haciendo el codigo un poco mas legible y estructurando para hacer posibles...

haciendo el codigo un poco mas legible y estructurando para hacer posibles funciones de analisis segun el tipo de variable que sea

connection_parameters.R

@@ -71,7 +71,7 @@ project_names <- c(
 resource_names <- c(
   "Harmonized_variables_2",
   "CIPH_numeric_derivated",
-  "20220719_HarmonisedUMFIasi",
+  "20220919_UMFIasi",
   "SMUC_resource",
   "20220919_FiHM",
   "Resource_derived",

valid_variables_script2.R

@@ -126,39 +126,53 @@ is_number <- function(x){
   
 }
 
-check_values_format <- function(valid_columns){
+
+# A esta funcion la llamamos unicamente con las columnas que el sabemos que el nombre es correcto
+# Usa codebook param. Si algún cambia el codebook agradeceremos esto.
+check_values_format <- function(valid_columns, codebook_param){
   res <- ""
   for(i in 1:length(valid_columns[[1]])){
     print(i)
+    
     current_column <- valid_columns[[1]][[i]]
-    if(current_column %in% codebook_labo$Harmonised.variable.name){
-      index <- which(current_column, codebook_labo$Harmonised.variable.name)
-      value_format <- strsplit(codebook_si_sympt$Possible.values.format[2], " / ")[[1]]
+    variable_type <- codebook_param$Variable.type[codebook$Harmonised.variable.name == current_column]
+   
+    if(variable_type == "Continuous"){
+      
+      ################## ESTO PODRÍA IR EN UNA FUNC DIFERENTE #############
+      
+      ### parse del formato de una variable continua ##
+      ## esta sentencia funciona codebook$Possible.values.format[codebook$Harmonised.variable.name == "CMXDE"] pruebala en el interprete.
+      value_format <- strsplit(codebook_param$Possible.values.format[codebook$Harmonised.variable.name == current_column], " / ")[[1]]
       high_limit <- as.numeric(sub("-.*", "", value_format[1]))
       low_limit <- as.numeric(sub(".*-", "", value_format[1]))
-      if(codebook_labo$Variable.type[index] == "Continuous"){
-        ds.dataFrameSubset(df.name = "data",
-                           V1.name = paste("data$", current_column, sep=""),
-                           V2.name = high_limit,
-                           Boolean.operator = "<=",
-                           newobj = "inRangeHigh",
-                           keep.NAs = TRUE,
-                           datasources = connections)
-        
-        
-        ds.dataFrameSubset(df.name = "inRangeHigh",
-                           V1.name = paste("inRangeHigh$", current_column, sep=""),
-                           V2.name = low_limit,
-                           Boolean.operator = ">=",
-                           newobj = "inRange",
-                           keep.NAs = TRUE,
-                           datasources = connections)
-        
-        summary <- ds.summary(paste("inRange$", current_column, sep=""))
-        if(ds.length(paste("data$", current_column, sep="")) > summary[[1]][[2]]){
-          res <- c(res, paste(current_column,  "does not follow the established format" , sep="\n"))
-        }
+      ### parse del formato de una variable continua ##
+      
+      ds.dataFrameSubset(df.name = "data",
+                         V1.name = paste("data$", current_column, sep=""),
+                         V2.name = high_limit,
+                         Boolean.operator = "<=",
+                         newobj = "inRangeHigh",
+                         keep.NAs = TRUE,
+                         datasources = connections)
+      
+      
+      ds.dataFrameSubset(df.name = "inRangeHigh",
+                         V1.name = paste("inRangeHigh$", current_column, sep=""),
+                         V2.name = low_limit,
+                         Boolean.operator = ">=",
+                         newobj = "inRange",
+                         keep.NAs = TRUE,
+                         datasources = connections)
+      
+      summary <- ds.summary(paste("inRange$", current_column, sep=""))
+      if(ds.length(paste("data$", current_column, sep="")) > summary[[1]][[2]]){
+        res <- c(res, paste(current_column,  "does not follow the established format" , sep="\n"))
       }
+      ################## FIN ESTO PODRÍA IR EN UNA FUNC DIFERENTE #############
+      
+    }else if (variable_type == "Binary"){
+      
     }
   }
   return (res)
@@ -211,8 +225,8 @@ data_colnames <- as.data.frame(data_colnames)
 check_valid_columns <- check_column_names(codebook ,data_colnames)
 valid_columns <- as.data.frame(check_valid_columns$colnames)
 res <- ""
-res <- check_values_format(valid_columns)
+res <- check_values_format(valid_columns, codebook)
 print(res)
 
 
-  
+