cute_little_R_functions.R 895 KB
Newer Older
Gael  MILLOT's avatar
Gael MILLOT committed
1
2
################################################################
##                                                            ##
Gael's avatar
Gael committed
3
##     CUTE FUNCTIONS                                         ##
Gael  MILLOT's avatar
Gael MILLOT committed
4
5
6
7
8
##                                                            ##
##     Gael A. Millot                                         ##
##                                                            ##
##                                                            ##
################################################################
Gael  MILLOT's avatar
Gael MILLOT committed
9
10


11
# color palette: see https://github.com/EmilHvitfeldt/r-color-palettes
Gael  MILLOT's avatar
Gael MILLOT committed
12
# https://usethis.r-lib.org/ and usethat also
13
14
15
16
17
18
19
20
21
22
23
24
25
26
## LAST ROUND OF FORMATTING:
# 1) Arguments: one per line
# 2) Description:
# first capital letter everywhere
# AIM
# WARNINGS
# ARGUMENTS
# Update all argument description, saying, character vector, etc, as in fun_gg_boxplot
# RETURN
# Update as in fun_gg_boxplot
# REQUIRED PACKAGES
# REQUIRED FUNCTIONS FROM THE cute PACKAGE
# EXAMPLE
# one example of the example cheet
27
# see http
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
# DEBUGGING
# one updated
# 3) Categ (see fun_gg_boxplot for inside code)
# function name
# must have arg.user.setting
# end function name

# required function checking
# see boxplot
# end required function checking

# reserved words (to avoid bugs)
# end reserved words (to avoid bugs)

# arg with no default values
# see boxplot
# end arg with no default values

# argument primary checking
# end argument primary checking

# second round of checking and data preparation
# management of NA arguments
# must have arg.user.setting
# end management of NA arguments

# management of NULL arguments
# end management of NULL arguments

# code that protects set.seed() in the global environment
# end code that protects set.seed() in the global environment

# warning initiation
# end warning initiation

# other checkings
# end other checkings

# reserved word checking
# end reserved word checking
# end second round of checking and data preparation

# package checking
# end package checking
# main code
# output
# if(warn.print == TRUE & ! is.null(warn)){
# end output
# end main code
# 4) example sheet as in fun_gg_boxplot
# 5) test the function with debugging_tools_for_r_dev
# 6) use fun_test()
# 7) write at the beginning of the function:
# todo list check OK
# Check r_debugging_tools-v1.4.R OK
# Check fun_test() (see cute_checks.docx) OK
84
85
# example sheet OK 
# check all and any OK # in if() notably, that does not like NA as a logical result
86
87
88
# clear to go Apollo
# 8) check all(, na.rm = TRUE) and any(, na.rm = TRUE)

89
# fun_mat_fill does not recognize half matrix anymore
90
# package: 
Gael  MILLOT's avatar
Gael MILLOT committed
91
# Templates: https://prettydoc.statr.me/themes.html
92
# http://r-pkgs.had.co.nz/
Gael  MILLOT's avatar
Gael MILLOT committed
93
94
95
# https://pkgdown.r-lib.org/
# https://rdrr.io/github/gastonstat/cointoss/
# doc:https://www.sphinx-doc.org/en/master/man/sphinx-autogen.html considering that https://www.ericholscher.com/blog/2014/feb/11/sphinx-isnt-just-for-python/
Gael  MILLOT's avatar
Gael MILLOT committed
96
97
98
# https://docs.readthedocs.io/en/stable/intro/getting-started-with-sphinx.html
# https://docs.gitlab.com/ee/user/project/pages/
# also register into biotools
99
100
101

# change everywhere: if( ! is.null(arg.check)){

Gael  MILLOT's avatar
Gael MILLOT committed
102
# For heatmap: see https://bioinfo-fr.net/creer-des-heatmaps-a-partir-de-grosses-matrices-en-r
Gael  MILLOT's avatar
Gael MILLOT committed
103

104
105
106
107
# New function: fun_pdf_conc for Rosine code?
# https://stackoverflow.com/questions/17552917/merging-existing-pdf-files-using-r
# https://www.r-bloggers.com/2019/04/join-split-and-compress-pdf-files-with-pdftools/
# https://rdrr.io/cran/staplr/man/staple_pdf.html
Gael  MILLOT's avatar
Gael MILLOT committed
108

Gael  MILLOT's avatar
Gael MILLOT committed
109
################################ OUTLINE ################################
Gael  MILLOT's avatar
Gael MILLOT committed
110
111


