def test_string_boundaries(self):
# See http://bugs.python.org/issue10713
self.assertEqual(re.search(r"\b(abc)\b", "abc").group(1),
"abc")
# There's a word boundary at the start of a string.
self.assertTrue(re.match(r"\b", "abc"))
# A non-empty string includes a non-boundary zero-length match.
self.assertTrue(re.search(r"\B", "abc"))
# There is no non-boundary match at the start of a string.
self.assertFalse(re.match(r"\B", "abc"))
# However, an empty string contains no word boundaries, and also no
# non-boundaries.
self.assertEqual(re.search(r"\B", ""), None)
# This one is questionable and different from the perlre behaviour,
# but describes current behavior.
self.assertEqual(re.search(r"\b", ""), None)
# A single word-character string has two boundaries, but no
# non-boundary gaps.
self.assertEqual(len(re.findall(r"\b", "a")), 2)
self.assertEqual(len(re.findall(r"\B", "a")), 0)
# If there are no words, there are no boundaries
self.assertEqual(len(re.findall(r"\b", " ")), 0)
self.assertEqual(len(re.findall(r"\b", " ")), 0)
# Can match around the whitespace.
self.assertEqual(len(re.findall(r"\B", " ")), 2)
def test_re_findall(self):
self.assertEqual(re.findall(":+", "abc"), [])
self.assertEqual(re.findall(":+", "a:b::c:::d"), [":", "::", ":::"])
self.assertEqual(re.findall("(:+)", "a:b::c:::d"), [":", "::", ":::"])
self.assertEqual(re.findall("(:)(:*)", "a:b::c:::d"), [(":", ""),
(":", ":"),
(":", "::")])
def test_re_escape_non_ascii_bytes(self):
b = u'y\u2620y\u2620y'.encode('utf-8')
b_escaped = re.escape(b)
self.assertEqual(b_escaped, b'y\\\xe2\\\x98\\\xa0y\\\xe2\\\x98\\\xa0y')
self.assertMatch(b_escaped, b)
res = re.findall(re.escape(u'\u2620'.encode('utf-8')), b)
self.assertEqual(len(res), 2)
def extract_tags_from_text(text, tags):
tags = pickle.loads(codecs.decode(tags.encode(), "base64"))
tags_found = []
text = ' '.join(text.splitlines())
for line in text.split('.'):
# Skip lines with no value to make it faster.
if len(line) < 2 or text.isdigit():
continue
for tag in tags:
try:
result = pcre.findall(tag['compiletag'], line)
times = len(result)
if times > 0:
tag_copy = tag.copy()
tag_copy.pop('compiletag')
tag_copy['times'] = times
__append_tag_to_founds(tags_found, tag_copy)
except pcre.PCREError as e:
print(e)
return {
'status':
'SUCCESS',
'excerpt':
text if len(text) <= config.SCANNED_TEXT_EXCERPT_SIZE else
text[:config.SCANNED_TEXT_EXCERPT_SIZE - 3] + ' [...]',
'result': {
'topics':
sorted(list(set([tag['topic'] for tag in tags_found]))),
'tags':
sorted(tags_found,
key=lambda t: (t['topic'], t['subtopic'], t['tag'])),
}
}
def test_bug_13899(self):
# Issue #13899: re pattern r"[\A]" should work like "A" but matches
# nothing. Ditto B and Z.
self.assertEqual(re.findall(r'[\A\B\b\C\Z]', 'AB\bCZ'),
['A', 'B', '\b', 'C', 'Z'])
def test_bug_117612(self):
self.assertEqual(re.findall(r"(a|(b))", "aba"),
[("a", ""),("b", "b"),("a", "")])
