cute_little_R_functions.R 871 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
# change everywhere: if( ! is.null(arg.check)){
Gael  MILLOT's avatar
Gael MILLOT committed
14
# BEWARE: do not forget to save the modifications in the .R file (through RSTUDIO for indentation)
Gael's avatar
tempo    
Gael committed
15
# update examples with good comment, as in fun_gg_boxplot
16
# Make a "first round" check for each function if required
Gael  MILLOT's avatar
Gael MILLOT committed
17
# Update all argument description, saying, character vector, etc.
Gael  MILLOT's avatar
Gael MILLOT committed
18
# check all the functions using fun_test
19
# check all(, na.rm = TRUE) and any(, na.rm = TRUE)
20
# fun_mat_fill does not recognize half matrix anymore
Gael  MILLOT's avatar
Gael MILLOT committed
21
22
23
24
25
# Templates: https://prettydoc.statr.me/themes.html
# # package: http://r-pkgs.had.co.nz/
# 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
26
27
28
# https://docs.readthedocs.io/en/stable/intro/getting-started-with-sphinx.html
# https://docs.gitlab.com/ee/user/project/pages/
# also register into biotools
Gael  MILLOT's avatar
Gael MILLOT committed
29
# For heatmap: see https://bioinfo-fr.net/creer-des-heatmaps-a-partir-de-grosses-matrices-en-r
Gael  MILLOT's avatar
Gael MILLOT committed
30

Gael  MILLOT's avatar
Gael MILLOT committed
31

Gael  MILLOT's avatar
Gael MILLOT committed
32
################################ OUTLINE ################################
Gael  MILLOT's avatar
Gael MILLOT committed
33
34


Gael  MILLOT's avatar
Gael MILLOT committed
35
36
################ Object analysis    2
######## fun_check() #### check class, type, length, etc., of objects   2
Gael's avatar
Gael committed
37
38
39
40
41
######## 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
42
######## fun_comp_2d() #### comparison of two 2D datasets (row & col names, dimensions, etc.)   22
Gael's avatar
Gael committed
43
######## fun_comp_list() #### comparison of two lists   29
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
######## 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
84
85


Gael  MILLOT's avatar
Gael MILLOT committed
86
################################ FUNCTIONS ################################
Gael  MILLOT's avatar
Gael MILLOT committed
87

Gael  MILLOT's avatar
Gael MILLOT committed
88
89
90
91
92
93
94

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


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


Gael's avatar
tempo    
Gael committed
95
# Check r_debugging_tools-v1.4.R OK
96
# Check fun_test() (see cute_checks.docx) Ok
97
# clear to go Apollo
Gael's avatar
tempo    
Gael committed
98
fun_check <- function(
Gael's avatar
tempo    
Gael committed
99
100
101
102
103
104
105
106
107
108
109
110
111
112
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
113
){
Gael's avatar
tempo    
Gael committed
114
115
116
117
118
119
120
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
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
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
# AIM
# check the class, type, mode and length of the data argument
# mainly used to check the arguments of other functions
# check also other kind of data parameters, is it a proportion? Is it type double but numbers without decimal part?
# if options == NULL, then at least class or type or mode or length argument must be non null
# if options is non null, then class, type and mode must be NULL, and length can be NULL or specified
# WARNINGS
# The function tests what is written in arguments, even if what is written is incoherent. For instance, fun_check(data = factor(1), class = "factor", mode = "character") will return a problem, and this, what ever the object tested in the data argument, because no object can be class "factor" and mode "character" (factors are class "factor" and mode "numeric")
# Since R >= 4.0.0, class(matrix()) returns "matrix" "array", and not "matrix" alone as before. However, use argument class = "matrix" to check for matrix object (of class "matrix" "array" in R >= 4.0.0) and use argument class = "array" to check for array object (of class "array" in R >= 4.0.0)
# ARGUMENTS
# data: object to test
# class: character string. Either one of the class() result (But see the warning section above) or "vector" or "ggplot2" (i.e., objects of class c("gg", "ggplot")) or NULL
# typeof: character string. Either one of the typeof() result or NULL
# mode: character string. Either one of the mode() result (for non vector object) or NULL
# length: numeric value indicating the length of the object. Not considered if NULL
# prop: logical. Are the numeric values between 0 and 1 (proportion)? If TRUE, can be used alone, without considering class, etc.
# double.as.integer.allowed: logical. If TRUE, no error is reported if argument is set to typeof == "integer" or class == "integer", while the reality is typeof == "double" or class == "numeric" but the numbers strictly have zero as modulo (remainder of a division). This means that i <- 1 , which is typeof(i) -> "double" is considered as integer with double.as.integer.allowed = TRUE. WARNING: data%%1 == 0 but not isTRUE(all.equal(data%%1, 0)) is used here because the argument checks for integers stored as double (does not check for decimal numbers that are approximate integers)
# options: a vector of character strings indicating all the possible option values for data
# all.options.in.data: logical. If TRUE, all of the options must be present at least once in data, and nothing else. If FALSE, some or all of the options must be present in data, and nothing else. Ignored if options is NULL
# na.contain: logical. Can data contain NA?
# neg.values: logical. Are negative numeric values authorized? Warning: only considered if set to FALSE, to check for non negative values when class is set to "vector", "numeric", "matrix", "array", "data.frame", "table", or typeof is set to "double", "integer", or mode is set to "numeric". Ignored in other cases, notably with prop argument (which checks for values between 0 and 1 anyhow)
# print: logical. Print the error message if $problem is TRUE? WARNING: set by default to FALSE, which facilitates the control of the error message output when using fun_check() inside functions. See the example section
# data.name: character string indicating the name of the object to test. If NULL, use the name of the object assigned to the data argument
# fun.name: character string indicating the name of the function checked (i.e., when fun_check() is used to check its argument). If non NULL, name will be added into the error message returned by fun_check()
# RETURN
# a list containing:
# $problem: logical. Is there any problem detected?
# $text: the problem detected
# $fun.name: name of the checked parameter
# REQUIRED PACKAGES
# none
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# none
# EXAMPLE
# test <- matrix(1:3) ; fun_check(data = test, print = TRUE, class = "vector", mode = "numeric")
# DEBUGGING
# data = expression(TEST) ; data.name = NULL ; class = "vector" ; typeof = NULL ; mode = NULL ; length = 1 ; prop = FALSE ; double.as.integer.allowed = FALSE ; options = NULL ; all.options.in.data = FALSE ; na.contain = FALSE ; neg.values = TRUE ; print = TRUE ; fun.name = NULL
# function name: no used in this function for the error message, to avoid env colliding
# argument checking
# fun.name checked first because required next
if( ! is.null(fun.name)){
if( ! (all(class(fun.name) == "character") & length(fun.name) == 1)){
tempo.cat <- paste0("ERROR IN fun_check(): THE fun.name ARGUMENT MUST BE A CHARACTER VECTOR OF LENGTH 1: ", paste(fun.name, collapse = " "))
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
}
# end fun.name checked first because required next
# arg with no default values
if(missing(data)){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": ARGUMENT data HAS NO DEFAULT VALUE AND REQUIRES ONE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end arg with no default values
# dealing with NA
if(any(is.na(data.name)) | any(is.na(class)) | any(is.na(typeof)) | any(is.na(mode)) | any(is.na(length)) | any(is.na(prop)) | any(is.na(double.as.integer.allowed)) | any(is.na(all.options.in.data)) | any(is.na(na.contain)) | any(is.na(neg.values)) | any(is.na(print)) | any(is.na(fun.name))){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": NO ARGUMENT EXCEPT data AND options CAN HAVE NA VALUES\nPROBLEMATIC ARGUMENTS ARE: ", paste(c("data.name", "class", "typeof", "mode", "length", "prop", "double.as.integer.allowed", "all.options.in.data", "na.contain", "neg.values", "print", "fun.name")[c(any(is.na(data.name)), any(is.na(class)), any(is.na(typeof)), any(is.na(mode)), any(is.na(length)), any(is.na(prop)), any(is.na(double.as.integer.allowed)), any(is.na(all.options.in.data)), any(is.na(na.contain)), any(is.na(neg.values)), any(is.na(print)), any(is.na(fun.name)))], collapse = " "))
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end dealing with NA
# dealing with NULL
if(is.null(prop) | is.null(double.as.integer.allowed) | is.null(all.options.in.data) | is.null(na.contain) | is.null(neg.values) | is.null(print)){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": THESE ARGUMENTS\nprop\ndouble.as.integer.allowed\nall.options.in.data\nna.contain\nneg.values\nprint\nCANNOT BE NULL\nPROBLEMATIC ARGUMENTS ARE: ", paste(c("prop", "double.as.integer.allowed", "all.options.in.data", "na.contain", "neg.values", "print")[c(is.null(prop), is.null(double.as.integer.allowed), is.null(all.options.in.data), is.null(na.contain), is.null(neg.values), is.null(print))], collapse = " "))
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end dealing with NULL
# dealing with logical
# tested below
# end dealing with logical
if( ! is.null(data.name)){
if( ! (length(data.name) == 1 & all(class(data.name) == "character"))){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": data.name ARGUMENT MUST BE A SINGLE CHARACTER ELEMENT AND NOT ", paste(data.name, collapse = " "))
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
}
if(is.null(options) & is.null(class) & is.null(typeof) & is.null(mode) &  prop == FALSE & is.null(length)){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": AT LEAST ONE OF THE options, class, typeof, mode, prop, OR length ARGUMENT MUST BE SPECIFIED (I.E, TRUE FOR prop)")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if( ! is.null(options) & ( ! is.null(class) | ! is.null(typeof) | ! is.null(mode) | prop == TRUE)){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": THE class, typeof, mode ARGUMENTS MUST BE NULL, AND prop FALSE, IF THE options ARGUMENT IS SPECIFIED\nTHE options ARGUMENT MUST BE NULL IF THE class AND/OR typeof AND/OR mode AND/OR prop ARGUMENT IS SPECIFIED")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if( ! (all(class(neg.values) == "logical") & length(neg.values) == 1 & any(is.na(neg.values)) != TRUE)){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": THE neg.values ARGUMENT MUST BE TRUE OR FALSE ONLY")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if(neg.values == FALSE & is.null(class) & is.null(typeof) & is.null(mode)){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": THE neg.values ARGUMENT CANNOT BE SWITCHED TO FALSE IF class, typeof AND mode ARGUMENTS ARE NULL")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if( ! is.null(class)){
if( ! all(class %in% c("vector", "logical", "integer", "numeric", "complex", "character", "matrix", "array", "data.frame", "list", "factor", "table", "expression", "name", "symbol", "function", "uneval", "environment", "ggplot2", "ggplot_built") & any(is.na(class)) != TRUE & length(class) == 1)){ # length == 1 here because of class(matrix()) since R4.0.0
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": class ARGUMENT MUST BE ONE OF THESE VALUE:\n\"vector\", \"logical\", \"integer\", \"numeric\", \"complex\", \"character\", \"matrix\", \"array\", \"data.frame\", \"list\", \"factor\", \"table\", \"expression\", \"name\", \"symbol\", \"function\", \"environment\", \"ggplot2\", \"ggplot_built\"")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if(neg.values == FALSE & ! any(class %in% c("vector", "numeric", "integer", "matrix", "array", "data.frame", "table"))){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": class ARGUMENT CANNOT BE OTHER THAN \"vector\", \"numeric\", \"integer\", \"matrix\", \"array\", \"data.frame\", \"table\" IF neg.values ARGUMENT IS SWITCHED TO FALSE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
}
if( ! is.null(typeof)){
if( ! (all(typeof %in% c("logical", "integer", "double", "complex", "character", "list", "expression", "name", "symbol", "closure", "special", "builtin", "environment", "S4")) & length(typeof) == 1 & any(is.na(typeof)) != TRUE)){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": typeof ARGUMENT MUST BE ONE OF THESE VALUE:\n\"logical\", \"integer\", \"double\", \"complex\", \"character\", \"list\", \"expression\", \"name\", \"symbol\", \"closure\", \"special\", \"builtin\", \"environment\", \"S4\"")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if(neg.values == FALSE & ! typeof %in% c("double", "integer")){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": typeof ARGUMENT CANNOT BE OTHER THAN \"double\" OR \"integer\" IF neg.values ARGUMENT IS SWITCHED TO FALSE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
}
if( ! is.null(mode)){
if( ! (all(mode %in% c("logical", "numeric", "complex", "character", "list", "expression", "name", "symbol", "function", "environment", "S4")) & length(mode) == 1 & any(is.na(mode)) != TRUE)){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": mode ARGUMENT MUST BE ONE OF THESE VALUE:\n\"logical\", \"numeric\", \"complex\", \"character\", \"list\", \"expression\", \"name\", \"symbol\", \"function\", \"environment\", \"S4\"")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if(neg.values == FALSE & mode != "numeric"){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": mode ARGUMENT CANNOT BE OTHER THAN \"numeric\" IF neg.values ARGUMENT IS SWITCHED TO FALSE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
}
if( ! is.null(length)){
if( ! (is.numeric(length) & length(length) == 1 & ! grepl(length, pattern = "\\.") & any(is.na(length)) != TRUE)){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": length ARGUMENT MUST BE A SINGLE INTEGER VALUE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
}
if( ! (is.logical(prop) | (length(prop) == 1 & any(is.na(prop)) != TRUE))){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": prop ARGUMENT MUST BE TRUE OR FALSE ONLY")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}else if(prop == TRUE){
if( ! is.null(class)){
if( ! any(class %in% c("vector", "numeric", "matrix", "array", "data.frame", "table"))){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": class ARGUMENT CANNOT BE OTHER THAN NULL, \"vector\", \"numeric\", \"matrix\", \"array\", \"data.frame\", \"table\" IF prop ARGUMENT IS TRUE") # not integer because prop
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
}
if( ! is.null(mode)){
if(mode != "numeric"){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": mode ARGUMENT CANNOT BE OTHER THAN NULL OR \"numeric\" IF prop ARGUMENT IS TRUE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
}
if( ! is.null(typeof)){
if(typeof != "double"){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": typeof ARGUMENT CANNOT BE OTHER THAN NULL OR \"double\" IF prop ARGUMENT IS TRUE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
}
}
if( ! (all(class(double.as.integer.allowed) == "logical") & length(double.as.integer.allowed) == 1 & any(is.na(double.as.integer.allowed)) != TRUE)){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": THE double.as.integer.allowed ARGUMENT MUST BE TRUE OR FALSE ONLY: ", paste(double.as.integer.allowed, collapse = " "))
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if( ! (is.logical(all.options.in.data) & length(all.options.in.data) == 1 & any(is.na(all.options.in.data)) != TRUE)){
tempo.cat <- paste0("ERROR IN fun_check()", ifelse(is.null(fun.name), "", paste0(" IN ", fun.name)), ": all.options.in.data ARGUMENT MUST BE A SINGLE LOGICAL VALUE (TRUE OR FALSE ONLY): ", paste(all.options.in.data, collapse = " "))
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if( ! (all(class(na.contain) == "logical") & length(na.contain) == 1 & any(is.na(na.contain)) != TRUE)){
tempo.cat <- paste0("ERROR IN fun_check(): THE na.contain ARGUMENT MUST BE TRUE OR FALSE ONLY: ", paste(na.contain, collapse = " "))
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if( ! (all(class(print) == "logical") & length(print) == 1 & any(is.na(print)) != TRUE)){
tempo.cat <- paste0("ERROR IN fun_check(): THE print ARGUMENT MUST BE TRUE OR FALSE ONLY: ", paste(print, collapse = " "))
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# data.name and fun.name tested at the beginning
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) # activate this line and use the function to check arguments status
# end argument checking
# main code
if(is.null(data.name)){
data.name <- deparse(substitute(data))
}
problem <- FALSE
text <- paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " PARAMETER")
if( ! is.null(options)){
text <- ""
if( ! all(data %in% options, na.rm = TRUE)){
problem <- TRUE
text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": THE ", data.name, " PARAMETER MUST BE SOME OF THESE OPTIONS: ", paste(options, collapse = " "), "\nTHE PROBLEMATIC ELEMENTS OF ", data.name, " ARE: ", paste(unique(data[ ! (data %in% options)]), collapse = " "))
}
if(all.options.in.data == TRUE){
if( ! all(options %in% data)){ # no need of na.rm = TRUE for all because %in% does not output NA
problem <- TRUE
text <- paste0(ifelse(text == "", "", paste0(text, "\n")), ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": THE ", data.name, " PARAMETER MUST BE MADE OF ALL THESE OPTIONS: ", paste(options, collapse = " "), "\nTHE MISSING ELEMENTS OF THE options ARGUMENT ARE: ",  paste(unique(options[ ! (options %in% data)]), collapse = " "))
}
}
if( ! is.null(length)){
if(length(data) != length){
problem <- TRUE
text <- paste0(ifelse(text == "", "", paste0(text, "\n")), ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": THE LENGTH OF ", data.name, " MUST BE ", length, " AND NOT ", length(data))
}
}
if(text == ""){
text <- paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " PARAMETER")
}
}
arg.names <- c("class", "typeof", "mode", "length")
if( ! is.null(class)){
if(class == "matrix"){ # because of class(matric()) since R4.0.0
class <- c("matrix", "array")
}else if(class == "factor" & all(class(data) %in% c("factor", "ordered"))){ # to deal with ordered factors
class <- c("factor", "ordered")
}
}
if(is.null(options)){
for(i2 in 1:length(arg.names)){
if( ! is.null(get(arg.names[i2], env = sys.nframe(), inherit = FALSE))){
# script to execute
tempo.script <- '
Gael  MILLOT's avatar
Gael MILLOT committed
323
324
325
problem <- TRUE ;
if(identical(text, paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " PARAMETER"))){
text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": THE ", data.name, " PARAMETER MUST BE ") ;
Gael  MILLOT's avatar
Gael MILLOT committed
326
}else{
Gael  MILLOT's avatar
Gael MILLOT committed
327
text <- paste0(text, " AND ") ; 
Gael  MILLOT's avatar
Gael MILLOT committed
328
}
329
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)})
Gael  MILLOT's avatar
Gael MILLOT committed
330
'
Gael's avatar
tempo    
Gael committed
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
# end script to execute
if(typeof(data) == "double" & double.as.integer.allowed == TRUE & ((arg.names[i2] == "class" & all(get(arg.names[i2], env = sys.nframe(), inherit = FALSE) == "integer")) | (arg.names[i2] == "typeof" & all(get(arg.names[i2], env = sys.nframe(), inherit = FALSE) == "integer")))){
if( ! all(data %% 1 == 0, na.rm = TRUE)){ # to check integers (use %%, meaning the remaining of a division): see the precedent line. isTRUE(all.equal(data%%1, rep(0, length(data)))) not used because we strictly need zero as a result
eval(parse(text = tempo.script)) # execute tempo.script
}
}else if( ! any(all(get(arg.names[i2], env = sys.nframe(), inherit = FALSE) %in% c("vector", "ggplot2"))) & ! all(eval(parse(text = paste0(arg.names[i2], "(data)"))) %in% get(arg.names[i2], env = sys.nframe(), inherit = FALSE))){ # test the four c("class", "typeof", "mode", "length") arguments with their corresponding function. No need of na.rm = TRUE for all because %in% does not output NA
eval(parse(text = tempo.script)) # execute tempo.script
}else if(arg.names[i2] == "class" & all(get(arg.names[i2], env = sys.nframe(), inherit = FALSE) == "vector") & ! (all(class(data) %in% "numeric") | all(class(data) %in% "integer") | all(class(data) %in% "character") | all(class(data) %in% "logical"))){ # test class == "vector". No need of na.rm = TRUE for all because %in% does not output NA
eval(parse(text = tempo.script)) # execute tempo.script
}else if(arg.names[i2] == "class" & all(get(arg.names[i2], env = sys.nframe(), inherit = FALSE) == "ggplot2") & ! all(class(data) %in% c("gg", "ggplot"))){ # test ggplot object
eval(parse(text = tempo.script)) # execute tempo.script
}
}
}
}
Gael  MILLOT's avatar
Gael MILLOT committed
346
if(prop == TRUE){
Gael's avatar
tempo    
Gael committed
347
348
349
350
351
352
353
354
355
if(is.null(data) | any(data < 0 | data > 1, na.rm = TRUE)){
problem <- TRUE
if(identical(text, paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " PARAMETER"))){
text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": ")
}else{
text <- paste0(text, " AND ")
}
text <- paste0(text, "THE ", data.name, " PARAMETER MUST BE DECIMAL VALUES BETWEEN 0 AND 1")
}
356
}
357
if(all(class(data) %in% "expression")){  # no need of na.rm = TRUE for all because %in% does not output NA
Gael's avatar
tempo    
Gael committed
358
data <- as.character(data) # to evaluate the presence of NA
Gael  MILLOT's avatar
Gael MILLOT committed
359
}
360
if(na.contain == FALSE & (mode(data) %in% c("logical", "numeric", "complex", "character", "list", "expression", "name", "symbol"))){ # before it was ! (class(data) %in% c("function", "environment"))
Gael's avatar
tempo    
Gael committed
361
362
363
364
365
366
367
368
369
if(any(is.na(data)) == TRUE){ # not on the same line because when data is class envir or function , do not like that
problem <- TRUE
if(identical(text, paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " PARAMETER"))){
text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": ")
}else{
text <- paste0(text, " AND ")
}
text <- paste0(text, "THE ", data.name, " PARAMETER CONTAINS NA WHILE NOT AUTHORIZED")
}
Gael  MILLOT's avatar
Gael MILLOT committed
370
}
Gael  MILLOT's avatar
Gael MILLOT committed
371
if(neg.values == FALSE){
Gael's avatar
tempo    
Gael committed
372
373
374
375
376
377
378
379
380
if(any(data < 0, na.rm = TRUE)){
problem <- TRUE
if(identical(text, paste0(ifelse(is.null(fun.name), "", paste0("IN ", fun.name, ": ")), "NO PROBLEM DETECTED FOR THE ", data.name, " PARAMETER"))){
text <- paste0(ifelse(is.null(fun.name), "ERROR", paste0("ERROR IN ", fun.name)), ": ")
}else{
text <- paste0(text, " AND ")
}
text <- paste0(text, "THE ", data.name, " PARAMETER MUST BE NON NEGATIVE NUMERIC VALUES")
}
Gael  MILLOT's avatar
Gael MILLOT committed
381
}
Gael  MILLOT's avatar
Gael MILLOT committed
382
if(print == TRUE & problem == TRUE){
Gael's avatar
tempo    
Gael committed
383
cat(paste0("\n\n================\n\n", text, "\n\n================\n\n"))
Gael  MILLOT's avatar
Gael MILLOT committed
384
}
Gael  MILLOT's avatar
Gael MILLOT committed
385
386
output <- list(problem = problem, text = text, fun.name = data.name)
return(output)
Gael  MILLOT's avatar
Gael MILLOT committed
387
}
Gael  MILLOT's avatar
Gael MILLOT committed
388

