wcsmbs/tst-mbrtowc.c

   1 /* Copyright (C) 2000-2013 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, see
  17    <http://www.gnu.org/licenses/>.  */
  18
  19 /* We always want assert to be fully defined.  */
  20 #undef NDEBUG
  21 #include <assert.h>
  22 #include <locale.h>
  23 #include <stdio.h>
  24 #include <stdlib.h>
  25 #include <string.h>
  26 #include <wchar.h>
  27
  28
  29 static int check_ascii (const char *locname);
  30
  31 /* UTF-8 single byte feeding test for mbrtowc(),
  32    contributed by Markus Kuhn <mkuhn@acm.org>.  */
  33 static int
  34 utf8_test_1 (void)
  35 {
  36   wchar_t wc;
  37   mbstate_t s;
  38
  39   wc = 42;                      /* arbitrary number */
  40   memset (&s, 0, sizeof (s));   /* get s into initial state */
  41   assert (mbrtowc (&wc, "\xE2", 1, &s) == (size_t) -2); /* 1st byte processed */
  42   assert (mbrtowc (&wc, "\x89", 1, &s) == (size_t) -2); /* 2nd byte processed */
  43   assert (wc == 42);            /* no value has not been stored into &wc yet */
  44   assert (mbrtowc (&wc, "\xA0", 1, &s) == 1);   /* 3nd byte processed */
  45   assert (wc == 0x2260);        /* E2 89 A0 = U+2260 (not equal) decoded correctly */
  46   assert (mbrtowc (&wc, "", 1, &s) == 0);       /* test final byte processing */
  47   assert (wc == 0);             /* test final byte decoding */
  48
  49   /* The following test is by Al Viro <aviro@redhat.com>.  */
  50   const char str[] = "\xe0\xa0\x80";
  51
  52   wc = 42;                      /* arbitrary number */
  53   memset (&s, 0, sizeof (s));   /* get s into initial state */
  54   assert (mbrtowc (&wc, str, 1, &s) == -2);
  55   assert (mbrtowc (&wc, str + 1, 2, &s) == 2);
  56   assert (wc == 0x800);
  57
  58   wc = 42;                      /* arbitrary number */
  59   memset (&s, 0, sizeof (s));   /* get s into initial state */
  60   assert (mbrtowc (&wc, str, 3, &s) == 3);
  61   assert (wc == 0x800);
  62
  63   return 0;
  64 }
  65
  66 /* Test for NUL byte processing via empty string.  */
  67 static int
  68 utf8_test_2 (void)
  69 {
  70   wchar_t wc;
  71   mbstate_t s;
  72
  73   wc = 42;                      /* arbitrary number */
  74   memset (&s, 0, sizeof (s));   /* get s into initial state */
  75   assert (mbrtowc (NULL, "", 1, &s) == 0); /* valid terminator */
  76   assert (mbsinit (&s));
  77
  78   wc = 42;                      /* arbitrary number */
  79   memset (&s, 0, sizeof (s));   /* get s into initial state */
  80   assert (mbrtowc (&wc, "\xE2", 1, &s) == (size_t) -2); /* 1st byte processed */
  81   assert (mbrtowc (NULL, "", 1, &s) == (size_t) -1); /* invalid terminator */
  82
  83   wc = 42;                      /* arbitrary number */
  84   memset (&s, 0, sizeof (s));   /* get s into initial state */
  85   assert (mbrtowc (&wc, "\xE2", 1, &s) == (size_t) -2); /* 1st byte processed */
  86   assert (mbrtowc (&wc, "\x89", 1, &s) == (size_t) -2); /* 2nd byte processed */
  87   assert (mbrtowc (NULL, "", 1, &s) == (size_t) -1); /* invalid terminator */
  88
  89   wc = 42;                      /* arbitrary number */
  90   memset (&s, 0, sizeof (s));   /* get s into initial state */
  91   assert (mbrtowc (&wc, "\xE2", 1, &s) == (size_t) -2); /* 1st byte processed */
  92   assert (mbrtowc (&wc, "\x89", 1, &s) == (size_t) -2); /* 2nd byte processed */
  93   assert (mbrtowc (&wc, "\xA0", 1, &s) == 1);   /* 3nd byte processed */
  94   assert (mbrtowc (NULL, "", 1, &s) == 0); /* valid terminator */
  95   assert (mbsinit (&s));
  96
  97   return 0;
  98 }
  99
 100 /* Test for NUL byte processing via NULL string.  */
 101 static int
 102 utf8_test_3 (void)
 103 {
 104   wchar_t wc;
 105   mbstate_t s;
 106
 107   wc = 42;                      /* arbitrary number */
 108   memset (&s, 0, sizeof (s));   /* get s into initial state */
