Python pyparsing.Word类代码示例

    def __init__(self, params):
        self.name = ""
        self.type = None
        self.default_value = None
        self.range = None

        if isinstance(params, list):
            self.name = params[0]
            for attr in params[1:]:
                if attr[0] == 'default':
                    self.default_value = attr[1]
                elif attr[0] == 'type':
                    assert attr[1] == 'string' or attr[1] == 'integer'
                    self.type = attr[1]
                elif attr[0] == 'range':
                    self.range = [attr[1], attr[2]]
                    if int(self.range[0]) >= int(self.range[1]):
                        raise ValueError('Invalid range')

            if not self.default_value is None:
                try:
                    # consistenza di tipo sui valori dello slot corrente
                    parser = Word(alphas) if self.type == 'string' else Word(nums)
                    parser.parseString(self.default_value)
                    # se slot corrente ha un range definito
                    if not self.range is None:
                        spec_range = '{0} >= {1} and {0} <= {2}'.format(self.default_value, self.range[0], self.range[1])
                        if not eval(spec_range):
                            raise ValueError('Default value doesn\'t satisfy range constraints')
                except (ParseException, SyntaxError):
                    raise ValueError('Incorrect default value for slot')
        else:
            self.name = params

    def parse_sexp(data):
        '''parse sexp/S-expression format and return a python list'''
        # define punctuation literals
        LPAR, RPAR, LBRK, RBRK, LBRC, RBRC, VBAR = map(Suppress, "()[]{}|")

        decimal = Word("123456789", nums).setParseAction(lambda t: int(t[0]))
        bytes = Word(printables)
        raw = Group(decimal.setResultsName("len") + Suppress(":") + bytes).setParseAction(OtrPrivateKeys.verifyLen)
        token = Word(alphanums + "-./_:*+=")
        base64_ = Group(Optional(decimal, default=None).setResultsName("len") + VBAR
            + OneOrMore(Word( alphanums +"+/=" )).setParseAction(lambda t: b64decode("".join(t)))
            + VBAR).setParseAction(OtrPrivateKeys.verifyLen)

        hexadecimal = ("#" + OneOrMore(Word(hexnums)) + "#")\
                        .setParseAction(lambda t: int("".join(t[1:-1]),16))
        qString = Group(Optional(decimal, default=None).setResultsName("len") +
                                dblQuotedString.setParseAction(removeQuotes)).setParseAction(OtrPrivateKeys.verifyLen)
        simpleString = raw | token | base64_ | hexadecimal | qString

        display = LBRK + simpleString + RBRK
        string_ = Optional(display) + simpleString

        sexp = Forward()
        sexpList = Group(LPAR + ZeroOrMore(sexp) + RPAR)
        sexp << ( string_ | sexpList )

        try:
            sexpr = sexp.parseString(data)
            return sexpr.asList()[0][1:]
        except ParseFatalException, pfe:
            print("Error:", pfe.msg)
            print(pfe.loc)
            print(pfe.markInputline())

def read_stats(result_dir, stats_file_name):
    stat_rule = Word(printables) + Word('nan.%' + nums) + Optional(restOfLine)

    stats = []

    try:
        with open(path.join(result_dir, stats_file_name)) as stats_file:
            i = 0
            for stat_line in stats_file:
                if len(stats) <= i:
                    stats.append(collections.OrderedDict())

                try:
                    stat = stat_rule.parseString(stat_line)
                    key = stat[0]
                    value = stat[1]

                    stats[i][key] = value
                except ParseException as e:
                    # print(e)
                    pass

                if 'End Simulation Statistics' in stat_line:
                    i += 1
    except Exception as e:
        print(e)
        return None
    else:
        return stats

    def __init__(self):
        # literals
        star = Literal('*')
        comma = Suppress(',')

        # indentifiers
        identifier = Word(alphas, alphanums+'_')
        alias      = identifier.copy()

        # select clause
        column_name = Combine(Optional(alias + '.') +
                              identifier +
                              Optional(' as ' + identifier))\
                              .setResultsName('column_name')
        select = Keyword('select', caseless=1)
        select_clause = (star | Group(delimitedList(column_name, comma)))\
            .setResultsName('select_clause')

        # from clause
        from_  = Keyword('from', caseless=1)
        table_name =  delimitedList(identifier + Optional(alias), comma)
        from_clause = table_name.setResultsName('from_clause')

