tests/uniwbrk/test-u8-wordbreaks.c

   1 /* Test of word breaks in UTF-8 strings.
   2    Copyright (C) 2009-2025 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 /* Written by Bruno Haible <bruno@clisp.org>, 2009.  */
  18
  19 #include <config.h>
  20
  21 #include "uniwbrk.h"
  22
  23 #include <stdlib.h>
  24
  25 #include "macros.h"
  26
  27 int
  28 main ()
  29 {
  30   /* Test case n = 0.  */
  31   u8_wordbreaks (NULL, 0, NULL);
  32
  33   {
  34     static const uint8_t input[91] =
  35       /* "Grüß Gott. Здравствуйте! x=(-b±sqrt(b²-4ac))/(2a)  日本語,中文,한글" */
  36       "Gr\303\274\303\237 Gott. \320\227\320\264\321\200\320\260\320\262\321\201\321\202\320\262\321\203\320\271\321\202\320\265! x=(-b\302\261sqrt(b\302\262-4ac))/(2a)  \346\227\245\346\234\254\350\252\236,\344\270\255\346\226\207,\355\225\234\352\270\200\n";
  37     char *p = (char *) malloc (SIZEOF (input));
  38     size_t i;
  39
  40     u8_wordbreaks (input, SIZEOF (input), p);
  41
  42     for (i = 0; i < 91; i++)
  43       {
  44         ASSERT (p[i] == ((i >= 6 && i <= 7)
  45                          || (i >= 11 && i <= 13)
  46                          || (i >= 37 && i <= 44)
  47                          || i == 46 || (i >= 50 && i <= 52)
  48                          || (i >= 54 && i <= 55)
  49                          || (i >= 58 && i <= 62) || (i >= 64 && i <= 65)
  50                          || i == 67
  51                          || i == 70 || i == 73 || i == 76
  52                          || i == 77 || i == 80 || i == 83
  53                          || i == 84 || i == 90
  54                          ? 1 : 0));
  55       }
  56     free (p);
  57   }
  58
  59   {
  60     /* Same input string, decomposed.  */
  61     static const uint8_t input[106] =
  62       /* "Grüß Gott. Здравствуйте! x=(-b±sqrt(b²-4ac))/(2a)  日本語,中文,한글" */
  63       "Gru\314\210\303\237 Gott. \320\227\320\264\321\200\320\260\320\262\321\201\321\202\320\262\321\203\320\270\314\206\321\202\320\265! x=(-b\302\261sqrt(b\302\262-4ac))/(2a)  \346\227\245\346\234\254\350\252\236,\344\270\255\346\226\207,\341\204\222\341\205\241\341\206\253\341\204\200\341\205\263\341\206\257\n";
  64     char *p = (char *) malloc (SIZEOF (input));
  65     size_t i;
  66
  67     u8_wordbreaks (input, SIZEOF (input), p);
  68
  69     for (i = 0; i < 106; i++)
  70       {
  71         ASSERT (p[i] == ((i >= 7 && i <= 8)
  72                          || (i >= 12 && i <= 14)
  73                          || (i >= 40 && i <= 47)
  74                          || i == 49 || (i >= 53 && i <= 55)
  75                          || (i >= 57 && i <= 58)
  76                          || (i >= 61 && i <= 65) || (i >= 67 && i <= 68)
  77                          || i == 70
  78                          || i == 73 || i == 76 || i == 79
  79                          || i == 80 || i == 83 || i == 86
  80                          || i == 87 || i == 105
  81                          ? 1 : 0));
  82       }
  83     free (p);
  84   }
  85
  86   /* CR LF handling.  */
  87   {
  88     static const uint8_t input[8] = "a\nb\rc\r\nd";
  89     char *p = (char *) malloc (SIZEOF (input));
  90     size_t i;
  91
  92     u8_wordbreaks (input, SIZEOF (input), p);
  93     for (i = 0; i < 8; i++)
  94       {
  95         ASSERT (p[i] == (i == 1 || i == 2 || i == 3 || i == 4 || i == 5
  96                          || i == 7 ? 1 :
  97                          0));
  98       }
  99     free (p);
 100   }
 101
 102   /* Test regional indicators.  */
 103   {
 104     static const uint8_t input[18] =
 105       ".\360\237\207\251\360\237\207\252\360\237\207\253\360\237\207\267.";
 106     char *p = (char *) malloc (SIZEOF (input));
 107     size_t i;
 108
 109     u8_wordbreaks (input, SIZEOF (input), p);
 110     for (i = 0; i < 18; i++)
 111       {
 112         ASSERT (p[i] == (i == 1 || i == 9 || i == 17 ? 1 : 0));
 113       }
 114     free (p);
 115   }
 116
 117   return test_exit_status;
 118 }