZv*31qpl@%VgZ(ez^MdH&Kz-oK^?47T|C0?~%IMcC{
z<S$%qJbR#)TslR>&}I&x<J|~{*N^b$y>Y~yvV(1e8F>Z}0h=xDE1qWpfcWM^16Fg~
zX+4(!h}c5ESEzp#48Mcp59^|iV&&6Yy%$y!O56ZRV4+NYYM<&xxF#Wp5p|;Z)|jgd
z#ZLXS<wjtgLX`FwmG7{)f1>g}Y37}V-6&PP$~`No0Y?)afQC~Te~;duClxd7%8<}0
z9H-G*6MmTr!j<|(=6abHz`hQ2MfXtDU!=4$U7!g#*-oCT8~9hWfxXFWKD%<t_et@>
zuOf$u^vf|urxqQq2;gguI=tdW!h}Wvdc06Ozx01n2~OxWZG>Sby*f(q_vui!=n(GH
zc}5+x{QAezl|OtOS`a&uKO9C#h7wz=Q_hvq;fxjy8|vt4gCsPy2^=?)L3MiJ6K_fF
zw)KdrK4vlP0X=5fb2-~cvG590(rzveN$|bW5xAd(RvZW2<4#s@Gj35~D7MGMx#-I<
zuprfVrN&MXjB<W4I3cSt)IAu0NfAY5WS@Sf_2z;cVSS1*FAA^ifrI%uw6uB>ZFo~c
zz3C|Jc5`ibP^RFH2kd-YP5XhX`*a4w1V!n_z-@VckSgI9^PhX_EEIw!Rk|j(lInM2
zRhCMml%)AyE7;>m!~z~<oZx4Y3I*A8A~=6VZLW+QJ9BI)hJ`YYXBJ~Ipr47mKk+h{
zHrt1_+!yo1Yzp5<MaZIurmY0^C$BX++4PMH6CT)6>Gh!+@OiL5vi3xHcW@%1<vOH<
z@K)DnuV>JkGlgx{Ps$_DgpPaFFF(1++`!X-s241Vd`X1urQg$BkltqibAJJ+WXRK;
zeah8|c1a{M)Pc1UgH>!y@Tc{6k1tk;g8!wCgDndR#v*(#x+wWYK)hlQ_O7g$3A`DZ
zq%y8|^)!(fkpH08jB24&M@jTBcB@DS-&z?tXoep%hTh)rc@epDQN+3f4oMTCKOgtf
zRB3w>bflHLOyuVvdGj(@bq`LhMl}I)tjhO_5_<`lRhOwBmGj5;?RpX|fL;d0k;~Nd
zS^42ai(MYZsLecN3zBqlSL|Y)-S|}3^QmG&m^=#$AfbjmgB=P}@nj%?4`J_+Xh5W2
zn^chA5QWKVLXjnpC~%hVL?{AnsefNIf8UsE17|Q9+E%L1bM5>DfUeeW$ZbcY2p&fL
zXHQBm0=vw?xEI~(=6yzse}I{g+q@PF8PS0&@zwTQ)CM*APB0P=d?W97d26%AR>~mc
z;Q(^Lh|Q6usnf}rr?~6?4JlRB^K9n)>7C3(8LtLQDio>i4*v@&J!g##g5<Eg=wnB;
zQ#rpG9~E{@yzvWS9B(b$gF6Cutv4<vm4u*SW1w*(Kg8b9(0i_Y(aVfNQD}x`%cK{}
zoXHryO9M5kP1<&|5pQeWRa&v2{~KtDU^f^~WyC&9V*r`3feL|G6q^o#7=us<W<Ilr
zlYfR;Qg_YhKk?c;6~($9!#fYRXwa{L4&;A<^qR@XS~2)u{d_anqHCAk^!ityUcP2{
z3quH}`U3o`6y^=0NZXcX8_}+0zUrYZe%GZt=GySL@vtXZh?wn5%jwJtI}}i&DF{B0
zEErm%@1CkgcAA*m@V|d6{UwG4BfU{eoedgDf`$-(u<{qC(#oLnY$Pk3bqEJjf!03a
z!lb9YQl;u6c0mFrxwB==m#gEVUbFc$nCsxQ?Ce3QH{;{5-u*g6=D~!8{WhfHd>H@1
z3@)ZJHPN<UV5II3N2W^#0kDTuw5SNJ6l{MQ1!^vZ_Y=REW*1iL)Tf#LbQkr9M)^p9
zLAaOAV7=-dbS??C27*u|uAjJwrnbmWNBPPM7I8?GAQR$ZTm^QHzX~Lh=*%P9$pCMH
zX;%N~y1m`OXV1Lrrc;KZrR$z`UYkd&u2q)*0=jY4t5dylHQq&2pi83<&5rFA*8KyU
zT-YqsrHsXLhUFiX%Bh9U(ny67zKkEQH<t%ZLd=FXPH>s~WXq)bIpjYGiVwv6X=dmo
z35y>$u>kkF^pHoyYVIxjWVD0@Iq;~B^3iM5yYH#4WvLGbU|D8=*#NC*Ajhs0`exgk
z0bRbAp%OB<0|#&{Kre37am>{WyWmut{5s!0u$L@eKW6Yq$({I=Ku7eGk@ZQGP0z}5
zcdb-+!k;*OA5wo=;W<x){<N~iW$yk|zI5nyxw~~yKa+QVi$&dGRmTpskkik|2CQJw
z=d~|})9)lVS8#u0rTQ>L1UG0aiZzrDHkK9)!PNgR7l9F{8e!YB!rD1Vm>mmn-#m<V
z)w;NoCP4ZXU}dveZB%J9i^}p$K$jmhaPPI5YA?^H?GLxsPilqRe+HK}mWrWh<}ZfO
zu6a(T?-wV7=7uz%^`G1(OA)#?{e>1jw&mZK?-qFx%~M0E7%WYlBqb)8&+Zk##)+F^
zB90ycHnj}7F8DeB!zlm@u%-hs(aYnNl)Mz#Ypsoy^GC9q<8D4O(AeKFB>7+GDty|Y
ztE`IThb{GVrWjYT=z-|2D7p5_i#DN$(nWT|o)_z~2V<u$E$vsuIXpZPQ$tC~>KUlp
zORVJAxF6viCLv4J?_TN0(VHwO<f%%OVxZt;*92nJCKv<Z?Vn3TmVeUuH_R}*Nkpq8
z@wKLLtt?&?%|AafSH*&FdISn{h)`_+I;N*La$&B#U%O4=-W=)?@fIk<h#Mo$O1*Mu
zd5@T@bE2j2`2vhJ+D{l>D}@DG%K)<*z)i{Ze2Nus5+jE8Q3kbOUcNCESk-IHG4QJ~
zB^N<G$K+XXWhsW>WY`FS?9)r)J@94rBOIhFr|=!8x4b&55u@mer!GLExK)p-NImD|
zxS><KpxGGE>|Tui`cf}9<g_1Z%PMGLNq@9)Sg^G#Z+N=qXd!2pvnp{PcnJOgGx>HT
zkwqI(NnOvhtRnxtlHs+dmg`@|DzGZ&#WDKrA>kPO+qgGrht%8OEt5JghqSN9|2kpC
zW`dm%WEqU4*)>XHac6Cc<u=F&dM7niTmGTQ<D{HKT#VK;U#2oelKyhLLZS#I>YPz&
zOU1fzxr}Q2o#k4%Zk^ppC{900s+C9#JSI@`VL8?R#MR56)rf-YDS%^01WsubxSUq<
zBZRDf{Om)7xlC*NjpSmKpYea^O~x6A?zIa!F;vQ}BBp85B%184n(S6pV)MoNvi_mp
zd)o<(-Fh!RA$Fl4n^XTpae8^q6gKIzrm!^V$FXciZm*gs%S`%7;Aqlkv(D=Lzy9gM
zs8;6RIM9f5f_82+org((RMWPj25ieivNAhQjT)Uwlp8{Bok1)bsrk(%NGnLBXAq)X
zP`}dqCimhLjH%`9l+G;rRH}YeT4G9+SAR1r@qZKHJ+uEwgkz0{wzZCKu;wL>GCeyz
z)=pEDMW_z(y`~3tGpW-a<l2|c#09!Gfz|srh^P3dY^M|Ks}~!Y$N!oFNoM{@jOUs$
z*x&%gu@Ff`A|#!0Fx-0#xbuC~&eRI}FMJ99`%-(Zd4e&iPxyv3=^x_nvra+|@F>kb
z<nUMJM`pTS9vIR#k?M%<nWOvZ1=9ZySKk~P$Q%AydyB1YyWQHh&8@AiJ+<v_-K}lg
z*4EwHwsrm9T`s>|CYj7XlT0%2%=<haJ#lb#T^Xbl|FUCTNjR#VY?EEQmp`o<S4#o*
z@^X9`h*l~Y#2SU?vrDXfsbgkHGBJI#B26bh)}U73u!#I9UX*R`n&H9Mo{J9%Lrmz0
z+*`xv7qoNlE2dymsKQa|IXoPxNXj#M)Xa|i)_zXw(|?9Ys5jcL4cY?N-!79hY03B5
zyjaP%*v_^(SC12%uNIG|4i^U7H*JBt_Kja(?H=Nk<=v{exfL~f{MNy&4nO!=SO;Fb
z2Br?DXy7}1rKe0OThFN+wB?7;<DVI9c0(J+?`0KP11PYaxZ{J1IHlBFCFv)3c%{1Q
z^>D@P-8c_6rTge-+XbtTTFM}&RJ!n#+8AOTF+=Cb8)o%ueDOiL2M`x=`cl9T$E5V@
zfb7cOrPXq_n3mne29oUp)a}IQ?bdzcLu-oRM_COxSxrx>@|VJ+6|{2qvL*Ge|I%=~
zYl$4OXV&-&gD5R4=<i?gi>uLx;sWQ5`7hqm){=$$kAG0az<e$;AuGuLwqi^zPZQ&A
zWKS-aScIh}IOc(O2@Z`PsK*4>qCGoLCqY)6@->jM&@HQ$WU%X|ox@z=h`nf~yVItk
z<<fj>Gf-3u^=#eOmLILz_V*@YQT|msiPsQHO~4Q_7aJ&C2B<c}i@T2_DSoxV`EnGs
z9_l6!&gbO+colUYoHhCt6-kqOo=)-)j-~4L<)Cd~$eBv<#2<=!M7scxH5&YP)GE2&
zi&g<PdpJr;SN7TY95ZNzRmai{C=Ugi&&LC^YrxLVw-T?`skdC%v5;!g_<|l=;Ar>K
z<T;nVGCzDYh%SXfP3lhIcPm+Q_1@CX0(qE>XF$n#OGno{Va~y@@fe6Y<Yr=%=E(xD
znmJ8}u2N`!IdLl%G7a#@ZF<OE>K%zEK-q&b6Ur!{1VO<2xd2go?{DfUN-F_%KgzQw
z^NXY%P4|Z4Z>l`Jia>MYmumAYTn=8{AAhqkxK(rt9Np6tJp=i!l%*JnQZ(rf6`-4(
z)$O8icRLS1ms!V9t^JFl2ya&J<a6ln0+UVOR9L@_@g<|H%mP5*Si;<?UKB$W6vvQ;
zDQ63o_wpObYw+&%+e&W4uJ1FF@}F*RJ;TkOu7j-i>Z@u#6PwaWv*9Gl*RlP^-$~Th
zl0tx(Hn1-sTLI%qFIW=aW6jml4g8)=rTmJ2iFMkIduF=a7(M9QQX)xk6ulIMx-3wz
zR7yUDcaePg2?gkx5hrAGD?e|x_+E1|u9x>p=FM-a+9cmytjh(@*teECtPcPMIgg=t
zVvY;Nn;Y+v{N7LA9&FrNpSalVyd~fLM!&c5S`7R_$yv~z4a4jUJ2ZRU97}W`5p$R;
z5MvUZx-1%UzRp$9w1bu-&yy#=%ifzfzhi{LTq2Y@1b`9&=QIy1=wM8&rPQ@)l#Rlb
zdW$IyPJjB;cJ^ly+6N-~^`~<YN?<T?C*UihR&9-kT(?I1=zKimO5Jy=3f};`En_)l
zwZvs#ap)L6uaj$w(@^|d1M$Au`aES$qgSW#$*S9j&RNmmAF~<rj`HL|oU^J)<v)mg
zYI|-;UO>tSZNQZz3EMES*+aLlz1LkJV`^$x>qJ_yyFa<yK;bC4f3hP1QFDR!kGFi1
z0~X^cQXclzX{dH)uNuwYk;N;{9#_H}_4OZNSA<$u(_PVhAG<ocgCBQ8&`#?RFzEU>
z4qa5Pmm8Vg?p^!;rgCa6_7A`?G$J;Af3K?ShXFNs)bHuIeRusXa`C%`1--xaaGqNZ
zHZtX19lrsvtx&CIgKSB9P3c2b>i;^F`n*G*p4m@SCY`^QhC`31&=TVL56tKGho?KF
zx@5x8sm0&do(2owo>?2FpDofO5yhJnCh0R!5Ys^qw!7W-KDT=t{Udbldj)GrEpe%G
z0s*)bXMY&Q5fycaJFImE+A@%X(HgjX+(3BD=L&@RaqBrXsK9cR<C4Ac##K^GpPOT=
zl7v|aQ^6`Inqz9x;JeT<%4DC-qd#NNaPy1geEFq+aCdJD%XG_4N__G4qp%8Trs=<c
zKFmNQQS7@TiGIewOk4a4dTB`g3VLCkc0lrDmxE-6Xc=C9Qw}tU`M!AWPDpS$w)mc+
zPKN!guk}w;Og}HPz1?+knRnZ4BzV>b249V=rzYJ!CWg!Q<!AbLw|AGc76+Ua8Hk6Q
zSX7%-`Ia<;duRmE?Pr0@ms}aP+ae6;6r0cJvFeAGC)~|@$Rh(t4{Nsa>`fF}3P94T
zgrGpcC+2z9h&o{GV8(zRmKPd|C2(Tm3wP`1lLN{9)1H8RbSm>wD!MbPL*45UU)cP8
zNcX5xa!UWG60QE|+CQh;XbNCJjC~q$xRm&UW<k05VzLoDS(+&OV-TAu`;kyhpwiPp
z`O;ZDSqeinPCV64^0^0JI<uggfm5_iWLtKf&K|kJpG8BAP1k+a${;$lz=ziBDdX4k
z@75;&o^Y8wW;YlQ=15pSXpR(zSUbLT<JrHyij&2BV<&um&c#jfH|v%W@cWgH8QpM(
z7UrCsm~VOIl4`YR@)D9<X`4wUu4t-uZ;|TS=Go}d6*Y3Wt#+vVKH&QfAep-7%@DFz
zR*wojp4^NVgJM(V%v(RikUwxhy)vA#H8y_?TE8ru@aSKexp+Ap1y<VGd$)5e6jcEu
zw`>-mCgz||y9s)b^kr_Lp3nm@n(C3;APCSl{W|?+$3}4jAAF9ZCdEHh)_gp+3eEmz
z`VW4xN$WRi>Dm5-YguatAbsAWP&ejS*G`{tZ^<7&jy<|e1gv$^d0)n+lU(|=Dp!lg
z!N-2=A3qOk2C&u0{%QZ&u2k)&L!kDZcYNX9le)EU120aO`R^sRF%yvePtLr))SOqG
zTC1nA{z{AJKg{cId!vlSHCSv1eif=<InuMx_tNjoP8V@iq&%v?1&lTNsR0Jh+yoay
zvP62)aN7izXmBnV%D=wlNRg(60VJP}%nWGs17Fn`n@i*54rgYGE)%mE-FJG!%xV0}
zW!+dBMP&3A+!f#E*-OXh-Q1De8PMieRHdyc_NO@;_SZ()YX#<~oq3P)8u;0$rQz06
zFZ4_hG)Ssu@S;V4gv^ymA4Gbk5Lv7*P3U;Vz)F@Li2bL8Z_GMi(&ql#vU&vGl;!H{
zCwhJMk!E;lztvp8dIBhCN!YpEbj(l=WHU@W0bD=BLGpxq<$YzCFsM+-lfwi)Bixjr
zjE!$<{LdjFXouoPB+w41&<Sfa8%+K%zQ<xp&@I5Fg*XE_W}I#dL+NuteV^^6`r(5+
zy(%_D{LZK2DzHZX3dcglFb@_UU<ytjd2f`;BP$n3IsS-25@BAr&DUjQ$CV0MUH(ha
z>zV+`f9gD&*lOJQcysc+S^C*p4;JLbX@RvI;t%hc*RrysfB6FGPt>E%I3_M~kKmUF
zUo)$S`G)}z*TXA{{3Dkuu7)Q@Hn@EVv#)YQor>=$Q>k=Sc5lb+?HAAT>lx(lj#&E7
z@YHdKCad8Z2bgEWK_r#H8_;)Q{gO5VfvM5=_hw4lBF5AU=@n&15t~9CIu0^@hW-^h
zqyFED1UkXvDH_O?hWHe->eFP>Me}n~MUrkl6|z=9FTq#xH1%o>D98#birckr?ECYf
zvpFhv%PXO+auGcvf8YLxXi~1Ma30uzr&7Bgw-UnN3<t^ESmKaIWS?PzRk^dX{veqH
zV%lXRy|X8UyR4eYQueiI{cj1bT0&*|WgRX@TjRws+<Ofy`I^>><8j(-nD-4KNigps
zvo(o;Ama3<$8w}|a_?5e$DtLIGK1!(CS!aYh~(!^{;D-iu?z3XO-S9>u#4H&_~BN6
zWDv}j)G{Kb*9^L~>nfr|kdtS74j-C-w9^<Vw&UNCA?E)+4=vGXTMPiVT4AvS!7(+X
zD1>4LE_3d&{JLpKl)wAgzgSgBTn7Vlf8iP+&^JcmV7i(0T~h~dtpQKvkKKn#j^a&I
z+3zh8*QugJg0*Nv+wbnGrsSbjATnt^Y|~bi^61~c(!HjiH?t?WOsgbp8WMHISM2hv
z{Th*XOuIKL@-Z?L;8YD>le{3<Rnl<;oSiJ?D*jTvYVA_m-`74&=tNePiN<JVpa(7k
zCl2V$Ld;Skv|w6VL46=Rgt~Xs+1Z)hj~pn2LMDeDb5(Q7NfjMc%QTi8V*bA)v<Gz6
zR^Ub%H0vXlK#qtrBa*<Ib-+4kbQ=(l%>&u)P#U<!aO7xKA3l?vhABIq!Nj~c5xjfT
z*taZ943Fg2AA0{h)D<F#^^YuG<$v*S1g)&;ZUa$c9T{VJ5!Bf}LK^b$Zo7-2=-wRA
z%Z|1R9^>LhT#d=YumOak))63ngbC@-ycs9<P9Q6*?gxn$(6>2=SpQN%$I8RV(K#zl
zzAINBq?Y!tg$JxHqYzzY{%Bt>LyLn?30<d@vb?8)siR+>Q+QEpT!Cu2^P8Cfx-`yg
zJ=dQ+2)NVM-ov9N@bz+@FTR*eh?~DUEE&@|jx^zuer31#S~#<c+=o-|^Q?djnJCyj
z2Bg?wB5Ub|r}x|hd4@J2ajwT|`MoxN?2xj8wK*S`_7}-kTc=fHy<%D0Ir#l8BBk7*
z42NN}SX%G+QFN(q14Z>tb}0praq@iam=8D8i25{HP+t65963|>)HbI&-3CU@XKeR2
z6vqg;h^87tZR-9jcD=_vON1*@<ZCYH$eSIh{%jkXhdgnl+%$Kz>-v(4bN5s%#jtuZ
zE$3FdxshUTOF;2fjMgR3Y1)oJXnD1)<-~1CesN-{z~m%3Hr1VdtkVmCPDM*a>u(Ne
zjN;>HGX)UfHXHuA9eJ(6_9RcCf<ETmShS-y2h$Qh94_EFY@WU{5wk_n(xFq-$gNtg
zr1eg$ZloA&FTe~iK|n6lf(m_CYcT>*w_yH@F`j29>#UOFy?AjSQ)<2WY#-eCP!kJN
z4Fw^R+{4GGe$La9d2R=68iwzd`A96s&#SF<@|Q4xd7o!w;{|~3GZJi9s^GjYj#3LO
z3f3DWu@|#b&s-n<-<n4a${;h^mJDS0Vv@9Q@q1XgtH&>7#-oZ?g6lumQw8S6<c%Rb
z#pd^a4`8XnbjRP<r~h3v;=FI9hpzlCrj+~K`ocVvSGFF2B9H(#&(uS;?w0ahiqEn@
zWyNyHBOSa!<4NtHy;epL#Ge2-sg=rZO@Hox<b|AMh0W-$GJw(!U0>Y*`E#=e93B~d
z=R7k3{cepW+3xdSCi14|3X7w^qU~3hit_oQ;Y6^u=Bxy*q=%$$I!-stnz!rV=5SQ^
zx=)Fy*a>iT^@$E6p)BSkMB!vKky=5K+AM0A`CvP-rz60r>-GcB8p<o&<MqH>T>lu_
z%_)(e!^Kt#9rSAu>E^wCb+YlChknBI<Gp<qTe|p9^UYfC8F!br-G9rK(<T|S$)fsr
zGzht5l<6xjt7Z3V^kJTstnsuq?}o*6y!inEy=wsKFv$!;P)9){`Fi;{V-A!_uD5cv
z7M*u0t2V;S^{y8AxKc#mw<3CG+jI`5v?e-&K$b#!u^&iS8lb#-%52w7A-b-rj%<qQ
zOws&hOEk}(Y-f76B`iz((00{tVe4oamJ0AG6P3sW!ClN1o2nr+*mK<opNd`u3UX#I
z%l1G7{V{;;tktpBzI63~Tx>mdT9D$oN}&$c&#OpE<|6MkS%$+Wa^Z{b<O<2N554zI
zix$?Fg)2&RIEU>(6Fr{#z)*Z&s$9D|OGw*cSrcbCV*M2{Cd$tW?flsC!SibKZZArX
zgm*C_g*kB4RAS;=F4Mn^m%m@|!c4ca3l+e>35jHeX46Wdfnw&XFKqEhf9!$H<^f|d
znt`zG+ORu>EZ5_CBDsH!je&MUF<gN~L7oihR?@dIkXR)laxz0Oi<E@&-ngyU`QgoW
z6hFH_L1WVCH2FuQ_CcBm^$+0U;c=gw(E8^FoZyO~LVyM+L1>SWZX-<Yg>J-rwE;qO
ztnQ{zw$M&n7AL$I=p%bP%q{Xo`rzf9f6PAWRe4Q_C((zNd#uA282yrz7HLV(SJzoU
z91&*c(ZOtRFQG}oj>H6iIiw8MDxBzZCd2XQO4ujJGWXP+OX$Kd&D52}Wbsv3Fu?qd
z?36?sxg?-xYMwB}c4iM_Bs~FcG6a-f%x4?->S{RSPR2mO$L+JYQlzybFER%}jE$L#
zg$n!ahh#CwS%k$wUB&HiVz1|ja$%EWUBTLzBJdS=4*YqKf$sk9R_KTp=7zpWd_m-Z
zbZqTD!P%4e!c}8y7Rek#6PlOm!a~Nu;1$}$695g*SpJjTxhgyvCZE5FbR94@bqJV3
zih4C4l)`!pMf&!;nJEPqixHED6Vah}JO0Y^bgzNZh>Vx8NxtoAd+=T${rJ|!o$X1m
ziu+Rx$?|4j`x$u|8Sw`e=wlKC)DmXrR(a}-7}t+Rax9E(j>LLO3bpdJR_B(|wt`lW
zJMYT%It_5-72C%}fy0JhAxl6j<tcP_m$rGsnd+RKYb`5+ohz(7e(TTFH#pchbY_@D
zeGqsI?+y?OjE;N#`-+WNgz8rpn^hRIR)oW5(PEz1Xk4FWb!0@z2;SlJ$;Hm4!`Oke
zPv?g2*QML}f;R^aj#1j#iC1&>P!4vHE%FO*@1DB^T+RmSL%n87u6>|oo4UEca7=`W
zY|1TmLYVhvfVAe4(9SDm?ScwtFq-l0YZx7(d&Pt6$AD3gjo<83Zr{k;!;c&so2jb%
zzcyAi9^wxYt2c%cKWDO6%V=mwA=xz?JxP$kEGAXMvQ3sf^IX63F+kY+1zHB4p+Bh+
zViNi9i@D>DmS;0-l<a|$PWA%G+bn&$;@wLl+Xf32#c3q`Kik&iIL%93)(!HV*R@+o
zA+Ih7L6~riF>B-c0VJv;=Jdkud33Dazy4)AEOBIHi8s3in4>xjVv|eBoEQ3eSDd1b
z#zR18cR@jlubfc<F<#m7SaCQ`>Ii<2$uxVHtGj;%MoxN=GXjAaY^mYI((sC;8!%Ef
z%NQ4m<=Cgx5mR>X0Q!>Ty^C-=Me;#=;k6;}ZzEN0X$>lTwqZXK#ViIfEO&5(6G+KE
z)P=%C5E5gs7feLf(~kZpPY;*fOMwvhY+UeF1WwZ_Y>B6RlBa6IBI-Akny-jpY+fY_
zcLTq8K&Odju%6}2aJ^eQ$S^y`Au(pqOCSr_b_?P%@sAO3v5sUhOr5J7cn{1Qn7dFX
z5s>ALxp<!ABCQ?k+b@tdnnD-eVGHx@XgG!>T(J11aBOS(emO=UAXO+8o{=NOd7m8_
z1rs=kzjn24?W70e<$-}XLCn<9X@LK#&;k60fKOPfhOBLz+4!UaPvWz?>IK||BJ$2S
zsXDkXC(46;4L5|Ci$nsEu*-;4dm&9%W6p&FS@WA8EAE4xU;U}_ogK37>DbzUP}gsg
zf1kP!KYuqNCIs7wyz+qXHv+scRG@JjciYf*7G>jJnNWz#Anku(F9*O$h=qSF?E)$n
zZJjzs5~z$l!#aAF3_a!y!eFt_kTHz<ewYmJVM2;15VXCJ?y2}k;EZOey(Hm!Lx13`
zQtYPqzu3LY9}KuE1tp!^&t3ENnpvNW>SzMW0W8)d%+~4Dd-LCU3~ua+<u@we#43hT
z<2ieVrlD;fDV2IKmd`<htQEn*F@Y4s50jqECaZV$5|)qka}3vGLcxKvb^qQXR*?kd
zDfCWBKkWis?=V@=k)Q&Vta(LDt#i{=0-n$ed&b$S-bfOaw*+jJQ~mp4D{-rM(W>$$
zFCWEPq%PNV2WghSjkESyne07I255tZiD3A#*(}PEK#4b8Hz^ei)08#>EYK4^V=HbU
zMggM%$zhB#2IM044<TQ%zbbyVm-`Zkojh-gc!xM2xSPYg!`;z7h*RYM^bY_(%3Lqo
zzB@PUCH;LG1IK%avc2ltl{<`d#}&}5gE`)Bq2wjE-+(Ah*_QI^1=k9dA^wVrAaEFL
z(t0su3`Unm?ZW|yTzsQ*2OuY8BBLqbMT!VkoENyx<fSduC7wBcO_|+)q;XqydtZxO
zGZGWW2#QB_3_>rhy?wwSjIyW3P{;By{jQxU6kG5A-3&2XRj2WHD&tuwh?D?HWTDiv
zbspWzsyi2l$hm8cGXzYds9>jruHa7Ht3H%&Cv-2xuyIwG@x}C?MxeBw`{u_1I{(Rm
z^BM1BCgYHKYMm3*m!#z73eRmg@Mk!WRv3;D&Oect6FgQ#7>lZ1h_1z|h1VT-=kmBk
zfBb|x_qP&9&8LEN)zLkv@wiiG2il^23scmzGE@fR2(Rpw?9{R$mM>X<y?S+^QWXjn
z!hlRk2gCRfjOR%f51>3Vc7Lz-D2!do$I|xQ&ecTGGUJuZMC+(A;vkWmy)CeKwUgRY
z=-0P2vU*S*!3;AMA)>MxX8aG_>GOwVpsc{l<#``f0al0PkSYOP?xhK$5`yg}fVz(f
zp@xXxp0|dGprOV<h!-THAYESj2X=T?ASp{OPML(=8Qq{<FmP8LQUEY3GDz8%c$-4n
zq#a6kjoUPzEr1McqO>)|>VM)4%jJ5V%V`N*IgYwO7V5RZ`03TCGP4#&)2Hu+U`mtI
zL^zmX**C9gO`G<)^7zHvx3n0w<NF!1+BB0ASh>?<sPsM?1%ADVbwrp5rgp!T;12)Z
zGvRA2gLeF{VN-$a>r=&yOBrT7%cjulun^`74mGne?gthX;yfF2egj_1))XH@wR&|L
zpjiq|)`H+E!{$j(|9^39`^J3sJJ6#-^n}%y1}7fvDCtm}dCQEPh@r!az3Uo9+`kJ`
z_1TbLKrPu-B%id;S3f%4#VeT+u*N)ql!f4DXoM9xi`zXm_Ux#G-A*Rtw&oY1_R3Un
zUmCm)G_%?*oXRH0COZqdxvTJ9<pmXcwHw^DW_#&moo*%I#rnd5O69>b>GAAc3e)K`
zTVb|swkxg%SNeW+m?OuOd=<((`R)dwMnV<g@+kUr9un?O!Ar~z59rXqR0`xxu7ABi
z4H!SgvY*2soO^{|tb+1b`lAM{Vq(3C7?er=V)?v+r*I?OE&O!15oAZGYK0#om1_w4
zRn1ioocl~)0QwaORYEbCuQAM4k5=BSXgApDmXgCxf4u@eZ>+@y9%m$}Zf(VeJ|q#@
zub|VioxV_NnM~yXjgUG(W4Gc5bBY}q7H`G*k6-`$`%^i`zr@*oIcWm@%PABe?^Hvh
zl4jd%9CS(><1na{7Yq_is{1R*dewS(adGrf;mp{MFa4hF1ZGJCZyP6v6@&#e#(|#v
zg0l}@q6Ms6D8)F)+Tcc}1&{2)!^}&+tW1oHVr+X}rDkSpUIobXgp*iKTkS6`Mzq4B
z=e++xD25dRs&0B14Xth(h=EZt1|^6h&qu$?btTK^Ggw+&C<F3!A4uC^FGW`dm`OT<
zh7BkwM+fJ<s%Ml)QMS=Err@xru5Rrid`}JbgEtBiTSrFt%3tr=-fYI@TSy48G`*fP
z;f)>x*F1NG8sHFK@w0dR7P2@S4u(XY4-x^<)eB`BFM8p8upKte9cit8V3V^+=79!8
zDEwTD>2+7Vu8!P+=Y7bsKJx0t7LLe0t6rD`oa-Dk)wMmP1~j$W+WMYNzXttfhzy}t
zn4I#lG<A!^q*?#Aq(;wk@*Vc1QFuFR1OI6%BTBh!3yAc(dK-Fb*?k+TaA2JMYInV!
zy}r4iS)v~KH{qQOiE`YJnxE@mrnNm}eJ)Y}DkX=35@9uEWd4VNYGr*dHI$Xwa%y}5
zCmu(BOYh-SS22h6T;P$P)~aj&Nk$(E{n|SzXO^Zpx9L4<Wwh8L;?EDh8+SFZnEt+g
z;pA8c^ySf>75Hj`rHOe81^FOuLhvfEwo^;5;SIaa`&I_dm>yTn{E?gmw~TD-(;6ok
zPaSq-1{_ElaTfH8isf<Q`!PMNSeheblGG02T)C1^BFQu23<nbA=vF3hvFKLl2*8_3
z;KlO(9%*b#xSY4jVlFaK)b_w=D9t>4el-jLO<$QtY;GCJ+9s;)r*RZ|V*b|GoB~4!
zB3iQACq&~UyJ%9KYy3W1hz?g6&t#wn`EDZ&91K%r#wa8gRii;zm|oq@#UFV9*|`Au
zN9Le$1NNUu5AeX*gB@mM&ebzmy6Pyuz?l%q08s(vC)mLlSW>XSoTC)8U@|BHX6^%E
zkrV4p)ID#xZR&Jn1NA$ywVj<bH2$t)@L1Q%dG=qM!vPUSPkS2He!A4NOOS$pj|Zlx
zvfPSQorYxNkt^=QdFmO0Y7`FO4d0XBq&ZkBzT*TU%iocXcwu$kYwuR*gCk{>KdWu8
zdI@y@t=Cm;S*ekAuSVW|R^li&ei@hnN)FBF8cT(8LL;O<MJt8j8nWxvk$3Q8yc&!_
zMJ48}Z>e3k;)bgO-E4h5oA=Q`wX^J1(j4SK>?4JK)b>uHK4*_9v)4-H{rvWipRw0p
zlD7t=mS`wlB?xnO0?XWt_CA_*EqDqAHzEYzqz^I!PXwE33W2Q$+kZEAT(95@xou(E
zB)FXz=z$bDQ4&7c!eZ5yd{+d6`c^85DqFzh7#)9FVe+Nv+GKXbH{4uiSSi0+M-%)~
z58{w6Q3+$U#*oG)h*y=%H6E-jwGBY0bW;0iG70as{hB+=w)G(NziRs#3tG+93FHcV
z3_*9J1>LKfRh0g!vr?0t{us*w9`rsRl!<{Qf}D@q3xm#&y~Z6bOQ%96e&pV+2OY=q
z*v7+P`4l8>Y*r1s3hWBCW_E&81(~|czZ@gy-p@5F=!*+pjVPizH$U~59*w!~pYXbb
zcP)!*Wh@5HYT}KKiUpaax%`p~EG5z+&_f)7<2C2cE9<P;&Xdx+{)?{|(U%?10>rjY
z5qZ0)S&4<0f^q~X4bpt)_LD;t+E9YK5n61$+K_J?O(h$uPsXpq8ZV^p!^<4H^;dbj
z+ArKJx|!<ZLEHHg&e0wrkskHm7vhoJl1}T)(rwft^=24Hajdtu?$_x@Ph;nt{}Hi2
z8|!_jOD|b>o3zX&A(46l9Y>Em6O4YAa%*DZ2yp#Gxv{2Iv<e(#D5B6BU6YAs#pkmN
z_*RLJEGI_^VGXy*bPh3fC=r4f?hd|Flb7)MpEKXA8o$fqCwo~}>2Nx*i`Rid#{{VQ
zqe?p|w)GublvA$Tsqa%fNTFpBJRK^rz-mK|bJ^pg5O~%-ogX>_?8dWuwWWV^C{4sa
ze~kWfP<rw*f#-9dvA2Q8Hy<=N%?X-_+wgHSP|3B&><AqferPzfw-@erBHuJj{erx9
zlvX{a!Fxx(Apl~{f5?WrLcy}8)$?q-d7u_cySvP5%As%wen}YKPwhfa&JWE8<W18J
zO0*1+QxDl@FKFWfPX?6q?r=@>(z^OENE^rN{`S~(VRdtVJyj6@)mAhJVx+cKpfy&&
z%t7T52k>46$V%3g8tXTvZNo9cHX(8gr(aXl)PljNW?n<xRvlMi+oK#!#1-%^<Oz<5
zS-k}ug<%py3>Xvl_Am)+Pio`O2Mzn-y~vw^FY{1&qa`o_#fb8SNh*WeCDNMiTP4z8
zo9Q1KCq1v8k};$3k+p!Km5lUCY;$`s&1&n`?qcMaSU?1=KiHBeyq<$uVsDXC(S@uC
z#59DXpsF=QZq{kTbyNO$JE?S1TjktD3fs&^<KczCWe-M$fJv-^RD#J7gp`P$*j=&&
zVfR~*^nwR)@-R9`6!k+1s*#&yEgR~VN)Q<Yjxh41!^Z!7-%0OQpF<mD)}J}~>2=6;
zR3FlT{oAiLg_+|ikG0uj@nT#<X+%`+b%!3oVq6BeBWLi`@6K(8;Cqg3mp|5y#i4nY
z*!~1YBQ}B9w4ER--8O~R==5lJxh8C&xZfsJgCQ>h?zhUT8L4*M`;Nqr0jyIlvOx%7
zCgkuCVTcGrP=4_8u*TpdSqc@B>%W$os+}d~)AB!0eN{v5pd*$DX}k`|F_2M|hJfvT
zL}K3W9sm4)eQ#z7F5~*_#|nB9m!W$i6%}O}?$#_MyA>_liU^aV4l-^T`H~(!@;+Uh
zX&-n1mx>L<_=ol@1t_*=C#!qO7VN=ER71|(%m?GpW1g~q>Lk=%HUk=!KZGE|q6xDz
zP+!0@1S=-92^w^KxC%X!W`^6ab%G$ivJ7J!4%0_HHm<iccI;*p#Ve^`r-+l?hY_y*
ztavGB8eJ<lLaXOBigx^MPqaF`RjHNLzU;<;HIz%HaLt)z7;3Qd9EsB)=7r~jGQG-D
zXA3DOQBmj9jHhJ-D#umfU-@U4A^orNWlll}=Xd1~4Jj4SEk!uX4TvXiPLe;dWu)|m
z&5-XUx`{Dmh)2_b0`uFF=kc@2S;Hel%zhhYzmy=|k`n+MVArq8_-zr1SFUCXZ<!eI
zGKn2HLosb;asyS#Ffyho{jWIqcX~U#5OnZZ(C7R1V9d^T2lvcz{d(mOgh|mJbM?Dc
z<U}8LXW!R|&J%mwW@8<^6?+ZIcLi_Ui{Fc?nt=rIV0fk$%(SnTmo~~uFc;~vO&}7&
z&dBN?Mu&k$s2WOHU8C7r-72O8r2ypiQ5;c<LS-wZo>$+DB1KoDq?i7G#~WDKkdY6S
zI27aJ=0BuXvx??l2Od#xiDfn4%~)PaG^uwIRubKRArp&4e!oAEH@Mm%a_H@i47{>M
z@%?r6=LWi+d%6ASae;WZ`=Zd%M~+`*fPoRp+pJjZvll>rKMsD&)Adz)9)26F)P@jL
z6>iZABfe729mohyD{!*bnV2$m?aI7bTwT!izZ&yqbarBV^6MZj-z6T|st;347B8<d
zlY(G>GQ}ug??n{y2af>Mq^jT*oKLxW%zbcAQe8`qH+3>T*exGKI193>_xhE-p>dyK
z;z5FLy#bjSo0y`XdAM&Z+_CAhm?l^IH!KjGd026CYPst5_gc=LTh@(ewOVSrjv+iG
z|8=ApgFynt_1LWoqs#Nh(}}x^<a1?<+UHR(U=)8bo9t<^Bo8kaL;O@`<0bT1F$U5S
z(|xN|GCWxHcqbb)9BXD=^4F(;qj~)9h2d3JfbEv<jo`-vyv$2d83x>wdz%4O3_*$O
ztXX91dlrtCqOn&|z2%it;=kEXH&ykvrpw3g+&;=)kQseCh#wIv6*IBCiwX;7_l+i8
z;otsRovX{0y7eYXim`*C_sp(gxg+4iDxy@ICx|cDMPujOUhZ*JK`8U#1jU|Fw1Qo5
z0^j^9qw+%a;-$mek@XUk3%}Vl7kqWHN`Y}>3N0ZPQPh4|V3VW__`l$Y0^<3w<{Bg5
zbRMOu7b>+MK;m4WRhz;apLbj~BXo$U_x#VPGB3Tc<1ZE*O7wv)oIA}%?=Mo{%H7tD
zDDhMhIzS-_oRkia_}_8#E8<0N)fY_(5EP6ZR_5E~2%FQAG+3TfK`Ms!5is208R8E!
zEcAxXN;uv_i<J+_uyG{?KbA+yM3V{v1Ddu)eDurSa}h1{R$A|8(bxC8VvnTlBpfY9
zq3^}hV@LT;r3uxTXzUjqZpra4y)j)kw>gS|g7~!;zu;Pn)IohR1cn(Ly@>h%*YPj^
zpgu}Lxo)$(fYxC$8R1(zaR%Anuc<TQUG2F+A`zd=9lxb_3XzZ@>W_`mw^mtN0#oq0
z{F=EMdKe$+<(sMeQ#JU~Zwdc~U(olpE|}lu7&bhXH4b4HIF^eh7qg?=LG;GEek5n-
z*kHsJx&;U9^Nu;@T>?*1(Dt3c#TX3nBkx8p0!~bvXemUU;YsW*VVmYnZJZka$c=y2
z*8g$AL>ER*ZC($V1a+FWn^gC?srb<)Zbx7mRr;*4et_YdNnrcNZG30BwaD2|o{UCv
zvq8t!Mh@AqW{UTDAKnSRrXTEE(O7dKJmM=2Y6{~jZL(rxv|mO71rh*J44K56Wnrzm
zXNP6QdI&mv&^~$;#Y+^-xR$7ppQ=I%q6!AgFG@?iwhanu)G=+#OGa)4E)NGFIt=~K
z^yxUN!wNu6WW}j<!sA~;x*P8FjrC&+m!-3<Q^yN;)z(a$m*xnc!kfc>=O*+D(<P&Z
zl_(xyigjWB^PZOtsH9DujwHRg>mX}hed3wDNJ;uJu3_@?sswwrvn#YeOqZHQW8<7^
za>Vg^6r<DfOy}R&<LZN7;0mi_*qfIGH8M#e56faf$v}VZ+<XG$QfHEkOk1%K7dQs(
zu!~xa1^%K$!GoQbFj{uy{tnK@HG)g9MM*nGr-US}ph{2&K)h7?oMKU-dG%G0?*>Zm
z&)DsXWTg^vN4my8<6~3prErt)era-%U-eWI$yE|-K^eWg375yPJg?R`i7r|P(`usw
z6D5hA>X(Oz?ImlX!2{yiu$G7NW?YPMp#^mbr|zPCzaV^2G5ppC;BR$r+n<$g9PwN-
z`=~j3ZHEcd4!BAXny}>-EYD$un>59O(U2Jl_u$G-EdefDHsZ?nEQ2x)xW%7Y8&lbc
zgJnBOQ>fdAmw8wDS3~CB>>-t!&vG`>>sWcb^LF<qM3^7A$I$e;lCSd5#W-Wa#qZ*3
zC_^E9pNY+ld;lK&%EiT|$K~yf?Y{bD2IJ-$E-ECY&~wM4<@)+nX`<sgX0&px>h}>^
zs~<Y_t#*wU&#pJ;WPgtty0$&2%+1X=<<o$X!$~{DBy?~-!{8q*zH`VGuH__dd}Fjx
z*mgQuzB#@L(tTj?A<%}6JO$)H(wZ>KxJ%Z2M(0|}9T3`V<ZoBuceE3ddizU|RuZ#0
zYm|JOxf6qmcy!=i@xKL)Lxg*(%b=U0&w5^wg%*j%95cq<F|{P-zeE#N_o>aGF`4WX
zxf==c=-B#iWH`c$?DSD}&HhWq$Nz*Ad?Qv$q&Bd=eL|+n-amsQe_XvMCgGwaaujN<
zL^GZWFh}!8y2dWG<#iid+4=DF?Ihyd)F+s0ULXGz&8Zi4I3*EIFEue~6ejwU&P*J8
zGO?mPrbe86I%u@(;(Hfy@yj>C>~G(_0YiDRaG=g2t%)NwIwPa*H|E3Bw}z9s=R&??
z<PA%)G$MOKxHwam0H9qtd6Lt_hBxk65Zk^1&<?COLn<8%`7-dLY-jDD6?3|huW5Ix
z{=~*?TfJJjhAmcnRa$-GVlZ&~i6UZco&FT8+FCLEy9*nZo_{S1Wkc6?wvbj85(4a0
z{`fiWR@GI{D6Y>!BUyDygJHay9`ju-L7+d9w*nU-$=KA+Rn14TvU;0-L1-bp>}^W`
zKo>;&;TUXHg&6LF_D-HiY`=VW9$#e5zbmOOOK?~O5pO@Wqui}mb*!1&t_x#xeSLE=
zijZtwr*)%X%JQDIT=9o?g#EumXT^+k8f@+)lVWwqWA#gKN*mME+M}F~YRbzoqR6%Y
zs%NhAQiH77RMnjJn$+R#7H<)Qgb8%_&kQq87`xyty!I@fwd#&7k<^(HM>R)~^%xPt
z5veb?o8b3ieuZU4v+F9IGdwLa-szpp*srv7xyqR>z+7tntNgpa%>ncxA(A}411Z{W
zEQFG0=4%ytFoK=d=7o&@Ckv?mRF?lGRwh>l7@(EVOmDyoVr3785~<F5z5!up5d$xh
zy9+E=hsj(+U0SOFOyAa0*jV`+Z=!h+sWt_BuEzK*lU?)S+LPOSQc*6fw+hAXDlbX4
z7i=$0QEAqwe$vnH$w*2tHl7wN9xFc{HvO&QSR<5bzAcn!yOBGr*QHF)rb`4gcTYmh
zS$CJxeHE6cEBG_)E$zUj6M=6D-Q5_aJ-HCFR3Znb;E5dNJge<VMABpE?%#=tB*D>O
z3_eo)a?g`Q+l_vaSp4!Yie_Q_SPUatI?bnC2UV}pfq~cx2jib0dXB8JD{=eSnW{s*
z#r%_y^^r;Jm|_&+SPqiaZPB)<UT!%>D=I7@jF~>ejLJkq+n;!vYXX3^tp;z58nmwk
zW4T2~qd!bRh+Yjs60(BkiJ-BgDn2y`<T<?s$+&~M_vp~M{8G6GnGO7@zabO1m()PE
z4V(2VbdDk)d?Q<i(tQVyXeW$Q22o`vpz7-s!%xo8=L~%}wHx?dWyh#cS2jEnOnIV;
z8*b*zD_$=SBI_TPA@DlDNvqh`q+s$56*}ZYrSSB}-MPijw&dG2iY)lmsU?-E=BiMz
zuymg}pXC7BWMQ_AZI}3^LxPineulTG+NBIy$9!BJ63D9(a?zYVA1uORodIG0e>PKV
zQ$jNbzD;Bf%2XXunC1-e>yy@`pio3qS~n?gMtbcYjP@a+ML?944v8!3mJil+_<B5(
zJ~!0@q=EauDavg<#KfEBSoUPMB1hKT9Pe%JZ`=f^jqEpEzo+yLkjFk7L}KbwZd&aK
z>)%f=tKGGJw5WS<g+`=B*KEZ2s|8d1V<vLz$b_)5m8CBVj$(+Y9*7b`*(+rEIqz?>
zgp*0(zelr`6M;nSoZnJ;7GL&?Pfbe)U8l<9pcQm2$7;l+;#TRQkn6yxC~R!LV{jE#
z=Qbh!KbB1D2Trzv(WlQnX;4@RiR_{3Z!it)V^nQ~K0|~}t2#kFS{dyYBQ~vdjH83&
zGKwKfxCynSxSxDOX;9AYh}wTqSGR)4i?%}t&gQkSSpoKhK1@L!TL{7=MBhpK742Fu
zO}j>JLaEBXA9QYBlkyoq*52)keKpqDpY)KUyiWsj8M7JMD{acbrPTtRG!hlApblZr
zshgk<|NFQ=Yy4o*b_9{jdbv851<AfmK^p#M`Z_1t>)>Yo@#euo=*7v5q&pWycST|6
z#FV%LdPCvhJ8BL+f*$p-m?OSP(_KX*J6;#D!%6;OxT^bg2l5TR0})dDBj1s_dRp*-
z;rUnBRD1(rA+pqx2q;9g%&0<Cd1PrI(&FQQ5ROn)+L9(f$T0kptx4r@_(G5y&bUnf
zR0GdnWJH~0#Lo^d9bkbCPcEdJVDLD`ns+V@pop?+V^irU8=2ZzW5IOKf16r$t+EF9
z96F{oLOscKE7b&deqv53E}Tq6;rG3eZ|N{QCqGwHv)CQ<7vesYHw@Qq5d))`M}CBk
zxr-R=?@Yov$Q%f{+#t7t>8jZ>4*y+(Wf+bV*SWP<IZ#QN*@1owy1Q5TbPP&qkMB+h
zaC*?1A@+>waC*R}5rIMV)uNfTUIwV?hd0JEn<iD2AuXhHcc>A}pUA5DUgfY<*Iv+F
zMA9LE4?>YIW(`WWx<leZ7Jy2p>6Q;g9zF|?HeyXej>^!9gZmdm2cnK<9Vm<jkBKou
zqf4!EL^zyiRtza0?R#ni_DFju=k+&$HHPjX`)`&F`8xJqE~WVW099vltwTzlxHDNV
z_kztXS~AIQl(|YD#4mFlcqh~#!PHb$7Jpdstl^r^dAs_!b*M0?*js#=RH0`lqA%uH
z+Dl45CL+8cx?+H#lP#tN@~u*a4$FM{V+)iFUn-;C&>m{=T2F`@DFw&Xn~E{OqGom=
zx-glwuv|9e=uMI~)VBF7tMLOP4!tt`s#V0V{^=||S%Qzlyn<c&H%F4c<-|dx&B^*&
zrDlWvsG3!B1g=z=ncJmlGe|yI4oWqKNwjX>??|c1U>%n6q$vmjxmIB@!9_wdBXC}{
z|F>x=9>Mj+!61MYz5#Zb=3GBp=|or-dHMVcYlz3(?c(yR%D~&En(hpHL14%89hxk$
zC_tcb$h4gInCPDRzV@p;C{!MXCJdza^8u$)DyeV2$oy^;eeH-<ln^HA)*l1uzKkjK
z2F#fP-Qfw-wp_n_4>w10QXO+ytf0DXR`u;Yu|V)ZE(%P3K>|ln^*>`*4RoNWc>myc
zYTW6gKp}z~tnBcpgJdIq!%d_|AH&3~YH~ed;%hpK`qG!ofoL&QEe&nVy3WQ8TJUUH
z@Z=71H}1%IMx*lHj!fh29_8*HIXqxQ>;xnH-c6SP4%#I~^lW)9VV<X&@Wn@v$OhlR
zgZwale>iaF_5%+wKBV{(!o$U6(Mm>w@M6{Q)4XTt;QTv8gIUaboW<tKVfb25O}bE)
zabYJO*Eb0neYhjs{Jm%Q1tg_SX=ZBsAC5@$B{T4ichc;Vr;BVMwUIimJ0z#^(3nDw
zDx?+KqC5W*Ic<?HrItQCR~NnTki4e*H|BKAf&)RX{sD|RoDG)CjX@n%wogCstt(!4
z<7A^-hr}5sq>AWkTj3u=o+YzkX|udxjZoUcdxg1vPNXU#z|lYvKNk0Dt`l4CWY^`7
za6|Ev=V-r!PXExm<u<?2SsWuTg*i?NHW@grcBe);EmIgJ4qZ9N+qn^tU$Y4}?T!EK
z1kJTWr2%9{I32R4Y!h^%lRm1D87Z||h<^OgXm-g!**hw&e=1q7%D15Eh@+j8f<~C|
zEySXtcrJ_{-qUZ?d)jt%$p4j#Lb1JvVhzTTRtg8tSRg`1O`7;|QhO~pk)A&-F=SKM
z+Wz?ZGZPjSrFHY!VdZ$a9HcJ4@MK^HEZaS<5CD8a<xMKe_1#F;$;))MTFDM?Ox>8h
z8vM-#dl>l8&>G!OsO-MH5sKzGG`S|_{+g~RQ<#n+X8}|{CVBz}*FYVQ&ZcDLwJWkQ
z$H+A_89BUoRv6*IPkXR%qaL4r{u!TEs)F}!(<$PtjPt)0wT|x4r5dId+WnVOEbAqM
z!wMJ+YbLF9q45?dq1>iaYw^QgHprYK)3%l1nix&YE870e|B7HUK)H1Efq7?g?p%U^
zPgb6gSD}kIJWRg5uhQF?6wj<Cgk1p}c&7K@$F)Z7yIp_k4k=?{AJsI>(-#qF4D32%
z2$~%T`?R*f`g=}Bygx-dXN{9t*=e;uK3*@`rw9rBgu;@72Ah7o3^(>8zCFF1m$v@b
zIEr3vppmlq?l}(?n-N^%ryhe%FiTBvq<S)eBC?(JgGhfV7Zk^<+W8b(ST_T|P;W5<
z15wIB2HMt&9LDA&cz?^CT#>S<QLd*vc)#o0ZzQkmg0zEvylChj1o>o@USqS=tVG`K
z5R=fr55oL|H~IsSECSw=4Le>x_oCz9298U`<@9A;fy@=v7Qqr{|I{(glN7==9e>HN
z5=T;HhU!xtMn0!ek*|jE@z~SLhcd5LYNeVeIJvagh&QNDmQ9A+r3H1Fd&cm^W?c7A
zI`Wd{%t<Gm$jVLZ!9WkGCyKap5u*`Ycx^cXg5G{ra)G=w0&4s)dK}J+zk|7g@4N=d
zx*M+3Ty*O9|0!Q!lfLM${qDpJk_J6r`rCm5!<t3)wbF3d-B&WHU%X8wDBv>sGQfsN
zu=RY1?kKb8x$<yHfAe#o`{WuK!zc|&L+c;J9Amgn%qaa6+K7wV%_^t8IU9sJD)*`c
zu=?i-3{O6bYUfj#!;3=$WCLV(K>eo&*m_%fSxwMB`AJUC&uxV}8|htdWYMg;I*_gQ
zKNT0KpHHQ*t((BIcw+pt;LbFR{?PlMQ-?ZZ+tM0Etx>}?+YuYjXx$t{>;>E@IJhv&
zI5PZnHKjg0%`8+Wx}~=IZXMwd4}b~-8Y5QPKT<)`WIky5Q)QlU3Yh5#4uZVO5OV{?
zp~RrRf%`ikc)s7`8?Yl7wDs~RrQU&ZmakoZOc=@y)kQjBV%U#jmadu6=T*wwNw`vv
z^QyS&b-9LjwHO;=DjN@)JmP`3d5ZlV9mo|Z5|F!C!0yhSll#DEe^)e*VfDBK%se^D
z4wIF$Y;D!3lHpq0*y_#Ok8MApJ~C$S_#xbkUhu8lesqg5b;nD^ONj{jQE>j=vIqcY
z??nobZQ--MMuZ3)37m1gc0gQU{${OdoDwz@nwj+GArp5;faeHyhJlVdQ>;_5A|gMM
zA(qNjv4BYo`~5IW(12j(R({S9K=%I3ChxX<v4_77V%>{}o*M^uF#FKW{(1B5&nP<A
z(3SZI(Nn0JV|Jh)3uQ6w{4dW0@=4bnX~r`;L~9I1nu_mALr1yi{!PQ#56ygr;);2@
zUDS91$#HN!#G2t*FDQt;+@^$&0;2`9mW|v$m*uQ0$?rWKW4#kV%44QkfKm>xx${ao
zmwoPd9G|woODhqZ&a=o9KX4<Z;@tS!NUa7~G`~7|e(;i&dI#d4zbi2;-ChdD6t<ek
z--(y%37-mR<iW8Kl#S!KPqK33qr@Gz(1Fiv({J*Mh<SLzTtSX7IHkL$*f%}Q$2*ta
zHE){fQ)-T4>piH*iLuuo02%v4;&1N`^&|`Y*sSKuuVk+!uF2)tI5UGIY533Dmqg`Q
zImJbe5;i4kf<OCD=%a+Ag!q&foYYRzG-I;ZX(u+6A%04unA9R5bdP9h#9LVuGF;9W
zUR@I4?u6i21Ort|7k-Q8uo}Hs!+MFiAs|AECB!5P$W4K~`B3r9fKzRZVpFb-&`QGF
zYET2e!g&vFYh7U7Vl&pr=DX+$b?a&YGOBiMmw*QvQxjS#Q+<FW_@FA`+j$HdfK|63
z^ie8X)Sbv-@~a=iKw-vz2PW&=(3rQc*+htSw0G|A9$yv4iH8W4C4F+_p?7`B1QIDH
zd)mbJaQBr6jNV#J0OIRnFLn>`)+$pyOxIH@=2pPpph2>7DWVa0LLzY^a-&n9xY76t
z4PIB3KlZM>AlGM0u?QE>x5o5mL?JnN8^7Hr7ULd8bKZhJmzS5{+Hh^3lm&uClm^@A
z+;M_Em9LLR^C`h9aO>qV?#i?H2w#oROmxoH3WP$Z2z3LPNBE@x-tkJ5UcFbbHt!t$
zD2cRmVZyiY-yeF2y&^GS>4KncDN}l>|35SQtxkTOkPIAlSrIoAn+T=N&3B|h>M2ke
zo|%BdEoIy%HSMeF5WnHad0IBCj<@Q4#|2%kX3nE?iSEJEd$L?@Qbx?NUt{5%&HLDY
z-M{uAW41nM->g;JQ#`Wfr<^#QF0-fs_Q4^A1d07(Y3F_%$i)<x_i@|8Nu#yt;=TR(
z2)o26_PWpd-pNaF;3K{k{{#f=+Y5Hf5QOdnsKw}7F+aTHgMK_X??<4GC?{>?ew|9O
zk)at^n_<Nz<Xl}M&!I&>XVX<Pl+jAZcp9<H#x)CstqI$2{r9Lh<>e%y$@!$~CsWez
z&SN-nCng3*0&1QV6=E;{x?Z#_6<1R*^!~2ADmdv_A1fq!0WTrfyHP)yKaU54zuTkn
z&5N)0Sb5^SO>;VU6LRvZ*(NcJHSUYZdTDc^3EGrJv+#Oe#~x3#5SF7lx`@53^I(hW
zJ6eV4!|vok&_CU<g3~-3wc?!``0Gji%ePnYkvH_w_RJr$JZK_fb-0Rw-WGEg>XWf0
z69YyOa*B-20RD0jIzd?`gl%*0EuX#(cr&(e&7?Ftf7c)14D-PJ+zLW<Yuf9_Pd>T7
zrRI-Ue(N@{OT6egE8|}G|DSyYrX0C|RF0`~ce^c{weruZL~WUq`#l2t%0W{Q9CnN4
zf~6WQ`%1_>HFSC3SO{GkwcuU|<^kks0GeWvXeugTGAcm?OwI0%CiSi8xI3Ow@AY(a
ztwop1_K2r%pcH~5AY3J!rQTwWGJ|}vK)A?M34$0Y7e*Mxc?QyJ(r;f}Y0gWi{U_ST
zAdhH=$N<m4duo(hn1eb%=N87{#>YP~*=2Q03yL<3e4D1*EF725jLM>T2Cu&lB$ET)
ze+5SM(to@bti1LxidA?W5luoP=Y&N!MZYRlVdAQq=T}KZsbyLf>TT|!BiNF+m>}40
zUrUhqh$Sk~+4idFG+fxVq$YDmfK@RMpym>oE}ge$l>I}Wjz*4LEG4JRv_)g*qxph)
zO!VbD2-F8MJ5AySGWiJed3Q1azE?#Mr4#*Ka#}~OpF4>HOU$mm0nvQh&)I&9p|*9`
z1{&&LVU;h(IGkQrOz-%tR-%EBg)q`6NhG5(@%v$Tmmh1ecFYg+w7<4cU0x!O5N_Vv
zA1f+EA23|K7iAl(hBRqK#w{FdPWfJTKw}cJ5<20w=JDE=zV~IobQr1u_%h=e_C@~g
z;dCYQCo7Pt^v#;I=s4(-<L@WUx;(m&7Ofl2m`t&5kpNLjXxF_RCP8Nlr{ySBT;Km<
z>m8#j`J=AE*mlyfZQDl2HaoVHifyA~+eUY6cE`5aF(?0L-g#$c&8%9hs_y-8SFNgB
z=j^l3{_Vqq{8|VmGc3r>l&5)?_}{yK8)0T5J&s)xq^529*Cdg2SXn!K>A92o8Y!ls
z@AodikY05--Z69|5#ZCIXTNVcB^v>lz2~kTGn~B5gd=KdEMHg^*~u{VZ<~i!sRYN6
zp=xF~t*jUOEm?cDi77z`)9kSz*A_!bL*cP%@4{~tglIV;nM5`!q5CN!_!z(|TZrEN
zxpz)b7zh0}6$9&~X89FWt~TUnT{;}KlUAIUY7~yBez;TWA(yJqV%Z!`1UCTGWSZlc
zB}=w%3ue?KXHo~=E8$=AubAzc2iIt|FC)eDl6f!XIG6jb!31q0y?5gq`GqaN>RJc;
zYQgfCSvP+?SG{Twjr>ix+VdlU6dTJ2IVA-6_3yQ#;V~n>28y*XHibE&C31=yE)ZON
zl88+pA*<-K-A8WUP|>Ips#5@)Vt5RXg5DstT~=Ss@^d@^4Uyy-3nh=-${b|DC3h!`
z^^WezJWaz_vA|S@I05FpU;&Q|E1Sh-PV}i~*eoj<AL)Xl($3w>jR#wg^#(s5z4pCz
zwH-Jn<#D(PI}sHS=1Ck31J51$5Uz#mOEif%;JzT*epy|0K01C-evHWCvuS87#)Zrn
zp1&EjONhgBIyBYcM($huSdXmaT~*W2q8J!Z7szV3?<1g)=BM^O^V0WrcUe}Dwkd!X
zU7fOWl3KlVftP53%$BqGRQX#a`d=SDq6%(r2W_37hII^-pK6qUwrU)eN#u8K;?#_j
ze<P`pICjzzgJ~i0yGdgEk-cLdJxd<oqv1|$)aE4awb=_Lg*4&TePa#hH5(ewUK+Im
z;9;XWQq!GWmd=tZ<;mw0to>wXZF}FTD&&3@P+e;q@Y}S$-i{og|Jukqih6eW{*yMt
zC{RMTV1A;fhHpcS=(sKq?_Y=H!YiwfZ#k`LGWFx|42Xb?K61f#Mh&=HRb<<!W}!a#
zK{3=@Cq!YXbaB1(5M*I04FN#XNy8)w*kXU@lKY%t)4ny-fMz35?c`K&8xDe@`n<Y;
zRf_kl+0d)o<Ud9gr)$7gAN#9gnR9z|7=uRSa4*2=zhbWEqMDhx{}I}~F-@shCTHeB
z&tQVL!4klu4a(0H@ETJ>D?*-5E1kt;0!}p;SOAAAhMBhy$hmbab|+$mI;{-=XeLfM
z;XG+z)INR{Yi0hp6(1rsauxc(45=diV5=+H<LJntWyh4De#W}qy*BKaQs=``G8t1Q
zkM;DR*)O)FGxL)|n5agA2|uDh(;ruJ9}rW=VkD7sNHOl*6iAdm$hu*FSH-eMee_d=
z@7mJ1#|R7!YSJ%f9Z)+CFVgP<Ef#(L^t0gho~Plv){_BYxLkY`?@I3G9#`NVFZc}@
zW<SAqcQBiF3j%b?Hqx^$2!F_FNf7M@tkCo|Z#F~`VtKU_!VzcnfhdgYzE|B9H*9<+
z27SVH3(%qk3MJK_J5L9jR3dA4t-}sLg^%>N4vb~fB_X3>j1)`s!U@9wCiA4+rTxVc
zEKUQ(5;4ti=sz$c>wd~a$6;kFbedW~Bw55Gzw2%>=GzUTL=R%LnB@;*aKt2W578{c
z9LtL2{;`QGK#s7|B%_IsZ$WH4em9Ac)+I%k&?Ni|n=Ru0DN0CKYC1qqPdD=AKdm0d
zWLByelXrT3Tw%&&?#6uzz))eAi*K9!W}NET&FS$k2X4cuaFqFmBXk;4=px1)0#F7c
z@Ff2-Tcg9;<0udOcW$yx_Zgv6DM7~OsLW2v`CFW<`^NaJaE=7sYw}&4vQ&h+v1W$>
zj5lg5Dl%A~A)v%5??hTIGv)FXB_@g}?I<msys9PxJCNJjVg-o55G4{+9)7u_Bg-|%
zNW~7L((=ZA$lXqKUNwn#$F{ap{QE06g@pw%*NE@0B-sm=IraMwXQfYSxt5LkMYNr?
zkkq*wr)aMCHBcoR@d%<km{IaoD=(4KWxnEU6Y(;%4Q3*m6PEV@LT&WH!?Khzf0Z%O
zl)m2?h$rwd*(t!yopOh!Laa4`C{b0HWflA+Lsv41^dCfjI@m!TLC}aAa^kL=3TuiT
zY!q}(*2pq{>RQ3+46<Vo)2QM<hgRyK!Z5B5XhNpvZijX932-zAOvSRX|N0@{BBSBQ
z)YD(kez=x=Te)D=rvHJTuq<5o<%;cJ?~TC0TcdFVe|7<Muh>~paGq?wA12>vQ>ts*
zqgQNQaaMWwbtt|NJVg7jR<slwD7n8$(#a;S5Vow?wt*9C#l_i^H>a;&@@9w;ew%{m
z=DbO!E>!SX_17(&i7~9wolf$2L%JQtu_k-yFS>PkKMK}wNtlu;cmUT8dNq(QC{|>n
zs}**TyXgZ21YXsX>Zz)fr+WI!{3-QGJaT91{sg}>57&g>$0%ZxEfjkNXS4nT$qsId
zM*)|E06&Fu0*!?KhLrb{7m4ced(DNvJy-<?-&pLUSJU4tIV7ZlEW4JB8&{UEpC%Y1
zTi7?A$=qP?FjO!NS29s=xn0tvwvmg%CSxk2@jC_@Cnx^NDbv8@()TsWb7z0;zTSjJ
z`S#^%cEsB?g}2lU<G}j5kD>7X0aLvmOnIyVlt<MG?`8IKBKHEf#_mn>ykf#%el}wl
z@$X8uWm?eDoiGisv51Gou-2(pX{D6J!Os@8B??cld6Sf#g9C4z`AGsL&vqs!Sq|R!
zwZIG>6Z=|=VjJDkON~&(-!nIBg)v3_<dRh?I+xfvtxt5nOge*f8eepCUV>j{cpi3X
zg~LLHb=-R$RXgkVGv`&ov1VK^>Ao{-mjxh^g7{x?sBdHB(vOtcnE*0r*$$6s|0f!>
zt0Zwi?>sDgu9^o%eE!>(Tq2ml%N@*q5O4~cz=<W>l;K)dmMRZvSvxTfZOG&F)R30Y
z!x1tOD>-cPN}ocf(g7vI)#oq~xIF^(A*qWLpNn;Ni#m<PkAQfS<=E~}Ja@64Fs$D-
zj4y}9xjy-Evqi<9#fKW71wmGFmg4G+>WCG|2g-2yb8t1rP*vIK+d^?zc`G?&G2m?v
zvOHnl8ZVGk7&6_6`Y=qSL=Z9Af2ae!fzP=F1_JiAqbgy3hAC$s{`ztWxkd5pvQ$A(
zu7<DUQK2o%CUGfDTff8O>;6s6+@@iW>*`7_NJbyoK%8Dfk^o@I-i7%i&7AA~cUezX
zhgbVML;lt*{aqz_I{99+G)I5~ep}F<Mu<|4C?;+;!p3A_R)uPr3?e5B4Bv{R2q<^f
z(O00J<LAzTzH0bd@A*-&cR{jANSLK`{lHwnHnJwxEx->^w|%iwa6eGN5fg4XxU%Bt
z<xD15&7E~8B;!ZgqQkFQyJ^e5&-f?tp_xEFTWsJFMu`B?kOV?Ega<${%=%io?F&Z<
z?IZukxq#-oJ2=!)Au%036;wRG9x^1#6oefinz@bN{sVQwz~T$YJ#XtGHY%5GO&+5M
z<w9z-ZSc{;c&HjAGCGq*cgNYijQU5Z<?V_&!cn$ji9Z~<RB6J{!#(vCYPR@+Kx-W|
z%3S}$Jq?=wa<8CqWH{jQTbRNpYHca_AMN_7lqzvx#aS{<?F*v6PiMI%Eg~*|l`b3D
z{=oqs2h`Ij2z-y)?+2O7@#3*EN-Y`K);}JN)%IF71g6EW+LsTP$7?EvAMMYrGYUkn
z#ufT)z<g!E=95=?L&q6*n17kSHeDJcQ=ro;*Ly47<NjBd7y?*Gqvzel&XR}W&r8A<
zRoY}NIQf>wUh8(L<EGNWDD2RcL)+BpC&8u|_d7bXJz;PR)${o*h1-Y;0R64YedI9$
z<3PJJrRnSyf{oRkYEqy_TLNal7e=lI1l0#J*`>(v`*|&K2$c^HI!h)Qc#Wz)I=a`@
zj>4|3<;9ySXaY*3Kxh@5f&i-!59|MGoOhGJziPARYv6l}Bz8{Mw#@RJ&x@VGIFS0t
zrR*f|SF)x4NAGcPQjUyoWl(qC<}0e@c)OK~d@-w=q~C1a5FGw3{bK*;WCw;~TFJAv
z>}s1KUIdRj&o*o5Njw}Xi{2nB{DfhgG4zKv5=fO4or7wT1d}U$(j@9Z!le=VEA)({
zB+3O6om5OBE&@SzA|2U}Yi>vpFEK&JL}--FkTL^*w62P_15taSQmp25PpY+Ah<852
z@GKs?>Q$?O4$WxFsXVDe`6-fz`zsAw6w*KOPVSougTI99VlzY24Va%Py9S{agO>eT
zGl0-ewe2%Uf%7jkwxbA?BER(}J0awzhm17j+EMdcN%U=1qiDN65;XDp+k?lP{DaJ=
z%XK-DzvF_u8;e7}^^(o&+5Qt3POQ=p=Wz7y@uj;os(Tdcafb6U!~mTU3j`P{9||2V
z!AW4hKE9)G>!1W=qHPuI)MU1-x2mzW1;EQzs>b=)<#9_okl@%Mocc%9k*_lQ-Rf?)
z(Np=b!%XyCAYh`!!rYFn=Vu<idLaaV-&d)=zWM@Pa`W22YYZ1Z=aulgmhSQ(TC^i0
z8shP8U+AmOfcEh4B4FqHqX~Hwec+6X7O8t@S2<%smE7r^?ZLZ>DyQwZ4n?}g3>fZ*
zu~>Yg1al}zOrA*P@LVFS{eAWpGGHSF26r_*hC$z3B}AWF@zNE)Abebh5B{d_X}#L^
zjaTg&u_6VlXEKN#l#8>Rj+)+mN%yiwV&FJSDy*Vg4ws9eab$RPogCcstBpoEkq!Rc
zCB$9a*B!!b7+Gu-$vG|pQ<Ki43224hXR-r%_yl2|L7NCNF=w_6Y{w34JXm6!{&q`O
zx|n+&9`JBi%*K%DHSJH~F#69X=<uJ+uFMTz)0U`x$o7CKLo);f(UVAcgbN1uUcT53
z6hY%5(aglOeDP+YPw~b{^8WxwyZP!rBU4#_bzISZ$Z<Je4UA`mlw>&pM9tS%9`nDl
z1?HdWJ#^=ZY{90^W;170+Gx~uov%Kt&F2bHP;{5eobti4wr=JS41$TfVd~1SO<duM
zRLcqXzAo5xytTv7SFU{!Mzw=ai{1>=$RU8)TZRs!E>H|&f7a)>wB7gn;AtQ7YvbaF
zuV^Y=laaWZPEFdvM3@My3Nfqr`1TW$wbYLe+<ym|LkPP?*(3|#n)iou%$Bz<`fBwj
z8TSZ{Ib?lgZ$P<PTXl}D2CypC;tVE7JI=e|z5D<P!8_bPtM3980Ogyn1ln$vK_3vz
zd_WnEG6rS%BI@jQ6|C9HF8;$eTse%{y<{zrV6K=QA7!=lW%z`h4s^UXYp@OB$q$Q<
zh$wkDdiTP6iSh^j_CQnQ;W#MZKXs(o__<@a;$oL{AT*wffrZ?-LneWmq~uL1u^pm_
zkx1I!UH=`bE7X|yiZN>$L2Iye$fDEvvv%zv&a{3iN0T;m`0TqpiO0Gv*Jc}q^0U2M
z`>={A;|}--sHIGWls88OMY7J(;X^GGOH%~9k8fv#Ickkfqw0|m+$1xA_U~y}&B{&P
zuYCINPR(t;hGpe#@MntGlzdM~Jf3J?(VCmu>icM;<6Nvv;x}-R%RBZVP^t`+O|^lu
zq(1Ls3aK;oj_l^NXY;Zoor=u<FP|jKEI}M0#D%6$bxWuwn%51!84Zl@u*jnAv=TkH
zzjwMk*z_R`wXVV4m+$QXr88Vl>Gayhzez?(YgOy-A<`G#onXc`S)7={HEfto1fzVH
z`(*A$^**5$bZf#l%WxIr94B%9Q`(V1(Y+ArnZ2G9n|6-}yj{XS+yr$JSBMur%58=?
zo0F`}V%fpNEl8ay7-)Bcpg=xw?jsc}LhQ=65$ojmLQ@OhQ(9C3Zv5_@Ec9*)T7i>e
z9SmZI2zhv>wUcf0cnwVITV#3Y9r0eHz>kFzm(#mbaL~8)_bjoeJ9QM8tbNsG_g?Sc
zUEu;oO5YDdOH}wqR#ifLG#90$Au^QRjQXOV=f8VkZ?{5(G)B@G@=7Z_u`(^OM(EX!
zJV9#&ZL&V6<l&V9$%Lub1FkB~3lJ>}zE!*QeQlW%Gv@2oSUv9`RlA}xD9yLC<6?A7
ztj_W->Ob{C(iyvhzO9PDZ7-&#>4L6%cP!UNAzi!{srXoJ75On$`XPhZcU#0LFq3pw
zggr`+58-o72c<{%;Cpmvi81O+>vfe4$VZ6RFNcG{Y<Z3YLnywezqh*dCjTfrraRvd
ziQg48*d3Sz1$?X4L%<RStqvoQje`KCb?v}0frjp4;4nMO4sEKFsPeV+(xv?!St^N)
zh>uDQq7h3tFHg#jp7`0T6iX<%U=$S|EYGFr0Q=R@XMw31@4^P)A=lYW>sDCS#)lW%
zU}0e9h?jo?czd)^wAi5&P?mpFr$!f|T$SC?!KarxaO#O}wz{KnA`H}4UuJo016eq6
zB;Q=A1v{MWoviz2=2Z9BsL00!`L&&06g#hVO>XD&G>5>iylasVI_olT#6KNkMz@s-
z-tynXF<#!kvtWuM8&ytXq>Z)Zoc+Uoo+zx!_3$SJ0&tJGub8`Ip+o*>gSZT!Eiq|$
z$m%mOEKYZpka6bYI650~wEo4}h*13>3qBEBWHF9>`b5w1=uV+;hw(4enl|74rxE%0
zwQz9T4utu+<PR!vA2|c`YQ&H!o2&{Thk^RlI(h{WyD<>GilAfeIT91kXM^@H@uQO?
zP{g;8%Rkv%@Mb%ma-S0S4#^^l99pIHf7Z30?oUpA8q&_2@nOooBBeVGyexOgszPYf
zIhv>G=_bk71{ml3AjL*n82m)krf!!Q{k0%nFI~qFuo0az_+lOpTf7a*><%a1)f=iK
zqrSxkbGvz@7mjXppz}e@YAtkUZ~#hWR&rM16heP(M#g}+yqkTcn7h)92|hA1poplz
z_4!lTeug*^A2+eyW^uzbE4{_}x$zLH?O*a<4`u}h+-)Px7r-678Ji<SQ9&V`^@k4e
zrdIH_DzxIo#Sz0qOxOc6@G8YRwc$K*f13GL;@sN+^|?BL`IM7~Z|>VC2!I()m~-0{
znU_R^KMj2v)jW9l#TMeTT|r_Ms|L|IcttO6c+vSGrtV_Yyarqn7Bsz<37(k7dcXef
zl4X(L@@y$tH8&({3t<nG{FGrG1}o!Cb>3!ky)`RVD6M9?ai2@!GZukW!OvB8bM6!R
z?EANZRrC%zq6AEUrTO7f6RizK@@LU+ud79J8TdNshY6hm!tbguWhA6LW-4c2Eb(eJ
z6Nu*Y-K)Eyb*qn_QoR?-d9(mr;it@LGSiesbaTxE%-w-O6Qf7X&M$=eHLUkyx)vd*
zQ^tT@X`Gg1aWYU>9F;w0hzd?6zIEl0O%T!vs$R+(gdzoiy~%9jf62|fBu>n8GnM<5
z=Xyd9H*?YSd9oEE7EJF_w32nt9hUAUjoUhv?pA}@w|Lv6EKUcXDyAb@3DN&4w@h-T
z#rE}9&VajFquGfg_`Q9uyrLa8r`qoPniuQMguX)Sx8}>0t%QIH#{**|9^2=sNC87!
zWRvCxWHv1hKOXCJ;fWJMP|#eX+0>J5qKEt*N@i|l4XB{9yM_G|5ZIGk<-wqD$^1#n
z3_h><)Y*g_KJ?L4i0|jL`0)<UfHJ0xB;j7`n4@ED&P!!Y=1=h*jGEoo+wU3rL_eKE
z*$MYpI!Ev*#v$B(x$=fsA*bBdNza$YAJWGHJ~C4KHbf1*Z7gwn(_+bGkNZn)7E6@>
z7@y-_1PrmyMXS^BA{UQ`9AE6>=~<LxZ$<^1@Q+Vxjhqg{C<&9N(fM5`532co!d*gb
zg&BfOBLa3jV&_CWr$Z(Og?e>$oH{u^xd9Ey4B%3Hi&-XQU1C7W=|R|2F6Mhf1vL(!
zzQ(vSlZEq$7(tyR(J<d_)G$dbFPFhZZ_h}ZR+YMF&I%$~t-^~bu!|s!kqXE@1%#zi
zq1m3*d5dIP&Es3Tc#+eQFD4;UL8FEt;_h;5v)tPuy$_R}Bzn%Ur4B}+quufO2*L(f
zh#IjcyXq6luocJum58{Nf%zsD9%~B_x$=oZcoRuFKue({iAv*P=CX$sCxUmYROmgN
z8wW-kLG#2AEG<$sOJlr4j(PWWPDqP*adthmjyGv1G~QKPDCVv<gfLDBOu8q9YGS#=
z5e)g~VHLZbBymuw&`^_;v(d$<oR!o|WPCRmuYFaw@S386uGh0RuM2x2_r(O}J|o@<
zYZA_*xcciA2$#*1wpIs}ELg}hTFCy;oJ&d)i~sxx<)wIeT&2He=iq2}i7eF3J(`ow
znJNxdtO{VIPG-B<vXjDqG9ly;jEn{HOPMk7LsMGPmN~@#+Ay?ebgeZS&PXsWJa`}t
zStzuxTB1OvSQGh@%8G7$nO6WKhK1@&M4TT-m4lSE1^PcP{&8OMUd51|>^q=1$Xju)
z4Y=36iak@_bzNLeod%yF%DMVjM*4at6l@9qI$Zsra~I(@w)w$HM-TsMVktWc&s@D+
zh!v4u#~6LE;9S^bJ#D$$NOHIx7%B;xG3PYtx~-AZGdMOFcE=&;$sPqbm?6c(n}C(f
zJIwYpYZ*ZbY$XG60~X-{PV)w_5Z6p+8f;)S%9g?lWftkoREQ&YFX81LU6IyN3mrgL
zr%dnTf^~33PluXRuFlyAS(NXNpD8H(tJY^m!TI~y4@x?2$6pXG?MN-6{oh~?f1p8#
zwWwfb!={o;^3{%sh)4mo3stceqI{u61lnBtb7HGJp#t}w<AO_8&`7GuMm`SHMfV~t
zwVK2OzVwZDSeV?A+r&R$Z+BNfr`Ap(F$pqU9PNFAMY=XoPxMWjn0&?Ps%Fx|4-kbw
zbH$Qz7hxLW-i6oy(QABKd1T-5HvYMc3Kpi0E-Cyj(@a5w6%Ya_)_q%t>1XR$tzph@
z0GDudPV=y?w$@t(FPj|7dS<rzWAkAj&u0Ce49NQV(!C{rFQRFUkhCG?O_}S3$TK##
zUVpV<iW5s*hBxOflt=P=GD;%^fmOtkucMWGA`e<dRS1<E%4$Tr*Edi>x$`Y{<H}4G
zd44lDihC$NglGXG{t3O2BZ==jFKK0y?wB+`9m$F^tW}^zS{hD3apfG+@F(`8lzqtH
zLNsRLpW=*K)S9y?on?xIT;c8jSXmYu{+ji&jXyKbd+d*1IVisTlObm>aU5;kOd$Hq
zUK0@rh<p#;Z-z?=zzuPI&D(~J7d0&cx+`@1NA3Lp7-!(B!&x20Ml<!mNNICa;SDks
zZRV|N$fQS$j@K$?5V{+mCuIqH5SAZQ=X(pwGn<C=j}PA{f2@36`T{LSzn2>kk9)I0
z$-OV7`y7rOn@wJDHNSDJILCZ6zKQm4C^XN_cmg3)r^@QB{7_0dNh6a!&LRH5A+<jS
zlThK7-_QZlQ*a?C!$Idt@9DKI$JY3Y)z(``+MrVtf0MrF%G%80Z+_J%sxc#N^B7AU
z^a2@qydEX)nlC48AxUJj(^4hM&OH+I4g~Zl&;c(IdYSVm$0OnG2eSZQx}~X?b5iE3
zaGTctbkvZ`;*cdvyaC*yWHgvR3ws4~I{R+HTp93}$VGrZU3?GqnGY8QXaC9!#I1=n
z%T#7Au@Fzu+Y_Hb2w`VN@m!VN8T%t(1X2V^DnP~g9aIGAkraA&7&0GaznFJ)1<taY
zj0nb=c+?nvRy_D;)!~78Nrx|vlsN+YD~l<y$lsw%PRZn*`oX?J`TYiEmz;RomR0ln
zASwXeX&!ZGOjs=;b?)3L2L7`@OLjyG;ni@^i2B)bQopD-IfU6OcK2)|pG$m$3=&<z
zq1?(fUJ5x-Q8r24K}rO?<>K^?|A!sZjDD>kQ578JFL^bFLHB=y|1#(=vIF@o0+6`F
z;n-sxNv7@NH=smWVXO!-UA&5=A?@Pm9VmeK>WlG*y1<$-n#cp+i!OL{=>eB}*fZ&@
zsRXC^npW?~W8)XYH`ac5v^Rxn4$R@6YV1+?edsY^GMN#tJ9+;>a!yyjCxXC$533LT
zp;5B`M@rlk|1TwK#7q2_5XGC4=SSz!kR#~Hp52tQ88+f%=`|ZHfoZv)uEQbGn50^>
ze}5AllROtddJ^kjZ`~Y%fh%}0RdD_J5F*o=jQUO+ohu^bX*fq;8mp}M-);O7!{gw|
zcySzkN>g2e2u2d_<K(W#NeI79tRy}{IkiFXcS9ZS{Z3g_K6G78%OD~S4s$UmlB_%N
z0Y5n_OOYFjW@QTo3P{6sFYJq!2%4eA8rGh2DYIUG>+%$eL}f+}AQ)}^Bec6JNRltN
zMY@_XF$w&KCkcB)G8kSkNNVlwZ;EeWU1}roH#m3H_`Kbn_qVdP2;s!=2K?Fgp3B_4
zq01eynx7kgcvcLdyrQW*#)-TBSVkhu^Dh>$CcBaGrDH{}05+}nC30mgTN|iVLTLz*
zPvCW~E}KVdie2}9Kb~EV>l$x0*~eQ`BRRZ<utNbPV@#Yl|LaKVmPG<#HlO;3ZDcjx
z9$66H5fRHPtyMTy>LCxkRTb7$M!8fhxxM)wW9MDMPBZ7BiRT!=WuXD%zPFI)jl59v
z@3C92@8$d?kX{i#cC}G3dx@k6uha{d!?X*6C<H5ZSF>@FG0}EL>h~{y1LM^s*&p=1
ze+|j2gvr%m5F!bcc)d(@_j13^!@!8*YT_;6K04Dx%YxwCk!W4yZ0nyRw6>O{qlt6Z
zsr~b`c&#cYGyR5y=EF8@F2ZYj{>+_&cql5?e3M22;NMJ4)o$_gpzy}?`?$NE=Z<s{
z1Cj`RbYnV5k@Ox5VUbU@GOxnq`)EiUatf|NVzC0zQAp#l!942osuX^qakoaVZA$u*
z&Y8y0J6yU4gp2{CUc`b*QEu_7BKclk#@(=|nZse3|3lnt(zh%K*Dc+I=E2j{@3Lq~
z8Fjjbzz?pvKg(HYv<<B$f`f@-#bI3Ry?zZ|%C$=Z*$xj};4~2wxO^+~eGNHRG!KT6
zrgKedOk9JH)yMG+5>DcMQN9Z494v(0@|xA`Oj_M$cTPlNzjrN)OgeOa=u?6lu?u}p
zRUD(1eXA>fQ|*UlujWsr)&LgZ$HkKmx#fa<0BsV~Xj|?Ze;iJ;h&{8Rzik}{&2gEs
zq_wb-r8&1+VEMQ|;w_SDC&+$;-6eLyu#AHTF=15>qDM9bY=+OR^`Szzqbv_8mM~+o
zZyKvqs4;TwqHh1y?^yZoFWIU#4@ZB8O@0PRw_xw{SP)f_WC5AWbQC`Y*5Fx@9jn&|
zSjT<|24x+Cyj%UL`9-j&i7#kB%+pK+MUrQKJ>P1dBYmwHbN3OulX;>o{6u{f&~g8Z
z=@4R^&0lX5-`f(NHwi~J!OA4WhZ1hJTtsiwYvh*7{Z5nCIZ>3zn=J#Mv&n9eu;P6y
z)^QQhFFaW^Ox19M{1+oE=wDf5o1YUg;Qo#Dv?cB6hllRNSqF2ff)DLX&sueUBMp*(
zm&Jw3_Ik-k$C}V(D>1F1KGuW=0)Zwo(#A8@9sF)x)n08ALtH1X(mr!P<k>zY%Z3JG
z7Dnw<tcaB;cT4vItT#L3Mb^c%;l}3whBgL?)w&#B&5Y5Yc7s9a{gD7xRFX33_%NeK
zj)_}t5j-C2U74X0n|5F3_lgo6DF_i_40v;7oZA8JKi+f=`Rj=*8?I~L5LWNwq?6&N
zdan^{Jn`R)Htgoth08WDp#vW~52AI{rT^Xy2O$>xIfF9%{!jkQ{mhI{q7as>iU5<P
z?0prS%EwaO6?v8YXz^=I^OZv=kc6FW5)Gq;KBpnaC&D@~{<j_Z&i(n6^Z7VeWQHO~
zC`%D!N*z=`qi{lJlLaM9wnN)E3C2W<odZ%GB;{B6oS#CV(ru!J0~?*f)APjd^RiNJ
zl$Hon*8h-2TV9`zc5Cz;4h}zbpfG)L2&i*HjANC$hTCyoiU>fw@;fIk6SeQk)I5fR
zzc;nCz1H}o?+v4vMDIrgaXyKa0Bz2q9!J5C#ZXsBx?fi@qbF{BTXi(JHT~%M8#LR?
zm3Oc~V-L)@yn>!{uY0i)@X!YAo^$J-a|j7sZpn$p2-r;ff=3E#)yS7oh~I6teiEo;
z3AyxWz3Lqw5<~&d$T|Xu$F*Qp%T{ktxU_htKP;xmlT)dx6U!yN6UF4)zz}<x#zlf%
z%3*(I##FWnU_+IU66fE9C#_f1KGi^$Ga{)k>7SG-p)<}VhzQekY?7X|A-RWv3;T)J
z%Kw15+Y?)NKK_-nf3KB`@7BdN%b?NGl*Ghywp$>*b_W1eVFlChPeZfC{d)5eWRgVR
zPW$iA^}ycAZar9u>jgvj5BUoR#;xKgB&4FOxlb}#_ASx+1trKJc<{aTQ1-9)+))3&
za3IFQ7wrEJ4lLln7v=v42XN>97aRZz<(pkR*~5q5TZrk3R`zhPzb3M#wo3g$^yo9X
zRT|R|j!XIi0H}8yAIFGu@zQ<b9C);N8+&sDW8NNdK~PtAoJS%yKICuC48vOL%+Kb9
z)+C&U<hf6#KR~F+KQeBy&_UVjnI~3b`GV@xpoRaHx0{l%=xRxmf2sJDoy38!wETS%
z4H;&L^eZiQ1&?E=V;^K;GfDJz@$(tTqJY&$$+UBAFFJx86MT;Z7q0pFZzf`~j5KB3
zq`PL+@^Wk$=KzC#T%1tGJg%U?X8rgY`ee|>JT5_VUBeT9Ao(9+)*n(!soaO@jWCkZ
zx6ceK|KACm{s9PClkTy2J*B{vU2a{T3L_@d_@q2!q%ZjVC8zg>hz~PHSM1e~!Vcu_
z1v0$ijLeMCj_&`-!i6R#pc&#&H^vB?kyMgAkkPEDF#;}I`TtiIPMLN<(XiSfU7XWX
zDW=Ia9a^3OyZY9$>NB%ZeyKSfxEeq#j6!_;1OvvFLdnJs%-5S4krzbX1w{JTY|}7c
zvS9yer|g}&PRXinsAjTI`HL5i(GUs$dfp_nzqgu>tr%eLEM1<H70D)in)anH1W7Zf
zZ>&=CG%MHP6w|CtnqA%u;7UWNXm;uWv2{YSbvj~Suq=mtHWzd$+YCfjnTAF$MBZmj
z!c^;YKMO7g+(oM~BAW={$(S2NNHrwu69<UNGSq%`B~J(ulrX<2SQ$G7-}5_oXQgH3
z%PE67oQw}etBi$n;Xnw*yNZ$Vm%@9?6>6xo(KWvtnGxidKj06h%iV0OJZHVlJqoaH
z1liMIDVk3nB|U;G#~y|~`gHCH1b^(U^}(j_pql3b^5e`qyL#%A7SOYaR3=dzpNEWB
za3OD$aWzd+9q!7#%)yw9)q{hsc8c~ig@jub@W&`8@2O+S2@-REkKC{qe#s_@SqH+m
z{d(S2VvwjMib$M1nP>;;2way>@ad7Aa-~`%d<&FC%>-?%VYD!czucTIRmYX#Ar5iS
z?xa@-bm%E#dyHLvHdX7p{<!fJ+!IPmxLI%K$~e65Oa;?ty!E$P%xet+A_5_LDm5w>
zQfC|eIj&6^a9&U|&RRhCLPJhWMkVY19W-fEGjPt;hl>~7z;c9s33~i?T&UyB5}v(l
z)KFnFS*!_RHz=)J_y|NR9WHjL?#TX;B@0^ukldi>+`#2D=T&KVs$spYf4x7;D^(f;
zraRNX<~i7qj<41_N`Y~SUNxs?Q5t>3X-$&n$}MTz>qU10-z65El5K;(|2V9}&qhTX
zTh@C4t|?Mzqo)jU8LymhJhv4zMr%x^2lrpjl&dif$;zh}-F~(DsWxNN$q~MH_ovH1
zzZK<4r$8(J<o!03jn411FK*^IvtljJQ2TnqWUiA_WqA+8;i7p6i4To)1kBm+;eq~H
zg6QY!7MV`6pOU?#e0Kp~j0<5I5BPlo4oywM*<S>lh3e;%uLaO(lRD#1X&SDtP7DG!
z3lbNFxyM#3J^J;mme!k8`Pxi{O0k>3_6!)?c57MPZ6;Y<&l=v}PQ{n~1;c&myM1fs
z)3J00GPbu6%QZQn4M*}k_y~%>M#sKohsOgvnFRwEwAPmK0$0b^w1xWtVu4~S`?Cbl
z7E8_-d9|R1bA3cZSeCt{1f7)*WiLuoJw7fBkE>~J2uwcteV4g48v=$lAh_=Ugc5;=
zB;Z3sI=fTus)a-@SVJq=)tvmPq0bAuN}tUxB_bA-5C}m69*CvKOIb1>X@&b$5YF}Q
z>Hdg1!%HKaR0^&9(LymDJbbB{ANOvSHS()>MqXNdyag&bD%`%yTg~c56$fNpW!OHa
zony8|7`pRTChdX+k)9BZG07B|v-|?@37gO!*5GW#VYs5L4zG{=$U>28fzWsn1IArG
z!K>g+D0KlL(-%|)j7V#U1;O6$rOQ%Vt2s!O(bDZbU(bfp`u?R);OFEpU2GgP&J#h-
z%=SDvYQc63u0hs*_}zqZ<~Z3qtBNJtj{;NPWDN5-y&y!{=XR{GnOQ*dH_%hY0VHL*
zZ|7*F_uD6pB{f^jB}+W@9#g0>Y^z;q%k$r6hF@aCXSIk%`8XA|gnrd|d|F6(Sl(Xq
zpe51&Tyr|9*om_<>R-)=!0_$`Igiza<shbp3;I1H9h<jfiB~P@U94!;8AwYvF~+=h
zLXq?E@l2x(zdj);KywA~_&Z(*655#Jhw>~7*PpT5dp!IsdFx~z>wS|HtUUp%w&O2O
zg8PO%69Y>Bj_?gRkU`JZc&^t7QYeM!XRi8>hKwGIi_giL&*om&6TAKm0%JTM*Rtn6
ze!OO(E%()$tNnwtS=@*Y{>II>^3l)7i1c5IgsF)nRk9Bhj}F<ug~8>lA<BXRn%5KC
zBNa2TL!5axf@hiTq=ie`lnasIZ2v-w06R|^uK5T%`Lv$(9hAV9=@3RlYQr=c>`4f^
z<?=%l)&A$aZCH#?%rI?W*qdS-*U4Xol0q*Js&xMk@;W)F=Li$&K}<ubJc{<p;r?Bl
z?~hW+oMIfsnPj$rMuvq<;s>{fELa<PQhgpNW^+f+D=NmFS3`^!UH7FR8rg>p?|Vyv
z#WJ&_dv;lX*mRPZ21k&DQGkm{fd0S&<B(2;wQ^KL@|88=u3dp>%(n^Cl%c!kE6N$<
z#9Q9J9-pyjBESSYx}AZ_*Qru`(XXqkx`=y`ADAw0HoXSGNBefDHaS8g^$lhUgwElX
zQ18{m<Xb?RGAftJF)67iphiv!!F|9VEQEZ}C6!$2l8`6ynB7@rd(0N!OYMq$PLGk!
z{8f7{m%7oGwI0`w-MrsHPosl%8(3@sOunr|X8eb>d!~UFg=!1*^D(E!I=r)%?-o2u
zw0~v6!_I)fZVT>4#V->uqp1e4Y9i#bP1*n$(S%L=i1aBexjz}f_bp!>7DX#KZ7#nz
zWyN*kC?l!$Uu$I-+Yt*7-d9$U9T-tx)oyw&rcBFEVT(3nN7ep^4%|Wrw-$MKILw$P
zB!zl@7ty3Vp-9hjlplwR7i1(Cm%W09w+RyWf}#L@yjgsv`-`tlSR9>b^`s#!ImNO|
zWlH50orMSzkies8qKX4ozJM(lUH<xI0CpbkimjbqZ<Ia2c3|s}EHagtFuWaZMEHkM
zTj*xG8@vmOYSN4ou1}&!uXFw9+W;X&Kej*vl1%;OK?r_Vl<0D5bPbEwbj%xbp$da~
zd>vpxrrD|3IB%QQ;<K#a(q9ZRWH>fK)bLh`conoUg${~K1MQDBo(=E(MCX9ivaZCY
zn=rQT{P|Hwlu4BN`I3BSw@z{W4aFeUI<`vAo_M`M{wc&@Kx8bTsZl~NkxEPGnnGYR
z%q?7705!r$qk~c^Yw&PyxUdq@i2EPvIy&IIWVFMmWu8C7>leJq!)DN+0BoheK|vbL
zfZorV^fFyF>2gDxP6e*pJFg9h4J56-yJs*?Ogi~7tUYWy87L}Bojc^_*r=Qf9j0TC
zPJazup?BrhiY;|05DUuZmM2(a1pgNu6w1!<jm8KR@?5jN4-`T|WcEP4;yjl4kq-DR
z@>?Vz--;B9&mib)7Kjqe4b&9*hH#M&zxT3TF3H9)#sL{o`g|tRb&6b!Vtfeu>kW=b
zu2uz#kMEnIV0-xRr$QFf$K^S^lhZJ)Q49u;PC(AfvGQ1Q5k|wDgWckG9GrU#c1iz6
zqf`3FFq5+A4;kGNp7=!pnI$@!yl=oL3e{@UF}WpW>Yu}-k?<~C0<rwacmFzvHby==
z54b4yk!F%wGK8@58yVZTJzig%f3Sh^wCRDFqHz(pm7sW5E^w>jc(JP0`Wo33Y23Ch
zaMP7t+NHjhU(gHj^_J-Jce0-HtRO%#ew}=+E<T_{dK$5jy+Z}(AoWJ;4h7tmsa1{U
zJE`T9269rVB%V*z+~Q!g&c_;uYdeuEV>J6oZaGpL1^#iIFMUf&FAp9?v2TR@k`-$0
zsOw@MRMGwuhEVL+J{9X_e<gey!A}JXE5A0k@u$uk?2T9z`%9?B7D8a?+mh$a0N8Y1
z8V-GI!+W(cV>sXgerICtX~18hGn^Qa25t0OT6yJlw{;H|z0qan)jHvi8}5{1%^JVz
zn1-<boBd=JJPqp{<?T#sJ@Q?C%G+pcnlxI~=NEMNXAd*Zt*SV>MVy;EYug%rVelA!
zt8nbGwvoL6y;S!22X{+t2KkuR!bNYq<z4f<6`xkWN!u9cfZ=y{4}BF2^RsqY2U~5I
zC`QNK#!4h0&1{RHXm5sN@Iz=I($ncH!89`q<b<AuVyy8@qkAk#CAdnPZBsJoY-({-
zhine4s-%v(p|7okvT#}H{V_;<OmYvctf!I^0zpLnv_5$fi9Z}CcY@&J8L&PJI?CLE
zd=64vWb*)=fQ3M<U>1Oi{0}aV;DHE?R7{Q}tQbe9v{c9Bt+S0<c&>ZRtTrM?P0IKs
zIA_@B;$qBq5Xb9!4NjI3L#)(JS5d$leHMk6cVpLl`J*J4?O3P`pAk0YrZDCQ4IBOO
z=pEv`<+9dz6g%DbSFMb%rqx=DS{1bRa<!uY0wB^;7Ih#%ll3zBPYqL%<6erAf8kcW
z#ND|-uA^Q%Jxh00^Hl++@1C2iBS}i=Un#DWz-{o@{_Nwzu6m{f>p`Nul{l2Eh=Yvl
zTUVT5zVT=D=(k9%`UCoMOqc+YNnYg)bI9NHI8U-IT2M0lB&xPLpI!Xk4Z2QtQpBTr
zaDd50W&b`-aC*OdZg2??*@poKUr>+n<Ys;HBDl{$ZjYBjaE}+HkBh~d-AeYJyPZo}
z@39f0`6>Uv_*!-+^1?IsHLWpL#ZmSbzuEb#PPu65Zi?@@vctK;Q`tf?$z$*LNYxLI
zZ#|d!0yt3{!;kE9>(HC4lwSexMl#yT6Nt8Ls{D7IfW`2A1%Vli-7_%x<WT>MQE}J%
zGq$c^U!cO6E>F-Tqy0#S|9w$XB_?%%6Ap)uk>VtKZ+Ss%`gjjj4Q1-+_?RoVVn$kn
ze3w8_28uA5itI#rgb(g4qa>26+pgJ@J;6x{-~U18LHVUdP=bN;afy6=C8BAb9AH1W
z;3l(4{wPep9vp=!djMvlQL9bTyKR0KC2P6){y|dS{8i4T6L~5Zl)IfGfOAX~zP;nw
z7p!F5qQ2VpT!{W&oBQPDM7dG${Zb1=<mOcymI0m@Z`6<7xh|Dz?bJ(^1zhNQ6uF_8
zP$N}@FN#d-$hit>lpY++U6AUY5>WcN({>h2I8{(JYvgKBT|W$>NF0QHX4}v~ar}4u
z5=7NMT%}2#;O?FW_M|alRnF$~09=|n7|nZsiUt31pLH<msFTOSPB?AWt`>1rG}a*f
z62<nh(`s3)=3#an^kYWxwdr~K>JIw$^XoIXqG6E_lkq14GCk<4CHDkB@cRQ-aDTi9
z2&0wo82}4hGU%7}4^m+NFB~QQ0ME8+PoZ`}SLj@F%mlsYp#HT_aZ1b@qN_90Gz_od
zxv(hlTsh8<5{FgXN2YlVQBD0A*8E?tG>Bt1w6H<mF^Pi|DjiC=>>7`C#+y3yCAq>1
z+l@t!vGp6BCa<%!*Z?9|j1c;DxTWVPC1j5A<Tn@ysoFezzR+!McfH!S=MGeLAh`2@
zEJ@gypHVi_#tZhE4vE%C%P!bg6%B*p>I)I*C7bN!onP(|GcTGdBb?>e-@KrhW+A<h
zAi1Z_aC=R$Q>cv=$i)BLFR92n;zr<oWjiL#3bO!D>7m782Vkk1v9W1$rQr%S)04Jv
z?RG5hw}TLo!Uzs1ZxX{XU?S(Zudg!8buMJ(wH-$APc-4a#LgAPjf{Rj0Z`LoeVp4*
zTv=3OZ}0DAI~zOPp7E-nvSjMSd&|oe2vZIJRNVI(s$CvuGd%11P)VDZEgpVHqq>z7
zNk-6a8Je900l?sYqQieFRc=)4Faol)7&y>-n4;kVO8Vi31?H+BPi=ig$(&H|=Pa)W
zX}l^XrdS!gTA$yIVsnfFK^6TToe7H#9|Nv_5{^u!g&*qn{jM6QLJ?BUWm20;kfxSi
zUx9UVxIteui1q2WUy=hf)h+7M{=E9eW3mP4*+nw502nO-{1W}nilj{^mG)C7umhk+
zindTEod#7#Xzmw%M5Fo*aH6?*ozNG)1^6f?LvkmLYCjr05x?mvWVOgS&=byBI`c`k
zFjpn@^qs>87c5J)<!qvCMCdm#Aq-J^+boj6{)Yv{_m#VjTrIL3&ZD`SGua|mTTB|X
zZM>ra9;SBOhyiBiM(F907ms8u_iDQp<FaYa3$?uzGqsu72vIWYVL+gXDN(#Y2Lt_L
zjh}hPl$9i}c&vz(VGw~{Q!#WO!AJUO)rL#ed?z8`)X={C9gQ#N;UeI7c-CM9$7?^A
zPT0h60=TE@$RXNG68z27RXQ$r?FPGd6D&V~>FoCf*!iC(cr(BF5iw&A(4Z)3<f)!N
zJ%o>s2>a|J*!arW%)<8V)<S#>$fW**Ht^~b<L2aa<*pZu7T`qN)n91D>;iY=AWw*^
zjy2OL>@OnsA$^@NxA+%7VIJ(0iKJ4}yBofZK(1y%fdCCZKepj)6sO)or`L4H@rMNX
zW_9`R-;M}TeLk|g+Bv9Yu|OxEo3ajG#1yf9CX+09ClM24aAC6>aePVABj(C%ldg6f
zO<=H&_Rxoyytz5XtFmOhhU~&4X4>#=I8I+DM@jR~2bz*eWf~({nyN|~VVWCl|3%bZ
z^bc&PQO1_(|G0E@TLho?*4eR!?*IUG`NPCm2JdW!IM8blf)ac`%;w02lq)6sKMEx>
zK{wazYgqDY`K*k}%DoR98}Va%UWAGVL=Ma(Dq&~oOwwgSepu~EP3?)k{?NI*v*Zcw
zA)Dm@m0y1w${KTf<RAdg(uHzvO>(Fz|I^>tszCwiRi2M~)?|XY%ab%*1RxA*<t65l
zkGITNPmF)XZ(jXt5RSez9f-kj#;rvJ5(M&v4wK17Ri^7DBE%*nE|LoiV<l{-g=}*$
z;^*YmvB?|fV^B8L|J_#FLk;N0rJSWU-DS(&JB#_|q3}rxekBTN^%f_Q#6zt%vdhbK
zMJk9668F_)=;hd)YWgu(z`C)oa-v1}m_!KEs>voWG*kRH>c~;V-0Fq16Bn`?6Ql0X
z1R`NON>z#EMl|%ggOHrmkR(ztKk-Q?zxvP62S$P&at$sVv%MXn@{|cO`QBXL$Wqf=
z0w$=JdOgEmr$t=A=;7pLj(MCWUc5^OV9Bb{#PNJFvu=U**uKmU3})Wyhly+#ep2-=
z3M?{jggG$@Bz|t1vXfkKx_<9ID)ux*>nySAnJnH~Nm%QFXmD3IZe%+v5&jL6=Ibsa
zo3cEyp?`9u6kmZV9B7LFp+tH4Asp>oC1I#qAYsU}R9Lreqg+b^qJz<VETZGe%VB0J
z*)RO!fMY?64I5q!ph0u?t|KpHROofvszJtO(0tg6)@azICRLsZUat<se884k;!IIU
zWbN{3h_2JMSDc3ZiH6Nus2XmBx|e5%Go@+*3nNA+5LbH$fVFOa+LGFs^Y;PQb5O4n
zw@?#es8TMSW~=V*sc=ABxFnb3GpA8#N-KrS?bNkT(Y0L#E?rQ5VHX*p(RdEWUNA9B
z7locGUX~_E6W$IGlYo_7)lUfhQXlw!6tGnsm(x}$EtDT~M|#E=uq~q@BpKTSwr;<)
zrqJdSTfgm2#5Ax!QhM^x@aN=v*^G|`a>QL^%rC=9;s@}bM%SL`1D)nNV!YfH*A@?Q
zq@t9W2lA|Tz;#rHLYz}Ms%QP!I&bVq=$rtd0DhxF-|miNmG>O8RD!e8IXA>l5tE5z
z`WJX-O<$;NlI+qL<CQG}&*1$??R}av3rmUIG<Vv;!?5zwI92TiW-nrO->zT<o1uAg
z3KC9%<6~7sCMyw60yODoNA)lT9GAFARp}u|blwLL;Kt>bPzKf>J~VzKmBH(+1uHmW
z^n!`R_1(X~X{rJ?Y;XkAK$^F0(Xis{kvMsffz~PWanBZGAyzBq^R_OoavG!|s2tFh
z%PcmuVDdr|4JJ8`F!CTt=LwNGAq;oi&1|^E<tcX$>W3~-=#BN+AKB)zVKb!GekYt^
z7X}`}K+iMJ^kKrR-7Rm-T@ma+-CPWoZ=7dp7X2UI`~M0SH|2A*I5a%Hu?DWr@QXTE
zUb|dHo9gJ}i2y4apCr2Hj!&_MeJY4zdm7_*A`Q0vIRqwkyqCpGSL*YNZ2LMUP9?d&
z<=-2sGlW{h%LRn}8s=W*nK#MlCEH``v0oB_=YhrUpQm~b1xa_1gt!j~``hyaIt|O!
z9NPQk`3;>2HyUy6+81d}f@dyuG%5z@c&#Szhje_!W-vct{y37LSdi}Y{m$3jhOnsn
zr6X#Ia{p2F)~i>kVN$@lPBch*^{P}6`px=K977UuL{$3=SW4T+cL_cyo}eqy$42@A
z3}GLXQ`mjVUiXFcoSFFQR?yr2;YX~yd~<&0FIcGTa@hCQDsnozb?{#3)WO(8^AaWk
zU#4DLe&{b7?vIDua4i1KhGY^Pv!V-b&yNUK(#vcUDny72^fdEddAAss)r_}ip*PYf
zkDUm<d`)bBw3B`4__4@Xv8#yfJ+TrnDn`rC<yff~7NvcbS_9w+t)y!I<u}L^!eU)L
z)=V5iCPQ9a+SvYz=HX%w83>K0jh;Q1o@<f|+e(T9PHum-Gmr}tO|m`*VIy=qB3EDV
z()kD%T~tuUBS%D=Pyu*pZ#SYJ8T6k&`RffmzNxmczzrxuJ58$(RXYSg*xmxe(}awK
zwnyV%mJ>l^y8kG$p}=P=^%ghTP8fDFSArg+=L6S5Xzo)MdrGH>NxR=Gu%io?<wLO|
z@y}jY^JI%OfGwlM=qYom!vI+WZ5&tP<KNt_M$CxIsYa@v@V!n<Olr%Sy^V<Stk?J{
zn|{HrVydY$&yvB*^nbE0TML0N`fBr*J>#o2GI-&Rz9tFEG|4xhS{(zA{NyHuX(LD9
zc3K5Tu~_sc2P%s27j2qcNpWo~5^t|*E;~PKBbi&Nb@S)rvjL~oT3yYfTxix+5(S)r
zOt)39rE{l$uBLwcrq-w?)#4^qZ0$}H&>;S!?M0p+HjOuDd?jaDUIQriWK>Teo4oT{
zOxqw$+!#f*iOQh(h)8XN|BJ27^nIGT8;iQo4>w#VQ0foDTdqt|_@rW6^1;x!se-p8
z20ej-+NrO}T@*E~FzK{`UoH7=i4mfrpl+roXeKvT=2LqD20z_w_Cmkji=jGQ_fz99
zqPlGMS~a=@g*2l!a}R)GOcn8$!2+12=UP1r%rGk_9TuYvm6`-JXyO{&OuKNe(jb<o
zPp5>rzH+AlsKi)h;rhNNloDS$ruv>bnRva3FD^b3&T)?q0)<rhX5#@uuG5rbx~f_G
z3ZI!P9paF|Pu(AMKMSxGP^LY94-?~{Obc^^N1^p+Ui`Twn05pLwrdhX(Dz@`ctxLJ
zc1(+t#l8-EI|g6A5GC_Rv<;n-F>yr4|L^9k+H7c}w#9koFuUCE_g3QCrUYS>d9%i>
z4IrF*tr%gQNw`h*L&!&9x2Ho3V{`z|OTKe|4n5(BoXW5$L?#Fko^TW8Whp9vWM2W`
z*F5|(AI<`ENnc%9Ibrr#PZ}{_;WXY3&7L<(Y~zW*K6lVHqrQ!gvoK7ayw$>3GJ)|*
z${A&vuFFk!<}Cg{D0>GO$)dJxw{7>dZQHhO+n%<&+P019?rB@ooVIP-w(<A8@BbxV
za!zt`vQycWR8=aq_o}RG-Os(AJF&_YnRQ+Lf94or+(xlLjPegDiI=Y|dtcYM=*|2I
zwHx6M<GQrf7M&RNL>(|=LHHh*-+Kp6Mwn|_fSEOAL&81w*U5rBB=TCO61rs*l=SXF
zVj#wV39(XcS;M;)0mcBj?}l{$hBoHEp=}7F)2qta4s9FHxVaRWdjmy;xrEV;!B`!K
z=n#reF!+!I#f!NF+05SA%-#%b+s%J8!4DZnb7z`AYSyaEp#fF1GyK=F=fkaZX&(?A
zkk1Ev$`-)um;||sDxakvJ)OI4b@~25#^^qs+l{>@;tdi{VdV>R0WJF-DY_i?7+jcc
zYN^19VmX~#E01HT;Qt85kBPmCIvKTT-{gwzltyiy3pDIY>fK~jw#;LZZlsKBl*8%V
zv8#(62m4P~`<bnH&tcW|>}}|y3FxGZns}l`5uSzf>~~hXhnz>@2Ayfq4RMt+%tuD(
zh%$j_Dzgrj5yI`oPZA`-jRF!R$&ONdz{J^6UJ8eg`6<=rlMgBciC+yGWCaxS6Iulr
z`+L>5yDh)nk2_&aqI>!bg1z&6Z9YCu!P=Vx!D|@|s*|O!(;j~eQNJ~N3@|D$iwkDP
zcU$^#9>JTX)cU(3ZgUCqBnb49>jn*f%PzT_zp_?=*)AF5LAj~f>C>Gl{_Ub$kd|rs
zQGTbfrsPSJaC|#MrHwAVQzW!Z2c{5AXv?wWic=BPrv^g4oodHFz)<ZXp^~+-9GnJT
zog0|OPK+$~;{quqTYgJO70_GMwZ_r{DUBqYk}f~?oummI*wPCxAAhDp23<fD9GW1H
zP(buYTEabL8#|c;wGbz>W4Q#PudFK4g-Y9Ay$0JnYZrU)-D0e5%P;jDticRR<%v-T
zeb^c0w%c|kDVKNDR!hP8=!!3(HE5;2vW{>3J7J2nO2Md^Rt2{o7Vx-|O$Q4dQ&>ko
zo_fG!0MlSLB?!M#8sms9^9v5|<<ECLd0JB}(mLNl&0Emni=FM${)rPEy__CZNzF1-
zRS7Md$kw>@Kc=}FHj!Tm+y>*d0Twk&$D54kuMh3+Q&UqRb*7S$ucR{Ix3TV;o=@Jt
z<+nB}rX1iOsvg(?4plWCJTvnzIww_V!5ut&u)Tt9B%4BYjGLU`clKDMbh6C+EGEJe
z0m5)FgC?vh3*g$4+TkOt|7zNAp%|STHQAt@z(Pt$VAiB^)~1LHz_~)&yggXZ0&o5P
zhVXF?#Z=WcnqG_oW<I*5^~QffQQB?R{MnT8Nz?lf{67J_Yr%cT0(7O=O7QFtdDp{=
zP3O9Jb892xvWqdY_!G6vY?mWJQ7Q;~0{xFKhex#ASUAGqAU~Ob)1{&Y&e_g*CrG3i
zDC3+fM@Y7Pw?F1@y4dh<WK;fvzvqG|{+f%LP%9_^>;7;<0$Kz!Oe(?RJ<w!>dF^am
z&V-tT&h!QBU&*FSI$~btiz2g&+qYh;2FHwPcVca#{``yggNU-%GIH941%yHqf)C0I
z>7MZTTlvW_=ldhw4Mc<@04Sf+DPv-ccLV};_UU|FlSK^r>kE=EgcryS<~lq{H6@7W
zbmVrYfOy^2%E&RHUaajyu)6+d3*j)7Dq>H*GlK&#Z`H_QaplKzUD;ZCE(D7IVnhjs
zmN8FdwZdFP-mL*9_8roO3=J*@B^A-|3#^FbnT=`Sb>QvvbVn-#^C`F};<0%F(ah5}
zn+$YdGw=^6%o<?^_n?Q~wg>Nks1gzfQBV<2_OVBEwe9enR#UYoT8`))8gxV~tIoR2
zO*lM2=Oh!fe-KgDfTs^<pU95-zVYP`Sr@kdNv)mUyarM5%YCJfE~2A3A~w=YpGU3;
zat_iOO<H_q{%n((OM^?15mgIQ`U7h;5GqXEAk-}>&%o*Xr9Az|yHv@Y`EkwP@d8#X
z^mJed4uUwWW(njfffvvef4oUqj>@?_$}kiFD)vsfF@#t*h{QFjX|W*g)3@v4^0GAK
zJDtlWy8dCgibQ>`((e;YpNm4@5#q6uC6d5FbU#|TNz+DMxd+Q46^XIpc#*5mPjEvX
zgQYT{N!;<#AEu#g0S8?n)jWEH63HQ+x9;c;X-)w{WLf0?B(5EN`Ac~G1IE&E(bN@?
z6!twNCOvt7jC_Nh9IJ}7uHokVx5e>VS0oX_&p6MsnYOG%y;@DC%6dhHrriLz3Qn6V
z^T)_nuV_-SbR1oddaPx<o7f=%OrSS(BrODyv`tGLTClSqn9PBFJaMJhwJp}e5|(#M
zwD2pGOK8ocq`_Ee%d95GpQs9%x?X^+ka0eKz&r^?*^peM*r@QL&cM^dF2uHRv1Ub8
zbYCXk&%YnEz7jZG9kqO&MY&~%7p!`;G5?qqAffY=dYJ|)R_wWAY3GG7PE4jtuG-N?
zsiwhEUfE}waQ@G{soFuW7DPv7@ow#G$bOc*u~Pz4<u`TcG(2x3_V(?6VjXIGxjj}Y
zteM@dwD{#cZ^4%&@+G9dCAhig<9A(-VQTiv+}R!LL#@Oht&X~*i_0K!>|;&iN38#P
zpImTJ$1fZ??RAnsFwfuw*11|%Ke1B~n*9YP=1a2;&xH3zQA2#>AmJ_{r40i?f$*VY
zG@Qesyb%rS{j$JwMB+ec0O*t*j_B}fFq(5xqqMvUI^SAWAwWQWN#L;*^~@iNHjYyk
zOKsLXN|eXE{He%13e~xZ!Gd2VbC>#;jyF<yC_?xY-A$(}zY->Fli*4>Ra8Pa3Y9sF
z)~_)2^#ls*kKYXIe%2g{D(|cI;m2cww!ic-ppSLh*Uo)$@_U|o0E+LnCwT;;wG1wh
zEy4!p>|KTWu<!yAr7JiHimC5W+`aB~3b+STuD1lYC7f(UXRc)>5Jc#H!z98dk3dcB
z!Y0d~)Ns%VYkEl@csJtb{~RFpZ@U_GJh~*lDjj4sGZqgF9F#mrV)j)}okq0xXYN<O
z>m9$Iv*O)8C*y|zH42(@H|;0MUiSy?_0<B#A9}6YG|u9Waa_cieDiPAk2sxE#09hk
z?8qiriJsI;lE;rw18=dV78rX{>|xrS{5v5%ipT8zx}h2ot{iRwDKZ67d+cVnU{fyH
zDONUHwyDJq!v`~?*9SQq+J8;`C<T(oUvVTEL3mLC>Unwqw4M8)YK)LeN*u$*+?U57
zZ0f{ZRnO?682{^4ID-j)rSE-0&M0ED&n~nI<eG(lWm#Qgr7eYfRkhix3M&hwU|w+c
z7iMTus=QHlwHaoR)Rf>^mD%9qqE~2sMbu>6OA~8pY3Wpj=bnUBY9{kv6lQ&I+b@<_
zTOvM!DXqu^2v+%)1cxK@bYhAqKYAFyQ!Km@wFOT0uIE8dbgH&|5UyYFQu_a(^559E
zPnKzR3H@h%A#c5q5WGxE^8Vh^)5(|!$E(K<%8!f_pCfrFO5hyrUkrJ9JM5G9ur|er
z&}FfdCQzcwMqNDWnRfBh?ZXTqw^XTSegMwKP$?fkznK<N9_MSQNG`o3)D3pS8q~eq
z5@{pGoFyV8k6=JtfDPYhJFP(-E6QcZ!dN(yNI${!8Y3}V*wto*`|X1R{f!ZmXaopA
zi|5J<*E_k`IEAUcxhWcwwp1edF7_lFoS!zZRUA7Iom{+Q-))RE<$3;SisBc8b9q!x
zstpG4s@1!}EbP{uGz!{qXXBnGKWf%oW^>?w21vK^_U=0*GR(NhcI^!a62!}6PIPPU
z1jMf?J^qH3&|fn>;LVhY0++k<5p)wie%0=MW!RV=o;!#2{51BJ!tYEp(9bb=uM87|
zEl@+R2{xJTr6y@q+G_IwD6tj$Tb^>+PcH%ZO^arrXa++>%07Yt#8c)ywN8WX!Qno=
zBFUvQL!+2B9zHa3jyafTkasHl@zrHJ*#dpqBjJ&nyMsA*p^LPbaD?>W5y$V^RU@$3
z72LxcAsbftgntKc8>7N%H>>%@U_`65TCFn*fm;1()9;iA&3~jT*Lj_!4IY3C?{)x6
z5fMb}*)knz%=y(PPYa*F&VAvXS0x)R4Km2CRs`vYd`!H@s&_?9^$rGFmshf@OlioX
zfL%aPFraz5xZbv|#w~%QQiFj&S(Nk2pDGi;loSLu<_zZWM&{+e1IzF<1Q}mu!?g0g
zx7GBI6-POxXz+anrjkK@?vb>vPi+HO1mz*dTmxS9HcOj%eM(3~HPv63SxNTx+*_2b
zxMSSO3Z}5L6u@(*x<tS2Eiq5zMHIZa_HJ1m?vahW_hO@QZ-<c3)ndT4`m_r;RVE-m
z2~cYIIo>+YHX)_BTwDWGMQzx~)1MxdDNygA5jBgU|MtQv6E}%h%$`d2HPQhTWTmBI
zT5CtzO^S@i@J)1<M4e3x^7g|*Vg~yMo<uT*S5}Tl(gdlNsO`QAG)GdmbDWW+)a_WA
zCbVX?L<L*Z;if^w+X}3kKTB8Ba}gV6+<6;XH*@4C0_+7!thO9z7VdrnI%2vNUA^#C
z&RDc`nJvzN_PR~kikx}b2yy`D#UhUQ8H4*7dYX20_l<=W<;kzwp6XGV7S7$m6B<K*
zD_h29!|&CAee1*j)Z3Y>P4MW++WLw=;%J-jv>ZEC!q~dnjl(!%h&FGKwdb;V5~L5F
zu}*4NwttMJXz9C@-0;SUU|%&EEZxnn^@_7ZeDt8WEY|NhF1p&e;u-_^d`RM$Lu5De
zzW4-Zey}QGU~({(QzcDa{wRwGGJn(H@{&=Zc*c>F2V|8G3g?)fDBLV##icr(g>DWX
z_|0x|%M(tx=19og$zy;<iDO3m!me;Op5JXPF{E96piQfYN#(^;$G5ypPdrneplAeo
z`bB1yFhH43b?~ay?kx*Qzc91x(i&Z;_lO6OHj|-k_;_J1@zuE4NXXG#4sc2ckU~a`
z94+JW^axR#$(E^<&4tgg8$y!Ewfug<&zI#+Q^>sH7nJ1{htu@?U4?*@RAt;a4f%|N
zC{7xg1}a-QPQ`O2d%%QOdEnUmP<O2cHJQkMm`5EFS<hN}>YoB&OuAxE09Gzmrb>-F
z#lOHJ?Ce{vaHq9hGo>(VuGsTIhXKFz{sqZ<BR0L2M>?ADN9Y&QJct`?&t;mk$r#U6
zObBChGSc<E0X>10J*}abbY<nyV0A$K+xl{M0bnOcpd(JiO_bI2y0RA=i?O2z)T&2N
zJYw0|-Zmd7tila&0_I_HJxtYIjf!JK6w6(S0R^+QJSvy(ua%18ABb0fYh)Y+ZIDla
z0tr2J!^6Prh6xOrYN!l++e;a|qtCc*bB7LOM>>#lZMZyyc`246_`ai`2;X~)HN}Vw
zOKuzz1Ws<;^Cx^S{Hu95I70Kp%&&GbEJE`Q3*tYo1EdDr(!;^q*B8&oGa^7y>%eBq
zDEcSyh2%FIZ17gTKf#E6hSrZsoH2xIk{+JIp9@)L|9w^mug%zCM5f_7N4F+McX>m?
zq&nNtD2}Qcq#dz7?~^4{tbL1=>!(CaZw8xQl_4Q${7OupOSP^NFV^ORU_juri$EL9
z{y7Bj@!g)mSIOlPwaQ5m-s<$ww{bt9T7bEj5;fp~4pN@3PEy9KlDOLIz+g$@9G>9V
zCe1XEE1t2GA@N~6KNRe<PXqz?g!OI4Jyf}vUt1__9$$&#{$eCxZ<bJ3@0rJUluGuo
zR$vSxmUt1pk*_YpE@>V_Ak=0wk`lKDA*Thz)D_$qVg2S>Gqjdwd)l3-*FxMVKx87>
z4-aCqzR<3kgVTRK5OaS5n%Yw&eOj$K^E|0)sAALvCH~;qCKp!{uTtf!8La6EP;DRf
zyAv|EthF4cL7Yk7(_$aBC-jD0iH)`5IB&XEs`4f6Rq#{$vrRu8-gq^M9aGNgoZJG4
zNn-j~uF7q}o+U&+Y!1iO*?qk_s7)Gfq%UsLr*9dJ$Z6=-1z<tu=l97jgfowV>B4HY
z7mV3w=9(|$m;I6a#Hr=sI`>-<=>p-~p92S0$f|+4kB6)bHZj`}@q_<yOoSL9=#Kcp
z#+)xv3OO4X5RC%Nw=dLG1HO!=q7MsT`hK$_=fM%btRUhGt4OaiV^0*^F;SkI%(a=E
z520{3e{wH>F#c^VxNWF!H2$qBbWuSVZ!xOZun6MI&ssr5Ps>^_+?y@*jA;@6&haq(
z+b~AB{4n18__WB2X%T$4)~Y2rW>iS8Vb~%Sh437{!0ZmSe(ri`zWsu*cN#!AlPGf>
z#o+gceR&a;H@?p)hTr}6VK)v`iqZ1h-=)SCpc6ng4hq_NDVM7TAiRb_{U3pm?&Adr
z84xx_avvJ`cLtwdhrMS+`J=hTvs8UsgP_d^<ZxbjDH<Z=l{qOL5csM<u~8scUcji(
zNML>~yS_jEfh6R*(YXsv01z?$+EEi{^@sl{joBRzVWkO9g56K1bc+l>Gl%(%X8FcI
z#5Vm%u=**Yw^uf|{rxnNg=ZUL^|M4;4)^!PmtadHSF=XWmqr%e1%0k#Ag<^Ft`#8=
z6Gf(3C?%C<ed}GM=<}){ZMRm5GIf|OxiAWp-JRIuQ)?=eWfnmOV1yqY<>^*bT{5`q
z7GqlDbJ?b=(O15~dfHN#6mn<!oSrSRk{)Q;7tC#lR}-8#7dN*?2f(19Sjo_DjR$Hd
ziGUgIerf!5Kt|o88{tCKJ@;Az<4gubv!kgsm7)m>ekd(iw8E!h90PZZa~{xWNLVP^
ztqZMqVFPU5wk9PF_;c?6AcjdCudcW})mXFy33FsTZbTAuO+Evhf6b4Oj|B6h-z<6+
zB&#XYiTq{zN)B(fjI{si+aRbmX??rT+fZ<hrmqJ*eZeav$Z6L6{9uONHXXGR@;Ut*
zxY3et+(x<Uaw7ykoq_Jo#H}E16;UQexfn+|fu6fyXX8Nu&}m~&)6Y}_EXq``Q^hg0
z=EOX^vSn(beXSarb(ZUdF;IHvRp{zA&s?Kns+yXJ;l0@pJ+dLd!XSXS@(w*xL=EQ7
zC`!^Q!WyWBZ&5Fv9G;t_Hbp%Y7tjfFF&05oe-R%0lBik@CU2h*BPNy03YuH#uyIU%
z(}IMTR+7mINOBo3sssvk@6aldGMXJ9A)gp-y;TZAh~?seoP8Kkl68#pDC+f>QrMV`
zny{=`;mQX>DFR||z#-o!4clNOIwBp+3YH9j=3*p*{Bgc$?3@VARjggL@pv_z(_mfI
z(1i(4_*t6|B%QFycN?6|NG88jx?V<oWCKsywtq7Ki1;-e;?$$h-*EaXHR62#3e`HU
z2$tMi$-Bzvms{QwTd}Kh@+;ngZt&JhdGnGDnnfOOO#=Z4o4ZH%9j<iX{Zw0)HKvlX
zJ>NXJr;OE``AXa4hn?RGbett&o{v5%zGs&Zy;~%Qhtbj=u)Td%M;e<*f;o1ahrJ5M
z;61-A;Eu^f<5$j{z8;VF4q<@XCYIc-)sf+~MKfN`usQi_6eBQ?RsW>5)8V4#&pe2l
zA39ayqs#R8VH_*=2!gHAFwd_$p_v{GI7{S7DHx{i*9av@s5i(Tj^GA6(=P!GqjYAT
z?yWuaL!)7Nmk1C0q)GlCw8)HtW_#|}jgNxO0E`svH8!8+0k<nWMnQ$*92bX_U%bm5
zP7}J?IF3qfsgQNxqytQRX3JxK_LA^YWz!(he!xqg62BJLGfuI0?3FSnrBk#g03aH=
zG5sVgN<VJ?MENu1m5j%C2f#h_?ixvTrvT_>?@!WZLv_>s)W*5|%sg5UuOi>I35#Wi
z1W>?bGw!_L&BX65-*fuLyqvk?96eCegf|R%M6TU$nu#y4+AcqkHPr|T@b!q}6pmu|
z!2Gsu!6ksI@27P~s{%Pw`Q7LIVy5DOg*#n?vnnmw6%_b~OmD7wYMpZk*#Be1^lR*s
z1EkBOx6ulJW(AE0y(~aiYTR>c8^sXN23Ve=Sq@BeWLdzUBYxsN<Jc2@?@Pun?kG&o
zy7(QABmJk|t+R#Hwj(nu>KN;Q`BrGLe{ZdYYdSrjNrSD_<I!#5F7};m+RB2S8*UfH
zcj6jV82nW;U|sw2kHbh6t|3*96;KHdF6L(0UT4LOD()gjREo6YM#|EF*A?EN7r+l%
z6R|E@Z>N<?Xr)Ru6AjWhkgk=DJ<n9{0pw~K%UJ-Zcv}gk{-@^(d#kwIka8}(<anX|
z-J$%-92nep(@wLf{KvEolrHS?x5I#I*z;L6`$e8&SDNuo*|~l;t;EJxhR7M_X^^uV
za&0&CwP|1M&ImY<rKn=BLNY=&4M1Mj$ENF*h|{z4KUUXiP++WSRr)Wh+o1b}xX;J^
zYFjIB<R-4C1@o6mV{|r3avzc$NaY)Bu0EK-#L1jvaWT$pIF44S+oadn*AK8n>!bRi
zNQfuVgtTu<5hFI7U5EMEZF<Lk=wZ$86kIPhOfr|w+W<?0J9jcw4S>d739#bvoagOr
zRu5n29x(IzN7|FFt7!CL*|HHiHB;fnzY8S}(Tjc9+(nlLZTUxx%}Jg}Q4wFRg0iEv
z+ttrHhDSrWu{MWPPE3=v51NTp`0}_2Y6q*(ujb3m^p=V^=SNt4$8-(T6{GBj$JV|m
zzJYtf!YWwL!ypP^1C8|8DE(J0H|3;lFWsK@EWF*a=ZA0$t|J-EAS*`tOai7`nCY@X
ziN<?*LP>EM>bqUO&MNNO^O`Cf`YCBU^9x;OlMcQuX%@<d`!JdBpUk)sUve*!w6~Qv
z7w0`$JZC+37?4jsGw+QLusB_P!?qvJyivB+cnc@0f^C|B9M*%%`JAeVo*%5e<cSJL
z`+y+=$WET`jC*WSvHY>N#1y@qO0M240@D1+hdw8Iq%TPm2wi7%MQFJ?Yt3P1ir0Mq
zSC67e$i_0}sM=bu*02G}xat|r;sDIsRnR3kHE-EmGKb%5CeB7AHU(D8_4LmiuAnjK
zwELV)R;-}_Kk`eQFHAD12d7rGOMDe1gl-_=%OI<`J6liPsKaX^TE1G1QtM*K!{uzm
z;Vb29#u&kw5-aks<P&w&&DK_s^!S5UEFHGvMc<!lG=W1v=lt015n7Y7+%a0OX=$bB
zIQ#C{VBKKTg+zv>I3XjYg5S|+FUmUbK!=m*&YtFg)6mXah$a6B7F^8QMBGZGY(|=1
z_+&;~TXFGDwJIt<m2y?OMeZ}%GVj%(Bgb?E8f@l2G@at1V=a&VJz}X*d%MU@B21Bp
zo%oWl5X{p3I7NtQ6W;X5G5rPZe`C+LR`PgYe*UpellOOrh|2~!KMmNf2knSK2KsRV
zJRsTxIOk^qxxq!`FDF_5l>qiw#ipCBSc^zcbh9J)r%@-RLEGnU`i-+&j1o^cd!%%t
zPOE~-Vm5d>$Q_UgTIfy2R}t)|R?@IF0}DCo&0{N8L@JPV;F$#MM+(zj-_VL{j4txs
zd9OTm+0Krs#C^otG&fu{Q$d|P>=Oh01LXVnzkGc6#m&bNVaa++tjnc)b;fEg%9t7>
z7&-9Eby9n3`J=jr<(pp7z(gFCZ^H_upnA5AmBC^g-wul1(UQtBVo(7-PGVf1V8AGQ
zg*eTy3+y$>{`QJ$o&i25_#PEbTO3oe-bV87(+@#&$Qs5=)NO-CfKKwXgx<Kk9uY|b
zaM07pXRcv8Ek13NHceuF;-?pC`Xwvvqhrcn=G><OzTI|>!zeH_j;Q`9j|CM1EFn67
zsBfK2tp_NmCk?)xO_xnQUSd1FGi~R7IP9!K0vvM)d^&E3;~mtTH3#x<P~o|(iOAi*
zv@NxM_=H`4cC~AKMjM;9XglP#9GDgX8vaTRHX15-VLyGCUtO9s)ZDKE3fJt5?sJxK
zs#w0Fv7{(FzeI#MmiP`3%uWE)a(52Jab9_!wh3RNs%&}jm!0J$=}rlG8I8xhVbxdD
zt~D!Ne?KccYg|x>Jc9}Q8OwYWz3^NBQT&f7f0H*mWn$ct<M#187hmsz&8~g`#I)l~
zZP!@Nc0b+{?Bg&H!E)xFn)yBa`ki6_9*N~$xsGWYp80-jl4H$ehvDR%Vl4owkH?EV
zm>t>x+7%Qoio*Tt^GYvuF24?AZ!Xd`Uh7ADXv?>;90492xCXBI&I?q|tql%uyMd)S
zVK`&=YN-3^Tm1KgeXhc>_j(gRHbWNp4Yt#(puJX?yg=ov95(^8K}?PK;D_L(hG?er
zco8kxL>6rI94gKvIZV4u*dlUGEJ1^@p~mrLAUL*`rNatxpg77>adlrSml)Y1XJBSA
z`e>U1xylN*Nzm4>9#OButX!kPOg>L>)>Sz~Zp4nJU&j%CBN5oM6~6a?ojJDAS&g;D
z$ze3bDO1r_KNT@kga4?JCmzP<+3+9>IpqJ=oEn9iia=$`Kv^sBY`;g0THbI+>wml?
zoCB{El&>7}@+jIh^+OsSB-mW%e*qa#CnS=C=A~CRdeSlp%3uf$adnFKIYRS<*D1)k
z${%S<hUfT9$a-5l0CetaA3zu9Lz=3orw29PVy%<{tnUak^LUH4I6Ig99bI|r9EFKh
zW6(pLBQYe#+8(U`#1^M;cDd1c{^%ow<vrZws#xT9JmS0#0|!@(yuY+@%?MwdOk`Vu
zvdst>VTeb|&`GBqM&vR6iCLmmJ%cw^AE9Ou4?m)v_s84CA7H2~72zRLalY!idFUW@
zY_8_+w`tj&fqYEduvp$4q?YsK2<^4-ohyZkxI1PAmMDF^;DwZy2Xdp&-rwyCk2KeG
z<x-h__sx6vuNQ-#Zrxc+9xL!w>P1<{IC2O}V~|7^OLjvsuPH1o=x{|DgCx0Y^Mx#(
zh;yb08&P)k00!I5Qp0HN;1{|zC_Z!y67=6&?PG;y+*vd0D+g^MS&>!e-)s0Z{`B{7
z0jnB)s%GsN!e;NQZkOWk^|oOhj78n(C~Q5lN7hVKga?3w*;AbpPLQ%LiD>|$9hWB0
z)*||{`re9>2gTZsKd32#)J8?!_b;~WZV>}slpM$b;0iOPHIH{Hv9yY?TOB<fRKPT(
zgM~MXmf8K8O!JJ_SI;_a=IaOf>%EMXh_`3N;0Vlf>}I{RC%U`)912Lr2ooKsG<@v!
zseFa#=kJX{&(6`v^>e1&UG`4*3?(*r95-L7WjE##>vge&F;LWYGj};EB^a#Kc`9;^
zI?`HxfSI>4{a@SPG(paHY>fnh#IzttEk!n~4_b&NaCpmkvX2HM@nur!>=PubcR13)
ziL@F$*-vafTX7$oJl%dbD}c=I(dJb*E;1%1ZjR>Wi^XK)ZCd*)yy28wnA}TqH5J|^
zHmj)nWqZN^;do6f-4Q*ojIl!5P<Gw1>1RmH?}UJd9zo=6$*yL`z8-p@cA$CSWpL;Q
zSi(@!wcLRRUB!(9*`}d;OZVI+bl21QUa(~4O{LKZHx<#FWcFQeHV%v(NGrpcI7&)Y
zl+8dd*-EeI+2GU|3kMCmu@M8%9CnT0Dl7CeOK`2*e0^6*GuBPIRBob08AuxW{b*1^
z0GX=;IwH!U?p(^34F#w6g?eR!7;TUdG?;-1zVv@Ym{J}tV;?g?p{3E;jk7JG(@-Ve
zwDz-Kr?%-zywj~!>3k6l*`VFixkR+9b%+y1|K!>H9_*4LfVA_sXZj!gN{~AqZ8g|6
zFqKGD`4O#kk}gib!IGV>FK+Gm-Rb}EUr-SZ;2-}{?8`DT2x<J6|71A*LgIiUSHDv?
zr7PwVA+;k=e*8c2uf+Es{s~(B!#^MM08^T4Wfd?*%KUq-W&qJJKngo5p38)&JYa-8
zaN=g(XCKNeDqi9@lcza`I;Z#{AYM;$d{tI;>=M=Z|4zf1CJ#Jsizu*^JdAb}LkSYg
zi!cS=ZOg;-`3mSOV%hl1V_MECyV}Pr^$XL`Kb@oM-Pu0nzQIs-{T~<#a{U9t_soA_
z_)IbE3`Wb)X2I@>uW)capj@%8ZpvM2G6`u*>^~p@WT~|TV55SG^KA6Pni44ZsKdaz
zO-0*R`i;b7A$TA1G$Ee=bc2(vWaO%tjXV=z64Tp)n~p<KYuc0E4XTNmD7UlX2c@*V
zo~xGU@=v8*`;QoL(^6(lJ1^ySu%vn4k8rk9?<j%dj*F>>D;&nJJE4BGN!yt)_Hpjp
z*6L_}4QQ+8vHAeqg})zz%(;H6R*P)E*#2g!9_0>WG>gkNo~E?TLZQn8r_sgXm(EE@
zxnG>r2NGtpNqF1^1!r?HrkNnTNSv1476yT<3aiNCs+wk?`YH<g@+u#Ja#IC>Gb|t9
z{|^x|A^amkAlLs*gn<8zgqa*nR&f4)<no~ZUqt9{`b~uC9`*a~R37Q#`NwmtNg|^z
z70_+GXYm<8Zq}`SZqj{b78@zVJZQ^wWbF}p1Cyhij|T^2<<EyGc@V(c>G<%#A{&$d
zS#j`PvGFdSvah=->vFD!-#KF}Djer2i1iTB4DyhgQk?2g^66+h5Ym|J>?DPo<%`MX
zh*-teb&FaQtLr(Bd~qeDn?&SLf0lHgt~H5qYr4;zzWTo7ij{e#n=5&skr?CVp~)eW
z!#N}g=-1TsiLoz)uKe&;{x%4>&ok{vx5v3~n|~|7t#8i(bh|qAILRaBxPN^T3{%WD
zyaLhKSK>%Doi8yxaIOPjHa+EpzzLMWhb*grsXg-2z|ee@(PVG%G{9TFNipq;9~cdA
z`IeU~30eCO6F)yRnY{r2p~I-@1M*w4-)vb2Xt1E~n(4=>ITHs+&cRW`o`Gs*^=qrf
z@>huq!@S5$Y<ZQk2p<eEui&@iw{PxK>Ovf$=Il>oCVV>gN6fZTEO}`H-xX-ygVjp{
zNyQk8d!y>f!EoBZ>Y4i|+=?K8h3RCi&?SQ9>bmwODWD3SF>aje1HAA46X=5&Ki|T(
zp+>+cb&dnR|8MxP@~X#Z5PEP$+_E9Xk8vV}fxFf3!7}{O1kOi)uwR(a63e03r@?)&
z=qai0=Q&fn(-}ABbgGUr(M89cH{ZHb_Ufef44<B7c?enIQ*s{H>rVn8wiaMulTNAr
zGK~DjL*?QBg@@=}|G)6C=JjHDK+*z`6m8X0<imFPHnF)eWNvWM^d;W1A$h`6!=Y^<
z#+Bjr6&7e;?LCS&I{THM1vuQ2CV1fgSRww3D8=F@+VBnu{tq7B$15S|Z55p9H~0cR
zH170INQG~qgu;!KU-Iv{k3Z3U&uIWuo!)8K7nB4ed|p)#C!nL7R1^=7kGGP8&3p~?
zn|}#T9zAXZr@4F`&Vr64cw4T$_4mK@TPneuH`6Mb>TPw5V=V!*IxcT%vkqIIWk{1K
z!-;eW&dkyXQ<LQuDSKg*&1oX_6NTniss7D>*IybMdKj}rHrDt9?lycImehUhlKKhV
zG~}@N*0G;SoOlYyM(Kw*-Q7d-Wb5w()^GoyeAJ#IpZW~@uytW$J%Ms<-NdLGAF@UE
z#Kz?C)d#;<F0=t;fA(p)yvScyuFBP#a8^%mfWxxZPjA+jAjgnaBWA!I9sWqsJR`AA
z)uueV6csYFiH7%!f<Uz*L0xSCv3cCbnz7`tvt_!@ofTBLmyw^3*S{!4%&S&P?OJVi
z6R-C7A}1Mr$f72GTRj=TWS0A`N*rmDm0q&m@JhP2Pfh>=K_pW^md7uPo@gae+czi#
z)584mNATnE$}{aI8$cJ-Wqf{l;&aIvFH?NYyr>eB$wBZl={0!8VCNS{<%`4U7w-HC
z#_L}$`ORJ8Oh?);-(~)ruc?J@wPY9A=CoH^1I^N;(s!t_RIup<v2yBt&LLahw)+ov
z?lLD4a=ADFwvTHds_^^2%ceUtD~)U5UCkS56i@ZN{qNoajmKg^o1&J0CvR;RLa}P0
zR5d384oQ=X%3r@M0Et{c!7upF*FJ~e^05<E&se24teLR4zXuR0rc9rj2PPYEFCk}u
zhJz)%cKcLTE`4jW7IJnTN2Z-Q@xZE7N1P1f2Zb0Q((tS<G_}fR>x7l#2o-;(loW5L
z$!k&I;G#6Du7dDSnHH9TP2m5P0Aa_!CBUE>)rA$XGM%e}nKiRjb5kjDGOWX)M7}A9
zn1dPBoC(XyH>V9_g#q<PiT-&~3usat*dvAu;Bkj2&k@W|lF#Hvui|BEw!r1fPEZxV
zszU$H0XsvPc$;<>t{+Ul_X?85Zip5c4a2bbc8(7r`R<1t)PdFxBTzwq?9~sPbEnGp
zL1>$iEMYkFu;4}W{_FxnQr;lEOi~`z#T-ILQSzSV?ewf~M%2zLl_Tzs{2xYq`({K$
zaLL+5&)2qj<H9ehe;S`Mz%ogECe0w!vwzc?;U6Rxq6%lDVCOKItPUkGl`v5O=Ir-6
z^v#+$9x)Xhvo_nuOEW{^+q+0@c`~ABByKHjxf!c;$S;e*P7b0}7V^_js?&7Ibh}@Z
z?W9`Y>z$k6x+a#URYv3YAsmwg=c;L@b9R?~mklb}sjC|640r+Z24ZgVrD!zL-}06q
ziHjA`dE&MKqzWaWMd}?|(j-CVNpvBI%tod%oNfR21*5)J4R?LoO5bg*meP6c%l!Fb
zpx^xzJ;SZP%$C~T80jIK0e3-^R4_Sbet~BH-J#e210m8d>?v6j8Ej1BUGPUs6(}j}
zIY=$6Al`8E;4Gl%V}o39*%<cXU4)-oZ*hDvjLD*$@3#fhUFA@gN(O4keXO#5a~3gm
z$}uAbz(H!+;~zpE!I>TeRW|uMY+asw1q{l@4>sEJdbgjl$b4cSH{y`v6*@<BnLi6X
zg7{Vn+S7Grx9qDU)FunlB7@GJbi<%={8Jb8%y1V0X*dJ+e2l=moqd{948Ug!_P3;u
zKxW^yeD!R5o&QRU{RR&eAP(Wa5p5G$JY-%?;QY+nVjlUIxAyWfiNX>#4u8|gs#~%|
zB}745sr#b(cFFe!+*7-zI}tm%l!?yyYy2KsSFZG)_$AwJH83Q_Tul4G?oLfynVIsg
z0sarXdm7KWk&GxAc(dQSVOV4h0nq<DIdXdZBS(M7f8-cA0ivb8UHls=Us3SHIeJPs
zzABp)oyd7m_$O#mOlihm*~>8UC^}J`tjX;-edSNtA^rByMN`P%7ezw1P~UqZnfEB@
z=U|TkNn$@L=Mr4GFf64j?R~`?8FLm{S||@Hl=#}%%}IRGS#VmzzC4Bj1e$3x)eGE2
zbAD+UdgOnfZd5bhi`84JhFzk2vJo|9ETW!%F4kAuC^t%N2fUh5eg*5?>6;GgNMuN4
z2=@aS#4RHI-0}6jH0l_}Kf@}=FuFvNO+pR?Y}BwW7%nud%+5o9vQN|5#oE2wm1D!Y
zChBG6qszg}4#J2E%>sga=EEv(ey{tLa5%N*bt-UhOk(mbxl**;JK!UGp{IYY<4&>v
zFo<k=ZFJ3I($?=AY)7OMP|=h#SABz)2&G0#>yAKZ+{TvF+T{O>Ewqo{^kX15Rk^AO
z@P4!`N`4){j#LD4zzk$+rom|wd;M+E*l==f$u<}eP;N=hE2GUv$u!=bARfy;4gxvd
zOW-D$PJN|?>ZWGKM=#`U6E{g2gc-I$)J<}{VDe1u$gq!_2#Ki-`E4}Tw!JkE&#GNp
zy4+s`4Embkf%$HdOgKMwpw4AVS<DItAl&Cqf7GN6*rKtmM|tKV|D=1-i$o!R{b8PI
z?sT4R@4JoXP3vE*S)Es>20FC8&F!i&M953l#nvtpV@#!Wj?VE(4AL%>LS4(lX8%>T
zLYJ}jcj@vAVn!EA#(Z+Vl>-BB`2CK(fV962+OX^ChLw>({bE7A%1fO-@8zxmuv2m9
zi>nT=ZWpB%Bu6Keg*|nH=widJWI}q+)3lq>%R6j38f~GJbdd)EX`gKhK1BfMDw1<e
z8w!a-PfZo0XIx4{WQM<s?DxPPjXq=Glfr78AU2jYz!lR$PTx&P{}lFWx?A)Wu8U4y
z2(qwh&~@EV^7Nq~eR*_(42(PoP&o;5h+RY%q!D}>iEwwM5ORzNPc#HD(;*-EP!W=F
zt@ilz-*L^rf~TbT%DRlj7T}a09oRD!NfSc>=MIrkuE=%TZ=3fAFOt;FMyHn(opi^K
z-)zaSDNZsRLkLAnSxaWR^<Y=9oNFe172|w`jDCB17J^_fe3RR>o93$k%&{Xhu*Vv%
z+;4avfvMoGE@b%2sV9;{RAa`EyR}nj@tN)#xWDWN8H+|oaZ4mwwBQnRX?n1Cw~}&Y
zmO0wK8<T6KIG!(OSxqCB$FMF4B4R3$01uUFfY>l`i30~km|%G+p%#~BN*8NEOF>3R
z>|JiiXXU)Ea5?Ll0B^hjIDzzN^UuDfVw`WG4G~^>JsDe$4&5=PTVLQj+-E$GJt($6
z`Uy4J7fG4Bb?ETPQ8rId)Pu->%<~;c_(35=)=a)Aw`|f?na2qth*C}taYi9RI~qRG
z)Z<Va%VSG`BP(!D88nZ8Gbyl+QIwlK)d}fp@IIHu{*21)HQR~=VAUh`52fDlc8A%<
zZZT(G1O!Hv$`K;PP0|1kEIivsr$oYPjWBref4Eh>^KwFTxA0Z1dTYIC#bR86n378H
zoU6Y;Q%we+g<_=S1(rgTzVk9M_KpdaMVyplt!mO|ykTipNce?vkh_>L2sNp7xq79n
zQcn+!tRw{4RE)+0f+6!M;EsWt)PS7*J!cK^)pWo;AUjHW>^!t|z!A(=<Lk`q&_uD4
z`d%;x*yEi*-2M``6P5Y#x(!uTfgmd&Od+c1fE&|i86l#n`Uj445AO5?MwQD^BE?N(
zkg1|k7D%6&Lz1cHKN+Vgs^te$n+^M;P*20xNtc2t!7B;_T#XHb(-f6Z@6e=6sTpKS
zkBj${6<-#irAq|~ysuGoh1CLdd4Twg1Q96<K+-|SlOXO&QK-9lw)x=g`^HeHe^Snc
z!hL@W-BCd>%rZQgs+1W=vSM-Ek)m)4d+L>vaFbn&9c4jcZ}<}imRbu9GdC->ArB`$
zwHqBXHzx$3dxYoW^SXYx)GM9I#~;qIsn3zzWzqUzc62^?I!_VGFa;MswOv2Au3WRa
z{h7MB`tFCUS|JP1fp(eKW_;Err*+3dc`{B#88A-B@*5{q-7}%4#Z8$-_gp5^lLwGQ
zZ_kbsdKypx0Wcv@d~L_I(K_gr8|*L=G*(6Akp3ND;RV6IVBJbM92umYWV!P~Q(p2%
zcp4f%>y%J`$QVpeI>mZrTngIeY-><uZRKPRBn<s8_5eX%ZiD-UQoFQUeivTQNA4Sy
zO~@f2kJ?|w9X7_IJiu7LN;-%(bAE1gC-&v?))BS=#}gUjqjdvshmRM62m%rIc&ZAD
zf8Shy8a|p&%>Ff>xG_A4f|SvdXwx}AJA_9pB*6l*{Y2esj;{<9q2RP$2OJF)z{$fe
z-MNMEFy1HxuEK|MTZjY^MVHmC*Wbgm>zk~IdNLQQ5Kt=fAB@R0&*BZ`u$T#ZVA_cE
z1J#f=fQFHndSJGJeu|Q7{<{|2U^Ijzc7S1jFdOL?1ZZ_bjv}R`QG3b4-*HFr;&OOn
zUmN504j?Wh?Zsnhw<$7x_tIPBmtKyqs~g63m8?NnbcA#cg_dQF*cQc?Ojavc;su**
zUZ`S54VqMG8QYSwMwkJCD$d1n_gIdeEQdeSg_<J<Pf5hl7iCTIC-V}%k(bVt9$+gB
zE6!>Zb?8x)1%aKuEDZ7TZaVpQXxV!0VAV5RlAfW>rRM2|v#mw7ELZu-z6!{)tmJ<t
zIB2z72x>~|`vMNfqomJS&yPxyN4Sm&QTwtVlo1EYWs7i3;$Q`969W%LN0Ps&Q>?`j
zLMclQi>`|h&m99j(5!K!0Odlvs-Y=73xh9SG%JmNJqRE*<SF09&oL8;HQ90-VE4?B
zqeNwQ+7eUk$mKdEiy1m2kvoM{DzVY)$EnQC8ne_63i=3<j(M>`0x#ro6fDe*ST!v6
z0`1}C?$7!^KlF;&ywuBFrs7V6uE2e4m<so`0f}E_B(@{vYC$h8fa+*-fRDH4agq>v
zQr__QgcZ`j{%M3-;`>{ho&GzKG5gD4{5wwrCX{=?Q?*S~Z+bk$=;&6Gg#y*pd`8qA
z;v<xKe^%#qOVWk%z^jZBmOBvRp{gcR1V8?#=ONCuBFDTGJw}(66y;>SeK&Lzc$bjE
zt7#xUW2Q+Ay5Sx$AhVSH1#&*29LMvXbr8ZU_jqn0C{t=P22UNL-Nb^FY~b>zO(`b3
zgh(0dWL)OTvWyfw#$e0L@fy_Hdu3wfnv^M+0`u=bw4^+dmD(-q@N<0i;Tru?id?km
zO%jQOV28*`a3cYmXwJA$^H6G;Rap>kv^1B$D!{Z4>^n(X0rDO}!=b?-?4;JJ(+pmP
z`xC|`m|I|X{PY3P%?&Ry8Zz{#=cB%{p9J>yot56+0WZ=OBy#Qk->i6{UbdY0J3aW1
zt}`!LHM(+yfLR<<nrVgqr>-XZLoLMb0AxSX*ff4o`dpF?s=taBER~ZHwL}|+fAcdy
zWxp8bi5xdz0LQ8Ce1KUxryHrI<*Mb^Uz?SU);AAgDrKxV`3yy4<#xtR;q%wlU<#_V
zkzLjmtJY)HFaN3F<uKE3vXKeVJ`}(B_9Nl6>A{vQ@WU~^Q9df#*p-ctQ9QUhZQkCn
z$zx|*Wm$JR`puU&i_|ivk*JIBCSVg++$5erw!{OragS%%30vrpLtY!K>sHa*P1!=J
zkg@X863OukhUPLp-ebUV8o1gHW~UNx?5k93QnBk9?;{h$*ZIZ<Pk}P}mM?XG{QPFp
z(b2a*w4>4}WKDE)|HPL0pFd|@zvY$`{}(Cqt|Wm1+@tZM1^x_M@+%Nw{v;8OQOM#Y
zWWboEHy;|{7XD1F^qA)BD9Opa#s$n7)V<fd$X&x~G5N0fn)Tk^`^qcN-X4X9J=1y*
z+4@ny=T+roLxMmyl2eKZY0UnuY|fE_)D}mDP24cRTYGWYM?2_NzE$F>1*b4Gv2<Wv
zL!`d*wEW=&Ur&Xf60lkDCXgA72cSIO158}-y%u_gd9_G!&~JWl*8?O*^9ZiK4vT)Z
z$4xY&P$f`scel|wulW{B(Z;G$%Sma@E3Zx{xWlwvDEG*am2~Taq3)x}#sldD;oa(E
z;ndqr7{tg%H@hP3-q7Lck$s~7a(*`ZDM2th>AGT84C+gmCUK^y==YL;M|8J92%vNn
z*cKrD5uPDJkd3jcDlW`~T6naQ`XY6;H;o}wnlnCTzZB(=Xaa0iP)-+_Et+g|(pTB8
zPSb|1M_)l*uU?^5sdjMk7el+^z(qYjrIqO|Lj30ExA<k-8P^9baU)cB06kQ#T*;yN
z=0|0^y5vrn^=9WffWJre`Dl3>1hDNzuwnS_EJa@~88S_=e6m_m3i8i<Rp@opU)~r#
z`<_Uu<e6xq7Z;yu+J*Zbkv9JKA___@fO(Lah39D@JS!Dm!QghC;FdgnHZeRwdyy>d
z90)l!Ej+;zH6|?!SVqa7W9LU}hKLD7;8O|ACS-m4JRLjDjf&&uP!cBz0C*YxJT-8m
zXE2L@T-vppJ{0E>Hc<T#q(S0P{g7$AAUsrWEXv+__>Obj0X1@q$F(N0qE$t+mnyPX
zxnbRsVBvrdD1>5O=uOcXZ9S5P%fADqs6ulVev9?Q5qYp0q+Q|cs5G@WE;=6i$L0&F
zPsx+sJA8`M&K?$f8Jt!s;N$=(jxus=L%E4qE-Aww3r(p%XNe{!l64_x#xD3E@b{J&
z-28Hkb#eUw7s}OMeqD&C-sllWxFjjZ1QOc~`RScGSau6E7tzTU5SRyn0w{>#0z#8@
zt>C|-rd==q$E#KADtWOl(?|qS6SqfyX=R#DF2|5n4c;}lWEpUc&;`9OP^Z{#5m5QK
zDMeOMag^dz=R>IOt?$KABF!@K;?Fb50AXOyM&tBElrR}RydW@$^n5YI5mT9DPdgrx
zG9X}19WX*RTR*@{ow${Ltddi?OIAPLn)9JVm^Y#}=aZjyjHro$5sdz*5^A-*J_gwF
z%=&)Z{k={_0LZ^1*7;9_d=T8)c-+QRCUQ4*`A~h_DY$$+i~QP0vVve8bdIi`={7~y
z1Kk6Tn*H=B`JAQ`O}kxrnIm;DS*Ff;OrkJ&Ccq}q=d-8%jkvEjL8QdVie726)k$<C
z7@|UlQ6cvgs#23GljBM_jdyp)n=4J%k0F8X(;3143|ORYi6TaEe(3vJ{q{Wd^om}<
z!>BI8u*}VtnTkIdVYqgRq^MxBDZ_xS6}XBt)rp}M_wl=ao?FT!v&D?u4Hh60GS`qY
zM+B<(cnF^QH0HbU%T0<4h3Bmx#q;$)dr80+qYhvjbC5}vk{J!C85zg&qA*Q|i5Q<G
zB7Rq(0b*Ev(ayLeClKWp#RD9Q?tb5Q7d~kMfjZc`^O-8-XAOph@bg$PiETlZ;r{%o
zL%jX6hv;WUQ(si5=n`t?Sa{pSew?xr5L(ioIi6jRP3U;801XP1G_@)aNNKlnocw^;
zs(^bdd@O4tezA^rb%ail;`NX)rc*@2XrSl<0JuH~rEKj;EJP6C<YgJtf*Zthvm7m}
znAkQkfcGehgm7wwe56IfIEevKqN15^i2$9w*<&{%Wb89ZvXhqH`SXr(=+vDiZR`V#
zK06aO)@^@?Qiwk|F^-c0yCX||^@#$wRjQdF#g@+nR+%GP1%3VG{%?o>mDE%6gmB*h
zfKbq~!#!~s(!FoS?#7gAP|LYXqil?YZ%jNw=}o;htVC7jB9#*4PsZ)`pB>jL{gbCm
zj#A&X2>Yr*3?V146u+<LxN7f4cH2LUtlEX$XCvQkfVItut<t5Ur<TE>Z3KDd2^Q$q
z9y)d*j$uq}mm^*?4<FH!`Ai+W+>k0&01%;OGndZdVZFLzTG!WIV;qSN{S_s~t>#bX
zHd*t9WO|W#r-r%KuD@LuzHI;9s1f#nv7uLIB{75Y^4*K&6XYWU_Os@i-p#WJS*G2e
zKIai{O$57xVu#fR4&%?VU87FGV#gB#<Bh5;Dg8AV4vBZR9_H(p-u8LuG#ub4%Rjlc
z?Gx9Th2B^0WkA>KXV_dz*wrpV6qyrKc(mJVhurAbo8H~unyYDpi7bUtyw}2)w7|^p
z$5Zid2S|{U5TK7Mr9T$pWfGT3Wv}AqM%-$X%2TvGbuRVbe8v^@d1~nikk5Ro>)tM%
zr@#HR`a##$-uV*ccl0W+7y`JFMhJ8uc|5)aIsicc;e`6d17)ZKb47TrzEKxf;cm^>
zuOXd^3V5sKux3td2TRlh+FCxM-AFE6(-JD)sQAlS(?|k)HFwk<HYXG1>@nw&Uowx!
