Python pyparsing.replaceWith函数代码示例

 def get_parser_atoms(self):
     """ Function defining the atoms of the grammar """
     atoms = super(LanguageMathematica, self).get_parser_atoms()
     atoms['assign'] = (Literal("=") | Literal(":=") | Literal("=="))
     atoms['consts'] = Keyword("Pi").setParseAction(replaceWith('PI')) |\
                       Keyword("E").setParseAction(replaceWith('E'))
     return atoms

def jsParse(inStr):
	# This disaster is a context-free grammar parser for parsing javascript object literals.
	# It needs to be able to handle a lot of the definitional messes you find in in-the-wild
	# javascript object literals.
	# Unfortunately, Javascript is /way/ more tolerant then JSON when it comes to object literals
	# so we can't just parse objects using python's `json` library.

	TRUE = pp.Keyword("true").setParseAction( pp.replaceWith(True) )
	FALSE = pp.Keyword("false").setParseAction( pp.replaceWith(False) )
	NULL = pp.Keyword("null").setParseAction( pp.replaceWith(None) )

	jsonString = pp.quotedString.setParseAction( pp.removeQuotes )
	jsonNumber = pp.Combine( pp.Optional('-') + ( '0' | pp.Word('123456789',pp.nums) ) +
											pp.Optional( '.' + pp.Word(pp.nums) ) +
											pp.Optional( pp.Word('eE',exact=1) + pp.Word(pp.nums+'+-',pp.nums) ) )

	jsonObject   = pp.Forward()
	jsonValue    = pp.Forward()
	jsonDict     = pp.Forward()
	jsonArray    = pp.Forward()
	jsonElements = pp.Forward()

	rawText      = pp.Regex('[a-zA-Z_$][0-9a-zA-Z_$]*')

	commaToNull = pp.Word(',,', exact=1).setParseAction(pp.replaceWith(None))
	jsonElements << pp.ZeroOrMore(commaToNull) + pp.Optional(jsonObject) + pp.ZeroOrMore((pp.Suppress(',') + jsonObject) | commaToNull)

	jsonValue << ( jsonString | jsonNumber | TRUE | FALSE | NULL )


	dictMembers = pp.delimitedList( pp.Group( (rawText | jsonString) + pp.Suppress(':') + (jsonValue | jsonDict | jsonArray)))
	jsonDict << ( pp.Dict( pp.Suppress('{') + pp.Optional(dictMembers) + pp.ZeroOrMore(pp.Suppress(',')) + pp.Suppress('}') ) )
	jsonArray << ( pp.Group(pp.Suppress('[') + pp.Optional(jsonElements) + pp.Suppress(']') ) )
	jsonObject << (jsonValue | jsonDict | jsonArray)

	jsonComment = pp.cppStyleComment
	jsonObject.ignore( jsonComment )

	def convertDict(s, l, toks):

		return dict(toks.asList())

	def convertNumbers(s,l,toks):
		n = toks[0]
		try:
			return int(n)
		except ValueError:
			return float(n)

	jsonNumber.setParseAction(convertNumbers)
	jsonDict.setParseAction(convertDict)

	# jsonObject.setDebug()
	jsonObject.parseString('"inStr"').pop()
	return jsonObject.parseString(inStr).pop()

def date(name=None, compulsory=False):
    """
    Creates the grammar for a Date (D) field, accepting only numbers in a certain pattern.

    The field can be compulsory, in which case the empty date, composed only of zeros, is disallowed.