Gael  MILLOT's avatar
Gael MILLOT committed
112
113
################ Object analysis    2
######## fun_check() #### check class, type, length, etc., of objects   2
Gael's avatar
Gael committed
114
115
116
117
118
######## fun_secu() #### verif that local variables are not present in other envs   11
######## fun_info() #### recover object information 13
######## fun_head() #### head of the left or right of big 2D objects    15
######## fun_tail() #### tail of the left or right of big 2D objects    16
######## fun_comp_1d() #### comparison of two 1D datasets (vectors, factors, 1D tables) 17
119
######## fun_comp_2d() #### comparison of two 2D datasets (row & col names, dimensions, etc.)   22
Gael's avatar
Gael committed
120
######## fun_comp_list() #### comparison of two lists   29
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
######## fun_test() #### test combinations of argument values of a function and return errors (and graphs)  32
################ Object modification    47
######## fun_name_change() #### check a vector of character strings and modify any string if present in another vector  47
######## fun_df_remod() #### remodeling a data frame to have column name as a qualitative values and vice-versa 48
######## fun_round() #### rounding number if decimal present    51
######## fun_mat_rotate() #### 90° clockwise matrix rotation    53
######## fun_mat_num2color() #### convert a numeric matrix into hexadecimal color matrix    54
######## fun_mat_op() #### assemble several matrices with operation 58
######## fun_mat_inv() #### return the inverse of a square matrix   60
######## fun_mat_fill() #### fill the empty half part of a symmetric square matrix  62
######## fun_permut() #### progressively breaks a vector order  65
######## fun_slide() #### return a computation made on a vector using a sliding window  76
################ Graphics management    85
######## fun_width() #### window width depending on classes to plot 85
######## fun_open() #### open a GUI or pdf graphic window   87
######## fun_prior_plot() #### set graph param before plotting (erase axes for instance)    91
######## fun_scale() #### select nice label numbers when setting number of ticks on an axis 95
######## fun_inter_ticks() #### define coordinates of secondary ticks   100
######## fun_post_plot() #### set graph param after plotting (axes redesign for instance)   105
######## fun_close() #### close specific graphic windows    117
################ Standard graphics  118
######## fun_empty_graph() #### text to display for empty graphs    118
################ gg graphics    120
######## fun_gg_palette() #### ggplot2 default color palette    120
######## fun_gg_just() #### ggplot2 justification of the axis labeling, depending on angle  122
######## fun_gg_get_legend() #### get the legend of ggplot objects  127
######## fun_gg_point_rast() #### ggplot2 raster scatterplot layer  129
######## fun_gg_boxplot() #### ggplot2 boxplot + background dots if required    133
######## fun_gg_scatter() #### ggplot2 scatterplot + lines (up to 6 overlays totally)   133
######## fun_gg_heatmap() #### ggplot2 heatmap + overlaid mask if required  133
######## fun_gg_empty_graph() #### text to display for empty graphs 141
################ Graphic extraction 143
######## fun_trim() #### display values from a quantitative variable and trim according to defined cut-offs 143
######## fun_segmentation() #### segment a dot cloud on a scatterplot and define the dots from another cloud outside the segmentation   152
################ Import 187
######## fun_pack() #### check if R packages are present and import into the working environment    187
######## fun_python_pack() #### check if python packages are present    189
################ Print / Exporting results (text & tables)  192
######## fun_report() #### print string or data object into output file 192
######## fun_get_message() #### return error/warning/other messages of an expression (that can be exported) 195
Gael  MILLOT's avatar
Gael MILLOT committed
161
162


Gael  MILLOT's avatar
Gael MILLOT committed
163
################################ FUNCTIONS ################################
Gael  MILLOT's avatar
Gael MILLOT committed
164

