posix/bug-regex33.c

   1 /* Test re_search with multi-byte characters in EUC-JP.
   2    Copyright (C) 2012-2022 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 #define _GNU_SOURCE 1
  20 #include <locale.h>
  21 #include <regex.h>
  22 #include <stdio.h>
  23 #include <stdlib.h>
  24 #include <string.h>
  25
  26 static int
  27 do_test (void)
  28 {
  29   struct re_pattern_buffer r;
  30   struct re_registers s;
  31   int e, rc = 0;
  32   if (setlocale (LC_CTYPE, "ja_JP.EUC-JP") == NULL)
  33     {
  34       puts ("setlocale failed");
  35       return 1;
  36     }
  37   memset (&r, 0, sizeof (r));
  38   memset (&s, 0, sizeof (s));
  39
  40   /* The bug cannot be reproduced without initialized fastmap (it is SBC_MAX
  41      value from regex_internal.h).  */
  42   r.fastmap = malloc (UCHAR_MAX + 1);
  43
  44                      /* 圭 */
  45   re_compile_pattern ("\xb7\xbd", 2, &r);
  46
  47                 /* aaaaa件a新処, \xb7\xbd constitutes a false match */
  48   e = re_search (&r, "\x61\x61\x61\x61\x61\xb7\xef\x61\xbf\xb7\xbd\xe8",
  49                  12, 0, 12, &s);
  50   if (e != -1)
  51     {
  52       printf ("bug-regex33.1: false match or error: re_search() returned %d, should return -1\n", e);
  53       rc = 1;
  54     }
  55
  56                 /* aaaa件a新処, \xb7\xbd constitutes a false match,
  57                  * this is a reproducer of BZ #13637 */
  58   e = re_search (&r, "\x61\x61\x61\x61\xb7\xef\x61\xbf\xb7\xbd\xe8",
  59                  11, 0, 11, &s);
  60   if (e != -1)
  61     {
  62       printf ("bug-regex33.2: false match or error: re_search() returned %d, should return -1\n", e);
  63       rc = 1;
  64     }
  65
  66                 /* aaa件a新処, \xb7\xbd constitutes a false match,
  67                  * this is a reproducer of BZ #13637 */
  68   e = re_search (&r, "\x61\x61\x61\xb7\xef\x61\xbf\xb7\xbd\xe8",
  69                  10, 0, 10, &s);
  70   if (e != -1)
  71     {
  72       printf ("bug-regex33.3: false match or error: re_search() returned %d, should return -1\n", e);
  73       rc = 1;
  74     }
  75
  76                 /* aa件a新処, \xb7\xbd constitutes a false match */
  77   e = re_search (&r, "\x61\x61\xb7\xef\x61\xbf\xb7\xbd\xe8",
  78                  9, 0, 9, &s);
  79   if (e != -1)
  80     {
  81       printf ("bug-regex33.4: false match or error: re_search() returned %d, should return -1\n", e);
  82       rc = 1;
  83     }
  84
  85                 /* a件a新処, \xb7\xbd constitutes a false match */
  86   e = re_search (&r, "\x61\xb7\xef\x61\xbf\xb7\xbd\xe8",
  87                  8, 0, 8, &s);
  88   if (e != -1)
  89     {
  90       printf ("bug-regex33.5: false match or error: re_search() returned %d, should return -1\n", e);
  91       rc = 1;
  92     }
  93
  94                 /* 新処圭新処, \xb7\xbd here really matches 圭, but second occurrence is a false match,
  95                  * this is a reproducer of bug-regex25 and BZ #13637 */
  96   e = re_search (&r, "\xbf\xb7\xbd\xe8\xb7\xbd\xbf\xb7\xbd\xe8",
  97                  10, 0, 10, &s);
  98   if (e != 4)
  99     {
 100       printf ("bug-regex33.6: no match or false match: re_search() returned %d, should return 4\n", e);
 101       rc = 1;
 102     }
 103
 104                 /* 新処圭新, \xb7\xbd here really matches 圭,
 105                  * this is a reproducer of bug-regex25 */
 106   e = re_search (&r, "\xbf\xb7\xbd\xe8\xb7\xbd\xbf\xb7",
 107                  9, 0, 9, &s);
 108   if (e != 4)
 109     {
 110       printf ("bug-regex33.7: no match or false match: re_search() returned %d, should return 4\n", e);
 111       rc = 1;
 112     }
 113
 114   return rc;
 115 }
 116
 117 #define TEST_FUNCTION do_test ()
 118 #include "../test-skeleton.c"