389
######## fun_secu() #### verif that local variables are not present in other envs
Gael  MILLOT's avatar
Gael MILLOT committed
390
391
392


fun_secu <- function(pos = 1, name = NULL){
Gael's avatar
tempo    
Gael committed
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
# AIM
# verif that local variables are not present in other environments, in order to avoid scope preference usage. The fun_secu() function checks by default the parent environment. This means that when used inside a function, it checks the local environment of this function. When used in the Global environment, it would check this environment
# ARGUMENTS
# pos: single integer indicating the position of the environment checked (argument n of parent.frame()). VaLue 1 means one step above the fun_secu() local environment. Thus, if fun_secu() is used in the working environment, with pos ==1, variables of this env will be checked in the above envs. If fun_secu() is used in a function, with pos ==1, variables presents in the local env of the functions will be checked in the above envs (which includes the working environment (Global env)
# name: single character string indicating the name of the function checked
# RETURN
# a character string of the local variables that match variables in the different environments of the R scope, or NULL if no match
# REQUIRED PACKAGES
# none
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# EXAMPLES
# fun_secu()
# fun_secu(pos = 2)
# mean <- 0 ; fun1 <- function(){sd <- 1 ; fun_secu(name = as.character(sys.calls()[[length(sys.calls())]]))} ; fun2 <- function(){cor <- 2 ; fun1()} ; fun1() ; fun2() ; rm(mean) # sys.calls() gives the the function name at top stack of the imbricated functions, sys.calls()[[length(sys.calls())]] the name of the just above function. This can also been used for the above function: as.character(sys.call(1))
# test.pos <- 2 ; mean <- 0 ; fun1 <- function(){sd <- 1 ; fun_secu(pos = test.pos, name = if(length(sys.calls()) >= test.pos){as.character(sys.calls()[[length(sys.calls()) + 1 - test.pos]])}else{search()[ (1:length(search()))[test.pos - length(sys.calls())]]})} ; fun2 <- function(){cor <- 2 ; fun1()} ; fun1() ; fun2() ; rm(mean) # for argument name, here is a way to have the name of the tested environment according to test.pos value
# DEBUGGING
# pos = 1 ; name = NULL # for function debugging
# function name
function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
# end function name
# required function checking
if(length(utils::find("fun_check", mode = "function")) == 0){
tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end required function checking
# argument checking
arg.check <- NULL #
text.check <- NULL #
checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
tempo <- fun_check(data = pos, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, fun.name = function.name) ; eval(ee)
if( ! is.null(name)){
tempo <- fun_check(data = name, class = "vector", typeof = "character", length = 1, fun.name = function.name) ; eval(ee)
}
if(any(arg.check) == TRUE){
stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
}
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
# end argument checking
# main code
# match.list <- vector("list", length = (length(sys.calls()) - 1 + length(search()) + ifelse(length(sys.calls()) == 1, -1, 0))) # match.list is a list of all the environment tested (local of functions and R envs), length(sys.calls()) - 1 to remove the level of the fun_secu() function, sys.calls() giving all the names of the imbricated functions, including fun_secu, ifelse(length(sys.calls()) == 1, -1, 0) to remove Global env if this one is tested
tempo.name <- rev(as.character(unlist(sys.calls()))) # get names of frames (i.e., enclosed env)
tempo.frame <- rev(sys.frames())  # get frames (i.e., enclosed env)
# dealing with source()
# source() used in the Global env creates three frames above the Global env, which should be removed because not very interesting for variable duplications. Add a <<-(sys.frames()) in this code and source anova_contrasts code to see this. With ls(a[[4]]), we can see the content of each env, which are probably elements of source()
if(any(sapply(tempo.frame, FUN = environmentName) %in% "R_GlobalEnv")){
global.pos <- which(sapply(tempo.frame, FUN = environmentName) %in% "R_GlobalEnv")
# remove the global env (because already in search(), and all the oabove env
tempo.name <- tempo.name[-c(global.pos:length(tempo.frame))]
tempo.frame <- tempo.frame[-c(global.pos:length(tempo.frame))]
}
# end dealing with source()
# might have a problem if(length(tempo.name) == 0){
match.list <- vector("list", length = length(tempo.name) + length(search())) # match.list is a list of all the environment tested (local of functions and R envs), length(sys.calls()) - 1 to remove the level of the fun_secu() function, sys.calls() giving all the names of the imbricated functions, including fun_secu, ifelse(length(sys.calls()) == 1, -1, 0) to remove Global env if this one is tested
ls.names <- c(tempo.name, search()) # names of the functions + names of the search() environments
ls.input <- c(tempo.frame, as.list(search())) # environements of the functions + names of the search() environments
names(match.list) <- ls.names # 
match.list <- match.list[-c(1:(pos + 1))] # because we check only above pos
ls.tested <- ls.input[[pos + 1]]
ls.input <- ls.input[-c(1:(pos + 1))]
for(i1 in 1:length(match.list)){
if(any(ls(name = ls.input[[i1]], all.names = TRUE) %in% ls(name = ls.tested, all.names = TRUE))){
match.list[i1] <- list(ls(name = ls.input[[i1]], all.names = TRUE)[ls(name = ls.input[[i1]], all.names = TRUE) %in% ls(name = ls.tested, all.names = TRUE)])
}
}
if( ! all(sapply(match.list, FUN = is.null))){
output <- paste0("SOME VARIABLES ", ifelse(is.null(name), "OF THE CHECKED ENVIRONMENT", paste0("OF ", name)), " ARE ALSO PRESENT IN :\n", paste0(names(match.list[ ! sapply(match.list, FUN = is.null)]), ": ", sapply(match.list[ ! sapply(match.list, FUN = is.null)], FUN = paste0, collapse = " "), collapse = "\n"))
}else{
output <- NULL
}
return(output)
Gael  MILLOT's avatar
Gael MILLOT committed
466
467
}

Gael  MILLOT's avatar
Gael MILLOT committed
468
469
470
471
472

######## fun_info() #### recover object information


# Check OK: clear to go Apollo
473
fun_info <- function(data, n = 20, full = FALSE, warn.print = TRUE){
Gael's avatar
tempo    
Gael committed
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
# AIM
# provide a complete description of an object. Using complete = TRUE allows data recovering. Using complete = FALSE is a convenient display feature (in case of large dataset)
# ARGUMENTS
# data: object to test
# n: positive integer value indicating the number of element to display per compartment of the output list (i.e., head(..., n)). Ignored if full argument is TRUE. Also ignored for the STRUCTURE compartment output, corresponding to ls.str() information (because head() removes almost everything)
# full: logical. Return the full information?
# warn.print: logical. Print potential warnings at the end of the execution? If FALSE the message or NULL (if no message) is added in the output as an additional compartment
# RETURN
# a list containing information, depending on the class and type of data
# if data is made of numerics, provide range, sum, mean, number of NA and number of Inf
# REQUIRED PACKAGES
# none
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# EXAMPLES
# fun_info(data = 1:3)
# fun_info(data = 1:3, n = 2)
# fun_info(data = 1:3, n = 2, full = TRUE)
# fun_info(data.frame(a = 1:2, b = ordered(factor(c("A", "B"))), stringsAsFactors = TRUE))
# fun_info(list(a = 1:3, b = ordered(factor(c("A", "B")))))
# DEBUGGING
# data = NULL # for function debugging
# data = 1:3 # for function debugging
# data = matrix(1:3) # for function debugging
# data = data.frame(a = 1:2, b = c("A", "B"), stringsAsFactors = TRUE) # for function debugging
# data = factor(c("b", "a")) # for function debugging
# data = ordered(factor(c("b", "a"))) # for function debugging
# data = list(a = 1:3, b = factor(c("A", "B"))) # for function debugging
# data = list(a = 1:3, b = ordered(factor(c("A", "B")))) # for function debugging
# function name
function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
# end function name
# required function checking
if(length(utils::find("fun_check", mode = "function")) == 0){
tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end required function checking
# argument checking
arg.check <- NULL #
text.check <- NULL #
checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
tempo <- fun_check(data = n, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, fun.name = function.name) ; eval(ee)
if(tempo$problem == FALSE & n < 1){
tempo.cat <- paste0("ERROR IN ", function.name, ": n ARGUMENT MUST BE A POSITIVE AND NON NULL INTEGER")
text.check <- c(text.check, tempo.cat)
arg.check <- c(arg.check, TRUE)
}
tempo <- fun_check(data = full, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
tempo <- fun_check(data = warn.print, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
if(any(arg.check) == TRUE){
stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
}
# end argument checking
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) # activate this line and use the function to check arguments status
# main code
warn <- NULL
if(full == FALSE){
warn <- paste0("FROM ", function.name, ":\n\nSOME COMPARTMENTS CAN BE TRUNCATED (n ARGUMENT IS ", n, ")\n\n")
}
data.name <- deparse(substitute(data))
output <- list("NAME" = data.name)
tempo <- list("CLASS" = class(data))
output <- c(output, tempo)
tempo <- list("TYPE" = typeof(data))
output <- c(output, tempo)
tempo <- list("LENGTH" = length(data))
output <- c(output, tempo)
if(all(typeof(data) %in% c("integer", "numeric", "double"))){
tempo <- list("RANGE" = range(data[ ! is.infinite(data)], na.rm = TRUE))
output <- c(output, tempo)
tempo <- list("SUM" = sum(data[ ! is.infinite(data)], na.rm = TRUE))
output <- c(output, tempo)
tempo <- list("MEAN" = mean(data[ ! is.infinite(data)], na.rm = TRUE))
output <- c(output, tempo)
tempo <- list("NA.NB" = sum(is.na(data)))
output <- c(output, tempo)
tempo <- list("INF.NB" = sum(is.infinite(data)))
output <- c(output, tempo)
}
tempo <- list("HEAD" = head(data))
output <- c(output, tempo)
if( ! is.null(data)){
tempo <- list("TAIL" = tail(data))
output <- c(output, tempo)
if( ! is.null(dim(data))){
tempo <- list("DIMENSION" = dim(data))
names(tempo[[1]]) <- c("NROW", "NCOL")
output <- c(output, tempo)
}
tempo <- list("SUMMARY" = summary(data))
output <- c(output, tempo)
}
if(all(class(data) == "data.frame" | all(class(data) %in% c("matrix", "array")))){
tempo <- list("ROW_NAMES" = dimnames(data)[[1]])
output <- c(output, tempo)
tempo <- list("COLUM_NAMES" = dimnames(data)[[2]])
output <- c(output, tempo)
}
if(all(class(data) == "data.frame")){
tempo <- list("STRUCTURE" = ls.str(data)) # str() print automatically, ls.str() not but does not give the order of the data.frame
output <- c(output, tempo)
tempo <- list("COLUMN_TYPE" = sapply(data, FUN = "typeof"))
if(any(sapply(data, FUN = "class") %in% "factor")){ # if an ordered factor is present, then sapply(data, FUN = "class") return a list but works with any(sapply(data, FUN = "class") %in% "factor") 
tempo.class <- sapply(data, FUN = "class")
if(any(unlist(tempo.class) %in% "ordered")){
tempo2 <- sapply(tempo.class, paste, collapse = " ") # paste the "ordered" factor" in "ordered factor"
}else{
tempo2 <- unlist(tempo.class)
}
tempo[["COLUMN_TYPE"]][grepl(x = tempo2, pattern = "factor")] <- tempo2[grepl(x = tempo2, pattern = "factor")]
}
output <- c(output, tempo)
}
if(all(class(data) == "list")){
tempo <- list("COMPARTMENT_NAMES" = names(data))
output <- c(output, tempo)
tempo <- list("COMPARTMENT_TYPE" = sapply(data, FUN = "typeof"))
if(any(unlist(sapply(data, FUN = "class")) %in% "factor")){ # if an ordered factor is present, then sapply(data, FUN = "class") return a list but works with any(sapply(data, FUN = "class") %in% "factor") 
tempo.class <- sapply(data, FUN = "class")
if(any(unlist(tempo.class) %in% "ordered")){
tempo2 <- sapply(tempo.class, paste, collapse = " ") # paste the "ordered" factor" in "ordered factor"
}else{
tempo2 <- unlist(tempo.class)
}
tempo[["COMPARTMENT_TYPE"]][grepl(x = tempo2, pattern = "factor")] <- tempo2[grepl(x = tempo2, pattern = "factor")]
}
output <- c(output, tempo)
tempo <- list("STRUCTURE" = ls.str(data)) # str() print automatically, ls.str() not but does not give the order of the data.frame
output <- c(output, tempo)
}
if(full == FALSE){
output[names(output) != "STRUCTURE"] <- lapply(X = output[names(output) != "STRUCTURE"], FUN = head, n = n, simplify = FALSE)
}
if(warn.print == FALSE){
output <- c(output, WARNING = warn)
}else if(warn.print == TRUE & ! is.null(warn)){
on.exit(warning(warn, call. = FALSE))
}
return(output)
Gael  MILLOT's avatar
Gael MILLOT committed
615
}
Gael  MILLOT's avatar
Gael MILLOT committed
616
617
618
619
620
621


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


# Check OK: clear to go Apollo
Gael's avatar
tempo    
Gael committed
622
fun_head <- function(
Gael's avatar
tempo    
Gael committed
623
624
625
data1, 
n = 6, 
side = "l"
Gael's avatar
tempo    
Gael committed
626
){
Gael's avatar
tempo    
Gael committed
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
# AIM
# as head() but display the left or right head of big 2D objects
# ARGUMENTS
# data1: any object but more dedicated for matrix, data frame or table
# n: as in head() but for for matrix, data frame or table, number of dimension to print (10 means 10 rows and columns)
# side: either "l" or "r" for the left or right side of the 2D object (only for matrix, data frame or table)
# BEWARE: other arguments of head() not used
# RETURN
# the head
# REQUIRED PACKAGES
# none
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# EXAMPLES
# obs1 = matrix(1:30, ncol = 5, dimnames = list(letters[1:6], LETTERS[1:5])) ; obs1 ; fun_head(obs1, 3)
# obs1 = matrix(1:30, ncol = 5, dimnames = list(letters[1:6], LETTERS[1:5])) ; obs1 ; fun_head(obs1, 3, "right")
# DEBUGGING
# data1 = matrix(1:30, ncol = 5) # for function debugging
# data1 = matrix(1:30, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
# function name
function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
# end function name
# required function checking
if(length(utils::find("fun_check", mode = "function")) == 0){
tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end required function checking
# argument checking
arg.check <- NULL #
text.check <- NULL #
checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
tempo <- fun_check(data = n, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, fun.name = function.name) ; eval(ee)
tempo <- fun_check(data = side, options = c("l", "r"), length = 1, fun.name = function.name) ; eval(ee)
if(any(arg.check) == TRUE){
stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
}
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
# end argument checking
# main code
if( ! (any(class(data1) %in% c("data.frame", "table")) | all(class(data1) %in% c("matrix", "array")))){ # before R4.0.0, it was  ! any(class(data1) %in% c("matrix", "data.frame", "table"))
return(head(data1, n))
}else{
obs.dim <- dim(data1)
row <- 1:ifelse(obs.dim[1] < n, obs.dim[1], n)
if(side == "l"){
col <- 1:ifelse(obs.dim[2] < n, obs.dim[2], n)
}
if(side == "r"){
col <- ifelse(obs.dim[2] < n, 1, obs.dim[2] - n + 1):obs.dim[2]
}
return(data1[row, col])
}
Gael  MILLOT's avatar
Gael MILLOT committed
681
682
683
}


Gael  MILLOT's avatar
Gael MILLOT committed
684
685
686
687
######## fun_tail() #### tail of the left or right of big 2D objects


