cute_little_R_functions.R 700 KB
Newer Older
Gael  MILLOT's avatar
Gael MILLOT committed
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
# 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("\n\n================\n\nERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
# 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
2033
for(i2 in 1:length(tempo.names)){
Gael  MILLOT's avatar
Gael MILLOT committed
2034
count <- 0
2035
tempo <- tempo.names[i2]
Gael  MILLOT's avatar
Gael MILLOT committed
2036
2037
while(any(tempo %in% data2) | any(tempo %in% data1)){
count <- count + 1
2038
tempo <- paste0(tempo.names[i2], "_modif", count)
Gael  MILLOT's avatar
Gael MILLOT committed
2039
}
2040
data1[data1 %in% tempo.names[i2]] <- paste0(tempo.names[i2], "_modif", count)
Gael  MILLOT's avatar
Gael MILLOT committed
2041
if(count != 0){
2042
2043
ini <- c(ini, tempo.names[i2])
post <- c(post, paste0(tempo.names[i2], "_modif", count))
Gael  MILLOT's avatar
Gael MILLOT committed
2044
2045
2046
2047
2048
2049
2050
2051
}
}
data <- data1
}else{
data <- data1
}
output <- list(data = data, ini = ini, post = post)
return(output)
Gael  MILLOT's avatar
Gael MILLOT committed
2052
2053
2054
}


Gael  MILLOT's avatar
Gael MILLOT committed
2055
######## fun_df_remod() #### remodeling a data frame to have column name as a qualitative values and vice-versa
Gael  MILLOT's avatar
Gael MILLOT committed
2056
2057


Gael  MILLOT's avatar
Gael MILLOT committed
2058
2059
2060
2061
# Check OK: clear to go Apollo
fun_df_remod <- function(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
Gael  MILLOT's avatar
Gael MILLOT committed
2062

Gael  MILLOT's avatar
Gael MILLOT committed
2063
2064
 
# 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
Gael  MILLOT's avatar
Gael MILLOT committed
2065

Gael  MILLOT's avatar
Gael MILLOT committed
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
 

# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# 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
# EXAMPLES
# obs <- data.frame(col1 = (1:4)*10, col2 = c("A", "B", "A", "A")) ; obs ; fun_df_remod(obs)
# obs <- data.frame(col1 = (1:4)*10, col2 = 5:8) ; obs ; fun_df_remod(obs, quanti.col.name = "quanti", quali.col.name = "quali")
# obs <- data.frame(col1 = (1:4)*10, col2 = 5:8) ; 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) ; quanti.col.name = "quanti" ; quali.col.name = "quali" # for function debugging
# data = data.frame(a = 1:3, b = 4:6, c = 11:13) ; quanti.col.name = "quanti" ; quali.col.name = "quali" # for function debugging
# data = data.frame(a = 1:3, b = letters[1:3]) ; quanti.col.name = "quanti" ; quali.col.name = "quali" # for function debugging
# data = data.frame(a = 1:3, b = letters[1:3]) ; quanti.col.name = "TEST" ; quali.col.name = "quali" # for function debugging
# data = data.frame(b = letters[1:3], a = 1:3) ; quanti.col.name = "quanti" ; quali.col.name = "quali" # for function debugging
# data = data.frame(b = c("e", "e", "h"), a = 1:3) ; 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("\n\n================\n\nERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
# end required function checking
# argument checking
# argument checking without fun_check()
if( ! any(class(data) %in% "data.frame")){
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": THE data ARGUMENT MUST BE A DATA FRAME\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
# 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("\n\n================\n\nERROR 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\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
if((mode(data[, 1]) == "character" | mode(data[, 2]) == "character") & (quanti.col.name != "quanti" | quali.col.name != "quali")){
tempo.cat <- paste0("\n\n================\n\nERROR 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\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
}else{
if( ! all(tempo.factor %in% "numeric")){
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": IF data ARGUMENT IS A DATA FRAME MADE OF ONE COLUMN, OR MORE THAN 2 COLUMNS, THESE COLUMNS MUST BE NUMERIC\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
}
if(( ! any(tempo.factor %in% "character")) & is.null(names(data))){
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": NUMERIC DATA FRAME in the data ARGUMENT MUST HAVE COLUMN NAMES\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
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)
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)
}
}else{ # transfo 2
if(class(data[, 1]) == "character"){
data <- cbind(data[2], data[1])
}
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)
}
return(output.data)
}