zkUcVmaS!X1ID2k#nFsHC<*z4UuJNqNF@q>R;|$vY04e`L9kM%$%PVNtWzFzwpYJ!j
z+SgDvpkcVi8gt^tRv3vx5OJzId9L$Wi6!){!zw&ccu9m+mF>+qD3O}OrdV7<6d~NS
z3RgJ|NedVq(&#?*dDl@bjZO;~6ZRt@xj!dcPU-psC{0o8yT+8INJXr*{|chRoAzKF
z&I^VAeQXL1XJ$aAa38~hbgsshiu5y}LR~DlVOu3lxJDZ&QZkde@g1Gmp9;hJS=v;+
za4e!Bj{3=yOuef9cHctt@LaV(-gb?=AlACRjdAdak#q1KLO6}@snZ$LB*trx|FKcX
zGfvFa<Tw@Z0T?2}zDGF5OVH<0^RmD*la88oj44=Ax(l8vPx&qZd?XhPv*bP)Jt9v<
z=vJV*jI7ari8`eAZT)iq8fz0)m!X?GZ{VYoaPygzesu>;SW;0HH^>Lh)Hi%`!D~M%
zrD>F#osf{2q8GWNkZR<2MKxx*`|*`4u|f5R&)z@)+!R1aP7r9uU<<eP+=Jz?QN}er
zM33%FjnN`Th&geLxfAQC^)2B&oMASfDFk2pkJ((#7OP2*)$876%79~62rZ<{*k8_p
zKa%V}?y_C{9T@#4+f&5F5<*0^RYgh+&GWQMcuOc0n0)B-d}N~+Jh6gL!@nInGri9;
zCwX81sf4-t;h07XdGi+n2c@q0jwshmShr`<t83JS`nP_FA?N=p5|3hrThcw71Q*`j
zj|_w%wI4Es8Z=ozAE~4#$^@0eiLN7}5mqqq6vz`N7<BxkshN`6D9Az{ag!Y)+Xv>F
z$xlI>MCITHr3jz>W1Fm(uB99@58RA809tczOHafHp%H|m!4pi7cl7Uh{4f2%|LN*0
zqv{BnZE<%E?ruSXYjAf69vp)EK?e=)?!kgXfZzvrJ0!R}90&y0V3+T%_wKvv&HS3x
zvu1u&SM{#0-Mg;H_Ut^azv=>F*;i*)W554HtBEMN{?w{#?`1eLXZeUjT3j=LLXhut
zkfeKX(a&mF<m1<%YP#w*_Uj3WqQ+GH+M>2AoMlcdO}ji=ucn#UMa`h*<{x-jgB{j_
z9;=!^@<l2|G6e|MZQw8EAX|nnRHN0JD8(2}LE$=el%y82=xP)3AQ6#3!Vu;Ub7#lE
z?wg!L<PF&_@KOP_gjSjmZ5UE+Exzrl`Vt6flR#=IphM5yIcm02MsAEmRK|Mvix6#N
zf*+#zf5nsbOF}Kw<X|sEPhv-+>VKF=_EgO-pSfavLI48CJ19<<zZ0q$E2^%6oRY1J
z<V_`JtEu(Nt`Fm#D{^Wjs`&FQZ(o-?R}Kc<=%4g|js*l8e%)orwsKF`!KD__u^u)4
z?Cs*<&B@Rk?A&ML#2d2G@fyT2hNR-?ih`Q5Zq!V2PL}cSXrV-}+Gto4N4ODlB{%i-
z*Wn9G!X3aN^3D|3%!_-V5y0SOxBR*fk3E!SPO^OybKZy~M5~OOI&q^{an~;*yKJAq
zp(Xs3&q+Zd)E^tjD8r-^4#d%yj#vL!R$PDy5t=azM%gghSY?jVoz$(-XnHR%?6OMf
zUMlb3p=Iy$QAA$szMRWrrS7eveG@mL5bLeYkoX&C>8RA@P18NjTZuIgXD0ZY<jfC&
zU`tlxKokm6Dn9XOIKl5ba*tuR@&>`>VuU`mIyD6`AXAP(!sJ+9_IkYw5gT$j>91ee
zDP&lWTH>C-4D#QWz<0&$VV9`>t)z$gdhf#8R(VY>ho=Bxu%PqX*^BseS?|=3U!xu$
z8pr@-qSi$ZmgTa!){`5I^u6SEX`WQ>snaB(QrnQD+t~zD9QcBaY>id!;JGQ|&!aWd
zZ(16rH>&$Bx_A6^F}Bh9uIDZdt_X}k11me6CzTEM=Ky0Hwc5(2zn=V9Aqv7uvsP)h
z6jA2{+T0)UOer)*6|hAX?|fW9TNnIT)b9Z;>^N>-jz@FY#AVJ20hE+I40*a0*cQ4^
z>NfeyR<A_f1MO|DPG5wvCF_UhrpOx#U@hZCkezqUU`8Te9xS#c9t`+96}oUijx594
z@F>Pd%BbI4)nIx6iaBp?5tWrPX7gB4T=bkv1rnS;abQ==BY|oDiZv#;f%q$vKdFFV
z;+d>p<h7%XU^m&6_j9*FJHvdZVgB6N+1aULiDIwu^;@&HP7Bcv|LaH=PJ6y;pDDeP
z`8t8>9(-rt(SuShDzyC_*e|!DMrOtiJG_B*$5aRVE5OrIvic(|d$BSt*A?UUAkq@~
zFS+mW1w(mXTqL>lzckBrx-C8pJk<j3n~r>d!N#b|xh+vZZ7GUC>pj9vl*B>*BjLpo
zJ2|X1)qb?VLp*y?X&K@R)eQOH0BJ6%zI5(fjfSV?DFXM>r1-seF(U1>C_~&0@eMZv
z)H?)4fy6<-A4M#&NE*03EKfjEHihSJu@3Gq#B!OF$-yx%9s;iI9%(h82?n^3q|2q&
zXnt_ihwN7Frqo$W4=}y&VyKh#n=5-cf)M;e7PoDTrB~D*IV=h-HI-to9%vrOJ?u5#
zHvPbXzi!J?g3&M1V2p)lE-~{thJ^x_(#?wOeWe!$q=@)5R+?vDzldnqOdF|H@~Vv_
zu<?BB1~yB}S-11EY6J)W=pfJ{OlPT5Bc9!r$rmn?tB@$<$-~Wj&E-l61SZ5QPH*{$
zz?_KMW0$KXT(dW5A-4#rcME@YY1(F4pxX{tb0_bvD-U+xVO_~(VXDjd?&MS`UoUP@
zo8%_8T9kBMQQ#?MP6?01hPh&cKIdS$r;*3}r6DCY%maZ%`^$8+0tcYPlOdD?|7eKc
zx=b@isq!G_50_?<`fDP6_TNYq=f|E<#bN)`B~>Eerz4AWH~UXkK}b1fE^xHE@QIVS
zhz2`~W}7@AI-d+Cw(w)qxI^NTWj`+i_Ll8%T6Tf%E=^1TN#6N#>?a$2nCnANKau)=
z9?A(L5{L%V87u>@&39mJDvgeDYE`sa<?+QqESKR-k=q~lO(}(zb00R;PH`5#3fF7R
zKZ~-@ZsqYK^6&@jQ#>Ev8yD}|1-C%M3IcRHi)$Y4)zo_zN*l4t>c#hJ@1B;B|8(#l
ztC;`6S&Mm4ljFB(a#)q@+SQ`?eh?*edTT2Ru-UHY%BcR*fWZKevEpmY>5}W6#a%+C
zZ!-1I#o9NV^mCd~dI-x<3luFv!#TLU{v6G)4GPQQC1BsATHF{CBCe_5TgIvV0(a%Y
zi=O4Q_FTQk6nu3Qtx%TjM+qyw75WNycH$dbKr`o1o_-}$-l15`pzhe~POd=bs}GqR
z&I%!PhEy#US@8fMqzDi*O+SXCRTjKYR(8%wePwd=<WB}<LUm`f78w~}?7UuCOUDyz
z7HFOqU*>|bY0dvVV|>VN??6btMAh)-AC$-*sQNi}cE<MV{ByKEeYO6E=t1Y9S&g;y
z;pR4DO+Tk+j_aORhUwv$#l(TH7=L6-=sEfuDUfC6d<>-PiOF}j{5hW!Hv45`sn!0v
ze+>0&QVnwWDUpm<`N7%w&)rk-ox@*|nU2oq^vzj2mM`yb`J_IPMZvd-4z_I}z%rp!
zspBLWu}VlqwRgA6D^jjt`HD4<d{Pf|*sjxIAl~fq+Y2%tbie!RVtx>WL-PsVMT?g5
zF8)bo4=)+G&w0jzW-i)b8>oP6eXj%#xD_@sFts3=R}m5Rw_1w{M*zoTUx?XX(m{Sk
zy2BI7qsg$$uFA+O2F!#CO-a(gnR1gmi&$#q^x2)|_?XZ&sthke3z+<z7;dn&kyUkf
zZ=IT_ZnfQ1lr&w-l$+KVp0tR3!y@6FTW8w<aV0><7PH$z6CrJ<vF88V8Igf(iRz~>
zwH9$fAKvkrNHj<m%o?|7{nr)YxVyJ1_Pq6EO*rn!s|0e9iLCQ=s7@-Z`VqJ7F~P;5
zk!p}qXga#Q+L{0&pKltZk9%<cBHM*%jhVtwQEdV*H0u~8&0d#cH(&Tc4g>{i)B^Fo
zCjl0D{!B(u9nl??xz3#BzR^09>EwKGJz?6z%zmDGOvVz{WZdJ8;lD7B1VN;@ig<Vx
zGlhAIXt<b$$!Bbv1S>6KZ+3h!eTWYaR+oGNIlsW=^mY$cj5mLMx2)~2G8QsFh%K6f
zg%Rt|pFnQeJNubpNRcJ(n}Q(o0qtj}NFKmADz20+3@RE(=8C{2?^=PKkX^G(q@WCu
z-a_x26lelt;_-A7_DpCT{qQI|YxuKGU5W=^S)I+tF%5_F-e|>b50ggUs%ejE5Bvkw
zW-qXY$uV-LnyG$37~KVFNYE{l*`;jn`QvmfAMPGVGlHTc&ohoQUsw(Xovn7q$q#rF
zp5wBw@N1SZqvx;Y3iw=L7WUYWRo^nU8dKJ|(w$>_QfU;Px|(F4IAZt_a<1s<YMDB1
zxRnQHd*TDn;S2BNr#m7<md|0TGoMRJFy;Yh`NVpYW=AJBdI|fZnlfFuHTp@M-`Kz8
z*@7`M+)GS<2%@TE%Z#J~euP{^5x}qC-%A~EaA{^7yc`1V(1dIl|8|@pClwY*To@40
z{>6$vE!+WZHili%kjLKQJQYRn3+a4r?zk0@`|QOd^N#C>p}mwr7<sY~9T6{A@;7P5
z{ZFlNV|afcQcHp%zmkdNYSxjcU+99uo4oBdrNepQBr1Co$;4||{+LmU$Nd5pasNfe
z;i>*BR`UD&v&ePzWk(*DQ<2Km2NPGuik^bxokYx5@4aUD+{**kaN>@R3NEunuv*7n
z`~@jW+9yd(b8-daQ2lgE&VLzQlqD*Z*&=BTf<l5L<TFjZ&i5fu;=$l@YsNWmjL!tT
zKH1T4B;Vi);!FBE<QEPiP(v~hL9EX1CO%2od*3#hXQ<7zcO7u!sTchlJwBx+VYfU>
zt2ZLf^RPGd@%$G$g4~}FZkzAc66{+=rJL`hF1-hSM|lVTRZxcrH;uh;oP70g+g_RT
zqzpEVugn#)5Q85k^}$*j?tw|be!d#&2QHf&y>U4x{vTG^08(9g5M5t|E1ZnYvVJOS
z_7iEI{SR1q{|=*v(ft3w73l)aaNTK}ue+VU9T(4X(~Y#cs^`ZJlxlZU<eK~z9X}ZU
z^=tYkvYiq6#H*zUk_co@i>0EK`=);awmnwXinab@+0>6q>d=mMw@V*+@U97!5M6H_
zP@$1WTUJOYYD;Ac)gXUdW*uVr_{jN0o6j8?x0KcXF1Jm@?;74;P>vo)5}tg*+q&o~
zX(lwMqHmttUJqp}w(r}RM-$xiC8=P6Fz1mWKVJYGzR`|hsl3Ub!c!F!p2+F!6N{ep
zx#-j>LPGj&Gz$ZYI*JVfWGwWND*Nn1Kv0>)04FGV|6@`Ve=?G`$^pvzD7w|)qZ&)s
zTd=D!c?w+*TIKtf9G*7BoRO)+R0iaF?A|o@>J>x>R>uY@#+G>!5=)Dlv8S$5nzbnz
zzsaK1XsIF~r0kx}!%@W#sb>}TGgQ%E(91mt&-urc@8vvqKrW_w10`y?d&=%PK-fi*
z-UKqKT5x-2Hx}xGXle-U)?{w<t`cXkJhYPRZtb^K)u%YYeIKF2q@(p~k>0)r8;eN9
zV@7Q4qac>)0GrBCTDy}XWv4~gcOQ6pvnizp8xIDfq&Z8zdYTKM*+v}ax;j=a2BI3p
zCWW)Rr8&@Ts7-&dcA7CW^^$i4`UR>Do?mdGiPGZ~I{k6l{%SIEcowQAcIX<z$;|CM
zTe0wxu$km|qa|4z>g>Zth>ISE%dml`Lg6c-P6_bZ0lDHiv5}})qyf~>G|U;oRd%6j
z`bkw%&u^(R>O!0p!BPY^ajsSDLi{d*^>gTv2fLr5BOq$y`(K`foX*_=a*=c`ImME}
z$@sWRRFrKpg;vWgOF~i|Onjen6;d0;C8;H;`c0{s6yrjmf>?Tr{Wj;Y*|6*Sq7<id
zF*LK>-StxVZ9Pd3!-u;3nphnvDUE#|8oz9@*nIG!i*_u`_U&62u*KZrU<+i|k*Im}
zPIEe)pUr3NPsQDyW;O*7@`**MM@Zuj5e_)tSJFzPg8bzdJS}tk5vITOGe>-bos7Q1
zDwI>+-<lQT%eMarb8H*#W;|D|EBO>{l&xp@#8x>XKMcco^B}+A5xM+<8SGSI+E>_Y
zzs05ZCa<+XahCJpJWIo)AUlWdpskzUe=^&usoUM3JCHV9@m>uG!EW=Qt-qmt#$)g>
zz*ahKoUb(JZh@~E&B0&${n=zedVrrOd#f{J%!f+(ckKCbSMG!D6_}oWQS)Sqme4#O
z>eLuP<*=`09;m6XjysvH1`^4iK9`_)=a8GexYB|;i`M#A;7jf|S~Z)`O{RedE7v=r
zz@)ZODGMHB@N659XFGZ28l2!h7*4Wb%_2V-b!zO*J2XVNK~2@ZQAtB!p<<c}F}X9J
z^ThQTg-DQ(x>C=sFL5C?GA#fWyxKEj&c~KQ$G<ihMmJcfov$a4#KA|EJ2|8Oe0esK
zYM;M&IW)gJ(!+XEi9w0o>+hew>1F87IHB2fD$kT0rLYFZGFLnK?fYMA)qQ$CD3-~d
z9>4xGbZI_r7F=eL*2L>dM$|0EQtC&&)czID<ooHPWAceq5bgu*!m*_2QT*a2q^T2f
zq84|D8nD(^Cd#dr)OOcn^H8-DWbR6AR_A9BO#J1a*7cIZe{Q?HJFC3fN|Eg0F`V%6
z6mqoTe*h0jo~tI81&-C;5Kp&4<JCLy+N;!$%^KL!Qs;4!eMe*VmBtA&QgEVkXlQFW
zQO}pj)qnARKn2S1n2SQsri)NqM-wqTwy+kBiZ;;Y5t^OYj_T@QAX41(XSagfhy~9V
z*tLuRO5X3GPYk09*}j&;YaDV)>zebtTCs8Uk%04vspG=b^E&j$Tvt%Xr~HT;`Wz(+
zbK_d*X1w__t1VM{?#)m6-xww$9<7Pd*|Z94lsx3grN4NShvUi@etnjd+j(f=B3Bg6
z9=l`?5Bv2iz2M+EK#zrxRhRHlNZu}yY3yJFM%G_w*fiEee*AEfzi&eOXhKT>cd2e7
z`Zus6w^-J{WAhy6)TX@}a6!2g2nUnuLzS^LsvGxC^8mVkIVDfS#_@~NzccL*rUCtW
z8Wp9mHvb=vuw8umMw60k{;U!Qhy)*x{G*xFJUiL$foMVQV7)SWSzjQNl7m8Zr~BUO
z<IWvsyGpa;`I@Ut;8DT{d=f#>xPP}vJ{926F}J^bm`Z=kR$q)WUAAJ|vOCVhI>GTL
zD1tE3MGLCWo!TCk0Omt9m7ozokD`6pGyZ4C!?r-J9MnO-VV>-6#7-mFK#MN=dl*sv
z+}(sf><e5BG<3V2;XSq`-JQV4K`wvaq9wr#+I(zHkSYWCxEQ07&jB^<u&S1^um@;a
z+3QdvZT4$guh0wih5nq|sRqx@siza!Z(a-sc2|FJCcQD#S=DhGg&2(#%g^}z{ptQf
zo+*$}TF<mOxPZHXtmXFc+vU!rviH!L=7FNZzv9?QL`xJkn0<uL5gOkJ719ZpI8U3d
zQk6aww`fMSg=LnsrQ-=4bG1ceUhx3bukb&u1U7BCT#OQZn>-F%H+>mcGh+WXf52u!
za*3F=wT@7aSkks|)&IQ3j0i{WHJ|&XXopGuAaDZmgU6ha#S~;}B)-jB#eH6wDeFc_
zet#!%j*KpaVBg#7QuFlOiHE~nNT7rgvKS(n^{cCIv%yh-oB$;{O|l;q<|77x=4->_
zdu{ZN-+fOcf#kB@bWcj!h1^}nv%9zro59)GP%)uxC<S;A=Pv_2KUxpZ89p@s=p>B9
zqwz#(=_&u29{F7eg{C^OITkHIe!HpivYi}58~&=LhP<HFT1t#FR{F3F^U3Ia$F+ls
zA2JU9(Qr7vOn4ZmS87*@D31|XpHHeg1e<|FpwRu<KYoFxUX?TKXYWiVPWDX}o_Vo}
zRWa*OLir-{Ck{pf1~|;2w5>9B(NT`%of=>kJvL?hA5)xfz`65Vh{?D2eF(qqw%aC3
zAn~vX#FhO+g!vp(1#l^gP8ZZ|+chpdJ?)D)I7WuJ_7XO}nwTW#ba}w1Gw_+xQW8Af
z*m0l<o*j7btBkp&jL2{lS!TMO#eXNVCIe%jB;Unt*?V%zBT`Y?lA<R|wakP#ANYa5
zq{Jz!*H#XXzt!GLkjs6S0C3df(l4rS+6m}^^6#F!qJ$`XwlH~U&50TuXDGCCy?tk5
z;#eOWUH{%~DG4-?1B{o=-5O`$AFr5qwZ~MnSf~U10gy$Muse$6G{)8DA~Y9*FBY27
z5BRAER0d^uSMPM-V6)B*&O~no+c9-CclR#E5&MOH6%~hd5^5%q!_!rLG~;Y^W1q3V
z-!Bkg{KUG5&J+=~cF;jBPP8>GJ_mv9z>B-KBL%A`V~<W?0qn@y+3O-_RF|m%kCmbO
zy<@EcNRh)IIL`m{9M8bESG3qwc9V841Z3g9jNtmJBJ<~Cw6k$!`2=-Ftb|LwZpB2P
znI;e^1?FKJuaLL_gE7mOSPwQ1pN;at(!e`5Z)LNaC~j6u6gRE+RBWlwY^m=F?m|@a
z_#8~%wgz`YfOQHUCO?h4BSvF7C3Edw>O4M<j}(!ZlMUuoN#o?l7+u*H1C3f87oEhU
zDDZx!IYNYu_{E2tpRO&(Rc4y~hP}57^V25gu^S?&z@i`T4cP@H6r<~+O*Us@vJQe&
zPmWT~9j-^K`gU*Q1$3ts;-FruUhN?FrfxCX4z93FU_@^5ADf`u;&Y?MtkW3{rI_7C
z0Vte5Q_20Fv>!U_anIiInde^R0pG*{ZbUuC-DU(yT0GR(OeEvjaChIwCE%(X6LW=k
z%V!#`KXty9-Dm&!^Y{d7pV3(1u^}BNqDw-5J#M7G-W_o+Dl_-&eg7IK*!m%??T9L<
zb7&+1V?H+)+jSyM`(O1ddghO8Si;shcUU1LR4+G)xoBz_7^O@iUs!735#|azv_QTV
zENi+~$A2m8T;d@5uG#F#Tc&^&SM?GnQobw$`Rc4Gh1O_(o-M8sa}}vZIM)AcEg3hN
z?Tu;J3cNx)SM?<*rV|CqAJ0rzH?tP&6YliOk8>ZXdi<ZK0b`fTWBtU;f#;j;TTkHW
z^d@t6_qgYE9e8!=d47Ao0JU#TQ<`lP-6fu#)^+<oU%2~%?r#6?&X4S_Z(j(MNA8OC
zQmH{Z2M}L<UoQTxZ$Fi*W4qL;`Cvy*Ohx9d=1yd`LbC4vJ@A*GnF7L_J^tN8xt+Py
zS4S7u1J-7jr`H460{SD+*E{#2hegr5V}@PNo1;?@;On;otX^FZSqiR81c7})0qv{n
zz_l}sMEBZWIdIGn)X^FE&^f*E6tnmoAZ+%!A7jP~4(uCv@bn4@7(`fK@5VSkcv%e_
z*jzj0nts77cz%jeZ|&b-5Pd#=(}M#xG1oTNsKqH?16z}zM&Q6Fcb})LiFisEAYP&S
zBhdEXDh}lJu&y<Up4M$Vzl=Q2KAdk84TKn*ZqJ*F#h>1secONa6JGE7_jI}q>h9MW
zkje9X_J&^^p~5P3<^n_mK!A|gmU@uzXZj<=m;3$g-HR8+o)^T7t59Snvy1A)$gAx?
z?xH<_pt1Q|q2R?`e7*Z={pJXG+w0eJAEp;K11_w1H(%cL_@B399tS9fU-bEJo{vb8
zE+5v*ucqpWFT1ynj`*LRl6wMvp}y<{cmzJ}W-LEWWUlV^Z*6Wq9XD*4z3!HeOkkeg
zUmv_i<31U|2OD4P0sZW+7ZU?+6xf;SIJ3a-fdU(g^$pPgz(c$c($mxPmGa4!>h=~2
z%%Vd5pYJadUp<6POkd`*HcFdsyg+2nU$=nq3i1YsTteGU;muj50&|0#%y)MIhbpLL
z)AzedYF_VzCC&u8x|Tj%7re|6RcqD~wGYIIBoT>Ii1_y&#y<)srXn=jYq|^~CZ;;)
z@0Y-!Tt1mJr~-?zvv|wB<4%u^7W)&R8km#OWB=m=D50&RApSTrYN-#(Bx8Gz^06_j
zM_s}6cT&>$%;^H5L9A2kCT1AXKIx{aUXsv*W30|Zt&D{`T1x*GUi`gTkm}F%OI#U&
zFn0kec3}3M?QRFYww9<?xb;P68Fk396kcX~H$!x4k9~?(J7W1KLZeYe{<<z;xu(L{
zGSdQmSrO`&WO5on)OZ}9DEw&Au0E;xk9YQ-Ateu5P^@5P?98w{Ca+<V_arp0I$(=c
z{qbPB@;9q{^)H;DR-30iJSQ7eVAjTDOS}Qmp?$a&htR>1n?(E%QS4=&bNQ|zqRa*7
zdVF8uY@n`QHYIpxvm<US*}Mj5T)<|E@%*|pbn`jtYTSdCAd7bPAEc1iB*$2oS4S3~
z>DsW}=ueEiA%vwM;NkauUft)k8(T#71hmGKOY#~Coq7(iZqXIIm6|ZWgCo}*Q2~xH
zTr9&6&1gfq5D$6>b0#zk>tD+){8=A4&@(|0{1B_@5u4e)0=VSmEe9gt{>Pc;R3pT?
z$OEmEg$^d=VY2<O`0HLTbZ)5S&1MY|U(#>-P=Ew%&1@<zq05JhEt+u@E#04J@ESHE
z@0^0*YC`IG_(2Z$6=9`)ALs{%mOBW~jSw%q-P)tE!|EV<gRTcKr9K_Pt9E=2adkab
zC_&Np!@l5n8ufkWECfKjB`W~xqpWAZ?da}&DWv)XPrQ6ggmQn0z!2(@<8)!{yzDD0
ze2x@0ZG4;VL7Xc+8Zaj`ImCeGqS2lo>c+B=h3jyMn^s$nwgWPDiS>v<gx3_6bz^Zo
zqTX;ZMne2rwZ>sXnG0K~OY*_0S6%~G$UR!mh^WWNOYwL9pAMjrpB8Cfv!WIcxv-Fw
z4T{`CBXclgFSRWq8L%@A+q&nL;nbk-jE*T2<=>5UbufL@cNq{xVe_HxLfZ1)8+7)7
z@)em4s&L-F>De25p@}_~F!8yxyNyzXj6f_)Z*~LKWN!mc^9Sv35*dSRo9i%T&*2|!
zo;KtW6Mi%Y&_=*Cs{9@WC+zYNN1=ZF!-kP_xiaH4)e>R4-D1eF`>OY_0VM?-OP-N%
z*r7%{5@q!GO7@(u4dkstLny<VPh~5uHuoolpiUQS!eh(Q<NL)VcG77ZMz1-qiF$}i
zQ3t%0TbK7c{#H5aRQX6qdq16ijdHZ@WN=OY{(+Ud={^8HN>(PY`oS7?wMG<7KOhei
z?m@rC=K6XNa2P)^6^ro2L(*cK7ixW(K89%P``wBu(bsem@j~>=z%_e8pQO$i2UV_Q
zL;GzEh3L53PI~%Sv{k{XYv^c{vQA3>rm@VA5idT{X*w~{!OOKRtSMG1qif}ct#>jF
zTg*k522lWB*jMJ`X*q(P)XAOWemTKTL}JIVW4oAq+xNjL5NYEL(O^1GkAtm7Y*N~(
ztK|*$90!Gv@~sv0aHz9MLq8kbxif^>CrnzSF1h!L01MoDnLS|ak$_;uQc|}y7=S>i
zH?qHQqE*w7JMLjD$!Kd_M@Ix@^YiH^;l`Ody9TyWF7tlTwf^NywjBteANf!!Uk%En
z%%^QFpg+GOS-|FOg_Rh7?n5a)Pr}8U?tvW>MMtyxvPhk$UGTQM+ZN1DIolCJ9r>7q
zo9?$DSH<&35Tx^0jx|BTmUO`5qa<3rYg;zh!einyDS_3cA*`nWNW7&v5K%q=O5%xd
z7y{_muElU)37FWsww(Ix$N8bTND5a@`IhMVihkm&uL*TUG#xo#h;<h@W;<gz|3*;o
z1tn>y7U6o(qwtV|ARKZecmL%epDe)ZmXldXgnCwU6)9vgFV)|Ls6LZ=T+5FRM_Zm!
z(~s+`KR{jQ(f+rl?lNzVd>%yQ2z{AC)(3p`&{PZ<=N+g@T@SXTJ;D93rNRkOf|QCd
zifj5(1^KoEfu2g7z4901ld?ytJ?e4MkEYa39!K=Ygd(bttD8C7RgmXlllHrY>AtJU
z#A$AKmm`nHJhz9&Sbns|w#b|F^Oz4rg=|ztsOcj%lqykBpMW9O$uqscn1F+-X+T!$
z{p#A8PByrE%Yj{V8p<XrbVo2{CF^gb67%(En(yY<v!E{ut^t_CF%00>UTkdwvk%60
z0geM7HU{?kOh7NaiD`reiG1mVmN~r!d`W$~pGUv5nL73-yVvHhAiR8^V5fY&x7J?7
zDYJ!eos&9COt3wzaGkT(U9gcr76AB{L}jKE=KK|MgukE~5|vSKfA>T%c!l__r}6+n
zjX>TA@_S@cGZ-FFod;dN^=F3jaf+nTl7TIpIPFbz`tuqc*SgCn3BgLyVuTrARdQTy
zZ5-@_YHBIa3yi}4F?rIMM-|Np&pY`%C<z9&QY10X5}ka&J_!_ub}cHD9+2Jqh&KKh
zr3|*^40bwq6vVO4NB_+%RD|QS{UL~2xs5|cK!`XDrI@xA<#|4-|Jf)Cj%hlXImT(E
zbg*19W4lp}@!XLzX<SGP26MYPG${p|uP=G9BiwPN8-TDZw2?^A2%?ODtBifjCvV!3
zc2`Q+tX(R*Ki)x_Hi9z{1xzK=5M73ne}Sm1U_7?7w>09lhQ1;z<~>HnfI<?#Ax+RN
zYFM8O3eH&(uGX<!p>YzmPqKv<NmaMqwtNOb$P~0ae@0lhCA#bgN}#M8yg87_c79or
z|H$s_uM2YVP|lFWrhXI+xr8WYhlMCH*H?ZmO+ChQ0c-1({2=3Ez(fC#VDL(ffx4e9
zSuzdHJSz!{ip;{tv>b5;#AFlv%qkv4Be^e(Ci5Fa9X8$n@o;>RiW)piTSL&Mp0-}5
zc%vFuyVerIN4bd3jF;JbVwakr%0Hl2K$lS;-$=Gs=>%^dJ&rlLmG==NLsk6Car!z=
zr{Hic!syObSZjkQFw%^>A<d>oL3WRtxMC#$uUU)7l;9H`17Eh~w`H;an{zaEJKsuD
zIPrRrEcgBql$!Q%k&DNsaBxFKZPMl`8U?>7Kt(r=2xCwTR#$op4osvVTkND{!a+_`
z>5Q#((lUg5dd{$KwGxuTXWnU3^$$<spL}a22R$iiVn2-li-^<)za20-JOWb>K@cp3
zEin~5bHgXql6bHZ8vTVSoapeSpIB0rn0%UHg-p^XagX=x1zyNJqIaBQ`UTqQ1hqU8
zEQmu&jZJb=;F(BTA^MFbYJ(pr4pGH4Lwp;mOha-7eMxYpLpR?9M;>)T5DMHfLJ9MY
zEK!w$kQEL9lXR`pOLFd{VUkj6w1WWos+;IUQoncY7*(}ksC<<&cvt%m1k0WNlPr=X
z0hN}|F^c>T3s(byl9TPE1D;KECJxUHC^ZMEHKOC#RTWoK>5q8IG;o>cxq$epNmZjW
z_9Qk?xJ@w~1bdNJ38%^9!9#yZnx*2BhO08kk_iC_CPawetS2z8@^!=m?qF(Hj+)^L
z8k~CN_k+C^WXv^TD~sO%cg4z|)UTi2hF^E>EJ<st?&h;Id*j*q{^F(8?&IDD!wxqX
zSuE!#&wRd}>P}a!BvN5}n+ScSDSTy~uRQvua+gHXXs2W_^L(aV*R~1Ar+wOKHu%bD
z!$80j;<|uZdJaZU8b0om{I}!oA%2ebC<nsg>6inxFgv47d~nNLB19<S&HPZoP=5qV
z5-{z9@%@4`Y5i*aZc#p5J3tnhN3cF-2g!`AB)tETKbUXh3SSn^jCjX8dX1C;%QQ9W
zx90$qiI#oK5<YCNnVdTN?$@xDJQPorlMQIf|G2e*r*Kfi`<*#vo=HPj+=E4kvi+1}
zGp}e6JZnrI-S6-ui8q}EYcT$}$46Ld=av2Gr|@TX^K*7Jb_);@a#V``*A_2R-cSy?
zp+|j$TVv@e2T|6jd7~lHYJKk$7<<u`_z!lCb$IF7@J+krK7%IsP&zxC3r878erCWy
z@k)tF#ZgMIGahqz%Il}%!Ea8Y&m&hB4Ge#CoYA2Q4I;U17#$9xWVoI6<5kPJh3;na
zVNf9#>U4cvT0!<i*bte5&iDa>h$4*Kv=5-Av8xQX&pY>yi4`Y;QFIKDs9nB{rd7d)
z7v?B7v_?nu0LKYV%8QZxv{bgO0Y>0#Gnx(M><}geG)R@nUXUHj#$S-HVnapIhg*K)
zsL&~p+ilNddLL783(MtsFk@<4QImSp@6&tA$ht9rP<k=`YvW+tXNvI?_T7afsYDl@
z0lR??c1nc)-ZZTGUKiz<?|}YP{5cjab^q?)JsdKY>1ge2A?mfz)JWHR4=_+G8VhfA
zx#o1KcGQS8b6=oK9mcCLcv^6W%S2_VR~MbVn2nbf<L^!#JHo(Xx50pTCIl#3;6qC!
zN$;W+^IT60d!5_ml|q@rdMbmNwa!v@Hhk)+(-(hh5AvNi$eI!4#s)D{vKUNwFjME}
zaL_M;x>=<Ng~}D)qEK0C@f%G^Z5c$FTNdY<X+(rmfDXfhPbK9Yadx7%Yy{!Z#!`}y
z9p?cpi0#`wh-c{BDyon-X#UQ(NFPNR-D<AX0nZry#2p&R{LIr@NSh9=)`B&969FyK
z4R9o^3swH9bhoDF@H3{|4&}DZRKQFSs;t_vBTdR4B~czZ>|rk{3{3mMkc@Zy(%vVD
z3rETK0H_o6CuWl_-8c3(sU0o}EeA(ES|z{tljDTBCxkD(zYy1@jxQ8X`u@q<8ty-n
zo%w`+>L&<lVQ9lLZSbgqib#Tn{;8X7(izd(Ov-QvM3AHB9+OZ<;Cst<rj1kqOl)NY
znAstz@3F}tu|Y2Y?-uGd9brkHO|}$))>6_!0kS>Irs6zWq|&QyQtDWvbIxWrwlS=W
zyst8l@J9TB$Ks~ge0jv>+|fJ2&&k;VOp>>dD;3nFVGpcj|7}gIq@k<D<?_L7A1p>@
zJWRO@7op(N3n&5=$VU!~+k-L%9z<LCcU^pdfpx(xcLERKnbwA30pGn-&~{YQ+Apx6
z11A*hia*EE5iLyN3OsZq^<ja$gyd@{NJsY`(cEKne^;gP*!~vE?c-$-S{>}-NgATd
z$TQh##->(Ubby#H8zp<-6wo21`0{g|QHZSR6ISP8ro~Q+zGlQ?6ogGO@Nf9!R=dPZ
z>GlX}Pw4{3`3Q|5s-U_Rh%j};BH-?!gzVtF;ZcWCs#}M=ZnS~eh)b#XIXYVOBSg%5
zqcE$l8Wb6XRIkMjTI%!3KG^24ZIuka8WbIL9Zr|W5X~7EnHaV;FCC*R5jFW~>~E_e
z$^%!N_r@PxV`v>_DL%YeGq6aYpSa>Aq10>`-Ul*yJhpU2EymbbADIREt7#uA-kY)@
z70=nnHWb+QjEifBLT<%DakdA+p4~n*@9d2ihH{2Pid(98aDj%6E471o<`N|3-M4&i
z;Sygu)`ORI@g7p$fTGVM@!Zr_dNR!#MAcYHoy_~R)8)=DebIMGxw2!qOA*0vQo<#K
z2VmI3GW51oS`)ewQ>wzE2FRl%L9U`@jdwdeV($l=My#o>d!3|GpkfRRr_aM^c)oG4
zR_7=Bus++}RfBV&DKDzx$;Wf$6l<Xr?E?&eF0T`&GEWx%2n<26ay!t7QH)6^D9O9l
zh-nsmf!z@#WA_K%x7IZJ(6TA)HjpF%5RVUtto4U>sN?o%Wl`H+-?fVQ(HU5(8I#88
zerRngw0p8z^)eU^Y^8T<m29seM(w_#>=5W~i08(l!Iw)>sL6_R=8l3w%o_so<<m8E
z2lrHY3QAyBidJ#1+x#UJ0XcVF2O=eXJx8hWq`HTU%!Hk54+h(9^u#g;m6rp3(?cu+
zY$hC%q?<#ZzUt|+Z$-y+$CrAJXS8DzLZf)mR<ky)URd9L$uf?i{X0J!W10#^Q&@Sz
z{kz*{XC7XA#y-9TZ(U7Ap1X;j)<2|U@%g;*+S4n8BVqk`wzaoaPL$Z>E}D7#!#4{Z
z>8&cbv!-fBVMuZ?v)9U(_-){MTr%8Nsmd$+>}$oht^gOU=7#g<oO#CSyAWMlwlkx8
zPG(yPcvMm1AGi?z>eNxEqK=BWuxjPEW9ai|1EoTtV|qAL4@yRN#OC+8NW{M0jTQ{s
z`C=cU_N#3W<#+wRRtYks>dJnn)l44p@cu-~0oDBwv-`{`30KbrbCd^M&9DC%xuF)?
zex~`P#+3N|OQ>YK=Ztgpo`TMwz7M&BbQd|bGq}1};%q5&<}Bj$v5j0m59UBfC8tR#
z(=7KD7JWlRLqst?qW>y#F)tsaU<Nx?zbyOz7EBSgy(hG(pESNvitkzYmUn2F+myEs
zr>Uk{hsh_?j#Q9mq!$JYxE*<}`s$c;>UAL;<%5mdZYxeCYQ<QXY8TAYIX9}jQ|7mk
z&Q_|W$<L<Xc|bOu30V)(bzV6`ypT7#^kdr(KK&rbnjFYjs}G9+y<p{z`;UrZ13NYN
zpIN4+aQ1Pd!-Xc7fUgY}Cp(*7Pj!s3KQWT(hf)%mwCliNW^N~dSNtMKoI2v{#_4eo
zM%V+z9iyH`06!h89%*r6F~UZAZ?bc_!n2<biYH-&KDy=i%R3@Tc|>vh29Y}dtigUm
zW<T(@D9Bj&TL!oQQKpiXuScT$J9WQLtH68i0wLKpTJQ*o!j1}x!htUub@h2Rdf!!m
z0{UYr9RY61wJaSV$(a}1muzI+A}b=@qp0I;FQBOVd!ynewB=n4t)?qI<t+aX9Baoh
z<ZdMRBT?Q(txtJut)=s1MslvBn|xv0+KF?G7^brhB=kA|WT;blpY$eo0)7&9Vq`LZ
z!aQSlQnxoj)VvjG#iO$&t1TTA^R0e77f)D%bn=qI;Zq&~u$s63T1a`$c<erq1a%N6
z^;W;u`j+2^5wRyShwzz-IC8Ku?YFkHo@kVkja<=So<yIO+KZ_aWILF!ZlbOai$9r3
z3+{dyuqp7d|D*^)Soo^*(p=6zwdLOs!VJ}FweRGSm?ar{JLo@XYwlN8-8e-pYPs+o
z!CJOZ%Xiuy$aQ08n~h5yv2h~VXY-JZ;VGZse+k5(PPM%wYs+f1;cy|GFzh~G(OK;z
zXu~$3xk?@3@4wS*VHjWJ|F?moW{Y)@3hJWtR34MP>pDyP60j+9cv$iH@{6)(c)53a
z;SYom5+2OAubN-~-x=cI%$*yvj&w(C1k}G%pa8gQIkiD5@HiBwFUUWkOvb%YSP!Fq
zGt-|*8XRbME3&5WZZPwv>k*CK2E&T<xM<<h5Gzxj*AI&B3>eKXh$b>ZJypu%0f~l*
z!xytT{+{DLr^0$p1vfCX;MWf6?YGG)w%lK`UXZt&?<N-E*k;OP5AImg)=_(ZM8dO$
z2TU!TcUg*cj`IE#^WT?2rB3zlF0kCPjfXb~s67)>Ex0)xV)b|R?=$R&{*+;`faUSk
zZ6JO_?p_hF_M=;FmM#nDU}YE6hXpg2N04hmuS(n6+}YwC5puo7T}B#L&5d*V86&J_
zx6oQASxkOE%jF;RBnVq|SLARfqi``50AUGM=`Ox2z?jauc`id0_n>2j8r{Nv*zyyT
z5ppO_w0qDG3I+wSz6lc)!pR=Ie9I5z3qq4E^gegq7SD$=ee-q|e{+Gmm3#KdB+w>x
zQoj&Y(ipEt;Ay@e8js+>+1q*;xC1GBbHn-{H{GH8+(=2Ibc@}^t~LU3w_-q7L(f48
zP3oet6q#_MdL-_^;%vh-lU4yf8`Q<}tnm94ugVv&yMT9B@+Rg4Hk4CRxHZK*f1g_z
zr!3dG_xPMbWIze}<RrN3i-1&m`Id?I`C~>l<wJEwIBzdB{cC;(1O<Eo=kNto<(VZ0
ziV8LhQ5Qg+M>ZJy3h|A%9e@RxV;w6Bi-XOOP%>S=pe?!dE=%cvor`{)B+Pvw`%7a!
z<9g_(>^t1X?+#-MEf*$nWSV^RLi=n6@RJF5UI_%e5{hr>t!{Wj>nHbiYfFkuDXjiu
z2|DGS&0M%fg4c`^cYd6t;oZXI?Yi`avOynuxJprBa%<uz!Q%N(a)4VwJ$gyhOqY<7
zVV+RjAB=k2H)Ej*e^Cf<BL6@Hk(PsR-lKy#-lJfZ<e6qmJYbwIibGn}lly^wL~x|{
zpv7Zc=Gx8Mp|8e8F2-JvOk#qx8QX<xx=BglMh6rYASGgrz=Z6b@p@1E3}rUKw-ns3
zmbDb|-6yTjL*$XG00QcF-f4k2{o=Af!@erjR(v7Mo#Vsei(wMB9pjeQumz({2$pgK
zyX~<272qS4Y-s&Rqw+TAgph2#8{%|U{62jBEvBn@_a=tI!K8l;Jvi3qluM8`O8?)?
zf%rsf_YvQ3*LB#Gd;P5M&8c@qD#w`h$Y;`{zTIC8NdksofMRiZqUkbKGPcb{B&&0*
z=l0f$-^&Hgz(FvbhJZLagk@7J>k&RBfGCgsbn|=c_|dE+ni*`la)OTpyrv&fuxA16
zn!K+2nM4-_*x#ey#6w-YOF$HWp}!KRp;l_~`u9wOg-8-`_Ik~7v5v#LG#vPZQFQUn
z@@vJ(M7IAiz?0bw+D@-yF?XL9xRcT__GjRv{M&E9+R0#h-7iLkvtNVy1}t%ze>Ot>
zhnGOA7RYC!#WHR6o?Q@v^s~#!G*)P|WE$zK96_{AiEp<fdc6STgku_SlAQc3tH3~n
zi!|y|tSg@RpI)4<XZgu@^@PFR-M<F4`}stYL}x8P;k>o%0^x50DkIFVXZXlF1TF>x
zd(~GpngsR6soP>doBG1<bNR_0D5u-eC5@59Sjvy#?Z*vyYG#PV<j==*K9sVGTgmJz
zJL9tvooVNPwikti*Yc(GR+9iRSVKO8YEWfBd>_mcz*P2}C7z2A>yaIg4#~|n&kVjG
zcf=%MXtyTXE&;m9Xn}2tZ}v#nj%LBuv%0d$tdmr6{l2e{tY974;vt$!lHlM+O>Kae
z=>p3$=fu}uk@N-e2~{Z>5RvU)%fxi(=nmnnb&31I*$VQKwipc0!$zrfANZkS{pGqW
zl)2oUGxA;1G&UUS=}^YN&g2ME!D!&9yXFPp@|Jp#b%Kv#Zv6y&RXTe&+Y*s?2;=X%
zFO~M>Krf_yI@?Tt#AxQ&xTx^!W#EqTmTN!r+E%f(^@<<`tC%VpNTu<9Uwql%r1JXJ
z{Rn4u;c=Jwj;xI`>JGfoMy~^rUlbq7o{-eqUZ0jED2)27|KL2M<PT3K@3;j<8`S`k
z$Eq6~fb4|nk1dJ~h&NWr>0SLpdPLQe#%^jC7DFtA%+&ECH<D?rZ$O&G!?1tL-)upg
z&45Ho1?-PBy7TC+*Q{i_Ru#p4Us3&OKd_3~1dMVQ0UZj)u>MeO?;g~ZgB~+4u$WF*
zBE(SLnWNvW+hV8!E)VibO~3gWIRyBq*HdCntG(f(GT7n(vVf4v%h}b%4cvYXhv3+^
z2;a5s@QHyK)9TW3P)<<UNQ7f()7AoYkCbP9XPwc(htu6CE9&^6!pAGyVLs;c_%xx_
ztY~EO`Sx^uef#nPC^UO%R=dm8_c5zKjO@gQ<_=V!T|9RGIbDyuDqNX*2GC!&?+1<o
zo`f!6W+REur*i9i->S~H-uc-2T)nB1XB#G)qT+G8*y^v{&o8bo7psAVuRAJ0m-x$h
zBGA4#fAQij`Z|4eG&`Vwef4mD{`h#j+nNdU=-+*L@NyZM+nKpZB@qaEb)SFUe%iPF
zRsMJk{5yK0FL&8_csjiVE<iqOmk$HmFC)a2lxLeySC94m-5p{*JuAS~^Y-@k!I8T7
zlU<L0p#RnGNNiyID|oA?4jU>V_B;|vwa$A`@cxd_|6k(3d=T%`&6Py=^MPs5<MQT1
z=H^rDX8(lC_NvR}Wp2PDNB0xaoeNcuDEV}&z#Z`K^v3L^9e4${J^lUv1%O{7UG6uZ
zu60oosj#u-Cy%#h|04eJ`uA_(<_fz$$nULgji}Yx-3|`Axp(yl^ttDKBh8diP6L7d
z_wN5ZAF6xSmtJ-wd%6d1w%vR8)wOT*i6tuUxrQ%EIy}7Ky#ikMPX~@}t&v_w7DF~S
zL`^rSsIZ^|`<?<peJ`h|O_!(tJa7Cz;l12kKfX49BQBY{WW46tIGO&}%`!GE5X7`B
z{nrcY^|gTG?;Bk}l7h0eCL4AdA3+UFDclOw!ZN`En#u$32_iNTU!}+Vb4zC%zP7-p
zNB=1LEPKIG>SOdnQ5tXY@8+tv5!4mMR}s=~2I*21Uaw+Z3ifJTASuR`g0<7k)T-yX
z)|>1{MazSB?CI9hNQ=1I<(Qe3^}p9N0yQ+l9y7FkW3V3yAN#f9Kk(k$0yCzw!s}4a
zctOt%*mVAjM#kBFkLIc}`i|`!yjQS`^)4qQ3TspgL*dG)no<J|gOK^xhB>Kbz@^`h
zaTE5#bTc)EXGpK?u_pPN5fifIET1OtpIWD&7zv`Q#qN2HKhAn^8U-1lsOW2p{J@?@
z+5MuF060dZ#|{ck)7bdyr;fv3aBu~s9RhAt>jwMfywB2Rt;p}bZ_|riclE%Hf18@P
zU8hLD`~57<E$ypfho?1D7fOG^XdWKmeFXWVF6lLT&c%J|N<%@M*0?$*bb&JZI0yw8
zUt2HgVe^m?Mm)aUv?jO0^1@>9POHMp4kcL}nj67q{%V7z1Qtz+<<F%Ini%&T)zIG0
zw2YrD?ln9FG}FsQ_5b{61n=M;D(U?2=;L@?3r|HD4#%MKDOEboS=K%Y`A{~?eC6}^
zrfy=~%1!N6FQxwcbS|+h1>!4T)$I?V_C;OpsnjHd`Xl!}#kJ{UW}q)>ST}+W-+T;Z
zwl-%9{T5c1c)5gS4J=E3#s3N6s;^GW$`)LW5koLtj7x|TGulMzu9%P=I#gIXpav)5
zK8{X&@y*O@hoyzR$n0r{C4MiVBQSP2xU7bTN#=-j7LL_>cm~snoQ%{($eJxWJdbPj
zvm50PQ-YsV%o(lZm+hQPqz+gKJxhNk4E}626r(3P(T?GbGFji_ZR_{BFHJF#`S?7l
zhaFX!iw~4miY)`oKLwZU*|me$cNxHH@>@MN*nis(&SX|<$}es%8w;i`$n=iL2TZaG
z2wxF1b30&H(0q?ex=QU~VHPhkflgR;*yYUWPFQC8|6%J?!oc9d{O=3T7Z%3H#oda_
zBhv&7OP0wGh9!j^&(s9NN@4Zp{@*cAUq|aq8Zay&*?(L`2Nnh<81Dc7QgG%kFswKn
v`j1SUE?99`noRXBSO&8H*#6}0{lD-qFgXAF$1o|GiCwUCNIyH?Mq&O7(G2T2
delta 98778
zcmYJ4Q*bU!u&!e}S+R{3+qP}nHon-lZQHhO+cs8`^Y61y)w${FncL~<n)mHzroJHB
zzaZ+-pkeHM`fU;{fq_VGlklM#kWx!LyWAF;uWolg;-xwXVUqNqi2<+}JE;5)<!=p2
zZB~Coc(UYh?Oz)nZ?4X-3j+g@yiu=};(V&BB^5p{E&)Grzh3Rx3f?V*dg1F7-TRM*
zmVoN%AqP)SKm2v)#{Pg72b(+y7i=vy-Rz}{<%Wq5&=-oc7_FYs-`3{}*+BpAT_Eo_
zj`H!jjku9swuc8$ZvYnGI#=7-I^x99E6xiApxAepyKnt_-UwjrBFW?CEcFeJu)+Vm
z@d5JXUUh=?b&C=<yY86dFvF*LC6ZS_Ktr?Fm7K@1^%q>P15G*sAY%V6>ZjiAL^3b{
zc-W?((*}JB>}cA~zBugAH)cOu5Wjt5bL?{{I>S$WS&bta2E_PJz6Ob=(hGJtj7&cD
zY1IAkhw?Fc5=$T>1u$z(fN<}=hUK}kGz`_W<-G=s?m}__du~!}KBArtcr}p4gq>WE
z<^OAvSD(6B%O?2Rq`0$tqTN@Qe@`7w<$!?XZ%wCn?lvjzuBYdp;ki6&m{J=yh{JqG
z3IhV6uFzq$18|ilvHt52K}?Irk+<VlvuZJG@?F4EEtKy2aA@GWZWf!F8Np#1p<xf^
z-Vtl^U5dH=2a)FWvRUpo05JT>LIiN^2J`MGQhAr`9LUc#uphw@{P-dues7tj-8FnQ
zfFMR%_j>FRDAL2N+v2Vx1g-kl`h>loGkfmuU3w4v093gp^bQGdqka)Ue#&HZFU9+~
zE?>E75W2A%ohd-u-x>mKvl@zHJ-FXVywY#v^|1Oojn-yQD(Bd{!t~eqrDxj4_?%yW
z4<^85p|thdL%9pu>Dl_iG$mCS(@hz!fXq5f#4vrZ03?3E&z<G0I6*2hQrin~Ns+jC
z+U(j^0GAjKbAJ6YqG-<6H15`)df$cW5`68d4O|B&j$7j`WBD7!Ce}_r*w<%}E+9K*
zUC%x9rn`WcA%`%~8{1k%YrSRePaA~inJI$^+fsZwUlgR@T$S4Syq$Y{^Kjd32<cj{
z^s8P%yydHzRYjygyLKnm0is<|fUlbVT;Vn*;K!;d%cYZ7G*n4HROv<VN<3&4qGl>=
zRZ=`q&^l<fM8MiiI3DXWoDcAIf)R93zqDGuh26rANvCpT*tOax;R;DkKu@w+w(tJv
zuo<-Ks!ai!Nz#)Natj?!0}&)M-KvW#Fm5~OC$7+Ra7sTilN7$tZujcVr8zN3Wv9^%
z5d11D=;6cUBXJo(>IMP$=_`%F(m*U2mc3+|HF3%aKOf@-onw=L81(!JHnEh&-G55?
z@IArOQIKjgY}+{V=W|=-c$pJx!@^ARGV{6ysR!wr;xtQFYT5*UFW$}aWuz6f0Z_nC
z%!>nm#+#QS`pQ^XfX@;HCrBJkfE8{5?DFIZF?Qqt?E~*YV2M84ND><gp~~>Zo*)~`
zQG^Q48U#9br!HpJgcD<Niz}t6&$PiP#<$6l?^Md1xiMH9avDKcogXyD>Bw*LSa&nZ
z9?qQI?6pV8v-piFw7lQ7SQ$2=HFPs?n!)khL27-g(j>>nc*5h@$WHBg_^?EPJ3Hkk
zJXMfOuB5Xrgr*ax$5<kB-tmtGrySATVyQ}P$-nd2=m^2mQ_ZTvccHt$-2S@{>+@X&
z?!tFtLU%51-OqXt7G|m81@1@YG!l1SMR3L-1;%@tyX_u<`9z89uk>0uQCWW22`tz#
zdd8?V3_^++UQpq(J<t?U6tLoe2>4H%H%KV`ww@n>T;aq~L}b6m*o0V9;SBMq2OD=*
zqE&+7vS=T%Z~b04eZfFX)~dhZV6Euf?om9Ue68r~=KBW;{hLdu3~^z1*iM{!P7gU7
z+w)UP=O?AB7h_sx5jSmHH;TL|dWmral8ec5Eb$?i(EZeI%NbC(Oj@mg^<3Rt@OPN&
zJ&pqDp`W$6NS?Hy1&l}$EWCM!01^X`IVf=)a)We#Ll9FVkhyThv`}c=SjGRgx&=?n
zLHIF?*hGwsmLVfVD7+XLFME12F<T-O35r}~#54X&a74U1Gey(LkIhib_ky&VrL%Rl
zjj=;r9*Du5c_f?3#Q;6d>r}}F>yu!u<_FISU`1g?qc4OlZ!(Vv;)<RXe<`(uaJcYj
zeQ<-pv=-1-Q{lC6;k7jQWD?BA70X*P1f!1xqa9{zrz^M_WvgTV@CS=IipW-9ieeX8
z^~ws5(Tk3Gp$@2?2X8Btr#FB=Tuf1R1NGI!KrvtU0F@$!)&TVSmYw@1!!PJEknN2a
zZNG^S2gq=ia0jsP06pJqEXmWh*K$oun>Um@6>D4H&(rB<O{S+uq{!rsW|7y#A?*a&
zk{olwpFAoG_SzZEy7ehuT`SltZBKuVIdZ#=Mbbxc6$*R9F!Fo8+fgf5EaB{BK+Sp-
zj$pnDjJq=E$^pKhvX8xsZYHsJ{|a|*+`T_-9w*Q3M)w@<b(mLJZnABd_X2=m5PPtj
z&&GEwRy#2&Eb`ii7WDME_?p*nk?)6t;o&J{9E9?MX$YL4;pw-ppG&toF5K9|=?SxV
zk^X+PPvH2(rPwNNWX`10!LvuGg7@=Ywr-7_^^p<qqW~Nz5YGx1I?LHr48biqkPaVu
zZ7X-ZoP96A71qMf&$T$SsS};CthWi<$(7jgni10(4@&Ml{AWDVb#Ru$;QvrW=vkH`
z;#e^sn(%`TU};O^&Mrx9DkB?=8UK<%0o~J4{3}{YKnK)#Dy~f&@_zur9`cv$SHrpz
zvb3sJWdQO95`4A5?(r0tXnboy7KDi64JRPk{P{m+eKbIib~SbaKjeh+>Og$Q)J8>m
zS7z^8o6}SFp*T{SUkfNagSIOI|25<Icb_0cgruE=?0hnK5cBJHxppxk(dXQEI=v!j
z)2)n8mjE^C_FWNFb1pJIL$5cmu=JfFSd?qLWB{GH67Sd3-+wrPnG`5nnGyofaxS9&
zG}G_IaP@Q`Ji3Nq2taiz!&S5)(3mMO=z*DB*#5^-j}QK|9K*F;a(}U&5j>_XEK8Ob
zb#Q~*qEVS@IX>X$##`j?@c^(YuT(D!Vk8<UTNzwz$-&+OR4`B-B2NrD5NT&;k3=*s
z<N)hXS>T3pe(tmhPeGS2xL#z9Cx-)Sf8)`2Z{ef6x1XCc`8>dnkw;Vm&`UE=ogM3n
zeC@fgzfCLJwFlh6!Wo|vPwm#)uVdtOW$L_yc+FVOeV~|t7PG>B{IuN9#I^03sBxfA
zqe10H@OR3Dp_%X3GWj&nn_1Z)=@=A<4ghG88g=xVF^!KDK2rr^3!4QeXcQsHR-a`P
z%N03dEG}_?E$n6BMMAXt!@pPHh3VD23OwMCDus#)OJ1P%D7JE5h1(;@`sZWa)}Xlq
z$Tga9Ui}^NK5|q@NH4?mr#+A3MCL<0_&=!ai059y;a9wMMCDL#bnpFt{4^*I@PLQz
zUE9^G9xwNz?+|CF*2nDoXt}w4_u;Vy9EBg=cU(P`sfW^4YKOnxvwWxP?&f?4VqLO@
z^f^zeTIV=ln1x3GRPJNxSIV!Y)(IxkBxM%8kp5QzJku+~<r84V8W1r>Ril||Ww4L6
zw6wzvhCIh@md!QYfXB!JAY38UYXBHRn^qHcS^1&IYc^O4>g(D_p;P4g6wPDRS*@ww
zm?W^6A<I3J<^K0pUf;Lx`^S~D>b+v!!ueh|+`eie5}PZPf5Pvj#jw9d6hiYc=74Iy
zbg5XsetQ;%#=k_YcjGjuKf(;#x~t{;c1&6B<p`n5!}W;F8@LRLY%|UsV*sjV?n^f@
zQ1G{;+BEGB^1N*s6r;qQ?g0e-VBr2+KX6@GT95s|`@PIinU3&q0<gbGhLE)`qHnt0
zW^Wn9t`saENZn@F_B`Z6u3iqLVJ;Pc!*l^wb+9u&0&!^_CQEnCIIMqa5}&{Q4<?5r
zAsUBsjS05q(b=JVTdib8r2%5m3Qojy|1!GW)nO4I$YJ{9@#pS~l4v|xhPYaVNB!{-
z!!|%vr8#b+RCLvTBrP~ykHM54IKiWz=Sj4x)6s{cpKJI`F!Mp+7&yQc!1_sp=O~eo
zlKmv?Sl)@=;8zmlb;N8?6|@rBi`}Mni3@BC_E^crX!f9?<xNg;lK=+ZvDC~$7<*^?
z237vva6cW6lVq>Zkx)LP<+k!FTPy>ye|7DA;X1SF(Eh$>K}5e(y3V&O2(oTyK>yz)
z2BES6p%U2S&XxE1tn#hMtZ8-6?UeyRV}&rQr1>d0+h;^~P%VuYd6edW!r%G6UHQW_
z3w@sY+F9;^;)D{Qx{zLu0<0sf5P2Erxn&X4S)-^uVw=t6Ydjc5*R9z(182$|zcR`y
zNY_BA4fJ)qZKzdngUY~N)c-ULok~ZiaFA5fM0$rduZbOlaJOaYlG-+}**LG`dXlMm
zjO)2Dwu496zFzKKA%28LQHi<HLfViJLDaUGsJ)obJhKZ(-1x_6uax_*qr>t!t3(N|
z*tBXlHn-UF4E)puSgp6ocd8eGxq>2;3jS<SNl_uwEAM<J?!n-tCGxo^hV`WBqKnFm
z0>|-vmfI_7U)mGwZUtjfj$G;i0xZs!C!Bij)r-JtI7Gp9t~vr`9aNK**gb1QQCzq_
zmsn!CY}*uYdy_m+Y$dT6ho&2;zCeDPE)i!miKLG9cWXq@;2|(ZnpBv~^L+cdEsWQ<
z(xLL@ThFw!nm1A-1sWF~uYW+L)!Qg6IgR0uTV)yvdCAnEv>1l0>^U)$X)p?Dc`*AZ
z4EsW+$O8ceJ?Vb%Fud|u6b3kY>`Zrls-X@YIolIp`o?Fo?Z8#@Gq<&j;;VaeE^!qc
z#B%BP6-wxpROx=7L=HjL+-I=uBwz}3i*4>X_aX6ylch~__8~nuHb$2S4$gZmw}$}X
zi_Bw+4a@?~as-3UE4!AxgDuN#0eQfhnxfmO0Ha^bdPjKcZw{qT_0NSqqg^#wi<$UC
z$FL`$@yCHu#y%OQ#CG)Gx?5-=vM0F{T5MbNdB*)tSJ>U7U{P&}Yrp>@G=zL~S}Gk!
zre3Sk-K!C@;nE55iG$0)Jz~7B#J3Q*bEXXob$ZZzg@Yy|acIA8Eh8c#CqC*N*dg$}
z80xNslqoXgxcAG5<Rv9D6DtK(dDK2U^;0-NiUOl$^*&z4fY?GxAQZLnTE;Kkgsfbt
zylJCq<*J*gwLtFVo@48zib6M6susnSF}xa==#CBm6^H5`x&I)V^hFRYd_W5(Mb}HN
z5&tLxt%W7V^sTk;>?Io)Xx*mJ#}h9x#=7=L5e*X&r=(vYZ%0ol4pI`F|K!N-^gS0)
zPl21RGQ95*iGRvL`evekw_y5aDHz!($ibUlK-jV~^p2_$(<s4<jecIbYsci^5);l$
zEZ2Y+2s#*%2*sMuH@d=mnd3>r&iSTnJ?{_7atekL8P*IXkNhIyOKU$f8pM*-Jl9PU
z|5toqN6SLPt?@7)Wk}!{j1Pj%ikk=U+LE@J;*Q-T*5<pjXm<2fFdU;fz8;PH50+%5
zS!BojWZD{H0ED9g%Y6s&pY7ABDVtivKE<I0qFlTSMNZkNyG|R<%qc`XV+%_~J$8!<
z6%s+fe@_HO>7y*$Z5MBN-`A7#%r{M^k}YxN>;k{8Sv+7J7olf=(ApS#9t9pi5aXl{
z_4#-jk2*#)@4QZd+MP>X3-HI?_tsn<->(iaX=p|kblpp$FSjKru|(|#UrSFJS*qSZ
z>^u2~Tak#U130smpIA`0a0Y@T-h_qF9jJR~%K_~}`mG#!-vL&~D|GLXJ)W@3Ws=O%
zfvS<uq$f)jE~g$63eSYZv?&N+szfHef&U?WmzhhmL&1ivv<j(XI4i_i)02E+OekwF
zk_SUNZ<Ewt^X!N<v$p%LY2%H|O%gNl5?Tvb|KQLefS|~SgpmaI0Q}b$t2@FeAzv*6
z^^VSenftvOZQVb<Cy5dXynMwvXn-MRB0U&B28E3+<H-|OY)}g$JCzjB@v<r!sA}d&
z+Hrx}!0kB}u24d#KgbeG$^ELpcGFa;ww$J=&KbfQ8V!$SEilSM%h#y!g3~|=aX9`W
z9qOF^_~>skz7T)ZMf7?{h+LwvL-V@?POLm8f3Z_dud?XgQ7pJ0jkgB9MzJlHfc_f{
z)_ND6fdE=T>RINnaP0<w+-_Y4VJA2prr}mtsscySquX)KFu+<|6jma296@{5sA&&5
zTshJVIodBUMl>jIf`S5TXzejI<I}u5#;fUfY?CB7(yVHr*Oqv6tEk$JI#z>EuJK=A
zmiGFZ3a9`2mz`CO#Sp7g!y}JB$j4yL5^mc6l$D(UG)N|gnEV%@A(<;>WmSpVUaVeM
zgy^$w<0t7_w9|{ZW6@LsdW6ypwp6<$6;^G-McW_l3JheP=-%I%P8zDTtrKXu3W4W)
z3~f)OVB<8Izjpif0&Reno=t360`2ax=)!2aa9=5K6`>+(JLbaJu$Vlgh_p>GU+=K-
zW*yW6b}Q%GuLle`dHhP`($f%J=^rIJ(jnx9Jr>23p@b*r8N<otnv~Sl!AM6)WfgYG
zdQ_t0!^El25sxS+Bi=SQ>pVl|b}3J5`#lh?;kGGnUs-zerxjky`*carzAHXy)%g@i
z4<%4yaW`{*uS$JS>+KNNHFW)Bj7}B^iITwTz>dD2vg!h8IY2KVum6{xsORvPW+osu
zpzOd8QCoN4Xd1*AJ$e4G#-W4W#BBn+c040`4N;oqYcf<mH@c1riL(Vrq***#3wOsm
z*EQSMXRy7`XMRlpVWu)iZukNYvZaVgZ+JJBo7J~%ZEq?!niDhBM57WZtKd}+pR`RX
z@OVb!bNMhJ&ToD@LBgUNW?^^?5nl%+a|WLa<g4-pXDBn#&=S^{Xr3(fzL*lW)krqm
z_PYqVXAHA9d#Zdk`*W7S#&Z3{F<a-A#R`FIjoR)K%@d7xGI_T|{w`-i&KoX<(8T4S
zh;-arGX8BC<K~NnV(jH1z9?8k(Mg8SZd{SoX_^Gksn*2Wt=PCy-%N}5C8sVW?MP13
zE~V&zk`JGUf0R>TjE(0Be9Cst(#c*Si;^fgO3tmy$d|f1K@ry>sX)_?bLe>F+NL@e
zXJG#rdqH(Cm+u=(&(YE;sH-I7b}8BX>9Mo07&bw2M+!fKnd!-&xnUpGRFYz%``)U;
zp*H~_rw$d}1L73sQsd1n=SWjMm#0FXMW+$hY*eO3QH;knVK#@N*%*F9P#G^F&$Cd&
zuBOUI>2)1-J*?#|s==qwUWmf&X#Iq0??Oi&+^L~zKlVUXXSCvi>D)tgw;i92&nZ<x
zYTgI&R<Y++k4wF}ZisCRfNU?$-Pa@l_eccb_45CIvg!(C4^HD;G72YT1PrW3ALEt0
zPXi@Ds)UU7-%<J`{0rk*!0%CDYbRQA9}J2s#M&-s<Q1-!@{i!QGaDx1LwI^3gV*h0
z2C#yau=tRhOp0I<n30ffsd)(&iHzk6+sqdgT=$%#Alr;T86D$keV4j2syLSFqyGRp
zUHC_&1Z`*&1Cw3ua8T6{(f+2Lm3t0?QZ1!WND;J?LhfKikTU|u4SNtX#v_gejuEtZ
zb>cE#2r^J0r);zawi4D6`6JY=rv%YtF#YDG91WG*?#gP&fNba7Herp=jHFs{Uk70c
z@bcyBAV|j6`1eWT6WP(Eb#`G6&>aD6BSxkLe?Ci58{`QPP#szkEB%&i*z$z>PptX3
zI_*}xK4m2L%B;XESQX(CM?#u;e*X3LO_Fn#*p~`4&Ec&Ul+=3F|EV7(H0j<jTqLUm
zGC3xf1~tqN%BIGmhfhh2<IW&vY+agZb<ubzMz6m-r9!s=&Vorxlj&fRY_$hK+)_0A
z`?}U=VIvnk<B6=xyv1-hSz1xH-+%QCyE=E)MbC5XJDa^xojh<>;vWtS*&??>nEyOy
z5|La-*E3(YzYm_how~%-)5O(By<i?3(xLd3pt*_FVs)5t*k2A}%<5qvId_ZIU(l@g
z>&Kn0=j8J4vtlRR2QWN@Qv3qU1nw~fW^X-<yONz|F<q$!<$1@=COFTb)8*YR-B_~2
z4w%|``15A+7uyU&W=&>XCjB**^K!m|&y3cOeU?hMGsB{aTQ$LkWYNUZC}kSdC@yJv
zb2t!LCu~=))@;Fn6MC~umK5Pd3N9F(|1vS?|73(ReZZT#tuLN;diny2jJPM^y3VZw
zW=N3>*KXOot*Avr*tT4Ffca^GT~jhb%LybF6Xe*kBNVf8{UcQcow{N3apKY>jE*|?
zWFonk<371}^rIvcjrA@GwRHc};KNxALDr8rxe3;J``kZjAHjBd;*}`SqvcH1*WDaO
z3RV7&M&&F0bcphnlZk+%ZuQ6GW}ZV7DqjC`ndn{AjqMw%n=P-KpIs9a%@ZWu=!r}d
zywr+*#k_WY1}nYUhXAQ~+`sdtMUXJS&D%+194U9}i^Ua>lEbVDsTmuvJTqhX-jF)o
z)>R@vy;5&RLIThZo@H;0Qm!DRvA3)|K-~yr1u$fOOjXS0xMqO5t&AqWrE@sz3!%TA
z>6Eq6{C>&Ry39Q{pqcaYAi=#dqJjdMJf-vhBg}USb~AH;xdXO&SLwNK^mGY`*v6)O
zdbI5*p_|*b$m*Em*jTYw)_)6U_$<HF`Eog`_RD-6Yp|G7!#8Vj5ow*i2A{oZ<k~9O
z-^SH4RUqX{Ljh<JV7$r2JYY+-(S<bB-*jB=4$I|CtjPo7Zh}ra6gqj~dc;CSlqjkf
zZ&o{s!Xr8O7PtAnP@*oW6G5`x^PQMhL(S~D6V3AxVrXVWlE#UUFj|#N?E!l}d8-_V
zy9|#qZ-1xvK7rTvMBA43t}0k88lrf(nym@Q3HYVH6ae_KUYM!Y1c8zXcR$u_TNpWt
zT;!AzatV}D&nU>L%E#n}$q93RSwKyDo$Ge2mfdo^F9YeB$8`BsrWp>MQ6gW0)Iaf`
zP~#X?S~^I>?eYYtI{eS@&D$K+@$8*tYk54@v~t91>fn8y<h1B+HHXpJkW{b&bx!+Y
z!UhQg9|7U<Q+87)>`E9Nxkv1ggBjj(QFzk78|Y0$dMS$EEjm~;hQ1Cx)~qIBok?+7
z3T>b1&Xk(1z6%!Y*3?Rj3?+;&6vJn`-T!PYAX3W8@F)>#o^4_7f#VhnEbay}Oy*kG
zn5$=v7s8BME*t&#d?`rJ5f7CYUh<q&P&7*Tga9djopO6~WcFf_xidSRWvqr^@!Lz!
zpNK2qVeGCr9x?R>$qd%t8oYr?ql=<OVe=b^$WgFKb?;@4W57FUh5ukpI=R8*ePNV9
zg>!GA0apYTS<zL^E9g!6UQGT2%YMWPI*MX6P5yRp7!)tKIfVG694&`&B%BjDt&>J*
zP5_`{UcQ1W-BbKsj$@d5Sf8tBA7F9-jjXp#B7>qHgB1$=@G<FDhk)mg9GXF-^|eX@
zdf)MYS3c;SGuumBCzEkgg=NW>raiITW1$z@k^)kmOmgDrRJew$eR9yuyHN=ee?*nG
z%CR$!k0pm$<a|?sKAQzrE4fAoJCc0CwE+NGO|P`6F}RMDQlwXT?pMBXegEX*S7QkO
zwJr(pyO(U3Lo8dn47H^{uwG~SCj*+s&F0rXcmt>Ptz{4C+^GEe->D?~z91_`)qdeF
z47pAuD$}w1?ThFJK-{;3qNKhGPE6ug#!LT#cASPBd4*1^!0cLr;DIHp)15ouzXSL@
zS39#8ARaf(Zk-6VHLK|JM(VJTdligYAY__-Tr=yCo7IpU$W~-w+=W!pNIOxM;#?SH
zTS-&6Yg(*kep;7@{#$b5zRU|4=ZzPga6uMoWa531SzU{iJj*1s!$@FyI!l2QK6```
z;FUfyj*2{P^{smE>^gAs2t3^A0lp`dSxpM%NCtbs)iXhAQJ8x$#Lk$g;Kk#o^CkAW
z=oEO9HR`Yu{W2$UacMop#Rjb+h00muIUqTgX;*<ReA%DNE8bfwy0kpTL8nsXrpN+1
za+HV*B{BOWd~<M?rqcBm0Lw-U)6auPN)IVh$-zdpLu4|V6mj=+{RLr{04mlXR7~PU
zs3}yCNi9P4Xa@-=?B-j-1L%%0T7;QsA-Ys`r|9wJhoE|kst_frj!eOtPpA=6#uIkd
z<b3GvTCmN4(Wp|s>#+xftG>&|UYWdvQky&X>nEW$aMb2>%_ljo){91>LloqBx8@Zr
zOy(g!pZultU_`Lb5Ic)-fW3WQ_-ssrX>`rHn*sB8mRnJoNe6jT-}yDu<XM$RvyuUU
zq=cnfm`^0N%~T9br4wzmvSY4=I<B87q<V3PWkq?heFROiX_&UrsiDlna?*6Mb<6WG
zM7E+#$8kke#>aw`FpaT&`i1o30jlTZhZ`tPSBP6BtA#va2o9+}pvO!kj~vrlBoEnF
z?71iBylO30`nBOZRrdX+9b!%y=aI#V{bfh!TCaw)+s>A56z7o?pMN&==+Ge<v@I&=
zKGg&DFloZxL#);}@a|n(Yizh_T+!CR#IsF!thcHAfoyQ9M4liA9KrK9lkvc^5EX?3
z&PUU7vePC>+_<O{@C52;I&$El>iX9xrw|<C%U{sGj!6_FKbx3RkRP{Q83PS}@W*B(
z%Be&P`nOG;sG~br%|*vgT|S>S2ajucD=PW|q%w$i;^okbDFHa(@V|NR$Dd&q-ICcq
zA`JPWU|{qnK~VV#HNldy$%+mnGG57%5X1?y2B`5!ndV}$0D(dt0%82S5nGZ><bwJd
zYlakm79jazzm!COo69C^=}TD<SuaqLlC@(;NzmzJjE!)1C(bljnNZ&z&Mzl)uI$e8
znMxGv(di1W8-3TGJqQf%YkFBTW8`V(zD!2zD0cdN)`@eH#wy-dr6M)%_@`1VF!|RP
zU+{xB@88!aK-nCB+T}65D~CrZs@F1i)J3?M5oW{#7fjt}J;%Oda|0xBj?w@uiann9
zNofB4Cqspxl;TvQ3($sYj~cbdzbZqyZ(g|_D{H6Q{As4mzes+vGa-AHPDV}#M`6m9
zI`3jd3)U=uwx?H${vc&}q2f18pOJK`VhX;y@*h*T0;0ZKnz@Ux(`u-^D5;hE?BGIq
zZD$XXEUq;W(wZ90`xINL=&7NxNKOZ^R6?$Jj3NjTkP;&x4kl?}ePPGh;I5J!PQ1@a
zKE1wd!}~cxFrV@s4Oi^kwrt5DgMkQN(d!5p%f0&@3ygN0!+Jl*at>MzTGw&+qHw_)
zQiLA`0UIbrxM0g}g>F;aiSt5vVvWcFZSKqsM^A&;c3nI8E3z%-qr?1;9Gyr)Sc_1s
z+)n?0Dwr1EbIl++D8#N3RV-**KxAY~98V_bZ^dtF3(@Mdbfd%K_LXv}2CrZc1T#o&
zws${xyBI_@Q}iqIVbjTYJZ^r&qX2;W-T>A<W_)gFF&ND{vNe06^X)9=Gnp#1I({rC
z!cXFlkVi6B8Qv2={QmEHLdlbF%3M(D+dDJ1x-X@hUM_)G=71=)>o#75!Txfr@7A!}
zv|>wfDWXSzaY(Nn7>`MV_p=;E4|;pxs=%xEfGYefyi;&7!`-A{{KkFq%LC?rrAYiy
z<0<+Z_Yvyp6@^Wbv5KY{T>hJ4`lx3)lA8c8eJ{3Oc#NP@Ah5bCsDD1dr%7LaQ_A8%
z8X^1_4y`3NeL$A4r-es}XyMTJO;uv+ZHK35Xf{B@`TLpcVbLB3sm5P{U<T>i1e-Un
z38w84vZf0>vhUz8d1E>q0NI6qJP2Exh`&>$L68n)jO?Y3p20AK$J1CGgwFIrA*4sR
zIfUzVt@e01Y|8lH8Q<gR6l3->l=nV@XbHN?$|>juf@#7PvI{>$+8ebRZ%|nZ5b+5r
zs~L-Xlz`Kw@6l~VUu1#jEgDZ~vTD=opIAfIRT&Gpw~9KO{x(YE0){;gegrnqTP_ey
z$d9@~4(s9^DJudUC~v@&@GJzDQ2V6d+`kVZ$~*Iy8>0DFOhLz(uT*;A?}z9$x7C7_
z`8xwjHmiRumn380Qyu#qw47MAjGO4#(lg{LZ_T9ma#SEes8$NfAivOfUW>55EB8On
zVq93c`cPo&YU7;z09Aq3>_b|r0U|zh))W7^CJ5H7t>&W-LzWXO7BRG)>$%_*O_NA|
zoMZKsP#H?V#MYD9bXRkDRApo$k+X5i@hk{%QA<ez2=JmSW+V!SoqmsCDzpKw$&O_(
ziwVe#V#Aoo&ML&@fcJ7XZpk!8NQxM01U6>1XVz_qOjqMoz>sacyrG##dE}N19Ae70
zvNA{JLBXZ}AROXK($xh!o}zk%acCH|wS+{1UgSXmneycHy+RDsI2@uX&i^(c>hF66
zJ#L%OuwMFq)A#f4Tpr9k%M#ITN**g=BfxP*SooEMAL$1Lr_1{L%hDV?EUe?1ygbWb
znzDAU)_`($Go|tjv`Q2{E3a2GrE~o1>Ed~HatLOsXZZE93^XkpC5uX~tex|{@}|;-
zUv38(W~ya@qW-}=7Alp8V0&NPJI1MyG^8|GpgEj(OX|-&YO|QN^BJtJ&K>i=@CfY1
z(z0>v!cmMx)>CW1zhUBQ)KDcTa}}&iBjaegaRBwBgUOSluT400x`TzNh^f#N95p@i
zX1P2(JD`VC|M}u!$va7@h><)T#Jq=lENX5pWWG{E3F%WrDXx!fXw7~Oi=d0Kn2QoM
zwFQrnD<+l&b?I@}=*iDc*k9z7m1Dn?@@n$8D^?!A50lc3j4+vH4NRkwG{vp(8w>nh
zU_cbF|2t^6QZMPZ{|iW>o^vZ<?yMpuhbT?>kaIC~zHV)wlOzx}x3V?cO+-F~fg~t}
z%dP3HbA|ud#Hd?^@7sQT&L_)oe4Ims)xZBWWj@w<MR^^e#k7TCPN^#O!)dC>w)g~5
zKlh1HHnp0af$F_7g&rgSyBT<*aY%ryT1>!ml;EQcVzRkrJx}#JoOsNYptJbH#q!mS
zi@xroW9|F6^uMXG{P*!8MZ|_iGy(!hxHgoDHKMb`LkO*T=nv>b!~cs%$YZ_MZ}H%T
z8hF=&a&s`mh--%H&Oqt<=qr`tW+?kPFF<1=0Ij1#5AHL){3lBZG&4Q(0gckkgpA~_
z^4Iau{MSx=A(h7F5N^S_TP|2*J!HGV?<W6WNusMzD1%{@(1EL_6-J=`)$Lv_M$0TX
z)>|pwl>wc+-zb&yc%)25!Ev-4E<sKc)2Sq;!{B(v^QNH-0(aAWZAu^{;?TgMDhx&A
zJUL*8C)^LF90s_P3KTSP><cWxqsf$IBFDWmOoHh|ME?yZz_=?!2ZJ-6f=ruE{pJLY
z&N&0;dC%oS71IxIuWf?cC*V8iJA~$!JAQzuALGMUP=7VuKPpX)F7kTCrj8n=)69pY
z$Zj%62mAhalN93DL(zLD6Ap>D?rPfIP2{<Z+^o`0+5jNUZlZC%yq!V^r{Wem9afr)
zDY_5G3kjUDt;FRGC6FnK59rd+Btd%2=4kQ^pT#Q6WKbi+EH-gvCZIpN9)a#~w;|Rr
z{gb^MixIru<OzJz14<Sjoil&~$UI+w+b)G2tq>*x>rR{yhJfsZNvoPkC&rW$NgIxu
zBX68nM*^4yZj&rSDd5FK#9$Qz7cT5*gV{qu747rWFkmHy8x~lP3uZR0u_wC5es}~k
z9vq3`8^ne{y+*8>pcG-Y*55Ur_>1hYX~pZ^#1Vw>Z#l3HZ4l44BG064<vTDj8iw_C
z%noiKfh_G5+C_WFY*Zn_Ma2xAv0~Jw4T)nSk^!1jJNCrCm*&{mq`+f~LFo)2h>rwt
zMYtVNcsr$Rx>ZjKu`axlP-#kpPy}T;FzVaTuR)&4K`Yz?lW`Nx*0YFlye!4DYW=^P
zb26PH1wd1B`-Ee(S5}UdFb$1p&MZ?k%l4`fx$6tp@ta^J{}U@hN-qyOQPn=Vqxc*E
zTtK*wh1;`FBhZG|E9u_!CdqX@s|6bbysaCDsr3?YA5cr-QQfM%NE73M1r*Y-%?Ehe
zG*mQv2<*xzbX`t(sDqu3)nNh30Fo+EX%8hswAs2voYxSZCQ%ufD8`Xq_wbQ870ogp
z*bi`s9F(f?EAH1iL<y0>DI|x#aRc~6JYXQ2WEjqxB#C(#69D4~-M>L|Mhu}E3#%H;
z(j3U?8?MUlWB8lg<|mYA8E!MFHmvMK<YRIaKaJ8_RzFxco*J&w!7$;_Ba8P?{_mKI
zIF3zI$O`K#T_+$scC-&m(;fX23<Z~&MfX1hf{CY<@hhcLWAt>*SUVspi-;|P01)o`
zc8nH~(5(?_{Q-B__$RkvV{_w^Fck6^g_&YwwUkDEc=J7<YQHsm20}pjIHAE%J=oH-
z1@Cu6`IYAjjalwsR7(ac0`1Rb(+bbpxHTKsklxIEb}h@;!qG%dVn>pHdm)e9e)I7e
zGJ-P&d!<sr>nG2b03Qfe4#n1*DL_4HSbZNE_FJEh^3uo1oiJRU=Ry|Kbz=W9v@o#b
zN+6wpRSwIqYe}|*$tU`b$WirH|3^ah2g(?cuOup`jHO;yuE_>^gpqM56A5_>a?G_u
z8s5o(J^y(kny5U)?+tHs-qQ<vh-l0>dah%y7;^L>p83vTq!)_ki=~na9k8cwfH-FY
z1SD{m^jrP@-b;!g($T&J{Ky!*EvPF;Vw90>yEo)e6rKxK$7lN9Ws-N&*+__@XF`87
zjMgoEGI673tzv(7f1DG=n48`0oPEPEjC?KxjU#r$XqanjE{Xexi#@o;$o`}=vN&1o
zj-%GCbKo37VxU|A4mw6r9*|F}demcr_)SC#8sqr*(S-AJBO@RiP25jZI|OWS%q5uq
zIBI4G1^}{fNwdTHHCOKloSo6b4WG-6RC)9`G}0}XBCbyD20np4q}@e4geA=vhXeNG
zczquAL&^f<LF4oaQTbJ@b7kJP`Uso8IX)75QI8DfXZT`svAN&01<2gU=73_(!pR?_
z`b)x<R4(xAp2db&Lu~~3^T*m?Kms#|lkohxO!3fkaRZJH_Qb<^Ef5s{b|EJ+1F@*I
z(Fca(E1c9X0rByiFnrbpsrORapzPU>s}HcjEezGh#b`_am?hm!`fs=pjqG89x`b`~
zWp}#t=R7YS7C$f@6wsE{E=^zxn!`e&H>|0Y)X!=JTMjc89O#Oa2!6Wbr$dQ&^g#z6
zX!IZay4pi2!!$9*IWaCBYNCQvpClPsF{%*A=P(9*GUy;!Fe=nZTnZLO6G~T34I{@H
z;<(}j43NX>4H23F1@5LT1aiIoUZ`3)zaNA5>OuWoAm%@N1N<-Gv#RRZc`QjlxI`-G
z%uElJ0f(xleNvaAobcH2G`;+g4YZwckEGn_VI}b-@=O$A)xxGm5>zQ`(xPixRH6Ep
z#b-E*?EEIBPt51&aj|qc&nK^Qw$UrlXf4Ey-OoTO=Kyhx%#RNmqdynOs8T)Lu$F9l
z#|V8aITGL_;l<oOK<?TUvmjR1J?g;rRSVn%Z37osUOOF%%v>kE4{k%TZ|hV%f2IRg
z1bHotb~w>&JFbDj^kYv>9*jtS!HtYB)O{#K;*c+(4ObpR8gc&;aScr9MUHE;j?z6Q
ztineWCr4<uE5xmg*rhGX@)@3hO&?%)X{eS8LIr4Pk|yG25Ya0OkNuSea>uo6e$xKc
zT!_EdLG737XGz=ooGyPrdhoTtB?Loy3xNf=2(aY^PCD&Ulc}}NibNi-Mc;{rBdi?m
z2FFL7-!v{Hc`U1$yoXmB#P)5C2ank}wB}QSE|eI%WqG{C%JUM9y-TrBa!p(ad61<H
z!~l}f6(eD4MeH&lFZX~$#@uI<k~<LZ=^0fW*^V5C#%H`AF&0?+-2=H<%=_RM!vfMM
z;3I;`Ef|JLV)QRcS!0K|ePQ)de!hJ)x9!<KcXA`EJ;Vkt^-Nm>uafVVUeYrjh8<H3
z<6fA{HVfSIo^vWZZz%wre`#XmyJadf<N<>zcO%`0ZT|w1ICxHcCv57#4I8Q6CkV$F
ziNnZ-^&w)zJu($Uhy!T5J_L?`%0%*TAM|)$0>`APYbHZ?AiF~|uz{OUj<ixBwv_?G
zzC{^gs7EDCW(N0W3>+IMk)+*VJ>ka?x%VK--1!*KR&XD)ijCynW|IG?UNGYAe*qy#
z&lDe6lG$xTb$fRFPd-Zi&*|;Y0gzLkXy!sf*X`gkr&)?^JNSQ2pC`uP%hZ#WtOfR)
zkm3+EA*jXm8biuI!}3<NHJ|`y(y;sfLq?}d4*|{UJ4_#ll<e5^a@eUC@JVQ@O6p9J
z-cTVYEzm`iqYvGV&6Lag$0M+u9sp&*-FPMSTuK)-_;GRbiKMwDJqZeyRX~QzAYRcg
zdj>m>>(BbEz3_gte8Wvh4Kcx(-E`9Mo&g(Q3tp}{t#ChaapT1ToFN~%EqiN>VE>Vj
zKtGK21-Bju{J|b$Ffs=;nK6@$7=FYI$s~Wld$`~SgjPIZKQrdmq{7Tm2;lf1g~0Qm
z4IC6G1QFQjr~;Oa@B%rdN8`UVW6m%3a=-*vN)JIf(>1)mcW3N-=rw09UU_IZ5!x%W
zuGa)4Gbxv1lnlD0-gY7Vw*6;>?hKqRr7T7*b89MQ<jxyPksI~a14sTXwF%E3`ztSY
z(*YTrcGPs34^Ex)JPk!^JAhZY7AUQ($ddg*vQN2MZ04YR1^0$+%0HCh$7gi^7xFZ*
zQt*9kq^X<xU!N}*?ry=p7s%hP->3U15GVsLfnFjI2`X>Q>lV4U(zPvM@7Bk3pX?xA
zJud9}ti?d}d;70hlXuL|_Y5fQs+O=9+9xKcEGG__K)g=lzl8UlQGh~df0`~I$ju=E
zt79~#v&Pqa4x~q4lZWXO5Z8G2omttoF5&E(D6y`LP2;~G?~boc$YDi<Zh-%=OWnf2
z8v~WGw^gFnr(S_NG{54AID+o)2b^4VP%L)}RlNOOuL0|<Ab``y&#`uIBS{KvAF!K`
zmA{|KDPL0_TKv-J7~l<`uXmwsE~qZg@B1Hkj2+;c8{6lL>)+2*S)D!r%dhVne}^X;
zZ^x@_Z};-{VX|*HheUCGImlD-R}LkF>>f?TFhhSP%N%e9{Irup*?w*XOtZc&Y<k#)
zBAYgijv#Eb52(HmyJ+7Vu<>DoS+2tgI#_kJRUkZZNEQVi2jDQo^NWpdIU5N8vh&Yu
z8S{X>KuvS<r;l*O8kpUjajVyEK4pt)i5Ed8A6+)@{zJPj;&aNuY_ZOcv$!Y<T2Ds`
zc5PW8T4aqf56bT#4d6C_yH(pMoAS&YmxLA6?A@^#SC(tR#LPM9dRy_Jn`$&%yQyX@
zo<S&;2toAg0Bg0{U7Rr+mdgL}?M#?ePMgbl2IX<*L+2iH=LTtUABy_&Q`o1FG2%rN
zXa3iG(GvN&7TVa>#-v|jd|Grl0~(90zf-f*|7lT+6ExmIP)#ObSIoM$tj7Pbgpq}j
zno9QjEMQFYHV1JZyD;yufyiFXdmv&g?jq*SZoEJMBl1{dL6)5i5X-t}6TiE%Pn|U;
zx*6&A%kOKv1dbqGHETaF6d!f*VlDB|M(FN)`j1&?S?nj5e3}d@=^A{#_1zCAvDAcj
z#BS{ViWkO|?`Fx3C*Z)-$A6k?Hn{V?C{oj@#@opWw7-yihYVI9jMU<z)aydzU&MMi
z*H^s(l6AI^MmoPT+Y9h{>BFV|c>mPJcl63eBuRo9VgBF$kBL_~uwIOQeovE3d-=nv
z^91Cp$bqt8!Na#;d8neF;Rgo|5wTvbQ(cyiZj&I-l71e{T35^ZX@iX_Lrt}M3tRAM
zw0m;&VV!DO(8eha;>V2se^eF&{Xxv-04ZK~cqYydVEgR)C{#*ZZXJ$Ivp(Ky_(_3&
z{FmA8lq<8FS4stF8<|*yuHJ+Q<{XsWAG7~#FZU!gs4T2iGmWjN##Zm!o%UqCShQ5~
z8MsG_Br-5RICJLgn;8ymIRz2xqUL46gltX;O`Z#R>1$+l%#62ncaJbrOo=}NfU1xQ
zvsq3+w52BFTv)qH;tqCd*Ra)#+?!TU5jBUY>dyQ)+2{7z4cG{&N_t)lX4L@~6G0B{
z%da-AIbUdo3EB>OO{psPG;;LTQw6S|WgCSb7;lg`+<y=!$!AgRh(6KuyGihSf(Q{a
z&Qp$XEF4jOM!7)vPHQ2tjQJ4~V6;WSLQg8qhIbxjd*mwkNV~UESo-;QZnur0Eynob
z=TGz^H`(451IkNtS0XiU`LH`4=bD?Bc~uv?sC4~sjEI#aSHsurzc*O_WRXcyQAtv8
zfimk+Ny<%#d*8EwC!paz#jllzx&deX4@kn%I9a|rHHBsrlyA=L9-NDAfI9YyA4%gM
zi0rqM!*HQ&_9>iHI(9o@=dw=sToy+l_1eyu4acQ&nWmKMI@8~+x%1iwOU6=nye9cH
z40dojLRB=f`0eOE^2t8fqtH!YjHRgVLz9;Eb78&(umPwhPZITgF|0_u9;J2G^B*~O
z*{hIB0xwxwPsiG<(L>8qKuUH)HHT-8FAhoK+hUcp9&wA(?wWf5s^&QaNUc0)D4{-#
zV}5n;+G;Ud{lt=8)r*_ziDI#O8IZo-Vjp#<EvhK>b&GqQT+4|=^l%G`t0yRe$s<~2
zj8+kSe0GzUw=a*`9|#-R$y;7+yk^N?Nw!TA>)yY@8lNifntAkQfOSsPKg;QVBriUn
z=<OGtlojHlV|TU=v<7!A&ZdgW3A62)vs$#O22TdB!tF3KqQ?(WoV(U5>?LC=efN=I
zows(%3yBi=ya<7~jA6Zbou@m}V9W6e*?PnP`HKTRvo>@D5JhYSYX>G!wV2-!mcQpd
zKTbKPVwGw=YF9T705*|x*Zv&l?JupjtFgGOjoIgZdmUOQq<aSwp@4qwb=C1zSDo72
z0)6C-3AKei^V>RmO`5Dht&PdOvvb}?*{R&LDigM|a;YDEZf<cFGNdQu4G4bP%4gwj
z%DBlAxfYjLDZCaDK5G@~1DK)y$p3J=Wttn=q(lB=2jsbLKo(t2Ec3j0ttq$3W?d`+
za@d~*x&V2^l_GIsZtREadw!#9g|+MV@%4v<29gM@6g$-#b`R2l3g=9F7fMg{%-GwJ
zv`IMy<nQ~bw~xC^(;InxO+;=JzTF4(yN#a;AK70qbC1~CihRX|k|aNb9)(Xn`Vac~
zzC{eZL3#xufL_nwkDxyG1zm2i{xjOorK<d(w4}>L`Ha+*GpI+A^kO~zvKmQz!Ye3X
zSNP`?rK#_UNJWf!#}RD`KS}?6H|qGOlWdtheJAq(nr#fGMoxcj{;D79vCVJezGAD>
zMmbWxUi3eD>fkuSef~Z3<VyV68)d^!5jld$#Vqm&a2UeGb}p4t$}iQZD4a)I>IHQa
z*=Pd%fg|c-gIr|zPqadL3&f#o<=WzKn`pv}Z3V-{%E}eT)4L@qO7<>fi!JOOnx8?o
zO(faiTKJk@zY?74CExIHf0&(uVv;8@V)K%{rL<L0<(CM+{4C(8pN^D@Yi$;d%1drX
zfpTREAf^6OvvArZd#Ra$<*W!EtUzVC^08B_SVMDN2rJ|IEoe}g_VeAOV#YNwahfZ+
zS_J}G(2)PkL>oE&bom_oR<5-Hl2BQRVHG1E4q`4S0iJ>g9~4O*h9*O!n$lq>50~WC
zeT}-6E<?hPqI{@VxQ(_zW~Wk{OOj&S8zOoFXc7H`m$LZ1Vh*2@!|Zs2%Kah2rIT2{
zzqZ0YGzJ-t0QM&`%|A+DNXp0ys$i**^chJ7k;)qR1iYIPG4)F8$>MB)LA<GqQSE22
zHKg0*96cQ;MyhVsbermd#{;l#RMf++bzu1fOSF-Qlu6ym89~2hW*H&Zvvu_Fw?rcg
zfX-(vftC>&1A9Yf{V$P@>`VkylK7)wRf^c6mF(bOr?O({(D0c2plSvEA@*%F8K4|?
z;^!bD{a4OAYUa}NB?^HXSS^@UbJd7L`KnB%zSNmvS*u2GYx!k6=w%7BYe^(#QBix-
zj5pZ>$ZG^h619e1wWiM#5x4LCf^B06U~7GXib%04EpydmV~qP=R>S?=Id!~Uwrcl^
zeGP|6gtV<EOo#0=0isr+&i#_Po;iF}V6>Df#!r$^&eP)6LLdF8Gdr7;_#s!L`Tj_J
zu+c&N@9_>MhD+Wav@fQYLh(P}!evOSid?p5)!veIwc<}YS05zdyRr-;2?Asuz{eob
z!@{QNR;aMq*df4O_pS;AU|bYe(e_`Zdev7S;#szM8Jh7>!tUgJ+J~5ZBYaA@Yxs0(
z{aA<`Oycx?j9-aqtt~-rf>{>$zBGEqT40yT{J6d9SMEWLH~y3$SR?8BUO@B;v{8<&
z=Y8Oz;%G33|E66Y-GI*GgE77VIs-xVyAJC7dCtby4&&==iS?fn%Ju{c>X_$ouC0@N
zZ5ih`p6o#LG3-bjN4ZMNiQZ2-KcjxUZ(O~pzx^Bb-^7-~Snq~5?_TE1BxUk#i#I=K
ziHI;}nR5qEx%h&!A~!?@PRO>+`_zA*X+_RKSAP8~gG|NU+TT2$&dMDD{9(QV069td
zYdO7E+eLkhe>0NLTrYO|w@;`GF+D&0?zV@zkKw$Ti?`J#%*X66p=mPB)I#UlryLPU
z+lALpFRo2a*8}J-oqlDF)72qnHa`^Erp#<(X4Kp6Ibb#>PmLB%8AgGH)8JY&ri_q&
z>Qb0Av2V&-I9<b2HL;U`l~2yfGMj4@s7l)%y}{CflR!)Dr!)_Sl=1A!6yL9;kpMbZ
z=|lr0FOL;0zYiALs6fX^IWuH<y(XvFMK>l`us@hQL<Wf)``}%+rpzAP&-;(u5yG3c
zN%goc);qldH_?*L5jH=lGSo-hMf${f6g6RcRunKJ8CyM<VLAx_ucrhZDNL^Q;#&i0
z6?GIAFTt?fb}&;c;*MfEXx4&Fc=OJqT@)rGoVdPo)a{mrpB&$`=9qBF^_MEEmIV~C
zgz<F?%r$~bUI`HSYPMm-sSa>&1bK03Ay6`PDaWP&Cg@JHq78L+?bTtG#DIjT7{cbj
zP_yBK{enL#K*{-lKWJKlP5)a^U<XnxPeb1YS!v?r5guKdc+E!T0|mw9X|#$VKCXu5
zSoiyC@fu6bBJTpfZN)YD&8d5LR>eC3l0XPrw6=t^eDQk8mo3zdUkW9GHO+RpGD6k6
z7I4^yV7mWq=Zdievnb9aYgue_^|neyRufoDB4KnM0XuW^98()M$5z=fxkrszY>J}I
zS2vj)tu=)<9YcEEP%il~oY0gl`ZX`~9>c|{<MbB2r#5W!KJYVZ^FA|MI{6=3=*~bh
zYh$BhU8+r50lU^e34y!TL<awb2BP5IE3#3gS3T*Cd78Y2p}FV0Si_7H@S14auQ!>r
z*$8UDcU_y+32sWfvzigb)Fe2$i}}~Qpu5{clK<^BS`PSp4xSd``Rr-l+<XrF&W2D`
zm54ZSXA>tmrHlEK%qVIP9D1g22@03R>xtCoeyfT_!inSkJTVQB6n7<H*Y#Ub%jEm9
z_+%(Z%cM$gp^AsLaeLLy*CPl&ZyNbwPsR<9vN5|K742;C@jm8xcc3JZM(I66qj$$F
z?W{W4SyWGxbK1$BNloI*Mssu)Vi%jzaKd$$VZ_h<_5F642gof_`miYiXG$cCi4tpT
z%mA%oU2)@2<T_XrRYukesQlHYNwF(Y)D%htF8ilS<FA}CJ+yTok2!~%XR%8b5tj<!
zT{AzPqqxlHrjKnTHO=@jF4$UknR;e^iT;;%ED5(cWsDZ7RelW|_r|^J0UKH^E)tB~
zp&UIP$u~3c$gw3_w^6R&{eP%>=iu17pnEvUjcwbuZJRf?d1Fs(-q^Nn+qP}v#`*HR
z@9$G})v5VoPMtY>ckjJ=_1Zn$H*y^o%RE<<b1;B@(HTHv!zmP<nO4h^B}WiS14XF%
zwDAg=;>&!pa!yaxSK8(7aXazQIEpR;H9%fSN%*UrD~uLFPGBAl*DJt(bOu0-Y=JD{
zX{ap_#e`oFit+sEJf!kl(BZGC<6kDt*8yBpLcdwi+PiOFEOz)>1J<E}8<V{jmOR}G
z)+HBTav;x_1e7)AKkUp5gD(O!=8G-)1-Ek|@xAla=kkXk<R9TcBLP>P-jA84MgYc7
z^HuNnx(w;jjzCZCiHaP`3(dhdRI>i$pOd1A?9+lE`ImC2Se#60-tRXQq1)Tp{USf5
zRR1ikSu98&kV?b=w7Zfh_w*SEwanKS@8O5w_j_BTi%5+ok^V_CaiaxE%jE5iGR#-?
zp6#nh(I>k64QqCYbc=kN%35DSX@Gg&Oznjo6*5YWSwAeFmVdw(=qHxk<*j$&Y|b9l
zlGk~zvZNNhNRjl+?Yp=Pu<1(0-s>omlVpKSlU~J!l^E5=Ek$h+YAPh{tbw+|@k=ip
zEND&;EQ{g`6h${h_Qv5v$F3?wmRtUZtGLOpRLNVdmNhGXC5V7XDTn=VW57EOUuz%K
zg0}<?Je3K*@+g&b0FdakP?Y*yY_cwU`6y!xYfvLklU-~^m|$E+-+jmwmjmZ!!*@#q
zZSeDJ8f2<P2(xf6F-E9JXY|Obb@yP@j(4h|)4u#fY>-k-G5IuY6y3-x48#Oiy0)<^
z=#(<f{KDoo1up4O$7!Kb5}-R@n;&`@Ae>V;8&SS)QbV9dbAMuY8Ic}&0waTUX4i^D
z^QR-j<n|>^s}Z7TYkPY$zryf7O>ZR4VA-qVK`ofk+MiouKHoFkuIc;z>0JGyq)&|!
zGMf6r88x$78%Z><W?=hsX$nzyD{n@_%}rw<R#o;!!15H`a$u+04qzXW)3oQ~&p%n+
z&ML=o9+LsaG)3Igai>e<qGP?6)hA=KPjF0d{2u*%%0gU3!y+BT_T1FqB&xlec!Y!o
zHezXjqnXO^Q8S4K9CQ>Ugx-GTie|FCL5FbG`ip&Z5PV0UY{Guc7@`e*9`Jh@laF`q
z_=+Ob+bDH33wt_(1~8kO6;4MPN8*K$vfpVEw$Eg;)~D&Y*mGb%GpP3X^<4jk(~-$N
zwG7M}?(Is3V5@AY0kQNH{s=4`B*29v?GWB=r!cE9kVU+(&UtEV?#)-O&wBBB{093w
zUT8=A^BDeZo1`GlPL7UU+e%K&qtUYcwcq-hk#n+bk29$Dh_&wee2L{zq0##m_hUrH
zhBnC!Qx7h4gsFHG<oUhn&ETrRed3znuF)G)(O_>S4dj8KE+@fI4h3LMU0r&{(B?bK
zY?2X|@at8ED;Lgawe9mIsW|UCZrV=f9QNJ9kn2?CgaZ!FUQr*%xm$0iV267}pY6FU
zCry-NU{z#SOoT_RScaP}hbo11_F0U41X{N^Kt90xJNx>2jW~O*>-(4Zqn|H2IA&~o
zAizL7Cy&b&*Czn)z76n>Vw~Vj0MW0pJ653#rGE@gB_>SGY*5>Ws5FWg(E%wNa(Pk*
zPlJ!FV+F6;)HRm$(&+s1lmN?oq802QnaZ?&+v#zX5XpQWY&+@lZaR@2pv`%TwYF_J
zbCxhNH~+f=4m>_jZ?BN4u^j8N{vdc!n}x{?i3a#F(M23z;4h%xZhnmz-b4PQmK1W9
z;FLh_8*5Ev8(9H-!~G-p(x;t$xu?NC3XX}tgAj`lYhFei_8aTz?)O^fmADO)V62Ho
zBXt|jL1-w2^stQQ%~0yLzafAGA&2~Nk7K3Y%<INp(4i0`6T#hM1=F`U;9zzln;oxw
z?k$|4`NPyu8v($=D<JaN7m|dPv=W<qvq159#2Js9r0-Mzp)Tz5Y;e$D_Q-p^gp=fJ
zaul1&J8c5EW`L#`C~ReJJZNObtJkVf2Qhzq;Etu^Tib-=BuBm(J32=PQ|AxZCQ%m?
zK1}5-fWKNO{#M7gCC({wGC3a)n-t?X#ryMkubjO8WCvjNCYZh_2i|HeW$!@CW&~SG
z5D7z6Rn&V>EK(Oj1{!QEH6ZkXwD5r<%;kX3ZXJsdrS5hi{&<t}n0Vz_zJ#v1`)#&d
z-o6<z*AL~Y|FQG*y02k>Q)uU#TKuClQoo)3Lq|np6d}#Qh9N=ne6E`yN91;adr+eT
zrzFaBQ3Pl)-^U0{U8A+;ZV{CR13r0J>F`=yQY^_8{nqQBz1eh|-EDr*6yazwDP7Sp
zndSJ^=<7>?iCJc4)VDvVjT@6Ii*j1j41p*Bc{eJy7X}`_27GK{Hv%B}O@2>K^4^LG
zh$|^AzOs_5QkS8d+gsVI1*bP4_mY0(FyV}NWd&qFdDC7uSjf+7ru%plRW*AB!a+qU
zFb3%hP-Rd_dT&JT6h=EbiLTJylc^r27bt!8lC0E=X>ZJP^~hkOto_br;!i(s=69_7
z^n}l>@thna{)?6Ge}Ffs5TftH=ejMnXM!wI-$6{FcoDbzw{-U}>iJ(ma&E1>K?wwT
zcYA=7blPdBCeghqRc8YT^W4ZL<e+K(tM1onIIjqLySq?sN|cUu>Ay0kjk7K5mecF6
z<ZT{gX(@YOYEvU@SjA4@eaLZGlm)-mR(c8a@H4m%!<jyc#SV%>v1vYYq!?su-coWz
zMRC(Ia`!b{?i-k#@^nJw{(X^`4jD{UltTbmt;rIo_909;)H4p5Dheuzo1$Fr9d2&=
z?&-9oMeE||{0YTTgdiJmE9p=hsiIsQ;1P(}tF13XpG|1H(r?z+x6;olz}ODd<mU<g
zRss89z~xZAK(-u$xI$@$Y`}!{)f4ezY(mLg7OPH&;B2ZWkd@VW>Al(<rR1Dn?SSG^
za>AHj@cMs9d-p}ego@D-siE(Stf8?2%L6{eq13>JMzcHB!19TJ$v`Fk@e2<WsVDJ=
zA_pYg35n7YA#OqVGlNi)NLI<_MgbYU5!Zr@L&luFPAH4_%JL*4qcXrMN&|^obRg|R
zN>h>nwHzQzL&-y0QIaA4)-3_L5eC?@wO{bewUq7MUMNcUf=wzXAyd*CmHd9jw3nKI
z(o$Dk!G4NCC)3$e*u^geIWni>jS6`0ZI$~YZg3$G(TD!A@&OcDt4UOlOm6eM&RkVs
zs^Mng?i3rMUsE%q^zVZ2KHpM9ImA-%49akp8Fu0FunMR(093yh;s|cZrv-3+$0pQa
zzm$jH_FF;n((~{U$E8$m+dfe-LnjJ%TstGPxiOmLt?*XQgd<217Y6*!9sKTASTp9h
zsfK-w`4{-w+2-z44K=&Hx}duNgE}JL@wQO{r6KQEzJkNb@SEGSF5-(oxl2lg9y>P<
zA(-M_Q^Ukr?HH?e6~D&vE(f5>4d4`M2cC~~t`=SM`{IjwY3S246uL54>+%$;Ds|?8
z{EQTOt<OH{9}ZUl^8hRrA^o{LQHIVr_!lDeF3R~9N+od$-$0tX_9A0uI7IN}dzE#B
z@H1Anfm5b?SkNrlr315`7FM8iqW4+4)_g9wX^)aHoHP?MWXts*UI+ju^UbN9)Q_WW
z&Yw7#Y>#nL5UYcY?Pd1%dAyFG6JDp_bUHZT*Y2~?<RYG?YJ+G(`-6bv$IQ^^k?y|m
zGKy`;AxS9q-+AnM>-=0PK-q9}C~Dh7%758!<(}`pT%;er@LD(1sZdCC0b}xTD^Svn
zrle5~LPbFT_&X{Z0F}f`NepaHR2~-`)!(;kt(w(c*zY=8bO4ANJm&CsP>4(wr4PO%
zjduV80oZ25Eyb=nu7Jis9h(Xm$}OfgFuoR1Ao_ra*J7Utjp(_us>FTwImH13tJTzT
zaQi(*vl?c|&r(8>Kc|k<q_ek0L5_7H3o0}XSZTPKwH}KA_;0W&c-VW+HH?{{Gs&Yo
z$buwqGwSjhcGwFseK0m><D31S!yC$p%%3q^R^WY}N=CJi#=zaf4Mr%@6M?;gf|3s+
zj1MCm{jd!7te5uN71e`bFmfQCrY3(57&RmFfW;CjfpcEBnN>}e&k%+7LPGUDZaq+E
zXSWz6a0d<p1PPP!Qxnhb-w>m0rY&;DMk7%V$+|;GkeBG^nkvwsY@)LZL!4d1NN7kQ
z+Cy+l##-tYa!F^4L*=?9Cd9vwSVWp$dZ=^teQD>Zvx#55&<#@RMMpH+c42Q@6EtnC
z##r#b;>Qp}wLc<5{<^3qaBjAe6(TfTK$<eB6M-KDAnO|;Ukg~eVZS*#xw$QMwRgAo
z%<n3Kb0+S&<8W~%<^c_oU<8w3_vXP;{9?l$C#-GX{-+p1D^umpnMj1P7KDHAJ;vR)
z?Cs>9NdiQ7$Q6S<U36k@R2I2fZk6!1g*#_4+t!Uz6+vg~#(P5&(gE}W4b;OUhJbht
z0;F&Yu)y}ZvAVUZab?<-P7-8$xjXI?VQXc@@o04e_5>9*08^d9n{$>?VUx(BqD-gJ
zL*MNOs}<U?MU+s4PTYW~ArZj7P2LE#m9SGRkwkE#r;euowfn>`0J4315r==wN=a$u
z`I`tQ_lKV};i3sRIBG_B-Z+q~o6RKfy|<D8%)Al<cH5T?p)1l%KRQ`Im5B9EhFhIo
zp@(2`>G)OYaqS076x({|;#4lFOn1Ts(?BjIV`@s=$~AT&VcH^5II$nQ)4BQIy!x36
zRLVt1v(VyDMx)Ub!yQBmXw}I2$MpmL2JHtESCi&_NSuv4#v;?ny%ey1)_rVjVty+D
zxM478-A9lzMb|(uE8PQ!kP@+y>45h4+bJ~@>1=ipyiob`LaLu(_mb*uh$Mr}3bt+4
z+mzKBhEx@vQweLFUYHqdtmDu7>3&~kio;&@fQVY>eQKB`q;c1v7lNiv%7S;1WgP1o
zz<!p^=Ct>jY`f|N4PMi+kE{+KFuZ~O5JQy!^Bu6Piry4T7ogXCl64=;d$Gw=k?c};
z!`lZ@k^1Wv4ZcV1ll0SvYS(Xqbaqj+MXYBfwoLT)_x1M#yx)X?0)1<QRKo56uE%!D
z_ZhO)wBAbwc=G|>D+Z$mA>#=(+ye&OupW1cUE~^f58;Ye){$l=Vub)9D-;FPdzu`^
zPx5776oOo-0#VWuo#%8iz&IXL<9LTE-O=&tPbg1A8+E?tN0M}V3{yOq1+(&i#wWgA
z<Q8FI$>OQfc%DO36ki$(XIBvJaVHG_CIeilHx#g`4B-0Wfk<q^a;)7Q>`68TdbgEz
zbGEaz+<+1jyL>QsMbTh~Q9wgWOb)__<&IVh6CpyEHyDhN-Q=D~x$_}DZZe%>jYa4c
zRI4}I{07$ECZW6y49w=5zF$Q3dIK`|C%kNb@ilq6RympYfdop+J1_(kGeawZoq!oK
zYVhf}dy?H!?~)+}HsL)C#`i8f8}9J33;N6nw;S?=p-D%zA+})#f!ZsBp?ZoN=M6S-
zmP=Xz-wWE?QKjN*Z*t5{g6Vlu21mEc-k=k3hmbLtT>n<qQDP5ZBCI=)xZ|&6y{mm;
zIXI^m7fC$_md!)B%+s+*hzWdSZNOe0KG^`Oc-_!&y$tu@s!)GXRAWw1c!=bC^@!=o
z6!dBio>24zwk3u}my?nrY(IT9)ad<(VXl8I>*(Rf9_0rIem{DQJ^m$Lf^|jodV4Ca
zkt0&%z?#D+9MBsMGwzUg`c$*d;X2;phE_GyfUcR)j0+3J#QzWg9b8<1{WaTZ$;`fY
z{xXv|N)l@`+ceakdB^7OX%3Mm%2<p<o>>V=B}n?*Toxg7Ga~75sG>R5mZ$j0vjzic
zsrh%(K0-+g#AjZ92v`tIP)10Zz};@z+9yc@tl~(DNy4bAl29ylXNkKUsy7?6&%pz=
zdSqGfR9IgRJ!g3~9`#ayRfCVL`XQqJR03Wo>XQTyx_w;DJ>tTxywKS~U^~k8m(Tu}
zr>J%!AG8JYYxAu6s>^`-A*!h|s^(^7d%x}j;?zo@ynEm$<fR0Bc2d}TJm%JSx!&=T
z?E%NH#f~;|kP$2MkCy6?f9n$W@D*umJtpKL9dp%eF%}%Ip`w(4j(Ddr*jYhP>_c_t
zUpPKQw>vGW#XB`i5Lq_6NyY5j`lxlg(VeCg_<Bq?L6eQh)PU4~RaXY;JfI#1x3FD$
zgMZ3Crc1CnFs32!4^l=*H_Sr9(R>+Xq-MeLRut-Al6aDGe411ApK6gajTLtNSdIR)
z6iJefAWzuOwG^2o9BI>PFWFm7O$`VgWTXz%h{(of?9F#J9c&vpMcMv;kck)D8zVsk
zdZl=9`u+!XJO7oqTGev_8iZS<xU89t-1T(ABY`95SlInOYQup&F%$efb#5D1f{hL-
zfD1bSgm}rtLR(9{&*&9h=Gj&SzKaSMAFR0ulo-I_zCyGTL+58u$;qS+fmx<=KT3fl
z@mCe;Iz<9q+G&_FAJu6WuM<f0?<%uDx<HleE%z5|6JdNh-xu=O78?()lUC(&!n1|7
zQWi@ZeSBv(w$<rQO^6Q{W84R^RLpB4;ICu=t#A1RH%#iBkhy;EROW;~M*RU;4<&~m
zOJ%|@>@b-?+Vev~6heN4=PRG8MVh7<(XN`b-U|k3iQ8hHr<|Cs0=9MU?4hs9$taI{
zf}x%M5^h0$x3Ry&evS|h!ro&or5RC2VVM_?1qMcFUFX`^&O#lg^0|0kw@X1E0O~h1
z!+Y&<PPK_^D#}UO3bKcj>Fz;_>TuNc(6bK%_9!%y_tyjC83%m_=LfLPp?(Mg6v0o9
zfe(&)p&n*}nK|wIrzn_{`&DzvoWIq)U$0H?F`F_~<7BZN*;u_f{jz@+D}p?olhm$V
zLI!Z|N2YGOMS2Ulld=f~zgPEa0YI}gb3mYvG=Rm?$4oz5%-wT@m~V4Pll$0UG4%pH
z%cV~$I(@JLZ(MRy$3-Afpx_hWLdBo}tX(koU$@eKi53X<L+JHGWUW=TtQc^{xT}45
zOB?+dh~~o|YYThhO({{<_M~JxZ}p)B7P2ahf16pVIO;)lFc4-cvipjB0l?ePv5kE~
ze6g=lr(EubJkkVyvzVJN5q_v^%fCZ=IQpvVKJvD?^UD(|OfyaWDi-SmlXnuk6%qq4
z!S>yQJOb>c1-k9f`N;s$1Gj?iB15WhV>mscqRwO`)*S#zg>n4q<QkZ;L)QCZXAYMx
zsMqO+b@;auM6lVy!&k<T3D}>ZETN&OCm(B0=a~$vYDQmNRkr4%PLYjVJwi?kJ^$9M
zN}Rs0U0iZ~%5U~av*-kY?16v#TrG=3eHGod#?D4t!SvCP<!Xp%o4`=r!kvk+V@@V<
z%vQeza@L7A5FJ9xr+<O;q{$~f(N#wN(+~>CtgsN5!R$d=CxJQO1T;!S7;vY~_*3(8
zW+}eu<9^M?x+Tuttj4w^&!e7%gNrpSTUuanEzRJ8@7$WbxV}8SY0(4qY;Njq>3<d@
zk=``LuG^+Rcfh%{R~+*;@mJ6ht^2{F8~0|TPgf=}ARNF}d4v3+iZtVgq=~p$E0UZM
zb`2p&1=jQ90J48G1MsV<F)ZpNOudhw<MP%|GHrAY%K5voL8pl@-o<fE#$=XgxWYnX
zqdG&A3$3HfXU9Ds&TxvXJH<9qtacG!;y*n9B|Lp94|J_=VZP#d`xij(CA$rAe@s!l
zMPxsn@N(r8Lzh~z?;ijSel2zeqsn*)2cpV|dQC0_7AU9_0Duljyr6H3g`={|GnfX}
zV>Www>Z*($&c%kZ|6zv$R`gMQcc*q!{ctH7{w>=$MTr&Ex;4)JKP66=OR39jy>J0b
z3p2T>c>al_Orl!)pH?oV@totMK5}1`itE%Szbr1sGu>}OJYvwBn{FRE2rK`yO~?vc
zS><9d8S9?;OHe!h)#J5LeGD-zmZ{*ob8Y0A$*!Se%pdYdEN8bm`swDf!({6h<ygS<
z4Y{9~Nx0yo!@5y}ZLpT%lEHqsu<sK}vB2s(rp<eJf~Fw>AjZ}hH43U+F+YnXA*Pfw
zPgJ`<q3r(-U=_4syK9`Jn#mC-*Rc;QjA=4dUJNWAzW{UJ!#_Cz=6mw%DT=Yx5jkF2
zd$%DUe^@$i5E;wl#W-5RzV5V7-YQ&MB(~^IS0F#f3D|Pg(W4y2Wdo$w5@Be$5OS5(
zDN}&~|4nZVDHijcUG1hoSRNa7mEdbc@ADj|X#EQ^ah4>43^cB02y;AM_TV7QcM?cT
zg}wANjp0AAAe8Z8`_Uc36=CDbng58){5JV+PKE~v>*LEx?4TX12}^w<h3T2{$ZQIX
zw`98*OMFJUeL9EhF4+H1Kmz{%5y|mC%!+;%E6k8niOD6WAoBjyW}mio#S(QdP8;P%
zZ-&r<ZX8ti6`kk&S_-mef(g$QUfiI(HjI@;SrLAcOtF{^GvAHy(|4u1wE3(QLOb7R
znsb}z_63$6XxA*h=QU{x10*ga3MUYr1}Ub47XWTk?^jj)#~O!4bYd-=qv=nG0jI>`
zHs*prQ#|hDnLhG7?^hL}R(I}V_cAyfiER_wK+S<W*JA#9Ze(n{6S^^H*9j2s{?9kf
zA=2FP48Yxiw}Cp(a9&@!yl=s-gF+{XC_^WKLniYW(?k?!b8}Ugq5g}GphgP7Fcx4w
z+pCbT?;6eUUusJVs8EdtULza6#JAq8Y3$g$;P%I0BhTTK2Xc?kbS>C}DzBnqUvC$s
zk3V!sr;s0XNdFdTsR3e`jACD$;}+S!I-(&u)Jd*hnc+(ICOG$lnrFBkU*+3lfQrqr
z)cB$~1S;FW=RfNZu5`ZWx}Q|F%2<8I=-;N;2NFMvwNcb#S}Meg-#LG_jXgB5gf~Vr
zkfEmaZGP*rFi{4Lw#M-w3;U!~V{KJ%4iA{-e|99UPC&cTbUYMcU7%^Js@ZF#!u+0O
zX6}<9Wr7C?Sx7dsrLl&dXb-1=eqB;X{W7>Yuo=|HDc#beYife;nRX8N4*N8zT?Si+
zt6l!6<lf<C7sK5uW`k<B$W{?tJI3+_B7>4fP9&w2Qjnr<mVXTEE&ppg>_y*zYQxZF
z*S703I-%J8Kj=Op)jv*>)TQs1|9N&eR&4<Wf*G34c#X%f|6RL84-EZU>z`Z&57}gn
zR>Pn2?!(PbM<}h_K_XG-$NSZB`URjWTKrq-1Ps%kh^&hq(SL-xUjO|)-F@Ll6V9FV
zW#GdTeR~9(hTNx*dmYxRL3<1IlswLY3=8(>?;`Ee1t!2yXhond%n9!`e2V{FPmzAe
zC|6P_K7g1HP0Xs5g(~S3iM|0W6!nqvbPi>?0J^p4wXJ9)Uto5icZ|_vq0Hk<xemrR
z;P3JR;2a|hoZ*d94BRAyVilKWdhv}4Q(R0<=Gp5>$htO<m&dAW+l<>@#p6Gk;pprK
zUZAe+QNI4Cov>+{0SLc)Vsx*5u?}CeYTCX2BGaW4ZCI`>Nol{mpl}I;0sv7w*^-SJ
z{4G8*_hbqU^>^G0@M_~o4d5ZHHVRNlVg>|ZG>_M+@taiW*q3X46VIewT7UnSE^Ih*
zQFIW1!$DyqvzlyU*O#_kR*+CM|9P~slt(+<6~1C$<_YqnhXwRI=lx78qP$BUE*_2$
zanL}(8>g<9d~$2Qsi>I0kFprltUw|Pkf=nMRavaYdQ%=_S8xPFJfGABW*a9ipdM1#
zeEY3$n9&Un4o4)OBZS>G$3}&b6uN@UlI^`}IV8fSaY>x6ewFLv%@XhYh7S6s-*YtY
zt8Io+8|l~~cif~tRr9)TB~8V20f!Z5$;`?W#TPFTjuJ<a-O%*l9ATqAfTF_2-~l2-
zVD$s9^E|-*tJ%v~k)0@<ev<VXuvV`2Ue2KpAJWlfE52H&91)Y-h@HeR8Nd=&&3a;=
z%an8|;Ig5L-fY-@5<rZ|QfHjG+UW+@X$~!l28@)9S!t8k&S&5~gvmI>ty9Lrf}a?C
zEw7SvkvYh=>bPjuXW6ctS=@oDAfzOvOO+5^)_$tS{aMXhE*ZDG3d~jj_|plWJ2v)K
z&3eoLv*DS#$K`oLK$S4;Xver9Vy0MOp+yo=y|<z=tEQH58S>L1W69hv9?$l3Xb<Ak
z0!E~(%;YHZTl6`pr-vxZ-OqSH{kCDRRlXN%bJw{pVLe0k?){Ksv=1grcfy2tMfeS2
z0+Ylo!qT24!R+ELjMI!0uw;Q3?3_;v8ti-oI=2SPgG;&L8vRM=5ik@$sXc6KTG;Vb
zZ7Ot=UYaPgeswiP*N#m*lc3WPds6C5Rqh$17x1-IwcOFu_Nnk#<n$dr*#=m;*tGWH
zet%a;S4^0)`6j^(DRp|kyD+zPkA34<$HU&<B)KBcaQo+$0!R(|Iasx%{k+!?BE=rG
zY*WmD+YCnf0B3(O=>I%${jS6X7oQ0}$}^Y~Jm_uALR~|URggplf^AaG5HtEWswvw}
zft|C_({VaRW#6Aq0v_+OD(EP0>^jS!FDAtID!)S|zRp4u9jn2City&AM9}Qfhq=gm
z2l)~+$_`P!zO8x&xP#aZ))yGFV)s>Wm6bHQ<$Pq7k9@6nRhhP!wSCV@`q)n(=Sk?`
z#b4Vu%gNfqupencHx*!Xx0yCWrlbBe{4(s`d5PtxvF-3r(vHigI3Ma-&bZvskUpJR
zQ7Sl?@zF3EKA^0$BC9=ZNDD{~za^0;jm;#22>FG}b^ZAa@KmTO_o3Q_heDcbMF)%h
zTvjhqKT}PfX!SAt7k&F0N%itb8M{pw3TqHFQM7&3Q6pNydBcD1{#<Pzoqe1td$uJp
z@cA6b%R#&FdaSjRjr(C2MUYL7>UH0n&b5P2_nz@-8wKe6R8m`Xd#uUBST#OsZY)pG
zK5ah0Qp!&P;H`W;=kg4n;xHBI`A~dnn%~*i?7UmHF4$MpD}HZm6Z#xt3pittxLI*p
zn5AOQ&4Vzzs>{_)cIPvH`kH8keTdq3hcRQE!DIV&b@La9Vd|MUe+C?!UndOsX5UEy
zHeu4w1sFYhS0D9;A35U3AkNiPL={Pfnt4iFnx4P_AoF3i<XY?cxYHpx*5x~>s(Trq
zqfB{o{p`o>pY=m|N9f4uhXx>*;gm>zSIBEt6&Qh%7*ronDBZq2>ICj!l{`@T@!&zb
zDwdZ9g6w37>XtWZ_Z0Za=v5o%jU*4odn*cNXL@HSfMC(I&>Q8wc(L5NDvJ9hPaR9#
zW0qQg8A85Ric9v!k|L>#XO?sa0C+LHB&J8$NXb0v3Ck4vK_)b-X~RQ{6^gFDmvzh+
zKM_n%4I%G`{ui>A#7l-H!ok)4Q+t8z*F4xZh!VS+|4OjEH3qZsW*9Yge6bgNb$@E)
z=L%LQRTxVl!(vS(6Vv^BQg6IrmVm~eM<NhVBl3tZSL+p1`)Yg<&iBfoVx~3Z@dx!`
zj=bH3M2m>-`DB!_WR>0G_m~VE?Ol=2$=kVp`>SKfSM>}=Pd;5|oXwWmkM!v8ltA;{
zOC*1vF3!cyn%ht%<!G=$wP+`#yM%LXf5iEiJF>r|Lm`j9czZZL<J%3zh}_a4nEWfC
zjPq#NI4zm9>LXlDN;!Q#d(#qGdwrncnU9z+!hsuw8?>mFg=#()DJyRN1umtY3uaD7
zUTKdE(PEe+o2&JbcWNd512?VRXePd9dTxi}y5@1WzCeY|3CHzYs$)UjC&vwTeg3ng
zqGO4rrgUmWqvC4jrlKjL+@D1M2gd?#kMTcqYTW*nRv^tt#Wcy}?agn(5og?`3`tDd
z!C;j>%TXulzJi^oIveS2)MYcxm{MD4h~rugh}s}&iovh>&HZ{^gQ7ZMa=b;bxnN-(
zi78O172$US%o}C?+bb}LCGoBmIv8v%D$NYi0Rote(kw_@r^AmnyIMC1<(>n;&bdg7
zrBlg&$~g$+(oM_l77}{HsJ>Z9Lx4sWrJydDCG8*Al=-tQ*%{kkrG2(nqeeZo+;09x
z_)^rNxcQTr?Z#Sq7l7$5O<_uU(l=UHsmJ2k67gaPzj2_@zx+pQ@l_~#lIQji@g$fw
z^rhPHigRBv)#A@3vf&}Z03HLsz*_$pZMu(T;TVXJ))S%FO#R~^5(GmkyL~H9zU_r*
zlCBv#3O`M(ia7XB1>297+z&%MN?3Y;K<3k&dJf@`w`{=yQ8E_nAbmOK%hM_+fn+~5
zm@3w9HtUEs@&ZFpOJKvsv;KB(N2riq<=0}S9A{q6;K$dS{ki%8FjzBQ2X>AF2hk8G
z)(8>D^h2O%>oS=QWz_fGv(CMqAjSnn(K{7f6nyff<XZ@k+y!y)|MUN9Nx=$%l-;2N
zDr*9Abbl~oyPN@#84wF5i;Xcem=zEkacSQ)KKX?}ARnG$fbu>VuH0^a1kA1l2$f!Z
ze-rwyEC7f}59OfGQP&!N6Ua`KGI|SlN?NSzw%Q$bQmcwB9V4EOX>SFK>_ehn(P$^A
zu*b}3k~mwC-PXOypIb1CudJXJ#jy|9)Hnv8_JW7$cIQTC%XMsGI8`>|OWgBz{C&P9
zR$YU@=N};FyLvde*J~j|^AOsJ{ls05DH43`7H}y^#!Uc=k#Er(%!a|6gTV?~YosX<
z{C}`mn*d2UUXxp0_d3Me3o&|V!=7^HPA=JpKy)2XwC&jg0vKD4Spy#wa?et{zG>!*
zdMM|?@F^k%Xsc8GX4>j7<_GGnip3ykhXQ-hwLh#iGiAb<ky+cjQ+|oMXV;U+E<ki}
z|8^vt?Sp7FMgIl6LI%Nc6SssA25#sZEF!~#`h{^S<LK2vFnFK!Mj`8TE7&&G?!_VY
zfDnP=7kY?ygmL$Um6&byO#D-{_T^sv&rofC6=dyAdsOWf?lEHJ(+S^;IXMhu?&jLJ
z$zEHs|3U+%HX2F}w2O==l=@g(0CX&nE}I@|4&cV)J7_QNO(4Pt4A?3O2ec+lP&jC-
z-d}pN(e+;D18?5YSALjhwFOVV)6;swEw5ZybD`4*-*2ycqOH|X4S%9Z6yZJ*SMsJv
z1w__5V(@eQ%LdVEM#V4`3n|uyYg&lm)n*QWeGe&a&>vKUGUztB)sRf|04So1>IO9(
zz)4hjRxN8k7-=)Jd~lOYh&3^ndTMYvq>ez+@qS}fc7<Nd8UCjrvoPbNPybK2gr~YZ
z^(WJ|XrWAxl_#rn+K%6ZyG|R^i0DiV_wpQnzi<Yl{ffDn<bY`{au>ZlwXMD<-46Pu
zL+Z<IF7of{Ed_|AjRhOvU*hj+@!@{B;yluZ20q0l>=ZF6Ru=ivTz76$C-e_EPVvl6
z$jYW%Cdm@-=g1(3!j`)v-=g&>8*8NL;hWA>ZDRbo;vM}T(Mg?2go4*OCWx9J^_*(j
zS&WwqJIenSLH@S}avCp*rKxbt9ICc9T@XES(|kwy@dgOzq~b*(FBLs?6~23}>?r-~
z*`n8GQB*AW;*tvlK+WX{y&44iMs93`{UMlOcN2gsIu2a(`^5^>IwWs2g7hJ%ZzaH7
zC)VXrZ-2>IQX%@{QT7sm3CBTz-{fgnI~yNn4ZKFtaoDkUBIxFwWL=Vm;jBd2K23?#
z1QOm;gaIllPF(YaT9j&*m#3&VEPn2{=%s)IaT~ot|JPRfOnvdux7@05>O%Ny#bnjz
zBHN^kWcBAutGxs`>KIVhqJer<>lcM3KYfC7flF9(Ii7{&cSS7)wNuO*?k$RW<kQp?
zYmbYNizLoy<-V0!an6&p%<;~6!r<ncI-RD#0Suv?g&<dukV?BxCcK5O8M3m|J!v5e
z{3Il&IA5e-K59*4dZFiOvT4{W;usRkw&~KSU)AfNB<m;i#$I(G4JRh~R90xI(0}n}
z)V`74KAvKpkX&Y;&|WvDfY_9JoOXl)H>S<6;C-$?4$dD|W=c0jIeIKxR<gcLX7rw<
z0Jj_u2t}x1kv#24T0L`?HPnBNA8RWFIKOfx`k2&YzUpd6SuYM--i&p-q04#P*iF{i
zont<l<-T=7nCaxXNOqL8*dCI?nlB&UHU4RE3<ykqpKij}gb=LkU{;!GJ_~HmJqxb?
zRc(I+z`WO)kI6}E-ePYW`?9&G>*4rm1MpkN>@3REekb_%Q0*IO&<5Hf2SW(max~+t
z46>>+h<F`u+4h$`_=$|_C$)m(f7qjLC;5HI1bWTK^ZC!eB7Ojii5pflmpA!yzRwg9
z)4cOnkK8@XuO&O_zAP<Q?zu=dPnNnpyvm$BXJ1v1xvpOzm}FeEOfHl#Y~^WR0j?u|
z(EFQQ$CtulTTf5dxo_9X+vyj$W!84du0orMHgK~EQ7oZ1D<t)m;V(g<Fk{Sqe!<5t
zrz#fpcf}h58=8*wt+$Qwg9Wf>wAyIUR+P|=k!Pe&TMs=8kA+Of`f95E+nJGPP*+{|
z)!dv^;U`i$#nD?_+Mk1jD<_fqfX@d85H|R)z!PWSELxR*+naW4hbd-L{|~A?y0x8?
zLX&;GEhJs*f!KdJiYL=Mu)1_^VvmHP&(tMm)UA*s<KzL5C-8KI;C*0F39OEPI8Ep;
z&*Wk3$_Wfo{wSvky!##@@H*0UTwc*n54=N%9r`arw0eZTtCDyyWQd&j0QQLfx*0(~
zsgd58hDvV3h~7Wvf``&_zW(p$#cRZ|KuLVjdv{XH!+nQQoz;ab^i9Mu9<E7!!oLuS
zKve`yzKDGX+yN4}ro3yo{5O#ZN}VVw#WnfvgbosYhM}uegwD7@ghpK**oTM`oH3Gb
z2VuxhQMB&D+lC5&MnK#{n4p_TdBBJ&SY|^LGbrN8cpwoX_JnyV6_(OX3q5_~V$@IM
z$Ip=TXE;X}s4^)9gQcUo_VpdBx~KH2GWm6yB1SO|N|++@hb3i$Y9VrL>GC;-9q|8I
zAe9;2L2Zc1c>6C%G|zE0YlAV*v-xQT6@!-!O3*~6j2XZIOp8fGafBh`L?VQv$uo>9
z%pm0wL%|$e%36=7|J1ZE*H?a8IHasvSS!}3zT?!Z-7M2!R7tKuchL*lEJAFK$`5}G
z?Zp-O+9ZDOp2p-4-9$tZye2|K{D>7v`>E}j0zRr4)o7foqx{E%Et3AX5O)>rfRqA*
zk23v!0!Xf4^t%*v6?X+zU-M2k-oOlRpC-15N2nR596cW{?TM3sj3@*WyMK%5GcvFJ
z=gN{TYLB?3Pk4dE^@jUyBiw#K(M^lPWn#=X5caW1ft&u$GkPge!>)@7V8vzjsEH{J
z;)DmNOm>otXYhsoCNEdj^qU~G1;ju;;yJsq0Y2QTpdwsuR8E=~I#%rFwm#|7AhYI|
zZuQ<vx1bF^R9sxI(CI=@j$Ztp2iF@3i|UdJ<0A)%oYs(xgE$aZr~d3BUIg@B$G{VS
zy82(1jnem0ix{*3F+;YuBd#9#@#q<7wAWsbG6M`Y#I8|8g<D4%bz#mfYH#HCwl>`a
z0LV6TVut*tejw*Zl-T_4<4jK&bM>><9;-}RNv=f#PnaesKlUKF_VKBp&mFAuc|Zw%
zrj^MFuUYJegRQ09PkWrEVA!`sKDDO+>h~`Z8L`HTntV0=M)BW_o_W(*#puBs1*mY}
z9zlaA7<77bgmyBR9P+9Ze6U4LfiyB8krJS?S&|z*?w*~x+QvR~ihsK!OYLo#ch
zby2f^Ov@!ox`aj6@ygYq@ynlm`qw%Nj{|7tT@`7x4laCJM>+WVLQ)8-GIc1oeZ0Sg
zYM-IivE}jKIiPY$$s{$utgm0RmsEcU#he#*#EOG?KDaqmb^7QI!<kfRs7%ZOb&A8_
znYh2=Ou(f$r4c6d3P}g;c7Q@a4E=0W9Ih907YN|iirp1Xz8n6rO<R&obbpTrW%8N!
zJbCtcuW_MiKp}?T{BeLznCfcUPWda;TxPIG;$XDnHvRo&CJsVM1SNG^@FOjso*EV=
zUwF8<KC3wcA03mbmopTn<3(Nobn_!{Fri8{je`)9|Kd81|Ld>3OQ*ujX!aW@!9Nvt
zQUEKgu3}5}yn_pf%Z&$%(vMs$7SB);4P4ghVJ1T3Gqr4zOC{W@SaO9d!3rc%>Br_?
zM8g0I%$UK#s9?#Nho>V~A@7xI)j1wFGM@O{C`{D~(r3*Z+CcEvY0ZG&@Gi#r8E=OT
zt~GA5y6@9~i5g6?3_d;!CTg<Ee_NK`BW|nF*`4}7M>K`sSQes(9!o5w(KG$XWB?&l
zzx_x<%(>fmOR{BfX6<qGoQuZS26;7@%mcYHKAhMsKE{kbUXN_{)6D0w%-06n8_=;{
z_bnX6LPq9j6UY-%Y3>0LQJwWMpLaR8adapcqL8Ia)8|t$7<Mxv5S!Nu(v~V~+CYeQ
zL^lp{#}f`J@YlWs48`9S6D~e7?yNx_ZDbJN5bidu->;W|AZM{+6NC1CV${GiuUe<S
z&SI&r(6NJ!_SSuBazt8cysNZgksgJROoge}q{s1$v55eDC}RMAN7|yB6OdzdoEu>F
zWZ+Yw?<8URV1gVXy{SEgp$07FQ>;yfusvjqd?4y?*pIdDFpUkV6vFW3SM={MrF)a1
z_<fFTIpj2JM_}XG{ky>A2Lfv;B|{mBgt!{3N(WtcIkO@M4HX&8hstZ&+-c;2m3yX&
zULKReBmVC605G7zlCmDaVt^Mbh^Yw~fya|^2G==`O1<Cw<~r=zura<|0y;b-O%gHG
z!~OA>YTWuVh_EC1CTbY>wrBSc>n0t5D!;v&fK+IcESnzFf&T9UuwUzrOYVkot{e$(
zcJPAJXidW~BC44A<KW=1N@#}IhGEq9Ahokh^10O3fDDL<iu<L+ky4`wZhaOiz+=*k
zXP3wNd5Q|ZCdTgr->1vzk1ByE(&p#vaG@cs^p*Vj>g@-NY+-kaR_uoPTlt`fM)?xm
z3l|cCkLwktRp4_$v}WQfS37y<2lomS_1vzM8hSZ!#h^<01aa)3?jtHV*x$tmg;{tP
zmlbug0E*RZ;GcLHckY+fb|)>?EjCs`K%590S3`%adsPT}QJ3XFg%L3j)UNJFj7qvR
zQ{I^Z9&PJK56N{4V=RRHWj6Y1K12hUj5`^5B{m%$(qQ#C9#fM3&lVDt``W?4_7%hG
zdyjdz-kdu2>~fac)fhIH#G9&ZLV9_2T>(Ax(3@qwKP8AbQ)Zp2`4BY&Nm^6Z?3rV;
zmogyxfiA9rQ{<&n)^hVJRnl?`Yno5L)Gn@FOxI$%WRb1h%W?&N>wlgji%mBlErSd+
z5rhCm!}!)}NgGvn!+0Q^?j~_1*Wa_XouKVzAGvAk?qwO-ophsJ2w`-eJhiMKuK@hl
z?VKC1RR<nZ@^k*kT*c{@Wy;ZOaIv+asj7$?72?EgGif2CZ?@DN!j$do<qYr8+F=?w
zd3j||9CXY838VF*7WOBgx~Otp(>}%(aSsQ>8(-BBX}3c7u_p^jC__?RSn+cR7fL@V
z#3El#Gg5BfPsVII&7bjnGn7ra1pJ=?y?8SOMx_Mz2ToJR1RW-1yysKx#xNLQg3SEI
z0{PQ}9W<mw#dW(Di2*co3NB-Eiv-HsU5Qh_ciktRO)gu*Mp0Jj$5c>Ngl<-Ceknz@
z$6br#@8+5dx8D7YMwq<4_7=6o*sxes;0Ul*+V?o^WIlNlZA7O8=nPbSsY%N=Y3kY<
zL%(czB@lhlSSjZP{uzFji(6Zqja{>y!2E~G-g96#bM{*8_eFT!Lah4Rw0-96K_+!4
zS{K2|*e!H5f;2DujWp!cIo<++VO8q_?(FTZsM$Xr`)P;;XwYs^!%R-OdOlM<&~E9N
zwsE!Hby*p82MfyrumZl(^sF8+cwsj(B82q5p5WQA#(7s8;wK7@M%!j_m6MqzDR|+_
zFNW1hv}z5l8k(k@Rzs2`2f`A?p^ktT8usokv>}PRcPE#DN+Y0&)b5zhEJYS#ivlgH
z2`cV=M(Ix-me!t#y1^Rb^Y0?R2}yKkZh#W$ui2JoRfQ-4p!(7m?*!ar9OAvCx3e5!
zH@}TpyBnH+Qa7F6Vt;^WQ%7v>XVv)NM$I$1oboDr)(PVgmQ{%;*8#<aBpG>)2o%n0
z{MIY}`}6ZkTQf5YfygHX0)O0ME5ha+*r-jVY8*tb+BQP#vt4VV@tTurI|buc(#M0*
zYy!T0ns^IfILdDCR2_3RxuP3Jr1t+=3=6cGfYer(je*J`T}OaWtx`@wcWL)I4NEJl
zureLfYl?Zr(Nq=<)1`C^YP^>(rWTTKY>{7&65vu<t~F+^e_DwTNE%t+q+zNGrW{L7
z#F8H7_Ys@zBn`Tw<tGH;xaweMB{o(`H~?1A_o@ZNx2LmC4=fH*S5-ND-)Gn6P2alg
zWZkEkKOT{se>*NSljkfX^UX?S5p_UgdLz~THfqthl{`J){kw}QrLZr`nIa7uW^5%F
z(pzt}&ZhqAq9QG^@v?BED5O-1272-E8Cm0(O)XE1^3N5zS@BO8wt;Svwg-ayb_wA5
z)lu>(+3q}HX1!~h$>oaYyXuH5Y@O5@)-5vY=D0UO%qGwJhJg1f9r*8+c-Qt*$fAg+
zZiilMfJxQ1U>_JsYyCi0D^M|JD+)u(-bb=xq+b-l<DFXt_(%i--Kf8rY$thi0@NkG
z=&dAlslCDUdVm_HCjEr_*GpmAgajaq&chQ!l*-`+#Nr6RmaJs9noB7|I<6kSWyH-<
zS<g*A3u6MZvOS&?XE=s=p2N)N=54Or)Kb0@f$Th;i6L4MgP4j@nJ^74$D-!opjoip
zIs`ikMYag79`QVsMn1`d6zqnXN1ES%h(j2@$5oOtHFa0bjH$!5fN{+#rU1aADil!V
zA9K9oEY2Rb9%0O-TWK<8B|l^$e`zuXss=4~x*zy3orrc9Dq~1lB-u0#*f?V9&L0WG
zFEGJBpz)0I<$MdH^^7`Y0=dK7X!2|rlcAVPb)Ym~IR-l1?BX3pWlEVb4vKueNFI5K
zz&8fR^}{F8Ynk>*YYwFxasgOdQJ($VJ`&mSMr^_*j8X9@k_nMnGJ@9dpQG=@N+$nX
z6zllsbwDzOAen!k`dCap%0UWsPxv9Krr%A)kJjlXOg$YFrqVCh*d{A$n-U!mF;Cax
zjDf%)UJV`GGH0EWRTjVJ(i3xQWZ<|GP%tbX;{kf={<AwR$Ur+25(B^#>XTV{DOi$N
zEdvsTVX(nz{~@smN|HnU4zW};$oOLUj0s?Wx`m9cRyP5z>-=+^Dh@lG#x$L&pNU0h
z^M~%T^w04mlQ>u+cxJvAc3#5(JJUU`!c#qwLkzUOU9diHCuRmcs@^1RIE=mfInm7$
z1-#Crxf9n<q$Fbe{IO?10vx55W(0aM#xoMFprjEWYl1*!=<~q5gbtH@^dGCrQ1JsB
zjuIojMt4k#1!*Fc!jFQ1<#-$uLrApKZL|~gUqVV<h`d6g^YoZlV(v)Ws;bQn>;bGv
zegED*lomzsy<CN(*U7|GzfN`6hx*u`1R=Ln_C`S{Li9gpwfO*p!xg)^LyWZPdv@ec
zTeyCikp<j^Dzc7Vbhpq^NEuJ-cej#a@nIltL2<e#m2=$D@Gb8yK{%+J$ibH}Wmuoi
zylLp!2mPyfZ*l1PQbulu>uC7lmLu0T=l+Q4$;a_vEBD}cM(ezog9C(mUqZaBlVtY;
z#Nc=J!?<WDZ%hEpsRZ~!N<%(gc~Wk9*MVpy9gp2)(#n#lY<yNjHrkMfXCyYiFe9q$
z$0GMUYzcJUI`kQYt|bxJ1OGsD>$_P*H<q%SA6L_<97>R9lDck@B<M!jD2u9+zAZ#N
zEL@GYxvJd0bVq>tKZw!=ZJiu%l-W)E@Ltb*92?-mtdIX6q{NUz7e6(r8@Ugg47OA;
z$e0d;MCh*)t?kUqN)0VY&BKE|3&WZhnvm0(ptFLcARnga(&}<kM+O7MJJ3b9@qZR(
z-^)PjEA2R3*6rBivu#0T<L<m(LDFj#3Wi-t6Z_Z|^6Cc?cCP!m>;VTcPC?l${DKPv
z92W8Y=U!V?@+QDFoqgG<;xOP|jH?OywGebx{&>Y<efauO7f;8Q6*<tu_n=;va=<}%
z>-@K|5w!}6<44K&Teq&<S6VeVSbG01AQ?i$Tq`+@ejg^sR=`Kg89SKBv@&ukyE~A|
z0zft%VFO71$Ct}hP5f&I@gn!J0F0Q#eTY&qSrN1C=*#Ufn`xBBQA)r?ASJ7M;GG-N
zN9*5`K0%6WkrkJ20~`DmE){z2BGuLHdJJftYx?|=5cZ?0`sgAg_M_vzlS+W%-PH)L
zGj{1HT5HV#&6YJ_T47Z!YJY0RFIYoDpSFV;!*t~A+M;c@iF|zgy!fUF04-*6>xw$B
zHkmIFcEH~upvF{S3)aFRwnCDBi{seFs)Ay_!Hv>^#Z(fY(E?b311Ht623PKAQC?_i
z@b#i<Fre)#ww0~m5IVaS%zBP9@SJ`CU8l}d^;eTslZvgn#zqRr^|PIjzQ92crtR)@
zqH9-~nbgv`>{etuWbkAIpe&EzM&1qvlw_H`q?e1|kq+cpDu2#US`Rf9q$}Xi&}zpP
zN%Uq&*#X`tIFyG!f1l~C4N>WRu{0WaI@YozLQ{c$)+3o4C6NZ$&rSox_4!8ub#dbB
zKA0#ZTaDNdYTT?7-}R~<wnk2UK|ve0)AkFJr8W!{--C;iBR-}Rumw9;wbCBZ5*&V#
z7f7_3lStiQkN|Iz#`9|otD5BqA47+OLEo+EE2!lg6PVt~<xlD{X!99XkZm7wyLQ$n
zh@QtpgfTYV>Pr4^$q6(y(MT<wNN%erM(h8b#nSBH7d*Pbr!&PqnX28NDYR%=;_q49
z%O!1HUjF1>2CSpwAqDQ#5lc(T_N+uh{Tm!WE6wGjtxsm^td1*-l0tGp|2<e+F~ko7
z4mn2p=Y(<VKgZQgJN#pw;Yk>Dz0C6U$djmmY6aG(6&0~Y3o_4ig>~uV4}A6kGH+oM
zX>B6`CuttTL`^u_45Ijq)*cvi$WNM$^L~nYdP^P+aPq=WW}Bu%|KH7(yiopvP3;KC
zJ>IzoA8br3YsC*ExUj0Rgc!<3d1Vl;#dr^17MMOGt!!|-yLR_h*vm7$3*RB-JM6Is
z_LNC~59uHRt<xk2cPH&}cN5z1ZQ*iavv-wlcAv=pY#Y(In-m)%5K;a$9uFDI8axfB
z0ifYnu`rXStBQ$mDf0G;G>5)uT8w`aWYO|KV3#qf;Z%f~2L5NWH(Hhg?UEG9N<W@X
zE~uUG`~YOaMWHTKj=S9j&l#Nn6#y;Z)7C6<Cfw)J9e$5OE$vFYTYeMvrSAnb?(n1;
z@*QY<wHoisJIuJ><2EReO0bm!x%)0P4v0nSd|TI)QF_ZJp>a1!e}7W;d`gRv-<V+Y
zltFinH0RFC5x6kn<?uZve@G}EcAVDbKgatbEv}|ZIevUj^AQ#J&94|c+X?VpEX!Yw
zV19R$dum)iuSCH0GM-MqdP`$iY1L;4=fB;;HJC8vgy8_t3A@?0cHosI(mUP#1@wNP
zsC1?x^NV8zQW(F7p)^a4G8mb;R3Rm?aEL(}Q3LZ<Ngv_nr!mChA)5+{wyd4I5)WMY
z;_ZQhwWOMzbnydFpKM}HTs%c>7%cmZnK&vE?#QsShDw2W%V(L-GTF3Bw$}#?TDNLQ
zHvM+9EhCQ<Imp|~{_nI2@RK)*HfzRU$b6o;GbizwB)6gFizQTuH5_ZP+Cnd*x}WAn
z9|*Z({ICItc8U&>mSYwTB@l8}Fyu~HA~4`s3tcZ+y6i)Xu<k*5L;F-HTVVfZf8s}<
zFzMzs8>l4Cs!V$TNv=E^hN0Qb3SB(MvI=JYBOD~M+qk6+SUh#;(3fZx;{?@_t$q8o
zJdS~|UQK@+&{5s(xLlgen$hKaDBJjJd5i1pe-mNf|4D@3IL$t~T=4Cge4MAtRx(Pe
ze7V@1m8a|e$>L6j|JA%{wC=JyJlOc;#^1@EMq(<MmQC22o(i|K<G}>6^=*XpT%HfP
zGdGz75Hc5riG~XKL*RyxteZF4v2kzy>>59B3ggDIm6v(gBbd=vXgEl~2Z7oG^k+k&
z{<hoG_qZuVqHeL-eaUa+|KjQ!gENV`tz+A^ZQHi(Ol;$cCbn(cwvCC6iJeT$FYmo|
ztG;`yySlonPxp_mK7ICHYwf)<_-eU3!77L@y0}dSug-o^I{)KY)zBUb`>fNb%CT<h
z+-MMsk7_&wdQYDFidD`{CQubpIW)J3WQXJuO}_wyl~YZQqax2j`sj5dUjDpfcPAr6
z9>UZ*#f4c&L{%*drOa%SnSz97&%qs*feGC_-sq;*S^Os^;<^^asz=n_TZM=g6%~Uf
zPE`<tA%vZ(90qAfhn#C)1O4vzZHc}Me9fP9Z!cS-&d<FxRRtN^o}lN5wLf-z3far8
z;Wt2(ZLM#<vMpH$w0p-j{+Q5J5GKbMb@Yj(U~)qmFJZo`LC>}(Au};JvqE7mVRD(j
z%F+5?mGNAvj==zdLZyB@Ie2~SV2fj+x$^MEuPuaqJC3exROxYZg*}9`6fsiC3~a{J
z^~#L5jV1SlCm)#5=J%vV+&{ft6y3Jn@F{@s9qFnCt0={z2ZrS7&Od5%Av|4kF&%ff
zuq0`~Jy)tSo{6IyDt)_7Z#u?s+*16<e`n)5^r_jh6-*DpZyn%4?+%MAo}4=(wTo44
zR6D%Qa^epXdYhCqJ+Y+ud>XdPSG#uL;`u4PALq3bj=g)5iQnbIew86LwiP}^ooxZu
zLe4O|u!Yq(mP6ZO{Q5Lh6=pxM20nQ#!QK;wqk&LdV4fZrq};lULfSrn&^h`@H_TA7
zEX6GpR}BVdKu63KJ=o#E*;SyU>lmzabJ;GIUHe0oK$0en%^YCl4q@wS{wxUYZ=hY~
zzN`K#b#ykdM3w)va=$I6PwTXbe=Y{x103uzzE6&iS0|BqF}rEVG$a4;9q%OG1q*yB
zq#|zop~*~_8#l31NfG-l%8+$_m+_}J&<|=arVQ_lp1P3zytDG>D}^&szvZ`zDlwQ*
zbt@g5LU-fp#z~9<E@1I~wvA3(`mQIXO7R%9BBBVNyd6qYJU0>(@arH*xJUr7;3d>y
z7R4Qw-e16%M4(I#@OW~RetWNUI1LSnvtC&hu(pWPgoQzuaIS$eraXX{M2t9pxf;n@
z9*wp$@A!Ph@fZDsguf<Vi0kBU9@a$!wX{bC*%$M-RJEDWKRL_ksC#xsP7kadO?}lv
zgQEt*W*8d{xd`U+;48VV)4l;j>~B-6=uhyMHaqLzFy9-Zqp2IV?s8s@b$SL>(itdm
zXb+mjkD6aY!;$%`AR#!=-KoLv?`7-0UiUW{UtF9B8HoR`b8H#l2*^pOB_9;bZWZ2=
zG1NPHmBNa@&F9TANX{xqQ@&#^FysFH>A{Icllhb%(U88`*;a%4Ow9totjd2#zIg$p
z49PV8-mbRe4u`_#(lm>b9{Um+rh;jJ)9DJPf+hHH?zC&rV?2qYJ?&e+Yxp9<^uQ?P
zb-+XqKkTjLFcIHEj@;+fP|npvZlq_o6uwOyojwy>zrx-`!!pX;l{^ET{5n4I_VqLQ
zqmF%lqpmr3*d#GFnjixH{E68WaJ|>$qh{d3KGbUIQ5^)Y`n}WJ4efAra=8JVTF@0?
