posix/bug-regex17.c

   1 /* German regular expression tests.
   2    Copyright (C) 2002-2019 Free Software Foundation, Inc.
   3    This file is part of the GNU C Library.
   4    Contributed by Jakub Jelinek <jakub@redhat.com>, 2002.
   5
   6    The GNU C Library is free software; you can redistribute it and/or
   7    modify it under the terms of the GNU Lesser General Public
   8    License as published by the Free Software Foundation; either
   9    version 2.1 of the License, or (at your option) any later version.
  10
  11    The GNU C Library is distributed in the hope that it will be useful,
  12    but WITHOUT ANY WARRANTY; without even the implied warranty of
  13    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  14    Lesser General Public License for more details.
  15
  16    You should have received a copy of the GNU Lesser General Public
  17    License along with the GNU C Library; if not, see
  18    <https://www.gnu.org/licenses/>.  */
  19
  20 #include <sys/types.h>
  21 #include <mcheck.h>
  22 #include <regex.h>
  23 #include <stdio.h>
  24 #include <stdlib.h>
  25 #include <locale.h>
  26
  27 /* Tests supposed to match.  */
  28 struct
  29 {
  30   const char *pattern;
  31   const char *string;
  32   int flags, nmatch;
  33   regmatch_t rm[5];
  34 } tests[] = {
  35   /* U+00C4     \xc3\x84        LATIN CAPITAL LETTER A WITH DIAERESIS
  36      U+00D6     \xc3\x96        LATIN CAPITAL LETTER O WITH DIAERESIS
  37      U+00E4     \xc3\xa4        LATIN SMALL LETTER A WITH DIAERESIS
  38      U+00F6     \xc3\xb6        LATIN SMALL LETTER O WITH DIAERESIS  */
  39   { "\xc3\x84\xc3\x96*\xc3\xb6$", "aB\xc3\xa4\xc3\xb6\xc3\xb6\xc3\x96", REG_ICASE, 2,
  40     { { 2, 10 }, { -1, -1 } } },
  41   { "[\xc3\x84x]\xc3\x96*\xc3\xb6$", "aB\xc3\x84\xc3\xb6\xc3\xb6\xc3\x96", REG_ICASE, 2,
  42     { { 2, 10 }, { -1, -1 } } },
  43   { "[\xc3\x84x]\xc3\x96*\xc3\xb6$", "aB\xc3\xa4\xc3\xb6\xc3\xb6\xc3\x96", REG_ICASE, 2,
  44     { { 2, 10 }, { -1, -1 } } },
  45   { "[^x]\xc3\x96*\xc3\xb6$", "aB\xc3\xa4\xc3\xb6\xc3\xb6\xc3\x96", REG_ICASE, 2,
  46     { { 2, 10 }, { -1, -1 } } },
  47
  48   /* Tests for bug 9697:
  49      U+00DF     \xc3\x9f        LATIN SMALL LETTER SHARP S
  50      U+02DA     \xcb\x9a        RING ABOVE
  51      U+02E2     \xcb\xa2        MODIFIER LETTER SMALL S  */
  52   { "[a-z]|[^a-z]", "\xcb\xa2", REG_EXTENDED, 2,
  53     { { 0, 2 }, { -1, -1 } } },
  54   { "[a-z]", "\xc3\x9f", REG_EXTENDED, 2,
  55     { { 0, 2 }, { -1, -1 } } },
  56   { "[^a-z]", "\xcb\x9a", REG_EXTENDED, 2,
  57     { { 0, 2 }, { -1, -1 } } },
  58 };
  59
  60
  61 static int
  62 do_test (void)
  63 {
  64   regex_t re;
  65   regmatch_t rm[5];
  66   size_t i;
  67   int n, ret = 0;
  68
  69   setlocale (LC_ALL, "de_DE.UTF-8");
  70   for (i = 0; i < sizeof (tests) / sizeof (tests[0]); ++i)
  71     {
  72       n = regcomp (&re, tests[i].pattern, tests[i].flags);
  73       if (n != 0)
  74         {
  75           char buf[500];
  76           regerror (n, &re, buf, sizeof (buf));
  77           printf ("regcomp %zd failed: %s\n", i, buf);
  78           ret = 1;
  79           continue;
  80         }
  81
  82       if (regexec (&re, tests[i].string, tests[i].nmatch, rm, 0))
  83         {
  84           printf ("regexec %zd failed\n", i);
  85           ret = 1;
  86           regfree (&re);
  87           continue;
  88         }
  89
  90       for (n = 0; n < tests[i].nmatch; ++n)
  91         if (rm[n].rm_so != tests[i].rm[n].rm_so
  92               || rm[n].rm_eo != tests[i].rm[n].rm_eo)
  93           {
  94             if (tests[i].rm[n].rm_so == -1 && tests[i].rm[n].rm_eo == -1)
  95               break;
  96             printf ("regexec match failure rm[%d] %d..%d\n",
  97                     n, rm[n].rm_so, rm[n].rm_eo);
  98             ret = 1;
  99             break;
 100           }
 101
 102       regfree (&re);
 103     }
 104
 105   return ret;
 106 }
 107
 108 #define TEST_FUNCTION do_test ()
 109 #include "../test-skeleton.c"