mdw@git.distorted.org.uk Git - anag/blob - pcre.c

   1 /* -*-c-*-
   2  *
   3  * Matches Perl-compatible regular expressions
   4  *
   5  * (c) 2002 Mark Wooding
   6  */
   7
   8 /*----- Licensing notice --------------------------------------------------*
   9  *
  10  * This file is part of Anag: a simple wordgame helper.
  11  *
  12  * Anag is free software; you can redistribute it and/or modify
  13  * it under the terms of the GNU General Public License as published by
  14  * the Free Software Foundation; either version 2 of the License, or
  15  * (at your option) any later version.
  16  *
  17  * Anag is distributed in the hope that it will be useful,
  18  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  19  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  20  * GNU General Public License for more details.
  21  *
  22  * You should have received a copy of the GNU General Public License
  23  * along with Anag; if not, write to the Free Software Foundation,
  24  * Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
  25  */
  26
  27 /*----- Header files ------------------------------------------------------*/
  28
  29 #include "anag.h"
  30
  31 #ifdef HAVE_PCRE2
  32 #  define PCRE2_CODE_UNIT_WIDTH 8
  33 #  include <pcre2.h>
  34 #endif
  35
  36 #ifdef HAVE_PCRE
  37 #  include <pcre.h>
  38 #endif
  39
  40 /*----- Data structures ---------------------------------------------------*/
  41
  42 typedef struct node_pcre {
  43   node n;
  44   const char *s;
  45 #ifdef HAVE_PCRE2
  46   pcre2_code *rx;
  47   pcre2_match_data *m;
  48 #endif
  49 #ifdef HAVE_PCRE
  50   pcre *rx;
  51   pcre_extra *rx_study;
  52   int *ovec;
  53   int ovecsz;
  54 #endif
  55 } node_pcre;
  56
  57 /*----- Main code ---------------------------------------------------------*/
  58
  59 /* --- Node matcher --- */
  60
  61 static int n_pcre(node *nn, const char *p, size_t sz)
  62 {
  63   node_pcre *n = (node_pcre *)nn;
  64 #ifdef HAVE_PCRE2
  65   char buf[128];
  66   int rc;
  67 #endif
  68 #ifdef HAVE_PCRE
  69   int e;
  70 #endif
  71
  72 #ifdef HAVE_PCRE2
  73   rc = pcre2_match(n->rx, (PCRE2_SPTR)p, sz, 0, 0, n->m, 0);
  74   if (rc >= 0) return (1);
  75   else switch (rc) {
  76     case PCRE2_ERROR_NOMATCH: return (0);
  77     default:
  78       rc = pcre2_get_error_message(rc, (PCRE2_UCHAR *)buf, sizeof(buf));
  79       assert(!rc); die("pcre2 matching failed': %s", buf);
  80   }
  81 #endif
  82 #ifdef HAVE_PCRE
  83   e = pcre_exec(n->rx, n->rx_study, p, sz, 0, 0, n->ovec, n->ovecsz);
  84   if (e >= 0) return (1);
  85   if (e == PCRE_ERROR_NOMATCH) return (0);
  86   die("unexpected PCRE error code %d", e);
  87 #endif
  88 }
  89
  90 /* --- Node creation --- */
  91
  92 node *pcrenode(const char *const *av)
  93 {
  94   node_pcre *n = xmalloc(sizeof(*n));
  95 #ifdef HAVE_PCRE2
  96   char buf[128];
  97   int err;
  98   PCRE2_SIZE eo;
  99   uint32_t c;
 100 #endif
 101 #ifdef HAVE_PCRE
 102   const char *e;
 103   int eo;
 104   int c;
 105 #endif
 106
 107   n->n.func = n_pcre;
 108
 109 #ifdef HAVE_PCRE2
 110   n->rx = pcre2_compile((PCRE2_SPTR)av[0], strlen(av[0]), PCRE2_CASELESS,
 111                         &err, &eo, 0);
 112   if (!n->rx) {
 113     err = pcre2_get_error_message(err, (PCRE2_UCHAR *)buf, sizeof(buf));
 114     assert(!err); die("bad regular expression `%s': %s", av[0], buf);
 115   }
 116   err = pcre2_pattern_info(n->rx, PCRE2_INFO_BACKREFMAX, &c);
 117   assert(!err);
 118   n->m = pcre2_match_data_create_from_pattern(n->rx, 0);
 119   if (!n->m) {
 120     err = pcre2_get_error_message(err, (PCRE2_UCHAR *)buf, sizeof(buf));
 121     assert(!err); die("failed to allocate match data: %s", buf);
 122   }
 123   pcre2_jit_compile(n->rx, PCRE2_JIT_COMPLETE);
 124 #endif
 125 #ifdef HAVE_PCRE
 126   n->rx = pcre_compile(av[0], PCRE_CASELESS, &e, &eo, 0);
 127   if (!n->rx) die("bad regular expression `%s': %s", av[0], e);
 128   n->rx_study = pcre_study(n->rx, 0, &e);
 129   if (e) die("error studying pattern `%s': %s", av[0], e);
 130   pcre_fullinfo(n->rx, n->rx_study, PCRE_INFO_CAPTURECOUNT, &c);
 131   n->ovecsz = 2*c;
 132   n->ovec = xmalloc(n->ovecsz*sizeof(*n->ovec));
 133 #endif
 134
 135   return (&n->n);
 136 }
 137
 138 /*----- That's all, folks -------------------------------------------------*/