zR}rJ5iC~`Hi?opi8jEVNB!sJCzasB<uidrETUzUP(l=tS!)FcQH`=A6T|IEULC1bR
zuPW{@*<l&cO#&51rs9HUV)1?P+$t&f;_MZXJ{9d1iK+EROfM6lhoVJ&0_-e1DLKSQ
zW5U3C%-3jI^l4j<+XNX^(&3pYp9aZz$j^Hb2lbr*{(ZSgbt%}(Zg8pS;LEfsAE2~P
zj(8Gf@_aq>8g-&Od-Z`GrD4xt<W%x<RaU{T9t$;<@F%Z>SWZxH6^+|K1_w{FHX+XD
zdgjcGa`iSX-X<PkqoF6dZk}azX+1K7OFr!pp65^`uq?-UorVZmjp=$dui%o<XA`IR
zO{DwkbZT3JT?eVL9WN;Qo*kFkacX4tzO>bdy&sQlMDXP!`9(;ehg;sRP6&s_o7ft7
z&s}aLjTsK~o6lTaC~o1Zu@J^67p{t^leyh7VQx~Wr<($xZ_d<?lMpM@t5?PEyg->(
z8%l%irv`Coou8P?q9O$-_f=VcjuEKz%@?xM|3jL|)5$DgSUK4u6MDY7KN7B7dA~LI
z_{)>|MJQP~&Y$grWqf^QV53lNZ)Ag{xh3jQ(4Q0pKB#B!VH-^dOBFyPPY4$B>8ayD
z(q%Tpkj)FQzTlch{dbIESTR|K+ru?~=CA-dRU9^=0x(v_{QAnHc3L$Ii-0Xl=n*qw
z{N@^PKMRS1)&TS)A?1Gmt3~WoV|B20GMaeWE&s5Ni|V`6Hw-|~h;P$qI)BA)FbOIS
zJr{mkJV_^{PbDe0MSx#P>~}FDVjZ7`+ic<3TKs8P&eKVo5!l;XoF_v@(366ajj8QT
zYdrmXC-U&}_l^v66anGOEA8>e$-!-S9Ne@24>XGrzxelbBA?RKhvo9r(TC-ON!TJj
z1A9J+x`<NM+q0xrvxK2iwR(5c%dt!04t~L+X&H7)?Ze#q!vvSHn*zMP<^CkErSI#$
z7Lf-)pj$Dax0-Da+N$SCx^Cy+^-HWiM*z^KN`Qzs;X2^(FCh=H(-`~LQ`ZA1R+`s>
z)RaP?lqNTICWN6!g@t^ngVy@SXoLxvx`uk3yH#10rtZm9xG#Z$8^f~V<?6RmC??FO
z@#0%1o@*ULpD#{);JtK@#f_E;!1Tuy!9W$@b79zBYFN|z7u-RkjWbMO?fhx9%}3Sl
z{4I>nFc?F~9V0-HVO@Q`(A%TjLnu1HyafxP-?5*go4q=d0Ydt6mo!bz{f)h4I}Osn
znqvlA>%*|<P(5pT`KctJt4DW!b&hMFp1-Llw~yJ-`6e&-ME0qvHD#ix?&HT@EPEMH
z>DS4ZFQ=ND3y3Dy0-a*(d~d;h^lm#+CLd_h+pbGXq`jO0o<L{)%w>l7cQUc<%8}(g
zYMPt~uEz_T2afWCdGYn=c7<Rt-i~kX^8h!$#S@Yx&>*v(Ae`|uiXO$*YcwEOG?30@
ziz;~~OM6vMWs9mo(I5@bNB3}<>B<8>=EtTwyp)WuC6C6R1T-dOF<lO2Eg)u=cTq-P
zw-zF29?L<%l>@sHt~_u*8CJOvBKZ*U9lHdP!LMzD8D}h-45!e((SjKja@t_JkY+sn
zva>oh2cU(N1Dtt{tm#EeApe5rC+ou~lSN}E=fancF5N$m<1aIFh{}~lm1O{)BD<9+
zK)>y+@#e1Mu%Esl{AGEHz9^Ty4mBv}VEsfc&jzEv)BdGEA<>{+9=3s*2BIjdM7;c~
z^Tae~;-EmpT{94>H-U@r6*G=ixdV<8m*sf-^X=(0tWpzTCG~6->Y{0vO)XFL(U*0e
zqD4o*Hv(}W+SbdLLNuyh>8k~BH8!w|6YBJzBPDF<T+`uqX4O#xa^Clsr0+U|&cG>t
znzdU$nt~&W+?|?*GoZbm6D*?ye5UZy3bls?hX&eVwdZ3-X3<|i)*2!~f!ID4+Onz5
z0bz9l;ljHypPMX=7<Y1QfLW`JD2u7!`Nt{={LXb=CvIE90k6}!DIEepVo{@}Au`b$
zB)_J3cEDLyxx>MR#S?0LFtJk+&NCkVu7rL1o4vIQ@cCo$=6qJ0Oi#`^tkF5*EP8On
zl6^(NQ4M|6px$ol1-^i@VvJ(<C|77EL!V<7>AE^LS&VVn5J#6$c!dQN&yVbcIrq4h
z2T<_fC<yNQUZ61FxIX}p>Rl^}K~IvQ(Fj#zw!oI7jxhB=hp~NQnP|Z-cw82256=q7
zwVM$%2GOsK#D0tILT_eUH*KFT6IrF97oV3*>V_fo`kt4ID$tTfwm`JNRj9itE80j!
z%nPEy^Fklqe&mGpNNHlj)t&-sAxMvc9%CL+prHdE7#S*iN}T~ZV?_)OR8z|o=sCDv
z*|<=^j)6WEFd$8R6OH&cFf7pdx!>-KWrIEUw#MtNbil$w{5cN3HHE(=J~;D!znm(c
zLS^CIDT8TH#yp=6{eE>U<hK2r(W+Gs4C#(R1;%@N@?!So%exrYGSVPF#ms(wEcggb
z?od?;Ens$H%X0>3*v49+2$WRUa6W~p4}sSESG{tsMw~#A`I7y8HIc1)>^&^OJ7J!D
zh0zJ21xsx3^gcsl-SAp01m3FHA!)>#QgJ8BnYJ>_6VsT-?tiaAXfxN{a#|^K7~Z1I
z9A1N<iHwSt#MVp_pCV2LSO0H7Jn!Ia;?)O$s*ES*K1&O@qM;X`n7oLrNpx35E3YUQ
zpH@$)iN6eDukwTsAQ-HW;&$vJ@2?sZzlite&=MsSu2L;tC@!@7Hstp39n4}>Y7QIf
z$BecDjg`O%-C43D<g;efom66rwuNWR{E<8JCV5Ftmrr+9;8ms{tYf+!H9K<J`)Lh$
zxCGFzJ|zR%Y{^npb-l8QUWA)Vp@3e9Gcx9yD!w=%<7;9p1tQ#;zRv;%{IK)8Behfl
zhbkF}AFcf}c)EXDh{xaq-_J@nZPv&DcNlrC`6VW6#03#;2GM^O8>(m71;`>o&0PA7
z?3Av?<o#eP9ou>1<68@l88<&frZ|yc4|&t;2Q{E9aSg7VXE&h)8NS5jdP|Z2Uz?Z+
znUaIDlB)oB(s}I|<&0iq#UnCEW?7F;-SXWpy2obsm82p!wNM<|Eby+>9E?p}x`Dc1
z9WbSp`}cm8ODX<}uJo7q!s2tTb{%YGZF90YzuZS&6gBA0KixW^=%ear<sYkQ<NvT8
zk2(P0UR8|bg^$uzLNTA$;G<W$lbg7fZi~XFZ#WLnYh)8n&>GhssuzE)`tgfDUwr&>
zwpH&40`2vofIyF?^@O=jf9!3FrT%hhf16Am<%^^7=E-<l+|o-76#0JQDl-gh3~EI2
z)L_C`pd9;IL>ruNIe6l_5wi+3O(LA*XMc>9uTq^U;>lh8<vLyb-{BNBmM{`pMnK%a
z&m*@x2RUDLY*e~sy@{npfs`}N9oGzh(?S^i#$dH;z&jrm<9wX3yfU~+?|jPn*3hy+
zd=J017GhF^-dZ$W?hT$=zkb3`<*MEA2*H=uZ`r;GJ}2p4?LC_y#!~t^K;GdgS#SX4
zWZ#>@v%yMOxmYKG;K%(>%wrBTy1pg>;&{2kr(#VX*lwMM33(SZ+j)a5h;t+7&7q-U
z)9&J}A!wHy5H#Mhwfi}JWl*(nljD_M<K_M50deOC^hklXcYq=CEc5U(c#P`evA?h5
zp#pEw3FsYB=3A*2m&$zy*#q1@axxWiup{uoZTyfI>mkFFseBBNm<>Km?=rf#kMMl;
zKn+AYv((zceKjg)PjfTScQtABq?l048jLBV%l7!AdgZ^-{hEW4!w%Cb?xN~vEHyv;
zre8L<!(`No5fHx(10BiK*)YvmC`WxKtew4N(MGLyN9B4SgNNKChX4&0^=avC4zrer
zDgKU}oM$eotEp<p@sl}E`rl7F+!Q${<QV|N?_-@Ls|U9y{3*y7%`w5*nC&;YR&0Ig
z!;ykPAQ|Vx4E?eioxbg$@-z<Aj=>c|_sKz08Y--AekVGR8KH4m9o;`YogepBLJ*Ub
zkC`Kb9^6GEK=^{W>VQVP60S9)4V1-HFVJACsRL#z)eRM~e*sSD{fp%4s17#c=91)^
zpefLcr$yd1iFKMoJM3{ZF8cm^Pm^(L@(meJ%|;VjF#F1<xI=I9=H-dRpHsJpEs}3b
zW~FhTrE7^_X~JnET|GYV4v0>jrQHhF6=ilN<V@3`>>?^B&4AVF$_3WLALW3hX1%Jv
z&`x|y;ygvtkd)hpwrQ+7a)t?kg<5WN<>Ll+F?d)^iBt7j&&}RARJp`LZmvZg<hQ)y
z>>AmiY0WQe)^t7h;VmGTp`HrCnb>jIPg?b`gMa$BMiK+5w9e?p$0y_=?z@S(Y061q
z0jb>+NcNJv_5ipAu<Ei{HG({2VPXg8=}2Q06>Y~bnoD`UGjFzHge&AvA?yW1T!iI|
zIn7`KY<;^zkrm!mJ{~liIxrUIH_fy(O+)GoUK+tY;G4m>qWiGP`it`7cV%o}C7*^l
zi}n(U8$4izhFF`K?vp=_P}v@#6g*_z^J0>ye!EAVV*p#Rb?5fX)f^!jjId(1kSuEJ
z7WJzDI<uxGH{qzTN|B5PPv(4pze>9LB%PdOVZsTm_Vt2SX&5>EnT_)3#E2{LU&}`@
zcVb{>$&N<P&qZ~23M;*#uTY}WVLl;?>~n=HZj5O+@wSCat$HAB22;zR4p?J+QmOoE
zUkz6(Fo2f*8a8M|D3lctB6V!i!l!g{l5!AmUcbiZyvv3x-^PvSEv=I#jbxyw<{kb4
z0?2cGZ|GPwC|^Y$=%z(0*57hxEN^W}QRlyn9>2OsJ?n29-_%7;GPNRz&P{p$@M}O&
zo__SV12nQjoB%KL`ChQ59f5bw4Z(kbF!2bW>i}&E3O*zfga3>#ilYb(^uCia1PsG8
zgAora{~12SQyA8!AO^zk3k4%PDxR4*`YpirdPD0ci>m4d86*3%a{HL5v~=dkhGkXC
zX2J^1t!~i!umdG*yE}ZkDBxOp=J+)F@10e-NlYn_Us+)C4((Z_?JXs|yG_)^18E}%
zGXc)P>x~A+V2w-$bZ^w>&OgU*I}OK5V~kR#9Q|coJO8b<FL>C1?*|l*yJWB1Zu|Is
zNkzXZO1^gi_!Z?R5j|Xzx>07j0TwQvrmSEF2eV_4a*Q2;qFK&6Za#^svADU;vN)xD
zV4?@U&*OB%q>6kX?_c|^mK|5Ih00qmg8)U4gUf(1Yljm>I*D-ztQbU^=xGPVe-#7A
zmpXzor4U(`W`JUPLiY#otBRV+LKYPx7xFSn5aNIpx!<SKp*5J+tACF{j95Q{WwQad
zd4qdgXR3fo(1=+48ED+c&<z_=9j4v05oew+h82k@4)Dj1d1jnmjY21%oF<8jB>*lV
ztb$d$;h2zT!R<kN&m}pP$Zy{jk`T;EmLbfsD`*eDrgbY<A4*RSV9jjh#O5UfnDMbr
zfdo8hq5LBoC8Ok-*!z?Ald}-u3TDd@AT6Z|k*`{QfzRM-1h@w7)AWQNf-rMjhMJK{
zCFxkpGUziL5TR$mO0*WFBYmBF3V={o`JiaFn_dmLtc;T8U0w{8V$}}OuuY(-B_`t*
zIdDJ^;TmQ-OIMcXSc-}GUp)ODw(THOx}d~Q5cocQ4`v_oYUmu1vbQIm<(<lJb_Xv=
z{eatsdJ6{<ldx;nI*>Lc>pak<X9TPts3WAqssJ;wo-hlj4-A|&Q<gY^88h{l!2Qd+
zFb%ZNmfDnr*wIeG*>&-?XXm4}ou`CzUg~V>s}{DORJ;H|i$_^De%p_uK4~xrG%*&-
zpm7Rp*?dz4%aE%*j!Xl@BnbpR<;;#W)(QUW<2f&1ViUU-8DCnu5_3@DTc=0aPUjD7
z23Zvl=&do}8=^{7NzH={nnW!uyIHsLK==LZOU!u+k*mr027;@LLHq_{pry839`CS<
zdhm13{JX^Jf=K(6qqz@sPxXB)H~<C6OO()mFSN9?d(3JN3ACO!^Hzb62!%w<g<s&P
zcC(Azo$U_9w|hC|7{vFOgzPrmaPQ>++;jg=92OV=UAP`aFTi6xP~g_Y9h)w3&77|?
z-{S1}a?w!WUEyydvv}kNZtjgNf7J5pw~6>2DLGYBC|*-)hKK@v11QBe#=+rCrP8pq
zFLnf|?T2r_I{*dTHgKCjs2zfx$<g?1PZ46;s%pg{nC9>9w!LxM7qHm$4nCGG4=Wk<
zk=_Cz3anhbb0BOSEkI6+XIUEFEn(T9^s2$7{q+gu3a6WjWrQ(<x#e0NP;F6+A${0s
z-L(pd{=R3E23<zRs*n4*GGPVE2PvV9$3Z=RcS=qA`yp(y?3Onu2<UJ1NIXzB)V4*4
zhFOrKx+5z_!3X3ylPw&uLlfd-U)TAGa&azzP+-{dhoABYov#z>GjT8@c|)h-3YcB5
z!BN;Hz~C$FxGwK^AwaIZI)U~65XvL<>~d@1i^gbKRa3agcP+H(=qro2skdqt9JqBc
zkV?I<=cDKk%J}1<3gkr7Cyq!w0WEwt+`t7kyrEzXi67}j&pW{YsVku2U*}xrx7r8j
zX{XmPGZV(*CYIDnz~OapxEpv$H8K9Zph;0Mg8jgSOJ#h#C*7i6<@Gw~4;BAF+NhP8
z+=nq)SQnG#!n$M+DJvPEBn!6WGYn3~&~;<pzwCo#>&$l?$#wk2r<27~MXG3OuwP`a
zFgEKl_iEznPR(lxWCl`(IOpWS_mTo=Px@VsTs**ETsO=7Hz44cM`6$AN~ec0R<aGo
z3V)55I%+JAChWIjrgVy2ZZlkm0%V(sPOJt;U<J=UOKJN_#+6rvHQf-EExXMl57rS*
z;Gj@<$mB(%ZFGLwP6sJ?jh>;Q8)I_925AxPzJ(-2N^6IrNa}Bcs$T*XrbWB}&&utr
zLQJ&P0BIK7dhli|4s@_W7~IAEzPl}Z!`CvIPR%q!fxF~-a_3kQ?S*0o*?AcyQpR)=
zxFI0egMXu-um{4huo%Oj`BIr@#o|c`YusEUpp^yQhoiG9TWV6@)>3bnpht-hp_7^#
z{h?m9hh5X|QwH!tANU0}@28J|I;SnL&efgdM^MHzJx!&Jk|>}xa9MO#)s3}8X?xEC
zwZcLu%g{UbuAE0b3fnf&Q?0lt+SW~AERG>}-h+bBRTF0PY(|bXd&teJfPAPX*mV7I
zT&!sZv^e0OQRjl~n^(SX&OKJG!TRp=IYN$4V34?1;g+$Z*3Eyn!K47)VLcWH4tSg7
zKIrqWY;6#1Mqa)W5rqFIxLz409IE9`ACoB-_*+(pEFGxeRf`U4uK{u}*k$<_m(UpP
z)JHgl@c5sqx)moG56R`pa=Lb#=1(25u|dDs-d~k0reOr`T!GRt+AJi4*Mg>F*5t;q
z*ANeYKAzyOQVxUjOYKDAK?N$sx%09aRHfz{jgKCUqFvs9*%t%B%0D>JIa~kw;GKqM
z=K5ep6w)llER10`J>HlP#D4gjjBY(?BqRnA#&G5-+WttQM+m2;CqAg9q0^qNHW4BY
z*E=Zfo@o#~lxB6n7|hEHSxL4RAO(n>TUG;LCPZP~BlB5IsaD0*cl$xB(eEQQ0`n4N
zVyOj;2#Z)907*~xa~442MYG2l3WSi2jwAm-TaQ2hPAS0mzBKSJ{CGC7iE9vj>ikL_
zNCZW)y_SZ-H@$I~n%vWz%7=Yc2G#UTB4}V}Ljm=_=il<qT`h8Lojop73@RD`n;Wu<
z<@dTfuS9$A0hA^lx(<}m&(kE^OMBdH?Q(p-hXsyTlC%wrVjcZPkZbR_L5WcJpjK55
zVoCL!LTKx8%(x}mAt?Gnri2f2En81YDs_@g3-J{A!C~PQ_(v+r97AF%__Mvw^FtSc
zR2BazhZtiSsk`rZ8mV*qG9Ca_+5(~NFD(}4fX*!~YG!-ubb<l>+?=nAXM4~^EtERp
zID*0CJQSs>STe?Cc251|3+!4wHiGWc3j~*CJcoIwG}M)m&Od{Rd-)o~Zz!Y`vJO7Y
za0m})ff+dl^fT&_aEss2@w_5{=wL~;f6u55^KN=UYulhZ3n=2b9>6dA5o6kARtJKV
z{9_{BBxp}gs|(V$=#b6P1y#3O3KX=;P3y`UzewO{)n)^Y>-gH)s$uvST(`hMZV-u9
z&DlB68`@vg{enFXX0v2<D=*)G7H#KjJMgA^f~o|jw;giMAv%tC)?pgv8D+)+t|P5)
zRY@{|Baq<NRDr#mD?l~Rcjfvqw!&uj)xqwb^S8HvX9!f442q|KRtZftqR*eoWwm9}
z^4u^4&+WW$kQ+~dEY|v$K8Y6j|42p+VcIu+K>R+VQFjtCyOPz5u-5Gvn2%*?RZFq!
z;#?N5j>Hnamq2yP(Hd1@mPCmh#s?yi{WEa-kd8MQvgM}{04{ZQ7%xJQU<@l9QQDgy
zjR=&zVsMlyRmxZ9-si>0+`b8A6fRZmV%l!CMroqh4O)%R*b;Q>!-+H9fdh#%2h5JN
zjY?=dG5+ZOYOA?ZuEy}DD6r9v=-_tWKM6xENwxPQR3%O0C;!5e;HVZeHB7A&2Mz>e
zqX<gW4U@zfV1=s@eV4dfq~9hiU-K0bPWUE<i^!1~jNSr@$4Q))Y{|q;6B9}pPFNB~
zcu1R1B8zz=o0AF4j)4F)uxKW?|9%pR3(78>OzHtOI8W&^or#3$CBoiH<ENDXYLTgZ
zOz?nW2xer>MR)8aKQVB_@o}l^JTroN`DNDjHTR4=K!NX}@VS(Qb=rC23qfeo&%uCW
zpcwRM16QK4J3Yn&&$<O#%^ZYNh*o)`&XshCrhpl0UraElJ8VI>j(zr={kFa%(-6J3
z=lLHm|CIReEcmIAXBg&OLN^zlQbo8JZ~IY|ESv|g^7048C7NQP3vL^5;+_5WpRRR6
zu4FwNfIJgO1g{Ss&(-5H4_yW1{N-1U-_WaIx2;%WsJNGW2{#pV=BQa{Z_E8x|4{J4
zhe2VHU`?{Q#XXs4qn#{TRd%_W^iS~`3Vy8#Q|Q%CFed~5omK+PM&?53+!Bw>3x@r7
z#XKhmr^i%GKwgXx%Utw(36b1rK0cCdKn#X-6@b+-D<h#d^@&fqiu@FFmDmHbD-R6z
zqaI{-jv)z2xZt{{vO4VCP_yavtC96JTF(Lr2PYM)#ig9NUQc~ShN^pi#JbH~e20Ad
z>)?AUE3BnxKAMR`S*0=Za9OuywfpG6h<@*Jb$3R1fv|F+Xh55r#GwiE7Zmx@rjNYs
z4WQ*?aBB9NpTz8PG@3b}<7V@>;j5W!irX)EYc>rrOW#<(M-x}Lnf!Yh+(j@|HEwR7
zmMzx;rz=Ea8UdF&)E@?|C>c~sI@3L*Ld%%yIjvXz8L@8#d%G%EM)1-AzLv38@;v`t
z@bGXH%TKLFv2)VMk)1@Qt;(OE!TD4;4$z&rfNrzqXc~1c|68yX;9Da22N3Ny#12cy
zQ(IC;Iy7R{B8SB=DYu$#*`CfxtBn8ZDA~balRBZ%vIhQ{SlxhuV6eNL*!@K%KL&4-
zB)D?qP~zJ2BpXUA|H@Q27UkbJ^Z}B-(yVVboX-RY@}_m%dG9KUBJApTw8>}X4%i_0
z>A26~#aI-}5Fy2DI014(g<Ak+76^HAvh~qR^oF>duR*)Nq}<VV-1!^S(v%F>agRz0
zovVNivV=A}w$=O3m4&*!NuJokSs;LbJSktjKKtJa@_qqj)ULmRB`bzJK^D)amx#rK
zV$w7`=I=hq*7;z_&(YR-Ck<bXCxG-g+}XAVBnSAL`+!e7otdJ~*AoRBpXNQj`vj0O
z4M9oHGapj)R{yQ0E3b_6r4QBot1bG*Q<tex&fA3#M3;5`KC%az#tonUyqujJJ=WW_
zif@RlUZTm?v3AXJAfo%9pfV%bn&G40w|hA|_%EpgK~%K-HJnHN_){{lW&o#*DH-Ld
zNFqBV9FlQR#_<rcES_WaDC}CKB%_Ob(qwE05M0^-p%H8crkl*`%sY(0SvcuQReYEb
zz2Z^E@K{e|5g|Sd6gW82;^pdPVR=uu@q!yOj)0!vK}lk0m-9U}8E$pF*vgRKkg=6u
z^mJU6EUPV$(Uo+m;obDVvH?9A>zPhZ6akN&sqa28`!6q?V$6E!5Q$tqs4immF`(*_
zCNN4Co}%ZOHfw17Cw9+Q$pi`+8`bQ~3aQD)qweN)H_DQ+Oh>?~!#vJlq>xM?e^7$L
zLMEAZgv!-&%Ix#>;0C2o&6TQ18?UUG;y}QF^KTo<XQwdM3<Or$ZvbvZ;7Isbsu6t-
zwrQ8@K$eS0W$$1)*V%3OIrgrd$fbu&1&i>IaIT&Eg#ikjj?}_sxEw+ZW$y}2KiBJB
z^jUe<PK{vuORKJt*d2VJy1=WCY$(t0134LAX@HDlI2lWNyv|Bh88B6dxc=&miH#Ep
zX*3E2oc0^p0=4fY1DH67{=5!HF5!FFy~}ur#ZbTXT}zrv0rEXR3mPpAOn{M*%|9>=
zu8{^w2d40`D4l3k%QOE8Mhk~KG!F*}iinZ9d_<|suqYP8)P}Dmi*SWTeY9A+%?01g
z)|%%&BZ%Yvar#P?$62r_t(7;3uuzqXFp(=mHSP7yuO4Hg0r;4yKhtZ`9ucYqti9}a
za2GJvLl19Uo(Xp?r>Ry-Li<%`HHb?YStT4ndE{nfpWq|<lV3x?n=eHs?%}CKcuXT9
zmXErC=-aHG`@P8cnbQ!gmFA92Nv4pfyRj@ZTPTfDgH7dzaw$tu(z9s#?$=r}F0;68
zNAU)MnruWN0&3DEc%48Ed9IS<j4oSFN@)DhiVN0dTvFK3fe=dKkL3gA!bn-(%6a~v
zxa-cxO|qYrp9g&Ly@=w+F8#n^>t4}tD4?fy7?|pgn$)jjPn-y+JfjYu{fl51(}DPn
zl22081j_EPE|^R$)EuRWI6tw$z>jE5PyUy`BF*I(z(LSLRkSW1fZB&h_!5@hoS4gM
zJNCAo{&k^^C@Y4a$axcCJPYWwygbxB)pt-+6f=hqYBw$LkcR20>dDlZaz!(pR3Ey$
zO#qrXblVBX%9lb<Eyhq>DCIyjm;R-FKk7LIg<hPQQwh0C)GAdxQ|=1?{};Rgek$H*
z1Afu1gqfJ#ntaZD#@mc~<67p9jSE)5&>B+|#YLu&Fl6vYU_v6ZG*c-%T!TR{B7k7x
zb#p=OV(eo=3eo`TtAcx?{0;G0zAg-CR#y*M#V_wo`Yqjc9tiao@(jiaCyy{w0MmwB
zmEu5R<RMwVu|O2TWxBP)rFq%_ts|Fw7AI5C4mJ8U=umaj&mx@9L5SnHjXVTFZ~{?K
z>Q>F?)~RAf!G;zM8~B$<Xk7U(nkk2!w_<I%puQ_@1^!lHnO{9EXAVggf+3<CuQCF7
z%R&W1{sYQLK(liluYST0gq;J7MgP(#T!kSF85l(aP$Pk<^}(&#b@RjlWstoZvdCHL
zBjlFtr2mzYYRkR2@MI<E=0wZqeDQH3sFT!K*DTMLGa9k@8Nas%1-gFO+d*-m<9jW!
zn+ivBXPiCrv}8~&jx5UgkVC@iQ*K;+geh#_>H1aktFq1w-Uyra)fX#PSQiAumBEG{
zoCaEB+O?){?p7p#vjbW!d9ok`IVbQx0g~n=Ht+t71&fp|n)TG(M|*YYGccdyrk`qd
z8iul|No!0sEB4FD-k5w5YG$NLz<50|fn)OtK<Ibg6jo|iE9rOK-wOVJWr&uGgkQ$(
z6KVHi7lGC$vAoyNE0Wt`7s#pAH}>3RWm4n_*r6&uLTtiE(*Ypi#PfUye~`<7NU+v}
ziS?_>>)rAmbhaza5rMQlgv_(FvKS4XYudpRMF`en2VZB?kk1U@!_FU1s)Xazr#$Ky
z>OxoDAz3<NU5Q@AxS0kAkq2{j+Hdv{&i_MG3Rn8>(40R}yS9>uWUa)i0J%|J?u*PJ
zuQ*+*;QM0v0mw7~O3l{YYDuEno_&+;OI#oilF`c+8;e0Rlr`a-g{jMPJ6MK-2CjmQ
zw9dh_?rVUz>SzX@G@4g$VdKBEKAPASgqNqZg5S{WP)o=c1V3nmHTrzptIH)x8zhNs
z+keUj(S<uF-7X%I+zh1e2MDnnrCF}H;B61PU}d-f#V>?IRTX3gZ$QFKSq!Prwe-Qf
zzt$5#J+nk%Kb&_DT%%Aty_S=>znODRcSy2z{W!j{C4B01Y_f%E+GNzctX20VknpTd
z49{LCEP%DWHL0{_i|8=lzQXccZ&xmAqmi8Fe%vbZ_o1YdWr%P~zqq~&qR0thwU2`7
z2-lzi|6K;Nt<`z@f$tzEv8LN9`$Xf*2|mt(UF%7_veCx>mvP?MJ$WDUjGmY^?n&EV
zee1<>Oe3KH-1xKy0(E^AKxcrpt3sFU*P{O%xRWIFD=Iz7x=Q2Ueds3=(W|6c0~>W5
zeX@8SVi-KTZ2uo*qAe8QM*=6{h^b#M;sEpqu?H}WB0{&BMTvK{l!NjpKYI5QWlINy
z`X7FKH?cz4?c)KmYrM0U!b|XTfTS+zY&*wZ#bXAcm*FH<GG(fw08q82Ndpj&v(z2X
zai^EanrZ)c!lx?#X*j34*y;S*%|)U(C+l1l(}#LxFD>2VEB*u?mszlV;lmX@4oE3<
zKH84`6HSMiXS<L!+B;I?TW5#JmCjEZHs0rv3_{G(L24==DZ>XxBTujGxGme!T<n9B
z%xcu&tfvVcwPNfiW_{C2DYn0^!%(}AKc}P_9@=TM;1kje!X`eB>ztG65HT7F{-1bU
zpj-{%ibG(@>nH#Jg_*IN5GY~~s8D735#@q6?25B%v#?|1-ZqXs2i8OvK=#Fe9^~pq
zV{dYCMWC^8M%m%rhk|W#5n*Wu2ioDm=ySF@0S!Oy%k7^2K=`;yO9?jLoS!(=yqTX>
z#$Yn{qg304;A7KO^Dy~A;W1GK9v@_J(Bhf4P<|nrYa3aQNsAh&e>QeCp-a6>j6Rpo
zt}f4%s%aF?_>ld-2w)gt9DGMDP1hn*c85!sVvr|AuQciqocH?@{O1RMH{T=!Dhudb
zHyqT^jcwQd;Yw=UP^ewo4T=p&D{<c#==;AWwYTsCOkmfquXnv}{y7hoDNMQ<vdN67
zNE6kSFW9ymb}o0GY)-T#$(41zgq{>+=j8bD8h;VzadvSlX}~>%p#K?vkru7rsEOQq
zq^e`GXt#5F^Lvw~frh2SA_!T&j!*qiwuhadTXj=$c(T|1L%0(O-q`?B7=>#vI|f`s
zq`&13phycDd(g=bAwyji1#XlJwF$7TcIj}`_MT~Oa#x{K8Rnw7xaBd#1UIN+n*6fq
z&|}w}rH^F<*_t|XCMBf3PN?2hwd*6#`F`EAL7T)Z2~tD$2&+m`T89(0;VS7%#Kd0~
zhc6(k6|bN(9A$f>+u%322>pt!=y<$>ic+)&jL*sB8jqAIqWaCE4AewHnU^LeEW-pw
zTa1SARkB*zP;35HtHq3pS$$og?HqfR&Xk8;4eqGcfPuHyu8W^2QS0t@WFOA1`?pvr
zdy@84Eb3W?ZXDF@NA&|sKz&`wsB`>{Z@3dW1~u<qu!f(_<S)j+Jzrar4DpCm%|2KI
zq}bX#s`#1fomvlBj!|ZdTuEB}J!ucv_7jJzsWCsmR5@xSuz9l}YkslyknnVA5l^dC
zLPvtC<}Eo9omE|Tix7q>(pOjEHteA5Yt?(ul>UNaysx)sQi@IKSXX}}Zs`qeQ-f5u
zth^8YRs&L=A4jk3;2~R4en^uG?KWNnv`J>xzDAR3Gkaxcp@{WzN6CJ=ca)k2UON-l
zkBm)qQL?JlPFPF`rc69|m^8)K5^erxvo+6(;e(Z`a>@LKdt#+)@?$)_aSS4Bu8@4U
zMI^ec2I-n(cQzzqwBEU6a;<uQzb#(KEui;-AWGQB6u0c8b_1&0bmB~H7zG&5{LQpO
zlL!>$M-`xDe3R$~W~g_I$+_Vu-qN+F+c)n~A=+pD_ickcL!GqS)TvD&Eo}D#T#ze#
z>-k&TdV9i`s3Fg;$(<PJu{I~A7pziK(NWRZMX^<*6(&eIzmm^~5y;pRgD^-*{Nj$Z
zkrw5GG)8l(KTq=*>)pZ%b`Zd3tGhD&vNXwg&6V#93Oq#nT&%tkL362b{-iI_UKf~#
zvC!5V6{6YJlO|toTjblKl*CAk-?LPkq*ErlowupbwPOn)+n!A$IGg0mnRCuD^KO*;
zGLEUj&7Fxhw0z??7HoR6?Jenm=2PU5`ufr-1_?%CDuBzVLzLU8z6SidQ3;}=*4&d0
z2_S-3A_{ob1qo_ZMY)O`iO3%T&*Yfwe$gcj(yNQgb%oR#uqQ?epoDp03O;1)>@3)M
zjpChBm^O@-aes3O9oLSeo~!|J{VF`u(XA=vu0w*xht@X}8I$WTuZB~s4NinUwZ`<O
z48F3*G}&V3&{*R|MFI?9V){;lg&-)e$qDT*eud25p)%})?*ThBERD!7gNh;gdwDSm
zC#wEQ454%##h1wnNAUDv$k~f<2?!@WpyQT{fbHawi{O1rZiiaLpXZH3l#z>wN^xc;
zXMmAn?mDLuYPu137e;Y)HVwH#{aY;ry(y5uE{BKprkGN@I1gwm@;lR*b;xpM_{D3e
z$i3T107GQ)B>ij*TdM}M>>0KyZjk3wo`leK%Q|Y{P&*!BZySNuUzCdE2lH&~n{AH_
z3+~*+?Z68uob|q**e_jh%VLn-A?D6m7XyYBkZ(4jdmrU?b==d9O_0e#VHdF`AMk9<
zN9is(=y{f}h6xDRN3N5co<m{!S4Qo_@h!6LDkJ-EI@sD?3dhsu-T~;q@*ZRhwuw8W
zxcZ{{7j>=G!G+EDC`hebJZ*AhO#0|*i0iK;o_DGwb~jO9GQQT*bHNF8Zn;BlJa8|I
zq;UVIf2@9Nsub2gc31KW)_&F$zYB?5yuVCK2ApPEN>Kq8nr0I{JW9F8G}ySKh%77?
zW)H=Pax?R15U<gt16mn#Ex_!xC!nS{Mp6nJcsh5a^BYt}r~VJvfHi9eWDrZc>q6qD
zqsd+}Oh!gwYy$_W(OBD$XDm~k`F-wR@`}HLF&VN#2~nC<EzPIbq+kqz>{^h0m3Us*
z`^;@dV2l7A%jdpiEeh{VTUYG7c7JT2yK>(2?Kby3gqiMZ5QBO>J1<)sYY?c^;b-;b
z1qN`lDVxjIaI#^4CKp6M2!&EOsBk&N$*l=!0vi`*h}GcFQLfj{&N24>%7_Wf#*v}1
z8C88vd~Ht;Zvm2!hVFJS_XK&z!+4mAgQ}sQ7N9XXlC?oa6w}#1HM#)GMt94D_Ukoj
zvZPx8biM0v?DfZx&_7v(Odk;(shfdZkhzwj6hT017ZuV9ti!~>^wX~$@50BVpg-rr
zCxD{YsBLeE;Uzt?L@DxhjgAVs(3~zA%t<<En|3awl~x>S=9hWCt#m|6Lya-fhYBZ`
z4PdjQ7XB2wtS(%>Jb0(X1b%*hOe;5)Ls8=8rGt0>KC>mQSyR0r1``tEESvW?x%EB<
zbh?u079rOkS`5Tz6_a!q6x?2)v>!9y!Js&D>q3(da%-WdprkgGwp_=1jcLf}q_%@H
z$n0*nJ-@xfnsja;N{9VJdZQpHET}P+bih$BtcDl)io|HLa%ZK0n&tVehL=(mtNQBz
zV6uC(v#Fi2kqHhBH(ARAgZAhE-4-n#g38#r28S{C1eSc&gbkC7u>lJnERh32fy^UQ
zkQ|c+a|BO~%x;%zu)?$fN46mC!s@e!KVGtVZn93>F=zAFq$lsh2H1(1cq**Q0wBdK
zC+||Ik@adn!T-HBp6F9Ts_lszNc$M!0SC>8i?_@DjYvZx%<j46Mu)x*9|j!lPO_3B
z<@}N7MW4#_To=KRi!MZ62s(78ML%D7TGGybLdkOHFAm+KUau>5k|LTtDDF;7`d0rx
z#h3}Tj0{wDP7e*!Dx9gi2qT;ef4~W={(_c}Ml;3W2?Fr!L6fBh!Q)nT!v>YrjuJ#E
za~5M`g@;3%$$~ezW`#P9JsEN4c2C+iNH~nIM7K^nyYp{r1`2je`w2Q#9#l`ZBN%jG
zZ9sYi{H@aX|14Uf1+-A2Bur4r_gzl8N4`gma1&bp9>7uuL#h6D0E9y;D}K`P%|Bt_
zpk#DWgX^Wn=|FhD;>{KkhU)iVe-^9|9J$N(H66Z(Hq^P2LNfLj;g_ups2iU4T%>sO
z+E-M|w;rw5fDy;94+<H9wxi+K8~`c6U@~yB9Q$R`#;51Z7|0Q*Jt=VnJ)d>fpWiyz
z4m<xg+Wo;`!ABemzy%jz9(AJy?8(DsE7z(<WvJJ_u~9Jeu-3@hn8GXtYSz(ifa@c<
zDK3gpbQSQkV_YmnGM=$3+)k4Ux<nTR#}*B(>rzExviyp*B>2p@At3f>x#}i8<lS_~
zA{LBCdqzY8(fRc@v>$-5$@>)JiU*CKm&hg&0Zfn9!@vjlS8s{%5+01^F&h6jJ6D?6
z-3)up8M|1Rr1E5BO4c66f#~7|tZ6kVvUtNJQ$6K<ioZEugmLK)?K)A7?kd(OgZ+9$
zRJA059@vC@OuaR7gp1-qGk-Uw{`-ZM?u`@QB|oB%tV0Gm0z|({pYOqZsqrexeP`?b
zqHA@-b(|=Gh5Z-@w}Xi?qjmV0VGH+|)Qh6p^`1Oj{FDL$_Q;pnP=uMP94lli55m+f
ztT@7FGLqy1>m}LO`Vo?E(RJn`yTd`7Hm#VXSQR4?xvuY^M69|>fIJM(qRl0I>YK~1
zi98it(s~yEFuX|5iDouuPAk;sK_hxgF(F+1s1O<Ox~R9kVkZ$_R5p81WYQH`Sa2&+
zD@BRV#PptGlAnH<#zBzev+TYfHZ>428Yqz-k|bFTy=7ENL)RUTkWd^g_f|DF3d(F5
zDL{zbG;w^ra|3XqkP|!fKN&JFGpz#|k4&vZ6@~RY5KjGi#OFz9sP*2Ca7`<>4?IcC
zo=O93k6bQ5<N^bCJt!$H9aI#xgVWdx=U9~oB-TKftc!CQj$=2%7S1z)!eTf;ju$lX
zvT|EBcpFfq=^6KIU0#{>(ZN-Nn%Xk$`~}1?2>!s%w`cYl2hAc7v#2%f_s#`)WneKb
zy16V@^j6c+(8kIZ?<%+P65`Xky%w)z-HX}`Euz(@h4%3g%&MX-@=-xijzQwxQ;)<C
zL-UUYsB~&F%)IgqRJnPYv#VY1UYxj2z5Zlout+Cuhp)EP@=$Xge+}W)&21AK21R;8
z{jDMW*7c!^B9OnWo1_O#!2#cJOo<3!xi`D0V?d(EtGg8`EA2K(v<3sUbbarS<?sZY
zu*$$Sc&vPds(kYmOBtl0gS=Kk-L=A?Y)=MfD~KWO9K14J)tvNriF>IC8%r`sYag&v
z19vj{3UW4C0h+K=ez<IePkhqMGhy5~h`;#6q;;N?i&QOyUT&c>5dpsh#AaS~r^9&o
zJ(YU4J%R4z7TE_5YZWzh)%+k$wG0<v;?_E9=r*YHP#O*oQY}O<l(`Ullx8viF)Pb7
z!~3}J=wu_9HtRMqOynM)JFZ9ns>xK(LfvM9_M3aa^C95;8<utCDIrjjLbaTcnLdGx
z5M4tNcIzxc#kcj`e*_440n*Qc@KW=y^hN5lKdokmb%8m1j+FMd`pbOLYBAMOvc)R#
z(9_E9!w8iEvQby&MT%s>qbx!3$!>ew$gTSDI-V|%#_^jIJX)Nk^G)bfk*~r<gj0ge
z&z|N@Xk-+rFggXHdm4X{{xZFws6Duw7<-IYQ}b{=3VvSmSqH!zTERcK^et~%!Dx!2
zhzOnmmN5(pk{4kU<P>Z<PW9y{wz9do8RBiDdwG~awdCtXyfpb;P^o&qn#8hV3(n(#
z2i%0VRyQo63$m$-aK^HsF^tuwie$enGMe9XID<X`ewMd)ZjySa*LyjsGP}wxTO)L=
zx<s~u28#Iw5(21=UvB=z0L+y3q3O$zcF!(ls>L@A{j7#B1Q`Pt35Np#vXAxfcEQ_k
zi`DNM>`};Y2I{EM_V+VmC^Om|bDx23Cy`O5ZAM2#H|;d;QYF=^^s>|U3dPT9Y?z^V
zG;-pa8^ZN5vNS0|7$678+^S;r7cR;Apm%>Sn8mZa-vP`#*sIOamhfzDR4UP+JG<Ff
z%(*Xae!+im7M=M)T@OF-E?xifk23NnO(IDO4FHmNWE)rjgR~7I1j}&r**L(5f=GZY
zxH{O!t2D(~sUM|7E<og_e*Kq9bjHth2)0H~!;!DprqvXd9nTrX;G$K<D1t&Z%Ny9w
zmw%E|JOm(m`Qn%LUwqibIDobsCq*nygWQ|`=-}%+#Fa>*b&lSj`w+fGtUBa_1U6St
z&@BJ^Kqi}cHJoCwrj55km#?NpUKqU~ya{R-$A9JEGZ>vOGUlhs1x8DZ9w1te!uv>y
z7vwf6a1BV5E?Ts9>#fgP(UCoWJS2D}1J=OLGY9-6-o=}`+UJvQB<ydlWpz#HeP@Nl
z=~f&VNkli^u98dP=Ji{xJbn;LO_uv2y==Plf36>-Q}X)lWZy(<Ed_6-)w3a3u!={C
z+?IHSNwH%N`)R+v_@{8P^YKKuL!6<Gkl0klrdSRGP5+GSz8bwYu;f%7Ms@~L5>XKT
z{RBwcrc<)Nde%@a@DT7i?EbdMu5-z3AS9mYpHIQ~GJT{hA;`+Fc2Ke@T;Y?Lx}!}L
zOcdl%*m6=iPF73KU}0KbSAYX9`uPSzfwZovnMC_}<x`zc>7IP!AZ`U9nE1l=>$ZVw
z=1^*bm}9#rgyDgM@|8rzDv50&y~N;2OaaFx_yxAY>p_j=uPxA4ez_ZgoX+Nu<~0t?
z@qO<RjjGNq(!><aCVqf>DkEEJ6=MtV1c=BAftO8un=dA_s^AaRjGz8n4&w)*^!>T{
zpd-jUxX~%EAJggZt2nPhV-r4l^h^IjmHz0^V9vc-C}$8#INCC0kfYo+qSFVNHUOY*
z8__2SC%#`A)hcFLyDAmja(TMd36{>~)ctP>#jb^}wW`yP>Ut3%%=E=m-a9I@LN?{9
z(Mfz-jU^KDXkdqU-QNl5&q?dc7j%TYJbhONTc;$=pv=P6ScW%_((M?0Le%ST$(JDp
z9(RfN9~BmOD+Q**i#L~m<XYt@y8#v02e}ZQ&%!K*6WVPRXE-+r1hv}}-j&OfX=yQb
zT9n{*+x%vA+$ptTptc=wCv9Tp!Kr90YVvMI)}ab(hvWo5C<SP$dp`KwKncffRr8)9
zJaF7ig{C!==d0n=QDuj-z0>?=oPX5ng|twMgaJeH-(*xL`zhAayHqxdQT7nT51y9I
z6>cf}J3-<$-x;h3yU3V4vbagbl+#dlm}+|Dx3o=}tdZ6{#fhm3^lb`kJ2ODj^2`k=
zpm%%?!3`)7;uB8K6ola&sQK_pH4mKIr*0gW;9Ho3Bs*oqdA&llfvE{^n`z}4#A<ya
z$EDdAPxnO^w83`Ad^(=|#uGb;qY+2G&R@cx+IBME!g|v4qY)XJ#~fbDjfLqUkA@$0
zg~ZR^@UxnY%1$<__U!VrtfQvop=^R5;5YQ*B$2uK{cs=>!5K+lO=@Sv&JX`(iO1E8
zh?iT*A%l6eaZg;zUxNoj#IY=aP0;f)=4!DOw+!9UN%DD}14qKh%`RF8cBd4C;rp5a
zoZ{P{7wcBvUrT2H<8>=gRxd!$ul$$IMK>Os%X2LFmZY0#zl0;Pu+J=1agmu7khe^X
zK}3NSddn(sD`0#B-jpduCnMJ;#G4sfuLRuB`$v>vZF}AP5s(uqyKuA;YRJ?k?P27u
zo^vgPze9s)?N&-7_@bB*RhpPrJjYV8yWN_}UctC3T|@rp?Ht#(a=-!vkIQEJ$7Zd%
zZ6#>F5{9x*5`>|(a?mIU%Ov~+0E%jZWGOLVAu&N5NW<x$@?XEP@ub?%$3r{4+H8&o
zeB%SPAl!lB3b71rwzH%e6dM)d#YU>2ct|C%0w|u-pdO2U=OU`Jp8Rb_nC}C8k{#j$
z{DUvqQC=~wD#29V;jAtK!ea}=me&-ZNYijHxmsOfsfBcCY^o>lx_iK6qG0b?-eIj=
zrr)1Szr8e)75<;%chGRfVTsMrPpZ`@_)1pgRkBeUg_b|H4)-vjttfgdp{=(sqzS!+
z6I7|KdlmFL9-LcKGWg=at5^rnN^y-BPuue9*O4a@;o=udiKw$2Fj)jCbr{wxjFbk5
zv?j9#EK(-3xroXHPP2gAA4H1Et?n)ny(63PZX)jji&t=9H22n5QRrfbW8Jf!w(9B8
z>(doJo8L42D;Ar*a3DwptOQCX;jm=fSuED$`wFZL%jZ0W>o$VtQ`{Be^=sQlLxaQz
zmRG>0R7caaI;FszjjQ#M!0!%7azbIk0Nl|UVcQ0JUmk3yz7h~yax}xaDAX;6p=SMf
z88nruTb%+E15<qT|JZt`;M%(A+c&mtTRXOG+ctLWtk||~+ctM>?bvp*lbn42bE?j*
zdvC3WxoSSFhf!<J*2b@o-n!qDPAz^dNVD$ECT#jx*N6bpl*q1E8*H4uZg#UNn&e(~
zn6FX*nF(Qb+HBR6Wc;*-lcy*%6WLLevQT9;lXZh6@<Dm^(1nj4ng`_QnxWxcU+R>a
z^U>D+D@lKkE(6DXgE_e<pwUmk?lINn>traDmiE%EdA_w6)9AKwP_05}G$FidX5H3)
zq1UFRd$XV{XdwA69a2LXlnfL;gYGWmMp2lK6S6^cy)1&CGL)|_go=gu?aYg7oZ`qo
z=6{pb6aSJIdy3Nz-FCVO+m4Hs(uX8g9@JV$6>9XTmWQJOO+a!2?NQ`{4f~fhbMnDc
zkyFp*e}Gv0R^7W(jHX1OX-fI5n`*4%UC(&D8nO1P-lO#VhF^V^gF~HA)%TQJ0G^vc
z9f(%Z0bJwxfnc(Q<)ef$6x^Ol<4|PuVAn{Q4#purbENRk{F+;62cHyDLtvm<s$7?e
z=NA+-nxy(9fNcRhvP;oGh{h4SmsZs+hM+2c!lJ2+M`8I7WS&hIC(O03{_$*8-Dg4n
zB*qv)`rSY-k8CrWjcr!M@jo!R=3>6$IcF6;#}|9=R-Ws%UOt8`NBhd(A!$@*AxEu5
z)IeC~a4^5Re`5@0S$nugkc{=c&xvtdR#KmhNEnxz5;lFW@0*TxB-DcwY=m#;;qaae
zPPV<4#Ec#4las#JHLb6RfB=>SGk)Fo<WfxW)0E4)^Lsklk`tzA38qBQpsJZC(=1x&
z#GNCv<0}7H?Pcimo}4&Uyq-2X`agB67)bz{q_Z3K_Hv}sTMvoKdMQK5N+iy0bR=Lp
z$XxECsNRIHs0W`SCxnRTBm31q<99l3gi?d+(W~$oBKVEQM>962&43WFk*&$8_Ku4u
zDK!cdE6Emqa!dBzucW1t-MOFqp<lUdo4#)c_Yu3-v(6*`xTxgOW$XJ3YZT9pc2;ui
zD-j=86=L3Wi_U#B`1n;*8l;lF4=nt1i)+IasW555S1Tad#WwQ(g6|hVy|qK+7s-&+
z$_zs0qgLe#6milrO#mD+zp_fcPcrD<epLl!#8T^Jm2#U11t)#KJO`Iabg$UgYB&%)
zLKUX0MOB(w*EGq$IXjC-#I?N>;PzcI*LGA)%Q}4v@7P<QmMxMu@}y!g#av_cXIBH|
zV)pxvDyI-7%Ak|YU^WDx?)NK(gcn3A-1n7q>e}o`Mh&)K83CXh+W!pU$p9z!^e$bg
z^hK+F=dYA3)%IB+Qz97abS8KonmjaZnddRc+Sk5U2Os{oFP=z9o5o|aFmBw%o_K0>
zQ34ZNk3H&*>u+d&u>=H~KNiFfTSAKSWTQYPH6!nU{r%AvF~DA;h+``n&uFgkJ}rjr
zj1wBCi4DdbK)cCM0OJ&xgZD|qk<DCiC^{=A@r$ynvFjy($2V3JhWQWh-EE}SouXjP
z;<e0-TU;}Fbs?f-|7EhF_Kl_(Vid3L4_TWr^rtxb$(Zz7SiAm~8~}jt5THl&6G~~i
zah(V;Er(U>T?ZQl4Vf5i?)~TaA)zBp7Dx=hiNF9%R>*iNMhk`M?M4eE;@Tk*&5^?z
ze~QH<qGl+znOH!in#RJuY8)`<*p0)*j3c#~6^$ct#HF(L(QLt7N{D99*v94{MVqS<
zlE=n2AhaI+n8b_e6Cw&}5zm8Xh`Ip8h=_{~Mi{9n2a_J9G((uoDwLxNt{#s|4QY+s
zc&>opBv@6F-%N&S$NG1F^aYj#-iA=4%kfP_Ycwa(MUOZHp^Qc0Nv*M3BEUGJ%Z>i~
z&UBjs5T#!(Ou%lZ!a>G0EJWUUt#_S2jgR0x{-Qx$BtlzXxy1m+A2A&n9;n6WUud6o
zDK3$fe({A8A3>aQo{~viR-cX%#A9o^1dO>AAm&k-_`atg&oW0&!VaO*^}u;6+)8rU
zHAr;9wy;y48z@Ysr$Z>z<69RY`$jP*f3bFy0g_8L>@{s7e#;C_p1!t^WPM%vm$8uy
z$0>lBqS!Te6RBL{FUT<!szlhQ$0a|ee;LJ9LmNM+OsWV}m=aGK8cv5kf{)5g0zBT!
zx2j6TTM!E3mbaUh!H-k7Cy<KIAo^0k_VMz7hSgExw_jCT5^rI{p>Z>YRRqv92rZ_O
zUHqBE7Ob3`D}ji>IyoZpn4Y;GG$<rPlJ7B<OGZ8SK)!}YLJlisd?Q#p7h*2m(&$oI
z5hQJilsYk#1vL3WadOnF9zven06UlL%_vzeH+~ON?{rC(bZrq!H||->UA?-Lp9mi#
zd>KpH%5{`Hensdd<F;`dwm;{+k?15OI8ZiZ?A-CEiQ{6-LG-g;rO@U}dF=Y==PgF-
zmgy}fc>EweP2gIPz4w&fI=&tTYPH9Yixoct=m&g%l`1NiWuR;1x0Ss91?1*_(nzf(
zEtenb@2LnR)yDJ8o~)h?_+cHa40(x{MI~4-bPvd7c>2!_Y)XU&kOu<0hP3++hrov^
z$m544d6%!a32+1~W#$}?espVD&yYigDbBQNPkVHx|NM09VS@F2{mHa_=2l&4!w_Xd
zrS{teZE_3wNK``FVluxAK<oJEt)v_cOd-`!iz0i*P^<LkNzkIuwN=lAc)OzblAf*?
zT+{g87unlms@;L1h*Avys4(NR#aTk^n&Z~ecSuxFj@#{RIeZ@bqHI;E4ieJ~UJD+D
za9#jwnRJ&?L5&~qVOd?E_!q2Tj;gnB<cF&`k-PAZy}@OMqsMC%U=f3vd8J*pnPT&~
zLMZw}Wm`&3%W5Z-uaZ(aMlGwqBYLON>!#3pBbaky4VYuP*kzOs2<J7j@3xkIU%y*C
zFAs<};(p5X{v29U^cN-k?>+SQ+T?{F8LJ}}MCy_~2GdDC0+gE+NzmXd7+j%}CsIQ3
z>!w69fWp@c%t;9F1SW|KTcRb~xwt4@8pyJ0b`(^X&G}DLW_$-{@JO80q`?<s61_q<
zs5t9y+mYa{2`He5K1@O(%JmJ(0+t{&)I*MQi(}!`&0g}LR?j4s92VE!{QLC=m0%_(
zTtY4|VZl|Jy90_dYB<Lq!(HCkZnUnVygkNRSwvMQ5kwil#{qPC#Jo91Af+U1u@2>Z
z5LbpEYHs9Q6J(E&WdjTZoMjV`q@3;prix?obMqohyUf*nks_~TC1>lSRA;7T+-9h{
zcB|`W{f;d>WndC(=!>k2MjqHeTwep@1p=h_l;%vdaqW$~<a}5k-0yFV1zNE7wiguW
z=Xg((9{?P&Is=ZhgJtRkk<oIY_U1}6OBIX75LsBkIkrXlKzK9GXq*R@1b8z+FPa|S
zP~S^;&qxl4h%=NP-sy802Uo;91O$N^x6ik8?|Vx)qeCqFm$vP^9SP*>cyjM~#rz4{
z^trSe4(vEiX=jq&8wjN{1$*AXWwD`kNP+Z&c!1Dfax@^G{NPDJe5F<$b7)juz(LMR
z@#wLrK@tSDVIbk=f$WG8jV=66-e}`SHopE|_&OI+;kd0U3TZq^=aXWs0}s|m!j*wx
z5tuAHTTk!BG=N2xw@c>G2e~q30q{gpm2u+_ujH3-xx#yb%~jBFOQVl>G-v@^gS`6T
zApnnIej?w<)y04(y8TlTb$tKQ(*&BvCqy5g-V$v}C|vGx4L0!o{e2D&sE?_Ch(3+s
z2bsIs!l4RkburlH8IP7aNA*f9)50g6ySMwJUBxfJme1zNPgiXHQthUIm71W#zwW8E
zttVKKK9#;|beXiY{`Su-zs(eHr!ESXK!CLj2L2ta90gde!USvqr9;NtBfJdGDz|$L
zcZD`)e%rQuiq>{-AvUGN-w|0|$^Ap9?tiY*c=c(0K!548AG&@91f<*^({}I<!ou!I
zHz+csC<E8wi=b2r2(9&}u}_la_I6)n5~`XQbCF8b@g7!qb9VjUwcx9(*|CPwDgfnS
z-=JDZDM3aa-j4?<W6t=J--!?Tq{C69Md-<w)U#OCKE-o8D{u|wNa!J-{4GhaoFSc@
zAoY!dn`Cn2Ad9BwsYqEV-^0B~B%+4xkO1Gx9oqJGVUYRO&JG;Tw47~c%gruZI3EIc
zhJE^r7wJH_Btny%$S0C<dfzH#7$8krU>T%M8bGe{uPRPIEIOI+JkKYLvH;g#OlkqC
zkSH+0rC4+?y0I>CtnerS6R~~<WAY@#`I<W7H^{0RwQ^;<7c!ll68!T?s!yT#6|Y8B
z1ZWyl7L_@D(s%xBw80c?0Z5<ZdkK66Mjr{c%|`0FM?fzV4lQCe23?z-HUMG^^|t2}
zCHie}bZg!pMRD_0PIB;d2N_w|qrEC-LF`RUi+GD63>3-U@1x6-;-mPR%R?1XpxwH(
zGo3@R<)Y=&)#)E@^k|te?tz%2i#ty-6pt9Ti!7II;2wHC7Dzxiet1fB?4#fw0|F<%
z&VFH_Wb1OU`LP^%PenZ)3m^wug(mA`o5wT3NU~iwf6|J$9e;VokJZajtDEXUmznsP
zp#Mm_sj)Rf|Bpf}jS@(%;g6lNxBCoQc>UDOVGJ2N<&`vCMSr;$E6|-53-tE7&$|=}
z2<qzUM##)Ls|tJ=aq57J7^Z(`U$$sSlG5v#;laNOFQMbK2Tia`2mEpx!fJv?4QQJi
zpE#Pt>9#@C`0MI7xYt4m9O8ar3WKq`Oo%b7`nf%3P4uV{3;e~v%XYO1k5BU&swy3&
zZ!VPSkBhs6j)u{7LI1K+Xyh_OI;6T&5}k#yZen6*pBTWDLBo=CGy`1DCCo#@&lAXf
z5JqeY$t5uyS(VbI1=s|=Ph$uA005<1M3@ORFsC;OZbb>M-`ilGz`voZ*eE=U2zq-h
zWTcMsTJ$Ay9KNv$J-m_K6~E@JI}mdS-|jVFXoCbI{>v8`=7Py~oGb7MLeP4MJ2A7M
zP`(-G`ws^Gk8xzTS~)u~m$RntjA$;w?RYUVlO0r^X$K%`zrOIC?amOK`J{HypCPgb
z9J`)QTU2Z&SJHR909G0;m7v1uFBdr#foE=BEg|ZJko1E!R6UruL*}Vf<L;Q#S-8zM
zw|@1K>lz1FY~j;kG{V$#2x9dVqergx6N5ON@);Jhm%j_0@S(giDtrIf-eUD>$txMO
zgbmC%Nx&5$Rs|owQCxzy=HZ?fmA@r~h$qwyazCz>0C>9`MT?U87wbg*PTnDh-1m$<
zC^sAHwyB+92Bikv{)8Ck88`eVE08d}^S$fZwtucazSVNyrpG0aH$-C(5GJGSelh;A
zMq3R<OUB~!fBq&*K22^nVFLuXGg9|kMI%)O4{^H<J?Hf*cx_0^`{Dy4Tmi1uEB{Wi
z)JdT2yD8^*5*YB6HJfAlJ@joh+ai_vP35hu$vYkzhbAV;e-?ojfXb~T8u|BrP#F6X
zBQ&79%hH&|5P1_L>umC~%YKt5c7WYozXA=XPJ}6B%du(OV>eawsEd9NM+Ukh81WS{
zo4kbJY`Hj9);cF^v#WP_nq=Yf?rbzwthshVKF0tp-3UN-d>2zOcV79fm{T#Nb0k#r
zNA(EuoBAO$!%K#cEtp@T?V`N-Bf)I15I2d00|IF4%qkL8o{_k_B6O8J=xI(ScBRUR
z!;1dlr#-wynI+^YC2yA1m(dTW(&D@751hHmbC-K=4OKo0LGW+2NUx)SA@>{ePN<*c
zQOL{si)}#e3d?OggPvIh`2bm)M#l$q!p4&e_}~hw9gV-X1&ujRy!+OK#AUzX3yg|J
zYruX3x^jxm0{$n51vw1EJBq5=`#!yQ+jiQ;B}mEgo28g}km^TKE&Ar1bYTM90vuju
z#CS?`c{nT%(5uN3C3^;PYfd0XH^UX$Pq??riacQN_id<+{B3IaV{ws<l7BN1Gkv#h
zSx_3Ewp#wU03&`d>RTZAPe#F?ip45Kuv_OB8-%?@4Iv`SL<OVGw{zedf@gm|p$yDa
z=_@=N8RTm{Eka^sL-FMR@L$I=BHMcYn<z+=6s10AL~MIg-MUjmP8|dXgqqM^Q->;k
z*69FwFV#WBMX7mp)Ft(Yn#d-wU2_b+G{-8f$7j&$NLX^Axk(XP#G3B$ai$qHM>k;I
z&RZCJOGs*e9Sv=J@ZT3l#$7br%{Ot<R|URt&pqYOd_e7YnM~{h!MCM2JEYW@&Rd+Y
z`V7xbTSvz59WKO<SGfo*#3%#CgfK)t2h;#eky|k!z>O1=0>`$&<Ol$1K5Vo#B?=nE
zzTSyR69~K>$Bl{$1+0AT$T%Dy@3M6%zQzILPcKHzea?|x1rTK>4j=1p(R?6HHZMt*
zh&fr&uMd7UFD$hnY61l2ZO$I@pX&WEQ8Uer{y%!O-;2>b=3}<BxH`KOL`0%NhRy)%
z=qjvoT{H&VY$XkLQVrS#v2{5(PJ<KUe!pSB1A{S<my*?m$VEHE4~fFMTP%1zoa`8>
z02h!@NN~uDr~0@xotWV|8+3@6kOAnUBX(bOa5J9vnvbG>_A!wJNx0qy-J>d<zKD#d
zLh<=mFso=%?7G#?8XW$Ls`c7=ivR$bXCHJA-LexP{qIjZXk~Kb$c7C>ly8FmBs%%T
z)lzmH;ZVir#yE350_2(#Jx6RNwLtkn)6&*vic5>qG-hRm|Au2=f}kDYe+OPAB^2KC
z;4bMs<?7jy*z2rHzg%1MFRSRyIdsP%<g48d#ZM^JEiYZG6oa<L1^_YFs^0gEi=p>T
zTq-Ln+|QW9P<&{${GJ}^vG7fQ*Z0Q2b%-ZC1#ragz%xR}&WGm|<Ub`jf1Y{+{Ybe%
zZ>9PNp!h%|rCXo;=;aO-s4kGp^P56cGNN;SXSnTkz{Rlme0Rx3gPQ_A=|h5-Sky1E
zjlFYqkYaG7a%+thbO4bm8zq}q7Iv^66H{13`uUb>*kx|+2qyu-S2j+j{TEl!R;~~U
z7c+^`YIWBb>o1$5d$SpM`xDABZ(3z)v_mr$u(6u(gQOxpqGBT6{7(d^C&kRiInj9+
zX2wt&9HKGs?t5AKrB&O7tJ^#d=_4N#_^*~Q+FKq0uF2OxUjSJDPfWy$_>?pz<ZaaV
zfcDkHJB}Ea`3@|dWEq&o&OJs=*Ms^y4$TO)_6zVfpRmP?g73l@&eJpHqzkv-1nc!i
zHP6;+9c39WG+Q6`D5&I;PCGIk<7!#)YIlbn8!muOC&z5!hXdhan`XYBCU+NoBd!#%
z3%5`@Y48)#0*o!Mw(H33k)t^!0v>j1q!Kao6E4$Qg!%GIk>o{0oyLn-thTqCwG4o2
zv(H|xdY5cp+H(ybh}Hz2slvY~D-`~xDAu|`I+x}>)4A0W`pw*Luk)BxyYz)s-USp%
zYP2Me5(748BpEY=RJ_4^^;-hj3pAWA?=`cTIhq|XKb`3Mf6kU2p_in2nu|kAh`nHB
zdV5Q|9NDrHtH;f&7|Yo=^yw|Kgl(Vm&)Xm__*`reWK^IYlyOr|ddN;oEHiIvvO|1m
z_!5+>Bw3OgpB`YobE&O8g1(0sCk8Noe_#4SJ8ww0-Wr?{?@kA~LmRc=E0C*!1dG4}
zfHC7u@3U2<10XK2$#TgYSe(B<R%^hKB!!`%J4Z4qCt61K`9Bsc2FQLB3G7okoj)P=
zsjxjf9Pg47gDVrpj1(F5e!0H~QbQkBje5l9njH|hf<Gvr$<NdD{z42{e;UBnkUU@_
zDDtjWDwI`xktq`ie-@<vjwG{txBRmJXiswy?R`eR@Wwshyyj)~^iuHiwQ;j!zDc|(
z-+gc$U>LTpNU)62hb!BC9!Ei4#)`XSnF^@AhQY9~x7=qs;~fwg)BV#qRPopT*6tPj
z@7C>VD$DR!LXke5I>Y4CyLHiW#|bnR&pPw(W|JCf_C}sZi3)xD@Efgp!mP+ZfKtm9
z%2#!S<F0==i542|3@pYJ++cIC=g|Ifp>c<{l^zAC;V?~!$~@^GAW@~avS36@O;3`6
z)~`y;<OqrxNA@&{Mqy+6_1{iNMr-lJu}SDEyT@}5yom%6FT5&J)ttmaeN87xK$a9>
z;w)0DM+vMV?~;1KRC&SmOfGBVfM=mLQqlYuqer*jV8qnH#Y%fpCXzT@{j@2QX0rqs
z?NDy<Lu-z9zQR$W^9J$NnAKg;Xvr2$Q4UqSS>V(%v)0Fl75Flzjax&ulv*BUMrCtK
zmoHXA+Gz}0&Lk;;!oti=gH$d0Fgn4|M<th?24<<NK2AVWCF;-C<qCru0OZfOwVQdE
ziCPa*xXr@=J&Ly)zJI0|B7s^w%sGShreMOuX9ndeg@C#l5Cs*LAf7a;&LdBPn5A`D
z?bBu$svu^r#jkBQQ6r*brGuokS4e9ZkpU%1Et?2E5z(F%Wt)~R`j02wLuEU_iK6XF
zaFs8Fl3>O20A(TB(oa$a!1W)ww(F<S0iageb{}VBWS!sPq4wbqXQKQU#NBA{&tz8H
zmrj9X2jRZW>z!);_MLjsHuu!=Fi7lo_(z-Ct3G4QVAR%B^QOVZ`d@CP!QDVZkw2;b
z9&hK3n)>vV+*QXu6b!h=2n3pzX#UvvW3rI;%j+`PvbK^22SE}=fLHP3+jb%fMVe2D
zZJ8)=^|=SWw-ZiVHz+L1t7S=`??!tz(m5Pg2vuKeH)!XD-?DXpaIdzk@>0#5kG86;
zLvMo`_VkS`8P~C+v?am*bKjQaYxr#S4%oxWW2#z4>)8OE>+MOw1%0K;4gAx2eN
zwT28uDdXmYOguCfz@c`}h9yO_3g<?Gs|W#KagNHGN?h5cbBx<xMmFfymt$USLe&c|
zaOUYTl1+1^fFvk{#8jeRx%hM4HTsJ40bGzLtsGd*7x)P~Z0sL{9L<gIKAxI8x2S))
zSTl@bkS^8`hxulWpW@uwr0Pg{u4*}<f*UnyU{XYN5{-c+0jH0h12IIrNmbi0pg25=
zV3Zb~Wuuj&C49dpb*nQFgV=kHZ=N}eRs`P4md4r!1)$k7jiW4^DGEifuVikq)k{Zv
z9G>?%(IR>jLrEtO=fV)p&w%6qnhrcx8V4(ogyYYvwyRcC(<<#Ak%J4(N@%+$X<;YN
z8bBOdDp<=a0-`v-hZOb&MH$><TBuPN#G?SUWJ`@XGzlKhEw;ZGt#SuZso@;Y^OA*#
zvL*^*18k4A*&mv7u|;2Vk-sbotX8hv!K=uoGA4E|+<FaEe$tY4fv#eK$J8mskz`v5
zC+6-3G<FCbY&U+m?tq4y4W7BL8N_&2CwcVc%1am#;N^#(fH%0QX(2ox#%v86C{o`|
z_i59uM70zKjY|6TPbG+W`U++DqvnIF;U)=~gd`u&mYrGvcaES$g0Gel61r$8h}*LI
z1;L@mG_W>3bB#};u1Knq_KXv!g4+@WoP4g`v0#)<zQb!<HX4IbVA!fmqiV{sU5(iO
z_Go1S5Et^_qP~dHl~tZ}f5KaQq*vO}8sw9dGaOT5WY?W%N_EbU!7z<@_|VEZ>J2Rf
zy;V4CSi6_x@rzJ#fn52h;*hWEkbSNUfH(q`$Bm~@!w2~JgzF%DwmEyUr&3a8bRfD!
ziCtG~@>=NmU_6|b7}_^e{lyeXf^n82CBX3oNPjY|4mnG)5+w8`0S5?u$wMe|&=zr}
zqTy3l?iGiP&EF3peO(Uoi;=qTA$sT{9sUNRB3~u^x&Cq@veyNI5t6SA%C7l6%5xt;
z1R03e3M9um-eIk;ubw#-=BPSaUr)#afg-AxZDqoWjUa<ZHz`Vh%eMMfu3V+3{>uZv
zj7*8Ilno~eu7y2AFxGzitGdF)XSn`b3@$VgFuKKd&Lahtv`W?63+E-W#HY+S{}>u6
z{^Ie7bXrWN_2h=es<>GzKby4je7|*c{?Jfqct0h&ENsi6z#BKQ(2G580&(i^TrRi>
zg}HzEyk<v47UX}d#FQoxsXBK&ey<L=AK$38{l;0#%udw&v)bFm81U1%oL3)Pybg!`
zvSwYC7EVz5J@SjRn`mHnSX;$tYsxF2ThK2ajy2IGTVnQ5Le^MhHEb6d>NM_nS1p?N
z|0#*`|4E|#lcvj$Ad0f`P{+{G-}#R-CmLY3tS?-Z4?CfI1CG0gr7UFIHHibk<>3DX
z5F`?6`$H~(f8#qaFwVKWBzm!2eX|sfR6~_Co?<@+y`aw+Z(tpv{kM#szOe-f$r{dI
zaW4A^f0IsyV)rJlG{J`Zvhh6~klbbZhinY_5oE3K&E(7MBHn<REjdrvtRFLbIpTZ$
zq&rl*saRM5STs)U%h&8)+d|%qXk9TkR)rmi-6^Cv#pxNG(?HLBH}lT~G!wy-$G2Z}
z6d0U!M#)Ho!ai8ibcN<89HU01=AG)Ie5Vn3qp>z+pm-_hhYhw{J@r3XZ`W^HmjwJv
zlk0prNuOr%Z5~c#aE&~^spCX5CVVt3S{#keRt@QZ9&;UoP8aQuZTyCJlZp6nfm>;g
z6W`veW}V%%RDrNB8d`u@k*!XKX3>kY((4J5dcfW(WL~(k%{DM=6Qm8|mte_a6<EGA
zp38G=Nexkuvd>rvJ-NzRqpDQ{E5kM}1?6>t424E__UtQ%2EL5a4Q=L-0*RMcjQ+{g
z3BwG)C*wW+-wr;^)pdW%?P1m28wtYPZ4#sC?nX{*hff#70te!TT*EK+Gt9YP-B<A%
z@4_$QvP9UxgTCD4*YaOpC*bP$z0-)Aed*MpW-9{-F9Z0^Uwnw}@gZhlloV}M*QqzM
zhumhMLkoH?W{A0eC+g#1-(|nWRk4hS`E1Yu=Mu0^fi|OB3^;L+CAO=fn6a!I`$)r$
zRNl~xk*ov<Or2vmC5yvbak8hcf*E`0QX7ws@>EgR<oF47zQe>gqc%rV;+~#0E%s_#
zb-z6&;t$pOLp15x&ZR1^M47)Lq#U9pp3VKvEpDLbI|`%(UwXi$WUBR{OUQ^L-a4NE
zDQ*wkF*Vc$C^odmB6wV))*d1Uo0wUcM_$`aZ{`4&_(09Xh9r7?yv?^<<L?ws3_h7B
zClR07$geeJGzqlo@l_YCpXu*YE+f(pfg>{FgfC2BboT5tsZ}W}X3Rk14-dQ9kpzyN
zcaf6<<UWE!lc*8KPZ~Y>vaD9IO=RtWG<1}P@le#8Fe8yrPI+}%-z**Zi}VVz{mgjF
z4eB^@pv;74ONz5j-b8ar!ip%QB3`0b;{s{k1WwpfU#uQ{JRw8K0KH397!{!de2Nt5
zpSCXoVEOHoteF4)k}~8q{lWmDM*Vu_t5$cy%5)kl^}laUUu$$oQE8p)<LH3um9-<A
zGBC<=i;J*9RF^%th1z`sbLwMMR_M>HUwI=xJ@LWzZ5*O`!F43>gblkzW)gaL;#jQ{
zpk2?WI^M4n)r^gF5BE)=j^UwvlOV^i>KYM?dBzo8yf_t28O?KNawZTNHIxF$o3T`I
z!@Sl$D7fE(az<2~02=R2Oh|xn8*!n&jc2Y~n^|8jtjc<G+5J}63=B=y{#=s93Iq5b
zJ+JJp&jWkd!uw^j2HTN2NAQEW9Dr~i(N`z;HTULzdDVj`0;z7jygkaI-EzLv-o_SI
zFh#J+luk~*#KA)C`t`r%tN&|l&?h?8k|Zt~vF60-IEk|XmIYMU<b}4EPOH@FCtRKR
z#3F$AC)OnNZ0Hp2NJmj5FyRKpXMbXWXqK8b6xd1pjBUBLo`HP5n{d{Sxl7pHxyZ43
zfsLQpweT(*dX7jQbvnJ}Y9XR>4s>fs+&%WU;B2$a_HFX5DrA|AjT=hH&Kq@nzZ>1v
z#a{B|ob9&(1n*z!uauq*)UDzEY{+&k&8TE7)v_@kf(~hQ_DONq*pIpkZ$)QdbGqnr
zA=I7KNBE!x70FrX6cI#jqADti;DT>aINPgeuVjQ)&+JVL!Fhdf(|*>KkV;XU6#qF3
zmk;df6tga@n1BIiB&cCwX)Geh38*%0GqtcUDQo~(BK+1_IyUk@e)Un!p#vSO$P1iQ
zf@uXIoWYb(C=i^$m7)c!88y^Qb3I9o`umSR;o16a51*Nd^EAA`8#y4cHPU%R>9&YT
z1qVTxJpwzL&A<PO1!2y_V9o$}s&HUqkN&<BzINVeo_YUUy9U(qJhM(*^2V|n)iN7m
zzf%B6|IS}t^uMO5kBk|0h%YE)N;O}RW!CZ>#EGyj#^f4yp`H8tQHaON8J5|fF;Q;4
z>1^a52?v4$pqj`zkKEPvKi^k$-A?g)XWavV#|BGIyr56?yD+ri1X_|foJ<ZC6yP7G
z5Bu7b(vLfuP-^oA-8TM7zJ5z=VjzfYmiq>@S!11d`y0T?(M*%aa2h{06fo${{vn?e
zAOTQ<f?SA<oBSc=u8uy|RCXyY4^7I&7V)DXzEoiqjvp!*0~bON(8nbH|JZ>U%TdbP
z|35nr)4&MB{bL7mknR4r9RU0Y8Bdsmv*nfnY$J0fU##Pvfu@R;H|zFqOgM_~<Sn*S
zsviS@DP(XEw}bQ<jwY*GR8Tn2N5Gd|?S4XKdAnooOwimaG5BlMI9d%n$$+8~!}vjL
zKSCfC;<xaA9JK=zvLGmIQ}uQ`DU^*T5Nli{4+ct8NQh`4Ap<751mH)O;YaB1{1xf`
zlVzTAA4gB6TF*P>!>*~OgAy=epsMs>Z%})9NdFz+l%GYN<ob@b^raE5pc~zc^f}Ir
zj!}{yD3m=+P(8-N=^Twg0L07lY8*p(j@X5aG4XQk=5x_IxgA>FY}sKMPkzTm97Doc
z9YBufPuQ-O&)WdrcaGoQsgIQJ3a6)4<HtFOuI!6e@tYw<8a7{ZDKbC$b@&Q}A~Pbi
z(x;~WQ(*=g3JiZKQzXLCzd0Vp$tv|!r~_*(FYwy`M}=k1{y`XpU@#ct&yb5{?sKU}
zZ`0lODpmEoY{i`W9pRP3L;VDZ1jfb@F^X6*ivURBWgx^+G9?ky+uI;=7nG5m@1kt>
zH{ld2CUB<SjVZ)x?j(`3+UsZevqbr8eLeTUG%3P@H!3nXL5uA>c)H*W&%kNB-h(AV
zE6rG3j%c7`uHZd>gj^bT6}BJar2nTXtKu7)8E<gR;(m_*N_6DINc=9yK&mBFV$;SS
z4zQ*imQLW@Y+wXI#U(Q~1H<KlV-A#h>f-76x2ObDv>S|p!J#?(P+DeH0o{p4ew$jW
z+f9|?;}Ilu`yk1|I%v|F-$rIl?%E~1i737T9BDl(<kNw+o0C&FLFl_nx=4STYP_d9
z>cikEfV6CB@6lI=yTfuS(2Jifn1QNv445c}LuL4*He%}rgzrgGs`=n>eP1g$?y$ec
z?VW6cc?8R%xi=^+E4zApm`ZYfm(%Hwkg#Kr!3(X0Cg-c~uJl<&##M~b{`({A*VmHr
z;d=nqM0Rg9Y&`Dx3|Pl%)sn2c@znrfIa&fX(dM}^#PjL(m92J4LNciEYih}qGN5F{
zcwr5feT-{rLo5v=?u6yb1R8rb#yiMLzfOmG?ug{3&HU3!<~0mUa<0m{L|YG<{z3y(
zU?s4CFrY&pmr-9bogvirk=uYjD`deK72aDC4JS`g1q-Hp8IwP3UlJmtZdDpRAs)Nv
zIAPANP`zuBxcGgAz)x6!S%4YZ1)%Pa+8_IHa&7xy@1^+$K6MOvf!PEwURO4HVZNMU
zP^=ya8-Yj0`RT*6c4o7l2rS>BiAbaW<<zzN2e#wDzTtit<Km<&`*xK2qIr};W!HI;
z&Iu?qPAWNh)w5hFjef6C{m#_kh1Q~Q28#!~WarQ3$%a}->LqW)xP(DnEPy`RJh67u
zWNJtk(Z0SE-1a{<lbg}``IZ9SM~6{kLmI+|aJwR}b6HPn<V;8OV&*WLXz(~t@yS=n
zw*GQxK0J)uImCJ5n8~7u7?H35(;+|OuQZdH_qE|fJWyQ{?O-L=M!33wYM{5oa2^qx
zy1fZb>s)&53zGRn^1gHmfd8Un1hw~mr^(~mU+bz7s<|ju(Ek*R6;-H(D9ZI&XJ#H!
z@M!IMT!fdn@9@_Yh7bMYZXL^?orKN3Ra@xxy!H@_P)8z-z=_70EQhe|Xy``2`gMkL
zi;n+K$=30hcA=o^-L3L_X{(E4)v3--YwWfW66f-IePjEf1tNYPpm4k7Xz|#VMgoE#
zD;;(uxW8Dr-z3BH=CE9nlt0rb<&4Hk28At&#EfJ)P-3D?lM|l$vmr#L=j}}?ZLc*t
z8~CSMo~f?0&ue+<2Y$Y$Q0n>b;$u2kwM5s3vBKlDwMrrcLm&jWpWf_pB|-;O57s50
zHqUgOV6lInc#nh&fZ=-_fd_Hs`@4*tQN%i(*z}_nF#B+u)Q<0wAHa#%+Qm`M#|V1|
zBYC>zhjPAOzm3FQzUmO^`&hGmmfsm3ytyf_V_T?RN4PaKVcWfM3KbNvJO2j9Acuks
zqo1``BIPI9hnG^pyYP$Dl|2cAyOc#r>3E!cEk8NVra)mHAm4?Yf6aV})Y3S>K&*_I
z%b>II4zJ>s1M0nsH^5ocV-rE&b7xz<>j4$>6wqyQ5d8G;;q~I+=HgC)m`6%HG1!5h
zmWy+}20>8`FD+Pb0^7gSo}vgESV#o*FM8<}+pl!Mx>({oxD>a%u`wI2DMx?7>LxD_
zrKP{eb}8u|kb3~RdVSEqKB|-+1P|6Ff8b^@t2J2%n{+wlaCnf((7!P%LQ_$HcRP*M
zzc5>WhqN|0f`M@bt9~o;HgKkuf)TZDv+^~&^{#LsleR;#wT)cbAmN@#I9(+4@nBcB
z{VuO>51kW`K1F&l6Bqdp5a;oBaj1li-B@XA;Pl-JSie3(_ykwlCQAqta0Y;Y-XCj@
zPWiicj@YtWtn<F?fzs>hZHgp0XDYeLitJC9h&%icDs|ZaP3#Na#CUm5$51-~Z{e?>
zF&^XXW0ut@mg6@r-bwP?$~2ko?oMYX3iq9+@o`k4qPlmkk6M9M)^v2e5Uv1LD;8Ce
zOs+wId5Bdi+F6`)8QE_c66g<VWuc!+A)$~^9_Cbf+u+4DDbs#m>7oIwRVX~DMVluK
z=OG=JsT!#!7z8IJ#2TCKm2!1TB}NGr&s*&=@eT&faC-x<xR$KKhyDFOB<ecUt**50
z*E>qLhVV_NLCi8~)9tpdDeg#<nTL~TX&wlGfIl9d-@cwo`r6SG-uqDPZ+s;MwqaFR
z#;A%&Ib|a!W#iw~*G%;sh7Qk797d0<FEnrc)xai{MN2Z~+YTG-on^7t_17HB7k)Fe
z0Wq$9G5Tezg!#Rk?dZ5e&l(L6_|QW}eR|)_x|_q+>;t)YlpmE1`mkoH&Nad{tcL9X
zf^nNr(hIQh0>l+O+Wz;u`s*5ya3Zj?t5A=1+XmJcE=_}|ABJVjlfOSvWRVC%7;#Bb
zC|-lXHs)sQR@U?j6)l!%m}6wh=^@ec8PU`Y7lMP_HZO}<T2W)lv`le;^I{HciiMKw
z?JX&M3?ae%dWj!p;rjaGfgDjw^@bEMd@uQ&ZdEqhW`5dEGCcXOL|8eHuS~k)z1JAy
z@D?)TfQ&t)z+c;UuPMCKfPZzwPtrZX6&kjci&y2(GduPY9<&vlAJj3^RpJxs%`{KF
z&F?{3^CBc&*2$8onF)bQB@&-`)Qfnq3w`<VhAOIHy^b}Uq?74!L{ge%<)>GGHM>*E
z1sUy(xmH)$HY=qpjN-qP?tjQrbsM)o{R6zn@d-|UEeg6jh{F2rj}@mrS)0H!WEoy=
zzTXqx65i?rtae8AK>C39)GAimI}afS@uz|nPTBC1;kamZhS5C(pFj;34(v4D)0a^L
zpv+TjoQkDv`R(Z{H-JJ}2nYB8#P0AI*MVvf@70SOr#UPK=$nr(autS(K+kxJi1f!F
z_xxQU&WtN~<5v`Ie7vl59or6JXN}W%tE-K6hy*7`hc3fk$Lk#wBVQ(do&lFz=I?{T
zVL{85?-<-<P?1AR4Oh&JeN>u~i;W2&7>fOV;JA2#ifCYZ`YDq;P)>aS5IitEFsMih
z<N5<~ESuSX!AK5svJ3F6onu8|+sD`ocCt$R_m0##XFv)a!>5mul1+qh$cGUrFb!j4
z-$y~-mj8B(-}Z0*^dezeWCM*#CKQ3MRm2gb@N!8b|3oXB!(W#hYIrNrY|-PopE>+P
zAx}<icFlgU+)i@oNxbg>1i(2K@<quznJaVI`T6;@PpvS4Wv<=4yt`dAHaV5$OFGNq
z?sE?shU4T7j|$WdwXx?Ry_P#+r&Aw>rx)Pl2e6mr%N5Uaj%=8<Xc<Qgv6ua=?#wcq
zbU*ILSgrr{y@$X%fewE4fLRD90Bs<~`j4zIcC827ukhAwxX#!DAPA6d`DwJ10H;B*
zbS6Qc*x90q6EdRR#&SzXmp~B{+O&9t4+FvAgr~Hd{N)+J9S*lgp>tb(7l^Lk;;PhJ
z$&4uamn^%4|4^psW&kj)L~R%jW6*jWb}w}qRn8}R6~HF4b~mE&QL%d8R(kQJc8S;<
z?+z<+H+r|)#&2o?kos(j9l%1%V%w`6L>ti?^TiwriH0xwggGRP6Or;--WZ$2viZOC
zllq{@zC5bB$i~Y1Pgu<NIz7)6T<P}6!2|nEXnr)yQSd+b{WItpbMLilR{+&+uwUDD
zkCyLmUoU~2JGrNS{<S9TV<ozqfD!wtf(5oLbU<oUzlnuLyMXCh8T%N0evFPNcLK0I
z@;Y&%XOw23Ox4`N7}Cc;3A(PfkT`~=Gmzcig-DAICj;&EJ9}u~53*7o9as%DLEQGr
z$U>~y{Yq;?_Oz=kakxMNWdPC$xtNsXtx(|XOm^<L(Pew`AI31qad>=xxTKGM1VQdj
zb3h)0pk=%sG7b{h|M>EB5v24}){N7HN?s!;b+v+r$)7^QcW2%Nj5x{<u(?3}MCawe
zY*o1~7l9OD44i<v5AI6DscdSmEx*Iw6~D+;Kj#Yia-Q5v6vX~DnV!l#6c{VJOXM3G
zQ&KJ21kw^0@>8^G(35HLiVvBAIr*vgFL_sL?j#jf%<?Lz?Ww8&Ob1oy4b-eDF*}oC
zvs^1WN_<;(gOPOOSQ2SHA6XnSkyl9t(q$2R4e0e#w8&Hl&O*ga1u-8}MaTLsPC;>=
z`@*{u!ohX^1ZoEWVGUNJhG?k+O$P?lsYaQ_52G@Vls@%M6=8OvueyN+PL}|PR%&Tv
zW0^dtrw_R_Gg<a;)S5`wa7QoUd93*AIWE_;R$14wH0n7lMt4(txa?Qi-Or;CRuoJB
zuvB;VRD3Ar#DUhKm?wDk{_xH6j}II+FS*_Bq`FMnv`9=db23j31fT9qx3>8UgpHtr
zH9H>#)EVEuHFksJh^P+DSYd#+arP>1L`183kdzr9Qo$fh&J6pVCs72GF}9rP?L!Qj
zBd3WarmSSwzCmC!JF*c?F54pih6ClqUanX@zQbsK0wM5upV(NxgN*lRQ!$dAy(H7W
zOrFe%1g_mh^|4x+9z&brQ4*loFT;Bld8ULVyXPuP#?An_!JHUQS~-Al(oQT(Bm3De
zFKC76=-<}gFWp!JG;R}J5WL(GYGWK%NJpPh+(jwny=14*^6NnXUye`&v6#HCXzDs8
zzFEJx=Wos<@-QtH@L`T(!I9AV@>1p+^NRwuBuHPhJsWFsInY^dLF4`*$($RhVK*0X
z2D&n|o%>HS6B(X3D?0%0hq3*-_)q8YI$_|&Pex#D;QoS<cr^UM$kGXvuen5EJaVUE
zn_3bw$V-Z6h$`=Cem3Cu4`!w+kl4W+a#UatsrIPNGFgEzh2B2i@TXfkbv;8EaI1?7
z<y*2woWcM5ec3FKb@NXiK=kc^IK%h%k*)Iu<$3nDAEh1;+B6LSz0BJasonapC|7Rg
zphNPwBCkHAToVQTc^#o<u-8X~%0w`8%olsZr(w_KM%~Nk7b1{|w3j^b+H@O+XxQl`
z*Hun~?pX@q@H&Aj^hY{5Nb3A(<>?XBgIVh{62r0Pt%)|=S#Q-ARoT4lAIy7uBdQ(F
zf$^2FOWAub^A$i&I{YY6bx07JopN$;?ZDFZzN3?O+slu;oq&}AkHxg_$KfR6s75nR
z--2vCo2vmd)Sw5t>e4a*{p_NFgvKP0(#u#g${-`X#;E93Gr)@1DP?u8Oszsy;jLJ6
zmq`!}P#7}FkzWUse%T(9IqR+#y!=UxCJR#RF?0%3TLhRhiLtXJxte6$iA<3wdz0oy
z$;HCn5(oZe!8?L4oKwD=vl^>`Zz|TQdGoZ19dF{zD%i#1JNJl&OZmbyT5qQe%aX@r
zS*9{Ds4J$;(n+gpu+GnTls=4#+&ENvW6ROL`4Bsb?OjjoYjF=G6F0X*Aa^9`Lt~wY
zSJQ!;fddfL_j5J$ZwO)uJV^tWM=UF|%zreN18LG!8*ZV_c-8$GrDR2c96<JkP{>Y9
z>)fJSOti?_MO;~f#=!c!<o@eK=-n0Z_0O5(w5_LQFFIT|cQ?!(%h8+gGF(J(;BZck
z;tPg=(baN0<)vHCqq~~fvo%-G?29H7==gFf6e>VOX>&-0jP8VtIji=3B|Dc9@Avhv
zd3^0f3k-$PD#5>iE%Yj)!-@j$xj&ebz{Kj^$qX_-rhB!1V7XYU>v44d&WEDSL>0D)
zW*v9Bmk}${*Jbkzem-Um{bNwkwy@0{Kaugr3yO#mw0O%~l)jDjdWd9k1(oa3Vq~?H
zwgJ)~0<!`*;Xp~f|N2w|I3mo(Hd683&TSXv+2p0fpWtGsSPRYnt+ITeNTF4VG0peR
zfTHqKpx#Sm^3p5{Pxm_=sY?E-4-2mAabr=#K!7g)IttzR)v#!_ivZ<QE)}iW^vkEn
ztQ~Hvz<)MVvwELka8IwI$H^oOeRJiP`8?pI5n+p=pP#QwaOK5=7tIZms66~}Zi{F8
z!VxelD39tgfb0qkdKg;u+@$hJv@g-!s@PJkE8Z9=y?SlTehd6368TVSC3jx$>&^Lr
zCJEYHa@*hJ&pk}A$bWo-y*=%WNbQ`~59*Jgu$lVc?|Pz~EqcgfO|`B;HW>pJd_ZHs
zF~0yI!6Pv2yP?Dg!Ef=6qAnH8l9GTmxP_FrLe~6`yM1OVX&uOWO&U*w1bDu#6xS>i
z&hiws%e*mFb@2v5GccPHRbK-mD>^_pa<z_%^psVe@+4jgNNI9oWhO{6O5TwL{vOXB
zz{vkn`NH;xr5lWl)~`Ci=HS6T0WALJQedPOH+;dNq)W&eG7H+byYW*xiF82!Ae0=a
zjw31cpUKE^Fgau?dVOV#piXfg5mc@ca}oV`dj&S{bY}?A)~A^}=b1apYSZ?V7Rj~7
z{Lbt?3zR;v4e&u(lK<L}nqYtWbd32&hR-lVF*XnDXPbjS>09BNq9~oP0EfhuRiI8o
zj~<~D_;2mALRE5CuAFyQc#`vtu|ZG3L{LA%%*_=@cLb*s%-O8F#7|ALp*ei~Z)ip8
z4r@l$RfvRk25L`_?N{BoIb-*cWQRVNOkaoUFOK0C=mwL+AEAg<Rl~(Zvx;@)*}wKX
za}kcmH7?`x^xs~am`s!flu_VTck03h6Wpr=uy<i-rb~h0Ir_w6)oe3hx<3d$?Ki4*
zZ~YB`ZP2z}>hb9Rd~R}4wLN4U2TiP<QW?>*yjeh&H+BsYVkh_tkp4NRet0pFX30&l
zduN=FIQ8^Y#qaeHX8Y%IN9)+5<c>4#!=1u}O($vhYub*&F)nHbAd*eJYla2J@DA9w
z**1Y+w8MH?Kz+O66Gy$b7l9n{mXGA1DA0*qnUa0af$!LsL9)?6r{HPQhXIQ=oUlkM
z7LDlH`oYg%kLftTXq=;a-=_Q5h|N@O+K+7L+K(yD^zV~B7)O~P`C>v*1jS-yDkALW
zWN^jbp+==n3BShyG>(P`Q=h+MGA2i+#MS<WXY9{XELpdxQ&)6egVym)pXM2Qh0g0L
z7gOh&kPr6+W>wmA_N3=)^;px{D9(x&F}g9+8bUG83yGQ`rss!7efuuqU$t@cKdC=S
zn~FazHsY9_^A1gJX1!A=UCB2#tE4Wl+E`Wot>`)u?-(fsT=7HRc{9OKBm`P)t$l$6
zv1*rJAKJr291;t_p^*H3N9g+Zpu+~|its9TzUEmK`3O2G`Kz%P6e`#B$z*Abw*Shl
zXWOM~6$;k=<&Ig_&o<MZkS7Ph^}XBT*<Uhm{Pkl5&RC!y8X}H^samI$tvDzN#=*Rg
zjKnli>kru?0HQ8@{SYe*Cjc^Qx@OTs7i;@qeu#JzjgG*+xe^oJ8I1a6I98+|o2-?0
zqOgz1eZ-<3yU}VAEA?|bfI~7jT`afJ-c9|`%BoD*);`?&oHGR-HVIOBq>X2-S)EQ#
z9mRMtD3*WuHW0&fkvJe3QNwI4nCT>)Gei}Rqre0afF$tp5oE>m$FZN-d%xwE0|m=m
zy(Yann1TS|sSgUbGgCh!M*1KkgQ+A8;Bp*{(RNd}G+G1qnKTGu$gj&gD05%Yb`CrK
zs+NnCNs<J|_^DCVM3TgktPSt*Rm3Q5ucyK=X4}?1%(#!%Wo`yOr&X)e^p&n~CC%RC
z$D%_5ma_5=Y!RThDe8tK1u_}`;mlYLoP=vsOfc7|{I_9I#T;SSqbe9k!SgQiUi*?R
zTRc8%9_<Y8g$qw@r0ht{`w&2r@e>w2Pelp+Fk%xSn7LIhDR8)w4;$R+J$rsPerOhr
zfkB^&d^7P5-vC|0#O;{{JCGYR4&H?FE+7PC2-wa{Rs4Cvppi28n_x66h`_*CjUF|l
zog@|ugHtty<Zu+8wQSFI(Zv6fNpi$iU_{$$*xM)~9oiqk2#*P3M1ev82iFNMZ}|LP
zMeYKjbkjF^GgV?ujzKf@DMWYQ`<<-aUSKXSYP9s5F8Rk;1|2@PUzpq^^WP0Gq^|*1
zjZYmGe30={G=5DQCrv1YxB*?edoYmpqHb%HIv@)*xhU09v=oznb;ff9FncJ`>5;9V
zIxF67a}?~_xjV!bAstA3uqms8(yQvF0cLMwgZY~1p5_&25=rm{jN~$jMM@WSuCkz>
zr5+mmJNZOot7TN8{RD81n|Zhw2`d0F;xxeLT1oJ*WW|Yq=FhUf0d9xfh&WS+35Ium
zyPdZBj9PQoy`#@v>CPe|oVxAhjZibOXnb9!uk>fHX$;>{o5CQE19VBhJO=t<U-~y{
zn;IcmH*&4-U;W#u-0sVaCJ28W#!cMMPJn*6wlTTl7BODBAf4!{O}|4(_EP{HE-&}?
z2i$6PYt$LiyV(pHtr&r7Ce5z+l6FvXGw40~sPNn6(v=mgwrw`xNCr3ayc(z?XH+c{
z#v6L;J}X$QpI+cGI@WaEj1kHYQ$mCw0}-M{AR!k0^`O~9+Wz9c+zW@~oY8LflY!0(
zoo#dc)C|-z5*=8>XP9fw>}dho)hXxn3PERnN3vV?91V78w}ag*yd#tCxBdq1stv8f
zPF1~?ipG50+gACfcU8t6s?z5hr}+mw$5KP2q&&(NmEs?Oyc@Hi*3S*EKGpI!2Cc%s
zDZ^36-V|jQC!Whsr!gTuuEtls3#T~zADA-9ZdJ3xNrn}R71`SnGp2x$$P;C7u0h#6
z3%hlCl=T?AhSG4Y{Wwj}E<xXfCy=A)L7M=QjWow{bPv{<{+kR|uklM49xeT9-j!kt
z(Se0|(UL!u;&tJLAY}WO%O}z#`Y<bh51UdVFykYflOp0^r3RWndbH@LOTN`w2BUs`
zeB5<P9Z0t*@0dcn>9_$j5_y+zFVyINhql4v1{u<sG#T$Tx(TbfAR%?s+(yG5eu&x<
zH<#(CuArA(perU<(@e_wncX%!Ybmkmvk8OY4yO(5`3LNDjVU?>pVRC7`HQ^@=guV-
zn$;sB-nO84Int5RI*9Ex54IN#X$uz!yayM|l21jPz0Oa+y9Wn=gxZ0T;uNZNs2~lF
zL*f_b=KAkKaE(shqe0_ai&AAXeEJ&j|0>Rc>rrKyjpka(Hi2z9!0vOyLW2>@QB<ve
zx`#K(P4<6+xbPf(H?5NHWJQSu@Kz^`70VFt0ei|j-LpxjldaYN-GeUx)Iebg?*N9K
z&8~S>YN#3AEA(=55S+*8J#|y<P$Bf&so^m^{}*HL9AirtJ#4mp+qP}nx^3Hb-`n<S
z+qP}nwr$(i?Vf(W-{i|oW-^)7AC;4;N+o;moc+|Z*4`^4z^(7ASuB|N`_$r@ZHzWX
z>@Qt#EX7*b=l<t87nQjP^_muj(17j#o?`}uv`O9a)^nT^zkAM>)n7(Wf354FF$~h6
zdzN{3ELo7Xi4j69BFdO^>(OHb*JT0Fy6C_mnC%K~aNnT~L2w+`{rSlKplxVDfKXO-
z0?P-$*a07)D67bDv4Z~%ZK0o`4Mbi`CKJ?<(F*~6;9Yf!6@%#i*BjUi_~8BZ*6j4o
zP-iEsH{h?gf%Ss5weingmyX+)j<>DM@(|@Z$u!Q6LATh^pEwws_8}g!069#kb(CB^
zlu;B8_gucc#<L-)E9mUw=^cF#N60Ja$~wMA3Ly9R<&xR#UeurZq23nqoZ-KNp94w!
zAg-V<8vUNZEaZ0s^78ohI!~|Ev`dIw%Z!epeK#fQ)iaXAyBveD$cK=Y!&pXWri<=c
z8GpPQ<OsAFAmKQo2XZ(BuoO|h>NP`5L^>5K!K*F?5TwgvLG+$b)kpz@z~1Za;51Xn
zC>L)<9VEJBf@bZng92jhZ-Y{esY9Z7j#{0CN-3Ox<VJMb-Q#9Vh4}f&h?a{)e@qE~
z4JJ1^SIORA6wQ!}6pdg%Jn1D&vT4~HAkJCabk@`a8)irNK^FzUFHV1#q&zTmmF>$O
zZNN&hZht@Zj*8>w+|5SO-@DrqhrS$u`5;{ocUNRkwWc`%V1Tymz{3q9gWBGp%6)iv
z#DMq1Wpfp6_^<Jpq_!6^N+CYtxxPn(vh}`L?KWm0!(&3JE#Z$z4U}uu&Kne=)XuA^
z?62*>D~_E#;)w#tWX?PI96fWvh*ZMP8offOaS7dMHh00=x>e&5v4t%#&khl>{VSZD
z)S;_|+*z2~^Tr~)tr#=c%Oz9tkIbIR4V$T|4JQH3l^<8;V{(kz$xO}&4W@p)6hsJ<
z$$V#H1(V57tgc_%D5*TDbQ9Pc13&eo<mcZ}%_f+KggOBRIa6!(v9p}m7yUMx_zy*l
z+Rt!{u~F-oC3U3ZvvD#1#gn5{3rit9)S4h4pY4?oZd_uF4fD>Kk$x67{zg;n!|Epw
z4*xG{?pBM16uazU2oA^X*(2)umvNqckLSzhrw*ocaAB!@y$a!tTS$^?kNNrIR@eLI
zvcis*KHy>bJ-h`)HmfITTT5uPW1)c%u8?OR4;A(xN}2H{S!lkW^nithd=Zy&FxP-r
zoJCbPH>fZuRp#Npn6|y!+{V9gu2~#JRB)kO@%cjWX9c`ntimCb%(&YIc{r%$+gG*p
z{r2D@$Oq-&ksxZCU~9^{j^<w)Bu7I2T$eznfSCqa&d9F?|8k@y1Lbe%7&z65=_RJF
ztjS4}t<l3dI2z@HY~EJRBgQWgn;d(eX^`wXA{9a)3KZY8MvY0mhUo>~_(%VH&XIIp
zkZs6qdAu9#1Ed+1H|^dVDj+iMu2mg0Y$>2kdq`jdSs-@z^giWi)q>m|>4>VZ-OSuI
zfW+6d${B9R9=sH)q48ldti0H2YvoeL9v|?_h7giJabz(RA3Mn(f7983Tu31Ve|+B8
zM9bxX3NScC;=6|u?ye+kP*5yLM#hm9>z=7c#@7uBAnZk=QW1ra%glSYMIu)_j_*gH
zn+#aT1sm1!p_ja=sE;cHyVQ8j>%|OrfJm0Or{nW=3+POG#5g|dZX7|dutMDLtoUR)
z5i1ZcF@?ZfD3mw_g}~TEP8umeZhw6g<x2g|nri&)q_Ds8lIa%q+W!=iM}{?-6)x#Q
zd2(xi#g`dJu7F36D9fON|3zVzY##`VHoq7T=FL*QkW!iZ&8i<4dn(y5)P)+e3aCx0
z<`T5|?0sxv=@i1BFBX>1h|%7!08WM9SQhg}6Hb+%86SB<G=I(oM&gS+;X<8*uIqAu
z9)qFQ)fVTFmcUMSzzNWlAAn-ayS0cUl&@E=oW8j}oS(^Y;<gX|dY{0RN-*@pGWhdw
zmul#VI`^I+pa8wPH|aFM{V&Xc62O#)R84KB07m-<Oh8BtX7De;#QXkrv=-DxQOhu;
zcFY#H5!=-HiTrMo<35_X!3=l7>z%mhG^J`plNlLW=3g^t;D&I!ES5{i#VaSK0c;Ix
zRC9z<#tQ8&!#fkp;nYZwJBTq9PkscZe@;w+e4;sDt=hi-7zQ_Me9_Hi)c_psseQ`V
znIj@`UWzr#qAW#&;K#C8r;rFUY^0XE6whhq6Y(BaM_Nsb<;h)*^&e=n<GD!%zvDg`
zWLP!JA6Mrsk3u0V)lI(g2wPKfM~x8k8(na;g*YKaYgMixsQSwxXOa~GS#_=xl!Y;Z
z=C9|Gk4=#tG*Qle9j)!CG=ND|QiyXTNj`@-kIh1jLRyn9RWE%8J0N<s@TT=~&dRl)
zaX2&Y+88$gyRjgBo@TWzw=gvSAG6-MHagp%`EV(yq^CqJ7oYUfD)pNR4^>1cIXf4O
zbIYKQ{jcQBp$jFp)p&>D8g@vt2DvpnAL`lj7aRHznJ%@)ojib~V`T<I3_jaCHQlra
zdAo%7G7Hxu!Qs=dL%!(~$QlnJHF7I<*OWPZ=XYnUivEj?2eP)!9&fQds8PO6QOR+d
z_PN3#K4QsO3C!OtzoZY;3+Zcx_3%gz^p>+Q#^j~Jj**xs4{=bJ{((ORxJByK8Tx#t
zc@>FgQ;&WT5|;v4Z#SET0xRV;m^gk{9I~3tHFR;kzZ^LIGmhb|gAeV!4;@HUB(rUn
zY&0(Rkhf~)WK#cdIBqDUswQr4g3Plz^?lt$-qAim393_oP5hNvj~PZ8XT+XB=3y~}
z=bpc8|K5L9;sVMxRC5*l^U*R$e(oIS+uhVLwcP@Y?L`298=6V}ddq|R5O)1T^EhO8
zz;oX8lYknU-n9p0I@VK4;)U@ssQ9Hr@l6B>R~Bj4OYyKmG};T{B~G+MQ0Smx4Zf=B
zk40?TVhdYAvLCej$pATRX7EAl7+{;{7}CDS_E!6L2&e~DJY4qJ!5J3r@p-YM#(9b9
zfwnxkfpZ6Zz6Y}5^<>DEN$C4fT)dMk?<7?;F`Juwr;_BaiFDN%^~`<HB99|mp)fU5
zrIk2D1=__Zbl!w|xG-g#IF>G5py*-31=)#Anz|)vxEf<<w4t_h+9{LE-^z?=rZqqR
z1Qy?eb?~d;^tas$o}_;QLn339J$_&$y)1X;Pip{GBsqao55?2ydSJ7wG<$*4JmRz8
zNh6Z$P*62fw@YxCWVY{|6pEx#@+0>q{%r3PqRJ<wf#jqzMWzef<Jjqiu{}b1cwI_s
z?G!}SPsqO%(Ut^BR2ObXQ@e=MU(<irkj5t^H7$jorumG9i=8d-x4+GcYkD+#(>YZ8
zwwD8-7uQ{-LNV(WBA7@`V<?6JN?io`G^eza|I%hRrJXkr`nRS7KqTWpR%Nfn+f1a*
zDSuL5?BX|kzz+Gj+9ZCK^uUsg^)8sQ?4r~GuZ5Y&Ju;0zX?X0?$4tIx_WXl8*!6Vr
zesdgGEl?`YbX|itU(r9_4a0tt(w(e@KnDk$?|W;OzOO<;nC3g-fyp4svp_|9SG1~;
zHgKUH9^R56m(K}KKzJAJOC~O`gR)@Loz3GG?6{HgomXTG|Ca}6;4w#u^^Bnboc0j5
znzX%LRlSQ%An#Ebh|cC86ej*QRL!s55dMW8zdc>Uw3DUtktU(oajHBhXl~C2Ee#k|
zCbKS4pu^9QgKaDx7|WBhjCnerc0br7C{3Iok@+HjLzW#m<k5qB1U?I#gR4Zthibz!
zwQv7(*d(hqUC`F3G0QpwX+jZLT|-v+J^w3;cd9$~gkd@5aE|8fXz6We#zUhKQOopB
ziW7pal35t%scPZ{Lsc0jv^e_!!W(cUB90wZYox=OxGw)nac_eTM`P|K5jsl2D;kFF
zT^U&yJkm0cx)N6B!zgeBdX|no{l-$66j>TdX1B>@Z!zWH2ehvg8#-OHl-}9fRvxga
zyP5?7bFV!q)Zqqoadoa-D7cuORFTXD6AL@)&&6PBBko6qR*~XksiQ|oIt_?;-(@^h
zE5K?oIyOG4<PR6jXOA$gGYkTW982LH`vN?72XomLYQEG=#=qF)CL?y&b(yN)5!F{a
z{Go47Xh^atGGT(q@*NcGkG7xwdf{Qg7K|q&>g|nOEIH@1CKXy-kbikxar$WNR3_1*
z&q}K3Jb&Y8q!hX(AQ(IH0SDv&J1}ch!!F+3i#HwZ@!%2l*N>m%VPO&UbR<bFSa#At
zdoPcAQQvw!a}npxT6b~~@$K#XY1rs&S1dfrV^H#0<=Bz_LkHvgWK_0ofrsqd5$kqW
zw05KW%r7~f7r|0UtX07@+}V=Cjh`X!)=Mq>il!HfT*juAq$Yt~YzuHRQj)ToY?Ix_
zBSF)ChT3Q+PsjOqC*7H`WhJKXCYVPZpXMiM1Cwe~v;069l4=IK$qbOJw<@T!luMW<
z1{r^I^Sx%8|E1dSB(XI7=#?7fvs*UxbZue;@ASbX`{AQqmd*a}z!ogHeBmKgD`#v!
zp34*x6Oz1wC~Gh5?G#|bfEWHS$~ruA+3#STSAwUjTC}(_-k`O}n*oY$)y@57+NGN;
zUdRA*I!(Mk-kfI2c4NZd{6w{8ZxMAhnLklSTf5YRL^GI$vgVR<>9Kt1ZGc^~z-d``
z{mh_l_IEC8_<<MRvU9~pcShPC24{-&8B%&yGa`6IbEWC%Zzy2tHi~x~mDwHe<%%Gt
zJS(beusK*-EW_QQTuA<$ex-kWN+*kVaKp;=&pRL6(K)k7V93~<S8Kcq$Acv0L(KDn
z(~ot<8J}*7M;DG3+mJNes&G0~F~U6Czs_Q%L8qEs>)uJDt)$IZs#=<jU=Cz6EV)n&
zy}+P2@q*&&nh0RnoRIfsMx-9GFZ30+nyljVaAv6U(l}p*$6D2ja)A3U7MGMf$C%|3
z46-bdsYtF!jzn(57cY`^D}CG3m-$SG->Taaf$0vD$>1%mnO9JzIebrtVE2X>GmcUZ
zp0SRZ;<A$^X~d~Ed1Os(q(6m-Q+hoWP>xDHKA&3-ss`wn>D~~9kW7*)(%>)kY&7<L
zgfOnzuB)|8$VyzRj(*UjL~XqZyS`h+5CQ-aDJAwH!h{uISLwhy^rxff?D;s*`&X5O
zdlvw0Ub$^`zGzf+_31brAmh)D#-JfzK?q;B{P1f?bFqC5Ff=B;w-5-0;GkG!Mk8%?
zejvEj>wqGZ-Ksv45}*z_o611W^IAPZN-txY6+aL&O(`H4m(lT%uwUd}e%X?^q`V_H
zoujKpV7MCozrjag3lM`n`~K8CX#+LJp{n->7!nw_VyJ@Lg?RpEhng_w9T25W4Ml@T
zPYne>AciOT${<D|H!4bb>*FRMH@Y&z?EUXLfDCQJkQn{a0jDOzgdARG5LvPWABl$t
zjLUtdTGiy|i9XRvbwuM(6MVTyLt?deH~hl(pR*QPuZPQkXysw8c4?T#?3IW?VP&v>
z(lk4E18GGa0!xx$-!(NmQUR~QI<lZOMO5d?m$Z?bc6k$alGPJp=jux%nHr|!ARgf1
z$xp&l#{C$v$XN}>`(@v<YAdNkg{Dye#rujBPD*G)PRg(mzcc1RZc=8SkYmR;!NP|v
zNxXnA>b9>ogYQ?zZy3}=y2rIYsq#(A_UZ`h%;ps5p95a2tN5Cl;L4tZWQzA?e0{j#
zxJM{MJZ;5RxEt6T0Tz9MWLRa;skH#s{@~#d)6`?9s4}Xh%?OSNGd`z~?UBe>Q26E2
zChI0S7Pm7DjIAwD-WZX%P3xZOlOm4k=6-`;p^x@{8JJ>NO-fIV*gX)CN>{`mQ=Vyg
zefdl^yrRazC2N!&ewTD6Jlus_wF~_cP0w*q0&k{!{vCoO^W8j_tmX5QirWBIx#36S
z7LRo+$pF#R4PpnIu#29Q0XgD@_QYk6woPgZmuV2kf2O#;{?9C`B;%AQf9OoEg6TUf
zJu+1TiuZ}%?0c^dYd=g8-e4YM<uSl`ZM&&wMeu8pqmxt--p~|sLwLbJw)hU!M}6`0
z@Fig(n5DoTV<DHikqy*k+{plgKQ{~RCFXoE0`na%z(LRSfiW=WhIp;h^`)@g`9T4H
zD8PImdW;UB*Wy1jdHhV5vBCKFiE^hz2z-YFeIR^?d41=Gg8*d{VT89wW&)4Vn*l~&
z{QFrEJ?dr=1ozb=c0G<U8QfDMhFCd1Viy-EhkG@oic#{%*bw~t8UW!i;uvoxrpV93
zcf69$KW7zxqVf|urxf`lHQ(!d2_^wDQ5dI`kn8(g0}gbYpZ5Z-gA2}z_{4OsdLp<l
zFT98199)pxh@reOo@3f7JsrW5r=oFGojqhul5_vw%kAWPY=pOv#GE`tDO8E13^iKW
z+c>!PI~=s1vK9vf_l5?mW6c4v_c9p*{Ve-3M2oaQOc5QEU=@`&-c`@N>IqyF+yGd7
zAUH>j$OVXO03@ev{BSHja;Vwra+<$Ex8}`O^<2L6?OYl=+=ihY5LF#>{eT8DEt0YF
z3k}BJr)iM)tRL!tD6*u=aO=~ceiJrMVXW<(Xc6by`j`O7KR6{f$51qhk>1C1g!JD;
zYTo*vxu(m73jW6726{JiEK=6D0~x#kui`xzpoO|QxOH3o>*5ruYe%ebhVv!jG9cV;
zi}-hPQVQlsHXj5ef6+rdnFLfBXl%-+ZUl~+N&_5c_M#Jp#+++k3pwxa7l{n)NG8}+
z1D1MbM?nSHs|QuXFboyc5;7E6h}R>a%#2_PiDK=~5QX&Z|KRi$rc^*oVX}g?S+*Jv
zI}GSqah5EJN9+;t=uqV<8%ubY46avqb0!mPc>n|6OI_PvNwGW?W-!Be_s)Qr&36{4
z`Tf@Jh~Z(--8dPu7Q}BQK+7hTWG5-sei|6;x*`C0{@G*_aU>H0acI=8X`9z}VocH;
zkPv0rr;Jri?>R1uS|E23wLB&?(Ww`ws41m)>5d{n>iRO-`N?bYKtY;Nj=0%}7<C9g
z2W94-IfkuB?BSrqSBA?E@l4&wKps5WOu~IEt<;9h*(S!0PUb!j=T_hCTu?E%BkZJu
zZ#@r?nP*C<2f^IFGfN^5`pqJd!Af#6F7b<-^Y9W^a)m;IX^3w>4hEE*&zXmTy&%f$
z#1}*|7DQvlE;Tj**JU6uvjEc`nHZe>us0A_p%y1%N|F39(W=R1qesM5uDogb+=$ip
zb-*8lQq1F}7sa7JkyE$8p`<*snl<(`egp$R5l)OT4(|CjSs$TAUZ2n-U&<FxpL!{F
z)f61I?!CQ|xcpo3yhX(k$;Z56U9U<t)BC=0+6QWJ^9y*!krH}dpu(=)Q1o}(lS0NJ
zUC~vC>BV|X%i9nIZ=s*py}Ov(;TuHTD%JLGkZlZPYg^N!o*JB3x+U*^nyfD7!UrQ@
zYGhV7%tg-A`RXz#0BqgensuxsE1)lD&B=*4Bkh!-AMCOaiDzOg4ad088@p4<rU{$M
zxWf%s&%A>R=!ltm_aX$xYQM`yCqqU}FLtYsCsSBvP~vd}%_~Uc1hSj5CGlwM2%-a*
zhQOztXE7l`2E?U7v==BTYNu~u9dr$#r($Kha%+gO(dF6;#Fy+c-=!LUZ}8C1=4fd;
zD0ZfWKZqWSI`Wuk$neob&PG}!49oZquJe%-rsZf}ZXMVmX30&ZYQ^>qt@D=+Ic8E+
zaqxg1q}wZ-Sp)=<c@Z`ugbY~ytBrA)LSNe2L3h8Of4<K$TMefvJ+uaodmR^$$@r%~
z@WfSABwA$Q;F0aPX2&*jxuTMA8h;PkZo%^eOLd{va0O$r8|2?37>;cuwKW(`YWbCe
z7vU&!!=`E@Or?Ap{LB%C?4Ic#!JaBh8lpFNnEhlY*$!^U16)X;XNrV>3?DmCuhGy8
z#qOf!>Nc8%5DsKaC*A?lzW>euMglbkuu{Xas-B_|n-7&cktlZv${vaL(MopB8N_y!
zd*ka*+wrw`MrP9m=7|%dkeVnl_B+Rd#{R=P%enh!8>TZn_bY-WS#76`p^lGq1M*O4
zr^et8)^!KnWGxQ9MNY-QnRb5++)AQfs>PN6YV6ReQ4guA>tS4Uxnf=b-pCC22AP|K
zPTIjtTa+x-a6_S+4}XX|-gybO)7X6SA5fyd1l4H&k=Vu|HXEL_J3Ap$O#4&?j;RC+
zqwkeM3Z4JD{;kxj72{0?WIsc#w0fK7O`=*Aqbsv4%8JSF`STu1@d-iLO4oFYH7y&F
z7t2R-=3QNe<nyRx-$fmOm*ul(N*BV|-pmk|5|Zr{|8VEAI9|*Y1+PJs<}2E^qCoh$
zUB&!)=x9A65V>yu{}t7fB}i;6<Ys1nSiFEmZpfoUC(aW*Ulp!ILD0_$cS$(5A&V8A
z$Z_206*OULBlU;3N0G+SiDG1oKPTyF|H_mIbO6Zzm_Qa^9V$LRZD<iWT%e|$fjgNa
zWnDDV=)iHMA@b<wAP>n%#7xpky*XTmR!6OR#lJ0zU;ie@zu$r6!fJA<T6T|1`sK{r
z8B;PpM%grOD^~3BlYwCnrA@b8raIAH{rhieg84goTQ6fC`c`a8%eOqfDa^&pJXFVR
zH#%Sinb2p7QOwX;A#*JdXSk^lXTA36J;;u0UrH^~lAf`Egy|A~thPt4=GBx$URrtN
z^+T|w5v%#`CCqAa>933By*8_L3+JjNJK6h9tn~YU13&po+G&)|wffEV`FI-lar-qY
z*r(^@b9)CgejD$I6P1}8_SzzM`9hhmQyqZYa$K{FM;j%8$|6vfG)Hdi8(sk3!xQA@
zj8!sRFvXdSrlVie(f8*gOEBvqWJrhbD{dIB^_VsnIo)8XHr6!yNeKAqP9f`;g|aES
zq2ZIm%Yz}lZho8;DAQpRWHna#b11L$woHq}@uv8)&O)te)}8%11opJ%=wF+83pl_(
znN5x_CMm@8BZrDjzCyxFKj6Sk&jtLWwfk1|sl5OV4}%hgB_)K(S}vT}ouXYsY`=W@
z1sQm%g(~t+M@Mi*qKPMzW~<pM&zo{Juo&P~Z$=lChRif4?1smg8O1e@(Hl-EU+7#&
zk=_71)ZqaTy!5XpYg;a$R378$qcT7l&htEu4lw!jAA?>ghg#u3{gqH03jMWp_*hSB
zO~tRWWlBvN7nRIK*Omy`BZl0S7Nfh>k1$yDJL{kj7@G9a5fY2&LzMiF9)xt5qm)SY
zv0t?o+m6u|r0`uI-WK0uUz<w41F9WPAf0S(ho}bX2w*9AyS9-)xsnK!!{q?emoTJI
zbGO^yc>Cy4q_y@{yZeFu-hZIO5Fq=Dw#qXX_cMNQ6qsK43~^sZ+%|ejOO`8+JP^P0
zV*wX)i{?y|?6HG8?JG==Q(=3<kcOD-ZH;y&1MZeY5hyup>Iks`GdxpAt!dEC`T?s;
zT8-pElyu&!o|)^hhw>C-lm!rRv%{+mzU9W)&{l(Wv6e58Vkn*dR{t8ObG^c(?(7;N
z&6YhElE#fkR1eiI9&jt#1fLzpH3ZSdliTNM{WsyWNvD^WxET8leBkj5=pQkDuDMM^
zjeudYCgZQcp3qUHb+`U1p8gpLn_45e2R{lhx2pYrUZ-E{ndlijihluzRTnTS$4%m0
zD!Qs+mCDE4BV?s&iNC|_!zU@6&D_d?bTGbvjw#aL`o_1ZK=T-PLqpd$rK6OJE*Mmj
z0~PeJ6ULDLwF`S^+o+FhcXFWZBm6rq>7m~!E%55Lp*g)4Kg+$qx5{To>6(5@!Et(c
znX0;H3Xj{e>T+LCPKX0cf01J?`2T)0+kDRG%8;Mxp11yvs_uxJch$0Ql>1~cBh7n!
zW&qdKeNMzlEczBwpXy2CKi~4UiTf5+VIzpU&MGfSb%@JJYdqx*p|_tkty$~t`rPbP
z<qnJG6-G2lRpF=Xf^Y8<&v_p=FLJ$5D!@BJX_KgFy5`TljLHW{Uc^P(cv@w30R$GW
z=03A?9q+f7A!lOT%=TV7PanGMG#zca7}Tvz*O~`$=L%O&-y7!b^#LkS=Flc%$)j9Y
zWFZt3O9|t92WS}+L~F;24X(0+8~T@@YiGMe86sO3{C*Epwhb$4UVL<egf181f(Uds
zeFVR=S=vV#c!mLQ`I?9|L?sA^=tmci1NA$F4@3`BuVyBj2!~!`g=K>><4W_LZM)Qb
zb;maOwH;0tcz5@&Of(QKLD$n7xp6_y=y*gHq<>Y<-x*mbN(Zqe6dD$VHDAU|sI^IK
zZnarcC1<=ASSB?ZMrK0AJI;ho{94pd`Q{k76h%CmV)y}@y7{!;O|n;!I>1jd;)xd=
zs-n+L2R|}037%qjdW}$F%W~g6_S}Un92`8B8oV}pOTjoD*QH}7k5(#<U<^OjTY+sO
z(-gA2D)|+tlGouE{v*MzG?~)8$u7!55_=38gvA-{DQC%w1t+bcC)0CT+4PwXG<018
zLz<=hl`S9wzqxxPox4R%6?`4aT7Z#`TEX@Uv?3wo5oZJghp`%-f?ZBBuzaq6AaZ|b
zuHLQ-yV}^2EV@Ql8h?m!6eOnYKgyHa<bwY+6ar2p?)uQuCC_haFp*^k(Ka)<hcyv1
zLnnv5Ux>!U9kxuib{fgIF<Qeik*!~*xW>mL1f;4c5pT~@cev!VylEr0>n&wvv!~md
zfwhlTHkDr)sGfN1`0G8@<ed)&yc%Kq6(W1oVGkaa2zI9YV)~;CI_Pu-iAQtd)h~kG
z<<rsJgLiw(rspDCfp$jzE}k&(bgwtM=Cbj(z~^kT3Jlb;_%76({pnQZW`H$K<i#lC
z5&(O_L5*2}Zs;??E*vv76dH8)a(!=+FIU3q>e6ksS8PP{fl_=`gg`EEAirLr9$uwJ
z5_+}G!>D_v!NaCe*{@kyfA4Ptwr%!kM7+HdC6kQ?4~i&tX(ChCM?W%7v4K_4LF0`*
z86BvVF=Jp$KDM5yy2AjOz<DM;Vw52G2H+7ly~dbnAhn`_s!akv?3LFfx`CS^kC?>s
zfnMg8*H2kDeG~|)<A2lATF}$W|632zUC>ggls&Yy-54ln<{&fSs}um**}gy>*JIi8
zSF5I;n8P!WqE+@L%UnfvAXrx)u{lkq9@|y%&|zlKZdF%VDp5F?%sG5al@_{E9<Y|3
zIKhy9Hom**1$zY<3l#x0NOzX*;=DXSB?9>$ykRh#SRoh7K7>Ah1|#Q@PN*`Tc>mgQ
zHST$VuhsW@CZMRTFKgD7`7M(;9aUx9!B&LH8L{CB=}<J{U(%tu1UBO&yH3p4j3b$g
zNU8>b;ea7@d{3E>_iKfj^$U0yAS~?7r~oDrU-N77Pc3XOJ&-eyi%|dNPfhW$X7j&%
z7%Hw@s#i#S>sjViVp-fxjKD>Su89r}Td0b>MNzMMlaubRLf9IQpfE5=fi1cCC>8hx
zta+vfKU}!s@E;dWC7r1>|8SuQz1z&~he@Lroq9MPk|p^d8s%Sb;aNbxr@J)NI8!Z!
zl#gq|rd?CzVoE`>phJ-GUn2xF7n&rqBI^BkZIXxI%c=~w5U+E~HWi9j?6`O6Hdh^r
zHM(}F9nNBdD|UKYlyUagOwobr+}&7wejmRHf7e*zlUHL@^nOOGE%meaCGH0*JD|V#
zGED2gIpe4kK+tw{z7GLX%z9kcJ_ygo^)rdk=AgNZC#7eX?pD3}L6?!E_4kZ@^P4CB
z<G;p3Kec?Yq0EiLEheQp)0o;SYxf^_bNpyYka5h)t)NxeGJ%$(Nw@~nxrhpk3Pb4~
zeWed|uOB~5Y$B%_YGqI#X28PBM8GAyMtm4&8JV*dL>3+Q2I!?HHM1QqH~q<0bqj{f
zStPQXs(laew)O*0#7k{XI2a|du@0K$)3q_dn9UM%K`McKHd@{cY(;-hv|)1s*__P3
z)Qtu#G$G-jaOA9y@+nM<e0>o16fajVaeqPI;{r!ST&Kxqk0^6_)-6)EpryfIrZxa)
zKpEU20ckBe17ySdjq$7W2OnZA0GFb{ou=UKiBzJKScu+x-VMs6f}hytDjImG=Aur!
z8U?pMsmsIAw3>|Me-O}vO_bf(34~EOTg=xew&oX^zW8V9Y=dn-|3}p7zO$k8)eU$w
zS8l}(;z5dIgv{m%(3)2`+9z+1{uD{MuCC6+>I76@$Z37=@DdAfk2v9(xurWtSxhld
zEVWVG2UQ`VRCyFtv-$@q_nJ-=c%DBh&g*)%R}w;*PXTXxF%X%<PY$Jue8qW`f1Vlu
zC(VMPK%!|LYY?C+5FjkA`N~Nd^ue-o|MTIWERbeZvWyJqeC}lv+`^ADAIms<9?jes
zO$S(GH&?QJ$!-Y_a77$P$zp@%=7cTu3W!4scj?$Ab%dY>@-h0xdfKlS@M`0)<a2o;
zN+VpN5^A3NVUcIkzUji?7b8)=vrZkxGH+S=)6Y<wko#t7Rp;O?{8`g-EgF>rcaa{m
zGg7Z>V2<{7_lc(NZW|mC0^%(GA0Pz$5aAthv3}3;_nSG_Ql{p@4;22#gpiuR!x$x?
zm87^dU=)+#e?ZCLa4zB$?NzO%mO>)`7ZW1eUdw&_^eplXA_~~1b+xA(EmCyKu5;VB
zVPmTx6+Rixauyg=s>C^10;tp=PN4wpx>_~~_Lb0$55YR1UZ7=>X%Dj_#);d^`gXMK
z<JFIDUz;8~ajYD-$0sRejvp8e;2|pnexnRNVo?Q5?Us@PhUTS$CUb|U4&I`jl%VxF
z4svr)rqb7cEIh!SZel45<Q!-B;VM-Kx>~uEi1O`2i`CUrk#LrCDt79C_lCTXK6M`+
z1nff_={zxg>=4S;FLub=SFW;VYs+8YI%Z{(9h#ahuSOp7g#rE##NF&p{9lH7MJ^74
z8T$*HF`xG13A4>_7JYPq_=$hq$LrVlk_u7gkA}r^5}~yI)pL(8a1{aOCUSK`*YFnW
zn_8Qs06GlDm`SXZ3ISjoKz{&wU`5QT0R`$LWOn#T)7i^I-iNiLhtBwz1c^PKeQt1#
z(|FZFY~|;EzVv`6LCVo$6=}M+vI&0~%YgFJQTr_<qrluGg4&zh8bJ`Sgy^D=o>d~X
z?sp&RZ>Xc9*fiR?8k)tEF&F1cIxl6SuUUJa_lx={?(vg}X8&tDKq7qmbsqV<!`g)o
zpiPJau(L@`P=L8swVY8h;lty=G{3e(YOKXbwG+AF)8P70wx;fJ)idYJq4fTrf%>f|
zSG@IAlv^>8{S!`p6gVw^Ynhkm>WHDc&k0T6Bbo0~mJFrBcuV_@fORx`^sD`nZKEY6
zC!<A*0x4bsN*E51Fb-L?BGj^Ss;vK3`DmwW)mY#~{Bqi~yak!qr5JsFb#?gvfr6vo
z|Nl`?1oi&~1p)t87PL+HVL@$K?1N40R}u%Ff{1bY5l&avAcB~NN59SeR^)%Rmq_Pc
zqh2iCSXeKhTswDBDn`d_;eGH?S$Oq<pOs7HGJmEto!;bbX*XnRjXA4lx3FMY>u0y~
zOOT>StKrjlCnuq)>y{-JX`2+5x5E6#0n1Q0pXk5H=f%kz&B2#1?^yEZ?ABL}cKM3@
zvQJYoONqMHB?zwR6_dNyi(SO4eLcuY#-B2%i9gowhA^3BLr{!QG{{P?S^v|-9n(|(
zAd*Qx9`3sEg;o-!V~au%Bg7Z~|IUM=54;ZPgSGNs1rKGS0I761FQay|Cu}xBX>*|{
z)K2jML+Fi&I(b=Vxf3bL&@8Vp8BZ&F?MC?_zU9TXrgG|8ag7f@OGQ7c2Lg?(w~`&s
z?p5Ev&YX3Bg^5+7(DBVs8KRS&FP8q`Nz?rGx@x<p3>GS;Tr!%=QN7GmvCiuCx%9G&
zz!7XxD^PMK2QUj79F+oj6hR25e)7H{?SCcgyjm7Kh=DOQ*w7_HUjgp}#~RT7Qr|ON
z1cM6L1=H)yW3#y8xA$n@UH1_C<FRhhoEC;^O>xIh-}jh<O64Clg00nD{V#W6GgID8
zvLN0maOGa}JFz8zez@6xx7f_ms|rf|e_S6O;{xD=@mIH!syRxbU1x~6eTrp999deT
zNh*Q#Gj9)DvIL4VcTwH-gK7jHn)X@_l5#>OS*#6hW=|s|V-M9anepq)tMd9ql2OGH
zlu^ga&$em;$RQXTorTzEMy3w0RERgGBAH<rW=WNuggAb#_kO9NFC0XP=0O8Mfyg}F
z$~J)7iMo4hib^tj7^YM_WUUM&$6)`sTR0bs_)7&%SN~qbG_VbSW4mLYI}7i8_AnqC
zEkXFtM6~<qOZ)m!UR1(9t&_70rHH7nlK*A%@?^2a^nI|7kvuE94mCHgqU50dN?AA!
zM_Lky4`&V+e#&-4EqEj?ZLgB>_St&e0C)?mu-Px){cER8n8gXk0_rS(c4kU#O+hX^
zVUSY|(35J_b4dCFe4*S-c61*xy~O;=z9TG8B@&M_C<`LnQ7-5jZtQ9Zs9P^>oFDeN
z<Mc*a<H}R3r2O$%KvA<rz_rr;06dvI-%QO(bs27eMKMb>46U=Z6jLqIPkWUDdX0&e
zw*bce=I!LRX9(8)M*w%=9TxEmH^UW;>v`&ZCR66d5Sd_Z`HLp?R^XE^&}SY>f4XgD
zBQZo;{X0;_&1U#m+W$dB!VPTcglE?bAj1f{#s`GAQzuOp9D?D@AK8*UG}(Y}FJ~%;
zI?p53wq_Szs**pT^VvlPFgzHV%nZq4iA_SNUk2&+?kM<%#AXq~UZ}r&ysDp}eY4J8
zuuFA{oaTB@-h*6&eXIj%Yr1vVj@RbuRf4G#g_A4+rBT@|{PBn`Nq;r;f6(u#7cBGz
z-k7fktWuzTEMXNiaZ9R5^KZND|2RZsz!C#YC($$4IeX2||L^G%`s=7K#-3mML<bk0
zBn)0v*PC8)>$V*-9uo9nIUECgL`EdWj@Hi>bNr9Tf<d+a0pkCi5959rPgcbIpM3Zb
zslVq))h}t{BKZS|mo(kokQQW83ja?-c(D6p2muA4!~ZElXm(f=O}~_MVqan4q8TQB
zSfOTzd5yqHD%ieXGVJ+g!4$_N5|rA3yv$K|EFHCjOn9DG*jfqy!63xge=s;2nbRu6
zkjoQ6sZm05VMIS0sIm2)Xko}*Wp1W8o>Ji9WIHJSK3T%<fbe}6`^{fVU!YszEDR|A
zM^E1_<MT6Pv%X1JIcUe7VFqyF)Fw}(-|SVI>aG-8B=mwlMJRkBwH#-S#<oN=$1n$l
z0(awA5$*PWyl)P;#o;Zps4(<z5~Y%p{KTIv%lqoh8;Z(|KfQCc*Wor^KSbG%&pA6-
zh3L{z3!~6N{R_V#-bw{%M=~5fvH<^^`k35)pwI5?2l^f-nf|q5Wmi*mqgp>%Ua%Xt
zwgyMJ5@`7p*QQUFKBGj#sZ!JXq2O8cvBlPRdi!wskMdbjh4axBX<5I`)P>3H^4Stf
zf{&X5jgD9US;pO+dO>bKI=5r$2?rF~(FrJP3X-w*4Q7ZXb1efP!52qFoY`~fo-`4C
zRZTW$LfoCg#HpjuVps56$q&{^?rGgww(!#&aWoxe2a=tKd&+PuIL#I7>VAVlJhI#|
zzK&1|m8MKoIgBbuSz-!~c5z(AX@K-7ubNQZvdDaD-*=#r$vjd`{=ncW)7E=0&$9-Q
zbe)cXVFg5V-!{Pcyg6FPP0GvIB?Ws(seYBt?o$lRC52kmz}0H=%eX<Cxdl+N_5CZq
z4KaB#vz*1Ao;&uf#8&uss545h*Uf>Y0e|IsMU~c5p$`AuiTQx~)CXrJcI7%m1w@8c
zJUd(V0m13AQ{}MasjJlxN|1Z(SR&R;KKX{gFZfN)F&IF$fZOaZ=g9_4e4EzNLROcs
zyz0P2_fUz~p$$eYQNW$)e5e>Yojv*q=Y4uBP(bS@V0Ofq;-AD}JtIDP>GIcmQVw6&
z)ecpg{)hyaC}vC%LN||=Hc%DN5rxF|P!`w_7qOK8zp+Nq%!eZ1$U_Ur2c#J@Q3g~M
zwP&VfIvzl<U1@f5$51SF7$KM&eoC$h$5Xps29mK>%%BvDQF(A7s0i~2fMQ9IOP~Y?
zjETIA$Z{Rrx=^mpSoSN$?h}JLWqMo{fm-KL@!V=jsb!p#yoNS+XKCzF{t-tDb#X4K
z)m*!n1aE{kVl8ZdUtQ2@TlY_BJJvufHH~8?@f5&>Nz$Vl%sLbh<3Vq@a%vZ7rLQ(z
zXKY_ABv8seuL8PdEE)xdTc8JBIdq5#4o4nsdMYNHm~Ta$WJX4cM~WY{;mTv{a!ye)
z=^O>_baMdYROeg$Ex<87M&ZkK`f%8{nVlw}BY1Mmvb!aA(tVKS`$7_6FsG3)3S{5Z
zngft@E0Natm9C)hl1T7I#7WuBeImW((bAa72f~Gz$4PKN%TGG!w^cK0R~pJ?g^MQ2
zyT=PyBgYuzouehFNZuNO_^W%jo4`Ft=K+)AO=C97k4S(s`~XBy&*c+g(F_SgpTF!U
z$ji;}nUr%hMJ@1x*B`8V`$u85`QGsc69};L+-h}WdSuR}*9jXNhkI44dqh#liBgTD
zr{V*XkDYt}Fxm^n09iy{Tw!Km+h%gDZcv5)hP9nK8`TB2r1rLarfO9oNI+pO0bgB3
zErN{io6VRDxu^y_34HuJz+J@_Z5Qu08=_^;)E131(NbhMxJeb!R0jUU7<EH%92&6t
zFZ?pN@O^I;ucaAAl$VQlN8T1Kw9O`%PelP3hPeVW1QM3Md^t~$S6Dw$K{Yp!x+()N
zQNeFJLQ`1P54k*%2vn+6kYkWK4_TB=5PB)j4@pm0Om<S8GOxUoHZMOVURn4^n}RYA
z#QU;O`VX`!xWg6DV<0eJZXk>SJR?9Jd$K^P6zInHg|+n;gLG+co(vB2&s&-@Ly)Nr
zI1&}elC~6tllZempjFp3x<#Ouinp5c0;OO`w)IU^wkk(0c1nG(cK?*FHji5TG$_SQ
z>&L(ZalyII@GUU&;1YaD+N@?plJrkiFkyf&k*}(Qo5r<=`*68Y+x31$BA|U*nNc-r
z`ZCeF=-g7o&f}o;POU=8j<y7%63yu5fN*d>z#wU;BpX=XdU*2y7#A8cWI|I1D9{JY
z7f7)GzW?JdLgxk@p$Mr7Ee*csl==xzYmh?;G?6&eWwgOdu`w5sBB7p|+YT=z5(YgD
zoKg0l27Wa)H<1(g`u=)uR{%a`Yp50^$la;eoO(gC=D6q19o*{0puGDH9_Vggc=CO9
z7}6D(Noe+cu$ybYGpIDY=l!7nDTH8fsJjXfq%R3`3=9ykFW8A9FxF4+io2`avc(U)
zV1}_l@sb5?886j;tYYqB;d?2Cw6YDHBRta)`2q>K-Z57aeva+E0eq&n{8L4O(U>dm
z^L!zrcqRYLHNu1L$Gs8vX@!7UYqfAo*7`3;q#0C+__LXg`hwA%2D?G42|0lhiVT9$
z_=5H#r5S+#w1q!PrEg>-5h!vc)e#S=hTm4KxGZ%CQB?X**I~0c(|)`1b#y_`HeDvu
z>yN9`|5~8&&S<>=44qlKMhw-@!<jYuHg-nk7tLE`CbowZnp$H9d90kuq6SSF73rAy
zlQ0Dtg2Bj51+g!e3?5CTA|POyWLOmP8<`kRu|k|*Vy(454+tM`k~bQNqBL6shqPPf
z`i0S%#LmbcUrIz)HS1Pq1&1E~d_=x@5X2OCN++E&Io}KeV0@E3um^_NY&t(?;}yti
z^8E9K0k`DWUi13gRXI+B*?ukAqrZ|s$m(>cm(q!bLkXQa+FH{eaeu26DOFL<BOWbg
zqGmt(@kt@p2-}kUW|PbuuXQL{=hU?&yLiU9IoKVC4%!LD6Hx-I7<P2A>!dY`3#HOg
zpRb=PEW;=RJQk@l<#Sm0s}2Y!bM;k5x^$Iqx?+NXWI>7u)^1rIj4jf_nV&nU%tk??
z{Nr!__QJ4?%2uJ?S_pXG<pJJFm#1o5mbe#i4yDs<vSR#JOPafkxp{H)<{dS<M_DW5
zZuql2EMU9eD6cv(JVg5u^Y9&^IsIYSr$E6OBV|4TfKBJ(xYWV7Ja(&;qf0-X&J|Dy
z*71=3!U=;f(qTX|6_U8k?|xEs90f(DpYa0q`I)RTCxYy?ki#1)asKoh`Pt7V@c+~?
zh&0pq+2hB}ceq^k3)s2zTpXV-s@5-i<lW>g*DT)EI3%g>dKLTk?>e4MZ0g{tIR)yd
zg#a^lfQyn6g5$x*pI7ftTOYOG%U7h#{uEfsI_U9uVro^|*THA_nu9d^q@}p%lj9`6
z5kVe8DT5CMZ~VF7!K?zQr<P>{ypz#f=Tt$e99p*HbSOH7Ohkl%agf=lPcga_@6GCz
zBkld&2~v8&Hr79jD#%ixUH14Tz7aTD^p?B2RC~ROmg0(ad(kbITq;y-ek;oe>r`W9
zBdWue4d&EKq)OMT3KE<o%kowFSsDZqgdG$K5Ms>4T_@_7E>NkNi>}7n)#}eq0v7`?
z&q4og!JMXl@_D9sDEz36D&wZm#ak;dc@^65ZmWuQtBLA2A9acafh=a)Z8kC?+NYv7
z??FVIc3s%ArELVJ4$5Z*YumgrGKwc>hwb}2HaYA}%M7cozaF#Y&EB<)DI{v*`*GOB
z6?gGxo-MJvxTk&Wge`PPLGKAxb?fLICTzh}NLV>3@#Oe<BMD`If1g?4I1OCwNAojr
zICfPkHObiZjF0|t;+uRE!{<P02z6U+l)FFpnV9tQS2Hn*RLV-ZSl~yil<n>2U;Ky_
zYT$>2NiPb3FxPb2Oocb|n#?wMpci>abs~aP4N3Ccvj_FpDb8|*!kqfoOqPRVg%g+~
zXiL9Yp{u&(aw1@%`Ihyu+VjT4Rb4#{4M(==4w3DH;F_PhYvVNFHxdOEn9-WEfHU`X
zXrQf8`7jjiANdLFG<h0@ZOWF@XAfHyW#p#R4>QSTb=}5myE(F>fcra$t*<50%U5Zk
zoa$=(IsGY|#SXdlzjxO!dH*Ps_<h>{Wh8{#@w0766^a7z&wwt9Cv9Jn$m=+j%2>$^
z6{RgXZtPH<wu<~yq~!d%pvXsQQVD^ZA$ShESlLZB(tEJdFl{dg2M@Hl+oe9~L{Dxf
zcB8)pc~(p@lHecxeiHVDh2Dk&rb8w<0&)lc`F_VL5Sl%E|IrIo#)2M_iOe>~Ii~1}
z5zz<Bb72NBbm%g4%SD4129-BcIv^`n7YUebHEGQ)ojPc$#x-e88C6lse|ODl(UmlS
zR82;$$+L!sL`2Y=UU=imafTy=|B~l)Niyo{TYr%zTIlWuUE3s>pm~OpzkQKVCNM#9
z(eP9xsCQNrqYmWnlUtD!L+kajOe0w0p-?LlOFRaokcN#X;8u79Hj+LZ?aFtWd^zij
zQo<xMC8w>m*^6&B+PJ;PpM1q~^d%;XMlROVx0yhO{^=Br6wR>c+=SBzcj$l~8J)Vs
zVL|K^P`l-i-72!H(+tZAf8fs^#Dea%GiX7*glFG=45?IsyfkVx>>;u+rDyasu`|UV
zzzqjjtTwftvWM`-41BWZy>_X-P`9P?1d(OtQ_`|cL_G`XZu2R)iQ+ZFvSIX9wZ?;v
z!V0yAlWpC~Cg5WjUQ*Y7wE9F_X}-@wr3vn33G}&o><kGgP+{<A8)l8inx^T&--hP5
zZqljJ%(JFJrz#?eZCZBCmq{vbVQq<P&|3iBbq58bwo28gaWYBn_~Eg&Q&)<-d;&s6
z9TwmeQS2-EYb1ItGh$KVzkUh&zTB}A0;$>RS3tmm{Tz?&u&AwhDH%_4mTZ++?2o=f
zNSDr8zy1^I;wVcBA#VP^Zgw*SzffSCJdsDOZn9MB+dZ_XFeW^Wnof#Z9Lrf30GYp9
z4?c}<kHi~@TL_TPCPQ_;jN$xt06r_D+-Jj^ncAA4j^)UynIv@DlLf?YQFZ)tW7OFs
z6PaskIpFtcI&4#(9${a;&Mx8>h61Lk5Yt_s7}*?e7}+b|u=Z!WgWrg-tC<_$L_NVH
zJ7d&kC>Iwi^8(+ZFEKk5QqOT_06`Mofm7sPGU#(`jQF*bS2M2X@>{}AR5@7!viI!4
zRr8ht?)O38c})992g6<t^4m*ZgfJn){?1n+9HjWAf69-Hn$C9qoG&?)HNb$i8~glN
z?XsZP{ZaoKjycF3ts@sjQ6qYvcYW<%S6YqK(k>ytaV<kDE$*brQ~t|*K(%YB15AOq
zl%WAnqg^R`b7{WQYRPmqo^4otF_VuT&m*wtagr#x;w&}hi)7ZXlf~(hu{BDlI~E*}
z`VdV08;d-rd<~JyVle0?(#lrK$f-ODVu<$v&^o8FkdQPkA*_K&M_iIKh;s9yAr6HW
zAKmd2wuGr?>Oh{ZSEispfD#OI;n$5e5}Uj#gTuvh`&heGH_5NGq=5+h&}s_S6eQ=2
zF@mCeAZqfzC9Z?uohXEkAT@HN^c7m*@QIhA9D<;DTA*hMyv&RJHvVp|Rk$7@Qf({<
z#)~<<n827Ku$^EEsRjvna1h_~4|Pn_PGI+)sx<7<VymQ}?$a0&0Kx1A*?9YyL2`ZE
zXf<2U;-A$6Hg|(4uqpp2;yg5Ujz3R)f|m%?=u<wxbvW5EaIZ#V6#c!R$O!(*vAS_o
zHQvaBI3(BzP_KEMq8C<kmC&{>oc(8q60WH(f`TW6d}rNmNb3-;(+f8C$CM%)j=gG>
zFw8yBqmT=ps|*kT@@2{Ebc$eq_>TJi4oqxU&EK%qD15Y|Y-vR?_#Zw}|NC+#(0#VB
zT^-n*wTJ?#p_uSusrFO{6zC;v?_eVM^Kg!_&X68Lw5)<{5gJ>}Ry`!{-60hzTiZFg
zp*Jf5`x>krJ1J&NXbok3e9bKIN87Y@7U+x`_f4HKr<p1P+Tj!@rsPiUpYPjWre5bS
zU+6sv)byQPhl=sUqz@-Ju-;)G&8J^``a4#S;>2mMK)7AkU2WhkZ?gjzDw%Y4jaugH
zz_lK>d3pzpnc`Y|jcwqr6kY62zg?pntc6)J?Ln0K&)*#<r#N5OjxZe0`tam-@VldU
zXp={W2$|ObXAkUD-#4yP-`$-;IT`A%<G<c3-ZX_|MLnGhd>USaxC(mqGNp_p0=&$l
zQ_JqAUY&|q?Gt(Pbtfz*9G*|hfj?6wHhL7WoasAsi{u$@0awr1TRQsQ#XD`CNGgV%
z@d3LVp?vMUf^EJ5K=A&j@F1(n{X9?`s;-y&=XlxxguA5pqrn~e%c%6oLwY|_^j_Cz
zE@_UVQx`P_%a?P8X_~9i{*Onl+r8KMecWx<>oY3n5m@v42ef~|nT4z$sUPODJ3Iy}
zerss_%d(FFijSKhH#xr4K~j0GtdKDoHS1xN?lEPui{DJ<J9PII*y1d_cqoIvX8*A;
z9=S2_`OHS{(c5F^7tS7mdML1c*hIxsb66LNtBb;inp9mXLj)fAb}ma%uN^VVm@H{!
zPa!;_L-qSSCcp0gRZJ1m1jYos`6Ui!Wy&hvdLvvOw`<_L^aMM>j;?VIt$=j9WNuqs
zC{Ph=9=_vh^QJr*h4Mn-{~X(Tqv`3MN!+KfAYJS5r6Pk2C{R~RF1VHnQ_hk43Y5&G
z&3q?kcIU#dJ{HzhYp%;^2;)A{Bs1@##jZQ(Ztm;mNW0FFSHzmP_Xtt;UQx34jwA5r
zv3+&gV<{(^!1w?(Q>lpgzfpN>N)LfE<w?_1fPaa3qb%7^A(tfCNNs8(r<pCfUqRcn
z!97#AgI<G;f;!4{lgI5_Ga*hM!vS9?2+IXUg^6<9(Bc!z?<brVYht>3rO8DIQL&n_
z+rMLV{NCy3?Y7^RO2wB55lN&*a8~(XA`5vcX&EB@UAF*4?3R*@#(P9rGYMg;q~sC1
zm0@R+!{u&yY`arTZU;F)XD9_LvsE}9@wwVV_Y6f4oMr);WJzm>4UmuW?N2j~^T6H1
zuUuDZ*l1$t$d-z~vi#G$&9d$@vw4SZ+q~|XnAA>85z9fv<IP234{C}4$fRO&$|M}~
z*<5KyfKGteo6-aQvL*fM9!l3hjOxTS3J&hf?f-HR>ll9CG~k|q#v9;`9$(b{X9ho=
z|D&t742q+Rx`lBM?hqV;LvR@g1oz+&G)N#g3^HgVA-H?62PeUTySoR1yHC&oCNMZ3
z?{}--x^=t%bXCjg)4flfwf5R)Zyslqe@e;4YMsqYY749+tfd?>dGbPK77aIvMh;A0
zROHdNr6ex!Lo%{IX>SntMU;~V6iD2V@cI8~SuEJ#o<Ht!s?9tt*q55aWR=XJDFdvr
z&P~{ZTp!y~r(F|^CTD029<OYsf)ZJk5JtLC^7vWnb%T=mFRyhHPT$){*5NaM@NCK=
zs1!(`FqpyV+>LR*T&Y(3b|7l+dgWJ0;dsiTUT{{BBcah?Rh0_?PgqqgLMOjlVG}aD
z8#PM*BS=O|Myrf&WOQB{p}X-UO#(i$*vh2KL)`69QAUZ52iLR4mu}x${w|<Owf+Rb
z`drGJiH(P-wC<}XQ$MXcU)VwFPWLh!FB-`jb8|W6=Q~5pdrIK0L<`nJV0GVHyz_Pn
zgZH!S5h2|4WTapLoJeSp1xPU<{FUT`S+4r?S8xy0U)_aMw_%Yg%N<pR9bm{Yh8tJv
z@-E99y_LH}$<<SC-sw@h@sL)qEFR^@E|mf1wDSG)iJ3EaM~wK1YsE%dhNu4`SzVx2
z^(R*1>z@KPy!`FW!jc)pmNLzj^-Wna&g(#Tx(zCuh}c)O(X*-*yqi~QkDkZc74uR;
zZ9w#H;5hy_!vVvW#gQlU{Q%y0=BL{pxwoH<6d|2V9hI{2BRT_n9-U-Io#&d`-rsFw
zQt^Q@H6_LPbE|yp8Emp`dS>h&mz+-C$-*+0@&70oLJ{c*5gaQrFd5ez_e9dvYEuO5
z;Y?n&-b(K8DQ$wM!||M^3eAdnCw1BQ=uYEWf=oS$LzDk>(28=CRRQ|1LJaJQLMN}b
zGI0#)-$x8##%UEI#s6rTdHU~Bn|$kw)}R>|Sw$IBo*}Ng!rl(vc!mB$JXOR@VgaKX
zW|C?f5A!?MQ6hE`SSS1Ve*t|PRTy_^WJ3&R*XkJ^?(lqkzWD6t4~EvZN$bUlj5|R=
zCqp5Fr&p$`-W*pv{!)M(tFO&j;)IQQy`&BPp|NG#V?z<eKyaBY+nhHZgO?x$rNA+m
zn6lh3P`A~2zH>5i$Jpu(?Slz7ZOEuaB~A>|dc-m+C4EMguH%uDWNw@sLvsAA5LHPa
zCKG5-FE;V3AttH7Z2#2N<2nr;q$3U$D|QamdP03X$f(D<X&ktG70M#sKNIO4e2TS@
zVr(a}+#SB|#>;D({e24ccg0&TL8VH&9;4l<kkuQCo{IvY%jiZcD1DMw#}$pvB>G|w
z3B~YEC-kt2qz7h!*=RO_pW@A-y2PY45Lyb-!T2h{%(rHAH(CFR6QQ!^wi}Rf$1=6+
zpR~a!l}!~7Qv~YGx?K%iRIaKG-dx?^xBJi96Z2C%{#f3xrc5}ad;AowMm>3CRzGjw
zcszLo`)Wd0Cw@2{@xB!j{9S5#fnxFdjXM3xU4(DzHT^s9y+1)nF@KXs#yg^Z<PZH#
zT49?&i{FZUrujXd_OWWlWli{*AE;OprJ1xy7rmr;$U+A!xH$SId8B2fb`o08gA_m_
zxJok1O1?W9LCk9GU%rGqM+hlNZvIyL;AT6)rpFC<3x?Bj*f5KK+tIAXdhf38WbO3G
zA1S6_zZq`v?`bMiPSI&%jZ)G)arsTa-J1|3zb&Jt@lk3TyCOl?3fP6S0oGPb@9!^!
