Merge normalize and reconvert l10n scripts, keep author info, format …

…the files properly, remove tabs from data

scripts/reconvert_l10n.py

@@ -1,12 +1,16 @@
-import json
 import polib
+import json
+import sys
 
 JSON_KEY_TRANSL="translated strings"
+JSON_KEY_AUTHOR="authors"
+JSON_KEY_MODIFY="modified by"
 JSON_KEY_OR="or"
 JSON_KEY_TR="tr"
 JSON_KEY_CO="context"
 
-def reconvert(filename):
+def reconvert(filename, json_filename):
+	"""Converts a l10n file.po into a file.json, keeping author info from a previous file.json if possible"""
 	po = polib.pofile(open(filename, encoding='utf-8').read())
 	translations = []
 
@@ -19,18 +23,53 @@ def reconvert(filename):
 				t_entry[JSON_KEY_CO] = entry.msgctxt
 			translations.append(t_entry)
 
-	result = {JSON_KEY_TRANSL: translations}
+	try:
+		previous_l10n = json.load(open(json_filename, encoding='utf-8'), strict=False)
+
+		# remove tabs from data
+		authors = previous_l10n[JSON_KEY_AUTHOR]
+		for i, value in enumerate(authors[JSON_KEY_MODIFY]):
+			authors[JSON_KEY_MODIFY][i] = value.replace("\t", "    ")
+
+		result = {JSON_KEY_AUTHOR: authors, JSON_KEY_TRANSL: translations}
+		print(" extracted author info from " + json_filename)
+
+	except IOError:
+		result = {JSON_KEY_AUTHOR: "", JSON_KEY_TRANSL: translations}
+		print(" failed to open " + json_filename + ", skipping author info")
 
 	json.dump(
 		result,
-		open(filename + '.json', 'w', encoding='utf-8'),
+		open(json_filename, 'w', encoding='utf-8'),
 		ensure_ascii=False,
 		indent="\t",
 		separators=(',', ': '),
 		sort_keys=True,
 	)
 
+def normalize(filename):
+	"""Normalizes a l10n file.po for better version controlling"""
+	po = polib.pofile(open(filename).read())
+	entries = list(sorted(sorted(po, key=lambda x: x.msgctxt or ""), key=lambda x: x.msgid))
+	po.clear()
+	for entry in entries:
+		po.append(entry)
+	po.save(filename)
+
+def decrement_indent(filename):
+	"""decrement indent level of file"""
+	with open(filename, encoding='utf-8') as f:
+		sanitized_json=f.read().replace('\n\t', '\n')	 
+	with open(filename, 'w', encoding='utf-8') as f:
+		f.write(sanitized_json)
+
 if __name__ == '__main__':
-	import sys
+	"""Normalizes then converts a l10n file.po into a file.json, keeping author info from a previous file.json if possible"""
 	for filename in sys.argv[1:]:
-		reconvert(filename)
+		assert(len(filename)>len(".po"))
+		json_filename = filename[:-3]+".json"
+		normalize(filename)
+		print(filename + " -> " + filename + " (normalized)")
+		reconvert(filename, json_filename)
+		decrement_indent(json_filename)
+		print(filename + " -> " + json_filename)