Python sre_constants.SUBPATTERN属性代码示例

# 需要导入模块: import sre_constants [as 别名]
# 或者: from sre_constants import SUBPATTERN [as 别名]
def __init__(self, lexicon, flags=0):
        from sre_constants import BRANCH, SUBPATTERN
        if isinstance(flags, RegexFlag):
            flags = flags.value
        self.lexicon = lexicon
        # combine phrases into a compound pattern
        p = []
        s = sre_parse.Pattern()
        s.flags = flags
        for phrase, action in lexicon:
            gid = s.opengroup()
            p.append(sre_parse.SubPattern(s, [
                (SUBPATTERN, (gid, 0, 0, sre_parse.parse(phrase, flags))),
                ]))
            s.closegroup(gid, p[-1])
        p = sre_parse.SubPattern(s, [(BRANCH, (None, p))])
        self.scanner = sre_compile.compile(p) 

# 需要导入模块: import sre_constants [as 别名]
# 或者: from sre_constants import SUBPATTERN [as 别名]
def __init__(self, lexicon, flags=0):
        from sre_constants import BRANCH, SUBPATTERN
        self.lexicon = lexicon
        # combine phrases into a compound pattern
        p = []
        s = sre_parse.Pattern()
        s.flags = flags
        for phrase, action in lexicon:
            p.append(sre_parse.SubPattern(s, [
                (SUBPATTERN, (len(p)+1, sre_parse.parse(phrase, flags))),
                ]))
        s.groups = len(p)+1
        p = sre_parse.SubPattern(s, [(BRANCH, (None, p))])
        self.scanner = sre_compile.compile(p) 

# 需要导入模块: import sre_constants [as 别名]
# 或者: from sre_constants import SUBPATTERN [as 别名]
def _compile(regexp):

    parsed = sre_parse.parse(regexp)
    parsed = _remove_group_identifiers(parsed)

    # Add grouping parentheses around the regexp; this will allow
    # us to access the material that was split on.
    # Need to set the Pattern to expect a single group

    pattern = sre_parse.Pattern()
    pattern.groups += 1
    grouped = sre_parse.SubPattern(pattern)
    grouped.append((sre_constants.SUBPATTERN, (1, parsed)))

    return sre_compile.compile(grouped, re.UNICODE | re.MULTILINE | re.DOTALL) 

# 需要导入模块: import sre_constants [as 别名]
# 或者: from sre_constants import SUBPATTERN [as 别名]
def __init__(self, lexicon, flags=0):
        from sre_constants import BRANCH, SUBPATTERN
        self.lexicon = lexicon
        # combine phrases into a compound pattern
        p = []
        s = sre_parse.Pattern()
        s.flags = flags
        for phrase, action in lexicon:
            gid = s.opengroup()
            p.append(sre_parse.SubPattern(s, [
                (SUBPATTERN, (gid, sre_parse.parse(phrase, flags))),
                ]))
            s.closegroup(gid, p[-1])
        p = sre_parse.SubPattern(s, [(BRANCH, (None, p))])
        self.scanner = sre_compile.compile(p) 

# 需要导入模块: import sre_constants [as 别名]
# 或者: from sre_constants import SUBPATTERN [as 别名]
def _make_match_string_from_pattern(parsetree, makebad=False, groups=None):
    collect = []
    if groups is None:
        groups = {}
    for op, val in parsetree:
        if op is sre_constants.LITERAL:
            if makebad:
                collect.append(chr((val ^ 4) & 0xFF)) # flip bit 4
                if random.randint(0,9) == 0:
                    makebad = False # don't error everything
            else:
                collect.append(chr(val))
        elif op is sre_constants.CATEGORY:
            collect.append(get_substitute(val, makebad))
        elif op is sre_constants.IN:
            if val[0][0] is sre_constants.CATEGORY:
                collect.append(_make_match_string_from_pattern(val, False, groups))
            else:
                collect.append(chr(random.choice(val)[1]))
        elif op is sre_constants.BRANCH:
            collect.append(_make_match_string_from_pattern(val[1][random.randint(0,1)], False, groups))
        elif op is sre_constants.SUBPATTERN:
            string = _make_match_string_from_pattern(val[1], False, groups)
            groups[val[0]] = string
            collect.append(string)
        elif op is sre_constants.MAX_REPEAT or op is sre_constants.MIN_REPEAT:
            for i in xrange(random.randint(val[0], min(val[1], 10))):
                collect.append(_make_match_string_from_pattern(val[2], False, groups))
        elif op is sre_constants.ANY:
            collect.append(random.choice(ANYCHAR))
        elif op is sre_constants.GROUPREF:
            collect.append(groups[val])
        elif op is sre_constants.AT:
            pass # ignore anchors
        else:
            raise UnhandledOpError("Unhandled RE op: %r" % (op,))
    if makebad: # in case it didn't get done yet.
        collect.insert(random.randrange(0, len(collect)), random.choice(ascii.printable))
    return "".join(collect) 