zK3GoUj$07msHNC(`S&@tx@(kU5}nr%piMCRDP%h9gh}NEp8J(E{4gXOKXN-gXOnO<
zYBAIpmq!?nv4@>&a%Y6eFC_?NTB#}pE72}vs5`t41OE-b-$_)z)h`dfSWNAGsZj5t
zn9mJ@1R|BDn#}VW>3)iNB~V*6H9|>Jq4gQY`f3}PVw<(b&sE36dBimxw8*!%o)x@V
z$48RXTIv6h!Yrj$ITzwQl~NPGz3K#szETz|NTH3WIp+8;rzQIO^H@|x=@*k#=uCx{
z0hf*Gv^P78)Xz6&A2Nr7dsV}lmQk)&D8)(NM1QQ@+u5R(9ZOvSGwH-6$J)jp^mHu@
zID2COjhMP!GkHiQODz99_8XpRf5!Z-u}r?mlk`p*q~|diOXS0?!ik4}=|8ZzYZ6W$
zSFj&JwNe&~u(pC?^z4-VuDo1fe$G5sA4R>%dTU#1HxG$qvcPx~aLtsyOok*Sk-ITf
z@hf)t(%_rm+^E0IXlxdTkKlf!3S4BG^H0|^@NxR{sK3sHIT~+CeWaRiFvG{$f58s&
zuMu>mO8noiyMBkbPlmM*n3+FjzerrB`Q)5~DBmeN)hDi5p0Xp1wT_O2V@WXHU5B()
zPfF-!N}x4cb)_W5EuXB2e1Ns+Mf?!Aif@0Un%*4C@+Q+J5h6wE50PVbTvFbv$euRY
z0>F>^WC2>$KC5Y;dHP%xzJ%H(XEJ<JQ_uBl`-f8-pY(NC?y=aY(?9_kLGnadL?Yx0
zJ3yLlKRAVf)KRD2(GVQzB=PtCSN>8ctVhy+qtFiPYURs5Xy%7Yvi&Ft%`aNaS>=s1
z!TPtKEyJ_xkG$Bk@bY9C=~N^6QE8|SdVtrTOWE!n9c-8cokwK{Q?f&Oh-NGV9*&bq
z$zH<@Ehz1*4{7>+vf_S}c)X=<^_`lrPci6xYE@Xt=XAcjzaQ(S;kk6&zjo|Ykk${X
zO2PRO)Y-|3NIT%W!hOP<QF-!HU8IitJMb2w2{4rp(H~^2`jzNUC)HF*fEuSL8i3_Y
zy))qUu64sHWs~|;t61J4E<5{9>fP4k$NTsfPYJM(5c(KR9V?|lxPY+}%tP>&S4_4C
zJM@SIZ7+RHi_ogt^-McCqDjtN>-6n{?&il9F8$F#pBlPWdWzb$&TW&-R(g4?g`Vux
z$l*ZdcSNo7EM1?w`)~Ifm*vPwHUQ&|hd5un<YwEjRQP_tr%&HI=Nhf=EXIS}whY@b
zYU4(+B-)iE;9It?N814^VayJsiHiCEzASbwF^%Z#H$c3L3K<X)wr1g!GG(P!c&^Lj
z%Rl*kRgjp?bAA-u`5oyD&e5wIc$tV!ywJVYc+buijz*$)TAY=)38_gD=0Jmq>?p=H
zrZZjST)5q@0SSaAlQ~fh&aRiSByDdS->)O42Oi00d}d(19Q?44oc5{K@%PqI5ZYPS
z?&|jw%vz->@XnzFRXAL`EMrsut9Tcn?J8>=W`lj?uqZPnM6*K}Q^E4_$}Wy*<CeZc
zH~%X~F(q_kl)TWT(B-H!8yNp$&O`=g#eC&Z_W5j!g6Dx2ZiQaK+9iMawR^NSAN*Ub
zd_sM;BwpQqWIXC^#J^>3&7VtFoe=Z>b@aAdk!!R3nD4=Czxdw%^r&J~yq=@e^{^Da
zKust`?cd3ZV;-hF7cx?E*TG?CC&rI%#WGqgz@uPdJ{;K+*V@FO?FGh(Xph3I*&LhD
zXOrQCH}Co3liaRhUt?VdVFG#*C#Hw~m40GaKL%uu9N09*>ygkV(!FVY<wnSsxo<JW
z`J6(jeCdy|R|{^AMvs3@^<~Sx)m&p8GvV;<{tai+Kod9D`xL?Z=5>);Sv*lspK1jp
zd!y?bsz&|O{B)WkY7#iYrO$*eZgu_YuW-I>vieckRC7S(e?-2$Ct|a*TbrIJ7(o1b
zZE3HbT<i@D=iMv5tFQ48J64+BZ4R$~5**O)1Z-&jduTWr+)$!DE6rNmgH+oU?U<|>
zmpz6WO63|>vQ|#Pt=abnb>?C(-+VAefzh*p<B(cjKJwqkb!uR<^L>1lN;~g8@?IHK
zD}X#)xm~X4zD#VR`D7~qYq(Gf4Y0TJ#V0eWs(y?kGRnmx8p5qK`1oOgLCIz+a&SV8
z<!_>j_yUi0-PblSp@=4f=UzCGV>5E(b0qzC<?9s$;dX3~7U%iyHv-4#yNn0|=6_ya
zpBRheQ@YBN=_mnb_O+(&8||-ZvzTjUa%e<UDoPkb2~1-AL8Oe0z*Edckr<gSS~TT<
zDZIYc9T+IA<RA1XF^My_^!GDkx@>EP*;AAb@NKy@@uyy_gZmdv(AF*b+pRBI*p07I
zDF-LaKJS=MzG65E2Spu+05Rz1PBz<vBx*SX^_cw9yn;a4!W_1^l!a8TzSY0?Ma5OS
zHyov;Zj8*YCC>MxLH|%flKd=Vv%V}7;6m6BMHLe}*4=~(Ut3DfrqoY4)d9He+oq;1
zBEq?t{~1~zeU3Mp4z3-_MQy&|3*MD`4yZlPf8fqGlUKw1U#yah*jsf)UhBqppjo6~
z15R~Pe<{q}|1Vg<!|1P3S^7V4<vh`j^wWHO*7l8y(CXFw=lyg&n1fect4>8r@l550
zut}Uj`$pxHX3IX~bF3cmh88`}{!T<B(O)u2*z#TSoK@D{Ql;D1p^w#lk7A<*;zLgF
z2SfEmbw`rkizHjKpY2660Hnh~O{(tEzq@d|mF*@NnMFSH4zZoZ(>?ak#S>wp{YDzD
zGc*1RDsK3X&B4t}{W}@TjTOKB<yvJ=eFsK-S-e_X`m0y@Jp<`2<cHcBcV&{g857%J
zw)Q1gQLh~vumZYDOQ?}Sgkyy0U`T&ot;0t)i76dL@~;+Ft%NLv8bEO`X^J=s-p+`K
zrNl91juam)Ma14n&3n$+?ovFLkOOzYJDbtoV^sc<wV&R--*HTw5~G~G6o)d60$-fM
zU$A@xljy*IZLB{0wXo!P8l$C2$?JuABQ5#P(rM=A<bS1c0_O#I;cb5I*R53Q`;oIJ
z7NFp&w9jpx@zcxn9WG#~Spz23n-V)N5k0>Sw^e;|9uk9yS07c_WCtH!DECK3=AC{J
zlj2*P(z2FggowYF<(S_@%(h}t`~THh6{c03oKP=SBCM?xf4h@7Ue0K?6xV0GeaA*?
za^z*vx4^{0s7AZL650Hx{-bgBzYN25qU`fY2ZL^4;C46B0~1K&-<``NNlN-^q)V}a
zlIqT)T8($JgY=<?H3IFDy%a$bb79bXm1E#7lkbmTAflDM+e5lpnsBi5DKrd&eni|h
z%yUpgZHf~coPzQCFUyQ>yc(t@VK^0Y4yRiDB+V4pRbRoan1NM6eTqjbHK0k;K|8Ej
zlBzw+{uaz^?Es*>zrmP2rMxi;$I$N|cnwPVW=!l&=>7ZHC$WKL5lVkF_(3qHlg+Qn
zfG-flF>uznwf}=Qr{jO-ll`rMTbh5D*B9)ERzPiCN7a%~(7H0ys@-REH^iiI7-U?P
z9(SJLxa+lTw*Qr9S=70W{Ec8Vg)9`EV5C-&St6tr7f}B>GR?BFktDpnRLb8gSkAA?
zXpaQm>F6(;(GkD##Mk)zov`PQ>Ju%+l>-?euwzh~M*ggNbdmHnSS9c&>1yqh_bVLN
z)pEmgJ7?;fn6#bYY_H+=&#ypib2)aU2a)2{e4}(=w&lv2HFNcRtWU&Ux?8AcIU|mp
z%H8s#rocj2a=Vk4gHh}$5?9>;nvalC#?|j%aSCiV)wY-G!#e+%9{gaRpO!&<*rQ&g
z{`JZVMo`~ur|sd7tvW^)q&hxm;8KzeO{n057Us~(=l#Ye=e~wM=li%7w(`gns-;T9
z{8o|9mMwbt9Ql{%uh?5cyih7Qi1xoC+P9y+`2faaL~jcvGuGAJ-NQvkl&XCku)--*
z7K_1)ft12#1F!4g1egM}Bc7tyF(D^WSNTg5sh$a|cSchZ4L;a)J~)vyQs6cMSHGVv
zY;{3-v7$DP-5=LK@+h{Yp}tFD>SJpCbq<zti>hJuAAzhMvJ!eI8DH*SS8IGf2af-&
zy8;?IUy|<cy1N!kPW^Z^y~jSJ%DdXgrj~~#yM;cG3T>;<r4yR4_$OM(cST-F5Ni4D
zUNyr{ss-Cgb^}r9ecU*-Vo6xe9&3!dezsUpt_@~Xn3HdXfA1odx~Bf0of;}dW#@RL
zS?Vr;R#3+h5n9`&j`7BHE`}0~%C|Na;7dLMebYLADE`1R*Z+YK#Qed{&*LL!2D5e|
z5<z1qN;8{Pcfg)QJKD*u#t$^r@~wL?I|}V3b3biM|CK>GqK`<{c~!Nof-4yOLH(*-
zN>e?=4tS8O7^O$C^L%T7EuiyxBlE3KOp=a|-c{@z)9!J>YAr>!7Z^Z@b9>di6vr^I
z`y>@rs-r|{czo+)ZPYZjtg`=E>Qq}uZgFbu)Y_^Pi?}%gPl<)&pL+4z-5<h2l!-}@
zu2WnkVy(}0Ae?OmA<vRB-*@@?s^O~ogF#b7E4wOnt)enB=JB<QR*Ojj67?GAmwI<~
za~fz(#~xM<Ei+y}ezh3Cv!*$l14Jx3Fun7UdiK^MSj9ZZm=#lReOm+I+9RsC4Z|C}
zC4O?`Dx&)l5mTvge&bs=Xi-j!_NA{5{(grO#7)dR8GaBRA8HX#fU;d2&%XcSwiz``
zJ$qD5dwbn=6DY^4%R>VarPa1*v!0a4z|XtPb2Wqx=kR*@irle8J&skdza%I9ah-MA
zQWrbs5z}VzX>dDx?pA-ohPQ~BPa|YeYh)6oP+3|$L;CwfwD3rBK5vwdQSlnN8kT0|
z2#QvNdoNLkX~yJ#--CBrEg>%06a7mgk8y+-<L2h<BAOk_P&ra_U2IqzyBz1_M-<jJ
zhl){XO_fM%b&6uuhLh;eG{AC0oTu@IlTY|cw?tvZr^7uT_JY6_JIUuo^l9qg3$=ff
z8q6iDa^4N^$1xuskBZiMeMN-m<q(#4h{2Q1F8ot{iIcnjGJ$(3LrG-Bk&PHh?a_5?
z(VJrKPacYO;>w*#na;NM*<DHoChu@vw)Z>mMxK#)mM0O*<EsnSz{Uyf-KmdxRP?@;
zzw=~&RIkbQKV#!<UB>HF$~7`dG1U+xC9|v2kL5R`NxqwPFNH%hRia;1qjas7Xa~Fp
zsI7JmCg#&4!9(wRLRsqbQyMPdxf%^^>`gfF`z<d9LN-s=b06;2SD5JUnOGCuP=X1=
z-sfeh(>;=Zi=(}b2duw6R8aW-{!J4>OO#u~ekK=1lVOm|>{7c&AKODWBOc%<T2T7d
zFY`lod<Y(C_b}x;5#=bdqNUQ@ABD`Y5&DQn(G&nuT!^*Zo%$!FagOH@U0&czkQmYe
zc`WZEb;<_vfPCBAZ8V~720FLjEy?I*Dn7!+ToNz}*R5OMTcDxMJm1jlr_i{h;-Y}R
zZ-&?NM8AS=7*{BR#wTc<xWJ~H+EVKX579>^CYoVhQrGDDu~{^ymwfl*Grabe6zuB!
z$o~Sx&MwQr*1@|9Ki%9F?9}SZg7>^kKv*tf$Ez^cGhwuWbOZU71Y>hOIgAIaHw%)m
zc4Cpe9_33v3m^lHny;2odx>%;{_W0?E$jEp|DVV%x(U$PY<NM4q$8BtFcL=W*`G82
zlrZnE_qfsTSqm;dJCB>*|9st4A8<=U&$lBj@(=TG>hAkZA`IZQmFDnkNLZ|LrUG}i
zXT)5oSzZHE(O4I^>Ah6TtOj>PR6rHqoFv}~&>wgV+(ksn_5FI44;Sv+3RGe#8TEE9
z@1qI-Zlq`RG*Eo2Wbk7EgpnqL#1^n1nINU<f~r3!*4m*J;)d^8{&&kofs|#qC-Ud-
zj2I`{7*Fb;R%je{`q)*U3L{f<D{o{1`b|ASV}1QiFo9_`@|Nx%>SZ13c(2}P#MFJn
z1IuO*@Y7P<?ea9Px4>Nkx8n}h<|)Jbljq~lnxSf8`p;jV-Q~^yglN2xJ5$9y$lY`)
z5c%}`U5iygXhu=E4m}h#!&`BUzSgR#L#}I?ddc7o>^J>T^!lu!lvV5)<@+4ciQzGu
z=P?$%!9YurH<RBf?$YKVmOc>^QFL>GZP<&EI=~VsKh_(Km=D9BP{B@_oW%me5^}De
zAAVJ)x;&hPWAp{*Wqj5QKl#djCrbX9DUjk>l~-$aWA|p?<4VqH8<Q`Au!|gbN=v8E
z(Q2`|fvL3Bg2ca#<$+qN&f&Zza93kqka*X3-P?>eZ>`cbA@Xc|Yrl4;@d+n>#;coq
z&J$Smk5nS~*ztz0V6r5cw(HyZ?LS)S7~V`gY27FJ#qT{ADTO)9wY`Z6mK%%l`)v!5
z+jW}F4p+IbYrLEF`p|QmE1q$)TZzoeI@NQ7OK96dq*`xw(YfQmcHaEs;Aa%o3860j
zW?Wa}CkB_hh)tJ^TI;KW9n>{}{nV$UXOxs@EV5>(2eLQ@w&TEeyC?=EB$^BkUt|WL
z9r-!zKf8*w=8>4Sc}4S6k+z{Q?fXGjr9JH^`{W3*VpSf3gH1bDzLKLOGR5U@&Zq{t
zgPlHSBatIy8557Izzuxzs)9ZNdSCJ9)Tidub>i4?cs1tF{`@JyX6Mu5yK2$ftBU)G
z{?q;KJ5T?Iv#Z73$7A4Tee-G3wYvjoe<uyvnm)7GCT~W-TzI?u9}l;;&(>F)5fu6H
zq7nC-IcM?BNbmBdL?bXCptljtq7m?X>CMhovGujrbk)+PW)@Nbfyc2?HTkaKyR$1O
z{Infv($U)5CYFDuMNfat3e%*2XKqHD|GYU)2DN~%m8lIK?!YH6uD36uT!HCf$JlL8
z|L4P_v#0%+1h4b$$PfQKQs|$HGHxyo?)G6_MsO~3#Q0&r?H{u4&yf4jcKG2A1W3Sc
zYYn2>nm=vEjOcEg(VBjGmG}4%ozUF3zW{nVhB9Y7uf1BcSZCOve+D-vzPwWBvmf{Y
zJ|1^g1H>uk{rAo^P>g;G(857*CuGO?bU$xjdrAki47o967(yBiU&75ECGBs!^EE|l
zx;ok=^Z5AZ`HAezyVra6FLt-#SN-$^^c(MT&d%SwqvKEhhXUOBAr2w;<rdEv<G#IN
zDQ_Nj{%i-dK*aJkH(&Gezi0#ZZqjDm59|AfKrWp`<zB2xK-Ug1^SSHs_%Y>iJ9@SM
z)zA}#$kpQ^F~;ToIut%%Lvh))b#W;2@Q~CU@EiMSC&0`1VK;sGKB8uEIJ~#J`wn5<
z+4%sSMi3ab?Cxeiiz=Seze4gDOnFAHe;BUu8TJB!0={KFMbKS~ksRM0o!wW!Z#|zL
zPg(zMb>IJLewPnuifxYHX8;fRe-5|zwc-vnKALFtul^N9d)8IlkMt7p7$AGD1iIzl
zN5wG<q_rFCO3`NvcWR+$)q)4S-D%w$U&TXGTJW4Wb*a35wz7|~2RQL&_T=w|UYWY^
zetk)QKgnt8vQafOd@#6L%vcCgQr-4|*ACi|JhLWSZMy<Te;OV%!m4_~1M^o-rrq)b
z_|oqfCHj>JlV3jszQ2tA_Ej(4ySaMgqm#*l3uOcQ;*CpP=)|a<N%2P)R^k-hL{f|l
z`+?_6$HLR}1co-M)b?A;+w37Rn#J7A<eelY?wCy;bZk?C<d%sm9+jL2jg-_e9<jln
zOYcQ$-A^Zh&kFn0?W86nni7j)CmzuHL^Icbxn+v+K&$xK{H!6qq&*OVRN-#NgKJb%
zI`n8~Z~BJ|wKUj>OaRb~Ge|NC|90Lky!$4@UQq2OTjUr85yfLPF5DGfZe$%Zdam7U
zlnV8fe7(rY5bbm5g7=hjR#7+hZ;R}C!B#yU&pHig3=-2W798NM#+1u3%p)o2A<bIw
zdCEDQ-J8uS;QTIDusF7^V+fk{(7f0iPZ)xFq8C-Z<Fa(uOH50JT0R!|Id50t<kB;M
zXKRhE{R^7TBYzVaafigI&QW9MFdO4aY7=sWO*J+4k~$c-u=q=(WV0bKBfBvyvU>ai
z%ea6PFJ9bSaRv=4<irW+umXiJ3#K%fw`p+rAdW}Rr;<SN1{@GM#tj)wnhwQW{h*lm
zvrv*USfH%e9!~Nf+>@ua*Q8(RHOs4NMn>ML(W>A<&dBDNag6{<&sR-M-{7bWoqvRv
z6pNNkJs(*K&eaA)rxu*?d^vskS<Y(*_auRCle)1s%M@_R^q0`Ib8(H03!|aZ-xB$M
zoMNCJ<5OI-A)=H42X!?$1xvK=4~nKf@$zaz!p2TXJ5+0jVk46i{&SMjCmb&FG0D;8
zri|R-h*k?x!_*Gk6=3Jr+jn^j$HG&t8Y^c+l@5)*NmOfCsk>SKsTO^vx0pson8XGo
z2s@{V-8{@0iR1oy6~dpPiHlQ#yh8%uQ?ewKI>mH)nAra(xd!*j6a_R1((R;9_$%Dw
z=&2K$`~9bt_(3hAN}&`_rKsky34Y4nrWnjugIAU-|NZT4_<pVVmkE&`=|vVb#V|G|
zR77uDDEMu#L~xhgbOAR>D&y+?+<+1gtei3&EaRtNeX-#oX(r&C-cwyZ(134~w%Q~(
zbxBF6yp1<YH4r=|e6FmwFjYJScb#%6Eoaa<d=R7YC^R4_cB}F*%0s2k7L0@X1bYtK
zBH?Er`r+8%%)e%cId1I#)hdjV1;X&0BBRT<GE)WdVc3xIioQvKb3!!|ObY>qFn!q^
zXcH0(l{5-N4)eBo$h|03&ehb#oaI9{Dlu-pIA2PQI<MF`oKaF?qTYfx5?V`DYIgxb
zxAJklT|$eVY&1OC(vByosZ`fwf#ZE6kykV=OeTFgU<#G0E}DVR9d0wG+adYb8{94x
zT12A{Gen~+VW&;MPo)%+HwFNa!kb(hatjKPzGa>kOBz!p-4l=&vW*Ab&gB6zt$UUr
zW=7y|!Y_H#*Pmxh<#0ggi{=a<x#MWEJVb;KSh1CcHq;{y&AicLt21qy_TW31g+YzV
z3;kv7Px&ti-saC4R&nckd7~b$i9{`$+exPUL+m$7Ff&@t{{3cd-U4`&?3zMYhOhop
zs<=s{%Vo;VV>w4qEfDbLBD)S&1)*s#4Piw=XM=6>X3zv09SIvarjvRFjb+FaOxI=D
z`=tgWRh=D}^{KpQLEhfXWW)J_aobt+4B>`cYwG5;mA3q^a0Gle=-X&O^s6g$A<SKH
zC$YZs9<C2AgPFJ4)qs%L8$>&f;$w*Df9Z1v9W$IFbf#H`DzX(QH7bx}h1Tq=Gz2<?
z5K~JSZDhv)g?OhG3dhHJ5A|dYn7o{N)rH|tkvc(}cv$Wn#`bxtqf<0yQJPSVBM?6_
zhdF_!lcGNmM7Y?x=6z=1tLYH$&_U1yem)4R03pz%xl@iB0RkPdcjt@}=2b{Kua)F!
z6DWdN&MFwDr@$N160AE2?G6F3_T3(g#CoTTV&ueBV$^R?GX)+Xb+s%)hkAEHx^WCX
z2X*g)W7kaXa=4-TT_-59q1<Pmq(B{3af8sYVT1SkM46bI+s3pOwK(v0{d}hU)+4Ys
z;e-f6QlAZ|VY;Eloj$n5(VCJ{5Hz0IAWtb|<gwid+iuUNCZ@FrtF^+0v7Yyn4jR6q
zpr&7Si$gH`x>4k@o-^AjEHG=>4ra2sRpZ&5+f&pXMpI<F6`NLk#tG>@6o^q04_clR
zmOljGxE<eCr!R53uW>t~9UoM`)n5fa@6KXNl9vG@n6i>|+887PHR&)b`kF&@N1AD7
z^y6313*?CqL!w~3yqUE~H!36S4rxr5OG<vRuWNG%6zrEtA*TXVO`yq{KojC=zNt7c
zM>-I8%c)G5$BLwLD}iZqIq?OneQA&g>YIlWIa`H}c0EF5;cG7b6-%^`L1m^nG4xV;
zhZukXxRx{^=qq_+=zAg6gE#Uf5LWsTVi(*Qjj2I~(DD(U_keJ*+cSA#&nTEB+G_PJ
z#|1~CZA)beHpmXM*dv})Mh>FDP`u|$L6+Fs5`R`nN-ygjk01}{&QmIoIob(&GhBI-
zn3KHg!{yEK)Yj5lpYme76oj_DChjakz{VfS9#cq{s9w?B&uv*^=u-ic`4)3|@8yWF
z9u){5U)XrC$#JYhGe&_IE@5f^o3px;hca&&W0v%_^Czma-E6Tbg7-cP?yn9tAsP06
zH!g03JV6l&|EX>1SMZ~xCNVROeHXSOSnH0{OHIRm&@j&OUZl(TyXGZ^e$hdX4Md*%
zF%y>HQ?eEJ4}V+AHM9;9+R&%6?2b;b!1~Y{kD@I^H?md#i?HqT3KI)fhQ)<`Mq=w@
zsuM_J&DHG85I1~<5<OtcBe_y+oih0=4H~)cSakB)$6;0GW1$DCfLIw}m~BeJil41r
zKW~UT7joUr!<IAmw7c_HM1QpYFn|%%S0u8`^G>f~AZqt@{*DYs3tfwmZv*b74Wyeg
zaf(^$Vu<ZYdV|F_=%I7~C)qa*Q<;MIHNj5L)#UI97+EHr9hO2o9oiR#vC)lcV}o8B
zV;Ic}DIw-SmW~={us<e9obso(g*(2$8Uu4K!ne6fBTNeyUdYM@|ER;x0MkttiM)DL
zVn|uCd?q>Y$DX++97UzW8Y!W+gOi}xVVbC9hl7YEeVP_8#%?_XZ@N?MG7YN7s!%bZ
zr7BySd{U`RSmT+kj34F0rcBP=M_p*I{5B3Tp;DV<$6p+5=q|-Rb>X7N5Mq~8VyOUK
zFbYlHL{9dF6_fq|Dd80Z?id`Aq2ilK4uLyG-{=&1!yLF^CrL-du;bLW`Y>8-gCu0a
z!HBtahCZT7O4Ef&_WQTRtrLhw72I4OLFE`lV`P)djw4(a4yPhaR#%SK4{_7p89ysm
zQQ`fYc97wE3mK5zPnpz-9K5Iw47NWswrC{au@b?-%P0AlpBD!Rq-<a01pC5Uh?GYk
z#JBc3=(xvu=oZH>jc4R5dpe?D6`$*~oqgr%q)8lE)?ItN_LN$WQ|%b%`A*pC69vJ8
z2_>*qse8Tom;%A%O6;}cJ(E`cxhTVFw=vU)(18)KT=Eh!s9v#tbcMih3|dFvV=@kf
z2K2dtIbJ9nAQ*sBeBtNjDSa++rwA#1+cpQ$WOpeXyK<(OB|}3N$Y94}saZp&3dr!r
zE4?sJSnnS58Bjvg2w2Au5c7}T!LUFs`r6kn7%a50Orn5-cygDNb^#M()_?}=+qU5(
zvQM{5JtV$^7|I8azuqp-w92Q9k2#eQwrHUxi#Nw|L<3f<*k4npO;zKE%luPBpp0)}
zHsnf$=c>|)18^&u^h;@R4Rnr=SFGe!D->#!AVZQE%Df8XLYssF=Ul^YmND8F7i8O#
zi-||n(5G6z3M1>Qu``;NFWXRyp|PM}C;;Aaedvg5;o1y9o8-;|M^0kc4%(X^`b6<y
zL4Vcu0xhlaBBP+vL4jk)HCI!y*&pX^b#`w%k_=g&I4G?ZLf@E1vhc%Mh2J@-{#Ehi
zM_Dx8{Uo}?5y1xc%);};Pbqb?_>#Wf0{H5*`KLr_>!|L<1HP^~Lmv1vZx-m$-4$fh
zPool?mj`$6T5>!V&;pkB%p}~=w*zTph!c)?f#gBBtNY*(IC1o(pH6Hi)ePqg!uXG<
z#t^Q>%pd9G@F~-kW4tu4I`bQsg>}wB$+Tc0^XFv~gIoM1wBv1QI^|ucGfNC>%#m!}
zdGv$$l-I^gQjTcA{2CUF1ceFoQxl&k@iIp&Eeo}eN6Z&1;@$A-7+6?_ZTdGGEY;ir
zV2FdgLl+PWlmB|=+w6Av_VfPtOkZIR*&i)$FzX=yl0py6bk6hO%`77B(m2KL`I!UQ
z&?oGZbpK+ONP(MdoJgiCUXC-VJsch1Mv#I!D!s`+h1%A?5j#oAjUZi=*&oaAJ>u+2
zi1m`d(G6;rrV<>S`kSe7&|2d$EQeNG0C}6}U1HV^Rvp)ijg4)1xMX>JXk*h!?*$)o
zD2*eEi|eK}cOv7W&qyoe6=g7LWJ-3@@C)PErMu8lS|48qF;^PSl0fqiPSuWeg&Ub^
zp?(|G`(u$lNI~w`@KxJwpSXp?EE)Z*rR;;alu+JV)LCB_)R7Mh;dz|XmMrqX-<*3?
z_YStaZ>bq1oX<GMDK(VLig^X-0%Qlll&N5rfp1YhMJQh$SVDvbIk`v(H*%5XKQ(_T
z9lR&`BcSFxwE~r6ANgR^b7}U;n~S|2bC!6x=yyZ1uBT^7OZnqopcT@UE7e*u>|$ao
zDqwKLav!uYMUIxCim7}c`Tp_%NY~qie+~<9GDrn?2YYTvq1w3PZlDXhur4#^{Yf0j
z`rvH^L4dt$;pGheB)4bC+yW_xdpj6tIu+>)2%6YQLAQ@&3{uMtRL${)T81*sSdu=c
zqbk2hQ0E~C%RNFtD5H4ro6F@54FS_atJlq~+>CG#dSahs>`R9ZSOEPq4w1f!eepB+
zn3&b4bz1k78eFQA;KfAnoZpWS^)7#gG9$7oRSqzDZao!9H@RM8jSJNnWHq|XYQBwB
z_^;{UB<(e+IOevv6+@OC{Bk3sKTyARO#EHz%>>-F^(6pc#MEkcy|MMKbjkE8MsVJ=
z5nqKN#%C4!^E4k-2Y{r#5iR@o<KRFFIwtdR-|u(lekUExb`n~8<_PoNnl;oclBy#b
zv2AzGIT#x8I)LvB-G+xZuDqDmgdCoosqa8f(T|RLrvP%r?;wG(KMLcGKHmnInHk3L
zpF!Kat-_i_3N_FvVlazwp3mrcWVA4EcWM@D%?^>8x%M(oD?nsUwmckk;m1!{LTwUz
z1-jS2z9sZ*LNHo^1Tl-@udN1^ck6f;xHk|DW@HIBF;{sAE?0RfVWSrc6jPs`e_@=H
zt=&50V;-l_l6>o%JIHEFVQ_J8q>dhyvrPd@1CvJ+r2CAK)6>9cxL*>Rk0x1);j$w)
zY4B78x1ZYsILcf1(!rj2THHI-n_(scZ}Oyg@5L}(aFanWS`K#@r$OR~H6BBlT{lh(
zE<&HRbjY##e#v^n*&>;HJSj)O?uKT24PAMkn_3y_#l{dL7@?!CQ&~*84~o#K+`k|o
z9A36uZVzs?P3zn=lBG_Vd}*tK<Abk)lG=wUb#_1juSPJ9xY3wh#aK8*k1iw)M`^4?
zml2hBh8m-0EsgpWlvjp!dKnoTL{Cr*KbzD^fr`Usy>&^I-s=PekW3f-m`&a_(<|w}
z{i3|D?R6;PxVZE(c1vLJ0HLo+mL>N+2+?%9ulZOQ^6)1!`3p&@S3!3gROW{c$g?-*
zgMdK8Pb*|Cy?@RuCG<@bfp}s{IWLssU!ITlOOsmKpEIM+CTt*2NRhMeKBf!C-EZut
zjx=)T-`u*|D(ssc)g{lqHbRK+>A862H>xhSI+$*oQz`Ga{%xC<(ym4(E22g*B!h^W
z3D3C4qsjmHW8jDfQ^jBc>FQLC2TaSM1E>3vi9h4QVPlmdzFn&o5+=SreqOTT!&J&!
zt-*0_8z#B(F1YgwC3L1{mmuCBLG*Z&MCwhUHzS5CZQa{_n?fxk^(kar*szlyvfLKJ
zTw>IlRC<FPS(TguWKq%FdG-CX_Tn`^wP|5K$i$F9WMybh&6&ZwYRpk=7(uTGAmJ;d
zPl7g9lHzT`s#K^$P`%7T7q#e=7GM4y(V9Sut#-j^oVuPbTs9;~=@pS0pxLWLhRfT;
z-t?hkcSdR5DRG(rg<->4XMDqbC}~c-zlPu6_TZ<X!$Tv+zURV9&i)B`qz99S^M8yl
z6nby1jv6mgt!#Cw=xC+!h*cQ|?5+1TD%oayIaLuCZp5$fyfA1?sl*qBESkUGfw<Bo
z_6n|g4NPd+RsG@4bFcm*u`wK8(t7!4Po!@*8OQrSqw~2gFKs*3Xj|m%`RBIS*-_IB
zvzP?G)=@#Q+b84NJ2U;gm3DsnpNx|Xg~c#b`KT}17f}c}GDm=t>naol_?;z}XbpMQ
znpN%z5pFT>;uWG{vfG$cEFmcs&$z+<D1)&W*2MMIR6-M{h7aj5;Ee{HjT2U%aJD;(
z+_-ne&7ewFaQoOQgih2I(HRnTF~QMCwk<U!Lu~h6MqpYO$vAx46e|xZ_i-topdga6
zgpV58W7`xjevPx9XT|ph&?u~<Xjl0eeXUqU(J=UdG%A$&BC{zZoZVVY`dpY&8Jg9a
z-nV1g%38ItBoHe1R7!=H?^m2|d;Z;goMC3}4N6wdcC!^Zw5V#ywne_z0b=*XU;UUf
zvaX^vjZ%Eoo|W}%)MQ=hpZkFKM(zw@=D9LKWL;77e4=lLY{lp~AlMO%X8!0pDLF{r
zVCgH5U3{_f$x&PUs;jZnHiBtgERW!erPfQYk3U^bm>%og20=xxWuK_~s0?XD;2moH
zs(sAAM2au@L>s?gY`VXa7q@QHSWUalnqrr>0kcTs{^2r9bH>=Fsuy8p=ekQB^VeLB
zl{3!zoC6YZ9B1DI!clcn#m$;7T+NWOK<%47gQ>!fz7<9bm*I=0D;p&}fg?Cs`Fu@8
zT_$xr+Fdy_h$dGFFn|8Q@zsAOs=RHmKtONzenVDCX&q5^aU(p15fp0RQ|-ovR?Ojg
z6tGXk`a&Vny9lHTsAtW%L7S_c7ah%BqJrl+atOwY6B_^#8C$M&M28kKZTi6uKST`C
zDHS<Og(-har$$hF$vZ=*b8tJrYal*c8FqT!h!&~C^_OXkL*Gp2z<<t9C_aC5tw^`a
zc(K+2UWrM#{kZrVXL{QCbd;R#^m>mL7JJNL(|;wQvOB+@SLmUsGM$K=Vc+x(_WF2P
zA+O+cnG1j_B4w??uV_)<Y15AEI0jCCxvW&;GguB=#a$qlE@R~+!c82li0-u1Y^~(N
zQ<lR>pz*K&1~v18kNMpDtwscc(LVR>mj=0aHH$&W+MA05c&dCV0#LCdFVnU*)<2w(
zY9<+41bdnrpGpaB2+t5bx@jWYT#|JnY#>9GFaV}<o<$=oK3GRkj=XrxI2J`5C_mo4
zggOA%MZR0QCuJ8v3?GsfU(~bRHH7)#A`U_CpbcSg4p@D6=aAi}Zv*!jA*55c!RmQg
z$6(l;ZelN0(5A7-8Ho>XT!No(col+-JK=4xJYDNGS-4>Prv}&ExJO&MS6LkgI4i$>
zfdGkE%C5ZlK%At@93ydn#{XF;H~*8)Zd!ZoaomH8QIMDq@@#7w-Z6K=ObGkBC#rb_
zY>CQ10_xgNHO<V=kY)!D#@lcd%Q*#=NqE;LB&=AM?8A5acjk+?b@w0qPUL!iOd?-^
zd?c?$u%B|aa4hOEjLMjf2?d`^N!H4NSvacb&y&|QJ0uAd--4akr&Us*h1-5o&v(6%
ziNa=Tigj@?tF_2N#!(K74r)v9jv8=r=B@9S7bwUS+c}s#&O>wDg6BbK{!C>|{QQ$N
zvE-(iHjN}WScvX97SE?DmWQ>AnMiqmW}CfzYnZ5NNFW*>U`DBsdN!#&yE6fN(zAm|
z;YMDDPixEVIPR9$y3q+xm$T9%8@+*jcp-}BzCxnD__jhzO#-22%4N#UV}=RaoZ;}?
znNAPI?Z2NcVkiCz+?b!lo!?#mjikA~|0e|T6vMz`0rG22O3CC8zf<;eY_d_Wj&dzR
z_PG<^zTfvJ8&2#4&sYx(2u=akA6bjA<(KtEu3PCI#Q*5i6%m}v=dn@>QqxVp`X#%s
zVK)W>q|(iBe2k|!{NlJELr&cEkVe;F_{HOA83DTuYu0VE4lod>E!#1hA%cj?{ijhD
z_3aoxaiF_Vs4BOE!1jguNkQm^r{Ma3#<sXH2(QHJAiT~iMz>ue3_zIA-|0?;xT$N;
zdut5ZkhTkBH_cnCzBT%=z4|R1_h%6^y{SwLof{l6GmnmYs>s=Tq+&SUmULqBi)MAY
zwVq9K_TB&Tuygwsl_3j(jjUZ)^#?ZX%s#T^|6HaO1^brSD{{7LhZQ$WDJXAiF|}rr
z(ONl(TjQyLo#Nls3J`?utZ=zs*i%gDEc1vLQJmcdv)gGPX}8786r@RQ3Jwi95=%m$
zM67**CC6T+<pF_qrF!C8XoYh(qJ_#rG?Q>!6G*4~XII~J%Y!Y*@V|+bAHQW%YYv>{
zXPT`l3ngtu13g$U=v=2Acwh}kb&%%AN?xwI>d4ZYB?tZhAYY8I3^Gp5W|3<bR|whv
zjt48qeZ@Ag*|0o<fJ~Vo?Fjr;5tpM^)?G;cl-BiEJk)TZl2QMHk(qPTG25sXl7EDz
zwo62rI0pX(zU$@U0Auu9YcTW`{lEtAJt~8#O4_}7CCT&upx+IZ)E8f*a0PSI*=fjw
zLC+#Y{^#F;o0a@NeL|Dl^@S(mf>8OA2yFa~^Al;pajgfWLjDL_=lu$Jo#vsBu?X76
zkA`bw7aXR(dND_Omr9TldCf0AnlwRRQMxqm2sv>S!N!AoFsXEUjmb+UFP;V#9|#Tz
zB2;$cYoaB$lNLFbKa)$3i3U)UZ#S0qH4GaT+eZQG74)Z}7awVjeFq}VkV&);zKB(d
z^G+K^RABCVn+ahhR)$Wvpl6Hgn)RY@yhXCPR2{=9%z4{+ZRmAd9(L_T!4Q*Q-950>
zb0yOOy&7UI@Q`kMDf?lHk&FWi!YI6vM{6HjOfp!ZSrW5Ch3d<Ei&i&}GrPYB2awlk
zL<2nKZ(=G>$atITjBF30!O;|4nm=TazksSS3-x>rF4VIA5MylK*IgTrj$v8ttP9GJ
zZ)Vj0)R)*bP>AhSG8^QC=_(xVD75hXk7u8~!dN2(emrmbp|Y#Bx{`8|!+3qZ<5Hn+
z-fBncsv^6xvP_D*%ir(H9^y?L>B0Gm5ZK4v!c|Ype~A{3Cv-7JHC30xfW5Nl=WMlS
zNU>uU=`&tg`NHueO`^uIn0QDI6ESWFC$09$n@%frR*aig7<g*6OKwQZBD(KSC)>08
z`^%mKgKifk)}=ezeA^1?U8Hkr@N2q;VfR0?eO}~N2g7E*n~2@vKesn8_5mdT__ne|
zwR#>Wx9|LN2`Kn_iey|j%*@v}<93IK0Sf>Sdp<wyng-lUyUpp|J;AjQZDJeJwA0TJ
z>A)`k_uV-JKHc4&pdf$jU)@ly8W;K}H!t|*{mYYK7S(N1Io(fqcLDyleDCt;_y1{t
zEH<P9BsOT91D@Pk9=8>K4g;&A&v#EMOB9Px1jW5;*SocKGpVQ12>KJBnwpxRn}3!T
zyYJ}FEiztIZ@TB*eu6V9a6+bk-^;OE?dD6;`-cSA?$(UWplhF}ebMLB=JQLFo9pb3
z2e!?i7{2bo$JMeI85Vvuf13Xyt^aw4fYEC$E|&2Pj{|q#p69gyt@ka@o-LAA<^=gQ
z^#7TcmC^I&1CXoT_T7AXhQ{%Vh{c*SST(OJJ}^X^_2%f}VR12{rfix3xM-$X1p*Km
zJ}n>+t=sLc-s9k40_RgqZ9vGGsj2MiO`!AEtt}|4x7(lk=4$@2|9*FUI{+e<Yw_zP
zD%1($<J)#4#Vv~c?Dr4|XfsqpNeZhT(8fs~0G_2`_QnqE&REg=)h<`nOL*L-_{|6Z
z-2UfoJGqXRcmWxcv*zwB<Y^MLAr>d`=Hx2mH^_j6T0cMO;3j&v+^B!HO?KV4Ce*a3
zUj>#57d$LJCcjkFh!N1$Y;O~&8<wxiRBLyRe~MGxynX}5v7WPsJV_`Me_{F<H~61{
z=I-0*9H`>uPm#o;*J45()inY${MsgvS})WEE;nrks$uz*qBO*-{-r1~`?KUNep%Bw
z1290}<Q+xB+jT>H9-L@W*$W<(8s_15@ne!;DEXT-*<2+NOiIc5+p?oysNiPr%CfQD
z=r>1q&c|CbREjF4S|8Wkqqyuzwq-m?#}TcI;ehLvMry7<Q>sXu0vwsYdIFtel<@Vx
z0k0eSD_Y=ierPy~Z}TxoTL}0*nKzF2Jg>P*jO#exr7Qa<yFu{k&2j<kk&K}zG2^sd
zk`vZ{AK?)lja=|WUB!r1WtrXfYe3WC)RC4<;K(_j`|lkq3C_dJxDoLSEcUhmC{TKX
zoQzG{O!W&n0uzmdIn=JXxEWax8PM%lK~)$`wEFwW1b4uB`n&p@J{y(9x&M|I6Y6NR
zmc74g!IJ*{AyKJaZ8oP+I+r!BXWVzv#NnT98aol)`*9o4$fOY2*YMs1mBu4f#^rXH
z;AF-GQ<@VCT{ZI;V+7^;_gMyl-!c6C4ZZ4ifqYCl-DOS3sb3|zEd2SCcn^%>-}_9^
z{=E0!XOk7+tWLN96im2My3f~gXph_%WMB=EJ>4bmt|}XA?L4FFh7=2E=20zpDjX-&
zbEKlB(99QO<0CtbHZi)&#}x+uDlc6#bR-hpkBmEbWz@7HGbvwj=H2AX<kazv{f7&p
zC1F;KF+b^ai|DDT$hPs^ae5Q}jYlNR=RKz7KcC?BFy*iyQqC;U&=rwQLl@T3n|MFj
zXc&`>!}hlfsy5_z29Ex3@I<mSu}mJAL9Ig@B?`W<ZCn2Ar<mx7ToM5dd*k^E+b&Gw
z<Wqhe9(-@=aZ*bJ1iR)B(M`>=oPYJLWRl3SZ$M{9r5*U;o4&-b8%3$zAb^}v-G;n^
z;}X#0l=KA|iRCOKxgA*xc_?G29hsfwe?xnIAR!SU{omU&1R2T4)x(AloZ;4iOp~F~
zflQ6uoMF>}EQ?%~QQCpbMDssGNe>wbDH!Gdz1cZqw*y%k<)$Wsxf59$89&3S6PcCf
gf9&Gq<@!@pBqYNByDcOjqp%a18Ka=><t)<w2h9|T!2kdN
diff --git a/other/cute_checks.docx b/other/cute_checks.docx
index a6e3246bb833db9ce19a4c81cdcb124bdd9d2063..4e0f92da7c81c2eb7200dd1b78f678a2f0cc72a0 100644
GIT binary patch
literal 110509
zcmeEsV~`-t)92VWckJ0QcWm3XZQHhObH}!ATRXOKv(NMX-?)poZ};)N&WP^n>Wb{n
zto&tWb(X9IFbE0&7ytwS00031AdfG97$5+E78n2kG5`dShJcN=qmi|vj-s2bk%JbE
ztCb~wJ_rzbE&$N?`2T<W57t0Kl9)w59YWAm&?k6-MP1Uptaw<0K#BpjDO4xrEgRxu
zFw^R%mx!7qg_vggfqt#YmPgL8=>|4+)YK6|zI~J7W_~|I?VQq>wD=>v`{<P#n%s)G
zsY#w8RjB1mch_l@@(+Dsb8HC~p?mmD5q3}_{asGg@&y_UKSylR4d@pV(jFU*EDoA1
zq1se~TiuT><OTg=dJpZneOSoUbfcJzE)*pOqj>MM5~%G8$lQI}xNB`Hir7<6>k1vi
zm*jH%!!|PMq!`u(TYLf~IB-iR8>XpT%8_8goeH{zL3A3$4tslJQ8k?A)&4)H>SCt%
zqm$rd1j4V<!st?i5JoQVZ)^wD`HVTj4nexXy)P?Vm?iX5-UzaMjQs(>YZ)1XdGC+$
zgbg;K5bcTxvU7Cz*MGo=61RZ#EVZKQ0q#FV$i36qfR<6X5D#HZ#muTU^W#_1u4u~e
zpju=Y#4WUP`!>M!^JsfCx5xi_^7-W^MW~kZ?j5_N54xCb+4aeO-~El#uP<N#+5ZE<
zKjN^OufEMv`Wv9o-w4*RH?nk~rTKgQ|H%Cx4A=in^vd{N^L{!Q-m8F*fZ0yDwO)(@
zX<CDsb&OST2n{hw<n<Nv<<Ae!l@%bZKYAkLbMpx^?hcv4PCJQO*I3C)u;HDMOE22}
z8gCu007*d|Mf9#qwtFxcM~}uIV?<&UV*wFrsNvICU~$i(sgpg(hm?Ykhb7QvgcQ>Y
z2E@M$vh9?XpHjUUbJOS-m&`>qJiyDjgt~%p+omwRp^SOzQW$^6#`JB~YYix#CbTd_
znNvJaV_DMCl4Qo|S3+T3cg&u;7j%~lh6PbW=0S$hzIhsJBB_5KhaX|`uv`(SHR$Zx
zU}tLa-Thzj!RA%5rttkl7aRZp2LJ-l#m3%{_Fs?~+88)leFwY0QSLv00sIbi-=qI$
zA00{K-v)yq^c3_NFx!=!rGhswT8vU}rEsELH{4W8f@wC#;<KGc5G{Zl7j$i3J)6~^
z700RbSxWi}bXGWvDX||Yl<S}%s%3QU^<n?iLm0x3o==a(&;&f39I>A9Ld691C$_@q
zjL?3T#(-J>ZJ>4BLbh+P7*t9}+#rdU(rQp6&L<q|nOw|x<B@A+Z-7x974vt6|7(rQ
zPtAR+G5Mow%iAiTC=+$tvQkz|C=EbkNj`!>ZJgm;@u6wP!u?nTQ4<dYpF?d)1BP}G
z#Khj+>O3)LgtxWjPw!Or$PNdLpg{7dp~e*^u2|_?5=0b!n?*3{%_bs2vKRchScI3Q
zJUP6Ds({e_f-w^R+sh-%)I-tEaYUy0aRf%Z0Vd*pR?97bIF>vY1&@3eo_M{v)YrLG
z5U=`AAv;-5c1}3t{sRPHo)=mg@Bgb&)<3O?bGm&y>;+f=#BW>s$0+}Ib3E5Nv)*S*
z{_O7Q0^S9~GP4}3^Ngw};S|MbFRp7`460t}1_1my-}EcKwD?5RxbBK+nog#Pn1nsv
zCLWceEsMV(K@?5uhxhRt9L0}Y8y}x0$$QaM3)wdXsR~x}(3M|VSdZ@t7sD6N{YJ2@
zIO__Q@)@4xoe0g3PrW{$!`s>1(q&x@p_6gsea<}G?5`zj{tn0cgFzLvp+TE%MkkSN
zjzed^8fi`}jZY5t!s`DB{mx@vuz@T1Wv6xHC1A;;-+Y8#!b3)7HFllgs*1MKHG`gN
zcl)j$>}*Sr74bQE=+3ufJYOncY3CrE7S#HjFiA^2@gsQ(T&_Bxm7ZQQpgWWhdovf~
zwXDhTQKUN*zM_0Ygxg)|#s@^EmU#P&%YNS_<aG#ISW*t$ydAn<ZFNChBfX^+0?KV;
zIaFjEhWqKj^r9JXZ|E#(87M(<`$~NL-n^?@uqrNB#1OH9EMAe|HH%M*+H$2vCv%L6
zVBb8eAIj7kWyWb#Pss6|kY)q@Y>_!T^JU%rg|R^5xRGh!s_4U*r*bUb(}`D&TfGm$
zd0RQ{3i=#ErRhGVd-7Y-0TgB=uNc86f<<6acQGsc6-WLO+S`(VY;Gu)?9ZFBCDZ%r
z^!`h!pDp)MqY+igHstJs-L3-0u4cEBF+K??>623}jLO?Po(8uedNPCHk_^Z(j%{2o
ze?Xg#!><pw(QgGN-l(w^GTQf$R~BJs83o!5!Sqh3+~lXVRxgi~HlwuVL8Uvs;I$s+
zKT4iAh**O5G~J&zec%qC1gq2X)q9!seV%W$w7agUTy|MSKR0UEpUVqiYmnA|zTQBs
zOPQJ;*S8w8R&Ad)=tN=GYuxXTS*PqrJ8#0TUkH-2=Ti*iXlP_!t?_s%#x091_qOGt
zh9IWYipXQXqujxd61J$w7T(nV=^r<4Kckng&ftFj7`{l^?uvG9EP}xe+jD1_dnShK
z+;*uLb+foX0xaP$f2tN<KS%p;4Z3h%G&k?8nps-_IOg!77rW+a{wi?OnYtNx@Bw>D
zrPr}lD;2v=4DCvle<%?Uig=5XUarx(K2avx3RjhKjW-I-J;R&xt_Shg@;=03^K>}S
z(ZF;+OK2MxNU4b`+g;}<b?kpzFW$x|Vti<qmZBUL-5i#X4xoXkX3ogSLegA6Pn7+5
zf?j3Gumm4hKZN(2q;i;?_#+FGG<({-jxK^>&JVH_AY&{?+waWYPBPt|`tbNf(58{w
z?vu|EZQcxKBKW5Crt~rQu9T~u+(=2>?tETbBY`tPN2Z5NJAiA9^%!suq@g3b%lXu-
zXw&#>(z^LLE_T;`OnczTHN9mx=Ofi^rNz><8mHmc-k1)Ieg*AOpJe$jJywXbhc@TD
zZ89IEEFw#TyFum+FTOCHM%kZz)sKRtUv$L^`Xb$k62Car?ZW&%muSNUE}JW+3wVSL
zP%hOW54C)~G`ys(x~+J<J(l4c3<}Id9>Tc1s$PUfz~PrCyfyA{twM}n*J3rVogvoJ
zW30{RipvLN$Bmj6>U~feTpAqw!>vl6j?c)8u1c>g&9={=^La8q^7lTUs$EvR6uDxp
z^_RIsG&${~CH1FY_HYoZm!}c42+B{eO*y!%bbzdYtxQDMKVVEbvT;=gvT(mHY<Gbm
z!}O+2;T!`Iv$(_g;!03Om&ChB7)xR6<J4JR@v$xFEc`7yax67jcv;}$VMm9d{YyaO
zVsJ)+8<KJKuR;0j;c(@G7dmP?*yxW+-GKXJnomcJfE_^`i3M<Hc}3us5Tjv{9f`}L
zMJ@cIaqfjE90h@K0dYrVMHB<RAI~AS3Vvt@5QyEf6531gFlw4p@ww%Tw(}`W)JX*5
zWpY8m=1p*kFrI%AT069EEq{O(k~B6j@u~BwGbARXiAwx3GxKlg;8Rc6Gt;|={D}N{
ztWv0eyt+cjuT>l@WSLff>~3?wXE+xj2`vaFNGcF#7BEtXjH$2!V@TRLA^O%kpyIx3
z8er>dYe<fRix3?USb(pdw{?e*a=Z%~UK)(ytcu+K$dV8{=eN4C{wRRbM7%{pG4`wS
z%snB~kyn}$pDgypHGg6lqfYUS`+)9aGGD>K_@tKn(du9o^tKYf?T8g^nytNwlfu^C
z|4ln3gCur?xKHn(b2}-ox)iJb7a0s>HvM^PL@n2Y>bubU>z<DA5KLL{D#HceMlQc-
znhxBxr+IdWqRGw+wHtSYqDiWQ8~S0|)FmdOV0J_YyBxf}IBxH$5?wldUPP!uLr;ug
z7o@g&lRIW)tt|fiGExQ@dH{sod`@y){k`eQLBPfaYQRBZGLGmx*yyOdJ-*Z%-DsSq
zH6jZc=xjF5Nfbkhc*1p}kksiLRopvBm;;krTn!MlIgxPAPN$tapC5=dLLv-kQH~Ih
z3pF&FU3K_uQC7sRe`N>`^@@ZAe`=;jI?`rsE#?Ftphk(&8F!%=u!hO#QY@*L>a!01
zM=c#XR$oxa5wGBDl+PFRAJRp<Y?u^$IaXs%A|)<Y=FV3^C_umW6DL43Ki`f5-%>C?
z-$W%|!HI1}iE8^e16`i*7y@b$!az)Se4^l=I$E(NIFS+^Li$@;Gsz2UK#+5}Kh?~k
ziA7V?HA)kJ<$9|nD#KjoNcc4aoyiwx!fXKpXc172dJqn+7=S2d@QKvMAPRZ$NQnS>
zupbThP)N`t%Aey`Pk%NOgiz4j$S-enC#QlUYvU7^34zSz@d?(WAc}@alMyoZ2XYx9
zSA;es&mriuzk~j)Wf$=0Tu83xm(Zq;7@37E9g9;0t<UFEs#i>+z75t3KGQO85lmaM
zZ=4M7KhfN9juO%DGAyG*7QPOUJcWpdTTSF0({%^+Tezkqhd8f@8|}jS*ubUwNYgXX
zJE?}~wBjVQFCoOAUO0e!SwiqO!SwE*ofCm_&WWVZInez)7_BY2hb%kIrLYT4Y2~%L
zYBAdzBQ_Z><X!tChRu-d*YKOzYDAUhA`=odFbfzpZjl$m>+>dqbAB>yYGwoaEv|$X
z|3}#%aD4(FXFYO^HgI$bO_eL#MDDpZ*ppviOP~n_kvP-AnTBN^WSX4t!c4$8;wYxx
z3JC;H1VLC!05L993lf_-SydCnHkcKl)u^lneo9-z+TQ8PgaB5fgk)<~v(*Vmnk&Dg
zj{@8-AT+%(qJPQ4;C7jS^>4!y`0C8BzP)Y}U>0R@TH08%P8wRP*c6vU^hNyGl!4p)
zA&`Ybb5U7@s?neUTbs6Vyw;txjp%o$o%~?xNXamZ2`h|t95GG&MXUi52|GW<v$<$3
z$Aq?4whR6x;8wWV!#F{U8bw{*cx?{+?l!13KndC3s{U&hhH+d!hV1eEdU@`(K-Lu|
ziU9+FoQg}NM<|19ImT-_A}aS}ZBk4C%g0hWRU%U(5~4sM?ZS(L&r+X-CgWW(JIjN6
zR{mgOGdL5h1nE90_F8LC(NC$n<sU&W!%Q9{xOQg^5m-5A4RMeZ-n@ftxrtFL*-#wc
zd4OjGc3&>=<cy<ROEEoEUf7tP`#_kIF<<cZQOrmuHvogb|6oE!tliI%aA_>%$Q5>z
z>JpW2zbh}YhHlMXv+WGB3+Cd9G4_QbO6|$|(~5k{?Z@%^KzihjGm>MwmER_<GYOE6
z+0uLZnrha>WQH%7zW!Grfl6rPGer%gotxiB{Z=kvEI?Z|p_XjilL8_v67P5=A-BS@
zD1YDDjgII}e9MXDpONMOEkQg`#hZwYq+o-2Pe0U!vLK;=a~XtW>L8Mc^J!IeO({n7
zh#@jI85C<oOCmXvoPMr3mSR(J(8owj{ezmK-Vu}$4rVt2Ym!)(<bx_KkOMEA+v4e6
zDOG#k9q%X*7n#~mIul1TU=O^wP69W?6%L8VDMGip%JYj_O`LHo+2%?(qf^uxJU?KW
zk|Q%w6wTv5+}HOUw~j+)jzxQa%&G}3wOWoc8GthwsviTa*lq0f;>!H2@Glk$VJX6S
z*VN)!3dw)OG~mHf!ekL`JyL>%QUCdad5h?m{k!lulJQYoXxiRMIC^ANgSwvGENYN8
zkbdkS5yIBR{W6$XaOaY`m1CcHK;G`jU)Oq<Q%dWN8L9@OuJ#S!3MZy-Gj<emBa6gE
zfb^KJT%L~E@1l^BkSKOR9TfAs$*?LdK2B)F8hOYIt74Eo9Zw638wQTKbXbqHGqtb*
zS(bUWQN}DTn%x@Jo1u1KAKcc?c}JFssSLNbf!~E1^YgGZHBud-pgVcx2yA%6m3R-0
z@mdgthOF195pl_BY~cqj13xKaKdC|??F;y?3_iom^v#S|8OZZ&iuQiH^r9?1yU`88
zrb5T`kVQIgr@*DqS-d%^w#;(ae!|f=d0S35IgY3@*1Bx|RYgtRUIz)9@)Ix84N2Mj
z1#ytkyd+t|)WT82_~NoQDwzbPZZ(EOa0o_bJ$`P|z{|AY%h*`ALS8jZkCFb_DCqo(
zASxexH6l(jn!ZSUB{fX8h?N!D$|#r&1=M$Lxkr2xqtFjiI`eT}400~!B7sv<S@+_^
zY@tiiiJ^{|k&bK7$($wE#WhMceB!bokU23u22fH3X_mBjyQcFa!!cG;LVHI);c0U`
z)LBbDcpzb#Ukc+;K10cORC~8s<<RTX#m7n`z(D|9^8mO=d@(3+jD%?n;BmDDr5d$i
zzj{LRD2^~@M{@+9O4Lr`IV7Tu!7?rG!H$BdWpytl(>NlSX8Q1;M4mf^G|{IMWOaTN
zl7Rd7LFh1Gzc@P_=DDp1AE+tMC23G+-+&&8rmSZnyJ?-^r}dTMcQ)}Ro9i_c(Oa8E
z-K1YtX~95GwG?Poyra1tXLLAj{)tjoDZk~u!G1&;4NJyM*_(N=7Y6kR^_}Xj)kxr8
z40ara>6=lxZI~;Fs%XRREQCt@B>CiVT%1<%n|r5)kqL8MW)T{H)A2S>ZYeNmXIY4t
z<tRtb_`6=&T8zl5E?MFu8|%KC3!yJr0oQ6@6#TmcsMxv*oauI^H*7w^HDdjeAF(E5
z7)6vq)W@qCRPSCi^PtJxUK@^Xa`{*f<+9(N>U|zw-o3Et_W9_NU>=*XF(n$B@);iB
z{W+IQ9Qiv((>GB?)+S~>-5J(--JtoVmg~IqJ}!21C~E2O**s%2A5X>kjyCALi6u+d
zR+#BQ`BX`QG?801mCN%u)$T^jfHFLOa7}fJpfVVAGu!6K3cl36gL~%ki6#Sg@~!+r
zru1g-6?xKo{c5aAS(ZCfQt%N@^=)!!aR%9o`*|}XeSn2UN7I1oiKg^b6maHulucI@
z+<nC6oXWQ@YWST>5r%pNL()fLP=qI)c{(?nW_TF^VM{^lu$gw6XA~A#J9rTvIxJ%p
znLlVEvNjSNOwEfXSF>k`HU(Au-KuYD=gnQb)@GU)ec=}v@HYplJ)&DMPO^v+Rvfz6
zL;}wq28wcQetL=E2%CJqCAr?31G?d^GPd&{VvWfmtyn4KR_K1O+^nM(sps0l@?8?c
z-X&HuMp&cmERjKfWW7OE+#t2#dUZ#^86@1-9b+%6tF&YNDB5SPI;=XJG?}IifUYjh
z{L?)p%9^&1UDYGCj*HeYuQ=t=XrFt5bU)`pvNwPVE7zO4HfYE<w$!@9Wc0gyeK2)0
zg#pmqfG6GG<(yr8Y<fHkX~AG?iVNIAGwkr(v@V;+%VG7bnGzAaFu`-1d(bop9Zh;>
zn)K@dm;p8=$Lr(fP{OejR2R6`Z-RMqg=Mht*F=u~t7J{x=Z=b%!Kwc6gzfAE(MM6R
z1>}a1Zl^=hk_ef7%E@r28P*k2i-cV%Q*+*u#xQ@-8OF8jwJd1hoa@O29*^*%gM8?(
z3O2%4AVUBK6o%Ru=tcu3HL`0<-&qUS^#LHddf=fcMe%#cMP+DM^W#}Fs~!obu)s-B
zrhA<0VPSc&9Rk|-8=-`I0*7peUL4;<PP*ZgVh|w^*sJwY(fQ~F^01q>%+|?Tt*fJ@
zor#?ZKgMS9lL_-dN;U>I`${8;gOLlQU-g;}(k45jJEL{UKjX+;#7{=iviMD5y)`L#
zE|X_sI^Sup4X+PW$CF1AJV@>hL*&5UG2Yph>Pc?*#~G$H=MM*noz9nFGZNVKoGjDt
z!z6oKds_{DY7ib@rp&aK5lvSGwOZO9inrD;k>}Aut5v^!QWv(-Y_kRE1n3N_H%WR|
zEg>gO)L3aQ3xU?@1#kp#{O!He97@t<Nn6G{M1w+b3gA!~&l+z*lf>Nt$%e_su~=nr
zv&Tdam`rwtR569m-%Jc!<`-7BR{KC7SpoJW_k<bvPWv<3S$t=JjVy^Tr0}soV$XrY
zGTQ)S0Yo4D1*IVXL;;9B_y~v70gwPjwPQyz(EuQQ4QWLU`#^j_c<a%L?SFFrS0j)0
zzZ$)C|2E7iD04w%>+wkFj-klIL?9*GfW-IYfPPo|#;YVgSGbX`_5&sZ9st613(6>f
z5&L=u9tO~2^v<_&A!ZWDFa$gZ0Hko)+DVR`ZL!?!SI};Y0HFJh|7eLd31Oh=LE;lD
zt*b9f`Dj`?K^K)rz5UK_BsamgV7D&#w0ZL{b-QBSfcj%xTzEkEH`?2$Q&U}BSO>gz
z7FyfK`_~-VoXNp$+27yGke*(4j~&5zSIk>Jn;g(T5>}d38DH&;q`u@AD_eJ!Nw~W{
zOQJ$6R;nU7bQ=e>!VX%kXyn1|!}<_(V(Xn~Sfo}fPo3$F?}~&34aAq~jziZ#q_h`e
ziab&D4w8I-92P^~*!87vB&n`O;~<^Gl~b9uz;aV)%HneEKMbePwm-`^RWF;zxRa#%
zKD#%vK-P6;fnKd)ZKZkX3{0VX6lWEfXSaCaqsM5dmlTn<ln)LD{kgjsK*_A=n6`K#
z$YYD;**Q2#2qB<D%~bIbb+W&w7x3^Eew<<3i9OkEajRl(cJjK9K6U?^i>{^}E&siD
zU+*$@*LIs#{28Z=Q}(&;EKd{puw9$BZrv6Imya~_^I4O|lV+n`s>*J$Rqf%DgM5U-
z)8XtfV(?Ef2yKte&}PPsc@O$y-wEzv&?ku!gsUER>l4`ldr4xFAZJIgZ;KvMPkwIg
z%HWdFBo+Do<>s83M2p0USpGl&_#v+#uV4v&KEowPb;e2%GKg=WZy*#uztILmRySMV
zlyA^q3h(8&&l#w%ixn!N!vTf^f}?~0AGKl|AL1jeJN5*m2KgpM7_|7ej^r!4U#9@z
zfZv1~5OBcKSL7ieoPwONP{$dV{oio&OZEMxa6U=<v(tJ+WTzHBD_0XRCJ-h?1T6WC
zf|RHbK{mywR?$Gw2dj$lh)tUwu^us!qry(AGw3lfmg=*7|CL(5aNlo&;ukW~)U!%(
z`;fDK048G2CCtswmzBd@R<)6$p>~$RjR_*WfTsas24p^rP9JNzb#H2rNDWRV#(Q00
z0^#U4ga<)xe}~tXfh3x@ah0yd?1ew=#jSjRw!~9pkG=M5Au83fvccSbv*3UjyaBVM
zQXIJ+Tvy*EextNpy$5?#8c{1vImZ{k7#(oo9MDcmS|ldp1wx|q4ZJ|gI06y%l`z_k
z|IhUVMD(J5zP*()wy+N}c4u>Ev+=qv_U)a6shY;tz{oaG{E7R}YS`_E0~{J0<`^Ij
z9M3L@^npKeYqUvA)iH3hb=$?RX~ynO<x1AVzu}~8pk2sy(qU}Lw#s9x&2l&pzv$$-
zH(5c($=Ft4_$Ydx!A~YyY`{Vv)uiQQIBA!^PDZJ(v!hkJGDT>+x(d<kt=!9LejUzl
zkBR>1;3U7%q@9kKExH<~vn+@CG@TQTsh8JCCn6w%04l;;Q!dp<EC~ihnHS39UOk2N
z%z__kdB8Y`-16eSMgH+P#QxzJfE<&&RX=f~9EL>2H#+FV6t7WuR27r_WzrZ#oJJOT
zJt@yHG(KYG2L=7r&?wzYx(~0SZvscpXwj*RaOUsL3u@B>bCpI1#r7?a4rhusTDAHM
z_3;`k_2xxNSr_N=tKeCL{O2DKdb1gm5PEW$<mHKavwuU=F<+A`p~&{ycvOb*n{h>f
z2v0T6W_O`@b!frcBSkd2B#l?7YAWOYi{+d+bkrsmzzlgY*u9sj?$EUagCPzEt|cFs
zAUG&*@X~c2fs1<0G)B|NYN1y&DXuPjPiPF{<Z$Skq-j2+BLF`DfnFe!SKvhL6a@#c
zB>;=^is{SBj;lVD-E6uqY)^1ZlyeWZghon8SEe3|NKWkI_!oBs-t!j#g>Pb(0wHo$
zf^LzwgvC?HnFco6loHNF=;Q~$hZaP`dw+9PZ~#g(B{{tJVwOi5cA9R)LPE)PlAS@a
zJw#M9pNf>fP|*zO8ZYX#JS3sq&FLCpwB)sHJgiehREXCb%{b<gCLK6F3H-zjN$-mx
znU!LpnTUC!djx{dy)gR*Rwia9yX_GOnJ4O0niFz6{V|L`Q2<fOcpP$Oh!UH~5iMLE
z<A29aj_=qRAx+Q###5b$=OS@3{14jyjb%t~_BDao=X9sOcma6d!B5cJQ*9{CqAfGE
zFnY61&w+ugGFmEuJC1^lmCfaE+}`*=`o-#WuqU574yIC_NM#X)^jaT#*mgC|Uw(EQ
z#rVwp>{Mwqez0uLB{%xMHsqwGh=9$~!r0>Qw|B}&6Vf=rO>097DYUIOtK#sx1Nw-r
z|LDK6fCRD>JNgw%vlQE)NBrKKv64sik&&^V6*X65P|u~*6%ic%i;K#_`1b&iGAE#f
zkJgFm(Ne^ecCKUtS364{J9%?7sBb>sY124+`K~&RKd`SPO(O5qX5g%xx11}n{dgB8
zIWXHVw%4)Q!-c7_+3~AG|FQETq0~GR_8g_E8*W#38cwYvK>rf-ZnEBa8U0lxx3&qC
z%ES^CeSUCG98S(_+cdrwcBk&kP28$^EcDIxMMxPXe0sNB)N71u)o%?A7%&7Nxx)NI
zl_Q*_2|`rNdvnv~_Q>07h065`0lmwA*}k-mkygpi0L8;uYxA+DMC|rVMr<<J-X*9u
zNm_TQgM(YceUr90>$Ve4l;)wTceBbl+U<vU<Vx>T?-SRv$DfK(<74sNtK-wquJ|f<
zEMm6vhlNj=u!vTGK6ZS-*jk0>7k@;~s`Z4_m1>^L#`R^;z7LW>tDOzI;GeAy9HfU%
zR~Y&}yuh}AE$4yIZqDTUK-Bgt(5iry<drZW+^29y(5)4-*ZZ5B@POz~mzRvBN1L0l
zc6e+~mzPNAg#{1mRdrZ<Wd0KO`%pBWMOwDA>Q=rIO}6xP<6roXiF6Df!N&*cfhpgP
zSkPdu3IX#8i#2N5EtsEQu!J;;tKyvH2kSt0&yYy}lyZc2W4ij46Qt6J>nz6)T=@0=
z=%#VkIc3m>O@@}8aP^FXDlpXNQ%$F4OeS|S9k8fWX8jrA{kB;*Xbc`5vSzK{cFc-!
z81^<+3zknhw3i~KI{YKKQ!bzS@wwdJ#fQSO^gHn~WO2OjI$D<l^&a*txPqRATII&<
zrNYH$@@bWF$?C=9YuATB7M&WVdC+7d6~|=jvTQt*&+V5`=#81x!N=MjuHE{U-`jKf
zs+B9^Qh2_BIf*dLQ!rNnq(xHg%7qip8>krYxlP~SwF(zBjPoUXoplA^?4N>bz+W#3
zCpe>E^w8tT088Q!GxYM=fAso8WN>`@r|?>T16R1&ekwr2rJl5;XUZUVh6IBA(B4QU
zj~3|%oyD8E`uafpL)r^t7k-4I#lFB)fCH_|Zr=geBw^D&rDRiB#!?cs_>RO6CWTp1
zmg>%>+<t$7a1=T^Cfh}^&8`gH=&Djl8c+e+%K;CeNp*D4tJE?EtVS<r5=sjZUn|n5
zJdsM<@^7VJe^}w{n9&kUZZlVBKNb~8?tTgmL5YAL02w!B{mzUKBHtcUPzjw{xBV7O
zt{e@UFY#!dX8|%HDbG{kh%t0;h0Xy}a5Nm3DpdXw5Rt!h89&q?|8%pJ>)H(>AAq_}
z7jPYS)w$EI-Q#x`^avkc;@^3q2q14<(j^Kte|>~^>N7MV-%h;35e!n#JgV9)owsOs
z+uA?%%9Z|kti8@*f%-a@0Is=v1Z%iS_g$B$s67iKyZ{?T?$RQB-<eo1-}+Bkf|EY1
zJ$t<o)0g6!l+Oslcb0<#KsD$DNYYVLYosI$A7I=FKv}ES_afS|9p~rz>9{^CP;Ii!
zmn@m2a`3H`TBVW3I&kgL7Qi{zvZ0W&x)qr2q%FYE3GyqZ##D`HD$eK>GX2UzZRj<y
zdBis4^4U#Zs(MBWNKWkmgrwuCER3=dA32zA7%240<w_H_Q%#MMyylwh1TBH#Ok!T@
zQtYAifWB2XFn=eVn%Y?NA;++Of!*D!@fY@BzX5%#&{$UqzcaTR6WVJvZjoiMsTgcA
zi(t5qLRl38a>yzjCy|SD&u^qU@Ko|8!H`m6ZYhvmHJbk4r~aUpE{5w-As1(H<_>Gu
z|GckZHw`H&>RR%!Nk~k;zCO`r+y!o$UrIV@w`KC^LeIm|ZALRS{2WR7P6`m0j7ElO
zki|UuwU<NsWd_@doWCbbuL4F?1<dbzL5o)Rgoa(+AFVHRmO^Z{2)4yK4(jQiuP7<?
zQ&w`%f}lBkYjsyyDKxD|a9F8*R?1`Hnq}X&7O77hV7#nPyjE+#xoTXvpswy`bPH#f
zxT$P+U@TvAU{Z|(n@69x(xtF!S^Y;oG5uei7|qo(?A1Z;4SOxg6<Nk;`sY7gnSGkG
ze*bY;7w`M0qBHs!KeMpcL)?jV+H;bHi>nIi1cp4AE(-H=U1f9|LXfBW{746C_2)(G
z@5Ur-Adc=q^q61e96K);i%C))%Or^-)p40m7j00ibhk^nV|6jE;d)un9W9xV_-zo-
zv6SL&h}~V!DRJ}yr*D?6PEjp<FZ`VQNysw4%=sjdT5!T6q1;hYPojD?zbBxeSTj1k
z92%{5(<3q+E<I9)$y^)-v&?2AH1eG^_&=J@j)aGIm6sfXAI_R}GHhQHL|k)po(_fZ
zM_M|U>ea{(w63E60Y0vwG#N^fX}nG3ZHXnML9Q!hF!?-e7k})sY=*D$r>hFgbO@wU
zMh`QO0s39^pnSK;s3NEzBEl8(qj8|Pcko$bW8-OXQ_q3v9<FL9K;$J`-wStK?vw)f
zSww&g?+2vAa#bq{n+jE7JRKU0ORr!F+|*xyPdQ*_s{92w(YiG9km*Kc*acs2>tC3f
z1>9A6_~{(k(q79ZJXu8<BC>2_#|4eCyOHDSm?&vAYQ>}fDPtyl3-2&<!rmO}BHTw~
z3!!2*y?it5-o%K0zrs3SYcVsq06R9lB~;l`T@a4sCQ%jX^tiiWPJwh2`Svwez9$Iz
z0(5dn$c-GG<6Ez-${yNfLfn^=18yn&&Jq<;8c*DpZ~65x8OWVmmn;oSjl)VIlFM{$
z4Xj7zaO4LICVbBVPs`Ffz-k6B+LPMq$bQrY!Nr2C)Rtg1t9xl~PtijbzN;~?*JVoZ
zPRs-fr$zKDRw0|Zk$?ei7!!E^3VzwPiTk7tZzoBg`Xm8?$z;`2DA|ojcW9flFPkvm
zfqiP4HGWp-{vJCs*{2z3gLBu343gKD#aXxxm0fEG_A%7`Me-bb3NI5Z5o-#>m>p*>
zluwd6_7)Go8aR8B1qF_;(Qe|h1Lh3O<<|oX#0s$yuG>B&b!J@pXeWl<@<I{_lg?th
z-qvidQzRq!AG?9u;Ft0~?)er<YOyJBk;CmAEiHOJ0n3W_LZrSKHqQRZn4IhchMMdH
zg7F^Tp2@PcA(V4p7iy+I!o}rcyeTkY>@2=TVZ<Tdg%StNFN}8v-#&2UCac%}=Iiec
z=C7~wENF)U(fX@ysJzwGS_3X~KzH{kHvagct*Ep4_1Tz~<01Eczry+nEJ$45&BqP4
z?T-auHGDvR3`^8^R%c^F-q5jEBV}ddxQ9D?rw2S9T9vVBiD71w@9j^_SYBER9O_mM
zC+Qp57USzYcWJnD?NP!78vCp-=e1v{<CvX2girTWRn8$=AG*}XNA@(IpFK481C3KF
z+U_)+eYPdVFCtc^HMj|EV`R4pJ~iErPchP)*96}EVb4!9hhJH1$rm#cJ{V{oBYXvm
z2q~>T*3spglW_zdzBBX@+*d~!YM+mGJ#aqPIj;fR@{ikGt5G!?xL#E4({LQ6(GA8m
zY4W+BxIjQ!UJuzn>)o<#ijNVZSWWJT8#bFhTIe!|YqED;@5LR2vF8n*w6LIWMJ_Hg
zxE3Fhm#m*0yMci^>D=NHX~(Ms--+sOB@-8>C9Y1<zkJSf$)s;|q7ksZ{BIj-X*QGS
zi)1?OLnSit3Z09Hy}a10qriwhoxV)qxfeg&-J0n-I6RVWTjjUdUjIx_p_2L!2!j~e
z8nw`<NmdS=h5YPXeXJp?NvxLeMi{O|p+p+aL|S_*e#%{;a-U(mIGoyKt&M8Qe=B`q
z`yn3+@V3iHf1cw{KWfO><uhr{h@Y<lCssm{-;5E@a!P9>@f@26gEw+7_cHv|s#uIh
zk2Cl?skQ1}rO9ZOzyL(m{!s_N{k;B+8C$CAgRmW@E5|3qmzVL?`yPv=@5$m77vUsm
zG@{T`t*h<|-PAtQrQU97;xH!Bg>plK<9z|{W@$?P=7;EBRQjf}0xsERXj3{Hjoi7%
zpzOw%Hn#LFZa|FlAZ}Yp*7(y$^H2GZT)1m&ZsdL&)J)`1@@>6CCLxfo3*wLF(>|cp
zL+bY8YHgQA^i`wl?RO3kkn}~L{g}Gr?#6dGi2J?+A^9%q?G+zrrkB_+XB+h}f~uY>
zD3=}o7_Y$}ksjuTb+_3`;QehX%9V_P(+?f02x#c*t~zE}luq;0A)?T{sNBfFN~oM*
zq2xLmKIw%13QqCTN=|xy=l02)wkwvVuMy`AY)XGEWb#z;=IxSQ={tV^XVzJr%Yme(
z;t?*iDmIVgqETc*t3!XQdQjw4*uq<RZk{?tkDUsCEsfP3-@`Az^Hm-<bf0KR;plU=
zEquK375}&}Ur$3cIbKKP&S~nsjwJ+1h0ocYsEC?!a<DZQE;J1RBlHioM~}m>dp8n~
z)#%uGR}ML3DUQc}DmXyv;uN+<Bx+1IE<F0EW0ki#z<CKhmz+ST#k6PrTE_-`cB$e1
zHFVmW*(PQns9~dkPWxv#F{qB|$dE6&UBGO>win%?=ueaZs1BcQg3toq*=8}uh7tNK
z;+BgN#TZS#0CHlbG3_Eq(AQq}>RkqTrDA;$m44QNjKdL^pxOj~F#x2_i<n)5&k)P(
zRjZZBz*caRNLn1K+7$WdFsEN9=iYKnaq{jZEa@If7_Q-5lYfwpa&ne_<ZcAFE-p|g
z8uO5%uplCp?}yITuuSI2519hAS}soZJVM*7Vv-4_PP3i1wpw0i_HpY<mPTjyZaQ<f
z$`%3MT(yrrwDzlPoe`Zj2<)(b4bg7r*USN{XTorCwdPm}<$ww7hw^o2X04k<>>{d_
zpFTNW_m@^)=kN4QDhI?!*}(aBhn!HoWLj7;VEgNb_?a{eW2zf4I6+5I!QoDfndSCP
z*)ce7Y)_URAR#Jc=K|%Dsyh~wht+>A&}4u7)5Wt%x*uI|MpROp_;CC-%BzVt;rFxU
zMQ}Mh*8AD;vnADVLsokzg!0Dk&Z6VT@@@ODlr4eGl8ZE|BlvoyvLXJoSdVoA&i7-v
zLFqQ3-xVwzqqIIbh{LvF`kaU=c~43cMQ51f6ztorHi7e_DN}{B1HN-X(*xvMsGO8m
zE3$8KRGR%J0|%NGQ5t((b*_l)NAG^057?|`s6HX3dCgvzyMBGZpwvQcxv^TJ&Q-=1
zj+4+=h7Y_S|DXu?Fn{}w_d`<Zza`+B`If*n*RQ5~%+A2NjDwjven{iI^i)<N(y?uT
zP(em5ef|Kw>B0G}jRNN+l|yYPP_@Q0{m`AB0PbQLyk4<1Vhm?^A?DC0qZ9rxZGLzF
zRXzkys4gX(@9WF(a*Mog@Zgb9esQS0It&%LUHuz+1xkyd7QuqVm53Kgos5ASNC!<s
z`2zOPIlK0E@VHdXJ~FKEoFo!4Z1rM{y3qXm&<*-#KF*eJ+(+5>wE8XD+ssj+IWKPP
z4v{&EhWGp7;tLEZf@x{hPf<~d@AEr{^ONIbOC|SHRE=(8WKHl*cP{2?3&uW7Kg=rE
zS0EV}xx8D}Po3JWxOsH6;>ltj0~oyWy%x+%t;`XU{xG(eFP}LH0Yz&&`N;hI5Y-$9
zmkP3zQ%Hp{A2Vc(tfnz}yl2ePo8Qg@2+e!Bf;Nywnf<C5|5JkZ$*@45^#^&$bboFd
z^oEF)rLlXWnr2giG9P?eL(BVlM;dRW&-RQOS$KNZ?7i>$BxmY9fVM8^fzN_z4G%52
zwUEaO9@De%s`@11iy+y?lkNJg=3}v^sv34$mOK&Dx{;w|Yb~#T#Aik(r6OfGKS$fE
z`7mbsD6ui^gHLGpNy%VdC1-FEN@t<IRoJs%7asqYiB8M9K(xU;;!0{Oc3Q|)-xmzm
z-~fY)!F>G+ZS8cpgV>3xWKOvJ1UWNCA#3%q50d=HmwK~|mxI;ydhH^{=aQwPaQB19
z<djX;Q0cs(TwdO<cf3cd@Xs|PMa-p#u?@$%wT*<szRX`u7TX>VWu4Eu+uI7=#zaz-
z5{<hBjZWA+*WhdKDLi}Nvk;G!)r=qfgLi&D#WYo{Ub=Xz?JJ)*@P5debQ)^4-J0fh
ziKEkB4JHjHk&Pw}dYqda(2qK=zGKK&Z6;rKfv%oP)@E9pDu>zvi|KunE8bp*En6b`
zv?z~$W9i(8-JLf9O#+&2t%11OYv--cF{<M;e(-w`KHiT5pCYgDJm@naJKOx2L4Ny3
zSA&<3DrLEH&;?uv%8I<z%Rk^WN=OMo4x}0JRn%1R+mT8FU*1AI0NH(l9RCX|@@t8;
zIp>X9SBeguKV(*lr}~AzjPcns91FCY7cs5N;V_@Jwy+O}ANjQ@bhF=`Mj~5|Jl!nO
zQqN4W4h7KI_ywiPDdQyI`%UId{3OWo`4q69h|Vm6r)?2mLZVUSAc1Ubu6ue{w3b3t
zwy66`8bi!0WFAOxb}cLdj&v7}(CATi71W}9he-RD|2VDSRidsEdT!bV%0Pf1a<db9
zz88%mhhEp+ARu-X5P8~_Q@aITmBt}V0STf|<QMco;x^7-5m<+`;OdI`gVv90IQKWL
ztxrWT4qW`>-<Yk@8{@P~?EJC+fTq6(*KAAwT$kGLqq$bK!CsYkGt?W?rWo8=u2});
zEsVaOdh4J->l(BvYrdP|`w5X1HtR3m!iF%*hltC_)+n5HwtF+ax&xZcE3vf!T}eY7
zt2I@BH*RGOhg&X&MKHtLJLO7CCnY7%WiHVvO*H-UfgdZQZ&M;3u3MtvslL_89*$M^
z<UV{%kDW1dbtr=Uyz7J(Y4dEpy7=i{l#JUm*d@p3WZ*vUq^`A`@K+zhdFaW^J6+NS
z!xa2O>16t^s2yXeVRQZ5fO@H<m5#_OW5;17mb$g0q~X4!1%lMH%Q1#*w@)3*pn>j8
z{`LLlU;a%3>pHagQ@NW69{CsH<etKUa|3w>LV+XzCG{XZ`DER(hmmZCkE^K(fa?(f
z+f?0yfRmps>sR*}?7iPjVFjy4rE1-<2;jB!E8r_yI9ZR090O>OcNs@Zd2DkHY82TA
zZ4;k5UaAgN-bh?kYTsqQXf^7YLxw?pKNT9Zl~QNv2++hFpM^E|h^mmHpS3?7n*Q+8
zNH8?u5RhDn(IK>*!m6O-Fpu=X`-Kt;IV!eT%qX|Fjj$0^N47BNQczaItl~yfHDkS&
zqPI4$o$Za$>PxamMMToy=xj&OAGW#}iProU&hf%5wsO2~PJDR}eP(!SC3<tFJCAH%
zNA(SPW$|p4MYB5@NG!TeN{C$_9r<ZN{L6g1L)w7octqqtJe6J!rR1)ig{<_Fnt>~4
zn?x=7lh@{I@G<5sDM=4;IzsC$K^Z8c<;8Pqt6@HV`2)w*{K}k^Ts9Ce+Is*(VHiEV
zh!1|Pm6lM-BBpwkir0e>D(71m!NPL$C--KApWV-@h)LD8Nwn#_RMJJYNn5z)``kXO
zoEWQ1V6dGvQO`sngzHkBWsRy!i9y{=jgtm?-zd=bP-wIz*Fe2$CQ6P{eWb6mRF(p-
zJ1;O;wHu>;iLl`iSRlZ|7Vz+MKnAIE+1@Za5%@URA{0d6kdoNq4As*5#(SpG@<Kyi
z+uQA&9G~SE9H5^uET_<g=9byB6b#nC=|(ZEOo`k9irg7--c7rq>;Bmcm%>1Wyx`#4
zh4OFIwH)$4!C%_#VT(-+W4o!osylvzceUudFw+ew9kK7g>kcE1sM?#^=I27XGS1or
z-quAB(Sujk-i7Xgv@voql+WB8_5q=fLWR+bTXYPU<xls$sW!`RgoIyv211H$p=tjp
zg+zbpN90=nQf!o7Z>X9<rA|yb?xihpHK<i5js6le_ymj6AtG%8f%Z@WBN+m%!r&`E
z$b@$5vqQmnJov<(Na-~u%-${^<rYuihG#1-vk{LJA_0qYd|ll?c{eWhnN#&3SSa5<
z_fp3GNSak5@8I7o7z=xUzPEEF)%H$?iUp0NVK#RYCn6k0P2T`(9ham*de4>$o<&%2
zT>jm%g(Awr>wJ3PLj(7{s6EMwrjuis4MB<nzqNs9xZh$p>~4l@KlzbFgb8_pY`5Ah
z-yg6u{Acl%`$A#I=}YxS+3svR&YB}W{xn<L&waNVX&*uS`0o0&Av926^E^g+;}Rr#
zk<gRa&<4`&7-oLKT}J(+Y&OTrP{SDZE^S*~Xu@z5#7w&hclxtQbhHr~lhr8fC})!z
za^-@e7$SjQM4*Q&1O(G1HQKMiO|`3W=(xBTjFa@yh-XhgJRUJg#(gOyjBbfw%&R;P
zYxPv8HIHK%D|^>qONU`aqf~gvN26+t)KUr`CDr(TrKu$<W4rb6j{r8u&%?QH&i?vr
z_$%S!I~DzTQxRUoJ`;TuK6Tn-2;ezWmV0kQv&ri;sEE?4a(Ak^%zqvKmX2VT?Zi2u
z`+mNG8&9q6Syxn-_ZwC{f!V97YK`BT@>?7SKi=leJ#jM&+1lONB;c}P#C3E(fxIaa
z>rd{$bj3!6K4tAPJ7D)e?eaIF64(qz3EWNZiB(cR)sE$_!k8=Cv@F@~9{gv95-Sa8
z6Peq!=}~T&n@-GeFOTy-Qjzhd>cKyLEnl1+RpZq<(BZdSrzxOmb|Ho*cKOo&7|kHV
z84bIwIQ;m;&Z=qKS}nO^W!+7qFa?rQGb_hzM4=OsmIF{&)D!1W5{r<BT%fWne|tGg
z=KMr*JX0dZ<QAibn>LO#A@h|l=~Zb3EJ8{(8%w$kk%umEGe_6khn?_MbB{QjppctK
z2{tE_@C?npKHTG4wsNNR$CtAB9d3wU3KM)PY~*Zi>>8Xt#wpZS#KKW<@-#sOw9FUo
z2{}3!?W)NXJdP|2NC_@Ci%B`n3tEaJSTrh|C(V#%G7S0yjx$rVMIDAFVq(-@*`xx)
z4+CKwvyP=%=ez_EeP;}_8ZN*>`W$oxNQ3?NqLC%$(rGV%rZ1IAi#un7p@-A*JF+(V
z^H>ILV=tG7gP69M>LC%V8e{@7KwK8x4_u#ejU%c|+8NNc{69#&_iaX!yyz!m88kn9
zN*q^fAj<t|W>BTj;a7PUxg1_qb^5&hYBwPF?cPe-M#s=Qvl%A{=_4YKBEhdPO%yD+
zh#q*t41gJRlfSoPN6Ku{uQfZ}9<U30>9I8!^kS)wXE6U*@+s#ydIzipY|_`;*Yo~`
zoSV4_``N~b4{Z3tUwk9`b$13-mkaGEG(Xt?K3xMn@LykW$J0%N0Nuu`!lb32#sy@E
zUrFBQTKa|Qr)sNAZNj-T%_e}4mWmpJ8S^v`IIm5>1hxPr6gB*wbB4TrcMfU_Y!x&&
zRPZ?sF#uE#M6f^W9gU&<5Ks#QG4EEo&B(PvJCnqt#kny32MO3nPPx)&)6AKjZ6}ha
z^IcTV-IbhYK%6%KPEH;OGL!+T9f+K-|MW88<98b_w$)Gi&n`bLY6`gp)!PZhcxf}M
zq)&z5A*^T<=_+6k`yTZkbweb#mnxym<|&03hv7^>nw}(9P0sBT4C}SIi5_TIA&+02
zMX3jW0t*1UK;8pLVtr-v+n#tE&w864Nkd64+@AStsBTZ;lsMzg5o|8-9bh`_ha*g)
zSGzWSp=}Rg(Wf!686|gnl(#II-Kw8UcB#*(hBf;<0qZ70<<ecPR~X11l6JWQj~(KO
zaa&d9XNmP{ijgoY?wFb&+kN~>$=3y^W!DFuIz4?t>3WhzL-Yf-4p1zSYblYHEsIcU
zfs`@p>S31O_>&p;VpmNJq`~0mf4ra@?85-S^`Ii?11R@U`c$s00qcTR+7E^Q6#Uy^
zQmOiisR~$;S*LGyjGzDrSvySrGR~CH(xj6Aq$dp(?`kv&sbT6X2c>!#iMnkJr|DPG
z(aL`Ws)^c*Do?vp4pAMe2dE-s8-xPgLg`O+pxUE^s7WviR2{wprG{p!1foJc2~^#q
zw53U~1y>EV6Mppo_TRVPV{V-PjtOlJs32_N@4)@bcJJaZTLc^MzieGO{<0ld@BPd6
zUk)}c|IZxMf&cl2I#&;fa$krQ2gUNtt#)#hCY3nktUKshU}t-#Axknt17#iQQnAnb
zkE0_LVGjUh!k!i5s)x6S_}hnz@h8M*Af47*p={-*eLtF+2l3@_q`~I-Lh{Anl;-Ab
z3f;?#3y}cT=H}%Aulw`G3#-ozs~_;G<*b0Ip1xUlQPGKsg$YaU4n0NiCUHa%Vl@uH
zja82!u}5;RN6=Dx#e;27-Nay7V8ESR7!q6870M;J%Oi77HG7CDEQYVV#WDBB;ZMAY
zWd`m?3$-ZyX*KA9EGWv)BLwX7!Z)9i&yqrh4JF1=O<Nc)(jt>^Sa2z>PARdg)eo`7
zptCHyl3Xg@G7WFuGGbd^&gVb4?B_};yVbX@x+xg@8Bd?0>+nrtpnowtBscwLQ@}=t
zdK2O4mgRKdBb4!^_Noyfx>`9y9j|3KbYkJEcvxHul0UO(yvsTXJ&$ZCWL8o+o*-XE
zR{0p;-^*Dm`lQWyl=&#=Ob}4WqC3HeE~-1OCe0PC^{o<xCa+Pq%s{bdA2%*trPA-p
z`y7FN7#0=yDT8;ZW;BW#mvPE=>^k4n46AswP|uir#l=mo*0`e1-e(GKM(2rs;N96_
zV3+tC8DPJY!3=;1Ab4bR83X<9MJpaO(I%OlnKOoMlqlH}7wy7gfuV3HuYC^N$wb=8
zCA}|MWJthyX_s`G5_ru=oT??_pf@FkiF4p0rstd^!?+$2;@jQdl-B++zko~yND!1%
z^^0F)qyr=kORaS$Vm=K>VE-NvR$7Zh$o!Ar*TI4Y(glztq^o2RjiUVnLdi>lh^l;9
zz(V#2RCpi~&Lpe3SzC|8+^^1sp{DXgbdqIBK87f;IXPau3rsF*2zi2O4n=)OT(47f
zk1?hP;X06LU|+BVkNTT@P`sb2yRo6la&;}JkI2t5PTm!ePmCk8Nu}yHTFHs&=xc-K
z=J3*FS<z|X=?OYCmO!_3hg4%0JJNy#(1SR$UD&-sw7QCRFNCVtyV}tj1iVbdy-AUo
z4!c!#$L387xrdJBTEF^y390o=(k2#hH|bF`;wE`ea#G9if!n<*o`U#_>Ul|Z(vOx@
zB;-*A#Xr9oXSMuWLs~^*P=5g_a*g@=`ah?DbXsFpKLt^n3=sUD9?FFI{v&xy>#oA;
zjtbxd8Gp{AobIfpl%Ipp_v>R<Gh4)Dk$G$jJL3^QG{~A$h)V$D4CF_rqF&{xdKA}{
z(vfw?9AY@Nm<%CG7$3IR<B$CorkLnOsUXp0$O#|OQB<Y95lzOaA$<REZ}&5AevpTI
z&)Arb>gIu)fa|3ql0_V$`vg*l%wzsmv)ksZToL)f-P%-u&-RTmiaH(F+)MU%mwN?*
zi`{DX1Kyu9*s27tE!J@i1J1oq^0b&{BVPnxF78JiV=vte6pU_djEAE5tx#hbetCEG
zGrxU!t3e%@|KKNEMjVVyEmF?Pj5KCyW|%vK#1>7Ip%t;zEq(9Mkfrx{!dN<Qs8Ul*
z;&gir)nI&Jn9P*V7A66+uCBr6QgT0!2WIugK3r-(=!vr2bX}|-?ut-Nrx|as)o9D!
zSebwp>b4|gk0Kw@8)fi&fu~@_V-Vr{r6jwEB*^#uQln7+Yc9h}<)+qqb98+X%rDxB
z{yZ6>#*&V^Ya|PV_c4CraS*WX+bq0J4N#?@embk0G`NbiiNajf>#h^OYo0_H205Qt
z0OO8lcd4^y`uo*5Ke+cuVII#6&sA1@y<B4^?`k8C#A0Lh>+nTO0U&4lqn~yV{yIiD
z!vS{#ExbpPy)Gnaf#6Boyp2V<%OoR@zYg;BvH5#DXfEjgV(lG(D{Z<)(H+~i?PTI)
zVq;=WY}@9<oY=N)+nLzT#L2|Y+4Fwi_n-5hI;ZO1x_5Wg?)CIqPiyyHz0hbRpB4Ux
zL1NUR;!4sG{hi?ym@Nk;*4MAlu^t~D3WO@x*RA%xdOn;!aC9ngA~4pFiY#Uj?>7b-
z$NTthT0L}Vhr7M<c^8xV5YmucXc1`JnUL#;VQD`(GID*!Wd?q?Lu|d<rN6;NxCx<j
zm^Xy_t(UM&_rroF{!6vW5Lq?W^3$)!4ptxHc?nOvAur<b$!qw-Ef!2smJN(F&r1We
zP9t4uYbpJ}M{v$2h720G6c`s4+^S)5-!`c`g2lSgBbF}5aCJABw5+n<<5v;<4yuRh
z6R+=_)nP8xVL&E%io+|ju9;aD4HxoWkfTRkN$Yd~qxc$?MOx8tqFGf_A7bfKB|g(M
z&DnFCGBPC4B=fHKapiMLRCVl+jqDUgBcI+&4V1$1n?7I^6&++QB;+qf08LEo{ihOz
zski|39o&$L-HguH`<s|H-(9SXZK$G0SW%Fo<y`a&#YlJe=w0@l?Kb6s+c0Grv@3Aa
zUix2#+)mO9ci`}g9aVUcLeF^RcpOrtEIHB7T7P8`7_8evt8V}}{X_WV>0VDr4FH{4
zZ8|*tY4uO;9Z+rVfki`mO=J1whL>L&$i$F?j^dS57g(<ry%4RAKPm=x46iJM7;2M}
zswy%V2S#hvki;V(e8iN4-Oa=4Rm$68iD5;>Rw+Cw9V;N<ZT*jvV&-MuL-3N84fP}K
z<one?8tOyVAxv>cDj!<)Wz76>m?zNKFuk!-Pe9q4bK!D9fV(}7`rJXZl>LwFPT&^&
zwbT|w(RCFT_f!bhA}Tjkd9xJ1NK(X&geZ03_~84rZ>>WXRWiscSAj2qEs_25tObv(
zCA6{SFE;z26ozWqD+^}GQ!;3X!z#lsbqIAQmzAT8SMOV>b<SXBC)oXf94A*_A9Zex
zN4cq;q$s0#Bi6&93^sO0QvGA5@42ORpvufoFGeRDOHG)m?c8txDPg5}_Un+cL=h_|
zt!TAkLv!gNr{61iw9K#^!L@vFdRHl*-aZt92tcpi%bM5#uD3(St|x)j3$Bw(H{LW}
zWo)|ZpC>M@S>p32F?dF$&EL4mx!T;KyPz9mu~9=kul<l~e@*9w@wbcQraiwrW~dm&
zbksf@Uv?+OOU$W8<4BNnroV<~m|{3ut8_3OtpjKg>3&t|Ak3s=vj#8<yx~*<My(M+
za7kPsY9#u3s8qeQzy-N`;MyU5A<~(GmHLAU;b`_4kXW=1fmP}az_i>_YB*p-OiQMi
zyz8k~Y9x=(`f5En)=M%>_DO2#3|iB#UA~E%^<gHaa@MKf4xPkM%{5YS@6TlR<Vxpu
zGgImPJA`@<=D=Tjdh|l=5BlgIrEIK>gzR3#mBXk_Te)umS$Zo!t5^=`kOBy2Ct|jI
zqrVBQPDxf86@4BCgD?Y<OKjGw5XJ;5Qr1H9^*9XX0NPu<7i!Sxri==d*Q6?6kfA~Q
zCZR8}LeP#b0x`sRC8C+_lj7+Xpuu=?R)wOO0S}14e#L&GO<)41>3GbB;2Hob^|#-d
zdH!~^y)<Y7H-)rgFiCV2I7LwQ+WCnpvUbd2uXmo*#+Tx6J=B)puQ&?Ih3c?GF?1Km
zeB??a_943HmF%f%0->7Z%Im)g*279>o1;+mszW@nCK}V?M?Jy{0-_kOJ+L%Dza-UC
zmQ7G#Obv7}@BxrQqekn7V*RCE6uI*gQJBjZDJ9S;K`9r94n|o;1dG~BjcIYg^~>QV
z<0<_Ap{mYBk}MNGAlJyI8p5>@RMfeh+uuc7ujn|+xt0->VP&uk@~S^d@J~@mNQRF>
z2fubdQ{H6VwO6Tq6oYQE#X-CCz4y$K5VyEmhxSmdee~$PExQD|OI}c4^n^g&@o&p)
z8Tt48gBk;?UNQrq*l^hYOB=)${0WbMWTm7Y^q`W(=Mkf*hmCka83*Q=Ft%GaQ_^lU
zP2`U_8;%%P!FMT@r3ov!xx=K<C4i^l>RE3}q54jDtq8IHXFpZ?23x+oi<MggE4!=3
z4b>HVHk0v7$Nh!}ikl#qc|z`L)#xRb0Tf3sB}~R<#A${haAo)owEe?bUSxe(CGyMh
zybFUIguK7fh0A1L`_X*E+6GB*$etyd)zk>JX9PFDeKd$Nt854Rgb2hq2-1#JpjivJ
zM?rN{>BUL^10V<esn$85w`Anc95pZhv^?0kf0n50Q@(`&@``(2s4w`}UMh&M0bKRa
z<dNLiuaPKUOn5EKqn*Du0t#V+8lxG2T1Y2|)IRaz$62v}KNUh0qJCpEM}=>z#~9{a
zT@ROdGQT?B4m3W&jyayMn%J9%QA~Pf0S7dw&=|N)(7Yux&(ueMBO!PZu{mC@npm)w
zb!PvEA$asFr=CPEG&+THW3+4q>6cA?XqQ3oo@x+cKqCAs2T#?@t_5Tqv^?i&B{?Sm
zRFIEkM!ovAGEC!(nHYG1YLx@*&`O>BGq@yoND~h6t}#^Bj&2Ew0_cTFm5sL5)LFdf
zMgzCLcF_1jfIAt)OP0UHk1A6knP-c-|E3T+qq`sIMeQCRGnfb|Vsy$-)gTzcjX>~<
zWC0jyOpADVA)sFP43&2{E|%1It7kN^WVQ~G6**9nmNJ7mj0s~;2G2_o(uy`{9;epv
z&mi6txFi@`M5{mMukmkb<N=EzpdIvB>#EogTVnY)rw>9hKM*oN>h6$@BF%PaQL0za
zaj7xVj(a(N_0pR~hD4VLDSZ5Z{jLgDy|>eUv;?R?uB|2D_{*rQa}io$gW$f~g4G40
z3{Y<RS1`BMR&lO)vVUKJ#>d7m%dd6edZPWJ9WgcRbQ)Nnn;3{={S8q!W)}i4Jt)P<
zy$|zv=MY=B)`BOMI~m8iA?8`tc1mUk9<KqcfwLF>ds|Rbiy=Zr4_Gs2kc}UbhalwW
zO^EEi(6t%(7a?^N=Rfw(Z2dX870~VO_U|}-O0%lqI)(;3|F&pcfBb*4XpsG%EgBS+
z*j#jhY=n1iMCUm!e_r^p!HhI<3b_7I`$6XlI%yg;HA{ceh!`P^u$5qjTUk?_Lsddl
zigk-_9`i-Bxyizd`O~?zx?c{nY$((4z;#N8V9l()5i8HQ$_^M#0OXJ|0~$KH_hIlt
z-L0T-aCee;b@@aFnhPO-hA6vMchlk2zAmWmX{+t+7B_bj-v%_7)MMK2<PEO0tceu~
zd0l*AmtaK4EV6)FN)}K{Y5vxatWNnDei*{7!D*T0W7>g{9b5%!YSJH76!M?h-3tK}
zXlOb!zJ_4a(I0m01vdvU*v)DvDf%y==Z5?E2jD79M5V7Zi~gP5>$=>S64$1V_mAh<
zP2kKys`#p~Wtd>M)-qd8G;J=C`yB<F!kdoBsQ8C6Ib~t}^|qesDhqmR$RHJ)_s2HJ
zO?TCIAU7O`qfJO4Y;|J3|78n4XZ_tEt-}6BX;L8Cv0x>NYo3|SkU&Ed_xv!Y>>rV~
zXsoSc2_O@K1!0WiMIHlV(KGR5IVv_g4asMp{yR_Mb=iCs!hi`u@Whp<dC=p02RT9p
zFwwdHog_yc#s8N{(vUu3eIw|UiYU1<2JJPH%wpB4352rG_;T1jBT7Mx!_WvmBeRnu
z70k(ltwm0?v`-u26>b-uSsmJG-8qEz24lP1n~3E|G{SJB=a~o8ZHLkBve^Sxm7EqA
zzC8xIa)F%~>flD)Hlsq}-55edzK|Bzg<uq<tuubzlujkSdS^7;JEWY%A09{)b7i|<
z1B;0R=ZBgXUI+m?Dp*5(*W3ojkk~ORdYS+5i7W`|F6NKrj@IvoAsigeF5ncv%rZ*%
zqthl(7!4*OdJOo5`0uL9nddD8Q!1gQjs~`Eq^%cIeu+=N9vDoQlp~gfzl7bj&B&e=
z^$dhi_!6HXn2J4$;lIJbmS@9;wnoVoFo!XE81wJv^if%Jg5dA~?aR@>SRm1JS~zAH
zm;*Q3$m8ChGs$37w023^h3@zF^(kl}p12fVkM+)A#v9bhr4v$`0CM`UiJVp&VPPy7
z7y<44$YAh(ik)iYOVFltQ69sAEJ+D}UVC+DxAnweMkQ3dOHpaijeri}*7ahFI|VIa
z2`$hKjL}UFhP=qeK~qLLR<y7j|GzCdqHvB+Sy|zUPxr3^a|^L=k&{0r<*wV6KcMO*
zXr&A8u=yd3g%dG_O_zY))HD6t`jzbB=9d>SoFgY&=17ARcwB&?DSFLdGJ9Zq@u<?%
zQZ-FIdjDFNfc}b$sC=({-WKx;%^cSWKi6RDVhWmNW_t<qT|hPJ+z@`{%e&_@D`Fti
zbrPB0frSY#;W4OA&0Phx;$K~<HhlN`=HY=CVzR62-5mMs;UT>>HoL36i$4T`TD1$N
z_W+E3D6>ZHwRurKTK>r~2GQr-wFD;=k?b&f6b6P%Z_XEAG#-s+R=gqa@wE&Bw>uMV
zK>!`uT+?JxwMmV)YZF#)+T@T3)Efp=RYI#_<id?Ipt{?026L5fhI8I2Y&fHVJr`}=
zR~vhJUwtMw2IV4Xm54nn($|8%^vd~hD*Mz^#>!1)@yh?2EC%x8$bHn0efZ>M=u>ZH
z6LuraN?z|GHu&g?pKn!q`6^I!xBmNl+Q6oF_8nVQ{MNa%-ZPt<RF6IX!uR*(kN!dX
zJzt}5rynVZTes)h23@sY+hn#Vn_XU;>GoAb!r#bAX2n?CG?9%wCTo-|>T^^-%2x!!
zjBt+V|A4%Jsb0klf@x1#&+zY+px^2`jQZ8cg-*J5H_tVqv^aKWKevxghgJ&pEGpeg
z%HV!=>QDY6A%H|WPD$Fdj9dC0iG?`-8iz4ov8BptB=UHBeyTxk>=&6l4ft{)w2mEr
zpZ`w38SNydsI!9Tnuc8zt|HL0xJznzW<%Sq6m|?<<5t+baO9F0)^8RZf*MLp8b+-s
z8hRvOc986HIk;?aK9%)FZ+ch4duW4Z)+&uk+70bBoO`(AF~(<fKGU*ulJ@OLLBP*Y
zmgsj(SMfO;)CX5h=03D4yz>l-k#5BzZbb*nms4bCFRjFl{uHOS=TKTo&eQ0McZ0u0
zwuZflTruS<Nb+s>cH<*WT8}Y}x6Ccu>52THsE=KAFq_4>Iu`dH31#|Ca77ex^5iyg
zUx;?_Z}a26eHeQBpj;%;pPxTGVZM75QFdTaTuKj-G1`pY1;`SeOrc)|Bvv^gOc+j;
z+hXBOdv_5}923_K5_siRuHHxFh1Jm9{c>r|nU8XyAe^p$^~^eZaA9Mm$NQ>`f|Hd{
zf52`kQ5t+;lS%!Ljgg|+dE>m4A7vnMz~#4{ZerP0^@@y4(P=(9T%P_nI+Jg&?ZvKN
zR+Mor%B+3dRN49?-A%6Y(2u@sg@kf5<Z;_rWvRg(sYpA~e&5#dIZY`KVodz@zRzH}
z2>N*#_J=4Y<OOJwU)DU}tZcgi8}{nbg=B8l7e&(8V_4R5rx0hD;tlusq0u+;dh1}q
z<rH#}m_DaFcsonfoN;l@I$%t=RTTsOc5gH=77s7Eb&c1Zas;#*ut!n^^9T}17a?|D
zuCS?qlq-0s>L(ppDG!i2m4Yqg-JJHq?!dW}eq$JLIcsJkLsS7PSM*raXMgcotR2my
zH*<O5_)pq8w6!>b+HY`akfPE~rCJ|EOSIQAUpp-|6kL?A$^_BW6b#mGpz;%c7sUav
zb`rwi)>ZuX`@aFbNeN+*<IP|@of=083p%&on88d32Q+`2`Q*)O{x`$F#ooym@%dj5
z#n_dCzd+6gZsnq|{+b?lAUer!`+u@VVX4KVv;tC$wcURO&SVkFKE`w1ORTx{LavY-
z4S>A}Su@{5a?-T!&RwQmU|^j=GMVg!Km~F$i7~K&{DngU_1tMZ2_RQ!(0KV6J55S<
zLSXevM9<Cn#3xt&Olp!u1_j8Z*`r*kJhd)R9o2Vd`lpNBB*Vv|<6rME?(b|Dk33ia
zK$ArLN33u_CF#}B0E7W10OCGnMNoI<BoyXdHdT$l2*+zw{CjS}sK$Ka>(bvqjfa`0
zSLFsk$G7(kuhu(mpynR@v)+)g7gj^cSnM~7+;~KEC7uR%FD#PT*JYqXEKHEnoqII0
zG*5|hb*nKKa-p!`hlKS1uKdJS|H>QFT^#%Q8~Z+FcUBja3qXZ~PnhUa{mpu&0ZSk1
z9KR0{>7YG(chRi41C(ipA`Of1%4ZkC?h=rWK6~BE6bkt)eC?QuMfkRfohGX~)h81I
zJa7(hjkM=}v>=E;xZ?=GA_RLl=7SkPahK|Yiv@Agh3|%7^#@@N@4ddRG4lj^X7CL|
zBz(B(T;Qd0O+#^Ep<^$9*h4o)G)Y-i+!2MZ88b_4k~-MX{`3|3nKw4}kOQ^zJ0@J3
zL+}tl%lB~k*Y9+MsDaqqZ*-~283InLDjX8J9*rYT9G4&~9SODOhdW*0?KpXuam5M8
zA@7e60fLnK{iv{;wKMd3<4f06@7KjK?L?!T(_2ms6bd1sblskOvr_ijNqW2`D4zCZ
z#n*k``(aQE9v-c(imqj6rS*}im^G{M`N;~Yi0s79d0kG-KG8l=DU<w_c8FB!HXZ~@
zSjuu?%807OfbfKhd0n&oADh#)MFW3{4D(k&8U9xy29)5xMEd#q`LxFQE36X#4OtWN
z7g-BnhzzGKdZ~w)k?L(}60cbOsp&&;MRH8*ytu5|yw|xQQgy%&t`kyii`iZ$_Nkyo
zO_!(&YQpeYuQJvFJq~NKNp1CvWiGuX21K2SYfkj2DBdsz)(dNLj%$60c_p}^Bw~!W
zD=ytI>HX}C%&k;dimp%gBS|^X0XGS$w@qw%lSwDPA_ZpaQcxL+$t<Lq2c$s*IwbY~
z=5nZ@hCr7Jd`(R%5##?curyEF$Etk=r#T%aVww}pjl(tgS6Ndvs<a8j6^W+6ajI*E
z0aiXqRP%oaL^NH1M8J|MfK~W^1^iRrzi<6tfJ22PO0|F5mq1PT|1aQ3Z3XwA&S^TJ
zOZ`>Lz~h|Ty!gS4wX-NM!a#7{?R@~U`vywxAFMI|`(KjNPcX>Zzhkglbi?n(*9mPo
zm(@idk_qxfhu~dTvst}IkEzVl@Abw9TuV>}l;Iz7N&|#0uP*UMQ4ePe45EOev$N}6
z-oe4nP5^hXy{n7zvcRmy*3L>||Av-Tr3;1?=nac^%7FqW*2}5kgwlSo$K(bDhi3fN
zcM&BJ3R!fWMk{lBa548s&S>cbOU#ar(BZ`C8*t5gys!En!Z21B*_lk-WW#?JL%0$&
zn=bZ2zo-}P$m=fF%MokuE$RrHnw_@YbQpfqR()Q)@%@xb!^u|caf^cBel=x;z9%m+
zoBrjBX=b_qChYNC`ldQ}yiG*l)nzE;5n4H2Y}ORp1XuLXg3B)wMRO~1Qgn_b42u^>
z^V9shRbx882YEY&uF^T}ENdHx^geHYf{FpE!4t>cUnkkX-KqgIvRmmnGFf06JLH77
zFK1njMp#2KrAAW8E`aE2W(4DYy^ZWov5sDX3;wS=$H|1~LJT|;RyXz|?g~4bZNRgX
zV75v{kxHBZ*tIlhw#v4sQb{2PJ0I(}cz8N*M>wK;l{SG9=c0wB%ZTi}<SABV2(}0g
z@FHywT||-I5*GZQcumt^8)mOUgg^DGi6V1u1y-0f*)*Q;z62B>JO}1V!Tl<9NVlZ_
zqqOhGn*HPw&^Y<beGRJn@iau>B$Pl%E`>!Neur-)^_#v9e)A~qsgT7;Y^FZl9Hp?2
z7t)LmDs}GyJx*a2gyG3GqE--SuO8e>*foKgRnpYb)pdndL^Y?kvb0+;KV=9PUvT(5
zI-{-@GdcZg&e`~#IA{Hg#v>sh$%IO9_|NU+@*Ve?;aCTslBiA9yro51l{wJkzzOs?
zv<E)YxHg_*6)oF}b}6vb<)8<-o=-j`mf5VzKZQh0^%UB8G?L}NeK}d#Pirt+y6&Qc
zVlP`<TwAsodle9e?ZJ&)yc#d*-ir%)4<cb{f?uYf(JfzPVWw!Nurm$z`BhHQs?5p^
zRdxJP!77zZR`yzC;gosEb^SvjL0eoz=S)V$|D5)zU0ICxt+b`{twLjMtfzdY8n?EB
zuTSd5txvk_E6o~1tf$HgL2DM}xaqp#6ME3PUaI0}iD#SH-N6a{z21kyeHoFZDB*_|
z1;HlZs=hnG`K=-{Bq$*D?lma>u_DrAI_@9wXs{xZ+m&C}lN*9gp<f+~(~d^^Y~AD=
ziwOTjarU9tNkW$%%wn*m^!&;HK_MQ}BM;dVb}+$LUbP}FQkmiT=D2JVx<*?{tD@?w
zYLld{x|%>vO_QkNOJS435^U_Hpjc|IsnV#dhA~u$POF88agnSk#=;iTo-Cu|Rivi=
zA3tlto+@yoM_mLpuE*FPNo7I^a+QE!q7ig-dR&-=W1dYR$Cg}TJ@2gYdc4vyM)r;n
zFtI8@eK2d(y45OFY+Cry)ChX`R0uj`a&94YZv=n;qA`Qr95B20y1;lj`KhDHZ7%i<
zHm2H;Ep+%!WNJpQ=d@k~Qbr>SdAi`;BRwTaw(<Zl;CZ>jux<U;A}4!O9A-IIhIASm
zO%xFw(*(lFlff+<O*&Ld=@^I)GtBQ64(tUb99WT7tZA;f>Lhm>ds?~6n1n$Sdpe1w
z9HBE*bvnzR`jR;F5)RWPQzUHD=|-R0pd7uf0FLB944|>g`nw%E(3olLs$t7K8e^Bn
z!!)0cg_KBmJ2K|q1KSF~%DJyCOxh{MIw+;`t~&GcE*L(3HxP>ZMAsK;ha!hY4Gg*B
z%>`Alr*%DB4}!DrgO=0;;C1?AO;1crLrq3x0??^VTtF4@1>mGEZ~XzJFPScN2Rei_
z`SzQeLG7rD#heOcT3>a9s_qD?gsbktv`#u05NK!1FuEp}E<QQ<W4V%!71eIHm_9!I
zK0hw(KA%1qRDC81L^KUEQhk{JxEMGc&+4Ogzxl0pd>jqNlfSB>PY(a!ZHAJ0Yvf;!
z0pPt-#$=rxKfE<{1NZGku|tL@Mc8&QU3TcHO6$YGk%MDPd!n(hX`tq_o^@4Lj#ttv
zX$Q*<xpXd=*OH4qq+f|e6j^KU=wu;IDNS`r>a>;oy0GH|bwQ3R)Aw*F38u)oN;I}1
zmPkA60t2H94}}`8p3uH>R04S86B6Miwrv?K8+AOeXn-WPsTGxMt+EDXO9*{SY=dN%
zf>yX)DGfJV%B4E?XuK0;u~NWLQ4K7oHbqDXER>K}LT`W+6Phb1rjF1I%FjJkR7M}F
z#JJ#|ZHjVcuwXe0OH>lyFITk7q$4-_!JjCjxxP+r9#kRGrz*lkt(in1#ni^{O`{cS
zeGW-n#=5#ijTlUCqVFHfO1%3@srvbd0E3Rs3o3cuc`Bfwth->Xcu!PI!({!SaZhxn
z4RNclaPOD0w$XknFzM<s^$<|;0p@uavelvf4z)54*K7lP<~<ym-U7zjMXC4J+V=va
z4T-Jb4u8HJlWF+gQJ1v3x1Lwl1UkhJL+#zv=bk#R$d!Du>t1O-`|?EKs)i*If*~Wb
zL!m;$WDsV9sw9G7WDsS8YA1qVb@1{*xYK~?a`5q)z@V3gl)HY$%H{YOOviaxKds`|
zRok;?A4jGYlQt!je8>rlyva+=#)(OQ$eu1on|7oXauwKWKkiA)1luVnJMi)eQ>di(
z=SzKk&u5R;iqb+6ng<j#pVOV5nDNL{a#@O9{tP3W`y(N7yE6FB*dyUZtNFH(ro+tI
zmq;@;>c(bLq_Q#hy-Ee_{y`;n0gWmDxA^&)Y{<$O_-*pL9LIs)aEQU%x0O-)(EDRE
z&zAM6sQst?-a1?^c|vPQ64@n$-mHva7j=Of5U!+Q*0?5VX)!ULyd=ixy{|Z?RCvES
zowj(c7?U{!ox}wHI1wZ*3kl+&EE9^T+KQ;9e(m;lvP|8(IVNK7TBG;vF~JU9nzQ$1
z;XxcC9N{X2M%EP$l7$f!3e-NVe!Rc@-s9GhNRSeiFgSL9Y7a-bjPhO@lz=Z0dqIF5
zHb6j#-SO4ax?mLk+TreLW#_m~)XQL_vEyRd`q1WDVKnniL128xKyEkcTBQbG4OD>!
zMaBGV7)~EFwK61p0|PVLHIayR%Jf^wSZKRK9`&8JALqrbT2T5F(Q2^Hm{$&Mu?b6P
z%aKm4ggL|ueenWa-czLbA}8wf0|+z7<N&BLW&v~nP89|`Fvn1cVtQ2+UPG7<!nZE=
zxI9O+q<~j#pV|?lTnKZ`NK8%rz}cGt!W)TR2zyO9EKP~P9pM%LR<8+zb}2E~u8B2o
z<>|UW0FFK-1Y09GBg=m4Mt>Z5XW>rBQXhrx<e$4aa9#hDd-psDhwu~_Hj_DpzZrpH
zY1J3w{3^H*V5xZ_Z?`3?wQ#|0a)mJ{fENbGHc#V%{h_~>i3O)YzJ<yg{<jThnBfRo
z6u$iXUPuO!(VWYGcnWc2cwdyNLXaBch*P{qtxwMz&b{cXbvEM5jJ#k{&@m_-uq?sf
z0drCyVFylGEEMZFYYXPr^6#;M2jI%RVmBg~YkhnGNW5DB#t}%9{<g&E`f0Cg`80q`
zE4^FU11rMIc#7e&Hwokfz51u*=fshBg%pG!VjL08D9tdnP<KrtmLNW|0R12hBP~gH
zOCy{<KDGe;FwF)X%hy~81q~<Su3Sky@HmmeE1MdWT*Oi$%u{qnpXwMn`r5*BG!igX
z{~W)Qc+hU99k3_BSIMsmVVc){+5Sj7B9vdX-EgPyiV(qTlyfz5HwX@=b#=r4oF(ge
zMH7PU*FH<w-kzcR`jk1Yd#)+(U}OTQ7WPCH^lB}6Gd=VnQ$b88(}d=9#^9~^@aj*$
z9kmfdhJhF=$ZE7IUN(+rx0$I?6;s7QJNn}jLqYu1D+fz)as&1CC)$ySB0>**Xd9~#
zS<eYX+X3eisyl_CdQhu_90af|kU~OKg=4>tkAYG3!ZdLS1)M$W$Dbv#*n?(Lqf_Cc
zAAL;o*?vFLEzb)=%?;K{%+=sDo3YR8g|R2y0Uv4)N85uiiUTbK&;Ut~IG~7np|10`
zbB{Ux<iS3|Oj47ufx}QoWC<y_7_%Pk^`<5*RQCDz`5Ty*Yy`!dT^;H}&4M^9L?R}-
z?y!8ljk5&^`y(SrbEd}Vh^h)f&E^TEK^`G*p{~S5BiIydwRfvQJ*B);njpS0^Ox}4
zTZisP)hOtc7IrU#?%pIuQidFlJmm7Q-{vqbvulZ)DVF)s9pKn(Qu~L?EyUs~RW^1I
zIl}k(2l5nL6t$97_N_cTk9D{@3vHCQ(ZU}3Py8{+>cAlUm5(?Ivvt|a`ISDh$X&gw
z!3EL+DK@z7<d$g?hcy;KAcJh_KJ{5tPg~3^%}0%oNB!~~row-g*jY%zAdN_!Z%vCT
z>HAEUT=cjE!;?ydHz4x&?2w~;c8rD%?6_w_nJgtf;D`UNfdTdWGeD&Gx|cceS^e2S
zqsR8kx?Gb%>o<*%7e$eX#gFDmX;4W(LKAU(8V^gbD0sB&_(Jmg&2L(i$c2<5SDAY1
zhp@MBx%7_oPSzT8qZ`7iI175!O1B5W5qH197$_0T?b)Vpbi3(G4HV+ckckC^3h+2I
zBbf$gz5<M)tRg*T@MW~5MOq&pALq<ZrL;tB!_um|`?=@dgEy;6DOeN@3d0o3D5`dx
zp1>V6j!H@*kP`(lf>ane6ItpQBVw)?^oj-y@$hn35v>kXU*VcjOt#^K#&JzJo8cO;
z3<$8e`1lZ)aqhvG*oc1SzYM#=AYpR|E^or)v6g^xfg>VCi=+|8L4~4>Ht|Gjbp$Jm
zvg39tm*ip3@1$3XBnY_=9Zm{#4&p>>j=gD}a-q`7JN$9AV1=!Y_qEa4^baiC`Nd=5
zy8y4_*C49vihWM<qDx;xyXZ)Zxiw_~qdLJ<T6FZf@uj+>omQ!O`jJLDwhy`w8Uy~<
zyOw1FS`>0r96E?Nw79td>VuU9j5f0{voJX*FnQLO{Cru-sYO3H!XQ{nJzBRz*PAR@
z@;fagI##FN&s7IJJch;Ko=W&(w&X+b5mHyj5Ax)8hs}z<>9XCRF0*tY0Q?Zd)v|H*
z?(pz8_FX1D>9w9(uUapw9Yw>A6fGurSEPSJjlL^W)q+;*%y*mJpooVd&Z{LUxuf<f
zITt$dP}}S&^|b(aBem-sM}%(>+*%Yi%R#>Lw!zn`=Hv+!yk9Mo>Udf2iq;75SaD0p
zgXSz*crGUUZ7Wv0xs{{ms^}5J_k3Eiu0UiUWQvlDq}OF~i;FCKmHL1IV_d1ohKm}q
zqJ{n$pcS-aImgokS&<M)^dccw)6KubpQny-C74Q=TEUuD(P-=kg#LyKG8T0Ui;t-c
z@u}l-)9R0>hV9Q%)yV+5!r;X~V4zK0l%ALwQ9?4WvaCr-U9vd&Z8@B-q=d2?(5phT
zc0$3SYJsyqK9vE#LJbV1psS^j5k)~uxh}irS;@TKRh-q+>!+h>-Ar=)X3|V5PzE0n
zLf}D?d64mqjkfTlp+usV*ka5P`sjVsj*SJ9TH{@g94(HIuR1-C7r=R%QASmAh!^<?
z_NZL;h}MQKApOswFwV~X*BVCx@EqW&>yYun2&5Kxe8{R`RjL>FsPjBFWZ>oqS`}+&
z6vgbWu>Sw5O0zPc5%S*z3>AcMjZi6)wpENOSO(#k1JFO>NfF4V8l#bZFJCOVD3jp1
z%%AvO*i7!$DrZb>(fJe%+Tz_W-8aDwHgD9?dBuHkp8smcE-C#nN_j55^{T?c5Xs%i
zu70y*8H*h#>Vv6#nx*wKkWLyKw++K6?q`Zlfr)c9@EB`9J}{=}W?TH=l-C2zQX=+3
zVu>5FH~@KNWA59W(NXOwyi(>*-2$60D&uEgzf(l#^Fc-n8G-gYpWmDgOrivLz22B1
zwE`HSvHA<uODO%&TT8WdHnu-$=sW%V{H*H#yuSl^Wh-x2PYr;n7j8F%|C>&1sG9Nb
z+%TePnOyViMXluuA%g`AZoq`o)xsyvbo2Mku0OmBT_~XB&Nb&#Zl{~6<Bl|E%^3?F
z3W{Gb7_#OGG-9spPr<t=G-5`dYfm*}R^I}da5+uaqgX~<^B<4$r6O-5m%doynsuAE
z-j7yN{V|CaV71x#YE^u^k~hJw#fK(f4j1}keE`d2Zu!GjzPQK2&-qKN+{V?7*V4+J
z&r!*GeL?{<+D>k(jz`eQ@PdlcOO(3L@DHkCs0+?1f~Ox3k1U@T>Ne(%4KF)sXs(WC
z-%_u!s?cT}y2iru{GLh~zFL0HTJBx_WzmPAjU&ZWOHzkwgvGBp7!ILn$(b3mEypyP
z=Oo_cWI3nfCGOXFdo-(vHgohiJ2@@qcJ(Bmtc>vce*NMfa{(ZL7z$8G5>o1+kR^#P
zu*#;_t{`!&fCDphVrPF@v=smcG#-<pb#RHxXg=KbOV>KRSsxgMiC{QR{(-@f_s@n@
zcS;jied|W|WKj6ZQbK%eL(=gcX0uT~jyLNVjb)=FG#>}CWA?CJ74j(Y0z1PSIx^lQ
zdd+wldr-#(@fc>BtuRV8ZU(7Tk~TO#baELE#lMoK7-U(dn)-qF#;6emAPH4Sth!1?
z7l77l%phxA)EQyo$gu`WOUKr>7y!%KhLkh5nc^}F%B)6HuI$3UD+`tiI0^`$v7!s<
z#czmD-BTrS>IcXIZhLzZ7Xywkl|Oji_yw(7kR$s}s?;KOAto&O{v%@AL*s*nava6c
z`-tGb7$7~9NrIyWh~fdS-AN}wonwfDCI=1Ypo}Luvqp{_ZXPJ_en*o9{n=ZHFPraQ
zV4Se;2q#aV0gid7k1}LP5&4g>8QDu(Ywm=pSgKWT-Y*LU%s?=IN=lT(RjbIW@tx{O
zH3vLbh%(Ni$5`nHXCNv1d8`t-#_aS2a_~!EDFJi2okFGG@l6%kyC6Ao8=9!nN+xnR
zxDxRVHnNlu@Fcl-F<56pG#;Av*NExO!URLeW<nc?)oBqHxo`qaIGADh;q%_%O3ZUH
zco0p375PP+-l!rAZ5FSjK4wLGnY|sRjN7U=dO<-&Xa!?&VL%zjAviz`gB*ao!$<>a
zKBrBpFW6wC5ty*2Unfbhr@!802@p~wo?W|ka<R5$p;<B5Z4|DyDOKaL!B8__e1OmZ
zQTtUVY*h+M_W%xrK`XT%&Kmh>^><_T!f{bn3ym4N7SEc0?&fl__d?bhAn?7oXx08B
zHuftqkjj_2!_bbZ+Jr`W;43sZEl`6Y2S8};CTY|TH;`0(7-whjlAsDIU_ckiXVuJ;
zwqOfvS|ss1%~Z<Ee(Ob~M3JS&K{K!WEaI`}cB3K&8&xo6RTv!jO<99z;|B}s51=3e
zh<Rdy_g-q(=0?%yVq?|;%8-Z!VPFIU+Mb!?&d|w_0pJQL00dYeUo*lM1~jqYfY18f
zoW00Q{sAyh5NRm@Y$__qAl$G%09%0rQXw^U9e)SvNe0R{0txM@;yO?;D1Tm%ID4;m
zLaB_*pQhy0NDcA~oEMx|rG6KvuaV7&yD01TSg(}Cb&fnuick|CGvU3^pbHRmKnsJu
znhbmu+nplG%w`qNiJ=9$?UCl4%7JTYxK5KpVMX$KyD(wN&lREFt&z%=+5Cpys^?gF
z)Om%n=a!;Dv2#V=OG$>t^G-!#KwuRV%;{0xg4waxzY2IP#|&rR6dZc%#UQaTmj)jj
zr$O%k1;nlOT#(-*kxpqaDx#VHNP&P8!57kto?OWmdD@n2-VEHNSgq?awdXjfr&Kv3
zj;;=|$!=Lz+<%LIY<~Y1|F{XC4Z`|E&kXK;Ly%(O)FwX8^BU{bM|6tA{CR+SMgxnB
zvG3T760d;fLdIoN=Xf!(L^I{UE4K|VN(mF2jK61Yw{|I~^97m$(n1RSJ*z_GYTr_=
zx)*LO`+4ld_6)!ar)A3S0>rCVS88Q~*p|>L)+DDWo9RjTZRPG6>j*w=<VVgF%9-b+
zcRo|H*K!C+LH$GmEu%`@)xJ~P&%(V{$(I11mF0SS#Y%tp2s$eWMn79wJMdcrGFR+L
z&gRG?u+ZLJTl-cWb9|s+=N2S_V+M&BFd(w8AnuC-zk*dA;mF)k$s0#RQ6*onM439E
zC00|^Po<K%w^!95z>X8uD4$NmF0y=U*C*tBM2#R1IT2W@&HbwLqiFW(@XQTIPN*lP
zX8Z8P<HOhPq)N9%=goJ9cLw^t;0L;ug0PDZ)cz`F(>)7?>X_p}XoU8t4~VpXLKAiT
zMCG$$pg{@#K#Vpx0$tF;fwX9~A+g{HDo=zGgao7S@GSFZEUa)yn{Eu9a@X+X=qw#k
za!7b!J9Z+ZkI*)^Po!2#ZHN#owO(##sVWSfAh$xdVB0SY7$zvd&5@uQegnl=Yz>x6
zO4Ntqd`0Oe{AB|2&_VStukXD&_AP@ONhtHAj*qbdukW#1L-?)Riiw2ijI1Fh<Q@9x
z;(aj4YOsp3{o{p!q%m#c)$rcen4PLYEv!=o`km~88pzG1!8X)c9$+)r!3=l?k?B@I
zsFQ)!E;60pctKDbIVG5y(gDt0)avl^vBURjUFYgdCtJsVhq1K+(k(KbHG~E0FF9N)
zV~8rvzska7mVgb0FepMEBBy7~2bQ5m(vQhZ_MbFs5am3~b;#Nm@FQsw*s+b`wZ$-7
zp7)xr=vABtV#Kb|f(Pg!w0lk2GKzlRxfeN}a~`UlcJYa0?F(>w@$a0l%;Myy5(qTp
zpkKYgQT>CHg6RD2Qn@R6z=}dw8t#4L919`L`ZWJbN;&WRpHhnNUnwQAX}^Z?^6C@)
zXrrg6hkf<w<0Jj~+eX&bwz@yoPRMnCc1iV!2%RH%F7xS^G6-r!9`+S%==Hm{bHkwy
zK1Hz#3zQxa+6euFTG|L2+7zHwGEKD+B_fx!lv+_N%eKKa$=z4#4Y7>6Q^gh|#45W>
zYoE^nf&1rHvEK7U>eRZEnN#fIU9H$I8Q8K<dg|7^x-Be~NU7&jnUX-tmCGc|tID(M
zZ>@XJh9yh9dxe=lln?Dy!*YOQ)x4PNoWwAR12F@2MBRItDtW~k@3vGsAH3-9suzl9
z-evWB)8Ez;+rWJc&dLr-dY(5zKRva4uOMqAFZ~A>Nqo6+$7n3&e%3=H&ASrV=zHF*
z{knCEx(oY3A}owvwA=bcWxMynKSD{>*{RwsgItT!3fnL94pTScn4yH-`Z#>4Hm{m}
z<FaUg^Fc^+!YnN_T6>0LBt1@Acm5+!F7+f+er1*;ONDnuW9K(QAlk@nbn13_%N_F$
z%)W*@M%f81hk3wtw1ai>BA0K=W?h#xXJS)qz!%?koG<#229OihkQVX3qjBJGgT~Du
zLl*VLSOS$W^`1h&<~?yeE@hczxFWkvBakL<wZBcLv#pG3so}vmJ*kOx$rpIYamP*2
zMr_m6MN|h*olros?YS`<l9xgUoob8yZW4dzdCkT3;=@)6l0oS@Bv3gE2O@xdOL&<3
zGAjF(1+GB>KZ0yn_>uQ^xR`%|viP%b6)An>gTH~FLzW1VCL#{PGb6tep5vp^z&Xk`
zA!5%>x-wtB%d`=#pzaTEA!*nAF*MWrs0#YJU+$Ss7tUH10q$pOn}Q;$g}LusHDTI)
z;rpYXj`>n*TK2Gx&4FcB882Z6ixGJ3`I@r-rQ=z(n=Wrn*V><1lrTN*n%V2!FyUKI
z&4cBxKlQ>W1>%Oavnr(%oW66?KE-I<Z)QmqjpUAMpC&`g4KJF%qQ5^%v-G;Qzlhh!
zO3ANu#*d}z|D^GI=D3+WF*>F<T}0?PI8T$|<!O9BS@E7D6wUed`o&Z4M-gH~v?++M
zGa*q28ac)3#hEz#zKOm5+GetDLG5@wu$8&_*0iupHs$#}l^*K{^52fiid&%vf_5qL
zLpNnmnztPUysBSolj!%DOkX<p*G&W61sH#1KZ`ggJ#k_1)2N#j745k>qUV3x94h(P
z`xzYVe!F0<of>n0?Aas3>$2Hi)?%?lZdaQ(b$IAmW3;g_6RtWBX8<quX|>c~-N5)+
zX1)~pRN%8!9<ra|e!3f=*@?wRH^iG_lrxN#BZ`IHK8KnExu^Wz$^UsY+w|2d{Yyqw
z1Sb4I&$mgG`M{mNyZ)E`*Q&;jW|#f&btRZ12+h4~>FuU$Z*F+5j}V1bvfuNO^UB+v
z#)huCE+vyFkHIK8soHNR(efe)^1~(UI(Av|vg;@U7rIiYwJ`!131J+NT6&+a>mQZ}
zs>_s|>>w5RHR)@ZIUa@!uSkDR4BO3<pBnGiGp}-NWX9<~k;|c9rE?w^fA3gCC$k^T
zfNwLKfCWx_RBSBEWdHtfH$IxpIph1ld|_LP=v&F&TG_9AgLu}wWn3}t6|Dd-M;NnL
z&TqcDy;ja$`mD-wDV+HAeqUau_r5l}-lfO;JNlu<rfb>uIGSyLy!A!k*`_Zfjmf|y
zrl#bfGd6k{&qUsO!pY|`K%K$O@$@U-3T8%yGWGndh;`RXSjj@}^JiIjNdS6%RkxWB
z$_sbVOuygZQo-93aWFznj<{`{scsU<=S|4TJfy(K38L6?D%)iOYc^45_*;*8?t`q)
zBPs`#!VHgXPPebm>GNB18yfI>=5b7h{N;@85d6M^TZsdck)(@!A{^c|q`oUp<(CLy
zP&aF&-~5v8#_C2)v#5TEg2-N{D&3^$8WGsO=%2s3po8-LpzaW4$#G^w4&k*q-HD!o
zP4U*~QmlCdUddEDw$|Hbo4I&4QcOzUPjE=8?a>Fid7<H!XOXQ3YbwR*`F>vfQ!&pj
z+N+;5Gu{LxMtH^!bJsjYY(o+rGac3?%>{*Vcm=+~=B!b~ZVbmn=NDcNJQY^@^&sDT
z!n#6ocY1BF=yQ=K5Bo2_KfDx#nR)pt7%ul}6(8v`Ka(L){=*cnUicOXzjGk;af$?5
ziwsp{nAWTpxSDF)7ppp4uKqk+w!MxyfHc!A>u$4L!V@A6*JCWf3r2unt{SSDNv+{`
zHyk&${LD^6Y-t<2xRTwt0p1?%DMI;2O7qKny|x{-pe)_?xzJ?Vk979^o(L(U9~gF8
zm1Ceqa*E2r%|c?|Up$62E8Ie0Fep_36T?42k8MZL%GuirxT#dqIw_Uj@oK`C)b{_)
zI^3o$!=^(VjoHahJB#q{5pGd`DG`uPZH$lHuA2AbpZsW>9{$*>Cu>%UPwA_HZm~Od
zG!WTbRd6EsQ(~8v$wuoa*sZs7V(}Iak--CMHs>$|VfWF6gC&5%hi3Dx^c^)*INuAi
zSReL5O-#IN3sD0fqS@e>rznK!jNyYTc?(S?acrw!Ik*J6*5CYfQRg^!lIc*xGq#oM
zJ7d@URv%|J>BUkMUU;9davtMj(_Obb?_Dy((kt+wij<yj#=9lX5<gyeWhBI3=zYz^
za*_-Le?;7*1udeD$a+rjI>=V~sAnv9uP6I*r>qaz@iX+e2DzP^w{1#W_D7`r{B9^v
zXx$v8S-_NzK9A3$$Hreg*OAHDRJ`;?Idyzao`LzZY>qiyeK)q+J^lh(KwAFEw{`0$
zJS_9?j*GUK%(tS9iN&Em`fxX^fs40aj~N!$#V^f+K~o<MwxDXyt+t>xhn$zqaW@|z
zDKb#eaXU%!XES56HxlD_j;AI(#XSxF;oojC$fBatXe6U&j@{Q1oU6CmbFE;!1r;jI
z_F`b+tlkWWGo`WZ!tXe5OD(p0hBm}WlqlkZSC#srI2&09)mkT%!gg)qH3cdt8-x4K
zXfkk^z~8(b?lEbK=@1&Jb2BZOjc-<#-d;Nci!0`?Uq7!*A;k+0=`J0VnSA{Wt8pg#
z!u~=fk_e^wZqX$IVM`)3dJKbUcL@6aWep}N=&6G{?W5QsjqPk?CR^IsMr|2P*HY+9
zAW=s{K9)>fPwlGhqEw>^#?Y0Zm;-)IwglxdLnr#&_bm@{oNP`=L*17YjV7_)T#_>-
z$b${zha;A;{hx=NcGu|#9Ckc1UdpfbQypXeRh~{7q$f?wPenQScyU)a7S$>5a&)FH
zBAW5N1aCw1gy7&CI62d;5Bjkwk$6<8g0RFv+q?&lr>g{an5Q-k=ZC-!Fm;pQ?x17R
z$n79FW6bOU4kl<&QB%S1Ja*v}#a6e7VgGLa6+U>F{EH(vNV|BITX$z3dnfjnKxs9$
zZdWwHweLsmTEFB}mSApuoEmTJ!;+rdU+Ky$lSo3MXg9~H<ekFSIinkf`NoEfypOnF
zuTD~%moGEIIP8Y5+K|#u%+ED-Zw|TRsov%e(kU1bQ#%=taII4fbo5>RTouX3nz|2O
zR_^%Guo@+}*HN-TdPaiqRdW{1+q$CNWB%nLd6Lh3hoW6HjJg{>y3H{V$c)=j`ebC<
z73_BDoQg{4Qf61y>Jv0By_XGVp5Lx8gfY&j$iFr`{SL91D2=UA=}?CDt<KSdvZ|k-
zQ(tW*%EuXQqbH&nM9cZ^*qW;*C-M2^@XFqewwe<jmzK}#ZdC6RzcsVt=W<VwfyqZo
zCBnF<)f&%*iif=p;j`11pRbw%Vw5CGEO~HLIe8Nz%kB5pxb@bdKFw_uTAFrKUAKaG
zcMdy<ddERe<FduPdLKOi;OFxb44@zb296GZ1V96yNdU~UgAGr>yT4Z~0RRkO%0E{v
z+n5+z7%*Ab7?_!`Gn&|#o@e=4ql=-BY!m8R9FBmH5(i`YiAuf&X$kd`szXakR|KY`
zi%_Sc!Ya~Z)CozB*92pK35C{B%=N?m?8O#|LPkToCW^}M8x)uNqBsYxOd&>s8^!;d
z*KRoD`@`DimX6oOdFsgdx(6@sZ5pSYo1I&^i0D+J>^)+nJd|{JdT=6rrLO3~Y|B}=
z79<)1s)-nIce3jgglNOyM+=$gdenaWS~hk5(C^S$kp;<$vVhYv;kR|-iAz*g*--_>
zkS#((>&J%=6H+_-S>^nRa8Y3nr6R7!2^1X+e)B??e9rbyeyRE2+V~-zMmHfOrd!Mf
zreHVxEQ~~Y&lvJF>%l5Y!-pe32)kfBrkt!F2cpCrd#Zmc30c=JSZNhyebr52+Ts?7
z=yX(QI2Uifu}E4s_OO}@N4szljb~Zs%^PXiAib!G)%tl%zf&hlOUQj#E|{WGTWdsq
zhcs+^CF;>IaN*(M;FE95HcK@^^m}Q2z$fav-TJrtmF=V}?XkM6%9p1I<BAd5ZnMk{
zzwII(ZcMFZBQ0Ep2qW2aQn&-`<GPI2F5NY&4$X=ucbl08K22Xw`kY5^l30^0%aN=r
zRT-j%_Gy9Wn-x3y$w{~3N26_+iw52nM}ibD=9UMD%<sRxZf?-hz;d)$$4oq%3IM9n
zY9Q}SEOMq(wqwC>4~ya_y}3PobdM+Ue9X6zE}(jSdO!T@l-Eikv;Oqe3Dz|@YnawG
z?B_7MGNO3Mjyf-ui4wqtE3sBNX>545dtPBayOW@p)aa<M>-#0ewHqhhxTQ8#JLzwu
zpE6gb|Msx+qH2)y?B3nznlg%6`|*1jHce~;@j#MiB+a?0`Xsv_F}r#=nUL7yYr15a
zO=g+%ro1wgdKFolOQGYQr_rEM+0T!u$#W}?m1ztmYU8986bIXd_ZKZMua3zsZ@J|*
zr7VrJmdJ}9njY_}5<c%&EpjJ4k&N!V!m9<I$_^hWSDW&;Kt(^jS9!Yz%RCje*)9zN
zrNhwZ7aOCdKqXC~N5lI<T0M&eJG_dq#BtbmtO}{}&cV`c4z*?KBn)RJHNWzko+u!-
zOT@pZ^JqUu|DwPD3oTdfU7^@xQ%OuocdT1!Pe1N(YeH7hC5%rs2FtT{Y<i!t_8C#%
z60f=-WdgAoxd`3l*H}M;rFQu{^>89(V&mKo<B!v%(%#v;CW$)2JA?@*a<)<6%~MwD
z2+<Ej!YAxraYC$ck2llfoq#rTvRDp7LIE5e_b6gs4{zbgeJc^DSQcZdShsMGOcv+s
z)}-@;y4cha2N5VFZ9AKTImt;f+v6QYS=Sr);M5UuAT{r(JfS(AJD)W$Lr?4dIm&~G
zI10q=97_Ve>t9piggvOY$IQ;x$oE=f`(B2GDewe*`c`DIPJ?{iY|hu%Vy-t_Vw3w$
zzyjVN^~quj2KnT1dE8}z^*JX$au}TLGeTV*+ov%k<ORN2EP0~PM7-Jub{{xm34CQ`
zeH6?Ajc*(Qh*Gq5KeoAJ(;-S$)b+!A)9SriFZ5}@T9>ywx|(|P-BxoTZ@5~Q<_O+y
z33&hxPolNZhrF(-2W9?;j@bKlEU#L&c>lq=pY?uB#mic+tysLs*SfN69alpeZC5T@
zygj4VSxb?1QCCClUNN&Y$1NxOi>@v%F47ryZWxcy0c=qm<Ja==>$M)*Qgv~zBn^r+
zHwF7*1HU=talh&<2l%MgtB!%|O2(o-BG}jOcJLgIyU}T_K6*c03JG+3wLSQ4Cp}o?
zkh#_}J>PE<J(?bMlr-#gyfZK&B=JA6H-~}_Q&<oU7o86lu`Rf0my>+>+!5@xI;*Xw
zIl#CaZ%a=a4;Qh{@%Hvzv4YWabP+OhNWC(1+()h~p#BRqq2jn8^{cbscd+i~1nzPO
z=4p?VV}r7__9WA{PIxe|KE<j#vF}|WzZh6a%vxy-dmEx$A>|<XXf;ARJ!a&Xl1Iz%
z%#ixFM3R`4bkfCo@?wwQDRzJO!v%WBB|6~@`G~eldvz#mdwuiU+tdf&_!KT%)=kzT
zjr8B4im{X$pvt(EdXjimImvFSf7mn(Uzr}JX~TZn5{>bcT1otOylV*ks{6HWuHOD}
zKh&k-@3v|O*_o2?aCvso4}<q1wzb(|eZn`m?U~<wt>X&(4IiuC1;fs~LUuL-+XykO
zxF(Wxq7u#wQ2gr%n|l{s*{YF@)fB1z`R8dM!KoEwW!H^2Z9gP+U3Ro>`&2Ho`z;eE
zb0O*Lb+h@<?XWdBBIVSv`3-L|70#Z#;B-ih+M2j293+j$-mx`TgMBC$;T0$N4(pOM
zV}<>K^^|>6mnjnNNptr#LT?fbH6G}38Bg*%I~eGR1@A0Ldf2@R^1<pLy@@!qvW|s}
z()sl@N+)fk+99TA1p;r>W;(h~W*LRtbyKD-Qpe+|ML8DYFOAN5geqjDj@j3V$(~r%
z14h?-tXJ)sZ{$GE=j%&jESCPPDa$9YJ3$q`?DZ!5#`!-;`|hwNy6s&RMWu>JZz2LB
z&45TJAJSA1M5H$%QbQ4t-jv=1k=_+lKzi??6RDwhi1beAAq2>sfZzF@bAI1_p8Gua
zADHafv-Z2!S~K&`yQV~UVY1i$s>vCwde^^OB5zed<uIr1W^jsx{&qdoy}vlwwPd$u
z5%wzsiynSI@|Q(g9`yGr@MPF6He(3)&MAG+yb$~aEU7vba_M6&%=(UjLy)_RL->l=
z1=cA>8ObC+Gf5ff8{BK+pv@|iBY{%`(um6yR1>(1|3npTs-$!X+`DjrWd7nmQiX@-
zRDrZqU)>S4B*n?f8gp>>hR3Bl(qdo!f=d1MLY4gD<KITDmp{RV-L4x_5MXX!chb6>
zr=$Bl#PuT2xA@CVaTzy5-nZUn{g&ZLb|tUPmi6~SMng`~2E!a_&X2W<r9{!SRQF9u
z>|~9Gy~Dv6WL+FRX_kU9MAwg`LQ$T>1~?^I>ZcIb+WR~2hPRMtP^6hMmFR4kb2ml%
zywFL4QqpY!sY9~P`xU5pyE_crcSZ=2W@BEbcN4JP+q0K`ZQDpT<?&70X-h^mPdaty
z#f-<YY?qN^Vv30iC@>k&()F4B?Hzg+6zTrz=ElnYv(pp1s*2T=fmKQ;Z>Kk&lbaim
zwL1!;Sg?m^y?eufU4B>$xE6-o+_1QKFtdLm9zr+F$y2{De1bGKdFMW*fksMD471u`
z($d-v%unH8jRSCVFZRcrx*cmo$D(bQkou0V-iC1un|N6Ji^p=8N>={3n=jho&6l}Q
z)ijNASUa(Sg9$g*_e-XnO@+s~keJu%ohqN@8Tk8^5z!|0kvx(9_CK*TcX8_#a=3zE
zK0oUS@^bXvDQ?ATDsNy2f#5o28cx)uIdqJguycJh%5s0RFCE^Qg*h$_t|}8)2gCE?
zD)*mN^f-&>HzDYy;MO6!neECf{@?d8ph06=Mqy*VgJF*z=hd6UzDK?vguonaR5p@`
zE4r(Bm_i21Ipc`4w#OBzC?EZeJk0PTV<Vm|TCH7(vEJq|vEG9?j`=$6X590>0|w#@
z1VnEhv<0kM946LRiW68MFAgw_s-*Q^2Th|uThN>2b1y4S`Dq-!mw3kudF(t<3Ip%Z
zqIPKyvmX?)%IB6VDv1F1_O>-#!}>suAA!(L;L`OeOWl2$wJEpciK50awc-Gl5*8g!
z`dug6?uapkT*^G~8Jfw(;W_=nkJE`PB@wq5CnN}mkK<{fQ%5y?qxydubtTUlE2{U=
zdWAJ9o@&EekOPVqi7`lRE95u4rLr||K)=g9IVDo)II*SGr35b04*5NDm$?mD>G&hm
zwV<Q~KRRIkI`f=t#D_E@0tVw2%bFIyj8EKF!e|*+GwwX>S5#%$s*{!T!I#___<A6W
zmwvbA9014uRRXA&QkZR(V>0o9=NbMRyMLW~Sj~nB<#W=jyNidEJPRsZgtxf#=V7Z}
zgYBIjpCiq64Qpp6074`@Dc09|TvUQJBdzDK19=TVj$`2s1y3cy3ThlbhhiwryxE|K
zAnY|`SQ~`#H)u=E=(S9kmUUeG+gtYj8>?RCW0RAmtAQfkqHuF`0ijoy*Dw1kOSFK`
zJTHT?&pq0^^p<(R0Kl=%7K918cjmj8Tsvk+cSSh`@tlb&NiGHi5};YO=<_&IT2`!0
zmhFm@1)U1*u`?6;t>4&SNhLk=qx7x0)-!PWHlE6*X{Q#y@g~RkL$*DBXErqi&=!sJ
zfoAT_6HN@A^s28p4E$u7O+XN&BMQr}0LX%tKrJ3M6=yh}o-{TjvQYXifeh^rQvSr1
zGW{$Awj>s%%~bZ!=TH%1{JJ|LxX~dYFgayoz4^mg+^AK<;DG-K5gxGhb&H-;86diR
zZJKf*W?}$o^Y5yDww~0@b<Im)Qx@s2?sVD+La+NIRzVaw8#gJ#ALkcNO_JB4p5vk?
zGytSMnCOyFLC*kN0OayB7Kh;fz&Pg%yq>OyZ}%Jzh7`Cj0C1Po?@K3TC7pFb$Z>fd
z5-;BiXBpwo*PEH5KeKLjJ}@HKc#8yioCRUk>taaOUs*RJz=zr9W^1MB*-j)t<4{&I
zB7&_aI0Gt_PAmMsflkGlB@CbR5-#bjuWuxTUl#zIf=j32waw=__U-gF-b33MrHwz1
z<bZTF;JrI%D|phrWpkdN4hiRxhYB*G9m$tu&SF<3dkbRmTY-r<a>b>|J>}#tydMc6
z#*JExheE-Dc8Wjd#Or^rUp<E>mrfhDAq1u(eAE&&<P>~efmL&~Xq0?xCwXF<M?C5H
z>nv=@R&e;_#71%q-clTIVtd1>;pTz%Rm=v#<k6$PWS1;+CX3F!1=|Cu6S`9_n4W&A
z%Djj`xEtihNVL&ZX9txOT2jWYOm@#JJxtH~(+9noidoB)dz#KqT&u?U2&$TFTd5$<
zzgOK~8sr}`Nz4<~Q)wQ<NHoJr?s01#Yf#$zBSW~o2tMREN5^xgat7eH4_SKmwyd;8
zZ%pX1OLuPw_a^&BqS0NwZs2pt&k{79H^$|dtgEDP!B#$4ho3bvNAL9>{+=39j&ru!
