libcharset/lib/localcharset.c

   1 /* Determine a canonical name for the current locale's character encoding.
   2
   3    Copyright (C) 2000-2001 Free Software Foundation, Inc.
   4
   5    This program is free software; you can redistribute it and/or modify it
   6    under the terms of the GNU Library General Public License as published
   7    by the Free Software Foundation; either version 2, or (at your option)
   8    any later version.
   9
  10    This program 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    Library General Public License for more details.
  14
  15    You should have received a copy of the GNU Library General Public
  16    License along with this program; if not, write to the Free Software
  17    Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307,
  18    USA.  */
  19
  20 /* Written by Bruno Haible <haible@clisp.cons.org>.  */
  21
  22 #ifdef HAVE_CONFIG_H
  23 # include <config.h>
  24 #endif
  25
  26 #if HAVE_STDDEF_H
  27 # include <stddef.h>
  28 #endif
  29
  30 #include <stdio.h>
  31 #if HAVE_STRING_H
  32 # include <string.h>
  33 #else
  34 # include <strings.h>
  35 #endif
  36 #if HAVE_STDLIB_H
  37 # include <stdlib.h>
  38 #endif
  39
  40 #if defined _WIN32 || defined __WIN32__
  41 # undef WIN32   /* avoid warning on mingw32 */
  42 # define WIN32
  43 #endif
  44
  45 #ifndef WIN32
  46 # if HAVE_LANGINFO_CODESET
  47 #  include <langinfo.h>
  48 # else
  49 #  if HAVE_SETLOCALE
  50 #   include <locale.h>
  51 #  endif
  52 # endif
  53 #else /* WIN32 */
  54 # define WIN32_LEAN_AND_MEAN
  55 # include <windows.h>
  56 #endif
  57
  58 #ifndef DIRECTORY_SEPARATOR
  59 # define DIRECTORY_SEPARATOR '/'
  60 #endif
  61
  62 #ifndef ISSLASH
  63 # define ISSLASH(C) ((C) == DIRECTORY_SEPARATOR)
  64 #endif
  65
  66 #ifdef __cplusplus
  67 /* When compiling with "gcc -x c++", produce a function with C linkage.  */
  68 extern "C" const char * locale_charset (void);
  69 #endif
  70
  71 /* The following static variable is declared 'volatile' to avoid a
  72    possible multithread problem in the function get_charset_aliases. If we
  73    are running in a threaded environment, and if two threads initialize
  74    'charset_aliases' simultaneously, both will produce the same value,
  75    and everything will be ok if the two assignments to 'charset_aliases'
  76    are atomic. But I don't know what will happen if the two assignments mix.  */
  77 #if __STDC__ != 1
  78 # define volatile /* empty */
  79 #endif
  80 /* Pointer to the contents of the charset.alias file, if it has already been
  81    read, else NULL.  Its format is:
  82    ALIAS_1 '\0' CANONICAL_1 '\0' ... ALIAS_n '\0' CANONICAL_n '\0' '\0'  */
  83 static const char * volatile charset_aliases;
  84
  85 /* Return a pointer to the contents of the charset.alias file.  */
  86 static const char *
  87 get_charset_aliases ()
  88 {
  89   const char *cp;
  90
  91   cp = charset_aliases;
  92   if (cp == NULL)
  93     {
  94 #ifndef WIN32
  95       FILE *fp;
  96       const char *dir = LIBDIR;
  97       const char *base = "charset.alias";
  98       char *file_name;
  99
 100       /* Concatenate dir and base into freshly allocated file_name.  */
 101       {
 102         size_t dir_len = strlen (dir);
 103         size_t base_len = strlen (base);
 104         int add_slash = (dir_len > 0 && !ISSLASH (dir[dir_len - 1]));
 105         file_name = (char *) malloc (dir_len + add_slash + base_len + 1);
 106         if (file_name != NULL)
 107           {
 108             memcpy (file_name, dir, dir_len);
 109             if (add_slash)
 110               file_name[dir_len] = DIRECTORY_SEPARATOR;
 111             memcpy (file_name + dir_len + add_slash, base, base_len + 1);
 112           }
 113       }
 114
 115       if (file_name == NULL || (fp = fopen (file_name, "r")) == NULL)
 116         /* Out of memory or file not found, treat it as empty.  */
 117         cp = "";
 118       else
 119         {
 120           /* Parse the file's contents.  */
 121           int c;
 122           char buf1[50+1];
 123           char buf2[50+1];
 124           char *res_ptr = NULL;
 125           size_t res_size = 0;
 126           size_t l1, l2;
 127
 128           for (;;)
 129             {
 130               c = getc (fp);
 131               if (c == EOF)
 132                 break;
 133               if (c == '\n' || c == ' ' || c == '\t')
 134                 continue;
 135               if (c == '#')
