def test_bug_926075(self):
try:
unicode
except NameError:
return # no problem if we have no unicode
self.assertTrue(re.compile('bug_926075') is not
re.compile(eval("u'bug_926075'")))
def test_re_match(self):
self.assertEqual(re.match('a', 'a').groups(), ())
self.assertEqual(re.match('(a)', 'a').groups(), ('a',))
self.assertEqual(re.match(r'(a)', 'a').group(0), 'a')
self.assertEqual(re.match(r'(a)', 'a').group(1), 'a')
self.assertEqual(re.match(r'(a)', 'a').group(1, 1), ('a', 'a'))
pat = re.compile('((a)|(b))(c)?')
self.assertEqual(pat.match('a').groups(), ('a', 'a', None, None))
self.assertEqual(pat.match('b').groups(), ('b', None, 'b', None))
self.assertEqual(pat.match('ac').groups(), ('a', 'a', None, 'c'))
self.assertEqual(pat.match('bc').groups(), ('b', None, 'b', 'c'))
self.assertEqual(pat.match('bc').groups(""), ('b', "", 'b', 'c'))
# A single group
m = re.match('(a)', 'a')
self.assertEqual(m.group(0), 'a')
self.assertEqual(m.group(0), 'a')
self.assertEqual(m.group(1), 'a')
self.assertEqual(m.group(1, 1), ('a', 'a'))
pat = re.compile('(?:(?P<a1>a)|(?P<b2>b))(?P<c3>c)?')
self.assertEqual(pat.match('a').group(1, 2, 3), ('a', None, None))
self.assertEqual(pat.match('b').group('a1', 'b2', 'c3'),
(None, 'b', None))
self.assertEqual(pat.match('ac').group(1, 'b2', 3), ('a', None, 'c'))
def test_empty_array(self):
# SF buf 1647541
import array
for typecode in 'cbBuhHiIlLfd':
a = array.array(typecode)
self.assertEqual(re.compile("bla").match(a), None)
self.assertEqual(re.compile("").match(a).groups(), ())
def test_dollar_matches_twice(self):
"$ matches the end of string, and just before the terminating \n"
pattern = re.compile('$')
self.assertEqual(pattern.sub('#', 'a\nb\n'), 'a\nb#\n#')
self.assertEqual(pattern.sub('#', 'a\nb\nc'), 'a\nb\nc#')
self.assertEqual(pattern.sub('#', '\n'), '#\n#')
pattern = re.compile('$', re.MULTILINE)
self.assertEqual(pattern.sub('#', 'a\nb\n' ), 'a#\nb#\n#' )
self.assertEqual(pattern.sub('#', 'a\nb\nc'), 'a#\nb#\nc#')
self.assertEqual(pattern.sub('#', '\n'), '#\n#')
def test_regex_compile(count):
re._cache = EmptyCache()
regexes = capture_regexes()
times = []
for _ in xrange(count):
t0 = time.time()
for regex, flags in regexes:
re.compile(regex, flags)
t1 = time.time()
times.append(t1 - t0)
return times
def test_bug_1661(self):
# Verify that flags do not get silently ignored with compiled patterns
pattern = re.compile('.')