zW)Sx(gVIQ!>;&T~4(QkY!4fgLvr{8+3Q9GYBZNVYiPuI4f7K2t=3vYL_lO^XYZSoz
zw%{~E9UyHD)+;@Jc}oqQ3f4>P|1Os>*ahi2`^{>S9k&*2k>VEG=e(bLCT=g)dvgW8
zf<4@)sh{0-WJaMDm5=w?xp$i%{XU#HxY!7?kdnf>VO}lWu}-s&EmWRyoL^N!X|I>~
zo;*_6mZ;}@x8|6v+cUk4@a0z$dwo#cOcrkq8g&kb7n{Tx_Bc)mP98D5w1HI}@HC-y
zRAH#-Mi1RECl5q4GjifPEm9?T=Y)W+tD@B$j^?2$TK8}u)$>C=`{5yIT9IHI+82(Z
z{5Ars3g>GQhG_q;nf5tv`wkZ#nUy<S6xKb2RbN3;XD}i424Yie^Or)mx@{@9JT-}7
zyz=6(DiPiebZV{hPnS+(>TkE#-?z?g*UfCEPzS+eY|zSwp>MY8)Y6u1FG%bzo*Wl$
z!Y^x?%&uNS2|LK?zVZ4{?9i%J_lf&hL$7;JuMZp(!=3-(asHOBUF)t)<?M-1vHcML
zSyl`=r0dn)XyIqjOftExd6-w}CL7J&HXIw1BP4LPeIE6~>W|xAW0g90&o+qB7u@p5
zQp>kWIvPJntvoHO_QxjGi>nZViBGlEM;M}in<IQ2KKr3NaVj>>llce@gu8{_snw8&
zczz(G#LDQ_+5yi&7dR<CuEIXkFbioLow@%|%C@}Iu-B{{x`{2x{%U)W20h}bKz+<`
zr>>oI&+w+Mt=Lpwd;(GCz~&Csb@236HTF0;eXvz<Dlk5e8GEf}0;?}dU4GojI(W*-
z`(4iI>FavPE%O?<QAE{j@Ugp0*_2_ev-3oEr*Oqo3NGDxbu!Id_x`w3&5CnaQ7l}(
zmpUA7qhl}CtQ0z5RO`VX(b^7~ETC$Kbn(hD&pEqy_`8PDt~2mObVnDZ!DA20K`_Gz
z&kyC;no)A~#U<-6e=&D<qIX6|5s+bOslC7six1^n?yi_e)~vrEt{4O6&?vZ-a=epk
zSe4Q*h?<8u(Jn1$R9KIhO`c}4NZ3gHyw_TX#3M<5Ly3rm1$S2-W#+XwW7l=MK8b^3
zewsy)acL)n^NHqS5#1=+j+Ox|4qhz~Vb|IB^lWUcWQlU^kcFZ^x!q9>L%uFZ%WLFY
z*LB&I4hce`l=E3RnlNYNY~KCbye?bo&ZwE}?kF!zFCHG$#}~0Jt_$FyEMoef{=v?l
z;X6BZ64aa#cDaCeIsy9vXZFqG;I|7zBQ`9g4)#QXx9Uo<#f?%XU1l|7w)>mzX@)37
zfEPNY`s0w4&q346+x+uOx>r4A4ZO0XjLiZ)TPPI5`S=;PeI=0!>v<AQSbE)(ncZvF
zW|`(SU<tL4avlCCj~l*j<S`0-?<UOJ)$N1z5}~e}oNlG<+__jq=Y#LVZA8g}u0(5J
z`1iT^l?{)&RZ|5OVfCnk@1<xjx>Lqw$|2H|d|h(8wR)_*6PXrC2Z<xu4YPn+M`!O2
z(Fc<!7abTE?uHIWZ3ORI&8nuK1h35b(D4Nh$G|B)7>{o0dbv*Z48Y3Q%rgCvMrqc&
zV-NehLcO|H1tz6RG_eRRvR&-Dchv8ArEBF=xXM*WuOu}A2@NNSEZhe#hRH(I4^?3n
zL29WSjCJhRQDb-wuk9^sDEB4F@lY8_9o6pfuTq&afw&r@0Y2@f+Ga2NFy5q-#)KEy
zjG~68Ev?#{0`oD_!9B5t1~KFgF$J2_kDtXrOg;7!w{|-=7YoI}h6_UbwT$~vby!hW
zmu<$>WzH|U!(J|-ID<9&vL8F1<S?)I^*|a9)=JrC&&Qbv^ju9mEsO<^czgXCR`iyh
z#$;~fq*c!7NWyg%nDVuiXRtPsyL%<d#0Y<;{06~<@?e`@v}9fS(vn$@+sW8EIThM`
z8zcM9Vi;|L6HTA{fZRxS<r}V|8jAe>y$4ZBvIGsQs`5fYYT*;WneU-ix45{k?MFsD
zy}sfr%3*Artk&m-KpzlQplX$Njg`W#FcT0H;#8EvO1O1UyLI~aw)Z_7D}O;=H~z*2
z2is}Lt$9kzIeJ#{x3wiJ-XPg~+qk?-I~q^xz%53IsiPX&uTz>`!}@$+=A*48zAA6!
zi4^4isp4TIhpef2b+(3#k2b8Eu({1Ov*Da>(Y~d$zoJ1BR>)ttrcyaJaq^-B#*h6N
z5h1puMY9H>8oHs)51_zJLOdoX9B&#TX-HPd37t|}M$r1|eQw@Oz{6B97(Z<o8{r@9
z5wA+8y%#ogf(uZ@gl?6uUpEHjq1>(I)=t_(R*$Cd<LC4jISt7}{RUtKHj|1;=9J4U
z)68NIXeYQrGA&3*pc~~v%aWfZwY^W}nJMZI*e{VFgsiD8V*_YMZymv2H$4AEr3HG>
zW(gk5MRj`3&9N@M(#s9x{>g@A0tUu@fgcuEAY%`Kt}nbF6uxG>MK|{$WEe|fThERM
z5D)O`*IT%jmdg6{{lziT1Fj1Ga$uM!NkVFp)3kWnr29H31#U6KO4Qk$ORqI%xsq5Z
zg<<a>ku_*?AGx;)%UVC5HU46EozOw#J%uh39ZEGM<e8hkE9ap<_*uc&_=_l!e93_g
z(K&7OxhTx4`MR}%w&bL@xT?xxh`jEVYTlH0-nRj?XCI_?oww=|LrKJiv+N)MF1Rou
zs~FfSuP7tFb&g7vga;pLa9&m_s@}D;>|$&=1*I>VqM^a(!^hxlJ-K}kZ;hv!DB?E7
zblU7a&osvmP9ik}V`iA!YIx3QaHG{FlDru3i1>Z6s|W=53gGW$npPKs{xh%MVAGhm
zLV%~XfLpD=gWt~FTKLm)0|rY}MxtLt*i}|=0o!~5=pqmZSpH8pn1V~vV<b%&r!1%t
zSY@cIE{SBTTfmS!GG)2grtt^zRZVpS`s2_$AQn~>tuA`Nc*=P{Q&FmCO%Il6X(OJ}
zdm?>*iq((VY7VSmeTuYi&8a@kKrtAfU98ZM;I9Rp9PqJbIFLc*{~X?|!f#xU0!`0Z
z6L}gw$O50`#nLNvgfr|6=5OI^5AdK6`M^+^=^wr5E569v4GqA8WDWqG<`sUVZBf;*
zS4S|1oS#eBX-hG{5}sRY0}xZShpO24bAJq=uRuQDYx^kh!d>{vIdN@#|K~EV`v_(P
z-s=N=B+%wj^{o$m;t;WIXFK8^23$SS4ziqwEa#^sc;*acK6NX8{F5i@-U7$mwtyFX
zij-~<06s+P^bnYVwV&hTcT+X&`)7G1S|vv;bxZBsNt%4Lq3j=FsK6XC*zVC8Q0idg
z2j#;4Fx#LH&%Xk@8l%6S5!@L326bIRbqKW5efq)<Jue<vy0FFcYoh(i*l%Y>YrTEL
zPlAc71{(QWaiq;dwIP<>(UCc%=L`p=EMTk2zGzuPaz=~87Cz*7-vepN?R#SC=<*G?
z&KI9w&Yyn4%a3n%wTW9gxanXra&1OPrgL($YW5E=_Kp;g^hAD8?7-Ono~U#V<g+El
zbS8KOpV&Zis<)<>PFO~q=Qrxp<V|GP;Ki1!diA(T?;o`kR#;6H{RCVAd9t1EDPMNX
zg#AhWG4ca<u3ec|M=8eWxIRs6C3nkVRY8K=EG-F@!>M7I<lY|Z9?rZj?4T0%wI^`v
za6NQEZ9&SObb`kyrD^S8C}Emyht4Qz>Ei1J{nm&|a)g7^^B)WHIhlI$FySX4m@DU1
z7_Ru7f(o)HsXD<b6hSaLcQT}*MdLZGT`5DQyLE&W`Jr$VKB&l5uDHmv26i9Ye9A-&
zb9Smd9ft2`W|!@vtM}NS9=4llMEh~T&T4gT<tbU!J*4}NK@8NHCsg~B$1my7);ag)
zsT{sf#%>)Y+(${B$ss)vH-=B*FXSlc>Hj`4&7aom!!UL9s8dLqR3^?;87rA<SKUv~
zB+;zpysB9nrKypB^t~=sKu?mhV`x|UxY({tdwh+nd5A4t4-{LKtCtsf;MVcBOb4K{
z6EQJ4a{UG>bouLvAC_cN>`f+OzI6Quadk(naAl^Y)h-uJ>7wsDe2Y)r84T%Zbn=i}
z^30et-;Y)pI3tzrZXGEVQ(6mf9p<$aE|jioZcwU#YnW*5FGv@|a57vzJyU{uS4%~u
zDh_V7^6xz~e?g*j=Zk>((l|qy-6_i%_M_chbLa3(59#mLfqi^N^TI^^enBuEU;#JD
z1-^EOVBu6&=42;?X{ji|6~i~EHchiLYRVN0XTI4DiNM-t45@8?mUu)w@zJfLPGZ;&
z&+DE^E_kv=<V6|-xB!;fCETBUM(XOcHQqs(#VP*PcP?{2|I8vSy!r`qjncyi-hF&_
z)CwPt4$R$cLSKNaa0ybEAjXD<zAn*!v4NCDe5osp)>aW!XDO=Kd7>)0;B9AkbcaQ(
z)_SZ#;-;=jRnk#^YU<7?_^LVTMBrGocWHP@(35i`uIF?zjv*m+XExTYqkALXgSh=u
zFf~sa%L<s&pwb#dcJ1iqs@X(0lKtQ_RF=<fjSiLQFuCnGL=wPFCsE&fe;DE`FmcIx
z=K0ryw{?w<T3(%b*qq+Y&5hbxMw;k90_Fa8O$FyL_uM*dI%~HS8aJq%rRjs(L{e;)
z9{Ev>I2`U&-@?>~jEuo1vUXqW(|PFajWVgT&}?Zl8S2jD3u)X-D-{}td-IJvYNpx3
zPP)}kY)8Zb_sSfNZHfcBsu2CN-4t?-GKmE9NdAhB1-~eL`O#$`Lk5pD6D%g`&89Hi
z**~C2A8XyQmzY2HFCh}gUC@;mV+1fI2jA`XV<q@(ItD5Q|Cme#IR%9?fN0#IQcqK=
zTbl4E7`$RrY4XrJ&J8Lh`jCTg*Zcjs{HVHt%sJOg@N5Y$qIJr_JF6c?LN6q-IKX6$
z4_ryw=JCpqyZOUV{f;gqg&_X)*ULjl@Z5CwOn6k4oQ~=5J^h$xq@>g)?c3=7ZZWgT
zeE4bjA5DTwC@i?D0{hjvBFlP+Jdb8*ZtC{YW;F(c{py$wvPI*CoEEXZKPstpdSX38
zoe*AsQQEUS(@h|GSqG>;Vk3u}(AjkCa+Y0|YUG68<HgDwly&qEXw|H}lt`2jD}0NR
z=_q6(ULRc^djEl=1OrGXMv1NHXEInCz^;*7)9bD9V$6R?qrmC#Ax;smh$8KvLIY+X
z^HV_aWH~A~??Smjy}G1);(-$oUTK0YXp}P6qw)a!*NmB$&jrPfJagRXe~FAQ>BYX8
zr=IDirri*w1KLAiW7eaI#t6BkM-TqXi)%J=5opf8-8e0C!572QrQ-jA6l7ecZ``E%
zOZ^!s#CUVV>q&WT`|}*HUk^hq{{IBV;w~t)*~-JPV9e)x@~6tan49DTkDY=J$Pt8I
z028ad9kOkFCerftx!Nvgdefijo(G}E5-k6zDmyuX_iiNQI)dLMwn{XQQv>~22Y9?{
zD!3koJujgpu_7Xc0KYT;{dX!{VtGCLkIo1%8@ywcN?e>`m77N&hBmrL0Alx6*HULs
zYV7xc2OJ!`9v`-~@Ms7TG0b)cdps!tAdoPCh5o;=IDkvHqiXCYezFL38!!ISXE;Y=
zFl&8=f)HQ__z~h<&;K_91meM??%4%^(E$IA0gi(Q?IQOF$ZtBYv+WEL02ui7He`4~
z^Y!turCC9HzXdI&1D?(NZ^;_x*oVfW82^-=y7hlo34ye9cKc5#HFeLw8d)W_0<_Sa
zwxIjn<cMT4`MJ!Br!pT{9C>7a%w0!y1EVLIf@nhl4RiZ}Qpq8)wiw`ON<!bFI$#nq
zfBa5X)fC?9^^1C%)ipNdIly9c>jHm`6!DmjG59eR`gsmZd({lN2LZgMPUfgY14U)e
zNT*N-8h~~bZkj9*rD0RQR_few1FOpCKY9WpopVgrn_^v9u&lW?@W6DPOwUeTM<YMP
zl6Ds>pa<$n&(+)31uiDMxbqK50cxXae6}u`EI&3jp1Vf=S^=GoU2EMpeZerl9M^P&
z0#`|aB`u4?R_WMi-FG9a!;~&nN@mLGeWumJ^|b}Hl+aSoa_uHD|B=WX!8)@d1lhp^
zW9#m5b#LB?osFn_2AEtvCtFwk|4mO@+17uhR`Hn&88&9U=>9FS+GIXyZTXIeVC)sx
z8t-8_?Z%;$s+$MR(BQ%!A&qTlke3aqhoG5BLUR1Z=2M@Eb!m57G1F0KWhz_kE}vU~
z8?_{0!jp9am$-gpPjDVFTtL*ig2B01&_#`e%IXGew_YE0g5j5SJ&*?f85tCm)Z9}s
zSqOoCITORSk8k9=)PKp<+wc4fucqxM-_zv3yxrM0WJ-7DOJ<6NkfJh^1<Q}6BZK}q
zLqH5};l;JkZ*@QT040?G5eQV*%d-otQNe45Il7hgn3~-;TP42-6RW|$ye**Mf<+Xg
zM6eNWO`OxN<)aSZhV(RH-6U`cZdk<O75P4oOQq<~Q09p`#{vO`5R28F<e<N`u#BNW
zrsor7>`oIKuM~#LT9HlV1FB`N(Rm(yRR<GgQ#MCTklY1O5{t_W3wP#~@yDdP@4Q}f
zBRe5>JG5x1+zzdSVkG=<Dit6WgG`~WxoX`vMyNR<p*>e^0Vyh%dzUs-mD81n|7?r+
z?&|}55g_2B=T;%M3MtT4ClhAz8xzv!IusyK)p}h0yRTOLtxgCcXt5~Hv#QT>)d>)*
zCN%!61**;>j$-Y93uKQJizE&T&6y`wJ3nfuRVeO^LjO&OntQkQkHXbbVl`mLAaLXM
znXCms9cBGnADf*Ag*!${I8~W5>n`GrI@@j)Wx}8Op}x2F4{q$Pos9LOU}_ab;|je7
zT14|#vev&v{OTkhQg{emE$<|j6LJjfAa1r;4S?S@J!d}8r&7{4PPdeVsJY^vsZ}H$
zy@W!p^dvMFGVOhL#b|k2s6L<i+LJKz&=Gzl`j<N@ZU*r|N_DCiRUH!=?~PuN|B;<z
zo$Bzq5=%hn{2$`eb+@c&0zo~pJWU5?l`XqvH7m@5qqquoJU?yWsCGtbVzI;`{e=1?
zeyH^zp>Eu}?Xk?8v=f$_TVU2vCmiCw4dLkG9&tpdor*q&wc~BGS0dr0hlkc-?qY>@
z6ZDFBTY0T&seazIQZs(CGBde-i@>8~dFE48_h>BScihQ5rS2mv0qE;XQG3{NaA?V-
zNi(<a%@qF994t6rw5jK)9)Z|vvojQO%twNQAM#*=>DLV<9ZkH*z;D~ERVLqBXH3Sf
zIY-7WfHwz{A)>EOqAhltQM$^$#IYMoI-U=OwUaY5yF}@*eW!NyZ8mNZ)Q96Dt}i>+
zK4ZExCX6|&;;8a9cXrYY(4#`#qJ*$2DN(tH-e07d%CSs*gMtaeSxKcH8{?&C^r}B0
z77&o&7YviX5A0ll6jdC)?v)opt&_ln?D1+1bsFZbv9TVj4(j{#Y+9U!L0E861LXBK
z#@pDK=T@GuC|PLQ<4*LQh!3qxfX>t?$-i&F_j4vR;-i>1z+`x7J(=->gh1bR8@H^G
z$F`a|Cf$>m5LGvm*}Eeb>hgVA-06_3)Z;lpoZaZ*qaWgbu9H6|ec~=Oe(mQe?UIM*
zQ%*WCxTzS9(2ezQdA)5ta5&g_a+BAKh}RGg66_z#zdyZv_PdS5_Z&-+M(MP(D;^%r
zhVFwx{3zMzsdRc6eyI<3h2Vtx_?8YVj7GuC*)3MNlctXHflDSW3tm9#G6A!xd5$M&
zi}510nSJf{vP(Tk36h?5(%3xCZ&+RQ%1Vi~Wgl=2ChO|A!1y#eaPUac;yKbwKruSJ
zD%W-xYhG86;ytb8u>p#JT+9Ta2plEl99N$ku2&a7EM6^2!*O!7MWyebMlL7+I2^^7
zP5YQGXDosW@#INt8V66W*Z+-o6yheyBD6u}>q2P2ii+$~0DcilU)e2I!*yQshEDqF
zeHiLK>3ypMORh+UBwsCX3k5h`v0h`9i73IjI+Vl#!s|eLlF9H=1|Crf1k+8Lf2l*-
z!y*wydJOYYX+Q-yP7ctx@QaKjmn3-;4bvL;0a|z|3@FXMw7>ulz>xKM&1g&g8Z3CK
z7e*rsGMM+}fyMHJTO>HgZE>&q%`rZGKxwx2Ti*^T=&UGzE;7_qv(U?T#+k8`2T5Va
z7J1JqvjQ#iqIAO#fSNTRxDDJT1QNz~Nzm|S3Ij_L5WI7VsMNz%YWGWtqphoy@5eEP
z_eJWr0MYOfP=<ZRd~=A{OA9|{k=(60yQdiooxAb;p{rLi`**JRVp<zqlz^2x9xD{8
zb-xRq(DNETcm%Mi`xd1h0CB+AWqmH<%ZWCdfdK<MLi+j-Jx0!)r!x-`!0|r<k_#`$
zSP}6exdnzc^;fNdw0m#}6k~xV!!{LM*)n7IC=k=vTMU;0ElCHq9ON~w{qDIG8!$xU
zVe&os`@Dud3k<URK45NqQ{~fJbEW`JcZ(mpviWBzR3v|EPF|Jd_!aHw|Hx@TrI<8x
zp+-tge~J_+zBbJQ_W}%K;w#*t@j$ZSl`E|WK(5$y4%+M;>`u3>vo*X9qdRZx&8U0z
zZqp49@j+~`F+i{|=?a~d{Chrha6%5RK|3P8#dEo+lZI);Ki#bn$nnO?XSX0lK-5l>
zl;F<ONoNOz$a{nk_r`|53HpQIK$Tsb8q2Di+Wk(+tpxmPaw{bO9STGJ7u8AvVQne9
zg9Af`Bg>W2%u#~<<iFEqJBPl1Rp~sO6BnqXCSLnnPP*BL4HmF3lP1JhQF9z;Ct{-G
zayI^mOHNnwEAR7iYs!^*;N>J*XH|{-!;fs|6<Cf-nzsBPO^HBaphBB4A*reOtSd&J
z&2NN+*P&)91R4<eR!HSwvi2%!bXM3+2PPV8hp4+~*k!o7r?~E{t2^XxlI5)bG4yED
zSs$^{d}O&X4L;P@<TXQOHBk*$7CY<no?5GMGO>6mp9GmLtUK)Hz>?((U1P1PmX;g6
z1kuOjGAZOU+ha7HnH`?<^ydt217L6prHJ(u`<}!z2FJo0CtB0Cl+tI(85e1rzaP3)
zrxA@3H+gVYzZCoqM>}!lr;s4cIfZwvLF3z!ip6v$OEp<e_OSqkGqeFH+;}@c;jpm)
zg<B+qA;<ow8j6<kKT<>pkDPX&F*XNsK!~dHzCUQ_21EU=AiYdRNxBwvkY0#w<#{4K
zdhuTZQcqb_0t@SZ@i*kxbv>#f<<yy~bnpAznzd#(&#K$0BTZB&hv=9buN5V63<h7)
z9Otnsjp}R|q7TkU(Id>Ke#*EpKT}z3y$YQG6_^Fyw{;{&3wt}Ku2*5ycAq^}5wD7Z
z)SOi!?Kb9#1yVwf?;ykFb5bM{%b|+v<r4$So3$Hl_rb$ES}NkM$PS1cfUh}JqD|y_
z$43ouSDwY9R404>h_U_w<>`u@`0hqLg<2wVwJOF_C{E34Py|!ZhSSEp0W<d?_0G;z
zeu{*~?iqdRVwGj?T*5~7K{=hz_)`Ux6TPGLwY(vFV1K+FA{1dA3ls`jfkI&=zwo7v
zc_#gMxIyT@jEPKD8;HB+E*1&B<cR<UUUv<{$R`2?$ngb`m)TmAga*-{QOO#eYMgeo
zx*?2fcU;#6`XtH+;~4UscBX$rR3f6TAqh$`j*rs>7E7kqD-xk!5Cuua|E3S^l}0I~
z?g9$gimW&$G{KV7c6Wm>#>rlI|GVVIe33U<?=;rr(BX^9p5*98`gl7Sy<lv~DO>^B
z;k_%vp;Oo*)$grb`*y;Z@x~r|bcMBX;6iXWqeiW}U0cl}qqB#;knhYhLIi(l$o}Y;
zuJXThUDUE9^=#iV#Ukz3P7N+N|30vj#x$>CH>OiHSnA!ees1jdO1u@eGlw|CDy8IB
z9~DX2B$x4*=8i476r{Xh87c4Kb?J#K>?!@ONbf&yM{~ULWy3b_C&w0Qo=CI&-6QZT
zhZ+&3z1lDOuU1LPXszPY_D&tjak~HfJi^pFGa>#<7cN}8{?E@N2%g<!s9vY2Cw&7a
zDLUu6IFAXImb-FAal`7N8$BwBUQSu0<#FF*=`E2L403EiF`w#hk6-io1|odf&b?vG
zrbQbh?el~v>HSA^@>t5gWGgZ27oT?W#qRYU_~!a%SIGsQrsgkKF7w*z(OoPK`f0fT
z?h>_6Rl|dBGYbD8j#eAcRdb>X4Bz6>7(p_^$3YurY$3PgKedRAFDJdK7E51}l_oNp
zw@N$i-92fEqBbL9tj^DQ#O0E?TYOKw<F^0xj<$K6u(_S=TR#0bUm^{!x-=7*>nQbW
zn{bB*BO1x*RruOyVzJpA<x#Hp@F~r#H^cKLw5%<%-y2h!aXiAWX?utg<Gg4VvI#TS
zO;JQfJQniS=uK2!TCRChIx0larBFltEFKkYRg(*j+m~t?AAksY>`k9Igv*oR8tYCf
zyb}-fTXciz_h^S^JKr=nJ12@9?9&PM+^SG|6X30~a1)(O^{|nDeEFGN{+$rYZVFyX
zrpC1>s|qS2g_l$_LBWmWHM+wBZ?DMQt(WFspppYSMoP83;DXLyUjm^*gH3Q%L!s+n
z1aiMH4L#XB81ep-?Q<?z{i|G;<Gt^z*G;YREvp|6t~(1$-aRVUA7iKq_d7tZV{GSE
z>W4U9y<W=M<l3&>6k=kDyCn>1Mq5q!p87K%XbH8v_R=J+d!PUH+bGB_IHaX#@Kf+M
zNdokqbg`!7#||l$Ri*r3D3{t)VV+lv8*gmEKR)q7*~CuZ4P;m;(LZO{BYcL<tAov>
zRBzPt|B%n%>U)@1Ue+d^buIl#qulZ5L5^=_&n#~H$|%a+y6a_n<7#Ny?3dogvFy;>
zL1gSDA|X+sp=!}Y-+kmcRevp=xSCD(nj2?dldME-PEFOIH|wx<h>I6k{K{ErnBFtP
zX+Is9TzLE|d%&#i0%Ml^MZR=Nzp(pyAUm(izh!BEQ;03*VPtJXRDacP`Y^tFO)XJ*
z+2Yu3T|B6k(#6^kc+(MX<bW$WRVL<=2bs@%VT<JtgA1>-wWP1`L4<U0$lk7Y50>_$
z7dQ=FH@5Er?TOr{3C51(PbG&3?Y6OoS4F(GVJ$=QZKL_F8h32s=eH)Uno<i+ZP9NG
zj6xA6r8_M5yGn4s;DIwVn%kR{iXMF1Dw2%#)t%N4@7Tx57Pyh?v55+MaX&3i)Fb+Z
z5-(26#8}NWIoebhqE(<d%+R^%NAAa4^asfOoe9!S6XzMb&oQuvNV`1UfC28zO5u08
zk{Zgg^>eX43}1QRQ2qrQGf{&%c|`%kxrHH%qpu!ulp|M7rc_HJ7R;QrR<#y7EB9u1
zf9+JeRi#GnpDr0NE3DjMi2;4<(wh~zw$E0t&@Aro<)BjCz%A?vXEFNW>m4sv26`u5
zm_Y_gAypn0T>HM3sN0*YRWi6uAf{1E-l4ZqIle49i#(_&L+ZL&Q9h%R5lxz}1j^`|
zYhdKVX=waA?4($0uRM<Ymdw#<`G>D8OvXWTx<<cvE18VUceVl@jm$y3Yu1Z6X>nTi
zVvuEo?!pPBeDLxTj9UIWx1TDarLVQ}wyp0+)o$r-F(c!cTN&(wNy`##nCnT8Yy*cY
z*X3jLKYK=<QvTD8*ZU9E$}`C>T#&&0bH)ms-*^qp)l;W}k0g|54c+(LvXr(_j0~ZW
zF1>YcA$BM0EyYWsXy@LY2iXTdVgo~T*a*`>7b#vq{-XTYrqH7Ez3od}3pwpm**-1G
z=f@3K=dSvY`|X(P*O#Eh95yG#ofo_G<J?uuP^Qi)xOJg5@18D9fp!UM%t3%+jJ4ZI
zP6qgR4W&2K_ZLkmd9uSy)t!XgX>NOzUZl3z5BYN0dZq$OZ6V$HiK9ss)9>%cPv5QR
zY2yYNMODh%_XT)_J9SRHMOef&ut}x$1~Q`u$^8e=lfx^PSIg393rVgiFda3H7YxD_
zYnJUcJKJWR>{LQ0YE4Hh45=W-7N)MH3k%$dz>|Nflc|{}E*pW+dL^s@Q!RB`+p?#&
zs@vuVmv!+{R9|Dw&}p5CA=w@TmwKS!_%?IgaDS`_r-~&IJJc*A2bNI-UZG^n(f-L3
zDsb52tB)+fs@NgmL%VUy4ejFv{a$WqOyWzJOB-neuu4SuiMZGne;N2?O~pKt9DeG~
z<nnC8aB!0ggKjRW&n1Y+O8Jg0u9-(Ut`FxA9wyO4w%;*gS3d8y-?V^%3w{oC1fk6!
zAXJ=@JFbHkHw4=J;y2<H<yRf-mOv}GP$sF|bUL<H4j$iG6Z+;UI6>6Yr&W^9`*u(F
zK0+wi1fDQw8-eU#q>(4H`anm?LKeQQ_fb@VAQg_KDDo<k4X|OPkq!U)>*B%u7d~C&
zHAlMPEru3U*_}m9Bms0ZY0eY)SINg{B02Q5lOdz0@JhHBb3s+vNl~NS{w;G%e^yCn
zdQU^DScxh8;#HJ?MQj53+;I#2!Qgu#aT1aEz$8nup0Vn;gX*1`PkIH)8~PAy4s>QB
zg;z{<k4u$!17lTqjZlaM!MksBxr}Ui)k@M$a)E}Dp_?Lw0d=BT5i<?+`NRrLWmb8|
z378NaTV6!$7Za-wW3k!TE)VsO6|uqov1887YM3zuCU}4NLQ(I9;X>i7q-2RRz_9xc
zm@WBUI{dhbodl5ty06<>RLm3}`t(o>CywfN0=LX4D3BGRmv%nOi-3y)>V)3Dns1;#
zSQHVke@!~#pOLR)K<5|$_;a^^C-K(@-UpD`%o4Tq9u2va4|Qd=V)n@tq7BnJQbMU|
zTFskf7UvuD{=G3w)91`*20EudF9RVMqXwsVKK1F@<OU`SzV~5{n|(cRfnV`_?5+2g
z$#RWs!?Emd{|tVbbvea|nDL~x^!WO9#?eb9yk24xe9~qW(S;SUV2%%SbNKCb*^`0M
zr|B#8FI@#D54XlUZ>>pHlNWqP+HCZa0+^o{725!yp8(LaDE5x^+&mcGLEL>Sq0=<2
za~kZoO9>s-B>=}Xd=OgiE>Z-{yww+3{b`a<q3~yW2T4+3d_fK`F?TLvvALsr&C@!I
ziMK_rSpse>cr$|!`tXI3!bZG>#(nj_XRt`We5@0h>z9!h7h46RKTAV!ihkmi6#ZDC
zqKeppFh)ILYrsaF**4HG7>Iz}nG;8EiJnbDy^~>(%&c(<Jfo5>R}*;EF=+Qqo>dfi
z`+jR;g~51YWF=_F@-PEx<t&N2VF;S9aUbq!<I{Cogn547j*GioJLhb+ba1>+zI3X|
zpS){(y6KABc_6jpLhC&r^uBo$I^nd$D!PGue7v9nR!odEV<sLh9OERLu1Z_N+0Q48
zvFWO_5j!sybPC5V8O^8dUfn&c4KKB2u6}m|+ej>xBEBwmR{~e%A_3N`w?QuI?kd_U
zEC@`bYPzZDAk*g4j-$MH3PN$Oh63)ImS4Skm6*kNoIW_f2femS6WvDkaY6x#yl1r0
z1sNatiL6?E3qm4pd3$fxmXtGzqIg2>;{6b-_9al@D`|_IDxb*LGG|Q7SU%+y>*K74
z6qI`vY(D7@*{Hckxmm?TJjp6KyzP?VH{SPPxSVuOjd@OpCt`q?u#TlC8$f~`^A>Yh
z6=Jeq4LU$R`Sm^1)X4KMs4{!oV~aFE2>z9=Zg0_%(hkOOdfVbwy2^Y7?D9{_4meBG
zcxEcddqTVx^c2J7`b9@{PM(!B*+f}S4CUovgcl|N+nI20v0W;$w0njpe>K_O>QjCZ
z=$mdRYDnShBGZRRT*%?QxuY)ZdVk?N-X*vX%it7OFN;9Y_mVdkFH&bC#T-RUlSMBW
zGji1R<V<HpOwS7xgc#Q7Jo4nYi<?igUxh}hA=W>sDU^b<Po`EY1c<#$EW5K?vZM;c
z?D(En31T>L^QaAQO4R|h-_>BhV!S@fGe*oq@+iiE7exJZ-*sra*t({=ObGYGBMCsT
zcc@|ptCkT%<eX|_H8xN$0#)Sd?O}Haw*#7NsPd+2S7ra8Rb%FcZ*WAM=yCKwYD&Sp
ztZ8+MOkNZx9l*`1yWc&T8Djfft+dMLkxmny_14`c3&|#PTxsk8>f@U2w4USf<p+Vk
z(Bqt?-~!)Fi!{j|#yg;DjDfftncXpuZcla|rrg8Yw4(xW{7AF6PJtvsY$U?F*Q<yu
z_1j6cm6aD;o5zYCc)V6W^T@8YUdVm1B-^bnBC4<)hKVj59Pw~T9x*SO$Ho<W_f{+o
z5T!n7hLAo4X)BTghXc;5C@ScRD^6S2HXiLWw?F%uSCTO|Pk6n(Q8GYFF@m}<Kyu_8
zWmMT7u*<qT6!tKY<pPJ!=>5_=HoMG-H@pKoNO(|e@=~6O&MB6*uW2UFgnpczLub)>
zqOMFyrsnMz6Omf{f`|NT43noV??5jfqt;oI=G)gZTf+j@KSfrv7Iqn=n<s0Cp`&VM
zEwVNE!R&KwFY}fR)AKSM4AO!6_!!N{0JvxinHm~7pc}!TZVq3#&yQycybclEr4yEt
zj`B+Qf1+BbP^aLJv83*+Ve#fPWTjT4QNGyHulhiDNuj@#N{R7@!>iRkXXh}-K>>8%
z<C>fdq4jJ*<Mgvmq8d+oH|ekIJo?krTWq8fu6EZvjuU`q`9A)90RWKtUPzB7T+9;l
z<odWLv^3xh0^U3)-1z!9)2A;c&8cE+9DkM=6`u5N_M4xZpur#0$=+Q#Kc;4xY})xd
zRXNiP7iRVZ1v>%fw}biLUa!sLgx&$uKX3c1GeCb=j?LUz8yM*yCsO&i=D&RT7W|;J
zbDS6O!MT3{MEQ;$gPX7aFDB~iBb;1s{ChOs7F;Ur6ngdVC?rKI122!KPzeE6{~;WP
ztkN}_K&Tv<b$y&>ek~PcIHtg3zf3{z_COTY#Ugm6Yu?W9j#AY0Koom7%lZIbZ~-4@
zjD%C^G~+VSS*8uDcIL9n)S$xt-6JmU)TNz?H~)P$sf^skU#2MW;k$SqzB<ufDn_kl
z$C<33CKCZ8<pV;kz%GQ2t6vD&nae{=-G-p*-!0|#qGqHMs<Hf}JR)}k8E_Pvac@L=
zSC@90s@D5YCm7aDeK$@OmOM?TvasAKbzBn2<7V}W?rn{Ujgz=kPY-*vA^p2bT>rZ~
zZ>@VXyHIJ1T*s5<sGF-FGv0k7sOL{rC_im9e(>mxICWB<ccp%%ik-trGj)};P1gwc
zRZU#GEcA9q1cOZlU%F<uR?@{pjID5@>&nh!!7A?@(|*a$&)z#Kz{6*AQ}a!9ufP(9
zMP9#7)J`5?jy^D<UV+y`^x$SQ6=SH7rN|+2Y6z^ffF*!^kbG*GG+X)FbHqMQJ(toR
z_lgsLlmw3O^&^vD-_>-Q`@?co-akV*3~BVcc1wcS(lblb^1MPF-};VUopXKT7}@&U
zRi<0NQ@6&Is;4+<ezBbD^N>{D?)`;Ci`t_sgyjz!?qMasL;@b)Y1B4K$hqfBBE+^e
zkqxaP1fPe7uT^J>Izj7YPo`u29m~=p0Dz1Ov}Z=w^pt=PIQ`_6Uzz(JQYN>VW*K^Z
z*q=v2`O>7zdOE3~DU;wKv8Vz^!tXU}8q*n`7Sx+YJs6*~Au(!S8k2j-98;)peg835
zt;}669F5M}W-ib@fzWvHcc*q*(<II;4TX5Frk@6%W}iMi1p+U8;@<7ArViGcu<L8C
zFsk|Gf4pZFbHs{P*>9d-3pDAUa^#k1qfcM31T!Gma#(oUJhJVlI$9i?KWKB-7JJ`L
zb+3u_bI2jKUe6Ix`HamJW5u6SkciW|#nbGcp*-mNLi>05X}8gVsOhPZzun8!yj#*U
zlE*lFZS6z2b(lW+8<1O_&G*&1FgSMNWg6haycdG43NO2(@k{_U1q4Xc*D~Cr<Z(9z
z%QGSP8)$f${vcybS$ze?@QplZn7+)baM0jnS#4QQ>pNZOw{hcWsU7Fq$JP{xPp@>(
z!fYsxGjP}-`6+iuoBc9${xqA~P75(9$19M2(#oSJ<NgA&E>ubl`C3{aSKc|Q^KC%V
zXkv(Mp*lk^V^8||=QJ}a%g>YiLwE6~dR+S<I#Jy0#YfOp-LGvx>V2FkR92|g%VaZx
z&AHKuBUQseKyK@<c_{3TM{RSH<S>j6-SE`W-BJ@$w+sb>NPT9e*t?+WE#VN+3(agQ
zARoDvU>YJ^Ynx(2kdiIkc}{4(c&dU1w^kUYze=NU4J6hrS85PAqsm#UjZLpKyb9j{
zUOe&-ZyzZNezi+Tc;Ui#hJU-VeP?cBZNh){7sTgcZMCkgXv-kSSkJc2B=2r}5|qNB
znsVQ1oqD`fb(VTxWSR-3ELl^MNXNNPTBdOKuB<d;RtxFf%dafi`4(CfuHTZ5Jfb>8
zth?1gB6Q;}t4e!r^~|BY!A@;3mld0?UJ8oOJKq+DdNd_vZ>C*+cwKk8+Y2=r5csX>
z>3cerC!I+*R`kb#XD~ut6y@%6TwA8B^4B%^_NrPv`w2^Ikpx2RbHntfP+IQzkiwhS
zJV-ns8rhmTYcaF~Ei%``KUjm0FbuDqAR^zSw#zdoYe-->+4p@AEngR(+kd|Kh<0dC
zM)^(P^>=>SodnVs!JAPv9Wo?}R~UE;ETs%gu*TnS8(Utt4$e1|64h6}5w;qj?C0xK
zI1IHi809@=xS$qVX;U$=GEC$}_}-I+r~4fjgrV-Ej(bPR<&J>RTJa&%GNU=g;}Pnu
zqz$K<DvKG?dkl}Lw-|=@+mxr87+yTG;#);due=-Pt-WZ{ynn?5)wJl=+{W;t#auw=
zuQeCyNUqy5J^3Ot3>T2`91SGuTqc_GX;~&!)N|r`dlz2&gi<~Kl=$sUkY|JATZtRD
zpch?2fA=ew&v*K>^AbNb?0r>Mi&i6H2l>{z98HC0M)bAUadM))B&xh|^V8&HKe<J&
zcfLZ@pUxu!S<=DG-uuUmW;V4vbKWPX8HQd*gQTxaewQFvP7db#O8BKtPbRLC9fmjf
z3e>7V?({j@hmpCE1c(CPadnXl@$$1sFy_3@u3Y%T?n}-6;@%O?_=Ucrx7Nn)Te0t{
ze_t|SMi$+Y3Tx3)y}D(2OSeI?M>~l)^SNeOZZS>U(`9PDHj~MVZ3=v2DOHC~aIX2C
zhrHJ2vOGd)IzPt1yR1Yjl2UhCq@9vpb{4n~(&+7ceO$q130ZyjG>t6RDv?6}E2za}
z_T`OPYm6z1+R|asxLz^b>h7y6rhKt}+BUUea3Z+!N4u_$Ur;SMK6Y}>rV2(rDp|3+
zVHq}meS6n8d4p1MDu>oBQgGef>gL3gugkyuYXe02*_<(1lrtiZt4dK>VV)}Cx8*?N
zOC8ylE4XD5tYjkE^qacTS~Fgb*2r&PzxVS<@AYU^KJ$57!^~+F6l`RWur>sGZZKR)
z=iig^RHUYjG;e1tj(jR*|D)BI_|}2_P?txBK|qo%f{E|a(W3K`zJ&6@^;>=fN9CgR
z);&1PVr_0J0>+N0E)?}qAs-KTUb{Ca`7Mm;=XJJ`Z!5@~UBqsM<{SZ8BJ-IPBo21e
ze78o!B7d(E{7i4YmU;V!>0LhoGp1+J_SZFuOV;P}gnyrWEiPN!dwJbF_rT>7Z$S)y
z(iH|l;%<MZ`{vIDwbVz){Rgh+m^4s*9}2S!?I(&#>F;7ga}1I;9aZ>^yGc@#2K&%m
zFGUIx4iqN42EFkk_zZtCQD^?7`_TGI#Y*VhYX2jl3AHD>>`zSWKC;jL=o5b){e#3*
zwck-(Gn}W$Y?-#yayW^3$8LNl;O(T3QSv2lwl0;aQU8udOwvw=ajNiWN^IHcOJ81!
z>`#_7o%b>XbVVr?O^-&BNH+pYKn;B#H3p2WGY<GIjhGjDKhJzRv~2INN<?iJ-I;wr
z6yEp#W9vo2I>(<rk{4sPem<!B!FlQa;Ce$>{&6kQqW5ll4nfJoCRdeycCoStzDagp
zeG_5^xlJ9aanIG1+S4Mr79O6aC8P_|kX1rIFdw}oxk0<aXzi+(6(mFRjBOQtnUOJ!
zr<Yh-+dSsR0N4ns-`<yNYa*c}-54y#)O+JPQHrvh&nM5A2@H{GSuOWsOr&53R_SZL
zQpwVLMvCE)p_=>+^oxfZ993*2Hd5vX63%V62mOzJq<R#Wx=zhCD!q`e|5^BQfWMCD
zkxFJ06lPfxUUnO}ryQf6)FgCeh2MPg)uELWY-5)oNXmGClHEsEB92Mo6JM*<!wV@T
ztxAceF;5RO!&OUe+_GOv*5?ZseCK4Rm*{ZN*?Qt>w#3#svLIn!SzWA^9@8(^)<u2V
zKI<!f+nr`p0rKNpQN*ObqJ`NJ%jK>1yY&4V`1e-*@7Gq@#|!-8qP=i|QU0IbMFBdO
zM7^V_T4Uh0^(=zT*Sc;uUmy~B4;;ZSGDaJkc^)TTFmiN~ZE5jkqjL(331-_CS+*3o
z#+q+tr|;z8lanvs_ddDgPB4e4DtWP`@$mCemtpp->$lyC@AF~U##+8ud15uTe_|c$
zHBsGIP_KG&`p@bdTW9Z81*F@QbH0o$l#7WT!Cvk3S9L9fsc#r@Q;7F?m-<eie3{5Q
zPa)Np;)F8+Uu9EfM-79H{llFNp$^F$lPR}Z-*beH-c7$D-2U{s>m5UW<6JLPQdZ5K
zHVQ{%b+kF|yPmVs{ig)qYj<C6t6{jZ<J~=)c=rvY9;tg;A}e3bzVb2-S@wtw|Fl1E
zs^axID`Pd{sd`FB#88hUhtO^QgroiCeP=ia;-#ZVJzCDP9?gD2AEJ%hj8rfXIlzGH
zUh}2N7{>dx4<`8OBN{nbl(UccC)0Y;3d$^=hkWi#45k)xz)?&E%--oSmsg?6yWL3>
zbXYpNRXohEfS`nTDWai_B^+)0P%cRhH#x&JR<V&eV`K`<>%>rVq(A-S=@DaI218bo
zm_`0!{lrwD!>kz8?^=wCCC1(sSNlZmj`B>gIH8?*(EgqL&vq4~TbjzIqDp-9RIU$d
z3|i=3a=g48W1K^d%>FWO<n7x<e$(!bo#7#rLNhpJ?GswR^YX^1;Py1?rnn8&isYS>
zpBnXA==>&%qM1Uc$e((_?rG0AF;66vC!xp1LEc>V_3w8kG4UZH_g}W;4u_1gmLqwM
z&8{B>6J=6zM~|2o4e|Et-sxDqzEk{I6nOgFSVR0?b*`73fyTY7MQ)+-L#!cuIr8vM
zqJ`TUq+})ZGqQ0*R8shrI`_59<l|%9Hdhdhw>6UPa!Xt#HX7pSw+USHaM_uEs+U28
z-hM9fn3sBdQ%r~LA+pTP|Bk=yv5iB7fq~uI{F)YB{sHhr=LD+8y)ubUdyljNX6#}7
zc8$Ysh12=h(dn0QxpwI3b9lRooti+d*JhgJ&MOzYW2n2-@(pO<R|Y9LS6r&m=0QYY
zu2aNS1wVJfh?MYmr%iK?PwIuOnlndnK@!MVxfD)u8^@&Xfz$ioCs0anZ1U#5q7Zc-
zI#Qvx;OWT6@(@xpD3;#jRDvQdc$9B$q@|~xVnujCWK><;e7aiIcae)ui<2*A>E*=}
z(u?*#`dT+nQrL}Slrp-Qt7EQXty|XX6%*EDNGJ~meie^63R50ngU7u8I&bQJGqzU!
z<vNx=a4f=EV2qoNWBn_G)`5k7zn|x`Vw~{>clKsK$#+DZQh7;<g@*iNI0(?hMuSx(
z@)(YFczz>_%jcOL$Mamyd#)mU6Nwk<)mEWCnaN!Ot62g&r}n+J78?Sy&bx)k{<qfm
zBe2)zz7}wvE`4*B+-CH9u{}bG>Qxn1UnBUA%qgBdJ^nJ@cVzc04eTtFscSft)Yk*g
zT+47wxY#rufUI^q-LBBCN4cbS5>xZhrN^lsMY$P}cQGX)-R{5|YEsRQOmK!OuPn)$
z1gYlOCm08LXT3Y;_$jHjX36#k8Na=GZJ>Q0@!coUSL4Q!P0rznyk<YqW8aL(8<B8K
z<cRiyxXrUVGVk4s2H>364{X|QPf2aCqD6e|=7u%r<Tv<7Dz^)oH!5dtJhC|u9fNoK
z-Z}krh$)_&an3olBcO9J_5YZp9BKF$akkvv?;N>2lUcHYbWD2bD=T(-C6C(qd3lKm
z)ixJd?V(1eZ=M!`kKyw_MBrWzIQk??Yp|>QtU|O=HlDcnz04nwofWC@_Dup$mMALS
zrQmBH*(T!0WOF;ExMoG~gl~VOV<$*M`h84rimZPe77l&f*70I>VpN3E(Ky;R^EH!m
zT~K<b>5yGUi)9rLR{JY%Hs$f4wzR^a_E*Lh%j!2mf=#(!2$pkSkQuA9YY)lC4_!Rp
zy%<_@$t-s=+AKL;@m*KNTXLN&fZs$2$rSZoha0)WlqJ<NQ7E(TJC*r`SZKyhp7zA|
z)3KAaNSm?0z&4Mor5+)N1e^Ef{tsPm9oA;^MFAEmP~0h{Sg=y8NQ-+Ylp-ze)&Rvl
zxEC*O#c6@2xCaSZpoHR3G`JK?aDv08-*0!HeRiMyGdJ^2CNp#IIp@whGl^BQCCsq1
zp6QB8-zq~CZ2dT1_|##8;?zOw!-tq?QSt68iCqh|hB%+v=$V+jxR}mHIq8yW9(uH?
z>m5=%5-E^^+`_MkBEIGq3q~sv&wWV2_78ykwI%ys39?lB>jH{;+2?yZNY<VXqLG4I
zCZP}aF9fe`LDG5Bpjq$k%lc&%8zd}d&L!i_6P+|Z<K0aKVyZj3etc^C$uFz=jrhwg
zs=dA%ZUyOFZL+QJU)cV>LPj}P{Nys1l(~m+-FtTg5kpI*(D5A8onH&e9R{H4CFaT3
z<gW*uc9&EfQtkN!_G{_p1&^_l{moS)?a54JgG6<<re>&n8c=1!<Tcf1e<@DA;_ba@
zeo>t_Cd&AUl|Tb8pg9LhxGudlx9|MU{Lc%F%{^A;U!EwP1!a|esCCmHcTyPPYshYG
zcyjrPn4#0GSGi+$$2rST6x=N~qoo|7{I>%=?{Lz)I65bINad-#i~LdMeA|@aNwao$
zJ;B50Y0_JCKc^s2eer&D<Xm!FyEbrT+}3}ZF$lVK(G~O2x$M%7$O%?ewFm6!&E`y2
z_TTqeQSfHgkWzH?Zz!rFI-3e24L}WBA@aOmdB$IzQ|Y1r%k4E#hfsLYkXQ3ksJV&3
z$x23K)fvJN{p!8w$Kg=PEJZ0ufOGfS<&}XL`}3RmRT($VS$O+_YulTrkIP4RqXzgk
zvx5Oy2>9xM=-}>-fS+wH=^C(cW(e4&#y!_xTe77Zy|=G=*_h10g?sz{^Qp_a@?&$D
zQo8P!_*|=(?~6VD!GKVGat@8H-pSgywY5FlOooM?xe72dy)=Fyo4vkOoZeLL`zv>b
zvM*E1dP@J9Z>M&)ELbb%qi3G}_~5kB_BHK5yvw+@{|T&Nul@Iuq+4bQ^G$%3<0pgQ
zTlosI%u(d7e5+D%`~CP+#QQ03>*<N}UMt7KAKr_(RZKO2Col_k6qqy4$u`ExprA}9
zUMf)~N-z4BXvuJIzyFJZ0dvg~&OyylVbkc2L8p!f^WP`VL@vzJUgMP-&XO^TR{i=)
z_s#F022^D4(M)m*#v;ECX4l*+ex7srY+I)<q&BO`HfT1Ju0D&APq#QRYUsY2H=hCi
z+Bv<WoLRT|AkOO>N6PaKyE%xLHx)WlU+}s>&$><^Qr(JrE@z(IOnVrasAKz|fnyf8
zvb*pjK<QoSz!TzrYvb_BZxgfZ9b2w9La!n#<}eYt`dnEy@s(=za)r9pr6c9rmyiwP
zPA!k5QuQ@Y`9txiT%3t%>v`d-mZ=1KmkA6GsRlARKuPBzM@r_;Cnu|Rf?`Wm15ZVh
zLT&YCn|F#Q=GbW`NFx&7U<&%>^xnzXypSn$12G8aY}|Fq1!Hh4bxE)Yv{Lu;mzw0+
zBB!9egy8Qc?V1FbmdKj2m?bTvsA@&VU|qpfDrsj=j06u=4zT74ui?vogBha^Fo@R_
zDt|MuT_+*e;zZfKn}S=Pofpkto|lH;q^O|n%hGj_YOOe2m^6GSjo+;*jj!MT)3CIy
zD(>>2I{tT)P2;N=GgF;So)(+orl6A3GY*hV>P3F^b)WE!&HGjNIn0)X)W^DLZbkF~
zrnBqH)X+(mKZBkaq9y$Qwd7u`8+l|}SP*FLL|OFv?nPE@bvn}JEvPOEgRVaN#%MdV
zKm7BpubWM<i&UyK=u1t3d`A075L;HQOM>8azf~5J6@#572Vq=rsjBLTVn#r%Gz~g2
z^_o?4l!snn6g~U0tnJ<a?o63#t|)x{Cnxv!Zf6yp+c!O~1uo<4Mzc-<nN54G+E4b1
zn+)$3Gy^@Je`hpqUTz7f7IwH@45z)?JD=0fPcuxamh!Eb(KnN7@h%%fK!h|;U!EqS
zv|C0`e+ymxnbZKA`-<6Ee#V>>P^ULt-*H?~-KM$(i(#}d-#3k5(^$U$t6jSv`0A)$
zL3&0F^<akbURrjKN1%zZ&}vG3T0h?^!kxSvg+Yg;vvi0-#2i+K($Frsp}B9Y?ia?w
zvvk|QP1Y}ULX9{L9ml=%HC0|56IuwYC2vV1-jhXziXfy~@wC3#<uAaM%rU{$%bnJ7
zJ+AYn<7kZ0ZBDUC$g(pY(<G~U(Z;t3O^mS?Gq+1H)%V%D1a9gkw3nJ@pQY11O_O?>
z>7*XA%>C9fbDEzmNK}?FaPyYA{os2`hwhcODGFy>c~`7tG%9egpek|N8jrvd{ZfE$
zbhLwW#_#9VZ{dP)HPl2QV&$S#W3}VBigIH9qu?+d_Y(pNlEiDJ*t(JD%Wjj)$cbJt
zAM*1m<OK1rc;n?i{FSrozpj1{x-1|)+z}<WNfp&n$)^l35y4+f*!tF)HZM3AufBur
zT7DHT^YbkW+68W6+AfyLIZc(;7T{%7ax#M9yBkw(N|}uv#~|PTfwYmiFL3MH;2%Wi
zg3huC>VKMium4ASDXjDk{-N;h`~RUdTW<x_;U6l$xBWK^7)~5*S)s!;Ljw+W(h0hE
zfiZk#eY_KJkKHaeXB4p$@OxTyO0==@cge}z`PWSb+wB=R+xATHuV0kEog5-+U*x-A
z2P+owVko5iPGM5dfhfc$$V2I~MDPVzcr`d-LzW{%7WCz?fYU~AaCz|lan_Nn%Fumt
z1>^)X@9!r2WHHuYx~!w+*F*Q_*3x<Z6_(F5xZBtaX3IJ<mmgwYuD`fce9eF7v-Rp_
z|4sFt-UVOSOPmh9jp3V=KGTe0EmF@{G*`QsYX3LD%BVPm{{vXa{T0gx<1DipC1OP4
ztLX#NyleFbg~xa5SM;8V;V0`F-uJ~39WTqv@nztL=ZJK&x^~$D>Ym0fQ3FO)(b=17
zl6x1kJZc=x8>h?L>V~loQTeED!4JC*i!NAG%ekF<zeSxnTBoHEx0UdNKrF9!O*1WX
z(g)AplWp~Rt$q>bY4vly4D>*fCgbO8BnP>x>ys7Pe^t_uJiAlms&vYyHfBfX1;5uP
zn`fdxtYlVLTkDe@&t?qc)9=?t_^1m<j07-)1ZL`%`6@(Kya&xmQyCieKdBWHJnz$U
z%09#87?qXb((-joygF%L-b|ZDs|*Vhjo85SgIg%iWpaG(^gSv|QR7^VH{P&)IgbAU
zA}ZM`U5aNxb2w|{A0$dYd6i@s&-54reo$pP5-Yt){D0tw8V~Cpvfo+R#yFyMYgK}w
z*Ps8H`Of}_iST(Q-QQ@H+Qj!eGydsq-fR4C5eflG95(9b7c4nProVT>)y{S>7A&V(
zpE1OiXN7rL3z-K=Q$!QlUh6i-WzV-{%d9(}@!6*T!Z)OzQ_Os6d4FX%6tvkjalsOY
zfi2@|GCZMcWBmVppEK{*hbMf^7k<4PK7<h?zEtR~KbXL{r@9`_8;ewTYX3eZe7&{z
zc0Ig;(lKn>;+;q`;0Lws&z7*Tl&JN(%5J^)c8=|471qooaZyRhWJ2w-SH2C&jb~*I
zmHO0MESTTp4xh!J+gqE<Nx5xEgPK{SA2ioO&C+YnVt1S8wl<(=KgpO6<<Z_=x1;!m
ztf+U}UmBohn~i6?i68LR*Pm@d&m{DEWZ+fI@5Grw=_~qg@*Z3Fir1ONwn^-lH~-+3
za85Je%e0m)uO2)<d2efNwf9g@lOt>DJaybUqDAdLm&~CYD&wLs)l1ZJHWsL9Wf`g9
zk5$g3`8M+Wqcc9w%ch<ty~jN1{az7^<_w0$)v5@;zJ|8Wz$@ch)t$n(@S*9wd@Q8C
zgPl{eIiJXHEnj|`G*UwG7RUB)zn6QbiEc!>vhEL3vToum_h@wQ`r)60--cbvrouPK
z=s9od=lqn?@A-w+g8{#8Fs$PEpO?=4d-ndd(E5E_&ka|<(0|^Dsj$m7Y)CPP=*>0%
z^Py#`-Et|=-WjVuu>P4S-N5oBcc+5glHYB|h0=3)wRT8iaX>Yf!!5I;U`NopLxvF1
zF=jv8*_D?n`?)5~)wa%die0D1;mN1FDW1V?$&2f>xf_Yq$9ppaS6+2Nd2O4j3)hZo
zwqbseS@U|8G~@qS)eNQOr>j%Bd9)E_P0@yofW4VFwS@JRm<KA%xp1{HQ(AYrnrfWK
zt1^?dGR<jft+keptuZL?x9I;vBuh+JYsdZ`{h#6m-Ad5pw3k+n24O0R9bWn~-GAlv
zT*iNIrqmrX#{w(NAuENLH{~52ux~UB35^TD^hdDEo$VRB!jr0;zgK#H!&~R6Fv!tt
zvaI|+0jlt2SfS@;ZI<`P*{8ETkPLp-?9~lbZB@+rQB%T^5c$~+R7T9ER4S9uo|JNU
ztM<%}@_wBN+2X16W0CAH_r{SlbUyLN$&k0>nW#UDpq>mCv$j8r?(!d|AG03@uY!KN
zIHEi%4!9--=gg6zy^pV{P1I4LI)U2$g(e>+{xCgrR5M!EoIa--DUSIUGk`7~Lr1&a
z80y5-mVd!en$_W;?}0T!!=X25JvT~t>)ranaJA(>_|Q!U=`vXA;luySk+_i8OZ*r}
zd{UmbVf_z04#?~0OE8~#hBhV23tJffzY4EY7}cKQC5w|cIz<%uo1_K#e>O{^AkfyR
z!!0r>UMeg|-QJOxLU0pw&(vlL6v1`WppEnFK+as9uOwM6_iT;^MYH-7Q`hv1Dg<FD
z)j!KW_H<7#=wDHP^H0D<4gS9c+_~%jN5GxCwlYO5Ga3-B>;adCMzw!#Mz@e$e=R7r
zxRrA2{C&9I3l(m+h<glWR?P@H*j-p={%KvZ{`{7qQ@-@wBU#PFiyNuiK+0QAC+(^m
zsnr8-9$8H>8_z)0=8vMgD(|EMu0QgNjzRen^N^@3UwgkmEo28m((bopCg{TV=c~;v
z%x=-azgv$!8b{9vXm1alpPZt;A`#odTS<Ee_<y1O7w}##woohgO^%hc{6LS})kUt#
zU#N;0TCMFImb(B|32gCxIVzGj_^SHOo#P+Hx~cwa6mV~66?Z!L^`<FVZ{_QXL(`8s
z1z%HIy{M?jYi#jJjEmSnRXUw4^nx9u`FR53?5*NN3X=^<>^}4JXd4%QUmddldH65P
zD`g9Y8J4{Mr-=Bp_TZ-;HRO8s!=Vz>v;z#PmoNz*9<-QC>Hih%@fB{Q0x!PIjA~DQ
z<bo7%O+ko`PCogQFp&9O=O$p%Q0JCatI$KVr77Lw?#3O{t+C?Vd2e&%#D#Axz<;}r
z6Iozyfv%5C|23+O3I3w5EbrGXhdL^2MS}9b!I%D^_w?h3f`$+OS6&-Yy-a3{W_X?U
zU-6vzAQGkl<N>qc6Iso43#taZ@NY5H@+D$aBQ(_i5jzrStm=pBv$1z~k^^+Pim!w6
zPw%$3zGB1*O8?TaQ}zEAE2x=MOXQYaJ^2^+Hjj5Dj-9R~mkE3h*SazY0ZHhPV`tTq
zUmb-K%BQmC?;;u$zkZmQ4@zG`mv<)8e@-k)L!kGjSdm2b=(+Thhq{k~Hm7sc9Xkv0
ztEp9&)&Wix$7Zl>BG;;CwMP7dL3*MDufxlEd8iNqHBSmadZ{Vn)?Oj*h@j9Zjz=~x
zB{5N7lDi2-XTTRLTt4mNX=uqmWB70wWl{1MJyq+f_k7uW+VZ&pqWZ3$bg(I?^cVfr
z%YPKOw6^xsg=hi`Behos#XbLZ(vQO`axwYO@z%O#j5-4$3nj-U&z~_;yCeOZ)zF+!
zbBlYou(|3tHgyN(D1DtdW=>&rW6G@)MloVE3)3jmpv-U(SyrvAzzFgn-u6F<GRDDs
zEpi<X`qd%+3&|X3T^IG)1nJQpHF@@fHU({_^JN+Gg_ErA>?)~U(J^3*J7pQ79TYWD
zads7Wl%tch40*t;dTEw8&^T3`@@T5lA}-45Ti$L?qlD$m)2=y^tCvLrUd24`i?)+y
z)}r|6Bjb+gfR_sA0<5LvyrR4p*L}~{+(LsW-j>kg&H?`auEh8eR_Iol|KI`hGR^~v
z|LaN&4{J|PJ7-&ue+$yRd59`gWpMrh&ITjyl6U!tu6B*%ki<jM2~vy34=mn}?&V5f
zHaK6EIuB|VbV7UgHAG|-p;W-uvDI}WhY^X-PqwbgJ1@7qT{<s%dM5it6AwCrR)bMy
z0Y`^-Q|o)82f_Dealsd;J6->yzJ5)C<<6i>^wDk4j6jgj#l@P--IeS;r1NTBpP5-E
zaDw^zBBEpMw!eC*{pzaEyFe!JVs}tLw&QAlb#XAUGw9~<(CiS=5Aha|-H|N_y5EFl
zYTlpTMwz$!1)Q&LLPR_LPtd~<<__YKbUN9<;q@7Tj$pt0DNXaEa&H$iY3ZPV8}#HQ
zTt9I+==$$ekLLZwO$@~R7$zG8lIgg*TkN^k9PPZ=oa`}{?peQ1TyDQSJUVG=zdjGj
zS@ACD+#tK3e|xjKh!_ptPh7eUo4H+IpWKrTT+5VNy6lRA^y`x`%i4!6wcp-^xGddY
zMoo@(Y-@JjI^3c)t1oVJubRC_@6nl>`S(ud<U#0%o9jKDD4t_X^EUgLm!xk_wgj#&
z(3!*oiQAp+my45w5N5gs&Gn=GDgEH<>+?iSf#7q!#3Hv#g=%!)$z=P5L*;R>Z$Q9#
zA9Lr`;o+&Ob|oL>LYx4Z&*ko9j;z$p`eaW1qmIkd9t8evZ6>qK&ElZ<)wkt<lf=<_
zG^U$7q>OhE5OY+gET*wh=he&wci^DgZ^b9e)yo~X%)#?P<FNUW`@nl}!1>{B!D;76
zaJ$6)-3`@N@Qq~9?domy-P`;7%f0)<;rl-G;M+sg;K6dx6>4xf@MPzx=W;322xqF1
z`N&%qbA!O@?a9bpL5g94FX5%~-7aSA{6eE3E;4NaC0opi2eQ)A9k(}IX0msS>k!t<
z(_+-gBv*dPRKYzehnZv9cY2HNZn=2Qg~?nt7ytflXIQ{|X|%{AXc5wRGjM+r#l%rs
zu+$cGfikQd4X-;pTyDF}jJm$vyT7{a*K|3JA9cB7_6Z9{VTn{-ZQfVio*|fJq>xdW
z1syXlZe_1(x3XqkmIE$^b0A!mIp@`lIQK=e=uLAC$n?@)V!{390W-5~;9b$k6zbq<
zP(YjFz1pe%ERjk)4g3cMoL5%;Y80;8$Jz0#K<|1Kt{Y^uCCgUinS(34I;TMT0`IHy
z6?9)9r+491lK;=<KC{(FRtHq@k4@?+FpC8(&qGOW5}(P11Nl$+l^>5D#1M5<ZamO5
zwYjz<y8X;c)~T%w)lc`9nHSz#^h7^7TcWx6toedyY37mUO7+LY#`C?pt^R8g;jQqf
zv%ldQ3^~23MO4)mqos+AHr@S=$i;1K=fgkT%TUkp5Kf<6PhX#15w_$!<QeCLg@(uW
zi*MrmkHeofwfhC`Gc7gv;E(G4{FH70s9Z{UD&^H8r*}Ew^0-MzsABv%2hF33JR_<=
z<LVc9Ub<J%hbL6IIKGS~yl7Qoe(USNFbCIHc;0;e3t@;0Nq>aLlZF=5EtOOrN4NeL
zlq(*o3K6E@-4paKNw(v~E0SDk57yq-1p`xNF{~fEvFr-PKfREY@;aq?&Gj9RhuxGk
zpJkYkhrM#N3_1Gdlgl9KNJP?`l(-Pg|5!Wt<G3ejONq?mm{<)w9(e)b5PLkX_0cpF
zt4=hHe)UmVc)j2s>@DyQzMoGjJ@KOa1g?Q(UIDM-f?spydb|%tEYq4|w?A*KMUKZf
zdtE>JcD~58w>?=iWBdmTRK4euK+h65NreQx&3qj2Nsxe|_zM9t<+19w(LeSFloSRt
zf+nHxMN8shrmh=cW(-J-fh_bq=8|*b`DZV8nqNjwWZLY0yGrzXiW=ww4jt{te4t3S
z&+-W(5MN>g7vG=sPvZm*vr@LbRXZ#_e=}OTWUgA`XTIDKMvQA8JSuN86G1fg%zg3v
zP8HMCuF&9VamFbS9!p?g0jTjy9UWxK+aS7ln&L8^p$iyh8sCr%vhnnr%tD$9xFh^o
zBVJf|Dtr>=iO|^*et5&mtz&S+jb!NHOX0|8-$0obSUtGRurjGBTMHbw1&2U1^y|+P
zfi&>v6l34CvroVKcHO=4bv6c+w8=B5m4s|&|0Z!4Nhj43Pp4TCemNFB?TR?&o5H1a
zFY}ysVAQx)uTRy&0-2(zKn2~2Y1Xx~qZx|0qnf1LvmcTaJ661ZKDZAh$mI~{9ed~Q
zNs`a^m+*j1J5@&edpp?O<Y6=}H^ok`+`bZsV*F3H1pF>cbgISa;C1rD4}~5KOI+1h
z^`=h^rdpEpMKiH8fczUxLqcUv;@LcN@j~?$1sTGXa$)th^6Ych4b24Ln2G|qf&mNN
zI7ee6+8MbkI`8o6`erP!-Vq2z%Pb$Bc7?~)3MX;d<;fiZ)KE5hI{P!Gt;lm`ke9cY
z0p*U<UBe;8KqD8Wi3k-e9!M|6itqrJ|Gf%p)!`qzvF+kV)0P7qJ+|&R<{>gzt#1d4
zc`~)SNolc2>eBGGz}$4RHYKMIg!Q*RO&Qbqj425P+Jv)L_rtK*!)8pkkkEqGRy>~C
zU0~Ce4z6cySs<b2$^qzY>=VHD%uDvMeET2t`4XQ=i^89{#nUuEo**!rQR8ut6;B}H
z=P^WWTDoP=KO2|TDA>1ly*1S)JH)jMy>GQSUYtg<+UPVkS!#HApkr9E(coc_?u-=G
zXN@KD{vQ!-jsOyyuf6I8c7=<2eC9p{Pgnq^4xv78Dro@r_+714q8H`B6;ByLndD7N
z)E{Z2t-}PFqxihZ0h-h~C`7j7@?l2^Eq}CnOvrL)-x|u0+z_4fig6oVW`Bey-O=~x
z=qf`MI-~Jczt;9^2`STj2IDM*?=1_t;fv|{_!k~6q5I21edESvE4TLZ5EVIp`@V3O
zqI81%7EUExn<{oA-IB1si@%kMv?Y4&jOoeWJ6Qj0S+vKJH7{pe%dXV$f$n{>2?2^_
z@rye97nnKs+_gygmI|25ozu|+JtweQSm?Hn<MJiMq7zJ671Fi{AMqCPP}dk#y^|bL
zD9rPUAc15NY^a)!hqm)!Lx*Twk}~v*RqoQUXG8SuFS3X<x_B0ovRq3Zq0#A#vmO)R
zm0ohJH*TBQ@0;JIkg1f4cvy#Wdg_w-9q6t<+%t*=3_Z2S(r@oJ_JqwMZDEXkV3(%v
z(vT)^@@|??l{*y9#j>-x+*>NULz#RlC&0>Cv*KekVT-ju*Y+pEOMF00@P5r9f_9bz
zP!moT>J}%6;MvINv#L&#T(Y29;11((gJ47d<~`}nvmOqW@!KM}-Ws8=$d#l!tX$5r
z68G+4xkrsK)&URJSTxW0i+YBp9$cSRQSfHKUS+q;e`Nd<H|@n?d{RYZb+b4wK4XgK
zJ7yX&x1WG4*q6A-Z*_4MIvcZaIUi#!uT%cLzANa6IY_BnvHT2+%WtOi5!(~VkG;|w
z@bujnsRrrIK{FTWMy!+6dFVM<v$!)R1i#((_e-RBi1KUPZ+cN(+FMfF2tt+OeB;o|
z0jiOee9?<r|3^tBlW$ALG48MpSX?@5mq-ejXG;p5al<F0Tk(zz`W3`XF-z2wZa0b^
zi?M&3xylG{T76UfB&e%}1)E-!t(PHRg1zPx_`RJ{8D3~kwsUm&C$z1fP2?~#@x`ji
z<8e$l{nEuxe}uh!mE;+k6-T6zl`J)D1F+cqOwDp7s};V`#!9kmHp518*cq=z|DMhE
zNuCMoJ})-cFV<6B{n~Rh{N&px@7Sx?ORiSYAp(BXWXWIOV*?3FaO4`S(y%3#pNW24
zD)Kz+PZW~#cx<gWGWsKnU5O(uSS2DtTK|B+#mFW4&F6X6Juj7BL_t=Sg4ukUp7X)4
z9f@AT?}(%+*}#zPF9a&JI}Af%Dp2dSQ@V@%@61VpFK)5#69{zP4XtbkvZt?lW6yC2
zzbkfl{o6h(d87NYD99!*<kw3M%N=jdZE9h0gCb#$XsgjQ>54Mo){Yv@w+^b3V>LRT
zC$`y))5aQwZZUE|KbvF6Q7P7_H@Izogar=$hSua`hbOu+vLebdfOwz1cJlHq!h&t}
z@N1!lk{cFkZ`fzMy^@JbF@0tz3R>Xx?vAed5QY7fSdXH{Boo9;&fGN!4TvDPK4t_T
z9#iqw&`4ca>Juczw<@%*cq<gS5fL48>&+&?1h@(JoRofxn7fTfbC%&4b1xlA#sExt
zOjphfCXg0ET2Twr>71a7%7#PlSGhzy%Gr}`h4x??<?<%68bQ67m2{&Lg$g6w8`g5i
zL%)h$xyXuama=C12>4=|=!Qkb{HoNVbI(p&d-xeN5bs(U*FLtX)7jX}WqI{xE+|}W
zA+$~y{IP}<+{`-P)Du>rpKO)bNHDuCnUcq$&`E^@<N@<C<x5PfYQE^kTd3QLez)Na
zS+fk92vI2}_=tI(&PdWO^mC@jgG!sNH*DE%C>?z9$1ps&@iHcA$u`RC1kL!*Pq@!m
z_tQ13=T@FvYV&+l6%2OY8m6GJL;Lzg=S{7hHSrB-%D17q$+^)Zly*>!Rn3Bj9{6!U
z<8q1?6mPg9Qi?y>a_gMQYL9<O$;CKo6!(`AIHqMyi6sT#KDQ*0?7Vdz8^ZtBR23|u
zS3Gwa16WL`jV3nkxLnuxv^*{o0U2@YG1wYJ*ZHEf5ZMyFBV8w4G^rhxvyjgOy~Dul
zykOU8K5k%SUKUH;$FmHl>u+SWXW@$c+Gs!hR6c*D*2*|Wt1bA&Br6U`B@fK`eWrL`
zrfQO<2Ynr90`+=U7bfii0IQ7_bRsFvnCe88_>HM3TTnhl=U-<#<S7nen8p~>WuXn%
z0zGT$19+jC>c#VV+pH|!K;&!`+{vFyDbkdNtqxW7V2KN#D^;_*YLqet92BhrJ<FQ}
z_a$)p7PRMm1yFvBfM11(4%2ARXEi{BmDy)?g8`lsSA<?pfU)&o`1rbnfH|~zoQUKv
zfIGrXF4=BolF6zc9C(;i%D(0#ZN@&IteXtFMgsA+@mWCz?W}UvNh%TIBv^oDKu-eW
z6?y6+-)nZnQGud`z$Y=(<2d+Fu32bgu+&QtJns2355T}k9`yj2+8FmaTnVodXh?Y~
zIXL2mW>t*%qSA1}4UBbih!rt%@m;ZnqCCdlD;U{6>w}&G%S<CS71bNSn6r31JO>iY
zVs*qAtS7=UXgO3+XSBDt{;XHl{<(p~<dak_7M@nK;LVyu3ZZGzuj4UGP%Nb|(_8Ko
zN0=F|yRnk9L6M-dB+h96A4^oP2x^y<`K9y=*_JNS47qRT12~|y_wOudRU15vl?vT2
z8rbVUOhwHa9x5+lm8<WPRxEA6^WRp@Qj97(F0Zf-2Tgjq)5Xj#(NJtZe<p--#4C^*
z4vRKr<-{I$Pj=Q<e2QHsx>M6Mqd}MyC6Hpt@qH5Sk;=p6*Mv=ZkJzS!e)>_>7S#gC
zM<s9@veQ_24B5MPc<Y%wGt<f#J!lLIeIdBFip4<xnqjTT=xoi0S3xm8Vhq}$5-*Y&
zDoS-2Q7yZui(I(1Eq*L<jde=HN5EL-u&uHgHU$EIM^gcI`{xp9@3fY)NTt11fn%%<
zkeRPfQVf|C*A7ZosjtVee*939=qRhJsgK0AonW?GT)y}9r<)l1bCZn4kE7t}kve^=
zk-2?JfEEhluD&Pkk-MXc$j?6%<vKQr7iGz?1VR)14y!h>8J{YB%VT_&A_s^G$;f!i
z5A_r;Ej)CZn$*^>1ACB&c$`7YWp625(1hF)d8jkJ1>DZJ;uMyEm1XXwa4GWI`X9hT
z4F%AwX?VaR3eVVYCx2uFRh#9R`9ei>xx>NgzjrunZEfP%VLTx!(37XYSkrS>cZ9Dd
z#!*|Yv~eA)#}y1E?s$ECc6Crs7?Sa>A@4d7ipT|kc^!U15`j;hD(Q+LdF{F(plPvZ
zk38vaejfZcU4SFxy!@k<<=Jk8#RB0eEp2I`etKPZ8UNIHe!1vSq7EJG_^27Lo4=Gl
zq|w;4QCJ7)EJ>zR9Ijh9gT>r|TfxXd6H2(LC>Q(YSvaz?Tk&a8<b!E{5{fp8m(*i@
z>^D1sKO@^&Q1eyJbOVK}Gd^?lqJSjnb%*CH)ed3L{gSbcR;cbm)4!DG`Far`d9x6O
z*ZQ$FgZ91W+@2Zm3EV7$w~*Rtn-B<XD5Pra47gEOPp%|^O!<o)@H$(_=R5~dLzxpL
zBMX?^f4`lB8zC(LARn~*I=|SU1U22;GiE<fjT8O2#|IvxGB#ORr^GUxqSPtB$S5)m
zWluq=2_+MP0|&?67!`+W{`lJ2wCUvi^Q=-1VnyL{o=07JifcW)C<&wmx+-<8xlHxU
z&D!&xW|4BOL_p>?zvVUYYq6;Mwcu_2^s=W6C5{#_asIgz&_cp<_syRPNC_;~JjXJQ
zG+_;6-jV3WW_j4=tItw-La5)gbsEcrtav+#qe)J5fn7CgArRg5#{Z4htiL?2<0%h6
z+>`>#{M0KK3|s-VPCTR?WPR~%=ll4Z$*Q{m&eJR<2QL5Fc>RQ=akovY^FE5wmA;na
zhe5|Y-Y}>ps5^_}c6|Q;P{(?WEK5KuoPlL32u_!%Sr<3K^f&xXuiD#5_qDYjg5*VZ
z(DZs!sIO~sxPChql5&Zp$;5X!6tK}SIBwoSf(Vg!g~ieG4tAUpuF&ni2&#iADn#0Q
z3pZmgPk6iN-0uDnZtKN0jJk$!ADc!65dRwDc?I2Fk!p8<I^dFi6|EdomoJuowQK4!
z8y_{xQwzi8a`MhJC1G{%$=@TjX9_s4nyX*b%pgvBumqr(jRzDydehVhYmSzy)=wtv
zfCRCVC6T;xB4grhXF7HoywC{jAq#C^#PRH)Zs``7{_r3;Ad%i`N?HM5Vp;n^fxRZZ
zXKjA=VtL;=lBKC}PfE$kTbn$z!YvtZsXAXETJDVaHW?VXNMQ&HS4YH!tI1N-%=>YR
z;63z9iCJ)+KChEYKxZRk9VmWlBE1kV6@zkZKFPHOa>c7~1D^x?BVnw^EZDZUt|E9n
ziqfg@hh}4|l0!casVaIX0JGhYoUp+B1lJzBR|@!hrnuk%0^zw7t4}23nG|fj;Yuu`
zR>0U_3A3!hw!K)khbp1d8nK1?@j7y?#5+edeu5A)%p-6*kUH_s#tY{b#T!rRZDxTK
z)=}p)Lf@t$5wX1zZriK8hb@ss8ALMLbifYf#g)FSk-6Ju;PVYtb?@*@Io#Kt-<X5~
zAG{IdC|fC&zT~!XU17@)1>_6cz$daOoV96K8C-vn+TN;B^A$W&7Ey*j%@PlN^{hSg
zmmCk%l8As9mhAv|G+$Bfcz8lSF5fz2U#L7>EzXd~r_;u?Ml<s**c>|<d+B!b$;Z4n
ziNHZTk|gcU0Gx8Ut%7hDs@i28+_rrlpIHP?*KPcr4tUa*(8q1MrG`WWe^2q4;IxTm
zU>RohAcj4LU^<N;*N{g+RwbE~>3ce00~S5|vJar+0@cFW>6{?B>Q>e1WWg4H%p(fx
ztRq;keqOgyY#Jd%sNcepX+hJxt`3qnu}LS9InZ@8)!$z5CIGoq8QF<WBlgz-Lh?+G
z_@a;FOjhg`Bp6WX&5I!vi%a3Cwm@t?zqk6(<*fx{=frI^`>ppOxcWu%1G95Rp^wib
zlbAYtaD#6WNJZ8drF}r);(2zvjFrV@H#t%~z3{k&cgsI?oz69>4U_8IPoURY*QSLB
znJS^}EZ9dCrqAIQ>Wfyu!u|$=zA|rmi$G-d$G^p%KY4YUw=_2J>{=Y~mkS9lbd<VA
zC~I$3y7O}0gn}dyg1Jj?{@z5(aWN!2RrjtuyH{s-_W1iVra!wr5RRcBElR6CW<!@0
zb5J$nEf<f>?R&*e`D8oquLRk+{_#v!1=)ZlsDTE)fMf?tiqqhk$rq;s-emg&KZ@oi
zg(n5|9vVF@-zi>mDIreSuov6I1~c<`I3`f-j3!t+0@xYNDorSD3^f#qb%Y6Py3*{M
zQH46sul{lAF_^LJ$I{G;M6F{8CFb;zx%cMD4Cq|%q$&5`3QobgVvV!Vg=De9<>V>c
z12%<%IWRSUZk?gOsNXIRjm8m2AcKy3?AXY0-P>qSn0722db_bo4ecI^`4{9s)royj
z2Bj4ZlH|cEL+O$ZFi00PpB>&H`XfJx6YVTjBJzUh(G@Rl0jm%@rClbdSyYdf&vTag
zYyEhp(-sfDc}dlO#9;s(6ypht-5`gRYmZS72m$-EoJ5(dyp@QQP3Au^K0O6fXVv&M
z`XJ$D1gDDPz5C%o=r+{D^hR=vzJgB2(x(ay^$Wnl!BCY(m;|H3I&C&8oibF0!>_^K
zm={2fpkejp!+xS4-P0|hGB91O<SdzK1PX5;7N37|#9H(AC$~KIgj|AsJ0ibAk?^x|
zK9uV%sOQpyl=#b}`&_LlhF8O*d9U|>x?=a8W;yjMG?FvI%ic-H%MXK1&sZC(crL1y
zo>V3gin_#xRCDIW`WQ0-m3aiUM#2a@)5LVe!I?f3DZj@BzC=7uS$uPq)ukO}v-LV!
zPP5ZB`*i+V`FVts3p4h$$S-GQTp(w=X4oD7md(IJm+Wt-VaaiASe-n3$mc|Q=XxTp
z^!cTDt$YiY{yMveG$;RDj;gPk87d_dl|3U+vH|(22spM~)(_!aILVeBA=b1!;h4Gx
zs4M5Y52BWvqIuXjO09v1a`Z^CD78?}3>2cH(Gp6wnHGSc<lc1Jdi_bKjZul*&LXo|
zNS;b{3NaT7Sn;B3!5~`*Z64g5=9v4zL+rdt;CDuvM|1jzUqO%9G1P@1TFt-(Bj2x`
zY|SE`5*jdX+0*#b0xcLfT<6|1C1d6DQov-D#FT>NtZ<i8|2&6{qTiPz#a4EH;ywW-
zFb2F7=O0Vgd?@}|BTVBJm*(^cHw8VESq~H%EHWtZNpyH%M1pbv%vrC}=<g82JC^Ux
zUHEWA3+3pa9Ea8nh5aV2EzeM>5>+WS;reLT=L`FNnl)RbU2HV3`Oc)S=pL}RPsoBT
zt)p<X0po*|Q&PZ~OxeRt1+Y81*NA2x1HctV#Xq^(6h5E?V0uU7H41NCU8gf^Rz#?<
z;>&^fOkI==$g7m6a7K#5AR5M<!#`T-mbx6xZtT=fcs^k!7<mJ?-g1odl)MrULfwqN
zQ<ZlQrF}=Zak67V2<Dt)1*^vrTz*sPvembLm}Ov(lW16)nI{0&(2kZ$Ywezw9zdf<
z%1HMEh#|f(X~dM5LClmg?724k9Q}I$NZ_Mi*DFb{DRgtp?72(a0(y}#DumV)XG0|N
zNqe3(3uPCB*@sUaRfJtlB~~*#F*-29y%uq%A%!O_ruOrNJTq!c#1t_7&C-b?K{!qE
zeQ94LWMpt8R~Gk_+>V+&a2yW`<BP$qIkmTQeKjqF3NT0}*#cLRwlN8ruto-G{k1Ti
zc~qNgl$nRXh}2SQjo~If@9el2M8h<&pGw7{Ov?)-v<4nH(_S7_H`K$DVp-YL{QktC
zmp@~c5`YifLl^9@uCzbBKG%)YnUz&H3JJ_^RDsqL7`&GF(6zomx$yLvRu`dDY4_zM
zrA|_u+-UPyU43}~P)b_9%5^#}M6+>|NCcj-dUvHL$Kco({>e#M9#boHNj1tbQjaP(
zq*Ct?)-aD`NFQlWaJ_3Fe3`r|fr<zP!Ul6x@RomVW}E7>V8+8W4<_um7QX19iR;%M
zyI?W;Cb$818?o1`7ZU<(RS5rxq<NMStOQS@Zdnn5t%W!i@;FNhYhd5i)jwBS$>2?X
z=y#L8N@|eoXd4pcLD$|+sf5h_swj8fd9377hF`<cm(EqWmm<$(QW6%m^PI{vww&_{
z>W82&h?2|i;kbQZqrugBE7?S}!q>`pnoIX)<+j!98_v<dh#yTT@dt`+hC4M!`fu-X
z3KkhDk+1WH!7i<eNMkmj#5Sw<7mE)I6lGRG{=?EG3crvzIY^WuOXi9OsizD@QjNE2
zx0#FgbFl7r0?=KoXN}-~gWOw~uqS}}QBs7DZS$=<dnkw77WOeJ<L`s|r1Hf{1<)#$
zBP@Yq0iwi0-aKH0E3q1zM7GIK)er((;nn#eL~}$Q)5hb;PuH@NONEtqV)`{R2MzGm
zfQ3s&^^~SZjrzEppHhWN9F2vCFGMq8Bbg}<PtIykepzGdBpuuNg}+|o_OmGVL&u0b
zwTMhT2-o$#en7O`cX!xslZ)j=y-m91#V%KcV7HCBjxB3_LWdC`Bi#01v0@9k6;6u>
z;KyH(Uk%}+z&_i)G_ON`M=EwLF@2T$?EZ=HZk$4*LUe~q1W#fM=^>A6CC<h%cKO<#
z+>&`NoS41^Vd|Ionl#G|04|AyhXvHKPSXE{P{X$vO6CLWfKP@^T9Bl~0MU<mMQUSw
z$|M$(2~So%nbPafkx1>w;e<_o(MkQje=rpkX~4|qSy_}JaVsPRr28B+b{6+)-iW+Z
zU?CL@Ugw|iX$J?HvLg(1g;bNHayr7N#cxBrSPUPHcW__2R5#JT@|m+yOF9txfUi05
z8Ry%qu=pWp*h3tbe4A2$vfEoy8{0gDAlD<+uQ6IAI(Eo2{Bwb)x8;`>dHx)Ni%-C=
zi5y?Kp^vIjN|;O|ud6YHV4;P^CjQ;~mkxLi_P@cAZmD>71%~w+*FoQFo_bPH?IhJg
z7arKPJl(k)lpH%GtDQj{r(w>Bjl%TO+T*=H*7NMR(U>I;voLKh=jw49XIQ=-t}>Sf
z454K(r-<vla}nw`NS7;Ao|NcA^LyQ>q^C+&)UYRKn%HQojIpd;&$OmVsmp4d|6z3+
zflnM1#^P-)IiY;svPDg>pvs~k(L%5h=hVsSkLQr7#JN2qsTE+UOh&Q8h=Xk5Lrv#v
zOABvjNT%>}T6zDZVeOp#j!^tL9fReP{e96z5fo16S>W&@#KhJ4K&p__`a>Hi^FeZy
zDP^t=0L<YzEAiIC(;oZQX$Cuas!}^rrXH!S9c$en35cUcq;g<q1syH05O=ci{CWC5
z`D6df^&NZXgfmuX)T?|ZxxbG6lY8B1?MI-h9DmmQ081a}*q7O*HXkeQze6!<#_O@l
zDcQ#o`Cbp}$-@^j)j{(?2Fcd?pI>L@GU8J|5+UNA7o<&g_6`i_*eADPH3!Fsz`q$j
z%jDx*Z#jLu1R28aRLpZND(~BW115;q=<|B#B6%-|yPH(xJ(BP{PA>6>PgQzTf0tx5
zO|SMtV$BuS8B_MTqY58>G>4X>OE2brb!FTurZ_q9>(g!olP5*<`#0=8s!}svKdE{@
z&Ispy=DjAS4VBbhw}XZCCbai^*fA#CpC8b)$Z17~u%FYn?E{k3=|*YZL)*|<;G)sN
zkuM1yAVd=@_T8SMBgtl3=-iQ0^6LZbRn8<uWHz;(i-kI4`v_DJSjN~*(}JqNT-s|w
z;5@ff{Wt*X^xenV4cag9NllYHjxi!`O9o`j^l8Jf+F9zM`XO%Vd|t6qq3sSU!VDkM
z{L>7dbv@@VvH!jp6ti|}t$Yo{==!9l#kuE^6OOkM`Ma7vVMb2N+Ml^ZaM<U-?F0+=
zZ0@XC;QaHQxChx~pW2)n^WKIzF>mZtS{cKlLOP0^tj^OZk7^WvNTYm3Bz&=S<LmgN
zcRb9auR=vh21ahAd(qI?1y><Vx{3fpp~a)Qy1#Td&5dXNh_LXl19sg}gi5|vie&63
zHv~6vIQiP@GsLl9<>8+H80ZQ?Bm)bHE?EoZe-R(Gq}Dnqrvk(B&Yn}sbN+Il%g4}d
zrx(AnQA@Uf;e9tY_s}ICt8058%lcn(BT0#HRILVoH+}BHLmruN>14r623}C9OJPaL
zisd~!%FpN390dKi_`!Kj46&HC))(mg6*&SE)seNdwGI@(4AWbBD*JJ68%cA+;?Pg(
zZat-0DNWb^l9U4K8{Z%ruIIz7P>75w8~&ourlo!ovz2XkzfPDbjYo8zH6LYnipnPT
zTPd1}S6b>80#B)MepRlGlr^Ci=~#J0rbsjcpQoX`TtC&_I?c)7DC4#g#yVdAD)%8~
z^Czy0+OGg)=DBd!$sG(s_Q~LNA7FyYF6Rbu5ckKgxy)F&>;91Y|E>G<t^M^$%QT}(
zt%tl)S>D&+eK<Bk9D)U2FXfJ0#_T~<T8T7CR{fA9AV%jN`O#deNvUU^`;io+g*wTq
zMdZ-yult-JDA{=#$Wgb7o2G?h8(1a!`J>xNf<P`}c%?#MY3>{qi#yw`6xl9~*XMos
zmqu0|@+_z)H)92*5=*dSrNmh$?Ys%9elQ1))R*#t9cDEnodp4i5%$z2)XE=f%-SeB
z8AR<(lAYVB|Dk*a?jnK=E53QVIAVEVJfR(PgKcTX!fGy$Z>XucCtgnz%{K6vmSq9$
zE+wE2>DX^zif1n8^)MgBJ~Mv}`RxaBVY}2@bn1UOs%{P5l1avb)%tCte4+#ULowVz
zJKa>1JxudQB*@#ACX~0#;6u}gG{E3KJ2~cl6gqo*aEF1^a6GErv>w6*e`g*XjnQOU
zE5((R_h^}PH-hOZ0yvQ3@WqrA=9ecO2lltfYw<*<*W_i%F?<zRIMb|ByWR~LPEhGj
z%_a3PiZjLGz!u$LQgAVX<!uETy$_{!7zqU<a+~UZTAh%tc?>Hy7OR99Xq>!L=AN)X
zi3Eo;O@!6&B`Fm)wf}2w>TK;6Bm%Dk!kM=qqiI~R;^lG_rUEK@+U0>y1@tv)?T$*M
zsX5ZA|2edYS-G}}`YWPs@;#kcqJrRJ*{IY41)cZ^9DFE9OCfpi1{7{K<(b>v#CSwj
zJnKytD<XNIw}I!oxC<?JS=)qfxwa6$=b?c4U3osj<GVT@V>H=_Ohq5#h^QELgBd$j
z9vp1YJnzNeg|TM@gUA{WaWufB7O}|sO@w_+xjL;Yl|x}PgBFS6agyxkHKX?<iUZbI
zSDAF<`)(3#Y?s<Z)Q&V^`cr824|T2Z^qTI!1=dpX`xaHmOQV5FAKmRW0QT)u;H&N5
z=WkfCQ{<T27eA@YpjH!}CqLLiqgJaPQ(&mWfk*0tBnA+iQ3BMs@TXA&6fS$0;Oalc
zY9YLMgx5KnR~Y<_pfXQV9)DGlWpG79)Bxta3uZKE)6|w_8Q3R!p2B**H{x1fLonyz
zZMMSPPLyZpUD@#3Aqrj=oyYg|x&<2>{;Ao}n?cXzoKd5GA0LU@=eoeYvQs3*-8BI6
z8?>9IVtn(Pww%N)LYj0;(a&VX>tLs@oYiV{q{@Nn5tyc31v{^eIg>bvxVFda<5|AH
zD~jqJu4BE~b5IT1JsfY2$ndM_>&o`Getau*RG~x7W&>6`XLv3S)l|jq)kHGF8j@x#
zO{OI*br?Jro8XH|v=Aoglz*N$X6gC*u>~txDL>keVWJN?B}uu!#y5d91_32z4cXLT
z8Ok7pX;Z^n9G_BrdlR<QMX_U&6qgOsOMEu6BvweqpNxPaF_?dq3U24%GYOoSkiRbo
zbx7lQVLb5!r=*`D2@_1=J5NC^>Y1o{gJeAV_kK1o<|9_{SqMDD0O3hw<9~(gn*^vS
z`#xGj3F&0LHpG>BEQ~w0Y!m$gA?ye{791gvZUYi-T#t&ePQTK;U{=8n)WMaK-HJ5v
zfqG8PzZt_s-Z5^Qu<LNLT8CCl3?-t2XxRD`JeA`dnOvh+DG?v0yYRNjPrB#C*4ySC
zqf=(@t|}-Q=cWFUE{iK=SWi6m4N{@#110pXk0m}#6~Gr2WyEd-0KlV4$#{M@DGRmz
zPM5@nhkzH5w~^)2(|e~iO>a)oc~1aoTP#-k{xKq$)kTw}O{gn=wT{JKATm3EqPYb4
z=^Fu$lYtA5qO5K*$+a!bu~v1;*C5L`x{>01Gy%^5g$ETAUFMCyzZ?D-*q7i<Vg26p
zV#{{wp@XL+zHPe(Gdz2iKU`;ND@=~1SRbWLid7sk<LbYi$tlDpB_Ex1-Pl9YT1T+0
zBB`Zu+tM<fAHnDn4REyUIFzAqMv#A{3hXD5nW_vV{E+wDN5p%#rVQM{0%#Xr80QiC
zsB2wjKFbk8E$Q_YT09#eR6Rn{@<QTa>#JzFmlp%4A(zB0YKZir3Rt|Wthg3C!
zv&wfDr0u=kC`!Sf&2d}5$99pJ4fxt_@x%dPb;iZBUKW7KB~0>nPJQQ6<1xz|`b&v+
zu)S)rqT{eUXja%;;tbKUA2p_}Y>VIkjAz6|&C)*y^i?U5hhBdst(8`LGtV%r5EgU+
zn1Xfz<*o_>o=M7NHdH&_kHiqoVHmd#0`Yvpk%x=1pZj)1=|sCNk1jBpPZ3yW%Hh55
zUgJbhv7L}<at4LGXP?I;yXri=$ZG2ofGhh8ib#z5t_d6;khHxn2NoKc;{u)zTqC9x
zNo(6v0XWm$Es`@#qRefKlGH6o$#LKvovCfY;tOKMZXP#t)%Uaq!Nmc^ciP3-w#wgC
z$R!K=J;}|Zc#nOh-)~$=Y-BhbEqW-iyyh0N^V(hKGHO_`@Jm-Y!Z5V4W*}ZAmNkED
z!0DeqbRzj#;OLQ8H~H`FI{ON#*=wdhT&Zqnl(=?3Q9FDQ6CjhLfqNWntoHTHTkp|B
zI`{)_TE(xo0yIaAL`K9oXh_FO7SH7s+&gZ(@esEptd0L+t4<8D%llZV!mnxDs{|Wc
z*J%w}UzIJ>I4`*k0gM{1F}a#7)#uVNH!Epr&^C>#RFa^3TE+2|JoIL{3($-ipZap8
zDzRHHs6SJ1H7)Jv;;{jb@U`0GWUMQ%tU-h<AhD@e$;=OOvc>b#sDbu~tNA<Xwg@0A
zhFE<acAv_MUo@p~U6{fAvG={oj<;1MSrDF$HFTa5IbK#D(c51Sj2wMOVc0<{*;*CS
z&azgyNE;!3=r3WX(AXnIyR-PPVOyd#5%?=WQ?{o>iLiP3$V&lpR#Q@dmjVVG&V$@3
z;B(HcpDfRRMeLqy3=*M@2k9EP9*JGawm!j}l(2SNqaH;nGR1O;+j#k-8XlVSruSOe
z*kN!cB32IHIhpx3UqiCk3qD6%DwA2+_&z|&VB@Tw!#L*=I3R}v<dVYxav8FTpvkMQ
zmZG@i6X5=)GSV5ruQ)rHl+U?fD9u;%9T>oo4PP%J0F0Fy2#`uVi2)rlpAsK2Q*1LY
z`-Nt^B-1(1o28C=Q*g`R0HQ}n$bF9xa5gsX-Fi7-Sv3=RXs%PT(t$FBV#yA13bn@T
zIT(K%abHv&?y|RGy2ExUlP&*IKYrCizKD!}!6xu2b2XiBV8mPUq?tUw2uu=7St+3E
zD;(WY2q4>b31nhEqw#H*HQU&7=unbb_O6QuLO{hx$*{)c)KMaWwG(cmA&)c7N^g>m
zkmRVu4ovFs<dFWVDhxnYfq?g)H;i$lwiTGLhc|ikico6Whf4bV(a<!7CH{&fmK{(B
zO&I^cFW&qifLwYn5NI6kO9_x2(@_?XF&h8%kwv-xaUn^<5oVR(7m*BHP^N~oe@Mo>
zMQL85t!1uzQta<(ix%J=1>oUBPO})$%@BIAJG_f@oDoM4@%gof9A^b$=-oJvJ8OdA
z+hax#U?IA5czy+(-+0~nG#Q__{mr6~Qf#6_Y}Y5d<VP|vZoc`0-noRtqo%pI0}Fsw
z_tp#l6B;W(l9K@TB1aJq!-AoQNWmhzkX+;tR-lNI*_i2H-j(rOBQf?u`H@Gy7fw6a
zr<lVIc05ju)R@l>Y<GSZy7*fhRG4&U7)rzkdFkcLFu&OdjUN8JQE$xAZQ7}d=kKpX
zB*DUPlOed7)`5~kD&l$;)1)ee+yvO-Qhdo>(&Jk(HXpAhh7%p7=vsoS;9DoyVHZex
zySt<&{^P6p!5Z2C45&qX>D`_3{!0<Mi<OLXKDwLK$mQOR-|`yAW$#-Bh+^B3<M>kT
z3zSo>j9T1N^>4(>SqYgt#H-h0;B1*uV_bJG#0W{lNZwbp`6rDGf`!MDJxEQ8-(+)_
zVbvaV@b7U~v|{lvh2?6N7619h5a4{H{L>HmKk(=-1DfKXY&cj1L#~go-}f!zeI$iR
zvnYZ`T}NX=5*j4c_NYr6yr0FXZiLx7C*tN$N?XiFs$h-&WO~tQhAT8Z|GlbIz6%?n
z<+Az+L^Ec~QY}odonahHQ-Lrc7nnN<$$)N=1&nL%7Ou)Z`adYT@<1m4Kc1QUoHGeC
zB6pGG+Rzpv${EUhOQ9U)KDIf_P;81KOt~s&h+IR7RWwJ)k^8>p*qHnG{QlS<`)kjh
z=kt6%@AvC^hdY^~<l8KyoeqwMa&8^S-+(NLt8GTr99=ier3m;A7m4gyesI%Dh-%(Z
z8kw(8GJv~$`LEwQVo6|Z)V7?q3<2MvS^ugiM_~?c36lEGnrLo#2`tE3M;I?Z<0`3O
zm#Pd!K_j+EEwn`h{t-T^c0oewLcE76+Ex{$TLj~1<_qVo_K@y|lDJ;D-jlpoA&;CG
zFj!=>n97$~A-sT7pnAJojMi1kg{_ErkyQO6Z#+56+0F5Ln+*)HPRpWnZG)TEV1jS-
zAf6yiS!vAh)9kZdLjE`_eZs%M;rWK$1>u>ZOHI914AzWuMa|@CQ~kV{{=ye0j&u&i
zyK{cuf+~%jJZS_r81Y-0@DgE>IeOztvK=40Wo#N=P~TED%R)TRkwFtrf;X>^r9$Hu
z)r1Y6`&q)kpcmUQEl!~i3q_yve=^eZwDvpm%RXJpJ)rppV!|9*6otu6YFdraX51Ow
zT@}UDmR}S#2#Pe`!gUDb7jPPm$nVc*4~B4gHNTwvy?5q4)(g?(eH@9aYuVksr&rX>
z8-N`tw?hvqXkr8mh|?<CF;U--lXer3m}^(}`$0lt9?udj@kdF_wK^D&fhLU5AXH^J
zc^EfN;9TjE8M8!qeX4l3$|}%aetZ4Y(G~XbQcenC{3KL3@=vSlOw>iDB%=E?ZaSvs
z=#9Jmu|k*_F2D~~PJ5)V*{ZZ_6TK${L7!Ldg^~|ZCOQy`TUr^5bk%E%CF<8#9G-3H
zr}fu~gCQ>5ASct_D~m(ibRI#8i7Mf@$F^?wLj)_dQBc^r{X)5_O^enJs>9gYIc6!j
z+#Wrm*xM3(luvFQ>%p?qih8mc+V2@YXltli^B7cXa+&o&cHC^ji76lA*8H3=^P5gG
zZT0MCwEFh7L|0r@)V2LCez9LV2tvYmqo7WMmvL$TcU4yiUUW1NlX45P@MyrdaN%#4
zM(B>8QpLO4>!%OXxtzPyc&0{b8wvLw-Q}-WwAnq_t?Lm#MFD-l<i);IRJ$CwQp2Bt
z(>d5TQfCVI(kA{jW4~rJU}M)JP2;Pu9qc^1ADt$-+BdM>DBPpLB%{MDeED~N(5DZ@
z!n_a9{g{6=&yc7yESL$9LqfIQn);|>=Qg%eD9cTk4Cm^GtDvdai?y6cSVZr4_dz4~
z-R!@2;W7p;o#mij%$QpWuEjx`eoS^Tf|n-+DVdB#3byUMd|vG`-ce#IA~B$M=ZjGd
zb%o31q1qd<fn0HPl2qKbOZk{Jq;34S;jOdlf3SO_=EUiJ_k_l=pNl;UEDqIZ()4KG
zO?WB{WKTC<$VwJRuwtzuGE=B!qIK}6qgB;G@~(*qmgSEVylGuYZQg#LQNiUY9tpQd
z%RK>Q<T7Wl0mHDb<UVnA3!pK!`U0M)H^I8*8}LRu{E+phNf`Lou`?*_+9%G=4G~>e
z7lA1(6E^se#c*ee<_1})QZw0fZUcxofvon?l$|oXQJE=~wQ<cb(Qjhs8JEr#>0c2;
zNY`@6j=>Asky2`tsIpC?G)&X$-`w<}z2W_?uOm4KzYUA89wZr<W~1+r#oAcTGRnDn
ze51g-yntczVlWL=30&_5nI;j53R8=M{c1bP?qWfYp`-JTs}gbX;fr^TZ5>@POn7w@
zeO12_f9s3*PfD(FI!nN-Smr_!CAapRG(qH^*j}f+ga&BGB!kaa0^+BWc#9M}cGt2g
zvPk<HEre^TskG^8%dX=4-3_TPz`?Zppit8AsrLR0`=EsdE!MzIXSd8+{@@L4bNDkM
zhCaFdKoH79Cpg#EMk38DlDJA-w+9BWr?1a2>Mr%#zNkGL%N&Li*HorZg*)zyv!zi|
zy^)Xf2YZZp-rw!9O&Q!*LgMJS_W6L$z2i`M==YEkF)w({NM4Q@ZV^mN<SZKjM>cZW
zY43>nbVc*Zyb*dkV!16}l8mQa>d#xmkIVM(f^+m<o`<qtnDQ*1bS#>kju^!^j@?x-
z(JR5H2wvgW@YL=}{PwHIL38weDKENJag&*`1MBFxiG+9WW+vy2=iO*q+wpv1&U0J@
z&*i(qqEq4-JA9r<B7Nw2L%;A>|3>E3wvIyJtkfb-V#i?T(6%#qqEbIPK1PUqYlWZe
z+sPD+Hi(yIGg=qB&zjxXMcOId$r1WnZFmnct%j#b%2u|<_VmqT2{6LFb+py;c<^hA
z>n7u%`?#kt=2*_Kt^EDVo871LKgM8Qa4$YwLn?UkH=X_x`+3sXCU8-9{g3E<$t0tJ
zhl7*sBPF&PtPqMPZ6kJK<c@(VfkhM9HuwpDt73l7EO+R8;LXqnnXjRnTRr(>9@~f0
zD6^u40Osf;;LpewlC&mVPcA+=)(WgnFbaAre}}y^U}o~J4D8puH^)ZCRbvdI0nWYK
zXzG>wEx8vPd5_rP$dYtAyv85<JxJzrcjCbGepnLb5%u}N#F4}*htAXA_;Qhz%k|B#
zFnO0<GbyaDH9c>Oyk%JLgu^&|tIoT255!jog~nt|x}L0xeX2PJ5}0@`**JMudfn~W
zZcP701xtYhW^Q@h+Igt~1jCZU`&mXVYb=H8n30HiRCDi$`SB!iQ7aM!r5aDIMpL>~
z2R$&`PFhTXuQxzc2aUGj`+1(E6^C^;<areX<M`q8kRwDnO;O(8SeZ&}MQfXKL_{RB
zhD>L%1}1vitx!j^IOwd9^-I*<uY6~L&Fsnc;#EoCXKDPQ%?{!JHJ1h&n>y=W&VP8L
z<1(bT80ofChG|iE+GW#d{fX#~OMmIoEZ`tPkNkPU&^>KdWhypGz+W_xvt%<`F|?ms
zZ1&@?`2+mvae)F;?75O47eNnazREzP+mSU|Dtv*1#pe;jqowvae{Uq;CI!ztWkOtB
zW{I$)e?7(fT2XzG{qQ<6x9S~d=YnWm?&DQ(Q#&e`r^Jl4=|RzILt9e!VO1YXWdCoj
z#y|Rg1X+C<d?KFNxB$&9SUuR%F2;%{zDY2`{P&8LTyz^J_y#DhX6rY=w7nh|d<GZN
z6P5Q(Q}DICvJF<ogUHhBUyHK-C*2JSjy*k0E=Xu`-kDW7i<EX=K8k&RuqI>n%8;9R
zd)t!^X8l-%x`>kuG~hMS$EAL6o~H*q)w7BhT@+l65VEt_=AiApmRv6QC3knwLVnQ-
z9`7(yaBQ2~7t2X66n2`32>;3B@Y{AV8-Bj{`-}90(=3qJehQTvBhn<IB$!SxtpS$=
z#k=~#INo8cBeDEGfh!vU5TqU@)`OSORKM6-h;8S{T$0n`2&SBV<dMKkL_#VpkVZ+k
zxu^Qe862uJ9$yc9h|3S9_sqHc`A|@YM|Zx62`=oP*^GUUv7|6)cHl~CGMMckXVi_)
zDu?{^a-0KwhUamMt<WJU6~Mz@vXWbEp?$YZs-xP2lmqXGdrcfBX5QID$WV~k==E<o
z+f1b>Sb2HKnuk%4Z(#4nRyOLa;&Lay?sE_6@$ApdJ&(jV>8Bxe<g0I+CHCeLc)LMk
z2C~JoU6R#|TD#C9zqbY=KOEOjk!P9e6~uz2qM$^}^^~QHPi_f7@Z?6(y_jwBt+a86
z)~QIBSj)LuHOhJ6=zsfH3N8_Nc1A~3R<o_@(0*B_*6e_)Bt!5xEN>DjxX>-%3uH+O
z5;$nL=(xFULpvm8pYdA&7DtQoiamVAJ$kbFWC45rWpz)%V;r)D`|!+9P3A0Lcp7J$
zSESvJdavxAEtPKj>)`rgI=U;j<JkSiSB-)&8ti4!serqaT(G2c&cB(>aBM+61#k_N
z%iu*1<ii%UUW%_va%o1Qa+7Cu^q5Mx>8K*txNQbhVAYRMbfoiw)}Ue2nnHi`Z*VWs
z^2<DKi-C3D6K7qB3LM|K4KAEZINd!?rbp*6?`_88jrdnz$*lKdB$PkyISP$xqLtic
z@lhQsDy-Y@Iqkk<W?8+IKy`m)7}bK&>glj*VH9TPjkK0ef-1tv0zCIQ!5}v*sm1GH
zn2ppeF7ekWH{=Cnvne{JrRgx}J#lUPJ$e7RSyA?-r=$#Je+;SQs0%qCD7Mapt%{~j
zzRByWHOzIvZAFy`Mc>--JDTuqype+huT8urEblO`o%>`uV`4dV|9u>wHB}iOHEtib
ztgF|N!E=mbCi%spzjUf0#&4tImLh$h{GL4zIaNL40vI@IrMnybQ~<R%beU1Zo@RAE
zLENFJNf6Pr8ryevYM+9?Md@xnd)t>rk2sg4{NUApDcfAQ#E}nHfd2DI&@!UJq&&az
zm3oNLjx~{g;|k=xYr|Op^|84gx82%_OaDO0@6WK)51xN&*{P5IBEEJI?HW|z&~qGJ
z^7Ew2<kmXsBU*~MSVY6CmOp-ecIC=c)wYT4Hc%l@BbMZ-)a8wy<*zg;;*uc(MlhD&
zWL~PL%-_2s==b@#UFs`8Sa-mU8RqF0C_MHoekUv0u7N5p9M%NayQ9vs5?j+)c*?0~
z>{wwFm{bJh!<o|Kqjvh-xL3RFX@}`B0$RH_YVoHr#u5-9h$Ld*)@EW*&mP#fhhxT#
z9jf@|Dr5~cTO_Pe+;oOf+esNVINF&xDaY>49EAbm0Ei**>doB39$XN*Ihi$Q)6@j*
z#d<40rL`<kF41*-W<qw!C^=JcEcBeXI-Xi968&=@@?B>=gWs@FuIbg-S2I@&Vckck
zw5|~)!c3^%#7Q+VV#j`?Y23X7_*t%x>M_h)AR@Pl@>GOWLLeH*JK7gEwWJjsYhefV
zAPaH&SFOkmuQ+e&!Zb!1ut$Th4#Hd%@FY#0s~^USReG`CtvVIykFODBO0BMu!__!F
z&lcP-S4=c2Or?PMP;DLxcyB*@4i*U;<<O)nUJ`2)y@ga=Q78CU+0Ah41j?D-|Bh13
zwzofUc18<%!CzQ{SA>YbG1-g$qMe64vAy64A5Z{o$PiZ}T2nO5zorj){=2_Zg+ap6
za)NCm3&utQkPUd*ji8=ADI449LkC+1QI^PMQ&-5QPgbCK@lG{$<lV2&c(4X{bc_qO
z?>d*_k8CiG6w3!kt(5k-p8$r^nZd5#(|h0pS5OCSEu+$@E0CdF!~Owm+-)%CBdmPq
z9NPv!ZY>{jT)v->(v)OJsQ4U0LZ8u`7%Mw$$+o{|`!(^ZFsu?Q{aCO<eCZ20!dI}O
zYYEfhUn;!hb?*v>ykvk+Nqm+L06KzR@E8?MtcZ*IK5k6z!KcIe&j-6^?t*AFUFYb8
zX4s4qesZ<KS9^c-5HkJauq^S$bcG&+xnM!p*ueP{Wo{mn;>f0Jo@&vUhow=sdr!4_
znevuAt?kZ@cxkrtco@<@u|7Ba_TNaZVy7o|_iq`6<*@$wq84}jjEb2*Wc;pCe)JR7
z;xP}NB=7K=u!$qDtv%jDC|PoPK`Ct9O@w<dG@Y_N^afU*VvavB3RtL%4f?yaE_E`d
z#ZbKK{ejoZUalQ$DnjOoWYdN5jqCaQKpo5C@cyMs#aK|rF8OJmV|`cnljJLMnQyv1
zYRY}ZcU!sAr}%cP{lwS$#w1Iv>~>hfFm!)<Pm_+prc5J%d~uMg%Ut?M6K4P7MYFTF
z83xd_Z>Z&+R{SixCU)hr`)((SxSE$pEak<`5?v8>N_gsy%w+1uL#YJ!qps%|<8g^I
z(T)73YzCj{519o8JsN7)9R<N9-6zxM#%u62vVwn$t>(|#gI!|GYL&zt=_)F)a{>sO
z2r%U3HwOh=KItjxJKU4IJWU|%KyDG))7us_7Sx9mp@*o6`@Ms^3x8XN`e35f7nE3X
zhn}0Kw;9o9UJI<=_#t?6c{gT8Qn<3c<5`OYfqP)hlCNaRbCvU0@p6#}k$WfWrQ*6!
zXDar7mKmdN683FSFwy%G<NV<;t=36pw8-`fl)$b<Kq??#+;@4T^E-GU`{HIzVpOSa
z-Zvh>;k>Hx>DkuP-AT$94a`32=E87ob`L@kTG1=nTHmmG8y9}>?&@77F7K;8OUq5E
zqOjUnwuJ~W71uLvVHug6Gm1bPV=vLPfx*4w^pb=@Z#!g=g^c20#l<E$AI;d0TR-w0
zY!q9Khw0Ya#LiVH2R??>!ABmKq#HSuUU}To|FnEmD6GZz)+%FH|1N`fuXU+gAhP7|
z7II?DwVvhB>86d|qz@Hf@=(&@h^;3cBTC<f#y;#i+Yo2<JMji^eM5Zt{z-iu(k}&2
z{|(sS&P~Y%-b-g1yw@Ads^XY2oRsX>6y9YA9^|i(+wc^NID-Wj$uZj4?Z^o^K=lx+
z^S)iurCkeCc$xhRCyE>rFWnB~XjFj-){Pt%gbZ&S(y7*t(k@GCE3q3}_Jzh`D+HFU
zmiBaho@E8&$%0=wZ*bJn+^tn}vDp_w3l*@5$Os#j=3m0UeSq$)mSgKn&MoNEgBiBq
z!xu7yJnN@x4~<(a4k`Rki;~YN5tfT9+HHkB3D#ZEG?MdTS3mp4M3FSY(KY-w;M(p}
zKFo2DO+Kn#Qw+po-0TtACEw!j|CeA%83^CoJX}Fu#LY#ODgoN71Z7DW!FH3*LoD-X
z`C~xjcU@9EkVYCILfhm1#`~sD-=k^a2rn3&ygM~`s=M@!4{JLM<^pD~UZ^m8tz6xX
zvl?6V!!}!~Mx<3}#k(RqN`v9O7{ta)j>^E~Hc11Pc7pY4ZhuyN>m6MNTbFDh>KpCD
ztDIb8#aocqIcGIxsgu<_7xxzPgK}JrQ%Sf<c7xbWJF4nLwn3FiHLa7~HAPE5t4`~`
zpD`4`H79vjH~Kn`CRqlD268le#ad$eHnR7TIKg6E-;<J`Iy&gxvmjx+b{7iIUi!*%
zEb1)Bx|-BVoesO>2`57AhS=*u<;~rrpK=P8-)j3oVIN@=TeyeuU$$IKW5#V27xMhD
zExsbH$6L%jHvl;LaLR9)T$M)Q2$>#H_YC$cC$kdF^8A*s$Es|`IBZ=3+y(qcwIK?W
z6Om_bFKBhg+PlZX_pILQ4PxQ{NAhDr4>b;<Rr9ZSD6u5(z~eu?4YGdIuN1RLXme<J
z<UOODf8|EBXB)-mRBUIuh3#IusDp%sk#I`fZa>eiS#x@e?dC%^P4UR*clw(zwifT_
zdEJ6(gt|noavUwAi%JsE9a6nKp4QEtME<zVahy%s;)pcD%VY+Shoa*%Ql@osO*`;E
zBW?oLx#W#ET|3t$ZlDDsJgmCQQ4_9$(YzXv;lV97iQ_z_GcfUYuh0(&)<*@C4X42@
zw^uBL@5G~>uWM`-@GWl_aA362G3IJ8jW+9MdaTrDfdK3@+KQl5Y#qheNCVgX0V=^`
zOhK%SIXD>4$?PRp$|fF)F5FT#A#y+}<l##_aPrmRdQM7Pxx>15Qi~Ha@NUbJ*?4q_
zv&Lu<x)^HO%&rxT%q4|b8ohhb&u@pvu5vhh=Jbq(5q@nmz`=)uilK(&BV+ikVlki9
z^w|kM0g$Hq#IA&h=`nzH_vqvXJmLIvl(O*bgkOY-z$3^{uf`8Xzqg-j^1}ofO^Y!>
zXF);^1E(8(Al-U?zPVigMqAbEb~%K$=XhxJbbhg8HP$-~;sHr_3hT!cY7KDjo=|v#
z57opCieLoNSrGB1iWA#Isk?Jgfs~ITERwh5#!OZTUN{w0hwrb6$7G)<$E6yMN@c%t
zH=VJgDC>nQek#`Nbhz-_Q=|G9+lJS^Jkq%3qjNJmLEqg~W+Xetbq(Lrb-`8|?y9l5
z_j^_=ZOReQTkRxz(-Z}!8V*rZcG5kl=DOCec^ckZW<w+lYn2xm)Lo3MFSo%PLyrcs
zKIo8!<a(Q|J+GXp{AX|6_!M#c5bz3Vd`Bh^D&04ejyUD9q7oW&(iSAY|D}-gW;m?h
z(XQ)y+M#q8*L}6FEMQP8`2W?z)P4irj`Uqc6C-Z}R=$FA8*cxjzAjctp_8LB9&Xh-
z<!^;`xwwaxUIF_mWK8~e2sp#$bCY-@JrXR@KXmUtxYS5u@4un7Go@`tWGRy0F>j7v
zvW{)xN{?d8c+G+2M3tq$I#uf|^TYqyfFJyd4b^y0*&f(`mWlbRzr`R<LD#K#?n&V7
z#&;}cwscRz21DF4&WMwEltESc_eCPZ!)|ZRYYn5+4PAb~UpeQ8J`YXT!<b$rF3Mfj
zTIxAW)?l<qse&qNF97W_=wAX7?&n7p>Z<^l6txrzV|!FJFG3@lLQf}Bwxx}T0`7M*
z(#EJI)$GDB0ob>BAAHLnQ2<4RqHu7uDrbrD^LL?8;rs$TiBB4wfAiX}h)yI9{vvFE
z7jl-kT6&#G8cZu%yHDCs0tML`H&R_@Ha?vul6Ed78GR;i|0yggJ6H5@;G$cH<G;-H
zKj*<>lo(rdP+zYH9DQEzTW*{-9;5~(^lNN}0*Xbuma7)t^qLM()5ql1V*7Xgyn(Nj
zC3$~;%_0YsP}sA}e|L146tMfq5XE%i4uWltz6p*0Z`tP?Zg-}ej|j3f=?uSYvGsVJ
zrnI?>pZBg`nLJ*&y5m>YLmz`Oeg4hYB=Uxilo|k3&&-ySaOKWh3q0k7OJP5ChnP`%
zW*qBNbHwzu0NQ1w@Gj<xui&s>!0yM+p}bCZ%BJBnlr(|E(AYkMBG3pn#6I$1+`^(y
z6X&W(9(LNoPGliH&YS&)e%<3p%>5J!b$%`Z^GIjAvdP10Cg$2m-e*n-Z8_^aC5pVt
zOgIPY=8@+N<;vgKy=43B2rPMS&i<Q6b?HPGgB>s^txQN)g+{A(;~RX9!ecmQ-T-7!
zQ{-3!;;#C|#wRSgBfy#)0w~G{;WL6PyO?~H`i@wuv3Y@Dzv@joN#cCIAg5)eSQxfb
zo0BP=hsqsqlG7OPH=hYj2@nJj!N+`5hg9lC$Om}3TrcMSSHy;nZS@YY>9gaWVs7Pm
zlQ@(%TXdyn+5CtX%c(!OLC%4H{Q5lLgrm#i>;^7a<K*P<Nw^f>9b01z{0U_fu<x!;
zp8=SN?<$+DGDn`T#x7p{<;f1R?&e{Be^&rrA;4SVQimH(daw?U3U-zu{k`RfDsv~#
z=^AqWwd}cif@?wl#(_Z5;|uHOP_hUj%FoxE6fcBofonxv`J`d+rsQ;Jz*dy6u7Z1J
z62U>M@5H``LgoDZaLo5)fO+<s1N;3qH?>7XCR1}D(N{f*%f&@^4Y4%|W~>4esytSY
zPA<A8cGUyISkaaJ3+x+^?MP5@UuroyqlcSFVbL(o|MN|l4tB1#yR#l<Aq9w<x7l*o
zKCs}{05&*55B}&^CyP{Xk*jiBuwyv%;*%GB^FRANcuZe~kDFg}kGfZVHKwZKD(j(7
z8e31k{l(FfFtTYpXIK`>=&7ocZUuYLp0^#5mFzhb_DBn;*PJ1HhNUs{9-rAE)P6VV
z0!}(J>-=&S=Xd%5CAXo3CF2}Sa%DnWn)E}}AFIKvQwTJK{P_QxWdywqD4GetA-Pw2
zvfX@BDH~2uSiUWud6W>~@%oMes1Go?n9XMY$4=t3#EI#y&)pfQ8`p<b2!}l7DhYVS
zUVrn?&u_;cx~LM|pi@7~r09g(|D_rWqP67xC(<Dsq9U-tc|+7O8%b15l!*K2EBK`f
zD1ZL0vU$}>qR_#Shw{sH9RGI*naxOZRR1=lfP`DAgh4EoSL|=PEK3p2Vcl@m{6<{7
z9xTadyj&uPcBEz?<cIs01GWc7$W%|03}iLwGcrPnvj!5H19P^TX5=D~^-{z=Cbu|I
zyt=wb3@?J<(|m@iEEf^^K#<#7xBz1zee2Uh-X*bXHK3q=Sa?%Z`CLM>2@;yV#k+t}
z+BEjQ3VuhXLeVW+l5Nj&bTNK^!fawf!i9#M6p6KrHnWpZo|Tdp7d^)S(Uf8<T?!~>
zY+oz(QlmuH0nFK7gJ|W8`<iLSVZHu?^}tN=GDifFgWmSyt@t|%k8{fp1o}`3l$N#i
za;D*YqUt)DwAs64xA32;=AzmkajW$YB6oF5V|1n#{(MW^o}HIskG|<C7#a}!?vqDg
z_wMh%z=Mlx^!5;2S2dGiU-b8c4Pk`s^b+j8+#9PlqG9`3Y4mG|adAaF>GiqstX7%L
z`@L%D+s|<+<lSp@iTB>G{bV&3R7Qr%8O8g5XxX(!JIT=8$!2Y1`Qrj-6-d3;_;9NW
z-QlNjxC>i0&F7r8Nt{8e-#7s+s?|R^bOn?&D>Y;dcx)#cupMB&V=1QhwCTL!M-4+R
z4@JTEr^UQ-70JC#hr$Q$eS4pio#u~*aCP6TM53#pAXg;_Ua);8u-P;+1J>JYsj{cS
zI<{DD;I?}707Rvhi;Z9V@5s4G)MsReM^N{AfnXeY^@%qCcp+Y&XOlV-$ayk;mN33A
zd5);di6qB%AL|OLFj3bGAnT70@3W-L`8jk8M+UTJ7Z|chNTHXY)6Mifb&joH=kD6g
zu$8a9iX^Ao&9HX4fH!ZGF~-x0=F>-qn0v7v`zjpo?viTw4{|n5?r%G4%{XF3>u=*N
zyQdK!W^jr)MwAZiq5CYdwEUGBT9TZp&AHN6?z30$e996XuXGaDD`k+N``n0o^fx*%
zgzjD_Fa`>80j67<+6hvE59S0`u0Xn>&0)sQV=plR8bn@a`ZdW$Z=J)5tTCj=+8l4s
zuQ!sTMLL(y6mc#d=^!S~H%Ns0B?3a9Tjt!6R(?u=Gvi*gbR<uwlW3SMcL+})O|UMv
zble&GW`AI{*b;SCKt}O-67RZkqS3{rw6SICS072udB2DjIXTedjFGp#A!FyEdchs6
znM+dVo|w<Xj358z-B^_jn>H%kNh*DrNJR*KfHyd}Gr8|UUU!{pGrfyS9=z@F^ZR}y
zDsvF<-VPQc3(L`hPPn5Q&OoXtOOk&<YEz_Lez(T+$tOloTL<_3vx~h+Tl?4C_ev_D
zaGhK3VTVTU5eRQV8GDi4#?K9@ZQ2LD4vhwaVOT_bS~%mA0CCaK5{4iJ)Zei>SMvLP
zpji9*SQ-QIIT}y>;&g1dk`P*I3sx3#q<^Rk`H`4;<7Vv8L09$H+7s62B3)h3x~=(F
zi7R(~j2qSDY$jg1OR*o<6g2I6s}ZRkaEF2TgC#S*jm%iiKl%6rOx)#}t&4p>3--w>
zu#s^m7{X6B=@wxF25$(St;eb_BwqDoKKp&KM4Mrle*P8a8OlA>?+YhTU`9KGi5(F!
zs9X)kPk`<-6W$z3h9S@_4V!VvGQg?1mc*-fMz|3I^tgJ?XY9XZTbv??UvrN*be@T}
z)tIVsMY$<Q1LBf#haX)d&td;U4mhW#qhRL*R5`OsY|Na~dmDA`e9!W`nXTD$U$GKv
zU;&=8Jg@NG6y@;Yhv_Wp<6t?8eKe{-9U*Mw{>@5N%0CI|R@TTlgG*3FpFidP6}LR%
zLNmscy;PrA2{62~&g*Fc%MB7+(w6@A05khYaQv5>Xl7XDyQ`98U%jCrKUc<<WkF+i
ziPKd^4rzJ|J64_FxBVIvUQ-UN4@SxsTlOIvLmtO%q$^TF2tJ#c*w1BuRJ_gg*NJAt
z!H5rcUlVp-mYiA@iHhR;^59-&<2A`P)A|^uSt4bFe1b>~`Ns{azv<IAe<Ja8cO23P
z*3X{Z;jk*9nm#PvM8Yi`Qe2~MYeQ|#-S<?pMeT=?l;0nxp}QDF*%^!W{b^JURg)9s
zVEp1gnq$$Kg}VxsSZovL%v=0H<9y)f0{{*alsO~18+|ko*wZTfa;;bdSxB;y-Q3!E
z(qA2R(9}J}#<Dw+RTG|Y{okJlP5YE=%DR0QETtSJxNBrneg42e46;z6WZXa5?zy#*
z<3_yf_eQ~zgT<xrN8y>xR_Rq><T?uSUPUnaNCYMs_$w!Q=S9*1M*bdy7i$~~L@vzp
z2Ln@UYS!cgVAtO^hW{8VIt8}EYV@0;kZv}W6oyo_L;1gm9Y-LHh1v+8_h7K!?m?bf
zXlhccXhF6t_l)pl-u6R`A77i<Yix}K`85C``;guo<OKVQ9N>NA#W8Chw+hjO>l+h{
zNNmK89d>8AOj*z(`76pj29vbp)%;HG5EK7QTQjWN6uv$e!)(&pEbvp|3Z|W913q8$
z?nMVeqF-qL+#t$AUVEwdw_2Qg06HFdKQ!-B*5h`+){KKXN+`Taaw7YEq3HhyDBy(5
z!TEm)3tRIZbZ%$oKdV9#WSN=Ncifsd%lEvA=X2ovU%t-r6}f+Xd4(vBqy8a9E@dc(
zURCIt5SH~DvlM@)p><jOujuNTLpXjrmF~lHkf(Lo+cECoXO)qFIZ+_+uTvTI`AyqG
zaEaBEqnU>_{5w={;kCUX*X{@R<LfwKJoKI+fwq?W9inMbvEy2LrHoMQ)0xb0Fw-!A
zIs6Nnn{vfdJjaKeJ4p>|)q8d>>$vl()E!$DweeOQickH3#B1n}QZ|D_D#jdLA03w*
zn4G?wwi-|w!^8yx&;A8~`JH;$56-cbpiZQqgZ6pfiO$Jl3VjIn#Zu0!*{YFZNPyQ|
z<;c;6U}~ryPZDme$2`qoeb}vqOe~;W4h&<+geK+tM`K>ZoEUSsJ^JYG!Z5h$0pIJO
zPx2L5^*3g*RTZ}EQeH*;L|V2hUX~V!CC?7)Mwc}Pu#C6=7De<y1yF5+O-YvPVhIzq
zhGYxcFk3h#QT0+lgEu~PZ8J9FzgN_Gn{u_SKfkhXONIu8+$Nju5361FZSy@xyFSCw
zllTu#$ZHTFv8cxs_r2w}oQcJ}GF<mb;kY0R84tj%UJs4BFwC*V5wb4#jS@NXg1t?f
zR1Ni*G}rjsYF$6UCAN{~viyVV&fApVA6_L`j{&lBmeuVa8c0zR2J5gHeih`$w5_N5
z^MIU(Z2^Tmg*B(W=rmpjKD(7sY3!B)w^^U`aa?Fp0%=`A6_%^>g2>N{OmOcg%MpEm
z*ww;rW{(U}DMar_?0aQfnSV)1h!UUoy1x4P&faGP!&>q7?jp}SY>K;Ri7{fLE28}=
zld6S#l$T^Q*RSQ4z8LOq|81$06|(4Ib2MTuFw>BQB))y2MG1wh*W?%&T}}vhZZ5Wx
z$lXea;?=!p@|&teqR}go=?_Df_M-eQgQplpM=ct`r|4-yZ?S4qD%t~!T)y<aVjASk
z=HA#MOs1)C!=#h(Oa?h=Wm(S(!d4DsBy2}>8jPq4{)ecN8_IP|U@AXB)c-!+5G{z0
zaXb|asT?~tJGyivdg-I$!wW`z5lOriw>(^O<+_4fVn4wbk`+POPo-J9{=-Tgn5kfL
z9ZOFLw?wLZRTYA~eywIx_)1oVpP~ebJ|Kv`m29H06^ENUKBbvM4rX9(_T}IzKcvD+
zTN?k({@z$4X?&FH=gk<&N3)k{dM`A5(x&vq@OG5%PTJ8_-+t~J<7cu@4^&%?Z(N=+
z!O<QNt;~NBWb1|n-kBHQM$hk!l#4<M{Nepag*4*+i|=3wK^L9eNI#e$LUhA04XHO1
z?)_0#WOq074&-O%-^Z<)YxHQoFT%4nwCc&u)xd>S`Jc<9b+IA$=8c;#_<gw-{DU5`
z#F56=koqdzap8-KKXk3pWxZ+!31jwdFW=zFkG@}|a{@l4_GSPvvRhgHMQmu|dr#t?
zRdKUiQg8X<+m8W|*9B^o#yf~f0IqCfQCx-`k6Gp0M#8?&kIB1+E{U{QKepNY_Z*cw
zpGE;S96_3et-tpFW>L+1q0{!!`U6=q-7I!w_od7z0tTH<@s((fu$B_6(-$p#fdF~9
zfbSd>VRH+buDa~;+6!b|#Ex?kdG;2ta`Mq4c#Tnk2O6le+p^eSgQ-Wp_zG?vRs8X2
zMJuRbvPF56vuP2j8T#1wz_@dveR4k@e~a)_2UWrJgOUDS`{hVLm^5GTpiI|?MQN<s
z=(rL2$W{<aQ1clhg}GAVww0P2b~GHyvrz-iT=J2~-}DjGUy%t?5H)u#7m5xOI((si
ze>|CUCbwBIsoXid&5=JTsvnT=#a(vq)z4MGbDRAP*4lflk9n#-JPhiOh1(npm3-iw
zX*QhF_9Lzt79F1PXc-~4g5)E|hF8mVr454C#$KG!TzF;klKy14Q49)tp@qD-+QYpC
zN4n({b9MqbJ`(A$?tXJHsq`AHXyMqfD9U@!u%4yVU|}*AbDqfWA&dEC3JK@zvK|mx
zB=e9c>os$7u8z32!F;?bLc-oXLM=D`B!>T~*=ATR>>SD5{j0R-20T7;%Gw<N%M`Wu
zhQLEIN01!lG!N)Eh_ek{+XVAHj-OS+b#RY{_|0gM2>ZhIKic1Rx%{HI2$OPOU?UNa
zk!9tf%Wtt`*6KnN={?%>RnAD<QSjPEY`76|&1cwj;scsdG_Jo9o3Le=3g5%ia#U<K
zwEj81OE^snSv-#e$`m1t#tkBGgkt;D!(8bNR{_XtQzcD|sTgrtZBV+8YHEWltn~+6
zRXG$!&#d$o$Bs!S0v(|Q+WHuw%90<cIy9J&WOQKa-a}UJa1lyj{#}#H)x}3S2>$}q
zSSg|7dlt7hbe&*B`{T%ug99k2I3SL2MJc2%moBLX{c(|jecLC7@44;0p{O5uCox_1
zQka$nrkL#jYuT_7OZf=vbvd|-6S&@&A%dW>t(K?u(LRdsnBB~%cGD_gvUbUO+4gwk
zR$;8UVBZhzpY2s;_nyFBSS*Y8$V<FI<k`0V2?-R3&>BR&s^JpK;StJ+3V$0X_*G?p
zqxPb`UGGv;Oq;FzUc%KB=FMa?2IDsjA;Vbp6FX<fpaZcz*U_+Vs=b*tW*%?PRLmnM
zT+K?uQ>r*S??UPx_5J)q&$$BZM9@^@_WAN}8eQX{Y?C$*j_uQWe-XNtU#tV7_(u(w
z<G)sUg-vkDn;ri6`GKjbLtMJ6n-Yr71dcSd=cA5|2`v};k0)i14rVsD>L4(lk}TB*
z>0r)#`iVw9C~DdYX*33aNItlEm^bbwtENQR6Gq<c3?x$#-4{E_5oMgnuy8J^z<i;y
z6m`UuR#ikoEAU2$kV+ZW3tA%{*YJe=vr0^SwHZC_A56mC&-~o733x9nkyj={+$ops
zC2zKLxKf~&XlA!;su4NOzhyvGd+#9Iv?b?-^vqp0n7dXQx3<z@*TNIVhmQF&gDI_h
zGAurM7hZG9rM9}*xi|HBK&NeIq6!=l@1JAaI&Ps9%>JHY&q_U|&&<h*nIhGb(}j$l
zra0h@d>@wU0^zN5OI3CIT7?T**C%2A3#%e{EZz>StJ-U{J#fk>B0Mc?{m<U1NU;Y`
zhFD=d-t=M6%`4@lvIfRU6T2Tuqh}uW<}@03qBn@@#JPYWq+4)@9Gt9H*)j~wQ#n0v
z>h{@%ldR_`zw22-IU&*ebt6cq>HADW4&F$tUv8q+Q}e;(bA?{Dg4AzgUW-w<ZcWRp
zVqAdEr={$H#C<|ipv-r%A;B=uf>`PCpTOkaqxP_GBSf=zRDy#XXkblph3EL13&g?2
zg~<62QEFW9)Qr{8HEPUr-&o#e@N{UwsR!B3jxgStz0Tbup8o$AI)CjE22RaijHdTc
zuZ2b<xx$eN@&vzTWDaTYSt6OEaus#<-L?f!vqyB(y2aq&T{=2haQx^8I99vsoYwp1
zpLUUAIEjzyd)9ODuVxS*tk?e(d~;%^O-5KIsjhz`E8B(+mLHxFe0@cAll9laAphT!
zw|727#vC2C@C=jitR{NsWB#mEY1o39#-q72D1mbcwL8pckNqc4C%Eo_GydKMsm$yC
zoNVf8R3dUcZjnx*PA^esSME&(uzlOP9&%_3)f?zzk-170F^8~}|3`g-TYPwhO+L!?
zjMg9P*cIxZEydzjTzWPdk{1plQPQ0M_p;#!cj_m#rk8U&QfMt?X%+?61VV>*9J|AU
z98She=IPB=&Tb7Lrf8sz0ACts6!S1>K9q@g^eGj?pb|o}R=)Aq9nX|rk5lfSpYhsq
zA;oUe<j{nsJn$u_au{UH#c70;dV~K+=e3or|6yk-`~f{|48|ot{$8RHk|#!~lKi}X
zqz+@9`kiM5z})`O7v6TS&;2vy$wjBI3!g`2Qe{LJ17k+$q8lkuI(Hf+LX9pSNx#3g
zv=uvqgh|kAB|tY{Zk@4UiRS)vbieScFnD40ug5c}5)u;vBeK_YK0}PFl-&L9Nm`6;
zcwA`iCqnDX$Q{^$e6c`61Kj5(9~z))%9WR^0>4;$WRS(vI0JEuRaY)OPiYbU1OAla
z^oCTf1hSXM)(9Bfud?sP7xCKtZ7Pi|&>303%>t+~wX38M0#4w<r#mR1?&i*tf!yq4
zw3oBIMc0$rAd3uRk}YxC<2GFs$i<uC*mrxvAM%5n$Gsd+D+994Y8pU+;R_Zc!?ZC|
zH|!M05HAoWrB39z(VLcg#aSN6Pb~lsPh4f&9QZr{STww%t~3e-+KU5XbRz5d1bZPf
zZiw>}lW>|cfE(Jo4WQ7Xd&Y3krycv)@{^)o<Y3D<AYSwz*~AjOSII!xo<MQCN0krA
z9#x(C^NW7i>dsDNuYAzFdEWYcUK{|tqu$KT(xtQ-Yo*@K<2eI4MWT~m!lW?P`e!J1
z!ylq8N4f696M4s*mmuf@HO-9(1+!P#tnN)~2A-mtDAAA~1dm#a=*A#Gf4%}h)idhn
zf-3!nPOb+(cyejAjP>1dgB6;gSUFKP9u!+Ws%By}?c1$k1m4R{I%r`Z<pX|uYj);%
zB&_s9^sS2JdtQ4^+%Qp$m!jnt7V9r$+Ev6S6Sm$2ME9X}mSBRxW8>gGVrcA+n80B#
zJ-XT;<K3)c$ath3xL!NS8?7=fz%UW{LAZy#%Hp@IgaEcJ&~4TI50FZ#l^S%~y43Aq
zPQ8(}fm&42#)Zuo>1@@%ex=M0Phq0Jw~ZtAt9jCw58(v;oxH5o;()6puSZKXoz~^Y
z&1^X(`$SaOWpGL-AD`-sB<>NeOvIjPxDjhv;8?p%-F_|%q#n5^-^zS)rm{7&e&Z`#
zF#7fKwQgZs`Dog<0&ocZ-?yuWE4+N|=QY-n?l%ex>W<+CCD!%1@uR0(t~A$0mxcxG
z7@8uVk~}y5uiL2fOu7G4^Gkt|^o$x}{lu8bjb0b^n{R7PNNkWwe_x;LgTY?gMP^0+
zTy)IHE<nF9DkkfUbJ=d<m5x67*QYC_Arwp>bL8XqXb2ilxbA-Wdo&*XWW(Hbgo<yG
z*MVGm4*TVyo=g)h71|I!L?+VC%GxyD)&PXU;>6V;^2Ya>Rgx(NVFvl)g>u{6V$!wO
z`Fq!In@Dfe6jJE(rtbTe1sk(hA6~_ZwL5;K`DnTx*8IzxF8(o+3CliuV<)c?a>TVl
z(8Hdd{`{74#0n5#y~q~7U3|);g~SB-WgGe3n{Ujv&*O+wr63}Sx-8-Z7w~TsF$(42
zYedOmTE_!m5@b_wuzj~hE^T+WEC^P888qYs>$89p`4la{V?mr9JNF{mT!iG>SW-3G
zX7gZj`P5~7SeMIN*ts7*jSkc65xv{PtDJRV2+~VI%0KZBW5TMvhlYxpn9_F?_FaWZ
z+QUo8GHL)pu(aBAnL3XHtN@ay-=6W4N;FdYDhD680XOc$`aBkrj0Q_h8^=-Z--eRW
z>rVeo*Bu@<B+IyeZFCqIi&X11ay4D#INE6(YRAPlE(J3!DR3br8>}?o4M-PR92-&g
zZ$#<diE~{~Y58|@nhIRth(}S%D_?W)0S36<a9Glqu^wFKP8;u57xL-OheXS6&>e*?
z<F<J=gQ?|@C?F74s@&i;Cp|;*lS#fseiUyQpVDIUp;2lFKvcRD`TXY#4q7~%=^G`c
zn7G^dOiLc(YSnvb79Yc!%gcRb<qi^l&}YXvIs!v90{swk@gs#kh9~Chg(__;_a91M
z6Z-4w+MMdQm%}%If$g>Xxm2W(fr};iRkGQ3`WFf)EZZi=FjEMr0=NJ}#8@I+z<Yqc
z#xEh|CF1Xfl7~Ek;HM<et;V>Fl_r}N9Xk~6{7XDd(1*dl!d-*(8+%SQ4qleuQ+5Nu
z%9pV2Fek8rasn@DN8Y6V-h(cW!Nfor9#NL&0?H-?Xgcp{24O{_*Q8(_A#5j|+MgX4
z=mvJ<15T1XQaESko}d0tY(NgUrJx?NsgdLFPcbA|2fL0>GUAJN!06aMgOeHJ-#Edb
zC^0kjiOAw?&C%bw#)ny)ogiu5%SGx$_JN4sQq@tGjFT`)M1YY))4<|-2B?1(_ABCo
z3`;Te72lWvL}pf^&Ve>0wi_+3ILN=L=D2W4JExaxyf21Yh6S%(QBxWnnU&*mA<q@J
z0<^BcF^Q?IHXbc}3*;4i<`xO9eKn3!%NZCJ0rt*Ftt@{iyf|T~&Q*8?Z+QuEMs3|C
z{h;;GHfMqbq#Xz2yU+mZiY9wRi#l&^@22OyJV9Qtfg&~_6)YCf$qgyY_XlUSV?ql4
ztW7g6ht~q`MkDcj(}jbDv2DL<+_{u{e*{|sltIPUNxz<P8XCXP?HDxL&uu>uQek&~
zJ^T+nTHD_C1Oj~xa207wT?){?P0Bd;fFFAKqX_sW+lN8=xZVmOQ-gw3$|l$Tkb{Sy
z09$s2*rr$lvs0SLdeOF@7<)Vb?@oJIOjh@(76{*KnzV<fYM9V+v51Oc*9eQ_;QUqQ
z*=^6aOx_=+Zt>OKt?6DYn=4n6*nm!d0^#|%uj)IigQo?{h5mFL{O&4*gks>S;}8iZ
zc%*}A`b%ET<wI`6g{AhhK-&MPCxaF3QQX?`5%rO#q_P=1VhOeyu;>D=0k%qTz?L}A
z%opk6je_DS8xrj>@s-u_9b?|z9b(hrK8(47LvJ{#!4TuGu7U4c0J*v*kusfvlFcU!
ze;_ZmeT<y^3TJ*7aK~4NcH=(>7=`S9XM0RXb3<?Al)h#j2?Z8FqE!Wn%T>|s8sAEs
zh#+_wsq=JXC9W=3hvxV-Zrm1j^8k^l0_$xNI!=@M%Yfxh9&-bUQBLFtptDLke85(>
zRowmXmV3;#{o;<0cz~qfN?@uFrbNvhM!Y&lIEc0P%<n&Nj*V!r;92*pjvVsHOlmoJ
zVmiUKSBq_a-or{KLX;*Rul?7;Pn2ST$E({9HP=eN)n_g!#g;628O1dfP0v3}Bvh~L
zZtdp5Brb43x_9;3Tf(2}KFAYK&1*6$&+EB&wPoiIFZ|%>r9ZT59+Gt3%Dh}^>SPoD
zdhQOpXddjxV7|dr!vi2ylrBeL>n&((N%nJRB?rTg(ieTAI=nG#mC@}w#!sKoj1)SV
z>Ygle?iIDBCxOV%3e0CRg(3PAXanaqOyZ7K@h#8Qr}nQV088?GFGqTh83Z9LyJld+
zD&>{WlKIb!Ab!Z%ChnfmGxCGyU=T*=Ilo!Z^rD<bKX)Zsxz7}%!$#!Bk&R!hb_9sG
zHmLLX4&S)QWH(*S<!VZ1TjzI<Yq+6sGAlfTZNlrc&y+_pE#|_UWDHmy$en5i4NqXM
z@DNxb?{{ynKkhOLKj6JCxzFj&+GIX}>!3b@_G(Xa=swpnnN2Wbd*jtU2RLbR^@}N>
z=K#a-TnmN2@%v@V&y4y`x6iguv=l*~VloP8T~L#{KUy5)-z>li7+?Vv-gli44*w@I
zAvME_<A#Liv3*RJj^<5Ma$?&r4WxuXHncCoSgW4{-<zv%b-(jCBd}K@tocmwT8(%D
z0F|Xf%XD4xghwwfR4mJt24JtFVE(cuU%tkvwuvhB-T8j;V8rQhbW?>*l;ooVFp-jm
z_$#Y&SwNJ?dvW!rsBa7Ea2JUq&!?Hi-};ayfRuK4%h{nU@j%)^X@cl<FDAsle_}Zp
z;C8Qtw}s@3C_VBk=e8|Cp8AUb;uso#j8?!IA&u%aOVP})scNhwFD}@w-+i5xYzL!p
z$Xl&JN^{k4)H7<o!r!Q3wc){4pqJ*sZ630r(ST_G<Cv@o_6I-W{iGlZsheL)HBU;x
z?3Wa4XhE<lp}g@-@qzTe$1SCG(=px=5Q4tG^m-WWC4dYH2g(TQ65iDIu^o^6hG4w(
z#Vh86r3H_N&h+vtu^lwQ0H?2J?SEKq%O9s2uAqa|dyRy{`+g+2?-lSVNF^L#92JRt
zsRtMq1IB#PM7$vyMEjI4xzJ=f)%B}Z7-V<xw385SRyulI_|#1yMM8Ga9kAqj8}E(p
z!ivZ>tC&m-ak)V}`NdnAPn@#{Sq5I$w&ElY&9<I~WDB%?M89HX8outqxmUDA<|SN(
zbz_a>W6}baPPoi<2g(!qdS@?ehLNsha22U_?*i^wT^NwD;dQRZ?-t+<J{V0=CY1g0
zKla{nA#42pG8y|o1M>cUlWg>ynwqrlGfB6LIh^((qZOE<vd4G9`}5=tjL})gop#n9
zZb${(xAD5%BU>=-=e407RnfP~bviQtNbN~P%9{@#E#q?ZY_)p|fY2hzwe~jxe@1V2
zS~ReQi@-tu_ilO*yVTGkp(dZQTo4sKe%A;RArGS)W*lBeDg(~gAEqthkCxdFCk8Ac
z09a7bbx{^smJ)p}3!(wSCL!yo_XQ=!EY$olziZSdCPF9x_9d1uW`b*L?lHw=nS*CW
z<i>ymsSG|?pXsGJ$taHynjE~@F`s!_g1Cfbfju9%aD>T@qtMScnKlMypcBX5DB7%R
zy}=0tnUT;ENBlTn|8adR?&UXm5F7)cIK4(&jKQ&O5qu~dnKpxnw)gzDGc6R&*})q=
zbR@O#acAJI86!B|Y)j__MPTebo+>2j<SHx_ZI!cCu=7<oF4;Y_%Vma{RVQ(WPCf;x
zumB2Ff2bw~h?-@KP;8rycyBQE`4l8|#%A1&76-PCK`Bs{39%`zN5V`$3B3;qBBT^~
z|Hh)nPg5I1Y}~`78~{hS^v0)u63Rcut^;BYv#pD{W*id7b3ean_#j9Ts+vu!s-TR5
zlqiuTnZw?}MDYX~msXziZ^Rp4H<^dR(YUi(S12*!Tx8*RGrE%<?3`anJH0hq)OLN>
z9=L1ysPx)fC^w)qd31HGG2~|5Y0TJDPP>RM9XV+SS#<d%br38Dw>@PT<Tb{6wikQ)
z5b}Qw1i4k5e|v)@u{;lis6*%N<oCJ7Aq&-e){P=bpJSiBN}XHXlaQph3Pzhx!;-8V
zMsnHmHChdKxUCPI1UDGZVF-d5Z0zA#kd)JXZqiN-d@2tH8=^8q7s<mnXIfUXrUZiT
zq=runn#PTJ{)CebAS6E}n?<R7{^2WEd1ivb@75$R@kM6f#doA>!1yN~j8X2H566qp
z9s?={<3d3FXXYLX$g5d=&h>MK;{*9f1#&hRAd-mzr<gVh*nlhdP%Sh~;d@dEr*GCB
z6s$PvM|RD}cd9Z5Xv>Zt)*RJ2UY}A2_1D>sbPY<ur;62qg_%yxV$a&kO5Z6k1cf;K
z{@&paY;nh}7C9T|2Td^;mzkK0%qwcUQlwuj*|)&*Hif=7x2{>mdlbuLJ-JmyjTOm&
zS1S8s<?qdjH61(WGndXn-j|E-?ozttav}2rO*9AaQWa7xYtOPT4gg^dGEkX1LgPRj
zNj(UEx?2D_k(Jo;GmX&lPGmz=J%y>R4jT#?f_TA;Zh-6GB^V*zI>9=DN51wD1`O{I
z`fqGsb#d#)_uckKEmddMRgy0JwLe4jb=u?rf}j`foRPWo{LJO1PSx-s*QYOSjN6l<
z80u$oTSk-+BpyAqUMgU6!Q>Dq>)9ye#nImdVlqXd(dNQ1j*^&OO0DZDOGh!;>|1Sv
z4*1igSUB23^BByKGk3jAXFT7s@9Fc%ee+fBk>B9;iRS|+L@QFFq>mqL9+Jz>GhZ*u
z!A;HzVO^ulHi4`Pg6%OUWqHri5!z>Jj#Bp|XrlnpnT?V1pq;F;O`kwDBU6te2npBM
zje2Sm0M_vAl9+KxgMqBCNd-u=Jp7nWov%y~`6B+ni=1heA8T5b&e}U0BolS$01b(B
zP}4c+Wj|z>*|;r<R9-kLe-#}Wvu(UK<IOUMOuT^JIq!0S34UsJqU$Z{{Vc>BzIn)r
zL^~hs{)WM-3)M_G)`o28RXiwerk%?$+2&CVFEF%%RW&_bJoKt_bl6K+Z$k9JM5mW6
zU_g34eq~A}8>zmUeF__|KwPPkK%NS%bUy4+{=;?ZYO!QvSSc31Y%jkD1lIlV5KJ%{
zE9JD??Hqro`9Hq?GAxcJiWY^DAy|+>gTr76n&9qEkl+p>xH|-Q2A7}-9)i2O6Ce=W
z2bZ9OyWW1!`R=*jd9KWldV0F5yQ`;rs&?(Y_FBlG`^r&cDD|Cxo_OXyapm|}dY2s0
z0=vDj@cV!%ji>uZms3Lzp#N-LA3uEsL&KiWgA9<sI$=4l@ozug1qt6UPvjVw-3n<Y
zBJb`NF;mRdn@?;7a9K-)KC*Q+_8d-jZC`9jGwqo^33h5ZUsKbgqgf-?uJm+GBVXPS
zFFH+Mqh)=SLb6hUh%oBV^}(0F1X-!Y_gV5rk{e@*n3LTD=^S#*X&n9BKEe0);Jr_L
z54LFN1=HBW6ZrptK;N}*kNO&>S!$FU^wTv4kpAh93xhX2w8QQ!R;O=1=A`lXJWzWq
zkeQRX8In}vklv+8Q%BqQ5<;>aL7@Q$XG+EcZ(Y_M$k$>|K^qssr&9hVVR7$!qB%{2
zG~Z0Et(R<tKe!EzAthlPH6@ZB-D-;;&s!FtPr1<TDRaI2a75a7LX8*d@us@FuOxL%
zM$7P*tRmIQBVTh@+zt+ZAzkXRv<cz+T`bysC`(^;TnuE<e&W4#kd4M|ArmO7UQ!eM
z+|_`6-(!VbaBdZ?99e1qBC~sQx9<oU&|VTjYFhLiC=nyNE&GRxFCvX?l~ol^@vM)A
zOP_@(kUp9wA@zbK;1|u(++_SqKuEmYBy5bYhvWO6XNGk!+ci_$6D1*Htd|9Vi9l-D
zxd=0URL-m=d+N-lpSDRu<ee_S-glkDdY-g|0IuVK8=H)Ap-@ZPWD-UsqxePag{!(A
zy8f0<V933C>Ru%Znr`kzz3Ih3ZI_PslCLp6eKM57xJ}Wv!QIfAqJB~~z&rD2tgQE?
z=o$)AuRWJNVPzcZ*3X!8GcWG&1#wVoy)yxLNC@=JgCjf$fdtfUZ)11K>q0bm7LnM6
zk$rF{Ejmt3$&<8K@KaARe<I3c8ms+8mo1nNCfHpso|tp8MD5b^djWip9?3QF)O0CD
z9XaP#gv|d&|6%C_7u+1;8ArV))twD%34XnpQHn|##65eLSH#dli<Gh0MrW+ess>2?
zi~tNKZ3+K6MnLkNBd>BMD!~w*Hsu6Ofv$RyD>3Zf4mnl6nT{9&kJ&O`x!azomD}9M
z@2>dOsGsVxwD~a2?yxx_WbMa{JQLZ<LJ7!To4-!LXD91)=|Hq$9h3u7@1jIhlmNB!