 136                 {
 137                   /* Skip comment, to end of line.  */
 138                   do
 139                     c = getc (fp);
 140                   while (!(c == EOF || c == '\n'));
 141                   if (c == EOF)
 142                     break;
 143                   continue;
 144                 }
 145               ungetc (c, fp);
 146               if (fscanf(fp, "%50s %50s", buf1, buf2) < 2)
 147                 break;
 148               l1 = strlen (buf1);
 149               l2 = strlen (buf2);
 150               if (res_size == 0)
 151                 {
 152                   res_size = l1 + 1 + l2 + 1;
 153                   res_ptr = (char *) malloc (res_size + 1);
 154                 }
 155               else
 156                 {
 157                   res_size += l1 + 1 + l2 + 1;
 158                   res_ptr = (char *) realloc (res_ptr, res_size + 1);
 159                 }
 160               if (res_ptr == NULL)
 161                 {
 162                   /* Out of memory. */
 163                   res_size = 0;
 164                   break;
 165                 }
 166               strcpy (res_ptr + res_size - (l2 + 1) - (l1 + 1), buf1);
 167               strcpy (res_ptr + res_size - (l2 + 1), buf2);
 168             }
 169           fclose (fp);
 170           if (res_size == 0)
 171             cp = "";
 172           else
 173             {
 174               *(res_ptr + res_size) = '\0';
 175               cp = res_ptr;
 176             }
 177         }
 178
 179       if (file_name != NULL)
 180         free (file_name);
 181
 182 #else /* WIN32 */
 183
 184       /* To avoid the troubles of installing a separate file in the same
 185          directory as the DLL and of retrieving the DLL's directory at
 186          runtime, simply inline the aliases here.  */
 187
 188       cp = "CP936" "\0" "GBK" "\0"
 189            "CP1361" "\0" "JOHAB" "\0";
 190 #endif
 191
 192       charset_aliases = cp;
 193     }
 194
 195   return cp;
 196 }
 197
 198 /* Determine the current locale's character encoding, and canonicalize it
 199    into one of the canonical names listed in config.charset.
 200    The result must not be freed; it is statically allocated.
 201    If the canonical name cannot be determined, the result is a non-canonical
 202    name.  */
 203
 204 #ifdef STATIC
 205 STATIC
 206 #endif
 207 const char *
 208 locale_charset ()
 209 {
 210   const char *codeset;
 211   const char *aliases;
 212
 213 #ifndef WIN32
 214
 215 # if HAVE_LANGINFO_CODESET
 216
 217   /* Most systems support nl_langinfo (CODESET) nowadays.  */
 218   codeset = nl_langinfo (CODESET);
 219
 220 # else
 221
 222   /* On old systems which lack it, use setlocale or getenv.  */
 223   const char *locale = NULL;
 224
 225   /* But most old systems don't have a complete set of locales.  Some
 226      (like SunOS 4 or DJGPP) have only the C locale.  Therefore we don't
 227      use setlocale here; it would return "C" when it doesn't support the
 228      locale name the user has set.  */
 229 #  if HAVE_SETLOCALE && 0
 230   locale = setlocale (LC_CTYPE, NULL);
 231 #  endif
 232   if (locale == NULL || locale[0] == '\0')
 233     {
 234       locale = getenv ("LC_ALL");
 235       if (locale == NULL || locale[0] == '\0')
 236         {
 237           locale = getenv ("LC_CTYPE");
 238           if (locale == NULL || locale[0] == '\0')
 239             locale = getenv ("LANG");
 240         }
 241     }
 242
 243   /* On some old systems, one used to set locale = "iso8859_1". On others,
 244      you set it to "language_COUNTRY.charset". In any case, we resolve it
 245      through the charset.alias file.  */
 246   codeset = locale;
 247
 248 # endif
 249
 250 #else /* WIN32 */
 251
 252   static char buf[2 + 10 + 1];
 253
 254   /* Win32 has a function returning the locale's codepage as a number.  */
 255   sprintf (buf, "CP%u", GetACP ());
 256   codeset = buf;
 257
 258 #endif
 259
 260   if (codeset == NULL)
 261     /* The canonical name cannot be determined.  */
 262     codeset = "";
 263
 264   /* Resolve alias. */
 265   for (aliases = get_charset_aliases ();
 266        *aliases != '\0';
 267        aliases += strlen (aliases) + 1, aliases += strlen (aliases) + 1)
 268     if (strcmp (codeset, aliases) == 0
 269         || (aliases[0] == '*' && aliases[1] == '\0'))
 270       {
 271         codeset = aliases + strlen (aliases) + 1;
 272         break;
 273       }
 274
 275   return codeset;
 276 }