######## fun_round() #### rounding number if decimal present


# 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$double.eps^0.5
# ARGUMENTS
# data: a vector of numbers (numeric or character mode)
# dec.nb: number of required decimal digits
# after.lead.zero: logical. If FALSE, rounding is performed for all the decimal numbers, whatever the leading zeros (e.g., 0.123 -> 0.12 and 0.00128 -> 0.00). If TRUE, dec.nb are taken after the leading zeros (e.g., 0.123 -> 0.12 and 0.00128 -> 0.0013)
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# RETURN
# the modified vector
# EXAMPLES
# ini.options <- options()$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()$digits ; options(digits = 8) ; cat(fun_round(data = c(NA, "10", "100.001", "333.0001254", "12312.1235"), dec.nb = 2, after.lead.zero = TRUE), "\n\n") ; options(digits = ini.options)
# DEBUGGING
# data = data = c(10, 100.001, 333.0001254, 12312.1235) ; dec.nb = 2 ; after.lead.zero = FALSE # # for function debugging
# data = data = c("10", "100.001", "333.0001254", "12312.1235") ; dec.nb = 2 ; after.lead.zero = TRUE # # for function debugging
# function name
function.name <- paste0(as.list(match.call(expand.dots=FALSE))[[1]], "()")
# end function name
# required function checking
if(length(utils::find("fun_check", mode = "function")) == 0){
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
# end required function checking
# argument checking
# argument checking without fun_check()
if( ! (all(typeof(data) == "character") | all(typeof(data) == "double") | all(typeof(data) == "integer"))){
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": data ARGUMENT MUST BE A VECTOR OF NUMBERS (IN NUMERIC OR CHARACTER MODE)\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
# end argument checking without fun_check()
# argument checking with fun_check()
arg.check <- NULL #
text.check <- NULL #
checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
ee <- expression(arg.check <- c(arg.check, tempo$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)
}


######## fun_mat_rotate() #### 90° clockwise matrix rotation


# 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
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# ARGUMENTS
# data: matrix (matrix class)
# RETURN
# the modified matrix
# 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("\n\n================\n\nERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
# 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)
}


######## fun_mat_num2color() #### convert a numeric matrix into hexadecimal color matrix


# Check OK: clear to go Apollo
fun_mat_num2color <- function(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
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# 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
# $text.problem: text when overlapping colors. NULL if forced.color = NULL or problem == FALSE
# EXAMPLES
# mat1 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2) ; dimnames(mat1) <- list(LETTERS[1:4], letters[1:2]) ; fun_mat_num2color(mat1, mat.hsv.h = FALSE, notch = 1, s = 1, v = 1, forced.color = NULL)
# DEBUGGING
# mat1 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2) ; dimnames(mat1) <- list(LETTERS[1:4], letters[1:2]); mat.hsv.h = FALSE ; notch = 1 ; s = 1 ; v = 1 ; forced.color = c(hsv(1,1,1), hsv(0,0,0)) # for function debugging
# function name
function.name <- paste0(as.list(match.call(expand.dots=FALSE))[[1]], "()")
# end function name
# required function checking
if(length(utils::find("fun_check", mode = "function")) == 0){
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
# end required function checking
# argument checking
# argument checking with fun_check()
arg.check <- NULL #
text.check <- NULL #
checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
ee <- expression(arg.check <- c(arg.check, tempo$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, print = FALSE)$problem == TRUE){
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": mat1 ARGUMENT MUST BE A MATRIX OF PROPORTIONS SINCE THE mat.hsv.h ARGUMENT IS SET TO TRUE\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
if( ! is.null(forced.color)){
tempo <- fun_check(data = forced.color, class = "character")
if(any(tempo$problem == TRUE)){
paste0("\n\n================\n\n", paste(tempo$text[tempo$problem], collapse = "\n"), "\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
if( ! all(forced.color %in% colors() | grepl(pattern = "^#", forced.color))){ # check that all strings of forced.color start by #
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": forced.color ARGUMENT MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # AND/OR COLOR NAMES GIVEN BY colors()\n\n================\n\n")
stop(tempo.cat, call. = 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
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("\n\n================\n\nERROR IN ", function.name, ": mat1 MUST BE MADE OF VALUES BETWEEN 0 AND 1 BECAUSE mat.hsv.h ARGUMENT SET TO TRUE\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
}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("\n\n================\n\nERROR IN ", function.name, ": mat1 MUST BE MADE OF INTEGER VALUES WITHOUT 0 BECAUSE mat.hsv.h ARGUMENT SET TO FALSE\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}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("\n\n================\n\nERROR IN ", function.name, ": DUPLICATED VALUES AFTER USING notch (", paste(tempo.different.color[duplicated(tempo.different.color)], collapse = " "), "). TRY ANOTHER notch VALUE\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}else if(length(different.color) != length(tempo.different.color)){
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": LENGTH OF different.color (", paste(different.color, collapse = " "), ") DIFFERENT FROM LENGTH OF tempo.different.color (", paste(tempo.different.color, collapse = " "), ")\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}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("\n\n================\n\nERROR 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 = " "), "\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}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)
}


