languages/cpython
languages/cpython
1
0
Fork 0
mirror of https://github.com/python/cpython synced 2024-09-04 15:56:13 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix 're' to work on bytes. It could do with a few more tests, though.

Browse source
This commit is contained in:
Thomas Wouters 2008-03-18 20:19:54 +00:00
parent e8c3d266c8
commit 40a088dc27
3 changed files with 22 additions and 34 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
Lib/sre_compile.py
View file

@ -472,7 +472,7 @@ def _compile_info(code, pattern, flags):
code[skip] = len(code) - skip
def isstring(obj):
return isinstance(obj, str)
return isinstance(obj, (str, bytes))
def _code(p, flags):

4
Lib/sre_parse.py
View file

@ -192,8 +192,8 @@ def __next(self):
char = self.string[self.index:self.index+1]
# Special case for the str8, since indexing returns a integer
# XXX This is only needed for test_bug_926075 in test_re.py
if isinstance(self.string, bytes):
char = chr(char)
if char and isinstance(char, bytes):
char = chr(char[0])
if char == "\\":
try:
c = self.string[self.index + 1]

50
Lib/test/test_re.py
View file

@ -83,33 +83,22 @@ def test_bug_449000(self):
self.assertEqual(re.sub('\r\n', '\n', 'abc\r\ndef\r\n'),
'abc\ndef\n')
# This test makes no sense until re supports bytes, and should then probably
# test for the *in*ability to mix bytes and str this way :)
#
# def test_bug_1140(self):
# # re.sub(x, y, b'') should return b'', not '', and
# # re.sub(x, y, '') should return '', not b''.
# # Also:
# # re.sub(x, y, str(x)) should return str(y), and
# # re.sub(x, y, bytes(x)) should return
# # str(y) if isinstance(y, str) else unicode(y).
# for x in 'x', u'x':
# for y in 'y', u'y':
# z = re.sub(x, y, u'')
# self.assertEqual(z, u'')
# self.assertEqual(type(z), unicode)
# #
# z = re.sub(x, y, '')
# self.assertEqual(z, '')
# self.assertEqual(type(z), str)
# #
# z = re.sub(x, y, unicode(x))
# self.assertEqual(z, y)
# self.assertEqual(type(z), unicode)
# #
# z = re.sub(x, y, str(x))
# self.assertEqual(z, y)
# self.assertEqual(type(z), type(y))
def test_bug_1140(self):
# re.sub(x, y, b'') should return b'', not '', and
# re.sub(x, y, '') should return '', not b''.
# Also:
# re.sub(x, y, str(x)) should return str(y), and
# re.sub(x, y, bytes(x)) should return
# str(y) if isinstance(y, str) else unicode(y).
for x in 'x', b'x':
for y in 'y', b'y':
z = re.sub(x, y, b'')
self.assertEqual(z, b'')
self.assertEqual(type(z), bytes)
#
z = re.sub(x, y, '')
self.assertEqual(z, '')
self.assertEqual(type(z), str)
def test_bug_1661(self):
# Verify that flags do not get silently ignored with compiled patterns
@ -599,10 +588,9 @@ def test_finditer(self):
self.assertEqual([item.group(0) for item in iter],
[":", "::", ":::"])
# XXX This needs to be restored for str vs. bytes.
## def test_bug_926075(self):
## self.assert_(re.compile('bug_926075') is not
## re.compile(str8('bug_926075')))
def test_bug_926075(self):
self.assert_(re.compile('bug_926075') is not
re.compile(b'bug_926075'))
def test_bug_931848(self):
pattern = eval('"[\u002E\u3002\uFF0E\uFF61]"')