src/intl/gettext/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 #if __STDC__ != 1
  73 #define volatile                /* empty */
  74 #endif
  75 /* Pointer to the contents of the charset.alias file, if it has already been
  76    read, else NULL.  Its format is:
  77    ALIAS_1 '\0' CANONICAL_1 '\0' ... ALIAS_n '\0' CANONICAL_n '\0' '\0'  */
  78 static const unsigned char *volatile charset_aliases;
  79
  80 /* Return a pointer to the contents of the charset.alias file.  */
  81 static const unsigned char *
  82 get_charset_aliases(void)
  83 {
  84         const unsigned char *cp;
  85
  86         cp = charset_aliases;
  87         if (cp == NULL) {
  88 #ifndef WIN32
  89                 FILE *fp;
  90                 const unsigned char *dir = LIBDIR;
  91                 const unsigned char *base = "charset.alias";
  92                 unsigned char *file_name;
  93
  94                 /* Concatenate dir and base into freshly allocated file_name.  */
  95                 {
  96                         size_t dir_len = strlen(dir);
  97                         size_t base_len = strlen(base);
  98                         int add_slash = (dir_len > 0
  99                                          && !ISSLASH(dir[dir_len - 1]));
 100                         file_name =
 101                                 (unsigned char *) malloc(dir_len + add_slash + base_len +
 102                                                 1);
 103                         if (file_name != NULL) {
 104                                 memcpy(file_name, dir, dir_len);
 105                                 if (add_slash)
 106                                         file_name[dir_len] =
 107                                                 DIRECTORY_SEPARATOR;
 108                                 memcpy(file_name + dir_len + add_slash, base,
 109                                        base_len + 1);
 110                         }
 111                 }
 112
 113                 if (file_name == NULL || (fp = fopen(file_name, "rb")) == NULL)
 114                         /* Out of memory or file not found, treat it as empty.  */
 115                         cp = "";
 116                 else {
 117                         /* Parse the file's contents.  */
 118                         int c;
 119                         unsigned char buf1[50 + 1];
 120                         unsigned char buf2[50 + 1];
 121                         unsigned char *res_ptr = NULL;
 122                         size_t res_size = 0;
 123                         size_t l1, l2;
 124
 125                         for (;;) {
 126                                 c = getc(fp);
 127                                 if (c == EOF)
 128                                         break;
 129                                 if (c == '\n' || c == ' ' || c == '\t')
 130                                         continue;
 131                                 if (c == '#') {
 132                                         /* Skip comment, to end of line.  */
 133                                         do
 134                                                 c = getc(fp);
 135                                         while (!(c == EOF || c == '\n'));
 136                                         if (c == EOF)
 137                                                 break;
 138                                         continue;
 139                                 }
 140                                 ungetc(c, fp);
 141                                 if (fscanf(fp, "%50s %50s", buf1, buf2) < 2)
 142                                         break;
 143                                 l1 = strlen(buf1);
 144                                 l2 = strlen(buf2);
 145                                 if (res_size == 0) {
 146                                         res_size = l1 + 1 + l2 + 1;
 147                                         res_ptr = malloc(res_size + 1);
 148                                 } else {
 149                                         res_size += l1 + 1 + l2 + 1;
 150                                         res_ptr =
 151                                                 realloc(res_ptr, res_size + 1);
 152                                 }
 153                                 if (res_ptr == NULL) {
 154                                         /* Out of memory. */
 155                                         res_size = 0;
 156                                         break;
 157                                 }
 158                                 strlcpy(res_ptr + res_size - (l2 + 1) - (l1 + 1),
 159                                        buf1, l1 + 1 + l2 + 1 + 1);
 160                                 strlcpy(res_ptr + res_size - (l2 + 1), buf2, l2 + 1);
 161                         }
 162                         fclose(fp);
 163                         if (res_size == 0)
 164                                 cp = "";
 165                         else {
 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" "CP1361" "\0" "JOHAB" "\0";
 181 #endif
 182
 183                 charset_aliases = cp;
 184         }
 185
 186         return cp;
 187 }
 188
 189 /* Determine the current locale's character encoding, and canonicalize it
 190    into one of the canonical names listed in config.charset.
 191    The result must not be freed; it is statically allocated.
 192    If the canonical name cannot be determined, the result is a non-canonical
 193    name.  */
 194
 195 /* Should be in .h file, used in _nl_init_domain_conv() function. */
 196 const unsigned char *elinks_locale_charset(void);
 197
 198 #ifdef STATIC
 199 STATIC
 200 #endif
 201 const unsigned char *
 202 elinks_locale_charset(void)
 203 {
 204         const unsigned char *codeset;
 205         const unsigned char *aliases;
 206
 207 #ifndef WIN32
 208
 209 #if HAVE_LANGINFO_CODESET
 210
 211         /* Most systems support nl_langinfo (CODESET) nowadays.  */
 212         codeset = nl_langinfo(CODESET);
 213
 214 #else
 215
 216         /* On old systems which lack it, use setlocale or getenv.  */
 217         const unsigned char *locale = NULL;
 218
 219         /* But most old systems don't have a complete set of locales.  Some
 220            (like SunOS 4 or DJGPP) have only the C locale.  Therefore we don't
 221            use setlocale here; it would return "C" when it doesn't support the
 222            locale name the user has set.  */
 223 #if HAVE_SETLOCALE && 0
 224         locale = setlocale(LC_CTYPE, NULL);
 225 #endif
 226         if (locale == NULL || locale[0] == '\0') {
 227                 locale = getenv("LC_ALL");
 228                 if (locale == NULL || locale[0] == '\0') {
 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 unsigned 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                         codeset = aliases + strlen(aliases) + 1;
 263                         break;
 264                 }
 265
 266         return codeset;
 267 }