zK_WU)85~8BI`_V@4KcyrRZmKHN4U6z#}w1bZ>S6pn-OOEiZP0!*z8CXS&<d3uO>>P
zAF2l=P`Ml+uI$WU3Jh4IuxwAib$zbKB$0xc?5~`JkncW7pVCGKR$ch_-PEV^`r@v2
z=Jl|51l1Nhyh#{z{{qr=-wLI8VTgSY_Es-6{%&$!Cl_(h`{R`v!*YF(iofgaI($MZ
z{m#Op4>pW{{9B)CJ9&Ak`{!44O8^>!#Z!T{;MycHzj0S;5vfs3*<7V5l7PLa*Q8ub
z(RL-laA6Pjy=@YIVM=_>-_xX7tlLr1V+cV<dcK8K&TGv7NVg9dr!|TvMr`{8DrBtW
z{Qx`mO58pPyBb(`cS$_zn1KxF^8@g$GG>m)MIznRO=DvWAVep>O8f*J*T5SF44RTv
zGB%{u_N$#uKZAI$_(Hi0(t@G0Bws(2Qc<ci66AQVm2+TCLI2>&bbf)i4|ZgSx&@4s
zSNR*s;spUNul3J({2~Acp|{{6*m1w2O1$B0s$#vN=&DpVl7L;3@=Rn_)+9c>+IMyU
zqt<$(YXB(;A_YA!PY4Fw1Oz~it^wZ_jb%>GTM0pt<Z&NT>HPU8X{L%y2fYB&2DDq@
zd0R5(nGH^dxgSJWC%Hxe^HsbqZ|3U7M>Pv>;~pWZoR`q1Ea|xTq)wh#5o7~8A52|<
ztw~GT<o8t+FreOjRnL5)jQ#x39!vNo!V?$2<1ig*DB)J78o=v}_x8#UA8H2NXT5Q~
zT-}oOb!e8mnMH>54K7JxKV>JiQ?MlY`HGv}I?%=63ZxQRYsCOd`gP{U^}|BI+{=1~
zj&lV7(4u>h3D2Zt;unzoe1>t$Wlud8xcDqN5Qn-<P3}}ty_g!%gSS3}h@Q*>sg<xZ
zZoMm`#$S?~eoYWCS8~BlIB)4rxMj3zM7vt)syYPeLuv{~r9t@J9%Nd3D^jeeGUX2{
z-n$gJhYGsp$zPb>Rc%A((^FcdD}Vq;2EcQQp)M{R*HoxwvySQ2FUv0ds^gGnolV~U
ze(1(qJ@_eLS_x^%J2i1*O+PjNz+4V{B0|e7Bs}R$s34zgH8<5+V)E7U^K<M!f_0Cv
zcXvmx0Uy{2`^>8(3B9T+61@LVR)h1~RhhI_Wc-qqZ9`{Qn9RgbiYU73OSAf*ly8`q
zTmk)-<K+gQYk$gJ0?b(xh5;RmEvI(ww|w7*fQxAF(N)vk`o<Su;k|ma9<6vd%9wv%
z+6}HeN6FEzk4hGK5vK2z(oNj2v(%FB8Qgu2eEH>A$eM`$-BTWuMcbZtLL-OfcwVEI
zilqaBg|}*@+?fx!`l(o^{##f0j)N1AIbEb*EqsSKjdx7uoMK~q)bRm16sq~JhIA@~
zb=pDbJQe;@Z(UhAjX-b9i4tWTi#$YkWX%97sYH0&yYi{rq9i%QE*<L;PbQ-((4T-p
z|4D6YAe9vLDy&`z{ja5MN9slHg(eA#zqWk$eRGUln4jtw)Hb932k3m+dWi9NQUC>$
z^W_>^wE*51!8r#fL~<wc=(wJK0)h7cbJX8od-m~ya2qncI}v542lo|=t|am1sJfTJ
z!-bBN-$w}ru~c*0cB~iSB4W1iK=&iNKYo`LQz@STf7}k1SvJCyrCuDgmL4i43evCh
zd%yUpAcRq|e{fd|bMmGLc<2SvwI}~B&~(&USs))E!%i<lONBQ?ynvQyXZ%0!KEw5{
zUl`f<sV2cuL#JSgL==usNf_lfsYO-o#%lOIMpkNJPy(n}15!tWR@t<6RYf5yj$=_z
z^7Peo&YEUd)qn7u_tvHz0>S~DPe^VB1o9^Gj}liYCGoETm@uM^-CP;`qKp-(wF}5u
z=t8Tjmf`?@kO@`)M7jZX2Oz;vqH&SkJO_mx$yHkFl~sQi2nb4HFN=1OZ#X=fr_Z(c
zqa6H|GSMoOL%}-N{(jQK#IYBfDtV|bnuAD#w~=}`yU83P#o!gc7GrdqO(b^ws28^I
zQ1xU*#kI$A4;@#;UtAZxZ9o?644Cjd2OGQIK*s}dhedu$Zts@;qy%D*FR8$-3WKVN
zy^A~ZwSm`v?MqO~VNFo|k#k^C&EQT^dPTOs$#}<I&-3V&)F)$b0)a3E^vk8t2IvOK
zJr5%4{nK20TOZ_$-C?(OwGD3m;2nm|a5yW~`$sb=wZq|Y6tQKR%x%*PWSZQeIpOK8
zf2mj!{w4M#WN)PG=cUSfky;gLfVx3bXDZ+a5TESCRI>}kDR@C6^yQ{47ti4bd?D|8
z#5PeIN1=!wC(r_6d6g?dLEE9^AYMR<>7SIZVFTJnS6p4Ew03VL>O16p4vtHIqo2O(
zD*@9!9Y+-0!S&<~Azd~zh-WBZKG7)?Kb{rul=xFSxU9;Gz%8|`{yHtCy=179Ka~~t
z1@|Vhgy++1Q9k<k+bllGh!{jL-&OoA@4XFSB+G&(=c@_H=N}fx^6QcriEafXpIyEU
zU3cgKgW{Bi-|wO!5Y>E{Z)lM+nQ0y;MYRN@dkG#5pvvPG0G^PP>YmHY;D9n%z$3-N
zw^=5bzjkPO0=iC<=eoJ$C1{u3_$|5iABS;oDS>VhYU_itE6gX>3N;kH1S_U(nzjVH
z%C8b5Jp4qVY*mgSU7mi#`Dl<%*qvUUB=+wpu)|W?+wC1#)WjF1ZPfF7O_zJm4!jvB
zsKJDW_Vc=o@z>j^tN^4gE|nokT@-JNXb@DhN`g{sg>K#2;8&ofL&`r}LTR?!H7$CT
zr>Vy97p=Tb|9u8TR+9iiA<PHcP6zGndv7s{yORs%Cdt#MGAX=;kAkmG^^ZXmJyIA}
z34-ox4WHp3lT0o5rr)|eKHdX32N=DW$Qm2LCm22J{l|!`W_R-*rsVyH^vH=y@7bhk
zk%pi(mwBqaKW)7lnX=TY&H8~qJepp8&1&m0$ae1p5sloDwogiOPP%<2G~Q*J>bY?3
z4YZm<$M;=jBp-S&@<H6-U;n>D2`YfH2S4C)#<^Qr$a&gQ?UmtC0s76d{GbYzAtNsk
zj0`VfUYaq>p|n;x1?`|dH_?8@TBH4{L?`M0STf)VC(5&0o74?Fu}0zA{CS9)o%@M9
z5pAY(nO!VK8dPowUU0mL?#x#RJ!&C?95s>Q^{+bKip$JkX{<Q+g=dzN73C<=1?R}W
zACWBxyzDCJ$#Tbc+aXx|4<YEL3-X=h4WLeO97~9B;WXc0evN-hnMaY7_hqkjC*+4b
zXcMyu1YKB8dIJ2xdm>M0UKb9Fon(<ih*VMk*+5#woUMcm_W*I`D}tcWP~I;D&-x+g
zpCIq?;eceNBN;9U9wOD5{WJ{o!ACk&%*11Zc6F_7Em^&-@Sz2|-(|lRDY<?B9gvP#
z=hjTz+<Fu51=9tc%ijFZ!ElWD*I=(}u_{6I?o#H<MMm_4BJLBg82`BVv=rZaB0so0
zqo@=HWateK_5-Ge%CU$IJx9llqS@WS{!_{keUjV-5FOm8E0-=j2`U=dy`YSdZ4YC?
zvofClSEM$=H(Kw4Cs$oYD)k^07cUkmHWvKx@1A*;+}V8)+Q)j_yB_rjoO4yPDQw4<
zW8BZF!E)Cks6af?CNapMBm5j%b+k&O9WZ^YYQnQhyNbh%l$9oJ6(3&fdmJ@%t+$kl
zd(l)VhxnC($&J#d-;xvhaf!zLp%p1{tL%@571dDIL@t1F>b#dJ_-ef5|5wl?z7e?$
zw`<R@6fyvN^U^;Oaw{%%-!OOnoJD~=TJ!aoVolbfO9FwNb&L5LMa>U;n%2qo_k^GQ
zTHFUO!;f@d0f&agQzX~P2mHIQw%TS>>ayv7fDCbDL5B22YP^j?KhU*!l!*E@IOl}j
zr33ZL$oC6?hJV|3N1CZ)^0miS9Q|Y*`{FJORLpTuw%t(`be=EGP@%FYawrFKiT5m@
z^gPBNt}nB}tac&Xd;`Jf+FZzQTB1BlnSi{P6$>rUxzGR4HoeV2ZsV1|z4uJ$JvCGZ
zh#$C@<u?aTRtqFi*YJz}%Qe(GSDxAUNO!xB3R%240>Aq&H#y>gi@sPMoB@P?0NtqF
z-W#G{a=)&IRRfs9Hjzn^3Qlm}<u1}w*e{??*{o9}!<kvA&RUEGhpj@4_3D7+2>_g+
zyN${tGlhcophOy#MpHs)98XjGUv5Hh=-)NeCW~jt(cvTg#K1Y~O1&p`PF0{^zon88
z60bUTkL<CO$!Z<|bc>ShR!RJFx+oUk-BICw;V+U%HFQWWSZ<&^;L`}O({b+1_-VT*
z!kODV$C{VaQZ^{e_LS?_gtYG+2Rx1OgV1qbLbFZoniu`uhfY2IpmKaj!#o!Y^)LlY
zTvO4{=w(kbIf(nKsCN>1osOYH06<Q#Nvx<QQJYDfVbPz{WZ$RohWTU{?Rt%Nlh*Sg
zHtgb1z4Ms!vsCVwCdk|J2+yQ52JJ3Xq4;tpuP;lp7o=Yo3eXdP>|*e`0DP^TZ0%NS
zewW1QV(6DRvU>XU-)1A#c0G^5kTyA<8T``dB(%PlV~Sb3PqpLThhv%<UpaqsQnfwD
z$<;E4M~S`2B--qTDMxNM5-ZceKKVe>45_rX8-Kc4%%vdOGJZ`Ma8alxb1n5?yB0!^
zd@oQEbd&^S_XR=Wu32h|c3q=1)nwe+)Op*PRs)t$bNrU01qfI`QrS`K;d%ILk8m31
zj4SS$M`nUt!%x8TlKHdc^^z{iW|d_n5O1V1l@+8!O8srX`7|W}XF{!OHhJilN`k|I
z(^zc)!UbUPX;PFM-*7aoe6d$C@>?W#M(H65eYf*P=1-s_UwSb|p+t*$YJpIHl|V0-
zL_|_^^y#rFqQ%Alhg7tNASs3g0ZjL-8M1am_e0$3=oAp4s}!Cb8NGYLmLfKFyh0<&
zq4T@3xC!E*kFc+q1@8s8VP~ii;sry64mckGAKe$GaJW>Xq4dyd>b}?XLn=`p#6Frn
zAdAtz;vGVr_2iRPx#sd!aq^hOPZU$3*xSYr$`I@~(&Q%4iR9Kuz%TQ7FWiHrrWA`^
zz5(%wi<JYh8^odrNpi2|0%qe_soLHVML8G&tECX5t)>t?70C8pvRU$IA4&Fge7Kb!
zI#%P<n9eTBqQqrIzRli*A5cIhs`Nrx#u~)>|CTSsomvx?CXA2JCp5F6f{h@U4uvC<
z#^6L#*}Z6u(_X{p+OorOW>u~}LN@Ql_uUZJFf7k%v}H`~7?032ITW&OLdOY}s)e3f
zYNTCQiqL{;W%5z;cgdpzCP<TEkhodTSSyFyFQQ?O0h~niw&y=&<$P($o)635nhQlK
z0jp%~j&+OH1Eo5{JjgkX$v_L!evc9$>1O?x8~O$}XclIG`ol)BN1Nx2E_#2ik$Esm
z;}n&5x#;g2F8`D-s#5;f;-LUEDY1GnOGiw^dLFP~8ZcXhFZaSf|3kV%)Q_oF(GT?w
zfvDDLZ1<|zI$bvs^ms0)g`?%CItY$A@K|pG8l0j!pz}I94@qK@#MeZBwJ;WV^h(xw
zZFg$h+Mrx*`I9v3BVa`I+LX_-^wDDU4U0C_oYvTg0bk0{?NpR3Q=WvIMsR<4J_Lg9
zbE4c%uUE2<P$5pdtNuYvCuSZnpABr)L(PJ~OfN-|z#A#c!CZjOTdfn#O@T<)M7Z-;
z^HR}57NlQzWe|$99Zt5ToKPolLJrq4Of=KZ#ee7YbFC$qk_CxlAj=V&kbI2TI?5VG
z=C4vxzxfLdkZc&K+~++}AbnAK!o~^m^S;0DJxOH17A#|f>NX#~Vz%<6<T^WM%B)^7
z^zg$-#NO2x9dHV=OTYm{!*e_WV7dViOg9E_!ksr1R}v-}+&o6fq|?=^jN#MhV`Z{<
z95iMG=z0r(OwjPr$wj@XrZH`ml|7`MI)Hze$o!3=bwEkBP$tzKH$!WUC+q<IeDF@J
z->t|WaNPH!U%7=CmKPyTatfm+$UeN|#ViP>Z{AgM)oVALHlUlROOe~s%Eil*3%|`W
z63`*?)^!@W#OToc=Y)OPvQcViD*-$*GHTjb{PfpmA*r10sKd&j?S!E0*$sBV9P2n-
zu_ZoGHmejgf&#JiTlBJrCTXJmJXKcF2c&k!R|`e762J{3eVT(@`fS!8=8h+FEI6R*
zVW}*Bev{=J#KY}^EV@2qmPe6c2ap>jWGkQP`8dXZ=|`Q~a3D7qbNhzol_neH&jKOa
z^pRe#C+8N^8ZDksaz98#u9|hDq|4hUHN*LcswpEaa9db7hU>Fjh|S@}bFxt>8gzTz
zn(ZGF?-DMRLzF8EMbr45%_&g*@9&%LDZVd3bqXM}a}vhUYjnBbwQo^|8J1(qR*e_+
zhx||$?}MZsZ(p=gmElXfU%}EAM8%nOQ#3ZFsfgZNEck92y0lVm<h!Ub;;dW_g!c<j
z16*3p!%$A=Ig~|Q2(>gj@w!jy2(>!PXsSVxYP*te2*49SxxbTl5-k_e97_n%LC~fa
z{f)5=OVN;4!DSfoWoAhj2IPbqdtXI*?eBIM4S*P@q6ja`2_2zFx-}s@^roI1xFq!w
zQtLN#IAC06aX*(tP-kNbhSyiE)+1mvp%BVu7CdFEI<gMimbmyvws%$L#b*d;fy)|)
zm4nM<NGH96Lc0!vi|64&<sop0oHdMo!4;Cw(P$X5xQcD_aU4$8Kj^py$Iye+%QP|c
ztcsHDgit^-X#@?{2Uhd?4pu%(ygI_3KvXdc!TCCfwpv5M-c05h>Td$!Ew=?HP^%DP
zq)%`6{HSj@gnyF&as8PNAj@Xu9#UK2Uq~60N-_ji)kuzEXnP!j>h-6ommn3@+`8%^
ztPJ<5Oz~-$dGM3HHfjyHO{UuF7(W!zY(k*8FD_AcOgygJg%ZLlWfBH8(}xhfYlO>h
z#?|>H%i`9$f|VoR`doDq#B<1MfS72ZVuZwr7KfN4@^KMs-wOSHM6$FoT+nzN6@Z8>
z=Ph)K`s!PV=j6%5*S<x={d9}|Ao{7)T@XuA*ZH&{byQ1X_{D=zAt_p>h>mwIgtdZF
zY;Sl^+6{k*(=ASlK@v__TwtDC=ORm56Yf(PN3(~1b&j0iYHt>*T*)@H%UHsUMtVRi
z(axtfH`?HO<?YmwpiTiS_1=BSsrwCGaS@Z6{hbo%IBBP@PIv&aTKRz*4?c7xdDQl)
zU_rzz5mw3P4rSGj6!QeM1jy2c>bd<B&3B7gjgwesp&D&S334G9OKH|4sfU+qjh@w2
z1D#pY)I5*4Z5Hupd-aExN8!$8m~&Q6gbsD<4uUyaX5D)ris$P$pr)h8Fv5Dc(y(~J
zVhuB-$=;J6bUeM@RzU;I+v^p%ZIu2hZ=I&Y5x$;8EHsLP0lZaEoB{-Pol}V`FM$p|
zcppheHxRdIROA%a5z;6BLT4)L1||3&Hm#5hwbv(Zu`|BHAUP6@m{^+pC4aK-5xOma
zlDJWVsUc55LhT&J0WW2Y=;kNI&|uNYvuK`_WTkEkU9w}91sx2P?{x!{7U4(i;cJDM
zqH_EM3s@D-{`Fr1!zNcXt;+<@w=UClcUYF1uW_4bklg=CF$iB-TVrp_OxcfT4TZZh
zGVHl9x%3PZtB2srgRZSdSC2_)@Yb`Z7foUK+AOPc+|s#e`vNIo6C%3xzPc1xTooRP
zliJ>!0TpEZzXDhy8#gm%hkJ)_BdYP3-G=cN-#ZM`CzMI8w<{kvK3zpDWD{Uvxk(TN
zF*!1a!5Xci++dCVXjaMZJ_-iA|Jt)_+xfuPXT<k3`T4bJBeQ#?8*I%BYLDVab8RG-
zNH?3DOICw+DTcS-xZ;vkU~$U@H}SFjm3gG+15phE;BtDA2@W1;qaF?(moaipW$(&0
z$oDu(#Oj}%*?}u3VsdFK1!tYKPsz_9Pub~T3S>=x{%6zVx}fcWSPovCTzQK``<-G2
zi`e>7cy;k>a>T+j3K;5x_s%KYOy&Eh)b|!AIpnDn=GEt5<Uqhg1Ve?1Xqe-YJttOO
z?M#|RYo?*UKU?BgWadiF2&~L)q*?5>vD9wFjMFaw!MDk{DL@3HK2k2CBcZ`7QVb_5
zrL3;45-J5_WK-+DPG-I&)E!8?{zR_vOLu|FuwZF&9748zo1bMqu5yM*H?&y?7bOhx
zjnOR7?8D!)T}PJy9V(!snCXB8wQg{m0~D-03I=FA42p)T0V+?Pg(HlsF#HUzS9pmg
z6b0s%RTW9Jav>1vJOb;s_eCu|a08BEQ+ga&bcY*vuz8-`>lzgua5YMD?Ous3!eZCn
ztAZOlK33`OzE7WE701E)g{=;|xB!JFgXYP`5n+uWK@<p9_Ev2}sMsG?^R+De_5Cu5
zjaRTL{2&Q}$r8)`Dq*b@8^h-+yWR6m6Iuh4^moJYV_d$^wxv!#=Vy9cwWpDMs?d(g
zzvD09msN76Py0?PjQ+3)@<xK%O(aN*iB@oTN75mkixURx7844+Yw&2cwhb#jbLY3z
z9>SG3;qM@qii3V^=9WHs+wcC#{&S+upjs8PnT{awPH&`!;FP0bk^o%O9V!fin6!0l
zR&yrLr;rb!Y6t)61c2Ac+N4hR(F-JhAeew;y3a%cdt~V)<=$zFGO?QuDt#|HGtzTd
znL#eHQJ7G6+Vwjj{n&cq1AFZVoum+xWl|gOzo-d!Qr#Bw5(rQAVOFv&kz9OFH~?Cv
zm7=Ro&yAX@{>_n>NFrKUlk||hF6$(qBy{7U7~RPzK07#D&<ShXcEf5K&q~GA>HBev
z_oivNdu_p`>1T`dtEX(dqqkm(Xh|!>)Dhr{)iG5OP<=_EAm;B}nI(GeUinF;OG6?W
z%nG!>)?qQciP&Wa9_ZKOaJIJAPPsZxlzCRAasOZ#Z$hZldU@;9t61aE&WMAFsMmt4
zYGzBbG{CE|C2fMbCh+cP)<+h{a)^St{8mt6e$Jf{_54an*jkF?uLo1L%86T$mQxYR
z&2koU|IjiC#6@~&>npLPb~0|HhY^sAx>xtZ=1-Iz)Ubtl=86ww7Ilobu-589lt@-?
z2f+<npa9<E6-n$iIYhy)H-vPL-+tMTXa4&OjMKGZq(@U^a6qVgzeqSCZdFAP72=Uv
zBFFmPVFQUw0f*u^=i3F=8a}eHc+5J*;fJ8`<u8qG2vG*PoL39*kEDqxyt!yc>y>%o
zjB+TZ%F^8jtN53Jgo}Xr?2450iJelYtOJx_=p6)12V&=`$B2x^><Iw5eUJ78eMC1#
zNt&UdITorgzjP@D75fzRl36oinT=QRJ7zrikwhYr!2K!y_U%R<@*!zYpS-g`EbNOg
zk_Ig}@AeRkynrJUnoSBJvCuC<yd{>gTh}E-{ILR6HU4N8#KN=rO63`lVh4tt;RFJ)
z6B6nN9DBTndh(Wu<{g_|fC`}A`y$`?JMV43tg{C_S>o%|-!%NIBJX6BF0HwHlmrg~
zFT6F1O<M{T*Db#v2Y0-u7-$`*{R8VKhRzD#ja|QiF02gW1X*oF?)E7DWTQ^Zrk3(P
za0(%G3&+Nre?6geg$q^9s99)uEk$aD!nw*-1U|T_%pi}o${!lYz{q6d{FIr6)?Ya~
z(}T<n`~e@|H9IIGa;{N>$A#AGGY4P|xbyFH?Q;m~=%4{VW|}r}6I3lqOfIrD0DRAq
zyR#IJ*H;UA(F(LIf|NO!{I8lpCD6a!ZBrO+>vHG<VFi*P?=|kk;k@%C;v#NOYdbP?
z1Dc${h)<hokX*24LjmU^v)R}0a6%3#!O97EsXhdG7*zIv0gxjSBKJpEq;D+iIxk)g
zNDUp=CWX{_(&L;7GK(wya<47gpJvFMhj6)CPb-XKL+YfhI?SVPMDv8J@m!G|2bdCu
z-%Y_O<st%`d{Yo>qtT>#%Q<>_>;DAvRciL3{8UHQaVK`CA`H4`gYe=vhcs~!`B2!Y
z082;w7(fxzfLD*pQ?6R&#@Y)FbT1hFx_Qzw5aWlfEY8h!9QeqTQ4!WTDSadWgQ$Ko
z2)#_o5?au>K`#SIXqB*bfc|kAn$RyqZh97w1;ee=5Cp=}_AK;W(8^1ZvX-SV7LGhh
zOtzNHgMF!y1bz>%g&XTp==giYIys)`!N-;%#F?6HJW!4W!d7F)-C+9*qDT+3Mznr^
z3R6#D!nn==CO||25|7bwk5$+;GT@eQ0@Rkm)CKyEFFFk0UdPfPIeS1<v25Gk*qxly
zG@9#!n6#n|Jo^8>yZt;kniWiS{gpKdUZ-8K*h!vfz9y<?H<pScl@HIs-<Jz#-8lIs
zj{Q@zH3m(p0cy|~pbz3ek*3C10@eCr*IaPpWSsgORJSokeSDx>Tjht1F(%pP9AehU
z4*Eb*2!p&4AtQ6E@#`j*by0Nsw)lu0pUf8UqR?&?8|NKD8KJ{tDa6B#Yj!!lZjip~
z90lBSuDN#$ofEWLi=Q+g194?w>qZUQDSHwvGgxOIT(O=uUn*BSL#~E(4>Ca|SpZuM
zgN$_UHdal>u@IIx{f=J1+r}P)rQ#);T9QLQ>I)JCL5n5rqW@tpxm`D-SkIn<EDr+3
zmVb-b8z1pwJr@>TH?e}LOdq&C(I`D4uk$PRBgHxx6V3I1UBA1s8TWwVvNP1$Q|-_+
zk_SNT*Cr=V?8rLh8h4DqdxcCci*ry47bJeE{<u;=<EgX+GO?>{)(bC}Tx{f1<VKFJ
zlS8tmO*Y?!AFK9AFuTAg0f^K!Bhn`y!B(<tz5diEoEmq4_M0ONO-|)g+WSPS2N$VY
ziaKgFJ3m0&idz7?#vU88ZoKe;4~E*PGWgY}LpqY+`Z-a}Ue*hqE|hupK|H*6hyv;c
z8>v&FdFwMj_<FG%g6|e1x03HEWGy1*0E_SZqU1DU5?+huHn5o9#^TavNM+l^%JERY
zQ5_JEHuPuMusY;h8$1pjGBGIu@+K!oTKWtbX_U)Txj^~HwI4C%BVaW&ELriK)oTvp
zkrId%f8*p>^#1czVYNNg3ysIWIpq(%V+u8kS-8@TS+Py(rPTZdTw+B^QxX8%+78xv
z5R{^!Nj*qTG|^l(?nr~e%A{m7s;Z~)BGRA{V2W#{jRY3j((7!)*!K{q)I>n7Vk5T-
zsB{2OeQ&HryA%TJHc~lRmQY=fnv25qp72)5_hRpooe??U0&kWelysHl6$oD_92YS8
zJtl5l$(^SE@rg4h9An=bD6ck_L}~C1yV~ToDC`w{kaG*)&V=jOfDHsmypNz<)mRpA
z?{v5t+)6ePHl|8M;-R%TC)sD1)S&WN>M*oT3dDjrzz_RUr!w#KkVa`5UyQbZq|Ct8
zfoeJ-&18i#z?}}?Or`K_bH5Cp{pppIt&po;9%?ebi{H}Cq$$xvdl|2YRBp^@LTW+7
zkT9pn;vp>z-eYZ13Fn9r?{B~-)^U#NXT1G2r4lI?KWXR^I$gVMXE_|nmGD{mXyMeb
z`1dYbEBD4UliM3U5Z$b#2@yb$N)&V-+16@t!-U^rKi)J{|8&)M51Okw$WtpA3j9#)
zA9VBU73$JwEL=oqXcj6%(nfn_-oj!Mgbr(NV*PbV9?Y2=GIITi5-F((Z6VtxFuSBp
z#&TX^l@Xs*2EZq}(pY<hp1s4sN)03B9g^eLTqxoex70{fW){H`JdiUBX=FClmRaDI
z6+Az{h25|vi7^+GFnysio&Ss%)cOtbZO(nwgJn;|UfQH1Me3nw*@<Cp%Gxmvg{P7!
z2vej&OOU~qijzO_Rp*Kx<FZBC$a+nSLEYBY%}z)bmNVKy85>m(v&)^?U~rU}Wrd{J
zKm4GJ^>gC#NoFxP33%w4prwb+L*65(23s}H!+3J>!*_r$n<hy-xP21^uK;aGPTS6-
zu~QCJ6rPf68V7Cq5`8`VIl}p8=>hP5n;P>6TF{ve@<nTSAH2Q`d|<-%ebh#$<_E%N
z_0B06M%iUCSYK3$GTF4T>Lut3dT#c_0*fkD6*bKI4uo6Iuhn@`vlyh9mNEZ|Kk<TV
zO}KdB0=f{g-xYUOz;iQ$Vp>rGVufOcnJY-!j?n}O4*JEp>|FI&QQh9Qldz0TAnisS
z)QTQSAegEYO6Lo1y9Tf{Pt%*Bb)~<M@X{^`+rm4CUOniKxb-$JA7u)HYt^BFmopr?
zi+bT>`)!0#Rd@ie46Y7p*kzd3t4!YtH8YhmMAKS5diawEp~oD2D2X1)s0`*VC@W^x
z8|!Z?2G78+c|eV(OQ3cX!f{yJzuB{?A1{Te0M4W}nqUU!2vb=(K)8Eo`8(fzHllwF
z8F%%M(hmny9_z6k>PTx5XjUkm!DDfdzm-~zDmPx#8Dmo51lf}$_^r|Z7-Pk#-X<z>
zYG-AQ&`PBKZZo15UnOMPh+=_?;MZ*BdwNpZ^{=2*YhS5vFEWUTY3SUbXIDI+TWa25
z3h@Xh6wXVA4kb>jQ)COghN8^Ig*3-f4l}bOY4uTdP<-%1+ba@&)tu$?ky`kwS#0xQ
zAF1tpU)-UJ7YrRfKyOVx%BJY=^RVZC2!->9V5xzQA|W1)1#)#hQ&6$LYT&2{_}AdB
z?MvB+26ZGacAw$;N=&nWC^ynW4tV*m73S>@Dx`alDs(%E4j$+`K-f~ZW!`*O488&J
zazed-8v(<G^F3lU)^DJC@yWb>pHEAqgG#$POGz;tY+_FlxqX<JYg>jkvqGGKyTik=
zEd-@o=#lh^Tb(vZi}9752}N6Al|448BBh75EI^2dwtp_?+QyMmW0>uU&~u1!evgCb
z6dkUE(?tTe-wWTl{6il`;Shhx=m|2m2uHANN85Yz&1M{GRvmvZ_p0Oq*Cz}b*ySH?
zJ^yZz)4LiWyK>{y8r*8uf%~ROV?*k@b+EB(q4}<n16-{-I#SG_8(5r2(c;lyI9m*j
z<Ix0cjq<r>w2yE~K79fl9hPw2+_g5j(Bf)cw(ns`87mNcQ+*X2m-LpUZjL7M71#)O
zf@*48PY5%T=bX#RSdt*f;jjCY8nnCBNWk!j(B*qoPthI5ncRqHGg+FkQ&JTIVKxq;
z^%EDk&hmi^9O8_rYWRYCN>Pd6-e`>Wn7y|zoHV>~)Bk!7GU~9#%-x)&eSqs+H`U0Q
zcVkU#l5=*SWdh^NYT39+e*HYH$(Asu9rR2OS*c_K!WUiGh+U(l*mfR8V*E$H^}3B~
z53t5P@UEZ<+Wu{?bafuM&T1p$O(gk;<*yBEg{8+F*Tj{V{R2B}P=N(ZYO`d4IH(_6
zbw|%EF#g;u0k%7auF|AEn_7>(hL3uS;gN~!$&6$<t4uO?g)T{_2KS$ZOzgX#aERsL
z2fdCmB+|+Byy6l_F*|?aKe(b2&5Q4P+HErK^7DE=!%@k3X;|y>dESH<dB#LEt%-E_
z-)`P`KeZpi=hsA@YP$SyG6EiMpB{<^yXe;*FX5KAv)Vb`gH1S&t4Bo<Cn6zfITzQ5
z+Et?d>G5k_j}L!OsIvWUuUrG}q8Tq{wd2=bV8)_{ZG)cU@hUIpm(N8wIeR%dlp>GE
zD`kVm0ih~6um3rA`9B@str>ScJ>C!IbUyvNXj>H)dV1vDEAp&*zEDwN-?}_JeY`u|
zrJ`CB$v(TgiSQKlrJOy<>0Elbvaas(xw!o6x$Efva=OuXm?nC6jr#I~eXad@W6<*D
z*YiX9ONY1D{nki8=M&9iZ!ach&ca9lWv5`*$AG61QGY*HUw^pcGn>CZ>+`+d{iELf
z6+kQB@OWsub78yFe!VqLMVYhO?te45=6g3EkUifO@bd5PgQ#fd{k-wri_iWvb-VWq
z)k)4tPWD>-jiacLu>a$2#Qd7@OF=;UM7*eouh&eu=;P%--RP62B4VSb^IqbB`};HY
z@vgh`v)RL4_O9F8jjOUIoPf#K&uU`#X<hyorxy=7*)PYAID!F>8z-V0vwasKtIr&X
z0XP5BuucQArQ@G(S8Q459R2+t;19exyEjxKPvyK-q9Oqow;LP#8+T9Bme(rmql`~`
zC;ml#g#jJyl)pVZZ+Es&@2dC$dS9|CkF&Tks+Qfh$}|k0N-J#r8h&Hc6|nYjyo|HM
zF1NNW>gJ{f*Pg2ofX~YdGJ>iiGKc`-6#^RYodzMR&@}xB7XcwI69EAS*yHKwV#e<3
z_SMeZmCeiE?#K}4vfT*wQ_*}5;#L{%({r7b(@!;QEpOLPG}E2>9nXdxIoYj<hi|0C
z{F2z!CrY$DL{o&Gq$F`pA6ZsQisy&mabxWkLhSnlPkZkq@{KQ#MSw)9lQf<j@9SvM
z<ks-sR^QH#pNZcJ9JbpKqlbp~GU$FfcHKR!b^3e%C>TeSC&$L6ca%IatCydK!$Ly(
zVF?pTou>MDXWwjzA{_n%{EkP>BMbP`=qN<wu##M9B{sbqvhU@~H}&3`n4r-}J>K9w
zJEnZH!d*x@h&)kxINaeIIlKgO2|xPDepD!l?lHelY$J><79)L_93gBbTDNy@U;EMY
z5=uuvoD;KS{J9!gXC0vsVf{S~A2ckSPcB>X+<ZH96^0WJj$YC1PQ0b(_tl64QTp{&
z>~o$3m#?tM;h%AK#$KsBDL%Un=W{t{ueL_Y?ho%RV`+bImwiNY{&vZgcRaiYcnz2x
zhQ7HDO7P+_@)mtkvp=S<-U!-*IQ^9Mf373TjF~(8G1A85<!mMG7q|D@@mM8nJ&5w>
zeG}KY{h4CJQrpL4uH~@f6vm4>Grxg$mHsZ3yU1^oI&K3v{*O*%>O`)N!Xy+_haop~
z1t%pZ-->76FRA7{Hq%TFhTgVqi5e)xPooN@N<W?~x<E9k=$_lyDHkzbb}VUmw++|O
zg2d{__9|W9J?D_y<-2P!cF|N)*KjUIB=pIOe+-jSYhCglwL4idXV#Fdn?JC6)}vrG
zD*M?7Y~JZnP;_X?;{Br+%uiJ5d<*GyK(HZ;3yQTOM_!!Zd&QRL{8~eN>rHy#?+<(e
zq=V~&R^%Z+AWmxkGTN)RkLrR}*}ScDKG<DOi5M>I{qn1Pe>D{Y6XCw>v(G`l-)KaA
z_HTwecTzd{=nQU^7L49?qm%p!xofsyiPcfaO!g&-`zJ%LI49<D+J7l$%1msHC^H}?
zY>p#yVQ|;BB1tEY+K=<ep5G2$zW;^jV(RN&xBO)Ws2%EVu(s<(g!m=NqDGw7&sTuK
z7Vp~-Ee{2Z(n^eWQ%bkFcKwI`%_J#Nr0>YV2)F_i)|<D&vVxYRim6P|YBU;-6WcPk
z*+-}pey9-0XQ_*etEj{4Veruus|rFldJ2)~eO*NZY~baaaq2ToS*3CTa?S5P{?wiN
zajv58J#CeGSvbkEuk;gZ9)_%hOtLwmJ6Q0*wDz^Tzw~O|d3kzpXa~LfbYG^bQlQD<
zNaXji#}-t2%X{IR`^y9Wm*=_*lP}8533fYIQ_^F1n$vH|Dkm_$v$_y^rS562>q2yb
zcr#7sg_MZUmwW`TM*SKUY~`5~1ro1Uu{csDuq}W5z`m)bKxDJd)cGkcT#ka%3EnS%
zUlcpkM7DDp1g{RCwh?90rN|DifBz@&YtB*V#?81s-bL-vv`K*?5n%n?UgMrjr@u31
zR`+L8nu<I>>4pDRDlSi9!d)r4fAnv1g*%`cPuQ<d4aKdjcAy)Q@?CR4Z*b<d$B<!B
z9O`d>W3Q7asuSt;%`gYO+$NSkTU!PewWGylvl_xne(lhpdnmJh<iLo$OHJ_lE>CM?
zj-BH50vTui_q&s#9+qA#`%%pl{cIJhiD|#Oo0XC{$Lp~s0ZYC;pzQ<Up+uwHl?h6D
zy<NIlSxzLc(9FvF!9RZ{ZgiREsmx26f5CgORPdRc`qV+zN+QB&FAYvLgv}eo`ayN>
zHM>DS=w`g?8hEoSM*g1J8K0bPs%+H+4ji&n7j=_Qe1GLCX3JzIP9zigmpRpoh+SuM
zvC}A;jrNR%Wmvxb#PmR)GaOWBj*f0I@)~b66^kxi6Zv&OY7vO%F$mCvJ1z1iNdHBU
z3$hp)FIRp^Jw8wFw!6y!)$LdUtpPZgzCTZ!(~$SyZzewKAJL~$$5W``N?e~5dRe1a
zZWr;k+X97JiHPI~m+VL9Hdl~|N(MVv#iF<mx=m{<*6m2{WNT}AxPr`G4T!X)R;W!z
zr<Rr+dJ~g=h4l?Mmo4;#hxU!^!Md-8!cjybvF6QEcdpK#=B_R|gqv69trst%L>$z1
z{ng8_(>`MlRwRAxF#_A3R4eaBOuwqJ-JE6_Ma3{~<I8j>PNg&iXE$M<a70FJ=l`TT
zx~MjENu$@MOQWFwoFhv?Pp`t9q~zP|w58GzS8aEmlmDBOZ>@GX8e?P0;Rkt1Lb+>?
z`8Q(C;>}F&>}Gw0is@KRi#vzy4<lc6%w&2dhGV8Ov@M_v!7GX58ly+)x$hpBv$GdY
z92masDJNb;#FV%{{J|<Zt_tv5k5o^_Eh4g<d-|<O|0n%wfqyxjMqrSFQ)(B%EPRh3
zGG?<l!AxeK{KqJ6R!A}l5=F&&OklK@DkX76u5XSN%BOp&GDO`#l6+SIlW;M7LW$f+
z0c!R4lDSZfejhMler_ZN&4g4Ku2p_7=(ViI+jv^0<P%8^=u|{8E{c){S`Y(M=>I(I
zzjp8p7}>!8e%K#hXM|6ALUF~aoj9Lp6sYng2)I7cfW8;wBK`j#{(rk%sW#RFKH;A&
zD8|kME@A#94U?4k|LETT*&hkC|IQOtEY?o^i3a0;ov*ZAmoKnbl@jRd8qmjIz(xPB
z%Kxjcs!}MM9Z-x<|MT7?pbRfFO>dRr7gbKnvVSvPx2?5Kg|sqv{V@#x?jG?+*8KOu
zGg~jomz>NcVjRob)xB2|+KA>ZkxNec9Wl&Qkzy;ikI(IQSUDW`-SO>vo;2IS>Ifny
zOz3kO|CHE_Q$f<J?ZeX}Nz{j#ywYp6bm^~lmBh8rIRf5>smO{ozZ*>sWqfs3Xk~~O
z!Q>D%NiSZ!Q8;9P7mDE!HAvqRMjL$5d71w+8#{Ve5mDlsU@C5^MJPj4Xyc#04|Fox
zKQq9sdgCz@voRK+WZ~HNUVY#5CWv?>uQ<62NakjqB5h>4zmq~PIk*G&9c|ENVKca;
zIVGP>`hlG$N-Ls#KfvA@q}`Xi+@S~dK>hfP6I0*+sKWwR7gF-kkbS=YK`L>@1JqmE
zq%VI~`fq<Y01}rBH7b1d{z(Zd&-=dyoZQ7plmD{<R9-?O!LE*+{#+QY+5t59{*hJt
z6H}p#ZZln+Vy{AQGkOah6myKehjLa;zSXipLt;;tRa}3EXHZko{Qu*QD)mY^-2?K!
z13mbysVJkqRr>3H74&r_&a-M+|Fd{Nwd4eP@c&x>KU$(H56J*s1kM!uUt#fIbAkV@
zp#S;b>Hb%dv^djjQ!@6y6~k6G1`RX+ze2LX?Nnp2RNiC+KIpY|`Bkx>w^@;W^!9+P
z;GQAZ?zV@$Q3J2nlrC=7LV+q3d5-t>&YE*?nZSyLP<AF2Wma~XRye#GkT;uka8G$m
z`l)hd-?p9Ril{B1wKT6_Br_65oOUlZonE~1>rYxMj6mVK$pwEifSm=I0310Enj<I9
znI4HRQyyE<$y(B7b<BO5Dxdjz<|{B)(Wp?8gSNPO3JA8HZAHn@eLDDblH_uha@$S>
zUY-;gnc<DD6hTF4*2)R5;Lh-s)bb5@v?S^xPQ?=#<t1az@hyrcWT;5FOsaRgjk(t-
z(|vkrX~j_S@)h~BxUqSz<6T0KV#i0)l7!jTw~2^_?c;$?E;|?(wKeo5gxu1`jhw>M
zo^Pyv#kut6(s&`Q@XJqj`PRn`KWqQ5T>NeC@1{C;_cE$B`Og=q$1_NTWksh=`dlpn
zD948xt3+FN>mNd!Hk3E^RWL~Zt?3h~X1{K(8CNOlR)1}b&HF~+W^Z#i*Sqb9EI=q%
z@VTBt(z)(;f%vZvGR89^vpg~)`I@;xAune%c$IerH{KdwuDY5rqH;tT#hu`#b9S~4
zqzMhV_@aU&BUM93S;i$e`J_8vGWf|5u}ecfiT?1DvQNyX+UKx(;<~7Xf`p_sU`bl;
zY*ZZNr4DW;Th4*EX7twza^1O1YY(bX+!4(RZ<+ger0d6wG-7Siq6f~xlv$L}%i1>f
zt@u?~v4^nqX4z+bKDd6gvA}kJO+w$!vVr{;-@CiRZ{^6=kjKe)vzS3SPxNPa;sHhO
z2%G*!XK6lWL)7?5XFzi&(g6V~#htU!h%#SWrXhJ2H`8&Z`3L$MRdsDRJQ9sC6$ifW
zTxz+iQ2H$mwd2z3Ti6F}Eio$EH^2S}D(cSt+jeVJ9q2@fA{-!FPvzVC=#jaEY2`D`
zSCCaItTbPl)sr6d)TL&dnX0EGP|5qYa$B2FuajCs`{K~GxcvK{85ueT#-9UuzHfcE
zWa2fBxY_-g?YVaT1U^8a6DG?$UOXbKy;A;eYG3Leb-w1f;C0+7|K=Mh%q0D|XcRh6
z1PkzpzTy}id~gUCk5Iz@`zl<?p3cncQE@VV;;L;g_BK!?G3C1o-f@G5T|PR2vA1r4
z^2N@rCE!=UpDn+ji#45bxD%tp@Ws>lL+^OD#fJa<J1gWJdUIyN_>*R@c$(**uQQ4@
zVgt!~L?0t+sB^shd5PMsia%&%%saIh*)JFt{boC~=uJH4bDUd_(DNnVok@!xUQ`uK
zPF|(oeyyL;9*((WlXfop-DaF<#ItT%p?$qaW%YRT<sW?ePt0pl7sa}7iR0VjtXtWB
zLPpzv{z#8}yLGGdAZD>}{wmfWtF-kn%o}TybV7Q(6WL_drP%(XY}MMoZC)>zlOMyN
zON;F9%Ld;9S55PwhSz`@3+sx&-{h4ql}0<Z&f&FxV`p_LBxe*F1*MM>*bfQz7OVWv
zUBk!xm#CgOeqEHUq)U*fu1c5kuG^&}x;Nh%bFT+RkQ}7JD&I!(n6iu3gz+0V!R#;Q
zXHEYRw#?BtliF%(aQ}O0YhTt*A%Aj!>aOY?t&|;8U1iled})|W|Ek+mORSp}{t*+^
ze2=(U)Y)4oYn&~)oPQoXJ;l~E{0&(>WO(wUtcrT{K1aanpBYn+^nkwp8Pm=oD!~s8
zI-~LWGTwYS^)YVzBlEA7SmO%s9-gr<Y)Rg!;HBCtP*71d#kh0uQ#QT!;1K>p8&WfD
zmk}V}y<q0woVMDzey*OTkJcsoT%$7Tx3u=V>#0=l@w)7%1N<^`jA;^OFo1Kh0=9p}
zT8#y8ojaGG{b6<cG3)lD!>b7~496DwMYs*#`pSQ|M%B*iF!CYk;c4xuVEma*WuD(N
z?S9nY^Xx|z7^kt4`gslJqO!`m=|b0Or1*2sVfCY)nT={?!=)q%Yi+Fqa_s;zS%>!+
z)vgsdHQ)q28n=vHc#(kV9BGqj{9VI8(Rr?QOZK>d0O|LRnR8dMp@sYAb3E=Wg92b~
z<;UqqSA*X~vYby~@-H^;sdjPPW-BUmTE%Ol4R#&Gv};C}J}71ggcUKd+!*{Re_e3$
zM}e~qtDa|~1rM7gL67W9%3-<HFXib!C!PCu9V35<enIZ4e6(`w#p8~WVLy!BJ`gyU
zU~Bgka(9yC#KU~z%!Nb@Y?!Qb9v4ScE^Z-$Hdpgyqdz;DJ?0eBqaaIyH_9{jR5NDJ
z+UI`*Q2rksBlwb3G8G^Mghd7f1ib&#G2-BEZ))yh?O^$zZ)91|(D5f9fuFJUi)5&V
z^>rRf$OpRbZpr_?%*-CyaKWAsv7PZW>=J0?FHYwFF{9^^P}?!=<H!1)Y)~HW+%H^S
zr12iO&xJ-*j^AT=O(bJ~juUlV%&&TU;Am+|QGWB8r!RY4&Ui6N_3r%k!7+zdxG`V7
zH|(>lNp*D-dS+*BOnz=rX<i*2K?Kf>do$Jj#a_=@$-($;rQ}TPxMgW2BI`5rdZ%Qs
z>iX*s`dg~<8<|ZuR-AFXy}C#ev%02~iA@x*im^;f4QJVqCcoJDi>0S$K4fmnJ#J&B
zm_D~+9HTe{(s~|vJJb^<kp3}w?5y1^Pm@^Q&67(r!c*`b7V=7YU|^$N)BV(&e;dcd
zm<{c*+?c{Tw^aC%A31FPs#ZkWRH<Xs0UyE3NIWK~{0N~eG^a4hTYkF;9Hd1<iwC(W
z#_Qb>u9JJ~v3UGDmh$gvABu%A!V?-oSOH~dhD7m3>akI*QZ*m@h;d67hKJShSF7<O
z?!hFivR7JqtPh0b*~FuwisVr341tbalnefBZwMsuAx_SiaZF6Mlm5?7t0FH_8FInN
zHRdJ$+bf|%=6qJnG)l&NVpsljx78jY$YUh_Ba;#x2IfkJlkpdom>8^3RnOvLm+;E=
zp(mNSexJoxbalE8Max-F^t(d*X$vx3rySAfwRYn6nWJ8|{2Nh<cpB^&LX%|fffLoy
zNpCjaaC#=(Mh=)tT=bAQ8$?b2KrbggPaR(gitv3@VI;d65>f~cIWaIYVEFpyvop?0
zqZy0hDA8H3IaDsR=DS|m2b~YT)tJU1dfL-syp&)i6}K_&M55U}ujlYnM>@RWw?zf>
zg><Dgi1bg;scHUX|677O7a186+GDm<#<#lnH1sCgxn>0KtiOdPD-%(aXVD<{z9I=u
zV-~36p+{3iR`@p%g+0qIU`4Da`72*C?VUiGNBGB?0@i*uV7rZn-i%;Lf?nZzAgWTo
zYmcR3f^_wrz?Df>+FM5HK!a+VXap-|1P+<E=#3?4Z>)H{Np-!S-=0eyMFwu*F5n)7
z7}FxL2x9{Ry$A@Okrn1E5D^3l++d_)P4HJFtu|QHZ@6;ZR!F<w)`yeLBj|l=bebjY
zuE((7xaLL(4x5{ZdzG-#(fR5PHn+*rd-O&F<o=&0<<CoA9LF~Ah2<7hdf1mve}wr(
zJ$BsK@d{@fF^EJ<Snt=$$$S_Lwy+_EwKHPs<;|M>ZThQ|^yoNxIhUTo{!)a3^?tK~
z-LeQKQ+5;+b(B6X_pV_n#vd{{W|5qUmA#fMx{?tv!(f?{ie&kX6J|Z0bF6CbAS+RC
z?@$~|T+zul@pl@QTlZC$mCy`1<&?9*kB7G{%bO()vq-aP)tjp19m_UlS!1=BH*sL(
z2RScpVS_+?v`Kt{r!YjU-}hFg4fX!Be{m-2`%(^(Gud>!fnUb26=8$s7un1I(cX1N
zHMMQ)&^v~vC{+anq$E_mB8ns+^}q!Y=`Hk-00Dxa93gb+gbvaJ9J&;#5~{+Xb0{KJ
zv4Vhrf(X2ghkJy`7;n7a@9rPT&KUdq_RL)0nrqIvc1Bu3IsuHPU*G#9<pIwS7zzf?
zxO~SD?T1d5ZZiZJ*fa3WVu+$uLEQufF|+S?zQC23d4wNO2st`?zfKTq&2yQ!U;3x`
zL^{46K}@XU+3{)b6W`^HsS_N7)hO)J9M|>5T)TX?6$om=WFQbPh?dOS!ySV{8X}z~
z&D{{r?vf4|1QNO*<0SS!@@G?$31@<UYyW>A5Bn>^+8~V7vpciXM0-@)OPy{Of`$m@
ziaK-s8C5k+>mEZ#hneO#`4=(oghThXbsL%j;ik~X1pCZFM;%TcCPi)P``bH?6k#2m
zL<#o67M{~JQ*Tv$+B?+gRu|d+$cD*5!gEn26TO^bx8orRJvT2Kidbi0DnB`27%}%~
z8r^zIHWgRM9Pq%G^Nh~U)wwHj5jkE@&U?{uF2DmlPo1PElz+DojVP-1z5$)3YMUu4
z45O=e+8FkmqBu?k@PeAUd0<Fy0VB%^5(fQL!T$K`xS`PQKSaQvEdUbE<o{m>KP{T*
z2>2&xIiM+EaH@LjuA1>Ydy34!$MMYDW3H^q&NP!EJl4yK!Rhs{wCCfZoB23`8%_^u
zZDpn}tW;+I9*-~mtLE*>>h!F5CO8n;@veB@f`_~NEg_p}+W^Lrz4n0{uT+?&cXf=r
z+3>HCJNO0ca9T|gqKkW3cyg{+#8}#>!g?dck3!uswU3fn`QdOzG(QajW>}{-A+wk!
ze@<xqs;em>uP8k&qB5pfq&FljZrbEe+J)G+UfWt8b1aGWkH_|^O*TA6O21m2QG5_j
zwUuMEImfZEzccdd-bw9hslfiW3Nzq8wqN}Be{x-M6p_9a0*3hsE>7j&1&bi__%mQQ
zQS3@={-j*=VuC=}Zj(~G<jN(~ER&wgqrNbjZx%_ShPIb!1$vvZD*SuApN{`dtE*ck
z#%6Gxak73fEMeKq8Z2%!!m=&b>|BKEEqq?OMAlBjjn{xc!4APU25&(+E}i-}_0=rB
zv#(Bzcu@EdtitZcD=@0Oirknw3$N>5t6tQ!V~R5C9sYQ6JG@9@VOevB`ZzyqFBejw
z0UtDzgFq}mG_gTh{}7L)HOdXKA4UEfK>nUPxL8w-HVAvrGS!~)^LvT8sZ+E_sta+%
zAb*aIhG=fPaxIP6dpkbEE=&2$xM`VXcV6vjxw<JlyqwJXR-PV=hLS21E{HeFdnfoP
zkd582tNJxO?=0M>{GD7wvpTF7hM0Y>FUv`2p`OkaXY&YCO#kC_r-AQ{=SwYYFM|2-
zB6VwJwt-F(&71US*v*F;XcM<C*Yk`Yr$ZvuMeZV+<lq;h$(6|j?d_mEI=)ssNkeAb
z_Hojc`Yz^e@G1tMyLPuuxYSM(J;$@}`-ROUgune#lr^Mvn_MU1CJ{&WNi?o^r_jym
zl;)U<l7D}8ey6{2?RtctaJ1L<;sa$C%APH&)uaoXsOg8C;{|WR5$i%O$-9|sil_5<
zo{62qu*SOH%t`BX=9IoM?c+ov*fq-R^nmd@aw02L!kb5XdF}M)<`A3!y;|BZ3s=O6
zX7jZRmfkP7ubtnBl2;KscP%2F-zjl-s|zu6>~tNc_^Pi2NYn_}`TQrNw?(;N^ewHN
zf9SLkj|`uOfV*b+UU~_8Kb4Evh$&(CSb|2s3~$>3LA8+MLLzF_-<D9Z2Btc{`IH;s
zFj8AyZ^n|}uk~o3BqnfOtWwt*X%n6ed#_*M;Xj=iPJq%o!*8iGlwke8yv$CV_=@V^
zkQpwjBul&PDSF5MbCD>)(*91LVD(i~H>%WRxmW|V(`Z)i8*4p9sJMnvC|(#=eS)r2
zd89kZM<0_w*Xd_p6Jv5~*Vi6P7G^YI68|OBU?#`FDK0@p+C}`eSh&h>|IUh(()AS*
zppt&^q4<N66cT1eAE`QCX5-4riDBcnBUkQeP-8R-9rz?TlXboFeSh^(P$|8Re>{1J
z=t`^oX!D<$8KR<=J|Y6OM3fngUsZ1Rm3V!Jp7SdcL+6x|T4dYrDYQVp(-m8>|4TJN
z&$^DyiQCn>uYkQ{*Fx583>BYkwMA^sQ+WbT6~8!IUQux|EmQs>)Z180$2yc%hNj0}
z0gDuGzuL1WRFdacWnpJ&)-|IN!Mye_RX(ikc=V<7QUo`u7}K+Jv@!zRET2KF*v(`Z
zqpZe!PDczsp>a>kYAB{X_6>NI16&5;chPO<Q5rWgoMeJw3I&2T8wH+~#D?tV2&{L~
zx|pPzWI@SU*=B@GmFBf&a<%K~^iL&^udeTY3E7Tvh`Mpz&t$HAWvij*ug=8VaNkhM
zP}LKs*`La3=E@|s5pJI9wQqVU#CA>ABo0F{R1N8ru#(!vwY`;9><eOxoE#>{zHAwk
zSX`7mR&cZ<-Ogoz;J%83K<vO3KlawW5LS8!3<jtn|K4CvdG0a^0Rti78?*nX&?k~A
zX&|M6?CUS6kuMs=25WS5*<nJF+v_~fwdZB7wi7<L#HlFvO0x3R#$Y{HEhZPI8Jk_F
z6uJn){hGJy>falzeeYTdX2aVl#+#bfluEDpI{Dsr-QAs<u&sfHFAGs(hwL(kwvtQU
zYyrQ$HDGTx;xhiGH<rpwpXgGK_<hpy*0|fI$JitNzAp&-lUkEFq!O5Q%gh!V9%d|&
z+HF$zp)W;3bG5_5NC%T{E)V?i-DD#i`6{TM&jisJ{AJveVAqrH6H8pgM<a`f-nEQ`
zfuws8LL#(X_La`tVl7c@{AH9l`7X7UCfHo<sKV>y65B@x>0I4#2K9klMn}9>>(*ud
zl2o6kPi<50==HmIP$uA>6=LT;Ed8aAF|Qdg{%o~Q-tY97G0JTH_0DMV=pK5eez$sU
zC~$Vs{C+_v;_5>KY~=e)zBlMseo_4qD&_{ZsN1C;1NXWcW<6P#lzJn>O&SXZT`Srw
zd!p#XGq{?h&4*fWj_P0nohtjKBvyTELDoo4G+0zt1)|6rDU7DY;S(G%Fn?vINXi9;
zxezj%4nr%rZc6cd3sHC|ExhV+7)9`iIvB1u%|gXYEkkK{PRl@dcGl2Uh-Gk}Gd^FP
zK0yUI!R1m8xdF=IL<@(q>xJbGvd}P(FOV~q?BCE{i881vO@JFZ%c;vxw5(0V!iW?(
zuI_<!8Vd8XXRF$AneFG9d0}krELzFl0t&)-oW<GU;6E;5Az8_<*#&uBS7sr`&@3DE
zYry4CY|29hs_wTFm35iJgGWMD1r#_{9!e*CCSJnYt6jD;%3l`LUE%CcYEA3Fe&g7f
z!7i2#*G&<tCd;P!`^Sbgx@$aAQziMlENQx``}d9&xPXwn`Vz44uL9{M`0&DSiAMi-
z-3MHDaHQ#Jp(Y`WO((|vAQ%+;6+R|%ct^C=J4<7Rc4U8w=7m-pwdj$|i84NfeCg1V
z%p+dD_k;_vR`O-(5|Oh3mA1;=&*RxGa*^>9sAtwY8;<ZWaG(ROhFb15<%5?i{g1{2
zYMFiY!SwMEOJ!zgy=<NmZzh|ize7Os=r?T`<4(qlnp><cCBE-lU`AR54f-+`O^RMT
z7uEC3CyJeW9LQqL-B!JOZUtR;n9_&Ok?+Oqj>L6#Mh?axsv~ZU6;YV=y><WdA=C%#
zL)%!y$piMOX{PgSHtOZ%flTt^PDxpDd3!Xam`bNW7sK6+)L1GyzBj~9Q%`ozH-9x1
zpi)_54f2|J;JYfCu^B(8U^5f1=3*<DLjm(MTWmrHw<Lz!xG_#>(=!=6Q=$TyTi#tE
ze!8A}_8Ki-jQ7<yS}7xQ$_=hAlbCq|ZS;%Uv#N7=NDL!FdBA$<#(JZ{V9poLuMo+e
zij=iV$?NDl<1|e)?TMdOD4FJ(I~qFtlEmE~Hw6<NZ7+CvD!vN%FzTtDG`x*NhoJO7
zd@Ead=jfAKvrv+pXj4PcFUsqweKW1E*hu2!mYmJy8s(E`xR)<w#8`M?PnQkTZI(?a
zyO(ZViTU!;0ZVK&(azu8VDNZ@zR*eN_&G($$i;zp|DPwU$gcf<BNUh<ra<n@_8&$3
z^KPsRBp?VEjJduy8sTobKWlU}juP<3oT*&~0NTJ(!tf7p|D+X|cYhb`Kk9T+;H7(i
zywC&gYytEDxQ_tm0b72oMt`G$gxv!P81YZVPs&vJbzEJ+8@T-l6OiE_KD_7-m=1H0
z!XcXNk2e7$#vMT*j>Blsy#E0liLi07lypE^+96~m5J+263fAA!yQPFdAja2ZBos{H
zzf#B?M=^!5la&N~NDG*J)C#x9zfwpaN5R%ANZ|)iRMC@Cj8y+hA$J_bhKybfff)o+
z*dnC}dh#oU)NvFQq55mozzF7Xky2Q69Z+EG5lDpOkJle{D5*Q$lBAxAa00SWA$4bR
z@2^%skF%n{Khw1&0|ermp&-%3#=-%G`=LaK)B@564}Z#H5JyJ(7$PZ{v^oD1`~iOe
zKGLcu#UJf|{X~O6T`6RT`XI-3!$`S~c2N#<^`;%>I@(hqWjWeQILyM7LCSLMh#oD%
z4-;@aA_M(Wl9S$_v^e@n(3o}L(Icf2DV{Xf{fQUPJ;0M>zN8FClZl^rpua*m^aT!5
Y3mpwAU<T~#p95qI?BGC)fItWOAFY+)J^%m!
literal 138628
zcmeFX^LH=7*Cra<=85g(#P%n)ZQD4pZQHhO`@}i1ZR_Uy&YC+j>)!w1o$9svhwAFp
zyQ=r|JbUjdc_~mZG$05dC?FspA|PzZI{gk{AfQPIARtsAC=e}SJ6mTHTW39G4|@|Q
z9eQ^gYr+CB5Xw9tke~7YxBWL>fu`hX+YLsfplk3q1d!Tb5sJ_PO4{SOlF{D+I!$UB
znrkAjkXKyw+dWci$XUwGSsd+9ab@YRZ430JYPe>R+yL`{+F?;vWN>$Uoc+}17tU{0
z9~AI}yXb~~VY(K-J<F%BP3N9yKk!Q4WkXRkeyMOza9IO%Vs~8&VaJFUHaZjKw^+@9
zt}JOjNqipv(u9W4G?`wZ%H@=Zb9nx(ct0!ATa#zqn2KNrjb5tz2)sqox)+fm!zMAB
zz0Gm%Z5|?;XJvGnPB(iKVjum7Z_<P0QRPrzuT3|4GiG?SR_gg^$0_=2VM}AwhxOZ@
z%6&G_tbCj35Z&BOLl_*SS8h(NM;ck|?>_#a{1rg|r>iOsh+C%6P-LTn=idv+aJUCn
zJi`;q-4$El@Y7k@?^3Mp$9f@WB!Cv|%NrQ?HtN1Fwx&)io%vPLEy18c`h2C%kiRPr
zd|2C7X%NrVw&TV1VD~-c057di7xu9KD<xnu(4+<fL$>UNC30RXRtO1EDsJ=f<JX0C
zjObn1$$ECEoo9aw0lA54Dvy`!$vt1Z*}<Aii1+W`b8yC!SCko_;6Om%-=IM9{|DrM
z#pATz|M;=&529gzAg||WV(r90|DXH+h5CPEeE)A#k4o5(1Qmn}xdz`h82iJ%0Ykj7
zwgt#{BR&TWO1}VhLzpUB@98ci?2zBkJ$g8vlv~aW%v|Iy<$<U*&{)b_&D0m)wsN;=
zjf9lf7}LrKFoPGLsk5-PTV~*iPs-<sVyZ0IYYY)2<s&v_sVmeD9_vx=h{KbBR*H?E
z!hMXY?C2SNPy@|KFIm@sw=laFXdYE((_2Q+VaM@-F}+@$3`ZY>G_YHvH>9?f7;%iY
zq<W;qv1Xhl%Z@j!f??aSC|G(F`PPe!38sb4hYp{;_jbtw)zn3FUMuhzt>q>l(xAoo
z<Ml2a*vMec_alD)UyKxaKN>~;JUN8~0>T4=0(P@=G-mi8Kryy6a<TbQ$p1*={{#*2
zk3Rm4{_owol4axu8GoDx@DWPwmf@o3ZN5qG*F*!?&nmUaRv=L^1p0IhT}heU0s&p1
zLp`i;DPD`;+c+FrRlv@uYMz`x`YR3!Z6#e^R5C~P8B>k4VN7|1x|)zotIn8-pFfjV
zV0iA+SW01I%tV2WjLh(laQPt_z|=@5(2>2JEg_#fBVakIw*lnDo-s}F@3i2N7iXbK
z3QFUY_UBTIYqIZb^}V)z+mbhl;BGvMGC*yE%(w(dIy^|aj+`S(@F?h)89<DP$g$df
zcv%Iq2!2?$hA$!y38OXH%7jML=_$Dvu3XTU1jUXdq|`s{jWB?@ELuCzq4u&A-DYeh
znzRVX+el#&lil-@XNbwBP?>OSNSp;;;vV5#T);WQ`KS@~!Jkyn<urF8+)jX;XsHfT
z(ObNWtBuzuqM7sQl<~?SE&$iYC5CAwflce(N=xqw7sSSOp|S6de%9Z}3=g5k+fe)2
z9<RRiTYYS-RCIx2B$sgR)oPT_RlMW9Wj5&w`{4ui|Ie9q7|+7Ki+&t=0UikX$4CFm
zng4e;zR>x!-CIrg_AUMc85u8TvH3gtmE36%ohbj>+{sIG^YoiobOf&tweC->%*mAH
zwneT%u61RM%^oNO9MP?sPIbz!{WPbTN}ora$7kP<tl1=r5&S`B1}V#SMa8soEt0`3
z?G&GX@O&*&5&M=$fJ_bYPFh};(ude9->=c_R7?^ny{Yt5>KmxP?dt1qCKhKMPs62h
zF?=RY(^RwY*;+BIu9F?C-^S6SAv1h*bbN2h)n(|n44GP7PIxcfJ*IpXvg5x0yzebi
zrlnu1>xYjs&Xef)UVpE=><@fF$+fbsVBxR~{s}Qt%jh?%np_CT#QX~2tHB8Jb=;*J
z`-b`ou)84MbL;NydXf|P*rS*+t%I0!E(!T;h&IZF(8_|4@?38YFMXwLv$LptjNaC=
z;hf=aJ6cE49i9Yb|Apy*MRFZ`#Bz;y^V!q2^hi;Wtdn`W4aH~J9Ks<L=gXo#uFW#F
zWE8`lakC2|-WAfr?__0An_`4p>$QG;w11{CmIhf>wSbm|xs~BG2g;`Y{I1-nA(4KL
z*()a<X*GylWnR9FUNokN*=J}TopbwFH3buht<WyDSh9Gj@+otrcxORPzgx0c)#@?c
z_w&7GO<im1-}FZBUa@5}lR4ZF!>S(Vm!Cn*j+~h_obQ5p%MGW}bQToP;P+P$yTi3E
zU!C-dKlg8O_MJZu)!^({YCUzuFP%B3*1Ai(RRio1kO<x1+ipXp_n18Q5}CkxE!_mp
zUuXN*fEClasvR5_{;!vqR-Zde$Xr9=mv#r+KwaU>XI1g0>wgCuu->-l&0FSbD>E{e
zi#U&|gK+bi-klt+PkH&eE&~rwVnxnEC?F9a%&5}G3K6~g?8ZT;RX!d>Xpyj^%g#Wk
znJl5V6sl)m4adGPcX8(q39E*69z&5notk!@VM$w<p_AYHqnnr4dbTZ>WQYukYxl&y
z@oWUpLCm2(8`GV`N$6HLN651GC0l2kRep@lws)7{Dd^prDmL9?5p&2k6F~2s?>t8S
zw7u`CXV8DtaJj++Zm_*aqsJGAopPCMh6J{;HTL*zV4EUji{~F){C4C>1kVtnX#9rS
zfh(5z6~y}?nWr7RV)BD~yKZZDpOw0pJ`T?qTt?w}Zziy?x9-#;YuvwWH1*m|qJu*U
zwOAq{GZ!#w4ma-t-j^)x4@GaGYiy|b@*nk5TF!(BhYE~Un}jfEXY@X9fR^C$M);p^
ztrv4G+Z4X_+-BLbKH~iO!ruqEhZD9ubV5$sK4L!q5fdkO^1n@dDz|@2wYw!RWbZnb
ze3g8We<^&zm*J@Ocmf>XR~>s08d+=FX}yP(I6U}Rro`Q`Gs@-Y^m9VFrD~UOmP{KD
zsop~NH9xQyZS(VYy|mVFtUfNzr#>Y+_sX4sOC_qjtr(R)>R`?2QX0G54FWN_9N3#s
z4r6z7rQOh{7foklb^#4=Dd_c0NbubsBX@@5A>Y5nNK8YjK>eu<-n=gNTe`wzgnTa%
ztHTgBc`_%m@0y=qk2a<SpnEk*4{gBSJ07cEo?T)!`ayZbKq4rGd_F6$D@sdB=xTG*
z3o$W*8Z~e81fpa*oufyEm>Zpx%2#j<IxTce-|mR@g~yp=(X_O{GLYm9zBE!LfL8`e
zubI_tV0HDtn*1QqT5_<OL|7#AxUZXBm$yc#7!iMCuG!qaQWGN2mNtlzMq?g<as)q)
zHX;gL`(C?1+Tq*K)FX)ITEsZjv=OS!BxAEsTs>$T)adf5Rn;3pzmTs!z1~8t+M-%d
zqGal80a-NRJ9@5OkpgYJ&pQ-Tjv(#&D>}mbL)g`867aAS*z#4LKNk@+Of#YZcMM|L
z445ltph43{7b7PE!`eSgXwf?pvv|ix!L>3h^t7}jw(;#G$%ltVx~{pr9vygyfFIhY
z;$SgCI;A>X`yz?NAfQ6Rb=NdE^jkr{h^epiqfmJXzTKNkqg=9L<~O!m**y~P6mkqY
z7&A>f=<X7;Nj?BHRr27Zq9>8mm`(v=hTu7OEaLGv$6~mIV#-Lx0WqYM;vz5u+~NdJ
zG%Keq;*9tof@F!Lt`C|AvrdEc(nRtpDFwPTB-2ZH5U=%Sa!BJX9AHu-8Q+L6IEu=q
zQ?E$Ex&&bJH|hj-L%HB=B;?nxyH$a|T8CHSL;74hB@2Q@qyh?x#mFc6$s>^HR8Z(@
zk?3}|Jdx@CX;SxEI=!=lE5e{FWwIc#kbs^|)DgpfMI(ElFpG!oZm^;-lNzI-F}t8J
zk3y$o8)ps6*MNe*qDg?fU-p<E>%at1XJWwR5FFa!GjN=dascc}DLHURIS^3ru;bB*
zX9n@u!I=kxP5joF2yjYP+JLk5aB|&s=<-y(`tcl~1C_WAUAhmDej$rn#BoshapX^o
zD#{;X*}fuaYgK0Zr6A_@KqhB$(-63=lr@R=M!ChHbocc?-L-N$qXd-jC!s|yyt?aj
zN$h9<-ALO&Yjf|Ot((^hV@U1H^~1Z^%L|Pv<^g#H+v7a=4%X3EbkzMuVA7?u^aF7^
zNk0$(z6T&ve!+cvKQYjZs3YTR<h^Ga3}9lAHadQDGQq_M3blbpK(#_0Lm#6)w38~F
zb#0$!Z5bD0$)`em^So(`?L*E0jVTej4sl^U+@(=wK!MX3Avh{H(3w>5GDU$g9F;&p
z(BJ?F54U0Aal8VMf%V>OvC*XD=42j7p6HhhyC6|G;~!mZsQ;!&$^4^V+5k4~*2)G;
zwy?9HLbG`Y;|k~8N5ufWUm)~SB(xiH2zTU`Yk(^4ft9FX2zEW5xWN_%b{g{wP9lxK
zCAh9E0I@}(ME-={mdbNZpxp6sxU#kIO%l~ZZ=zvWGN9k#np0=QyeD!`?|c4j*uk`7
z>W@erLP`3_ulRW_pZ+`FCBa<AWFfTtn<<YTUX$N5D-V4Ew6)JBF+^Ux^v?GimhPMO
z)4_s6+P~RhpQ7yTM5Vw-W;C>fTbZDnUci#c{ejTMSPGuR#VmMa7QZ5PrWpu0yO%m+
zy0kC0NQ+zqH?NdklQ;_3lTZu}RT%SvqUquJ0;S&Ydqb&E&|zQ~pE{0wEuqLk2z;SE
zg)Su<Bp*a(Yz>dgf)W9A7le<_Afpc;ZDP@;1U^DHKnlmKpc~J-k&j}ILLw1?OzqDn
z9R)T+GUy$V&r{mJnGC`?kPaIdBbtOe1D{Zkrq*#JaEFG7r^^&`PdF>$Z;ebHN=}qe
zgp}`ivL7_Inx-JPwz`>d(-Z_};I#24#Q7gKny2~B{lJ9zut&};Hkhn)aFOJYbkAUA
z0k<-`j3Svtd4k9$Bf57`2>5jlW5;({L1WN-0LbXPcfDn!fs+*o$s#^wej0D@b5*Il
zr?%Tz`cs%FzTiAjDCd}Qs{0^nY#gC7jxHFq&>vQm)5>{?wwpgHuWb_qUsJ4JmL~{7
z(9$*_6@l2L5`>&C1&w;;TOCk4>FYf>*`#OH!V~IEcWnb1SehC*S=VJ3^wII$GX9u8
z7HK3W=5#5bDllvZCkBTg%E{plf034grHQ;uk<AHo^q2D^Oh7J4|Bx%!d#6$G2^)en
z*i4}TkhLM)0es&t_-29Yh4sw0i#|xSlzN6)lolWJIQ%LUi8laQ_a<ryWs2)(;@S^5
zrnti->Tg}GgKplX_gLz6QxpU@TlEj%<adFfzK5pgErAn}P9bys7u8G}VYU`Rcqp~w
z0#Qc@i{d!qqEE0yupWPUpP0&Yu0jRwQ5^EVjm2)eYA=H10btD@rQ=A|V9RnRt2B-d
zHYs83=0Vdr8dawJ`yA%KA#h?X;U#$T)r{bo;9xm%^gg8s031hn{DSVC?}eRjADl(?
zQ}<h;<mK}cDc+C{&p0CK<Yl~dN4|hL5FLBV+sTE&?BW&6Xv;>Ge97~{p~yOD+EH|J
zK{!d?Vgr{dd{lM%>#l!wO95g^tB&&{t1%zL_Q>S0W+X>gsEO98yHqF&{_W$g8jU)D
zwyx4kq2lFG=u|g|6Qe`u34xQ2gQ9*{<a2pETu<jdKtXdH3^w^FF1fOv4%lFjqI@Di
z&_*Sp>=)wsz95(SO}TfdKtEu=5J(W3D>hWNf|?%0xr7zCSw*6tf&+|xzJC#Y!5A=(
z{@8#xU>6TU5Hab~k!HcgjOSIt3(F9$Hcz7p2>Zihs5ESgjFg$QX`Hl4dKNEyM_JBO
z2LB8r9ir<bUK6xuEEK_>$aJ#Pu%;S3VR}g7+!>9vjsdPz&2)2kH}DFZwa*xFvk*EC
zB)s$YNwVDy!P-t`K~bIhC|rgdk?kP0jUoGu0u`=JB|pfbiq4ci#&}5D`MN|2Rnnj%
ztlz4XaU(;UWYuGqQ40mU6nA3!xYA}q2F{q%Bt^t;eKL~Hs1wBb!)doFp##)>$O;R9
zvguLJj}aw+NoA{Rh)1fCJBgmh`0xkA%yM5)s6}8^!X>w;flQ-A*d@1JMnM34HGt|D
zqC8q7$lF1X0(%@&8y>(X@WO6xnzQ17P&=nwZ+ZT{LKB@DNGFwe8Y{>rPF6@`aig(J
zg&JtO5$!gf3!JD=ASlRWbQ^8#Dl}lWF*tg@zr;|m&>e|#HF@MG0yC@l;^AT}Q+%X=
zP63-nWeq=@?M6saS)>k$agdyy@yS(kK7&4@31nhibdMnPp2>4ZNg;js?tRU(iAst;
z<l=zOZuDh<D+>#E7=b)>7`~+|SJsl)$tHEDP|!-(x@qwtRxE+)3*CxH>IIHCk43em
zlNP!@98SnCaaSnb6BW89o#zhU8}2G#hpPY#sxSBb37D+sK+uQ(ZNJbEo24?ZST1jm
zU&>utLVnkMebY^$Es1&QI93)3mW4>QP)8&%L5{($@54qr<(8DUeegFc8g810H4}Y|
zRfsGCY+ZtSMY+J;;3pP7G|QPKjAk6;H9H!O)}AmY+JA4(g@YybqZ+WW6I70M_g;-Z
zLx#>t3_1>?c`DeYFXUo1H7kP3>Nbv-DP?W?md84lM*<U+7`WQ6&x}BV?o1rI3W`K>
z3f@#IX?LQNsj>1EMRIvq_#iYguHlSLVls>ptEDGYD^EzQ<CeH6k(}W<j#uI?r|-CE
zA5()v5oVA|maJ7((OOy%^oY_jG!$lIn#x0YCFOT^CqPM#(2OJ$E6Y|UzARcKm2?#j
z=d5&9r14!e^Npc+INr~Qu^>ZB=i$l@lH^>eSE|m+$Z8prO|$wU=JelF+l&cXZBfWr
zM>gbcbG~dw7xLd~FJH8RCz|`5k$7@@_81AkTG{MX+~94_5TP@J(!JZ~cW7sl^WWz-
zPlbu@#^AtA)N;Za{J0J_?FfPWqnnd<S0tD)jx>gvz0+nb`qw2g*1m+CP20riW;Yh?
z&#q6?h>>Fq!6ja;A}?9~`)@%+Dc10SiBFeEnr!BfBE+FfHLL^G^2N7g*J4z*>~>NM
zJ9twV&X~<Y?1<-op6YmP=`5&3o07t8QN4I^tZ`jseOVd9fT*b?oZh^?+o8yl-{cYe
zywE0^bDmL<zODB7MnglqyEY?)sexJ4!{dHciVo&Pq{y5WC=LtCs5O6dpcj)9kDfVG
zK9KZ(ou`#6@|qF>v2@2^Q;+c|%(THjo>AFPohX$=4kg>o1#0^)lJ8{Cb0}HQ;LYs^
zvy32>RSRH>OHuYpMe-M@g`~?tv`Q~NLodXJI8$$u0^}BP6Gbk`J&c6>(>Aa2KgH4U
ztNR8>-`BD=Z{nFX_UUL8cj5>+HRI$tEr`INIh7I|N=UWoI!~ovCj?2<_^8<VMG4U5
z#{|tn<AvseRi(>6R5CYusdi*?m_pZQe-Y3ZbIl3uSbF}USBCN=&V@A#OQO2zSOw|8
zA9Dm=46&SzKeCKq{YX5mIX11wsOh#(t`~HQj>!cm6jRF!U5V(>NF@1RJ!4iWk7JEI
zy_0u@nNm;vNn;rWR(s&Cwfi0y_0m)%l;?25{HOfk<YaID%bCrLe?IzJKXRDGtwIn4
zu7%so9sht3H(tfQOGQAXB0ZiiSx}}LPLesPS_kfMx)l=sQF5`>e(d=z0@_C{dRe9#
zQ?{6$&<yS-AX*1c`QL-3SiV0RrSlhf{NeFmOagWC)x$~H>3R75hQ(Hle||#^4A$Y3
z8Bfw?_Z3%T|8LHJJXvSld|8an<@<Q+HUwmHtr@qBkn4jDgjifyY+y1OnbwSyxrDf}
zUGNd@9=T*`G>2vF6&YPZQ5~2X_MM&L)<%?mK=ssLIXoK;691p9Jpv0c;hvqr$mP;7
z?CsEQsQoo{D3SO20XN<S&oRaWO@0@~8fm?Ly`1&z7i_89qt@sl1h(=9U$TqJDZIu9
zm>u-f8ve)2Lht)$EW`FnYEJ2M$wxy^=R<Fr&F>?<R5mn72iHJY|2vqZ^T3V@-RcK4
zu2vZXwv&~2cm+O`fG5G?<zEj4Gcr9oP%PhyfV65AQ5<?RMky2#UqaPXOyNn-EUS;(
z&PoT&n{h0E)i8c@OD9(i&$h+~&1wZ3Hm#l+q(?}@H3nb#>!$Ukkh4$a2f471&Bxv{
z*Rvj7V$fn=n6`R4mX#H3FJW^qM6{*CCCOSn%!!1J&p{^{%Mvy(*Qm$!PW?u9A&RG&
z5q>_0Ddbpq`e5^}{m;DbXm_=*;<-+k%1sZI)q}tnG4Up7P%H6+)21_-W`r&WS_{!W
zpd_N}VBmP9_klt_xR>R$x~3=leOdY{=Zed>=p7+FtId_-wn)C;@k<`$5@&gbk)?gt
zUXo3P=Yau84Nm=mK831wZ*}?N)qR%{NlJf8IA_}T_%<%I)0v$1WM|Qj&JG`Yp|>ak
z5*{AzAGzHG2)Ks2UM75QD!>`C`=l~;Qu5HBfR{~4&DaDdRKN<M%DZvr*}$EV1@tZ#
ztw480HTj8OafvWLrN$>kx8Y5o_{X6y1~8+g2#-(UAC(xe;Meq$Iy0}`w#j}+G({e;
zzfLD=s#z*sJ{A_GczZ=*)>g1Gq(0+otBd0ufW&FPkm1`W%zQAPunM9@|Hhrc`<{yf
zRum)&wX^9==poW=?4GQlW)h3nE#6~>?D`VI9uz?=Bb(ir44!{#tKRD|jF!Vt2WqZS
zEfM#@50=?<GVy1Gk4EKb3VGt0YZv#NbuiWjEKO!|M3S}dWyP05!mU@!RwH=|R0VFt
zp?<v<iD6y+3ogY#NY*pEvLi@N-)%F~^ZB<m@rogWd!B!0WaOn$8l9jTA&8?-#O6ti
z$#L<%1%rEcnZv*2HGt8{ba*r<8zQWGC;ZpH7Z9*CH4sAT0F2hy7|-kJd*q(Nf!6K6
zaUT@{v$b|O*J;OZyR4S$@DSBs)PTR|wF`L1@M-!NG0(tT02g2<9erZiqY~nv=sY)W
zuwWe3pgL?7o*_H>hw1nANgTU;xMO*es^H^7`^5m6Csh?PVvQalJTLf5y@KCONc*`s
z*jGdpK`ZqD<nnU~2xpig<AM*Md%FA-UnOpX8bvgqBz0^B>rIPWa0|i~#-b}K+#dSs
z99~zoNmF17kDf_O*E%H|eFWcGYm$g1@<Q6sm~y(W>2IqmcD%A|uyW0DTs%|nIu0rY
zDuvNO9af6^b4$FiY>pO*Vs(8$HmRe*OIx}y-;-rFfEbkmQ-ZkFacrEe)q2lpZ=d+q
zfjiWjrvYP*q_c|C#Jp^pJSsK6YFM)YnTegrEk3O?QTV9E8c(!SAt(8gM;u0yNDBgZ
zs+E#|mhF$T-I64N2lkNT+OZkZl1UUtVi-BH4k>bVgm(d0`4^p{xO8=uaG>@yBkDh_
z!<ZMF!w;^Gzs<CKkK;$BlDn$YM&xPZ<R<e!I~KMPYS11ENo|c8quPv2RbgN(r)$9r
zjrVDA5GvA~Q*&1n=?}{BDvJ?S=Cvn~6|1l^fmF^Fd|9CGqN3d761fs!%#^3x;1b*O
zaYPYMAr^h`#0d{5&>gm6fHDN>hlBNn13k1*<ang`3sRIO)^I==z!e30xe4ii^w6de
z5)%=xNsy*R%83YxG$aBT=QG2C!i*{6^#>S{p^?XBiIL08M94(QI?KUOal<9UBwdxD
zId$abYUtRZlQss}(?Uemh1B;9pp^%q5UuW*^pbISOY=UDAUz!IiRXO_f)*iMH>u3l
z%d8Ou!*&*HejflMwHKvmU!)(VbhF(CZe2d(5@SlT4@#fXVFO>3+Bjh5i7to<?IR3l
zy=SbKm6Lf(Qy<}8S8ewl%?8}HUQ%Wzt`^w9?%D3y_HXLUWHnT`E9oC9e(!yKg$}kk
z`F?8{HNL-F9r1Le{PE2V`S;))8@wG(*M6()jF`)xyuUUcI$MMQMYN!vD4zXBvRSiO
zvL1<7M!TK!IbJrJYYMo$32O2+FIV6}*j`s1K2L>@rrGGcxWcZ>*jFpx`Nt+cUn=c8
zMD%!-ba;6vREhg<**2MLwic_zitcF2ncWl0ILm4lebRX1M#GyHk5o^lKNed_pjmgU
zmM5kS_7vJ9(YoGwki(2LJx|cCwt(PY$iwVE>8pJHNu{KR{90FB-2$azmCEg=w>5_f
zYbOEk(`rs4p=+k5S-%Og1vz;4s{`u?%H-c`&M{lYu1`)iWyg|C6t<I!7S}_U{>L8=
zr|j=XYqE-n7n+3EQ*(~L8Nr>B-2`vwCC8?oIyvAtjt)+Aa1nssR4k#$(E6Cjd1hHM
z2)g2S{1XdTPJ#h}klG0ftmI#DCtrD<#i~qG)Ad%`CrgvtQ2r-+LQ_rMKG7kZ0sPtZ
z^<Eki5-WZe05H3W`;G&z%*S3)(B(4O3A@HOrDu@Mvn#r4Yzc=hFru9MXwh6;Bg*_C
zC#fO&&34+y|5^E8QG7axUY{3;I%)^vn@6gtNY>r=3@*DS6x<46C)r>y4dLAXZPxkg
zcP-@E8E<f#{Fj)0zArSQAE}<C%C;U_ET=-%(15`!@AV_nSSzXq7V{AZM9Bj2dQ=~T
z`6t5M2SkMUGh6ItXS{{eD%tOKKy0uq#qS9TDQ`VGrnl;2#IU<nH|<=D8L<cc4HS|L
zJm{dJti6CCsXIs^59{<RzKtoITCXRlV79Eqh6E7=mWFXvK7pMSH)mMG+@b1nl52hQ
zo?t|K+vtO?K6$n&ZtGKcYqN!*lv1nNIBYxn2d&v$@*o?I`w&4EQ`vu17D8Wmlp5R6
zq34Ho090ZbN+KDk_pT%OU~Aj8Ai*HPG<6^<`~q$vA4l{1QfLF*-7bT<F-<tLswz&#
zpe+;BkxBV#GrF}fzwn=n>hXS1^a-n_W(Jp0qs2;{e0VCLe4~L(j7{^%D(bG3ApGdW
zV~|Kbbb=4VKSDWXg1!_AM|93x)W2&b-`is+GRTRYsQ#DM?J<1f4`peURAQ5BSQr<_
zC#S&E$V?lBH2;UD&O&Kf8y0&|(M-T1?$V6#HM9XO>1LS{=_>vCK^3Sb&nVz;RCBrV
zd3sT;<C*;spGuDEAsD5<G)I!Vm+I~=ne@X#!u6{?8f-DjY#e);oQ61X|16PTz?jbP
zL@e&oIg#*n)#QRBho@D}+qzg{TLQ0vvmkpr6m%CKv_xE(5<f$)NB#&L2zy!eoYZl`
z5RoA)l6y7!R#B~x$X{H`)_C%Wn~W{EXJLQ1XRmyL9^Md`_)xAV09>-%-UFm~huU37
z>P9pRZ~~<P2(i4fDa-wStc}k^9|JL`6!WhIUQ-a#cqxPN!!+nP5&B@VRCQFsCD%fi
z&B>WMkPIMfD4_=yBsa{IPGKV=G8Q`J`dBoShDr~Nz1;ca4Gu@qaUI30^AV4l+b^rV
zU}~8w=`C1HGe?SUCtyQCB2(s^WIRTxgZ^>>oxrCG-Uc)Cm72`Qujw;GD3gPYV20V=
zreZV435BB@<UC;P5Rn>Z^YU9+yq@Oj%d1w{-8jCC(df$Qd(F-srZ1Mc5vPZ}PGluH
z7u&WF1i^t5hx8zdKaiBWH%|%ikjJs&QKZ#7hEQRL8{kNx5VT^yYvnkRMlUC^@kF^a
z?2hT{nAKxyf{md-EV|XMby`&o*$cJ)v2d=h&?%D8J#{yjZ!8kPcT7Ufj&3Z@Qk$gF
zy%s+Uu3M1?U9LOywGu%k+F>Fc=}nZ-w<jhr7qVUb8`gyi<ZQb<Pqm=I4z8;ZIVVb*
zIr4E8O7m&4X{1cJyXE7z-7sX^HcKUZ{VBSTMihB+`{I|!nC0}kStY;f?`rm_yF-%N
zZJ5XGsL#(X!ZtV9lwymTEF?fDeA=J?66Pm})1JlrzVbVW1ybK&%Op|XY8j5&V~R*~
z8#H$88JJS$^CRatL0!Sy1nBhJA}=0@^|-g)q)ny&Y0-HyMoW|8#gbI(Sl3j_!Hvg}
z={X;6A^HZr>QYn4Y}O1(aLU|Zq+q{xpp&9`7+{|8FT#i~*CbVPp%F1j^esG$AV^%E
zz3TaF<5Yv_-nhO&rg6kQjVseuJ70;{@VmPtc0AXkRy<hd&z&-6b`G`46*<b7jLtP1
z)QrmO`KalYfCD9*s2|s*EMwT#RbCj=OptP0y_FE&*Y=7L$9DW0VM-E|O@!l+$qWCe
zWqEFmg>ltnh;3&xGaGzsXSPQ2G;f?tSRWzJlnMi2vVXzJbo1gMPX~p>F^ZuHVZ-F%
zyDx<zAq479{tAv_-E4lr&WALqwGWahiq4S3t$`?<o$*rzG{hO-S`_};85vk~09wLE
zD->AadK~oq4nn;2B|SDW)@-m&z$+}pm}{HP07tz<!LESM^(JM8=Cq#|*HJpp_F+AD
z!E6~xKYfFF>jX0Ynr;Q~#9Qe<?}dAJ-h>%<7`vMJ3;0anHDR{j`!&Zqg2<B^ynG<J
z4?nd}9BhCaT<n&qiYBQ_vpN~3B~B_9uH9M7>CMVIe2LmZk!a2d@H^mHffFFUC7s41
z?WB+*%u5+$ljQVtqzuJF@gC&YS^``WZ&+Cr(#$_@-slTMS<%M?lLD3n8e~2;uWX`T
zjqNN;p?ja|u3x?`)sCg}TE#_>rJ;xccTA<$GE>sJvfGm$xJCnNM`;?N5+q(na6P9v
z-sa4*Q1bJ4=~(Glocae`nx4>-wXJazSB=wE4gs0OvgC3IS_RA1+^NMr=H~4pSEV9v
zS2iS}Sv;*`MVgR_z9yLGx@z6tUU0FrYvbj0iJ6=S`~TUrOCn`1!ZeCfHI>;rpbP9K
zF_KHajdr!u;yf_U)MC43x@BrG5DYkOy~Mk_9n%!>G}LgK2sv$l-Z!yfYfP~gnUa~(
zAw9Y}h-;_AO_Ul>l}Ty5VMiC06?Rbj%332#ndIXSETcCXN*DPZenK2>PB}J7pqHts
z5ei|HVSvJo%8eP3eV+zx+8}|=<<5X*v03DTDRn+$M4vX@PbU_T4(5~(=JYGEKQE#@
zs~7IFCaW<O03IWW?mfPuB4YMZ<7PtkMrl3#3yW-r7r8p>r(}+tkcZZ&T$Gm2Z$qgd
zTxLwgFGC8$)Wrhdyf(B<rGIWNBaoy6^YDGQ|AfJ{WL2z82`R!%MVMYwwhyZ${uMlc
zI+*fj<0-wA+1t<=rvkgFs!McbCStx-!PiHbAIbU1aHcJPGA1D|CzC>zB3>4~TTql?
z85pHh*pqA4^%Ky<CzmCg^~;`16$T^XFYN9d5dsAQA_pD~6nN<M;wr9XA)A0(K$>2n
zOm_<nInU)kQ+gcSTpEr!%MHH9*DHWhKW%(o^S|*9PyL{o#%-3GaaU1rvfVVbkK9;@
z%OE3HLi_SUVFT*^di5rGj8q>8U}5_b1gSeDh`Sn*^9^9~pb)^sv~C%q5~DcH@aJcJ
z^L2F@IbFhn5OKE1+a%c709v#WGhV$1VcQreQNYg1v?sr#CWplR+-u$j$>(2B9*Y>!
z(xIcFpMwE+JMsXF*|tKab>u&=BpM8d(5giDDt#LC#FQYLB3Q$J4`2=ImEO3~Ub@UL
z!E-izDo17=roejxwGbk-oKm(|lND&<sadJ5MUPDvKnMA<q!b(rf(laV#H?Jdl2~k%
z*+RXXJ7@A#YW0#Ej4E2l8W1fiJRGNs1Z_yt<bbEdSz37;^<O%%gfDP^qbSAE5^C6M
zG;lg$`5G3}zguYi0~hpbcLAd_MlNV_>l7d`(Q+ykz$*UwnKmLJ=JD58<A~|=^U9Qm
zGu``_E?XXPzUWU99_V@z*6ZCF_VS);uj3<|YY-zao>N~ByWFW+2}`>?6>cC)FBNSX
zNnX}{4aSP~b?Reu6z0J8Rw1*BUX~hL5TA=XN}jyk&c9$bSD~$@%iJjZW%By;J2A|5
znie66TEK*{tnZs=!TAfOrs8sEBHoZH;ncIaV!fg!jo#V)j$KzwrgzJ(OlUH8hryVl
z$$UGrTi3%ia>oxJC>Z-5fm2nR?@j=C2&db0(SKsQ=v<&UJbb}lm0B%%#47Ce9{m?p
zR&toZUb;bt%fp%qO?OY<iF|ahr&6_>F8x)>T+DM&99Ug%=PV0G*FN{)VHI@?Y89B{
z1l_qq&F$!Q4aD2@-2j$O$`GQVc<2AxYFRsTUnGJN|A@O_bZ_z%i_7bAHTii1Io(^H
zQq%YLo1*-XNDIA?Iv>At-tucEZz@mCM)l2L@XvKcpP@VBR6V+r$`|LQm-WN?EFYyi
zL}%-q-Qt-{J@}lMVa>!P^QNbvc72GiP%R8N`_M6t?1Q?OkN`frQ1XrWAc?tF9CnwD
zV0C^fvNp_o#dpSs9oT*7mUe?VHPg2iEZ!?&&pfh?Vmz(Kpg5^-q9OHqgok*1YE%?M
zAO6TLDtU+G$4v@1GFI)wPqL%@^->Efn#&YTQ&TKWTWqGqNG&~307$1w1Vf%2H8;#}
z><mM0=G(cz>0XYeM*A{yVf}`#U0K2@v#C9|w9zAgH4<6q&@><2aAsAmDM50K$MTgq
zgylV&POhA@YuC|EGmJ)en?QRgr_^tBG-3|+jY`gEQm9zz;cMp0HBbt-d{ON$dAavx
z779=3Mc!j*>H(orxcdp0^}b5_#GT``r76(6i_ZilmHimZTaoK8e$Rp$atejN`Tpmq
z<oEZ*pJ>~Sv4pEiq>N$zz)W+%*>)oa;kcuH-Fj9Z3oyniEr_P&Bu|M5oy4dt{ZsZ6
z_EqBMnJx2G2_T{X5Xs-X%Hk@LK^&>4=KHDQvLs+fHJch^Rl}#(tlZ3wJ=H8|^-(*U
z+Bu8xX<(~nA0T?wU8!2jpJ&F8wb9(JbMN~tE9dyqJ`>&jsLe5GpIYLvl(8lkBY}5_
z>S5~3zSi-cORw`P;@Z#swVgicLvJ(by3?=?AGvj)d|6KC+V_d=T>&2`XZirKn%-}Q
z@z3rCd(G#Cqqy|zYMb@LSWeQnVKjuLoYNNE35?*6>q+Rk_H|5$|C(*>c=xTbzl*ZV
z!hCZe#OCMeXC1(v-#0x+*!hX0KbuEyYwc-&#QxX!1KemJ%k+{;rChJQmGzj)`G($8
z!{5`>&oB9Xx#_K>Z<MP5sdu&xFV?Vci7HN1_($*4Xy5zqSk2jeHse}?9%gHnxAxL5
z>nbTXGuqu6Hw1_9`|PHXiO;`JnPzFK6O%89t|wOoy%&4gLUx)GuR2?4TWN;dG+Uqw
zlo#lJIYu=EJE>lWGukw7&4$~biL|`z&eJ>1(`#|N4ACT)gbiRwuflWe7UAM1CH>Eb
zvsHnJGEK3>YrA-t5gB_M{oE?vDTXs@>AZ1=g?F*OoysW~OnAeM$<5W$t}P~O2@DZg
zQ%JPoEC9yOf=l|%+}f;NXB5`9)2CYB`Kb*!(urP>LbEIjQ&NSfo~wAuZf{aa_dO02
zGI;kJdt%7UX9!ETA@eTw5^F09JJ;<dm4$66Jq?asl%7f4zg{eOf16vQTzNMlT&t_#
zX6>PA$?1DQ|2E)1cRSj>^!6Fdwl#*m(?9UMgIVu1Mt_$7?G5g;1nG>6WdtGrXBVC#
zDvHXuhD7|+@q1N7e7gKlMB&1c5w)uTa|O`%bO`LgGtNkT>+T6#mnGYm!97Q8!@jUI
zexXt~A<*|h^V1g+Ppw+&4JPfr_W_lb1LzB?rOon#-bNSg6WH@4(~tSA`j}q5_>$Lu
zph@Bd#nwv41+lE@{;7+!S`(BQO6SQ0%K1mUwo+Ld3}my1MU00al_I=CWpZU$5PKk?
z%a9UK)rW5xx{6Y}{zj)c{*NId9abJ3y8_(gc^v7_rApBWgcuBy0>l5YW>_p<xk`(V
znyq)fbcKgJn*hC0dIh#zJRyKQt=*oP+*@r5^h(ND;+Y=DTAJX&KF<-YGBFs5CXVIj
zld8cK7D$|l@I;&<7H#ZnB8sL##>A(9aG3{;d%A?i!f_2wof$2^Lubd3mJM*}sJ$!G
z<?I=aFh&~A`&xT0&cBNQ*JT<4Zq#3#Azh>;x}xh{+~+f)Iv73D5<Tj1Ea9lot?73O
z^QXqY9J+r~nH0A35uEyfrF<z&WJWd;G4^pgS-C4a6dEWFYViAmq`GU|{wl~7^XD0-
zNO)Qv`N*_*+e+91oLoN@*81(WeYDPbVVW0v2O)#(9<qk1PK5jz?_JbS&LtU<rssBi
zmIXhKl7B*%2C^_wXkp^i@5ogxvLVsbz@n^@49*9x=~3t!!k(xnl*jlz&KJ$;_6;-M
zbm!7=Nafh?OlwO#m-f(>3bZmCuL|Pr9gu9>DUjfD0y>IRpid!0+>Ul}6#d=r_>>&o
z&3rvgW~3Mj4Y#3(%t<WtOxGpw3YdbW6+Uw*0m~&pc`*?V0~6z0&`0eTE4U*4!y~|#
zy*B+#xyD|!re9Q#w#mEQ$4vPLB;JMgj!|5;lIyLAf^PSO#7ApE#+uZYY!q+SeSF~(
zKG-a8K?U<88C0XJu6X(pZx!U28sg&7fc_Bj<8AO;o#`6AWz*uE>Ieg2wJv?^w$^F&
zw9fkgP)gG4>mV{NZWuQYR!V{kB7;_6sbGp#5Pm>HCWc&-@$p#`uBEx<y)yS@tPGKQ
zkD(v#ymDth;O{D35+sdm$6_q`qBiD_?Ls{4?hD2RLew5;53*e}qo!=v{jBMJfK`*y
zrkk<(aHLcdS8S{<?WVWs>jV{-ht%jWv>IV=M#X?%tv<@ccH8Nh;psFu2)UKEiCynk
zMYfqUD-&3GfxJw+jP_zXm5Up0x4*+1Z4qM|QEF(-ww^W74QkCX%o@`IcOmnKu?IDS
z<)dk>p~+wvsHf`VInf~fS5@z>{sQK+*_$`pBMBv~EV8UoJdo?0CKE&+*jHyeDM1hF
zD7+TSoV>2vgB7hB@4`kX4+b^Jd*;=gh;nA(3A9hc8Wondc;#Y%+HjoMFwAThC#M~~
zI`QVC!U3KQMey(6Q08STN$9Xd(S>KkuRrUM)fjE#)j%d?{%9hVYoC*u1sjVu2&rLE
z1n;mM9wpr-yQ{4EQj_Lo+Bz9Q=1?-~D3aj5=~7zzq%w0}z?lw729!`kxV5M*$K|xB
zV&?SSIpScFNs2?YwfWNR_zVhlVG3Fdt&gdl<||ewobm$^+mB3ykf=U{LlS9c)SNnz
zd#;g41l0;*)a*@>%`JV7wBnr1gTL(u)3tGf{}u~dk#jP|e6ab*@rK=(ly@(CDqKqb
zbrxqqrD^Z4%5CsghwxcXM2c`<{vMc`u39;L`XuW{J#PDRd-zD^Zsa+dw?ny6I*kvu
zK-kqnEahaRE|^B=Tn#llrD#UN+ZY9;P@EjgXq}V@g>mU>Ed1||eQ`xU!*GF;#D5@)
zJa<dKa~Fq%i4Pr$h*_@I5M4uis?oD+DXb~^Z)onc>tEY%Q${(NG~?WntWMuxJSZjO
z<b4I`O1z&%6y8Nit>dj`&)Q-nPfBoMI%{F-vWtQ5PZaSNXlqTvoRvK#?kD-@$NsDQ
z``R~Ox4hfin@y*nkA&k0mbTOaigNlFN=g=v#6O{JIwYi5Kp=FOby(n#3}1J@Yhrd*
z8+lx8(yE9ZQ!b&yU>tt-qf-3`fPfy60?&N32vEL~eU80BnZ*jTuSW#AhoPZ?&DPgo
z0<6^GP3;I7vC64b6eR9rv<I=5Jd@=JsB#QQ19~JIxI~}XB;~YCzpD|5+7POcURYaj
z=`LakE{*oWhCu%eGolAQGPkv`MD}0;fsT`N&Qc5Mp%sFRs)f4TmcC&F<Dw#<y3NvS
zLZHMIdRV@ThscDYXi_!?N+)3!H;y?zvvPIX#C4Xn9N+!1>Ih2%#dB!E+JwTv_K1?l
z5l;hp9x;5)M7pLlA-7wqqt0G$SV;^4?`^%iez(KeDRp&NrXV1^p!f#OVc7*5NiOV_
zxuTUc?%Ki=F`_#V@vlVqEm_&8UF?p}9kQ5pkd}?SYm7Bi{QI~#j&(s{n;xe^P=x^%
zyDA$dZE0nQIX9IXuA^ZmS7f893j8O{LEk5_mAtUs=0bVM0Lf@Q>bhxk0;|(IZ+JDd
z8g*9nj|e)gAl&9;(MfH~cAQ?-tt`2>aUBWWEY|tP)GY)>z<PaT9c=#DtB(F_>2|A9
z@i{4a!~P0ILNVaR9@hJ#gs++Z#pe;d$*7`Z+mhqk8L%mTUvbxE_v|)fTzRGSn5CaA
zrrO`tII?>MHJH*Io}bv0>Li<Q#+blDuDe0MT<tW&6H@HU2cWZVbmm4eZKIf28`eDd
z^Ev{+RYu@O?>Sdur$YoRH<|VIs3exGEI2|!{RHPvg1#^N#f1cv`A|SBn3KM!xCp6n
zg-BKY^M3%Y0{_LFY{&@}iV4)dVbJ&L*xLW4i;BT5iGU^RWtGR3Zfop{=vOtQzP9(j
z?*_<eyc!)FdL_qkE8}cEVvWBv3O$m{qK`nKGp$(?dqG(x&{2MI#we40zMg2}x;7U+
z3kbB?-6y4~kpENLg@YeDM7GY!ul+2#UgvyTC#r!}sjngk*Afr1dl+#;VESdDQk^GO
z-MBtEy&)F>8aDiR2n7%3e?Kn1YF%9(cuvnSzeGqR!xjWEKOiwL2ulXd7}MTd7*4hh
z)8gBI;YOXeoyrzzz}r3sBj0n4ye=Wk{l!8`VtV#_6YXIv;<PZ&(DQ(9A)MH>0nU;S
zjQ`mb_^EY2Hm2T-F?HAQw_=R~f#pS+5BStc_n$bCM%HE0f?(w#XJoxNZMNDI*h3Pg
z5C%{^22J8Pc!xfb#LSsGt&U%!>Cz7lXNhnlktgag)gJCi;U9_M1v(mNChYV1;1SmI
zY4e!7Pl|N%#FIp^-r;_ikDNQZPdr*X)N4a&)lj#M?RUQhCZiKAJhW9Z!G0=q_D#7)
z1oOU5wNsoUsuv!4Q`0dw3+xK9826>})wJHO`a7R?1PpXz`*69lr8o`G9v?1-CD^N#
zjg1#Kj&z1G;kA`CMz4OL)$wi|1%DATDATk8a?$$UDeqw=E(mVesU<VwXyqc?5@Q*!
zE1uu~^lG|<rwrKTi%v+qsdrIm!@>50BoW4cjbr>M=hUwoT=)de(%*2oteJm%L0r;$
zjwdEyNoa0kB-O*!enb-nT`EsjzpPGy>>sm+W)rNNOF|Ofs0L`%oOs;R;i|zCMfN!G
z2s9B7u=zAW?}tu3zDa9`+Ol?(Up7b5Lr4uO)AY#7C_JzRQqCcXbql(*O2Hc2U*=b+
z6p9}zN&Jtio|Kgzm@Z%=aMQXf5u*{A(U>#{I`yMa4ekp_sT>h5xZZE&Z5x0)H8`?;
z!*JTSl))^4s#M8eI>mDnf*)6HknlKS%%s$-KFu)M9BTyykTJ3EZfzyBXFurB-=av=
z3?ZR<l|suQUM>@Ih&C!lM>~pN*{p(#L?7c@_OdN@fM!n*IWRnfxOxnJ5$}UP{TX!0
zudVr^Suzi|H0P1juO~a+olcO`q*z8VG9I(=6QTJcrQGt2xqMB=c<-P}eb4KX(dY2t
z%rT}F#4VL`(8!zxwke>PpA3&+NXElbSwTo~{)x-d5U)JDC-&XltGhis&Oar=|A(*Q
zX<;BWT_M)bpN>|n?_E?O&VC3K0!QV}D=($)p!H_Vfs|}Wlr(|z<5glP`ijLZHAHmb
z#2Zv5^VA1*=>)1;inie)iLOaN<{&WJ=er|wwPlj|DlFMOdeChUw@bhz7g8R_zd&&>
z&0XTQWyHq|>s|MD>d?sgCIcof)K1i&u><(~+c1OyE#GPcld5-~t79hV)?OdLH}I0^
z7msr5UqF@kgTI_YApLVFn=G&`s{d$bu(b1-?cM2=?U#4}rd+br<E#lG1nu}TNsL}?
zA@dM^fJuLX95YUH;B{u}li6s?wFXhb91m;+?*}WBw%uOFh)Se2{}}Pt{ow=?A#p><
z`6c}{-ERVli3#WK-oE>$RlC1>y>vdGVt)g6C%CJ<pkX&4H(uY*DvDhw?lar@?h)gX
zPwQhtyXWm!&AOXhQxAp%pYv$-^s$$en)zNVOkP31+&OTfLwJB7%{I3KOLOm?=$=gK
z-~yHKpSHQ*^xnEUCzHEB;ApI0Jmt$@p0;}J2(^@VrJ6Pm@BX^+!#{>RA(k?1wLKUK
zUL=AydD))^HM%hLiy-E-c7A-ixxEUog>00>Gjc#ox~~?&d^<1qzGG~t_nud*fB!K`
z(|WP=62s&--n-hF+4XQ2Gq`)seXAZQ)E4%2IUHwV;pv;v7BWExNr3oph5o*`gZ1Hl
zwcNPQm2nsRSh%77a09nX=%wM1?A0=mpt}ivaC!ALnA6&2FxzgceR9p)I0R)<wbgfe
zrSARfE9vH7mfnfGnkyYG?yHt1MF|f867<^xR`1Uv)M?B4#%(?OUJ6B3!9V?r*qFoh
zKBoeCkC>?7c4Lq?11VqfbF!gn?R%~_gz5GGrOjcmV=KL%X1T+w<zJfAsOzOyE&LyD
z-`4AE{}r$u5Oa$P#A-FFdpB3BxPlwc$QznJP@c6lja}du{cV2R29T+(-g&(X-K$^6
z_II_QQ;fAOd8UpA#A;VL-<!Jel_|FeYg1SZ&od#qkhb#v3b3^ZAvCx?7xV%7|Nav3
z4Ts(VmLBLCbZxXDe0!X<uSx3J=OAW-Z+H8$8zynL`xl)f5xeZZ(cWydGJ8m#XqQ#r
zfMknu>)uP<F1}$9`{?%&ZMc37`3_$9;^BRME?xIuKj6eE_8q14Z49tf-2Ha8Z%LIK
zUE^ac{_Oxx07DCXd1TkbDNBL%Xx8{EDqnA$u^cL14ku!PRw0h2lyF78PK3)9&pbbP
z0C|J^z&??m%F?MDb{L5ytJB(&jp|Cgw4vlywAm56%qQgc)LHK`z|G#QmIvmp+KliY
zAa{tXMntN3)0h5-q6B53poJaq<(FJr&XZY}&!a8+*j(^Zn2qytMo<992v5nV$J|w-
zK|9t7ikvu45G*}R1QFibgCP8;7ie}wwMdMxUtGqb+D_fPY%o*pPG}_X@O=LNNti}7
zu(5>ZQ8v3^vq$v)Mv0<Ih^K`={8rdukF~_GDX>uMP48M>LYF2bFd)C<=SneK7H~&B
ziY+*4DBCpobIusw9Y0X|Y^G@kqJgUY28-6y!*?lr{3ivCgvxszsUt`*1S1t3f@EwN
z#<`OghM;L137t>2DG{Ab`K3$*dVOpSoi9aJPNjr}##>mOsdvImd%&=RirQNOM(>|I
z1Z&L%OfJuM_&eGG2>^pQ#nrrkwqsKOS6ZxWLkB~&+z|6F{Uw<;s_w9d*^6*e{E*<=
z6}o#be3w$5L*IrbG22Gt?yKo<bLmk4xQ*NCm$yoJ)9(i6W^C9nQ3=RI4L#iMnn(}%
zBg)O0-heuksfrCz)&<-$GV3CWpH<(aHzMgSHt7t<KMC5n+vf~bo-j|QlhMs->Lk9*
z2DkPvly2LkRYm*umcFNvhDr{7x-S8E8<9}1SC0t`&k29jJ`sf03+&qBB)yRT1XmMx
z@#L9hg;sUK#C|<t#HUa!3U+*LIn#`uk+h6iR=CKFR`luxo6WzoNcBkZEnWm7o3iz6
zS?m?PePU`QaDc_~^-KA`MeF*MStSZJ&*1YgD5o!qI5kSe?fSv^`0qvps~3uZ41RL>
z&!M{1orUnz?|{v&+uS-LXA_P}zu}@hhFko@5KptkaJZgH@m2R5eiVQ=YCu2(kw_{L
zL3w7l@oRu*(CTCUPg3E5e|O5{8hHc96$u3HcBlD@AO`jT>Dg0M{tq7t`;yfPr_8Xi
z#7MkgN2Y-;*?bxN-1H&ZByFTRD=ihvD^5__glMd2TVn(f<NLTeZk2%G*QGvi1}biI
zOe*{~9_1u1CJ0b|xPDRHR7rO?emFUf^gFfPHu6#K9Wq<pXKYM13nTcBBlHG}DccZc
zc%e7(Y}WcerT|bX^NVAx|A(+|4$|dW+MJm)bH=u9+qP}nXKdTHZQHhO^Nelo{O-Nq
z;@gebf8MCKs<Se?Bf2v4d9tbt0r<AGu2IVG5k|${pgIhT8&N^upFWzu9Z6XS#`J_V
z`)9rQ`RxHNqI{$HF)D7dQ?X`OJnY5**TsSiA*?WOd;+ZbFsyL=XY?pzm;DBTOw{c-
z;glla6F^=z{l<Voo*9$y)pMyf{6tczVvRZF*wpdR*4c?6$!1imyg|8urVeebC$vD`
z+g7W&85*#Hq{Fr$cY;Z?9)sUQ4$G7ZE$-sbo;Cx6(yYLk<OgY()C1sua4vna^rt3$
z)97}FvvGZ$7(GC7dOK+%O)$`P`k^kuq=oGDv=}wWWF$SLhdIIaJ6dH{q9Y+mV0sKi
zrR#CObkl1IfQ{3$2AcubPYG``JHhyRosyLlPVXWAP;fI2%uOiW7RZ}a{*R6%*}n}4
zJ^gh+{x)6to^Hi~bZeVoUqUQ_HKqAmKp-41V{-7og(uDL1Q*H5VJ5epG(?{76FSZu
z++DnKZgz3D<|z85IFz!WF|<+C`ja8gl;PDu6bt8%ErqSiM(d}L@cC3b^#Kez*{B&w
zjV@zokihM)?$R#nr(y67f-aX)s)yUI^|_gm)^il!aV@R7KotXrARLac{Ia8avHWX#
zHO1$jugqIOgBz}3B=Ir1VFXfz=Ib}<^swWhsbQWK*qp4(d|I#<170CE7=5Q<<1ai5
z0X^OY);3vtP*0L)Oy<f;&5vd4X(tZN?`jIo9a+`e1s!|l`jps4XA8|;=5(N83M?NP
z63?Z^1<77g=9FS%eOWf>7Mep*$C-l{4ol7Pq~`$<h_4U7XD%6Fxy)F9S{s*ScF<4$
zlN=%8_dRav!2~YJAOb+O7Jh)pMV?5nL^eI5)pdGdTK)TckX6Xhqkb1>Z&HxgaS^H)
zMv9DouXOp<ri_)dQ~z|AaJ_B-2$R}@Qsy}<H%t74GlG!_iDaKx?Zq`vIb;!!?{Y(#
z$Rsd<Z-tOw_hZnkpf3KzR7upZTMm)f(ei_Vm`}qG-X}6_1B`iJoK-~Md>Jrl)&1ie
zg7SZ#1r(iC2)wO>Ksv#ksJgq{<`I7<!#&`=KBVr=1rmX-!5Qeu|AWPuv#y&xVHhIF
z0efH;vsOND(2h;=s@v9=$tiJ;O|nMup-m8^J?7yoR<agkC<u*7b)9O>bo?5=AReR2
zq+jNgW(dd<&No0wOd8R+2?<|I!GBGd;w3IcDv~!wl@!L-hT{1{rK&|2k&|~w$QA~w
zBo_`2iVMNWxlS$lNtVNaBkBEttcZ+Eyg2ONCzz5}e}YFt@`K1d2?~$L0IJEX4F3yl
zUcrX0ezc{$*_JNxxTi=3pXP>${DAQPJ6;}DH+AK)j#ly-@9XVsO*L+IZ{FiEVA(Th
zCJ?rsT3y~nQ{#p+s<(?9kK8LHo=EY;6-GrKGWosUi+nq#-F3CuTd%F=hx0*Fs%`V9
z%Vw5<Q(c-<kNtZN)*CxvMS*JtJ_b%(^1NjSVMk;BCT|BjAdtz;{;LSFS|AFVQ!ZiB
z_U3DNb_@{#AWIPY7{qm$3-Kg@_X*-r6fCY1K@3Bes1^H_{spV`ud_m#FbHCCenQL-
z63WE?&(lBEUM6L^F#`%eB%Bob`r(nV6`!ik(hU!95~c_ZBzgQ-B|@lugm5CN!ac~+
z@WaXCV{DObN3F)B9#A0BC&BuZX@k577K;9GIP*qvW!mOqH`LtzelvyAv8;G4JF}>9
zKt!)hl%C>B@d4>`FxGyo$1`(XnxMS&O8_gzI7K;aIc^byI@=_!B8-O;xhDA|Z9C2!
zK<Y`IGIicZ6qB}Me9h`(o-n{o6g|IM0&CAF(*|gRRf!xFlz!)fv_8L~Oq1>0h^MYw
z*MkVUtV{DT!ANyJ(lP5IQaPY@nJGqq=Dr3LkQ_4z6*yi!5G_Wo*ikEK1Y~#9b*b=A
zYF1ifjx9D2{!f~dHu^1QTVg5@r`tVYWZY=4HHA9-eAlgi#_OZ_mHz#@k`oUb_JDMk
z(Rq}xsB~_bpiV=JgEwe$+9C!qqc=(lz^Z7bEjG~rR`Uxroig7g`FiT2doD1Idn9!`
zgG=LrskJ6o<(+&aorx~}fdV<EHM@e!yaLOYYeE_tvtXYVbrsR8@<tJ4A@y*-{Ihfp
z5O&CRSr)VYvH!?w=<osJt@zgCD6HLlNUpV|c<3<u@Hx{Y8gaFj>%C}~o;YV(VM;5G
z1w$*~_-p*}1rOYd3{8ed!DQfw5~TPLsM~G3w*%?bHB)Sx+^3<Fm#5c#>5+!VOP%U4
zi9El|Hz=4(bqPh9x*!Qisy^W<g{=M;LX+YvSbK3SfI23fa6-;!kpx|`St9|s1hegO
zeMR;mZ?n<+-fH>zDhB3Pq3puUN%*aRhP3eXD+3EmyZP$_lN?5ACEr2@rMR#Hm4!G4
z?q&j>#Z@t&>Uh%po#f%GU)-_>J0gsJ@FGKTPjx3u0VFaUH-P*EQnW-u!C**o`%n5M
z|B5@Oj~VCio7hbaSKCqPYM$wchmvm?heXrmd*LLQD*i&d5H~uh8x`YGk#c~sQRzxt
zB!U;xVyjP!Oc(!z?C!J)1l84^R4gDY^U4!4DJuYU1|AkQtN#cjPautc=Pg&pJRx_H
zQC-h~<8=j-0vjR;7YsQt{2Ke9OCm)8q}eUj$+>m3J+(BI*$81>aw$>VsB)m}Rl#bm
zV7Go0<88Ld9EDJTszFpBwCms0-<Y!HU$_*u9dXOf6UJ}hh(5wrIIYB^4qn{^|Ap$?
z0_5)@Uw=#K3upt#x+KQtU?k1pCJ#C|EwJj$KIZ&gjWwMOW*>rce-V0g;$KGEbCr^c
z<h>_635f8^uAdB){S*Y2IIf+p+5(1WFOnM!vUs|o)i>U)iTy&Cb3;Q9EQ!+7)7@~|
zq19s20)}s#FEsDK^2l>RC*dNf2^YHS*=?RyHgi5*+#NNa7o(2vKN~o#38zol25#g;
zWL2JTVuuGk!W)Vn8woCR2J0FY=mXLFsoeXG7<gvs?XR20BL1RXLLWY)X%1pQ7_vjw
zU?e>1u1B|d^VP~9+*zOto2ap{bO{(T@Kyr%mNh}E7!ycmcVs3y*m4GWx?3ZP!;&RM
zfuv0re&LpY*NslaW(eCeW{rLLRx1Xo38CgwVrWYpgF0pr3=*`%Y0sVPn?I4o%BRNw
z^Nuo4(tpdDDtyBt8Q9^c3AT;B7(r%%e<7KY4d(0gxvU`XO{Z>u7hJA!F6zr;yrUZ#
zqX04&zkZwTtYw?pz*&bj3(O5G`Pf>K4c2e1wxW^s7tuH6@Iy7c!Rv$IRSpz-&phO6
zvHy+-a?Zz5Hv#%3KIb9AA$4YYvX5viuVZD;eFCGNMKr>!idO3xvA-$&W;jdzLjfe*
z6>0bh7;|d)DMl+!=v$ZL)-mwk91I7)9nrN$Ni4-;+%rJv|61JbTfYkU#PX(b<#oK-
zUj7~uvEd$hMO9olS)E3Z0x%sTP!u<mxV9{(ak{N8*JZ+51=Anv!U#;`FYi<HvI*Pv
z?;hx2t=L{&ZLEQT<gUhc2vUTX5mqnBH*;w^sdX(dY_079+Y7oGhoIHtpejrqbTne_
z)kS&3Rmp*ke8scyfR#hI>}yuBE9S#VG+{@l;>=JLMOU6cTZ#vX$KK~=Z`HlHY|(fh
zYguwxw37-Ij&@1D#D^8YNlXGA@{16Tj6$Z_o3(pYzH^|AS~srSHyv1Oc7Z_(=X(yi
zqO5qwplzpOu}X8f%-Z8_EEI-H+C52f0c7W0{6oujjf7C6FIh_=B3|$VE1SowA6DDh
zzpw$x-<?v8^?rxyCq1Lmw@|()YPjr|2ZffKY+`Fn@l`1<=d-!AI5LIS0;&=7BV8t|
z{i^L?cp(W?&e9WREP_)J8)2NcgY36&nXL<9GkRQ8ODGeJ6m9l=F!(^5V&U(_-(>)z
zfl*=l1*jiXJC3fx^bGKy2K;~Zd_>j?s2xF3VmK!sQsEo7e|gXhyV%Wahcb~TX7VM?
zEDB&X6y`S+n&ksdE3io(aY@#(ID}FTDsQ*upZ@-LqkmR1E(idl13iX0QP3yoi^9ey
z2p{D^6$B}Mo^IH`l>$f=&4^zM*^e@^9<MsQ7a6(%&qj1Ro7G0N3;R<U@GJb${a<pX
zK11!D_|<T#ZU+#L&Y7SQsi1bEzxK<)wDPv2|Cy$=h43Hwv&+y50^1;W&i^(9(Ti-^
z8r)8Hmd}2kU%7;_4`G+YS}kWY;#M2#2KVSZs0`mK(1mTM4E)n6`j0`28}O>ayFbeQ
z`;af?a^r6&`9)!eWnuOA^wxO$D9Vx)(Shs&PD4G9B#CA^v>p%ID9pj=ol@#jxl}h4
zD~md;b?`N{rWTjl$ESsqp?s(znmKmmd3NEJj@&7Yjb~ZS-WA;z-Q&=jUgfg0e`-4%
z_55a<BQH7VV15cZ?ns-UUyMUsLLyE*_;s9{5AfTwY3-n~@WwvfJP(%Z1htlB5OWvP
zwY@*zhKQrzAJ0Vt)O34v5TEB7Ty%SmcdYJRud1EesnN(_>2xM9-^!QHVOue8MZ+Bw
zyIu>vJ}f`3%+6>UjjCWIHgo(CWig^<`^0JZRtd0r!Wj8{;ocG;5JE6FWtzHK(&qB^
zTaV9Bs>mlx{FsA8es`w!*f3e_`EQBK^SO0L=0WrjKiQ_raDJXjSFFSGTtUCZXPD2w
zD|79;SJ!*~B>FcL;m>@DMTYLj2<h_}({JX2xsnY=*-X=g!`_4)1K3*6H65R1I~2I$
z3umudo+MenB#aBX^oyT3&!%#@wux@Zdkv$t_nWi@ROWAdM<}9rxgLP$?goOMWL^p9
zuHS7a^uRq)DmTv?3_TuZeI<@?MwSwL)VFbQ6O;DPFFi(9V<SeOsirckF0E*o9=im-
zSNwDyv)l0FKzV}musRL;2>L{qJkRH1T`o$nFIRVCw}$TWnddaFla_^S{Hhex3K<T(
zHX(VXo&J1cJ4rlcReVky9P>U-O#8f=eKW;nlS6aHsd)gH`qhtKL6;oZS6aYGn-~Iw
zO!LR>0(!*Z$ob19ncUE`Ieo_&c)4t0F>etZbDqUa%KJn*E-u+wd+4y=h%b3M7#^1&
zjN08;CCa~?XP<SJL1J8U>m+h9?JY`|NjWYh#myn)_*Z{J<@ocJK4C7&(-fK$!IIc>
z(fuxWPUaJh>*rg0SPJOft$Rlq8gl=X76{692eT|cbLDd}cg3;W&lbOPOmZtf5yr0s
zEE<tw%necoj@E{_8IRJ2>+;j3QGBs?OLLvbt10W6mM$En%$Zh6@*i$IK*v0G1_2%;
zc-~f)TzMZ@f3@G?L3V+Lq(EspVMw6igYf`=-6B*PWGYhQQ9~@9J+e3l(eHoHBPC`d
zx+`xV2t*+>|Kv->lT!PjBf&M~KaXnk4M<=X0LhC7L)H)|G>&6l>o)-j{Y{_;M;GRs
z*Tk<uMU0W>FDD)dLH#_Zf#?w+HU>+cq>dDW+DIijQR@sVvV8Hy#TX_3@O$n=zusx8
zVtLZrUTU8#XOs~+Cb|0K2gY+^by$;4_+$k=<zew*%j65du=wz%)(_Iemc7Bg3#>t!
zpb~zlNDen&LV7sSgHV)0mqO9VC@+aO!@x#{Q~;q!|Gmdc%|$7#=}|*fiXfJdcnA!%
zw|0)i9q@N){UJ=T^W?0x!%I~4bw%|4xKX34@Hmo|0%PTDY01vBNx+h*&rpd0q!509
za;cw~p%SyFHc8gT;y0i43sn(I=H6+XCkp!%`cP<HrG(!|)%3aR=HS%Hk`~TIPlTSf
zsA&&45xaCq)p%$2$&xi0C>hhh`wSY_4_{Y1n6@--_d$jzr{2P<k|vKr@qxgdU4GA`
z^lpA`_os`+H{vM@G`Unc17J8za&f(L!hLFe%JI4Hf!Md6g9lOhw}x*-H-|b4oTdGk
zXgVR8639^Uo_kBIw<uQqn^ol)5(Fr3eQF1*j^q9dAM*-mT>>wvct5OWlIlqm=54lI
zfBb@xM6EzoZJ_WuWDuw)pGf)X)|qffffsYQ8u9|ci1>jN-lr3$`_djvT?E?RrP8r^
z`Z+U-U4lBY?m=sS0+Ck^7ZKXautc|bB0&WzpVe8R_NiYpYmBPr{P0EbhqE<EZ}myP
zbIgX)@AZQ}Skx8A6`{u~J@b?Ouo2f~N}OAA{M0%AZ*fbH)wQt{Ei}mUdKO1}8W-HB
z$lkKOB`Dg>8{~H4DpM^i$Je42ckke`*-IqWb5i(_WvFM$EX--S>vCwSsXe%*9y&2m
z1kAhF%pS6aB1?p%XJ?gVX83Qd;#Kh)PaeyX+*ffF<C#y61A6sk`;_4(*wUjIr7L3R
z%<NyNr%x4J6eHxO{=4Ys`{EUh$~_6nnWG45cl#5xaM7G&>YT=TG%7><ovblRpo7Sk
zahYloT+JHV(90H4uCi#B<hqr{mvHq}wL(zPy4RJY^Woa{%86|JMg*Ou!|EkkaR;Ro
zZcA>WR>DQvXZndCT2<i+qV}JD=ax=|Q)k)@gFeH%#43xgmu=m<Ze(N?PZ%Sg&KEjD
zdv5-smB*A3DCn?@v89_CP@W`A1>G0Rk#<5oeN=ts;C+CV^ahu}#jzKP*!{JC6Du6f
z1K8pfr&Qs&l`M1pnfAGt_(Al!jk+(b9M1Vk(QkEs4pSIO(g|TT<XY0h);WMj;#SA6
z8O)NhBBfPQf=FmMyBXJiS+0=2EeNrIc0Iabc7nrbFTHGC8D^RhH)W`iGd{T0L5w|%
z#3hOlm}qfGQ4y(ssi{>wAKrGS`C%ip)8CVBW<H}+0m8m0PvTv#R&HgS_2S44yrj<`
z-Wp2^1Sta~!WD1w@S?{{8rE~aL>H^Bc^K!7{;1xRptL<#YhXCf0c&PrgI^QN_=%;;
zlR{{d0Ok&$RAi;6ZYXCM^_jt3a`S<!oBj26!6RU4KXo)oI14NtiNb@<>q1~_wX@Z2
za{f=PBIm@@`j3l7+wj@)P9xMHYPtRrY78~Qr{&zZg`=EFVP~SNfwo2qU^Tynx(4)}
z`<EXE$&C2#et7&RO(n@sNJh~pkwm3|9d@?<v!Yt1<I~%QwXrLIwq)-W8)p#Z#(gFe
zyk5gxd^3GSXr7$F?D{wEK%i&2c%_DA|J-(s%OD2k)zO6QyP(y{wGH|iiE2a(Z48Co
zSo|2XiizR=HU_05;iP4-c@n=-46eMHc{qI^ekFk^va4R1+539@&W{RL--zs|zEUOF
zl<YaW4nKVFYINR#c%aaZ)*rGugH*ieps+5E$-9@E-m5V_SaUsDdVzDTUsSooLiPxz
zSuFhBd7y4|<d-U!bJ&VAxVpC4bs{g)`}BXj>?)sCgP-edE1=1Rm9?ouUfDJBTMrCT
z+^h;j9H=WazkBEfO)_d#m<Zw$i96(1=`+nswvXH9M~U~^EA96?k90K73D|zZ;ntfk
znqGwi+T1DO@1`!c#Wzq_;UU#UZ==#rvfQT?wY*0{1uE#nwGmj}piro{O<3KpLKsYw
zs>6AZN56`_4z5}19{DpDd7v4WYB7_zdrV>hE;C7Re>k9O=)`<N9W7WJdDp6yoz0>>
zw?RL0H2q*I_{1GTL!d-rjA;y#VPkNtBpB$C>HLWN0*H9<dOWgqd1@v;<p_J0@uunP
zY&fkUqg`ZnXM@K<kgbe)cYeJ*4grvTtS*%hpmR_lNWUb;+IA#0zLhbIzM$Q4bS-WO
znuy9Vlz4pnZhttz&?Bfio_k#;yntN++asuIK#-O6Rm?#Y14_!~;RyVoIQ>hUNF;Et
zAW>U6XIi6)9d+OkB`0J!dGzG{N7%zid@m5J1@DTW2BQ017!6YF#PAC!HOQzXgIsZ*
zq8nwf29CbS@k=ZdjsW#@J3+*5y7B=2znw!aFbyyv=+c5hx%x0v8z<R@i6P;)SZ3NV
zOg+pjZ%m1}^%Jm4L4Qln=#{W7gqLwG0{S|Y7e1KDZr3y>d8c``Ck!UhpL>v~up<{5
zStvD#`1ZBM_qlau0U)1oypFl`96)Rc9BYwbscTd{bn8!APALvkln=2!-#C&^E;xuN
z#FXk0kum-pRGajk(n5E=8jln)chOa*?-~7`9J{HXOKu~k27yR|U8kaq#00kxX9}9h
zJ<k)o_;Xx;R5JFZNfczoM-~`l)^@%(6p#?{XE}4}q?X%`-iSuj`hKUZWUYcrYXv!f
za&H;|hEp?5){&^>f}MOw0kckVtjL+E3SKfyVXqo$jv81k3LrTz(|Aq;f4@THg`*vS
zBp-5NsW}e<+fmT<4>2Po=xUq9F9H|ds18XC!^3*o?)YfW!Z8pcLTJ?u`jqO!jaz0)
zNxXLCN3*((zZsg*ig#QBj)D}dqRKl8N(MYQQ*||&OZZMl?wLv<LlT*y$&=t8!VR6Y
z4A}{2Gc{-aYGH^NsQA`3`@^mDD*K)rf1pnsWbfM4!V>P(CN*L^j*eQw04VJY?4`Vq
zPRee=lh|nOra6tLBciPgiDfcD1i20GGeGJPNGJVInjmg8e&+LgTtDIvjR{I(vy)d_
z%(wOXX|xE@P`erux>*&NMaaWcCWm$aG+Y5is#6*e%WNJyY2>(-8|f3v2+fU=g+Iv^
zt-eNw-78CFwm_|P1R=Z*;#bTJr%PcQaMk;r=|{;ps)4Xmk@VbC&vc5Ed~;eo6fJS+
zgVI%53M5c&C-t}J1Lje@vCz6V8?Em99wc$DT{n7?Dk0|!bHUrU22Q$xiwAkYu<HyZ
z2u3H}0ngsXXM=GDYERcft~?yp-a~o--MGTqKDf^Ifu4oY)fbd2Hl^soNXBCL40|C9
zix{NU!dKsp$^UkuQ5SW40p~udgCYWgst2-%QV+HiWLPv$dGjc8SQCZ`!HKHcb^z9D
z6~J?G5J=-ol?(F02Zr}g!yhdGfzp?Qx>OZKloQD%Tj#8V4>qFS&<bbnk3IR@F+7N&
zKRIUjrY3)Cs74qGqc*52|I&}4A6m_q!A@+9YZ<8BWb%|?h#>kZl&xu{F(|N?X?+A#
zi!{O0hOm0c+{T(s74t+M8&iv+|0VhW<C25TGGV1^`{S2Fc-b|@!ySuqs0=$`Dkb`f
z-+`96>+zzj!ZdY`%{b=U3uGXxPg;-5r-M1WeHHYky3m4cn8Gv9mV+_Ai_Jc#YcZ9f
zEsR}*?XS@=<GV?e2KIsu5JUpFE{{%TC3t}mb$O#uRLU<9qmKq0tRC^z=LuxMx=DO#
z1;vZ|{pjVG#{+SF4Kg+!l`B?oZ;rIc)pYie6w#~GUy$lBu^Uj;1ZJ7~GuSLgnY4Ap
zn7wA|A$T;$x#}-<D{fnXUDIV}^DQ;gb3^3nWXQ_o?_ZBo6-0z*=uYI($TO_86;;;#
z%Bi(q3&3u<L3_|7B~e6?NCZ-ud_xB>Iyvem;RK4><i?bkzyKGtdS^cYL=yo7N|{h8
zMvx!)3z-1O72L9)TS<nQ?LG`l2M-Z9EIW>6{UOG~!qS{n<(QKChZtIbZ-L%<12U^#
zlq3OwBN&x|(fTfj-yx$;t7xq!3W-H*|A=R50y!-Iy<AWbp>};A#_s+UbP)dB=1~;c
zyWFt!e4p}om%8yZ_c*?*>l(WCsXY?;9Ohl1*IhqeCWP6?E7^iW#s{rUA}~)805UW(
zAQ7ILch`t}lHFfr^TJB!fl2Ln%GIwYk%M0CULc{(Hf&^_<(b+fu!m!}@=&$nnT@5D
z$8z}eo&DLM{V`4*vJ_s-mUSy&*iIU&enV}Yx<WP2(UWY~Tz+7lSXrz`$gx4F%)O8w
zU>O>NKoo_oD&B1-riY^sjt76q@O0L{U~yA)LkO0B=N}Al(W!(}^LIMD)iC#{aU0r6
zmpX*9{#|$#o&^lE9UxMZB82jXQ$5pwBqd)|YaM1V`ILF}Tr@1w<o$uOb6Mz+1af@Z
ze-~!@EJa!scUqBXt$4c4X>bP0A-6c0a>JilpnS3rbBjg6{zl56?xX5wo+tu38g+_2
zC$huFK!%fyt+4I_?KTbVCQbRX?4+wdXA2lI6FP^7UCU^&Dj!Eps3{XIHeYBCgCqSJ
zg!7U#j}r(xkE2t=^e1VWq1x-^=68W7bn`4zaPAR^<NWNBBeez{&YbV3uH9{@s<ryT
zzC8d@7GT*UZp0YU5blF0=*L{edgbl!+$w766V2WpvVM(Noa(Yr!ofK}tgi7H4b@o`
z)EnCH0v-fQqUEu{w<h%xlJW(}7gS+5iSUyEn3r5tKaF!n+=rFR6!xr%Rx&R}^$&JY
z-av1%HP|kSg>~ONu_8DOI#S>XPhdD8GIYM0v&LagBL)CFI}K|WI6g2u7(8Bwt(1UP
zl&RYOAv0*l+<VbOdf7H$_*bPZ`H19S@ZiLJ0*zQ?PaH{F!Lz>tp4Js=9<Lrdg$*l?
zYEog_{uokN40FdW;#T||3;PK)5qs-<(9w9nDOjPN34Ml(RsXZZ5^o^ejh;2sktl|s
z#aIQHNqyX&B-e-LzZHy5)bLyF+0Joj9~agE#%ig2Q&V2XC{{{5T3WkLxibfKB#(=#
zb*%6>I$l1aQ%0FROL%bg9VoZ`MU*Zr0CobM>AT=>YqbKatE>2@QnkE_Q++-85||k4
z&mTSg;yWIbQ<iJhdc~^@UWW^vvYkVPlsx9UjYk-s5aHqPKd#ZQ3dzim>{4)DKqE90
zjPGOg|KjpD(Z{CzWj&R{Kt}u{??=NBOaarEcpU;>3CQF!$MgLiJS@8^#d_te;V@Zj
zRDHoUl%d}Svn-934mDvm1)}&Psk3*Q6KMOHQ+cAZ6-sA@if0|MZ5X1E^Bf@(^VQj!
z<9@q#abH*80z(Fw;Cr;rMU~9SA6=KK@v~VeLqmJZCNN-jhel<C)U`2B`h!&hb)qy6
zi0JvLF+t@^i<XDi;U|Hx%nl3hg}Hs+?)E2d&?sWT83ge|WjGSHLi|EK_h}$`Dqe%O
ziJi+4=&XfjJG&}L1_&)tTjjty;2_t~!!6aojG?t+$wmzbDGaT5#OE~^fXyZZWL(YW
z0#(3SZ95xd>T@944N|5W2tF<m|Ld29>5FRR7yUH^ad-pp4E_a@QT#XY6KS;_kmvVC
zrJ;Z#oHeBBjj4EF<QUp)Ej{P)R2{KNJ9S`~8!NaXUP66c5~sl>bZzj|&P+3WO*eR-
z-Thy0*jI){BMwd8&V<6QUlevr>fe4N5josW>s-Z;1=eY3;5<cN`~+_1H0`NcFeeOG
zbThkwXBeS0=kr_WlJ-mM?1RM9F7p_)jwIE;+Ao;%A?JjHda;fmvq+0uxE)?sxAVXI
z*XDF}kCSt&8E?NtH#Xe|`7`HkBe1>`Jv&o_3a5d6{wL*;2i;WYFPWG>kBYYxK-oSC
zO+$QHB^q&p_QO3rkCXC9HD7&|NUHM0xfQi+<Yd(*8o7VRG@D+p9}YQ0KK*v)$Y@08
z6o?|V$JJGZP4gkTH%<SJdCIzUd-^>W9F{Lk5`HH22s-jxrwZEGo7$fSIwix6mNjjC
z2JbkX7hb=0bJC-2O<$qM)cvGKhQcW_kt#giDRpsXc8FBs!GO#J3D3ka_2imob;dfP
zGH(IQp8n84y=~QXPM@NuP!=qVq>@}fSJ7!3B7gU{vG5yYto3m4uR5rRL2j$5aC*^*
zpPe|fnnsT(bD<iJYG|gHscgu5`P9$Cm=RlSv&vCyCqLxr(Pf9)<T0nZd$S=1iu5lg
z?m_f@Dx1tY-zdROQqTiGHpyci$$BtoYw?I~NH5jx?g(L%NaT4BN3AsM71ufAy_A15
z1BD+=MA*kvud1e&H9Cd!ON&z|*W<A!Kk&Yh$j(cLejg9rgkx94eFw1=WED}U!J#|`
zs!6Ygne`iIW6FZWUIm<qlRr8%q(kZ{lu6ZyG^BoTL|-RrK#$(?9^L&UG=J{iINkj?
zoymG3zwp&eJC+$P_Pcor{vw=3S{{qetQmo*a!lUY4-mcIHUICR{#SGC$AL%|7pwfj
z4Z(<J&YtzZe8Vi7-1)S-0ps8No#I1iRFIQ329M@c@k9}9_K)A52u0YkKac=L&;C9D
zDkwWGj2Tk=fc5BZar5K0hcUg-cQEIBK>BpsGFe`nI7}=2uS^&-O!~&MlkM5Aql;~|
z0VEjKA78=x<11K!vF2MkPcujcVO(!~YgAFwhv`~6XSXTa-(bA3-`YF9wk)~$4*-Jz
z+&Kbt&UsN1$gA;Z{~ABKp%gys^gjvj$1a%~N*DiBo39OZrVdq#w3NRt)Zluq&U45>
zRt?~TJHas{(}$%_hL5QocKs$~!ifUQ)K9^$i9w{3>2;Ist;vIsj#Q@F)g?nU%o|hz
zYywdm!SD(G{MjP<DSIHceAQ#%rb0OTU}X3gGF7q=AkXDdql-Fyd1<NMf?eF{#$dm*
z**yL0{F>buQ!A`zpJ>oH=T+aA4Z5uSFZ<44rP!L*u2&8bXv{*gYSG~o`Z11pr;cPN
z8#f4QF{MU$$6qujc%1^Vgrk=j_BL%0)KV&VhyF^!9ZUG5@Q2ROZ7o(oD20@*iI8{5
z7=gFtv%P%6eN7Y)!`T(RxeEc2!Y8f$vL`}v0O;&8TWTIR#1Ac-K{SUnyQG2HZ{K9T
zQoS3XH`@%T?yC^+fnj^3FkBFb38Nd5%n&{Fq>Ih`{si+sKL&g#IPA!r&J-`$9Qen9
zi-9<WB%^Ufj;qxR`X4-5a1Q(y1W<_Q{1*QGgNF(igxq;OPKqb|?m)lDZm83pkVu@(
z$~)Q}<e!t%W_K76Z>#~>=c@~~U0~W;CIi2c(|m2<?caR+2Z7tN1h1$W^LmU)b~OVt
zmkC>vHkYwGsy=A4dxK8E!oXj^`AfFH?w8RNsY%D$wI?{dWo1J<@o4>Igh<6C<PefQ
zL$luDK8Ca5?Dz!;pi&(D1A&7i$+YTYY{7fhNe8uKqeE>`3kkto0z0~Puc|cj5etb3
zHbKb2`?2@_(6)M<r=EbB|ETeT_)mwVq@#bMQ0h}i4x9@TFFvjFU!4n0oqWVYVghW;
zLK3m%xXO))N)JgHNB={izz#fK{#L@xOAr?u^l+KjyiiB0*h0DQ)S25i!q1Dz9@ChM
zjY&t`=xB0Zsq&V_Gwk20%fRr7e`1Tkim*Q(j^(kuO{lgw2#gn|Ya#A~!8sKSNZmxD
zeP9(GO{#`~H>VRQ8W9oNEdptnYBhFpPynj=VB?Pq%uhKplc&9DuvbVO<vo_2FVzK8
z+^ULBcYY-<PBe6M)ZjRDb#%m#G+bX*r#|m;PIPooXz`UfwU^&{I@{13FB^DQKfMLy
ze)WD{7>p%(Q0IJ=euE?q9C%KxtTf8h#Dtu?FJBxtcvLrUpq6MnH{QSPBftOQS>xe>
z-}yq<Z4Xuj)kfB@g__l)Ho1AXdcyt!CpgNA9pFA0D;nv!YBqg60O`ku8@R>62kp5t
z7VQYxn7eA`ESNjrKC&)-aqiCBldOZ@;+`n4QzcE`KDP{dmhj}<skkXx)nUf)+=mIW
zS{`<PPMoljxkbwuYALZgxr_KVVDx3v;&HFu*65yia25-Ry+Ista}5+bM%@7(IC-5+
z&#J(tQ9?N;<W%k1;#iO=DR~A>M>&fMWdP(IM*X~cBu>F6AE5Su{8j{x`yCkR4gYsr
z3^PjfV<j*V&<@Va{ws~2$`2+&#~^}N*n~IdRqzI@d+M*cPciG2BPglnSDKBw&)!-6
z0;cslDoJt16YH`w_h$9x6g!8?4W5b*?On}ojaW2TB{t>PFcMiH{zH=wf~densDk$d
zAtS0~55YlA|DSxrvX4AA-c-wsWvffw5#HVtf%DABUqY~^(bvS*>f@q{mDk}DU(aLy
zX#<qsR0iv*6|1qPeij<4!Ua0A=S1apx8e}D6?o!Zy6~!!F}tIS%ljgmgGL1}g?*ZZ
zSx+>LtdR80!uyEb!UqZ7qAP3{tn|OXIisT8#!+}2o{zM^OlG?4mO#L?yVU*bq7ZYK
z&@}f&22V+O%x_RmSF)a3KP?DH<cN6oG#9TQ`T*Bbz4Okt^H6)Z=Y$5R6hmX0vX{Fg
z)(LLK@$&pxRVNL}h$GA*wkzL9xkelHSo!7k2uj{%f^wCYUH0h{n0Lf?ny!%;HzYYF
zc1WgWnIcu8mQ;JB5L6A(+>rO|d#7_!7ed%KkD-cLFm5u`kGOc!(eM!b5Levx6WYZq
zYIOT^0asYklrA3KSJpF*kodqV%Muj1#m6IT{=TDhD}j>t^mY4XRY%bzG>dumw~%?B
z-e@9;SW*9BOjHOQ(*(v3OZDBzvuikqjg&9hN!++i*`W1Dl19Q^f7~^&{d6A&WF0(+
znn1d}{Ma2xQGmg@0}L5^MB;oc8I^y)gEwyABdjL-RALhX#L!=$2#A|XNyZ>}x29$x
zex?XX=ZjQZZFv$cdzt)oVCX(m3=7}}au$5<+*)jXP@nEGSj_3;9@wjD5bOFp?FRm-
zEWrZJOD0QJjI0P=*aSpgNP3$Yanh<bS@7qqGWJ+vv!GA;yF3Fe898UP<__{QPHI@k
z9>Pl7RP-34n8i>;lNeawIInP>81cwu<NJJu83aZ5)X626zrCH|x8$}&NasC)wA+}g
zv!Ip6xJIm_s8TMRL}I!fs)VP&-6Dr7Z5HW@LG@i2cLAw8uZHV+jSTl~lz@z$>9%)y
z<ydkFp_n3!qg$a={j^yzSJLH0a5*{3?eK@{BCHcqNq(NpzEZxLQGMwcN{N@(?0Fnp
zmd3foP477U&c80s5ouzO;U&uv0fAaX4x<3(A^a{ojf+n;Kz<U%F`b9vHmVdD9=(nO
zcpMa)-S(q5yT|x+x&9(gh97^Ju^Z?|DGDXD2O?A_2i2~&kgLa0HYgS5C|BicNZQ5o
z#M%bM5r6@F11m8;66U}WN%{Qq-2_m__i~`;a(|_wBUZo|FD%&H)os7&Ck3UO=(+cM
z`=oqYYG-j=OhBKKAPXKbMku7q>HW&Lhq%DcsrZ+SuYztU1pl{z66wX~cont`HDSCK
zAB!rZV=RQ?f~iCPuQ|YC>kmB&B^?MFXR1KBi7k;}Up;Iw8c--KU)<R|<|gSGOZF=$
z_iKsPnEqxV4`Go6uE&u>!QNDVVyHuKBY!`}H$jV4t7#OvN*EG!IA$?RCLPONW}vv`
z-VV<&B%$!(1fXknxjd)Z$F2HHFFJupIVMsFO!Nf*Ao7mAs1`)qMlMap@9f=rPCyOG
z*4-i?ESl~4%&Q?V;!}kCN8=P44bJ`uYihNL25)$i{!^!p&@TyU1H8$D9be7~y~b5j
zhiMu;xc6VQlQL<-CT3h&!NpVkw<U#@VB-AT(iA$b-Qc$d>k}Rck(dmBT+x`2X}QP)
zT>HWzfl>V>S+rc|f%fh^ZQNidt@%|yBJ~7o$AY0@Pe$|wt68<e0Gp|D&Nf{;*pxYM
z%~&J9EpGkph)ls~*X)Bq{n>S%luZQc(m94KT*+RLY2T{{@Gcg?Ik{*)g0r|$Dd3zf
z0<p73Lg6{PnKvZu==zFeY{A_r1X>$OmVs!li456h(1Nc7^;fMcn;XMlFHR1MS>(|;
zV6XC5UuupHxS>5tk*iX6L!X$M;F-1l2**dyjZFCh#H<hilhvsa`667Jc>(ki)`9^@
z+#o^ZEx`hErG|s0-b4e&149q>S=urU0uG5`LIC4Deyr6?+e45Q#B1Z`--rH%)GrFX
zZFSZ*w~(!0R5mwXW6cs}1EgFgbW3;Z3l26c8X!#fMK-0K+IkF^V8W*4%6tW?yFUE|
z!cYc5)ZW1%4f_+9Q&2-*7ZWe)RmNAQ&^71akYC=2okHfCSny7%oufKnGZ>CWj#1&A
z-JDRAl9;X;@OrISarYbXMnDvd)l5HCaB`8o1L3RhxZr?u<`y_Rj|396WnfkC2rAz4
zRsYTKwlr~*R8S}^tWQFUdG6j1;&2(oijL1j7YGLRW;ZLf_uVY<d=3NkA;dpOP#`Eu
zEK6+X0{94MgIi)W{|);px%Q=yGUDrXQ&;90G@0UgUo>IZi$zP)<CZO<TjqpBxMwaN
zKm+^zmTa@|f~_FJJ{~IS#jc`J82`dWa2LDHRhTz-G`CgBf8?%fdG6Ij;064eC)QdE
z2IZ7dR8<xZ&G1?wkzF9&P5D~A)=df9JEI*zvKCtwYe;dk+A;A#niAg+oQzd_llngI
zOy(&eH|7Tc+mgXJ&li|zRruhV`?=<bT2%x|9A{j)vAV^lq)vU18=5zF0O+`p-k3}R
zRfqb)MwQAh>xkY|G?<Lmg~aBHs-cUiT|O7Fu`5vE5vq@QEebjFzb$zO$3Kf5F^)l+
zmix!~^jV=oK;8(k{}C|A6~i4M0USiTGfFooskytN<)<gC=b4}wNbjuM28DyZ*F~-@
z==zB}l6b)Jragi5A`N@@(k3O6JSapuY%0bn!=l^DmY?%_5*Y>On1*zd$x>p}@Sl<h
zF(G9w)VNTDn)<=^AS#7~(DLjqDT49FnI$udHfRvmTqU<QaLtc2j$eM2rRDG7WA$sj
z66gr>%1&4Du-2ULPHqgX7fz#d6_uhno+ndI54C&|LP%ky5q-?h$0_{Lz*ji_+;wX%
z-&#wLU*Hmr9(1aeLmyg~kDwH=wJN;9sOPrCloAS*5R$;+LUEx4@Q=yQ9JV@vm1OiN
zLB`QU`C|X-Ri2Un`88oFF2pP@BoC-j$suWfJ=5#|m+I~5(YeNh!a+di4so64kCavw
zylUM(S472~X>nM^%GAWI@8#i7pja$}T^SrK{4KPco8BHTPpnXBY$0+czWkO_6mLR2
z<<1?tAG#!4TiiU1rtt~M6jmv;oF3csRTM6yLSQL9s_~uAO+XpfL}FCs^NJCdG^&AE
zzuXfWF)CqH1+IRsJ1=BJT)!Mx=>{UDgk-m?u99e4#6{`gL+4~>rqri&%i;VZ%}i7w
zgDfSqjPYc-QtvNb0zvZA>Esk?LkuYkokfuvS}CL=J}U8Zi3&jlkx|jV#+4E36+jUG
z(MVmtUi^2bb||vV>S-;r0Wu{G)|1R|y=#kOi3L6?Y<MZ(jzYbxIqb)#q`*q0jeks%
z5Kt+%Ca6bHKf<#NvB>z3h9>F%XjuQpf`2qv|D%Bd{Qo7c_m4O+KBY1pyn0w=d~DKx
zGZf3{^4|vjBkuoS+WY=X`~Oe;|GLlrHBCSzclm$BEvG0`zrH-12PNlH<}Q8zT69!_
zs2YjVdO6<?K~~j7(JY+|Q9f;X)_*>>_(XRN`rMnM_6!2nK(Awb<$E~Lx1iiRG_By4
zdM>wPcCK~(^A=sxb(I-sse)6q>hFa_HVdrcn*T|~#qoLJtwr+*{SF}{Akre(_qOtZ
z{WFSW9-I5&**Mc##BTieY~EPM7Pi_4mKASOBO=Vc8jKIHQD{`OJO`$6?olksq-j7_
z7n;zT-(G?J6+lwUoSKT4!zlit#ufER6l0Kdz|fr2`}}<Y=)}w8-bu;~o~Y+JqLE(<
zvo=R^<Pn^5#&o+znKG|U4c)(KRwMn38$hs&tE1x`VmQFX<$2ENX=@Al9>>$v&7}n5
ztT{BVEou4(cfF&5@%;60<2quB!&ol(EqOy$;dLw<H+s{%)ug_AzP|e@QqwS$quaAP
zaT1Z|5k?f&ou3zXlw+C@#Bw$HSh*%G03}HG7hzJH{;=;CCiK=+>I|ushOyP|6nb!8
z%ub?=?rj=qUGXV*1`hxy@^?15;>ulBnK|V@(W4Pu#CVDcf!ZH~LZqBg*n&HM$;~iK
zKndtw-MlCdo(#6>fr3*W&s%A_Hhf18PE}4)vw5mX(a<^1_Y<w3*qY7&<1(W*EosmA
zINr~Hth;RFBY9-?LX1=8OCd9?guG|tOV@{jae-BHM|<_>lW^y8erxR9_NdYCu7yo8
zyks>y6h34eA0Km0*~w^qE^BvY?u9=G^O2qLD!KMgl4Dfe)ZEa5T||#@ETVgLZJlRJ
zAnmwsbyo7JS{d8_Dh<sOHz3`ajm*=L5ocf$L<gj=vMF7N0aii1<p>HAZ44#1r>7MO
ztL{|>JXB)UPANC=A1w*jOPlGD&dfoB1$m5xO-D9I7zlYO49yb_lgnMp6GYZKbP(Lv
zj?r0p-z-ev0;OroGdD_d0XyTn3o18bx4fp_0Xv*IX_n@$3k*tWm!jeRTH62K;a9;|
zUP4z7(6#`%Lj`J$9h!XTE`VcPwTXvhCA35f&ji3N|B3tVfzlY~PBsOlqO2ANHcH3z
z8A1)aDyB3N+NI44x8MTVT}hgi$x`cV_~j4{X~NG(%TVJ%09_2(q@b#TXrP0=dC6{m
zrF*0C{^iY0SCWSEv7urK+zM~D&6owv3tK6TMA>%t@<9Fw0+4pclI9Ri2oFQ8ZGi3y
za^1g<C^C+Y0oO5rcMF&pXw}ETD?=1jAmvkt%@(9`VoRnI9AR|6QMk#9?uW?U2WSE=
z#MX&i8YBuQYR5Z1s!IrOtiRacA|O&y$s5vC2bu^44)LnCAms-acR*IEXV%VV^beH$
zMFAB*cejp?Kbok<{<G5<v!x>|TS!=Fh45g4&05%A1AJ^m6+|~xDg+`GrsLdLUH<om
zka(lTGc(c@?bwwLm_J=PQN+08NCbK%E!F(dm{y*;VTf5TJ@`$D5e<46i|lHCH_yqV
zqZY>XBHW_ngF&S5{)Stc@(Y4{rSut0xH@`&vGTA3?MAZQx((^0F<)A?0nnC7bd41D
zhTTwiB!zLkWEIz%kbE}vN$o?`8YGRU^%%V1?DR#}$!CSP)5WCrRsYYoD(`5I7Qn(^
zakFq&%Nc({al`#!BzQXJWa1zu`#5fdLvi4`)!`s9$edByKuBw?B;cRg`QhW{6`t4?
z5{$mcu+!*VbRum~RgtX2Dt2M-<3||d%ZZozf|%+GLizh<Jv5uZag9_n;V>t`cwE<a
z*MP5@T1hCxBmI<BumxGvRJXeToQ8dEgScFz!ji)daumqQFzB=BinCG<C7^w@;G}^B
z>E@!)=Ul-D8?}@hn@cH8WcT@5o6=8GEB$vS1e7{R4{4<5sk)WJ<Wb{vjG+V+WRFHn
zdO;me4|qK}embiIzl`O0Vg1))5qOZ>Q{;Wb{tztbnt^@ksC_^irw3;8v&3Nov1A}-
zD+{dP&j^2YB1I;W*TQh3oUrUaOkmWR9n_rM=PS|dO1gtG%=so<=^DLe1XHioS@W@M
zcQ&gA!&)rwR?}-BlN>&~ZSRR;O48=^#QO+##F@v2dI6*5c4#1+hfu>I*t^u(hkw{<
zMk(g&4ztF2EAv!!%MHX4;u6Iuc*WnDiO7*=np~Ig<P*s#c!fcoica2k3GXDvD3j(W
zxyu-Z#~7m$;+e`nFdiG19h=s$2)3y^1qGOVVHI^ml!ZcS2q+vuZzO;;LW@Q4Rmw>n
z3aa3qMAMZXO$vTLLx-gs2_6Lt*^iB{m-6%I)Xq+_3P>A<6yCre-p2X&A@Tb|M?oA8
zj5L(f?BD;|IsVlz$5&Y(frYO2TV+FyuCThCt|w>(09fzzPCwz}eYMI40J3009+CSM
z6ql)Ne^kNsj$@kX$%Y57KR>v4*Se;GY)||F5@9>oT*jk`h#$uT&Qzv`<_4(65_m(|
zXq&+N{n)}I+Dj;QE+WS362wsVFO4CjF9Mg)IN&(R6Jj)yN&u-?0}k#hL`*&_=ruWu
zR%-5w)`|W;ha8X>4ENH2D=J?3Odl$Y$aM<`9fX&`0dx5WI)oD+W?}#V3~BX!=dJ}m
z+%711Vw{~k-~qu9>M)WOKr<jHCLS4<p0xj2npTF?%5R4NyNkeh^P98fT)cf3WXu3D
zS&?HBGcg1yram-KF0|A}2wO*tmL{2Bn&)S_E$I|6er%yA1@I-vvj&M*_*qCNh5DXE
zU%LZ)N#J846r&+aW6yz*3=FfQwrMLG>2!fenaBTcIFU2yFjU-?zq4~Y%+}*xxl?$T
zZkSYBzDA&GZg`!DlN1$3Wh7CefAYG0S_1X&)v?So!p<}({Gf>t>4oNy4c4k>E|Wy2
zt*uzoxgCoxM)>*;QhHJ~O&_2B6>Kp~fcmL_q2F=b$*$e3-BPelGUwB`M)%C-l9LKa
zmzIV67{4>my$*PA&-aec_Sb=~u8lreXYHL(fk2|??Vaas*ari>1Cb-q7s4~yp@0x<
zo+Gn6eR1GW%Y9Ttoj8xNg4u!58TW?Xafj`4QXFO)DbN{>u#gqUu6Ft(O8T)*Nx|mA
zi8H-6326~MKUry4sy-npuu5?PX*rp^I_vb#7*c55pA=j$6Sfzu-)KJf=t$Ave7sO}
zrToxRz3EYHI&mcQ8!Da(gXBM`1r9D|WE%>sE2U2@39;L}THlxoZ&X`zrNiUAn?3b0
zMGhKnu7<dqcA-YpiMU~PUv$+&VoGpy8;QkoRQjqjp{v@RN9<W+nc+Zawo?HqbVGF`
zdhA7XY#b)7Hzh_(;udI;{_>1zF)Oj`L(}-E#v~q%NyGJ5n}lD!OV<)HF}Kw$on<~*
zPak)lP!O30c=-H3f8`s;`8@-I)dfrVn|>`Grd?I{sXEUz<_%OD##vQ9t<ZF2U6HR)
zaY?8-{f32nkPA0ny_NWVEA9tmUGGh)|7mfFL1J{%UbIA%z$D;wqS=tLB#YZ5B&K1l
zsdx4*K_jka*62==?r}+q_yhe|p`!1Ii<$L84NVn)U1K6J(~A(Wlb6_vIpepazR8xF
z`{If7L(7=Knk=05(JgC22G^ma>xk8w(^%<%)(yn^K%2q@2`cL=TsJFztLimHQA?hc
zLl91woqg}j-QIZnt=$sp_#J$<Kd9(VRF3zV6KKkQ>h7iJ{Spp_kCU+nWf0!Lbv@(d
zuW3UW=#c#+tz|~F+K~Aqy<kG+^vHRW-L(K{m5}<h(b3_WdCoO*8>&bP{yh@`3Mf>i
zU1Mcc4?8rVT>+UyTIb5m&>ec8ZD=66Tfk*WVe2;D?L?y;Ef*w4Xgt*tLAz|Ku~wW?