######## fun_mat_op() #### assemble several matrices with operation


# 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
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# fun_comp_2d()
# 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
# 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("\n\n================\n\nERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
if(length(utils::find("fun_check", mode = "function")) == 0){
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": REQUIRED fun_comp_2d() FUNCTION IS MISSING IN THE R ENVIRONMENT\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
# 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$fun.name))
tempo <- fun_check(data = mat.list, class = "list", fun.name = function.name) ; eval(ee)
tempo <- fun_check(data = kind.of.operation, options = c("+", "-", "*"), length = 1, fun.name = function.name) ; eval(ee)
if(any(arg.check) == TRUE){
stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
}
# end argument checking with fun_check()
# argument checking without fun_check()
if(length(mat.list) < 2){
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": mat.list ARGUMENT MUST BE A LIST CONTAINING AT LEAST 2 MATRICES\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
2495
2496
for(i1 in 1:length(mat.list)){
tempo <- fun_check(data = mat.list[[i1]], class = "matrix", mode = "numeric", na.contain = TRUE)
Gael  MILLOT's avatar
Gael MILLOT committed
2497
if(tempo$problem == TRUE){
2498
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": ELEMENT ", i1, " OF mat.list ARGUMENT MUST BE A NUMERIC MATRIX\n\n================\n\n")
Gael  MILLOT's avatar
Gael MILLOT committed
2499
2500
2501
2502
2503
stop(tempo.cat, call. = FALSE)
}
}
ident.row.names <- TRUE
ident.col.names <- TRUE
2504
2505
for(i1 in 2:length(mat.list)){
tempo <- fun_comp_2d(data1 = mat.list[[1]], data2 = mat.list[[i1]])
Gael  MILLOT's avatar
Gael MILLOT committed
2506
if(tempo$same.dim == FALSE){
2507
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": MATRIX ", i1, " OF mat.list ARGUMENT MUST HAVE THE SAME DIMENSION (", paste(dim(mat.list[[i1]]), collapse = " "), ") THAN THE MATRIX 1 IN mat.list (", paste(dim(mat.list[[1]]), collapse = " "), ")\n\n================\n\n")
Gael  MILLOT's avatar
Gael MILLOT committed
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
stop(tempo.cat, call. = FALSE)
}
if( ! is.null(tempo$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]]
2526
2527
for(i1 in 2:length(mat.list)){
output <- get(kind.of.operation)(output, mat.list[[i1]])
Gael  MILLOT's avatar
Gael MILLOT committed
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
}
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)
}


######## fun_mat_inv() #### return the inverse of a square matrix


# Check OK: clear to go Apollo
fun_mat_inv <- function(mat){
# AIM
# return the inverse of a square matrix when solve() cannot
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# ARGUMENTS:
# mat: a square numeric matrix without NULL, NA, Inf or single case (dimension 1, 1) of 0
# RETURN
# the inversed matrix
# 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("\n\n================\n\nERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
# 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$fun.name))
tempo <- fun_check(data = mat, class = "matrix", mode = "numeric", fun.name = function.name) ; eval(ee)
if(any(arg.check) == TRUE){
stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
}
# end argument checking with fun_check()
# argument checking without fun_check()
if(ncol(mat) != nrow(mat)){
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": mat ARGUMENT MUST BE A SQUARE MATRIX\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
if(any(mat %in% c(Inf, -Inf, NA))){
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": mat ARGUMENT MUST BE A MATRIX WITHOUT Inf, -Inf OR NA\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
if(all(mat == 0) & ncol(mat) == 1){
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": mat ARGUMENT CANNOT BE A SQUARE MATRIX MADE OF A SINGLE CASE OF 0\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
# end argument checking without fun_check()
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.2/r_debugging_tools-v1.2.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
# end argument checking
# main code
if(any(grepl(x = try(solve(mat), silent = TRUE)[], pattern = "[Ee]rror"))){
tempo <- svd(mat)
val.critique <- which(tempo$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))
}
}