# Check OK: clear to go Apollo
Gael's avatar
tempo    
Gael committed
688
fun_tail <- function(
Gael's avatar
tempo    
Gael committed
689
690
691
data1, 
n = 10, 
side = "l"
Gael's avatar
tempo    
Gael committed
692
){
Gael's avatar
tempo    
Gael committed
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
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
# AIM
# as tail() but display the left or right head of big 2D objects
# ARGUMENTS
# data1: any object but more dedicated for matrix, data frame or table
# n: as in tail() but for for matrix, data frame or table, number of dimension to print (10 means 10 rows and columns)
# side: either "l" or "r" for the left or right side of the 2D object (only for matrix, data frame or table)
# BEWARE: other arguments of tail() not used
# RETURN
# the tail
# REQUIRED PACKAGES
# none
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# EXAMPLES
# obs1 = matrix(1:30, ncol = 5, dimnames = list(letters[1:6], LETTERS[1:5])) ; obs1 ; fun_tail(obs1, 3)
# obs1 = matrix(1:30, ncol = 5, dimnames = list(letters[1:6], LETTERS[1:5])) ; obs1 ; fun_tail(obs1, 3, "r")
# DEBUGGING
# data1 = matrix(1:10, ncol = 5) # for function debugging
# data1 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
# function name
function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
# end function name
# required function checking
if(length(utils::find("fun_check", mode = "function")) == 0){
tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end required function checking
# argument checking
arg.check <- NULL #
text.check <- NULL #
checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
tempo <- fun_check(data = n, class = "vector", typeof = "integer", double.as.integer.allowed = TRUE, length = 1, fun.name = function.name) ; eval(ee)
tempo <- fun_check(data = side, options = c("l", "r"), length = 1, fun.name = function.name) ; eval(ee)
if(any(arg.check) == TRUE){
stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
}
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
# end argument checking
# main code
if( ! (any(class(data1) %in% c("data.frame", "table")) | all(class(data1) %in% c("matrix", "array")))){ # before R4.0.0, it was  ! any(class(data1) %in% c("matrix", "data.frame", "table"))
return(tail(data1, n))
}else{
obs.dim <- dim(data1)
row <- ifelse(obs.dim[1] < n, 1, obs.dim[1] - n + 1):obs.dim[1]
if(side == "l"){
col <- 1:ifelse(obs.dim[2] < n, obs.dim[2], n)
}
if(side == "r"){
col <- ifelse(obs.dim[2] < n, 1, obs.dim[2] - n + 1):obs.dim[2]
}
return(data1[row, col])
}
Gael  MILLOT's avatar
Gael MILLOT committed
747
}
Gael  MILLOT's avatar
Gael MILLOT committed
748
749


Gael  MILLOT's avatar
Gael MILLOT committed
750
######## fun_comp_1d() #### comparison of two 1D datasets (vectors, factors, 1D tables)
Gael  MILLOT's avatar
Gael MILLOT committed
751

Gael  MILLOT's avatar
Gael MILLOT committed
752
753
754

# Check OK: clear to go Apollo
fun_comp_1d <- function(data1, data2){
Gael's avatar
tempo    
Gael committed
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
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
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
# AIM
# compare two 1D datasets (vector or factor or 1D table, or 1D matrix or 1D array) of the same class or not. Check and report in a list if the 2 datasets have:
# same class
# common elements
# common element names (except factors)
# common levels (factors only)
# ARGUMENTS
# data1: vector or factor or 1D table, or 1D matrix or 1D array
# data2: vector or factor or 1D table, or 1D matrix or 1D array
# RETURN
# a list containing:
# $same.class: logical. Are class identical?
# $class: class of the 2 datasets (NULL otherwise)
# $same.length: logical. Are number of elements identical?
# $length: number of elements in the 2 datasets (NULL otherwise)
# $same.levels: logical. Are levels identical? NULL if data1 and data2 are not factors
# $levels: levels of the 2 datasets if identical (NULL otherwise or NULL if data1 and data2 are not factors)
# $any.id.levels: logical. Is there any identical levels? (NULL if data1 and data2 are not factors)
# $same.levels.pos1: position, in data1, of the levels identical in data2 (NULL if data1 and data2 are not factors)
# $same.levels.pos2: position, in data2, of the levels identical in data1 (NULL if data1 and data2 are not factors)
# $common.levels: common levels between data1 and data2 (can be a subset of $levels or not). NULL if no common levels or if data1 and data2 are not factors
# $same.name: logical. Are element names identical? NULL if data1 and data2 have no names
# $name: name of elements of the 2 datasets if identical (NULL otherwise)
# $any.id.name: logical. Is there any element names identical ?
# $same.name.pos1: position, in data1, of the element names identical in data2. NULL if no identical names
# $same.name.pos2: position, in data2, of the elements names identical in data1. NULL if no identical names
# $common.names: common element names between data1 and data2 (can be a subset of $name or not). NULL if no common element names
# $any.id.element: logical. is there any identical elements ?
# $same.element.pos1: position, in data1, of the elements identical in data2. NULL if no identical elements
# $same.element.pos2: position, in data2, of the elements identical in data1. NULL if no identical elements
# $common.elements: common elements between data1 and data2. NULL if no common elements
# $same.order: logical. Are all elements in the same order? TRUE or FALSE if elements of data1 and data2 are identical but not necessary in the same order. NULL otherwise (different length for instance)
# $order1: order of all elements of data1. NULL if $same.order is FALSE
# $order2: order of all elements of data2. NULL if $same.order is FALSE
# $identical.object: logical. Are objects identical (kind of object, element names, content, including content order)?
# $identical.content: logical. Are content objects identical (identical elements, including order, excluding kind of object and element names)?
# REQUIRED PACKAGES
# none
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# none
# EXAMPLES
# obs1 = 1:5 ; obs2 = 1:5 ; names(obs1) <- LETTERS[1:5] ; names(obs2) <- LETTERS[1:5] ; fun_comp_1d(obs1, obs2)
# obs1 = 1:5 ; obs2 = 1:5 ; names(obs1) <- LETTERS[1:5] ; fun_comp_1d(obs1, obs2)
# obs1 = 1:5 ; obs2 = 3:6 ; names(obs1) <- LETTERS[1:5] ; names(obs2) <- LETTERS[1:4] ; fun_comp_1d(obs1, obs2)
# obs1 = factor(LETTERS[1:5]) ; obs2 = factor(LETTERS[1:5]) ; fun_comp_1d(obs1, obs2)
# obs1 = factor(LETTERS[1:5]) ; obs2 = factor(LETTERS[10:11]) ; fun_comp_1d(obs1, obs2)
# obs1 = factor(LETTERS[1:5]) ; obs2 = factor(LETTERS[4:7]) ; fun_comp_1d(obs1, obs2)
# obs1 = factor(c(LETTERS[1:4], "E")) ; obs2 = factor(c(LETTERS[1:4], "F")) ; fun_comp_1d(obs1, obs2)
# obs1 = 1:5 ; obs2 = factor(LETTERS[1:5]) ; fun_comp_1d(obs1, obs2)
# obs1 = 1:5 ; obs2 = 1.1:6.1 ; fun_comp_1d(obs1, obs2)
# obs1 = as.table(1:5); obs2 = as.table(1:5) ; fun_comp_1d(obs1, obs2)
# obs1 = as.table(1:5); obs2 = 1:5 ; fun_comp_1d(obs1, obs2)
# DEBUGGING
# data1 = 1:5 ; data2 = 1:5 ; names(data1) <- LETTERS[1:5] ; names(data2) <- LETTERS[1:5] # for function debugging
# function name
function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
# end function name
# argument checking
if( ! any(class(data1) %in% c("logical", "integer", "numeric", "character", "factor", "table"))){
tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 ARGUMENT MUST BE A NON NULL VECTOR, FACTOR OR 1D TABLE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}else if(all(class(data1) %in% "table")){
if(length(dim(data1)) > 1){
tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 ARGUMENT MUST BE A 1D TABLE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
}
if( ! any(class(data2) %in% c("logical", "integer", "numeric", "character", "factor", "table"))){
tempo.cat <- paste0("ERROR IN ", function.name, ": THE data2 ARGUMENT MUST BE A NON NULL VECTOR, FACTOR OR 1D TABLE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}else if(all(class(data2) %in% "table")){
if(length(dim(data2)) > 1){
tempo.cat <- paste0("ERROR IN ", function.name, ": THE data2 ARGUMENT MUST BE A 1D TABLE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
}
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) # activate this line and use the function to check arguments status
# end argument checking
# main code
same.class <- FALSE
class <- NULL
same.length <- FALSE
length <- NULL
same.levels <- NULL # not FALSE to deal with no factors
levels <- NULL
any.id.levels <- NULL
same.levels.pos1 <- NULL
same.levels.pos2 <- NULL
common.levels <- NULL
same.name <- NULL # not FALSE to deal with absence of name
name <- NULL
any.id.name <- FALSE
same.name.pos1 <- NULL
same.name.pos2 <- NULL
common.names <- NULL
any.id.element <- FALSE
same.element.pos1 <- NULL
same.element.pos2 <- NULL
common.elements <- NULL
same.order <- NULL
order1 <- NULL
order2 <- NULL
identical.object <- FALSE
identical.content <- FALSE
if(identical(data1, data2)){
same.class <- TRUE
class <- class(data1)
same.length <- TRUE
length <- length(data1)
if(any(class(data1) %in% "factor")){
same.levels <- TRUE
levels <- levels(data1)
any.id.levels <- TRUE
same.levels.pos1 <- 1:length(levels(data1))
same.levels.pos2 <- 1:length(levels(data2))
common.levels <- levels(data1)
}
if( ! is.null(names(data1))){
same.name <- TRUE
name <- names(data1)
any.id.name <- TRUE
same.name.pos1 <- 1:length(data1)
same.name.pos2 <- 1:length(data2)
common.names <- names(data1)
}
any.id.element <- TRUE
same.element.pos1 <- 1:length(data1)
same.element.pos2 <- 1:length(data2)
common.elements <- data1
same.order <- TRUE
order1 <- order(data1)
order2 <- order(data2)
identical.object <- TRUE
identical.content <- TRUE
}else{
if(identical(class(data1), class(data2))){
same.class <- TRUE
class <- class(data1)
}
if(identical(length(data1), length(data2))){
same.length<- TRUE
length <- length(data1)
}
if(any(class(data1) %in% "factor") & any(class(data2) %in% "factor")){
if(identical(levels(data1), levels(data2))){
same.levels <- TRUE
levels <- levels(data1)
}else{
same.levels <- FALSE
}
if(any(levels(data1) %in% levels(data2))){
any.id.levels <- TRUE
same.levels.pos1 <- which(levels(data1) %in% levels(data2))
}
if(any(levels(data2) %in% levels(data1))){
any.id.levels <- TRUE
same.levels.pos2 <- which(levels(data2) %in% levels(data1))
}
if(any.id.levels == TRUE){
common.levels <- unique(c(levels(data1)[same.levels.pos1], levels(data2)[same.levels.pos2]))
}
}
if(any(class(data1) %in% "factor")){ # to compare content
data1 <- as.character(data1)
}
if(any(class(data2) %in% "factor")){ # to compare content
data2 <- as.character(data2)
}
if( ! (is.null(names(data1)) & is.null(names(data2)))){
if(identical(names(data1), names(data2))){
same.name <- TRUE
name <- names(data1)
}else{
same.name <- FALSE
}
if(any(names(data1) %in% names(data2))){
any.id.name <- TRUE
same.name.pos1 <- which(names(data1) %in% names(data2))
}
if(any(names(data2) %in% names(data1))){
any.id.name <- TRUE
same.name.pos2 <- which(names(data2) %in% names(data1))
}
if(any.id.name == TRUE){
common.names <- unique(c(names(data1)[same.name.pos1], names(data2)[same.name.pos2]))
}
}
names(data1) <- NULL # names solved -> to do not be disturbed by names
names(data2) <- NULL # names solved -> to do not be disturbed by names
if(any(data1 %in% data2)){
any.id.element <- TRUE
same.element.pos1 <- which(data1 %in% data2)
}
if(any(data2 %in% data1)){
any.id.element <- TRUE
same.element.pos2 <- which(data2 %in% data1)
}
if(any.id.element == TRUE){
common.elements <- unique(c(data1[same.element.pos1], data2[same.element.pos2]))
}
if(identical(data1, data2)){
identical.content <- TRUE
same.order <- TRUE
}else if(identical(sort(data1), sort(data2))){
same.order <- FALSE
order1 <- order(data1)
order2 <- order(data2)
}
}
output <- list(same.class = same.class, class = class, same.length = same.length, length = length, same.levels = same.levels, levels = levels, any.id.levels = any.id.levels, same.levels.pos1 = same.levels.pos1, same.levels.pos2 = same.levels.pos2, common.levels = common.levels, same.name = same.name, name = name, any.id.name = any.id.name, same.name.pos1 = same.name.pos1, same.name.pos2 = same.name.pos2, common.names = common.names, any.id.element = any.id.element, same.element.pos1 = same.element.pos1, same.element.pos2 = same.element.pos2, common.elements = common.elements, same.order = same.order, order1 = order1, order2 = order2, identical.object = identical.object, identical.content = identical.content)
return(output)
Gael  MILLOT's avatar
Gael MILLOT committed
966
967
968
}


Gael  MILLOT's avatar
Gael MILLOT committed
969
970
971
972
973
######## fun_comp_2d() #### comparison of two 2D datasets (row & col names, dimensions, etc.)


