wcsmbs/tst-mbrtowc.c

   1 /* Copyright (C) 2000, 2001, 2002 Free Software Foundation, Inc.
   2    This file is part of the GNU C Library.
   3    Contributed by Ulrich Drepper <drepper@redhat.com>, 2000.
   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, write to the Free
  17    Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
  18    02111-1307 USA.  */
  19
  20 #include <assert.h>
  21 #include <locale.h>
  22 #include <stdio.h>
  23 #include <stdlib.h>
  24 #include <string.h>
  25 #include <wchar.h>
  26
  27
  28 static int check_ascii (const char *locname);
  29
  30 /* UTF-8 single byte feeding test for mbrtowc(),
  31    contributed by Markus Kuhn <mkuhn@acm.org>.  */
  32 static int
  33 utf8_test_1 (void)
  34 {
  35   wchar_t wc;
  36   mbstate_t s;
  37
  38   wc = 42;                      /* arbitrary number */
  39   memset (&s, 0, sizeof (s));   /* get s into initial state */
  40   assert (mbrtowc (&wc, "\xE2", 1, &s) == (size_t) -2); /* 1st byte processed */
  41   assert (mbrtowc (&wc, "\x89", 1, &s) == (size_t) -2); /* 2nd byte processed */
  42   assert (wc == 42);            /* no value has not been stored into &wc yet */
  43   assert (mbrtowc (&wc, "\xA0", 1, &s) == 1);   /* 3nd byte processed */
  44   assert (wc == 0x2260);        /* E2 89 A0 = U+2260 (not equal) decoded correctly */
  45   assert (mbrtowc (&wc, "", 1, &s) == 0);       /* test final byte processing */
  46   assert (wc == 0);             /* test final byte decoding */
  47
  48   return 0;
  49 }
  50
  51 /* Test for NUL byte processing via empty string.  */
  52 static int
  53 utf8_test_2 (void)
  54 {
  55   wchar_t wc;
  56   mbstate_t s;
  57
  58   wc = 42;                      /* arbitrary number */
  59   memset (&s, 0, sizeof (s));   /* get s into initial state */
  60   assert (mbrtowc (NULL, "", 1, &s) == 0); /* valid terminator */
  61   assert (mbsinit (&s));
  62
  63   wc = 42;                      /* arbitrary number */
  64   memset (&s, 0, sizeof (s));   /* get s into initial state */
  65   assert (mbrtowc (&wc, "\xE2", 1, &s) == (size_t) -2); /* 1st byte processed */
  66   assert (mbrtowc (NULL, "", 1, &s) == (size_t) -1); /* invalid terminator */
  67
  68   wc = 42;                      /* arbitrary number */
  69   memset (&s, 0, sizeof (s));   /* get s into initial state */
  70   assert (mbrtowc (&wc, "\xE2", 1, &s) == (size_t) -2); /* 1st byte processed */
  71   assert (mbrtowc (&wc, "\x89", 1, &s) == (size_t) -2); /* 2nd byte processed */
  72   assert (mbrtowc (NULL, "", 1, &s) == (size_t) -1); /* invalid terminator */
  73
  74   wc = 42;                      /* arbitrary number */
  75   memset (&s, 0, sizeof (s));   /* get s into initial state */
  76   assert (mbrtowc (&wc, "\xE2", 1, &s) == (size_t) -2); /* 1st byte processed */
  77   assert (mbrtowc (&wc, "\x89", 1, &s) == (size_t) -2); /* 2nd byte processed */
  78   assert (mbrtowc (&wc, "\xA0", 1, &s) == 1);   /* 3nd byte processed */
  79   assert (mbrtowc (NULL, "", 1, &s) == 0); /* valid terminator */
  80   assert (mbsinit (&s));
  81
  82   return 0;
  83 }
  84
  85 /* Test for NUL byte processing via NULL string.  */
  86 static int
  87 utf8_test_3 (void)
  88 {
  89   wchar_t wc;
  90   mbstate_t s;
  91
  92   wc = 42;                      /* arbitrary number */
  93   memset (&s, 0, sizeof (s));   /* get s into initial state */
  94   assert (mbrtowc (NULL, NULL, 0, &s) == 0); /* valid terminator */
  95   assert (mbsinit (&s));
  96
  97   wc = 42;                      /* arbitrary number */
  98   memset (&s, 0, sizeof (s));   /* get s into initial state */