        # select statment
        self.select_stmt = select + select_clause + from_ + from_clause

def _construct_grammar():
    logical_operator = get_logical_operator()
    logical_expression = get_logical_expression()

    facets_expression = get_facet_expression()
    highlight_expression = get_highlight_expression()
    sort_expression = get_sort_expression()
    aggs_expression = get_aggregations_expression()
    nested_expression = get_nested_expression()

    # The below line describes how the type expression should be.
    type_expression = Word('type')\
        + Word(':').suppress()\
        + Word(srange("[a-zA-Z0-9_]"))\
        + Optional(CaselessLiteral('AND')).suppress()
    type_expression.setParseAction(parse_type_expression)

    base_expression = Optional(highlight_expression)\
        + Optional(sort_expression)\
        + Optional(type_expression)\
        + ZeroOrMore(
            (facets_expression
             | aggs_expression
             | nested_expression
             | logical_expression)
            + Optional(logical_operator)
        ).setParseAction(parse_one_or_more_logical_expressions)
    base_expression.setParseAction(parse_type_logical_facets_expression)

    return base_expression

 def set_param(self,line):
     # define grammar
     a = line.find("]")
     b = line.find("=")
     if(a>0 and b>0 and (b-a)==1):
         return
     else:
         modele = Word( alphas ) + "[" + Word(nums) + "]" + Word(objectPath) + "=" + Word( divers )
         try:
             pd = modele.parseString( line )
         except ParseException as pe:
             pass
         else:
             obj = pd[0]
             key = pd[4]
             value = pd[6][:len(pd[6])-1]
             nb = int(pd[2])
             if(key[0]=="."):
                 key = key[1:]                           #expect ".keyword"
                 if(key.find(".")<0):                    #a single keyword
                     if(key in ocd_str_param):
                         setattr(self,key,set_str(value))
                         #print("->  ocd[{id}].{key}={value}".format(id=self.id,key=key,value=value))
                     elif(key in ocd_int_param):
                         setattr(self,key,int(value))
                         #print("->  ocd[{id}].{key}={value}".format(id=self.id,key=key,value=value))
                 else:
                     keywords = key.split(".")
                     if(keywords[0]=="likelihood"):
                         self.likelihood.set_param(keywords,value)
                     elif(keywords[0]=="posterior"):
                         self.posterior.set_param(keywords,value)

def parse_morse_letter(line):
    """Parse a line of input from the morse code table file and convert it into a parsetree."""
    letter = Word(alphas)
    token = Word(alphas)
    morse_letter_expression = letter.setResultsName('letter')\
                            + OneOrMore(token).setResultsName('tokens')
    return morse_letter_expression.parseString(line)

    def open(self, id):
        self.id = id
        item = 'BODY'
        typ, data = self.imap.fetch(id, '(' + item + ')')
        assert typ == 'OK'

        inline_parse = Forward()
        inline_parse << Suppress("(") + Group(OneOrMore(Or([Keyword("NIL"), QuotedString('"'), Word(nums), inline_parse ]))) + Suppress(")")
        parse = Word(nums) + Suppress("(") + Keyword(item) + Suppress("(") + Group(OneOrMore(inline_parse)) + ZeroOrMore(Or([Keyword("NIL"), QuotedString('"'), Word(nums)])) +  Suppress(")") + Suppress(")")
        p = parse.parseString(data[0])

        #print data[0]
        #print p
        #print

        self.attachment = []
        for i in p[2]:
            #while 'NIL' in i:
            #    i.remove('NIL')

            a = {
                'type'          : '/'.join(i[0:2]).lower(),
                i[2][0].lower() : i[2][1],
            }
                
            self.attachment.append(a)

def nexus_iter(infile):
    import pyparsing
    pyparsing.ParserElement.enablePackrat()
    from pyparsing import Word, Literal, QuotedString, CaselessKeyword, \
         OneOrMore, Group, Optional, Suppress, Regex, Dict
    ## beginblock = Suppress(CaselessKeyword("begin") +
    ##                       CaselessKeyword("trees") + ";")
    ## endblock = Suppress((CaselessKeyword("end") |
    ##                      CaselessKeyword("endblock")) + ";")
    comment = Optional(Suppress("[&") + Regex(r'[^]]+') + Suppress("]"))
    ## translate = CaselessKeyword("translate").suppress()
    name = Word(string.letters+string.digits+"_.") | QuotedString("'")
    ## ttrec = Group(Word(string.digits).setResultsName("number") +
    ##               name.setResultsName("name") +
    ##               Optional(",").suppress())
    ## ttable = Group(translate + OneOrMore(ttrec) + Suppress(";"))
    newick = Regex(r'[^;]+;')
    tree = (CaselessKeyword("tree").suppress() +
            Optional("*").suppress() +
            name.setResultsName("tree_name") +
            comment.setResultsName("tree_comment") +
            Suppress("=") +
            comment.setResultsName("root_comment") +
            newick.setResultsName("newick"))
    ## treesblock = Group(beginblock +
    ##                    Optional(ttable.setResultsName("ttable")) +
    ##                    Group(OneOrMore(tree)) +
    ##                    endblock)