z66<N^FWvomL>?GHvTNXyQUa%w?wbNlE2|DT#4^?tzVjKI)>Rqdq@R)PGl0VQTNU1i
z<?ltm@$oO&+c)9iqcg9ns^>W^XKqWBYiPI}By?N#08Ix3(vfgZJH(fwtcpK2vy$s<
zRR!8nk0;<etv9T>h#9jAT6P%L-mHQYE&b9Jk&HdB^Zt}d?&;kaj$vql(n9LKaUPdr
z@?;>QbROdm+0wX`Lt3}jX$*GW6eN+&zSS_sXRFbWL&PYXU?E-zx0D<_QAjD=9K1I>
z()_7zlZegEsF6rBg5Dx13RaDFn9GdfgiQ=FxFCuDXFwTrFmGCH)7;(hgTwcaYmg;n
z;<Q1IcyZ0I%Nf7FRo#W;56tkR6L(G+PY`xid_9KSnaJGQi+<I-bky3jdai)&IQc1e
zc;B$@YQL~2uEYefA`0bP*u3z6Rb`p4`}D)AAJ%6-iNEjf6U^7OQK)-*2m<$NkabzC
zv%-#yjhi6!3(X?bHRS{0XuH(VtdHH4MgmQ1e!(HCn^z!3%DE*w2BLa;%eiis@Z}2~
zTGLF4mV7QmOq{HS9h=%W)_zO;Vpm{?1e)LFQIC(q3kVgoo^~bM=0{u*iyuFG0xjA)
zTguK5&Notj5;}Yt0V><y5Qi1d5780un}iIV6$RuycK$ee%Ao%$MX6O8q!To&3T%-f
z%@y+WP?SgLNYaF{EE-%g+!wzFM00FHx~Yg)UF!vLf<u?T&2E4FzyubiKZgZG;IBdf
zUm9AB04pKz54f>&^|Z2mOm*$e>zu6yMGZI~ZJQcf4C(>IE({D*q~?qp_>9{mJq%z}
zid?WTE)?3efKzHEkHeU)0av0suI`4DYQ8Y6;gquF5OLj*2k#(EB<tF4-AdAYy`f@S
z1?;4Fb+=u=zH3JnQ{k<boRrI{ICdOI@z&j13Jr8L19^Xi89xK42u3op@4aUYlHCF8
zUEN>%N<I)cokwAa^c^d4yP#O9Vn$A7izEOpo;sG3ThaI*guQcoCPC9S8g97Z1{-6e
zjcwbuZ95wq8{4*R<Bn}R8{DyZvim$o@AsYG`E$CbYp$NYX1c4os*5iz!}mCDV7)tQ
zI4*^)j3_Rpq_D~NhkH!+2BC<KKsILW#F!3b^D%D1wM`K%38gwXK)M9U+`_BbDt%gD
z6I72!x#v=ogH0o=GSx@cj<Ch357IVl+wLmq&<Xn{)JY!I2^oL5=gL46Zt=Q8H!#)u
zh-Puad*%^Zl}QghikBo&$c<k-)uz{s3n9vl9(d3xP70q&x8k+~YCrIIKjuFc&bHW*
zm;V|#qt~qPYmTGap0^t@n)$RvrLP}nA++>*FcDGn&q8~US|+fQZ{#$9S_?9HLfFQ%
z{+Yv;FcMLZ{2J1G@&+CJkw-60F*2o34Uy@V2&9<^G1pQLBUov&s1ODf7(|WSU8!NB
zf_pO_GFqQXh8L1*1!+Z{I{$$3Rxu69LaK<eKXLJbs*pnJ&`s*#sULKvG9Q^NUq`MZ
zYa>?a2(-h<Po5k038KyQcW63-NZPaxJ3&BtZtLw3Kj#Y_*)XE7xUjxd7WMn<Go&_a
z&4;7J<Ot&SeE3hBVMM*76BH3Ih6gCckl@L9osy-P=E|@}Gip-J>1IMK>SWhku<B5?
zXdav3H|o~iQSIE^8z)?%x9+41TjdvN0>r*yi0l7r!dhl*Dm`ob-U2f09)--`RUMb?
z;Z8ae)wi_2s^#x?Krnfe2YuC<sEssKoe2rA;>E-GX$(+?2kj0MgolNEqC7AX6(C!{
zVsy$n7yd2lmW?HTBY+_)iY+;-h2Gsit&2KLINLly9B(bt<W91XlzWqKI+bLt;A<Sx
zIEaSqJY{MeVfG`sBSE_*_a^WH?n)m@JC!~n<Pij~7xXRp+H;%2WINza4D@jWEo3PT
zQ8nM@;#nclBvLTHA9c{z!L;hrYA1|YWtf^>CfRsZ{cQbattFQzIVpAssEe5ub&?o+
zMUd_8;=6W!t|=a%v=@+Y%<0X5K7?zkfHX1>`(A=7t{KzhYmkZXnfd{QvFLv8ccVA`
z(n?4<-_HQ0Awq+ag9-91)Gv&q>P^eCTg;l#MV!s|vqsYU1P;T(2w76Wt1^IxOwnDa
zWkr6&uj<N1%DKveiql!P&y}eM)FU~0%a&@mTYC5zM`gO~wasD7MpCi#B+2xH@rxOl
zFJ(2PGDu|myAZNkqN929)P{WoXXK!m?H<3`_%mX{QMI!U-+Q%&&r*M-2FoN6NDK?Z
zWS!m<^0YU*eA$Jt7od^3)M7mfY)x7F^<6=xfIwkb7&hzho;v#xqzb8<#SVj5vXBBr
z8jpT{UC%1nw3!`W!-xmC6D4IJvY<&CERt+m8x(%EKWS@@fJ$Q$vh6zoTR{;X3Zu_p
zj-Ey}MKL8&)_LW=NK?(7(!Que<3cbM&Bfcw#jE48&G0c&xGL@NmPe8J)#c5twfEEX
zTQ1>5z~SUx=*LcM8MZ9<`zv_mTz6YsDSsD$5-LmDbw;CY8eF|ilIKo@q%9?gw6uCW
zIz&EDuqB`npm^KFY)&oRlAs+bg_b>KmwVSZLn71f%x&p{O1Za+>Asp%c~#pMgWLEY
z6BUGN(pBw4B7)|S$+eIs^1wykTP&Bi>nndozd8gy1f=(Y>0NEaH3&+yU2%xD`}@q0
zzg~I(&+?mBI}i$$y;|7<%=%bFa~=^4Q^kR9hlF}Iw23gJnenP&Z^2l+K6w9jlvC1_
z2)J4NaTK@6!u=6UZ8C0LKXsU)-7@aEeeA&do%oANcI7TloTi&%RECp^ZKY@tN^Wyq
zN7!@-nG=?EEl-tqLndA+zV?xMDZeq_Wo8u~YJ_cH9wO|~CBK@J@k^3wlOg?1L7=l3
z9ha)Bd@%NkYM*ykK)t=MWuPK0Tdy2(?{qCL#&i!{$JWt1a6YDBaoSSgYen}@+OM3d
z8=E;H!MlsG=NhYPJMSF<PHX&UG|C#^)e?mWR<v6!29o(87a@MAl`8g7#Qy^J>zYee
z%&D`ia?lWo_ePbBHtwZ+t~<d*I}4i+M?ne*Uw38CH0%)FsA;SqX5kcVg5&o0xVUB2
z>3$=m#SIDmav)PG0a;46_&ao_CLA+q=Dm0{nVUdu9bLd~kY;z0PEm$OFl!2l9;%CU
zt6Q7;h~T-UREV{~pF(*iuH@|J4OQ4@raxRwlpT(I#d2$RMTpDi%UZoECXSV#ZM8GG
zzVqK@W;+tXeJ(SvYMbWAC@#c}jkJHl(bJ}WvAG3b4LENpmYg5)D-2SHi>qom5xes|
z>rlWzP*$~GT9O;HlKsreR#aputVQ4Dp9TR=21%n)m_}U2Ubi)>GrDPKYqnOn((mSM
zln;>NbKh-&BJUPw{f?~xj3t>LbUSGeNks&S{!Osx0*QF+gvQ!B%6|%;0X37D35OB)
z9cxi30Wl91VP^csTYls=iDV8d2`Zsie_Y}36qAFW2&<xgi$=jhj#p787S3`6UVPGb
z7?m|^k01^jA$|=`BX~-%NW;LfX|-fk4RMxowp(Mjw+YJjQnQ<P08}uV_Y{TawzTo>
zA${Vf&%nFmA=hZ{Tx`P}q*W+Svp@)#RhBzmah3lvW}_^eN3wvLP_dpDc=cli_A7Vu
z=26=(=b5Opce*O273HF(mjc+z(JEt85-0(4;U5E*r>n6@7Ls4g1*fPThzOmIxmwX3
z6J^i72VK3H?WiA>3->F^dWIoR{oP6_SAIN94HYP*0aF)XZ`Kgz6ib@sWD^nw^Pcf5
z5LlhQfo&1_5hw=DFVLqk&Iht~*6vfK(uVM}j|+y-Jy6aDhz^tM0iFA9?8^10<9VUc
z8Q_4TO249n9CzrstpuJYR6{=CyAQ|eR1q~c+CZWqQ%CJxB6CJYI+>uGibs_M6;#Wx
zve>#;6Tdi9HN-K(&W<ItCI1-I6QlE}U%4~XC)+fLs9TqmLjGW+Ni@QJY0Vmqb}05c
z#S6n2FVHD)|CUz&Gdva}IE#~(YD6*7`s&96T_}^#j&8$I_*v)FPVmlBN(i3=A};6%
zWLNG{qz2qehp;tS=DKD#3q`O|MWNX3aiQkntr$R~d0e5{1IO$TWyCjx-uhd<TZFtA
zP2pI|!ClLCMf}p;bQhiz=;<tuon8O;s59SQ+$P4QRmytK@RXw3vr*;l%I0?DCFANA
z>T@qi$L$gOlrM}N<qsg#wGrO7c3wo-`7groC?ouuc~Vh7l((5ex^5clFS+dsrED!K
zlR3{Dp&3LQ&KVH3;(G-%J9NO<NtBa+EJzDHK#2d0Q^MR#kQ2hr><A8^BOq#hlP_Y;
zFq3XbN?LR$d3E^0yNaLSV)^F(K;DluHxV~Rd7V6-HwSCzj24IQ5kSo8KKTBlYO`>o
zwxU>eH+*10LWPSxRwI61K_}JcqOIH#qYg5U*NYimj5-zd2uf3&ngqXU|2e4uUA)pr
zPZU8|u0$m{$4SSFC&+^!L`>bD0zCEWedhJ()WJCOK?2dDph_&H0J>N_RZK^?{qZYl
zB_AV!ySS2(SRrTJJk|G&j=CKWHM|5QjKf9a%@E=5MV9UYsUyO3BCORek7XO3T)4ya
zlR4M~W3Mq&DHf*GqHW-L2I4=H_TP5PUCqSKk(gNL7bhrBlVn5=SXjr3g`8%I;_kxa
zz9r=4d=2+^xsf<8jBh;w^4dmTHUD)IAyOnHY4#D*ifyroYZ<j9iDB%{K(mOE0Wn?c
zn8igBmLYc#h(*O%OSMq=wLD>YnNxwafalQ7*GGEt1xn}!i!gLi$TF|d=t?bLG83%I
zbYd8qjJQKZ#7(Cd#N2Xrv7oYWZqZ%&Zg%kFL^aX3azP#0^f)S1puq??&RLD@v?h7!
z(awYUkM7-%vY0Ns@47?&^VAH4+-FZ0cSE-QGqwixAW#~`KF%cl%eSS5oHV}~wlfiy
z0_MDvx6Zb=>P20`rY4mgIgH`DtKV@ax`>(qM^ez?jS>~(a?tnjn#nJjW}d#v7fxUf
z6RZ|yw>4;n3USH*_E1I6B5oL*bOe{tS|<6h8<<~(9}W#{s1J`Us(GP9d*GUK2lCrm
zLzS|EUTzS$*8CbkHsParkx&fs5ju5~|M(<X$e@x`S&50+tY2Sfn@nm%h)-_R?Va<T
zj?Q;E6ZP<@$f7Uf*JW<TnBi21ThT;@NUmLJe)_lZEw80l8}>o&fO*3C{9&kWWP1D6
z6XfCU;`*>)SoJ@5Xi#*ZX>n-Jj%MxO<iepx>sm0f(+9EUoCHp4c%Vb23JztEI5
zru1T4-EZ)!GFSEkQpC=w#r84xv)!Ms!ev_+G3!fEqA<L8*-jtWwDBy%s*-6BguOcW
zc+t!VoY2W;1=d4bx~cwpMup{Vi?n2LIFuS(@m>Kvw+xcsNcqZl<Mv<{hc~1V{Rvp$
zhzNv_a)n-qKX-PF#t$V01rJyYi;u&nxG-0lI!))GmN6jxi$({YN>g%BKCRuC<iu_#
z=$K4h%)fduxDQ-!?4Oda%?n!2W}f^$Y%J2V03Dv{q<v-f<WvwXLNKD(tkIa`=sw<W
zaRpe5eQ<w`00fFO07LIFLo1g8TuL8`$$`0qE$feJ-o(`7^va8>r{O4@O3^R(ySKyI
zZ0CE(Qb=z(9p2@yxqmKiM3>w(3_w<m$MTP3pOI<j-a=qyQ`m*K0zW9<X>jh5x-0`P
z+In1{dbfOCcaG$Ac+p88^JVW@UkN@ICz*h}J0Y)3M4&av1jSBlS%IP3Bs5$+d6%D|
z{<8f3HmdS=#~{~~2uR*r%F#8o*)gF`%Wh$Y#2uC3AvB(hL-?vRX9O^_YA^=D<bIg1
zthWYH603Lyz65w2NYdv-G<1cly&}A4BpchI&#YGThBWD3pY{<jlyXEa^aHYDodBGu
zaJsaqZoiv|R9w11*5EmnUQ~g}-q<}vY1c{I^2lf&sv@|mw{U;qwQc^T7qSbqlFnN;
zZmRbW1;17QO*iSZ*O^L{8T_SDF&KXd0VyA9J4MtIJu6`^!j#Y;o%V99QWz;ag)9zc
zFP1H^pQVZG*y*)Uiabyy<|0rGRP#}{F66nEIyo6u@E=8(Y*5r7h7qRD8c}dVOrUd+
zo^lbOR((9!l-te4V5p{)K|KvCI9eTxV6%h}QecOiPOnCwsKwyID5x)y|5W0V-{KVY
zj?OTN1>-@KWw{E1&|>oo>V4*1KL@rnwy4)g5r7k>7A5*?&<Hz!uqSg(Ef2R)!$O6Y
zFFN$2==vAUA;RpY;mKl~nTprgFpVZ+nE~|ke3ml(?ARH?O^tfgQ_sRu8s=`BYIO0s
z^FrNXdoZF-zyB)9`B7>EA<HrHu2ogeU<GMaUvAq1pNAWgA4pq8_OW6uzv=0rqF*U7
zpX7;}(1Z#J5fRNC{@A5(`7v|TdB9EA4B45?NQ4?|-fdJS)$d?}L6u@BD7ZKB&L-^d
z+l1spwfUMWZ?E({+i&+P<<%F#y;g5Cf7>gKMrtD#!Z@Vwr)ZO?VWrCQShDGkLJFW1
zXfb6Mj3DgHlVFeqSh51$L~&(e*;D?TSOTVK(<Almwtv>8DJYCR>KQwG-^KQPe^_b?
z9<fUgAWRAn?M45LM$~ezfa(k-B+x{)1SKwL?lBe|z<g}%6YC9EofpApUHHS`l(jC<
zd#gM4(0sB;2n~(fh9<8LNzlW{2o3tPkz#fL4@mAQZ4OeAFoEK=4*$KU9`@6N$PS}z
zk8E}JnZOrn0vQx_vtB$dCzs9bt3n9dGK-2di(df?X!O?)tCd4^8Gbe%nv<yFQkFk=
z*T3e#<%U^wXsMfe^P6Y`6N!clS584s{n79QAuH6P4+6gaP^~jWs?#?%b8>V0rnfV$
zoGV9+uFD=$e)Ix-fArrHl;@<^FtEzW5hhv?&}hoiXzB-VqK)BA;gKu3$LP^+0dok!
zl}`-c++T(}=FgO&<F7KNCMc6E=`}mE0oI2PxZZ$*og|97>4Ly1zL*bbbf=t#dY}`l
zx8xWXh{&yl?``#AtM>J}&%B6j^vtKyQ9$UH{UUTID=-`Woq~Yq5b*AlM14Q!I<ZI$
z4{R(!H4X_+NZJ1HiclRA#(mlSkmBa@Hv-~)H($xtr(#P49(v&=RYS@k#{VVj>e6%0
zc7n=Qorhf5Wya$3M@;|;>Q;1=2r{3!ro>9m-ahy3!uP0#4Xy;bEZ08H1l5=9kce6C
z1f4i{)^As^deF}ZYD5pp*&X&8%8{_wU=~R~9G2NC*d_T0!wWi6;fkyk<SP#6>H&<R
z49KT9g83nNsc%s!NEYd*PTPZcJ~lZjIsBg7_kQ%GkMNGJ&i@%olxQYLfdwB)GX-9q
zv?k#wWp~8Ta>Ami&G*e6rB8mg;wL+_xVg9jV4`ideC+1haVhV($S9m;0&$PRJ^4wN
zCXo!~IX&1rD_nOKfrJp|MY^A7sxO_9!6OaYtEFn`&lS^c{V|P(B!)+`MDy`<g4yO(
z<BqoIF{+n^_X9NmPZo4-v7nJ@2Ud-+MxlfMOyJZqU((8Fn|pPQDbzfmY3PNVfY9_m
z6Rw=QS=<tqYTMRNH1VBxF}zp~nwe@DxtL;AJLqHDmizH0t`9Rq{+bE?Q;&}Vmkh*o
z!5>E*e$V_M;9`0sO9P<iQe41JNUWWXHrzNh7$wn+^5#WiMVMm&bJ4>$ic%FV=^q>_
zuk*;qbHsP~r(D#`!x>+;j!*2rZ-=NGT^Eg{Zw_Wksu?j+>~J8%PbmX`T)w%oos=H(
zXYL!I{ANIb|J=3B|2g&0`t)-n$?PY5?NELPQvrQWIVwh8ie(P<rY-y*&?+)Y?#Yuk
z7h@>v`O<Qx8lk|F`t~xuyeL?T9A(u@x%N8M6FskyYLT^V;U<iVM5{X#nGnyA#|xf$
z$%YOkm~_$LOM^>SwRh6u#5NpoN=fKDcUWS7;kL@n&sG0tcnCnIWU3N0T!s?&4l7{}
z=>+CZzf){fY`ks6<11?8KCac*mxmiPN{X{B+oyJVd#3TVq~!Y@2jcAFCSj`BMX5&n
z&5L!I`}P`px-d<V)g|;Dc2}YTu#Y#-HZKrf(}l<7oBw;iG-zxbN=HU>k){lwOo3%$
ztSYE`vAu&f<Ix(D=HS#(CH<*q{*cE&_{^;_q&Je69ZY(n(u$la=8RRVXO{9!SG$x+
zLGMxW@lVO_Wj#jARER`+NrO2?q4cWD%IoxfxKKnw(=lj8ThVGAg$d$3i}_mN5KKr3
zYsnBz8OwOXH$_y$P{DD=Wc#tw#dr&DRx8cP0YVe+vAr4!`57EhPUWdC-5RIt9WX&a
zEBJx@EWYy{KR#z$lg%@an&CQ61*@;L3z~F`{l&KFxud)KwJ5-7|F-0Te`XbUP0c19
zE7%%Zn$#1<WAhUXKC{k2_=jsl?4|P)`=7F?g5zJwj<(nKjFY%?3JH;Ddy}W$%5Rgn
z^%PF$6P&KUvy*uryxV!ZFo&)A(UCKZ-+tTqZ1&2n_x@yP0Lwn*tj%s}(JiDX(o-){
z8WR{-7yhC|7xSXfCLbUd2XtSAB*8SHjz&@7Tp=%wSnVYsuMSJaJH`wnXNDIVE!anx
zcQ2FGNTy)JeEoJpW`t$M0&mfoGYluuX=Pf~xUjb{D3lxM*tk&zRVev+#=mZcTQB%+
zW7+!2yfck{w^U1poQ`6`lJigfWH}jdnEtiO-%<0ZFd?YDmjdnV)v(C}jc3Gxj_o0*
zGbD4RmdR2t0kbwUucq|+iR$;55l~`T(Ppr&h3D4dc!neScmo@{*h%3iu7qC<JQT-1
zT@5Lni5(fnBN0rbAnO&1wvDbo&u9Cq6osm#eJVv?3w2Y$`A(Jw<(##z;ZjDbz6XuX
zhsRe-c2(hUU_`)Lh^SRL5?)D%G1i;4Gp>x*t&FN2$}?_S2&oZKwRkp4ZMYUzb%?vz
zj4;q7P8NvO!OxV~Xj@tjGZY~yDXbYw3&?G>sUVtjv^FHP9SjL_QyZ1S<;77O-Q8z5
z4?V5{AIGTYD?X1APn4Pf9a{G_iQw(_?xm#FmsLr38dgi{Q>(HZ`F5<ukaTKAwN#@O
zo~Q6F(Z4r~`77~KjYv4OWMCz&ekqZ}CVy&sCOKbK#kVS}uatRW{c*vS)xnke6!VsA
zAZYX`j+SHVS@X>D{im^4#vS%DAvB#ilC&&mMu8)-f-R<n0NQpMn3n&zmXSMlNC$+o
zm{~qXoKFm{Ym84V==k`qw9~mm=~iV1UG#F6EqP&5Iz=}T5g4`bI#hsx+Q0h6AX(a~
zWwT~6Mm4gCiB)l3UhGV+saw*=w#w?c5$SnSm(b#xht*#pgd94!v^3y$-@TcGo^^ai
zUXf}c(e!v<htS+;NX~UbX_X|Ky1MG>Yo+}EDHmd1xq8nc_m?VFev8L(mUe%==LJ=Z
zve-g`eo4hkM*B2Aez?Zi?pE~<@hu;Y&%DI1V@vHzv;nAio_^nO8YG#yl_q}bB2~+0
z?f7iN7Y)K-WfD8bd%_A7HoNHNxIl{v4EUeZ=<4<9R(Z^h5h57eZ*|FmL{~6m8jmPL
z=2${Z_QW=QnAd6-Sq{$a8v2A?+{Rl;x~~}NermhllQhR%Q#TPx@oj(u_IP;G4?7XI
z{;^k{CKHnjlN4F=eV#@~C(W9stH-yu?|i3$$b4?->FqpSe_X3+mANV#evSLUy`!cR
zFLJ#t)9&JeO_}Dff2IfKz0+tBO)xCM=-ql|$@W2QFg&Bp#p!N4yA>gO1yBCqlw49T
zdW+ruvzT1Z0A=~skB>{)qFCO|ie5ARf;Qrp*BE&e!kjTRBfNm71*evhR4+$b6j5F2
z=?_h8Sal%~=!-6)gh@oVH!J5E2c~ZBY~k<=&RTw8dmbTYfVz#xTK9I-W_m0>cz2Dx
zp)c7$qo9E4tO>PHUscB(6P27q=*&u*r==8&@LP?$fu$S42<zv9-BoL>0Zk1}g69D7
zAT~9ko3}32n8x9NKrcTB2Ofzkya#7z`QmmBNTaG$x5od162-hqG=syJ8_G&X1`9y@
zb^s%K1*Gt3uCUuT-(^;;T~gY4?02xTRJIYL#1B|-doqwCrgu}}SWph!tIZJr%M^zf
zSU<=yh!GuL9MJ17H9#^V(2T{Jh2__ZfO6Rx$lw{*K7z^6qeJH&475!-<)HDHfHW%5
zQ4(PZKE+lG+LLc~b@q>1wjhx5?5j-23hV#c1l_0c%vSp<>#5L8{nM6>HP|fOYNm7(
zsMU!=ty&6ei|84#U5E>z!|_PJm|&q4iZ$ZDO+3_Bxd0=$#W?4~WQj_SkPP02WYcPA
zJ(oriUG$KnwuP4bb`gJ2w-P?98;#@Yj!w9;k5=;oLNt(!g4|MI|FNJvvLQm%We6hW
zZ!8hh$)8mMdVT7@fk1f9q8i;6giWP}0m|P^fH{{r8*p>FoKp7zglKT!7d08WB{IuL
z5_B66=??CnQ81(3??+OptK&{+Tk?ld3>q9wK<GYl^?f9@oPJ_|tO7HZ+HYnT5*Hhf
zO1ukO`)+1^p|nGRzx?-Mefkk9sC0B{OpG%Fb0bn>QiRxG3THgd$rI08VRCW=^tc~z
zzq_ZjL{qa54rod`6~Er(nNw7dA_WdhP=Kb8<VUcm;tTs2%h(H8MMC-*I*ev>)4V0y
zIxcB#hO{>3{n|tHx}Ote=cw-}w0`%wuWD+x#)0`jBg9AOMBEVbj`V5)2{XiiYzray
zwS`zmMYg&cCzBWNg7~m?kxGab&p<LJGkdrQJ1u~ICxF1tI0bJivXx1afXq9!bHr@t
ze7RYoyq$#J)igALzBrl1*>SAh%?k8PK;h|$!q-7uN3bnV(8e{9l+5#QZ&WrSjO8d4
z9~DXJyWl50_oqYGFWM)x9_5Ga(0R`HOV+WsHqVnN!-zzXXz17_SNiRkv2G$>w@;7D
zi5|m!>u8@S`o1;gk|&Zi6fy22AoH_`5LcAm-Bes5%tt^t<7(#aW+wS4pOeQ!qH}XO
zIw0i7%Wt;k%?Z6l+&XUIugljMPHmi>*Crdta9K|9`k_y*{<aBoXR}+WsGfl69kXfq
z9voHbWPO67YmHHL9Aq=sfg5BV&_Au4AV-I6CsDx)VMTkwDC;H&kjh^taY}=EtKVLx
zW7DCI*&YFVWN~})bYn<yqHxpr^*>ZhyMFX8Wz6$sMDy1{5zbycx#MvK)=c_{&(eut
z2R^Q1Jf?1BQ^HoM9^o$GW>mkC50%Hm7{3Iox`)T+M9*q#f4nDmDPxaD(t4W46PCPh
zgBHjWu&6@|`{mQWm)h+78I+B{w<VT}@nk9zm_R#}pG4C*-au$)z7$BKaSoByj*YLE
zT&M1=4Z4Jw8U<?wk2^e6oIEa`iFo=cVLD?WJdkflh8=S19*P#I40PuElC1(v)+!+u
zU;J>QsnWuzx!EhYkm0#s-?`meC}DkNb1Lr7T5G=~-Q4eXVs+Dx5BbyD-6o!I5Y@Sf
zK))m(L*icdts7$GnM<7}pMXjQV=?KdtZ{Ism_aGHJ-8*<8pB>JKd0+Z-7aJh>dMa$
zqC9-t4<Q+|f+IR28#{$T(*P;c0YaMXrdT%~Qb6@BQY-~?D`_PyD*%B*&u|>~F}c0J
zv*f_Kqo##>!r_WSY7shRyqDm+$7u8eKEK87_hj_s>9yTJ(l9_>cqZAUD+ZBYNHC_K
zFnz#y@s3d<yA;WlG!58q$6sh!9pP902B9ZNyib)9m`G5^-S)3kM8|HF?DiQ~-{o-q
zNQ3Yo<+T+3PGs2X4+3XG!jzm*QmJSA=TBXVi!lbALv8i;Juo(C*!HvsUsPA?M0c>-
zb%R!rK02vOJjY^R?a0=TY0%5Ltv!OB)Qy^Is4UD)OE^@+4+UPtDa{XEXiZd`aM9uT
z7f_i-jA8O`e_Xnw+YDx<WZBQ{t(V}Et^YyGv@5tN%<x;&dRusEIw=tZud?e!i>`8J
zHFrWy9&ND)QD@}MVM<VFhV{3eljWo7(<C@X@TPFqbo`n^O;(X1E%1k---=xyMY}zK
z-xXVuSU5m6^|z_cq?2OtiLhXCn3=Dp9xJTCX6a?Wq?$d$scdyId>0l+20K8;Xr^tT
zsD?aCc$~+FcO`*t@fE~GBitWT{~r3LGpyKy<=FL?13*P9ssC#W+^)h?rM|2+(r2)<
zUft7VH>)=l(N4r4`oPklf@pRHVv5k{q1{LR7$dmH#N``YnjIAlwq6m>z?tQqy@^Sz
zaz2Ko;7Mx}$-y%7{c=x35chAp+&z2ND0w8Y!j-Apf@HBPedWXLm{Mew{C6fg&%BBJ
z&TSEcKdYo&EWQI{LXl8;Hi3fDq@|S8MT0pOfoW{6Emm^+nrJQo0@4!Z?`I`ngqcoM
zAfqe|BvM7rdhXhmLIz3B;GC!1R*=Y6>2>fnt825rU?vSf?R7FBjfLySA8DNCwj|hs
zo(YzHA?lBt=;L<>-x`U%j+oGrxg$Ai)+;y)g|Z@2PzZsDu=ht*x8D-kpCQ_}yxmOR
zt{#*)5Hl`Mps<&)EIh}$)@kNEOQZyW=bCb6r|Q7L*4NP?l#&6JN>^}>2Agv`#4Swe
zCg~z_Q95^-(D51+gI1vvo@j2y89{XI%sw}$9lcDNvFPN2%vU!r29e0^ini;tL=wHK
zhooR}=yX)sK~UJYnot!8C-Cc&n$Ws%o-geUi`2Tn+88)EVn*{`MKof=Y#9-<;2Xk$
zpfmz*T!^i8G`f!c2&{^kNJ<haVk>cJX3a|%#$jT>scgL95C*m<*KNg~ySR;ZCoSKS
zwXA__V*LEAA~VKsp!bCC7{we$Ky!hEMJj+Fs~H{2v565nHY?ut2y7zxbwivk#RxyB
zGj%Fk;H)*{P8N>*AtSP*80S?e?MLVRyY2_+Tq6@v#0Frksj_82$x&=g2A(!Hn9dCq
zMchoz`hhtPvT~z&BGiaEzqW^ScEHx08qIPt(5Z4m^*%pEWg?6gz1)|9qF|4`zOZnk
zO0jzBp(<pJV_NhKse}v(p}grO4TM`_s%Aj>n=v=@(R7EL5l*E(rKA=`iv_ij%QjQ!
zz-+Kqkg7z=fG;ONVDe&@FzmxvFbgu6xVn9@6$1!|(}7U~E<))+LIX#hb{%>mW~`mX
zAyV0s`PJ@}Mk=9IPz71>B*a8Homyc)ENhl+_uF=qcxby9-Au=8;IYX`iP!3>Hch33
z`L6`B;>dARF{?yDph2nfc98VUY?b(xBIn<MMEQV=^d7kHu{j2b;H??C=iAcQZc+Tk
z=KVDV8U?3`n+^!>k~ve}`~ov+{iQh3@jNBFd?+98CyGWnafK(IkJWE3ozT&Z3MAw!
zzcK#TLET~>M~K1kLmz5>a<~Thk^UsUAq7MxGsC8u@XgFYMPHV=cf08g5gxN244N7!
zXp+COam{nnISO-4B7Dt^L^UKoT*7C&$CX{`b7#D@qj@r%2FEw2yLKO6sR}YX5lLi>
zkKFqK%P&;ja*D?bWZw;i{uDRiDXG;`^^7j1D^^UTwTzEdx#8jQ)QIm=j->I9@F<z`
zqrt>IT@2qOVS|l`6kzWV;C5=QFB5^4kG-Xp%ZF^6NRDkC?yK5@H8=^rIqr4_fV{lS
zS5-O}OH?nU;qh`ko;qPfEC`r?uXJ%onJ9C0D||D6eK;{gjGwRg2Biu2Cnz0c5oInj
zZ-B{<6*AfTp469RANElRZnUa%(n(3iOqr~32cRnhFKpe}?<GrNru=M^Z?j$1&m|fv
zipEAvQWR`ag;-hr6_lztp2aQt(^b(Di)abqo8O+LId8@t&1Ei~xXCx>RiguV*M1S7
z%+w@X#rdj<^yzj}{G50yWQ_m_FW8ZhW6oIofq>`^*o=%2C%h(+7PD=P5}J4IG^dF_
z%$h$RKfb+!krJNYTf3(uXVY!WgRvfjxiX*WYBGwCkEPmRMUZ>l;4Q8p3XH0>nG4)}
z@CFxE{CI~5;H<)FqhN5F3S8-vnmN8TefwU{Jort5zWd&v%|i3S_Zs|gS;lEzAzf?T
zsrdZ1Lhs3g`n3y#@6AOKp%_m>7;KZcIbNHrP+2b@VA~P~Vqv-;Wm6!_()j^I<bqtl
z<xcQCh{=P(L4>}vCsp(KpzM+WIZG*E+qx;x<!$rni4V`d5ehZxeCvcw)}UPY@>QP4
z<&Q(rD1&%R&-3ezx8PI7pF~ewConHY?L?Ja%@z!Dh8K$dBaKT)=Wi^nk#cIN&NlLq
z=#_j&aaf*gvglh+Ak9|f>rdA{|I*NdgUCsOLwp1I0s{ScO$b6KHBc`B3j(ri0Rn>d
zS>kHzU_@(eY-FxaYi_M?YRpVyY;AI;ZE2+>{4I-ys-ZE6+P_L5=TS^@H`)M=(%*`Z
zfT9xCKmk5iPeLd_?X;X<I^4_vI!#zfC;@=DhxnU*2%3-{^ux?k7&R=-EH4kHNa8c1
z6-&L__ME%Dzh<qoc`eBL0{0p?p33feKF(@7IqsFnP^d&pppVM{4MRJwR0dU|z1$>L
z%5ZnbEdj)-oF^w&0z1_n+^m&gCj4j&Rfxt0)Ia_KCR$t_$velNs8}i`YZRwltrbRh
zpmZzqv8NFpm&EccJ&jE#(KKXhv&BjjA{7eG9;<$D3&(RRZ=b<F_R$y-&C|cH#VUJ{
zefy$4IiS7NWSjHq)JGBB?Afb$!a`qV?KJw3bn#uV^$k??;>8l<c(0&q&kz+8d48=+
z_LO~MD#P^Z@QFUD`S;1f#;`I^{(-%ASM@cp`Gq4fT&v}nDO%DdV_57>v7W^li+cZ&
zUJ9G{O;#>hqo&4(>%f)Nnq%U6lcAV<ib^N>#4|<k9H;;4sMYwjocs4j>teO&%bCk+
zllki82+`W)$%BlAjd4Z1tnP30pUx9kGTi7yV=0rrl}Xu2xnm($NsZ1(Z8eJTgsaUP
z9XtMT)^i#D3L#@Zdh6)>J|@$QtJ<-m(ti^{brnKydUjDwKJJ1QTnh}<Ub`oHI$BLS
z=qQ}U>vk+(_I5{dv`n9BExdD_s9crsaikpfc5hj8fT=>AdTKDP%pObH3V*CWD2(k}
z3|r>CxqW8AdB7|*8Porj5RUb0CZ~FCuqgz&266d(&eh^gtBa9+HTxn%T{-3A?1S|-
zQ_}8hJO~!ZxR)}V-6kqMU*E<s!H>`k%cyyNdW7x!P=p#(+z6f<O;9bJV(^Rpb>qwc
z_c>)B4XAT-9TB>@R4qo5Bf%{X<F2dqI9;1Z&XaEu(JO3Z)y3|JrQl{@%Lqg8DDm#O
zd~A+P4AzJC9jLv}b{BjbAUXIJ3`pQV#s!GJ47a2*xTB>$H;s3&gPcl(*z<uo!^HP{
ze5>OQvg5=6sDhNsyF%0$-+0CyrmQqAE-i&<FaQnHH9%L>=38O!;>}`d1)pt=65}ex
z*Iuu8Ru=gf%=IuUDr+NkK@B8vOJeG*Ip;ZHYhJMj7`$t$2~(5%gUU1iw2z=2SZM+6
zT36BZs#G+z<R^y~aA{P_!NwPqA}B{!TGDnMlq1+m9525;+f4F*7CeJL-KzRPrj!Su
zt0vUj-&Kt*<bpUh&g_OeuvB?yRpG(WscKbEu#2%{tFEs(AUMq73;8snz0nAM1W}Hz
z6R&Am+uvk6-vtaS!*iwndmtODkAJ!DPpt+0vt5vCRZNu%wOm8<Y<JxHZzNjTJRt87
zLf#POCy{}tnBAVQ!QS!FTcFqcsh+<<{f7bcx9Og*urY|E`}ru)MEIgJw%1q0d_3gw
z(e`{uPj2{cr1-jE`vDA3El8i$wmDn-d(i)?BYX+2YrlS0zx)#>^~HZ!l=WGfCl2H(
zM~bHilz)E@{ZsZj;$PW&4g{f|^ZB2RO?*A&Nb#NdrXROgJy6sX4$owTN%8Z+M**}n
z_-~;jP{!Bt2oN(oU4q@__yFYizrbKWpPc>o$(jRyk(B(!&HpLe^LNT$VW@teNT5EE
zKq39TeGldHSym$5BLJI!x2J$X#71-e?H7c>-)>7S{Q*=%+;qY)QXiNOsR~qp<sCh#
zTjQ=GK@oxU^F08TKUzVa)B@Cuu6y4<k^EsD0pKKm3O4*A2hVdNPYe7~%(b`4{e<yL
zBiIS!TZy9o#TVX-n>Uuq?9;cVenifT^pR$wr!GSCP3`3u>qhLphqZx4DeQ`1{P<S~
zVqE6g+%85@LGQ_X{oFUB*XHqjyRN_+lH2U4&vS--Fwb<(hUQAY#qL7@SgfhMgLgIU
zo(|=@8j~(up&qjyAhUmcMn+x?X4tU3NV^Q5>R25g%ECRPbKRTyH*+f!Zkbc}Ucuym
z_Gl?Kg1iV0fWTbabWulCH!a<=eQU+HwX~GHm{TXqz8UZ>_x(l(8uP}L9Y!ymc|cP{
zcn)3#l~>~Hd{$kNwZ&cN(KX-Oq0$47qZ+Iky8@Eqb?jnC^L5chJ`zDZ>)XV4i#6-b
zBTvcmUKd@Xk87_8?>yo+>bAvZYt=4c-q%1wNgeBlkjkKy^Buw(?OJYxpaQt>YX&cc
z4&0ho6|^5)-<Nlcts=V_bMAn(&QkNKj7%!khS*NsP7qpO_mfxUCtvcb!EYt;LH1W~
zB`b)*;P;cUc2fT&|MbnR(N)S9gn2AV?pUtgNpI+~?#{IxUGM%$xYaN>wv5#!e!6Ce
z{Y=_PcAV&rKbZ9+YGv$gitkqunf&=6>jpUYGig`9<qPRz>=f$+R%_R>G3~wOBWb1W
zn(d!-iCI?QHA7Rd?d<dYa&VWqU6(oXYdFT%z$wc!>3O8p)&R%#Q_-p8Bk5J)vSQ|r
zYJ_tnC$@EA_hI`KE+1-76QfDYXApCQG&2heN6>k7Ln=TTLJhx4fAaY~%`+3s1e9Kn
zhR+?r_q_}P391VG9pn(y2ULOAX?CfDD>;?+nA}MORuyVSH2&`&!~^4L{{)GE3nH>S
z^1uI^JA!O2?k^CDl#`tQZfOI3+7BNdROTKn19|?UC^{=Y4lpJ^^#^m7xKzKV)LGOt
zq%FJy-ns0MV807KD0P{xJd5Dnx?u<Tz%hCzdEIkRHyN%fF`tQDAFut8eu;A`xfmae
z$*%XB<-2D@xrfu-k$tq{f8{t}7;e(t@_+r-bD<B&(>g@37V&%%+tM7Iw*3cW_n~G#
ze5?O6S?dmZNbzUBHiVFfuH~>5+O;Di&6{kT!cSKF<=~8MJ$aZsmnAcAwG1sj-=|$T
z^{*_|2=3H}y*_+21r);~v4Yl~VbeOy#Us-~dH5ECoOFh^K+?A+nW8E{0gKV-zyXQ+
zBn%ii4eA@gKDZX7`*0x3rCp1aPKKj_<J^-|9>cRNBOx*#*~r2i>0WzuxZ%cI$Imys
zzt*G?JU?@EJvVuUJU}rnQ4Zc~BYPbU%sDP_>kKVkYHMuaX-<-T2@~8(+;Y{1uj_cE
zIYCMWYr2TzczvuYr<WPL)+W=FT18sVHh7}J|KRP_p}n`(injDx>l$<=(Lm{#zwcnd
ze=4bQ6}V^NJfht9B2{<P<1wv}jI*s4j??}5JlK{M9)MUi25*wxY8wig4NA!`(9}7|
zlwfaLi^m3X`71gb>Bg2kmUTr_3|D@?5zHh<o1uaDhQ{;2JKIC-<AQlz@;9iFY_Hm?
zYWT~WN5j<AyNIVy?GhjU1=sl=+y2@Xre)4zvlj#)li#4)ayu=Xm*P(*V4%PR7!P}e
zd}JAAkH7;7&jjF*pS%O|>%sPLcd!Er5aMDiqo_q!0v@vy&y_9jI5X#Q5tkju8V>Bq
zv!Uq*K#D$~KtMh|{^#NoMK*I(h!q3`ejoIIyZB`KxA@G2OVCzMtdjkyTmUW1_3cL{
zdJu$h3MIKaGFU+)2xNj2t8`yK86gA;0iiIoUbGF9%PBCVgplZG7^Xk@az3`1PIxG?
zozoWH&sEIqVd5`Bx0^#<UVxVV@_p|u9!vCep!kBg1zMPLqv*u*lGZQaQ#JF0mLsr!
z$}qZg!D1|Sxr?Pq<=6OVL+2@Y<LVw@!9sEK6+pIVVQ8VfHZ)%pRldNH$KS_1KVe<H
zU}I6fP(_7%aZX(vU7h&~jKcb25kjnpG>*$sWGV$jOE?wN^ty@WKI0a*YCj9t0A_|-
zBc2?WILjZo*i9{WUEgLk?Yfn$EOyviOto}*h1@ngB<ijymUpJ1$6JfEgZ(_{_ptB+
zBA%?2S93KEc*#v|dfinnQ#}-~d*3{FMn~~5jbh`MEEMsooeOT>cXRM})xBln?p@rs
z>eq7()x9}<;J591dGT~tl4Ag>+?|hcobTs3PLdZafT34><O|S%Ri8?hYc@-9U5HJY
z!KF&6HOlBC%pbe+hBUi+(fH9qznf8t$re-=CQI)YlNO@(d~AE|$Uh>xW!jJrq7<GA
zt9XODZVBd)pXJm2{~#1bQOP)9qe;e=;Te|PzcAylexHe2GcB6+r(V_M#aH*Q;6s<Q
zujRn>!1iEeXuQYyFtp@3GO$Gp=)WNaj0Wa5=dUGe;oR#Y$Rp5`0fLsbLMJ3qf~#Ch
zCm9f|v(niHrX3xq0`{D&bkpc=X>PUlrVn=q2!pfY4JmL;_0pN-IZ@%gz}l|TiK`k~
z$?Y@Z+2$P`G7tgk%N_8Z*q9!;+)Hc?Pv-;nF1vD5BldVWwt&NfZO)$Q3F6kF+zD6G
zC|44MyWHJr7rW2=sw7nnrO~5Z3RAuI<V~5TxE0oO3M<~CUZ&;~^%Ihf2$j!@9P%e2
zUTM)S2phgQB7>xMc%NNs;~Ak(W;GCv!bT0wa*UiVdP^3B=OzQ(E%V^4v68@N-lS3U
zG`d^uA_-27)PhDW>A&2-2(r4APC86Bx=a117%ytc@Wtd3n(h|n&ya1@^1|WzfW6Z{
z)VMfvweE0i;P9PMS(Z4fpPh`7z&T-O=Ya%BS$TtIKnC03jW8ovPs-qklK$^{gM1ok
zInLx}Fe%L4x-VSauT$n+qBeajdn>3rDXJCH74f^!0L~2rv&}wV)?xfW6b0~r^8f0I
zYt2R!5dX)2d(!Ij5Aqcq4({^1eBU&&j$?+SwF?;n{cQ%$mcJpb)dwaH)IR|x-1cY5
zf5*fVHA=>c2lo7Hu#TuJ-XR<QYDNs3PSU?B@TjZhSC|$=%r^HQpTo71$aXJ{b0RpH
z3*1fsXC8Z+F5>z5BEbzMy8jf#ZH=qp%#d%>g~0BU378Df_0Pp-ef|N!3Dw%ciKo2H
zunc4YcH<cQVQ1G0z#DA3BKn8pkQOBb(?ceTh3&V`Q7xtihH+WlC+5%dFa9#@x8EMU
zRk|!b)Qi(o3{GwNgeeMO3RN`MbpP`u=}q13h|}G24E`nG+;aS$soo#w@K&}aLrY49
zkiBBPKO3p+O(appC?U^s_GWAL5Kw|+D5|UTqQ<(VoKd|FXSJu(gRFFIho{rE@XrYM
z6X5t@g0}e;pRd~ZVr=+I6h_dtII7I*|NIO>9t_V14*$!g!+_7Yr%x{7u$}ka+rtP>
zwThVC`2G6?`d=7a4PV?Y@#t<ds+AjB+3h{4?MsgKQ@#Q)YTdyeNj^avAL&Q2mrT6W
zk^oeDr%w|c?Z^A>JC}c<J{r^Ao{!@^kVIAg_(z$hJ?PJ~6(&Pi?&McgPofNKaDJU|
z+*T342hA|+hR~Qy;~Q*A*nwlV)s}t&_d^TIXVw41z65Q(>cjr`9cl#v%WFD+{->-;
z_TLE5`0n*D>tud@xhEx~f4Q#xAFjhkMeDsVT;}Wj6X4BY?<ULpNgc@@wFux?Z~0%K
znIY_lIV{zKNjTXcg4K$XgZ4a4OZpeEmxg+O4r4QLmVQJ<ll^}Kce0FiPOBD7$GmO*
z{U&0^{9kJOVWR*{QG++5yvpkAmwniU*atFjvU?9vth{OdgIp)=A0s@mqHb!u9Q&pn
ztB_Z<691K1`R;jIZGq81{u_jPR(q|uzx1sLG5pTb4Wa`QtW$i(HBJxJ=uY3XuLkpZ
znGvzaVWk_|=>tLwCU`M9Udv@)A@+~i)&OLjZ}l07)+hOO)$_OjJ$8p`L8sE?rrBCD
zpw_AYB9Fs-1kZ;ZJg_b1YItQ>YoK4ana@(v1e<g2qXzg6jsyfo;6(JOaw3%y4>ei7
zWE8Y#l}-sZYa=pzzv80FeJ35T9IcT}r^-I=@vIH9E%QjF9mTb~kkLJESAKgeOMdGz
z344$Fp=)h$cF>|=dsrShkFA;wDOy(DM0fv*DgEgrcE9{0T2I#*h4PchtTzifpLguV
z#$u^<sVd$ldNAtZ3U+pRfHNp~HOAE@dlhbut5TLMsU{N;FR0o_%5|BT9hPIqRd<*y
zqv>XfWSt(Uy<2pzcX@G6H&-^sY41I{;#8;5lw{65boLlb$-MvSKXmFr?cV6p&NqKa
z|ES-AjWB)t!={V!pxrN%`W#kX)is-@0d!w|sA6j&@4nlY#Q<vGY}}_a-H~4%I&~ae
zv#@@~@Vt;M&JHWQ^}BK_EnC7r1QD~!m|r#nKTfwauwSP>lK^$Rm0r`by`a#7Te`};
z-q~(+(_g*qjbu*Iwsy44pSH`yj_u7c@1>FddPDg;={+>QQT(ojN{yn+TKzUqOYZ!m
zsZ!}|{O-jFVcykdht@i&x^+o@MmiI?&PtIOnHMCU<ac3s^z(8r2l%f4+6Acg80-T4
z0+xO~JtRJeELE^7R$vETfeQ1!#LBtA>6$-sbbb1Tx@SAMg>6>c%?qXShRgEh`avoR
z_xOEtHm@6B=KWnC)h8s<@0GjU<udf4d8wyzz!e+#`+}wbYfC>s`|U7(<qZG$3h3-a
z`JN@n`L+?D^eJR{p}IR7=g##?$A|L$o?G+H%N*6Gx<Cf|_3qTR+FaDb^Ox6Fc(<3W
zfR7Je>5%KPo@)2|{cel0Pqx4so)O2apA>&X^?dAdlR_=ylI6rXVUxa<4Fe{5yw*gM
zVCk*K<X*>g)B2hBslmxsSgaecpbIB8#Ex6#l%JG1A_gv&WVqcN|K5DgwOhH$!7G8&
zj1kAn`ODsdtFo?p=c~(!Xj`>O^6uxEai8=W-=|zE4o!_kSJg)QTafsG7+DTOysWy0
z>9bF{C()T&-K0=moh2e-p|t^69_#wc+Fzwhst?4Ugj@PAYb|i^o=-H43Z5<44QXHg
zHrh?=eG%k5EjN=bJLSADY|&-<+C@_^e(Le<?KsC1>*bfZUV010t6yznTHT_Ai@j!j
z(w>f;J#t9p8{icYxR}*aYj5ki!w4MgT$sjgXj~m5i}bOcp?iz?RZG^{U2ldD+;mde
zd&F3cri}`HQ`UNO+RFkmH+T8ueHBD}UGmt|InPu5daiScx@Eh*hBw~1o!hpVG2RmS
zwK){(&51xkR7R+S^y~|9FlnK~lg<g%$z;KXX76AQ%dU>aa*Yj-!seH@<hJr!Y@AlZ
z2f!)z1@HXq3np^;+rYX0z42-gq7LCXxr>jNiS8LX5IbAO{BdS4>*YbDBI`L`gi81~
z!ngMRW3S)ZkwLkmhV4jjA-`yBPr2W;>_uzkMh%5uEU4I`*TFO+gJQD)o}G&lc%1$&
zhxA*`OMnBeqOYgVV}^_?-H0ep^UU6=h8T>4Nu1GlMUB;lntN~K=-11qXT!f%(HkN-
z!#5Q0C|6YtU74bR9#rtkcoAxDQ=C{!`qTq`r<!3w|EcYD-$!Rl16WvUwBp_9NKm?4
z;Q+@CH*$Drj_9q`mUQuM(rhpfYl$%zC--afyqH}?Z<cF@4d?!+0F5!jwM%tM_cQy|
zR6v&Wzvl1>0FybB&(`1JElsc0X1cK2^WCk~_UUyU;4$F}xl|gehXnSkduF&6HSpRI
zx0>Ss=ee2gXP`p($u}u16{i;-XJYq={alB}hrYF%9@^@k4bIK*u+iPe3q5+&$PfM^
zLMQTpj5@4cyk2I0>`LUim;VG%C0OPzg$;f{zh#OY>^^?WrFWL}{{$fD`M<Vy@Z6HX
z)0E)6S^im&256v{!{ZERS>Qr{c;%$LsQzC+K{B7Fx>wvT&0|Yt-EltnLk@%1vNdnK
zyT{_okOxi-lGaSGv6wo<;@WCBbHO~Rc#Oud`^U!PBTO?LD@$&q<mf%!s44mZuNzdj
zdTOtz-k9%nv<Zu$pFN@-z`9zO>B8VTi0So;(jP^H`d|-cQ$288-ZIq-w%EjU9ZJ+}
zm4(V%Lhi1<0-2Rwak+O@_=yisBa}ouMK~D{5K~Ru6tU#2AJDSCbJ|wF{DAIFVgS?a
zd4o6li*OJ5CcKUn<?VDzc1Gl#e9c?+Xk4y)>D$(t)lO$oE4+CRyP}3<#Qg$$zV0Ux
zwI#c`huP;|0(z%XH<+P*KUj<TIO-w40Eo5(b)sI$6tcke`&n*cuhFyL>}e)Sa1|r&
z=GpbmF36+EYA!$qDx=iS&%Vz+S=ce4OaT(>J8+lh)ErF|&GEGr4PGW*vN(6Pnmzz9
z2Zj|9Z|CXZvtd*sL}a(B7%w!^K)>DZU8|wB@>;0Ii-lbbhFUivX9-ZeXnS>M=tuqU
z7tunxw*kh7Ue*Ssn~q3;w(6(>7kJ2rR7V7%6EUmmJUYnMid(K#$FZq=<7}R-nRl1_
zL;&8p&R$Dq@U4PMCGqcW+xDs6+DJ=FP9uKcRBe44d~2S3E8K^Vqy3d)V54mVuC+v^
zbh`Zqnla-v$&j-PcJbw<0n5vgiuG2>-N%fuy_K#o2f;D+yMg1b`DT^Q@2-pO)x36{
zCw~PR?q6Cn+4*bX*_dUGWPQgEOsm;x_ua34H{QU9sSo3^HmD^X^?}KIlPz9e<h7fa
z<6&`Ik1-hYOEWD1T_6D2(149=>p>dchTgChE8G1|eZlaZp7!NX7wTZiPAXQGuleN0
z>u$x<2<Ez&yjcH5=~8Q2OdDji2z%s8)7&SH?VW1pdO*KNAThH4eT;Cl7Giva9)IoG
zguf4NM=jJw2cqt`VANokU_$@Z{=OR}&i%`6;Mg`BZYSlZm4c1TKGcCND+9;YuirZx
zGz*j_NYE*>@aNc5%--4O)N3Eg&Mp(&KShuYlLeV2yEW)BWIudWHL&yW<pAkZ^Me4a
z)wrYoc{ra85Imb?Ltx*GWqb8(#<^m*bunf?GDc_}#VJ`~yOmy1bvQ2Xb-)S#dZZ3z
z+1sGB$SN;O<38g-Kke+{yit0>S*NKTQa1I_D0_pvl-n)hP7VG6;zs2L?i}LRbS&af
zA?TCH#*<=axI5e`GW>r~_TFJlbX(uQih@eB0}?<*MWl&H30+Z8X)3)&K#COUASIxp
z0s;X<nskvSz4rv9gkGeC9w0y{fdByl34wP4_kGU&ocFxH=lcD_b+KnMv-a9+&tBjC
zS(%vZ&z_l$p_`l0^|ON##82h>V;o$%mXghQy8J=N>wvsPMnqt09ft4-<qALnrm$Db
zM>z{3d;ju7#W^N2g*drXzv&PFB-LO*;5nc7kBPttO-US!<2sh%hZ1HgWO@-f>)qn`
zYC0xD0!Unak0KdtjZbtPmwSY6S_#I4nP~$B;hRnY3QbpaVyVi>0Ypm2@l}F={L<8D
zeyCwE#q_?VcdwjbPdU?g$e-pq{81tuq8(nCY*!TyTIw+h!~~t<dIF-FfJ1<Ofau<V
z!7FuT@JjJ1_em=yKz+3ICP+L4gVfbvqf-TF;tg+B7!y@>fyg)h>ZO<jj;K>tTW4Q?
zb<U7`=B?gzCLj%Ykhu=q9Ph~WYO8U<3ORn7=COWGxHyTM*Qv+k>I=oi5T{{ui2Hwa
zC8@C&NV`*}!`cBsWgSJBIvg}GAgCTDqA`OhT0A0DNhPw4LE^~ZZRMVwWY~8*4wByV
zTd|bN2QaEle{U>k4EbD$L#S8@>QH;I7SwpP?dsgjHZMr7es2Hiyla3b%LCzd=?Kj5
z;gO-!NVB0`=2b@r20&;7&PaR$l2fN@?!=FnQ?{~nvZp-H^)H>&0t6FC-Aoyl^xi2G
z6>bPpe`&?H<#S^1!Wb;mom<nK%^x&ca0984e5hyd(!FUHV}fdKMOI;0`H`H}Oh(=`
zsWxz?T2ouhaCQE1t{<I+5>rfJeerf2l6M?I-+IU*P9@xan>*gN{P$-H_MHNp0+t^O
z!sHsa5{BZw0*4Rm<5U3#%Tx^}_YNay%lXxT>CkM>$A`^N9L7RAb>0)-O}M|HdZ6%M
zTJa>H6?d#xl_zbRQ;e13{lP0wS{&udX0~tN^7rIuZE`$PhHsiknT)R2bI5eJoU0IW
z^uFPk$$Zhjj05;h!44<cjYz6GA9rMkGE+a8F`_GT+{$5cO<Y{Z)K7@NW5FDE9i5K=
zIZvveLnhrY43}`qJL~kf_#zZSFN(2TuQ|!INgc4I^ZvJ}VV~>Z%^H`%Q*(rAQ}ClH
zb*8uiGex1Jb?uUb?1xT~y3h~plh2agu_C~xIxuxpWhOLFm;#`aTZfE;Iz<2tndU$p
z5KI_GAJ@8e>KW6f<kuj~b=&-g3B5^V9bxJjpvN3lr?=luKjXU1WG4liOpdvc_kVT9
z&Mb!jWU&xTsB=B@yX<;ZiTmSXz<z__UUuVJ{YUZZFgG0YAXb>^aJ~RTAUb6Hp%d|B
zn+*_$Kt~O~1JebVtoTnDY1=)c-|7u1z#dVUnb!0f>J!WG6mn;1*UN#o;^?6VY6CG}
z8&eXgGhg>}C-ZxjcS~%*NW4~X>bMUUY>oo)-3RwC0TF<x+xj$AeY)mk4HMHQAm!W7
zBfm7J$ZCQ{+(7mBg@J7o9a1Ka&lM@G3maSN(Uxl5x&rn2;G}h^E&bgYkS<p-;l<&V
z0fjfKPDoBa>>YYe;wqC_Ir7kpd<}VRbrY)dTX)OTgh^NF&CeX!(4u9}u-2BOm-B{;
z`%-3W`uw=Qo9&kL+3>7IbNDUyx>_k{8eV%7lijyEoW|X+6St}g&I5#GTlbSoj-nr+
zrD@ZAn-=i(VrkN%GLY9N1S3E?i=E!H<WABe3(!EgeUQ`Bv@VbqU#`Bmv%d7WZJCdc
z+Q^3=UC8cJEXaS8vrSFvzo7DQgzqLg(~I}kUkV{bK9Lg_<O_!eUvx(;m!Fv^s4PH!
z8kq-HD);R~4IM+dm4q?w?VZQvhG;vxtsE1C0hCO~-i@>}943(Jy9#cQM!pB~u#ox-
zGxoBVYb~cPZ7$se-sjc64HJhzyrb`p*uky6gh5H4Y`q8xqm1JkECD9_Njxy$AWs}|
zhs8%+bfjsoM}sKw<81ue=*);i^{i+J`nJ6lV=ZZ1+wL2|{+ra({IfVn|GsRg+$SF*
zu!q{L0qmjjron9OD(o^KLaKAR5Ol>S$VTt`#~f>tP>!<d`=d`pcq5r<S#Emp(bYyL
z(AV3@k2cwxd^}K+okPx2k!@@(vX3{++!K?=XYs1Rh>N}m@g1^#+_sc0!wf#QR9Hl@
z3Q=A%j~|UdAjTTeCT3TByDx{}r$y#-8_Sal?<MVa-`v}B4WcSe+;d+RbewGy+E>14
z0By{pNhr^U>aP*QQ}^EN0B<X^G|q`u-CbVF(B55!BEu6AP)h$(Z}LRg-rnfVJtlVC
z3XwE(EANfLCh?h(+`VecrBbqCqd*UC{_S@=N9e30-4$b$Xf>iXhU9K~eZ@XoNNMhB
z!ei#?P0ZTNJbnH7diP#gTwgkVugq|kbj!WRLV~o&(439`1^@0BI-y8;OpHtl39k4<
z4I5EzAhHq^JAz9p7JnTC#cpD}V=4aR)_|C4)<I;IMtVXwFG#$=EW*DWY-vo)WRsN&
z^7FC|&L6x1F_Hq}vgRNzzu3^?nU`^(;4PIYxLIY%5?{6A!#LOrkAmfl`rc39mNhHY
z%zTBwLt?4j8gHblx6<AwI_<=7u+S^G)`G!Jv^^~TTMP0(VetWPRfiZaD_n@1P)=yM
zZ&^TaSOxXCISPr|>WXvm+J{JGNx={GoXw3isj=q#*=@#Uqqm<|Ox#{d$f^%Z1bPoZ
zaEfvDW#|FjRgvm8PC&D<(>JVnwB;s_j>l3(TF5?Kn^}Gapk1tbHlBBxjTGb|q{SjA
z<SS>1T_HvxMoa=_V}<P9#{!7_2HQpi)NuIJw)CvFnRplW<tX0zP@b_u6W4?C!^sAH
zZkALrChqK*EM3L|=rM9yycTfP$$Yf5Sde`MLxZ$iquLyxelUH*j7c8e(tR`i3Et_g
z&AfhI=}(HgyJp&yO)`o%Mv7B4^X6(z!jS3Z=jhWcT`w$W7fwn$(c5n7aReEQR_MN+
zCQJEfD{UGs&yTHiNzXyMwAY8sN<=~_(mNIL2Qny5HO?dkX>lhfq4HmXl2p3G=^`Um
z9Qi***jzWB`(o|<-Vk(tyhR>m=HAyVy(RkWP|n4Lek)(66mky>GZ2ROk=5t(`2dkI
z#L;Qia4ob_lllLagtSyiDEk1AWEpkVoPEvm9=2-BMN=_Im2R5mHw${PfLg`@+Rr2$
zIadPo?|&*v0~o=e)ePI$CGB=9GqfFbFe*!hsY8%jPOmJ8C89qmm*1C#s14E_V8NXC
ztAy$1EtVxI$q-|#Wm;Y}cAE+wLpe6MDj`IBy7r7}(TyohWrOG2UUzO9JB*3T72avr
zIc2{w`p7$JK6Y!`%Pd9Q%41^w=i29vuAH-gX~qGfrs|%FcQaNW;oX<5I3Gf@m)%hm
z-UHFrSy6SSd+Mhq3?CNDy_fKQH$p0K++RK@T<NzWovTzEw`F1e0G3yw7i%`x{eAVu
z78!V{$jV|o+oBgc5b_fQw9Va#@~f22_m6PCUy{+s_Gou6>Qc4X(0DIMw*}rz_@;Ad
zX~B4Yz_k>F<SE)e@)7(@nqGU<|4NqzKKfg`u+ny(JQ_<{h74tMW-m|hZP(tN>GbuU
zHt{FkOJtLd^;D@Te8<veljCjIO>f6}TgGd<Y`xh3wb}ky+p>5*nmn&4fZ#Fw=Ii=_
zWFjL}BA02I4>!x$a@mNK@){Xgj&_57zW4I-ukMxCOE%7DLQhPvO=QVn+<rkTlPIC>
z6QyE#gXFOSy%hqCM5xB5m3>Y{)l_BF+Seq=m`N>0uNk}z*t-{fnQOna>TDlpGW=v^
z!WB~0`)_4BoZZ4L7d|u(ayp+4v6JJ}OB~5%LlTWfY)a$q_?R!eb+-pYy_aX~-5G)`
zS%oa5+w6jSZK{585{e?6&&)El<xjDiNg5J-g&^;@epgvi00lLjctUeqZ|ug)`HHd@
zJ+KdR*C7>iJ{=@GH+`7wfx)|D>kn*=zus}#O|k=xNQ*lLq+n-sX^{&<@OvinMr_EG
zPC!mkA;AlYuDUJjRh7ROQpKOSAB|I%P5^aslm%Iu#C_BAR^s1~^vrX#B8&%OYyofp
z!q%_3*+Q7$09|6I!m<U+tl3p5{}dp@Gc;%F=XXp~{a3!4GUSlxT5w54+1ySMYH41}
zI=?w(3e$Si(DHIH#Xqm%q`^jmY$rgJ-pDhhnx2<nW~gPXW&hvcm?ZG>L_$iZ+4%!P
z!{|`MG@7K&>Z#N0Xw`=uXgQ#LGG&w+Lg(b!m-FX)LeNx1QyJLMt?8c_(XasK7%hJ%
zQrELIPepp*KuqRA{-oj1(KyFar$^-lRCI%D1ccK;u1A*v%G8^F8b?i!HbJ0ptkhP4
zt5tmh!*>Bvh7guJPXYa?`9<Lg=_3r|6}>#*xpnS9j2A#@l*)xFB@T2EkmuoF7IVbO
zJ!bXn5`7RyK7?P)4ZZJ|xo$AXU@xb1t{hMrjDZ90(=wMIrtW|x<nX{K=llT@W&e8$
zy03KKcl64JWEJ5Ka<^yhW8+%Uf$8iH!N@uFI_Te+641AJxsO0<=M)Kk_CKh>wQl_D
zrDt{o(crut$(dbOPAO)eQ=i|AQS<F9PM@4r@(%7=0_X-HXdD<lFN`(^9h!`Ie86Z%
zChzPeAWi3uzGYDn&CVgE6sCnC3+!Z}AY9+9f>UFjSki$2`ji;1kRkV82IbEKK$;8A
zAqsKoX6k_YNkQ?>O%K6xb02fSl0KvWLSZOMA`zgJ0L7!S4h8|>ZTlKrk<hdi3EK8T
zV7JUYt``6_$h-4WL)h9J|GJ*hTT6h^!{ob8y8u%h%}Tw^?hk-eYSKSNiIpncY5tgk
zHRLYYd3gYlgy~?_1yI97LgRhqU{b8YJI04(Icw;E0kFAf_eKo_QKpJEi-xsFwuwi4
zPBGM~4SU3xF<BFQY+Ki~06;ct2r4ds3T;gs_hDnGl@dWh12KGo3eZf6l4Xexex^}`
zlz+#6VinW)1~CubiAo7#+wdPMKR0&d^Bdbm!0wlirsq`;9Sev~DhOpUA>X#me|p#k
z=o%`1q@UcrQ>MeiP#Yw9z#omH|0dkrLvV)OTZbnnsemT>m=EUjpefmD)%xEc&jtW_
zJV2-eATK_2*>$M;*e)v-oAMw<12LHY54$M{XL$GgUq}pBeb$iiuPB-P+_*LjfOxZ{
ze-lWgWkKcfxFqF6bE0n^^e5>vAYjn(z%v6ttU5Oi+Sn&0xPZy0Tz?Q{I8{(I02AXu
zOH}D1^p52n5EfF*&K>vdjDrQh+8=;_vC|byqb#O)uDYPQo@x<iesvJvzrkq}KUU#j
zTK=V4&KTV;%XPs$k%}(s@dH4{k8V3Ig=)c9(oSM`S=JnS1$sul4mWpg$k|r!|4X${
z=756N5m~w%?mWvy!mZIfy4c{ioYBvG&rge*4`rxEbzd@t5B(O`s_kHzWvXz`$ln<M
z<>Nij?gJ@)=&STKUrcz*$&gAU6c<CXyFUPM!GP5ZfTEq=n~&Ecb<%)_gvO@_4W%rb
z%>!=%+R#R{1_}oNAYld&Fnf8$U-{a14>6kc2>?i(15UY}ceV}1(CO`}upWtGsGiR5
zYjLgxwL{R(7j9h|)#n=)h53Sgg7z5WM3GTDY|VW)Rb`Be#rHGU$X}z{N5_8F@GK{>
zB61QJx!_AFeEedc<8Hs#_9frjW{Dqv%hZr?-q%PPRgOT*M)Kw9RaoGv#WqPd7Nwrf
zqETBJce((c^ZTu}#u?;~zkv=u0^GA?SH2o8#{6(G02Q*7IZtbcg+IyizEg3h_3kPZ
zXL!N~RARKt`QQ*kv>}|u{W!>1dPB!OWZk0?;=+5{yS4A#NMz@i1BcRLW+;3GFCai7
zPtD-D#Ma!0SX<D=ij?tn5!X?N^tJI%nwZfgE9^;MDzZx56O^SHcq$v~C$j0z{9>~l
zHJ!UnZh7uozN=z{E*?Nfi>(pcE>V_6mmhbdjnU=j&~Wh-gj}d5M#{Dz8}04XueR-#
znaA~gT){RK;%LWd<>n&YSylmGuh?5D7qr-y5<Zuo&>ziB2vhqB!~y>UV8i+fJ{=0g
z0h*#+*|zz7CArgKV}LDK2dxF2-CQ5d;#z=zH~PYNz%xZ6@(q)&St#lodN5vae)gYO
z>M#bVBIDj&KxD`^8y|4daphWg6&G6;yz28`TyzPD3e+_qaN5}pBk3ZFVS;N!5Aw+C
zg%P??sEP99^{VStb7NHMygJ=%-w;}vX0%4LyPsmx={{0%L%Pnscdz^jhUWrfA7bL-
zL=%R10gcw%b<2gi1)(#2QYhO$WVdvLeNsI~NksJ$llk{jWR-D_arzbfu+kQ9mx6EM
z9ZghR#FV($*blCM$P`5e)u*%TwVL?z2~VoaKLp*kYpm@rFd6s$S;Oce<f>yPE;0D@
zsk>8cXkl@8|BXu!3p!d)nDy?UPl>&^WZINcXej!5o@|3%@(_m_-V6r_i%mR@ke53*
zS7{h>F|qx9BA3P?WV-i&^l2!-Tl5)!35lV6>ihp>O{o81O*QcB!k}2w_5|cKanb;q
zF6i-EPU`Nne^3>anF({7nOf>mvy^YhFIxk;4xE)#E;HB7&?`5)B#cih`HF1r?UEj3
zv`z(tg^7``N6xU9Z^M<)<?aHy;D6I9#vo}odA@(bFeQSnjfX9}tjƄtu+6t_}Y
zSjhYJA=?p#hNZ&k#3nVv(2Ib!k@}AO{D_kf)$z_opZB2U%nqN!z0lGq<VpiEcc#$G
z%97LTNbldD4=-W+?15uu9R&DN)!*s{FhH#GKa>p>FG(Q&11}YV4?oWKKjD~={f$dc
zJ#wrdA)i_x?^8W;N~Oy{SU}1|{ppe}JDoKrYzLq**bi0F&?cGvt~T;gMo$?)v|X-f
zoLT1k3uwEWz^3Z{dbqw;KDhyHesQz%yIrqksML{OPQ2c3m-{e_hRz-nI+rzPu}m0&
zI@_)_zD3AH%9V09p74JDb#%Bq6%Q}(wYY4G1}SZ@N!k1ehQ#^6-p>Ith6d~n+IRBK
z5SUN^7v35DgJoqZSx~F75E{`TJjgjaV7bV&v4{SqAnQ0VpmUCkyiY3M;+9of!})MD
zVQ+1uj>0DP<JKFR-En)s3uLj;hQ{%~;e=<hJaF_y>`zahAMk-K1yTE8DYMqxF&++M
zrCoS4&13P|05Ig9!PfZ1rxvFfaHEIAnRVBX2snVMQvIrS9(x3J%fvr2&+hIYzlcz4
z@Y#@S)|27M=uH>jgl5W-&pY2V&G|&^fTW;lzl)@mmRLa&L=sub>$FjS?3taqHuhew
zI$OAR9>qY}|DZv-jeCVH#4|#sd*^w(-xrv~adaNVzQDv*g>1SAZ9u~UbTDub(ucLf
zwe25}YMStD*)M)}xh}D=l=HR)_lmv3f5TM2$9M-zVl<w!yyYd2V=b<X$Tr_E6cQ?;
z-Hc@Gz6@b_3-U-|RTw&n9z{_!*vTs~c7FEO=QV~sDI5owN*k^nAk9KHeq?gmNpB&x
z4lCAXlWCl4AIY(7Zwc|hX{}J7J9Nkou8pD+W#m7Btf)b^mJcIIW|pD|nzCzd_;cSO
z_WLg}g!jTr`+7=+H>wVI&48_O->4JF$-_-E?`SfxX|_K(!e>)zRlf6U*=uyg7d^7~
z!<F!PC5s-Y>Gw`Agg`3nz@j9~z-5GGJWhHLI&HX#m>Arvc{`ePG1|G**CrpjwDUE!
zx#oB6wtcbX$nrexM#)85xuUgf9Zi%u1$yl+b1W==BfUKS3`Qwe=6x3OEMt6Q<SKag
z*U;TOm}d}##p|kDAO`++`Bo%kk@Vha%CtG9vGlc8Xr=Bl(MH1J?zu#cc|CYW@2crY
z68+^YdU?u7$jr7;$?Q<LnfU`(aPadhhdHR(6AKu9;cJA@lT7#-j(tmZe=!ywrsv}T
z!P1Q)*};nU7c=~fu7u5cTM<CPtyh0Z0SHSO_<)X|<-5fnHot%ZMYmE9w`tvPRB(fB
zn8Jn#(R3dhXp)QpD*C*2rdOptECQl}P_QFrYs^IAkaF#n{z^Th1R$NU#kJU#l*zE{
z4r-0!Z*g--Yg#^SyW=gf_`@Pay}0LSJ$-^L;YF?jjZ8bS)WzvMMfy_BQE7&acxrm;
z!R<CvA)HbLke#d4imiv69<ks&WI6@Tn?jnJk4N1{c2NmK3AG$?fCHdAj6)^pB=yWa
zNf$Gg7HaL5%U=gJfdk@x`E&zKr;-o5r8yk~)X@PoS@n*`#q5t}_$``pbO7~;oUsyf
z2XrT6GUh#y4~pFXt5flj61$+<Dv7@YO{3QKzbI9$V@T}h+8E~p7zez?(wgHkFhq6?
z-t?4lu3bTiOZVPvJb_j2o^+epHU@^sDNkX9s%{qPUd?T*EfSf}{VH#$_uz2Bbs#@4
zg^5)c^7?14ymfMt4pt>oQa>(o%%9B%cU6iJl|etMFs-d?x&3%9KQ9vlqYi3mo%ztW
z)I@4$(2ZqDMBfXz<G>irOLsWRmQ>nL$AldmdS_;DyLTMK{43~CTFM5d@;OqeTXcN=
z`|q>fAMM-VU)up9!aW!V6f7>{_269w*cifO^r?XmcIa0#rWj1g!q_=K5v=6(e~Fxt
zom*lF20tnd0Ym=9s960}G3d%0o*~cXzwugtu};!)08cnA!|k5{W$>CqAxq;9=v()7
zSQ1bV7Y`aOBz_c|@;vuVqZTGX&Aq{iiZk16hcstRx~%M4-l*J(Sxvm6lap;jB1jz|
zS5%tQ4(hQ0bCI6@Q=?I&+fiOO-0!&IDbnAb9ctlsMDr4`kcmT?hLV;T6Itkh*EEnt
z+)W9Lr?58<YzGt?m2JHBofEX(x*i`<5F|-0Ed()BK`21`Y;7k(Ix%%#^56j~BC-9K
zPOArW8YiG&sm7OF1u6hRL;$Yke>j9|B8478TWUuT3~gBx39%S><67kZ3wX%59(bLP
zb;F;P+%)<5<6?6<P^C*3#RH2x1-Sk4)xX{DcvUR__uP$BLQi9;SO$7%I)B5Qf1$_g
z-}L9)o+}BrgQGE3hh$LqOK0dyXVLBF&(t<g1yJLEcKdqS%xWT>Dm1&0bNaT%w8o5W
zS;k6EAqp^?LLMZK{2{O*6jU;KMf-8l0M7s4*%Qet3eU(uGHiyQGE8_@%XqL4(cI{0
z_IuJ6DXTU7uYHK6LO`pD-`G|8O$BIxSUYert!<)zXEzV<HAiY|_PK*$GLsSOM$UCk
zz5tA;Hvfw?NlHYS5H77A1KK73e4dH%b22;-Y=?eN`aSn)h<xnr*;^T&?iR4wK(j~x
zfM&Da^L&5r5Gk#=1_)jpjc~!8`nP101O4xkZR>j>Zu2*&09L`2*_~=z&H3u?w1;#w
zsr#C7&?o4~s~r`(xgY^=je&OW=PU2L+(}Q-xr<1hS{%aVBgy)_nGlj%k7)N39>C`S
zlz!3wRr<yG1WwvJxuwNJ$O5){EZ~dD6BQ%uY*~(?p?@XjN*ek3dnbWZeO=b@Fq@C}
zgboA+YeL3b{wgsG>0jc@XLD=P$q2`8i1<Rg?UQwFZE|0sB=%;v$Nj`@2SKCr#lS0P
z{qds?fv8ZgrMQyWy*D<+*wf0TsrV`}H5pVo6zwYG8os=oZP35J;%>ygJT3KWPx0e&
zZK6vMzUnh&i3cV(MjFD~KinQ4uwo%AaM@r?ZFB?BS3^e(jQh4#zZ4l$-ejAFbuYQ*
z=cbN5zilJ4=$kZ>^Z?-%wr~=YrFtI}^g_vWbQpC6F53%vGsR-Z8PI<=zYnNPWE|;&
zM!eX0>0t8Je9~*U@^lR?AnFoL4n^IxyO3I%6KxUmIm%E1*7E;HDWaG~^J8LpK3G2x
zFI*x5C*J?HMy)xtmyPY*3tzE^l$T87xnE}>&;kD}Mr`j^GmhG={S^q%qZu$JAL8qU
zcKnGf&rbrb!&t&M*uLLAv6IvC{E<lwTQ%mPQkF`y<Gs7rKiqlp@CV!e{kjne`Uu<4
zXD%=VU#$q!Q<=D;e&y=%`!3k#NQ>DAA7a9E9N6zSf0BaI_@b932YAFqwaQ1>`dm;q
z&|`A0i<;xk2ofp1s!GtK!=BK40q<_mNR}=GOVH3xN`(1drHLBpLj779-g4YBp6YSp
zc!K=(G`}~qGJeb8cAFN^-J)fU_#ukjKzTj*4zp9iv9dar$;%ZVYIl(7HHciZ9XCGa
zg4kwVqjba<BT%+-iSGk)4$^vj?P=kQ+S}UpmkR%2BOEOT-<F<fdb;^bcflw3?dBK*
z(mHxC&Nxu#mrm0n^T@?)x%!;-ML{U>;6rkKdAgKi7NXka3)P{Ki&}+MkE%^KFUAb$
z+V#0}@eO2~nmf~ub6K<V6)P>aL9vE<NoOrY9A@{e5zWGv-Vx7p*+eW$v9`=5qmxE7
zGM<c&w1&c-aIi30!z+JF1h7ACmT{b9ZK<Q}iFXJelPEF#Myc&DE-`9e;a=}<nQd4}
z`_5cybVOEkT$v-f+ATOv@>WJM1kLVh!z0&r9;3qH-l?cyGV6##&i+203;rG$EC9UB
zALp<BnK7StS?=e?Ep;gK#(t@r6R<z`grjdC^}+?mi2G^gsrI07OY$a{b#`B2`OqH!
zsv%c&b;YUI2-l7YlQ(F+n>7u)0ct-3tKWoVzvOa8O_1fXZ0b;ssAdg}{a^Q6h!IRx
z6s#Z)(~)E7zodShO@O84GNyX*p0?`q#+`BdQ^-B@r4S{&^UNRtmSFxv_LHRBw>RQW
zs0H1E(8|$;p>3OCl`xn<nnV3TOY+j=)^#=H9~ax`n6&vJ1;9+-i^5bZRVB+{ofj5|
z=v}fyU?<2i)d?`wPq2k36>gSz>i_iVkSKS_sB|;;K<F13<zs>ij`LY?Ev(glWO2^S
zo=~z5d^pIZI=uRg>?NGpikqa^Mai<SzfWYC)yvL(Scs;YzA$9>A?b%1#FU4Fe7W)9
zu3JoE8kuH7y}4qN|N7XL)9L@X=_pVKgp@_9NxE^|YW9((IyEk|bGkTk&;6a;x#S9;
z$e1T+kz^;sXbN})d!K3JC|qljX>`jI_>NedXu*+tf^Pd9a^gAC$u+HI91O3_x`m%H
zETmr^E`gv=jJn@quNR0Q;5R}f5JI;HRZ2*!1+^2cvzD;eUP1V0AD}t(TWrEvHay+E
zO6(DDy>pC&3!M@u9ylil+B@)Cj6C@?TS5O;?5j03!xu$_F!dACG^E;QXu;!QiE_hl
z%a#6ii@ttb1qLUWkL*9?5ayD^)7SBMtxTC+)<CJ{1x4WfBFz!BxZPYC&Inaow=d4+
zHgn1NUN;*a=1Nnu7vg2=i<i%!-Rpl~o_RCc7B2IyviDq{<C@GrO1K(VoA+MXHrqf7
zJhw)>rF?|bzx3rSS?fyTQq{V7kPzbHt)!*sS|gVA<>gtjm*lFz_Ud%=0*Wy8y`HZG
zGwl}^(k9;j_BPkBw1W*Zdd-H}ER~?KzJ>5{-lmxF;m^Cx_U*Lzh{|cMcwOWAHoNJa
z$Dcpv-|qBf|NKs0E{?LZy>5&ZOCplL#FBDKa|V>A6LYbY^$M>9=ST6SUfoH>6)plL
zO#5REI|^s>9}~0o)hMrY89Vg3*yTo_#Ug9op2%hhU~%5D@=nZ9_10{A*S3G`zkTg)
z+{IGV7dUeA)PH~HgUG?x?zB2>H13~C`li@d!fRmj#Wd*b1?fQpp}~8(Vlh6?Uz}|6
z`$2ai<!Z9=1)KZQrzrF#Eced{m43)rht&=T^(dU+d-PCx=GxoWi3s$H+WN^s=CBX*
zb@)!4H*N*D(h(HcBAtAX{?Ui0>QHd_r7O;EyYY`z&e>gKF|j%Nwa)e3Jp|-cvewKg
z2KQ@gCAG0np8CiLroISQei4;@_oIF6P4H2dBd1hSoTRVeOMNbT@tW<cm~b;F6i5nG
z)tElM$fh}Sxu|z{Neyw8fp!8sbV@;OaiKn1sr;7gIP34gs@kybGPb*3#@ee%gdOBy
zQrGA<f34FR1nS)%z5Y%o$D?}9%sd~GH?$Mw#yPU7g}7{6PU~DjUNQF))Zk*$?aRos
z&HYRv>^=6I8)Dd*{XKqf=cy<7XYC@&E0HVDsIryMeoQVjYR#KI3sSMazbBU$huHK!
zU9r#K!`%3uq0aCm&8=F0)1as7l7r&?W8zHpq4q&PWf|m-u_m8(I19HT^$ucYQzsuy
ziaBr%9WPP;gV(zonZ1$jq}9SO)4-Vux>|BhnI$<c>`7}fyIN|H$`fN!GUoDfqH_oI
z>^E0LPEZ7^`w>H(&CW_iuOZ8v(<n)SyJx0uA?5xw;6jf$yEe~B<mU*1N@}*(f|pN8
zy0w29n}3p6#<Z6z^J-(FU{ePX{x!&cn_!Pwzv=5<^NQv4L=~qVf0ePwy6IKbYFK1G
z8_I9*6Q{QPo?bFn01ewa>{C|xCpPIGHKWxeJw``P8t?qiObXqXZ__-DI8$|WMcF1m
zj0Q*JmYs6WHe?ZOPLWDcfcDqx|B?Q_^clVqSyg(=m*&4k?B+d<vgvb2j!5tP_lUiF
z_(f5hU?Z)INOdlar@WHT`>*TOl=-+H^8ZK^8X3K0mG<Z{FI2%V%Ku6D6M<(JT!UEy
zntukoyG--l>fV>LOSeR6LFZ&&UN1b|{QdDXoh++d7)hQ}Aa@@P_v(!2oa?F;$ywUy
zz(wd0<d)*0sQndaw{L=L_c(Uqxn7iOo9*yj$#$_;+nos#Klw|}@_Ke&OVRV;KjKBW
z_owv46d8{X=M5DQzVRDKh_%D>kv>t$m9y$7)s{h!ZL79rbuoJ}3^!=|Y0v%RCBla9
z58Il=&$;C@gZscn&0?kID}uQvr(Rs}%FEoM+#*4}bRVGZ;D<f*mMp%>^Caa}hg-{~
zwt!PbtVNN_P=QNv6b<u6DI+(yZIx{*j&Zr>zQoh+?vYT`+cF#WWe;1Q>fG{7)mZXG
zmJi1Z><Y`Csukt(Dd>JUY4<0o2C^M@c96(2fb}21B2II0m_X&yWzi|Jp%LNEN`HjW
zd3$fdI-~8lOrQvDO0900YT2ugn+X{2BAjv&8+Q&~?wRjCZ|A;d=3U!|3l!!>!u`!1
zq2CI?5v6`CPd`@p*8<bM6#QwnsN63IJ&T~79l{sD9%EYz)#``p4@wQhGzM>v-u6<)
z6kH)8ON}?O+U?|)gJhSZt)(3KiWZ7Gizs&&%}w?9`<v4&wC4+kbT0w#BFGgf!)9IW
zR#)fcM{?`4m*qz+e+&8QT$#8Z6Cfi?epOuDK(bfHIvJc-eN*dh<0H`Rs5NtB>LmQK
zmtzl-??UyXqdN4RC<En*H<cTHF*e-Ds+RZo*84_fio#j#@&xq(xBS9ubCKaR_t0?|
z?r@rqH2n1buTRL7E>)@rW9&zJD{}kNC&&;PY5vZ^b=nYNecE2V)h3)PuJO$)B<2{>
zO@ag5Gwk{$FiQM%C$V3zkjr6p<gJrKhqDZWfZv>#b@C9iL<JMdmZ@vge5z%sL`5I|
zF#pEEW%Qyw5~8D(n9jf^rJRzpKiw^<d29vG=nC+7J)?UGyeb3kxhTq~PYYc4h;Wwh
z)R9i&T(a!v10$8#N^_<dmG|O3G+|iq`DNHe`kiCAHQ<sn^q)=zZPJX_0`;esjs00S
zX6T!d-s()B7=c#Cf|cB(%^V5u9o|iVPP0kKOX#-!ey%8xn!O+?-^(_Hu^ID#%D5<Y
zl(ih)+J=GV7x|m!YNQno%-Tt85$1E6^a6U<t*PQd`yHSA-uW8cNxJz?M>oLjN7{cs
z*A@`!VNEqZlW+oJmaio)AQS|;0;+ePk++P#C8lqD>mC{~%J4E+Fa)%n2m%p;E}W;j
z$>!w~>IG`SmjHJ#74U=12i9pWTb3~VD4r}9BuEPRia5P<IF1gdNr0+;j^euT9iJs)
zDtitLKTGwK#2i9Up*?wL;__Pw;5?vj%#Z$cos4_UR%{O~>n)Au(q_9K{^jmsnY$(4
z{ypC631<l_Eo%$ZxTqKQ%{Uk}o0s;7=B#`H%zvpT(e`VW=bx9M`Er8pq|z#N?7LrC
zc_0q26CI9eJ-p7oGaGf>2J3_l{nL99EH6=4<nT?az2mHs<rPVKc(YOV8+{8|#=rez
ze_K%B!spvgq9pfC=MR&UAIf&94pO?vb>N_PUKb}1JQM=~ZcV3q307ns3+|a`0D6WR
zvKdfq@5l#2gZp66B%2;tvYXF&Ee6LJaMKaeJBc%sG3cK|R!8ok4QLDwZ7Kv)U1dJS
zni+-4_T^Yi+AH#15_x#PyC^hdq{*@90WlwXt-27KSYBZHZa>NuN$0Ypq4S}67Oulp
z8#KVhsV{sfUc_>z(5#Z8WT8#zLPIW<K`Q48dwj^>qI}mXu?o)F<m#lX72oNcUl1ti
zbOm|hb>p&6)@9!n`g|3n?@re3@Alh<OM94AxpB(+(8{PQDXvH^xZQp3^POS2n_9mK
zFLU(ZUd_1E+iRss%#2m(T3o&oTUZ_&ZH`e7%NWm55o8p$850|wK_2p)n-1xwK43M-
zi@IrR9%REVmt9Td^#${JLaY94@fTaTO1{$IO>TFS+`9&4i!vvi@7M{WolKu&uiHAA
zr@dHcjcbhOwTTRgB0O)3%W)kk#0OW%Dp%Zhd%X3{_S1+_3N7m7Ya-@UM*bLdYk^Oo
zId-H>EeayhT`-m!;%>=eQNCbXp|M?vb2H)|Pu(wX9gQ9T$rsGBXy~Eg7NkFx2z{{(
zj&wAYQWd+2Egs)0OmZC=l9V+INJ3i(57}cPo#fk&g4!3}W>TQ1%{h`%I<iox)~s>K
zIUhdmXu&^6oRj2*GtQZP8R>7!@+t|9mkcv}y3(*-DLmUX<&HCu<;!;KBM70$qbVJw
z7iWuGi}W{_ibsg?I8!z>wBUs<&}tWVJ3HDY*C)+sIrDj?zR!(({m$6#@-noOW6;P&
z;wZ+-O#gAI^)D?A*?dl#CSqlJxFXW>7U7~kUI-$SM-gh%71Z=^J{7d{g*CTGf5$p{
z>p8pF*Ng!3%3FxU^%wZfS;f%o_niJ&(8|U54bRA>&C6}lat3%bxt(CBvmB+WRju{(
z4)%1AM<T1`FV$RW-_WfAc%El*UfRU8lToB+zx9i*vpa;3cIu_S)a|}HsZ=_3Joc*n
zd^bZ-M>alZSLC^39>0(FK(1YPyKJT#egC`eoG;)=R*n92ZME@Ks68TY#8<Er)g8s|
zvES|4maqVoLMG1!x^VDtFo$q5E5zC$(#*;bhUA>BeYGvEVc@rjM}M2<-DuX-t2H#J
z$s5rryMg?w-kPoLJZ-kK8@F!vM<<n36dymL6B>ILqEoBnk&N)(tk~MuOV7$0KH*s}
z$HPCzXIj!`J)3Pb5)IrRlK9jki%43KlF9q3*5_p^MXO&lf%&E9;dN^%55kMwh(=bG
z?4MxU!O>iP2cZx02xg6VRl2V0zcowH^<R1RQqdw89Mfc|SgVy&ziwyCwTyFB#skoJ
zkiT|BIBJ^Q@7T%Nozjt?>PpE+?2>Yv2Azv^T<-`(g=l%fvUk1RO;rZU^mklCw`wT>
zmD7s?#das;#C3Za8I;FclkL1@Su(s0U>L+O>FJt(nK5VEu3PO_E#IA7W$<ihkxl=~
zwPB{%J+&75#+aXWrRxU%Tj6M2l@U5N*F6vH!JmX!2e)%4iQ3e3_Yo2WHL60DE7D3c
z(#@sJN;J|*Vq?ntD*<zT_RT1alzl*J?S{j<IWMPGeMfuB^{d3exNaLoUm4PUm6cq7
zKd&X=`4p2pG^Qa3{;`ut{CPgYYvEwEn>6+WmZ%zhZh1`SWP$VZpo#KwU}gRm5;eRv
zU7Aulq7&X3y&$X?r!vkoc)O%&1<Fq<oo0I2-pGIsJsa-DT8rSJfvr+l&rYrGB{E3#
z+}$}UK&V~bPL%Yc5omZHHC0EPlW&jblZPvDZS=EI<Qp6gmJcfyVEM2Y=nJ)Qj4Aus
z{>N(P=?(UhYC*s}tcvLmBMEte40!OC*vPDpU5PDq%?@;%l5YpPn4@i5;^q2~Amgg2
zQbJE>1sA6S?)9<3p#4%Y8_2x>F2eCsRM0%IdZkWX^K`O3N2mXpMj(#5oc%DD=qUmo
z`nziEK-<Ae`4F&Y&dWI&xS>Q%sjYBaJ+N2{X(xP$idH;4q5Z(I{F=5UH7Ws7K&rqd
z+<O;fHSDbpQ@+5?C=`0IbO(Mu@TYt=(7m@d!=_?>slocP24=xg=<~hQHSj!R!_n{f
zR$+z0UZ=P-0&%`yyc5nD=0VS%7bri~OSQ_9S4=;V;~yt?zIU5CGPOEb)MQHwq$3^n
zhAFdpXAdp714^cLM37Z~j#Kh42Spt^sN<P!i8K%3NMN12OzmbI&GPS~zo&A|gN7Mk
zcdb1Fn^^@Qm;Z62*22eyz3Wa_^~I>k!^3umWOKm#mu`#c=RN#?dx=0HTY>v@f#d>M
z1+7?_fF&B1`XwkzDd<1kt#pA!AoMISQq)oWZ+@D%7ML<+ci6LXrGpEpPRqBW=&61I
zdZGW>%wVRDI5U(vWVQV{U@qx4B+h7pc>d6*|KTR@_swhq2)TdsKLP^<%&N>d{#O(j
zNt8VTVy~P}P(P9DA8zRU$AX<(TQwviu0mj!vLkPgc`Cn=DOGo#=v#XxL;Km!cqDB<
zu&5?4Xatf^;jorxCsWeAT<<n+6+W+)uouM;qepJ4(;sKwGp5vDCuEmM%BjcJX-LA|
zH*Tlzm~WAlvJnNhclW|et8JY(*9PH|#pNSBoR-TB%kEK&l)XQE-8IE%$S;ZAy%)uE
zt7@nU2V2eCk%`X7F(N>gA9oD{0_#~Jd)OSIQ4W4M4>n{o*c!>VlN`qC<{RDg5-q9c
zGv#VNoE_rjOq9d@5sDrItt4>_*~iX1Fm`{UX$r<#XpdPe56h`5a`o=M6ixN_pS|AG
z;rqK{u(u;F&hy^+5f;%1pJM(2Wh>xED*a_D>(?KyfviVcp=F7S{t-WX-j;?N#$<Cn
zAe#!#xS9*AY$^=u<@-!}8|3@UpHBU5j<tD-d`*$PYG%9X*F=53E&4Mpr}|ixyeS~S
z2WXN9PA1!x+YS1#4xZY#d&piXd9_5yntR|>mfB<2->Z^%14fY-i}CNb+zaq$)qgDA
z&KLbkc6NPk1?ix~XX`O&7-=E|t8R8N&f}}v?uRUxTZ=2uR2y#xgb9~bwU|1{q-z_M
z=#bY@THBi&21N!o#(IS#*n2Wv)|}a&I?BbkD=a)~&yLSnr}w}!<1|ImzVi;KWLZqJ
zT9q_krWHl9hs<;tcAgz!an2aH{5fkb6XJ}sN}iKQ2ELtB>XlFH7PQ*_pn1p|<{hjr
zs^mYS;e1cu=9$vZkq3UskCKew!vV5RrH`y2{Yt+CUf6|O2Wc8G{44{?dP6PBxoAtf
z!WEr7@&->W2UU()gBmoK`tx0HOGX0~)C(S^J7KnEv~Fn*NkCupGk&<5s+=GE<+^1J
z*&{v0W7VOQuaEF!-=;S&X8YpwgG?N~hGj#i;XF~psZ4Ywl}40~&1t?dn13@#$0koF
zZ!d6}=cE5B>61CJw#HVEsLaROmh-$*dYsZV=gHv~vt;fO2_GUk#3;RlvL(eS|M_a5
zPRZ?(Q^xd3p?YAk^`Lzo1ByJS=|T;uu)2=c=_-T*4)e%Qf&SuhQ4j8tiZgfHy&x)T
zgi*IMw3%1}EobY<xnwaW)jBZu8cwl%nz@tB#ov70E?Y~^#R?rr&?e=xf7z^C%{&j|
zwD+32X#6$D#^OfD!W(&Z<mFz)JhhX{K_ib*PyV=`TbKW(;E9E!+G5KRK$c5>V4+d7
zAn<Tm#&%~e&FCh{u>*^&+}6)Q?J<cas}|$wipKUwKAgwzFJNV^W%*t?jF*j&rh;B#
z0QZhhva^=PKz0z5wMw_9)C%2q9{vgDSHiSX_G$U2TjzdW$j^SprPuk&PSK(BXYaGx
z;cc1wDbVNQSr$)4HGajuU~CkAmcwp6;LwmfchdYp?N}_CpC{@fk8yOFe#4ys(576w
zUax$oqdaoje*oqcpnWQcc!v5_XiHJQXP`xhzt{zA?uL%d&ygSHZi6)!LrCwegL-5K
zF7d#bmGK$bcvXWx*azHOK4#fhFXsx-j`*<lU#m&WCCdux1by6UGPDs5F!}Xp5p`|X
z&9!!kYuwY$7?ph2-z|QG1>2(^%xkTaeQCI1$sk6-055K0Gr(;~EU>xawoLQN%ZF<z
z{YIC9-$3B+kLHlgIw#UZ@_ne{L-D1>>qItSUiCkznh4Kx|GC7$Ge_HgzEe^uT4wL1
zvt{uLBWI%}zW{YYgd9$YOP{lvGa+WUo;p;vFjHn{>D@f{v0MDc_!eZQmYK`EBTtvi
zLiO)t!O2G5?9Z8{&=}MuZ>>&wptJDIP)&oMBi9BlaaWhI(v}{}g1-^$7&x5*6=u;7
zc^PB<iB$Ug(d8V^(g)GTaURwv!D?MMgo>{WK~sPlLppq8>3HVhtfg~Moztaw`&QJk
z{&N<#-FUNc>-}lb0Qc8uO*ZB1;LUwtpZ34JJ<TcRZp+0RM~;9L|NGn1gbz2xyj%<i
zp!jkmCwF@@XZr)TGt9$b+J#S|OkX`3(l-yb8yKnz%%7G}i#ze<cIK6s%+u{|Ey~ir
z9Q2|M1<Q?YggB?1a#ht`nhsNn)A~B=5K*?!X*X^mnsz;9QbNz}@hu8B^yfSBP8S#v
zo7)f@%k80PCS5$WM3F!cq`Z8)?UN`{WUFrD$@pqE-T8T!6(ePS>{$2{?8^QlY*iU0
z`abqj1w9>0#Zs4}pwWfajEUm5O2;mWjgx)Y3nAR1(R>x&c;?xpAf3?>*QHA{JxY=^
zFJjHmH=MjQj$S9%%3@(Ot;X@n)#m4}7=AU-eN3Eru|r^)4OKBJm2_8|8u~u=<AJAL
zS7m!x>u0B(TVIW%oloopIS0`%3`~>?bqVl(<LO;MA*1EKS9zxnTz$+%{?5L$jiN|u
zYV55L)V((|r1mk{t($bEgjvh>*6dN%`3>;eCV3f3DrR;P|FAlsexWbggIIC{`={C~
zVU+2WTnx<MB&oIic*gJDr9e5QdU5IMntAK4qZTMhtm1|FUG0}p)6)ic$jlht8P_ne
z1N;D}hf?;akMy$7uxb^t;*|J`7y<v032daXHS$1|Up0yMC$`#~$F=o6r@V|I%QNSX
z46KZf-#$R-dTUCr>f5swj_`Kn(NN?iFY@W;yof}e=*z_{%agiJ*xbJ~c3crGk#(=I
zwY<(!-25x5(x)=5ucanx)Q667RDE>i>fYm@Ap=%Yw+U0nNlE7eO$ut(ZsiFGsM&a+
z#h=sVVI}X=jvafG{D$Gg&hEX}vve)Wf^<cb7IzeP#$SSTp<2W@u!OvX<ZE2PaaI$E
z#I&W($ci}rhJoWqd%I_OOnq+Xyl<hHMUysj9i_`L*_PJ~dAX1S#0PnaH{NJgFV<;$
zSHF;jWxD~#^?EW5Ya`3Z2WOm_owjDN(#%VLUELkI?r1pvsIuaZgak|8=Y{I?@$LI7
zJQ->U*KU6#Yqne+Gc9Ke#VJZdLaGOtxX8mBM<LjW3~|gSDfc_Uz6-hz8}MpKv^s8R
zudv4ZnGDugAHF2`%V+WNEw0Bbz`MC4(({R2&1$tj6hBRoczoPH%ku4yq@g``cZ4T6
zw8!_pw87+jjQ4uFb&QkbAQ<>>nRXrfXEo-dL3|f{r)owa;d}nSNn_Q}-TtAy8uE)(
z4Ndn1A-NO#<&*ZMZx>3UeRA7$){(D$WaE?H7q2}jj=!;TAy~<Ah8#}R=uwQFamnFU
zfjp>T5B|7)i%#w_sXXuo(Q2tuN8R}C%D{TYPX&CSLLGCF_=yEQ{KPk3l>Sps;b*q4
zzp%CAPh^~u>0T=5(Pu&*(CTK25!Z9SI_w{Dn%*NcK3`E)I=6yy_K6@*CbsD)|E4(3
zG*bR<_W~nYp><F?@0(Nb6JSSQ-T)_i$*&G*)C_#JCR!-=BiXQplHcq=Z4^rS@tW8K
z{(gOoi=lEKB0ScPh=e^Ig9sxF-FS$6be{5}$ln!}g%?m^&|A7W&!gv>XbBfxjHOOO
zACw2W(9a)dyL0-Dd?2ZIJv1CPcf$-B3H+6luiTiEzZCqjr)Ec@fOJJ8Mp^l4A7e-K
zx;cM>q7Q5ADVJ3QS!tL1$Ah;Io=RLtRe#5ymvi7Dd|9T{?RvFm3!a96%Qt^qxwvby
z=Z#&)c=%4tkd!_U{eF1c?0rlob#d72V{4N3r-^>VZ-x2CcXCJFa=s<c3hWiDtjZR6
zY~m((F1<<Ek?q}duixyNAqMJiiVSDjs@LOGiS4A%&6=ChVlTx;&Zc*wldkNS%VQBI
zcCrJ~8g`+5g)v1}C;e-RyXCxi$8c_I=mM{}(IxG#(ZVO6E|H!ZaLn#^m!bPfGP~7D
z3=>IN7P7i>ynXh9QIicza*@93YCZPr>-Et;-kBxHK`1v#oD(>4Q-oDSkp}`T{p5b7
zkMqeLA({|zuR8G~``z>6f8OUt<HRwb(Z#Yg_oH#+Qeh*@Rb!*$mz)&7NtT1=y*G`T
ze*d|Xzn59lnrY~P2-T}<iW7O_^?CsHO2WMhx-JO1FeS^IGp#G;FY6I_+red;1CcIq
zKdGZ#iTtW3*p#bOIQx(8R;3=yyjK>y*1r&HF7?`-Y2(O>=%tG#3U+Nr=y`5s&W(L-
zzUTp(C3P5qkSb%A^(Q7x7n41Qb<HHBo@ra>uuTmaEO6jm)}ZD9jmfa@|AO!&HKsZ^
zv9lhLreEI*zrlHwuAcQ47BeDpjRXR-SB}uIIO1ulo6a|n%p0~U;rxOa&gir$Ek8B1
zJ6T!vb*Mb)4eYhF@=?UtpV!SP#Z7rLQiqR->9FiU+POtuTs`*bjzk_@p^`M&S7)o(
zou*intz>oh+@KttonCwnJXv0k5h}m1({0vBo@^ZOm*>svXr50Od8Hy7!olQhd+cay
zKPkcWdc0ZeHknX)dCz@YprS*{CVGY`y^Q~bVVj)KRMl@o_2`)&dxf>dE8k}g)d{1H
z6Z_xSm(#S(jH0T`_<iwT%5KNuj<Xfe)P#NzGi_baRf>rUCN-XXd;YCBUH;X?>rwGL
zOX&)-lryV-os3V^2?h>zaotd+YY`6n&0%B#;W6cd3CkSyp~=K{E5qGKx%};G8Dvhp
z@`Xg#fp6%*3DEpe!*183M#?Qm=m&h%BRlhEu05&nJ)t)Ym0JH5ZBP6Y{*i3lW#a_#
zo|C5IdQ!1Z56t-|XJ(CoX`Jr}O!;s!a*F#Dk@*_W3gYCCgFBMLf%9kf)V3f0rW|*g
z_PPA9G8&lPI;2ZDa=~}0H*}}jl3kSF+;2B;T<0oDf9%>LO6iHEPaLnku=uu3YI|J?
zGPaJ**n1u13u{w(h}4cSscH4~7e%KU#&UScVPn*tO)?vrd5b;?2ia$Rx@Q+JDJQL-
zWIQ0)XF^oigj|fumYKeP_q)2BkV^L0F89|RGqxMQ3&M4=B_Gtnj$RVH(R;%9?NmMJ
zx$N!?d1vE7Ebp4YH%WMana`+JK10DB71Ix=^~a-!j!v(Vi==oW^$BjoqZ?z3dG2KC
zz~So|dAFZp`!r@UI7tI*-fNM3+V(~fkSwzo%yxT&<#1QgTrP<soOvbJ8*ke+aP=NO
z`P7Wr&KJmj#C_s5i5(rk6?U)v?K*BbPUrk*aNJBZlZq@cjYde&fG@&Pz*lmf^DXfU
zRH!uHZsA%k*<&WczJ*JJOy(tT41IxqEyqZPTF)&_w68JyRzT68_VH;=+&_8BN>@go
z>hH+k4D(vdNB20*Pi#TyxR_kz-r0_!vwz*7Nv7LfAG9fJiQ<=5{<O#0a<l8RsFPf}
zz;E(PsYg6PbGbrJ6+QcZep29$9EKY-Yd>}CP4(_9n)xjZdid-cD$CNpIhGnPeX#mF
zP~Noli@WAQ-Ynd^D+P|Ck$G3MbnI79*wNlHMMAnVs<>pMOKB9J{FR)Ak!j?*TDGK<
zy`0(M&+bvM`<>`eSmP_BjH+6W%RIZxCv9VYHP?eIJIxStlS4cdLCd}E?n%O%AeeNa
z?6G`~;-?^<u)sc0%Q1|#Ks@8UTRQ$z_2qoCn2M2#I|bR<f$WigCjxhvLP(SOXT)#o
za7+=hm|=J2ftmbf=Kk$GAHRy{m4DDz$k1ohCVw8sO|*QflPJd&b+XIW6ATSsgW~CR
z8isJktJp4E?<M#tH-1($<mQ2gczym}BZRAbrR9U>%M&lLf9l3PX9m53`YJWOH!p-r
zE2{COu;cx`^2Xfd*J>Lz_7UQ)uh!m9`1ctMve2#(!o7%faGG^uf<f)3x4TRI>^Ffl
z6@13Dx@^*)`ncg^<<S@eUkB&CiizBdOfN1$(dW@{gw<=#PK_urO4Xjt6Zju;sKoO-
zNpdY034=a6y-}Bh*17$>p3kU(hCzYe-DD3~$tS1l^aTFsun9)Rx?<4w$?ji7r*-wf
z8<vE%bOmqbqjFYn?%(%pgZcx@{4W-w|J3zU#C=-q^q$(|F9$0%B5_IJ1cTJh5bo#M
z%4qxek7NefqKf_Nj^DicfQP`uP2wj@nD_GavHfezcO3U6{%V(X$5pFVGfz0%%9x~K
zDO%Fk8!1WWl`6$Ko>buNN1*j9n}T7DE9g`B>AmCBoZz-2WJT&*SE;3MKO}QNmsOm(
zoxAi7kzeZ6gITau0*Yw{)dCqiZl&}d`}~i1ohz3kt)`9YbZ8In*pDDbdvB&{-WO0!
zab*$P?l55wPRmaSjR+NIRnr1(N7=#aB;rp)xdJaOz8(F|V*V}7E_C_kN4pfz-0p2(
z-w#EMtn23+6QuW@Pd@1aR^9rPuX}sOuucorA1h6^WYye#xHHpIl3+TU<?Q!H{Y}5&
ztzyruIG%;$v!u?eE0n7YdYA_6JP6GL+hSkni4nc7u=Zdz%6qY2!n{;JLLwSFZb*Jb
zH3klkvN;#A)%t(f`s%QzyYPJ#5d;Mk1SF)TySqa`8j+T6X^=)jy1SK>oOEx}W0Z6=
z1xCZ@F~;CG-}jr>@B7a#_$;ooeV%ha_kBOlImiDZhrk)h$TDwtgqL(@5U;xn9TYu9
z#qa*2a;lW#@2#IUO?zMWUwO^vZ(b`mA+Y{@v+TrML*_*cxoW#y*;;vdmSD$V5b#wF
zPK;Uoq?QBxBOqq*W{xqo8b)V8-48@BkcW?mmaJHRV${1q(Q1rEEsiHm_|m91yw)}`
zRV}eG*pE)YxoB2e-(x+prvZmb${^dMK4DHysuOyKnJ-B{6aFE_E_Ppu4LYMKGuRUX
zqzv3QzFvP#2%qPFmAMYNW}yF@;HY!<D=);IFkkXS>IzYsA7Y2tQ%?b85M~T0W!=XH
zjTLP2f51^`czvP_rZyjiDoiTZ^(r&6jXupGIFWpK^8w!0$?6=8PjKrqv@L~`;^I~k
zz^EbK+hpCph)aZ!-uy;?<m*ZJZz`EDv164{><huq9D8k;#&Q&rm_&DU;rm(ps0@>I
zy6Lvx9);VDdmg9s9~sLZNy||ElbHS?odxPXcc0{rxIddLqK9Hi{>@BzD-*wO$0Jq{
z<qwXe<^R4K6OOyaKj3`?#Q^H2d$kp*M}jg|se`W4E>aKA79$ZlvlHiv6B$QbuYwO{
z0`^P`**lrnNY-_oBy*+$8eX6&%z-nvuKT|ng0(6oVMZR6iiT{oqP2aD&nI%$Od#UI
z!Q4Qd`3poro{QuNghtXgZ!fFVwCs{FP`^OFAh;k19UNrlZQko?ca+h}6Lc_kAb-D~
zxjeshIcj*sw!F8?dC!w%2SLTDfPhSkE^E=E9I=I?!gQ~S$GT5rfkF69Ml^5J)^If&
z?A&el?<smpYyEid+g~CrLVoLX?PW6Ycy}66(@gspy{Yo`au+uBvpzx?0iI|NwThI*
z;d%@O*fb?e?n=I2LYJs>yic1eY}!#+R^}_U6pVARsENj?I^23wkdceBja0tOo#)Rm
z7<%8ZW6lMtz+{g(jJ7kfX$u0gv-v5o+t%)RVTdk-{v+o8#cndJ{&fMIhM#S@v_Vec
z9^Tn4D4()GDg#1Yu1>M3pC*#T{~$1s%;SQ?y<|xrr)7QluR?3aJP!#|g6-8dt^PyE
zasyy5h9pC%;y3oVMAA($XdP$zblR6$B<s$+asQrX?SePwtZ(xgI~=&YS<e?mw#d$%
zVHi{fDFNZEt*`?cq`B5eG(Lo46ccQoW2DIN9BtiKK#R4Bzp5!qS^tk}LhVl}v7r8n
zCaTp|h}Vv!JMI<TZY%DX%N4M)3}#?#<lX2c2xITYHw>s6>kh1DT#*jVZ@ZKD4=}vQ
z2a#|PY#6yYh^4Jn`@3_Sb3R54r4S<~{u2>lH?I*r{K`P{=SRlL#0bni<SFD?;0IqD
z#vuaL)9TLuQ!52|0wk<}BA4SCsNIOaI?44Y!(h{lbwg5dsfc)0p)gqM*g@$H3a**d
zV<*6!Bb9t#RrxHH=BtUnrwrPWF=A_M=<_U24yGR;FX!V2U^=(euE1#@cmdTPSgVFI
zKv>loAraGeR*L(~uvS<k6V3Hm*(HP>(z>Pj3<vKLwL95;3Iy8u1i%(A)VehV5D@%>
zw?0hb8HcRd-4<1mz;C?z=(4BT2|f`HX9#BSNb4lhRq2+^ayR?1qp9_5UZC#=vQN7H
zBA>hF;Y{NX-@;kmYnN%rY%9l@AOAH5YX@Ch_NR7mqGnv30TPYNeCXHQG=Z#mB42X-
z_Jt=A$6IE8mLM;XU#f;y9h7Q)<sbbL3amZ7iw|a?zD^W=cthK%O<wWi-Tu_p>A1vl
z{pj^(P6Jnh6q5DCF#yK2ljeuCFPaI``>}8dsTRmooG-(CmjsL;SzKY~e!txE@JQ_<
z=UHOGQAd+UkQMW_=T{t_N9|m+;zWEu!@Bbn&xC>%hZ@>kVW$Px^4O&Ns<yu6%a+r%
zXSN%}N%Y)L?(M-mXz`Wg;-6)mxgU1qi(3Us{g(t`ofnQ&+TP<6`QgjzUKTi0*FT$^
zTy0K!d03HWH4^H)z}@OaK`Cy0u*c|f-u~CNAWk2t3w{{?FHBGWN8G<EHZhTWLXlCe
zy>#UHcjhpp$FYASF|f7rntsXmx~Om)!DH;?-L&z*G2Q-lJ!E9N5e*&Pjs-?7-nw2z
z`yszi*qxD0qM>ZaK+NO8^_w7I<4J&hZ6Va~K-MZMY_S4!MzQrzf@3MXrD&!KD{2cP
zzRu6ry=7mgZe8P&Up@#<E?`yB{4rN$H|SefmTue)L1w?+K(sqT=F>tB4>Kd~@RK_q
z0sW%~!pDSnU#TiTzn$NZj5`~UaZd{gUM6^zdVD9~v3{pW1Tz;)JqGo-*q;&_V5XjR
zXHppB4Q2*i-AU6zKwgZ0Kv2`kO`5vETd11U<GX+q;=9&Rmb)qKn2_2#KmNvtYAKu-
zHzz^WFmN+g<P$BiMW8jlDZ!%ISO4<R$m!KAXTeybpQz6X)ns36gp^VW8U)m2@f?;s
zw~k~<Ycg27aST71AQoK3t=w+c0u4%uzVtjS<z}2i8$ICA^CjdP!&+^Wna?u!Kc9HC
z3U|ivIm}`Q;Z+yH2CEy}uD#;!o_B4`t33ky^qGLQuZ;p(SA<ZLtki^^R#?(mz-f`+
zr5cXc_(0B_6O$E4G-PNQ*a&U+tpo>IW=#zN*<KE7a9$<ccyn*Iu6vu#s)2N+fA~T_
z8{d@)U(ihz+vQ)ue2>J5v;=orxXImh?>tvlS30?x3#_ub-JzfL!H?Odf!5*kN8(?b
zAdo}Zav_4S`@gWpFD1diNs5DkGVZdaaE{#8Tb%jG+o9T`tP5e`o;wyL!8ayx&B}k7
z`Gqrzx+!m5yRljMGJkQ2^gxz$Y`)|$&mM5?#SFfekd#Y!h<`R+y0N8fKKSLL)0AfF
z&B70J#K%|xpxa)&N7@^!k^iFmsbqY#khcP^9b1h_dh4nyZ>R1_4)r@=xQly+8KnLZ
zY>!9c>L0A%wh(Ekj>C|O|K#bE?zH{0kJgo7RpGZ=|G>SA=_@xug_Z9nOr`<nA8)gj
zeFT*-|GiZ{!0_|V4nbnKGIuvFlc&?I%)iF&#+sGgen%NiG*cp2xs3J^-Qk+9bNdx}
z`>(7k&I&${K&agbkAW4wojb3r<AgqD*W(!8ZJ$7s*;{dP4)moKOfG4Op=CG1?84|%
z26MI3=2It_ax6Y}QHjRsOE>&BnkmdoV7?XWJ!k6g?7_XVy74)T<BR5|SS1_q+6fB8
zWQD8e>mm4$x1cXy+m80lpP9|Zq59mCBT%k=?QNh;N)p%_CY*zVz<wCF%ryt2$R+aQ
zQ0Oa&_EFJeI0pR#@>a;U@HU;8|42nna2~;#693CD`x!LVY)1CGFAjy_ng3peZUTRH
zuVJSyps&?`q79-;R`M{4It#Vh?qTIpJJ_4B@c9ORCAt@GK3t!H@ys{~6{J*5crd<r
zGuG~OYx3K#*j`~z$gzdLDxl4fj<pxp6d8BV`~ikBI@e!jqnyZYA8C5;KdQ=G`ZMxg
zEa8S{=rvKYx@WBcQWDJ8-8br8YtSD20RaI;&87qNrAI6^8ziwoPv?YF>n`@bSxJ6Q
zbmhO?k`r$Rb@Ni6GDLaNac6@-Sq+Od7tdQ84~1Lm`nf`iK9ZWo*6j+NG3_ZK3VY$i
zTS=p*MO#LI93ksVy|LgAVrl=<?MiD(A?u0DL>?FTa1_|-%==LG5_*z1-Xsa$@_=qF
zG3PmTuOnzNh(7jrU;)Mesghgw=jr?WzV0^_=VXVnKGY#1x&TS=UXQtC(7N^~vPbvM
zJ249K%pJt4+#R3fG~qH{(BS1?2$~p~Kf=KyonJ>=5MEzMJ%jBD|51T<QWR7(I(7HY
zT9HS$UCUcGzFgE4zTtBgTh+uEN6{1e(4d*AI*qA8OxkV0NYR76AU!z~YFeRJ6baRu
zegAUl?!tO`lkSwi!n1MZ9}XiBLRV)-%QcX3p{!sF&JN7I6&dBt1AV^;c#rw5|MF=@
z{jJ-DZZX}>eOUNWXbENvM?DA$U^eo}wNhnJ^Cyz5eE7mxy{c39o*!s$I3%^%Y&|3w
zC+_O*W&Oe2R-PAOb^TnmHu8@3rujbd{dzNly6^Lmk3*qXH2psZ?#N<FZq#rFUL1l}
zi~vfoJMB+uy?1IvjJMD@8sdM@+5BUzLCNXeTEtBV_S67oo7oqb%WNm+4z*cVh@fiO
z)5Ck9U=5U{#cS-KsqRPLlbi_5IvTq-W%Vxu*bdl9Ot3CKy41Br=~YU`&(W2Lv8FG<
z9W+p4kryWjm0Xt~2;^?Eds%Or@j|l;p!<Cq&<KD6CaN#J@g@q|0n^3jC&J<g@89tK
zbL`UwC@U>x3io$t#f)o`Vf4{{sb?-qOw^YoalENJ;p2LFf)3!#x=~FnMMkjO>lQS@
zQg=~@Ai|ps2-WO+vl6&Dc3cWA?4V4)F5n`_2&{nq!*<(#*Ws&}VGJ*3mh7=$EJ{&r
zt9$ZJe_W+Y8)-0ZGkc{3*Urn=)M9nKy1Q*4bX%l{`?PZYeTZf%4eiJKqRH68Snh(d
zf$a6W%H=#XZ%#0<_&G23EG0je(2ma_s`A@ELP}D-y-iFG17ZK0;Cjq{uKgDj4>8HU
z;j4DLYDY^|J3!adocR^&_#75`iWe3O3TUxH4H2+t8bG3?4eR$gRcD;cE1cfd1gF)}
zpaZ@%p7@TAauXpk2ZeJQnga3EDx7x2t=oY&^w^z+doz#uWPrcWBu&rMG={`_o6t|E
zasB?3P~FhFdxcO=e?vg%e&e8)L8~N>JpY_H1wq_5=9i~16FVcxI3@u34r2h`6Lr7y
zTW3L(V^r3@Xb9_js=?;pQ(lln4$~tH)CC|N-?T|5=C_3uia@$|ReFQCw}smY3q?=S
z7+Vm;N{Sc@D!>ri{7}MfJ7Z{UB11-@Xp2V0)R`2S2XtkAINSATKb=*FcOT8Pu@#`R
zvrp)4ly41G-p<H_zl62j)gj4SsK(UF39Un?Cc8FqSSmu=skA^8kH7hy>3BM@T6cU#
z>XcpY^6(Mw{JHZp-fwj7A#67~LH`Vv{{6Q+!);(nQfNMToqE$kYv5;IQQT#F77OCH
z2gOmX&Q}Jhd0%=rAY$6<E?Bp-7J`VGGwaF&yKi14ROZ?1Qu0t9;!bbD?@0ZSxsI*D
z`&Vq1Cm%urEX2HSF;*)QWkquB6x6Y>>Hv1}%v=0yg*_CJK}$Ud8R6Dn^!PB8fj}0y
zK&vjG-^P3-iGpA@?&)4|<}~h$yp84~<4NC(tu;;ZQoFQ0G7JZr;V#6O3v+avz$XH)
z5AR>!LhkslGrf*$SQlFdMdwyfkDseXh@KA{%rK`Np%wn|*GrE|!UkFpx@9L^yK9<4
zYl;}p0bYUm)CRG}&YyU3cS~2S(56)A4QcvS@QF)4|2-oa0_z@J5nO>m^O!D7+#VO`
z&7Y2N7S#PQ*$xC*U6~wy>xL+|FTF)Rf>0K)CnVS!tp7E}NTW=epXZGU<eExiOG_C*
zhTJcxF|{?#aEvf}_8)vSM5evXpG1Joz&{gZG9bD9-=27fXv>pdZ|(j9-S&T^lv-$C
zn}U>jH|N!sgC~W}mdNI5fl(Rcne$Vh%I>0L<mst)(ZrTWae%b`DKho;l?`hV-1)}V
z2q2n5y>dt8^GnHDX!Tk7#m5RM-Un+Bo3=J7);at6sPl+<sqtAU9|VIb3GP&qfS>D_
z&G1O0>MwE}$GLMfW+$g$=ksZ(3_ZTb2*58=cL3;FtOxgBkLszYs|U@2=s8s&<DXk@
zxg*Wz${4$vamDGtVJV2mpmcj+gZ3YPyYr4d`ro%G*ZuXq#sB)=#NZf|!(Ti6udOAF
zIG4CYvy26_{!31!G~-MN3?-NKhDd(=TG<|M=K_EF6N%$FIIIcO6nihO>t9rHd_}L}
zFP&`u^{3<rG*8+jXQIQR_^}Gvc89rI*!m6VVqtriE;!3r$N?pRPHLLFUklWEG1c|$
z@T?RHZwV^&8duDjt7fzU?Nhf-q**f>e=Mq~PkG9O@Y}n-?3f8^P3SD7F*&IUD%7(@
zNw9XucBC(FzX>G#mpKL%1{2B!#}}Y`u~#NM^W!k0(dkc)rNhUd11Zf0!URla+#mjY
zz3}1MtI*jbNEZEp6C9TJlZ9E6?XB}^mD+z7f%Q%$OI#AR-e%R=Iqa^In{{SfiTxt1
zTD=svhX<c$8pnBwxwLMp8`%Ar&=qh!ESP9<E$%h}9Ix(^PTfKc)cayi_wrIlK66}k
z>i4)Z97+~F5jvRf&MTg7PGb?C{A(CjI5{(Cv_1DtvuX1u7*{>S_Gnq2wP-sz{2Qsa
zIZ>&>p8L@M;RdHBhE`|vCqVR-uLAeNOP*+9Bo@~#hpNx)+kc2=+{^@)nF;JKFWrqc
zWu>t*=p?`E&lv1!cN2^s+bfWtxI^X|Z0qp%4h0JZ{p4PtJbL3L1W<eq9h5wuzX_UX
zYjPzYL|kazd@Kx@?IE1KDGyyE-3^XKbhlxC*wdUGl!Uiw)7Nz=%`Je*@Y>HF-*oJc
z+yU|~=$<hgy>ef?$nf-pL&7@O;M^rg2R5O;iB^wkZt2rzM)O@X^WKos+1xD)w?r)d
z8R=_=#7`OEq<~XNc+~e6EscEdh+Va4-|%P0;?-f@%yYlZ6%mCz)KlUQZ-~%GlYKp9
zp~~h`En>6(%}j@Wn*`K3OTcF_+>vnw;Ob%iBLF#wDe35nkRkfKV!yBcy^auS=XT>Z
z7k`YH_=#b%C*FXqzf2Yg7!LZ3Ds=+gVvi4?51-I#cmC_}dYJz<gFkq45G)FRxPD9g
zPnDSIXe7v!O?Yn3kt04Y$~3|SEJebh!G}|OM*bY%^H-iqYbd4;Q}X)}MP)D`uz4;x
z$Wi~t#PmM53@c(NkgU!$d*j_|hisXjdO*~#pnO13VXhO%HaMJRp73C8@AiF1?rHO$
z-QusyXuOMelr^pk9km{}x2|-c`=IT|>kcGa_H9caL$<BOtT)PT(7cKI0f$+8Jj+1+
z{q_{m8h%ly`QX_q!23J3Do|n2{+swF6R22euq<@SOl#x3mfwY&<$`x3k4?i+?69l^
zVNn-!s=kO1ne!+}Z~_%0AV+rRYjf2T$FW)V+}~a6o?M14DQs8Gn`yzWm1?i@TtSQG
z69SO#_NPzo9Sp!^NWp|ap@COF`>EsXq___ZX{A2_U1|&eLaw_LXbnqw<j?oXcGZSU
zH4{^a3`5=AGt}j@iq8c}%M1>m3`RgT5yn=t?^v%p61$gpT;Q%;{nOHL8F~L-qgHYx
zsBXRRuz@|G#e9MvNMbNY^Ud>1{BXqyS(K;{6#6~S2{fKBO3;W6RoeR_Z&TlABhv#&
z8;d#kYT`-d(x0kJ{nvf<!Nd~l!{BHO-)UtZ?m|&!OR<?bOig-6_}2xM37*#EF;r&2
zIJMb){dPOvbNyE7=<`v^m~zhFnE8?~+^z|{R9k8NvJw(10zrGa<wl@@n393t`^AHe
zdyec<o?QRSh)r>l@_bNav;uvM=@H1otYv(HvCwwE_hWYSb_3D~Mgx6*J!r~Q?3llo
z@P<PErYXVdrd#jNGO<3#XcFdJrY<j<Z^uO6AwbQ@pr}}@{Uwu}pdvY>H+bParbVep
z4t@JdwXs$+wWE1+Dz3q;QpnzwiGES0TsZ4;skt=ya{p&71OS*44PF8?21eE2oct~#
zpi1`TWF_rNya6PKgrl76e?*vReo*3bBqQ?>z_KQ*G&#BDs8GNBjxpNqDllzS|3@EH
zeT?2&&7*tw$|(QG;zBQb8y6eye;#?cKDjs_7;3swDw2R0PN<avT&<XWD3rac7V0VO
zr<LlBXEr|hc8QN2j62%q&j`=FN=YWdcK`ZZIGISRGNUb>n1)H~q2ivw@3GM%e`d3^
z(uA*JTfdeXL4J2o*R>jM^L8?RFPVr^35S4-I6=ac{_Xp3ds}>+7C=9?=qlZwe;6d5
zx-3`lHxOD2+U7Gecoy?G3>{-K<if4PW@>JfD#c;tmr}v_+0Kf~vgUU+Zq|d(gt8w;
zGc%o<XUoW^`{f@e${z4sBPoe|JSEDdPJ}#mB_FXpXSxWR=$9FJy_or2XaV39kq_Y_
z-nrM<f)o27ScDOg9z~vs_o}Ho7kQQ`#KryR!kJhX|1$IPV$8gSdzt8fo3enO>V@h)
zT4{gk9Ub&ucJ9!HfA>d#8JPdx0qS>Ua~VhDq<OWkM-EdmT7<jP*lCV}3&<E^Grsk%
zV%njF&$vbltAU2n^Qw8t(aLkNhcbCUc>6eE+_trI@Ap&iy217_TCH~WOPui6=t=7k
zf?DDwxZM;!IJ@ys^@`%>yyMHjFJX=>Rch7bX9}EE&E!y6G2y+Z*6$}DRywE-E=2Ji
zt$$)<nZr94>nUUQ8P4)lGa^d`eywy2fLD+kS-ytD8sTPA0bz+p(vlr_8<XKU%7~C3
zVfm)ZB9BKp%>vQGMM+W$Ari>Ffxd~VlZN=a3sh1ta(CL1wrOHCf4fNf_HcJ&+dS|F
z)k?_zJMTdf<d`lYMnM=%x3s}4Vq3g!q@lF#EdJ$LIgC2*4U2nNb0!{A@`dOQrS53M
zfryV+G-c27gPjt*Pu?E$yq}0-Vv1kn0u<}9R{rFBw|$TA6|Ai$*}@e4sx#u=yljm+
zk;RlWH1cmQU$Y+Dnq?+#tVWC29`WU~P^Ct&lop)?$+2~H%2yK4Q}5w(A>V%!C-LT~
zeKMS!nD3bDvQ*ByX=4QA-&~Fv8+V!cQv=!O^!QExjh&~OCs-w$c(D3Yv6~5#N3_*D
zf-CQuVm1aG-9K6Gl9ja6bQV+>59n=X3-cb_a5AAO9t~EhqX_Z_xpY`M%$d%_wz}e1
z)={gtLk=%|WYWIJ2WcM3$5=Lv=)VYz>ZC3P$-i=I;8|^Db)O32qg*NeW;InKbDkFS
z9Sqa&Ex4vuUnD1%7nIo}Fc=a*im%&}WG_34NUuMt3cQ7t;B`XvAZroR)m3&aNzX3m
zSKR%NJf8g$E-Gj}W?W?YMRcY-W9$*EHog{8m``?bhSWWckNo`gg2NKWh2Ligzv773
z#`$Y`yh)brai-hl!=U#wsL_lxlGYyrVM!fnvY*58v4sMXfA3*c70X8Gk}vHSl0Jv1
zd$DLkm`q5=Jj$FCqI%31t8I17<v2W5YJ5I@$TB$h2%cx@C$E<C{5JNY*8YBlev1by
zF&?lxwxL64Z)Rz4P(VI5v8C6B+R~FXDw(XV$MV@*az~<gIU<;T+qu)M+hR}rL$M<v
zWj`Zz(s6!8zTW%89;N8Q3GG!zRzQZRauZDp`T4*}#P^J*=B9=DUp~2v8MA_<-$xn|
zVuld*Pnk@~K3&v?Iv+JMzxK7vJ8}t2Zf-OkCj2RULbE~Az;!_tV~FdT7s%wiShO%&
zK4Q1%3`=Z(s}*HHBXs0RATE)NSK0C--q|5r+3@iX1*e8fZehBbZI}!;Wx;rCh9JwA
zkTJaIk~|~6SH$KSX{JV#^^NBIY%pTQep)<&$>_KJufCk`>tKgGVEf}8W$_S=y*9RM
zpQp?vYWLP4n8DV~hq_Uq_vPvy>6rLp?^C)LLfNiNXFpq2mlHhgd<U9tMB8<^HturR
zuO&WXCeP>bP(t_@Y^4s}rB`HZ>p$fj6wtuXo*p+GJ^U_c<dBx$@t36=Zt&+@=w*I}
z#S#Tqieu6j6)lk@Jx1I`Y$Lu|->@J4K0byi9K_aaRn~RtwTlcMG7DOszpJE{B}rr}
zDTRjZfBsIKFG5M~f7SjTmG1!(RezV?-GXzwJ3J)BF9JLL@zVbEitIJ%M$ds>GivPe
z=73YYm}YdaMrrNE*BUR9riPXfMdI;KGe(2CXP?@0C~*w)5#*$CpIIV(RaEY<wrO5S
zHG+J0u)M*DP~&GepT0%PF3RYe{|LvKdgwr=%p<a>Xsm9O{?tWCZ1+1X>OxerBXZ5E
zXD+kgHPr5j4{ttZ4V;<8#0R@)K9(-?q+clb0=g6}9P-Vr)h$Qi)-dyis~s*zfBazn
zY#ye!yghKe2YpW@d4leZxeDgiQik0BztS6VFB4WDX19kXy#JBjTx_fzEVvz9EbMHA
z{%I3_CuyIiP0}5cFu9Pfnb)*5M-e5bkQ_^5($D?IMZQE|73-x~lb(XC<pC`LX^9An
z*UrI=B9TH=sv*_D@W248YNSq&60^37#CG?}Im3IW%Gj@D8f%>)a6}<ufSFz~v;hm;
zOZ2Ut3zbG6OzX2ZnjEdEN|%CvjJ)p!^N_3~ClOtvKfRY|s42AyXrOxodIJ723Qi~4
zEM`15=E(`Va>3^D)kfXdywhAB`jPoyZi2IB&QVa{X-+HZ)S`N!C*Gv@qt{)+*SB7;
zHA}@`n(ZDg9l79RwM$$yIHA)Pr_A>JYwE4$%x`8~PDzgz+fIKb4^N+DhI4&XV3d-a
zJsLB(+wK%{wKkq~mLGDxq@l<Wyk&sHTO~TfI$g~Lw$}rr-QF869=@v<Az7n=DGv2i
zJ6G@vmpH3U4p5*m>m(*NysG?fb6JJc`cT^|{&#<h+U`Omr4r8^rCM$4J~<&TT<dIn
zkG{1#?W$+zvbP#~2O>)=pUyT9o*sVVey?-bA#JnXocYDIOucLH>6SBaaM{2v0Y4RF
z^}6bO!4GGD|GmG*@`C>SWxJKriG!PiTC+*a67qJ-X#A^^$yXZOd%Tzp7)0)s;eW#K
zqoklS>n0R)s%fvgBA?WlBro^HA~=3#A8A2o912YIzq_~O{`(ORPR+gRF^iQFEfLXa
zVvRlWY-0^9GmBNN*B_`d8lU%N**xJs6@8N@c-{-d|1ctWSREc@$_>pHIe!;=R@LnX
zFe9k7)QAwWE{svR(}!Zgl7S#d4vpIm`-BBC72nCeJD9p2-soY&K#=e%xRr^3lkql$
zkdK9yYzOCo%##m>&j`M*!&qFZ00ym;?&(&?6GI3QRbX-ncBmXd$35tyIyKSQx|iW0
za&sH@1VBNbSAu~I!rcpxO%ao$BSKg7_Xh4S++nEO|9`Z&F1H331>oMj>s73K&oH#l
z*VV(C+tbF&%i*J)=U+nCUtkPp7WOEFV{M9ac!Ayt=UMyWXC_EJBCy2yaX=1UJ+Pmn
z_(ae<Y7e`pn3!g8czOU#vRHWRQPziI4smXHuV1rKbO)0L1)WtxyRVNi9pS#_A-uPX
z#{&Sf?f~TN@;v<ZWS{%?7Jg*TZGN-2Q9%X=2Ayo7I}OZ3PEZCkpsNwk72pmw5rTQL
zTC?b?F$rWD>q`UjN1g6-15xL@C;;&0bYlc=)*W<mGO-qNw@3Zk6&mshwN8A8{6hl@
zM6LHpqmR)5b6^Mx9j6HhIlCU)*96@)JdwX!-*#9FK@SkptOZ@1k${3x*RUuA4fO7E
znt!tz0YV?{X90q5hqxJ}ZUTZ4m~G2YS<+oW7e~|E<~OH()xtG@(49%p?%Uf9?#6C(
z()b{AFk*xegld&q_jQHdocc;9_JJ<4q(hF9;L?GpsAqSDFi9-n-SI7~I*CT~YpZMU
z^-3$Z83@%UV1-VLx&1L4GWae8Iqe#PR3A8Yzq)a~>o`qvy|oS$IV{pR3F*2Wiz1^5
z7PH0uZ8)%;WsaVe1_s>y8E{>b(UVTRhy#Td-@?$SR?ziHHQYSt(=c5SdO$4j1a@|3
z-gR?+?F+d5)3+bB21K5tgaKF8WYCa1VeU1)YXArtH9>j4Vc=>ep@&)~1Oe|Z@o(3C
z0c%01+dlIG^N`8?Gbj}3cLutZ*_{>{{dJjS-W}i>h}eTk%ic_c$hLFe?Vf2$+-*o7
z?*rD**Pz2Y=@2v=galpfLxI<aaWrdz*PUb_<kmC_UL+aR(=*X`Iz1PNK*2!)7e{AI
z*U%qP=Ep`$npc>+8Tk9yEa++rMaC`Kxq<G~q*((YVVKys6NXDu{|F=0+M8gI;{US!
zE5z~{^rqjM{N(7&;_f^o#Fi=qeY<QhF?4(U=b}ifMCWXK)ercf7!?RaMy>UcNO#>{
zz9_ognzajSGrSST#MbQP(emAG-;gH2La_*RBV8XB;&lIj9rKsPw-C(e%vS+sf!Cj1
zu0zhPiZs_T2gsv9!RS9&zFVuUCO_{El7f&UbW%ex5y;C8?z_`gI5gxs68DP*aHuw;
z=yDy|7^3*af(5|@mKY18I%5PsX1)sC2{{!vIDGkXd)LnKZi9DYDWv>NS~d~5y$JaP
zSyE+hz_9_kv;>))!PQ;bu@^NKTOEpQu&04Ve9QIDFDv`lQ12>agDd9bflY_YDIpcR
zOJe<LL+87xi~j5ld%;}N_z}`lhrtP_;%Ae5`U!m^ur$gjPq2VkqI?#wRfaD45HkHm
z{C+yZp9&yYG$FF);bHYMa&0iIG@*h=qw#RODO25bqiKXSe%c$hEHc*uM$EL>9QA&A
zXXtIJyg2Ev@nuDDdigzEal9!5J&u|yXV+a{C>Pg-FI;K^o07t7YO{1%>NS<Ry9pZh
zfOr536mS@LQSvxpe5cgUS0=(|t9{8C#Sy?`-Z<tnj6BYL7Hs;Z(oU5_-US%|`4R2R
z(dQ2fV(+U3c|wuu_*vr6BEDP>)(wt|rye$}SS2hu&qvx`<@n=AaW>YwBR5rZNGYec
z0=HE^|EAZ^Xo%j@g0zoRRk)A)epB}zPJh2A2FBYH#gClXC)2-7V$`jiXKZYi-IQ|}
z9F~pfOc&aRkGep_A#|I01J0el>^t-F9k{~}?BPoDXreXaYtdo6Em>=;oKEl|A9vcX
z9cR9_(dA}XP+n3hlWvY?isv-XQ@<96@5Kgh``ls1d{5^K4R{hZ3U^&7_EY?31ul7~
z1;ObKb`*%^cW(3SN6UJHZ!W*8_YRcv4RhwG0+1gyZ3gR-*T`w*=5-C#%3CX)I}h&f
zea@c{5Rh-X^mEZ|l+0dwZH^b7(OuWa5s>hO+uSHmWxMiYoYsrzn%cFfn%ac41KGR-
z^<{V5?q*%<<C!5YEB|Kbu1nhL?sB?dR3(ko7MCtagVNK=EkR%juKV5B?67Iu)Y^(V
zR;64_-ME}cMfrPcVm97-mFy<=Bl;QRAs^Ukr4?D8Jarc98Cx<F+@9K(+VcQeICoNm
zoC|kGa?R)vEEe>7vcZij8xwXd2`g<#V*#)=YoS*+$ijEoH-+;;FaEfC=hg+P?KRal
zluVm$2JUriqOhu;n_<%%dDT_~If!rG&iJh5FwFYI3B3=<%5pl)JDyzf{R3$>SJbfJ
zYANwt<O`|h7iKBib#8biCc$sodcTZ*eD2FW8BPz)ZnulGrLtq!`_m<(oc_~kRnD9$
zBZummS}IFv`vL1W>{<3KE)<!g8Ot-a$XA8ALght<LYD7$!hWheW%Vo%f7d#aJVKPH
zvGdj9<Cs_I%J|LK6WInFunV%#huUK)DA~Y!-*(oedql0+yU|$a)oe~u%i>zX({DKj
zt5LFMAF`FB0;OmLN7|*-Pr3eF=sK&X8F<qy1r6~Z_R+6^2=f-iP;OvxeB!K~lc0Qs
zaD^b^7=ni3ti(o@?L<jNUHZ~Sn<MDSi_OkoBLa4@IM2rCM_y;9j-N448%K4LPr)*#
zD@wrGVsojQopYjRlV*g=)HSt+%6qlKi`q{N8%hpIR?8T|`JuZ(nZ7*H_Q~Dc^sNEr
zFqb)$S;Lp9?Xz?gzqzF>{A|{Flrul@)sK(;X?cHe{FGA~ea!^dtiJWXj47NTQShqI
z31DL<Z|#8TTZN`US@oML+DG34E2ugT5*_=mxD*3=mp@^Iv?gXP$JQ>@W<a@GjShcR
z)q!zAq}*0VYzD=Mf69v)wS#$2{^i4!V{cg6#>bJI78~$$w~ujMt=qPMz4iiqpNiG>
zp=^uNh88K#9CQAy7T!4f)-#jxz;M})8jAf=truHt@}!d++kV6WFQQepm3y+d3jo^o
zcEF^r7HstISdxY(s6v}zzE_q>=7Va^_tgvsgt6&;?y#pdrLdFPLw3VtCT2R-^L9=~
zykbGYmEAB@<4p6m?)yD1(k1Na51AEe;dGn2_7@O=4mV2J<`R9!UD$&h(Ss?j`e4qL
zY56&yxXY$<ZD&fgqJF+8^PTjx{n_tT6UBbqLCveXomUQ7!O_YtyA|L=o7eet@bRuV
zbI%-2!)4WMjq>mp6Y}L(-Xbl5n)81$QH_bVWrNJz9n#>G-4E2;xq|u<v(Mui6SHb5
zDxB6Po7JB?e4aHoQ)1`(u71L977-XSXDRH&*{Z@%6q8|8A-lFrdzxv;y9cRaKHCIj
zKjp7@zfZq**IY+`f4<|5b+3_TjD^uMEnKvTI)$9JmvwG#0Kc^td5W?-Y+U8AjN@ZA
z-_|}~^Dcb%&51<uoA&I3IX`j!ZkN8M&1RQyq)+P~v9ns}k0W;a!W$O_TxG77Kl-C4
zFRfGHFURIv61Eane{|UGt$yMLAN{6FZ~k*RhOlGL8uuCQjNtVMr%||cp?Bkfvv6nD
z#JNnr287jWoTe_Y(evDCnD)a*oixrTpW=VbcW^p|EK<x>bcNXh^J{CL-?cQG7Apv=
z+aiTNq8A#im$v8;UU|$XSReUw^=}LPc*U{}>bG1Py9Ha3&uQ+Yvh$bSRqUR6(AbI|
z2K_)xNF7)PEDb3#PMeXQl)!DWzC)-x-vgP%ft5A02h)rxlE0!<&Z#SNSlx?!CZiF%
z>F+DMb0d8KE@S5{^RVQz%8>$BbfS%h4-Uw1%Qc|!eIGS*+<|>Kt5Sz;eE(LxQ`FcO
zNlja0|E=x-am7dQ!%X|5ZsNwwkcx>G@%Q&^0H}^EiDe1g?Ttj0hii4hSLCrb<2@6b
z@cSv9Zcoso=RRNEYy!Z%1ies#O`kQabs7b(P?i2M?Q%|Cuy$3~xy-nFjIu+qInB;V
zgX8(9#mR<f!5fweaHJIlp&_L#ry&>BG+WCB))75(RJ(y~pM19KoVKzbjH)>rarZs0
ze~a4w%CPX&i;mdVO?%R3-c7ch(_6e}tyo!3Tdstb?C()5PT~fXo-r-%O~#Cmq8Eyj
z#%;jtJBkt!JNMRWMf}MM;6FuQcTW4(57D5L$xbI-W00b_92t3n%W-!yks0E`uDWxy
zFBzm_tk#(JE<y+Fjywn&`^yUYUcbuoTxjjuBUXE3cJXE@vgQLv?|?>e9$#c=(<nfb
zgQ+D<p8nDJ$#WC|INj#>%}!S!p^4!;tpiiJ_{1(LjR#yd0Ok#x?WQ-@Lecgn%v5v1
z+av|+jP@G9rsDE}&IB^6(cD$Tw}pidH>?Q)*bJoh*)j)-JJq&a?HFQN0Ui1$Dn(Zz
z*3czZUUfV10o9-uwLlu?&#F0b1wkpTql=tQk2k?xdyIIDBay^dE+d`c;_r@nyVRZx
zK2}bpj7KPbcLXXx^3OP*36s4EK?c-)CUd+gbozn&j>xaw%zD^nsa>hJ`4mos%7FI@
z9(jaw%kiADn%IztALz#_l$e$Jf{)S%E^AGmYr6k2l{(FOPQno&!(v|^j;~pNVBak9
zBQ8`8NiB!RpJ3+wqIxLOmU%nemCduXy6}0u&#YR=Ht8zNF|EL%lhNt2x1Oy+=6mXa
zY*9a(x5eY}d@6;T9<o0Wk4D^PqP>FZPcHy|q!X)xuFu2W$2!5&l=Bu_T*8-Y>aL$c
zHqlu0q61f(t)l?AI~me5sr9&@GEF{{&(|=5;HV*{70#7q-4=QtxI}Ad)A^<%gSQlF
zedM8eqqRvQ4BQJ0hdUJVAFdF&mbd01^69qPJDq4+yM$7EU~2U|kM&cST1T9-jz6`1
z?h<t6kZve>>k##md9k%!Wku}PK1Ktyt=M%c9Go-C>8iuMVfuUU>Ef3MVeEW|lqzF@
z2gyt$(>TN0oSP0$a?+$9pDc#mm;2;zw%lf2V99v0881x!&{wfA0U_7L6u;=<MGH>Z
z2%ury=X&3!(mla`FJ=bye&AG%N<o5^)~O}<^MNRk43nVTzY7Lf$!1a&;fBGaLs)&6
zlb3fA0w`QJR3gqenUzC<zhs$3BmfD=+L#-_>mro8U?X&rr#)sde1b!mDvAwGVIY>_
zDp?D*GZwHqs-ckpQ}}V~r#WtQ(+129ykZ;zhf;~YBikHJ$&Pr`;=J*oCz(Xkg3(17
z%5h4eDkYkHFfBiNs*ChhidnPn&lRLg=2Xs2vf5n~!lm~iOWug5kOJq}42z_ir?Lj;
z`?oake&Olw!b#g~m$GVrV_A!D29`6}4e)He+Nl0OwNINGgL|G|G?2|Y6cyT30}my#
z$wY*5=*q{skyZKP63Wt%vi7A3VK$7tc@VSN{7i*|L|V;^vhhtqUdq%HBq!EX1De8m
zSc;U$^3i%#<zC8}!%I(FD3_uGqt0XNQGW?=x@0x22>I!f8_-&nBYW=5OxH<nyeZvq
zyoux_Fe&1jHI&Zh14MpPKW&j(P-Re{9iOq^jt1X~%k=(5_bDD*HL*1W>G|HbZH2Gg
zhFi#rF`R0XM|G3Q0^|uJ->w%JCc?_`IFM4AaDAIRuv1Gdwlyw<qY&dz5b2ds6Zhiu
z{i#~<=&Ze+nWC7k6MX>pGwDXKQejhZ*;}iTmT`9qO+Rk+hle(jDN9clD};6OWPPG&
zlk-I-G{DoNgfh;)uM=6O1fNNZqSBz1NK-=k`dV7!am$D8o)6bVW=+jFh~F?8Le!}=
zx5SB443M@x$p-;TbP^}pT8Uk*L6G`d7Qs}}n{>tz;2tIzk7+x*mb82${43*W;q7AQ
z2PX=ZDY*=U4c3*0F-_nuGDc&A?6*~TVfmuD7Bwqrfq1l{^6|XuoH&b~LjB>ma3++%
zGO;$2k$|T3F;iIlWa>b+p)+N(v*`0jvmTI^Wt`!)_e3Xlk}55T#Aa1JK9v`{qm5=>
zWT||O6B&zKlU<e0^b)U>OS{HcCbA?Qtp-{Okt|B`DG6<Yp0CIoT~2eWO0B@~-{)f{
zHRDc+1XTNhVB_*)#cczPf-)}3mxQ)V_l7T>2Az8YD(@JC@%<+Rc~ctYFHfYOe}d%a
z8<o=+JU%B*h=b&b_C-c%4{JvkrPRnuz5d3bDZ5F%@*y@SNeB-B&e|(rc&*N1n~;~$
z?%~8~@c0`?Z@y#-t?ii8+Xm^Eo-j6XLi(EoEmpsY&(Rt8w7KUplm=P?8;vTA-ly->
z_HjRb#+eZMxzJpcwioai-JpLaFw)_{AS<#=Oevaz40v^!iM(GYUoUX^hon<&C^f@{
zfYuU9H!6E|UnTKAW24+DqfBBhoXY#Lhs3}_N`>K+PQA?eq?x*|;>%G{R!z=!s+^}f
z@WApJX|Ov%{H`b*3$q))N=Xnso@uhNnXaD2^Q5=TYR2zOxAU#|FFU~K`zZ(^l|)X8
zn0uq{1cdaN91!+68_d23joB_Ef?cl|=Pe$zdv4Ay1U%a(D>G8Cl;}y!frl<2&kfJn
z@N^*66q<>v?S5%zvfeojTL}(YYfKM`hMjz0d$N1Na<2Mo7{>az01u>;ov(Nn4w$Iz
zAUD>Fxb!AZ$~n4x%fCBs#qcwk+CX_cV3g|q-JhI4++Iq;GCM<k3L2LyMKnJcZIYB;
zVwy1qRnDIn)|noZq!U97&*SNG#Q^sZnsX7h5MmqSd(kT#h0DdLvlY_A%=UwT9|jn?
z{PxrOitn-$mwcuzzselzmSI(4{N@deMIqVr<1in|%kXlJIdtUGgdrEl!^BO>E-fG%
z^cq_!#g{}OP!u1TkmEzD6LjWcqCzq%u8PK*oQru!D(U_$(Xd>B=#T!4I(U(-Un=dI
z9RA<}&x7%$m95`=S{ZZZhlYZI*9vH?dc=B22j=Y^dTg&MJ(Bk|!nAxuLRcf~Z3vIW
zM}{xd%x*>`Vx3KF5B&VFS5j}vHmzd?O0H!_auKQ@Q|aQsk>0e-ymbjikJgnd%F|Tr
zs;F=W0OSn~vq1xw4D4U8WrE*m1<6Q2o#+x6Lo{XIJnwHB&(J#<HJx)v_nm{Bfbm{e
zwZmSh$E`%Zz=ql`d!{c>W67`Wu;12<D$Dkhc1tOBcI~khyi{_0366uwS}X6{cFOZb
z9NdpSU@0?-@S&K<xsh~9jV9Qm+3XP6Sh7E7XsoXYI}mM6TI5PG+(~K&;^)FRC+*k;
z$!wroyhDA7SCI<R@7?okjvAi6J>a6uk~$L}HxjTk%UyrFw6jRt_>3@5(}IDUSr=EI
zK;fgSwUn(HcIrMy{HAEkcx#G#fu=>?CmRL4#XUA|Qlgmi<>NzJ$%hJSo|LfR#{6ZT
zpD<%NiEk@KwQpYBx$l#lkDD(sVa4$3b*J)njwx;Z&;rLN2u%p>XX%)8c&ar>F;aLl
z6iS@6;tZ#^5j$sx*wM`8Kjl5b@}^uz9E5L5d^o4c3|tbg=V@qR7NWZww|~dFnQAY$
z)6yP`@04Z)pq+V>{^F<K!*<wPHLjGd^UJB0ny7=W<g>iT3P+b_rKhkqVqf+Oo(z6n
zZqc0e>T=t#ogd=b7RX1s{IXuWj<20h%UqHd-tEzVm!DXV2xqh5Jjd_Q9n;%MkRuez
zn^7)yA4*}T4_|3d-EKq3(#Okg&bIhE+vYs)oMGJX6-v2K$X9jVPh4a;n?(D$ZNC^r
z6v8h*hx44Vt=D;IQc$Hooa={*K043|b6)qOZ8f!)^lNFr8Dc@I>#4sV!1Isq{Nxt%
zGkKnFmSmIqMo*UosW$Q&_$xe?L&!GESk5LQS9Cx&T|?>+3lKObB*Q4Q{x!KGpLGsB
zO88xslP~)kM@iiKc{d(moLd%_q}lhZ=hQhgR2P}X9!>hi-wFJOhk+J7;K1}fEc%>8
zjqWpbx`;9xR!=QOLtv9Dr`pf(2!(H0av<v``sq*G2Vy;STqOYsD<Lr%^yNA=4`~}p
ztY62xk~;#q#S*{EK>Y51z*0Tfkc{}vsHIO78*PIp`gj;(=dW@3NaFG6IEkE3XKkp7
z4zf__%_c5#=ecae+J}I910z{e%Cu+Ujk;8VIug-tz_c}L!zjsN+?-SAHs{SI)<GvQ
z6CzzD%p~g`)3cOU25u>Ia}QHgp<S}PYB1gndccEC^Ey5;DV19tqWX@sMNVH7+Y~4T
zT6ww-Qi&OT9^UmarO99sc5z=imyg<HW`9)YEH~DTQKY!5Tt>lmvP7ZFiVUIz^xTY5
z>%s@qZI>Wzy&Or8G#|C^;8=&p0VotKdA@rIj3f%}+K8FGFKa06KBlu9_tmDBV*#oq
z3HRP`os1-Fa|pSfnkUq`D?+O5;?@{7S3WW{dqvQURvd@t`37y3Ab!M_ohS*}O8IXn
zO^ExaeaV1NytTH=<lD@|XAZf}wA(A=N=m?>!C8M^c6(sR$LlA<S)|h&uGguyKu_WV
zuIlpG?1cIiMgg_@Bj_|8FS#DWxah&gm?s@n^t;%+W~s{|6YsLG)pjvQk1|`KW$_~z
zS7@~Ke)-GTgFOAXm#$R<;H)aZ^-)xp6qvR(O@5c8s@H&14>RBu8n#i~LoF=1A+Y#<
zA;MYNf735>nYQ1uoF7}%(~CG-ui^eTImN=UbttZJX-<B;VFPzM?}#@fn{$Yi*LK47
zWwQ900=>8Umn|r(j%)HWyE-}9v0A$6op9^Q$9!94Bf358C^8wN;P$ZS%ZGA%Znsl!
