def test_bug_449000(self):
# Test for sub() on escaped characters
assert re.sub(r'\r\n', r'\n', 'abc\r\ndef\r\n') == (
'abc\ndef\n')
assert re.sub('\r\n', r'\n', 'abc\r\ndef\r\n') == (
'abc\ndef\n')
assert re.sub(r'\r\n', '\n', 'abc\r\ndef\r\n') == (
'abc\ndef\n')
assert re.sub('\r\n', '\n', 'abc\r\ndef\r\n') == (
'abc\ndef\n')
def test_bug_1140(self):
# re.sub(x, y, u'') should return u'', not '', and
# re.sub(x, y, '') should return '', not u''.
# Also:
# re.sub(x, y, unicode(x)) should return unicode(y), and
# re.sub(x, y, str(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'')
assert z == u''
assert type(z) == unicode
#
z = re.sub(x, y, '')
assert z == ''
assert type(z) == str
#
z = re.sub(x, y, unicode(x))
assert z == y
assert type(z) == unicode
#
z = re.sub(x, y, str(x))
assert z == y
assert type(z) == type(y)
def test_bug_1140(self):
# re.sub(x, y, u'') should return u'', not '', and
# re.sub(x, y, '') should return '', not u''.
# Also:
# re.sub(x, y, unicode(x)) should return unicode(y), and
# re.sub(x, y, str(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'')
assert z == u''
assert type(z) == unicode
#
z = re.sub(x, y, '')
assert z == ''
assert type(z) == str
#
z = re.sub(x, y, unicode(x))
assert z == y
assert type(z) == unicode
#
z = re.sub(x, y, str(x))
assert z == y
assert type(z) == type(y)
def test_sub_template_numeric_escape(self):
# bug 776311 and friends
assert re.sub('x', r'\0', 'x') == '\0'
assert re.sub('x', r'\000', 'x') == '\000'
assert re.sub('x', r'\001', 'x') == '\001'
assert re.sub('x', r'\008', 'x') == '\0' + '8'
assert re.sub('x', r'\009', 'x') == '\0' + '9'
assert re.sub('x', r'\111', 'x') == '\111'
assert re.sub('x', r'\117', 'x') == '\117'
assert re.sub('x', r'\1111', 'x') == '\1111'
assert re.sub('x', r'\1111', 'x') == '\111' + '1'
assert re.sub('x', r'\00', 'x') == '\x00'
assert re.sub('x', r'\07', 'x') == '\x07'
assert re.sub('x', r'\08', 'x') == '\0' + '8'
assert re.sub('x', r'\09', 'x') == '\0' + '9'
assert re.sub('x', r'\0a', 'x') == '\0' + 'a'
assert re.sub('x', r'\400', 'x') == '\0'
assert re.sub('x', r'\777', 'x') == '\377'
raises(re.error, re.sub, 'x', r'\1', 'x')
raises(re.error, re.sub, 'x', r'\8', 'x')
raises(re.error, re.sub, 'x', r'\9', 'x')
raises(re.error, re.sub, 'x', r'\11', 'x')
raises(re.error, re.sub, 'x', r'\18', 'x')
raises(re.error, re.sub, 'x', r'\1a', 'x')
raises(re.error, re.sub, 'x', r'\90', 'x')
raises(re.error, re.sub, 'x', r'\99', 'x')
raises(re.error, re.sub, 'x', r'\118', 'x') # r'\11' + '8'
raises(re.error, re.sub, 'x', r'\11a', 'x')
raises(re.error, re.sub, 'x', r'\181', 'x') # r'\18' + '1'
raises(re.error, re.sub, 'x', r'\800', 'x') # r'\80' + '0'