    def not_begin(s): return s.strip().lower() != "begin trees;"
    def not_end(s): return s.strip().lower() not in ("end;", "endblock;")
    def parse_ttable(f):
        ttable = {}
        while True:
            s = f.next().strip()
            if not s: continue
            if s.lower() == ";": break
            if s[-1] == ",": s = s[:-1]
            k, v = s.split()
            ttable[k] = v
            if s[-1] == ";": break
        return ttable
            
    # read lines between "begin trees;" and "end;"
    f = itertools.takewhile(not_end, itertools.dropwhile(not_begin, infile))
    s = f.next().strip().lower()
    if s != "begin trees;":
        print sys.stderr, "Expecting 'begin trees;', got %s" % s
        raise StopIteration
    ttable = {}
    while True:
        try: s = f.next().strip()
        except StopIteration: break
        if not s: continue
        if s.lower() == "translate":
            ttable = parse_ttable(f)
            print "ttable: %s" % len(ttable)
        elif s.split()[0].lower()=='tree':
            match = tree.parseString(s)
            yield nexus.Newick(match, ttable)

def parse_ampersand_comment(s):
    import pyparsing
    pyparsing.ParserElement.enablePackrat()
    from pyparsing import Word, Literal, QuotedString, CaselessKeyword, \
         OneOrMore, Group, Optional, Suppress, Regex, Dict
    word = Word(string.letters+string.digits+"%_")
    key = word.setResultsName("key") + Suppress("=")
    single_value = (Word(string.letters+string.digits+"-.") |
                    QuotedString("'") |
                    QuotedString('"'))
    range_value = Group(Suppress("{") +
                        single_value.setResultsName("min") +
                        Suppress(",") +
                        single_value.setResultsName("max") +
                        Suppress("}"))
    pair = (key + (single_value | range_value).setResultsName("value"))
    g = OneOrMore(pair)
    d = []
    for x in g.searchString(s):
        v = x.value
        if type(v) == str:
            try: v = float(v)
            except ValueError: pass
        else:
            try: v = map(float, v.asList())
            except ValueError: pass
        d.append((x.key, v))
    return d

    def __createGram(self):

        if self.notNeedSpace:
            lNot = Keyword(self.operators['not'])
        else:
            lNot = Literal(self.operators['not'])
        
        lAnd = Literal(self.operators['and'])
        lOr = Literal(self.operators['or'])
        lImp = Literal(self.operators['impL'])
        lEqu = Literal(self.operators['equ'])
        
        lTrue = Keyword(self.constants['true'])
        lFalse = Keyword(self.constants['false'])
        
        lVar = Word(alphas, alphanums+'_')
        
        lVar.setParseAction(self.ffactory.createLogicVariable)
        lTrue.setParseAction(self.ffactory.createLogicTruth)
        lFalse.setParseAction(self.ffactory.createLogicFalse)

        factor = lTrue | lFalse | lVar
        
        expression = myparsing.operatorPrecedence(factor,
        [
         (lNot, 1, opAssoc.RIGHT, self.ffactory.createNotOperation),
         (lAnd, 2, opAssoc.LEFT, self.ffactory.createAndOperation),
         (lOr,  2, opAssoc.LEFT, self.ffactory.createOrOperation),
         (lImp, 2, opAssoc.LEFT, self.ffactory.createImpicationOperation),
         (lEqu, 2, opAssoc.LEFT, self.ffactory.createEquvalenceOperation)
        ],
        [('(', ')'), ('[', ']'), ('{', '}')])
    

        self.final = expression + StringEnd()

