mirrors/cpython
1
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2025-08-04 17:08:35 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 13

gh-70278: Fix PyUnicode_FromFormat() with precision for %s and %V (GH-120365)

Browse source 
PyUnicode_FromFormat() no longer produces the ending \ufffd
character for truncated C string when use precision with %s and %V.
It now truncates the string before the start of truncated multibyte sequences.
This commit is contained in:
Serhiy Storchaka 2024-06-24 18:07:07 +03:00 • committed by GitHub
parent 22b8a35d6e
commit 6eb23b1311
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 59 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

46
Lib/test/test_capi/test_unicode.py
View file

@ -419,8 +419,29 @@ class CAPITest(unittest.TestCase):
# truncated string
check_format('abc',
b'%.3s', b'abcdef')
check_format('abc[',
b'%.6s', 'abc[\u20ac]'.encode('utf8'))
check_format('abc[\u20ac',
b'%.7s', 'abc[\u20ac]'.encode('utf8'))
check_format('abc[\ufffd',
b'%.5s', 'abc[\u20ac]'.encode('utf8'))
b'%.5s', b'abc[\xff]')
check_format('abc[',
b'%.6s', b'abc[\xe2\x82]')
check_format('abc[\ufffd]',
b'%.7s', b'abc[\xe2\x82]')
check_format('abc[\ufffd',
b'%.7s', b'abc[\xe2\x82\0')
check_format(' abc[',
b'%10.6s', 'abc[\u20ac]'.encode('utf8'))
check_format(' abc[\u20ac',
b'%10.7s', 'abc[\u20ac]'.encode('utf8'))
check_format(' abc[\ufffd',
b'%10.5s', b'abc[\xff]')
check_format(' abc[',
b'%10.6s', b'abc[\xe2\x82]')
check_format(' abc[\ufffd]',
b'%10.7s', b'abc[\xe2\x82]')
check_format("'\\u20acABC'",
b'%A', '\u20acABC')
check_format("'\\u20",
@ -433,10 +454,31 @@ class CAPITest(unittest.TestCase):
b'%.3S', '\u20acABCDEF')
check_format('\u20acAB',
b'%.3U', '\u20acABCDEF')
check_format('\u20acAB',
b'%.3V', '\u20acABCDEF', None)
check_format('abc[',
b'%.6V', None, 'abc[\u20ac]'.encode('utf8'))
check_format('abc[\u20ac',
b'%.7V', None, 'abc[\u20ac]'.encode('utf8'))
check_format('abc[\ufffd',
b'%.5V', None, 'abc[\u20ac]'.encode('utf8'))
b'%.5V', None, b'abc[\xff]')
check_format('abc[',
b'%.6V', None, b'abc[\xe2\x82]')
check_format('abc[\ufffd]',
b'%.7V', None, b'abc[\xe2\x82]')
check_format(' abc[',
b'%10.6V', None, 'abc[\u20ac]'.encode('utf8'))
check_format(' abc[\u20ac',
b'%10.7V', None, 'abc[\u20ac]'.encode('utf8'))
check_format(' abc[\ufffd',
b'%10.5V', None, b'abc[\xff]')
check_format(' abc[',
b'%10.6V', None, b'abc[\xe2\x82]')
check_format(' abc[\ufffd]',
b'%10.7V', None, b'abc[\xe2\x82]')
check_format(' abc[\ufffd',
b'%10.7V', None, b'abc[\xe2\x82\0')
# following tests comes from #7330
# test width modifier and precision modifier with %S