######## fun_mat_fill() #### fill the empty half part of a symmetric square matrix


# 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
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# 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
# 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("\n\n================\n\nERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
# 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("\n\n================\n\nERROR IN ", function.name, ": mat ARGUMENT MUST BE A SQUARE MATRIX\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
if( ! (mode(mat) %in% c("numeric", "character"))){
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": mat ARGUMENT MUST BE A NUMERIC OR CHARACTER MATRIX\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
if(nrow(mat) == 1 & ncol(mat) == 1){
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": mat ARGUMENT CANNOT BE A SQUARE MATRIX MADE OF A SINGLE CASE\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
if(ifelse(is.na(empty.cell.string), ! any(is.na(mat)), ! any(mat == empty.cell.string, na.rm = TRUE))){
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": mat ARGUMENT MATRIX MUST HAVE CELLS WITH THE EMPTY STRING SPECIFIED IN empty.cell.string ARGUMENT\n\n================\n\n")
stop(tempo.cat, call. = 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
list.diag <- vector("list", length = nrow(mat) - 1)
2689
2690
for(i1 in 1:(nrow(mat) - 1)){
list.diag[[i1]] <- numeric(length = nrow(mat) - i1)
Gael  MILLOT's avatar
Gael MILLOT committed
2691
2692
2693
}
sector <- c("topleft", "topright", "bottomright", "bottomleft")
diag.scan <-c( # same order as sector. Recover each diag from center to corner
2694
2695
2696
2697
"mat[as.matrix(as.data.frame(list(1:(nrow(mat) - i2), (ncol(mat) -i2):1)))]", # topleft part
"mat[as.matrix(as.data.frame(list(1:(nrow(mat) - i2), (1:ncol(mat))[-(1:i2)])))]", # topright part
"mat[as.matrix(as.data.frame(list((1 + i2):nrow(mat), ncol(mat):(1 + i2))))]", # bottomright part
"mat[as.matrix(as.data.frame(list((1 + i2):nrow(mat), 1:(ncol(mat) -i2))))]" # bottomleft part
Gael  MILLOT's avatar
Gael MILLOT committed
2698
2699
2700
2701
2702
2703
)
# empty part detection
tempo.list.diag <- list.diag
empty.sector <- NULL
full.sector <- NULL
warn <- NULL
Gael  MILLOT's avatar
Gael MILLOT committed
2704
warn.count <- 0
2705
for(i1 in 1:length(sector)){
Gael  MILLOT's avatar
Gael MILLOT committed
2706
tempo.list.diag <- list.diag
2707
2708
2709
2710
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])
Gael  MILLOT's avatar
Gael MILLOT committed
2711
2712
2713
2714
2715
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)){
2716
empty.sector <- c(empty.sector, sector[i1])
Gael  MILLOT's avatar
Gael MILLOT committed
2717
warn.count <- warn.count + 1
2718
tempo.warn <- paste0("(", warn.count,") FROM FUNCTION ", function.name, ": EMPTY SECTOR DETECTED ON THE ", toupper(sector[i1]), " CORNER, FULL OF ", empty.cell.string)
Gael  MILLOT's avatar
Gael MILLOT committed
2719
2720
warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
}else{
2721
tempo.cat <- paste0("\n\n================\n\nERROR 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 = " "), "\n\n================\n\n")
Gael  MILLOT's avatar
Gael MILLOT committed
2722
2723
2724
2725
2726
2727
stop(tempo.cat, call. = FALSE)
}
}
}
# end empty part detection
if(length(empty.sector) == 0){
Gael  MILLOT's avatar
Gael MILLOT committed
2728
2729
warn.count <- warn.count + 1
tempo.warn <- paste0("(", warn.count,") FROM FUNCTION ", function.name, ": ACCORDING TO empty.cell.string ARGUMENT (", empty.cell.string, "), mat ARGUMENT MATRIX HAS ZERO EMPTY HALF PART")
Gael  MILLOT's avatar
Gael MILLOT committed
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
}else{
if(length(empty.sector) > 1){
tempo.cat <- paste0("\n\n================\n\nERROR 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 = " "), "\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}else if(any(full.sector %in% empty.sector, na.rm = TRUE)){
tempo.cat <- paste0("\n\n================\n\nERROR 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 = " "), "\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}else if(length(empty.sector) + length(full.sector)!= 4){
tempo.cat <- paste0("\n\n================\n\nERROR 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 = " "), "\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}else{
Gael  MILLOT's avatar
Gael MILLOT committed
2742
2743
warn.count <- warn.count + 1
tempo.warn <- paste0("(", warn.count,") FROM FUNCTION ", function.name, ": ", toupper(empty.sector), " SECTOR HAS BEEN COMPLETED TO BECOME SYMMETRICAL")
Gael  MILLOT's avatar
Gael MILLOT committed
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
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)){
Gael  MILLOT's avatar
Gael MILLOT committed
2761
warning(warn, call. = FALSE)
Gael  MILLOT's avatar
Gael MILLOT committed
2762
2763
2764
2765
2766
2767
2768
2769
}
return(list(mat = mat, warn = warn))
}