    :param name: name for the field
    :param compulsory: indicates if the empty date is disallowed
    :return: grammar for the date field
    """

    if name is None:
        name = 'Date Field'

    # Basic field
    # This regex allows values from 00000101 to 99991231
    field = pp.Regex('[0-9][0-9][0-9][0-9](0[1-9]|1[0-2])(0[1-9]|[1-2][0-9]|3[0-1])')

    # Parse action
    field.setParseAction(lambda d: datetime.datetime.strptime(d[0], '%Y%m%d').date())

    # Name
    field.setName(name)

    if not compulsory:
        # If it is not compulsory the empty date is accepted
        optional = pp.Regex('[0]{8}')
        optional.setParseAction(pp.replaceWith(None))

        # Name
        optional.setName(name)

        # If it is not compulsory the empty date is accepted
        empty = pp.Regex('[ ]{8}')
        empty.setParseAction(pp.replaceWith(None))

        # Name
        empty.setName(name)

        field = field | optional | empty

        # Name
        field.setName(name)

    # White spaces are not removed
    field.leaveWhitespace()

    return field

    def parse_string(self, string):
        '''Populate a new object from a string.
        
        Parsing is hard, so we're going to call out to the pyparsing
        library here.  I hope you installed it!
        FTR: this is hideous.
        '''
        from pyparsing import Suppress, Regex, quotedString, restOfLine, Keyword, nestedExpr, Group, OneOrMore, Word, Literal, alphanums, removeQuotes, replaceWith, nums, printables
        gr_eq = Literal('=')
        gr_stripped_string = quotedString.copy().setParseAction( removeQuotes )
        gr_opt_quoted_string = gr_stripped_string | restOfLine
        gr_number = Word(nums)
        gr_yn = Keyword('yes', caseless=True).setParseAction(replaceWith('1')) | Keyword('no', caseless=True).setParseAction(replaceWith('0'))

        def _handle_ip(*x):
            a,b,c =  x[2]
            return '  %s = { %s }' % (a,c[0])

        def _handle_diraddr(*x):
            a,b,c =  x[2]
            self._set(DIRADDRESSES, '  %s' % '\n  '.join(c))
            return

        def np(words, fn = gr_opt_quoted_string, action=None):
            p = Keyword(words[0], caseless=True)
            for w in words[1:]:
                p = p | Keyword(w, caseless=True)
            p = p + gr_eq + fn
            p.setParseAction(action)
            return p
            
        gr_name = np((NAME,), action=lambda x: self._set_name(x[2]))
        gr_address = np((ADDRESS,), action=self._parse_setter(ADDRESS))
        gr_fd_conn = np(PList('fd connect timeout'), gr_number, self._parse_setter(FD_CONNECT_TIMEOUT, True))
        gr_heart = np(PList('heartbeat interval'), gr_number, self._parse_setter(HEARTBEATINTERVAL, True))
        gr_max_con = np(PList('maximum console connections'),
                        gr_number, self._parse_setter(MAXIMUMCONSOLECONNECTIONS, True))
        gr_max_jobs = np(PList('maximum concurrent jobs'), gr_number, action=self._parse_setter(MAXIMUMCONCURRENTJOBS, True))
        gr_pass = np((PASSWORD,), action=self._parse_setter(PASSWORD))
        gr_pid = np(PList('pid directory'), action=self._parse_setter(PIDDIRECTORY))
        gr_query = np(PList('query file'), action=self._parse_setter(QUERYFILE))
        gr_scripts = np(PList('scripts directory'), action=self._parse_setter(SCRIPTS_DIRECTORY))
        gr_sd_conn = np(PList('sd connect timeout'), gr_number, self._parse_setter(SD_CONNECT_TIMEOUT, True))
        gr_source = np(PList('source address'), action=self._parse_setter(SOURCEADDRESS))
        gr_stats = np(PList('statistics retention'), action=self._parse_setter(STATISTICS_RETENTION))
        gr_verid = np((VERID,), action=self._parse_setter(VERID))
        gr_messages = np((MESSAGES,), action=lambda x:self._parse_setter(MESSAGE_ID, dereference=True))
        gr_work_dir = np(PList('working directory'), action=self._parse_setter(WORKINGDIRECTORY))
        gr_port = np(PList('dir port'), gr_number, self._parse_setter(PORT, True))
        gr_monitor = np((MONITOR,), gr_yn, action=self._parse_setter(MONITOR))