zuHSt<c`CX^@{r%XuvQ0Go~ZIjah*Yb3DZxuUtG#*od^rSqvv){mIP_aCK~%hsnJ03
zZL%NVj#Zbtny2T+r;I-#Oht!k7Am#4f95cRV;eTklUb*UD^CNQ1+7Oj4)sep-qdhK
zTUUkWi%M<sr@Bg}q%Gnb6gDl6rl_2bn&j}dm2M25WJAo@2N%XD{Iaa~27V;7m+?@Z
zasWb&i(E+`+bup&d3XOce@?C4zLZsI#UpU8=R4W{Mm!JOvHOQ_8!Dt3v<t(CQu9C+
zsu(ZGL-F#-XDwB+38nesz*dK_lxB*xj@Oh9;dvVw_#ewDpnQa>-SxDB#vqyRO$8tv
zG_CFndaX~m>m15nTFLEHh99L5_;e+ZJME4{jRkVuM4Fd19wB9~AmN*0I}-&H@3hSD
zzjM1oU$HpdziRhLr(`IF+z01Zi%CWw8XGp~$MWh_M3jA#+k*FK6(R2r1fCjySrnRV
zvroxBdyjX!Tw@;YESgENN&4~4xbIwaQSm(wHQWj!GewceW9=PD<A$q(?h#-!zcIzj
zzG~K@2M-uL$=2Zwu4H^Gp7Bq>83ANrD~pZSw2NeTbOgQCEae{pLIdbL>nmo9wAMK;
zo!<Q*k0k2kuHcEQEthzCl5;j`r4fuH%I;x7UnQmXq`|`@<I|~kJbRiV<~13)>~Dy}
zS!PQL>qbO%9xFIre|9b&D$&_opoh?@4J1{FPp2SRhTAVCO9i)$<?M#povvtG^DFgR
zdI63S!C?wE*bzF)#uRh=`bU>Dt|iIAc5(a%MtjzuM_WqLLF2xO>w}ARCTym%jXr8c
zCkG{^K_{<I)3pfy%<J`!&Tg`>1Z?I?3e+w%xZ){v_%T?Q=cimn(<~RTwV%d|MzE(D
za0;a}!^Axq#m2W6qU8h>FX=-$PPynx%>=%wBC2ZG4b&YB)fhI6&Z(30VY*d|)!-W%
z<GB=y$(M%XDUt=<^m_O|vq{TdyNoD~3$8bxntruy{k0qpuBtD<lY5ypiY}UCLz2iu
z5EvOYW>tK!BT*w&snhHsuk=Zb6^J=}+*}1l98aaCl?-W=Un#99WnRqSkeDKeX#vP4
zzn$Sk9mIQ<(pQTFaeSX*f|^dm(z(m3IszxsmI@5s(t2iwK6!Q+^XuK@0}A&l<RfA=
zI|NQhrj{$nWA6Dkeuc>`c+LcsjTX3V+_1bTrAxf@$KM!t6B&F#MD7cN(}jp{<7#~!
zc5a>78Cwe(OmBp}i04$6=703UeDKF3`36AlBF`f*`BEBw!<H&^R4J=pqMT^!HJE9t
zCQD6#%SbQ|JNez1$G*5?z-9n0)o`w<a_LBwcE*`x*>3DY8!lmrFsH2EB2*>!u)|Jc
za=EbkE?zWA5-upkBaa!hf7U0dO`YKsYr<4d9hp(mr}RRd8b;*@7hTZ<@gDo4W(O(d
zn5>D`Gwam+5d<_WO@%liSl0@76HIV1#%`Kzc|VfZWRSn$<h2dg;0bv7V7Pi*54*t*
zFQ*c~5ul|WS8m<%>C09OpBXaXyH6n$KSDG*pW%>aC|G-^Zd*c<=ydrEk0p`00uwP@
zGiWkXzt@f?!BM%YImW4Fb^$V^@3w9=xms#+9eL&G4xdXtOpxp1Hu-Jh@>tv|CF>Qb
zO1=FQ*gp&Yt-~F7vFRE&8y;1Cm0Xl8`ppUAg4?YfW6Dyv#^%?;ye&R5jV<gT6&aa&
zhTP1wcqSSAQpoRfiZPeh+irx(=WieKNQM<Yws$4_3C1cghnv)()C?ox9Ey9o=>ymL
z;iDcQq)nkUPg9e)mL5fS#V7Mxx|>yb5tL>L3wqM~8t%;Hum)GA%8onh@Yiu9Fn+%g
ze+nw51wWDRFYoyEK<JFEVR<0#Dg6?y=s~Z=u*Jl)Lz{2eEh{uNGai(zDRiCkX&j3M
zelG!C;(98ri;T>vD=c_~LyMpGx7b|c9W0mMKSo3HEt$r--~>5)qGm5`O+9KQi7GGf
z8ioT1(C(u*{y~A&euTfYtbJMFshSp?Tf@w{lFBNX89QFcg56|$TK}1Cun75$KSoB}
z`Pa+VT|Ad8yuI*|s)gaYL1I<9D{vNVpx>4AU5A(6iR+Rkiw>xQgch|`Wry*8aNtsV
zHL&wVYNadL=Cx9tycvMLU8j0?jwLF}=zm-A6y%w?vrH?ved79|A`#k1H;MgC%5Tj{
z%!7^IyKi`$K7<o+Z{AlbvR1D>0la0Bca*YZdRco=9F;7~<@)6Gp`utCf<;K5PjNHz
z7qMO!J?(QfnF6j!q)bHLNAkyf25UPduRC>_PFtNBt7wp$drU?Y1$>GRKGv#`tSdPL
zd{c;h07iVE8a4Vk_bGN;KW#IOiMZXYBTRg|KK+pOS$0U&DMW7Gl<f~+-lA5EhOjP@
zS`s_{eoox8NcuwX)8#J*7f+JL*uVPJ)+A_fBGrAH^N+&Mskp>N9sRIrjkok?P|5GZ
zs0}baT*3F$E*x6R?d6ezGaHD;{M;z=yNtC})sbvktPA7r8z#XHiQ}pSX663}!azO0
z{|7la`1qFPDTqjCdbdmTGl)<XQbF$AfmJHstb<soOm85<kq&4l>&hU)6A#ro4`QkF
znLdcDnv%bnst)pe^7+sS-geaLp%9eTOO2;}5WJEp{wVt{-!nCNXOV|2J}?mR<s7(|
z_q?{ch=={aNtt=b(e_-W_tD66X3&G=Obw7r;>8Z0&T>Aa<jXC=<Sbv>7qOHa#68K$
zaY=;YDRLlGg<I1QLgVr&VCWO-<$Wr9FiTgDa?ECNK5z$T$#js!n6(WM;iD(%1Nm;B
z=)L)SPW0X^oyn2Tm6IqDB9+)d=pY<-l<aA_C1g*F2lu2sQofBR)f6(lotSR(X~QeU
z+jk-vGQH8!T&!#-EjseaozpIs5^}-Ci%mR-Sh7IG+jkNxC8u2}zm`vdRtB=_T1C#0
z4K6Z^zUAT-IjN*3r(nGcakx^cF|V48QK3;@)k1H?aS82EZ(R|Q7$+6dWo^_<5Za-+
zn9wK{&hp(R<3Tj86Ad!UtP{Uk-UBM_iB;;Nl+GRZn%>~TEiZDqGVi^pfSS+eK{CBv
zxWMwhUGyTc-`T~FK^C%j?&umWekQWcg?65-^L&5kX_)hA=_anbLn{(FA3RVcp911X
z_6?g*zFBm#78@~{&=SXW(Oxst=!JbJ)7ypPOwMQCCEM$oFS*+w@?znZid70ktVLus
zOYF)G#b*h4KYdQ)DUj7HQ6M)pv~0Fr;>GBTEzvVKcGygBP_^jSWE+crflk)qk*1r3
z+_@X=HlHbL7Txm3OkrcTTnV?3yDSmLw{;IGHxeVuxx1dZ-UegX(|o2+8w^2Kk=r+2
zvNHFl%8zi%8=q^w^;EbKX$F!rq05ad$uf%!WzpY=nGnw^J+8$A_4K%ozk^~)8?&@;
zAU#{MKqS&{qF;5UYmhO&e?X?}*90MCI<F10Wt4M_IWr}Kz)ACL#I(nIiW^emkwBIl
zsq`S(ng_DxHxtO2H7<}VXG?;R-=4o`D!DG(M9_G)!-8BnW0b^l3P<GMA<$b+fqGn#
zhcMe-qkaH#=J+aN&(@qH0CL$WkoYLOLvEjZPd|I)2NV6BbA}6vC&ZHF-I=B531gHF
z<}Jq`Fh=r_Nn@d$c3c5jz`0H^V-BGzvVgO%fH7B=r?^!m)*XZnXL)TTY#P$DHG<g)
zjg#4=?lEH|I)na^W454!NJk|Qj3{{p(66kGwv(9>iO9+_7)dWZQ&|`}mU#u30x~D(
z5OO3Lv(34FFUTsK0z^3Z*u?5i^HGVW1DUf|9>i|rNU&DauklB>Y})`4`}Kq5SkNa8
zEc>@X7U3=kCOd6HPV>oOsL!vdN3Y9o3=o_jkq)4b9M#X5P2>W>!rA&Ykv^T@Od!%_
zfkcF6E1(aQ>;d)TGe01qDO+_(k{fe;2*}){?hhh0OCRLUH~2}e$TEv0VEa;fUF6&N
z7`9~(+M2z6R3yU#luw(!FPZyOupsN#H1Cxzb0BlJJ<*7_WZM%6rtP>k#AddFGG<kT
znumEdYz}!xVe<LaMiOJmHb@Y>67oBT=sR4XTvv+R8<N=bY)F|*iHymfyC;*!c83Pw
zvIks~e19P99Qn)d)NE}9*+epdWNQRtup7!ATw~U+2|do<mkAQdmMjo?|A0u3QII41
zF+i?dZHX(7nJs@0N=?=RXpNU^c=5oH?nod~h5SK8zn+@-=bnh|>`elpEL#4aPD-yJ
zYqr!g+m>tS(@L8npLw=<*@Z&rf-ZYKuh~|^??}YWgBa}~gY&JgLlqf}6I&>rOxqm4
zgMoO<F(nh~n`mitWS09=x%HVR(yLjtPON;QMe}GADN{-H&2#{gt>JvnneBWJl5-J2
zsJa*_guq|>9#Xqy52i&pED7WyHF1~Z`qxvXu%b{%r5{ROb(P*X%ywl<{e(s?x@B!g
z&LC&Zk)z--A`x1X*G5v8oGD3N`JNKXATM!}J4D;4RMg8l`SZk(eVg@W&0Dtaa}n$n
zN>6B6^^>8TwWo(EXX#A0XrcH<m&Qzq#4_ZLSCo4v2R5D$u7d18Wz3r6EkQ2PdmFJ4
zSwA2~G)E>(jY<z$nmw7rvO5!zo%wrELry_xydnsdBLty@qagE@WeGj9vM-13S%<tE
zL2_imMCNezCV@zYT#(uJfT(`iH*yGN4`{nAw9~6tPRTYFF4j7|1M`mdHT#e!T2$R~
zf5;gUgrqw;A8cItOZzZn|H9MVEc<pRlB06JT@XS$<Cl_yEBa_Bx+&jn90H<Kbn*f7
znRe>DWm!V;Lea$t#mUv9P~Q&G$?5N$BfMx1NNPjg1MIz78|>4aBgxRavC-z)>L40q
zuT&hAF%oI<==WlS#E~K#2}C>(9Z4%^N>?^nTWexdB@0;O9NrRn*V98;cooHSbKW7X
zmFa4m_syz=Fy_k9j@0^VVhfmuE^B*&YxDh~Xq~)5LF6t_TtE9YpH5|4J|FT$<u+0=
zA!k%gAi2i@2<lVZ(V2I>A(PmIW{gky*3U(vsOaUeXCEa95fyoDwaHx|)hfq}O;2&-
ze7AA&+X@h8ate)L!@M?xiR>G1+DXRHMa&T4i99I9!bA>kj&^*CEp7S!%$9>YkqIaI
z0Z7hTVvN)p2jL3mFYV;xWrX1LtRoTPoO7Tk9wi-eA0%fRgU~@sR&WRx!<s1C?Sk7k
z(;IoN;-6dC>byG(cRTOFg51uU&_V}cTfruJ$Wn$q@4-R^t4#3=HD<D|Ob)J`&q6sA
z`<+X4J~C~^>b^{?cp?{z0epHeiI-=gAd7sdWx!7KYR>UHAj_F~2&ZFbiv2pj8<*Zh
zhXoP)8`-S#+FhtDpZUR)=bE(|kR10kEp&~L*$`(k+r$3jom;r|vTrjv<T8dHpMu;u
zC!cChW^*6|wjOadkRyL7Rf0u}UPq}8P8rWlB<(sLqCxTsZd+a(#Zfb-pq{w+w?V{O
zw9-8>^AIv%#fv~c<XjzMLPHXLbS0!DpB#v-jeaaPhVBq~cbe1|p4LVIc^QLNA;>Cx
z^vPzG^J!a-m0|^_cv3)e#M-oC+K|&G7eQ9t>He9&hY~X~8@2W_Ph5%J&zzhN#-b~e
z%_{LO<gAK5x(%mzx;7qDSts9kYUWeeFe@7(HLr*Oq?170H~KA#-Eu3tL^qLtnloVe
zom=ej+a%qqCJ=cuZPO-pv8_t|#A&NWI~&QlnGQfC%3vctUiO+=6IsrYF`6?0Cc5S1
zx}YFUV!!C2oHK(MWA8cgLd8Bx_NT}!@+gf)=UqIj8(L@9o~ZLJTEIS>61hkorD*v3
zxQjH~$s);Tx>I*0pWIG`CK-dymv0F@dXFqmsYE6Ito`v@#XGl8WKs%ph#s;NRC(ty
zb<NojB$o_1uMLFDooA!HjQq}HlF4H*$p{^Equ!?vZkB)`@;2KEuexLfn`n?b?LxB_
zKrWu>@zm@VyBJ80Tc5~G)!*5~+PYJxQRr>HWXys5$tOpL+-}Gvu`34&Az6o|mZ`j%
z4th6b>OBZ4$rOJSI-tgrJ5XIRy&dRcSvDTT31vz=>QZC{b(eS6QO}&qLu<af<nuXN
z5i2zoF44HC<s<ZVv`skQ(RPVHINBjR<jEz?obMo(K)wefan`tqajQ0)X$}X~Nb;S-
zTc4}a(1}E92NLOV${x^bR-)X<rp>vr6M2*p5yhHwJz~a44?d7Xd~IY{%4wftm8fJ8
zp(=7Og%VHNP37|e5eo*LYej!MiA<2$$eYUC<-`w{ZzJ4I;{DI*T(ZQVgNsCGoV>Ga
zOcl$|(;-*ZMx~+r5<f}qurY_|6m;K~_)(CY(>0wq`(y=3=P`22lOXHN6RFfMQEn$3
zNzN4SWtPEbBX3q#aOT{;>4dvvSwdE(-1?KA8nOq}B$HF<yLdS01uR<h$?H(oMrjk-
zo%s+u`9<ZiOo<?pD*|%my{1dByfrWQ&J|CA_`6&i2){#<GbLM6v>FicQC`!SdumYK
z-{KrVa<&&^7I|Y_)Edv+<<c8WA6z*(keri0UA*+<<ft{CDdeiR&>N8}q6uC2`1AcC
z(nIbx9dadq;X+jw+UfZFGKF;MO*CE961CZGJB7LC4tNkX(GhtC8uLRgsnF;>3?l21
zJ`y!?54eWn3dlc|R}NXG_B&5HM4oLVY(fyJmqxtWIc2)OY0#7@@x~&Pxj%@!oyjtl
zn0oYkv0sxHDqaM#KPAcyM0{YkN_3)K#HLEe2Jr*lbry@}U3ZDaxpCa(cgaoRihQ@p
z7&WSDmcKy8sBp0-FEne>Hx^A<f%-9Sm{``LZ_5UMB0VIh>!y@WmZvx6fMwq>x(g+e
ze;WB4^)aME2FRGJxPZ*r#}6V^Q$eJI7uh8QAf$T=GUd!}ke>6vK+b#{A0$UNQ72#I
zDM-r}Gms_QxE>^H4j@N9Igm5Qx<81mn)Hsyk!VdcWuEvTIYI@50&w{hh|pIgTbPJb
zBK#}QMq{3MOBgff7Z`|e|0f<Q`{(ffOxeEk_=OHxTcfpOUNx=PBBx-sHCrPd#LkB}
z2U$TMGY_F-Lp!48@bq%XZy6A&jz`VMF<TlzFpcFsptQW~&RlcU5wl6}+y}}2Mvx=h
zqKRa~85-9{n1h@gNlV$oQ7P${^^gb2)=7{<=-}z9oA-@yitIbABb8gDh5{lrrXS?W
zr+@=0>xqoHa~?94j>;ajClK*bg3Pu%_>-c{BHCZNSAoD{^O@4DEb})EGPVa*A{Y6c
z%T{f5L7z1ng>bj?Z5;b#^!7AL$zKX0bU^vHEVn7mkaZ*qQHg~K<jT=>AQDGRAPzz+
zt{N9*w)-=t3l}3GH+w}$z9}_RBGE|Mm%|vlb3`X*&o^a?$NVY)nS@`H4mf2>q{U}`
zQ8L>m8VY^Wvo(S-XRcF10+Q4yd{UFHOgl7V%sM$^2$IQX%1dGs<jnq4QlOTslXJ7h
z?hhiqHV`7L@(S9K{eX<QMAp;tYR~tSwm8`uQ45)MnIlo&GDpG?P|`)eCfp^ze;9*D
zIDaW!C*@4>Kt=l-iCdHCZ{+%B`zT{hx!aE!Nq%RP7>r7tVxa>vQ$-_2$mHnSij_~t
zNUuz4o(w{xq`Z|9_ZlSovzV>t2o;be=VF6g@?HSRuic6{UNmBQrivys5ks4AFlh>+
zMPuC0o=9}{Gxu<*Fp4Ad15%T+oq{n%I3How*~-S4mhI#q60=RTWfSVf-B|O@A`C_7
z4KKQAPcS(99YkZ6S>$+TJ1osq2hmg?L{?2J-mLQ>mqapwWcw%`-j<vf3L+J_K~DP~
z4BMGPYLl34s-a9dmW^auq+`6NmBJPI1x8ydp?Gp4h4<2dRe0~yn>F98Iv+XngeL6g
zlBEpUk~yM~^wXMs;vlqLiF-g!WXXFCk~7hmZOdK+5UEtm!AV6U5Sk$6wT<HcVT{y|
zuZ@x)nd0fdDw4KgInA=MQMMs#sv!56S3MCOA&6OI;G#JSgx*L5N#@yTAe?V-tModO
zF>A{*i|V6x+B}2hdpK129j24tAu>V6bev#$)qTj32~@kKlc$VXMVd7%82L70#K{>h
z{LUf#nzpfW>uKhYW54ibNvt8rnqxshHqqZOFlDbVV~*@$2RU=jH^`OiPF6BGg|6u+
zn=680mm6|iz=Py!j3AgN<C${SVhiXaCUhBtQXvR^r3Bez-}<=cnidEdLg({rybfh;
zB(rADM}rQL!BkVVRnr<T%Q?&kXTHIGf)Mhmi4V+jq0?S7^F#uOY&OgknTru+Cz>iP
zkg~U;_6}Rjt0wu<kWF+K+PR7ilH5yq9b!Vs@yaxwE=D_eN#%U-zQ`$HcgZ{ve~r-2
zjElp@thqWk>W*}1e`Jij+OQ(!nj?&vW#8seGI|+<-#4BqnVd(aUa~UBEM+LcD$<z@
z+?<y~q1R1bdb3r_L>VmEuL&X@ra@-0*%37;_BRU0W*<L3($ZyjP;0VmoGGUud)*}3
zp5Kw`l227Oky(V$=IAEIbm1YSO$ooI^rq-8Go?WCttUEO_%%rGx`M|@;_zvMNLlDe
zqYKqYGJkuzkLL5C`>5z|AVO6O#`VlY7H(s{3uIX4cqDQmHu2ajquK1puoPLpU?9z>
zK&(lQ7$8!26}#or<<w@|#rLt`c#yrO&VuX?^^x*jSgdn8w#l-|?>zO&GKEZs_)C^a
zdgt;xSI%#IkQ_aO<}Y#UAV<!<d|nGVdjRAXpTt6?l+0Zg%O0roj1c)lm(v`z$Cz7e
zkYq@UwRPzdm0TB`FLH8}Ob`m0rBh{H=m?W_q-Cape58Yw>y!6wRS(BHAc17I|3lZA
zB*_iKK=4b`+I7_W|0g-+b7uHv1T}1~AW&5Z?BqoC(eOHV(1G!_U+pY*Y1|laJR@VX
zQ^Un~FjY<?*17Wxj=u4(_nk#m(s=D+aqRK5ql)P#ITi0CkFk%z2Ip(f;T{)^(TEzt
z3yt>zmdGmZ(-NKoHB%i?Nc8O&_Z*h+k%$poF>|P^#QYSBiXOZXncK;voW8Dk(D`nJ
zF&dEtvZvuA?MoomroB{CKTEv8cHG<O8?S=+J!{}P`8yCld<<}`JUWS@taDcR+UXo=
zEO*4B$xMZZe;~5sXBCXd$9E%a*2&+&uvxv5C>tU4qOly|tslh)q|F^S?LpGjy(elc
z#G46A#Cxg!<5>3zKCPpLu>@;&LfJ89?ik-WMif-#qH70kY(#ajIDc-!;+RG(btb|K
z3P0qSuJAgEc?8k|25ev>jzVuvpXKo}fWP2p{6TrZSiQ@<_{n#GnD*UxEP*+l4y5ht
zYoRd!BjpSa@}T;TpF1a8&VMPExHYLO94`sjxv@e}`El(ts5j5gX91?|<Y(eJ)VTxG
zHnK10$g@Ca`>R-5mOJ7*#I@r@Iy35yC2~{L5{i5<bps<Gd`^3;L_EvD{G8Jn=%y3R
zQeUHReZ%WK5wUbn^@Kj>UJVRSSNxdH1(6J&AMi-AMnM?I-&wafDTvcY+#gFswy!_b
zH??nAB8E?$3CGl<Iti@BzlQ7Lx8?<O#rFc}ozeX&EP+G0h>Q7cM$OcSG+yX=`(1JY
zX!pAWOT=z3MA~CS(RB5d2(K=@rwfh1+edVK0?iF_KkyXSG{>!ZO^<VICb+)NM1%BR
zdFP8LTZ{pV$~`-HyWx3X(57OAEYuG4(IF69me*B6Ksf&bIdcC6_k}q9FYs?SB~*L|
z-$>K>Ubyl3e01(%bY2@#ip4eXu%H~Fqk|=0{P#fp#Ax$4Ck2KlQs3UDYK+nbA@pdo
z9=_-95xuf#v$@(If9G6$9iZVK{sER)qj%m+)C&!~C2ZEfXx<Z5-u(A~TaRCgq-%`z
zf<Rw<?TFX=y`aSrZSL?K?&S>+|K_C^W4%^n;U)|lJph{*D{*~dmpp&e0@2@t<#u0;
z%x>hju|y^XOCVVBH6mihzKU1D^CNfXz5WF+6qPJ0a7UXfP)B&9)i;Xl_{=>6f{^}q
zNU`|*z}}7=3zm3mVu_5v16nKQl)@r@!osHwpZuBrg5B|eDfIpGfF+BygZt*I_u+Po
z`9Tag#+u~I;A4%#nV*)=OfW6lED;ZZX&bDVhjYh0E-J8w=kP!;iQj{!MXV6ej<Kp%
zC=~vC$Uv3li8Jw8-^iPxUmsj6DzFBM7fa*{UZ~6dOW#sBBNbcOvG;e~?vK8J_TukA
z?kEf36&=xc*a89Z4p*QIsOK7IB23%3F<$hVeO*(VHBROCQ&99?!wvGifIi^je$`9l
zzbtRW5751g=>Ci5ukSgSw&DA~2!6yEklQUw<riL*9{falQJ9I*SwhW+Y*uiwKTG7M
z;M2xSkJ_xY4xlsX4o0&3&#@6$0-1mXfQYRSEQh~;uv{L+0X`H}n6SW{^1nl&S6x-}
z9~YC}C0I6JiIi7^YU_XJt6vqC$#H{xjcj$aiD)=&9{0s&93hV%d=}S~#S%JZSfWnK
zXYrdB%kIdY&r)N}n>NN^hHd4eGZSVD3lxg7`0YG@A(WU?V2H7L=hJnv_!_Z==s{#~
zY4bY9nOXsxXHu|4P4&-G`xMl8^lb^9Q{dN!qieCeo<jy(3+zb80L$oe`s1Wf4~4Jo
z9?koJ+5T$#(7m3|!ty%S128vY*4}GKa=t=P4R=3ehRy1@Xe<%=K@C4tda+Qt$@_*5
zhM!`<-{Kslo~virXHHSa9%G3*o>)%z)<28K@Ud{$`tO+szk?}S#5RdGbzcAGz%cQ(
z!=o!$$S}m;39JXqzQDR;33L+{sy6uFSvHTd(dKfGc>YngPa`r|sJQEY2Y_OX=(_q#
zkFOmbufh_#D_CxS@pHoe93O&Zao=*L2CTp5<^cs%->1W{yPr4{vU6V<3kQOZ8!S`L
zUeacXH9GLx)z=hl0`mhQr0%%^7;$`!@U@?#z)(DI_7P89hZfK}>fAmC^Bea9!HDo6
z2XIfmv#@yf1&k<L%X_dKaZ(J()kK@M&Uj*9aW4!MFsZLio_D4=;OiK4V(y{0j)gy@
zzZ8o{TPc<{$IOi&B>fJi)~svS^0m6r2_E~a`lA_n92N^8TK_%h5r+kksC#x_DN$eX
z;2&I>hhqXfs`mh)JD*c(zs0@qRbX}RiBgE32b`(3?pPf3SpNJ)`B|LXhXv=-e-A>D
ze#U>)3in~h`HtHeZftk~R0FGCFPFzDP?@`HewN2EX7J&mv-&Sx5eEv(<hfnGMp=(&
z6VVmg<hX%zh*{wJ@v$a)?XemngNx^;u+(kFdoo4`%dK%(kgfdFG5EwlqSvsN{4Rl0
z93Jr}tKp+F{m#?pl-gAuF{9Vz!(#Ca;fLW3ujZ5Da32ZrjJVAZKyU^h#K*&H-eWmE
z$ASg5N}mO`qpwj2>ESs5^>EfX)XO8hL1bv-jSj5S`9Cv!XRoP&bnf8X7c4kd{!2k(
z#tNB7N1yRESa<QYsIN(zr$%%^Aj!W58P0De=uxgUF_UHToZgJ*@5lr!kjT9U>+P}i
znZQ`QX$gN_{(G=IexH)8hzJI<`>x?d4+3QXVBycB;aE^!(O)|Mz#+PYMdjxX%Pl-1
zoY?RjaAM<p!?|?M9c5Oj5Ey+s;|^xQ$4LR<AlAVQ<R?aF4Akq)=8m@+>}k*T(Iyb>
z^H)*U{VGgxGg0m%&JZllW8q_Oyyqwp5}w^x9oBcryd3e)LZr&GNksV&L-aXKp8}5b
z7^Ab^KmQt7?zk7I8ymjM!q)b6fY%!*ENQ^-I)R{j?t@cwJA7SZair4M#`WxGs>8-@
zwhUY*I@Tb}Zl6=M@tg(B-#M#q!+ZP52M@VMbUd%e`!+st&n@Vy`27R&D{c(-h%*IA
zfc?eEw-AN&GZB|OqHG}Z9OsS1K;-nWoN=4MDGqNPPH|*x;1oyf78HJXC<}EWW5okH
zik-Ed5y3z%I$mv*;G2Es$PYg5p?uY0{r*8IEZ!KT+8m)m9`ZhX#S^lNz0LE-Suq@9
z9AmnAL8ZmJWF3yk43Wrp#OJ(@zla}vfsz=v@r_&?8ve0%c(+C@;3Kp8Sp_8})*F<R
z{5zNH3Gs4%<5Z^3YQ(FM&WQImLY{FqQZqGtichH6Pcb4J;U8deY%6uOuDAiQL~d%!
z_vr?n!e?RoJ0tqF{avx%HnJ~%FHnIsGF@9K9?sNR4ZjLYtlo|CS-yJ7#YU81``zIQ
zZ8*2Sw>QN0;_nP3>~#mj({M)oj3r`_R1AuY0O+cS1aIW3-K!zn=$IdL2+zsS>kWQz
ztPqsp{5IPN)%YC@qc?WRMx;D?4M)KD&xR^M^o;`PF+VW;{e;~XfEzw@%X&r+VD!cq
zpbFq`vyEfSBPG;kU1PniQKOf$@(quR3asH*?VJaL{!&=L;aBZ{aA2cN-9J0!Y5hFd
ze^B`RYrj3t5V*dP6~W@TXs|$Mcxn6ZK78iD0!4%gOXQHzhKShHKHQO;g6kVE*FASI
zjNY1sk(voUq7Z3f768}!O5BMf`AXbTDfE2`3ph?oViDEr=?E`ip8`Wm&T5U^`;KyX
z^qTU{zPC{Uoc+As=O<1VxV{m0*%zb<d={1|PBFA^@@r@07f@TT6_2b`%n#BczJrgd
z4h#HlKC4ZzCFdjW`mb)R-!kxNoqfaCPRGz<xx%M7hNu%c9|(rV?-{m;JsnhJzI9Fy
zGR1x`z>SS~`~mto#sKuQ@8IK8P4FJT#*5zr1J>6yO1zO1KiDO{b`B6xeq$W-<q9*O
zV$b4t@PUL@^ct>wjOYRr*~fsOl<)0>XO5qI)T#*|7Rwv+gKl2*V6C{zvF>qa9SC36
z-+6w1Hy-P|_~@Ky;9WX8SRz+&0ygwHJ^w=Y3+&7tS^2L*uFol!`|-Z<D2U%fKFh}d
zB@pLu@T9I0J%4XxFHayU@pq7XcmLp|0G{tG>g~o#gdZFp7um*|Q<z&ZI&g}8#Z!ql
z&P4Wa#9?9ddKEzS{N;RS^m2lo`zh%RzU$AWFj=Un2Bz^`&&zl4^@kL#t}1HHMs$UW
zvd&VNuV3iqVOhcpxBwXYseJwRh?HCdqDtOtG913%IA0@bNF-qJbwDJ*t91e~fYjwZ
z7=v?1n=P_0*BH8>RDdnK@&!!D&noy-fn=t3?h@;Os(F5AT|8xCtf`t8ZRWAo&Rp29
z3m3`nZR!fg3mJ@#tDc}{7it$+LZJx@fn}f5mF!elwzvT=bfsb?qAL|^hpKtigQzx$
z*wYX<#!a;i-&wHQ0=xfNe73Q~D-^}wKpCL@8GaR($kX1%dxv%cL|$hBK1f9KZiEqh
zL~nGZ{HBFN5*c<BM_lt4Et0~iQXMwpiT8)1_1^;@-PhYaLj;R9#3#dYMy>#aOhm&$
zlmsq-8nA)<#S%Q08{HMZW$xvSNIjNX*GolcVf03%9u+{>-KUi57;&VV>hUq66tjy}
zMO@iu?)eJP>HIu@1P;RzRr5&JpW(kfe^2;{59*SBUSkQa2>j%T)jTAM{d{`>bi^1?
zQp0_|XQ)ghyxmx%Sb`OUj!opmu>_Wy7U9)AF!+3*K16kW9bj?Qt@*2I>+c^b*Lr0w
z+C;rJEU0Vw+Cgi`y&9_B1_I*YK^Zew+D+8a*?zIE;T6ORnS&!utO)$3{~kmEuCOd#
zGnKEM!{KdEhKpZ{!gKr{pvTcSQeE}6BYew;JZ*g~bfxBJy~F3DLT+Svu*4nwlBe}`
z4c9lKMKAADADtJXk)JLv3BqbK#~BZ&IHI60xhY@QXmSS90!!p&U;(?j%KQ$+|9N?Z
z`pn^@MI;M;aJ<pq@eI%V1#lVj14J@Llu|9e=e|n3?mfMfw2j|GZKAYzy-<qs@IBpg
zZ=NnNl<W5-it6Jo``L>n;uP==B3AR_N5{AUIgK%xpLnrQB{(>QFuEe&h~<g-$+HUX
zaxew4)H*<<J;n`Z!+lHM`|)0Q*Zp<h<z2717r;k`U-jZv#aOR_s>B=p9lA@fkN~ok
zhy`FxST^Um(guCJXmdJhk2aS_rLa66smJoVAOC4x_}alTdgKR7Ac(LWo=3;>_?h@w
z{MEMnA)FMmFofnJZHSQihz@}u`r3Hpbom=0Jp0$i%l%ORfUCzKv4BnRuT7yiMVrgd
z#LwcHZ!GVqEB>?idtu?K_qZjNz*mt&i0aCp#dCUCLZcAN<2WH+fZ;`BS={sfEY6Pl
zEPiKU*&REHh0K}H93ea3!9Ny{uSB@mj+Fvxzc{uCOXx;?nl!PuN%ndE^aHTQJ_R)H
z?*&TifGF27p!iOV4v#~RN`3e)*DYFp?2;JKstZRO^1j|T6zqNGKJrbpng3#E0q=G8
z+6Vk~FAd8XyAgPmd(o(vg=RLE#a}E~4v%?bx&1BkaV}yE_<zR|&}NJg#d7H@lr|Li
z()q#S@kH80eRM3VtJYx&CER8BQMAr`@QYUtZLYZMvBX^uoyp%aeC_cnb4b7<E#iFx
zv#p-cSD(<)fmrSu9aIj~iv~^(<;{CA{`^f#n?NdI*}_A?;v6;F?Eb3$EOCaAu<>XS
zAoyS0!2?AS{cfawuxB{<+U00eEN@xhk{&*y&~IW+=coe(bboRr8J2jLU^(J88^{3Z
zYnL{+XYLj##hz=!68Rr2DAoHI0Ovl#GS#?I3oCXo9)A3u_3<}4)oGvbgs^ywYZ<v{
z{d^mdG;+^wBwylVz?O?sdHmaA#gCA%{G`C*(c@)Ac*VyW4!dJ@X;Z6;0^lA)f%bt8
zh0ZBXK%BiW)#C;vUE*;h{thuF|2srm{jMj);wT-$z?&m`u<YSI!8(66Dg~)6Vyaj!
zk1%}}&mm($Q_p`X)iyLXNTd^(L9{{4CM;0M!*a%*g(!gM&3F_1#f?5LyW@yy6MFbq
z&?5A{k(Y_lS&w+RQo8W<Q+fJ(oUP9TBkCqCFq0e?fCVXB?KRjL&tlVNAN>^jxY)kl
zrr+b4NWS(sTM}uAsQ8PeA`8@ncJ>XrBEQ#X;VVz>tLiH?9-~7+W<NtdV1l2c6Rxtq
zv(P1d;%m_LV??JR-X+MK#`*RYtoCyUi^t!fwauQV#d3J&4$E1q7u1B`jnlk&ZkKwP
zr^oEL>8VlZt9SjyNzTdZ8Uv^>QM@C*wmDjuHhcALJ>&kNV1gs``PzF;+5?!~9vl3K
zf}YR9vO9Vc%jG!<EO&Sgr0e5#h~(7l<Ms(y`Fca}DQ*m$&bWW(-|ZDtK!S3e`M9B{
z>Sq<q?s$db+|GU;aG8zqOUJs+5O@jC8`+b%0pWZ*yL~|zA$Bm1cicbdK*fCuL#;;V
zWB2)p&WB^eK{?DG4gc^D9*@UzM=TAfl<UY)l6qYHw1lWv>lzH8pUMPPW1kY}n|)Og
zX7k(!$YUx+>bMb*@vP#9R*twbZ1B-<3@ncKr)D?;R^9`EWPX>R4Icgufr>b<zd(h1
zw^(A=^9-$DiiB2-HR4fa*(3f&88ug5gX|27)<*}}li&3?*p6oS>M=Xg0!ye0VyWB*
z4qU`GxELa8h$vdzv?SqT&vD2fu@^oS6(1d9>#-7vyZf2Q!j7A1fy5qds`fM@w1`Z5
z-@r7+iU-peYbT?VMB{6Xa)unygDXzH1!kS^IUrf#Uz661*DieYcvXMWj`4Shy8F1n
zVu^bW=P9BD6eD(I99UGuO>}h7M|Tb(ZD4{$8<5d%p$hF4792&l;KRn(aNyiOpk8&{
zjg(!kIR#7L`v>q$yfq0H#d!_8JtFAv(N90cs6Fl2NkBq-oMP(|jCk5m0mw&`ti12w
z6=4GRVW~YGyOGqp^C*{feO+f^0q;5c{ez-koMKomAGgoqGe>n9XRlF~A@GS<UPo1a
z+IY8*8yVLaYb<fPP<z^$0Q{ZGtZsz!e8p3H$5EA^#!kfJH<T59XTcK>&tan$z3<Zv
z{;99G4P`|?#kQ&PfK-ci<`!wyBi@bFk#_vfvdzEn52)f?D}?ns>;Boe82m2TxB>k(
z+i*H##X}<TJqH`3=4S=*uYKE*yY)2+v)j3(6ujSDVFt_L==9HWc&}k9I+ksLaqa9V
zEL3gq--DWhkJ~=%5qH6}jQj$YICs8^%rS0McJ$L_NB7I$3p<*>@q1YCev0iQ^{adg
zpbUj&3N{{<V#v(+*O0@HUc;>ReY(TtjM;|!5~DMp&UmA5SIsGy+?eh8*<;+6`SKHK
zM=Q)vBrKbseESmhz{oOD&cR2NgTh}QR7I%Yv+|eXYxSZewh}%pmWaO{)ZFrQKn>-X
zx#N%9QTqlyCH#1p-H}5+ep|fj;e5wAN~PrRVX;J%`rsWLy@q`gV}L$Vv>}V_H|;^%
zs@jZ?uU>fKkq<tojTe6hA!*-P2Z8eVJ(OSYm3U0ESLncheBxDoP`Sd-L@bdvgIn&L
z@8y`cBgSpsvkmooeI?FkoNv@njtCQ$z?2;G8=0<y9Il@saM&YCeGsXLUy8_@-wOu?
zBx6pO{q(c<K&~eG2Ejj8A`iz{A^0J)j`jI>#OZSWfM<A{-xlwmlMK`^+ML-YbhhDO
zT`{8PKpxKj&ag&25zd8k5nyB7;XR$1Lk5u*UJdx6c(I%yK7ECt=og3^szt|cJSj})
z>)-?&Se7+#4yf<M8~p^d=lhf@&T*2T^BXtdNyx$P#uL(PteurapwIR?Mbrb<dF(mr
zOUFHTu5V=S;G;*zhAPgHL%tBK_Pu@O9en-maVB1Wrz`g@*WVqvY8df9AJOYahE7NH
zg2&<>7nPqQ3VIQHioZjJ0iW%Q;C}rbTvhKIwY_3Qhp48yT*GtuUid0Z$JYR&4Bi7c
zHhwBoVLI~8RG9W^o$y5okjCFR;@yZ|TmGHH`N901>u^7WdeyFMwNS4*?#2tA-r>Cl
zITAC6YIckpvcs_wVZQrm3FbAhpcjIwF>@>A3hy<iOSK6NR&-YP@JFvTSm)g5i!;%%
zY>}^r!yd>Uszk<laPvm@--F-nsx%9Au>+M1?>)|I>S71Z4)yEUS@6*xUy0nn)q^qW
zmQz(RDkflg0ts^?ZQ|?go{+N96>?KZz*qcD*x%0(P>>PTB@z&x9kqKS+6o&m;@7bC
z<KDhWzWb?+CA=pr`L)8o-MGpzZu9ZF=Wx$=L;zv<$LW6)K|g(dIN|&?1;0M>7&mDI
zKYKrZeXP-yFoE{~8CXBxZiHL?6vOhy{H#};@pn}zp-#&i%&q5d3OxCPgqP18mhi<M
zN(cJAO=a%j%RE&4^>KSBm*czf0f+A+N}cb>pFTjce1)Js5E%||zk!scMZ7*x5(q3c
zdI9nNdBD5G7yzdB6+)YcrBUS}upTH0MBePl%mbEqNnoiSjODEpafmbV0YvNm0hZVc
z4>v}P4s}vug^X|9r_>;ibCeq7;ah_Lj(jkY=-OG+84NKvjPa_*65b$|z_Ou_9`E{x
z`lqqp<|pvv53jb^jlfO(ta`A-83LdpW)6$rrw_=%`a7Xqv5;S{Ss=OV_rgPR*YE8I
zEP?mn1<&U<#`_0;M(+hIk@ZEfAYO-9;)LZj7qgAUwT>2)D#H_cVJ-P?e5udmGlyb9
z%;^iCHb!UQ5$2<l*dijgfueec!z)^VO1t7+{}KuIy?`a~iC7|&LMAlcKkrv&&VMO|
z;bKI`Q1>FWJl8CYfPUidRCeqIHtu@?1%`-9QBg2Zl`jy<_!`QsWeLTpg`E&5(hIuC
zPl^}3f2=CXoB6!}x;I{JFY;BfLP!UzPXSD5@pA`D+)PMAkkGQ3!*aQDId~4l!o1B8
z`n|N-Jx=~vE<NKv%?3vqVBx{*Js3B~@6cxVj3<`MF^%}@;a6dy7?t+`c)XrFD}+%;
z_F$PkCbZ%Z``h`mxQCA=)QPYxeu`l^-KW43oD||0!P5RLo(IGNV9dvDjh7y6#zprJ
zq2tLhKUkvH()#;Rtyz13#dFBC!C8(rgyzDscqWps1!EM;<(U*REb&6d0xHXUFv9N$
zOCa>g*HFAY+QiMoGyKoJCoD+f``0F%q_C`$-X(mE{BpDjb`%J+!97SUAaDI^U>2fp
z#Xse1N9<q@SjYSzsSz1MEU#w~DA@Do@i#0abNu%}%8WS$k`*gt-Tn2>-#I)Uf(0$y
z=s~b^sc9G}11zKKeqnhWgMsCp{Ff3xcPuC{)u`Qpg(Bqod*Ho0qMbHx^=<e$mU>0(
z&pl#*vWqKHVxgFi_YE|i-;Ky%#2t*qqil$EMkP@!p-zP5@%Q!$vpZG@fd%LC&<2$?
z|D_awa8HOm?=K?&#=A!Vu{_RNA`3Xy{H$Z_AKG~4fv+u&+rSc4SFr4kgT!(>UhK2P
z7>qawewzXKjMEZ}*9Kp)IL-kJwSjczMz~}8`hZFl87M5fze2Ix?klg{&FAAi7=kxN
zo84pHSPswSV7Wcwj3wCmtf{?v4P!LsXIf&#LlgH56uG3}y?;6E*!5I~aD+Z>9%qcM
z--FWkDT+Og=%7uY`mr1ymBJDXSiHokW*d~VpA@jPovrdQ$HI4kzwPmOP6KM-`Fca;
zI`+bPcrFw-h^jRDwdwElTmfx@xrOC&3@w)1b-z~F;Nfjx3BUJ~DT!W_81mf+v)ePC
zkah`_`xtPg`#lFDI9`WX@@vRu$KP3EwyD|eIB$CW{Nk66f2XfRDs#tM2F|zZ$WZtB
z&(+t~X+CcEqQU1o6OA^HBXU1Wt+#b^R_2NVQQVC)5_o>Tft-vN3q0i5r|^&?N<i)D
zxU*oFxGoO76J+aj9ek|jvTV+T=4)HLs^{Mx{u>rY_VBgav3XdYxEsL&d1P*-%w4=`
z=TBP1#{hn#@9mjF1~I2#(EVL9kD&!Z6@fRR1bBR&XZlKiI_e0^<Cq^}?i5k>UkX0N
z_YZM;kI?cp_>%FpEj(f{d+}nKzv@x(wIh5><Y7I!4?&FRpnq)%%r<Ri$Jt@oJnoMr
zc*s~D&vbq2@!^vr1Qn|niA|3LQx}yYh5AcjX}i7)Z9o@Ao2aV@@EC}fxA{?BGAyTO
zWBJ+{Yv*HJN8d<v`5TY`pyzTR2XoJO->~@2#B0IPXMAnRHWlN7aqT!rzDC8V=-U<1
zA{MV>viRB^UNq(OJdZ(}Cr<x`3nThAUU7~RfsFeUp%Twcq5Vn80w05E^E@DJsOcAN
zF4y9v%^fj(QXe(jd0KvAAkhPaFk)44kh({VY(u<Hsax(S9dzZX)a7F^pK<n*`EYIp
z)Yzz;{#oM91TxR@<g|&J4Ok-kj|ETOe-G1LmaS%vs7JMVm#`34j+25{PR!hThHptt
z^;)A4Q$5N?>LK1gh<3+&j`)%@cGeaAj;|1s1#vSGTz5PY)aBFBGgyKNjfKpn&MD*R
zXcI<yiU=U(7h(s0k97A`L5szUg)7)~sR<BAhW)?Ay(gZO&c2~0=#iE)q68nL-cJ`E
z8R13a5=YD%cQCwLcy-S4{EDr6-Ivc29wgf*Zkcrng><6vgqU=Nz&elI)K-<#(V`!h
z!N2z75g3*wUPc>$8=vhB$x!!Ds5Bngq;1&a#=!FWY=i2I7bTWBLtwXuF9sjo(Wooc
z$s-ocJ1=ggjoQXB))a@0-!qL7`Ptyh#QYG=^HU6c&3LhVwa(qAfR7$;nQi%n?*bn^
z@(UYUh_OPJ!;#FC7kFI@p6aKedV=NWIBzW1?c+vua>x82>M_I`-6_v>=`Pv-PDg?8
zwel&jxPIcg|DuW+mZ*M>g(QpjjW~tB3U+GA`P!jwQ_LKeh@oJKl?cE7_VZvzea-Ls
zk1OwMbSF$5Un5NI_vwygX#5>Q8`WlY&JfIsGj16y&Y+;pJH=R2dB%4MjL2WCH~6$3
zk)cia2gHu!v?SsX*%!F-zt>k4ILJ9Hm5JP!GrZ27SU{{^sw?{q2!lP&x1H>J^o@sv
zpUUv-9sfq&yRrzN4`S`WpNKQ$<GuTdba1@+J|*`O>+Jvn;j5SGW|0|!!=9GP<-l%_
zY{fxR+<nVqAn2_t<Up0i_X4%f!%I6xXg>ZmSf)PP$0uGJZN|XJz-n`x>WU>|4zSxJ
z8+%MqvzxC4ZhbvKsQ6iRpbivkXW1f`14}!QSJWtp+!U6``chFNUX;g5nJgb`^4or1
z<5k6+p6o0?6VJbjwqkLuqxD2<+;_=IL73`WWgbrI8Tsmkm+EQ?$RV@sb#zXmhCXhv
z+v7|;$w~NWNz6SW!PE?oI1-F_=RDBsBmBe@|G>{t81dl+oWI-IpZuL*a#HCJ5W0`s
zutog(B)}A-11Zx_Bq9Md3nP3<@4;9(mJ=Nezq8=5M;4Cg!QlQ5mdJ&E7C-q=^@waN
zHR{9XqjI`)u~#am$IS#IJ|YYF!*RO6VUH0-JwBrDRA-7?2Bv!Wy<pE`PT`}wE)ICL
zd-3yt3hfaOx$v+3yhevGZs+R<Utd?%Rm=wJ<5m>(;_d8S+Qoh9`}7(@y@S7VhlhUw
z6!QC&8ZzOJQYSfXTKMRZ-=^Ms_(;@l51;RvqMihhy@~{1^B%~>3y7xgdQSLwu~2RO
z=er&@c*Jh0!yX<!(b4J~Ov>sT>`Px&U$8SQ!5l{8Z3;gTZh3fVH*Z?M8{vt^tL=v6
z<9)*tHz1aXN@0l%;a54`PyZWGnBOutvW7mQaLZ#(Z}>ma2ANa8nW&uZY6|NOVmd~2
z+FdVzHm9>c|6A&7SD;bv(m$ZcP$5U|+za<YQ72uAH@8gmZK=#RZQ`cAmost&=-7le
zh$UiNH!!Z4Q-bTUdWk;8FI}J8*BcehBc2GS*i{aY(~BCTSgwfbQiI&Nq_n}N^MfVu
zF3;}`kL#;6?q?$W;D~HL!x?Cxho$3Vjpg$D^cjC=^Z-Dg-<mLb!;gP($$gDJs6)6P
zPu=u*<$PRUKar?)?rc_SJwYDx`5_G$zZ9BLthYI;V64xK@8EyutJ4^kA#Udf)kQz~
zh*3W=3m}$bbUqxS_W;mqeeI6OHnqXSg9IfMZ<&|q@vJ>~DK_i92HO--Ht_7RPv5`V
zbMR>+dkLR5?o;@*)wf^y{x2kj!G9@lKwnj`b7Pml&W*eb6&oTG3GXrdc$l`~t;4j9
zEF9Iz!>@u*8{Rr~%LC2*j=Qr=*OzJ=F*;Rgd}WtL3=&JAQD6LC-wQ9Hb>9oHW+Nhl
zCGs*aMYjFqg8>`av3EJbQ+=rl>Sq<IY!UZ=Ntef-`*7pFMo~S{dtv(nD64I-fa|xf
zVL{eBEK!@4Hv6mdgXMJJ1zF$0W8PRGNqY|fSx2uwf>yNIe$^(_bLm06C^sSw9v{Ur
z)GviaAM-<*NIggS+U{9WWZi=ki3KW+j{ya*VnoNy{R6&69xlGNdsdM)kH2;|IK_?u
z#4`AKfaMxu+#qu5F4_J-QN>E6BIfPChpcam&VrK^ZNOfIg{lo@al{QlzNjbuS={Ty
z5{yJF_oQCh1`oN$03D`kb2-O`uifsgZ<N%Cy$!X(b8WQw`CA6d5W9rR+%emDZ;saD
zYmetXHtKKtdmGE>J_VNGq+q!{!-3`THxod%*>zX2?0#oqx&4KV1>ZqeA{Iy2l6u)4
z3yLK$3y4Afc_a%<=pOT!<NWu&0TT5+$1e5v9A67A0+z!g!B|3<n$j-elVdp??TF=e
z1}rk)q4<GicI-YD@XtOvYeXYx^Sag%A$KST`ZX$ZJ9~{byZhHzj@b2B9Q8_@s7iq4
zar_%)!X|$=VqsGK_W<AV6;GY-SO-{QXAN}0Z;Sy+KfiwlD7`*&p9E{Q**%{{F_g$N
zePvQSE3$!l@~k`-fLuOqBk1F>Y+e(9uYpnVHgjN$wqcIC@>l}<g#|oa{GDy}J9wC>
zoe^zz$4YSlqgmi>sK(`<5G85C>x7H#c>k|h&EY;D7S}SMj7`)$!7|n?tQ1A_-vbLu
zqaJ)MqQ_Xst?JjPi|y~=uaJ(vjIcyS=KmI-ZL$k7KUf^m&ezDpMX#e8ByAp#lwb+1
zFbV?!nfAX!rr380{B4hvZ1A_8*M=qDfTWVzviX~qHk)U<K8t5ku$&$Xz(QQt#{ed$
z?;lv&&fKD)(ixr*mN=29zwH>a4RNn{u^{eM^Fwyb{rHjhLaZu^3dSyhPV0Ma#P=^t
zy|WO=sNVw&#m{(NG9Hfy5)9_kXKvl&RY15m-X#-qOFnKeyPaQ8a3^pmSZ<G;0zE!F
z6Zu&@!;a++|7}7%%Gc<G6o>C^iidc03ZM++)O1AQp+z);q_5{2Dd6?QKAlElO#XL@
zHwBnt_RJlY#aYc*4(B3ZdE%r%&hGZ}fH0rO>^P|4KKk!L*w;OjX+3MUtHLyY=XCrv
zmOFgQi6D-z5cC@zxk!P;;1ysAeF`jl_?FN<JU2y~+cPN?WeZ)<&vN)l0p~m3S)aw{
z2a9JO_&dQ+2B1%zKxZ3E?0P``o)zJ1Q7IV<nxX!8Kur6|_X#Uj8!8s~*FZ%1Nik28
zN9wnE+TsO<C88d1zQePl#<gqu@poRw4gR;dheAp`Vx1fzUQ2t!5;q`R?1+zk4}SOg
zuy}S1xx-L-!xB0RSPs_)+LliYQRft1wKH6312iDoT&`+MoBQ{88+3krFD$<^X7022
zZX|k8UmK9m^&Swv_SFlox<&_lV)S6xVrMPG>3Ek7&_u-j7bJguHv&3xjgD<WR4L{Z
z<-YhDjDy2R2T!%yY?u2x@HkiBSv)A?M8XnxFtP;BI^U>^8eZDR+x4BbP=X?UDO!6m
zr)w~SAV&d~@^ORf6uV@NDvf;Yir<6uX2fohGxaPsgsP~)hoz!<>)#vM7r3zz#rdj5
z`#SjKC4Fxr3p)C$0x=gi#+q?upah;J-2kwL4-4eSakmsxU7Qz+CG^QRBy-~aAs7*N
zJ>uZu)gYb`D-q6W_@jhWBL}up!qV3@0tn$-La>h*K9!#xA-PegG9u{EW}SOU8)TGq
z|9q)k_j@U5;mQbn?Ty#rmxl8*p2S(?rnbM$73{Z<gX^pJO91)VOBx|gJ}mKi->7a`
zU%MiY0iV_}SsNDDv%_*YPXbG*YEUqUjHAvR&+3TBZv@QzJm9Gp>usC%c)4!VS!2M?
z@)L<F&WG=XjUt+Ul4HrfJ&p?7w&{&1==aQtF(B9)YZPIg=r#2yqYdIWu|}6rj@`B;
z)MBtK;a9=vjrak)UB|L*9G9^ds4yKq(vE1h-={EbT@`I3>f(6IZU6l_3K7c?u?<iG
z*@m*=x)<Q>I`5so!_gOQ3e~dz4p-?VRS_DWSh()}?+klnNq6c|`rd}O>sacIx|@!e
z#xg}zm#RFI`(p5114{s#HF6UB^u%dN%-wO`8@R<9gE`n<w0Q#6vI8KBIbC2l`Ycep
z)^{+9%;81DS*>*q2Hd{}oAvcI3THL4m(*$wWWoVn*-uN@tP$_U;{2@bK#?VS&2IFQ
z@Axs_I-*#@hdqAO8;myZ;WzCu{6)<BKqTIG7MAeE;Ks(=>==%SCmzEYdCdc~qrYjf
z#2P(t<4+%hF%&kpgP2gP9R!hM48UZ?oQ{D)<nK@|Gy3*87L+!xt3yDUB~0yKo95aj
zUy!)kM6BkZHlFW#Dp@<ab~|vdV|0+S^RxHhtcuY=6u?iH1DHe1998OL#V@yGXgB!5
z;rk!Fonx%mGxE*{Tr}_7`FD&y3+G?uf6m`8&Uh>lQzf$-e#pm~J#-zM087&nXUGWx
z!cR-;JG-6(X~;F=JH)~xf`KKn^6=dv`bDi~=Xh>su4^pL#~=U+BG$(ZIaOcR=X|+B
zI$wKgPATc+tM{aush^1_jCeo&PkdNE6Hls<#B5{n-T1L)V+>Y`OZw=LiL4&%&P70o
z_lWoDxnAKTo$C{EiVJ+A_wD)%J)sRtcn;K&j%d*ZSKe<7IJj}XQTaL6C?`ndcB$`t
z`WbRjLfOw9EY2a^V9f>=hnz^foi9A%_&cOaeC@!rjkhxuMr-CqW>r@eF@ktI;|Ilg
zfF=B@i(+<u#-pZMV?9;G3>8h+CD<;aUw&G`w2jEzMb0(G0H54%S~v=>>#<$S8z|n3
ziXX8{2<Kfs3zSlbaU)*kd*KE_<Tv__G@bvR8*w~8VR<nH&Vh=|v3jvYmK4h!J^1Z`
z=zvcfCoKEQnZp}Q+sHfL@ZO^b<6pc1@39cmRB?_yM-}H99n>LWeh@J6)l1#w`lZxX
zIdtvZq&)qeyHO(ad+sJJQkKxw-Oxpd?DCD;lCNu6vk@`7iT75UH9R{;AR@u=;Ul_F
zW#P$BSSs;GtcDqNJ=N{uMd@>jCGG_{xZ!cZvkl)LK5aw|(fyB|^;!HjgHIc~9@cEc
z4^RdQW&?cMh*MA@*ENh0ehy_TELXhiKZ~#UXUy4#9OdU5^=Ts`{Sf)_l}LTsxSgLF
zP`mC%EOEx8s1)zh2Ov!Jnq$=Og$GQ%?_eyE$%knhFYyOHZG4UL!O;NRKqJ35utn~V
z!kP_#6s$@36fZfJSw|GsY@jNMB1Lo`f6E!88|qat3-9lWEDvnfc!9mYCuZRVh#$WP
z{a!z>;kia;2(F5&gluqD<5Y&T8ZZ8rn5wV%cf6b@zG10fIw2hHGe^if#-KERH>$No
ztP@Kh+Fx)1?s2`y(fdAqmpNV^FWP5*j=rRQeJ@b8HBRMsjSM03qoF^FWpuv^OH_>@
zDYCel0+v9Ue~FY)SK?=JelV7JwP86tkAda%S16X?a3H~oRKBe$EVvF~!5a(9>!>4$
zJTa$O0_FBu-1Ek=d*l>LR3*UjI7VhCA?>eSEW2l4K8r`PupGX6u|zKPOQ^XQfF<f<
zV7Wa42#`9edSHpFtyrS^D3+*Hi6yj~uw>t;sjjyf#lTRO@>K;)HP#z9puRpxDh4VH
z%kEwPmdmpUNH(mO_kc)^Ywgj7oQ?h-Ebc3RZCu+P%XQS(ZubI6pTcML9)Mu=z5V0X
zRBw>4ov}u-c+BBzbLyRi<?^fusa+f^?=?Jey)kG*z9QOOXUrTL;QGcOQ5<MGEUzQ8
zflE``(R%=ZG<r>rKP~>A+qr%`PmAUDC=Qm_H8Dwi6W;e;1M$+ko;HX@Mw`uV=N<N^
zv;MH4e5_x?621O7p1yj)nLBP{r(~V`qo2h!a<Ndf+xzz0oo_}P-g@2!)T>6Ab}+9`
zS&*@euaVFyi$^{9J5hBK%i+2!Sn#?0?~I#gQg$$+{>s4;ih5XH&n_ce{ulScFj1Sq
z`!+>Y>Yaj0ar(pl{Q0^jh3xqszIN5RrnGUagZUZMyX9-42K-q(i;d-V#LZ{%*!l-h
zapV;iL~eXU0qe#}TmtdQ*8<y$<@9Jcme*g^JMO5zKCm1fN7~_dI(r(+J^S9qvYa{w
zSR9Y^ea$h>KygNYn_+>a=6{C^$nSdin$B?9iKu!M6wB!7Kr9qu_Z|!+^}>RnLRgO2
z3nOQUpKl}h_?QKFp}x0Al9$oz@p8?pJ$a#6-gy6v(-6K37U!e@ucG{bj}F<F7&lM`
zG1e%<`5BL(Yn<_*ydAAYD4cMs_YJ^W^qNv`F+U51zT<0~XMO1b4})lPI?sqU!6M;d
z1{}!02B17f6sEeLy?|%pUVswm5odZ3RadaQ=cRq)sp*l%9RU;1*JGLDjEBt{zm%H8
z;ZtC74B$Tf_IUqHNIUutCZ6d!+XV1`aayu{!bh4oNPX8&l%@8y15Va+ITW}K-AOEK
z>}@=9=e5y>2&vC0nlQe9CUhgpa>XlW;$82jKRnlXGfi_aDE0|Jq3>X7aXETsN3$W`
zSsZV!yF!~lv||B*;N!MFjz3@-POg!QC9*s_)qBH7nvqNM-H7G#RR#NV^79SdSVwE^
zpQYN|j+&rNofPwkbh(b(3}(T13Hs=9_Cn8!H9DiV<o*sYA<w<<NT`Gt0AtkI;5!J-
zxS8gv!$F%d&Nu4q##?h<2FLFJJ+3z(evs!1Xj5+{i0aihD)GkrfP3=2O^seh%F=@;
zybWaL;%0)G8quPe*Za>`6&OH2DdzQwmmb6h#|7+IJm-q#j1v}Bg4iV#_>9=o=b(76
zk<E*9l!~&RtsvkK*w%fSriheKp2^YNw1G>bV@(a%@CFIc*62X2_L+kq=KSCtMWr}<
zVP%H*^zk*L*ZdhjM<H@LD~6OJ)ggQg5K?phji`Tk(O4q#Ly$U7K61;>`Gx>Bd!`G^
z>R2``QK93rczzqp6|Y_T>=<X-yfND=LV$i=<Ar!O7MRizX9%KB&Xa)hFuJx0mOxQ`
zmdJgqJa_%3UB~h*ami!ttm6~@0kMdPTatQ@I}1y!D##uAwcWGW5JITHpeud_H6CLy
zUyjJ#;e|R!4a+q7TV^c{FWS^=m+ezMU@7{Y1*;(5GGDRTY7<P2of53!$784wrDCX`
z{#<j8RO0WH*U74mSSK&q+s{PsDUJ%;H+)Nf@$)8#orMZhcwAIdjn_NEU(Slz$vVgR
zh8R^uey~K84K0OuFTf>@xAO+QF~*Hpjla=1>Nom5hon*Dw~+vf2oqvD@oxP1r+)vy
zGIiF)4rA0&M_2%Q>#pDCJ@&#zF+0B-xA}@VEb<+(>;H=kcn_eR$1JQ{yo@&LQ+VG<
zV@LdY|Jx!8x`UMUu|^{?d_L;M#V#Rv7(NoIu88+i88v(gQcw}u{!&AJ(|&0YKi^1X
z#4Crh-!XIhUR9;AoRJ5Ff8;o>{R=t!nfQf*{e4Ot%e6E3+8x=keS9K@vQL#s!4gqJ
zSf=$$KglUw*WZ)>3d?lI%yIDfsSKhde54&Y)oAlM^4<d!`Tcy`mn-u1)Ow0G2Sjy`
zp`aEJ>-r$Y?mPJSRm6<CK=BIwEbeXK4@a)x_=PIk4$3Z269@jNk2RH1<Gh9!8khw*
zp4DqY0lpUwJV-yQK-tAji#|r6T8`<AGvt`Ac)e4xG|pc5$&sV{7>O}%<S2cY9ONi{
zjbe#g=9u3%R^q~8rF-GvDIRNQp=4aNnFD7BkJ43K_Jg`sF{jJvh)>!)fh2=rSz~}O
zj_-PEvpTPOKX8^kns=-mvOWd?En)|+XG9}-gT@Pt3L*J>f}6CTe^G%53kzF&4aYNH
zlvtcsuv1Mn5JX>PR38JXoJNH91kT_q5zFiA4PI!xWzGRvbN{77Fh=)tM3%<ej4*1$
zQupbJ>b>u6EOD1$3FOF0O&Y%&;gUM5Y(M8c?$dL|=D&wG?d=vKw7%luP6uZ0Bs3Re
zu<$&LzY{Ey9WH4^%!o~fM~uF6<f>77ikBX|P*(+@2WNOSFbKWsF>RtY_)b07c(q-B
zMDcZWu*95R&}n^s$mGN?z2FG=nRt=XbAR;0!}r}tb|Fsk3&gaqgX_mJ;=l9<)3q#5
z^crPsU%eMH=)Sk%#*Y3f_$<ClumqX`25g+y@LVHHO5IpzukA2kBU?e;*oeDOH#TtV
z%yq=p$$W%Ie4*Oyr}BlgW2_KVyTdXgy6AICr^`})?=?T+!(MY#wAj(4iEJO1$i2gr
zbjB!Oqll&do~7^<*FrVZXcOvdJFMC2!8$vyjW&Daf37w7GW-1(^%t-pe5}1j&o%b;
zU5XMrysd~9-BlH5zXwEey$9&X_^yXF>pFmZEp#Wb;9mOQ8OJ~>q4pGKB1-*n)81%J
z_?d{sS=tn!z#EIcopHLLXB4Pc__Q?()X(xA44*b|b~hp7=-V9CYW5oqecwO#9JM8P
z*tv1n-%wKg451EgWdCp25Pr+tU@-hXr4DYKmP7&q(+J9Y@cRekO1xaD&J=Gz@VKrf
zx5MZi{k(quw!n)$e{a0&d1-pZ1n9g-e&|Xh{uehOyxsZmHk3p0y+GzYeE$b#L;TX=
z3UvCz@#eG5ja9#Ngoow5o+?}Zpt<O$<uipQ^9~2sx$>0w3B^M!6wCC!p$_4v7#v(@
z4)e9ps@xxBlw%A?R{KfuEK9sKsYwui;zPwN-z5);&9cl9*?x$i`{_cqKK3@%Zfo^Y
zf7E+Gwc9v%VD!do>Sdq$%)Ml>eXL(37yS(g*Eb?r?_a$r;rd2I=KZawubua|#r~mk
zZH)nOzUVc)UB5Bj;f%e2C6FVa6XPy<QDgLT2Yzs@11y0seu)zJx$}-=#BN{8$oTp8
zeh^mO!PL&pHe^4d%@MEAcRa&)fx8^Yg!dm^XOb-`3<JR{{bFXxc>g1H(Cdh8<uka9
z2U$|4ASA|2oPLwrVtC);u8Cs85?^~deleoc;p%JqJ6$zLk>iS|&GQi}7**)#8$J}@
zL*Ds~Tj<MuA=)f;%R%Mmh!L#{pP2W!tb87ze^5PmBH!g*StkB8Dkj7VCT$`{6k+N_
z!tVXp;C`08>+VW~`%B7CtnKyouW-OEcMmN{)N2EY)d+HV=Hh`H)+=;;$fI)<1k#Xw
z1rR1rAm|PUa=Xokm?@>Xz0JVIB#=01d0NzWKJcJ;oEC(LmcM5_{H{rx(@#y1+cTAp
zziIItT9CuzOvfK3M`A?D^VT_v2rN7w|3Ab&;%S&jy6^C`(^(`R;&BX+%l#oBx96pS
zgx~T*Ji@^d!&Th>DV4@`)zbzcjA+BQ9LVNOPM(GzG}_#Y?xVB`ZxQ0T2$ks5(4UL<
zfW)l&;TrTFd?LL*ci1Rmh7gwVdlshC={p|?3S<6%&{XoVM&i?T*?3yywi2p7V&@}>
z?U9rNAvL#mKvutnf*hVx0&>>>Aq!D$q~k|9u0SX_RmkF=U<{Yt&p8m{vHlfs;p^>?
zHji^SKwi&tU=yX%kiTcyJok$>xBCW37#RI_2ohX`5Al3Uknrw<oX)iZ3DvC+alZ@5
z9XsEEB#V#E=l1s<X<%*p3WmGp{%evM5OdS7Fkm0}?*COV_8IcA+oKKqQ(wXG-aU`_
z7!VBb{S7Tf_lDEv@f+nakP#KVM%lNJ&EuLp4J|?+w-I5SK*G;Q;r7*UdLYEQ;%SfP
zMjQhiff(!QaGxq|h;aLR$WM&XnQqSu;Ax>&2J(8ADAMWBV!Q`)=**FD6Ok?u5+VI*
zaBS3rL7OdhauR!7>6<op>^on|kMEWP-qz>=jDy*GO=%#vuLtTk`tA=B*7^X)#_gR?
zu-ebyuk4Gz=c{_`GlZy_XQLiCBitKtP+2%EA(HxI=Y!bo_h6p3#u#Ar#yTZ<8EXzI
zoaf9@<bk@n`sGFzgMYb^>TkYZQ`<P^C`FGwT5%wG<$Pz5xNn1mej>>0H*qACqaqo|
z6l-Cew%FfBd?+z@P!#taiK=7uwE1-3?E!~9_8kz9uv7ftZ-})3c`UvhCH;J5QRm1b
zp1dcNmqDQB_^evUObeuJOY-78yOu(;xbgjI)U1o|L++RFRU^I)9~~IvvAbX$cy=ub
zl8>tb5Cr5Vti2|57#1;=l3lZhr$ufa<s_ow)rYvZ6(pj49}>1QynBxb(1v2b{!gho
z>8B=SBKLmrv^)GeaOK16F^xax+#%IG*4G}VxP8Egcm5nti;BO9n1xao$QrlziFj7b
zkm6(Tv^(28u~&W4>G7U9*zKhEUhdKSkhm4V)_31JZ6ZJMpenU{wn4V=;!KclVg<t~
zc6;<QI(-j;`|sJcq#T7l6386qHpK4OVW}|fY;el4Y@_$>LwtXmDLvr-bf&C`@4Ybm
zJ-_OpR<lRGK|<y2L*gz2^WL>H4*2!qJti~natmd~8{uab%=_?L&d3SH7<?O#?o|YR
zC%1&=iK}?Df$tT_6}M2R%bss}%(%&YAEm~e=O@yJln(E~9E!WNq5PV+LH#D?+p>GS
z{=n|`Xe>ymAbm)!AI@zbQH13^SNB*Tv--G!#8<$3?RmF64ZB81hx!j;$1kYbMw@~3
z41d~ih1Y``%5j^*86JK<c>LiHAz3)q9MY)Hn>wfh6?@2nf#4@P41bRh^0Yl>$bu4-
z?_wC8u3&$Fc<u@a@{&H*b9svv=z#Hk*y4RhLdH4vkOhLhuOEciotHz2s?hZNidT6g
zAEd@$#T)Kxn`P>`9z5;!DE2{-1<#QDkeDGz>BL=T{Y9P~^f2ga`sF~ve?VZh`Zhdq
z3x#<fmMRX~(eE-Ko<)W9+UEH?AOsfu{}54&??Y5RkQt#<e_E+A9_xI1X56tZf8pz-
zB8|tYX+v>V?*Ud?^m>6m9c^yUB!d_<)Qn$A3ihY1n^$w-X~FIQc|5}y<c<6G_V*Vt
z<P8~eAA=1sZSOVpoiDe^Vcxqg+p(eG=(`KrpK%K%U)n3&92<fK(d*wGv1)kyF%LHK
zk$jF~#fL@A%n92C?tjG5Vfe@HLgn`Orm*$H@4t;(_)uWpdp6KPj&;Q6VBUw1YvY>Z
z77BHw!<)BJVZiOt4RNhPqOQmRCqHiD8@g=uv@I<BZ4Tvf-a}0?U8f*ni&4cn_8k;d
zJa^^TmNV|OM7LcXf;MkhLLir)KOkXwqcPx_iwDXXo9~t&VbxGwGtO-a7DViW=(cB?
z9|*{={uMx?2m9acdK(8~@DYhb>N`Bn)P#<?!#yHAOz0BCxd5Lx>|SI>W3R#s>$$oI
zY<;h<2~xWy!jqmqdcflk%K;>8KBNdE28Om*ST!*GV;@C=E@G63c*R+dAXoVCK_ZGx
z9cj-aKK7+@N_Mj8d@bzD5gv5B%8}==lL-^^dL3O=gOK7J{vD9W0zn)oB5ELU2gI`x
zb|N)f!(&EDRAd7kRL%2oL*-!fbDK)mVFi$$5%=wb+U;(kz-Dr`{sFVx8Ok5xnasyA
z)aZc3i4NkrQamkGkPbq3aXTak7V`}x;?9H~!bXCL9hMz6f+PL|67dCYW-)hAMhO2k
zh$}QffO>20Oz-12<O9X`L<YC)s1_Z5K9GpH;`;Gi>f<2raH{2S%y;+(i86&H^eGm^
z`hoT5dkBgJvFf=Ngs1UX&usI!3KDYsxQq0kq9hPj=lM5>H3)}3tQxZYBLa2)k>%_?
zz_|6Veg1+Qdw@hX&{xTNiT6;`&k+)Z{QHo*8Bt%VU&oyZEd%G=9cUSZ@98A+T|KxW
z-gu5f#3)aK)$ueAGxs^11g+vLP;a-8)3q6nlPn!yStn|LzDlUSeZ?G|j<`*o(-l6U
zPfa1lV0yw&fnH9H=p3x2<D4r)1?{Lg=G(kSz6)v&&avS=I9Po4vRB1F#cLRAZYdA?
zxtx(<jB0~xbR5+A4jbuQIUK#$>sl*#y~BpRNI7-Oft)aBksKF?v(Mh^cSaT>ssM3r
zU&NjLn_j~e_XW8BaRS2qk4WT2VnF;;LekL#0=2$=sCF19;8ps3l=#8-qhIct_Q;i@
zR$4?JuDK!{7d+(nJ~Jt~-h=rJ?<5R==j|RB0R~?`AdByb<a5NiO_lA~VNnu@`wvLi
zJ0KB#q0ac@miI-ze~cS<nP{_)5y{7{i@o<+A)xwKz~LL;2NSm1U`NN(&bR}T6;Shf
zee0X9Z}`dYKU|I4gB#+9&x4x;dH+A?9|VU3Bz!_QGE6=b(M*Wl^2Vd-e%O1!HRu0_
zy~@u8>WRmxNe*vh+k?cua}Q5MNvSj*9*lbo^?`%hmd=F(dBgukd?S2mHzgo^j-r_m
z@gJ&H*9;+y9rIwG!}oM^-Hz{rlOX0Bs>8nb-jsduRZnCp?sGQ+%keaBAHEt<EO1pP
zN;uzP#}feyPd&b`@WVcTW90m>yM^8Ru)X>IhMv*SPr!$pgnzjY@zHsHYuqHrdWw72
zGYoMDe1<8ais&Ci<nEz#l&|`Swcu93!!hR;*F(;!&z)yF>zkr-;IsF^@{X~l$Z5QX
zgqQd~nbD4xL7a2wNXHKQz}t;6Ktn(JhI(PlkR_@k9aK<`-G%&z@Oe{7E_P-3ykYY_
z5S?PIkqV1muTjzOK(Qd!&-+&lr1$R#-z}Npu8VU}u{+paAi-vR$&tDBG{cH_y&B#V
zwedhuNb;vaSMl{jJwI3CIH*n@8M*I(szto#i8}_m1i=KLqIvjGs3_>lqzCE)!G-&(
zp!=Fb6(Br6FOn89Zqr!3CPgydQyQ#X>hCxK-{}cv-J98GFPRB`o`Qt!Li8tYu2hCt
zVus+tKmBWyFcz~H+F{HPl8df*T2whbUUJ?2+y;poKY7ey1-#2!bB9Es7*VW^LeNUF
z|8r&~c#u#h0@*!h0R-P^yyx-EY>wHV&yWvsix`ATxc(jrJSXDOB%T(#*oXKn2P7&p
zfSex3IKMJ?I(t9Fvv@&}H}-$}l?4rivPFSxi&_r6=UjBB_z>S+&cBhw4Sx@n>2xj7
z2HiSuL)vXTjnpqaX+LN9y!g{9Q|W}^@3tjKWG0-D&ixkpA?_Ol+1~gHAn_GeQsn*r
zaK+Fa>HPgsxI)`-p!*$=HjCe;KvqAQK&V_1?;#H)kjpcsd0Oaok{UR9t~&@3dw&l#
z;(hMGuXhga`5OlJ$biJ{0|W}W_h7*A3<L$zKpt00;XR=~4N^NfMq11r82;{mBVcpJ
zevR0y$0*MaiLXuOys!G-G<((&PkY>-i2OBNG#@wY1K$t0Wz`(TKJeA}gC!T=l*H(2
zLn#pN8zQiQz#@qmvXXn{Pa~?RHwi*Kkw5w&ey=|V@)P}^!mp2U1DU+nkW4)iiRcRT
z9Q6Odd-psCoClHX4g%HIp9bXLXP`h-v>9=G_|qWn`j;EV)8o!3ym!B!f<(ou4~bRs
zRdx6MZJ01Oyw_abk3crhC_G`eyM<3+m9UjRjq-`n1FHPE#Wmn%t2RhW#e2d7_yv7E
z%0tli&vT_fBC`~Pkf--x2|ZQX%yHHu%@BJ$Mw;t~5ReYF3y`QM4szAXTJNv|;J3s%
z_t8syEr7W47;>&#>_{N)8>CI>uAC#%JU(|oJoZ7G>GV}HB1YnCfnZ;p&Lj)ROdK%}
zeD;FWh>nEebXJWci!P9*#$epT+8{(3`!&3GXG@+S9@hk6#(CeS-6I^dL8;N(5S90R
zbR=lyYYvxcSdg&j-6wQ%#c;ne$P?Be=@emyU~70T4k;M3b0R@Vcl0q>7WXvLCUW;d
z!tZj9m0JYz#FwLT`sCKr^hYRCdyOv7?C0snNB1u`L5=k>fT`{r+>`0%?7|Ou{9GWw
z?z-V8&aYVY(@4&M|I_h`GYipSkJyk1L#DSs4L$GZHCcPr<_&A`E2!k_hg#=hlY_*L
z1mfN=<elFs`i6@C=>L>byuMT5LGk$MNs6<_OAxZgLh$#XyB{-=>{{PpA?JGhhxcq@
z5mTYveYdm;78b}As~+hy_eIkN54QeO*x+6x<otNzewxjs;TMAXbG#>X=6KH&Uk(~_
z+$ceu$8v()?!iE#*R^zLLm3(GHHyeSL*`$o29f#|RoOs7V-4ht8yJaXo~d+FY{O%R
zAW#MU|BT^RrcLBHgOqIvyFG4GAYqFk0THVaI&FOg23tMoCq&!W<3Zx~ju9}&=)hF3
z_w3<wARWoGf#}=e`XL~9oX#W%dNdX{W7LxXdE!<;igWm3VawFKCh5rcHx%ccL2(j~
z_Du8h6CU$@6b07U$rti!q75!4-~E?gXeOK>ac(b2IJ$)bzrK;M<3HxT-wGC-1o3}x
zVHAR|L{~5d<m)>*nYx}yMgcE~@A0(A0i@08QBn}%aQ-ypy?p&FWElIJTgD}#A`8hG
zewGk6i#-ISMrZnkg@TemctI)V9s3)}fzG)*k*}(;p6<9!!Cvr;HHtHlT;yX-^s<mW
zJete=#O?wz$9$V3L;obeQR7BY^_XvPpkf}ZOW6LOVu7E*FgQFz@}ypmXW)U5NUS5e
z4w3y#BJ(dShZRSkTMiVo@W?k$gJcsu*e17okwT1bx>9Aq_XBi_y+R@B*QD2a52!5c
zUv8a9O!GDqoSi_f@RY2`EchM*J8O-v&FSnX6O$?;kzbMRY7=~>6Z^jA&Vtl~|5G-R
z7@Y-oc(j3y9S9~uAa{84R>U)Ww?xZeioOw5jQ>;FwG><oF3$<~-x*>cp{{bG5fVEe
z^*tkF0PeqMXPnzmH?R!_N8c?sELHzL@aspPiF~)PU8rpDb^A_`h=r1g6S)K+5&egg
zA2DPiIdPw(=Dp|4oz%SdY*Y|(3UwxK9O*va;FgCS4-#tu)_M3{HtIaMmtg;+5TW*s
z*h|=A)US6&;z|Ab$-g$0A;Jpa?jHV~9dTn{+dCTVe$rC)B_bTuybtRM#4~?SnCfvl
z?*uY@E$joLQ$C`@<DBo4yP@YNf(-TLaNfq3Lxa_4)lTsp-xH}P8<r{@_L?D6D8HOL
zQ#9t=^oFN?pCf<tq&i^4D5;hc@nAUY5!;2s?%A~`9QIgq+*>^_?S$PPu|xRiVX49`
zkNsd@%dZR)SyU9Eh=?cU9KuS2kM3N>6Fz#JjquS2U)vOi7~SGJep}o^57eWg4RR=c
zo*vk5KC2F70Nvs`xMu}2&YrJK0rE9w{Kpe_i|as3w2;ssf@p=rrtTC65t~BJxV^&@
zckPT5amUE?gTL(>lC*h__&)RNS=Q%)Y+bZjC~e~ZblKvphYcQH-D5ezrZ|>6e7A^_
zMdS|dXMJrhMls*kJ?w*HC7Id(=?W(?5Gv{^@+EfqtRiX7`LriUoMv##omEDgEBsaG
zU)1rW%@bDV$5eNF=S%+bJ%kjZ@Hu=bMD?`L^M+JFjcFfI?1Sh5;eX$E&f$!k5#he@
zyPV@0w%EttzPxYa6&CS{eI0$9f<aCxJIbm0d$>3HygrdLb&Cs*boi@I(kbFSR8)$w
zrl6nOAzU0?k>#X*ecbf$u|;MaNO(P{MIZNiYSBAO?xdbl_y*DbjQci7tVSxgN0tF~
z*qt+Ueu(cB7v7aZP~+8ka1pKWuW(^Ox^+tx^tkPk%W&1xo>(Ptv1{DmN%#qPP4!K$
zDHxU~NMt}<RAly(2_$y0Yd*t63*#Z~B_I*gqt<z3Z-Ug^fg|91$hDqf@36y1#FL8V
z(S!RhYRO?fUJ-q{u?GFj0*SSK|A;v19EI!a+6+i3ym804QOftZb5n98-h-VO2wPP=
z4dt=Zd(FJ|)p$>TaR!qK8v7ea+~=t07kkJ}fZKQF5An5qvwo`AweQ>np!{^Es-m-}
zPpY6VzUHWP9=>R5vW4Y+&s;bCE5L8`^M{%duG)ZX#Tk1Q>$kqo@<a{>UY9fG4u^BB
z5^^I7K}pm9Dd(Jjxd#K%&-&-@2)p;8QnJ72K~>QA_-CN}sP>Jj;gJJ~R!qGI#K+(v
zKmPFdkeuf$7#?R>Tp)3i00}P+$W!l$`kLp%9ung=zOh#^r{azd5<Zj%ogbg^4?Bgg
z5<JndBRy!2xHU+eE20$-v4Z$MbIu)Zlri7%;>5V&JT8PvxIRM&aYWy^21gIpH*%OB
zbTE8}rPg`mIK%n~KN3jzI^hSqa>WVPH{#n=Cy#$hb@IrfdST!AxV^&=d(}IT7^Y+W
z)f4yq0H$sI(=qBhpYON_yYQtDZ}fm@g`Z3>n!i2|-m3d@ewDdnbYS@V-uuoRCO$fI
zu`XWjS<!2f>7v)v+x0#Er2<5Jg)jW?eS`AtYk~UPVJj2eKI&;_aL3@rUcMtyWjlQM
zR6L0Y)VofR6a21Aa8}8binD|rB>eF&bZ6cDxq3wX`VVnC9whFKAW<Xn`s!lqj0YM0
zT)4<4cbgn!axXOqLbNtRK!|k(f*4dFm*4mi9wWur+d$XUJ)SnBS9bxK+!6vIH9OuD
z-aN`oaL{=h$}s8drOoOVB*^1CIgW*3e}0H(419>&mLF0hN*xYer?lBT#_}P4UjW(t
zl(`VQa2p9EwCO;?!vyjSF{{W*jyVdgI@akLY-1cUhg(nA$7a>tAA}**83N)y)vry=
zQ4(&PbwTmQKled`kl5)x7$VAZ{UF)+==>;7{`k{KxQTyChL^8kc%d=l(KxvMoBoJa
zL=Q;ptTxN(tC5e8?|D5ob&<H}88;x2*#q+WSr4`F&to|t5pxAWD9gVBQ3O4mX@k~X
zw0S&A`iU`m1~>>V41L=0w@v=FQFDkLX~0tT^)pbbS#1{_n)mH<lytEegJG<(hRqav
z$nX_f^?y3<uCs6rV^n3kKK7^2L{bUddrTYp?LENi^fiZ$jmO$}8tPj9G?LwY?hIc-
zd9<ORT_KCFM(Py1ck-f6v1fooy+NSQ-?QRKD#SC*c^d9kZ8PEz@i7?x24@05{9j@h
zBa`}an}U&ce=i&SFZL?RfyF-^sa57{VPq3=3uVM}>N9ctv0eG-z_9eamqp^f+Y9-S
zuuw)mw|`TTTAYu;(}IUg@N{uc2}s0PE~F6a%S}-Qoi?Yl@jyb4i=tzHu_KXWQ1gwj
z=IdX9>g4L7<vrIDZO~4S8A7Qt-$TYFYJXni^7w6vkWu&oKE$J!AmmN>KSe3U*Y>y)
zMUFOC?1`KtVXecXjJ=n@YTR-l>${iWf-M;q3Q6#B29sVDcH)<K<!9sP_V;-}h1@uu
zsgUd08P^1>Io<<Dz;_C)Ugr$)w9uyjd7P7fO%xsWISOCXbKPl!cAw4z$}_p2??R>|
zZhBnW;+6x$a`2V)aW8%LP864PD{X>c;=3}Xqg>O1;6f<-f*c-a0)ar_qr({-2vkL1
zSzp3{uST-TJ$vR8x%OPp3p$497F{z&pTB1wqHZj0P~O*nI*$Yo6Y+ogRjw`M49lDP
zw63a-j7?dp-$mT*DmMSP>B0JQjVR>2M`u)h$mKH;B7iGJ(q?^pUe8?T{M_cQ8+T2R
zIOo0spD_lo3*(zo$vS4pytxPcn#l!?d9c*JLmV>pc)Av+42Wm7&;!JQd?qgY9rKO5
zX8coD@aa!u1^a#tDKq?HXg0VW@%2%l{QE4p0DVtn4~zX^`6Jk+y+%0JE1G}8^5J(`
zxD@#A&j#yp83-v$%mbvK<6aNqK4P90I)E27N@9-(@f@?SjWc#YcF%+Skgx)%;u+Q;
z8NnV!K#)JQbU>g6`Inohg%}7jM1kD(wJBcZb_i_1$5${yp5b9yC}a>%`$(SNx8;hv
zCZz|$PXQ7)<YkSR9`2Z^Jo6!8Cqi_Jvk^vdSj6im;i&(Vc#X$tuN5b8wV6HtfHqOT
z5rnE}(SzONa~Bm7U7zAZJR%J;xth<1#BFM&(4Nl&t}=-r2IBvRs7N7XyhNMZbu_Q9
zgiAbagpA-%<2vQ5gd43#2zk#Hdm=kk+}_vebml!}D<WSE1X(oi^^7ePNN8(aP*$D2
zcEQv0=rTyuGQFty8?mYF57R;WMx7SVKj3Lc_;+CA#fgq?McgDd{5x)~gB(74w_k8<
z_|$J$#2{h&gM@9lv5nq7I&kIvn{K4C`8?R5`ua|>A*K-jbf~)-s{5V=e^Cp#zV-<B
z#Q)D28o4~}iSq|vc$|$J^+ID-kwzS&LmV*1dS1eUq!#cUBZ_uLj3^pJ(SwB)FCQKB
z{e7=Ojo6t%NbDd?6Hj~VJ(Qe}F(4(&ci0WWX#AfwR7Nj!5j-yq<PQJAhG3}Ac<TB(
z3+n=L&fSN21_TI&q;=eOG6wuy00|!oHL0C>f9(V|<7vOe8NIZ*!@`FX8DDN6*4Qm!
zONOU$AFk8qHF}hBZ~UqR`<mOwiu!`YPJvEw#4&bicldnUsa6v`n710ynd)Btwde9V
z?DOraCp5U@#Gn>%%@ERvea#VnirsQ27Fund*mvMWx)R3)5;00@^oG5&uUGxk;&5CC
zi8cP^J|3m-ZwE^Ge#ZcLeMiD$7xz<;$loE)FLp5)ma+SzTM@n*D*hfmt6)uqz4O%q
z_Sp+ZJA4}l-n@7Zx9VBvHFl~qb}p*-4l$w#T>5-F2+a7IMcl!wIq)8mAiQtm8?m;7
zka&#F6xz?wt^dNikKRZ4#ZZ8V-4ae@_^aSVMtlJz)(;k}XVYEOT8~qP{k3|%?%{(x
z@I75V*6SS+LYR+X)toF~KeO-^#SQF)Chq6yhqyI({`SZi0Et!qAwEOSA6#f11GtJY
ztB~LhI|MoH!S8?wQ-==)9o*PmPSkH=M2Gt><f(r;E}rpuo#PokE;y0ZYwB;uxQ$Ol
zt*C<=H!$Qf<Ms{`cXW`jLqH<>4+kqejR*lojtod-NW#j@r=e}-vufTWT7J%?M?{+^
za+uDUoLq0i9uhMVWT`gwwU;ZrY*aXicow|Sh{m3<w4>MS<H|4BMNrXCGnkwaU4}Ik
z{$48NMic?W)y}UAQK#qu0vUzWeSwsl@C0Az{P=nLF**IqQHR(0KU}5Wh>pSWj1wIu
zr!#i04~c(@SIXxb0ms-AFYep%w7EQhoFw6Cgcll~IYKGURfHJ(hX0M=M|~fr^{nfM
z`n1(%-NFi>0&92{ZoInwp8M~0J|Ii}&ujUBV32CBZ?e<f@}>f7_${er8mIG(q`05k
zH?dtm>tScbPH~eK;CD^<jhFA1AaS3AFB<uL)IW`g8ouKzdOhzk6X8UL2c5{_>lPQ8
zc`>W*ss_n6&PB_=>9U2be3P?XZJvlXQfD<z+IuCG;yoZA*2j&yj&Wmo{s`IV)6fu$
ze~Q+H^IS1p_VD~XKkkSA9^#t5=I}4QK5h>wCBBQ1W8l2d3ncO!o&jYx?;AN7F`}sK
zL>uTBzP8CDi*xQ7pC?9#%!K%-Q&e4HQrG?T!Fq~b!)%CNvtP#yVZSZ}6VgYNTIYVA
z!m19R!^2vL_Yh<$1cIaYU@bP$v!Y?5ZC>a^zVm^E7Y8Kf8~jFRR9$c0=U&w0{L8_?
zir72+#>j|*-xyw;_s8kzV=z#vE0AnMOv1myOX-Z)pEeTF3M5oHDG|($9c|dL3khE|
zRRm&Jeo4;rlNPpQL`l)yil_)k&5$V+L$7!KbKc>HM0|k=SVU}IWLsmbQGN7%=bg9k
z=AokL3N!SLO2z&a<|poRF9}h;C&Il9i}+o(H~~Q-SBkqwoFyM8(svg)kzvDv#Le}E
z&>R1Uq}NyrD?SVxx0_m~eq#ZN+JztD{$3C?!QwsVq%)p2w<~dgyw171sekI0DhP?*
z-h*H87ibgtqadeq;XtBZ8^SNaH~^V_)qi3!p1F9#%Jk?z2->m!Psxzfy_YtR=hRcQ
zJX9Y+M%RP^!65Y>kQN<%gG#C^n5Pl2@Tbk>B?rQy3*_)DYh;S?8+se6yXeZ|SPGRL
zkYCN~W%a7mTn@?f_8zby#q3=p+Q-u#k3fM$&H`7>!S`MeT<F?2Dx<nZOq(rwO)XQm
z$+^-}PTSuDaZ7j08`H&o4j`LnwQxaR+(P+~*XJ8WkNu=YQYmWVeTe&EK|*;Mq~<6}
zRGzy+8*b@7Zp)van)h#+W4|U5#_dts9G=Ypf-g9Fa67MnHjlH)ZU`YBqXY?^;16;3
z4G2@-`-XUjuPm(V*eQm;!F{TTVa<NW_>cHGH&A3wEul|Z%l$hzaiWgR=iYgJkN=1O
zzFVSb?%YeBws;QHhq%WXWOr`gO&rB-F%V>f^!q^aaV8CIR`-~JY<}AXL3AqK6II!0
z6YPN-o~!$ALFl)?2PEdfIGiyCWANH|&+8tHo7<enSU^VS%7c)Q<h`CE4tSHISu-9o
zLOg8_ZAYFqyKfLA@}fZ~aOu6KG^?*|JcI5nx~YHa%6%XdUiSCEMTjp)dc5yD!$MgN
ze;Q2n*bmUvjUH4bH$?bq7lVWAnMyaZU$OI1mo#QC+*sFTyGJS^`iai+aUR$W1J-j1
zK%nF4xUqcP<4l{#O1=?2aUVX&&#NDROs?4j!u0jNQIV{W;H2=h(<AvH5zqQu_<o}V
zfqvtC!?g4r38roAA(UhB2s=gyZUTP~OFQNPR*kO(GBgk0LqI(D>mIqZ#Xm*V*4NL7
z+S2R2UbeVnAaN6CBb1cb{YP@Lq6eh=M-SFFcG&5U!lc@3!hkWaVdsV|20P31<8egu
zB7YCLQa*QR6DQO3LvbUX7WEG4!RyhA4+%fTH2gdp2sdz@+k7|o^U-FD{dywv%D>z+
zLZI`1O2)LGOt64mRsH5}IQso`V$~;7V=#p}(f!qPt*4O~?f-{Mj-SEQfQ?fV`9zPK
z(Syg?tT*@l*b}ME8v6lA&BSH!3<!dktRVlVU+J?zs5KGD<+>5P=XqiUufJN2Yk8%n
z$|OUEmeT3oH#i*rPd`Vn+pr)Wxx2rzaF@Pk{v$6rkl?xUw9w8V=o|HAK*}dH|CZ`C
z679Yt%_KMZT9|)N>|!8s$C!}-@~=R33|9;#h>eVkj~n{RzV~91dlZhqH`E}18aq;a
zx$*EUog4P_5pxH3h}*Czxy5RPP3v52-a~C5?=^*vVhpA${0G#N^_VNNvB<did$_Q~
zKSeXn_e3PkYsQne9=)D7=OWOz+C#qDtbYE$EsXo=Oc71DLtu7$o)8iTt7kKS?6HqR
zEw7aY$1}bu1rHPPY{z?s?i}-ao$H^j7}EN+;bMCX?4|~+NBclvsm0Sa&m5smL`gv?
zTNh6&OK3r>GX4)q<G!}x;$@pNyhRI=N`3-DT8i5)oNw14LL7V#%Wk2zms?Nxo8v}F
zxkG1Z^Pbn^bN9k^%Gb{_-r?&6iCqjf_~L6DB>Z9v5;M_*X$os&A&V-$DJn?uG_-3M
zpH&M92!RCCmaup50YPTZKLCN{>F=SIPxOF%ifVJliB1w#oWU!_U3_$4HHX!?{<eq}
zK%a|oTYuC#(tk>_N$jH^d)oggm0n$+4`LD_LVpi(#PzhOXT;N@jv5G4B;K=zok*qE
zxNn1mRRc>qyx}0uvAHP<>UpXl95deQ@rauq%<foOgy+JFW(4AfOliqjjg*!Q&o)Th
z+ExNxe#&6qdyN0)01Us&I)|DzZJxMQbNPt93c~2>JKxH|5WPmB$*rdqMV3HVj`1|O
zq`tN{g2>*t4Q&QLWwyU3PNwY_bwzF%<hQR9m^=}M+kTs?>)#upl;|}w7(ND6tq70I
zHoS2LgT%S8Aq40r#zq!Z{L>M3pRTzLUTE}sJY#<&b25A=aK1gS<_2+f6_CRf7(kv_
zSsR3Fw?{W>ofmS)s^4Z*^nG-of%%C}ofhYp+;G0*MhQzhPNt1)p!h!I_r^EHJLzW$
zEbTbw;7G@bvEvKye+m+_cmF%XrwZdbBAzg=!xrPzcIDW6r<#MGr+cnbY9~hK$WHFD
z&yXFzWg?LQfl^rPy*q4`==Cr-m*<9Y7dK>h;y<4UJ2fc$Ua}($>ig&pd%$NrO7P+R
zf_0ux+fSdpJ8{i;4~mD;>*?@3yL;y*9%DduhT9bT9Q-WI?x-RILSnS;2Rr$RzWeXU
zAidFM^Xe|Vhf31XhM8SRSdja2c~&y-xg(ctr_x09dWHJvJwP%idd*(tqjUTr(xFc~
zenTQr|MZ5DAJIO%mf=4DiL((lxU(hikJ;_M!2=Ok|31gCgq?U`A4Hpx+6+EA1RP`B
z#;BBhALA3=WBA))ZNNi|j41M$BO(CbI!=sZ3eM_%%tUwnpQ6AW|CB_SDcV2<C?sr(
zV?H87lDkLTWeyY%<I62;&G==FJrNIzYx>;>wY1~^RJ}vmAQSInz=grrDR-&ccV%+Z
zFW+I0^$joRiHeKcAty<VzQ<F;`SgDZyFK<d*zI8-oP;&wX}>+rQ}Rlk^>?2{5quW|
z37e0M?Z}};A8_^&J%=^CUoh1p%laJl@NA#>-=fzPzb=HdZJj&k@Weg%B=c(UHb}n5
zJV2rLiZ(-iQ<Bf-XoE#sNcfScO&&Li6H;cpXF_(%-&6HI?sLNSjy6a5#m<SsE8d1m
zAind#h>yMMoY(LK!(p#)y8MRd^+Hyo_l>g4F>Vm9-KIE~Ir4W-bh3OGJ4suQxwA%X
zm7C&Gai4?T?phPH*&^HIV&C!q1MfZd>yP*DYmVYKVcCI%9RlL~lKaB#;$Pt+rP0r<
z4~f1F1S$1vQ&c1FbJQ7kRhpYTGS?p^0gwtA{vNU_{bWM=Kk}Xs>yIpJkorFer2EN)
zczXD5k@^n*0lfFP&tZ|rZHh_?5eK~H^3Yy$`Dzf!4eJS6+pz3VV2z0MwVZJ_Qqv={
zn?NGE3|rsTRc_RKofrBcv0Gm25Ud#LJ4dvfQ!b)bP=mq}qH?aQ_}@1HcAuj+E=b?6
zsTLhKiJN3CU#B1K!dJ<ci0|u^n$QuWB$2qDMijyKx10E!uM*Nx!<$E9XLuS(0*t*1
zBy94V{3_r5zl5Y1ooR7pp8H0gFTNbyf43m-`Kewn6kgU{>;{SNvp}qlHfwlfZsgyp
z&3XChOs(d)H{Qz~cNqdRk<meY_BemY&#!UAaT=quZjmE%uPl5Yo%IPe#Ep`TkM;96
z#Cx7U3@d%llWa+9LPs7k`UkOsK_ViJGD75!K13pY&7pq~9zawMVkV+;5VkT_RrO$i
zB<7<tFmi>2#|-7#u!tZ2PCRX-z_Y&x{diyX58BT$ZsQrZsR!qZ&rz6{HA6ri@7sg}
zI}ivafl%Z+5DG8^vJJi;JQFu=Z!`ZzW)GFDYle_^6yFDJt$5G!i{0f}5J>XAp_Api
zG8v~eqU#nG{)2SJPv3(!L(GF0s;$rX7piSNZ3^}pESC7%FMbN2A@2`wwD<aVE<QRh
zOby?~&~tFDmitZn;2p-;L*8MFDE13E%4aXtuU*0MeupP+MlaSz{8MrXd@Z~qky;<O
zG3tTc#QI`hgM<YM61j$7E$#TG6PFq98>OgxfBWih`<kQvcCZv)VxiH7Y{!_rh|NY1
z<}L0rFFH)V>d}CRbAh_p;n_wPE|_)Poa3hVqPQFXhZ>_X6OsO}HpOduQO1syMa)%a
z;zNB;=LkH1zsR%zi5UU{8P4Vo$l|;l%$+}vQGSNdq$}%_i1OL{Upz`j6lC^j5=f}H
zfb4!>c&HZYUN#U&@jeDr_*9#ZHjn2{Vz?ku=uabCSl0q=f}INj@i3kil^tobJAVNr
znB5PoDbK$KAt~5<fZ<RGIbP8wqUA`11-k=ei*ft?{al;jVYYcT5J>QyK@L9~pWknE
zFDM8FQ1zc8M&hR#ZNlsE{D!E|4ia@^L2mbCfxON;d{E8vI}^y}XCuh&XE4a=o_dhb
z4S!HmnDs0HApqnfI>lLnWE|x~ybY9)_&%h5=}nxcg+43@RfFO^$LwnxJqNd|cv@6P
z`VjYVf!u!M2YH;M_WV$z_p{_jE<ua|83VqyNpf@B66f2WpJpJ7pUxnW8~!1#*#i=H
zT99jud9Xe)M+ahu{!fR$!}+uistqwZr0T{#I{ZD(UwBY$aN8eb_uLhb!~HHF;&ED#
z({IQh;u(0)z*(;r3dj;CEeZ^7_tIu_-yle^d_bb263FZEQyds&p&-Ll^I*h`*0nv5
zspNarCtTsy6B${aBlCcGMhD1JUxDOwU%|r^zIC3qd93<DwV~$QS76h}fJ0NQTb_1$
zh9q)R5z7Gyl|+y$EG{BBu7FCL+bw*M#}x~3WX*A+Q$Qtbmto!gmc!FRN9G~n#ytxl
zp<wbMv(H2j&%~t7?*8i!IsIfB|IWi_FICXv_JN*5_3ih@UWJ}RSQ~h&{5HjAeOIr)
zlIA)`Daf42>5;n!`j7jfpOM_-c+a%O&PU<r*pWay2KJ0I38emiAcn>~Kqb*_q>()Q
z!=J_x?B^W6dzADc|J!5yAcyByfuI5&J$UkI;j4MR0-y1JphLuaxE%U@Zd{{E7Vkkp
zQF}f8O<`Ao#C{Dm*5maLh(}Twny7>b5-M*GT3GJU1UcQe{vmOrr0UD<BRV1d;yVRr
zW~_yY<><C0N24osJ<~{5mVbrubT8=FChiMpCWNIr!K{m~O$qQAgXwg?<@1r6<7up5
zKToL^di&ZYEx`Rq4`pLLF94+OW*_1fF^Ff+@U+v_6ds69ahIV&yL(${LrG@;rpp#L
zS3ZJkCOj0`_8fGO-F?v@M|cS^o}9yu$TdY9wAb@rosA$dqBA<^^|Za-L%2je&R-zK
z;|m~>_XM)IpB$uaMl;5kpBRu#J<H(1w2akAnFr6~qD|x;gOnGC%X{43sW=~g%bCDK
zjOchdn+Qqss1OB0EX8NY9Ceo;RH-jt_49)D?@#00j`u99XPZ3pa>iPinY?Zd&ZI{8
zI-MzQ96eY^kBL9%u!rwyCWP!eED`>=eJp>WO7pNGdF&kocZ7~8Stw!0!=gX^`;fR8
z79_DtkDos9>z$ABP)ROq7fNY|eGL-c6^fXJ4GZGAy7XX)^9N48XIju^_q<M!%ee?3
zMBDr;R9!d%xCozk8%!iWY4MUfdyRAsZcY9){PAv~EZlZ|HBtpVPUi(?Q1orSJc|8r
zG=^tk!3=SW3nY92sC8DaA#nRnzHkD@KgA8}yCsCgh;UFWI&Ql-_rtqFhM98+AMVGK
zkLXHntM5o4UZIdO50K?`9w0dq`vEdco|}iO6~SnK+UNENWOa2(p7w;NkyNR$-&QsX
zpS>UNJ;rSetx6IgcGt7}kl0~ix5tUTBIS1a7!XnOnTVi7tgIEiLHECr4eL2f&{C;b
z?L8nd&S(5uH36P=AtLNgTi&>zuDI4?hA??z_R`3$^>sxe!+Stok(k$OWQ{*mXuo|7
z;L5)~Zsd1GbZjHD)c0!;&q93Q%Ex?zEAP3~&o&Uu(J|PH3rCyvihqjL>9Y#nd-&Et
z!gsrotrhQ?h@g6};eEJ;vQ01;q7Av`g}9Qy1G_yeyA9t{j3}n6&v@#ux;D_WQ3XBb
z!GaIadrh`hi~$7Y7y}Bl`~J4EPW{x}D5c=LC5-sX_bTk57`HuddOHQy{QrQ2&yFmv
zxS#IB=vi*hPQ1kb&xgeS8BWjNp=bpqnf3cn*U|I&XaoH-+K^e}$W%MUj;!t*ZJv4$
zb!^;1*}1;^D*390`wqL$OL)U!gNN-;rc}fmcS@Q1?1d*@{|EJH-^q!DhGq9L$bHT2
zE2W!!MB#Y)`hlfA`uf?)Zi;WZj<Hkh5X9pDq=yz2&HAQmFy0<~?s5JccxZiw9Dm#5
z<3=U9$=3qed~s$SRHpN<4HA156~N=h4^r<b9oh$mwDU1Q5RcV)aFDI#yD~_uV7&g}
zM>+@^#kh@-eZ2=`RBe4=kcWMJR81P%%=NV=XiMG$a!P!TQWrJOIU-<@BXf|s7_*A&
zf}f}49J!*zb9~}--fQlS@l8upmZ!auyK=0d>+yhuO@XpW_}^e$N8ZNycZ8P><n?_A
zE_PVa#5dwZKZ#!YxSc<4GdiLlOWVI3-_13m9$4C84W8T^{rsWQc${-o8jtk@v)dJB
zkdv64WeS2Km;cjY3m?h}k<-`0$NBbsl+3T{0X3?8j-JTx#P=Z>8ed`D;?8u!y6}Am
zE_R%NAhCi`w~UCuiTG#CQG}dNAL}V<z&_{nM&<~N>xdbhxEkY|lClteTOu?2IdN0?
zuFTydcD|EPj$01ra!0Iys-!h{*3H#bo(l$r&$o+gEuSG55_rDm;8n-0g6|wTSr;ji
z@ifXV`&n|qhVb3xVy6g%)US9Cs?a{xRF!u%669Da{~9KA_zy0MUHIIA2_0XX-5}x%
z7g?gwYh)T@493tGd&uSsTkIOCNU3Y?Lt;d!qUmSwg%wb37zKY1i^O-!3!(fN1L~H?
zdnkHb2*Jqc!5cC0i-3vmi5C(#v9hq^fBqiaeDO4#c5k$ScOFQnSUwc{i#XFw2|gcD
z*wc~ge*ZnOCsKH9_FjK{XCDKQ%fB{=GijbX_k;T@)TN_CQK#wwk)djHgctOt$dB(;
zH~CfmeLyasZ?LD&7z4b=ewI+RqQ2=E`n}H$RngZRH<`#>q?khNE|fnAFAhlLu27{u
zB2d(#4}Ux&?O_E_Jw7aAYQ)EGdBbmXo8ptKUVQE>WJd%t*Q`P*CB~ZQvhS96=}uDm
z;C5XckT;?{RGg1EElAv*Z*s$ZzI{|0A2*QrJ|7p`M-;zl<gPFaoEu9?4^m$AD}aP2
z7=4-3*Bqpj@KS?>#Rb8{6$>D&;(+tMacA<m^PplGeM4h9o;IAP+y1b@6S>3d4zoK>
zS~6E6UyQWM*ms`saeXleUFCNn+a>m1lJg>eheW&jKCq{KC#N1}?AIXm<tBKu{!L-U
z__>WmQS)uCeE%=C$9D>{T_RfXKq2%KgM7~`X2^0zwEQ6*DgNnlMJCKcoi4W=9y}$n
zPFKVXd_<wO#TZoe(dU8T5_{EpG!kd<%Z<g?-21CYG0M*|ZXgkJCATR&n&eBzI(=b;
z_#X07dDG{?J8%cnv4+_lw|6RX#<~3xlki#f!qOi6J>-kUh?0)v>-5cg2_Dt(siOWC
zS$i++4!6Z#?j?`+8j)*1o#Ags{tkD;@I6s!HtrZ8adUmKLwq&9h++9EdEt`A_o0fW
zuTvuLVIP3R9pmF~$Co1!CEl}ACDF%vUB?(}6lP*xlZ{$Wi|W@ngh<-99)OIVnGF)T
zUmxO?bU;?$6F<c5y>(a|S@-YTI0O&w1W#~xcMtCF?izvzg1fuByE_Dj;K41pyWb`=
z?{DVK%suy>zt1Ti`l+sFeQWJiYj3L8r*;*$H(|ogBcbz|!wbl_eO!ddj|#WBz}w8-
zND7V1!bN_o0fS+t@@I?ezpf(7(b}P~5~#v9ya>gKoM?Ci=Y1pt5it97g7c+2+h%H$
zeOHR5e;}hjsJteD#dRnLe%~a#lW<Pq;;EzsNJr|tE%8?&Ve$#Z=qCBt>ayo5E|@!5
zTply7IP@J0Kjd-VNg_+%T=+z07Jhy}BCxpKO@|G3LvjZpj96Xa?de)5C5&19!MCvj
zZQt|-^zB(RhFkD04+Wl9JG!SUIROFd?e3S%{T-_b;FUIHV46pM<E70`5m0Je6_Ea|
zLAzahEL2t#3ui3Dw-;`sOsXE?(XQM0jsnU9$S;oUX<A|_1LI8zA5%=?eQlv1Uig`2
zWch*#8Vk0BAttb(2pd)r_=$-31@SYW*!H>o;tcJB?{;56?0{37h-|<NZ(Py8t*4g)
zufM$&<SM_DFQ-<9v6!@RuZ3wNNBZi;1G%|Xh3lUQgqT}bnA?GkcrWjcSM?4=#rq#b
z>NHkA=^424K*%mG;hZ6ulIlF;e<Ny_DY85%F5;rzU-DUV2*2UN_t`K<$2UQe*t?yI
zA8>0q@s?!%kvW=q&1@xbgh*0-tsU&yl<_FIbp3+V!Re=|r{L{>C1g3`kk{xRjz0|2
zRrwv6RJi{Hwzf$#XH#%#2k`bx#D5C8=?p(~V9#XptN3~gZ-V*wsr$Ov06wow*T?)(
z`djNkk#)KB8x-Gjhfk?~qrekmv_(E5o6|4cT8!W#`W{npilhVw`V_>&C`xvGzAbR}
zy=uJCFF&ulR04X+;E`_N+fjU5+I`nYfb%VX2noYR4t(w<c-tUIBXZ2%*Hk?{Rd{m9
z($`CHb;6hxcpg#ipk=-UoqbyJ@o`MKvaAF}K5ubLzD4{4mrp0C=7GkbzS;`UDrT!8
z<<28m9f@R4dV2b?Lvx){eno3lkfZNhRaoSRg_ayB%++FEKxgBure!${2XL}qaub@a
zW=fK<&RwR-XMrlIvN`c+oN=F(qhu&_iYntnkWsBrFK?loYj4pprg}8-f+%AD?}dPZ
zfn?{jpB7U-Vh<i9&`wLytM8XJziDh1k&y&njSz7bY<hU(X66swh`*oD&%my)!m{Xo
zF1p6utuK#4=K`U3d;+hL5(Nly5Qmgjbak<upbR%ATKo{w@S0WIwCiGFu@I-~F4z%x
zb0ql5Z!eqP6Ai*)q};o$WnrA(<U9QgABLmB;Gl6*8l_AD4%nLk^8?uf0WFvhgtpBw
z$@rX_3d!??ByTs}6vU)+@IX$y6=&S}jF5BQF^MwF(aDC9{VY?H+$A(7g2j@<Jz5_q
zUstclA&EKh>+J<)*l0V(_TwM;5-?!uS-_;j*|N`c%%bZqJx!eBlBI3N_>tP(coI~3
zZil$-V`8BqO9it4Z$63CB_(M7sER5{zeb>T1?7I!8eJP{RBdy!N-W#5nj=|o%sP+R
zLwpsl)7vQnrcM`@+34Vq<h)U4$NA~znE2)u$Rb>YO~#3xMOFLmwtjL^eXbb_dWUzc
zTB&ZJ3cU1NsKL`5$$tEE4k@GMV{=f!QHi@M;RVG<2%p<WF}G^Q40Uz^*3N;B09GA@
zpr1Yvy9NE)f+#kZsyn9?b!0y;zRySEqJC|``Qo0}E=*RN-X+HLGxi3CtXW{rR*RB-
zi!Q6KUle#1wz)Sa%puPck0Rd;4}q=ZCUI8@%LP-xd18H=tuao{oOq6$sLv7b_P9hv
zBRAqjCwZ$(%uO!dIzbC)Ta7rzj+IL8I-om?y&Ofo2bUc1*su~Bgcvxzyml9XYwg-*
z4*S^2CxXjpM;g(ZEK%-ktJpEZ2WbL1WRbCC4JkWiBA8IknJJ<0sWxE|h!D$)R}X~r
zXCK_<Q%c1^5QGmvxa?{^iRydsko4n6I?dij^3}`DrSt30XNY}3!Js_Vs@={JYzb<{
zfK8QTuX!U7re5O8u*r`e((M{478;?5L~7fiU~$by0qtovEPCP`M!k_NviUwTv3Orh
zN;HHo(7A?DAe(+nGp410j5=q>nSX&WCkM4@lBFv1n@|0aqZ6xII+2BMeH#Rl^u_13
zBlVyZ2Tbsb>82u5i5QwlS$R=n6>VN-O}7|IiVFK0n$CChyKs++C|#ML>E$uPoC~`B
zVKS#l=?>de%#8r|UJEHYA0`D1^3EC7GuQh{oD=k|+y&ZL7M8EQyBaXYG7n00kMN#>
zQgy~PknUEdR5)*U)`^R1MK!D4=8{!u)D21`eEohLSL*Dgk|f@0ByqsP_UJeq3&0ff
z?l$qcNit%{V;E6R&GxbkjjWeMV!jYcjBt%Ed4dSFRe6wJtzN3jN)72&uWh9)e{zaT
zNm9~k5<`C*Q&(JWK8oDAZZR6HO`mx`{7$@%Ecz2YzM;_(KXZw+A;MCHaS3?yfCZ02
zp;*GfRiMl%_PasmOO<#Ii8cUAgOL6?X56xmQaLtp2ut4i70O5d87`b6sEhnH^I9}N
zqui3XGw`^s4yB@+AhA$h^{$L5O31cF=1szh7d_+j<m3>MRWY#S)pwY>$MG2bex>#b
zFsu`S%bDYmrkYHOXz|i_(g!6?K`JvU2n6ydOw;X>(&zFA2&o<hCBU|XN^`q#a=RDf
zY(YP|*I4&_JsGL-@fW|`rP5Sx&!31)hh_F>ZHp(Qe^6k%Gj<AavuPzqa_Ha8o@xxD
z9KhdF7mdP{LSJy-EK^8YT+`F*#gQ!}<0__xc#9${^xX`x!?<jYu_S-M$8BnQU2l)k
zx)Q&Nw|O$n(z-gq_B>77B;X0ld^h<Aif>tknq(H4q4Cf9(W*#cR{AUsTKN3jF-nw@
zEK%{|9RWdemGRR4pA>r$))V3IiR@M`$!Z1+xYMqT#1cv@ej(Po5}8(9?U_OKEI}6)
zDZP$X@bG+UWjXMr3}rU(O`uP8I7k*85i4{%OU!vmq>k9Eg)k&!*5c*ky(Mdv7BtS@
z5XK(okd`HG<^qc%C91M%x%8aG9D`Wk7LsXG%H^=sbQavR(>6>^xNL`Y+8DZNDNnMn
zg!Qws08GWjUUc1p*$?*~2Sc=-3f~M;R<qKJ;!GEZC2Y>Gbm1J+q9^3PXnL_%Um3@l
znN4OflSqs-f`HL?fKASZ@_#owHqwnxZp3`+2WvW&bR68XOmUbsz`^`TRc?3JkFM!`
zXJ0f-(?EC{8lJBOeZRmkb;8~GiM6njg>p&d8(d9FOC97Q_NT}c+Jpo$2GLE)8f%x!
z-LkC|XO%eqWMX;?Wl=dM7gmM6cb(L%u0|vx6=ql_hfbF5`J&`3I#a8)U?w{xX;oXb
zB#fMfz=1hJy7v+;T)KCpadJs=Q9Jo-(vlu4t5dyTY~mycs%1KShNecsH@&&pzUoE6
z1SGlXdl6|hhXxYIpV*ybKSVdUr7iucZ;a!zP+JBwoYnMNoiMlhwR;HXlpx%&6mBRb
zq_hL{H_40xd$-=`tC=!#xEZCx+h@czmU4Y=hli!CESXKmi~WEi7_^nEY%T_n+Bp`v
zVswx;*iiZ?8w&3ukaaMzQJeq?0^;Y9Wr7Z8ACnUbcb=Jn?b^IzFyXa$%F~@q-{f0q
z3V)0qRHd{Q5~l69z*sfbnL2fk-no!8Ifz}80h_r890T733_=53?Q&u`Gg6CVD&%em
zG7>iJ+BlxiFPxi>3I;j`E_C!dQfTUjX`g=HSMN16Y&Lh?_F|naS|v1W?zp#c4%-Ai
zJh45h*k3q)LifBq{lI4ZFiIBsw5(<;e{f`9$AkZ4{n>GGw7}EDwc}(FD~GEY`}pP7
zxuQjn6@If~^Lg*|-eTOq>-KZQ#_8RTE@b<%O<M|2dBe`;K*3_lg;I;B$6ds(S;PIo
zFs9Az=EEr$ZNY#ZXYbkF{0Tlyy3W9^=OfZ{Pyr9^3CrfF7T1n~7h9=p!4l6%IW6l!
z;DDL)AR5X`+5OqYUdCeJU5ce=^qEd<+VkGYMvC3_BEgye954I9QN)cw$VFA(L)DfC
zl^bt-dPLsIK=nlx75*m=cDPOBwg->TcSqx{kDeV1hZ%u0HYevVc<y-qGtc8sirmr<
z#vR<DJ!iK!_tx$k5mQKb1P?u(Dkm8%C_@c8=?JIdvQRYbRDl&!SKF29vI!UiJ??c!
z9ojl$D{sU)(%SZPu4SiOcKSK{Vh~sH+ElL;#<vR)@0R;K=6B{YJiX#~w#^$ga!Ofg
z7Z;xb5_m>^2()!9YItN=d$I@~%y`p{c+-wY(gMsjJ`s4`VE_OxFJOR=l3?Jd07w8d
z004Lg!2H_jg9QNqP>})vZ-HA}Z0rqbt(~m&jqJ^=O=w)LESJ=+Y;r`9p0$f!x&*2n
zpCS{#7<J9E7o?(O$ZM4pOM3yj4lwbkOuMePS6$+n{awIh%4>BtFr7_y^h`2Lj86w-
zS3ld>XL8wt)hkZa6-!Y#_=XLqCk(Ad98gLo2~~}<)ML;!D{7nOdNkkN6+{d`EI!D_
z_H@F`>rQT9a!49s46N(3grwqK9qs;b!NV8oVrh2<^+LC7C>Jyt@U}5d4xZlXh=f82
zBdV;aUg!#sjbZOFr-pv-S62^mbHvt+aC<S*aM$qJ6=GJ9(>X(PJWQvvUP7L+OMbzj
zo{ilxj-4c*Dp#-|!V$6}vrtWCtQ9u3O;#<nlx=ewi~tn<Xz3KK|Dy%KE}_}c6wL0l
z&2fodsIgibvS}E9$?gGK<IKELzx=13<rc|m%sBty$w7HMHBlY>h?%H=&`2N+4eFc}
z_nB3Kk#PWb@?*~zVmomUuk|jTktLYr;Zp8On<Q#GhEJ~f(OX;FS;B2_is8pPoSeBs
zy)Nj8<5X+cF;Mt;$mw6dDq4F}9~MyjtVelizajr(gP4++O(v+iWxcl2b-gy!CIg*R
z5u22Fc+xiiB!M{n$pglW!}#XBcX`xiBc1$su&Ro{eeWwXJvCwSC$|C4uQ)UwU(wcA
z*)F{enS{QX%j6<<I&sMq5VIqGwLUVzV4b%6{7{QI@1DC_^GU%mjryu2Ci!faB|f2t
zx6yl(!AC(8a%hH1;kLSF!9k#z!xc2sWI&h!n@10sp7Mmp-ASi!UsGw4s|i#S9M#b~
zFd|yi9X}ic_x+2+X5#tC6(~s=D@o=mNM!As-0=jyd&W~M<p{9~!QSN^tqU5nd}V+C
zQiPV=gm*;;K9TBv2y^RZ@EmxXM5!|Tfy1^dw7a1zk#t00gji@|Cuokj2y&k$^h$)_
z(|i#aGGZ4<m|JBhlAK)C{l*9dP2Vjjicg=6Ix5?U3DHgp6gYstcd`qd_yn=s`f8VC
z9)ykUi{SX4e9r_;dq_@uy{bw?sDn5<C-B2#nR(Q>OMW~(H`kBg?h55^-Ks4|6hUQ%
zi#qGl)Z#49qM#9U_yDsd&<Mh#p#nX4JsqhY5z5nKh&yq78FGE>x=t`(kjQ@OQ}-Zj
z4v1w}2z>BnuZb}AHTW7K*tD_9LWuOWx6Dl7TCzTKA4s{h%M=Y&8>QD!7HlY1W*)Ad
zH2VtB`P%QF5+s~*6!pHIG)X))u2r}09C-Id>FOl#f9c55z;Q<{$Qn${&(_YB+mXCx
zyc?|b>a)fY7knovw}PDn7v8|fQbxU^ae&}|DKyXEn4H(K*S?2$Apwc!=!Pj_m2;>+
zmp+;Fc(5TBOWktu;8gB*sb~K99Vd0B`~t&|k;7UAw7e<oQcuF&z)#trL_j+<%&~V?
z;@y+vr%?)02GXQf%!CExxhC(_>x#4SyzdwWHTSv};R9w?RjWgx&9mv1@ZPhG7Ln6Q
zHn2!m;S=f|akccd)ZTNLG2^*yXR_8ObuG-72rEL)pDt31(T$26jg#%czfpQhah7^v
z`$$;<a_w#)=x7AYzs0{0e+I#edd-6bsXKc9xsJ06jgR`5)=r@<w0siN6YqEX7cl^7
zdmVx~$FQA{@ZL<v?cQqH0RfpMAGlfx0xzJv`bs<&cE{VWU6D?jV_fbB4;&K_h`(&0
z=Zhnkr^TaNI87Nof8?b8%&Wm#b{^xeLLS2kQxtQq@I18llK?nO{&OJh?vZ(PKmY&=
zjR63}{|uxKj&7Dl4zHu=me#WUA`5!^YSL$@@MAp5SgM8e4|p#^(#EONb!iL3c)|Wu
zET~1KMQnpOSM2xt+-v}fNJ8+=+?o_?Mg<KdZ$viURdxW#eL+Tt<=Yd+!4Njd9lyzv
z^-f~5L}9yoHUbY1Ropa$)@-=%6k^q}BTzcx^tX3cyf0@{Q*Lse(qilUP-b@RFgt!u
zz$$!1M2M|43;)@N#c0%aO`|(9igkl$*H)Nw*_I=@M5}s!EZR*HnhyMjo^{90mO(4`
zLFE!Qk~y`;#mL&;>j8H0%mFIu*|U12HtvPX>)7o{mT>`jRJuLs?)yb>J9Q_<WC}d1
z4Z87QOQm+-LEIW-ys7fcJJYWnR?$8UFypJ-`D*x7C`c<)rVOs|(IM5R$)&zp^5~Za
zo+%d*k0{TohGuH6vOE09{bwGwXqNGAxNw>N;KFWly<HxsNH%<61m9v@60ATmPp9tg
ziKzEydIZW`rQSsoe)dPt@H6C5-4+dLao<szWC+h3AmnZCz~t*Oz^%G{ALgMk>NjI9
zg@Ovc-)y1`CO}w{LiH$={qVLu5wY{U@^rIhk!#GpGW2fkFcf1YM*5K@oJb%{QBi8n
zPUyLgY;V9<=;3@7-PEpifNn8w+ha}EgyKH;>xu`=VEm6aYk6IenY<H_nbehLFi=pn
zAvZaC5b%mrj}hLuav(m2iyvo6GOf4QZxKP)OI^}@JUEEmK<6u_R0oz@5*<%Aaz9~o
z(rFpdAFMz`n$5}4Ek>4p*Df+bXV{Ff(m$Nlu6jRk9n8;OuTl#0<3KBXOqCwr0JcvV
zc>VqqUHwO@u6Y5u$1Zd|Z`8Tk26~B}U{L%GhIh43{KeQ#B*OSSt&4<+_!qh@*QQGr
z2oFeQD<|=J%O9g$Gh|C{jMDQWM1pQdgwsb?Y*-eZKK4fTVhWwOpgw^~OSI~ovoQ#y
z(QPXH%)Mz{B!M2}kR*!!gt!GeRVes?nB>v{FYq;?zspJSK>D$frtN_l(u8<kk=U_~
zMD>N$&WC|*S4v@2Q7TPi&$hTq^7GD#IO++D028{~aBNkJUos97TW=3h@<+5H8Ou}%
zCGSyG)kfP4&}$odPjt<thXRgzEM8dQ@xxX9M=D$H8n4=jo_i<bWD8XVcjaSdhwJ4M
zs=@Oa8e|^b+`yz7C>11U1zYEhVlT?*MD|a%VLWXibpFigVQCx(`r|nW6l43k6uXw!
z4tB~Ce%B%hjP}YO!G>Kr_xtM*_{MTSiSi6M4i7P75=yEx<xi#!V8if1R}W=*EgI#q
z2VOI$nuMIwd<l#5b#*p12-{i6w)>=iHpV}&44MJ<u>pD)Finlh0Wdoh%(zgbpTy7o
zWb4>zpTGSfIBP+T@nR$?iIT7ZX+mMC_FD(jN=X9*v$c9A8>LuS3OdEu53}Ct57z3s
zSLm$Nl}vJi6~svjT4dX}tVWq1P%0DUc(<%TrV-sV09S%5=zPnAi*Vlw+%$CA9R&M|
z7=WAg&=T@IAxL5xAR-xA>FPH>i4Yy@(FS5Pa7Q*k1lRZ~p4ao)&4!1!Qdd?6hmbN=
zekiW#+z)J}_P&2-1DomjZ4tDP;JhC2Yg2Pf16D5Yd3~q-&wyYBBj9HH{J=M2WHzrH
zp&PzzD@^-kW6R2Cqm(eNmHK09-4x;U2l2Yj*Kn2)*}zhjZs^0|!iFsx$v;Il^o9Pm
z3z-$;$Y<aVhm&wV$ixjUf__WF|3H`o^d+?^E3S1CODFYJ6)iXV)pciVe|6yV*W+t^
zu~pH4P0ng_{N>2GCSd~BYcv@DYR2)F735w7u#G^W4aHxIds(2-zjlG6K8W*qs}H!|
zr^f%mO(T=dqJY2fRgs;t&__uw4MO=f$o>HfuXdF~jNTI=n$xc_ANeCP|0}nB(rdn4
zWX}#Lh+-30a+?s!Fsp*yW0Nj`p2Ms}p$)Dhpwed`>_Jl)?|K)~BQCzzHeMhfKi0ib
zH6Wb*c*YRK$o2gRO?FRA1--up%;cKDv+aDPx`aW3ld0%_5T`MWM29N7!RAySJGADp
z`O7f3hFx{)HDp5gZXKnk8Z>3QUy0x&D3X<w4LHAWD{5cJSrfHr-|Bk3CdocrU{1;2
zULRTtvgUSn+E&=G#{DPtQJcz_jg;Nxj7rZC8<k5M%<$ki@rFuU?0_H@1&E^^Lw}d7
zt)gRE@^$EV5Y3|dxRpqG@ThLH<N9+F8Qvg!NPDy`%gcucvTV?f+ANIyJ=+aLH|j<e
z5tEys$RjAg0LXVlRy&C|WOUAoxb8Y{8*QM~Gg1+-oa-s|47Ywd5LKSRA`zU_D;Xi$
zenEC$0Pirl({RrT$Bt-pA@{lqSnbPM-JIdP8!k3djg=2fu+0dEw;uHW+-HpZeJWo+
zc5`oV0}ci<g=n;gl#5HJi8eoiP%kwBb~eM1tsq;70hj8)kga{z)`KSpA8vpZM~0hP
z@O^tKbr*yZ3;Rs`ss7MQE44IwLC&y+dA3!3FPEt-`xN?cSG7z$<E;8vBkwTQ3tPP3
z^ZBMm_ySosGx%C1340x{^kduT+r~-wC=TCK%@6U}O(X#UCH?50YL*>&o%=K8RD8&x
znka0N?)3Yere_<9m7Ix8uBt(5^&CQtb0e+g&DKjiBDV8J-Pr_>=QMe0P@3ig1|(E1
zLG9qBl()~pB!dqC1EVCXD<zcbY-C&xY<N5I`Cap*IHUT>hVirh*m^LeQZCHvC>a6T
zGsgzC=v1BS@AKdKXqNIwd@Q(x=#d(!MdsAB`pO7&se_hw<Ril(zf!rtv_kV^O&dj$
zU@b8W9M}M#E}^DG{VBSIyVrD9Sy5IMgPh!Cj;qRY)N}1KNEfy?1>a}y4q06t?+O%M
z9Y&mTqJctE`$`5RD$qR$O@km?6>pH0DOLz3MyC%W5?hKaqp5h5uE=MJ%B>Qf>M)$@
z(okx_;PRgddc4FZ;z)Ijklh|mpGA<#sV{dM!sX4yC-!}DX#3|15=vFVkhP6kxB?z>
zPCa4GgIK;>FlE($z+i?$x1KnE`(7M<7!1R&Z0lC9VUS-{{4ueZN_Y-9&!D+IkG=U+
zmP`zayAiV-N#k>f=VsV7&kSL$2&U-5>{56ccU3DXg}p%`b@{fFZb?#Yd4qq^H=XAr
z74tw<6bN3$))X7aOabiX*c+ZLQeGD1@SG1ABLSRoQ|?s#(*a~;7hN!|#b3_fwWdKb
zfL>#XKnHv_?bCL-<jxp6M$9V<tuzER)d(05*jyJDMwtLP4h&USp39c<EI|;D@Z=9?
zgz8ie(32<qrW5o@vG1pFV>W&B{PQGp{pvN6MC8q|keVPp!O8p8L^Ol6q|AscCJm>l
zIF_d9eYt6|2rtTp!tTSA%fehU!t!Y%`A*pWjaE{Kgq!F^F`1vdNn3{J3Se=rXTfG)
z)U@6To3?GRm7&d$kx8Sj#VIj*gbMO59hs8BNp6Pri8&EV^geS47n32vk@;?V04^_y
zv^&}IW?&z!2mPSA-1X|oV4mE%Lgb~XAr<?dPqmlJO)Bio8O;_Poyer&_Yjs$Ao9O_
z-nD8ajo?L^#~{y<Y^V5Kk3XM#`cemPCrxr(&v>W3JdvNxJTQ4e{bM+0A>|C>!t;2O
z&-3=4@Am!q+ohzeK?P963FuZS`XPyXyOb`0T}xLYf70Y3vjkqf6Z_(G8Y%#HBbm0Z
zj?g(?fm%7qYOv@rLDa<kgD6RKPn{)g=WzUl{5>&88e;Lo-1Nuv4wol=o<bhBMW}dP
zh=)qYMMM{*Fg+2Lh(T}^BtQCZ92M!FYFI)Rk~MPBV38<1a<dt7Y}?+qx5do2Zdy46
zjLvao5IfJNk~=5Yx<leVx2Q#0KAYBiumvG6DiF?WBOeuX<FJ=|Nwj+2N3`ZHb`K@r
zQa=TkyC$tNIkWw6E>UOd`CLMZ%s`e^_h<zDLXFOZop4D7@XCzHSw*xXCu0AMLC@u{
zzdl!qwt!!4?Qs3m=&AnLN(ND~^9M}4eA(M*2XPJ&dbTEuX4$XrRB1UbIErWnLm0gn
z+Kt=x9q^jfYR9d5YoOj^6)Rz;g}W7JmaLDHd8ml07;vT8!_LkmrhYXbZ^o{a9G=3|
z3fnU@#N4wy#1V1dmxUwd#_39dle1~I<BJBseiV$WVeC;A&0%!ycWuNV2(Tz*NUypC
zOEt+jnnmMEV)?EiSe$K``6!++8nDju9u4-p`g0fQ!_!4Xz5$Ehyw&p#!N~T@V~;_E
zv;Tv7WKK9};&JJd&dK4pHX|r~o%Nxo7rbSRB8*@y95)}UL!}~|nWEM$Ec{zn*DCaD
zgu^#BF4UDDue&QxpFNH0YY&qu>-d~?>s?OAKHu+q-Eu;hPjo!6zMy*s55&W7#&0HD
zzpGf?HieO09HuB3G8mvaMFv4!%N*Rm!4=Jo3NG*I1+$WZ-=gN0F#b3&fj(y?vAEwE
zkj0#iqo^LTi5t+*_C>@%qRL~xFBxB^i$`GobLSL~z16Kf8u@7gWcm0muQ3td&r#}7
z?4l`A{y>8PXzfXt#(v9hh~RrffhV%?$V0qeZj1?puw(-;q4^2*WzIL63PjcQ-@#{5
zis|1vAWawJcF#uht_R!EJP?wlM}Fg%ioFueLRcxYvk2py+XWQNjuyFLs_Ux|GV0B7
zl@8R6XP%kl?c9K*eh9!|?Y_@FD)qj6w!nql#gUm+7R;Ld2~tv)q=vu7%1q_pQV5j;
z@rWFc<wBb7@PZ#41lmD}<mBSnnf79u>d>+9#SXR8Evok0JTixMS{SYRk!&1EtS%ds
z_6x9__s{RAhsq2|ryu~p85jV71b_xHv@!T-Z)58~t7mKb`URCR8ytip69D{+|Ff4=
zg)ZwwI=D8(d0xzRR*6(5cu<kPNc~+s4cK0*@nk9TZcFjV`OJm6h+v$7J?CWEE#jxk
zOR=sajYY#xaFc8;L*zMj=~*-749+b>8ABB+AzUH3mU-AFc5wuBKO4?3$X(&a3<&+`
z4c_3W+=YLm{qog4uq>FEMx%V1rZ$)jUSoa#hiR+BeVwpmHfF>Q=qj|;mzlT|n2T_`
zrh*sJvQ(M-4F|VVDpi4?G8@#a$EYo)oFO^bY|$*o+~n`I6y!B#U2!~tqIG=~u#hd#
z+M0#gsnqBSNLp|QX>TM@7AZnf`Ys`R0Y0=VtRj^9A=@rFej(-ghwG*y@tAyY8s6|f
zGCkn~5bWXf6T4XrVv_g3g)pkyGwb>9v|=v~NUCM^a+LfmSOjK^gP=FyI=PFPwnU9o
znGVgGM5SpACm(;(W4K)_2uod;owvOaQPam!&`7}D>=%7g&4ikmVxi>;7iiTpKSCn8
zqm4j=gEDWn)}6<}lht+m@nU(GK%ZQ&N)$eA+9#ENKI6>C=}5ieaGmQN{sw2P<_lVe
z$KLUGgtinn7zF3Fz6tBixS#yeA$QD(+pX|^*p{vql^bX2nB#z@nu{#)f%6XMkeQ;L
z6M-KUJI4?g*Wvw-co_TsXEK+vlFz_Q|7S08<Yjm726g}n;HUV&USi<n;Amr|YGp~Q
zV{c^XKx^h`WJUklU1(n4{@*_H$5`P%-W3c=4+pu{vc?&^ls4vy6t(B7hTa;~xK`)L
zz}RP|kW6uZjV?<`bx03$_nf=T+g*B`xS+r^p~y@T8tzUY3dXy0GJ}p964sXXKHG<!
z;jm>pn|s|dkLqj-r6gO3g)S&JZT651E}l{XbJ#Eww=|X^f<w)vj(Fcm%f|89itk3R
zSkQ)$;2v`NE=l72g$3Tv<yQC@oo<K#`Ny&<g=V+MNNnRbmIw;SR$bSI<l-gRW{ax<
zuVvVO%K{wCQlSlC)0BW&fC-=g{FMc-$N$%``6~$svq=CHnS_4?;eYX(8YykjLkAad
z;`aoV(_(Q$oTRZ4TYsE6SJho!zDW`=k5JFZ(*OK`>%1S%86+y;uB~{)d+5F5^tGQ8
zVHg4;H|;#k4lnTFfYI1nKe{~1&n4OKqOyX3pqT6!JIPRrXIJt&`gXwJ6%m^US|2~O
zyL-1u74>O87oCS;0kg4avwFE?J-1CSYi6nQoz&Ju0n;&bQ1P5{{}+O~;htG2N`?Lj
z<kfl+v-KInI!QxeGx6!ztLssOkK=Bh4LXiECwAd15EnYbqf3yRErbKlC}LVsyv0c;
z+Hx<zR{ygX6@Akz4h4Sl1Ne*bkNEv>g|1q>`pp<EwS)|{GPKO7xHOHpGz?AAh<Ke+
zupON3bO`jdoIC*bmh>kmgpa<&8?+{Nw?1X@dI>C2QV8Yy`@TV}qNboe_dAFpH@>FY
z$Lyuvw}cH8TD-i})Qs#iq`XZ`tjrZctKT|1rFh>nz9(noZKJ3n%z0~$sT5jWGa2})
zKRf}LBmdIKr|@^W@}K~~EY#n~nX!$vqoSU^rP1pc(Hb{t{euoQ;Dr1Lk7zZ4?VSY(
zvmzeO_gpg!luv+7wqA8h(R#t5#glm%@8C&fGaTsmkMkYjycsX2x0giwrsK8i5<e-D
zbW!QWmAbZRM0r^bTCjpcNvjF#qNzloj3=hfpIQd2<vy56gFob{K_}-7{V+xz2vC+n
zVw8iAIW;J%B}8xE*P{CRyujk37!;t|zRlba$kum@f5mc-j-eEvS}hw~?<;<AGmt#9
zR+wAlY5>`AjNTx#mpvKGV6Zlljo|5P>sBjX4NKfrq%0EW3>p|DC|p(JdjxYWn=KNR
z8A8ih#I!KmSuV*?d)OtsEbta&BLLPo+om+6osvISX9B4RPEvO?eVV=dee1~?s#hnL
zhrH|1$HP=p;aa`xMkJ^v65qK%{+X4z3Y*5fnaq=*pXHU61E+Y86Uo$LlO~~d_pRDF
znI!D$KIe49(c0o!g{R3kLzjhoQ2cr;1GM74h!ks6C+AN6Szl<rmVTWVFj5g?1NHGI
zj(&!E%f?L})kJJ4x-Hb5fpgo&uYVl0BBL+(p|ORMkVin$kUQ9t1bWKOpu0`dh_%@$
zMX;y8R?I7EN+ATdf`w@mlMeHtzZq{tqAr>7NZ;Yh7+7QySll!e8O{uu0GiaW_l}aj
z#B+BrvK`iLs_J7*<23HM{g<8}8LM`-Z45LtCR*ojdKUdQ8ED2!R(p*wNxkT4bTo!w
z-<GY}@O7y=_-)%qH_wke;W}kuK^a;ajA)X5Us%9(5&%8#!^Mj|V6#0h*qcODta<#`
zQ#$3ZhQ$5A3^bqy08oKB`itGV80jk*IXVL2?Jw?dm@H`nJe~1A0lOvuUDh~wf=))K
z2!;o1E}L&&#v2e1U5M8!I;+;<?Kf+2Ue00%pYkdA=1v#Qt3V{2uJvfSKbU_gxV?y_
zR}d>TRDT*$tTp<rVbO&=UP@qOQCRz-J~he;>s#O5BCAP{BgKx+FlLrXqIWEdREoaH
zEWPs-6>ML#<+<&COZAujypt5JZi9}Y<klCvxir-Fy>r#klIGyWP=(^bw!~*<`A2&-
zwCy<B<K*hF&D&C1i^$!@%JSw@LMzv>+OC_B@*80+(9(jUoAcyWbAmp)JAsc1Xl;m|
zSh9Nb3!WkiGN^DSj#SLdNlbm0J;hg_c`DZ75u~CWTtA1Hqa1sY_tdJ7e2q)S8XKJl
z-qHU^sm>=~S{n>2`iP^arG+NpEGwR8*0V_Yz)NxU{@7?$k|4+Cxj1l7ljeTetC!b6
zT~{X4;Fg*A)QB%@7*@4MmB!-;M9CB5$rJl3CJB$l{@GF_FzhI!{4p|X;hV1JW&3R^
z<YqUFOdKC>pa-~U&$}}8aJ4Wt`OdB{cA>tWgyAH%rbDHSg?NFufSya$6f_qZIqFz}
zBQbtXaDxCfe!CiuZ|R)z&d~1*6(n+u;vf+=OMfb7u10|)z%bQDW3!f!*isnBF~xFn
z4Joh<#W{6jQ_FEiPas%<Lp5kJOoyY^_F$Xm)$PU=tV60-+Jgkt6ym)%4V=<sv>j&O
z5Ca)NfS7qxs<jmbjws^NkGA=6iwc>kjdjg2%M4@i1FN517+-w8Ig}F`O5P{k4SN<1
z1vGnG3a7PELXbDt`8x%FlqfbA>Y=i{hTk!8t(M^}0<w>birHzUK-an+?zxzMprPW2
z_mc!teu0dUU4&^}578v9d@}FBfzPDG48FJFx(>=VSfL@ovEgl4<<*V|YI4(hSpPZo
z$<>=jRz7Fk3;1Q_pZ{Lwzw5)@15sO&832I&pS=GAwW*jnnhF~m8yPtMN!<=!soQf0
z;{yWjlS^p|>7qj3rsT9?zqr<7(o>V5Tx4G|QC;i2_)6jUB*sBkmoOp!S8p*B)@9@T
zdO`q!EzSQxH=>iIT(!jBnw2%_j-4nrv3R{lwu8@}i3DlatyDLgc?6{NNmuRBr8gIg
zUeB8a2^Ul^*ij>X2pflwMmQtx2VAS^1Oz5s$(E0YqhxU3NhFo;QV17t$xD<lBR3OB
zwD1uh{CJ-0rDMDeu`;#p&T2BQTwQS58Vn`%JJm_|`p0Wn#qV8P&!!6WKY_Q|JaK9I
z%O!e!X@5YOlx8?UGxHdOawLHgN=dB&IU~pxE*{N>OsH#>@zodHz(tNwc-w^;&888!
zxXgrb1fSoCF6h3(8nh!#vfpM2WdJWLvmgRBSM&l}JnvNKq(U4@YEc`Mh+F9CqF#2*
zwtJESHwS$r5Z+}BZ6rk>9Y)Y!ozf$e^9JKFE9~=2G+CcAf(cIj!I9u2$wRjF9$ob_
zIJaz?`p;>fpUpQGUXt!BWNa^|nB+wrUv-6O{i;N%t8dT?^^?Ho47V}l(kCPKtcG~c
zWp54Ad)<Dvr9P-3`MQ=rjuOE?;m0EJ4%1doWL)8Q!+#tjKi^m;#UYq}uy1ewT4R3&
z_^I&9{fKxy{n5ugnGVXn=iF&EIjGe_CrspI;4kZ3X0FqRt3E1<w8vfO)_s@)7XY|7
znlQn<az7GRqTvVJ_QWf1^QzN}z(po0`L%MjlRvoMEsrJ*jFFKXy>_7Fc?EM2uZdw3
z<%?B#>bFmD(L`e=ZR>`?*%8&-EwmsD^46fO^Q(YM@*jY&U}BnGT_9-d)=Z_DzQJE!
zkR!o95Jp<!0V_vL;jV1YZG3D8*~}A>V`6E5>jS1)xd)p5qZT%^Rfyis`s&2`6VU3w
z`W)T*EY-X=`3n-={(>L9-^$vI6~d30{p>9()KF?!TX55=jJd+7nKaBSvywy%U`O9#
zYnTSFE)tzaZ4_U#2$-qLbqXemBK2dl*-b>Fy4{{yKTUf{JsthXv8&)X{9Nm<LuKER
zMRbNQQ*z^$JYiYMPbB!02&(3$Wz_k|<Q>VVG#|C_3m+rONdlAqVKvCUKUM|{dCktU
zh$r8ycr8iM4PQH57)LGeK4mu=yq%XVdbZRv?kGIv4v5h#*r_<A^z_inR4}tI(urAc
z*ZogHdLN`;*`I!;q{9xuAOqDO?9cy|{ZRth-xCqG0@|L;5*utWAmx3f?K%&{FV@%C
zcq8j%$>@n5jPM8h!<*;*G2r>c!8po%=<W(+e@;O5M<VfE2^ogKG&iuQ21*jT<vkl`
z)!}L|N&XnVX144J5wvcYi9x|w^ZZ#MSa1yo<{OTbC(YD2`K&gdp3qb@#V%|P-MUS6
zC_<C(ZN%g6mUDnla{cz5Vb0QVCyE~w$+y&g7pP`4mK6v&RwI0e)YC5^!o^t|)pcPx
z*~nexx|6x->@yBhs0Ho_Fvoa<=M8tQH|Az2RSRWCfQwg|YNjSa&%|kAoX){m18xrs
zf(Lw@*W>V7o3X~kX8tRcyKFQALbNmqsu;T|(2^S9JEO1I#&|`k`wZ`Q8jC;&r8@OR
zmVpHCBkL>wg9n1XI`R|a$m2ig8A(C$E3s12jK6;?Mhc{JCf2%^12>^UOFi1xwAP%2
zHn6BB5PQE!9Ucmhbm^gCNEYgfzoNeYAmnO(MSs?>=nv-={k?Ab5#z!r<^r_X9)XWm
zQ2K@bpkQ7}9Fnr#dVu=5NQIZvD@#<4lv#P@mPJU#Qu09H28?Dj(CmwnW~h%rmR*Zf
zF;Fv6?tlDtdgYc!gDqIOoWH;#<X{mH!U5Q){J`b^;B)4KIB91yfABeuKlq$LVe}5w
zU)tB-lYgPPEIt~>S8t7fMdbgN^5?C8D8FLDSFaUKd}dxFn+2?ezY<P+0FrxDNHQ=u
zAz1(4+Pmx;Y_#T|xtw(AKo=b#Hx&Jt*CsMaH28SBIID@?Yu*!&76__?24is$VvqDV
z2PcT1Z!O>*;$y+}Vt>#S((@H<?$~&bw>{Kt+PpLJ+*6^l^2u3A#3GPPiN_7BA!<5A
z`0^TlrPtr?r%r=ylL@6v{M5?YcjPwbK3Z74K15rmuR4J2u)Q$C!mkP0uJk+&+l9BB
znuuD#98b?{=JxJ(ea7G_O{U~#+vmzx_h?K%Z)%HIbzc@(Cm(D_?nm3TCR=pbJ}Boe
zb$6RFNomTU)n-J4tbCC#Xl_aMCsw_+&D&RT7nDE|f-7h`7|NmZe7g;|H6**R>vAJ4
z1mS7cLp441>^i?wx2BDiwv6&_Q5FymmP=sWooFbfdsxVJ*-mInj9$ybbHsR<`|Xwf
zvA><ah4>obqES)})mRPNX8E`pbIIK%CmJTY)S}SgF>tamc_Np$pHOWSxh?T0{OkG!
z|0Z7HU(Vp*<KN*Q<sb0RVKzhm75>rVQdt<W3FfTB-Gy?{rRvqJ$4)WCZX7JU)yy)E
znX-bR&3_zU{O;<>(b$?#rfwoR;^!R8SXA|D#!j5G9ok%zIvHWWzDvv7zRk2KR2cxk
z-6F}mHu>7oAP!C|=q=Xff0}9$uiyAe{{~*^UjvZ-#r>jxOGs=E?)2v4K>8P}b103?
zRoY9XK|r)*VXRGFfio9@jO&bCwF}kZ@zrehtecjl=Fs@A&qS;1MujAp=N^;t!qB1|
zI|?mqZDKMrLb6*zC4^)=CamdmFl&A&p11ByS7n6bTjQXCK4X4)&zB)xVhrJry(GRs
z_Q&`K`?Kn^!q<Htw6y)o{z%nOkz4Nsai;SKgkk<<f81(H??1`6EDyZ0KN;1XKiFUC
zFZTERi~aF!6;%Ut$UAxHapI#4KBBOzVnL((jDXTq34Juw^}e~32APpcd!rDXI+UMW
zV*p_z&XtcQJ2XzACt70mPBMyb|HR;y7DiqDT&bGmgn(zobR+i*{?n|S1T&1fnr+VA
z?)_!-l{sB^Nv@b+Bk!A&hRk?V;x-8s^AdBpKHWgB3d*wb-t*2r-8@`gv=mf-v@a1=
zWAcu88}!8o>dAf~>w}nknP3<_X!>O~lQykgV+9}lm0>n1j-nT<>d52gz=f^h^7`tU
z5lWe3eq6=Vp<NAkJV{s)SY)p~B)eRU0u;rB6ZTgEo3{qYy6qgdeHUyC?A0T_B>tG_
zHJs)2QK|dL<KHmFC(_B&1Km}5Q#L;w-oB|Y^ORK^kBlw_a2h@}^P296WCFok^Zb@+
z>_j~@-LM~dyNj{a_!FQyM3<}xb3V6$uVi;YqzU9RmM^>5i&meYNee@1hCT$}dC1=7
ztI^LU!Q5j?)F`uBnZ-~R;R@sIo!M-oGmNx{MA#=oZb(>(Gr00<CeG%r?Q7fP4Hi8X
zM5ju4x@W$n)izVt8<Nb1wo=9|sFSkR|1Jcgt;XCpz;@ZA#m}gxicL->vX3K2be-^C
zp4Cm44jI>kJF`IL47C4!oi?3i0@v4019W4n=!rCp)(<nW`rb^}ff4u4+@InU(!@*h
zEf~zq#az~SDCwvVdn`AZ-5%3Bd3doj&xQB8+;wBRUlLe);iDRt$F@X23#J6uvxj@y
zOlqe~t%lJg-Z&6Io%hTE?{gZJur>xKolfNHEwZ+qHcV1L&;Uc_KXy+ppqgEj><m9}
zs3pU2(hWM{XR$$gD@?Mgmzt}7UgoxSj0{uO@kuC~O>^`bwAB)ZI*&7(@h-F7OB#Or
zwCzCUg8!*&fu{17ay~F`(bE;VjDDo>C-8bG(C@l%#beu3w3L2@RtuJ>-GFI-*;pqN
zR2g<cp7vsRPwmCsflw_Ii>@Z#AK&j<xD)N=33-HzRn`i%?z{IkUX*;O5g21pDHL*2
zB6}^^wxfPJYN%?UbQI`JqS0QY3?YavTPuYbH`>gQ2!x57zW3Ve@w#9S>u2YN?IhUU
z+;tJ85+W%$0R#qqq<Unr6rK$%U5stRS!yl&l=8jjZKbnso?@B*v}Sa;9GuKi@-d1&
zG)qXiTbg40R^NgxSF>?j@-Ol?{7U{vu|6nariQx~>lV|ehToNk7xkMmtfN~N)AJw?
z$Vf5kCd3!e=VQm5W$NWlO;+mV7WCNlh>b&@k_k?ynBhi}bb{7q4K1kks6zPx(O+Xn
z!aKNztRd5(?})bPTJ;*rr>^qiHaPNV)L`4P&6X^1F5*S1k@4q<Gz%vgUW#}%*uB<M
zyf3;mFFOSIpK6Jh=6rFbQ9B(?j$VvScwI-Kdlg%%9-f#h!}O`)CK)aPQzKpS^tp0Q
z>$FT_YwLaT^NTf<t<>haS%-0Y>$Fm)HF-u*-SMCc=?g1et0kFU=Ah4Xf%umv&!Zh*
zGB&D=`Wj~%jn%)UxUiddPhHz5$!}K~VBCK7hO(V7yoL`a{w^jwr?=t5k`K8bk59^2
zIQ7c@)HU36b6(ls^eg*=53_p{*B`5YH(kTbo$C7;yz;@j;Pdp4)9nwTYqap18IxWk
zLf#`K8knT=1rfg21K}X8**0EbW*G#~48q~-w%?Jm0t!x74%;Yl><GZ#K?{wDYjL?;
zib+!CzH_{jgZ4u3RgV*#S3f1!d^cu<(c6pimicQ?^)bbCR|~wgmF>=3ytU~c?97OQ
zu!nC9$qthPuPQIGm^?qad!F<Z>ATtym&)#sy`ebi=-2aZUpFkneUrz6sA@GjVvpFd
zr<c?s$c{O>J`OJ3M0u9_aKk5sdJExG*m41O{3yDe`XG($FG(;K(yP`TK4EuI>LF+0
z)l8BFyp7cLHSNFFD#^~TrEG!4?0R6e2>I{u&%nn1SE=&<k_=$nfQx^$j)fl`YTqi<
z146Wmw7!Oh5OryeL8X9Rcv#z)t$O$=3UT#Cn-0^GgXd2Cww<jVc2@D91J$PpnPH`h
z!KQqXP{Aq!#a6-E=Xtz>=~YrN%8GB~9YOtmE*wdBA0(<^&o&qHYryn7kQnJ}7nqt@
z;Tc%oB1jm6?tGM4QfxMy{k%_Ij`gCe(fRHjGEEml7gZ1SL<JO4Ga&>)<drrC9-l27
zL~Leh7gOf#umyj@&vCFc`)#T`?wRr_(qm?g(;BwM2oZ$-d^2C;=|0=L_nsQu+%8I=
zi+*U?@u?BHjxf_{>>W-@=Phg4RL#A8>p@y)L6ulV*>DyHZs#(DkBJKN%xzPKjZocI
z6e>fOW`cQ{Sr<R#`%OT=geU2qX~YaQB0Dki#^*<h<~M8dom@h$IR>7STorB6kOz+-
zTv`$yn9pyw9dRQHOteq-lbbsbI?uoF=xoe?u|)Y!&<mV)nbpxzdx-L%!VU;1HL!U3
zpXn%A5|z5mz(gH}d?g3JFB*ct^16|=qmH7Rt&xM)YmNP*<bMm`M`1Ty0PSRenf_Y0
z|7*Db){_2!c)!@)?}oV^R|X-#J?{a_`M7^s0oO2qTYi;R|F8yvb0;gH6R&&!S0ydF
z{jV>%K$Y+Cz|QvfWees{rN3+Z?g*V!Z;iwo0ASu606_b@HQ;T<pXOFZhGu%SW>$J8
zM$9xuR>r?eyp>HG5@H1a;I=@1lZa^grv%f#Nh~{>Fna-G2n}@kzrBDw`=<osze!v%
zDb#NO=M-_>{Vvf6JVL)B`nQHK{F{V*BEvT{;LIdkk>49K3JLNL3HpDNs0vcNC<SIy
z9?tJxpi%uP;b>}PWkmby^%tA^J)9`rq+mV3SkwZK%74e=lHs3Tu>6}B3bZmb*ucYg
zzYO+UPV6)PDdF(9s{QXC{60(Tue4y?{?q)w=4}0L|NG>GzwABs|Fir5nX2%+@qb@A
z@|P*_u-t(Fe!q_7ckBNq>VLNucm(+yS^YP@{=3$HbJ@RZi2-r`{~M+KUFE;&<lj}k
zzWlE8ulevh`TI-43IO^?R{j?U{M~`yxy@hpC!l}Y|3-0sSNOdG{+GfI#GeZPS`YtS
q=fBSj{!#$|u+TvNI<)^eOZX@W0qjWt06Ore9EhYkF@U`S@c#e?GKuQ|
--
GitLab