Gael  MILLOT's avatar
Gael MILLOT committed
165
166
167
168
169
170
171

################ Object analysis


######## fun_check() #### check class, type, length, etc., of objects


172
# todo list check OK
Gael's avatar
tempo    
Gael committed
173
# Check r_debugging_tools-v1.4.R OK
174
# Check fun_test() (see cute_checks.docx) OK
175
176
177
# example sheet OK 
# check all and any OK
# -> clear to go Apollo
Gael's avatar
tempo    
Gael committed
178
fun_check <- function(
179
180
181
182
183
184
185
186
187
188
189
190
191
192
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
Gael's avatar
tempo    
Gael committed
193
){
194
195
196
197
198
199
200
# 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
201
# The function tests what is written in its arguments, even if what is written is incoherent. For instance, fun_check(data = factor(1), class = "factor", mode = "character") will return a problem, whatever the object tested in the data argument, because no object can be class "factor" and mode "character" (factors are class "factor" and mode "numeric"). Of note, length of object of class "environment" is always 0
202
# If the tested object is NULL, then the function will always return a checking problem
203
204
205
206
207
208
209
210
# 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.
211
212
# double.as.integer.allowed: logical. If TRUE, no error is reported in the cheking message 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 the data argument, or NULL
213
214
# all.options.in.data: logical. If TRUE, all of the options must be present at least once in the data argument, and nothing else. If FALSE, some or all of the options must be present in the data argument, and nothing else. Ignored if options is NULL
# na.contain: logical. Can the data argument contain NA?
215
216
# neg.values: logical. Are negative numeric values authorized? Warning: the default setting is TRUE, meaning that, in that case, no check is performed for the presence of negative values. The neg.values argument is activated only when set to FALSE. In addition, neg.values = FALSE can only be used when class, typeof or mode arguments are not NULL, otherwise return an error message
# print: logical. Print the message if $problem is TRUE? Warning: set by default to FALSE, which facilitates the control of the checking message output when using fun_check() inside functions. See the example section
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
# data.name: character string indicating the name of the object to test. If NULL, use what is assigned to the data argument for the returned message
# fun.name: character string indicating the name of the function checked (i.e., when fun_check() is used to check the arguments of this function). If non-null, the value of fun.name will be added into the message returned by fun_check()
# RETURN
# A list containing:
# $problem: logical. Is there any problem detected?
# $text: message indicating the details of the problem, or the absence of problem
# $object.name: value of the data.name argument (i.e., name of the checked object if provided, NULL otherwise)
# REQUIRED PACKAGES
# None
# REQUIRED FUNCTIONS FROM THE cute PACKAGE
# None
# EXAMPLE
# test <- matrix(1:3) ; fun_check(data = test, print = TRUE, class = "vector", mode = "numeric")
# see http
# DEBUGGING
232
# data = mean ; class = NULL ; typeof = NULL ; mode = NULL ; length = NULL ; prop = FALSE ; double.as.integer.allowed = FALSE ; options = "a" ; all.options.in.data = FALSE ; na.contain = FALSE ; neg.values = TRUE ; print = TRUE ; data.name = NULL ; fun.name = NULL
233
234
235
236
237
238
239
240
# function name
# no used in this function for the error message, to avoid env colliding
# end function name
# required function checking
# end required function checking
# reserved words
# end reserved words
# fun.name checked first because required next
241
242
243
244
245
246
if( ! is.null(fun.name)){ # I have to use this way to deal with every kind of class for fun.name
if(all(base::class(fun.name) == "character")){ # all() without na.rm -> ok because class(NA) is "logical"
if(base::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 ==
}else if(any(is.na(fun.name))){ # normally no NA with is.na()
247
248
249
tempo.cat <- paste0("ERROR IN fun_check(): NO ARGUMENT EXCEPT data AND options CAN HAVE NA VALUES\nPROBLEMATIC ARGUMENT IS fun.name")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
250
251
}else{
tempo.cat <- paste0("ERROR IN fun_check(): THE fun.name ARGUMENT MUST BE A CHARACTER VECTOR OF LENGTH 1") # paste(fun.name, collapse = " ") removed here because does not work with objects like function
252
253
254
255
256
257
258
259
260
261
262
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
# argument primary checking
Gael's avatar
tempo    
Gael committed
263
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.7/r_debugging_tools-v1.7.R") ; eval(parse(text = str_basic_arg_check_dev)) # activate this line and use the function to check arguments status
264
265
# end argument primary checking
# second round of checking and data preparation
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
# management of special classes
basic.class <- c(
"NULL", # because class(NULL) is "NULL". The NULL aspect will be dealt later
"logical", 
"integer", 
"numeric", 
# "complex", 
"character"
# "matrix", 
# "array", 
# "data.frame", 
# "list", 
# "factor", 
# "table", 
# "expression", 
# "name", 
# "symbol", 
# "function", 
# "uneval", 
# "environment", 
# "ggplot2", 
# "ggplot_built", 
# "call"
)
tempo.arg.base <-c( # no names(formals(fun = sys.function(sys.parent(n = 2)))) used with fun_check() to be sure to deal with the correct environment
"class", 
"typeof", 
"mode", 
"length", 
"prop", 
"double.as.integer.allowed", 
"options", 
"all.options.in.data", 
"na.contain", 
"neg.values", 
"print", 
"data.name", 
"fun.name"
)
tempo.class <-list( # no get() used to be sure to deal with the correct environment
base::class(class), 
base::class(typeof), 
base::class(mode), 
base::class(length), 
base::class(prop), 
base::class(double.as.integer.allowed), 
base::class(options), 
base::class(all.options.in.data), 
base::class(na.contain), 
base::class(neg.values), 
base::class(print), 
base::class(data.name), 
base::class(fun.name)
)
# tempo.cat1 <- NULL
# tempo.cat2 <- NULL
tempo <- ! sapply(lapply(tempo.class, FUN = "%in%", basic.class), FUN = all)
# for(i1 in tempo.arg.base){
# tempo.class <- base::class(get(i1, env = sys.nframe(), inherit = FALSE))
# if( ! all(tempo.class %in% basic.classes)){
# tempo.cat1 <- c(tempo.cat1, i1)
# tempo.cat2 <- c(tempo.cat2, paste0(tempo.class, collapse = " "))
# }
# }
# if( ! is.null(tempo.cat1)){
if(any(tempo)){
tempo.cat1 <- tempo.arg.base[tempo]
tempo.cat2 <- sapply(tempo.class[tempo], FUN = paste0, collapse = " ")
tempo.sep <- sapply(mapply(" ", max(nchar(tempo.cat1)) - nchar(tempo.cat1) + 3, FUN = rep, SIMPLIFY = FALSE), FUN = paste0, collapse = "")
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" INSIDE ", fun.name)), ": ANY ARGUMENT EXCEPT data MUST HAVE A BASIC CLASS\nPROBLEMATIC ARGUMENT", ifelse(base::length(tempo.cat1) > 1, "S", ""), " AND ASSOCIATED CLASS", ifelse(base::length(tempo.cat1) > 1, "ES ARE", " IS"), ":\n", paste0(tempo.cat1, tempo.sep, tempo.cat2, collapse = "\n")) # normally no NA with is.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 management of special classes
339
# management of NA arguments
340