# Check OK: clear to go Apollo
fun_comp_2d <- function(data1, data2){
Gael's avatar
tempo    
Gael committed
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
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
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
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
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
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
1193
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
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
# AIM
# compare two 2D datasets of the same class or not. Check and report in a list if the 2 datasets have:
# same class
# common row names
# common column names
# same row number
# same column number
# potential identical rows between the 2 datasets
# potential identical columns between the 2 datasets
# ARGUMENTS
# data1: matrix, data frame or table
# data2: matrix, data frame or table
# RETURN
# a list containing:
# $same.class: logical. Are class identical ?
# $class: classes of the 2 datasets (NULL otherwise)
# $same.dim: logical. Are dimension identical ?
# $dim: dimension of the 2 datasets (NULL otherwise)
# $same.row.nb: logical. Are number of rows identical ?
# $row.nb: nb of rows of the 2 datasets if identical (NULL otherwise)
# $same.col.nb: logical. Are number of columns identical ?
# $col.nb: nb of columns of the 2 datasets if identical (NULL otherwise)
# $same.row.name: logical. Are row names identical ? NULL if no row names in the two 2D datasets
# $row.name: name of rows of the 2 datasets if identical (NULL otherwise)
# $any.id.row.name: logical. Is there any row names identical ? NULL if no row names in the two 2D datasets
# $same.row.name.pos1: position, in data1, of the row names identical in data2
# $same.row.name.pos2: position, in data2, of the row names identical in data1
# $common.row.names: common row names between data1 and data2 (can be a subset of $name or not). NULL if no common row names
# $same.col.name: logical. Are column names identical ? NULL if no col names in the two 2D datasets
# $col.name: name of columns of the 2 datasets if identical (NULL otherwise)
# $any.id.col.name: logical. Is there any column names identical ? NULL if no col names in the two 2D datasets
# $same.col.name.pos1: position, in data1, of the column names identical in data2
# $same.col.name.pos2: position, in data2, of the column names identical in data1
# $common.col.names: common column names between data1 and data2 (can be a subset of $name or not). NULL if no common column names
# $any.id.row: logical. is there identical rows (not considering row names)? NULL if nrow(data1) * nrow(data2) > 1e10
# $same.row.pos1: position, in data1, of the rows identical in data2 (not considering row names). Return "TOO BIG FOR EVALUATION" if nrow(data1) * nrow(data2) > 1e10
# $same.row.pos2: position, in data2, of the rows identical in data1 (not considering row names). Return "TOO BIG FOR EVALUATION" if nrow(data1) * nrow(data2) > 1e10
# $any.id.col: logical. is there identical columns (not considering column names)? NULL if ncol(data1) * ncol(data2) > 1e10
# $same.col.pos1: position in data1 of the cols identical in data2 (not considering column names). Return "TOO BIG FOR EVALUATION" if ncol(data1) * ncol(data2) > 1e10
# $same.col.pos2: position in data2 of the cols identical in data1 (not considering column names). Return "TOO BIG FOR EVALUATION" if ncol(data1) * ncol(data2) > 1e10
# $identical.object: logical. Are objects identical (including row & column names)?
# $identical.content: logical. Are content objects identical (identical excluding row & column names)?
# REQUIRED PACKAGES
# none
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# none
# EXAMPLES
# obs1 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; obs2 = as.data.frame(matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])), stringsAsFactors = TRUE) ; obs1 ; obs2 ; fun_comp_2d(obs1, obs2)
# obs1 = matrix(101:110, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; obs2 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; obs1 ; obs2 ; fun_comp_2d(obs1, obs2)
# large matrices
# obs1 = matrix(1:1e6, ncol = 5, dimnames = list(NULL, LETTERS[1:5])) ; obs2 = matrix(as.integer((1:1e6)+1e6/5), ncol = 5, dimnames = list(NULL, LETTERS[1:5])) ; head(obs1) ; head(obs2) ; fun_comp_2d(obs1, obs2)
# WARNING: when comparing content (rows, columns, or total), double and integer data are considered as different -> double(1) != integer(1)
# obs1 = matrix(1:1e6, ncol = 5, dimnames = list(NULL, LETTERS[1:5])) ; obs2 = matrix((1:1e6)+1e6/5, ncol = 5, dimnames = list(NULL, LETTERS[1:5])) ; head(obs1) ; head(obs2) ; fun_comp_2d(obs1, obs2)
# obs1 = matrix(1:10, byrow = TRUE, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; obs2 = matrix(c(1:5, 101:105, 6:10), byrow = TRUE, ncol = 5, dimnames = list(c("a", "z", "b"), c(LETTERS[1:2], "k", LETTERS[5:4]))) ; obs1 ; obs2 ; fun_comp_2d(obs1, obs2)
# obs1 = t(matrix(1:10, byrow = TRUE, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5]))) ; obs2 = t(matrix(c(1:5, 101:105, 6:10), byrow = TRUE, ncol = 5, dimnames = list(c("a", "z", "b"), c(LETTERS[1:2], "k", LETTERS[5:4])))) ; obs1 ; obs2 ; fun_comp_2d(obs1, obs2)
# DEBUGGING
# data1 = matrix(1:10, ncol = 5) ; data2 = matrix(1:10, ncol = 5) # for function debugging
# data1 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; data2 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
# data1 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; data2 = matrix(1:10, ncol = 5) # for function debugging
# data1 = matrix(1:15, byrow = TRUE, ncol = 5, dimnames = list(letters[1:3], LETTERS[1:5])) ; data2 = matrix(1:10, byrow = TRUE, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
# data1 = matrix(1:15, ncol = 5, dimnames = list(letters[1:3], LETTERS[1:5])) ; data2 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
# data1 = matrix(1:15, ncol = 5, dimnames = list(paste0("A", letters[1:3]), LETTERS[1:5])) ; data2 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
# data1 = matrix(1:15, ncol = 5, dimnames = list(letters[1:3], LETTERS[1:5])) ; data2 = matrix(1:12, ncol = 4, dimnames = list(letters[1:3], LETTERS[1:4])) # for function debugging
# data1 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; data2 = matrix(101:110, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) # for function debugging
# data1 = data.frame(a = 1:3, b= letters[1:3], row.names = LETTERS[1:3], stringsAsFactors = TRUE) ; data2 = data.frame(A = 1:3, B= letters[1:3], stringsAsFactors = TRUE) # for function debugging
# data1 = matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; data2 = as.data.frame(matrix(1:10, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])), stringsAsFactors = TRUE) # for function debugging
# data1 = matrix(1:10, byrow = TRUE, ncol = 5, dimnames = list(letters[1:2], LETTERS[1:5])) ; data2 = matrix(c(1:5, 101:105, 6:10), byrow = TRUE, ncol = 5, dimnames = list(c("a", "z", "b"), c(LETTERS[1:2], "k", LETTERS[5:4]))) # for function debugging
# data1 = table(Exp1 = c("A", "A", "A", "B", "B", "B"), Exp2 = c("A1", "B1", "A1", "C1", "C1", "B1")) ; data2 = data.frame(A = 1:3, B= letters[1:3], stringsAsFactors = TRUE) # for function debugging
# data1 = matrix(1:1e6, ncol = 5, dimnames = list(NULL, LETTERS[1:5])) ; data2 = matrix((1:1e6)+1e6/5, ncol = 5, dimnames = list(NULL, LETTERS[1:5]))
# function name
function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
# end function name
# argument checking
if( ! (any(class(data1) %in% c("data.frame", "table")) | all(class(data1) %in% c("matrix", "array")))){ # before R4.0.0, it was  ! any(class(data1) %in% c("matrix", "data.frame", "table"))
tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 ARGUMENT MUST BE A MATRIX, DATA FRAME OR TABLE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if( ! (any(class(data2) %in% c("data.frame", "table")) | all(class(data2) %in% c("matrix", "array")))){ # before R4.0.0, it was  ! any(class(data2) %in% c("matrix", "data.frame", "table"))
tempo.cat <- paste0("ERROR IN ", function.name, ": THE data2 ARGUMENT MUST BE A MATRIX, DATA FRAME OR TABLE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) # activate this line and use the function to check arguments status
# end argument checking
# main code
same.class <- NULL
class <- NULL
same.dim <- NULL
dim <- NULL
same.row.nb <- NULL
row.nb <- NULL
same.col.nb <- NULL
col.nb <- NULL
same.row.name <- NULL
row.name <- NULL
any.id.row.name <- NULL
same.row.name.pos1 <- NULL
same.row.name.pos2 <- NULL
common.row.names <- NULL
same.col.name <- NULL
any.id.col.name <- NULL
same.col.name.pos1 <- NULL
same.col.name.pos2 <- NULL
common.col.names <- NULL
col.name <- NULL
any.id.row <- NULL
same.row.pos1 <- NULL
same.row.pos2 <- NULL
any.id.col <- NULL
same.col.pos1 <- NULL
same.col.pos2 <- NULL
identical.object <- NULL
identical.content <- NULL
if(identical(data1, data2) & (any(class(data1) %in% c("data.frame", "table")) | all(class(data1) %in% c("matrix", "array")))){ # before R4.0.0, it was  ! any(class(data1) %in% c("matrix", "data.frame", "table"))
same.class <- TRUE
class <- class(data1)
same.dim <- TRUE
dim <- dim(data1)
same.row.nb <- TRUE
row.nb <- nrow(data1)
same.col.nb <- TRUE
col.nb <- ncol(data1)
same.row.name <- TRUE
row.name <- dimnames(data1)[[1]]
any.id.row.name <- TRUE
same.row.name.pos1 <- 1:row.nb
same.row.name.pos2 <- 1:row.nb
common.row.names <- dimnames(data1)[[1]]
same.col.name <- TRUE
col.name <- dimnames(data1)[[2]]
any.id.col.name <- TRUE
same.col.name.pos1 <- 1:col.nb
same.col.name.pos2 <- 1:col.nb
common.col.names <- dimnames(data1)[[2]]
any.id.row <- TRUE
same.row.pos1 <- 1:row.nb
same.row.pos2 <- 1:row.nb
any.id.col <- TRUE
same.col.pos1 <- 1:col.nb
same.col.pos2 <- 1:col.nb
identical.object <- TRUE
identical.content <- TRUE
}else{
identical.object <- FALSE
if(all(class(data1) == "table") & length(dim(data1)) == 1){
tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 ARGUMENT IS A 1D TABLE. USE THE fun_comp_1d FUNCTION")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if(all(class(data2) == "table") & length(dim(data2)) == 1){
tempo.cat <- paste0("ERROR IN ", function.name, ": THE data2 ARGUMENT IS A 1D TABLE. USE THE fun_comp_1d FUNCTION")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if( ! identical(class(data1), class(data2))){
same.class <- FALSE
}else if( ! (any(class(data1) %in% c("data.frame", "table")) | all(class(data1) %in% c("matrix", "array")))){ # before R4.0.0, it was  ! any(class(data1) %in% c("matrix", "data.frame", "table"))
tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 AND data2 ARGUMENTS MUST BE EITHER MATRIX, DATA FRAME OR TABLE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}else{
same.class <- TRUE
class <- class(data1)
}
if( ! identical(dim(data1), dim(data2))){
same.dim <- FALSE
}else{
same.dim <- TRUE
dim <- dim(data1)
}
if( ! identical(nrow(data1), nrow(data2))){
same.row.nb <- FALSE
}else{
same.row.nb <- TRUE
row.nb <- nrow(data1)
}
if( ! identical(ncol(data1), ncol(data2))){
same.col.nb <- FALSE
}else{
same.col.nb <- TRUE
col.nb <- ncol(data1)
}
# row and col names
if(is.null(dimnames(data1)) & is.null(dimnames(data2))){
same.row.name <- NULL
same.col.name <- NULL
# row and col names remain NULL
}else if((is.null(dimnames(data1)) & ! is.null(dimnames(data2))) | ( ! is.null(dimnames(data1)) & is.null(dimnames(data2)))){
same.row.name <- FALSE
same.col.name <- FALSE
# row and col names remain NULL
}else{
if( ! identical(dimnames(data1)[[1]], dimnames(data2)[[1]])){
same.row.name <- FALSE
# row names remain NULL
}else{
same.row.name <- TRUE
row.name <- dimnames(data1)[[1]]
}
# row names
any.id.row.name <- FALSE
if(any(dimnames(data1)[[1]] %in% dimnames(data2)[[1]])){
any.id.row.name <- TRUE
same.row.name.pos1 <- which(dimnames(data1)[[1]] %in% dimnames(data2)[[1]])
}
if(any(dimnames(data2)[[1]] %in% dimnames(data1)[[1]])){
any.id.row.name <- TRUE
same.row.name.pos2 <- which(dimnames(data2)[[1]] %in% dimnames(data1)[[1]])
}
if(any.id.row.name == TRUE){
common.row.names <- unique(c(dimnames(data1)[[1]][same.row.name.pos1], dimnames(data2)[[1]][same.row.name.pos2]))
}
# col names
any.id.col.name <- FALSE
if(any(dimnames(data1)[[2]] %in% dimnames(data2)[[2]])){
any.id.col.name <- TRUE
same.col.name.pos1 <- which(dimnames(data1)[[2]] %in% dimnames(data2)[[2]])
}
if(any(dimnames(data2)[[2]] %in% dimnames(data1)[[2]])){
any.id.col.name <- TRUE
same.col.name.pos2 <- which(dimnames(data2)[[2]] %in% dimnames(data1)[[2]])
}
if(any.id.col.name == TRUE){
common.col.names <- unique(c(dimnames(data1)[[2]][same.col.name.pos1], dimnames(data2)[[2]][same.col.name.pos2]))
}
if( ! identical(dimnames(data1)[[2]], dimnames(data2)[[2]])){
same.col.name <- FALSE
# col names remain NULL
}else{
same.col.name <- TRUE
col.name <- dimnames(data1)[[2]]
}
}
# identical row and col content
if(all(class(data1) == "table")){
as.data.frame(matrix(data1, ncol = ncol(data1)), stringsAsFactors = FALSE)
}else if(all(class(data1) %in% c("matrix", "array"))){
data1 <- as.data.frame(data1, stringsAsFactors = FALSE)
}else if(all(class(data1) == "data.frame")){
data1 <- data.frame(lapply(data1, as.character), stringsAsFactors = FALSE)
}
if(all(class(data2) == "table")){
as.data.frame(matrix(data2, ncol = ncol(data2)), stringsAsFactors = FALSE)
}else if(all(class(data2) %in% c("matrix", "array"))){
data2 <- as.data.frame(data2, stringsAsFactors = FALSE)
}else if(all(class(data2) == "data.frame")){
data2 <- data.frame(lapply(data2, as.character), stringsAsFactors = FALSE)
}
row.names(data1) <- paste0("A", 1:nrow(data1))
row.names(data2) <- paste0("A", 1:nrow(data2))
if(same.col.nb == TRUE){ # because if not the same col nb, the row cannot be identical
if(all(sapply(data1, FUN = typeof) == "integer") & all(sapply(data2, FUN = typeof) == "integer") & as.double(nrow(data1)) * nrow(data2) <= 1e10){ # as.double(nrow(data1)) to prevent integer overflow because R is 32 bits for integers
same.row.pos1 <- which(c(as.data.frame(t(data1), stringsAsFactors = FALSE)) %in% c(as.data.frame(t(data2), stringsAsFactors = FALSE))) # this work fast with only integers (because 32 bits)
same.row.pos2 <- which(c(as.data.frame(t(data2), stringsAsFactors = FALSE)) %in% c(as.data.frame(t(data1), stringsAsFactors = FALSE)))
}else if(as.double(nrow(data1)) * nrow(data2) <= 1e6){ # as.double(nrow(data1)) to prevent integer overflow because R is 32 bits for integers
same.row.pos1 <- logical(length = nrow(data1)) # FALSE by default
same.row.pos1[] <- FALSE # security
for(i3 in 1:nrow(data1)){
for(i4 in 1:nrow(data2)){
same.row.pos1[i3] <- identical(data1[i3, ], data2[i4, ])
}
}
same.row.pos1 <- which(same.row.pos1)

same.row.pos2 <- logical(length = nrow(data2)) # FALSE by default
same.row.pos2[] <- FALSE # security
for(i3 in 1:nrow(data2)){
for(i4 in 1:nrow(data1)){
same.row.pos2[i3] <- identical(data2[i3, ], data1[i4, ])
}
}
same.row.pos2 <- which(same.row.pos2)
}else{
same.row.pos1 <- "TOO BIG FOR EVALUATION"
same.row.pos2 <- "TOO BIG FOR EVALUATION"
}

names(same.row.pos1) <- NULL
names(same.row.pos2) <- NULL
if(all(is.na(same.row.pos1))){
same.row.pos1 <- NULL
}else{
same.row.pos1 <- same.row.pos1[ ! is.na(same.row.pos1)]
any.id.row <- TRUE
}
if(all(is.na(same.row.pos2))){
same.row.pos2 <- NULL
}else{
same.row.pos2 <- same.row.pos2[ ! is.na(same.row.pos2)]
any.id.row <- TRUE
}
if(is.null(same.row.pos1) & is.null(same.row.pos2)){
any.id.row <- FALSE
}else if(length(same.row.pos1) == 0 & length(same.row.pos2) == 0){
any.id.row <- FALSE
}else if(all(same.row.pos1 == "TOO BIG FOR EVALUATION") & all(same.row.pos2 == "TOO BIG FOR EVALUATION")){
any.id.row <- NULL
}
}else{
any.id.row <- FALSE
# same.row.pos1 and 2 remain NULL
}
if(same.row.nb == TRUE){ # because if not the same row nb, the col cannot be identical
if(all(sapply(data1, FUN = typeof) == "integer") & all(sapply(data2, FUN = typeof) == "integer") & as.double(ncol(data1)) * ncol(data2) <= 1e10){ # as.double(ncol(data1)) to prevent integer overflow because R is 32 bits for integers
same.col.pos1 <- which(c(data1) %in% c(data2))
same.col.pos2 <- which(c(data2) %in% c(data1))
}else if(as.double(ncol(data1)) * ncol(data2) <= 1e6){ # as.double(ncol(data1)) to prevent integer overflow because R is 32 bits for integers
same.col.pos1 <- logical(length = ncol(data1)) # FALSE by default
same.col.pos1[] <- FALSE # security
for(i3 in 1:ncol(data1)){
for(i4 in 1:ncol(data2)){
same.col.pos1[i3] <- identical(data1[ , i3], data2[ ,i4])
}
}
same.col.pos1 <- which(same.col.pos1)

same.col.pos2 <- logical(length = ncol(data2)) # FALSE by default
same.col.pos2[] <- FALSE # security
for(i3 in 1:ncol(data2)){
for(i4 in 1:ncol(data1)){
same.col.pos2[i3] <- identical(data2[ , i3], data1[ , i4])
}
}
same.col.pos2 <- which(same.col.pos2)
}else{
same.col.pos1 <- "TOO BIG FOR EVALUATION"
same.col.pos2 <- "TOO BIG FOR EVALUATION"
}
names(same.col.pos1) <- NULL
names(same.col.pos2) <- NULL
if(all(is.na(same.col.pos1))){
same.col.pos1 <- NULL
}else{
same.col.pos1 <- same.col.pos1[ ! is.na(same.col.pos1)]
any.id.col <- TRUE
}
if(all(is.na(same.col.pos2))){
same.col.pos2 <- NULL
}else{
same.col.pos2 <- same.col.pos2[ ! is.na(same.col.pos2)]
any.id.col <- TRUE
}
if(is.null(same.col.pos1) & is.null(same.col.pos2)){
any.id.col <- FALSE
}else if(length(same.col.pos1) == 0 & length(same.col.pos2) == 0){
any.id.col <- FALSE
}else if(all(same.col.pos1 == "TOO BIG FOR EVALUATION") & all(same.col.pos2 == "TOO BIG FOR EVALUATION")){
any.id.col <- NULL
}
}else{
any.id.col <- FALSE
# same.col.pos1 and 2 remain NULL
}
if(same.dim == TRUE){
names(data1) <- NULL
row.names(data1) <- NULL
names(data2) <- NULL
row.names(data2) <- NULL
if(identical(data1, data2)){
identical.content <- TRUE
}else{
identical.content <- FALSE
}
}else{
identical.content <- FALSE
}
}
output <- list(same.class = same.class, class = class, same.dim = same.dim, dim = dim, same.row.nb = same.row.nb, row.nb = row.nb, same.col.nb = same.col.nb , col.nb = col.nb, same.row.name = same.row.name, row.name = row.name, any.id.row.name = any.id.row.name, same.row.name.pos1 = same.row.name.pos1, same.row.name.pos2 = same.row.name.pos2, common.row.names = common.row.names, same.col.name = same.col.name, col.name = col.name,any.id.col.name = any.id.col.name, same.col.name.pos1 = same.col.name.pos1, same.col.name.pos2 = same.col.name.pos2, common.col.names = common.col.names, any.id.row = any.id.row, same.row.pos1 = same.row.pos1, same.row.pos2 = same.row.pos2, any.id.col = any.id.col, same.col.pos1 = same.col.pos1, same.col.pos2 = same.col.pos2, identical.object = identical.object, identical.content = identical.content)
return(output)
Gael  MILLOT's avatar
Gael MILLOT committed
1339
1340
1341
1342
1343
1344
1345
1346
}


######## fun_comp_list() #### comparison of two lists


# Check OK: clear to go Apollo
fun_comp_list <- function(data1, data2){
Gael's avatar
tempo    
Gael committed
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
1411
1412
1413
1414
1415
1416
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
# AIM
# compare two lists. Check and report in a list if the 2 datasets have:
# same length
# common names
# common compartments
# ARGUMENTS
# data1: list
# data2: list
# RETURN
# a list containing:
# $same.length: logical. Are number of elements identical?
# $length: number of elements in the 2 datasets (NULL otherwise)
# $same.name: logical. Are element names identical ?
# $name: name of elements of the 2 datasets if identical (NULL otherwise)
# $any.id.name: logical. Is there any element names identical ?
# $same.name.pos1: position, in data1, of the element names identical in data2
# $same.name.pos2: position, in data2, of the compartment names identical in data1
# $any.id.compartment: logical. is there any identical compartments ?
# $same.compartment.pos1: position, in data1, of the compartments identical in data2
# $same.compartment.pos2: position, in data2, of the compartments identical in data1
# $identical.object: logical. Are objects identical (kind of object, compartment names and content)?
# $identical.content: logical. Are content objects identical (identical compartments excluding compartment names)?
# REQUIRED PACKAGES
# none
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# none
# EXAMPLES
# obs1 = list(a = 1:5, b = LETTERS[1:2], d = matrix(1:6)) ; obs2 = list(a = 1:5, b = LETTERS[1:2], d = matrix(1:6)) ; fun_comp_list(obs1, obs2)
# obs1 = list(1:5, LETTERS[1:2]) ; obs2 = list(a = 1:5, b = LETTERS[1:2]) ; fun_comp_list(obs1, obs2)
# obs1 = list(b = 1:5, c = LETTERS[1:2]) ; obs2 = list(a = 1:5, b = LETTERS[1:2], d = matrix(1:6)) ; fun_comp_list(obs1, obs2)
# obs1 = list(b = 1:5, c = LETTERS[1:2]) ; obs2 = list(LETTERS[5:9], matrix(1:6), 1:5) ; fun_comp_list(obs1, obs2)
# DEBUGGING
# data1 = list(a = 1:5, b = LETTERS[1:2], d = matrix(1:6)) ; data2 = list(a = 1:5, b = LETTERS[1:2], d = matrix(1:6)) # for function debugging
# data1 = list(a = 1:5, b = LETTERS[1:2]) ; data2 = list(a = 1:5, b = LETTERS[1:2], d = matrix(1:6)) # for function debugging
# function name
function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
# end function name
# argument checking
if( ! any(class(data1) %in% "list")){
tempo.cat <- paste0("ERROR IN ", function.name, ": THE data1 ARGUMENT MUST BE A LIST")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if( ! any(class(data2) %in% "list")){
tempo.cat <- paste0("ERROR IN ", function.name, ": THE data2 ARGUMENT MUST BE A LIST")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) # activate this line and use the function to check arguments status
# end argument checking
# main code
same.length <- NULL
length <- NULL
same.name <- NULL
name <- NULL
any.id.name <- NULL
same.name.pos1 <- NULL
same.name.pos2 <- NULL
any.id.compartment <- NULL
same.compartment.pos1 <- NULL
same.compartment.pos2 <- NULL
identical.object <- NULL
identical.content <- NULL
if(identical(data1, data2)){
same.length <- TRUE
length <- length(data1)
if( ! is.null(names(data1))){
same.name <- TRUE
name <- names(data1)
any.id.name <- TRUE
same.name.pos1 <- 1:length(data1)
same.name.pos2 <- 1:length(data2)
}
any.id.compartment <- TRUE
same.compartment.pos1 <- 1:length(data1)
same.compartment.pos2 <- 1:length(data2)
identical.object <- TRUE
identical.content <- TRUE
}else{
identical.object <- FALSE
if( ! identical(length(data1), length(data2))){
same.length<- FALSE
}else{
same.length<- TRUE
length <- length(data1)
}
if( ! (is.null(names(data1)) & is.null(names(data2)))){
if( ! identical(names(data1), names(data2))){
same.name <- FALSE
}else{
same.name <- TRUE
name <- names(data1)
}
any.id.name <- FALSE
if(any(names(data1) %in% names(data2))){
any.id.name <- TRUE
same.name.pos1 <- which(names(data1) %in% names(data2))
}
if(any(names(data2) %in% names(data1))){
any.id.name <- TRUE
same.name.pos2 <- which(names(data2) %in% names(data1))
}
}
names(data1) <- NULL
names(data2) <- NULL
any.id.compartment <- FALSE
if(any(data1 %in% data2)){
any.id.compartment <- TRUE
same.compartment.pos1 <- which(data1 %in% data2)
}
if(any(data2 %in% data1)){
any.id.compartment <- TRUE
same.compartment.pos2 <- which(data2 %in% data1)
}
if(same.length == TRUE & ! all(is.null(same.compartment.pos1), is.null(same.compartment.pos2))){
if(identical(same.compartment.pos1, same.compartment.pos2)){
identical.content <- TRUE
}else{
identical.content <- FALSE
}
}else{
identical.content <- FALSE
}
}
output <- list(same.length = same.length, length = length, same.name = same.name, name = name, any.id.name = any.id.name, same.name.pos1 = same.name.pos1, same.name.pos2 = same.name.pos2, any.id.compartment = any.id.compartment, same.compartment.pos1 = same.compartment.pos1, same.compartment.pos2 = same.compartment.pos2, identical.object = identical.object, identical.content = identical.content)
return(output)
Gael  MILLOT's avatar
Gael MILLOT committed
1471
1472
1473
}


