tests/test-mcel.c

   1 /* Test <mcel.h>
   2    Copyright 2023-2025 Free Software Foundation, Inc.
   3
   4    This program is free software: you can redistribute it and/or modify
   5    it under the terms of the GNU General Public License as published by
   6    the Free Software Foundation, either version 3 of the License, or
   7    (at your option) any later version.
   8
   9    This program is distributed in the hope that it will be useful,
  10    but WITHOUT ANY WARRANTY; without even the implied warranty of
  11    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  12    GNU General Public License for more details.
  13
  14    You should have received a copy of the GNU General Public License
  15    along with this program.  If not, see <https://www.gnu.org/licenses/>.  */
  16
  17 #include <config.h>
  18
  19 #include <mcel.h>
  20
  21 #include <locale.h>
  22
  23 #include "macros.h"
  24
  25 static wint_t
  26 to_ascii (wint_t c)
  27 {
  28   return c & 0x7f;
  29 }
  30
  31 static int
  32 sgn (int i)
  33 {
  34   return (i > 0) - (i < 0);
  35 }
  36
  37 static void
  38 test_mcel_vs_mbrtoc32 (unsigned char uc, mcel_t c, size_t n, char32_t ch)
  39 {
  40   ASSERT (!c.err == (n <= MB_LEN_MAX));
  41   ASSERT (c.err
  42           ? c.err == uc && c.ch == 0 && c.len == 1
  43           : c.ch == ch && c.len == (n ? n : 1));
  44 }
  45
  46 int
  47 main (int argc, char *argv[])
  48 {
  49   /* configure should already have checked that the locale is supported.  */
  50   if (setlocale (LC_ALL, "") == NULL)
  51     return 1;
  52
  53   mcel_t prev;
  54   for (int ch = 0; ch < 0x80; ch++)
  55     {
  56       mcel_t c = mcel_ch (ch, 1);
  57       ASSERT (c.ch == ch);
  58       ASSERT (c.len == 1);
  59       ASSERT (!c.err);
  60       ASSERT (mcel_cmp (c, c) == 0);
  61       ASSERT (mcel_tocmp (to_ascii, c, c) == 0);
  62       if (ch)
  63         {
  64           ASSERT (mcel_cmp (prev, c) < 0);
  65           ASSERT (mcel_cmp (c, prev) > 0);
  66           ASSERT (mcel_tocmp (to_ascii, prev, c) < 0);
  67           ASSERT (mcel_tocmp (to_ascii, c, prev) > 0);
  68         }
  69       ASSERT (mcel_isbasic (ch));
  70       prev = c;
  71     }
  72   for (char ch = CHAR_MIN; ; ch++)
  73     {
  74       ASSERT (mcel_isbasic (ch) == (0 <= ch && ch <= 0x7f));
  75       if (ch == CHAR_MAX)
  76         break;
  77     }
  78
  79   if (argc > 1 && argv[1][0] == '5')
  80     {
  81       /* Locale encoding is GB18030.  */
  82       #if (defined __GLIBC__ && __GLIBC__ == 2 && __GLIBC_MINOR__ >= 13 && __GLIBC_MINOR__ <= 15) || (GL_CHAR32_T_IS_UNICODE && (defined __FreeBSD__ || defined __NetBSD__ || defined __sun))
  83       if (test_exit_status != EXIT_SUCCESS)
  84         return test_exit_status;
  85       fputs ("Skipping test: The GB18030 converter in this system's iconv is broken.\n", stderr);
  86       return 77;
  87       #endif
  88     }
  89
  90   for (int ch = 0x80; ch < 0x200; ch++)
  91     {
  92       mcel_t c = mcel_ch (ch, 2);
  93       ASSERT (c.ch == ch);
  94       ASSERT (c.len == 2);
  95       ASSERT (!c.err);
  96       ASSERT (mcel_cmp (c, c) == 0);
  97       ASSERT (mcel_tocmp (to_ascii, c, c) == 0);
  98       ASSERT (mcel_cmp (prev, c) < 0);
  99       ASSERT (mcel_cmp (c, prev) > 0);
 100       ASSERT (mcel_tocmp (to_ascii, c, c) == 0);
 101       int cmp = to_ascii (c.ch) ? -1 : 1;
 102       ASSERT (sgn (mcel_tocmp (to_ascii, prev, c)) == cmp);
 103       ASSERT (sgn (mcel_tocmp (to_ascii, c, prev)) == -cmp);
 104       prev = c;
 105     }
 106   for (unsigned char err = 0x80; ; err++)
 107     {
 108       mcel_t c = mcel_err (err);
 109       ASSERT (!c.ch);
 110       ASSERT (c.len == 1);
 111       ASSERT (c.err == err);
 112       ASSERT (mcel_cmp (c, c) == 0);
 113       ASSERT (mcel_cmp (prev, c) < 0);
 114       ASSERT (mcel_cmp (c, prev) > 0);
 115       ASSERT (mcel_tocmp (to_ascii, c, c) == 0);
 116       ASSERT (mcel_tocmp (to_ascii, prev, c) < 0);
 117       ASSERT (mcel_tocmp (to_ascii, c, prev) > 0);
 118       prev = c;
 119       if (err == (unsigned char) -1)
 120         break;
 121     }
 122
 123   for (int i = CHAR_MIN; i <= CHAR_MAX; i++)
 124     for (int j = CHAR_MIN; j <= CHAR_MAX; j++)
 125       for (int k = CHAR_MIN; k <= CHAR_MAX; k++)
 126         {
 127           char const ijk[] = {i, j, k};
 128           mbstate_t mbs = {0};
 129           char32_t ch;
 130           size_t n = mbrtoc32 (&ch, ijk, sizeof ijk, &mbs);
 131           mcel_t c = mcel_scan (ijk, ijk + sizeof ijk);
 132           test_mcel_vs_mbrtoc32 (i, c, n, ch);
 133
 134           static char const terminator[] = "\r\n./";
 135           for (int ti = 0; ti < sizeof terminator; ti++)
 136             {
 137               char t = terminator[ti];
 138               if (i == t || j == t || k == t)
 139                 continue;
 140               char const ijkt[] = {i, j, k, t};
 141               mcel_t d = mcel_scant (ijkt, t);
 142               ASSERT (c.ch == d.ch && c.err == d.err && c.len == d.len);
 143               if (!t)
 144                 {
 145                   mcel_t z = mcel_scanz (ijkt);
 146                   ASSERT (d.ch == z.ch && d.err == z.err && d.len == z.len);
 147                 }
 148             }
 149         }
 150
 151   return test_exit_status;
 152 }