Python pyparsing.Keyword() Examples

def _parse(mystr):

    LBRACE, RBRACE, EQUAL = map(pp.Suppress, "{}=")
    field = pp.Word(pp.printables + ' ', excludeChars='[]=')
    field.addParseAction(pp.tokenMap(str.rstrip))
    string = pp.dblQuotedString().setParseAction(pp.removeQuotes)
    number = pp.pyparsing_common.number()
    date_expr = pp.Regex(r'\d\d\d\d-\d\d-\d\d')
    time_expr = pp.Regex(r'\d\d:\d\d:\d\d\.\d\d\d')
    nan = pp.Keyword('nan')
    scalar_value = (string | date_expr | time_expr | number | nan)

    list_marker = pp.Suppress("[]")
    value_list = pp.Forward()
    jobject = pp.Forward()

    memberDef1 = pp.Group(field + EQUAL + scalar_value)
    memberDef2 = pp.Group(field + EQUAL + jobject)
    memberDef3 = pp.Group(field + list_marker + EQUAL + LBRACE + value_list +
                          RBRACE)
    memberDef = memberDef1 | memberDef2 | memberDef3

    value_list <<= (pp.delimitedList(scalar_value, ",") |
                    pp.ZeroOrMore(pp.Group(pp.Dict(memberDef2))))
    value_list.setParseAction(lambda t: [pp.ParseResults(t[:])])

    members = pp.OneOrMore(memberDef)
    jobject <<= pp.Dict(LBRACE + pp.ZeroOrMore(memberDef) + RBRACE)
    # force empty jobject to be a dict
    jobject.setParseAction(lambda t: t or {})

    parser = members
    parser = pp.OneOrMore(pp.Group(pp.Dict(memberDef)))

    return parser.parseString(mystr) 

def parse_line(attribute, string):

    Grammar = Suppress(Keyword('mpc.{}'.format(attribute)) + Keyword('=')) + String('data') + Suppress(Literal(';') + Optional(Comments))
    result, i, j = Grammar.scanString(string).next()

    return [int_else_float_except_string(s) for s in result['data'].asList()] 

def parse_table(attribute, string):
    Line = OneOrMore(Float)('data') + Literal(';') + Optional(Comments, default='')('name')
    Grammar = Suppress(Keyword('mpc.{}'.format(attribute)) + Keyword('=') + Keyword('[') + Optional(Comments)) + OneOrMore(Group(Line)) + Suppress(Keyword(']') + Optional(Comments))

    result, i, j = Grammar.scanString(string).next()

    _list = list()
    for r in result:
        _list.append([int_else_float_except_string(s) for s in r['data'].asList()])

    return _list 

def static_function(self):
        return (
            (pyparsing.Keyword("static") | pyparsing.Keyword("inline"))
            + pyparsing.OneOrMore(pyparsing.Word(pyparsing.alphanums + "_*&"))
            + parsers.anything_in_parentheses()
            + parsers.anything_in_curly()
        ).suppress() 

def _typeof_transform(self):
        """Removes typeof expressions."""
        return (
            _OPEN_PARENTHESIS
            + pyparsing.Keyword('typeof')
            + parsers.anything_in_parentheses()
            + _CLOSE_PARENTHESIS
        ).suppress() 

def XXXX_cast_expression(self):
        """A function returning a parser for parsing cast expressions.

        Args:
            expression: a pyparsing parser for parsing an expression to be cast.

        Returns:
            A (pyparsing) parser for parsing cast expressions.
        """
        word = pyparsing.Word(pyparsing.alphanums + '_*[]')
        nested = pyparsing.Forward().setName("nested")
        nested << pyparsing.Combine(
            pyparsing.Literal('(').suppress()
            + pyparsing.Combine(
                pyparsing.ZeroOrMore(self._integer() | word | nested))
            + pyparsing.Literal(')').suppress()
        )
        typeof_expression = (
            _OPEN_PARENTHESIS
            + pyparsing.Keyword('typeof')
            + nested("typeof_arg")
            + _CLOSE_PARENTHESIS
        )

        type_expression = (
            typeof_expression
            | nested("simple_type")
        )
        return (
            type_expression
            + ~(_PLUS | _MINUS)
            + self.expression("expression")
        ).setParseAction(self._create_cast_expression) 

