1
/* Determine a canonical name for the current locale's character encoding.
3
Copyright (C) 2000-2001 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
# if HAVE_LANGINFO_CODESET
47
# include <langinfo.h>
54
# define WIN32_LEAN_AND_MEAN
58
#ifndef DIRECTORY_SEPARATOR
59
# define DIRECTORY_SEPARATOR '/'
63
# define ISSLASH(C) ((C) == DIRECTORY_SEPARATOR)
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. */
73
# define volatile /* empty */
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 char * volatile charset_aliases;
80
/* Return a pointer to the contents of the charset.alias file. */
82
get_charset_aliases ()
91
const char *dir = LIBDIR;
92
const char *base = "charset.alias";
95
/* Concatenate dir and base into freshly allocated file_name. */
97
size_t dir_len = strlen (dir);
98
size_t base_len = strlen (base);
99
int add_slash = (dir_len > 0 && !ISSLASH (dir[dir_len - 1]));
100
file_name = (char *) malloc (dir_len + add_slash + base_len + 1);
101
if (file_name != NULL)
103
memcpy (file_name, dir, dir_len);
105
file_name[dir_len] = DIRECTORY_SEPARATOR;
106
memcpy (file_name + dir_len + add_slash, base, base_len + 1);
110
if (file_name == NULL || (fp = fopen (file_name, "r")) == NULL)
111
/* Out of memory or file not found, treat it as empty. */
115
/* Parse the file's contents. */
119
char *res_ptr = NULL;
128
if (c == '\n' || c == ' ' || c == '\t')
132
/* Skip comment, to end of line. */
135
while (!(c == EOF || c == '\n'));
141
if (fscanf(fp, "%50s %50s", buf1, buf2) < 2)
147
res_size = l1 + 1 + l2 + 1;
148
res_ptr = malloc (res_size + 1);
152
res_size += l1 + 1 + l2 + 1;
153
res_ptr = realloc (res_ptr, res_size + 1);
161
strcpy (res_ptr + res_size - (l2 + 1) - (l1 + 1), buf1);
162
strcpy (res_ptr + res_size - (l2 + 1), buf2);
169
*(res_ptr + res_size) = '\0';
174
if (file_name != NULL)
179
/* To avoid the troubles of installing a separate file in the same
180
directory as the DLL and of retrieving the DLL's directory at
181
runtime, simply inline the aliases here. */
183
cp = "CP936" "\0" "GBK" "\0"
184
"CP1361" "\0" "JOHAB" "\0";
187
charset_aliases = cp;
193
/* Determine the current locale's character encoding, and canonicalize it
194
into one of the canonical names listed in config.charset.
195
The result must not be freed; it is statically allocated.
196
If the canonical name cannot be determined, the result is a non-canonical
210
# if HAVE_LANGINFO_CODESET
212
/* Most systems support nl_langinfo (CODESET) nowadays. */
213
codeset = nl_langinfo (CODESET);
217
/* On old systems which lack it, use setlocale or getenv. */
218
const char *locale = NULL;
220
/* But most old systems don't have a complete set of locales. Some
221
(like SunOS 4 or DJGPP) have only the C locale. Therefore we don't
222
use setlocale here; it would return "C" when it doesn't support the
223
locale name the user has set. */
224
# if HAVE_SETLOCALE && 0
225
locale = setlocale (LC_CTYPE, NULL);
227
if (locale == NULL || locale[0] == '\0')
229
locale = getenv ("LC_ALL");
230
if (locale == NULL || locale[0] == '\0')
232
locale = getenv ("LC_CTYPE");
233
if (locale == NULL || locale[0] == '\0')
234
locale = getenv ("LANG");
238
/* On some old systems, one used to set locale = "iso8859_1". On others,
239
you set it to "language_COUNTRY.charset". In any case, we resolve it
240
through the charset.alias file. */
247
static char buf[2 + 10 + 1];
249
/* Win32 has a function returning the locale's codepage as a number. */
250
sprintf (buf, "CP%u", GetACP ());
256
/* The canonical name cannot be determined. */
260
for (aliases = get_charset_aliases ();
262
aliases += strlen (aliases) + 1, aliases += strlen (aliases) + 1)
263
if (strcmp (codeset, aliases) == 0
264
|| (aliases[0] == '*' && aliases[1] == '\0'))
266
codeset = aliases + strlen (aliases) + 1;