X-Git-Url: http://git.indexdata.com/?a=blobdiff_plain;f=dfa%2Fdfa.c;h=340cffc0ed5db3e76e42b23d5c5a78f066d66582;hb=58e3c5132f9fe86fefbf2e130275ab9980eeed1e;hp=072c39700919854db1a1cb7c641894c785ff1a8b;hpb=75b086d1cbf7d4134e599d2d3b0cd51abb34292a;p=idzebra-moved-to-github.git diff --git a/dfa/dfa.c b/dfa/dfa.c index 072c397..340cffc 100644 --- a/dfa/dfa.c +++ b/dfa/dfa.c @@ -1,78 +1,26 @@ -/* - * Copyright (C) 1994-1998, Index Data I/S - * All rights reserved. - * Sebastian Hammer, Adam Dickmeiss - * - * $Log: dfa.c,v $ - * Revision 1.21 1998-06-22 11:33:39 adam - * Added two type casts. - * - * Revision 1.20 1998/06/08 14:40:44 adam - * Fixed problem with signed character(s) in regular expressions. - * - * Revision 1.19 1998/01/12 14:39:39 adam - * Fixed bug in term_Tnode. - * - * Revision 1.18 1997/09/29 09:05:17 adam - * Thread safe DFA module. We simply had to put a few static vars to - * the DFA_parse structure. - * - * Revision 1.17 1997/09/18 08:59:17 adam - * Extra generic handle for the character mapping routines. - * - * Revision 1.16 1997/09/05 15:29:57 adam - * Changed prototype for chr_map_input - added const. - * Added support for C++, headers uses extern "C" for public definitions. - * - * Revision 1.15 1997/02/10 10:19:20 adam - * Added facility for open character sets, eg [a-]. - * - * Revision 1.14 1996/10/29 13:57:22 adam - * Include of zebrautl.h instead of alexutil.h. - * - * Revision 1.13 1996/06/17 14:24:08 adam - * Bug fix: read_charset didn't handle character mapping. - * - * Revision 1.12 1996/06/04 10:20:02 adam - * Added support for character mapping. - * - * Revision 1.11 1996/01/08 19:15:24 adam - * Allow single $ in expressions. - * - * Revision 1.10 1996/01/08 09:09:17 adam - * Function dfa_parse got 'const' string argument. - * New functions to define char mappings made public. - * - * Revision 1.9 1995/12/06 12:24:58 adam - * Removed verbatim mode code. - * - * Revision 1.8 1995/12/06 09:09:58 adam - * Work on left and right anchors. - * - * Revision 1.7 1995/11/27 09:23:02 adam - * New berbatim hook in regular expressions. "[]n ..". - * - * Revision 1.6 1995/10/16 09:31:25 adam - * Bug fix. - * - * Revision 1.5 1995/10/02 15:17:58 adam - * Bug fix in dfa_delete. - * - * Revision 1.4 1995/09/28 09:18:52 adam - * Removed various preprocessor defines. - * - * Revision 1.3 1995/09/04 12:33:26 adam - * Various cleanup. YAZ util used instead. - * - * Revision 1.2 1995/01/25 11:30:50 adam - * Simple error reporting when parsing regular expressions. - * Memory usage reduced. - * - * Revision 1.1 1995/01/24 16:02:52 adam - * New private header file in dfa module (dfap.h). - * Module no longer uses yacc to parse regular expressions. - * - */ +/* $Id: dfa.c,v 1.29 2003-06-18 11:46:33 adam Exp $ + Copyright (C) 1995,1996,1997,1998,1999,2000,2001,2002 + Index Data Aps + +This file is part of the Zebra server. + +Zebra 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, or (at your option) any later +version. + +Zebra 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 Zebra; see the file LICENSE.zebra. If not, write to the +Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA +02111-1307, USA. +*/ + + #include #include @@ -423,18 +371,28 @@ static int read_charset (struct DFA_parse *parse_info) { if (!esc0 && ch0 == ']') break; - if (parse_info->cmap) - { - const char **mapto; - char mapfrom[2]; - const char *mcp = mapfrom; - mapfrom[0] = ch0; - mapto = (*parse_info->cmap)(parse_info->cmap_data, &mcp, 1); - assert (mapto); - ch0 = mapto[0][0]; - } - add_BSet (parse_info->charset, parse_info->look_chars, ch0); - ch1 = nextchar_set (parse_info, &esc1); + if (!esc0 && ch0 == '-') + { + ch1 = ch0; + esc1 = esc0; + ch0 = 1; + add_BSet (parse_info->charset, parse_info->look_chars, ch0); + } + else + { + if (parse_info->cmap) + { + const char **mapto; + char mapfrom[2]; + const char *mcp = mapfrom; + mapfrom[0] = ch0; + mapto = (*parse_info->cmap)(parse_info->cmap_data, &mcp, 1); + assert (mapto); + ch0 = mapto[0][0]; + } + add_BSet (parse_info->charset, parse_info->look_chars, ch0); + ch1 = nextchar_set (parse_info, &esc1); + } if (!esc1 && ch1 == '-') { int open_range = 0; @@ -523,7 +481,7 @@ static int lex_sub(struct DFA_parse *parse_info) { const int *cc; for (cc = parse_info->charMap; *cc; cc += 2) - if (*cc == parse_info->look_ch) + if (*cc == (int) (parse_info->look_ch)) { if (!cc[1]) --parse_info->expr_ptr; @@ -759,6 +717,7 @@ static void dfa_trav (struct DFA_parse *parse_info, struct Tnode *n) n->lastpos = mk_Set (poset); n->lastpos = add_Set (poset, n->lastpos, n->pos); if (debug_dfa_trav) + { if (n->u.ch[0] < 0) printf ("#%d (n#%d)", -n->u.ch[0], -n->u.ch[1]); else if (n->u.ch[1] > n->u.ch[0]) @@ -772,6 +731,7 @@ static void dfa_trav (struct DFA_parse *parse_info, struct Tnode *n) } else out_char (n->u.ch[0]); + } } if (debug_dfa_trav) { @@ -842,10 +802,12 @@ static void mk_dfa_tran (struct DFA_parse *parse_info, struct DFA_states *dfas) for (pos_i = pos; (i = *pos_i) != -1; ++pos_i) if (posar[i]->u.ch[1] >= char_1 && (c=posar[i]->u.ch[0]) < char_0) + { if (c < char_1) char_0 = char_1; else char_0 = c; + } if (char_0 > max_char) break; @@ -961,7 +923,8 @@ void dfa_parse_cmap_clean (struct DFA *d) if (!dfa->charMap) { dfa->charMapSize = 7; - dfa->charMap = imalloc (dfa->charMapSize * sizeof(*dfa->charMap)); + dfa->charMap = (int *) + imalloc (dfa->charMapSize * sizeof(*dfa->charMap)); } dfa->charMap[0] = 0; } @@ -981,7 +944,7 @@ void dfa_parse_cmap_new (struct DFA *d, const int *cmap) if (dfa->charMap) ifree (dfa->charMap); dfa->charMapSize = size; - dfa->charMap = imalloc (size * sizeof(*dfa->charMap)); + dfa->charMap = (int *) imalloc (size * sizeof(*dfa->charMap)); } memcpy (dfa->charMap, cmap, size * sizeof(*dfa->charMap)); } @@ -1021,7 +984,7 @@ void dfa_parse_cmap_add (struct DFA *d, int from, int to) size = dfa->charMapSize; if (indx >= size) { - int *cn = imalloc ((size+16) * sizeof(*dfa->charMap)); + int *cn = (int *) imalloc ((size+16) * sizeof(*dfa->charMap)); memcpy (cn, dfa->charMap, indx*sizeof(*dfa->charMap)); ifree (dfa->charMap); dfa->charMap = cn; @@ -1065,7 +1028,6 @@ static struct DFA_parse *dfa_parse_init (void) parse_info->anyset = mk_BSet (&parse_info->charset); res_BSet (parse_info->charset, parse_info->anyset); - add_BSet (parse_info->charset, parse_info->anyset, '\n'); com_BSet (parse_info->charset, parse_info->anyset); parse_info->use_Tnode = parse_info->max_Tnode = 0; parse_info->start = parse_info->end = NULL; @@ -1102,10 +1064,13 @@ static struct DFA_states *mk_dfas (struct DFA_parse *dfap, int poset_chunk) if (debug_dfa_followpos) pr_followpos(parse_info); - init_DFA_states (&dfas, parse_info->poset, STATE_HASH); + init_DFA_states (&dfas, parse_info->poset, (int) (STATE_HASH)); mk_dfa_tran (parse_info, dfas); if (debug_dfa_tran) + { + printf ("PR_TRAN\n"); pr_tran (parse_info, dfas); + } if (dfa_verbose) pr_verbose (parse_info, dfas); del_pos(parse_info); @@ -1118,7 +1083,7 @@ struct DFA *dfa_init (void) { struct DFA *dfa; - dfa = imalloc (sizeof(*dfa)); + dfa = (struct DFA *) imalloc (sizeof(*dfa)); dfa->parse_info = dfa_parse_init (); dfa->state_info = NULL; dfa->states = NULL; @@ -1141,6 +1106,13 @@ int dfa_parse (struct DFA *dfa, const char **pattern) assert (dfa); assert (dfa->parse_info); parse_info = dfa->parse_info; + + if (!parse_info->cmap) + { + res_BSet (parse_info->charset, parse_info->anyset); + add_BSet (parse_info->charset, parse_info->anyset, '\n'); + com_BSet (parse_info->charset, parse_info->anyset); + } do_parse (parse_info, pattern, &top); if (parse_info->err_code) return parse_info->err_code;