lib/uniwbrk/u8-wordbreaks.c

   1 /* Word breaks in UTF-8 strings.
   2    Copyright (C) 2009-2021 Free Software Foundation, Inc.
   3    Written by Bruno Haible <bruno@clisp.org>, 2009.
   4
   5    This program is free software: you can redistribute it and/or modify it
   6    under the terms of the GNU Lesser General Public License as published
   7    by the Free Software Foundation; either version 3 of the License, or
   8    (at your option) any later version.
   9
  10    This program 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 License
  16    along with this program.  If not, see <https://www.gnu.org/licenses/>.  */
  17
  18 #include <config.h>
  19
  20 /* Specification.  */
  21 #include "uniwbrk.h"
  22
  23 #include <string.h>
  24
  25 #include "unistr.h"
  26 #include "uniwbrk/wbrktable.h"
  27
  28 #define FUNC u8_wordbreaks
  29 #define UNIT uint8_t
  30 #define U_MBTOUC_UNSAFE u8_mbtouc_unsafe
  31 #include "u-wordbreaks.h"
  32
  33
  34 #ifdef TEST
  35
  36 #include <stdio.h>
  37 #include <stdlib.h>
  38
  39 /* Read the contents of an input stream, and return it, terminated with a NUL
  40    byte. */
  41 char *
  42 read_file (FILE *stream)
  43 {
  44 #define BUFSIZE 4096
  45   char *buf = NULL;
  46   int alloc = 0;
  47   int size = 0;
  48   int count;
  49
  50   while (! feof (stream))
  51     {
  52       if (size + BUFSIZE > alloc)
  53         {
  54           alloc = alloc + alloc / 2;
  55           if (alloc < size + BUFSIZE)
  56             alloc = size + BUFSIZE;
  57           buf = realloc (buf, alloc);
  58           if (buf == NULL)
  59             {
  60               fprintf (stderr, "out of memory\n");
  61               exit (1);
  62             }
  63         }
  64       count = fread (buf + size, 1, BUFSIZE, stream);
  65       if (count == 0)
  66         {
  67           if (ferror (stream))
  68             {
  69               perror ("fread");
  70               exit (1);
  71             }
  72         }
  73       else
  74         size += count;
  75     }
  76   buf = realloc (buf, size + 1);
  77   if (buf == NULL)
  78     {
  79       fprintf (stderr, "out of memory\n");
  80       exit (1);
  81     }
  82   buf[size] = '\0';
  83   return buf;
  84 #undef BUFSIZE
  85 }
  86
  87 int
  88 main (int argc, char * argv[])
  89 {
  90   if (argc == 1)
  91     {
  92       /* Display all the word breaks in the input string.  */
  93       char *input = read_file (stdin);
  94       int length = strlen (input);
  95       char *breaks = malloc (length);
  96       int i;
  97
  98       u8_wordbreaks ((uint8_t *) input, length, breaks);
  99
 100       for (i = 0; i < length; i++)
 101         {
 102           switch (breaks[i])
 103             {
 104             case 1:
 105               /* U+2027 in UTF-8 encoding */
 106               putc (0xe2, stdout); putc (0x80, stdout); putc (0xa7, stdout);
 107               break;
 108             case 0:
 109               break;
 110             default:
 111               abort ();
 112             }
 113           putc (input[i], stdout);
 114         }
 115
 116       free (breaks);
 117
 118       return 0;
 119     }
 120   else
 121     return 1;
 122 }
 123
 124 #endif /* TEST */