 109   assert (mbrtowc (NULL, NULL, 0, &s) == 0); /* valid terminator */
 110   assert (mbsinit (&s));
 111
 112   wc = 42;                      /* arbitrary number */
 113   memset (&s, 0, sizeof (s));   /* get s into initial state */
 114   assert (mbrtowc (&wc, "\xE2", 1, &s) == (size_t) -2); /* 1st byte processed */
 115   assert (mbrtowc (NULL, NULL, 0, &s) == (size_t) -1); /* invalid terminator */
 116
 117   wc = 42;                      /* arbitrary number */
 118   memset (&s, 0, sizeof (s));   /* get s into initial state */
 119   assert (mbrtowc (&wc, "\xE2", 1, &s) == (size_t) -2); /* 1st byte processed */
 120   assert (mbrtowc (&wc, "\x89", 1, &s) == (size_t) -2); /* 2nd byte processed */
 121   assert (mbrtowc (NULL, NULL, 0, &s) == (size_t) -1); /* invalid terminator */
 122
 123   wc = 42;                      /* arbitrary number */
 124   memset (&s, 0, sizeof (s));   /* get s into initial state */
 125   assert (mbrtowc (&wc, "\xE2", 1, &s) == (size_t) -2); /* 1st byte processed */
 126   assert (mbrtowc (&wc, "\x89", 1, &s) == (size_t) -2); /* 2nd byte processed */
 127   assert (mbrtowc (&wc, "\xA0", 1, &s) == 1);   /* 3nd byte processed */
 128   assert (mbrtowc (NULL, NULL, 0, &s) == 0); /* valid terminator */
 129   assert (mbsinit (&s));
 130
 131   return 0;
 132 }
 133
 134 static int
 135 utf8_test (void)
 136 {
 137   const char *locale = "de_DE.UTF-8";
 138   int error = 0;
 139
 140   if (!setlocale (LC_CTYPE, locale))
 141     {
 142       fprintf (stderr, "locale '%s' not available!\n", locale);
 143       exit (1);
 144     }
 145
 146   error |= utf8_test_1 ();
 147   error |= utf8_test_2 ();
 148   error |= utf8_test_3 ();
 149
 150   return error;
 151 }
 152
 153
 154 int
 155 main (void)
 156 {
 157   int result = 0;
 158
 159   /* Check mapping of ASCII range for some character sets which have
 160      ASCII as a subset.  For those the wide char generated must have
 161      the same value.  */
 162   setlocale (LC_ALL, "C");
 163   result |= check_ascii (setlocale (LC_ALL, NULL));
 164
 165   setlocale (LC_ALL, "de_DE.UTF-8");
 166   result |= check_ascii (setlocale (LC_ALL, NULL));
 167   result |= utf8_test ();
 168
 169   setlocale (LC_ALL, "ja_JP.EUC-JP");
 170   result |= check_ascii (setlocale (LC_ALL, NULL));
 171
 172   return result;
 173 }
 174
 175
 176 static int
 177 check_ascii (const char *locname)
 178 {
 179   int c;
 180   int res = 0;
 181
 182   printf ("Testing locale \"%s\":\n", locname);
 183
 184   for (c = 0; c <= 127; ++c)
 185     {
 186       char buf[MB_CUR_MAX];
 187       wchar_t wc = 0xffffffff;
 188       mbstate_t s;
 189       size_t n, i;
 190
 191       for (i = 0; i < MB_CUR_MAX; ++i)
 192         buf[i] = c + i;
 193
 194       memset (&s, '\0', sizeof (s));
 195
 196       n = mbrtowc (&wc, buf, MB_CUR_MAX, &s);
 197       if (n == (size_t) -1)
 198         {
 199           printf ("%s: '\\x%x': encoding error\n", locname, c);
 200           ++res;
 201         }
 202       else if (n == (size_t) -2)
 203         {
 204           printf ("%s: '\\x%x': incomplete character\n", locname, c);
 205           ++res;
 206         }
 207       else if (n == 0 && c != 0)
 208         {
 209           printf ("%s: '\\x%x': 0 returned\n", locname, c);
 210           ++res;
 211         }
 212       else if (n != 0 && c == 0)
 213         {
 214           printf ("%s: '\\x%x': not 0 returned\n", locname, c);
 215           ++res;
 216         }
 217       else if (c != 0 && n != 1)
 218         {
 219           printf ("%s: '\\x%x': not 1 returned\n", locname, c);
 220           ++res;
 221         }
 222       else if (wc != (wchar_t) c)
 223         {
 224           printf ("%s: '\\x%x': wc != L'\\x%x'\n", locname, c, c);
 225           ++res;
 226         }
 227     }
 228
 229   printf (res == 1 ? "%d error\n" : "%d errors\n", res);
 230
 231   return res != 0;
 232 }