1474
######## fun_test() #### test combinations of argument values of a function and return errors (and graphs)
Gael  MILLOT's avatar
Gael MILLOT committed
1475
1476


Gael  MILLOT's avatar
Gael MILLOT committed
1477
1478
# add traceback https://stackoverflow.com/questions/47414119/how-to-read-a-traceback-in-r

Gael  MILLOT's avatar
Gael MILLOT committed
1479
# Check OK: clear to go Apollo
Gael's avatar
tempo    
Gael committed
1480
fun_test <- function(
Gael's avatar
tempo    
Gael committed
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
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"
Gael's avatar
tempo    
Gael committed
1492
){
Gael's avatar
tempo    
Gael committed
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
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
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
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
1700
1701
1702
1703
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
1753
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
# AIM
# test combinations of argument values of a function
# WARNINGS
# Limited to 43 arguments with at least 2 values each. The total number of arguments tested can be more if the additional arguments have a single value. The limit is due to nested "for" loops (https://stat.ethz.ch/pipermail/r-help/2008-March/157341.html), but it should not be a problem since the number of tests would be 2^43 > 8e12
# ARGUMENTS
# fun: character string indicating the name of the function tested (without brackets)
# arg: vector of character strings of arguments of fun. At least arguments that do not have default values must be present in this vector
# val: list with number of compartments equal to length of arg, each compartment containing values of the corresponding argument in arg. Each different value must be in a list or in a vector. For instance, argument 3 in arg is a logical argument (values accepted TRUE, FALSE, NA). Thus, compartment 3 of val can be either list(TRUE, FALSE, NA), or c(TRUE, FALSE, NA)
# expect.error: list of exactly the same structure as val argument, but containing FALSE or TRUE, depending on whether error is expected (TRUE) or not (FALSE) for each corresponding value of val. A message is returned depending on discrepancies between the expected and observed errors. BEWARE: not always possible to write the expected errors for all the combination of argument values. Ignored if NULL
# thread.nb: numeric value indicating the number of available threads. Write NULL if no parallelization wanted
# print.count: interger value. Print a working progress message every print.count during loops. BEWARE: can increase substentially the time to complete the process using a small value, like 10 for instance. Use Inf is no loop message desired
# plot.fun: logical. Plot the plotting function tested for each test?
# export: logical. Export the results into a .RData file and into a .txt file? If FALSE, return a list into the console (see below). BEWARE: will be automatically set to TRUE if thread.nb is not NULL. This means that when using parallelization, the results are systematically exported, not returned into the console
# res.path: character string indicating the absolute pathway of folder where the txt results and pdfs, containing all the plots, will be saved. Several txt and pdf, one per thread, if parallelization. Ignored if export is FALSE. Must be specified if thread.nb is not NULL or if export is TRUE
# lib.path: character vector specifying the absolute pathways of the directories containing the required packages if not in the default directories. Ignored if NULL
# cute.path: character string indicating the absolute path of the cute.R file. Will be remove when cute will be a package. Not considered if thread.nb is NULL
# REQUIRED PACKAGES
# lubridate
# parallel if thread.nb argument is not NULL (included in the R installation packages but not automatically loaded)
# if the tested function is in a package, this package must be imported first (no parallelization) or must be in the classical R package folder indicated by the lib.path argument (parallelization)
# RETURN
# if export is FALSE a list containing:
# $fun: the tested function
# $data: a data frame of all the combination tested, containing the following columns:
# the different values tested, named by arguments
# $kind: a vector of character strings indicating the kind of test result: either "ERROR", or "WARNING", or "OK"
# $problem: a logical vector indicating if error or not
# $expected.error: optional logical vector indicating the expected error specified in the expect.error argument
# $message: either NULL if $kind is always "OK", or the messages
# $instruction: the initial instruction
# $sys.info: system and packages info
# if export is TRUE 1) the same list object into a .RData file, 2) also the $data data frame into a .txt file, and 3) if expect.error is non NULL and if any discrepancy, the $data data frame into a .txt file but containing only the rows with discrepancies between expected and observed errors
# one or several pdf if a plotting function is tested and if the plot.fun argument is TRUE
# REQUIRED PACKAGES
# none
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# fun_get_message()
# fun_pack()
# EXAMPLES
# fun_test(fun = "unique", arg = c("x", "incomparables"), val = list(x = list(1:10, c(1,1,2,8), NA), incomparable = c(TRUE, FALSE, NA)))
# fun_test(fun = "fun_round", arg = c("data", "dec.nb", "after.lead.zero"), val = list(L1 = list(c(1, 1.0002256, 1.23568), "a", NA), L2 = list(2, c(1,3), NA), L3 = c(TRUE, FALSE, NA)))
# fun_test(fun = "plot", arg = c("x", "y"), val = list(x = list(1:10, 12:13, NA, (1:10)^2), y = list(1:10, NA, NA)),  expect.error = list(x = list(FALSE, TRUE, TRUE, FALSE), y = list(FALSE, TRUE, TRUE)), thread.nb = NULL, plot.fun = TRUE, res.path = "C:\\Users\\Gael\\Desktop\\", lib.path = NULL)
# fun_test(fun = "plot", arg = c("x", "y"), val = list(x = list(1:10, 12:13, NA, (1:10)^2), y = list(1:10, NA, NA)), thread.nb = 4, plot.fun = TRUE, res.path = "C:\\Users\\Gael\\Desktop\\", lib.path = "C:\\Program Files\\R\\R-4.0.2\\library\\")
# set.seed(1) ; obs1 <- data.frame(Time = c(rnorm(10), rnorm(10) + 2), Group1 = rep(c("G", "H"), each = 10), stringsAsFactors = TRUE) ; fun_test(fun = "fun_gg_boxplot", arg = c("data1", "y", "categ"), val = list(L1 = list(L1 = obs1), L2 = list(L1 = "Time"), L3 = list(L1 = "Group1")))
# set.seed(1) ; obs1 <- data.frame(Time = c(rnorm(10), rnorm(10) + 2), Group1 = rep(c("G", "H"), each = 10), stringsAsFactors = TRUE) ; fun_test(fun = "fun_gg_boxplot", arg = c("data1", "y", "categ"), val = list(L1 = list(obs1), L2 = "Time", L3 = "Group1"), thread.nb = NULL, plot.fun = TRUE, res.path = "C:\\Users\\Gael\\Desktop\\", lib.path = "C:\\Program Files\\R\\R-4.0.2\\library\\")
# library(ggplot2) ; fun_test(fun = "geom_histogram", arg = c("data", "mapping"), val = list(x = list(data.frame(X = "a", stringsAsFactors = TRUE)), y = list(ggplot2::aes(x = X))), thread.nb = NULL, plot.fun = TRUE, res.path = "C:\\Users\\Gael\\Desktop\\", lib.path = "C:\\Program Files\\R\\R-4.0.2\\library\\") # BEWARE: ggplot2::geom_histogram does not work
# DEBUGGING
# fun = "unique" ; arg = "x" ; val = list(x = list(1:10, c(1,1,2,8), NA)) ; expect.error = list(x = list(FALSE, FALSE, TRUE)) ; thread.nb = NULL ; plot.fun = FALSE ; export = FALSE ; res.path = "C:\\Users\\Gael\\Desktop\\" ; lib.path = NULL ; print.count = 1 ; cute.path = "C:\\Users\\Gael\\Documents\\Git_projects\\cute_little_R_functions\\cute_little_R_functions.R" # for function debugging
# fun = "unique" ; arg = c("x", "incomparables") ; val = list(x = list(1:10, c(1,1,2,8), NA), incomparable = c(TRUE, FALSE, NA)) ; expect.error = NULL ; thread.nb = 2 ; plot.fun = FALSE ; export = TRUE ; res.path = "C:\\Users\\Gael\\Desktop\\" ; lib.path = NULL ; print.count = 10 ; cute.path = "C:\\Users\\Gael\\Documents\\Git_projects\\cute_little_R_functions\\cute_little_R_functions.R" # for function debugging
# fun = "plot" ; arg = c("x", "y") ; val = list(x = list(1:10, 12:13, NA), y = list(1:10, NA, NA)) ; expect.error = list(x = list(FALSE, FALSE, TRUE, FALSE), y = list(FALSE, TRUE, TRUE)) ; print.count = 10 ; thread.nb = NULL ; plot.fun = TRUE ; export = TRUE ; res.path = "C:\\Users\\Gael\\Desktop\\" ; lib.path = NULL # for function debugging
# set.seed(1) ; obs1 <- data.frame(Time = c(rnorm(10), rnorm(10) + 2), Group1 = rep(c("G", "H"), each = 10), stringsAsFactors = TRUE) ; fun = "fun_gg_boxplot" ; arg = c("data1", "y", "categ") ; val = list(L1 = list(L1 = obs1), L2 = list(L1 = "Time"), L3 = list(L1 = "Group1")) ; expect.error = NULL ; print.count = 10 ; thread.nb = NULL ; plot.fun = TRUE ; export = TRUE ; res.path = "C:\\Users\\Gael\\Desktop\\" ; lib.path = NULL # for function debugging
# function name
function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
instruction <- match.call()
# end function name
# required function checking
req.function <- c(
"fun_check", 
"fun_get_message", 
"fun_pack"
)
for(i1 in req.function){
if(length(find(i1, mode = "function")) == 0){
tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED ", i1, "() FUNCTION IS MISSING IN THE R ENVIRONMENT")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
}
# end required function checking
# argument primary checking
# arg with no default values
if(any(missing(fun) | missing(arg) | missing(val))){
tempo.cat <- paste0("ERROR IN ", function.name, ": ARGUMENTS fun, arg AND val HAVE NO DEFAULT VALUE AND REQUIRE ONE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end arg with no default values
# using fun_check()
arg.check <- NULL #
text.check <- NULL #
checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
tempo <- fun_check(data = fun, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
if(tempo$problem == FALSE){
if(grepl(x = fun, pattern = "()$")){ # remove ()
fun <- sub(x = fun, pattern = "()$", replacement = "")
}
if( ! exists(fun)){
tempo.cat <- paste0("ERROR IN ", function.name, ": CHARACTER STRING IN fun ARGUMENT DOES NOT EXIST IN THE R WORKING ENVIRONMENT: ", paste(fun, collapse = "\n"))
text.check <- c(text.check, tempo.cat)
arg.check <- c(arg.check, TRUE)
}else if( ! all(class(get(fun)) == "function")){ # here no env = sys.nframe(), inherit = FALSE for get() because fun is a function in the classical scope
tempo.cat <- paste0("ERROR IN ", function.name, ": fun ARGUMENT IS NOT CLASS \"function\" BUT: ", paste(class(get(fun)), collapse = "\n"), "\nCHECK IF ANY CREATED OBJECT WOULD HAVE THE NAME OF THE TESTED FUNCTION")
text.check <- c(text.check, tempo.cat)
arg.check <- c(arg.check, TRUE)
}
}
tempo <- fun_check(data = arg, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
if(tempo$problem == FALSE & length(arg) == 0){
tempo.cat <- paste0("ERROR IN ", function.name, ": arg ARGUMENT CANNOT BE LENGTH 0")
text.check <- c(text.check, tempo.cat)
arg.check <- c(arg.check, TRUE)
}
tempo <- fun_check(data = val, class = "list", fun.name = function.name) ; eval(ee)
if(tempo$problem == FALSE){
for(i2 in 1:length(val)){
tempo1 <- fun_check(data = val[[i2]], class = "vector", na.contain = TRUE, fun.name = function.name)
tempo2 <- fun_check(data = val[[i2]], class = "list", na.contain = TRUE, fun.name = function.name)
if(tempo1$problem == TRUE & tempo2$problem == TRUE){
tempo.cat <- paste0("ERROR IN ", function.name, ": COMPARTMENT ", i2, " OF val ARGUMENT MUST BE A VECTOR OR A LIST")
text.check <- c(text.check, tempo.cat)
arg.check <- c(arg.check, TRUE)
}else if(tempo1$problem == FALSE){ # vector split into list compartments
val[[i2]] <- split(x = val[[i2]], f = 1:length(val[[i2]]))
}
}
}
if( ! is.null(expect.error)){
tempo <- fun_check(data = expect.error, class = "list", fun.name = function.name) ; eval(ee)
if(tempo$problem == FALSE){
for(i3 in 1:length(expect.error)){
tempo1 <- fun_check(data = expect.error[[i3]], class = "vector",  mode = "logical", fun.name = function.name)
tempo2 <- fun_check(data =  expect.error[[i3]], class = "list", fun.name = function.name)
if(tempo1$problem == TRUE & tempo2$problem == TRUE){
tempo.cat <- paste0("ERROR IN ", function.name, ": COMPARTMENT ", i3, " OF expect.error ARGUMENT MUST BE TRUE OR FALSE")
text.check <- c(text.check, tempo.cat)
arg.check <- c(arg.check, TRUE)
}else if(tempo1$problem == FALSE){ # vector split into list compartments
expect.error[[i3]] <- split(x = expect.error[[i3]], f = 1:length(expect.error[[i3]]))
}
}
}
}
if( ! is.null(thread.nb)){
tempo <- fun_check(data = thread.nb, typeof = "integer", double.as.integer.allowed = TRUE, neg.values = FALSE, length = 1, fun.name = function.name) ; eval(ee)
if(tempo$problem == FALSE & thread.nb < 1){
tempo.cat <- paste0("ERROR IN ", function.name, ": thread.nb PARAMETER MUST EQUAL OR GREATER THAN 1: ", thread.nb)
text.check <- c(text.check, tempo.cat)
arg.check <- c(arg.check, TRUE)
}
}
tempo <- fun_check(data = print.count, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
tempo <- fun_check(data = plot.fun, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
tempo <- fun_check(data = export, class = "vector", mode = "logical", length = 1, fun.name = function.name) ; eval(ee)
if( ! is.null(res.path)){
tempo <- fun_check(data = res.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
if(tempo$problem == FALSE){
if( ! all(dir.exists(res.path))){ # separation to avoid the problem of tempo$problem == FALSE and res.path == NA
tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE res.path ARGUMENT DOES NOT EXISTS:\n", paste(res.path, collapse = "\n"))
text.check <- c(text.check, tempo.cat)
arg.check <- c(arg.check, TRUE)
}
}
}
if( ! is.null(lib.path)){
tempo <- fun_check(data = lib.path, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
if(tempo$problem == FALSE){
if( ! all(dir.exists(lib.path))){ # separation to avoid the problem of tempo$problem == FALSE and lib.path == NA
tempo.cat <- paste0("ERROR IN ", function.name, ": DIRECTORY PATH INDICATED IN THE lib.path ARGUMENT DOES NOT EXISTS:\n", paste(lib.path, collapse = "\n"))
text.check <- c(text.check, tempo.cat)
arg.check <- c(arg.check, TRUE)
}
}
}
if( ! is.null(thread.nb)){
tempo <- fun_check(data = cute.path, class = "vector", typeof = "character", length = 1, fun.name = function.name) ; eval(ee)
if(tempo$problem == FALSE){
if( ! file.exists(cute.path)){
tempo.cat <- paste0("ERROR IN ", function.name, ": FILE PATH INDICATED IN THE cute.path PARAMETER DOES NOT EXISTS:\n", cute.path)
text.check <- c(text.check, tempo.cat)
arg.check <- c(arg.check, TRUE)
}
}
}
if(any(arg.check) == TRUE){
stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
}
# end using fun_check()
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
# end argument primary checking
# second round of checking and data preparation
# dealing with NA
if(any(is.na(fun)) | any(is.na(arg)) | any(is.na(expect.error)) | any(is.na(thread.nb)) | any(is.na(print.count)) | any(is.na(plot.fun)) | any(is.na(export)) | any(is.na(res.path)) | any(is.na(lib.path))){
tempo.cat <- paste0("ERROR IN ", function.name, ": NO ARGUMENT EXCEPT val CAN HAVE NA VALUES")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end dealing with NA
# dealing with NULL
if(is.null(fun) | is.null(arg) | is.null(val) | is.null(print.count) | is.null(plot.fun) | is.null(export)){
tempo.cat <- paste0("ERROR IN ", function.name, ": THESE ARGUMENTS\nfun\narg\nval\nprint.count\nplot.fun\nexport\nCANNOT BE NULL") #problematic arguments are -> put everywhere
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end dealing with NULL
if(length(arg) != length(val)){
tempo.cat <- paste0("ERROR IN ", function.name, ": LENGTH OF arg ARGUMENT MUST BE IDENTICAL TO LENGTH OF val ARGUMENT:\nHERE IT IS: ", length(arg), " VERSUS ", length(val))
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
}
args <- names(formals(get(fun))) # here no env = sys.nframe(), inherit = FALSE for get() because fun is a function in the classical scope
if( ! all(arg %in% args)){
tempo.cat <- paste0("ERROR IN ", function.name, ": SOME OF THE STRINGS IN arg ARE NOT ARGUMENTS OF fun\nfun ARGUMENTS: ", paste(args, collapse = " "),"\nPROBLEMATIC STRINGS IN arg: ", paste(arg[ ! arg %in% args], collapse = " "))
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
}
if(sum(sapply(val, FUN = length) > 1) > 43){
tempo.cat <- paste0("ERROR IN ", function.name, ": CANNOT TEST MORE THAN 43 ARGUMENTS IF THEY ALL HAVE AT LEAST 2 VALUES EACH\nHERE THE NUMBER IS: ", sum(sapply(val, FUN = length) > 1))
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
}
if( ! is.null(expect.error)){
if(length(val) != length(expect.error)){
tempo.cat <- paste0("ERROR IN ", function.name, ": LENGTH OF val ARGUMENT MUST BE IDENTICAL TO LENGTH OF expect.error ARGUMENT:\nHERE IT IS: ", length(val), " VERSUS ", length(expect.error))
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
}
}
if( ! is.null(thread.nb) & is.null(res.path)){
tempo.cat <- paste0("ERROR IN ", function.name, ": res.path ARGUMENT MUST BE SPECIFIED IF thread.nb ARGUMENT IS NOT NULL")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
}
if(is.null(res.path) & export == TRUE){
tempo.cat <- paste0("ERROR IN ", function.name, ": res.path ARGUMENT MUST BE SPECIFIED IF export ARGUMENT TRUE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE)
}
if( ! is.null(thread.nb) & export == FALSE){
export <- TRUE
tempo.cat <- paste0("WARNING FROM ", function.name, ": export ARGUMENT CONVERTED TO TRUE BECAUSE thread.nb ARGUMENT IS NOT NULL")
warning(paste0("\n", tempo.cat, "\n"), call. = FALSE)
}
# end second round of checking and data preparation
# package checking
fun_pack(req.package = c("lubridate"), lib.path = lib.path)
if( ! is.null(thread.nb)){
fun_pack(req.package = c("parallel"), lib.path = lib.path)
}
# end package checking
# declaration of special plot functions
sp.plot.fun <- c("fun_gg_scatter", "fun_gg_bar", "fun_gg_boxplot")
# end declaration of special plot functions
# main code
ini.warning.length <- options()$warning.length
warn <- NULL
warn.count <- 0
cat("\nfun_test JOB IGNITION\n")
ini.date <- Sys.time()
ini.time <- as.numeric(ini.date) # time of process begin, converted into seconds
if(export == TRUE){