def _typeof_expression(self):
        keyword = (
            pyparsing.Keyword('typeof')
            | pyparsing.Keyword('__typeof__')
        )
        return pyparsing.Combine(
            keyword
            + pyparsing.Literal('(')
            + parsers.anything_beetween('()')
            + pyparsing.Literal(')')
        ) 

def make_keyword(kwd_str, kwd_value):
    return pp.Keyword(kwd_str).setParseAction(pp.replaceWith(kwd_value)) 

def kw(kw):
  return p.Keyword(kw).suppress() 

def lexical_analysis(self, src):
        delimited = re.sub(r'\s+', ' ', ' '.join(src.strip().split('\n'))).split(';')
        result = []
        for stmt in delimited:
            if stmt == '':
                return result
            string = pp.Regex('[a-zA-Z0-9=_]+')
            nums = pp.Regex('[0-9]+')
            ws = pp.OneOrMore(pp.White()).suppress()
            lp = pp.Regex('[(]').suppress()
            rp = pp.Regex('[)]').suppress()
            c = pp.Regex('[,]').suppress()
            q = pp.Regex("[']").suppress()
    
            table_name = string.setResultsName('table_name')
            create_table = (pp.Keyword('CREATE', caseless = True) + ws + pp.Keyword('TABLE', caseless = True) + ws + pp.Optional(pp.Keyword('IF', caseless = True) + ws + pp.Keyword('NOT', caseless = True) + ws + pp.Keyword('EXISTS', caseless = True))).suppress() + table_name + lp 
    
            column_name = string.setResultsName('column_name')
            data_type = string.setResultsName('data_type')
            length = lp + nums.setResultsName('length') + rp
            nullable = (pp.Optional(pp.Keyword('NOT', caseless = True) + ws) + pp.Keyword('NULL', caseless = True)).setResultsName('nullable')
            default_value = pp.Keyword('DEFAULT', caseless = True).suppress() + ws + string.setResultsName('default_value')
            auto_increment = pp.Keyword('AUTO_INCREMENT', caseless = True).setResultsName('auto_increment')
            column = pp.Optional(ws) + column_name + ws + data_type + pp.Optional(pp.MatchFirst([length, ws + nullable, ws + default_value, ws + auto_increment])) + pp.Optional(pp.MatchFirst([ws + nullable, ws + default_value, ws + auto_increment])) + pp.Optional(pp.MatchFirst([ws + default_value, ws + auto_increment])) + pp.Optional(ws + auto_increment) + pp.Optional(ws) + c
    
            primary_key = pp.Keyword('PRIMARY KEY', caseless = True).suppress() + lp + pp.OneOrMore(q + string.setResultsName('primary_key') + q + pp.Optional(c)) + rp + pp.Optional(c)
            key = pp.Keyword('KEY', caseless = True).suppress() + lp + q + string.setResultsName('key') + q + pp.Optional(c) + rp + pp.Optional(c)
    
            parser = create_table + pp.OneOrMore(pp.Group(column)) + pp.Optional(primary_key) + pp.Optional(key) + rp + pp.OneOrMore(ws + string).suppress()
            result.append(parser.parseString(stmt, parseAll=True))
        return result 

def __init__(self, cfg_mgr, pre_defined_actions, scripts_dir):
        """init
        Args:
            cfg_mgr: Instance of ibis.utilities.config_manager.ConfigManager
            pre_defined_actions: list of default action ids
        """
        self.cfg_mgr = cfg_mgr
        self.logger = get_logger(self.cfg_mgr)
        self.pre_defined_actions = pre_defined_actions
        header_pattern = pp.Keyword('action').setResultsName('action_header')
        body_pattern = pp.Word(pp.alphanums + '._/')
        body_pattern = body_pattern.setResultsName('action_id')
        self.pattern = header_pattern + \
            pp.Group(body_pattern).setResultsName('action_body')
        self.scripts_dir = scripts_dir 

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()


# Stolen from http://stackoverflow.com/a/12017573/268006 

def parser(self):
        """
        This function returns a parser.
        The grammar should be like most full text search engines (Google, Tsearch, Lucene).
        
