From 9a76d603e7f2d6f9e3ab97fcfd135f00ff3a991c Mon Sep 17 00:00:00 2001
From: munoztd0 <david.munoztord@mailbox.org>
Date: Tue, 12 Mar 2024 16:46:42 +0100
Subject: [PATCH 1/2] feat: :bug: updated the handling of 'to_values' in the
 mapvalues function

Now, 'to_values' is created by splitting the string x[[paste0('VALUES', use.language)]][[code_id]] at every occurrence of ", " and ",". This change was made to handle cases where x[[paste0('VALUES', use.language)]][[code_id]] is a single string that needs to be split into a vector of values. Also added a check to ensure that the 'from_values' and 'to_values' vectors are of the same length before they are used in the mapvalues function. If they are not of the same length, the function stops with an error message. This check was added to prevent errors in the mapvalues function and to make the function more robust and handle a wider range of inputs.
---
 R/as.data.frame.px.R | 50 +++++++++++++++++++++++++++++++++++---------
 1 file changed, 40 insertions(+), 10 deletions(-)

diff --git a/R/as.data.frame.px.R b/R/as.data.frame.px.R
index 6fd6785..9f033f7 100644
--- a/R/as.data.frame.px.R
+++ b/R/as.data.frame.px.R
@@ -11,28 +11,58 @@
 #    20120402, cjgb: warnings can be either errors or warnings depending on paranoia level
 #    20120402, cjgb: adapted to the new px object format (where DATA is already a df)
 #    20141222. fvf:  bug in "wide" direction
+#    20240311. munoztd0: updated the handling of 'to_values' in the mapvalues function for language translation
 #################################################################
 
-as.data.frame.px <- function( x, ..., use.codes = FALSE, warnings.as.errors = TRUE, direction = "long"){
-
+as.data.frame.px <- function( x, ..., use.codes = FALSE, warnings.as.errors = TRUE, direction = "long", language=FALSE){
+  
   dat <- x$DATA$value  # already a data frame
   
+  if ((!is.logical(language) && is.character(language))){
+    if(language %in% strsplit(x$LANGUAGES$value, "\",\"")[[1]]){
+      use.language = paste0('.',language,'.')
+      default.language = x$LANGUAGE
+      codes.ids = match(names(x$CODES), colnames(dat))
+      
+      for(code_id in codes.ids) {
+        if (is.na(code_id)) {
+          next
+        }
+        from_values = x$VALUES[[code_id]]
+        to_values = strsplit(x[[paste0('VALUES', use.language)]][[code_id]], "\",\"|\", \"")[[1]]
+
+        if (length(from_values) != length(to_values)) {
+          stop(paste0('The from and to vectors for code_id ', code_id, ' are not the same length.'))
+        }
+
+        dat[[codes.ids[code_id]]] <- mapvalues(dat[[codes.ids[code_id]]], 
+                                               from = from_values, 
+                                               to   = to_values)
+      }
+      translated_colnames = names(x[[paste0('CODES', use.language)]])[codes.ids]
+      translated_colnames = c(translated_colnames, 'value')
+      colnames(dat) = translated_colnames
+      
+    } else {
+      stop(paste0('Can\'t find the proposed language. Please choose one of the available languages: ', x$LANGUAGES$value))
+    }
+  } else {
+    language = ''
+  }
+  
   ## maybe we need to change values to codes
   if (is.logical(use.codes) && use.codes)
     use.codes <- names(x$CODES)
   
   if (! is.logical(use.codes))
     for( var.name in intersect( use.codes, intersect(colnames(dat), names(x$CODES) ) ) )
-        dat[[var.name]] <- mapvalues(dat[[var.name]], 
-                                     from = x$VALUES[[var.name]], 
-                                     to   = x$CODES[[var.name]])
-     
+      dat[[var.name]] <- mapvalues(dat[[var.name]], 
+                                   from = x$VALUES[[var.name]], 
+                                   to   = x$CODES[[var.name]])
+  
   ## do we need to reshape?
   if (direction == "wide")
-    # fvf.20121222: The order of variables rows and pivots-columns was inverted
-    # dcast(dat, list(x$HEADING$value, x$STUB$value))
     dcast(dat, list(x$STUB$value,x$HEADING$value))
   else
     dat
-}
-
+}
\ No newline at end of file

From 22f5ec109035e9678ec27c6eaf02957d3b4d47c7 Mon Sep 17 00:00:00 2001
From: munoztd0 <david.munoztord@mailbox.org>
Date: Mon, 1 Apr 2024 18:34:39 +0200
Subject: [PATCH 2/2] docs: :memo: Update as.data.frame.px documentation with
 language argument

---
 DESCRIPTION             | 1 +
 man/as.data.frame.px.Rd | 3 ++-
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/DESCRIPTION b/DESCRIPTION
index 511da41..373655d 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -13,3 +13,4 @@ Depends: stringr, reshape2, RJSONIO, plyr
 URL: https://github.com/cjgb/pxR
 License: GPL-3
 LazyLoad: yes
+RoxygenNote: 7.3.1
diff --git a/man/as.data.frame.px.Rd b/man/as.data.frame.px.Rd
index 787ea54..3960e50 100644
--- a/man/as.data.frame.px.Rd
+++ b/man/as.data.frame.px.Rd
@@ -10,7 +10,7 @@ This function extracts the data component from a px object as a \code{data.frame
 }
 \usage{
 \S3method{as.data.frame}{px}( x, ..., use.codes = FALSE,
-                 warnings.as.errors = TRUE, direction = 'long')
+                 warnings.as.errors = TRUE, direction = 'long', language=FALSE)
 }
 \arguments{
   \item{x}{ a px object }
@@ -22,6 +22,7 @@ This function extracts the data component from a px object as a \code{data.frame
   \item{warnings.as.errors}{ If true, the function will fail in case any issues are found; otherwise, it will generate warnings. }
   \item{direction}{character string, either ‘"wide"’ to reshape to wide format,
           or ‘"long"’ to reshape to long format (default).}
+  \item{language}{ a logical indicating whether to use language-specific values. Default is FALSE.}
   \item{...}{ Additional arguments, currently not used }
 }
 \details{