Add input encoding detection for Windows

R/functions.R

@@ -355,7 +355,16 @@ guess_file_encoding = function(input_file){
     host_os = guess_host_os()
     if(host_os == 'linux') tmp = system2("file", args=c("-i", paste0("'",input_file,"'")), stdout=T)
     if(host_os == 'osx') tmp = system2("file", args=c("-I", paste0("'",input_file,"'")), stdout=T)
-    if(host_os == 'windows') return("native.enc")
+    if(host_os == 'windows'){
+        # Determine encoding by looking at the first 2 characters.
+        file_connection = file(input_file, open='rb')
+        file_start = paste(readBin(con=file_connection, what='raw', n=2), collapse='')
+        close(file_connection)
+        # UTF-8 with BOM starts with hexadecimal characters "ef bb". UTF-16 LE with "ff fe".
+        if(file_start == 'efbb') return("UTF-8-BOM")
+        if(file_start == 'fffe') return("UTF-16LE")
+        return("UTF-8")
+    }
     if(host_os == 'unknown') raise_error("Unable to detect host OS.")
     encoding_type = sub(pattern='^.* charset=', replacement='', x=tmp)
 

R/load_data.R

@@ -177,7 +177,7 @@ read_parameters_file <- function(input_file){
             # Verify that a parameter name has already been defined, and that the parameter value
             # line does not contain any ':' character.
             if(is.null(parameter_name)) raise_error(
-                msg="The first non-comment line of parameter files must contain a parameter value.",
+                msg="The first non-comment line of a parameter file must contain a parameter value.",
                 file=input_file)
             if(grepl(pattern=':', x=line)) raise_error(
                 msg="':' characters are only allowed as parameter delimitors.", file=input_file)

R/rename_samples.R

@@ -40,7 +40,7 @@ rename_samples <- function(samples, session_root_dir){
 
             # Case 2: the data must be split into 2.
             } else if(length(new_name) == 2){
-                # Get imageID values that corresponds to the 2 sub-samples in the original sample.
+                # Get image ID values that corresponds to the 2 sub-samples in the original sample.
                 if(file_extension == CELL_FILES_EXTENSION){
                     stopifnot('cell_y_position' %in% colnames(tmp))
                     image_id_split = split_by_coordinate(coordinates = tmp[,'cell_y_position'],