From ecb7cfcb59cc0717921f09e03e41e873a255063f Mon Sep 17 00:00:00 2001
From: Rachael Stickland <50215726+RayStick@users.noreply.github.com>
Date: Thu, 25 Jul 2024 10:31:42 +0100
Subject: [PATCH] add file conversion function

---
 NAMESPACE             |  1 +
 R/convert_output.R    | 37 +++++++++++++++++++++++++++++++++++++
 man/convert_output.Rd | 21 +++++++++++++++++++++
 3 files changed, 59 insertions(+)
 create mode 100644 R/convert_output.R
 create mode 100644 man/convert_output.Rd

diff --git a/NAMESPACE b/NAMESPACE
index 4399581c..8a0cae22 100644
--- a/NAMESPACE
+++ b/NAMESPACE
@@ -1,6 +1,7 @@
 # Generated by roxygen2: do not edit by hand
 
 export(compare_sessions)
+export(convert_output)
 export(domain_mapping)
 export(user_categorisation)
 import(cli)
diff --git a/R/convert_output.R b/R/convert_output.R
new file mode 100644
index 00000000..306e3925
--- /dev/null
+++ b/R/convert_output.R
@@ -0,0 +1,37 @@
+#' convert_output
+#'
+#' The 'OUTPUT_' file groups multiple categorisations onto one line e.g. '1,3' \cr \cr
+#' This function creates a new longer output 'L-OUTPUT_' which gives each categorisation its own row \cr \cr
+#' This 'L-OUTPUT_' may be more useful when using these csv files in later analyses
+#' @param output_csv The name of the 'OUTPUT_' csv file that was created from domain_mapping.R
+#' @param output_dir The location of output_csv
+#' @return The function will return 'L-OUTPUT_' in the same output_dir
+#' @export
+#' @importFrom utils read.csv write.csv
+
+convert_output <- function(output_csv,output_dir) {
+
+output <- read.csv(paste0(output_dir,'/',output_csv))
+output_long <- output[0,] #make duplicate
+
+for (row in 1:(nrow(output))) {
+  if (grepl(',',output$Domain_code[row])){ #Domain_code for this row is a list
+    domain_code_list <- output$Domain_code[row] #extract Domain_code list
+    domain_code_list_split <- unlist(strsplit(domain_code_list, ",")) #split the list
+    for (code in 1:(length(domain_code_list_split))){ #for every domain code in list, create a new row
+      row_to_copy <- output[row,] #extract row
+      row_to_copy$Domain_code <- domain_code_list_split[code] #change domain code to single
+      output_long[nrow(output_long) + 1,] = row_to_copy #copy altered row
+    }
+  } else { #Domain_code for this row is not list
+    row_to_copy <- output[row,] #extract row
+    output_long[nrow(output_long) + 1,] = row_to_copy #copy unaltered row
+  }
+}
+
+# Save output_long
+utils::write.csv(output_long, paste0(output_dir,'/','L-',output_csv),row.names = FALSE)
+
+}
+
+
diff --git a/man/convert_output.Rd b/man/convert_output.Rd
new file mode 100644
index 00000000..5a2cede7
--- /dev/null
+++ b/man/convert_output.Rd
@@ -0,0 +1,21 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/convert_output.R
+\name{convert_output}
+\alias{convert_output}
+\title{convert_output}
+\usage{
+convert_output(output_csv, output_dir)
+}
+\arguments{
+\item{output_csv}{The name of the 'OUTPUT_' csv file that was created from domain_mapping.R}
+
+\item{output_dir}{The location of output_csv}
+}
+\value{
+The function will return 'L-OUTPUT_' in the same output_dir
+}
+\description{
+The 'OUTPUT_' file groups multiple categorisations onto one line e.g. '1,3' \cr \cr
+This function creates a new longer output 'L-OUTPUT_' which gives each categorisation its own row \cr \cr
+This 'L-OUTPUT_' may be more useful when using these csv files in later analyses
+}