self.assertRaises(ValueError, re.match, pattern, 'A', re.I)
self.assertRaises(ValueError, re.search, pattern, 'A', re.I)
self.assertRaises(ValueError, re.findall, pattern, 'A', re.I)
self.assertRaises(ValueError, re.compile, pattern, re.I)
def test_bug_931848(self):
try:
unicode
except NameError:
pass
pattern = eval('u"[\u002E\u3002\uFF0E\uFF61]"')
self.assertEqual(re.compile(pattern).split("a.b.c"),
['a','b','c'])
def test_bug_581080(self):
iter = re.finditer(r"\s", "a b")
self.assertEqual(iter.next().span(), (1,2))
self.assertRaises(StopIteration, iter.next)
scanner = re.compile(r"\s").scanner("a b")
self.assertEqual(scanner.search().span(), (1, 2))
self.assertEqual(scanner.search(), None)
def test_bug_764548(self):
# bug 764548, re.compile() barfs on str/unicode subclasses
try:
unicode
except NameError:
return # no problem if we have no unicode
class my_unicode(unicode): pass
pat = re.compile(my_unicode("abc"))
self.assertEqual(pat.match("xyz"), None)
def test_inline_flags(self):
# Bug #1700
upper_char = unichr(0x1ea0) # Latin Capital Letter A with Dot Bellow
lower_char = unichr(0x1ea1) # Latin Small Letter A with Dot Bellow
p = re.compile(upper_char, re.I | re.U)
q = p.match(lower_char)
self.assertNotEqual(q, None)
p = re.compile(lower_char, re.I | re.U)
q = p.match(upper_char)
self.assertNotEqual(q, None)
p = re.compile('(?i)' + upper_char, re.U)
q = p.match(lower_char)
self.assertNotEqual(q, None)
p = re.compile('(?i)' + lower_char, re.U)
q = p.match(upper_char)
self.assertNotEqual(q, None)
p = re.compile('(?iu)' + upper_char)
q = p.match(lower_char)
self.assertNotEqual(q, None)
p = re.compile('(?iu)' + lower_char)
q = p.match(upper_char)
self.assertNotEqual(q, None)
def test_re_groupref_exists(self):
self.assertEqual(re.match('^(\()?([^()]+)(?(1)\))$', '(a)').groups(),
('(', 'a'))
self.assertEqual(re.match('^(\()?([^()]+)(?(1)\))$', 'a').groups(),
(None, 'a'))
self.assertEqual(re.match('^(\()?([^()]+)(?(1)\))$', 'a)'), None)
self.assertEqual(re.match('^(\()?([^()]+)(?(1)\))$', '(a'), None)
self.assertEqual(re.match('^(?:(a)|c)((?(1)b|d))$', 'ab').groups(),
('a', 'b'))
self.assertEqual(re.match('^(?:(a)|c)((?(1)b|d))$', 'cd').groups(),
(None, 'd'))
self.assertEqual(re.match('^(?:(a)|c)((?(1)|d))$', 'cd').groups(),
(None, 'd'))
self.assertEqual(re.match('^(?:(a)|c)((?(1)|d))$', 'a').groups(),
('a', ''))
# Tests for bug #1177831: exercise groups other than the first group
p = re.compile('(?P<g1>a)(?P<g2>b)?((?(g2)c|d))')
self.assertEqual(p.match('abc').groups(),
('a', 'b', 'c'))
self.assertEqual(p.match('ad').groups(),
('a', None, 'd'))
self.assertEqual(p.match('abd'), None)
self.assertEqual(p.match('ac'), None)
# Python imports
import optparse
import sys, os
sys.path.insert(0, os.path.join(os.path.dirname(__file__), '..', '..'))
import pcre_re as re
import time
# Local imports
import util
# These are the regular expressions to be tested. These sync up,
# index-for-index with the list of strings generated by gen_string_table()
# below.
regexs = [
re.compile('Python|Perl'),
re.compile('Python|Perl'),
re.compile('(Python|Perl)'),
re.compile('(?:Python|Perl)'),
re.compile('Python'),
re.compile('Python'),
re.compile('.*Python'),
re.compile('.*Python.*'),
re.compile('.*(Python)'),
re.compile('.*(?:Python)'),
re.compile('Python|Perl|Tcl'),
re.compile('Python|Perl|Tcl'),
re.compile('(Python|Perl|Tcl)'),
re.compile('(?:Python|Perl|Tcl)'),
re.compile('(Python)\\1'),
re.compile('(Python)\\1'),
def run_re_tests():
from test.re_tests import tests, SUCCEED, FAIL, SYNTAX_ERROR
if verbose:
print 'Running re_tests test suite'
else:
# To save time, only run the first and last 10 tests
#tests = tests[:10] + tests[-10:]
pass
for t in tests:
sys.stdout.flush()
pattern = s = outcome = repl = expected = None
if len(t) == 5:
pattern, s, outcome, repl, expected = t
elif len(t) == 3:
pattern, s, outcome = t
else:
raise ValueError, ('Test tuples should have 3 or 5 fields', t)
try:
obj = re.compile(pattern)
except re.error:
if outcome == SYNTAX_ERROR: pass # Expected a syntax error
else:
print '=== Syntax error:', t
except KeyboardInterrupt: raise KeyboardInterrupt
except:
print '*** Unexpected error ***', t
if verbose:
traceback.print_exc(file=sys.stdout)
else:
try:
result = obj.search(s)
except re.error, msg:
print '=== Unexpected exception', t, repr(msg)
if outcome == SYNTAX_ERROR:
# This should have been a syntax error; forget it.
pass
elif outcome == FAIL:
if result is None: pass # No match, as expected
else: print '=== Succeeded incorrectly', t
elif outcome == SUCCEED:
if result is not None:
# Matched, as expected, so now we compute the
# result string and compare it to our expected result.
start, end = result.span(0)
vardict={'found': result.group(0),
'groups': result.group(),
'flags': result.re.flags}
for i in range(1, 100):
try:
gi = result.group(i)
# Special hack because else the string concat fails:
if gi is None:
gi = "None"
except IndexError:
gi = "Error"
vardict['g%d' % i] = gi
for i in result.re.groupindex.keys():
try:
gi = result.group(i)
if gi is None:
gi = "None"
except IndexError:
gi = "Error"
vardict[i] = gi
repl = eval(repl, vardict)
if repl != expected:
print '=== grouping error', t,
print repr(repl) + ' should be ' + repr(expected)
else:
print '=== Failed incorrectly', t
# Try the match on a unicode string, and check that it
# still succeeds.
try:
result = obj.search(unicode(s, "latin-1"))
if result is None:
print '=== Fails on unicode match', t
except NameError:
continue # 1.5.2
except TypeError:
continue # unicode test case
# Try the match on a unicode pattern, and check that it
# still succeeds.
obj=re.compile(unicode(pattern, "latin-1"))
result = obj.search(s)
if result is None:
print '=== Fails on unicode pattern match', t
# Try the match with the search area limited to the extent
# of the match and see if it still succeeds. \B will
# break (because it won't match at the end or start of a
# string), so we'll ignore patterns that feature it.
if pattern[:2] != '\\B' and pattern[-2:] != '\\B' \
and result is not None:
obj = re.compile(pattern)
result = obj.search(s, result.start(0), result.end(0) + 1)
if result is None:
print '=== Failed on range-limited match', t
# Try the match with IGNORECASE enabled, and check that it
# still succeeds.
obj = re.compile(pattern, re.IGNORECASE)
result = obj.search(s)
if result is None:
print '=== Fails on case-insensitive match', t
# Try the match with LOCALE enabled, and check that it
# still succeeds.
obj = re.compile(pattern, re.LOCALE)
result = obj.search(s)
if result is None:
print '=== Fails on locale-sensitive match', t
# Try the match with UNICODE locale enabled, and check
# that it still succeeds.
obj = re.compile(pattern, re.UNICODE)
result = obj.search(s)
if result is None:
print '=== Fails on unicode-sensitive match', t
def test_bug_3629(self):
# A regex that triggered a bug in the sre-code validator
re.compile("(?P<quote>)(?(quote))")
def test_weakref(self):
s = 'QabbbcR'
x = re.compile('ab+c')
y = proxy(x)
self.assertEqual(x.findall('QabbbcR'), y.findall('QabbbcR'))
def pickle_test(self, pickle):
oldpat = re.compile('a(?:b|(c|e){1,2}?|d)+?(.)')
s = pickle.dumps(oldpat)
newpat = pickle.loads(s)
self.assertEqual(oldpat, newpat)
def test_flags(self):
for flag in [re.I, re.M, re.X, re.S, re.L]:
self.assertNotEqual(re.compile('^pattern$', flag), None)
def test_bug_612074(self):
pat=u"["+re.escape(u"\u2039")+u"]"
self.assertEqual(re.compile(pat) and 1, 1)