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))){ # normally no NA with is.na()
341
342
tempo <- 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)))]
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" INSIDE ", fun.name)), ": NO ARGUMENT EXCEPT data AND options CAN HAVE NA VALUES\nPROBLEMATIC ARGUMENT", ifelse(length(tempo) > 1, "S ARE", " IS"), ":\n", paste(tempo, collapse = "\n")) # normally no NA with is.na()
343
344
345
346
347
348
349
350
351
352
353
354
355
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 management of NA arguments
# management of NULL arguments
tempo.arg <-c(
"prop", 
"double.as.integer.allowed", 
"all.options.in.data", 
"na.contain",
"neg.values",
"print"
)
tempo.log <- sapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = is.null)
356
if(any(tempo.log) == TRUE){ # normally no NA with is.null()
357
tempo.cat <- paste0("ERROR IN fun.check():\n", ifelse(sum(tempo.log, na.rm = TRUE) > 1, "THESE ARGUMENTS", "THIS ARGUMENT"), " CANNOT BE NULL:\n", paste0(tempo.arg[tempo.log], collapse = "\n"))
358
359
360
361
362
363
364
365
366
367
368
369
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 management of NULL arguments
# dealing with logical
# tested below
# end dealing with logical
# code that protects set.seed() in the global environment
# end code that protects set.seed() in the global environment
# warning initiation
# end warning initiation
# other checkings
if( ! is.null(data.name)){
370
if( ! (base::length(data.name) == 1 & all(base::class(data.name) == "character"))){ # all() without na.rm -> ok because class(NA) is "logical"
371
372
373
374
375
376
377
378
379
380
381
382
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" INSIDE ", 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(" INSIDE ", 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(" INSIDE ", 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 ==
}
383
if( ! (all(base::class(neg.values) == "logical") & base::length(neg.values) == 1 & any(is.na(neg.values)) != TRUE)){ # all() without na.rm -> ok because class(NA) is "logical"  # normally no NA with is.na()
384
385
386
387
388
389
390
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" INSIDE ", 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(" INSIDE ", 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 ==
}
391
if( ! is.null(class)){ # may add "formula" and "Date" as in https://renenyffenegger.ch/notes/development/languages/R/functions/class
392
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", "call") & any(is.na(class)) != TRUE & base::length(class) == 1)){ # length == 1 here because of class(matrix()) since R4.0.0  # all() without na.rm -> ok because class cannot be NA (tested above) # normally no NA with is.na()
393
394
395
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" INSIDE ", 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 ==
}
396
if(neg.values == FALSE & ! any(class %in% c("vector", "numeric", "integer", "matrix", "array", "data.frame", "table"))){ # no need of na.rm = TRUE for any() because %in% does not output NA
397
398
399
400
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" INSIDE ", 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 ==
}
}
401
if( ! is.null(typeof)){ # all the types are here: https://renenyffenegger.ch/notes/development/languages/R/functions/typeof
402
if( ! (all(typeof %in% c("logical", "integer", "double", "complex", "character", "list", "expression", "symbol", "closure", "special", "builtin", "environment", "S4", "language")) & base::length(typeof) == 1 & any(is.na(typeof)) != TRUE)){ # "language" is the type of object of class "call" # all() without na.rm -> ok because typeof cannot be NA (tested above) # normally no NA with is.na()
403
404
405
406
407
408
409
410
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" INSIDE ", 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(" INSIDE ", 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 ==
}
}
411
if( ! is.null(mode)){ # all the types are here: https://renenyffenegger.ch/notes/development/languages/R/functions/typeof
412
if( ! (all(mode %in% c("logical", "numeric", "complex", "character", "list", "expression", "name", "symbol", "function", "environment", "S4", "call")) & base::length(mode) == 1 & any(is.na(mode)) != TRUE)){ # all() without na.rm -> ok because mode cannot be NA (tested above) # normally no NA with is.na()
413
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" INSIDE ", fun.name)), ": mode ARGUMENT MUST BE ONE OF THESE VALUE:\n\"logical\", \"numeric\", \"complex\", \"character\", \"list\", \"expression\", \"name\", \"symbol\", \"function\", \"environment\", \"S4\", , \"language\"")
414
415
416
417
418
419
420
421
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(" INSIDE ", 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)){
422
if( ! (is.numeric(length) & base::length(length) == 1 & ! grepl(length, pattern = "\\.") & any(is.na(length)) != TRUE)){ # normally no NA with is.na()
423
424
425
426
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" INSIDE ", 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 ==
}
}
427
if( ! (is.logical(prop) | (base::length(prop) == 1 & any(is.na(prop)) != TRUE))){ # normally no NA with is.na()
428
429
430
431
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" INSIDE ", 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)){
432
if( ! any(class %in% c("vector", "numeric", "matrix", "array", "data.frame", "table"))){ # no need of na.rm = TRUE for any() because %in% does not output NA
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" INSIDE ", 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(" INSIDE ", 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(" INSIDE ", 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 ==
}
}
}
450
if( ! (all(base::class(double.as.integer.allowed) == "logical") & base::length(double.as.integer.allowed) == 1 & any(is.na(double.as.integer.allowed)) != TRUE)){ # all() without na.rm -> ok because class() never returns NA # normally no NA with is.na()
451
452
453
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" INSIDE ", 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 ==
}
454
if( ! (is.logical(all.options.in.data) & base::length(all.options.in.data) == 1 & any(is.na(all.options.in.data)) != TRUE)){# normally no NA with is.na()
455
456
457
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" INSIDE ", 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 ==
}
458
if( ! (all(base::class(na.contain) == "logical") & base::length(na.contain) == 1 & any(is.na(na.contain)) != TRUE)){ # all() without na.rm -> ok because class() never returns NA # normally no NA with is.na()
459
460
461
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 ==
}
462
if( ! (all(base::class(print) == "logical") & base::length(print) == 1 & any(is.na(print)) != TRUE)){ # all() without na.rm -> ok because class() never returns NA # normally no NA with is.na()
463
464
465
466
467
468
469
470
471
472
473
474
475
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
# end other checkings
# end second round of checking and data preparation
# package checking
# end package checking
# main code
if(is.null(data.name)){
data.name <- deparse(substitute(data))
}
problem <- FALSE
476
text <- paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " OBJECT")
477
if(( ! is.null(options)) & all(base::typeof(data) == "character")){ # all() without na.rm -> ok because typeof() never returns NA