# in python2.3 (etc), these loop endlessly in sre_parser.py
assert re.sub('(((((((((((x)))))))))))', r'\11', 'x') == 'x'
assert re.sub('((((((((((y))))))))))(.)', r'\118', 'xyz') == (
'xz8')
assert re.sub('((((((((((y))))))))))(.)', r'\11a', 'xyz') == (
'xza')
def test_bug_449964(self):
# fails for group followed by other escape
assert re.sub(r'(?P<unk>x)', '\g<1>\g<1>\\b', 'xx') == (
'xx\bxx\b')
def test_basic_re_sub(self):
assert re.sub("(?i)b+", "x", "bbbb BBBB") == 'x x'
assert re.sub(r'\d+', self.bump_num, '08.2 -2 23x99y') == (
'9.3 -3 24x100y')
assert re.sub(r'\d+', self.bump_num, '08.2 -2 23x99y', 3) == (
'9.3 -3 23x99y')
assert re.sub('.', lambda m: r"\n", 'x') == '\\n'
assert re.sub('.', r"\n", 'x') == '\n'
s = r"\1\1"
assert re.sub('(.)', s, 'x') == 'xx'
assert re.sub('(.)', re.escape(s), 'x') == s
assert re.sub('(.)', lambda m: s, 'x') == s
assert re.sub('(?P<a>x)', '\g<a>\g<a>', 'xx') == 'xxxx'
assert re.sub('(?P<a>x)', '\g<a>\g<1>', 'xx') == 'xxxx'
assert re.sub('(?P<unk>x)', '\g<unk>\g<unk>', 'xx') == 'xxxx'
assert re.sub('(?P<unk>x)', '\g<1>\g<1>', 'xx') == 'xxxx'
assert re.sub('a',r'\t\n\v\r\f\a\b\B\Z\a\A\w\W\s\S\d\D','a') == (
'\t\n\v\r\f\a\b\\B\\Z\a\\A\\w\\W\\s\\S\\d\\D')
assert re.sub('a', '\t\n\v\r\f\a', 'a') == '\t\n\v\r\f\a'
assert re.sub('a', '\t\n\v\r\f\a', 'a') == (
(chr(9)+chr(10)+chr(11)+chr(13)+chr(12)+chr(7)))
assert re.sub('^\s*', 'X', 'test') == 'Xtest'
def test_bug_462270(self):
# Test for empty sub() behaviour, see SF bug #462270
assert re.sub('x*', '-', 'abxd') == '-a-b-d-'
assert re.sub('x+', '-', 'abxd') == 'ab-d'
def test_bug_114660(self):
assert re.sub(r'(\S)\s+(\S)', r'\1 \2', 'hello there') == (
'hello there')
def test_qualified_re_sub(self):
assert re.sub('a', 'b', 'aaaaa') == 'bbbbb'
assert re.sub('a', 'b', 'aaaaa', 1) == 'baaaa'
def test_sub_template_numeric_escape(self):
# bug 776311 and friends
assert re.sub('x', r'\0', 'x') == '\0'
assert re.sub('x', r'\000', 'x') == '\000'
assert re.sub('x', r'\001', 'x') == '\001'
assert re.sub('x', r'\008', 'x') == '\0' + '8'
assert re.sub('x', r'\009', 'x') == '\0' + '9'
assert re.sub('x', r'\111', 'x') == '\111'
assert re.sub('x', r'\117', 'x') == '\117'
assert re.sub('x', r'\1111', 'x') == '\1111'
assert re.sub('x', r'\1111', 'x') == '\111' + '1'
assert re.sub('x', r'\00', 'x') == '\x00'
assert re.sub('x', r'\07', 'x') == '\x07'
assert re.sub('x', r'\08', 'x') == '\0' + '8'
assert re.sub('x', r'\09', 'x') == '\0' + '9'
assert re.sub('x', r'\0a', 'x') == '\0' + 'a'
assert re.sub('x', r'\400', 'x') == '\0'
assert re.sub('x', r'\777', 'x') == '\377'
raises(re.error, re.sub, 'x', r'\1', 'x')