def getLogLineBNF_DBpedia36():
    global logLineBNF
    
    if logLineBNF is None:
        integer = Word( nums )
        ipAddress = delimitedList( integer, ".", combine=True )
        hashipAddress = Word(nums+alphas)
        timeZoneOffset = Word("+-",nums)
        month = Word(string.uppercase, string.lowercase, exact=3)
        serverDateTime = Group( Suppress("[") + 
                                Combine( integer + "/" + month + "/" + integer +
                                        " " + integer + ":" + integer + ":" + integer ) +
                                timeZoneOffset + 
                                Suppress("]") )
                         
        logLineBNF = ( hashipAddress.setResultsName("ipAddr") + 
                       Suppress("-") +
                       ("-" | Word( alphas+nums+"@._" )).setResultsName("auth") +
                       serverDateTime.setResultsName("timestamp") + 
                       dblQuotedString.setResultsName("cmd").setParseAction(getCmdFields2) +
                       (integer | "-").setResultsName("statusCode") + 
                       (integer | "-").setResultsName("numBytesSent")  + 
                       dblQuotedString.setResultsName("referrer").setParseAction(removeQuotes) +
                       dblQuotedString.setResultsName("clientSfw").setParseAction(removeQuotes) )
    return logLineBNF

    def init_parser(self):

        INTEGER = Word(nums)
        INTEGER.setParseAction(lambda x: int(x[0]))

        header = INTEGER("species_count") + INTEGER("sequence_length") +\
            Suppress(restOfLine)
        header.setParseAction(self.set_header)

        sequence_name = Word(
            alphas + nums + "!#$%&\'*+-./;<=>[email protected][\\]^_`{|}~",
            max=100)

        # Take a copy and disallow line breaks in the bases
        bases = self.BASES.copy()
        bases.setWhitespaceChars(" \t")
        seq_start = sequence_name("species") + bases(
            "sequence") + Suppress(LineEnd())
        seq_start.setParseAction(self.set_seq_start)
        seq_start_block = OneOrMore(seq_start)
        seq_start_block.setParseAction(self.set_start_block)

        seq_continue = bases("sequence") + Suppress(LineEnd())
        seq_continue.setParseAction(self.set_seq_continue)

        seq_continue_block = Suppress(LineEnd()) + OneOrMore(seq_continue)
        seq_continue_block.setParseAction(self.set_continue_block)

        return header + seq_start_block + ZeroOrMore(seq_continue_block)

def func_tokens(dictionary, parse_action):
    func_name = Word(alphas+'_', alphanums+'_')

    func_ident = Combine('$' + func_name.copy()('funcname'))
    func_tok = func_ident + originalTextFor(nestedExpr())('args')
    func_tok.leaveWhitespace()
    func_tok.setParseAction(parse_action)
    func_tok.enablePackrat()

    rx_tok = Combine(Literal('$').suppress() + Word(nums)('num'))

    def replace_token(tokens):
        index = int(tokens.num)
        return dictionary.get(index, u'')

    rx_tok.setParseAction(replace_token)

    strip = lambda s, l, tok: tok[0].strip()
    text_tok = CharsNotIn(u',').setParseAction(strip)
    quote_tok = QuotedString('"')

    if dictionary:
        arglist = Optional(delimitedList(quote_tok | rx_tok | text_tok))
    else:
        arglist = Optional(delimitedList(quote_tok | text_tok))

    return func_tok, arglist, rx_tok

def load_variables(filename):
		"""Load random variables definitions from file (in C45 format but with class at the end).
		File must contain information in format 'Variable Name: Values.' as in the example below:
		A: true,false.
		B: 0,1,2.
		C: c1,c2,c3,c4.
		D: one.
		"""
		from DataStructures.randomvariables import RandomVariable
		RV = []
#		variable = Word(caps + lowers + digits).setResultsName("name") + ": " + OneOrMore(Group(Word(caps + lowers + digits) + Optional("." + Word(caps + lowers + digits))) + Suppress(Optional(","))).setResultsName("domain") + "."

		variable = Word(caps + lowers + digits).setResultsName("name") + ": " + OneOrMore(Word(caps + lowers + digits + ".") + Optional(Suppress(","))).setResultsName("domain")  
		for line in file(filename):
			if not line[0] == '#':
				dataline = line[0:(len(line)-2)]
				#print dataline
				rv = variable.parseString(dataline)
				#print rv.name
				domain = []
				for value in rv.domain:
					#print value,
					value = ''.join(value)
					if value.isdigit():
						#print 'lv#', value
						domain.append(int(value))
					else:
						domain.append(value)
				#print
				var = RandomVariable(rv.name,domain)
				RV.append(var)
		return RV