478
text <- ""
479
if( ! all(data %in% options)){ # no need of na.rm = TRUE for all() because %in% does not output NA
480
problem <- TRUE
481
text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": THE ", data.name, " OBJECT MUST BE SOME OF THESE OPTIONS: ", paste(options, collapse = " "), "\nTHE PROBLEMATIC ELEMENTS OF ", data.name, " ARE: ", paste(unique(data[ ! (data %in% options)]), collapse = " "))
482
483
}
if(all.options.in.data == TRUE){
484
if( ! all(options %in% data)){ # no need of na.rm = TRUE for all() because %in% does not output NA
485
problem <- TRUE
486
text <- paste0(ifelse(text == "", "", paste0(text, "\n")), ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": THE ", data.name, " OBJECT 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 = " "))
487
488
489
}
}
if( ! is.null(length)){
490
if(base::length(data) != length){
491
problem <- TRUE
492
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 ", base::length(data))
493
494
495
}
}
if(text == ""){
496
text <- paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " OBJECT")
497
}
498
499
500
}else if( ! is.null(options)){
problem <- TRUE
text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": THE ", data.name, " OBJECT MUST BE SOME OF THESE OPTIONS: ", paste(options, collapse = " "), "\nBUT IS NOT EVEN TYPE CHARACTER")
501
502
503
504
505
}
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")
506
}else if(class == "factor" & all(base::class(data) %in% c("factor", "ordered"))){ # to deal with ordered factors # all() without na.rm -> ok because class(NA) is "logical"
507
508
509
510
class <- c("factor", "ordered")
}
}
if(is.null(options)){
511
for(i2 in 1:base::length(arg.names)){
512
513
514
if( ! is.null(get(arg.names[i2], env = sys.nframe(), inherit = FALSE))){
# script to execute
tempo.script <- '
Gael  MILLOT's avatar
Gael MILLOT committed
515
problem <- TRUE ;
516
517
if(identical(text, paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " OBJECT"))){
text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": THE ", data.name, " OBJECT MUST BE ") ;
Gael  MILLOT's avatar
Gael MILLOT committed
518
}else{
Gael  MILLOT's avatar
Gael MILLOT committed
519
text <- paste0(text, " AND ") ; 
Gael  MILLOT's avatar
Gael MILLOT committed
520
}
521
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)})
522
' # no need of na.rm = TRUE for all() because %in% does not output NA
523
# end script to execute
524
if(base::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")))){ # no need of na.rm = TRUE for all() because == does not output NA if no NA in left of ==, which is the case for arg.names
525
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. Warning: na.rm = TRUE required here for all()
526
527
eval(parse(text = tempo.script)) # execute tempo.script
}
528
}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 # no need of na.rm = TRUE for all() because %in% does not output NA # no need of na.rm = TRUE for any() because get get(arg.names) does not contain NA
529
eval(parse(text = tempo.script)) # execute tempo.script
530
}else if(arg.names[i2] == "class" & all(get(arg.names[i2], env = sys.nframe(), inherit = FALSE) == "vector") & ! (all(base::class(data) %in% "numeric") | all(base::class(data) %in% "integer") | all(base::class(data) %in% "character") | all(base::class(data) %in% "logical"))){ # test class == "vector". No need of na.rm = TRUE for all() because %in% does not output NA # no need of na.rm = TRUE for all() because == does not output NA if no NA in left of ==, which is the case for arg.names
531
eval(parse(text = tempo.script)) # execute tempo.script
532
}else if(arg.names[i2] == "class" & all(get(arg.names[i2], env = sys.nframe(), inherit = FALSE) == "ggplot2") & ! all(base::class(data) %in% c("gg", "ggplot"))){ # test ggplot object # no need of na.rm = TRUE for all() because == does not output NA if no NA in left of ==, which is the case for arg.names # no need of na.rm = TRUE for all() because %in% does not output NA
533
534
535
536
537
eval(parse(text = tempo.script)) # execute tempo.script
}
}
}
}
538
if(prop == TRUE & all(base::typeof(data) == "double")){ # all() without na.rm -> ok because typeof(NA) is "logical"
539
if(is.null(data) | any(data < 0 | data > 1, na.rm = TRUE)){ # works if data is NULL # Warning: na.rm = TRUE required here for any()
540
problem <- TRUE
541
if(identical(text, paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " OBJECT"))){
542
543
544
545
text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": ")
}else{
text <- paste0(text, " AND ")
}
546
text <- paste0(text, "THE ", data.name, " OBJECT MUST BE DECIMAL VALUES BETWEEN 0 AND 1")
547
}
548
549
550
551
552
553
554
555
}else if(prop == TRUE){
problem <- TRUE
if(identical(text, paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " OBJECT"))){
text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": ")
}else{
text <- paste0(text, " AND ")
}
text <- paste0(text, "THE ", data.name, " OBJECT MUST BE DECIMAL VALUES BETWEEN 0 AND 1")
556
}
557
if(all(base::class(data) %in% "expression")){ # no need of na.rm = TRUE for all() because %in% does not output NA
558
data <- as.character(data) # to evaluate the presence of NA
Gael  MILLOT's avatar
Gael MILLOT committed
559
}
560
if(na.contain == FALSE & (base::mode(data) %in% c("logical", "numeric", "complex", "character", "list", "expression", "name", "symbol"))){ # before it was ! (class(data) %in% c("function", "environment"))
561
if(any(is.na(data)) == TRUE){ # not on the same line because when data is class envir or function , do not like that # normally no NA with is.na()
562
problem <- TRUE
563
if(identical(text, paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " OBJECT"))){
564
565
566
567
text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": ")
}else{
text <- paste0(text, " AND ")
}
568
text <- paste0(text, "THE ", data.name, " OBJECT CONTAINS NA WHILE NOT AUTHORIZED")
569
}
Gael  MILLOT's avatar
Gael MILLOT committed
570
}
571
if(neg.values == FALSE & all(base::mode(data) %in% "numeric")){ # no need of na.rm = TRUE for all() because %in% does not output NA
572
if(any(data < 0, na.rm = TRUE)){ # Warning: na.rm = TRUE required here for any()
573
problem <- TRUE
574
if(identical(text, paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " OBJECT"))){
575
576
577
578
text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": ")
}else{
text <- paste0(text, " AND ")
}
579
text <- paste0(text, "THE ", data.name, " OBJECT MUST BE MADE OF NON NEGATIVE NUMERIC VALUES")
580
581
582
583
584
585
586
}
}else if(neg.values == FALSE){
problem <- TRUE
if(identical(text, paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " OBJECT"))){
text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": ")
}else{
text <- paste0(text, " AND ")
587
}
588
text <- paste0(text, "THE ", data.name, " OBJECT MUST BE NON NEGATIVE BUT IS NOT EVEN MODE NUMERIC")
Gael  MILLOT's avatar
Gael MILLOT committed
589
}
Gael  MILLOT's avatar
Gael MILLOT committed
590
if(print == TRUE & problem == TRUE){
591
cat(paste0("\n\n================\n\n", text, "\n\n================\n\n"))
Gael  MILLOT's avatar
Gael MILLOT committed
592
}
593
594
# output
output <- list(problem = problem, text = text, object.name = data.name)
Gael  MILLOT's avatar
Gael MILLOT committed
595
return(output)
596
597
# end output
# end main code
Gael  MILLOT's avatar
Gael MILLOT committed
598
}
Gael  MILLOT's avatar
Gael MILLOT committed
599