res.path <- paste0(res.path, "/fun_test_res_", trunc(ini.time))
if(dir.exists(res.path)){
tempo.cat <- paste0("ERROR IN ", function.name, ": FOLDER ALREADY EXISTS\n", res.path, "\nPLEASE RERUN ONCE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}else{
dir.create(res.path)
}
}
total.comp.nb <- prod(sapply(val, FUN = "length"))
cat(paste0("\nTHE TOTAL NUMBER OF TESTS IS: ", total.comp.nb, "\n"))
# creation of the txt instruction that includes several loops
loop.string <- NULL
end.loop.string <- NULL
fun.args <- NULL
fun.args2 <- NULL
error.values <- NULL
arg.values <- "list("
for(i1 in 1:length(arg)){
if(is.null(thread.nb)){
if(length(val[[i1]]) > 1){ # loop only if more than one value in length(val[[i1]])
loop.string <- paste0(loop.string, "for(i", i1, " in 1:", length(val[[i1]]), "){")
end.loop.string <- paste0(end.loop.string, "}")
}
}else{
loop.string <- "for(i in x){"
end.loop.string <- "}"
}
fun.args <- paste0(
fun.args, 
ifelse(i1 == 1, "", ", "), 
arg[i1], 
" = val[[", 
i1, 
"]][[", 
if(is.null(thread.nb)){
if(length(val[[i1]]) > 1){
paste0("i", i1)
}else{
"1" # a unique element in val[[i1]]
}
}else{
paste0("i.list[[", i1, "]][i]")
}, 
"]]"
)
fun.args2 <- paste0(
fun.args2, 
ifelse(i1 == 1, "", ", "), 
arg[i1], 
" = val[[", 
i1, 
"]][[', ", 
if(is.null(thread.nb)){
if(length(val[[i1]]) > 1){
paste0("i", i1)
}else{
"1" # a unique element in val[[i1]]
}
}else{
paste0("i.list[[", i1, "]][i]")
}, 
", ']]"
)
arg.values <- paste0(
arg.values, 
"val[[", i1, "]][[", 
if(is.null(thread.nb)){
if(length(val[[i1]]) > 1){
paste0("i", i1)
}else{
"1" # a unique element in val[[i1]]
}
}else{
paste0("i.list[[", i1, "]][i]")
}, 
"]]", 
ifelse(i1 == length(arg), "", ", ")
)
error.values <- paste0(
error.values, 
ifelse(i1 == 1, "", " | "), 
"expect.error[[", i1, "]][[", 
if(is.null(thread.nb)){
if(length(expect.error[[i1]]) > 1){
paste0("i", i1)
}else{
"1" # a unique element in expect.error[[i1]]
}
}else{
paste0("i.list[[", i1, "]][i]")
}, 
"]]"
)
}
arg.values <- paste0(arg.values, ")")
fun.test <- paste0(fun, "(", fun.args, ")")
fun.test2 <- paste0("paste0('", fun, "(", fun.args2, ")')")
# plot title for special plot functions
if(plot.fun == TRUE){
plot.kind <- "classic"
if(fun %in% sp.plot.fun){
plot.kind <- "special"
if(any(arg %in% "title")){ # this is for the special functions
tempo.match <- regmatches(x = fun.test, m = regexpr(text = fun.test, pattern = "title = .+[,)]"))
tempo.match <- substring(tempo.match , 1, nchar(tempo.match) - 1)
fun.test <- sub(x = fun.test, pattern = tempo.match, replacement = paste0(tempo.match, "\ntempo.title"))
}else{
fun.test <- sub(x = fun.test, pattern = ")$", replacement = ", title = tempo.title)")
}
}
}
# end plot title for special plot functions
kind <- character()
problem <- logical()
expected.error <- logical()
res <- character()
count <- 0
print.count.loop <- 0
plot.count <- 0
if(length(arg) == 1){
data <- data.frame()
}else{ # length(arg) == 0 already tested above
data <- data.frame(t(vector("character", length(arg))), stringsAsFactors = FALSE)[-1, ] # -1 to remove the single row created and to have an empty data frame with length(arg) columns
}
code <- paste(
loop.string, '
count <- count + 1
print.count.loop <- print.count.loop + 1
data <- rbind(data, as.character(sapply(eval(parse(text = arg.values)), FUN = "paste", collapse = " ")), stringsAsFactors = FALSE) # each colum is a test
tempo.try.error <- fun_get_message(data = eval(parse(text = fun.test2)), kind = "error", header = FALSE, env = get(env.name, env = sys.nframe(), inherit = FALSE)) # data argument needs a character string but eval(parse(text = fun.test2)) provides it (eval parse replace the i1, i2, etc., by the correct values, meaning that only val is required in the env.name environment)
tempo.try.warning <- fun_get_message(data = eval(parse(text = fun.test2)), kind = "warning", header = FALSE, env = get(env.name, env = sys.nframe(), inherit = FALSE), print.no = TRUE) # data argument needs a character string but eval(parse(text = fun.test2)) provides it (eval parse replace the i1, i2, etc., by the correct values, meaning that only val is required in the env.name environment)
if( ! is.null(expect.error)){
expected.error <- c(expected.error, eval(parse(text = error.values)))
1868
}
Gael  MILLOT's avatar
Gael MILLOT committed
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
if( ! is.null(tempo.try.error)){
kind <- c(kind, "ERROR")
problem <- c(problem, TRUE)
res <- c(res, tempo.try.error)
}else{
if( ! is.null(tempo.try.warning)){
kind <- c(kind, "WARNING")
problem <- c(problem, FALSE)
res <- c(res, tempo.try.warning)
}else{
kind <- c(kind, "OK")
problem <- c(problem, FALSE)
res <- c(res, "")
}
if(plot.fun == TRUE){
1884
invisible(dev.set(window.nb))
Gael  MILLOT's avatar
Gael MILLOT committed
1885
plot.count <- plot.count + 1
Gael  MILLOT's avatar
Gael MILLOT committed
1886
tempo.title <- paste0("test_", sprintf(paste0("%0", nchar(total.comp.nb), "d"), ifelse(is.null(thread.nb), count, x[count])))
Gael  MILLOT's avatar
Gael MILLOT committed
1887
1888
1889
1890
1891
1892
if(plot.kind == "classic"){
eval(parse(text = fun.test))
tempo <- fun_post_plot(corner.text = tempo.title)
}else if(plot.kind == "special"){
eval(parse(text = fun.test))
}else{
1893
1894
tempo.cat <- paste0("INTERNAL CODE ERROR 1 IN ", function.name, ": CODE HAS TO BE MODIFIED")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
Gael  MILLOT's avatar
Gael MILLOT committed
1895
1896
1897
}
}
}
Gael  MILLOT's avatar
Gael MILLOT committed
1898
1899
if(print.count.loop == print.count){
print.count.loop <- 0
Gael  MILLOT's avatar
Gael MILLOT committed
1900
1901
tempo.time <- as.numeric(Sys.time())
tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - ini.time))
1902
final.loop <- (tempo.time - ini.time) / count * ifelse(is.null(thread.nb), total.comp.nb, length(x)) # expected duration in seconds # intra nb.compar loop lapse: time lapse / cycles done * cycles remaining
Gael  MILLOT's avatar
Gael MILLOT committed
1903
1904
1905
final.exp <- as.POSIXct(final.loop, origin = ini.date)
cat(paste0(ifelse(is.null(thread.nb), "\n", paste0("\nIN PROCESS ", process.id, " | ")), "LOOP ", format(count, big.mark=","), " / ", format(ifelse(is.null(thread.nb), total.comp.nb, length(x)), big.mark=","), " | TIME SPENT: ", tempo.lapse, " | EXPECTED END: ", final.exp))
}
Gael  MILLOT's avatar
Gael MILLOT committed
1906
1907
1908
1909
if(count == ifelse(is.null(thread.nb), total.comp.nb, length(x))){
tempo.time <- as.numeric(Sys.time())
tempo.lapse <- round(lubridate::seconds_to_period(tempo.time - ini.time))
cat(paste0(ifelse(is.null(thread.nb), "\nLOOP PROCESS ENDED | ", paste0("\nPROCESS ", process.id, " ENDED | ")), "LOOP ", format(count, big.mark=","), " / ", format(ifelse(is.null(thread.nb), total.comp.nb, length(x)), big.mark=","), " | TIME SPENT: ", tempo.lapse, "\n\n"))
Gael  MILLOT's avatar
Gael MILLOT committed
1910
}
Gael  MILLOT's avatar
Gael MILLOT committed
1911
1912
', 
end.loop.string
Gael's avatar
tempo    
Gael committed
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
)
# end creation of the txt instruction that includes several loops
if( ! is.null(thread.nb)){
# list of i numbers that will be split
i.list <- vector("list", length(val)) # positions to split in parallel jobs
for(i2 in 1:length(arg)){
if(i2 == 1){
tempo.divisor <- total.comp.nb / length(val[[i2]])
i.list[[i2]] <- rep(1:length(val[[i2]]), each = as.integer(tempo.divisor))
tempo.multi <- length(val[[i2]])
}else{
tempo.divisor <- tempo.divisor / length(val[[i2]])
i.list[[i2]] <- rep(rep(1:length(val[[i2]]), each = as.integer(tempo.divisor)), time = as.integer(tempo.multi))
tempo.multi <- tempo.multi * length(val[[i2]])
}
}
# end list of i numbers that will be split
tempo.cat <- paste0("PARALLELIZATION INITIATED AT: ", ini.date)
cat(paste0("\n", tempo.cat, "\n"))
tempo.thread.nb = parallel::detectCores(all.tests = FALSE, logical = TRUE) # detect the number of threads
if(tempo.thread.nb < thread.nb){
thread.nb <- tempo.thread.nb
}
tempo.cat <- paste0("NUMBER OF THREADS USED: ", thread.nb)
cat(paste0("\n    ", tempo.cat, "\n"))
Clust <- parallel::makeCluster(thread.nb, outfile = paste0(res.path, "/fun_test_parall_log.txt")) # outfile to print or cat during parallelization (only possible in a file, outfile = "" do not work on windows)
tempo.cat <- paste0("SPLIT OF TEST NUMBERS IN PARALLELISATION:")
cat(paste0("\n    ", tempo.cat, "\n"))
cluster.list <- parallel::clusterSplit(Clust, 1:total.comp.nb) # split according to the number of cluster
str(cluster.list) # using print(str()) add a NULL below the result
cat("\n")
paral.output.list <- parallel::clusterApply( # paral.output.list is a list made of thread.nb compartments, each made of n / thread.nb (mat theo column number) compartment. Each compartment receive the corresponding results of fun_permut(), i.e., data (permuted mat1.perm), warning message, cor (final correlation) and count (number of permutations)
cl = Clust,
x = cluster.list,
function.name = function.name, 
instruction = instruction, 
thread.nb = thread.nb, 
print.count = print.count, 
total.comp.nb = total.comp.nb, 
sp.plot.fun = sp.plot.fun,
i.list = i.list, 
fun.tested = fun,
arg.values = arg.values,
fun.test = fun.test,
fun.test2 = fun.test2,
kind = kind,
problem = problem,
res = res,
count = count,
plot.count = plot.count,
data = data,
code = code,
plot.fun = plot.fun, 
res.path = res.path, 
lib.path = lib.path, 
cute.path = cute.path, 
fun = function(
x, 
function.name, 
instruction, 
thread.nb, 
print.count, 
total.comp.nb, 
sp.plot.fun, 
i.list, 
fun.tested, 
arg.values, 
fun.test, 
fun.test2, 
kind, 
problem, 
res, 
count, 
plot.count, 
data, 
code, 
plot.fun, 
res.path, 
lib.path, 
cute.path
){
# check again: very important because another R
process.id <- Sys.getpid()
cat(paste0("\nPROCESS ID ", process.id, " -> TESTS ", x[1], " TO ", x[length(x)], "\n"))
source(cute.path, local = .GlobalEnv)
fun_pack(req.package = "lubridate", lib.path = lib.path, load = TRUE) # load = TRUE to be sure that functions are present in the environment. And this prevent to use R.lib.path argument of fun_python_pack()
# end check again: very important because another R
# plot management
if(plot.fun == TRUE){
pdf(file = paste0(res.path, "/plots_from_fun_test_", x[1], ifelse(length(x) == 1, ".pdf", paste0("-", x[length(x)], ".pdf"))))
}else{
pdf(file = NULL) # send plots into a NULL file, no pdf file created
}
window.nb <- dev.cur()
invisible(dev.set(window.nb))
# end plot management
# new environment
env.name <- paste0("env", ini.time)
if(exists(env.name, where = -1)){ # verify if still ok when fun_test() is inside a function
tempo.cat <- paste0("ERROR IN ", function.name, ": ENVIRONMENT env.name ALREADY EXISTS. PLEASE RERUN ONCE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}else{
assign(env.name, new.env())
assign("val", val, envir = get(env.name, env = sys.nframe(), inherit = FALSE)) # var replaced by val
}
# end new environment
ini.date <- Sys.time()
ini.time <- as.numeric(ini.date) # time of process begin, converted into 
print.count.loop <- 0
suppressMessages(suppressWarnings(eval(parse(text = code))))
colnames(data) <- arg
if( ! is.null(expect.error)){
data <- data.frame(data, kind = kind, problem = problem, expected.error = expected.error, message = res, stringsAsFactors = FALSE)
}else{
data <- data.frame(data, kind = kind, problem = problem, message = res, stringsAsFactors = FALSE)
}
row.names(data) <- paste0("test_", sprintf(paste0("%0", nchar(total.comp.nb), "d"), x))
sys.info <- sessionInfo()
sys.info$loadedOnly <- sys.info$loadedOnly[order(names(sys.info$loadedOnly))] # sort the packages
invisible(dev.off(window.nb))
rm(env.name) # optional, because should disappear at the end of the function execution
# output
output <- list(fun = fun, data = data, instruction = instruction, sys.info = sys.info)
save(output, file = paste0(res.path, "/fun_test_", x[1], ifelse(length(x) == 1, ".RData", paste0("-", x[length(x)], ".RData"))))
if(plot.fun == TRUE & plot.count == 0){
warn.count <- warn.count + 1
tempo.warn <- paste0("(", warn.count,") IN PROCESS ", process.id, ": NO PDF PLOT BECAUSE ONLY ERRORS REPORTED")
warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
file.remove(paste0(res.path, "/plots_from_fun_test_", x[1], ifelse(length(x) == 1, ".pdf", paste0("-", x[length(x)], ".pdf"))))
}
table.out <- as.matrix(output$data)
# table.out[table.out == ""] <- " " # does not work # because otherwise read.table() converts "" into NA
table.out <- gsub(table.out, pattern = "\n", replacement = " ")
write.table(table.out, file = paste0(res.path, "/table_from_fun_test_", x[1], ifelse(length(x) == 1, ".txt", paste0("-", x[length(x)], ".txt"))), row.names = TRUE, col.names = NA, append = FALSE, quote = FALSE, sep = "\t", eol = "\n")
}
)
parallel::stopCluster(Clust)
# txt files assembly
if(length(cluster.list) > 1){
for(i2 in 1:length(cluster.list)){
tempo.name <- paste0(res.path, "/table_from_fun_test_", min(cluster.list[[i2]], na.rm = TRUE), ifelse(length(cluster.list[[i2]]) == 1, ".txt", paste0("-", max(cluster.list[[i2]], na.rm = TRUE), ".txt")))
tempo <- read.table(file = tempo.name, header = TRUE, stringsAsFactors = FALSE, sep = "\t", row.names = 1, comment.char = "", colClasses = "character") #  row.names = 1 (1st column) because now read.table() adds a NA in the header if the header starts by a tabulation, comment.char = "" because colors with #, colClasses = "character" otherwise convert "" (from NULL) into NA
file.remove(tempo.name)
if(i2 == 1){
final.file <- tempo
}else{
final.file <- rbind(final.file, tempo, stringsAsFactors = TRUE)
}
}
write.table(final.file, file = paste0(res.path, "/table_from_fun_test_1-", total.comp.nb, ".txt"), row.names = TRUE, col.names = NA, append = FALSE, quote = FALSE, sep = "\t", eol = "\n")
if( ! is.null(expect.error)){
final.file <- final.file[ ! final.file$problem == final.file$expected.error, ]
if(nrow(final.file) == 0){
cat(paste0("NO DISCREPANCY BETWEEN EXPECTED AND OBSERVED ERRORS\n\n"))
}else{
cat(paste0("DISCREPANCIES BETWEEN EXPECTED AND OBSERVED ERRORS (SEE THE discrepancy_table_from_fun_test_1-", total.comp.nb, ".txt FILE)\n\n"))
write.table(final.file, file = paste0(res.path, "/discrepancy_table_from_fun_test_1-", total.comp.nb, ".txt"), row.names = TRUE, col.names = NA, append = FALSE, quote = FALSE, sep = "\t", eol = "\n")
}
}
}
# end txt files assembly
}else{
# plot management
if(plot.fun == TRUE){
pdf(file = paste0(res.path, "/plots_from_fun_test_1", ifelse(total.comp.nb == 1, ".pdf", paste0("-", total.comp.nb, ".pdf"))))
}else{
pdf(file = NULL) # send plots into a NULL file, no pdf file created
}
window.nb <- dev.cur()
invisible(dev.set(window.nb))
# end plot management
# new environment
env.name <- paste0("env", ini.time)
if(exists(env.name, where = -1)){
tempo.cat <- paste0("ERROR IN ", function.name, ": ENVIRONMENT env.name ALREADY EXISTS. PLEASE RERUN ONCE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}else{
assign(env.name, new.env())
assign("val", val, envir = get(env.name, env = sys.nframe(), inherit = FALSE)) # var replaced by val
}
# end new environment
suppressMessages(suppressWarnings(eval(parse(text = code))))
colnames(data) <- arg
expect.data <- data.frame()
if( ! is.null(expect.error)){
data <- data.frame(data, kind = kind, problem = problem, expected.error = expected.error, message = res, stringsAsFactors = FALSE)
}else{
data <- data.frame(data, kind = kind, problem = problem, message = res, stringsAsFactors = FALSE)
}
row.names(data) <- paste0("test_", sprintf(paste0("%0", nchar(total.comp.nb), "d"), 1:total.comp.nb))
sys.info <- sessionInfo()
sys.info$loadedOnly <- sys.info$loadedOnly[order(names(sys.info$loadedOnly))] # sort the packages
invisible(dev.off(window.nb))
rm(env.name) # optional, because should disappear at the end of the function execution
# output
output <- list(fun = fun, data = data, instruction = instruction, sys.info = sys.info)
if(plot.fun == TRUE & plot.count == 0){
warn.count <- warn.count + 1
tempo.warn <- paste0("(", warn.count,") NO PDF PLOT BECAUSE ONLY ERRORS REPORTED")
warn <- paste0(ifelse(is.null(warn), tempo.warn, paste0(warn, "\n\n", tempo.warn)))
file.remove(paste0(res.path, "/plots_from_fun_test_1", ifelse(total.comp.nb == 1, ".pdf", paste0("-", total.comp.nb, ".pdf"))))
}
if( ! is.null(expect.error)){
expect.data <- output$data[ ! output$data$problem == output$data$expected.error, ]
if(nrow(expect.data) == 0){
cat(paste0("NO DISCREPANCY BETWEEN EXPECTED AND OBSERVED ERRORS\n\n"))
}else{
cat(paste0("DISCREPANCIES BETWEEN EXPECTED AND OBSERVED ERRORS (SEE THE ", if(export == TRUE){paste0("discrepancy_table_from_fun_test_1", ifelse(total.comp.nb == 1, "", paste0("-", total.comp.nb)), ".txt FILE")}else{"$data RESULT"}, ")\n\n"))
if(export == TRUE){
expect.data <- as.matrix(expect.data)
expect.data <- gsub(expect.data, pattern = "\n", replacement = "  ")
write.table(expect.data, file = paste0(res.path, "/discrepancy_table_from_fun_test_1", ifelse(total.comp.nb == 1, ".txt", paste0("-", total.comp.nb, ".txt"))), row.names = TRUE, col.names = NA, append = FALSE, quote = FALSE, sep = "\t", eol = "\n")
}
}
}
if( ! is.null(warn)){
options(warning.length = 8170)
on.exit(warning(paste0("FROM ", function.name, ":\n\n", warn), call. = FALSE))
on.exit(exp = options(warning.length = ini.warning.length), add = TRUE)
}
if(export == TRUE){
save(output, file = paste0(res.path, "/fun_test_1", ifelse(total.comp.nb == 1, ".RData", paste0("-", total.comp.nb, ".RData"))))
table.out <- as.matrix(output$data)
table.out <- gsub(table.out, pattern = "\n", replacement = "  ")
write.table(table.out, file = paste0(res.path, "/table_from_fun_test_1", ifelse(total.comp.nb == 1, ".txt", paste0("-", total.comp.nb, ".txt"))), row.names = TRUE, col.names = NA, append = FALSE, quote = FALSE, sep = "\t", eol = "\n")
}else{
return(output)
}
}
end.date <- Sys.time()
end.time <- as.numeric(end.date)
total.lapse <- round(lubridate::seconds_to_period(end.time - ini.time))
cat(paste0("fun_test JOB END\n\nTIME: ", end.date, "\n\nTOTAL TIME LAPSE: ", total.lapse, "\n\n\n"))
Gael  MILLOT's avatar
Gael MILLOT committed
2146
}
Gael  MILLOT's avatar
Gael MILLOT committed
2147