        # This is a complicated one
        da_addr = np(('Addr','Port'), Word(printables), lambda x,y,z: ' '.join(z))
        da_ip = np(('IPv4','IPv6','IP'), nestedExpr('{','}', OneOrMore(da_addr).setParseAction(lambda x,y,z: ' ; '.join(z)))).setParseAction(_handle_ip)
        da_addresses = np(PList('dir addresses'), nestedExpr('{','}', OneOrMore(da_ip)), _handle_diraddr)

        gr_res = OneOrMore(gr_name | gr_address | gr_fd_conn | gr_heart | gr_max_con | gr_max_jobs | gr_pass | gr_pid | gr_query | gr_scripts | gr_sd_conn | gr_source | gr_stats | gr_verid | gr_messages | gr_work_dir | gr_port | gr_monitor | da_addresses)

        result = gr_res.parseString(string, parseAll=True)
        return 'Director: ' + self[NAME]

    def parse_string(self, string):
        '''Populate a new object from a string.
        
        Parsing is hard, so we're going to call out to the pyparsing
        library here.  I hope you installed it!
        '''
        from pyparsing import Suppress, Regex, quotedString, restOfLine, Keyword, nestedExpr, Group, OneOrMore, Word, Literal, alphanums, removeQuotes, replaceWith, nums
        gr_eq = Literal('=')
        gr_stripped_string = quotedString.copy().setParseAction( removeQuotes )
        gr_opt_quoted_string = gr_stripped_string | restOfLine
        gr_number = Word(nums)
        gr_yn = Keyword('yes', caseless=True).setParseAction(replaceWith('1')) | Keyword('no', caseless=True).setParseAction(replaceWith('0'))

        def np(words, fn = gr_opt_quoted_string, action=None):
            p = Keyword(words[0], caseless=True)
            for w in words[1:]:
                p = p | Keyword(w, caseless=True)
            p = p + gr_eq + fn
            p.setParseAction(action)
            return p

        gr_line = np((NAME,), action=lambda x: self._set_name(x[2]))
        gr_line = gr_line | np(PList('sd port'), gr_number, action=self._parse_setter(SDPORT))
        gr_line = gr_line | np((ADDRESS,), action=self._parse_setter(ADDRESS))
        gr_line = gr_line | np((PASSWORD,), action=self._parse_setter(PASSWORD))
        gr_line = gr_line | np((DEVICE,), action=self._parse_setter(DEVICE))
        gr_line = gr_line | np(PList('media type'), action=self._parse_setter(MEDIATYPE))
        gr_line = gr_line | np(PList('auto changer'), gr_yn, action=self._parse_setter(AUTOCHANGER))
        gr_line = gr_line | np(PList('maximum concurrent jobs'), gr_number, action=self._parse_setter(MAXIMUMCONCURRENTJOBS))
        gr_line = gr_line | np(PList('allow compression'), gr_yn, action=self._parse_setter(ALLOWCOMPRESSION))
        gr_line = gr_line | np(PList('heartbeat interval'), action=self._parse_setter(HEARTBEATINTERVAL))

        gr_res = OneOrMore(gr_line)
        result = gr_res.parseString(string, parseAll=True)
        return 'Storage: ' + self[NAME]

def ipi_name_number(name=None, compulsory=False):
    """
    IPI Name Number field.

    An IPI Name Number is composed of eleven digits.

    So, for example, an IPI Name Number code field can contain 00014107338.

    :param name: name for the field
    :param compulsory: indicates if the empty string is disallowed
    :return: a parser for the IPI Name Number field
    """

    if name is None:
        name = 'IPI Name Number Field'

    field = basic.numeric(11, compulsory=compulsory)

    if not compulsory:
        # If it is not compulsory then it can be set as empty
        empty = pp.Regex('[ ]{11}')
        empty.setParseAction(pp.replaceWith(None))
        empty.setName(name)

        field = empty | field
        # Name
        field.setName(name)

    field.setName(name)

    return field.setResultsName('ipi_name_n')

def audio_visual_key(name=None):
    """
    Creates the grammar for an Audio Visual Key code.