600
######## fun_secu() #### verif that local variables are not present in other envs
Gael  MILLOT's avatar
Gael MILLOT committed
601
602
603


fun_secu <- function(pos = 1, name = NULL){
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
# AIM
# Verify 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 the working env will be checked if they also exist in the above envs (Scope of R). If fun_secu() is used in a function, with pos ==1, variables presents in the local env of the function will be checked if they also exist in the above envs (Scope of R, which includes the working environment (Global env))
# name: single character string indicating the name of the function checked. If NULL, fun_secu() checks all the variables of the environment indicated by pos, as explained in the pos argument description. If non-null, fun_secu() checks all the variables presents in the local env of the function will be checked in the above envs (which includes the working environment (Global env)
# 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]], "()")
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){
628
tempo.cat <- paste0("ERROR IN ", function.name, "\nREQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
629
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
Gael's avatar
tempo    
Gael committed
630
}
631
632
633
634
635
636
637
638
639
640
641
642
# end required function checking
# argument primary checking
# arg with no default values
# 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$object.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)
Gael's avatar
tempo    
Gael committed
643
}
644
645
646
647
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()
Gael's avatar
tempo    
Gael committed
648
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.7/r_debugging_tools-v1.7.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()
649
650
651
652
653
654
# end argument primary checking
# second round of checking and data preparation
# management of NA arguments
tempo.arg <- names(arg.user.setting) # values provided by the user
tempo.log <- suppressWarnings(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){
655
tempo.cat <- paste0("ERROR IN ", function.name, "\n", ifelse(sum(tempo.log, na.rm = TRUE) > 1, "THESE ARGUMENTS", "THIS ARGUMENT"), " CANNOT JUST BE NA:", paste0(tempo.arg[tempo.log], collapse = "\n"))
656
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
Gael's avatar
tempo    
Gael committed
657
}
658
659
660
661
662
663
664
# end management of NA arguments
# management of NULL arguments
tempo.arg <- c(
"pos"
)
tempo.log <- sapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = is.null)
if(any(tempo.log) == TRUE){
665
tempo.cat <- paste0("ERROR IN ", function.name, "\n", ifelse(sum(tempo.log, na.rm = TRUE) > 1, "THESE ARGUMENTS", "THIS ARGUMENT"), " CANNOT BE NULL:", paste0(tempo.arg[tempo.log], collapse = "\n"))
666
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
Gael's avatar
tempo    
Gael committed
667
}
668
669
670
671
672
673
674
675
676
677
678
679
680
# end management of NULL arguments
# end second round of checking and data preparation
# 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))]
Gael's avatar
tempo    
Gael committed
681
}
682
683
684
685
686
687
688
689
690
691
692
693
# 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)])
Gael's avatar
tempo    
Gael committed
694
}
Gael  MILLOT's avatar
Gael MILLOT committed
695
}
696
697
698
699
700
701
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)
Gael  MILLOT's avatar
Gael MILLOT committed
702
703
704
}


705
######## fun_info() #### broad description of an object
Gael  MILLOT's avatar
Gael MILLOT committed
706
707


