1
/* Determine a canonical name for the current locale's character encoding.
3
Copyright (C) 2000-2002 Free Software Foundation, Inc.
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)
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.
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,
20
/* Written by Bruno Haible <haible@clisp.cons.org>. */
40
#if defined _WIN32 || defined __WIN32__
41
# undef WIN32 /* avoid warning on mingw32 */
46
/* Assume EMX program runs on OS/2, even if compiled under DOS. */
51
# if HAVE_LANGINFO_CODESET
52
# include <langinfo.h>
59
# define WIN32_LEAN_AND_MEAN
67
#if defined _WIN32 || defined __WIN32__ || defined __EMX__ || defined __DJGPP__
68
/* Win32, OS/2, DOS */
69
# define ISSLASH(C) ((C) == '/' || (C) == '\\')
72
#ifndef DIRECTORY_SEPARATOR
73
# define DIRECTORY_SEPARATOR '/'
77
# define ISSLASH(C) ((C) == DIRECTORY_SEPARATOR)
80
#ifdef HAVE_GETC_UNLOCKED
82
# define getc getc_unlocked
85
/* The following static variable is declared 'volatile' to avoid a
86
possible multithread problem in the function get_charset_aliases. If we
87
are running in a threaded environment, and if two threads initialize
88
'charset_aliases' simultaneously, both will produce the same value,
89
and everything will be ok if the two assignments to 'charset_aliases'
90
are atomic. But I don't know what will happen if the two assignments mix. */
92
# define volatile /* empty */
94
/* Pointer to the contents of the charset.alias file, if it has already been
95
read, else NULL. Its format is:
96
ALIAS_1 '\0' CANONICAL_1 '\0' ... ALIAS_n '\0' CANONICAL_n '\0' '\0' */
97
static const char * volatile charset_aliases;
99
/* Return a pointer to the contents of the charset.alias file. */
101
get_charset_aliases ()
105
cp = charset_aliases;
110
const char *dir = LIBDIR;
111
const char *base = "charset.alias";
114
/* Concatenate dir and base into freshly allocated file_name. */
116
size_t dir_len = strlen (dir);
117
size_t base_len = strlen (base);
118
int add_slash = (dir_len > 0 && !ISSLASH (dir[dir_len - 1]));
119
file_name = (char *) malloc (dir_len + add_slash + base_len + 1);
120
if (file_name != NULL)
122
memcpy (file_name, dir, dir_len);
124
file_name[dir_len] = DIRECTORY_SEPARATOR;
125
memcpy (file_name + dir_len + add_slash, base, base_len + 1);
129
if (file_name == NULL || (fp = fopen (file_name, "r")) == NULL)
130
/* Out of memory or file not found, treat it as empty. */
134
/* Parse the file's contents. */
138
char *res_ptr = NULL;
147
if (c == '\n' || c == ' ' || c == '\t')
151
/* Skip comment, to end of line. */
154
while (!(c == EOF || c == '\n'));
160
if (fscanf (fp, "%50s %50s", buf1, buf2) < 2)
166
res_size = l1 + 1 + l2 + 1;
167
res_ptr = (char *) malloc (res_size + 1);
171
res_size += l1 + 1 + l2 + 1;
172
res_ptr = (char *) realloc (res_ptr, res_size + 1);
180
strcpy (res_ptr + res_size - (l2 + 1) - (l1 + 1), buf1);
181
strcpy (res_ptr + res_size - (l2 + 1), buf2);
188
*(res_ptr + res_size) = '\0';
193
if (file_name != NULL)
198
/* To avoid the troubles of installing a separate file in the same
199
directory as the DLL and of retrieving the DLL's directory at
200
runtime, simply inline the aliases here. */
203
cp = "CP936" "\0" "GBK" "\0"
204
"CP1361" "\0" "JOHAB" "\0";
208
charset_aliases = cp;
214
/* Determine the current locale's character encoding, and canonicalize it
215
into one of the canonical names listed in config.charset.
216
The result must not be freed; it is statically allocated.
217
If the canonical name cannot be determined, the result is a non-canonical
229
#if !(defined WIN32 || defined OS2)
231
# if HAVE_LANGINFO_CODESET
233
/* Most systems support nl_langinfo (CODESET) nowadays. */
234
codeset = nl_langinfo (CODESET);
238
/* On old systems which lack it, use setlocale or getenv. */
239
const char *locale = NULL;
241
/* But most old systems don't have a complete set of locales. Some
242
(like SunOS 4 or DJGPP) have only the C locale. Therefore we don't
243
use setlocale here; it would return "C" when it doesn't support the
244
locale name the user has set. */
245
# if HAVE_SETLOCALE && 0
246
locale = setlocale (LC_CTYPE, NULL);
248
if (locale == NULL || locale[0] == '\0')
250
locale = getenv ("LC_ALL");
251
if (locale == NULL || locale[0] == '\0')
253
locale = getenv ("LC_CTYPE");
254
if (locale == NULL || locale[0] == '\0')
255
locale = getenv ("LANG");
259
/* On some old systems, one used to set locale = "iso8859_1". On others,
260
you set it to "language_COUNTRY.charset". In any case, we resolve it
261
through the charset.alias file. */
268
static char buf[2 + 10 + 1];
270
/* Win32 has a function returning the locale's codepage as a number. */
271
sprintf (buf, "CP%u", GetACP ());
277
static char buf[2 + 10 + 1];
281
/* Allow user to override the codeset, as set in the operating system,
282
with standard language environment variables. */
283
locale = getenv ("LC_ALL");
284
if (locale == NULL || locale[0] == '\0')
286
locale = getenv ("LC_CTYPE");
287
if (locale == NULL || locale[0] == '\0')
288
locale = getenv ("LANG");
290
if (locale != NULL && locale[0] != '\0')
292
/* If the locale name contains an encoding after the dot, return it. */
293
const char *dot = strchr (locale, '.');
297
const char *modifier;
300
/* Look for the possible @... trailer and remove it, if any. */
301
modifier = strchr (dot, '@');
302
if (modifier == NULL)
304
if (modifier - dot < sizeof (buf))
306
memcpy (buf, dot, modifier - dot);
307
buf [modifier - dot] = '\0';
312
/* Resolve through the charset.alias file. */
317
/* OS/2 has a function returning the locale's codepage as a number. */
318
if (DosQueryCp (sizeof (cp), cp, &cplen))
322
sprintf (buf, "CP%u", cp[0]);
330
/* The canonical name cannot be determined. */
334
for (aliases = get_charset_aliases ();
336
aliases += strlen (aliases) + 1, aliases += strlen (aliases) + 1)
337
if (strcmp (codeset, aliases) == 0
338
|| (aliases[0] == '*' && aliases[1] == '\0'))
340
codeset = aliases + strlen (aliases) + 1;