  99   assert (mbrtowc (&wc, "\xE2", 1, &s) == (size_t) -2); /* 1st byte processed */
 100   assert (mbrtowc (NULL, NULL, 0, &s) == (size_t) -1); /* invalid terminator */
 101
 102   wc = 42;                      /* arbitrary number */
 103   memset (&s, 0, sizeof (s));   /* get s into initial state */
 104   assert (mbrtowc (&wc, "\xE2", 1, &s) == (size_t) -2); /* 1st byte processed */
 105   assert (mbrtowc (&wc, "\x89", 1, &s) == (size_t) -2); /* 2nd byte processed */
 106   assert (mbrtowc (NULL, NULL, 0, &s) == (size_t) -1); /* invalid terminator */
 107
 108   wc = 42;                      /* arbitrary number */
 109   memset (&s, 0, sizeof (s));   /* get s into initial state */
 110   assert (mbrtowc (&wc, "\xE2", 1, &s) == (size_t) -2); /* 1st byte processed */
 111   assert (mbrtowc (&wc, "\x89", 1, &s) == (size_t) -2); /* 2nd byte processed */
 112   assert (mbrtowc (&wc, "\xA0", 1, &s) == 1);   /* 3nd byte processed */
 113   assert (mbrtowc (NULL, NULL, 0, &s) == 0); /* valid terminator */
 114   assert (mbsinit (&s));
 115
 116   return 0;
 117 }
 118
 119 static int
 120 utf8_test (void)
 121 {
 122   const char *locale = "de_DE.UTF-8";
 123   int error = 0;
 124
 125   if (!setlocale (LC_CTYPE, locale))
 126     {
 127       fprintf (stderr, "locale '%s' not available!\n", locale);
 128       exit (1);
 129     }
 130
 131   error |= utf8_test_1 ();
 132   error |= utf8_test_2 ();
 133   error |= utf8_test_3 ();
 134
 135   return error;
 136 }
 137
 138
 139 int
 140 main (void)
 141 {
 142   int result = 0;
 143
 144   /* Check mapping of ASCII range for some character sets which have
 145      ASCII as a subset.  For those the wide char generated must have
 146      the same value.  */
 147   setlocale (LC_ALL, "C");
 148   result |= check_ascii (setlocale (LC_ALL, NULL));
 149
 150   setlocale (LC_ALL, "de_DE.UTF-8");
 151   result |= check_ascii (setlocale (LC_ALL, NULL));
 152   result |= utf8_test ();
 153
 154   setlocale (LC_ALL, "ja_JP.EUC-JP");
 155   result |= check_ascii (setlocale (LC_ALL, NULL));
 156
 157   return result;
 158 }
 159
 160
 161 static int
 162 check_ascii (const char *locname)
 163 {
 164   int c;
 165   int res = 0;
 166
 167   printf ("Testing locale \"%s\":\n", locname);
 168
 169   for (c = 0; c <= 127; ++c)
 170     {
 171       char buf[MB_CUR_MAX];
 172       wchar_t wc = 0xffffffff;
 173       mbstate_t s;
 174       size_t n, i;
 175
 176       for (i = 0; i < MB_CUR_MAX; ++i)
 177         buf[i] = c + i;
 178
 179       memset (&s, '\0', sizeof (s));
 180
 181       n = mbrtowc (&wc, buf, MB_CUR_MAX, &s);
 182       if (n == (size_t) -1)
 183         {
 184           printf ("%s: '\\x%x': encoding error\n", locname, c);
 185           ++res;
 186         }
 187       else if (n == (size_t) -2)
 188         {
 189           printf ("%s: '\\x%x': incomplete character\n", locname, c);
 190           ++res;
 191         }
 192       else if (n == 0 && c != 0)
 193         {
 194           printf ("%s: '\\x%x': 0 returned\n", locname, c);
 195           ++res;
 196         }
 197       else if (n != 0 && c == 0)
 198         {
 199           printf ("%s: '\\x%x': not 0 returned\n", locname, c);
 200           ++res;
 201         }
 202       else if (c != 0 && n != 1)
 203         {
 204           printf ("%s: '\\x%x': not 1 returned\n", locname, c);
 205           ++res;
 206         }
 207       else if (wc != (wchar_t) c)
 208         {
 209           printf ("%s: '\\x%x': wc != L'\\x%x'\n", locname, c, c);
 210           ++res;
 211         }
 212     }
 213
 214   printf (res == 1 ? "%d error\n" : "%d errors\n", res);
 215
 216   return res != 0;
 217 }