libidn/iconvme.c

   1 /* Recode strings between character sets, using iconv.
   2    Copyright (C) 2002, 2003, 2004, 2005 Free Software Foundation, Inc.
   3
   4    This program is free software; you can redistribute it and/or
   5    modify it under the terms of the GNU Lesser General Public License as
   6    published by the Free Software Foundation; either version 2.1, or (at
   7    your option) any later version.
   8
   9    This program is distributed in the hope that it will be useful,
  10    but WITHOUT ANY WARRANTY; without even the implied warranty of
  11    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  12    GNU Lesser General Public License for more details.
  13
  14    You should have received a copy of the GNU Lesser General Public
  15    License along with this program; if not, see
  16    <http://www.gnu.org/licenses/>.  */
  17
  18 #ifdef HAVE_CONFIG_H
  19 # include <config.h>
  20 #endif
  21
  22 /* Get prototype. */
  23 #include "iconvme.h"
  24
  25 /* Get malloc. */
  26 #include <stdlib.h>
  27
  28 /* Get strcmp. */
  29 #include <string.h>
  30
  31 /* Get errno. */
  32 #include <errno.h>
  33
  34 #ifdef _LIBC
  35 # define HAVE_ICONV 1
  36 #else
  37 /* Get strdup. */
  38 # include "strdup.h"
  39 #endif
  40
  41 #if HAVE_ICONV
  42 /* Get iconv etc. */
  43 # include <iconv.h>
  44 /* Get MB_LEN_MAX, CHAR_BIT.  */
  45 # include <limits.h>
  46 #endif
  47
  48 #ifndef SIZE_MAX
  49 # define SIZE_MAX ((size_t) -1)
  50 #endif
  51
  52 /* Convert a zero-terminated string STR from the FROM_CODSET code set
  53    to the TO_CODESET code set.  The returned string is allocated using
  54    malloc, and must be dellocated by the caller using free.  On
  55    failure, NULL is returned and errno holds the error reason.  Note
  56    that if TO_CODESET uses \0 for anything but to terminate the
  57    string, the caller of this function may have difficulties finding
  58    out the length of the output string.  */
  59 char *
  60 iconv_string (const char *str, const char *from_codeset,
  61               const char *to_codeset)
  62 {
  63   char *dest = NULL;
  64 #if HAVE_ICONV
  65   iconv_t cd;
  66   char *outp;
  67   char *p = (char *) str;
  68   size_t inbytes_remaining = strlen (p);
  69   /* Guess the maximum length the output string can have.  */
  70   size_t outbuf_size = inbytes_remaining + 1;
  71   size_t outbytes_remaining;
  72   size_t err;
  73   int have_error = 0;
  74
  75   /* Use a worst-case output size guess, so long as that wouldn't be
  76      too large for comfort.  It's OK if the guess is wrong so long as
  77      it's nonzero.  */
  78   size_t approx_sqrt_SIZE_MAX = SIZE_MAX >> (sizeof (size_t) * CHAR_BIT / 2);
  79   if (outbuf_size <= approx_sqrt_SIZE_MAX / MB_LEN_MAX)
  80     outbuf_size *= MB_LEN_MAX;
  81   outbytes_remaining = outbuf_size - 1;
  82 #endif
  83
  84   if (strcmp (to_codeset, from_codeset) == 0)
  85     return strdup (str);
  86
  87 #if HAVE_ICONV
  88   cd = iconv_open (to_codeset, from_codeset);
  89   if (cd == (iconv_t) -1)
  90     return NULL;
  91
  92   outp = dest = (char *) malloc (outbuf_size);
  93   if (dest == NULL)
  94     goto out;
  95
  96 again:
  97   err = iconv (cd, &p, &inbytes_remaining, &outp, &outbytes_remaining);
  98
  99   if (err == (size_t) - 1)
 100     {
 101       switch (errno)
 102         {
 103         case EINVAL:
 104           /* Incomplete text, do not report an error */
 105           break;
 106
 107         case E2BIG:
 108           {
 109             size_t used = outp - dest;
 110             size_t newsize = outbuf_size * 2;
 111             char *newdest;
 112
 113             if (newsize <= outbuf_size)
 114               {
 115                 errno = ENOMEM;
 116                 have_error = 1;
 117                 goto out;
 118               }
 119             newdest = (char *) realloc (dest, newsize);
 120             if (newdest == NULL)
 121               {
 122                 have_error = 1;
 123                 goto out;
 124               }
 125             dest = newdest;
 126             outbuf_size = newsize;
 127
 128             outp = dest + used;
 129             outbytes_remaining = outbuf_size - used - 1;        /* -1 for NUL */
 130
 131             goto again;
 132           }
 133           break;
 134
 135         case EILSEQ:
 136           have_error = 1;
 137           break;
 138
 139         default:
 140           have_error = 1;
 141           break;
 142         }
 143     }
 144
 145   *outp = '\0';
 146
 147 out:
 148   {
 149     int save_errno = errno;
 150
 151     if (iconv_close (cd) < 0 && !have_error)
 152       {
 153         /* If we didn't have a real error before, make sure we restore
 154            the iconv_close error below. */
 155         save_errno = errno;
 156         have_error = 1;
 157       }
 158
 159     if (have_error && dest)
 160       {
 161         free (dest);
 162         dest = NULL;
 163         errno = save_errno;
 164       }
 165   }
 166 #else
 167   errno = ENOSYS;
 168 #endif
 169
 170   return dest;
 171 }