mirrors/cpython
1
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2025-07-30 14:44:10 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 8

Issue #4426: The UTF-7 decoder was too strict and didn't accept some legal sequences.

Browse source 
Patch by Nick Barnes and Victor Stinner.
This commit is contained in:
Antoine Pitrou 2009-05-04 18:32:32 +00:00
parent 2827709d6d
commit 653dece278
4 changed files with 265 additions and 199 deletions
Download patch file Download diff file Expand all files Collapse all files

19
Lib/test/test_unicode.py
View file

@ -521,19 +521,28 @@ class UnicodeTest(
(u'+?', '+-?'),
(ur'\\?', '+AFwAXA?'),
(ur'\\\?', '+AFwAXABc?'),
(ur'++--', '+-+---')
(ur'++--', '+-+---'),
(u'\U000abcde', '+2m/c3g-'), # surrogate pairs
(u'/', '/'),
]
for (x, y) in utfTests:
self.assertEqual(x.encode('utf-7'), y)
# surrogates not supported
# Unpaired surrogates not supported
self.assertRaises(UnicodeError, unicode, '+3ADYAA-', 'utf-7')
self.assertEqual(unicode('+3ADYAA-', 'utf-7', 'replace'), u'\ufffd')
self.assertEqual(unicode('+3ADYAA-', 'utf-7', 'replace'), u'\ufffd\ufffd')
# Issue #2242: crash on some Windows/MSVC versions
self.assertRaises(UnicodeDecodeError, '+\xc1'.decode, 'utf-7')
# Direct encoded characters
set_d = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789'(),-./:?"
# Optional direct characters
set_o = '!"#$%&*;<=>@[]^_`{|}'
for c in set_d:
self.assertEqual(c.encode('utf7'), c.encode('ascii'))
self.assertEqual(c.encode('ascii').decode('utf7'), c)
for c in set_o:
self.assertEqual(c.encode('ascii').decode('utf7'), c)
def test_codecs_utf8(self):
self.assertEqual(u''.encode('utf-8'), '')