############################################################################# TextIndexNG                The next generation TextIndex for Zope## This software is governed by a license. See# LICENSE.txt for the terms of this license.#################################################################################################################################### a new native Python QueryParser for TextIndexNG# based on the QueryParser by Sidnei da Silva# $Id: PyQueryParser.py,v 1.11.2.4 2002/06/18 20:29:25 ajung Exp $#######################################################import sys, refrom Products.TextIndexNG.BaseParser import BaseParser,QueryParserErrortokens = (    'STRING' ,     'OR',     'AND',     'ANDNOT',     'NEAR',    'QUOTE',    'OPENP',     'CLOSEP', )         t_QUOTE     = r'\"'                   t_OPENP     = r'\('           t_CLOSEP    = r'\)'          t_ignore    = '\t'def t_ANDNOT(t):    '\s+ANDNOT\s+|\s+andnot\s+'    return tdef t_AND(t):    r'\s+AND\s+|\s+and\s+'                return tdef t_OR(t):    '\s+OR\s+|\s+or\s+'         return tdef t_NEAR(t):    '\s+NEAR\s+|\s+near\s+'    return t        def t_STRING(t):    r'[\w%?*]+'     return tdef t_newline(t):    r'\n+'    t.lineno += t.value.count("\n")    def t_error(t):    print t    if t.value[0] in [' ']:        t.skip(1)    else:        raise QueryParserError,"Illegal character '%s'" % t.value[0]# Build the lexerimport lexlex.lex(debug=0)op_dict = {'AND'     : 'txI','OR'      : 'txU','NEAR'    : 'txN','ANDNOT'  : 'txAN'}def p_expr_parens(t):    """expr :    OPENP expr CLOSEP """    t[0] = '(%s)' %  t[2]def p_expr_op(t):    """expr :    expr AND expr               | expr OR expr               | expr NEAR expr               | expr ANDNOT expr    """    t[0] = '%s(%s,%s)' % (op_dict[t[2].strip().upper()], t[1], t[3])def p_expr_noop(t):    """expr :    expr expr"""    t[0] = 'txI(%s,%s)' % (t[1], t[2])def p_expr_expr_factor(t):    """expr :  factor """    t[0] = t[1]def p_factor_string(t):    """factor : string"""    t[0] = t[1]def p_factor_quote(t):    """factor :  QUOTE term QUOTE"""    t[0] = "txQ(%s)" % t[2] def p_term_1(t):    """ term : string term"""    t[0] = "%s,%s" % (t[1],t[2]) def p_term_2(t):    """ term : string"""    t[0] = t[1]str_regex = re.compile(r'[\w]+$', re.LOCALE|re.UNICODE)sim_regex = re.compile(r'[%][\w]+$', re.LOCALE|re.UNICODE)rt_regex  = re.compile(r'[\w]+[*]$', re.LOCALE|re.UNICODE)def p_string(t):    """string :  STRING"""    if str_regex.match(t[1]):        t[0] = 'LL("%s")' % t[1]    elif sim_regex.match(t[1]):        t[0] = 'PL("%s")' % t[1][1:]    elif rt_regex.match(t[1]):        t[0] = 'RTL("%s")' % t[1][:-1]    else:        t[0] = 'PML("%s")' % t[1] def p_error(t):    raise QueryParserError,"Syntax error at '%s'" % t.valueimport yaccyacc.yacc(debug=0)class Parser(BaseParser):    id = 'NewQueryParser'    parser_description = 'A TextIndex compatible parser (native Python version)'    def parse(self, query, operator):                try:            return yacc.parse( query )        except QueryParserError:            raise         except:            import traceback            traceback.print_exc()            raise QueryParserError, 'parser failed for query: %s' % query def test():    import os, sys, re,traceback, atexit    histfile = os.path.expanduser('~/.pyhist')    try:        import readline        readline.read_history_file(histfile)        atexit.register(readline.write_history_file,histfile)    except: pass    print "entering interactive query mode:"    while 1:        s = raw_input('> ')        print s          try:            P = Parser()            res = P(s)                    print 'res:',res        except:            traceback.print_exc()if __name__ == '__main__':    test()