    This is a variation on the ISAN (International Standard Audiovisual Number)

    :param name: name for the field
    :return: grammar for an ISRC field
    """

    if name is None:
        name = 'AVI Field'

    society_code = basic.numeric(3)
    society_code = society_code.setName('Society Code') \
        .setResultsName('society_code')

    av_number = basic.alphanum(15)
    field_empty = pp.Regex('[ ]{15}')
    field_empty.setParseAction(pp.replaceWith(''))
    av_number = av_number | field_empty
    av_number = av_number.setName('Audio-Visual Number') \
        .setResultsName('av_number')

    field = pp.Group(society_code + av_number)

    field.setParseAction(lambda v: _to_avi(v[0]))

    field = field.setName(name)

    return field.setResultsName('audio_visual_key')

    def wrap_as_optional(self, field, name, columns):
        """
        Adds a wrapper rule to the field to accept empty strings.

        This empty string should be of the same size as the columns parameter.
        One smaller or bigger will be rejected.

        This wrapper will return None if the field is empty.

        :param field: the field to wrap
        :param name: name of the field
        :param columns: number of columns it takes
        :return: the field with an additional rule to allow empty strings
        """
        # Regular expression accepting as many whitespaces as columns
        field_empty = pp.Regex('[ ]{' + str(columns) + '}')

        field_empty.setName(name)

        # Whitespaces are not removed
        field_empty.leaveWhitespace()

        # None is returned by this rule
        field_empty.setParseAction(pp.replaceWith(None))

        field_empty = field_empty.setResultsName(field.resultsName)

        field = field | field_empty

        field.setName(name)

        field.leaveWhitespace()

        return field

def Syntax():
    imp = Literal(u'⇒').setParseAction(replaceWith("Implies"))
    conj = Literal(u'∧').setParseAction(replaceWith("Conjunction"))
    disj = Literal(u'∨').setParseAction(replaceWith("Disjunction"))
    iff = Literal(u'⇔').setParseAction(replaceWith("Disjunction"))

    #op = oneOf(u'⇒ ∧')
    op = imp | conj | disj | iff
    lpar = Literal('(').suppress()
    rpar = Literal(')').suppress()
    neg = Literal(u'¬').setParseAction(replaceWith("Negation"))
    prop = Word(u"pqr")
    expr = Forward()
    atom = prop | Group(lpar + expr + rpar)
    expr << ((atom + ZeroOrMore(op + expr)) | Group(neg + expr))

    return expr

    def parse_string(self, string):
        '''Populate a new object from a string.
        
        Parsing is hard, so we're going to call out to the pyparsing
        library here.  I hope you installed it!
        '''
        from pyparsing import quotedString, restOfLine, Keyword, nestedExpr, OneOrMore, Word, Literal, removeQuotes, nums, replaceWith, printables
        gr_eq = Literal('=')
        gr_stripped_string = quotedString.copy().setParseAction( removeQuotes )
        gr_opt_quoted_string = gr_stripped_string | restOfLine
        gr_number = Word(nums)
        gr_yn = Keyword('yes', caseless=True).setParseAction(replaceWith('1')) | Keyword('no', caseless=True).setParseAction(replaceWith('0'))

        def _handle_ip(*x):
            a,b,c =  x[2]
            return '  %s = { %s }' % (a,c[0])

        def _handle_fdaddr(*x):
            a,b,c =  x[2]
            self._set(FDADDRESSES, '  %s' % '\n  '.join(c))
            return

        def np(words, fn = gr_opt_quoted_string, action=None):
            p = Keyword(words[0], caseless=True)
            for w in words[1:]:
                p = p | Keyword(w, caseless=True)
            p = p + gr_eq + fn
            p.setParseAction(action)
            return p