######## fun_permut() #### progressively breaks a vector order


Gael  MILLOT's avatar
Gael MILLOT committed
2770
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){
Gael  MILLOT's avatar
Gael MILLOT committed
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
# 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.
# 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 correaltion 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
Gael  MILLOT's avatar
Gael MILLOT committed
2784
2785
# 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
Gael  MILLOT's avatar
Gael MILLOT committed
2786
2787
2788
# 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
2789
# lib.path: character vector specifying the absolute pathways of the directories containing the required packages if not in the default directories. Ignored if NULL
Gael  MILLOT's avatar
Gael MILLOT committed
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
# REQUIRED PACKAGES
# lubridate
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# fun_pack()
# fun_round()
# 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
# 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
Gael  MILLOT's avatar
Gael MILLOT committed
2807
2808
# 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)
Gael  MILLOT's avatar
Gael MILLOT committed
2809
# a way to use the cor.limit argument just considering data1
Gael  MILLOT's avatar
Gael MILLOT committed
2810
2811
2812
2813
# 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)
Gael  MILLOT's avatar
Gael MILLOT committed
2814
# DEBUGGING
Gael  MILLOT's avatar
Gael MILLOT committed
2815
2816
2817
2818
# 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
Gael  MILLOT's avatar
Gael MILLOT committed
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
# 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("\n\n================\n\nERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
if(length(utils::find("fun_pack", mode = "function")) == 0){
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": REQUIRED fun_pack() FUNCTION IS MISSING IN THE R ENVIRONMENT\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
if(length(utils::find("fun_round", mode = "function")) == 0){
tempo.cat <- paste0("\n\n================\n\nERROR IN ", function.name, ": REQUIRED fun_pack() FUNCTION IS MISSING IN THE R ENVIRONMENT\n\n================\n\n")
stop(tempo.cat, call. = FALSE)
}
# 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$problem == FALSE & length(data1) < 2){
tempo.cat <- paste0("ERROR IN ", function.name, ": data1 ARGUMENT MUST BE A VECTOR OF MINIMUM LENGTH 2. HERE IT IS: ", length(data1),"\n\n================\n\n")
text.check <- c(text.check, tempo.cat)
arg.check <- c(arg.check, TRUE)
}
if( ! is.null(data2)){
tempo <- fun_check(data = data1, class = "vector", mode = "numeric", fun.name = function.name) ; eval(ee)
if(tempo$problem == TRUE){
tempo.cat <- paste0("ERROR IN ", function.name, ": data1 MUST BE A NUMERIC VECTOR IF data2 ARGUMENT IS SPECIFIED\n\n================\n\n")
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\n\n================\n\n")
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 = FALSE, fun.name = function.name) ; eval(ee)
}
Gael  MILLOT's avatar
Gael MILLOT committed
2871
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)
Gael  MILLOT's avatar
Gael MILLOT committed
2872
2873
2874
2875
2876
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)){
Gael  MILLOT's avatar
Gael MILLOT committed
2877
2878
2879
2880
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"))
Gael  MILLOT's avatar
Gael MILLOT committed
2881
2882
2883
2884
text.check <- c(text.check, tempo.cat)
arg.check <- c(arg.check, TRUE)
}
}
Gael  MILLOT's avatar
Gael MILLOT committed
2885
}
Gael  MILLOT's avatar
Gael MILLOT committed
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
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
}
# end code that protects set.seed() in the global environment
if( ! is.null(seed)){
set.seed(seed)
}