# 需要导入模块: import sre_constants [as 别名]
# 或者: from sre_constants import SUBPATTERN [as 别名]
def __init__(self, lexicon, flags=0):
            from sre_constants import BRANCH, SUBPATTERN
            self.lexicon = lexicon
            # combine phrases into a compound pattern
            p = []
            s = sre_parse.Pattern()
            s.flags = flags
            for phrase, action in lexicon:
                p.append(sre_parse.SubPattern(s, [
                    (SUBPATTERN, (len(p)+1, sre_parse.parse(phrase, flags))),
                    ]))
            s.groups = len(p)+1
            p = sre_parse.SubPattern(s, [(BRANCH, (None, p))])
            self.scanner = sre_compile.compile(p) 

# 需要导入模块: import sre_constants [as 别名]
# 或者: from sre_constants import SUBPATTERN [as 别名]
def __init__(self, pattern, flags=0, charset=CHARSET, max_count=None):
        # If the RE module cannot compile it, we give up quickly
        self.matcher = re.compile(r'(?:%s)\Z' % pattern, flags)
        if not flags & re.DOTALL:
            charset = ''.join(c for c in charset if c != '\n')
        self.charset = charset

        self.named_group_lookup = self.matcher.groupindex

        if flags & re.IGNORECASE:
            raise ParseError('Flag "i" not supported. https://github.com/google/sre_yield/issues/4')
        elif flags & re.UNICODE:
            raise ParseError('Flag "u" not supported. https://github.com/google/sre_yield/issues/3')
        elif flags & re.LOCALE:
            raise ParseError('Flag "l" not supported. https://github.com/google/sre_yield/issues/5')

        if max_count is None:
            self.max_count = MAX_REPEAT_COUNT
        else:
            self.max_count = max_count

        self.has_groupref = False

        # Configure the parser backends
        self.backends = {
            sre_constants.LITERAL: lambda y: [chr(y)],
            sre_constants.RANGE: lambda l, h: [chr(c) for c in range(l, h+1)],
            sre_constants.SUBPATTERN: self.maybe_save,
            sre_constants.BRANCH: self.branch_values,
            sre_constants.MIN_REPEAT: self.max_repeat_values,
            sre_constants.MAX_REPEAT: self.max_repeat_values,
            sre_constants.AT: self.empty_list,
            sre_constants.ASSERT: self.empty_list,
            sre_constants.ASSERT_NOT: self.empty_list,
            sre_constants.ANY:
                lambda _: self.in_values(((sre_constants.NEGATE,),)),
            sre_constants.IN: self.in_values,
            sre_constants.NOT_LITERAL: self.not_literal,
            sre_constants.CATEGORY: self.category,
            sre_constants.GROUPREF: self.groupref,
        }
        # Now build a generator that knows all possible patterns
        self.raw = self.sub_values(sre_parse.parse(pattern, flags))
        # Configure this class instance to know about that result
        self.length = self.raw.__len__() 

# 需要导入模块: import sre_constants [as 别名]
# 或者: from sre_constants import SUBPATTERN [as 别名]
def _remove_group_identifiers(parsed_re):
    """
    Modifies the given parsed regular expression, replacing all groupings
    (as indicated by parenthesis in the regular expression string) with
    non-grouping variants (indicated with '(?:...)'). This works on the
    output of sre_parse.parse, modifing the group indentifier in
    SUBPATTERN structures to None.

