def _build(self):
builder = AcoraBuilder()
for idx, item in enumerate(self._regexes_or_assoc):
#
# First we compile all regular expressions and save them to
# the re_cache.
#
if isinstance(item, tuple):
regex = item[0]
regex = regex.encode(DEFAULT_ENCODING)
self._re_cache[regex] = re.compile(regex,
self._re_compile_flags)
if regex in self._translator:
raise ValueError('Duplicated regex "%s"' % regex)
self._translator[regex] = item[1:]
elif isinstance(item, basestring):
regex = item.encode(DEFAULT_ENCODING)
self._re_cache[regex] = re.compile(regex,
self._re_compile_flags)
else:
raise ValueError('Can NOT build MultiRE with provided values.')
#
# Now we extract the string literals (longer than hint_len only) from
# the regular expressions and populate the acora index
#
regex_hints = esmre.hints(regex)
regex_keywords = esmre.shortlist(regex_hints)
if not regex_keywords:
self._regexes_with_no_keywords.append(regex)
continue
# Get the longest one
regex_keyword = regex_keywords[0]
if len(regex_keyword) <= self._hint_len:
self._regexes_with_no_keywords.append(regex)
continue
# Add this keyword to the acora index, and also save a way to associate the
# keyword with the regular expression
regex_keyword = regex_keyword.lower()
builder.add(regex_keyword)
regexes_matching_keyword = self._keyword_to_re.get(
regex_keyword, [])
regexes_matching_keyword.append(regex)
self._keyword_to_re[regex_keyword] = regexes_matching_keyword
return builder.build()
def _build(self):
builder = AcoraBuilder()
for idx, item in enumerate(self._regexes_or_assoc):
#
# First we compile all regular expressions and save them to
# the re_cache.
#
if isinstance(item, tuple):
regex = item[0]
regex = regex.encode(DEFAULT_ENCODING)
self._re_cache[regex] = re.compile(regex, self._re_compile_flags)
if regex in self._translator:
raise ValueError('Duplicated regex "%s"' % regex)
self._translator[regex] = item[1:]
elif isinstance(item, basestring):
regex = item.encode(DEFAULT_ENCODING)
self._re_cache[regex] = re.compile(regex, self._re_compile_flags)
else:
raise ValueError('Can NOT build MultiRE with provided values.')
#
# Now we extract the string literals (longer than hint_len only) from
# the regular expressions and populate the acora index
#
regex_hints = esmre.hints(regex)
regex_keywords = esmre.shortlist(regex_hints)
if not regex_keywords:
self._regexes_with_no_keywords.append(regex)
continue
# Get the longest one
regex_keyword = regex_keywords[0]
if len(regex_keyword) <= self._hint_len:
self._regexes_with_no_keywords.append(regex)
continue
# Add this keyword to the acora index, and also save a way to associate the
# keyword with the regular expression
regex_keyword = regex_keyword.lower()
builder.add(regex_keyword)
regexes_matching_keyword = self._keyword_to_re.get(regex_keyword, [])
regexes_matching_keyword.append(regex)
self._keyword_to_re[regex_keyword] = regexes_matching_keyword
return builder.build()
def enter(self, regex, obj):
self.lock.acquire()
try:
if self.fixed:
raise TypeError("enter() cannot be called after query()")
regex_hints = esmre.hints(regex)
keywords = esmre.shortlist(regex_hints)
if not keywords:
raise ValueError('Failed due to performance reasons.'
' Need more hints for RE: %s' % regex)
for hint in keywords:
if len(hint) <= self.hint_len:
raise ValueError('Failed due to performance reasons.'
' Need longer hints for RE: %s' % regex)
self.esm.enter(hint.lower(), obj)
finally:
self.lock.release()
def enter(self, regex, obj):
self.lock.acquire()
try:
if self.fixed:
raise TypeError("enter() cannot be called after query()")
regex_hints = esmre.hints(regex)
keywords = esmre.shortlist(regex_hints)
if not keywords:
raise ValueError('Failed due to performance reasons.'
' Need more hints for RE: %s' % regex)
for hint in keywords:
if len(hint) <= self.hint_len:
raise ValueError('Failed due to performance reasons.'
' Need longer hints for RE: %s' % regex)
self.esm.enter(hint.lower(), obj)
finally:
self.lock.release()
def checkShortlist(self, expected_shortlist, hints):
self.assertEqual(expected_shortlist, esmre.shortlist(hints))