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
warn <- NULL
Gael  MILLOT's avatar
Gael MILLOT committed
2914
warn.count <- 0
Gael  MILLOT's avatar
Gael MILLOT committed
2915
2916
2917
2918
2919
2920
count <- 0
round <- 0
BREAK <- FALSE
tempo.cor <- 0
if(is.null(data2)){
if(length(table(data1)) == 1){
Gael  MILLOT's avatar
Gael MILLOT committed
2921
2922
warn.count <- warn.count + 1
tempo.warn <- paste0("(", warn.count,") FROM FUNCTION ", function.name, ": NO PERMUTATION PERFORMED BECAUSE data1 ARGUMENT SEEMS TO BE MADE OF IDENTICAL ELEMENTS: ", names(table(data1)))
Gael  MILLOT's avatar
Gael MILLOT committed
2923
2924
warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn))) #
}else{
Gael  MILLOT's avatar
Gael MILLOT committed
2925
2926
if(print.count > n){
print.count <- n
Gael  MILLOT's avatar
Gael MILLOT committed
2927
2928
}
cat(paste0("\n", ifelse(text.print == "", "", paste0(text.print, " | ")), "FOR LOOP OF ", n, " LOOPS INITIATED | LOOP COUNT: ", format(count, big.mark=",")))
Gael  MILLOT's avatar
Gael MILLOT committed
2929
2930
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
Gael  MILLOT's avatar
Gael MILLOT committed
2931
count.loop <- 0
Gael  MILLOT's avatar
Gael MILLOT committed
2932
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)]
Gael  MILLOT's avatar
Gael MILLOT committed
2933
2934
2935
2936
2937
2938
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)]
Gael  MILLOT's avatar
Gael MILLOT committed
2939
if(print.count.loop[count.loop]){
Gael  MILLOT's avatar
Gael MILLOT committed
2940
count.loop <- 0
Gael  MILLOT's avatar
Gael MILLOT committed
2941
pos <- sample.int(n = pos.selec.seq.max , size = print.count, replace = TRUE) # BEWARE: never forget to resample here
Gael  MILLOT's avatar
Gael MILLOT committed
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
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
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){
Gael  MILLOT's avatar
Gael MILLOT committed
2955
2956
warn.count <- warn.count + 1
tempo.warn <- paste0("(", warn.count,") FROM FUNCTION ", function.name, ": NO PERMUTATION PERFORMED BECAUSE data1 ARGUMENT SEEMS TO BE MADE OF IDENTICAL ELEMENTS: ", names(table(data1)))
Gael  MILLOT's avatar
Gael MILLOT committed
2957
2958
2959
warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn))) #
tempo.cor <- 1
}else if(length(table(data2)) == 1){
Gael  MILLOT's avatar
Gael MILLOT committed
2960
2961
warn.count <- warn.count + 1
tempo.warn <- paste0("(", warn.count,") FROM FUNCTION ", function.name, ": NO PERMUTATION PERFORMED BECAUSE data2 ARGUMENT SEEMS TO BE MADE OF IDENTICAL ELEMENTS: ", names(table(data2)))
Gael  MILLOT's avatar
Gael MILLOT committed
2962
2963
2964
2965
2966
2967
2968
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){
Gael  MILLOT's avatar
Gael MILLOT committed
2969
2970
warn.count <- warn.count + 1
tempo.warn <- paste0("(", warn.count,") FROM FUNCTION ", function.name, ": 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")
Gael  MILLOT's avatar
Gael MILLOT committed
2971
2972
2973
2974
2975
2976
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
Gael  MILLOT's avatar
Gael MILLOT committed
2977
2978
warn.count <- warn.count + 1
tempo.warn <- paste0("(", warn.count,") FROM FUNCTION ", function.name, ": 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")
Gael  MILLOT's avatar
Gael MILLOT committed
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
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)))
Gael  MILLOT's avatar
Gael MILLOT committed
2995
2996
print.count.loop <- logical(length = print.count)
print.count.loop[length(print.count.loop)] <- TRUE # counter to speedup
Gael  MILLOT's avatar
Gael MILLOT committed
2997
count.loop <- 0 # 
Gael  MILLOT's avatar
Gael MILLOT committed
2998
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)]
Gael  MILLOT's avatar
Gael MILLOT committed
2999
3000
tempo.date.loop <- Sys.time()
tempo.time.loop <- as.numeric(tempo.date.loop)
For faster browsing, not all history is shown. View entire blame