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 /* The following static variable is declared 'volatile' to avoid a
  67    possible multithread problem in the function get_charset_aliases. If we
  68    are running in a threaded environment, and if two threads initialize
  69    'charset_aliases' simultaneously, both will produce the same value,
  70    and everything will be ok if the two assignments to 'charset_aliases'
  71    are atomic. But I don't know what will happen if the two assignments mix.  */
  72 /* Pointer to the contents of the charset.alias file, if it has already been
  73    read, else NULL.  Its format is:
  74    ALIAS_1 '\0' CANONICAL_1 '\0' ... ALIAS_n '\0' CANONICAL_n '\0' '\0'  */
  75 static char * volatile charset_aliases;
  76
  77 /* Return a pointer to the contents of the charset.alias file.  */
  78 static const char *
  79 get_charset_aliases ()
  80 {
  81   char *cp;
  82
  83   cp = charset_aliases;
  84   if (cp == NULL)
  85     {
  86 #ifndef WIN32
  87       FILE *fp;
  88       const char *dir = LIBDIR;
  89       const char *base = "charset.alias";
  90       char *file_name;
  91
  92       /* Concatenate dir and base into freshly allocated file_name.  */
  93       {
  94         size_t dir_len = strlen (dir);
  95         size_t base_len = strlen (base);
  96         int add_slash = (dir_len > 0 && !ISSLASH (dir[dir_len - 1]));
  97         file_name = (char *) malloc (dir_len + add_slash + base_len + 1);
  98         if (file_name != NULL)
  99           {
 100             memcpy (file_name, dir, dir_len);
 101             if (add_slash)
 102               file_name[dir_len] = DIRECTORY_SEPARATOR;
 103             memcpy (file_name + dir_len + add_slash, base, base_len + 1);
 104           }
 105       }
 106
 107       if (file_name == NULL || (fp = fopen (file_name, "r")) == NULL)
 108         /* Out of memory or file not found, treat it as empty.  */
 109         cp = "";
 110       else
 111         {
 112           /* Parse the file's contents.  */
 113           int c;
 114           char buf1[50+1];
 115           char buf2[50+1];
 116           char *res_ptr = NULL;
 117           size_t res_size = 0;
 118           size_t l1, l2;
 119
 120           for (;;)
 121             {
 122               c = getc (fp);
 123               if (c == EOF)
 124                 break;
 125               if (c == '\n' || c == ' ' || c == '\t')
 126                 continue;
 127               if (c == '#')
 128                 {
 129                   /* Skip comment, to end of line.  */
 130                   do
 131                     c = getc (fp);
 132                   while (!(c == EOF || c == '\n'));
 133                   if (c == EOF)
 134                     break;
 135                   continue;
 136                 }
 137               ungetc (c, fp);
 138               if (fscanf(fp, "%50s %50s", buf1, buf2) < 2)
 139                 break;
 140               l1 = strlen (buf1);
 141               l2 = strlen (buf2);
 142               if (res_size == 0)
 143                 {
 144                   res_size = l1 + 1 + l2 + 1;
 145                   res_ptr = malloc (res_size + 1);
 146                 }
 147               else
 148                 {
 149                   res_size += l1 + 1 + l2 + 1;
 150                   res_ptr = realloc (res_ptr, res_size + 1);
 151                 }
 152               if (res_ptr == NULL)
 153                 {
 154                   /* Out of memory. */
 155                   res_size = 0;
 156                   break;
 157                 }
 158               strcpy (res_ptr + res_size - (l2 + 1) - (l1 + 1), buf1);
 159               strcpy (res_ptr + res_size - (l2 + 1), buf2);
 160             }
 161           fclose (fp);
 162           if (res_size == 0)
 163             cp = "";
 164           else
 165             {
 166               *(res_ptr + res_size) = '\0';
 167               cp = res_ptr;
 168             }
 169         }
 170
 171       if (file_name != NULL)
 172         free (file_name);
 173
 174 #else /* WIN32 */
 175
 176       /* To avoid the troubles of installing a separate file in the same
 177          directory as the DLL and of retrieving the DLL's directory at
 178          runtime, simply inline the aliases here.  */
 179
 180       cp = "CP936" "\0" "GBK" "\0"
 181            "CP1361" "\0" "JOHAB" "\0";
 182 #endif
 183
 184       charset_aliases = cp;
 185     }
 186
 187   return cp;
 188 }
 189
 190 /* Determine the current locale's character encoding, and canonicalize it
 191    into one of the canonical names listed in config.charset.
 192    The result must not be freed; it is statically allocated.
 193    If the canonical name cannot be determined, the result is a non-canonical
 194    name.  */
 195
 196 #ifdef STATIC
 197 STATIC
 198 #endif
 199 const char *
 200 locale_charset ()
 201 {
 202   const char *codeset;
 203   const char *aliases;
 204
 205 #ifndef WIN32
 206
 207 # if HAVE_LANGINFO_CODESET
 208
 209   /* Most systems support nl_langinfo (CODESET) nowadays.  */
 210   codeset = nl_langinfo (CODESET);
 211
 212 # else
 213
 214   /* On old systems which lack it, use setlocale or getenv.  */
 215   const char *locale = NULL;
 216
 217   /* But most old systems don't have a complete set of locales.  Some
 218      (like SunOS 4 or DJGPP) have only the C locale.  Therefore we don't
 219      use setlocale here; it would return "C" when it doesn't support the
 220      locale name the user has set.  */
 221 #  if HAVE_SETLOCALE && 0
 222   locale = setlocale (LC_CTYPE, NULL);
 223 #  endif
 224   if (locale == NULL || locale[0] == '\0')
 225     {
 226       locale = getenv ("LC_ALL");
 227       if (locale == NULL || locale[0] == '\0')
 228         {
 229           locale = getenv ("LC_CTYPE");
 230           if (locale == NULL || locale[0] == '\0')
 231             locale = getenv ("LANG");
 232         }
 233     }
 234
 235   /* On some old systems, one used to set locale = "iso8859_1". On others,
 236      you set it to "language_COUNTRY.charset". In any case, we resolve it
 237      through the charset.alias file.  */
 238   codeset = locale;
 239
 240 # endif
 241
 242 #else /* WIN32 */
 243
 244   static char buf[2 + 10 + 1];
 245
 246   /* Win32 has a function returning the locale's codepage as a number.  */
 247   sprintf (buf, "CP%u", GetACP ());
 248   codeset = buf;
 249
 250 #endif
 251
 252   if (codeset == NULL)
 253     /* The canonical name cannot be determined.  */
 254     codeset = "";
 255
 256   /* Resolve alias. */
 257   for (aliases = get_charset_aliases ();
 258        *aliases != '\0';
 259        aliases += strlen (aliases) + 1, aliases += strlen (aliases) + 1)
 260     if (strcmp (codeset, aliases) == 0
 261         || (aliases[0] == '*' && aliases[1] == '\0'))
 262       {
 263         codeset = aliases + strlen (aliases) + 1;
 264         break;
 265       }
 266
 267   return codeset;
 268 }