/* ---- This file is part of SECONDO. Copyright (C) 2004, University in Hagen, Department of Computer Science, Database Systems for New Applications. SECONDO is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. SECONDO is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with SECONDO; if not, write to the Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA ---- //[=>] [\ensuremath{\Rightarrow}] //[<=>] [\ensuremath{\Leftrightarrow}] //[interior_A] [\ensuremath{A^\circ}] //[interior_B] [\ensuremath{B^\circ}] //[intersection] [\ensuremath{\cap}] //[!=] [\ensuremath{\not=}] //[emptyset] [\ensuremath{\symbol{31}}] //[\n] [\\] //[star] [\ensuremath{\ast}] 2 The Parser Specification This Parser constructs an nfa from a regular expression over integer numbers. The Grammar is: expr -> term | term [] expr term -> factor | factor term factor -> atom STARS | atom atom -> NUMBER | ( expr ) */ %{ /* 2.1 Needed includes */ // #include "RegExParser.y.h" #include "Algebras/FText/IntNfa.h" #include #include #include #include "Stack.h" #include "Algebras/FText/IVector.h" #include "Algebras/FText/Functions.h" #define YYERROR_VERBOSE /* 2.2 Some variables holding the return value */ #ifdef __cplusplus extern "C"{ int pregexlex(); int pregexerror (const char *error); void pregex_scan_string(const char* argument); } #endif int successful; IntNfa* resultNfa; Stack* stackNfa = 0; char* pregex_last_message = 0; %} /* 2.3 The Definition of the Parser */ %union { struct IntNfa* theNfa; int numval; } %name-prefix "pregex" %token OPEN CLOSE STAR OR PLUS OPTION %token INTEGER %type expr term factor atom start %% start: expr{ resultNfa = $1; } ; expr : term { $$ = $1; } | term OR expr { $$ = $1; IntNfa* n = $$; n->nfa.disjunction($3->nfa); } ; term : factor { $$ = $1; } | factor term { $$ = $1; IntNfa* n = $$; n->nfa.concat($2->nfa); } ; factor : atom { $$ = $1; } | atom STAR { $$ = $1; IntNfa* n = $$; n->nfa.star(); } | atom PLUS { $$ = $1; Nfa tmp($$->nfa); tmp.star(); $$->nfa.concat(tmp); } | atom OPTION { $$ = $1; Nfa tmp; IntNfa* n = $$; n->nfa.disjunction(tmp); } ; atom : OPEN expr CLOSE { $$ = $2; } | INTEGER { int number = $1; $$ = new IntNfa(number); stackNfa->push($$); } ; %% /* 2.4 The Error function This functions is automatically called when the parser detects an error. */ int pregexerror (const char *error) { std::cout << error << std::endl; successful = 1; resultNfa = 0; return 1; } extern "C"{void destroyLex();} /* 2.5 The parseString function This function is the only one which should be called from outside. When this function is called, the string given as the first argument is parsed according to the rules given above. */ int parsePatternRegEx(const char* argument, IntNfa** T) { successful = 0; destroyLex(); stackNfa = new Stack(); pregex_scan_string(argument); yyparse(); if(successful && pregex_last_message) { free(pregex_last_message); pregex_last_message = 0; } // kill the stack while (!stackNfa->isEmpty()) { IntNfa* elem = stackNfa->pop(); if (elem != resultNfa || successful) { delete elem; } } if (successful) { *T = 0; } else { (*T) = resultNfa; } delete stackNfa; stackNfa = 0; destroyLex(); return successful; } /* 2.6 The GetPreviousMessage function This function returns the last occured error. If no message is available, the result will be NULL. Otherwise you shoul don't forget to deallocate the memory of the result. */ char* GetPreviousMessage() { if(!pregex_last_message) { return 0; } char* M = (char*) malloc(strlen(pregex_last_message) + 1); strcpy(M, pregex_last_message); return M; }