toutf8.c

   1 /* toutf8.c     convert strings from system locale into UTF-8
   2  * Copyright (C) 2002  Simon Josefsson
   3  *
   4  * This file is part of libstringprep.
   5  *
   6  * Libstringprep is free software; you can redistribute it and/or
   7  * modify it under the terms of the GNU Lesser General Public
   8  * License as published by the Free Software Foundation; either
   9  * version 2.1 of the License, or (at your option) any later version.
  10  *
  11  * Libstringprep 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 GNU
  14  * Lesser General Public License for more details.
  15  *
  16  * You should have received a copy of the GNU Lesser General Public
  17  * License along with libstringprep; if not, write to the Free Software
  18  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
  19  *
  20  */
  21
  22 #include "internal.h"
  23
  24 #ifdef HAVE_ICONV
  25
  26 #include <iconv.h>
  27
  28 #if LOCALE_WORKS
  29 #include <langinfo.h>
  30 #include <locale.h>
  31 #endif
  32
  33 static const char *
  34 stringprep_locale_charset_slow ()
  35 {
  36   const char *charset = getenv ("CHARSET");     /* flawfinder: ignore */
  37   char *p;
  38
  39   if (charset && *charset)
  40     return charset;
  41
  42 #if LOCALE_WORKS
  43   p = setlocale (LC_CTYPE, NULL);
  44   setlocale (LC_CTYPE, "");
  45
  46   charset = nl_langinfo (CODESET);
  47
  48   setlocale (LC_CTYPE, p);
  49
  50   if (charset && *charset)
  51     return charset;
  52 #endif
  53
  54   return "ASCII";
  55 }
  56
  57 static const char *stringprep_locale_charset_cache = NULL;
  58
  59 const char *
  60 stringprep_locale_charset ()
  61 {
  62   if (!stringprep_locale_charset_cache)
  63     stringprep_locale_charset_cache = stringprep_locale_charset_slow ();
  64
  65   return stringprep_locale_charset_cache;
  66 }
  67
  68 char *
  69 stringprep_convert (const char *str,
  70                     const char *to_codeset, const char *from_codeset)
  71 {
  72   iconv_t cd;
  73   char *dest;
  74   char *outp;
  75   char *p;
  76   size_t inbytes_remaining;
  77   size_t outbytes_remaining;
  78   size_t err;
  79   size_t outbuf_size;
  80   int have_error = 0;
  81   int len;
  82
  83   if (strcmp (to_codeset, from_codeset) == 0)
  84     return strdup (str);
  85
  86   cd = iconv_open (to_codeset, from_codeset);
  87
  88   if (cd == (iconv_t) - 1)
  89     return NULL;
  90
  91   len = strlen (str);
  92
  93   p = (char *) str;
  94   inbytes_remaining = len;
  95   outbuf_size = len + 1;        /* + 1 for nul in case len == 1 */
  96
  97   outbytes_remaining = outbuf_size - 1; /* -1 for nul */
  98   outp = dest = malloc (outbuf_size);
  99
 100 again:
 101
 102   err = iconv (cd, &p, &inbytes_remaining, &outp, &outbytes_remaining);
 103
 104   if (err == (size_t) - 1)
 105     {
 106       switch (errno)
 107         {
 108         case EINVAL:
 109           /* Incomplete text, do not report an error */
 110           break;
 111
 112         case E2BIG:
 113           {
 114             size_t used = outp - dest;
 115
 116             outbuf_size *= 2;
 117             dest = realloc (dest, outbuf_size);
 118
 119             outp = dest + used;
 120             outbytes_remaining = outbuf_size - used - 1;        /* -1 for nul */
 121
 122             goto again;
 123           }
 124           break;
 125
 126         case EILSEQ:
 127           have_error = 1;
 128           break;
 129
 130         default:
 131           have_error = 1;
 132           break;
 133         }
 134     }
 135
 136   *outp = '\0';
 137
 138   if ((p - str) != len)
 139     have_error = 1;
 140
 141   iconv_close (cd);
 142
 143   if (have_error)
 144     {
 145       free (dest);
 146       dest = NULL;
 147     }
 148
 149   return dest;
 150 }
 151
 152 #else
 153
 154 const char *
 155 stringprep_locale_charset ()
 156 {
 157   return NULL;
 158 }
 159
 160 char *
 161 stringprep_convert (const char *str,
 162                     const char *to_codeset, const char *from_codeset)
 163 {
 164   return NULL;
 165 }
 166
 167 #endif
 168
 169 char *
 170 stringprep_locale_to_utf8 (const char *str)
 171 {
 172   return stringprep_convert (str, "UTF-8", stringprep_locale_charset ());
 173 }