708
709
# todo list check OK
# Check r_debugging_tools-v1.4.R OK
710
711
712
713
# Check fun_test() (see cute_checks.docx) OK
# example sheet OK 
# check all and any OK
# -> clear to go Apollo
714
715
716
717
fun_info <- function(
data, 
n = NULL, 
warn.print = TRUE
Gael's avatar
tempo    
Gael committed
718
){
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
# AIM
# Provide a broad description of an object
# WARNINGS
# None
# ARGUMENTS
# data: object to analyse
# n: positive integer value indicating the n first number of elements to display per compartment of the output list (i.e., head(..., n)). Write NULL to return all the elements. Does not apply for the $STRUCTURE compartment output
# warn.print: logical. Print potential warnings at the end of the execution? If FALSE the warning messages are added in the output list as an additional compartment (or NULL if no message).
# RETURN
# A list containing information, depending on the class and type of data. The backbone is generally:
# $NAME: the name of the object
# $CLASS: class of the object (class() value)
# $TYPE: type of the object (typeof() value)
# $LENGTH: length of the object (length() value)
# $NA.NB: number of NA and NaN (only for type "logical", "integer", "double", "complex", "character" or "list")
# $HEAD: head of the object (head() value)
# $TAIL: tail of the object (tail() value)
# $DIMENSION: dimension (only for object with dimensions)
# $SUMMARY: object summary (summary() value)
# $STRUCTURE: object structure (str() value)
# $WARNING: warning messages (only if the warn.print argument is FALSE)
# If data is made of numerics, provide also:
# $INF.NB: number of Inf and -Inf
# $RANGE: range after removing Inf and NA
# $SUM: sum after removing Inf and NA
# $MEAN: mean after removing Inf and NA
# If data is a 2D object, provide also:
# $ROW_NAMES: row names
# $COL_NAMES: column names
# If data is a data frame, provide also:
# $COLUMN_TYPE: type of each column (typeof() value)
# If data is a list, provide also:
# $COMPARTMENT_NAMES: names of the comprtments
# $COMPARTMENT_TYPE: type of each compartment (typeof() value)
# REQUIRED PACKAGES
# None
755
# REQUIRED FUNCTIONS FROM THE cute PACKAGE
756
757
# fun_check()
# fun_get_message()
758
# EXAMPLE
759
760
761
# fun_info(data = 1:3)
# see http
# DEBUGGING
762
# mat1 <- matrix(1:3) ; data = env1 ; n = NULL ; warn.print = TRUE # for function debugging
763
764
765
766
767
768
769
770
771
772
773
774
775
# 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"
)
tempo <- NULL
for(i1 in req.function){
if(length(find(i1, mode = "function")) == 0){
tempo <- c(tempo, i1)
Gael  MILLOT's avatar
Gael MILLOT committed
776
}
Gael  MILLOT's avatar
Gael MILLOT committed
777
}
778
if( ! is.null(tempo)){
779
tempo.cat <- paste0("ERROR IN ", function.name, "\nREQUIRED cute FUNCTION", ifelse(length(tempo) > 1, "S ARE", " IS"), " MISSING IN THE R ENVIRONMENT:\n", paste0(tempo, collapse = "()\n"))
780
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
Gael's avatar
tempo    
Gael committed
781
}
782
783
784
785
786
787
788
789
# end required function checking
# reserved words
# end reserved words
# arg with no default values
mandat.args <- c(
"data"
)
tempo <- eval(parse(text = paste0("missing(", paste0(mandat.args, collapse = ") | missing("), ")")))
790
if(any(tempo)){ # normally no NA for missing() output
791
tempo.cat <- paste0("ERROR IN ", function.name, "\nFOLLOWING ARGUMENT", ifelse(length(mandat.args) > 1, "S HAVE", "HAS"), " NO DEFAULT VALUE AND REQUIRE ONE:\n", paste0(mandat.args, collapse = "\n"))
792
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
Gael  MILLOT's avatar
Gael MILLOT committed
793
}
794
795
796
797
798
799
800
801
# end arg with no default values
# argument primary 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$text) , checked.arg.names <- c(checked.arg.names, tempo$object.name))
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)
Gael's avatar
tempo    
Gael committed
802
803
804
805
}else{
# no fun_check test here, it is just for checked.arg.names
tempo <- fun_check(data = n, class = "vector")
checked.arg.names <- c(checked.arg.names, tempo$object.name)
Gael's avatar
tempo    
Gael committed
806
}
807
tempo <- fun_check(data = warn.print, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
808
if(any(arg.check) == TRUE){ # normally no NA
809
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 == #
Gael's avatar
tempo    
Gael committed
810
}
Gael's avatar
tempo    
Gael committed
811
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.7/r_debugging_tools-v1.7.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()
812
813
814
815
816
# end argument primary checking
# second round of checking and data preparation
# management of NA arguments
tempo.arg <- names(arg.user.setting) # values provided by the user
tempo.log <- suppressWarnings(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
817
if(any(tempo.log) == TRUE){ # normally no NA because is.na() used here
818
819
820
821
822
823
824
825
826
827
828
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 management of NA arguments
# management of NULL arguments
tempo.arg <-c(
"data", 
# "n", # because can be NULL
"warn.print"
)
tempo.log <- sapply(lapply(tempo.arg, FUN = get, env = sys.nframe(), inherit = FALSE), FUN = is.null)
829
if(any(tempo.log) == TRUE){# normally no NA with is.null()
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
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 management of NULL arguments
# code that protects set.seed() in the global environment
# end code that protects set.seed() in the global environment
# warning initiation
ini.warning.length <- options()$warning.length
warn <- NULL
# warn.count <- 0 # not required
# end warning initiation
# other checkings
if( ! is.null(n)){
if(n < 1){
tempo.cat <- paste0("ERROR IN ", function.name, ": n ARGUMENT MUST BE A POSITIVE AND NON NULL INTEGER")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# warn.count <- warn.count + 1
tempo.warn <- paste0("SOME COMPARTMENTS CAN BE TRUNCATED (n ARGUMENT IS ", n, ")")
warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
}
# end other checkings
# reserved word checking
# end reserved word checking
# end second round of checking and data preparation
# package checking
# end package checking
# main code
# new environment
env.name <- paste0("env", as.numeric(Sys.time()))
if(exists(env.name, where = -1)){ # verify if still ok when fun_info() 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("data", data, envir = get(env.name, env = sys.nframe(), inherit = FALSE)) # data assigned in a new envir for test
}
# end new environment
data.name <- deparse(substitute(data))
output <- list("NAME" = data.name)
tempo.try.error <- fun_get_message(data = "class(data)", kind = "error", header = FALSE, env = get(env.name, env = sys.nframe(), inherit = FALSE))
if(is.null(tempo.try.error)){
tempo <- list("CLASS" = class(data))
output <- c(output, tempo)
}
tempo.try.error <- fun_get_message(data = "typeof(data)", kind = "error", header = FALSE, env = get(env.name, env = sys.nframe(), inherit = FALSE))
if(is.null(tempo.try.error)){
tempo <- list("TYPE" = typeof(data))
output <- c(output, tempo)
}
tempo.try.error <- fun_get_message(data = "length(data)", kind = "error", header = FALSE, env = get(env.name, env = sys.nframe(), inherit = FALSE))
if(is.null(tempo.try.error)){
tempo <- list("LENGTH" = length(data))
output <- c(output, tempo)
}
885
if(all(typeof(data) %in% c("integer", "numeric", "double")) & ! any(class(data) %in% "factor")){ # all() without na.rm -> ok because typeof(NA) is "logical" # any() without na.rm -> ok because class(NA) is "logical"
886
887
888
889
890
891
892
893
894
tempo <- list("INF.NB" = sum(is.infinite(data)))
output <- c(output, tempo)
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)
}
895
if(all(typeof(data) %in% c("logical", "integer", "double", "complex", "character", "list"))){ # all() without na.rm -> ok because typeof(NA) is "logical"
896
897
898
899
900
901
tempo.try.error <- fun_get_message(data = "is.na(data)", kind = "error", header = FALSE, env = get(env.name, env = sys.nframe(), inherit = FALSE))
if(is.null(tempo.try.error)){
tempo <- list("NA.NB" = sum(is.na(data)))
output <- c(output, tempo)
}
}
902
903
tempo.try.error <- fun_get_message(data = "head(data)", kind = "error", header = FALSE, env = get(env.name, env = sys.nframe(), inherit = FALSE))
if(is.null(tempo.try.error)){
904
905
tempo <- list("HEAD" = head(data))
output <- c(output, tempo)
906
tempo <- list("TAIL" = tail(data)) # no reason that tail() does not work if head() works
907
output <- c(output, tempo)
908
}
909
910
911
912
913
914
915
916
917
918
tempo.try.error <- fun_get_message(data = "dim(data)", kind = "error", header = FALSE, env = get(env.name, env = sys.nframe(), inherit = FALSE))
if(is.null(tempo.try.error)){
if(length(dim(data)) > 0){
tempo <- list("DIMENSION" = dim(data))
if(length(tempo[[1]]) == 2){
names(tempo[[1]]) <- c("NROW", "NCOL")
}
output <- c(output, tempo)
}
}
919
if(all(class(data) == "data.frame") | all(class(data) %in% c("matrix", "array")) | all(class(data) == "table")){ # all() without na.rm -> ok because typeof(NA) is "logical"
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
if(length(dim(data)) > 1){ # to avoid 1D table
tempo <- list("ROW_NAMES" = dimnames(data)[[1]])
output <- c(output, tempo)
tempo <- list("COLUM_NAMES" = dimnames(data)[[2]])
output <- c(output, tempo)
}
}
tempo.try.error <- fun_get_message(data = "summary(data)", kind = "error", header = FALSE, env = get(env.name, env = sys.nframe(), inherit = FALSE))
if(is.null(tempo.try.error)){
tempo <- list("SUMMARY" = summary(data))
output <- c(output, tempo)
}
tempo.try.error <- fun_get_message(data = "noquote(matrix(capture.output(str(data))))", kind = "error", header = FALSE, env = get(env.name, env = sys.nframe(), inherit = FALSE))
if(is.null(tempo.try.error)){
tempo <- capture.output(str(data))
tempo <- list("STRUCTURE" = noquote(matrix(tempo, dimnames = list(rep("", length(tempo)), "")))) # str() print automatically, ls.str() not but does not give the order of the data.frame
output <- c(output, tempo)
}
938
if(all(class(data) == "data.frame")){ # all() without na.rm -> ok because class(NA) is "logical"
939
tempo <- list("COLUMN_TYPE" = sapply(data, FUN = "typeof"))
940
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") # any() without na.rm -> ok because class(NA) is "logical"
941
tempo.class <- sapply(data, FUN = "class")
942
if(any(unlist(tempo.class) %in% "ordered")){ # any() without na.rm -> ok because class(NA) is "logical"
943
944
945
946
947
948
949
950
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)
}
951
if(all(class(data) == "list")){ # all() without na.rm -> ok because class(NA) is "logical"
952
953
954
tempo <- list("COMPARTMENT_NAMES" = names(data))
output <- c(output, tempo)
tempo <- list("COMPARTMENT_TYPE" = sapply(data, FUN = "typeof"))
955
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")  # any() without na.rm -> ok because class(NA) is "logical"
956
tempo.class <- sapply(data, FUN = "class")
957
if(any(unlist(tempo.class) %in% "ordered")){ # any() without na.rm -> ok because class(NA) is "logical"
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
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)
}
if( ! is.null(n)){
output[names(output) != "STRUCTURE"] <- lapply(X = output[names(output) != "STRUCTURE"], FUN = head, n = n, simplify = FALSE)
}
# output
if(warn.print == FALSE){
output <- c(output, WARNING = warn)
}else 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(output)
# end output
# end main code
Gael's avatar
tempo    
Gael committed
980
}
981
982


