tests/unicode.c

   1 /* unicode.c --- Self tests for unicode related functions.
   2  * Copyright (C) 2002, 2005  Simon Josefsson
   3  *
   4  * This file is part of GNU SASL.
   5  *
   6  * GNU SASL is free software; you can redistribute it and/or modify
   7  * it under the terms of the GNU General Public License as published by
   8  * the Free Software Foundation; either version 2 of the License, or
   9  * (at your option) any later version.
  10  *
  11  * GNU SASL is distributed in the hope that it will be useful,
  12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  14  * GNU General Public License for more details.
  15  *
  16  * You should have received a copy of the GNU General Public License
  17  * along with GNU SASL; if not, write to the Free Software Foundation,
  18  * Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  19  *
  20  */
  21
  22 #include <stdio.h>
  23 #include <stdarg.h>
  24 #include <stdlib.h>
  25 #include <string.h>
  26 #include <gsasl.h>
  27 #ifdef HAVE_CONFIG_H
  28 #include "config.h"
  29 #endif
  30
  31 static int debug = 0;
  32 static int error_count = 0;
  33 static int break_on_error = 0;
  34
  35 static void
  36 fail (const char *format, ...)
  37 {
  38   va_list arg_ptr;
  39
  40   va_start (arg_ptr, format);
  41   vfprintf (stderr, format, arg_ptr);
  42   va_end (arg_ptr);
  43   error_count++;
  44   if (break_on_error)
  45     exit (1);
  46 }
  47
  48 static void
  49 escapeprint (char *str, int len)
  50 {
  51   int i;
  52
  53   printf ("\t ;; `");
  54   for (i = 0; i < len; i++)
  55     if ((str[i] >= 'A' && str[i] <= 'Z') ||
  56         (str[i] >= 'a' && str[i] <= 'z') ||
  57         (str[i] >= '0' && str[i] <= '9') || str[i] == '.')
  58       printf ("%c", str[i]);
  59     else
  60       printf ("\\x%02x", str[i]);
  61   printf ("' (length %d bytes)\n", len);
  62 }
  63
  64 static void
  65 hexprint (char *str, int len)
  66 {
  67   int i;
  68
  69   printf ("\t ;; ");
  70   for (i = 0; i < len; i++)
  71     {
  72       printf ("%02x ", str[i]);
  73       if ((i + 1) % 8 == 0)
  74         printf (" ");
  75       if ((i + 1) % 16 == 0 && i + 1 < len)
  76         printf ("\n\t ;; ");
  77     }
  78 }
  79
  80 static void
  81 binprint (char *str, int len)
  82 {
  83   int i;
  84
  85   printf ("\t ;; ");
  86   for (i = 0; i < len; i++)
  87     {
  88       printf ("%d%d%d%d%d%d%d%d ",
  89               str[i] & 0x80 ? 1 : 0,
  90               str[i] & 0x40 ? 1 : 0,
  91               str[i] & 0x20 ? 1 : 0,
  92               str[i] & 0x10 ? 1 : 0,
  93               str[i] & 0x08 ? 1 : 0,
  94               str[i] & 0x04 ? 1 : 0,
  95               str[i] & 0x02 ? 1 : 0, str[i] & 0x01 ? 1 : 0);
  96       if ((i + 1) % 3 == 0)
  97         printf (" ");
  98       if ((i + 1) % 6 == 0 && i + 1 < len)
  99         printf ("\n\t ;; ");
 100     }
 101 }
 102
 103 struct nfkc
 104 {
 105   char *in;
 106   char *out;
 107 }
 108 nfkc[] =
 109 {
 110   {
 111   "\xC2\xB5", "\xCE\xBC"}
 112   ,
 113   {
 114   "\xC2\xAA", "\x61"}
 115 };
 116
 117 int
 118 main (int argc, char *argv[])
 119 {
 120   char *out;
 121   int i;
 122
 123   do
 124     if (strcmp (argv[argc - 1], "-v") == 0 ||
 125         strcmp (argv[argc - 1], "--verbose") == 0)
 126       debug = 1;
 127     else if (strcmp (argv[argc - 1], "-b") == 0 ||
 128              strcmp (argv[argc - 1], "--break-on-error") == 0)
 129       break_on_error = 1;
 130     else if (strcmp (argv[argc - 1], "-h") == 0 ||
 131              strcmp (argv[argc - 1], "-?") == 0 ||
 132              strcmp (argv[argc - 1], "--help") == 0)
 133       {
 134         printf ("Usage: %s [-vbh?] [--verbose] [--break-on-error] [--help]\n",
 135                 argv[0]);
 136         return 1;
 137       }
 138   while (argc-- > 1);
 139
 140   for (i = 0; i < sizeof (nfkc) / sizeof (nfkc[0]); i++)
 141     {
 142       if (debug)
 143         printf ("NFKC entry %d\n", i);
 144
 145       out = stringprep_utf8_nfkc_normalize (nfkc[i].in, strlen (nfkc[i].in));
 146       if (out == NULL)
 147         {
 148           fail ("gsasl_utf8_nfkc_normalize() entry %d failed fatally\n", i);
 149           continue;
 150         }
 151
 152       if (debug)
 153         {
 154           printf ("in:\n");
 155           escapeprint (nfkc[i].in, strlen (nfkc[i].in));
 156           hexprint (nfkc[i].in, strlen (nfkc[i].in));
 157           puts ("");
 158           binprint (nfkc[i].in, strlen (nfkc[i].in));
 159           puts ("");
 160
 161           printf ("out:\n");
 162           escapeprint (out, strlen (out));
 163           hexprint (out, strlen (out));
 164           puts ("");
 165           binprint (out, strlen (out));
 166           puts ("");
 167
 168           printf ("expected out:\n");
 169           escapeprint (nfkc[i].out, strlen (nfkc[i].out));
 170           hexprint (nfkc[i].out, strlen (nfkc[i].out));
 171           puts ("");
 172           binprint (nfkc[i].out, strlen (nfkc[i].out));
 173           puts ("");
 174         }
 175
 176       if (strlen (nfkc[i].out) != strlen (out) ||
 177           memcmp (nfkc[i].out, out, strlen (out)) != 0)
 178         {
 179           fail ("gsasl_utf8_nfkc_normalize() entry %d failed\n", i);
 180           if (debug)
 181             printf ("ERROR\n");
 182         }
 183       else if (debug)
 184         printf ("OK\n");
 185     }
 186
 187   if (debug)
 188     printf ("Libgsasl unicode self tests done with %d errors\n", error_count);
 189
 190   return error_count ? 1 : 0;
 191 }