        gr_line = np((NAME,), action=lambda x: self._set_name(x[2]))
        gr_line = gr_line | np((ADDRESS,), action=self._parse_setter(ADDRESS))
        gr_line = gr_line | np((CATALOG,), action=self._parse_setter(CATALOG_ID, dereference=True))
        gr_line = gr_line | np((PASSWORD,), action=self._parse_setter(PASSWORD))
        gr_line = gr_line | np(PList('file retention'), action=self._parse_setter(FILERETENTION))
        gr_line = gr_line | np(PList('job retention'), action=self._parse_setter(JOBRETENTION))
        gr_line = gr_line | np((PRIORITY,), gr_number, action=self._parse_setter(PRIORITY))
        gr_line = gr_line | np(PList('working directory'), action=self._parse_setter(WORKINGDIRECTORY))
        gr_line = gr_line | np(PList('pid directory'), action=self._parse_setter(PIDDIRECTORY))
        gr_line = gr_line | np(PList('heart beat interval'), action=self._parse_setter(HEARTBEATINTERVAL))
        gr_line = gr_line | np(PList('fd address'), action=self._parse_setter(FDADDRESS))
        gr_line = gr_line | np(PList('fd source address'), action=self._parse_setter(FDSOURCEADDRESS))
        gr_line = gr_line | np(PList('pki key pair'), action=self._parse_setter(PKIKEYPAIR))
        gr_line = gr_line | np(PList('pki master key'), action=self._parse_setter(PKIMASTERKEY))
        gr_line = gr_line | np(PList('fd port'), gr_number, action=self._parse_setter(FDPORT))
        gr_line = gr_line | np(PList('auto prune'), gr_yn, action=self._parse_setter(AUTOPRUNE))
        gr_line = gr_line | np(PList('maximum concurrent jobs'), gr_number, action=self._parse_setter(FDPORT))
        gr_line = gr_line | np(PList('pki encryption'), gr_yn, action=self._parse_setter(PKIENCRYPTION))
        gr_line = gr_line | np(PList('pki signatures'), gr_yn, action=self._parse_setter(PKISIGNATURES))

        # This is a complicated one
        da_addr = np(('Addr','Port'), Word(printables), lambda x,y,z: ' '.join(z))
        da_ip = np(('IPv4','IPv6','IP'), nestedExpr('{','}', OneOrMore(da_addr).setParseAction(lambda x,y,z: ' ; '.join(z)))).setParseAction(_handle_ip)
        da_addresses = np(('fd addresses', FDADDRESSES), nestedExpr('{','}', OneOrMore(da_ip)), _handle_fdaddr)

        gr_res = OneOrMore(gr_line|da_addresses)
        result = gr_res.parseString(string, parseAll=True)
        return 'Client: ' + self[NAME]

 def _make_valid_state_name(self, state_name):
     """Transform the input state_name into a valid state in XMLBIF.
     XMLBIF states must start with a letter an only contain letters,
     numbers and underscores.
     """
     s = str(state_name)
     s_fixed = pp.CharsNotIn(pp.alphanums + "_").setParseAction(pp.replaceWith("_")).transformString(s)
     if not s_fixed[0].isalpha():
         s_fixed = "state" + s_fixed
     return s_fixed

def iswc(name=None, compulsory=False):
    """
    ISWC field.

    A ISWC code written on a field follows the Pattern TNNNNNNNNNC. This being:
    - T: header, it is always T.
    - N: numeric value.
    - C: control digit.

    So, for example, an ISWC code field can contain T0345246801.

    :param name: name for the field
    :param compulsory: indicates if the empty string is disallowed
    :return: a parser for the ISWC field
    """

    if name is None:
        name = 'ISWC Field'

    # Header is always T
    header = pp.Literal('T').suppress()
    header = header.setName('ISWC Header').setResultsName('header')

    # ID code is composed of 9 numbers
    id_code = basic.numeric(9, compulsory=True)
    id_code = id_code.setName('ID Code').setResultsName('id_code')

    # Check digit is a single number
    check_digit = basic.numeric(1, compulsory=True)
    check_digit = check_digit.setName('Check Digit').setResultsName('check_digit')

