posix/tst-pcre.c

   1 /* Regular expression tests.
   2    Copyright (C) 2003-2023 Free Software Foundation, Inc.
   3    This file is part of the GNU C Library.
   4
   5    The GNU C Library is free software; you can redistribute it and/or
   6    modify it under the terms of the GNU Lesser General Public
   7    License as published by the Free Software Foundation; either
   8    version 2.1 of the License, or (at your option) any later version.
   9
  10    The GNU C Library is distributed in the hope that it will be useful,
  11    but WITHOUT ANY WARRANTY; without even the implied warranty of
  12    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  13    Lesser General Public License for more details.
  14
  15    You should have received a copy of the GNU Lesser General Public
  16    License along with the GNU C Library; if not, see
  17    <https://www.gnu.org/licenses/>.  */
  18
  19 #include <sys/types.h>
  20 #include <mcheck.h>
  21 #include <regex.h>
  22 #include <stdio.h>
  23 #include <stdlib.h>
  24 #include <string.h>
  25
  26 int
  27 main (int argc, char **argv)
  28 {
  29   int ret = 0;
  30   char *line = NULL;
  31   size_t line_len = 0;
  32   ssize_t len;
  33   FILE *f;
  34   char *pattern = NULL, *string = NULL;
  35   regmatch_t rm[20];
  36   size_t pattern_alloced = 0, string_alloced = 0;
  37   int ignorecase = 0;
  38   int pattern_valid = 0, rm_valid = 0;
  39   size_t linenum;
  40
  41   mtrace ();
  42
  43   if (argc < 2)
  44     {
  45       fprintf (stderr, "Missing test filename\n");
  46       return 1;
  47     }
  48
  49   f = fopen (argv[1], "r");
  50   if (f == NULL)
  51     {
  52       fprintf (stderr, "Couldn't open %s\n", argv[1]);
  53       return 1;
  54     }
  55
  56   if ((len = getline (&line, &line_len, f)) <= 0
  57       || strncmp (line, "# PCRE", 6) != 0)
  58     {
  59       fprintf (stderr, "Not a PCRE test file\n");
  60       fclose (f);
  61       free (line);
  62       return 1;
  63     }
  64
  65   linenum = 1;
  66
  67   while ((len = getline (&line, &line_len, f)) > 0)
  68     {
  69       char *p;
  70       unsigned long num;
  71
  72       ++linenum;
  73
  74       if (line[len - 1] == '\n')
  75         line[--len] = '\0';
  76
  77       if (line[0] == '#')
  78         continue;
  79
  80       if (line[0] == '\0')
  81         {
  82           /* End of test.  */
  83           ignorecase = 0;
  84           pattern_valid = 0;
  85           rm_valid = 0;
  86           continue;
  87         }
  88
  89       if (line[0] == '/')
  90         {
  91           /* Pattern.  */
  92           p = strrchr (line + 1, '/');
  93
  94           pattern_valid = 0;
  95           rm_valid = 0;
  96           if (p == NULL)
  97             {
  98               printf ("%zd: Invalid pattern line: %s\n", linenum, line);
  99               ret = 1;
 100               continue;
 101             }
 102
 103           if (p[1] == 'i' && p[2] == '\0')
 104             ignorecase = 1;
 105           else if (p[1] != '\0')
 106             {
 107               printf ("%zd: Invalid pattern line: %s\n", linenum, line);
 108               ret = 1;
 109               continue;
 110             }
 111
 112           if (pattern_alloced < (size_t) (p - line))
 113             {
 114               pattern = realloc (pattern, p - line);
 115               if (pattern == NULL)
 116                 {
 117                   printf ("%zd: Cannot record pattern: %m\n", linenum);
 118                   ret = 1;
 119                   break;
 120                 }
 121               pattern_alloced = p - line;
 122             }
 123
 124           memcpy (pattern, line + 1, p - line - 1);
 125           pattern[p - line - 1] = '\0';
 126           pattern_valid = 1;
 127           continue;
 128         }
 129
 130       if (strncmp (line, "    ", 4) == 0)
 131         {
 132           regex_t re;
 133           int n;
 134
 135           if (!pattern_valid)
 136             {
 137               printf ("%zd: No previous valid pattern %s\n", linenum, line);
 138               continue;
 139             }
 140
 141           if (string_alloced < (size_t) (len - 3))
 142             {
 143               string = realloc (string, len - 3);
 144               if (string == NULL)
 145                 {
 146                   printf ("%zd: Cannot record search string: %m\n", linenum);
 147                   ret = 1;
 148                   break;
 149                 }
 150               string_alloced = len - 3;
 151             }
 152
 153           memcpy (string, line + 4, len - 3);
 154
 155           n = regcomp (&re, pattern,
 156                        REG_EXTENDED | (ignorecase ? REG_ICASE : 0));
 157           if (n != 0)
 158             {
 159               char buf[500];
 160               regerror (n, &re, buf, sizeof (buf));
 161               printf ("%zd: regcomp failed for %s: %s\n",
 162                       linenum, pattern, buf);
 163               ret = 1;
 164               continue;
 165             }
 166
 167           if (regexec (&re, string, 20, rm, 0))
 168             {
 169               rm[0].rm_so = -1;
 170               rm[0].rm_eo = -1;
 171             }
 172
 173           regfree (&re);
 174           rm_valid = 1;
 175           continue;
 176         }
 177
 178       if (!rm_valid)
 179         {
 180           printf ("%zd: No preceeding pattern or search string\n", linenum);
 181           ret = 1;
 182           continue;
 183         }
 184
 185       if (strcmp (line, "No match") == 0)
 186         {
 187           if (rm[0].rm_so != -1 || rm[0].rm_eo != -1)
 188             {
 189               printf ("%zd: /%s/ on %s unexpectedly matched %d..%d\n",
 190                       linenum, pattern, string, rm[0].rm_so, rm[0].rm_eo);
 191               ret = 1;
 192             }
 193
 194           continue;
 195         }
 196
 197       p = line;
 198       if (*p == ' ')
 199         ++p;
 200
 201       num = strtoul (p, &p, 10);
 202       if (num >= 20 || *p != ':' || p[1] != ' ')
 203         {
 204           printf ("%zd: Invalid line %s\n", linenum, line);
 205           ret = 1;
 206           continue;
 207         }
 208
 209       if (rm[num].rm_so == -1 || rm[num].rm_eo == -1)
 210         {
 211           if (strcmp (p + 2, "<unset>") != 0)
 212             {
 213               printf ("%zd: /%s/ on %s unexpectedly failed to match register %ld %d..%d\n",
 214                       linenum, pattern, string, num,
 215                       rm[num].rm_so, rm[num].rm_eo);
 216               ret = 1;
 217             }
 218           continue;
 219         }
 220
 221       if (rm[num].rm_eo < rm[num].rm_so
 222           || rm[num].rm_eo - rm[num].rm_so != len - (p + 2 - line)
 223           || strncmp (p + 2, string + rm[num].rm_so,
 224                       rm[num].rm_eo - rm[num].rm_so) != 0)
 225         {
 226           printf ("%zd: /%s/ on %s unexpectedly failed to match %s for register %ld %d..%d\n",
 227                   linenum, pattern, string, p + 2, num,
 228                   rm[num].rm_so, rm[num].rm_eo);
 229           ret = 1;
 230           continue;
 231         }
 232     }
 233
 234   free (pattern);
 235   free (string);
 236   free (line);
 237   fclose (f);
 238   return ret;
 239 }