https://github.com/python/cpython/commit/89be6f7778edda0399950aa368658ecfb7315028 commit: 89be6f7778edda0399950aa368658ecfb7315028 branch: 3.14 author: Serhiy Storchaka <[email protected]> committer: serhiy-storchaka <[email protected]> date: 2026-07-26T19:19:06Z summary:
[3.14] gh-154582: Fix test_invalid_utf8_arg in non-UTF-8 multibyte locales (GH-154584) (GH-154604) Arbitrary bytes round-trip through surrogateescape only in UTF-8 and single-byte encodings, not in a stateful multibyte encoding such as EUC-JP. (cherry picked from commit b1e530ebf3b68e884ab37882d690d95e5d82aef5) Co-authored-by: Claude Opus 4.8 <[email protected]> files: M Lib/test/test_cmd_line.py diff --git a/Lib/test/test_cmd_line.py b/Lib/test/test_cmd_line.py index 2f4fee0443cff38..adc665135576d46 100644 --- a/Lib/test/test_cmd_line.py +++ b/Lib/test/test_cmd_line.py @@ -331,8 +331,26 @@ def run_utf8_mode(arg): ) test_args = [valid_utf8, invalid_utf8] - for run_cmd in (run_default, run_c_locale, run_utf8_mode): - with self.subTest(run_cmd=run_cmd): + for run_cmd, encoding in ( + (run_default, sys.getfilesystemencoding()), + (run_c_locale, None), + (run_utf8_mode, None), + ): + with self.subTest(run_cmd=run_cmd.__name__): + # Arbitrary bytes round-trip through surrogateescape only in + # UTF-8 and single-byte encodings, not in a multibyte encoding + # such as EUC-JP. + if encoding is not None: + try: + lossless = len(bytes(range(256)).decode( + encoding, 'surrogateescape')) == 256 + except UnicodeError: + lossless = False + else: + lossless = True + if not lossless: + self.skipTest(f'{encoding} cannot losslessly ' + f'round-trip arbitrary bytes') for arg in test_args: proc = run_cmd(arg) self.assertEqual(proc.stdout.rstrip(), ascii(arg)) _______________________________________________ Python-checkins mailing list -- [email protected] To unsubscribe send an email to [email protected] https://mail.python.org/mailman3//lists/python-checkins.python.org Member address: [email protected]
