From 41124f80953188acc6de6140702402203e979e36 Mon Sep 17 00:00:00 2001 From: Zhou Wei Date: Tue, 21 Jul 2026 13:24:21 +0800 Subject: [PATCH] gh-103925: Fix csv.Sniffer for a quoted field ending a CRLF line (GH-103926) "$" does not match before "\r" even in the MULTILINE mode, so such a field was not found and the delimiter was guessed from character frequencies instead, which could give a letter. (cherry picked from commit 70f7c6c0f2ddfd3b447946f1b926776b2a344703) Co-authored-by: Zhou Wei Co-authored-by: Serhiy Storchaka Co-authored-by: Claude Opus 4.8 (1M context) --- Lib/csv.py | 8 ++++---- Lib/test/test_csv.py | 6 ++++++ .../2023-04-27-14-50-03.gh-issue-103925.khC-El.rst | 3 +++ 3 files changed, 13 insertions(+), 4 deletions(-) create mode 100644 Misc/NEWS.d/next/Library/2023-04-27-14-50-03.gh-issue-103925.khC-El.rst diff --git a/Lib/csv.py b/Lib/csv.py index 0a627ba7a512fa8..6efa889ed84677e 100644 --- a/Lib/csv.py +++ b/Lib/csv.py @@ -283,10 +283,10 @@ def _guess_quote_and_delimiter(self, data, delimiters): import re matches = [] - for restr in (r'(?P[^\w\n"\'])(?P ?)(?P["\']).*?(?P=quote)(?P=delim)', # ,".*?", - r'(?:^|\n)(?P["\']).*?(?P=quote)(?P[^\w\n"\'])(?P ?)', # ".*?", - r'(?P[^\w\n"\'])(?P ?)(?P["\']).*?(?P=quote)(?:$|\n)', # ,".*?" - r'(?:^|\n)(?P["\']).*?(?P=quote)(?:$|\n)'): # ".*?" (no delim, no space) + for restr in (r'(?P[^\w\n"\'])(?P ?)(?P["\']).*?(?P=quote)(?P=delim)', # ,".*?", + r'(?:^|\n)(?P["\']).*?(?P=quote)(?P[^\w\n"\'])(?P ?)', # ".*?", + r'(?P[^\w\n"\'])(?P ?)(?P["\']).*?(?P=quote)(?:$|\r|\n)', # ,".*?" + r'(?:^|\n)(?P["\']).*?(?P=quote)(?:$|\r|\n)'): # ".*?" (no delim, no space) regexp = re.compile(restr, re.DOTALL | re.MULTILINE) matches = regexp.findall(data) if matches: diff --git a/Lib/test/test_csv.py b/Lib/test/test_csv.py index a0a1e91777d8934..0f5235d5b2beecd 100644 --- a/Lib/test/test_csv.py +++ b/Lib/test/test_csv.py @@ -1402,6 +1402,9 @@ class TestSniffer(unittest.TestCase): sample18.append("v,twenty_one") # 'u' was not skipped sample18 = '\n'.join(sample18) + sample19 = ('time,title\r\n' + '2020-10-01,"Pocket - Save news, videos, stories and more"\r\n') + def test_issue43625(self): sniffer = csv.Sniffer() self.assertTrue(sniffer.has_header(self.sample12)) @@ -1481,6 +1484,9 @@ def test_delimiters(self): sniffer.sniff, self.sample15) self.assertRaisesRegex(csv.Error, "Could not determine delimiter", sniffer.sniff, self.sample16) + dialect = sniffer.sniff(self.sample19) + self.assertEqual(dialect.delimiter, ',') + self.assertEqual(dialect.quotechar, '"') def test_doublequote(self): sniffer = csv.Sniffer() diff --git a/Misc/NEWS.d/next/Library/2023-04-27-14-50-03.gh-issue-103925.khC-El.rst b/Misc/NEWS.d/next/Library/2023-04-27-14-50-03.gh-issue-103925.khC-El.rst new file mode 100644 index 000000000000000..858ce6e5bd98e70 --- /dev/null +++ b/Misc/NEWS.d/next/Library/2023-04-27-14-50-03.gh-issue-103925.khC-El.rst @@ -0,0 +1,3 @@ +Fix :meth:`csv.Sniffer.sniff` for a sample with ``\r\n`` line endings in +which a quoted field ends a line: a letter could be detected as the +delimiter.