Python regex 模块，MULTILINE 实例源码

def __init__(self, **kwargs):
        """
        Args:
            annotators: None or empty set (only tokenizes).
            substitutions: if true, normalizes some token types (e.g. quotes).
        """
        self._regexp = regex.compile(
            '(?P<digit>%s)|(?P<title>%s)|(?P<abbr>%s)|(?P<neg>%s)|(?P<hyph>%s)|'
            '(?P<contr1>%s)|(?P<alphanum>%s)|(?P<contr2>%s)|(?P<sdquote>%s)|'
            '(?P<edquote>%s)|(?P<ssquote>%s)|(?P<esquote>%s)|(?P<dash>%s)|'
            '(?<ellipses>%s)|(?P<punct>%s)|(?P<nonws>%s)' %
            (self.DIGIT, self.TITLE, self.ABBRV, self.NEGATION, self.HYPHEN,
             self.CONTRACTION1, self.ALPHA_NUM, self.CONTRACTION2,
             self.START_DQUOTE, self.END_DQUOTE, self.START_SQUOTE,
             self.END_SQUOTE, self.DASH, self.ELLIPSES, self.PUNCT,
             self.NON_WS),
            flags=regex.IGNORECASE + regex.UNICODE + regex.MULTILINE
        )
        if len(kwargs.get('annotators', {})) > 0:
            logger.warning('%s only tokenizes! Skipping annotators: %s' %
                           (type(self).__name__, kwargs.get('annotators')))
        self.annotators = set()
        self.substitutions = kwargs.get('substitutions', True)

def __init__(self, **kwargs):
        """
        Args:
            annotators: None or empty set (only tokenizes).
            substitutions: if true, normalizes some token types (e.g. quotes).
        """
        self._regexp = regex.compile(
            '(?P<digit>%s)|(?P<title>%s)|(?P<abbr>%s)|(?P<neg>%s)|(?P<hyph>%s)|'
            '(?P<contr1>%s)|(?P<alphanum>%s)|(?P<contr2>%s)|(?P<sdquote>%s)|'
            '(?P<edquote>%s)|(?P<ssquote>%s)|(?P<esquote>%s)|(?P<dash>%s)|'
            '(?<ellipses>%s)|(?P<punct>%s)|(?P<nonws>%s)' %
            (self.DIGIT, self.TITLE, self.ABBRV, self.NEGATION, self.HYPHEN,
             self.CONTRACTION1, self.ALPHA_NUM, self.CONTRACTION2,
             self.START_DQUOTE, self.END_DQUOTE, self.START_SQUOTE,
             self.END_SQUOTE, self.DASH, self.ELLIPSES, self.PUNCT,
             self.NON_WS),
            flags=regex.IGNORECASE + regex.UNICODE + regex.MULTILINE
        )
        if len(kwargs.get('annotators', {})) > 0:
            logger.warning('%s only tokenizes! Skipping annotators: %s' %
                           (type(self).__name__, kwargs.get('annotators')))
        self.annotators = set()
        self.substitutions = kwargs.get('substitutions', True)

def test_constants(self):
        if regex.I != regex.IGNORECASE:
            self.fail()
        if regex.L != regex.LOCALE:
            self.fail()
        if regex.M != regex.MULTILINE:
            self.fail()
        if regex.S != regex.DOTALL:
            self.fail()
        if regex.X != regex.VERBOSE:
            self.fail()

def test_dollar_matches_twice(self):
        # $ matches the end of string, and just before the terminating \n.
        pattern = regex.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 = regex.compile('$', regex.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 regex_match(text, pattern):
    """Test if a regex pattern is contained within a text."""
    try:
        pattern = re.compile(
            pattern,
            flags=re.IGNORECASE + re.UNICODE + re.MULTILINE,
        )
    except BaseException:
        return False
    return pattern.search(text) is not None

def regex_match_score(prediction, pattern):
    """Check if the prediction matches the given regular expression."""
    try:
        compiled = re.compile(
            pattern,
            flags=re.IGNORECASE + re.UNICODE + re.MULTILINE
        )
    except BaseException:
        logger.warn('Regular expression failed to compile: %s' % pattern)
        return False
    return compiled.match(prediction) is not None

def __init__(self, **kwargs):
        """
        Args:
            annotators: None or empty set (only tokenizes).
        """
        self._regexp = regex.compile(
            '(%s)|(%s)' % (self.ALPHA_NUM, self.NON_WS),
            flags=regex.IGNORECASE + regex.UNICODE + regex.MULTILINE
        )
        if len(kwargs.get('annotators', {})) > 0:
            logger.warning('%s only tokenizes! Skipping annotators: %s' %
                           (type(self).__name__, kwargs.get('annotators')))
        self.annotators = set()

def regex_match(text, pattern):
    """Test if a regex pattern is contained within a text."""
    try:
        pattern = re.compile(
            pattern,
            flags=re.IGNORECASE + re.UNICODE + re.MULTILINE,
        )
    except BaseException:
        return False
    return pattern.search(text) is not None

def regex_match_score(prediction, pattern):
    """Check if the prediction matches the given regular expression."""
    try:
        compiled = re.compile(
            pattern,
            flags=re.IGNORECASE + re.UNICODE + re.MULTILINE
        )
    except BaseException:
        logger.warn('Regular expression failed to compile: %s' % pattern)
        return False
    return compiled.match(prediction) is not None

def __init__(self, **kwargs):
        """
        Args:
            annotators: None or empty set (only tokenizes).
        """
        self._regexp = regex.compile(
            '(%s)|(%s)' % (self.ALPHA_NUM, self.NON_WS),
            flags=regex.IGNORECASE + regex.UNICODE + regex.MULTILINE
        )
        if len(kwargs.get('annotators', {})) > 0:
            logger.warning('%s only tokenizes! Skipping annotators: %s' %
                           (type(self).__name__, kwargs.get('annotators')))
        self.annotators = set()

def make_xpath_ranges(html, phrase):
    '''Given a HTML string and a `phrase`, build a regex to find offsets
    for the phrase, and then build a list of `XPathRange` objects for
    it.  If this fails, return empty list.

    '''
    if not html:
        return []
    if not isinstance(phrase, unicode):
        try:
            phrase = phrase.decode('utf8')
        except:
            logger.info('failed %r.decode("utf8")', exc_info=True)
            return []
    phrase_re = re.compile(
        phrase, flags=re.UNICODE | re.IGNORECASE | re.MULTILINE)
    spans = []
    for match in phrase_re.finditer(html, overlapped=False):
        spans.append(match.span())  # a list of tuple(start, end) char indexes

    # now run fancy aligner magic to get xpath info and format them as
    # XPathRange per above
    try:
        xpath_ranges = list(char_offsets_to_xpaths(html, spans))
    except:
        logger.info('failed to get xpaths', exc_info=True)
        return []
    ranges = []
    for xpath_range in filter(None, xpath_ranges):
        ranges.append(dict(
            start=dict(node=xpath_range.start_xpath,
                       idx=xpath_range.start_offset + 1),
            end=dict(node=xpath_range.end_xpath,
                     idx=xpath_range.end_offset)))

    return ranges