    # T followed by 10 numbers
    field = pp.Group(header + id_code + check_digit)

    # Parse action
    field.setParseAction(lambda c: _to_iswccode(c[0]))

    # Name
    field.setName(name)

    if not compulsory:
        # If it is not compulsory then it can be set as empty
        empty = pp.Regex('[ ]{11}')
        empty.setParseAction(pp.replaceWith(None))
        empty.setName(name)

        field = empty | field
        # Name
        field.setName(name)

    # White spaces are not removed
    field.leaveWhitespace()

    return field.setResultsName('iswc')

        def convert_to_literal(tok, val):
            """
            Converts a value to pyparsing caselessliteral

            :param tok: the token we want to find/replace
            :type tok: str
            :param val: the integer the token represented
            :type val: int
            :return: pyparsing caseless literal
            :rtype: pyparsing.CaselessLiteral
            """
            return CaselessLiteral(tok) \
                .setName(tok) \
                .setParseAction(replaceWith(val))

def char_code(columns, name=None, compulsory=False):
    """
    Character set code field.

    :param name: name for the field
    :param compulsory: indicates if the empty string is disallowed
    :return: an instance of the Character set code field rules
    """
    if name is None:
        name = 'Char Code Field (' + str(columns) + ' columns)'

    if columns <= 0:
        raise BaseException()

    char_sets = None
    for char_set in _tables.character_sets():
        regex = '[ ]{' + str(15 - len(char_set)) + '}' + char_set
        if char_sets is None:
            char_sets = regex
        else:
            char_sets += '|' + regex

    # Accepted sets
    _character_sets = pp.Regex(char_sets)
    _unicode_1_16b = pp.Regex('U\+0[0-8,A-F]{3}[ ]{' + str(columns - 6) + '}')
    _unicode_2_21b = pp.Regex('U\+0[0-8,A-F]{4}[ ]{' + str(columns - 7) + '}')

    # Basic field
    char_code_field = (_character_sets | _unicode_1_16b | _unicode_2_21b)

    # Parse action
    char_code_field = char_code_field.setParseAction(lambda s: s[0].strip())

    # Name
    char_code_field.setName(name)

    if not compulsory:
        char_code_field_empty = pp.Regex('[ ]{' + str(columns) + '}')
        char_code_field_empty.setName(name)

        char_code_field_empty.leaveWhitespace()

        char_code_field_empty.setParseAction(pp.replaceWith(None))

        char_code_field = char_code_field | char_code_field_empty

        char_code_field.setName(name)

    return char_code_field

    def __init__(self):

        # codec used for encoding of usermessages
        self.codec = lightpile.codec

        EOL = pp.LineEnd()
        SOL = pp.LineStart().leaveWhitespace()
        blankline = SOL + EOL
        comment = '#' + pp.restOfLine + EOL

        self.comment_stripper = comment.setParseAction(pp.replaceWith("\n"))
        self.blankline_stripper = blankline.setParseAction(pp.replaceWith(""))

        # filegrammar
        ws = ' \t'
        standard_chars = pp.printables.replace('#', '')
        text = pp.OneOrMore(pp.White(ws) |
                            pp.quotedString |
                            pp.Word(standard_chars))
        text.setParseAction(lambda tokens: ''.join(tokens))

        lbracket = pp.Literal("[").suppress()
        rbracket = pp.Literal("]").suppress()
        # identifiers have to start with a unicode-letter and can continue
        # with any number of unicode-letters or any of
        # ':', '%', '+', '-', '_', '.', ','
        self.identifier = pp.Regex(r'[^\W\d]([^\W]|[%:_,\+\-\.])*', re.U)
        sectionhead = (lbracket + self.identifier + rbracket +
                       EOL.suppress())
        sectionline = ~sectionhead + ~pp.StringEnd() + pp.restOfLine + EOL
        sectionlines = pp.ZeroOrMore(sectionline)
        sectionlines.setParseAction(lambda tokens: ''.join(tokens))

        self.filegrammar = pp.dictOf(sectionhead, sectionlines)

        self._init_sectiongrammars()