        Grammar:
        - a query consists of alphanumeric words, with an optional '*' wildcard
          at the end of a word
        - a sequence of words between quotes is a literal string
        - words can be used together by using operators ('and' or 'or')
        - words with operators can be grouped with parenthesis
        - a word or group of words can be preceded by a 'not' operator
        - the 'and' operator precedes an 'or' operator
        - if an operator is missing, use an 'and' operator
        """
        operatorOr = Forward()
        
        operatorWord = Group(Combine(Word(alphanums) + Suppress('*'))).setResultsName('wordwildcard') | \
                            Group(Word(alphanums)).setResultsName('word')
        
        operatorQuotesContent = Forward()
        operatorQuotesContent << (
            (operatorWord + operatorQuotesContent) | operatorWord
        )
        
        operatorQuotes = Group(
            Suppress('"') + operatorQuotesContent + Suppress('"')
        ).setResultsName("quotes") | operatorWord
        
        operatorParenthesis = Group(
            (Suppress("(") + operatorOr + Suppress(")"))
        ).setResultsName("parenthesis") | operatorQuotes

        operatorNot = Forward()
        operatorNot << (Group(
            Suppress(Keyword("not", caseless=True)) + operatorNot
        ).setResultsName("not") | operatorParenthesis)

        operatorAnd = Forward()
        operatorAnd << (Group(
            operatorNot + Suppress(Keyword("and", caseless=True)) + operatorAnd
        ).setResultsName("and") | Group(
            operatorNot + OneOrMore(~oneOf("and or") + operatorAnd)
        ).setResultsName("and") | operatorNot)
        
        operatorOr << (Group(
            operatorAnd + Suppress(Keyword("or", caseless=True)) + operatorOr
        ).setResultsName("or") | operatorAnd)

        return operatorOr.parseString 

def parser(self):
        """
        This function returns a parser.
        The grammar should be like most full text search engines (Google, Tsearch, Lucene).

        Grammar:
        - a query consists of alphanumeric words, with an optional '*' wildcard
          at the end of a word
        - a sequence of words between quotes is a literal string
        - words can be used together by using operators ('and' or 'or')
        - words with operators can be grouped with parenthesis
        - a word or group of words can be preceded by a 'not' operator
        - the 'and' operator precedes an 'or' operator
        - if an operator is missing, use an 'and' operator
        """
        operatorOr = Forward()

        operatorWord = Word(wordchars).setResultsName('value')

        operatorQuotesContent = Forward()
        operatorQuotesContent << (
            (operatorWord + operatorQuotesContent) | operatorWord
            )

        operatorQuotes = Group(
            Suppress('"') + operatorQuotesContent + Suppress('"')
        ).setResultsName("quotes") | operatorWord

        prefix = (Word(alphanums).setResultsName('index') + Word('=').setResultsName('binop'))
        operatorParenthesis = Group(
            Optional(prefix) +
            (Suppress("(") + operatorOr + Suppress(")"))
        ).setResultsName("parenthesis") | Group(prefix + operatorQuotes).setResultsName('term') | operatorQuotes

        operatorNot = Forward()
        operatorNot << (Group(
            Suppress(Keyword("not", caseless=True)) + operatorNot
        ).setResultsName("not") | operatorParenthesis)

        operatorAnd = Forward()
        operatorAnd << (Group(
            operatorNot + Suppress(Keyword("and", caseless=True)) + operatorAnd
        ).setResultsName("and") | Group(
            operatorNot + OneOrMore(~oneOf("and or", caseless=True) + operatorAnd)
        ).setResultsName("and") | operatorNot)

        operatorProximity = Forward()
        operatorProximity << (Group(
            operatorParenthesis + Suppress(Literal("near,")) + Word(nums).setResultsName('distance') + operatorParenthesis
        ).setResultsName("near") | Group(
            operatorParenthesis + Suppress(Literal("span,")) + Word(nums).setResultsName('distance') + operatorParenthesis
        ).setResultsName("span") | operatorAnd)

        operatorOr << (Group(
            operatorProximity + Suppress(Keyword("or", caseless=True)) + operatorOr
        ).setResultsName("or") | operatorProximity)

        return operatorOr.parseString