Merge pull request #136 from IDEMSInternational/fix/windows-linebreak-and-utf8-encoding

Fix Windows linebreaks for table export and force utf-8 encoding

Author: geoo89

src/rpft/cli.py

@@ -21,7 +21,7 @@ def create_flows(args):
         tags=args.tags,
     )
 
-    with open(args.output, "w") as export:
+    with open(args.output, "w", encoding="utf-8") as export:
         json.dump(flows, export, indent=4)
 
 

src/rpft/converters.py

@@ -37,7 +37,7 @@ def create_flows(input_files, output_file, sheet_format, data_models=None, tags=
     flows = parser.parse_all().render()
 
     if output_file:
-        with open(output_file, "w") as export:
+        with open(output_file, "w", encoding='utf8') as export:
             json.dump(flows, export, indent=4)
 
     return flows
@@ -70,7 +70,7 @@ def flows_to_sheets(
     :param numbered: Use sequential numbers instead of short reps for row IDs.
     :returns: None.
     """
-    with open(input_file, "r") as f:
+    with open(input_file, "r", encoding='utf-8') as f:
         data = json.load(f)
     container = RapidProContainer.from_dict(data)
     for flow in container.flows:

src/rpft/parsers/common/rowdatasheet.py

@@ -38,8 +38,12 @@ def export(self, filename, file_format="csv"):
         """
         data = self.convert_to_tablib()
         exported_data = data.export(file_format)
-        write_type = "w" if type(exported_data) is str else "wb"
-        with open(filename, write_type) as f:
+        if type(exported_data) is str:
+            # There is a strange bug where on Windows, csv files would get
+            # exported with \r\r\n linebreaks. The below fixes this.
+            exported_data = exported_data.replace("\r", "")
+            exported_data = exported_data.encode("utf-8")
+        with open(filename, "wb") as f:
             f.write(exported_data)
 
     def convert_to_tablib(self):