raises(re.error, re.sub, 'x', r'\8', 'x')
raises(re.error, re.sub, 'x', r'\9', 'x')
raises(re.error, re.sub, 'x', r'\11', 'x')
raises(re.error, re.sub, 'x', r'\18', 'x')
raises(re.error, re.sub, 'x', r'\1a', 'x')
raises(re.error, re.sub, 'x', r'\90', 'x')
raises(re.error, re.sub, 'x', r'\99', 'x')
raises(re.error, re.sub, 'x', r'\118', 'x') # r'\11' + '8'
raises(re.error, re.sub, 'x', r'\11a', 'x')
raises(re.error, re.sub, 'x', r'\181', 'x') # r'\18' + '1'
raises(re.error, re.sub, 'x', r'\800', 'x') # r'\80' + '0'
# in python2.3 (etc), these loop endlessly in sre_parser.py
assert re.sub('(((((((((((x)))))))))))', r'\11', 'x') == 'x'
assert re.sub('((((((((((y))))))))))(.)', r'\118', 'xyz') == ('xz8')
assert re.sub('((((((((((y))))))))))(.)', r'\11a', 'xyz') == ('xza')
def test_bug_449000(self):
# Test for sub() on escaped characters
assert re.sub(r'\r\n', r'\n', 'abc\r\ndef\r\n') == ('abc\ndef\n')
assert re.sub('\r\n', r'\n', 'abc\r\ndef\r\n') == ('abc\ndef\n')
assert re.sub(r'\r\n', '\n', 'abc\r\ndef\r\n') == ('abc\ndef\n')
assert re.sub('\r\n', '\n', 'abc\r\ndef\r\n') == ('abc\ndef\n')
def test_bug_449964(self):
# fails for group followed by other escape
assert re.sub(r'(?P<unk>x)', '\g<1>\g<1>\\b', 'xx') == ('xx\bxx\b')
def test_basic_re_sub(self):
assert re.sub("(?i)b+", "x", "bbbb BBBB") == 'x x'
assert re.sub(r'\d+', self.bump_num,
'08.2 -2 23x99y') == ('9.3 -3 24x100y')
assert re.sub(r'\d+', self.bump_num, '08.2 -2 23x99y',
3) == ('9.3 -3 23x99y')
assert re.sub('.', lambda m: r"\n", 'x') == '\\n'
assert re.sub('.', r"\n", 'x') == '\n'
s = r"\1\1"
assert re.sub('(.)', s, 'x') == 'xx'
assert re.sub('(.)', re.escape(s), 'x') == s
assert re.sub('(.)', lambda m: s, 'x') == s
assert re.sub('(?P<a>x)', '\g<a>\g<a>', 'xx') == 'xxxx'
assert re.sub('(?P<a>x)', '\g<a>\g<1>', 'xx') == 'xxxx'
assert re.sub('(?P<unk>x)', '\g<unk>\g<unk>', 'xx') == 'xxxx'
assert re.sub('(?P<unk>x)', '\g<1>\g<1>', 'xx') == 'xxxx'
assert re.sub('a', r'\t\n\v\r\f\a\b\B\Z\a\A\w\W\s\S\d\D',
'a') == ('\t\n\v\r\f\a\b\\B\\Z\a\\A\\w\\W\\s\\S\\d\\D')
assert re.sub('a', '\t\n\v\r\f\a', 'a') == '\t\n\v\r\f\a'
assert re.sub('a', '\t\n\v\r\f\a',
'a') == ((chr(9) + chr(10) + chr(11) + chr(13) +
chr(12) + chr(7)))
assert re.sub('^\s*', 'X', 'test') == 'Xtest'
def test_bug_462270(self):
# Test for empty sub() behaviour, see SF bug #462270
assert re.sub('x*', '-', 'abxd') == '-a-b-d-'
assert re.sub('x+', '-', 'abxd') == 'ab-d'
def test_bug_114660(self):
assert re.sub(r'(\S)\s+(\S)', r'\1 \2',
'hello there') == ('hello there')
def test_qualified_re_sub(self):
assert re.sub('a', 'b', 'aaaaa') == 'bbbbb'
assert re.sub('a', 'b', 'aaaaa', 1) == 'baaaa'