Gael  MILLOT's avatar
Gael MILLOT committed
2148

Gael  MILLOT's avatar
Gael MILLOT committed
2149
2150
2151
2152
2153
2154
2155
2156
################ Object modification


######## fun_name_change() #### check a vector of character strings and modify any string if present in another vector


# Check OK: clear to go Apollo
fun_name_change <- function(data1, data2, added.string = "_modif"){
Gael's avatar
tempo    
Gael committed
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
# AIM
# this function allow to check if a vector of character strings, like column names of a data frame, has elements present in another vector (vector of reserved words or column names of another data frame before merging)
# ARGUMENTS
# data1: vector of character strings to check and modify
# data2: reference vector of character strings
# added.string: string added at the end of the modified string in data1 if present in data2
# RETURN
# a list containing
# $data: the modified data1 (in the same order as in the initial data1)
# $ini: the initial elements before modification. NULL if no modification
# $post: the modified elements in the same order as in ini. NULL if no modification
# REQUIRED PACKAGES
# none
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# EXAMPLES
# obs1 <- c("A", "B", "C", "D") ; obs2 <- c("A", "C") ; fun_name_change(obs1, obs2)
# obs1 <- c("A", "B", "C", "C_modif1", "D") ; obs2 <- c("A", "A_modif1", "C") ; fun_name_change(obs1, obs2) # the function checks that the new names are neither in obs1 nor in obs2 (increment the number after the added string)
# DEBUGGING
# data1 = c("A", "B", "C", "D") ; data2 <- c("A", "C") ; added.string = "_modif" # for function debugging
# function name
function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
# end function name
# required function checking
if(length(utils::find("fun_check", mode = "function")) == 0){
tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end required function checking
# argument checking
arg.check <- NULL #
text.check <- NULL #
checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
tempo <- fun_check(data = data1, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
tempo <- fun_check(data = data2, class = "vector", mode = "character", fun.name = function.name) ; eval(ee)
tempo <- fun_check(data = added.string, class = "vector", mode = "character", length = 1, fun.name = function.name) ; eval(ee)
if(any(arg.check) == TRUE){
stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
}
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
# end argument checking
# main code
ini <- NULL
post <- NULL
if(any(data1 %in% data2)){
tempo.names <- data1[data1 %in% data2]
ini <- NULL
post <- NULL
for(i2 in 1:length(tempo.names)){
count <- 0
tempo <- tempo.names[i2]
while(any(tempo %in% data2) | any(tempo %in% data1)){
count <- count + 1
tempo <- paste0(tempo.names[i2], "_modif", count)
}
data1[data1 %in% tempo.names[i2]] <- paste0(tempo.names[i2], "_modif", count)
if(count != 0){
ini <- c(ini, tempo.names[i2])
post <- c(post, paste0(tempo.names[i2], "_modif", count))
}
}
data <- data1
}else{
data <- data1
}
output <- list(data = data, ini = ini, post = post)
return(output)
Gael  MILLOT's avatar
Gael MILLOT committed
2225
2226
2227
}


Gael  MILLOT's avatar
Gael MILLOT committed
2228
######## fun_df_remod() #### remodeling a data frame to have column name as a qualitative values and vice-versa
Gael  MILLOT's avatar
Gael MILLOT committed
2229
2230


Gael  MILLOT's avatar
Gael MILLOT committed
2231
# Check OK: clear to go Apollo
Gael's avatar
tempo    
Gael committed
2232
fun_df_remod <- function(
Gael's avatar
tempo    
Gael committed
2233
2234
2235
data, 
quanti.col.name = "quanti", 
quali.col.name = "quali"
Gael's avatar
tempo    
Gael committed
2236
){
Gael's avatar
tempo    
Gael committed
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
# AIM
# if the data frame is made of numeric columns, a new data frame is created, with the 1st column gathering all the numeric values, and the 2nd column being the name of the columns of the initial data frame. If row names were present in the initial data frame, then a new ini_rowname column is added with the names of the rows

 
# If the data frame is made of one numeric column and one character or factor column, a new data frame is created, with the new columns corresponding to the split numeric values (according to the character column). NA are added a the end of each column to have the same number of rows. BEWARE: in such data frame, rows are not individuals. This means that in the example below, values 10 and 20 are associated on the same row but that means nothing in term of association

 

# ARGUMENTS
# data: data frame to convert
# quanti.col.name: optional name for the quanti column of the new data frame
# quali.col.name: optional name for the quali column of the new data frame
# RETURN
# the modified data frame
# REQUIRED PACKAGES
# none
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# EXAMPLES
# obs <- data.frame(col1 = (1:4)*10, col2 = c("A", "B", "A", "A"), stringsAsFactors = TRUE) ; obs ; fun_df_remod(obs)
# obs <- data.frame(col1 = (1:4)*10, col2 = 5:8, stringsAsFactors = TRUE) ; obs ; fun_df_remod(obs, quanti.col.name = "quanti", quali.col.name = "quali")
# obs <- data.frame(col1 = (1:4)*10, col2 = 5:8, stringsAsFactors = TRUE) ; rownames(obs) <- paste0("row", 1:4) ; obs ; fun_df_remod(obs, quanti.col.name = "quanti", quali.col.name = "quali")
# DEBUGGING
# data = data.frame(a = 1:3, b = 4:6, stringsAsFactors = TRUE) ; quanti.col.name = "quanti" ; quali.col.name = "quali" # for function debugging
# data = data.frame(a = 1:3, b = 4:6, c = 11:13, stringsAsFactors = TRUE) ; quanti.col.name = "quanti" ; quali.col.name = "quali" # for function debugging
# data = data.frame(a = 1:3, b = letters[1:3], stringsAsFactors = TRUE) ; quanti.col.name = "quanti" ; quali.col.name = "quali" # for function debugging
# data = data.frame(a = 1:3, b = letters[1:3], stringsAsFactors = TRUE) ; quanti.col.name = "TEST" ; quali.col.name = "quali" # for function debugging
# data = data.frame(b = letters[1:3], a = 1:3, stringsAsFactors = TRUE) ; quanti.col.name = "quanti" ; quali.col.name = "quali" # for function debugging
# data = data.frame(b = c("e", "e", "h"), a = 1:3, stringsAsFactors = TRUE) ; quanti.col.name = "quanti" ; quali.col.name = "quali" # for function debugging
# function name
function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
# end function name
# required function checking
if(length(utils::find("fun_check", mode = "function")) == 0){
tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end required function checking
# argument checking
# argument checking without fun_check()
if( ! any(class(data) %in% "data.frame")){
tempo.cat <- paste0("ERROR IN ", function.name, ": THE data ARGUMENT MUST BE A DATA FRAME")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end argument checking without fun_check()
# argument checking with fun_check()
arg.check <- NULL #
text.check <- NULL #
checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
tempo <- fun_check(data = quanti.col.name, class = "character", length = 1, fun.name = function.name) ; eval(ee)
tempo <- fun_check(data = quali.col.name, class = "character", length = 1, fun.name = function.name) ; eval(ee)
if(any(arg.check) == TRUE){
stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
}
# end argument checking with fun_check()
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
# end argument checking
# main code
tempo.factor <- unlist(lapply(data, class))
for(i in 1:length(tempo.factor)){ # convert factor columns as character
if(all(tempo.factor[i] == "factor")){
data[, i] <- as.character(data[, i])
}
}
tempo.factor <- unlist(lapply(data, mode))
if(length(data) == 2){
if( ! ((mode(data[, 1]) == "character" & mode(data[, 2]) == "numeric") | mode(data[, 2]) == "character" & mode(data[, 1]) == "numeric" | mode(data[, 2]) == "numeric" & mode(data[, 1]) == "numeric") ){
tempo.cat <- paste0("ERROR IN ", function.name, ": IF data ARGUMENT IS A DATA FRAME MADE OF 2 COLUMNS, EITHER A COLUMN MUST BE NUMERIC AND THE OTHER CHARACTER, OR THE TWO COLUMNS MUST BE NUMERIC")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if((mode(data[, 1]) == "character" | mode(data[, 2]) == "character") & (quanti.col.name != "quanti" | quali.col.name != "quali")){
tempo.cat <- paste0("ERROR IN ", function.name, ": IMPROPER quanti.col.name OR quali.col.name RESETTINGS. THESE ARGUMENTS ARE RESERVED FOR DATA FRAMES MADE OF n NUMERIC COLUMNS ONLY")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
}else{
if( ! all(tempo.factor %in% "numeric")){
tempo.cat <- paste0("ERROR IN ", function.name, ": IF data ARGUMENT IS A DATA FRAME MADE OF ONE COLUMN, OR MORE THAN 2 COLUMNS, THESE COLUMNS MUST BE NUMERIC")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
}
if(( ! any(tempo.factor %in% "character")) & is.null(names(data))){
tempo.cat <- paste0("ERROR IN ", function.name, ": NUMERIC DATA FRAME in the data ARGUMENT MUST HAVE COLUMN NAMES")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if(all(tempo.factor %in% "numeric")){ # transfo 1
quanti <- NULL
for(i in 1:length(data)){
quanti <-c(quanti, data[, i])
}
quali <- rep(names(data), each = nrow(data))
output.data <- data.frame(quanti, quali, stringsAsFactors = TRUE)
names(output.data) <- c(quanti.col.name, quali.col.name)
# add the ini_rowname column
ini.rownames <- rownames(data)
tempo.data <- data
rownames(tempo.data) <- NULL
null.rownames <- (tempo.data)
if( ! identical(ini.rownames, null.rownames)){
ini_rowname <- rep(ini.rownames, times = ncol(data))
output.data <- cbind(output.data, ini_rowname, stringsAsFactors = TRUE)
}
}else{ # transfo 2
if(class(data[, 1]) == "character"){
data <- cbind(data[2], data[1], stringsAsFactors = TRUE)
}
nc.max <- max(table(data[, 2])) # effectif maximum des classes
nb.na <- nc.max - table(data[,2]) # nombre de NA à ajouter pour réaliser la data frame
tempo<-split(data[, 1], data[, 2])
for(i in 1:length(tempo)){tempo[[i]] <- append(tempo[[i]], rep(NA, nb.na[i]))} # des NA doivent être ajoutés lorsque les effectifs sont différents entre les classes. C'est uniquement pour que chaque colonne ait le même nombre de lignes
output.data<-data.frame(tempo, stringsAsFactors = TRUE)
}
return(output.data)
Gael  MILLOT's avatar
Gael MILLOT committed
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
}




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


# Check OK: clear to go Apollo
fun_round <- function(data, dec.nb = 2, after.lead.zero = TRUE){
Gael's avatar
tempo    
Gael committed
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
# AIM
# round a vector of values, if decimal, with the desired number of decimal digits after the decimal leading zeros
# WARNINGS
# Work well with numbers as character strings, but not always with numerical numbers because of the floating point
# Numeric values are really truncated from a part of their decimal digits, whatever options(digits) settings
# See ?.Machine or https://stackoverflow.com/questions/5173692/how-to-return-number-of-decimal-places-in-r, with the interexting formula: abs(x - round(x)) > .Machine$double.eps^0.5
# ARGUMENTS
# data: a vector of numbers (numeric or character mode)
# dec.nb: number of required decimal digits
# after.lead.zero: logical. If FALSE, rounding is performed for all the decimal numbers, whatever the leading zeros (e.g., 0.123 -> 0.12 and 0.00128 -> 0.00). If TRUE, dec.nb are taken after the leading zeros (e.g., 0.123 -> 0.12 and 0.00128 -> 0.0013)
# RETURN
# the modified vector
# REQUIRED PACKAGES
# none
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# EXAMPLES
# ini.options <- options()$digits ; options(digits = 8) ; cat(fun_round(data = c(NA, 10, 100.001, 333.0001254, 12312.1235), dec.nb = 2, after.lead.zero = FALSE), "\n\n") ; options(digits = ini.options)
# ini.options <- options()$digits ; options(digits = 8) ; cat(fun_round(data = c(NA, 10, 100.001, 333.0001254, 12312.1235), dec.nb = 2, after.lead.zero = TRUE), "\n\n") ; options(digits = ini.options)
# ini.options <- options()$digits ; options(digits = 8) ; cat(fun_round(data = c(NA, "10", "100.001", "333.0001254", "12312.1235"), dec.nb = 2, after.lead.zero = FALSE), "\n\n") ; options(digits = ini.options)
# ini.options <- options()$digits ; options(digits = 8) ; cat(fun_round(data = c(NA, "10", "100.001", "333.0001254", "12312.1235"), dec.nb = 2, after.lead.zero = TRUE), "\n\n") ; options(digits = ini.options)
# DEBUGGING
# data = data = c(10, 100.001, 333.0001254, 12312.1235) ; dec.nb = 2 ; after.lead.zero = FALSE # # for function debugging
# data = data = c("10", "100.001", "333.0001254", "12312.1235") ; dec.nb = 2 ; after.lead.zero = TRUE # # for function debugging
# function name
function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
# end function name
# required function checking
if(length(utils::find("fun_check", mode = "function")) == 0){
tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end required function checking
# argument checking
# argument checking without fun_check()
if( ! (all(typeof(data) == "character") | all(typeof(data) == "double") | all(typeof(data) == "integer"))){
tempo.cat <- paste0("ERROR IN ", function.name, ": data ARGUMENT MUST BE A VECTOR OF NUMBERS (IN NUMERIC OR CHARACTER MODE)")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end argument checking without fun_check()
# argument checking with fun_check()
arg.check <- NULL #
text.check <- NULL #
checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
tempo <- fun_check(data = data, class = "vector", na.contain = TRUE, fun.name = function.name) ; eval(ee)
tempo <- fun_check(data = dec.nb, class = "vector", typeof = "integer", length = 1, double.as.integer.allowed = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
tempo <- fun_check(data = after.lead.zero, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
if(any(arg.check) == TRUE){
stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
}
# end argument checking with fun_check()
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
# end argument checking
# main code
tempo <- grepl(x = data, pattern = "\\.") # detection of decimal numbers
ini.mode <- mode(data)
data <- as.character(data) # to really truncate decimal digits
for(i in 1:length(data)){ # scan all the numbers of the vector
if(tempo[i] == TRUE){ # means decimal number
if(after.lead.zero == TRUE){
zero.pos <- unlist(gregexpr(text=data[i], pattern = 0)) # recover all the position of the zeros in the number. -1 if no zeros (do not record the leading and trailing zeros)
}else{
zero.pos <- -1 # -1 as if no zero
}
dot.pos <- unlist(gregexpr(text=data[i], pattern = "\\.")) # recover all the position of the zeros in the number
digit.pos <- unlist(gregexpr(text=data[i], pattern = "[[:digit:]]")) # recover all the position of the digits in the number
dec.pos <- digit.pos[digit.pos > dot.pos]
count <- 0
while((dot.pos + count + 1) %in% zero.pos & (dot.pos + count + 1) <= max(dec.pos) & (count + dec.nb) < length(dec.pos)){ # count the number of leading zeros in the decimal part
count <- count + 1
}
data[i] <- formatC(as.numeric(data[i]), digits = (count + dec.nb), format = "f")
}
}
if(ini.mode != "character"){
data <- as.numeric(data)
}
return(data)
Gael  MILLOT's avatar
Gael MILLOT committed
2439
2440
2441
2442
2443
2444
2445
2446
}


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


# Check OK: clear to go Apollo
fun_mat_rotate <- function(data){
Gael's avatar
tempo    
Gael committed
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
# AIM
# 90° clockwise matrix rotation
# applied twice, the function provide the mirror matrix, according to vertical and horizontal symmetry
# ARGUMENTS
# data: matrix (matrix class)
# RETURN
# the modified matrix
# REQUIRED PACKAGES
# none
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# EXAMPLES
# obs <- matrix(1:10, ncol = 1) ; obs ; fun_mat_rotate(obs)
# obs <- matrix(LETTERS[1:10], ncol = 5) ; obs ; fun_mat_rotate(obs)
# DEBUGGING
# data = matrix(1:10, ncol = 1)
# function name
function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
# end function name
# required function checking
if(length(utils::find("fun_check", mode = "function")) == 0){
tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end required function checking
# argument checking
arg.check <- NULL #
text.check <- NULL #
checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
tempo <- fun_check(data = data, class = "matrix", fun.name = function.name) ; eval(ee)
if(any(arg.check) == TRUE){
stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
}
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
# end argument checking
# main code
for (i in 1:ncol(data)){data[,i] <- rev(data[,i])}
data <- t(data)
return(data)
Gael  MILLOT's avatar
Gael MILLOT committed
2487
2488
2489
2490
2491
2492
2493
}