983
######## fun_head() #### head of the left or right of big 2D objects
984
985


986
987
988
989
fun_head <- function(
data1, 
n = 6, 
side = "l"
990
){
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
# 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$object.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) #
}
Gael's avatar
tempo    
Gael committed
1029
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.7/r_debugging_tools-v1.7.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()
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
# 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])
}
Gael's avatar
tempo    
Gael committed
1045
}
1046
1047


1048
######## fun_tail() #### tail of the left or right of big 2D objects
1049
1050


1051
1052
1053
1054
fun_tail <- function(
data1, 
n = 10, 
side = "l"
1055
){
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
# 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$object.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) #
}
Gael's avatar
tempo    
Gael committed
1094
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.7/r_debugging_tools-v1.7.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()
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
# 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])
}
Gael  MILLOT's avatar
Gael MILLOT committed
1110
1111
1112
}


1113
######## fun_comp_1d() #### comparison of two 1D datasets (vectors, factors, 1D tables)
1114
1115


1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
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 ==
}
}
Gael's avatar
tempo    
Gael committed
1193
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.7/r_debugging_tools-v1.7.R") ; eval(parse(text = str_basic_arg_check_dev)) # activate this line and use the function to check arguments status
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
# 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)
}
Gael  MILLOT's avatar
Gael MILLOT committed
1329
1330


1331
######## fun_comp_2d() #### comparison of two 2D datasets (row & col names, dimensions, etc.)
1332

Gael  MILLOT's avatar
Gael MILLOT committed
1333

1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
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 ==
Gael's avatar
tempo    
Gael committed
1411
}
1412
1413
1414
1415
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 ==
}
Gael's avatar
tempo    
Gael committed
1416
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.7/r_debugging_tools-v1.7.R") ; eval(parse(text = str_basic_arg_check_dev)) # activate this line and use the function to check arguments status
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
# 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)
1594

1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
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"
}
1607

1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
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)
1646

1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
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)
Gael's avatar
tempo    
Gael committed
1700
}
1701
1702


1703
######## fun_comp_list() #### comparison of two lists
1704
1705


1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
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 ==
}
Gael's avatar
tempo    
Gael committed
1753
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.7/r_debugging_tools-v1.7.R") ; eval(parse(text = str_basic_arg_check_dev)) # activate this line and use the function to check arguments status
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
# 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)
}


######## fun_test() #### test combinations of argument values of a function and return errors (and graphs)


# add traceback https://stackoverflow.com/questions/47414119/how-to-read-a-traceback-in-r

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"
){
# 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