AMR/R/misc.R

193 lines
7.1 KiB
R
Raw Normal View History

2018-02-21 11:52:31 +01:00
# ==================================================================== #
# TITLE #
# Antimicrobial Resistance (AMR) Analysis #
# #
2019-01-02 23:24:07 +01:00
# SOURCE #
# https://gitlab.com/msberends/AMR #
2018-02-21 11:52:31 +01:00
# #
# LICENCE #
2019-01-02 23:24:07 +01:00
# (c) 2019 Berends MS (m.s.berends@umcg.nl), Luz CF (c.f.luz@umcg.nl) #
2018-02-21 11:52:31 +01:00
# #
2019-01-02 23:24:07 +01:00
# This R package is free software; you can freely use and distribute #
# it for both personal and commercial purposes under the terms of the #
# GNU General Public License version 2.0 (GNU GPL-2), as published by #
# the Free Software Foundation. #
# #
# This R package was created for academic research and was publicly #
# released in the hope that it will be useful, but it comes WITHOUT #
# ANY WARRANTY OR LIABILITY. #
2019-04-05 18:47:39 +02:00
# Visit our website for more info: https://msberends.gitlab.io/AMR. #
2018-02-21 11:52:31 +01:00
# ==================================================================== #
# No export, no Rd
addin_insert_in <- function() {
rstudioapi::insertText(" %in% ")
}
# No export, no Rd
addin_insert_like <- function() {
rstudioapi::insertText(" %like% ")
}
percent_clean <- clean:::percent
2019-05-10 16:44:59 +02:00
# No export, no Rd
percent <- function(x, round = 1, force_zero = FALSE, decimal.mark = getOption("OutDec"), big.mark = ",", ...) {
if (decimal.mark == big.mark) {
if (decimal.mark == ",") {
big.mark <- "."
} else if (decimal.mark == ".") {
big.mark <- ","
} else {
big.mark <- " "
}
2019-05-10 16:44:59 +02:00
}
x <- percent_clean(x = x, round = round, force_zero = force_zero,
decimal.mark = decimal.mark, big.mark = big.mark, ...)
2019-05-10 16:44:59 +02:00
}
2019-03-15 13:57:25 +01:00
#' @importFrom crayon blue bold red
2019-01-15 12:45:24 +01:00
#' @importFrom dplyr %>% pull
2019-05-23 16:58:59 +02:00
search_type_in_df <- function(x, type) {
2019-01-15 12:45:24 +01:00
# try to find columns based on type
found <- NULL
2019-05-23 16:58:59 +02:00
colnames(x) <- trimws(colnames(x))
2019-01-29 00:06:50 +01:00
2019-01-15 12:45:24 +01:00
# -- mo
if (type == "mo") {
2019-05-23 16:58:59 +02:00
if ("mo" %in% lapply(x, class)) {
found <- colnames(x)[lapply(x, class) == "mo"][1]
} else if (any(colnames(x) %like% "^(mo|microorganism|organism|bacteria|bacterie)s?$")) {
found <- colnames(x)[colnames(x) %like% "^(mo|microorganism|organism|bacteria|bacterie)s?$"][1]
} else if (any(colnames(x) %like% "^(microorganism|organism|bacteria|bacterie)")) {
found <- colnames(x)[colnames(x) %like% "^(microorganism|organism|bacteria|bacterie)"][1]
2019-05-23 16:58:59 +02:00
} else if (any(colnames(x) %like% "species")) {
found <- colnames(x)[colnames(x) %like% "species"][1]
2019-01-15 12:45:24 +01:00
}
2019-01-29 00:06:50 +01:00
2019-01-15 12:45:24 +01:00
}
# -- key antibiotics
if (type == "keyantibiotics") {
2019-05-23 16:58:59 +02:00
if (any(colnames(x) %like% "^key.*(ab|antibiotics)")) {
found <- colnames(x)[colnames(x) %like% "^key.*(ab|antibiotics)"][1]
2019-01-15 12:45:24 +01:00
}
}
# -- date
if (type == "date") {
2019-05-23 16:58:59 +02:00
if (any(colnames(x) %like% "^(specimen date|specimen_date|spec_date)")) {
2019-03-15 13:57:25 +01:00
# WHONET support
2019-05-23 16:58:59 +02:00
found <- colnames(x)[colnames(x) %like% "^(specimen date|specimen_date|spec_date)"][1]
if (!any(class(x %>% pull(found)) %in% c("Date", "POSIXct"))) {
2019-03-15 13:57:25 +01:00
stop(red(paste0("ERROR: Found column `", bold(found), "` to be used as input for `col_", type,
"`, but this column contains no valid dates. Transform its values to valid dates first.")),
call. = FALSE)
}
} else {
2019-05-23 16:58:59 +02:00
for (i in 1:ncol(x)) {
if (any(class(x %>% pull(i)) %in% c("Date", "POSIXct"))) {
found <- colnames(x)[i]
2019-03-15 13:57:25 +01:00
break
}
2019-01-15 12:45:24 +01:00
}
}
}
# -- patient id
if (type == "patient_id") {
2019-05-23 16:58:59 +02:00
if (any(colnames(x) %like% "^(identification |patient|patid)")) {
found <- colnames(x)[colnames(x) %like% "^(identification |patient|patid)"][1]
2019-01-29 00:06:50 +01:00
}
}
# -- specimen
if (type == "specimen") {
2019-05-23 16:58:59 +02:00
if (any(colnames(x) %like% "(specimen type|spec_type)")) {
found <- colnames(x)[colnames(x) %like% "(specimen type|spec_type)"][1]
} else if (any(colnames(x) %like% "^(specimen)")) {
found <- colnames(x)[colnames(x) %like% "^(specimen)"][1]
2019-01-15 12:45:24 +01:00
}
}
if (!is.null(found)) {
msg <- paste0("NOTE: Using column `", bold(found), "` as input for `col_", type, "`.")