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


# Check OK: clear to go Apollo
Gael's avatar
tempo    
Gael committed
2494
fun_mat_num2color <- function(
Gael's avatar
tempo    
Gael committed
2495
2496
2497
2498
2499
2500
mat1, 
mat.hsv.h = TRUE, 
notch = 1, 
s = 1, 
v = 1, 
forced.color = NULL
Gael's avatar
tempo    
Gael committed
2501
){
Gael's avatar
tempo    
Gael committed
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
# AIM
# convert a matrix made of numbers into a hexadecimal matrix for rgb colorization
# ARGUMENTS:
# mat1: matrix 1 of non negative numerical values that has to be colored (matrix class). NA allowed
# mat.hsv.h: logical. Is mat1 the h of hsv colors ? (if TRUE, mat1 must be between zero and 1)
# notch: single value between 0 and 1 to shift the successive colors on the hsv circle by + notch
# s: s argument of hsv(). Must be between 0 and 1
# v: v argument of hsv(). Must be between 0 and 1
# forced.color: Must be NULL or hexadecimal color code or name given by colors(). The first minimal values of mat1 will be these colors. All the color of mat1 can be forced using this argument
# RETURN
# a list containing:
# $mat1.name: name of mat1
# $colored.mat: colors of mat1 in hexa
# $problem: logical. Is any colors of forced.color overlap the colors designed by the function. NULL if forced.color = NULL
# $text.problem: text when overlapping colors. NULL if forced.color = NULL or problem == FALSE
# REQUIRED PACKAGES
# none
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# EXAMPLES
# mat1 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2) ; dimnames(mat1) <- list(LETTERS[1:4], letters[1:2]) ; fun_mat_num2color(mat1, mat.hsv.h = FALSE, notch = 1, s = 1, v = 1, forced.color = NULL)
# DEBUGGING
# mat1 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2) ; dimnames(mat1) <- list(LETTERS[1:4], letters[1:2]); mat.hsv.h = FALSE ; notch = 1 ; s = 1 ; v = 1 ; forced.color = c(hsv(1,1,1), hsv(0,0,0)) # for function debugging
# function name
function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
# end function name
# required function checking
if(length(utils::find("fun_check", mode = "function")) == 0){
tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end required function checking
# argument checking
# argument checking with fun_check()
arg.check <- NULL #
text.check <- NULL #
checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
tempo <- fun_check(data = mat1, mode = "numeric", class = "matrix", na.contain = TRUE, neg.values = FALSE, fun.name = function.name) ; eval(ee)
tempo <- fun_check(data = mat.hsv.h, class = "logical", length = 1, fun.name = function.name) ; eval(ee)
tempo <- fun_check(data = notch, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
tempo <- fun_check(data = s, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
tempo <- fun_check(data = v, class = "vector", mode = "numeric", length = 1, prop = TRUE, fun.name = function.name) ; eval(ee)
if(any(arg.check) == TRUE){
stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
}
# end argument checking with fun_check()
# argument checking without fun_check()
if(mat.hsv.h == TRUE & fun_check(data = mat1, mode = "numeric", prop = TRUE)$problem == TRUE){
tempo.cat <- paste0("ERROR IN ", function.name, ": mat1 ARGUMENT MUST BE A MATRIX OF PROPORTIONS SINCE THE mat.hsv.h ARGUMENT IS SET TO TRUE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if( ! is.null(forced.color)){
tempo <- fun_check(data = forced.color, class = "character")
if(any(tempo$problem == TRUE)){
paste0("\n\n================\n\n", paste(tempo$text[tempo$problem], collapse = "\n"))
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if( ! all(forced.color %in% colors() | grepl(pattern = "^#", forced.color))){ # check that all strings of forced.color start by #
tempo.cat <- paste0("ERROR IN ", function.name, ": forced.color ARGUMENT MUST BE A HEXADECIMAL COLOR VECTOR STARTING BY # AND/OR COLOR NAMES GIVEN BY colors()")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
}
# end argument checking without fun_check()
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
# end argument checking
# main code
problem <- NULL
text.problem <- NULL
mat1.name <- deparse(substitute(mat1))
# change the scale of the plotted matrix
if(mat.hsv.h == TRUE){
if(any(min(mat1, na.rm = TRUE) < 0 | max(mat1, na.rm = TRUE) > 1, na.rm = TRUE)){
tempo.cat <- paste0("ERROR IN ", function.name, ": mat1 MUST BE MADE OF VALUES BETWEEN 0 AND 1 BECAUSE mat.hsv.h ARGUMENT SET TO TRUE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
}else{
if(any(mat1 - floor(mat1) > 0, na.rm = TRUE) | any(mat1 == 0, na.rm = TRUE)){ # no need of isTRUE(all.equal()) because we do not require approx here but strictly 0, thus == is ok
tempo.cat <- paste0("ERROR IN ", function.name, ": mat1 MUST BE MADE OF INTEGER VALUES WITHOUT 0 BECAUSE mat.hsv.h ARGUMENT SET TO FALSE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}else{
mat1 <- mat1 / max(mat1, na.rm = TRUE)
}
}
if(notch != 1){
different.color <- unique(as.vector(mat1))
different.color <- different.color[ ! is.na(different.color)]
tempo.different.color <- different.color + c(0, cumsum(rep(notch, length(different.color) - 1)))
tempo.different.color <- tempo.different.color - floor(tempo.different.color)
if(any(duplicated(tempo.different.color) == TRUE)){
tempo.cat <- paste0("ERROR IN ", function.name, ": DUPLICATED VALUES AFTER USING notch (", paste(tempo.different.color[duplicated(tempo.different.color)], collapse = " "), "). TRY ANOTHER notch VALUE")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}else if(length(different.color) != length(tempo.different.color)){
tempo.cat <- paste0("ERROR IN ", function.name, ": LENGTH OF different.color (", paste(different.color, collapse = " "), ") DIFFERENT FROM LENGTH OF tempo.different.color (", paste(tempo.different.color, collapse = " "), ")")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}else{
for(i in 1:length(different.color)){
mat1[mat1 == different.color[i]] <- tempo.different.color[i] # no need of isTRUE(all.equal()) because different.color comes from mat1
}
}
}
if( ! is.null(forced.color)){
hexa.values.to.change <- hsv(unique(sort(mat1))[1:length(forced.color)], s, v)
}
mat1[ ! is.na(mat1)] <- hsv(mat1[ ! is.na(mat1)], s, v)
if( ! is.null(forced.color)){
if(any(forced.color %in% mat1, na.rm = TRUE)){
problem <- TRUE
text.problem <- paste0("THE FOLLOWING COLORS WHERE INTRODUCED USING forced.color BUT WHERE ALREADY PRESENT IN THE COLORED MATRIX :", paste(forced.color[forced.color %in% mat1], collapse = " "))
}else{
problem <- FALSE
}
for(i in 1:length(hexa.values.to.change)){
if( ! any(mat1 == hexa.values.to.change[i], na.rm = TRUE)){# no need of isTRUE(all.equal()) because character
tempo.cat <- paste0("ERROR IN ", function.name, ": THE ", hexa.values.to.change[i], " VALUE FROM hexa.values.to.change IS NOT REPRESENTED IN mat1 : ", paste(unique(as.vector(mat1)), collapse = " "))
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}else{
mat1[which(mat1 == hexa.values.to.change[i])] <- forced.color[i] # no need of isTRUE(all.equal()) because character
}
}
}
output <- list(mat1.name = mat1.name, colored.mat = mat1, problem = problem, text.problem = text.problem)
return(output)
Gael  MILLOT's avatar
Gael MILLOT committed
2625
2626
2627
2628
2629
2630
2631
2632
}


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


# Check OK: clear to go Apollo
fun_mat_op <- function(mat.list, kind.of.operation = "+"){
Gael's avatar
tempo    
Gael committed
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
# AIM
# assemble several matrices of same dimensions by performing by case operation. For instance add the value of all the case 1 (row1 & column1) of the matrices and put it in the case 1 of a new matrix M, add the value of all the case 2 (row2 & column1) of the matrices and put it in the case 2 of a new matrix M, etc.
 
# c: case
# i: row number
# j: column number
# k: matrix number
# z: number of matrices
# ARGUMENTS:
# mat.list: list of matrices
# kind.of.operation: either "+" (by case addition), "-" (by case subtraction) or "*" (by case multiplication)
# RETURN
# the assembled matrix, with row and/or column names only if all the matrices have identical row/column names
# REQUIRED PACKAGES
# none
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# fun_comp_2d()
# EXAMPLES
# mat1 = matrix(c(1,1,1,2,1,5,9,8), ncol = 2) ; mat2 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2) ; fun_mat_op(mat.list = list(mat1, mat2), kind.of.operation = "+")
# mat1 = matrix(c(1,1,1,2,1,5,9,8), ncol = 2, dimnames = list(LETTERS[1:4], letters[1:2])) ; mat2 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2, dimnames = list(LETTERS[1:4], letters[1:2])) ; fun_mat_op(mat.list = list(mat1, mat2), kind.of.operation = "*")
# mat1 = matrix(c(1,1,1,2,1,5,9,8), ncol = 2, dimnames = list(LETTERS[1:4], c(NA, NA))) ; mat2 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2, dimnames = list(LETTERS[1:4], letters[1:2])) ; fun_mat_op(mat.list = list(mat1, mat2), kind.of.operation = "-")
# mat1 = matrix(c(1,1,1,2,1,5,9,8), ncol = 2, dimnames = list(c("A1", "A2", "A3", "A4"), letters[1:2])) ; mat2 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2, dimnames = list(LETTERS[1:4], letters[1:2])) ; mat3 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2, dimnames = list(LETTERS[1:4], letters[1:2])) ; fun_mat_op(mat.list = list(mat1, mat2, mat3), kind.of.operation = "+")
# DEBUGGING
# mat1 = matrix(c(1,1,1,2,1,5,9,8), ncol = 2) ; mat2 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2) ; mat.list = list(mat1, mat2) ; kind.of.operation = "+" # for function debugging
# mat1 = matrix(c(1,1,1,2,1,5,9,8), ncol = 2, dimnames = list(LETTERS[1:4], c(NA, NA))) ; mat2 = matrix(c(1,1,1,2,1,5,9,NA), ncol = 2, dimnames = list(LETTERS[1:4], letters[1:2])) ; mat.list = list(mat1, mat2) ; kind.of.operation = "*" # for function debugging
# function name
function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
# end function name
# required function checking
if(length(utils::find("fun_check", mode = "function")) == 0){
tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if(length(utils::find("fun_check", mode = "function")) == 0){
tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_comp_2d() FUNCTION IS MISSING IN THE R ENVIRONMENT")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end required function checking
# argument checking
# argument checking with fun_check()
arg.check <- NULL #
text.check <- NULL #
checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
tempo <- fun_check(data = mat.list, class = "list", fun.name = function.name) ; eval(ee)
tempo <- fun_check(data = kind.of.operation, options = c("+", "-", "*"), length = 1, fun.name = function.name) ; eval(ee)
if(any(arg.check) == TRUE){
stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
}
# end argument checking with fun_check()
# argument checking without fun_check()
if(length(mat.list) < 2){
tempo.cat <- paste0("ERROR IN ", function.name, ": mat.list ARGUMENT MUST BE A LIST CONTAINING AT LEAST 2 MATRICES")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
for(i1 in 1:length(mat.list)){
tempo <- fun_check(data = mat.list[[i1]], class = "matrix", mode = "numeric", na.contain = TRUE)
if(tempo$problem == TRUE){
tempo.cat <- paste0("ERROR IN ", function.name, ": ELEMENT ", i1, " OF mat.list ARGUMENT MUST BE A NUMERIC MATRIX")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
}
ident.row.names <- TRUE
ident.col.names <- TRUE
for(i1 in 2:length(mat.list)){
tempo <- fun_comp_2d(data1 = mat.list[[1]], data2 = mat.list[[i1]])
if(tempo$same.dim == FALSE){
tempo.cat <- paste0("ERROR IN ", function.name, ": MATRIX ", i1, " OF mat.list ARGUMENT MUST HAVE THE SAME DIMENSION (", paste(dim(mat.list[[i1]]), collapse = " "), ") THAN THE MATRIX 1 IN mat.list (", paste(dim(mat.list[[1]]), collapse = " "), ")")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if( ! is.null(tempo$same.row.name)){
if(tempo$same.row.name != TRUE){ # != TRUE to deal with NA
ident.row.names <- FALSE
}
}
if( ! is.null(tempo$same.col.name)){
if(tempo$same.col.name != TRUE){ # != TRUE to deal with NA
ident.col.names <- FALSE
}
}
}
# end argument checking without fun_check()
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
# end argument checking
# main code
output <- mat.list[[1]]
for(i1 in 2:length(mat.list)){
output <- get(kind.of.operation)(output, mat.list[[i1]]) # no env = sys.nframe(), inherit = FALSE in get() because look for function in the classical scope
}
dimnames(output) <- NULL
if(ident.row.names == TRUE){
rownames(output) <- rownames(mat.list[[1]])
}
if(ident.col.names == TRUE){
colnames(output) <- colnames(mat.list[[1]])
}
return(output)
Gael  MILLOT's avatar
Gael MILLOT committed
2731
2732
2733
2734
2735
2736
2737
2738
}


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


# Check OK: clear to go Apollo
fun_mat_inv <- function(mat){
Gael's avatar
tempo    
Gael committed
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
# AIM
# return the inverse of a square matrix when solve() cannot
# ARGUMENTS:
# mat: a square numeric matrix without NULL, NA, Inf or single case (dimension 1, 1) of 0
# RETURN
# the inversed matrix
# REQUIRED PACKAGES
# none
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# EXAMPLES
# mat1 = matrix(c(1,1,1,2,1,5,9,8,9), ncol = 3) ; fun_mat_inv(mat = mat1) # use solve()
# mat1 = matrix(c(0,0,0,0,0,0,0,0,0), ncol = 3) ; fun_mat_inv(mat = mat1) # use the trick
# mat1 = matrix(c(1,1,1,2,Inf,5,9,8,9), ncol = 3) ; fun_mat_inv(mat = mat1)
# mat1 = matrix(c(1,1,1,2,NA,5,9,8,9), ncol = 3) ; fun_mat_inv(mat = mat1)
# mat1 = matrix(c(1,2), ncol = 1) ; fun_mat_inv(mat = mat1)
# mat1 = matrix(0, ncol = 1) ; fun_mat_inv(mat = mat1)
# mat1 = matrix(2, ncol = 1) ; fun_mat_inv(mat = mat1)
# DEBUGGING
# mat = matrix(c(1,1,1,2,1,5,9,8,9), ncol = 3) # for function debugging
# function name
function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
# end function name
# required function checking
if(length(utils::find("fun_check", mode = "function")) == 0){
tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end required function checking
# argument checking
# argument checking with fun_check()
arg.check <- NULL #
text.check <- NULL #
checked.arg.names <- NULL # for function debbuging: used by r_debugging_tools
ee <- expression(arg.check <- c(arg.check, tempo$problem) , text.check <- c(text.check, tempo$text) , checked.arg.names <- c(checked.arg.names, tempo$fun.name))
tempo <- fun_check(data = mat, class = "matrix", mode = "numeric", fun.name = function.name) ; eval(ee)
if(any(arg.check) == TRUE){
stop(paste0("\n\n================\n\n", paste(text.check[arg.check], collapse = "\n"), "\n\n================\n\n"), call. = FALSE) #
}
# end argument checking with fun_check()
# argument checking without fun_check()
if(ncol(mat) != nrow(mat)){
tempo.cat <- paste0("ERROR IN ", function.name, ": mat ARGUMENT MUST BE A SQUARE MATRIX")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if(any(mat %in% c(Inf, -Inf, NA))){
tempo.cat <- paste0("ERROR IN ", function.name, ": mat ARGUMENT MUST BE A MATRIX WITHOUT Inf, -Inf OR NA")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
if(all(mat == 0) & ncol(mat) == 1){
tempo.cat <- paste0("ERROR IN ", function.name, ": mat ARGUMENT CANNOT BE A SQUARE MATRIX MADE OF A SINGLE CASE OF 0")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
}
# end argument checking without fun_check()
# source("C:/Users/Gael/Documents/Git_versions_to_use/debugging_tools_for_r_dev-v1.4/r_debugging_tools-v1.4.R") ; eval(parse(text = str_basic_arg_check_dev)) ; eval(parse(text = str_arg_check_with_fun_check_dev)) # activate this line and use the function (with no arguments left as NULL) to check arguments status and if they have been checked using fun_check()
# end argument checking
# main code
if(any(grepl(x = try(solve(mat), silent = TRUE)[], pattern = "[Ee]rror"))){
tempo <- svd(mat)
val.critique <- which(tempo$d < 10^-8)
Diag.mod <- diag(1 / tempo$d)
for(i in val.critique){
Diag.mod[i, i] <- 0
}
return(tempo$v %*% Diag.mod %*% t(tempo$u))
}else{
return(solve(mat))
}
Gael  MILLOT's avatar
Gael MILLOT committed
2807
2808
2809
2810
2811
2812
2813
2814
}


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


# Check OK: clear to go Apollo
fun_mat_fill <- function(mat, empty.cell.string = 0, warn.print = FALSE){
Gael's avatar
tempo    
Gael committed
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
# AIM
# detect the empty half part of a symmetric square matrix (either topleft, topright, bottomleft or bottomright)
# fill this empty half part using the other symmetric half part of the matrix
# WARNINGS
# a plot verification using fun_gg_heatmap() is recommanded
# ARGUMENTS:
# mat: a numeric or character square matrix with the half part (according to the grand diagonal) filled with NA (any kind of matrix), "0" (character matrix) or 0 (numeric matrix) exclusively (not a mix of 0 and NA in the empty part)
# empty.cell.string: a numeric, character or NA (no quotes) indicating what empty cells are filled with
# warn.print: logical. Print warnings at the end of the execution? No print if no warning messages
# RETURN
# a list containing:
# $mat: the filled matrix
# $warn: the warning messages. Use cat() for proper display. NULL if no warning
# REQUIRED PACKAGES
# none
# REQUIRED FUNCTIONS FROM CUTE_LITTLE_R_FUNCTION
# fun_check()
# EXAMPLES
# mat1 = matrix(c(1,NA,NA,NA, 0,2,NA,NA, NA,3,4,NA, 5,6,7,8), ncol = 4) ; mat1 ; fun_mat_fill(mat = mat1, empty.cell.string = NA, warn.print = TRUE) # bottomleft example
# mat1 = matrix(c(1,1,1,2, 0,2,3,0, NA,3,0,0, 5,0,0,0), ncol = 4) ; mat1 ; fun_mat_fill(mat = mat1, empty.cell.string = NA, warn.print = TRUE) # error example
# mat1 = matrix(c(1,1,1,2, 0,2,3,0, NA,3,0,0, 5,0,0,0), ncol = 4) ; mat1 ; fun_mat_fill(mat = mat1, empty.cell.string = 0, warn.print = TRUE) # bottomright example
# mat1 = matrix(c(1,1,1,2, "a",2,3,NA, "a","a",0,0, "a","a","a",0), ncol = 4) ; mat1 ; fun_mat_fill(mat = mat1, empty.cell.string = "a", warn.print = TRUE) # topright example
# mat1 = matrix(c(0,0,0,2, 0,0,3,0, 0,3,0,NA, 5,0,0,0), ncol = 4) ; mat1 ; fun_mat_fill(mat = mat1, empty.cell.string = 0, warn.print = TRUE) # topleft example
# mat1 = matrix(c(0,0,0,2, 0,0,3,0, 0,3,0,0, 5,0,0,0), ncol = 4) ; mat1 ; fun_mat_fill(mat = mat1, empty.cell.string = 0, warn.print = TRUE) # error example
# DEBUGGING
# mat = matrix(c(1,NA,NA,NA, 0,2,NA,NA, NA,3,4,NA, 5,6,7,8), ncol = 4) ; empty.cell.string = NA ; warn.print = TRUE # for function debugging
# mat = matrix(c(0,0,0,2, 0,0,3,0, 0,3,0,NA, 5,0,0,0), ncol = 4) ; empty.cell.string = 0 ; warn.print = TRUE # for function debugging # topleft example
# mat = matrix(c(0,0,0,2, 0,0,3,0, 0,3,0,NA, 5,0,0,0), ncol = 4) ; empty.cell.string = NA ; warn.print = TRUE # for function debugging # topleft example
# function name
function.name <- paste0(as.list(match.call(expand.dots = FALSE))[[1]], "()")
# end function name
# required function checking
if(length(utils::find("fun_check", mode = "function")) == 0){
tempo.cat <- paste0("ERROR IN ", function.name, ": REQUIRED fun_check() FUNCTION IS MISSING IN THE R ENVIRONMENT")
stop(paste0("\n\n================\n\n", tempo.cat, "\n\n================\n\n"), call. = FALSE) # == in stop() to be able to add several messages between ==
Gael  MILLOT's avatar
Gael MILLOT committed
2850
}
Gael's avatar
tempo    
Gael committed
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872