Validate language files for … and non-matching formatters

.github/workflows/style.yml

@@ -40,6 +40,8 @@ jobs:
       run: scripts/fix_style.py --dry-run
     - name: Check header guards
       run: scripts/check_header_guards.py
+    - name: Validate Languages
+      run: scripts/languages/validate.py
     - name: Check languages
       run: scripts/languages/update_all.py
     - name: Check dilated images

scripts/languages/twlang.py

@@ -7,22 +7,6 @@ class LanguageDecodeError(Exception):
 		error = f"File \"{filename}\", line {line+1}: {message}"
 		super().__init__(error)
 
-
-# Taken from https://stackoverflow.com/questions/30011379/how-can-i-parse-a-c-format-string-in-python
-cfmt = r'''\
-(                                  # start of capture group 1
-%                                  # literal "%"
-(?:                                # first option
-(?:[-+0 #]{0,5})                   # optional flags
-(?:\d+|\*)?                        # width
-(?:\.(?:\d+|\*))?                  # precision
-(?:h|l|ll|w|I|I32|I64)?            # size
-[cCdiouxXeEfgGaAnpsSZ]             # type
-) |                                # OR
-%%)                                # literal "%%"
-'''
-
-
 def decode(fileobj, elements_per_key):
 	data = {}
 	current_context = ""
@@ -45,10 +29,7 @@ def decode(fileobj, elements_per_key):
 			if len(data[current_key]) >= 1+elements_per_key:
 				raise LanguageDecodeError("Wrong number of elements per key", fileobj.name, index)
 			if current_key:
-				original = current_key[0] # pylint: disable=unsubscriptable-object
 				translation = line[3:]
-				if translation and [m.group(1) for m in re.finditer(cfmt, original, flags=re.X)] != [m.group(1) for m in re.finditer(cfmt, translation, flags=re.X)]:
-					raise LanguageDecodeError("Non-matching formatting string", fileobj.name, index)
 				data[current_key].extend([translation])
 			else:
 				raise LanguageDecodeError("Element before key given", fileobj.name, index)

scripts/languages/validate.py

@@ -0,0 +1,54 @@
+#!/usr/bin/env python3
+import os
+import sys
+import re
+import twlang
+
+os.chdir(os.path.dirname(__file__) + "/../..")
+
+# Taken from https://stackoverflow.com/questions/30011379/how-can-i-parse-a-c-format-string-in-python
+cfmt = '''
+(                                  # start of capture group 1
+%                                  # literal "%"
+(?:                                # first option
+(?:[-+0 #]{0,5})                   # optional flags
+(?:\\d+|\\*)?                      # width
+(?:\\.(?:\\d+|\\*))?               # precision
+(?:h|l|ll|w|I|I32|I64)?            # size
+[cCdiouxXeEfgGaAnpsSZ]             # type
+) |                                # OR
+%%)                                # literal "%%"
+'''
+
+total_errors = 0
+
+def print_validation_error(error, filename, error_line):
+	print(f"Invalid: {translated}")
+	print(f"- {error} in {filename}:{error_line + 1}\n")
+	global total_errors
+	total_errors += 1
+
+if len(sys.argv) > 1:
+	languages = sys.argv[1:]
+else:
+	languages = twlang.languages()
+local = twlang.localizes()
+
+for language in languages:
+	translations = twlang.translations(language)
+
+	for (english, _), (line, translated, _) in translations.items():
+		if not translated:
+			continue
+
+		# Validate c format strings. Strings that move the formatters are not validated.
+		if re.findall(cfmt, english, flags=re.X) != re.findall(cfmt, translated, flags=re.X) and not "1$" in translated:
+			print_validation_error("Non-matching formatting", language, line)
+
+		# Check for elipisis
+		if "…" in english and "..." in translated:
+			print_validation_error("Usage of ... instead of the … character", language, line)
+
+if total_errors:
+	print(f"Found {total_errors} {'error' if total_errors == 1 else 'errors'} ")
+	sys.exit(1)