    @param parsed_re: the output of sre_parse.parse(string)
    @type parsed_re: C{SubPattern}
    """
    if isinstance(parsed_re, sre_parse.SubPattern):
        # If it's a SubPattern, replace each item with its processed
        # equivalent. These classes are mutable, so that in-place
        # modification is allowed.
        for i in range(len(parsed_re)):
            parsed_re[i] = _remove_group_identifiers(parsed_re[i])
        return parsed_re
    elif isinstance(parsed_re, list) or isinstance(parsed_re, tuple):
        # Otherwise, if it's a sequence, check for the tell-tale
        # SUBPATTERN item and repair the sub item if needed
        to_process = list(parsed_re)
        if to_process[0] == sre_constants.SUBPATTERN:
            # replace next int with None
            sub_item = list(to_process[1])
            sub_item[0] = None
            to_process[1] = tuple(sub_item)

        # Process each item, in the case of nested SUBPATTERNS
        processed = map(_remove_group_identifiers, to_process)

        # Coerce back into the original type
        if isinstance(parsed_re, list):
            return processed
        else:
            return tuple(processed)
    else:
        # Don't need to do anything to other types
        return parsed_re

# Replace any grouping parentheses with non-grouping ones.  We
# need to do this, because the list returned by re.sub will
# contain an element corresponding to every set of grouping
# parentheses.  We must not touch escaped parentheses, and
# need to handle the case of escaped escapes (e.g. "\\(").
# We also need to handle nested parentheses, which means our
# regexp contexts must be zero-width. There are also issues with
# parenthesis appearing in bracketed contexts, hence we've
# operated on the intermediate parse structure from sre_parse. 

# 需要导入模块: import sre_constants [as 别名]
# 或者: from sre_constants import SUBPATTERN [as 别名]
def check_anchor_state(self, matcher, arguments):
        # A bit of a hack to support zero-width leading anchors.  The goal is
        # that /^(a|b)$/ will match properly, and that /a^b/ or /a\bb/ throws
        # an error.  (It's unfortunate that I couldn't easily handle /$^/ which
        # matches the empty string; I went for the common case.)
        #
        # There are three states, for example:
        # / STATE_START
        # | / STATE_START (^ causes no transition here, but is illegal at STATE_MIDDLE or STATE_END)
        # | |  / STATE_START (\b causes no transition here, but advances MIDDLE to END)
        # | |  | / (same as above for ^)
        # | |  | | / STATE_MIDDLE (anything besides ^ and \b advances START to MIDDLE)
        # | |  | | | / still STATE_MIDDLE
        # . .  . . . .  / advances MIDDLE to END
        #  ^ \b ^ X Y \b $
        if self.state == STATE_START:
            if matcher == sre_constants.AT:
                if arguments[0] in (sre_constants.AT_END, sre_constants.AT_END_STRING):
                    self.state = STATE_END
                elif arguments[0] == sre_constants.AT_NON_BOUNDARY:
                    # This is nonsensical at beginning of string
                    raise ParseError("Anchor %r found at START state" % (arguments[0],))
                # All others (AT_BEGINNING, AT_BEGINNING_STRING, and AT_BOUNDARY) remain in START.
            elif matcher != sre_constants.SUBPATTERN:
                self.state = STATE_MIDDLE
            # subpattern remains in START
        elif self.state == STATE_END:
            if matcher == sre_constants.AT:
                if arguments[0] not in (
                    sre_constants.AT_END,
                    sre_constants.AT_END_STRING,
                    sre_constants.AT_BOUNDARY,
                ):
                    raise ParseError("Anchor %r found at END state" % (arguments[0],))
                # those three remain in END
            elif matcher != sre_constants.SUBPATTERN:
                raise ParseError(
                    "Non-end-anchor %r found at END state" % (arguments[0],)
                )
            # subpattern remains in END
        else:  # self.state == STATE_MIDDLE
            if matcher == sre_constants.AT:
                if arguments[0] not in (
                    sre_constants.AT_END,
                    sre_constants.AT_END_STRING,
                    sre_constants.AT_BOUNDARY,
                ):
                    raise ParseError(
                        "Anchor %r found at MIDDLE state" % (arguments[0],)
                    )
                # All others (AT_END, AT_END_STRING, AT_BOUNDARY) advance to END.
                self.state = STATE_END