2019-01-29 00:06:50 +01:00
if (type %in% c("keyantibiotics", "specimen")) {
2019-05-17 20:22:04 +02:00
msg <- paste(msg, "Use", bold(paste0("col_", type), "= FALSE"), "to prevent this.")
2019-01-15 12:45:24 +01:00
}
message(blue(msg))
}
found
}
2019-03-26 14:24:03 +01:00
stopifnot_installed_package <- function(package) {
2019-06-01 20:40:49 +02:00
# no "utils::installed.packages()" since it requires non-staged install since R 3.6.0
# https://developer.r-project.org/Blog/public/2019/02/14/staged-install/index.html
get(".packageName", envir = asNamespace(package))
return(invisible())
2019-03-26 14:24:03 +01:00
}
2019-05-10 16:44:59 +02:00
# translate strings based on inst/translations.tsv
#' @importFrom dplyr %>% filter
2019-06-11 14:18:25 +02:00
translate_AMR <- function(from, language = get_locale(), only_unknown = FALSE) {
2019-05-10 16:44:59 +02:00
# if (getOption("AMR_locale", "en") != language) {
# language <- getOption("AMR_locale", "en")
# }
if (is.null(language)) {
return(from)
}
if (language %in% c("en", "")) {
return(from)
}
2019-06-01 20:40:49 +02:00
df_trans <- translations_file # internal data file
2019-05-10 16:44:59 +02:00
if (!language %in% df_trans$lang) {
stop("Unsupported language: '", language, "' - use one of: ",
paste0("'", sort(unique(df_trans$lang)), "'", collapse = ", "),
call. = FALSE)
}
df_trans <- df_trans %>% filter(lang == language)
2019-06-11 14:18:25 +02:00
if (only_unknown == TRUE) {
df_trans <- df_trans %>% filter(pattern %like% "unknown")
}
2019-05-10 16:44:59 +02:00
# default case sensitive if value if 'ignore.case' is missing:
df_trans$ignore.case[is.na(df_trans$ignore.case)] <- FALSE
# default not using regular expressions (fixed = TRUE) if 'fixed' is missing:
df_trans$fixed[is.na(df_trans$fixed)] <- TRUE
# check if text to look for is in one of the patterns
2019-05-16 21:20:00 +02:00
any_form_in_patterns <- tryCatch(any(from %like% paste0("(", paste(df_trans$pattern, collapse = "|"), ")")),
error = function(e) {
warning("Translation not possible. Please open an issue on GitLab (https://gitlab.com/msberends/AMR/issues) or GitHub (https://github.com/msberends/AMR/issues).", call. = FALSE)
return(FALSE)
})
if (NROW(df_trans) == 0 | !any_form_in_patterns) {
2019-05-10 16:44:59 +02:00
return(from)
}
for (i in 1:nrow(df_trans)) {
from <- gsub(x = from,
pattern = df_trans$pattern[i],
replacement = df_trans$replacement[i],
fixed = df_trans$fixed[i],
ignore.case = df_trans$ignore.case[i])
}
# force UTF-8 for diacritics
base::enc2utf8(from)
}
2019-05-20 19:12:41 +02:00
"%or%" <- function(x, y) {
if (is.null(x) | is.null(y)) {
if (is.null(x)) {
return(y)
} else {
return(x)
}
}
ifelse(!is.na(x),
x,
ifelse(!is.na(y), y, NA))
2019-05-20 19:12:41 +02:00
}