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 <bruno@clisp.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
86
/* When compiling with "gcc -x c++", produce a function with C linkage. */
87
extern "C" const char * locale_charset (void);
90
/* The following static variable is declared 'volatile' to avoid a
91
possible multithread problem in the function get_charset_aliases. If we
92
are running in a threaded environment, and if two threads initialize
93
'charset_aliases' simultaneously, both will produce the same value,
94
and everything will be ok if the two assignments to 'charset_aliases'
95
are atomic. But I don't know what will happen if the two assignments mix. */
97
# define volatile /* empty */
99
/* Pointer to the contents of the charset.alias file, if it has already been
100
read, else NULL. Its format is:
101
ALIAS_1 '\0' CANONICAL_1 '\0' ... ALIAS_n '\0' CANONICAL_n '\0' '\0' */
102
static const char * volatile charset_aliases;
104
/* Return a pointer to the contents of the charset.alias file. */
106
get_charset_aliases ()
110
cp = charset_aliases;
115
const char *dir = LIBDIR;
116
const char *base = "charset.alias";
119
/* Concatenate dir and base into freshly allocated file_name. */
121
size_t dir_len = strlen (dir);
122
size_t base_len = strlen (base);
123
int add_slash = (dir_len > 0 && !ISSLASH (dir[dir_len - 1]));
124
file_name = (char *) malloc (dir_len + add_slash + base_len + 1);
125
if (file_name != NULL)
127
memcpy (file_name, dir, dir_len);
129
file_name[dir_len] = DIRECTORY_SEPARATOR;
130
memcpy (file_name + dir_len + add_slash, base, base_len + 1);
134
if (file_name == NULL || (fp = fopen (file_name, "r")) == NULL)
135
/* Out of memory or file not found, treat it as empty. */
139
/* Parse the file's contents. */
143
char *res_ptr = NULL;
152
if (c == '\n' || c == ' ' || c == '\t')
156
/* Skip comment, to end of line. */
159
while (!(c == EOF || c == '\n'));
165
if (fscanf (fp, "%50s %50s", buf1, buf2) < 2)
171
res_size = l1 + 1 + l2 + 1;
172
res_ptr = (char *) malloc (res_size + 1);
176
res_size += l1 + 1 + l2 + 1;
177
res_ptr = (char *) realloc (res_ptr, res_size + 1);
185
strcpy (res_ptr + res_size - (l2 + 1) - (l1 + 1), buf1);
186
strcpy (res_ptr + res_size - (l2 + 1), buf2);
193
*(res_ptr + res_size) = '\0';
198
if (file_name != NULL)
203
/* To avoid the troubles of installing a separate file in the same
204
directory as the DLL and of retrieving the DLL's directory at
205
runtime, simply inline the aliases here. */
208
cp = "CP936" "\0" "GBK" "\0"
209
"CP1361" "\0" "JOHAB" "\0"
210
"CP20127" "\0" "ASCII" "\0"
211
"CP20866" "\0" "KOI8-R" "\0"
212
"CP21866" "\0" "KOI8-RU" "\0"
213
"CP28591" "\0" "ISO-8859-1" "\0"
214
"CP28592" "\0" "ISO-8859-2" "\0"
215
"CP28593" "\0" "ISO-8859-3" "\0"
216
"CP28594" "\0" "ISO-8859-4" "\0"
217
"CP28595" "\0" "ISO-8859-5" "\0"
218
"CP28596" "\0" "ISO-8859-6" "\0"
219
"CP28597" "\0" "ISO-8859-7" "\0"
220
"CP28598" "\0" "ISO-8859-8" "\0"
221
"CP28599" "\0" "ISO-8859-9" "\0"
222
"CP28605" "\0" "ISO-8859-15" "\0";
226
charset_aliases = cp;
232
/* Determine the current locale's character encoding, and canonicalize it
233
into one of the canonical names listed in config.charset.
234
The result must not be freed; it is statically allocated.
235
If the canonical name cannot be determined, the result is a non-canonical
247
#if !(defined WIN32 || defined OS2)
249
# if HAVE_LANGINFO_CODESET
251
/* Most systems support nl_langinfo (CODESET) nowadays. */
252
codeset = nl_langinfo (CODESET);
256
/* On old systems which lack it, use setlocale or getenv. */
257
const char *locale = NULL;
259
/* But most old systems don't have a complete set of locales. Some
260
(like SunOS 4 or DJGPP) have only the C locale. Therefore we don't
261
use setlocale here; it would return "C" when it doesn't support the
262
locale name the user has set. */
263
# if HAVE_SETLOCALE && 0
264
locale = setlocale (LC_CTYPE, NULL);
266
if (locale == NULL || locale[0] == '\0')
268
locale = getenv ("LC_ALL");
269
if (locale == NULL || locale[0] == '\0')
271
locale = getenv ("LC_CTYPE");
272
if (locale == NULL || locale[0] == '\0')
273
locale = getenv ("LANG");
277
/* On some old systems, one used to set locale = "iso8859_1". On others,
278
you set it to "language_COUNTRY.charset". In any case, we resolve it
279
through the charset.alias file. */
286
static char buf[2 + 10 + 1];
288
/* Woe32 has a function returning the locale's codepage as a number. */
289
sprintf (buf, "CP%u", GetACP ());
295
static char buf[2 + 10 + 1];
299
/* Allow user to override the codeset, as set in the operating system,
300
with standard language environment variables. */
301
locale = getenv ("LC_ALL");
302
if (locale == NULL || locale[0] == '\0')
304
locale = getenv ("LC_CTYPE");
305
if (locale == NULL || locale[0] == '\0')
306
locale = getenv ("LANG");
308
if (locale != NULL && locale[0] != '\0')
310
/* If the locale name contains an encoding after the dot, return it. */
311
const char *dot = strchr (locale, '.');
315
const char *modifier;
318
/* Look for the possible @... trailer and remove it, if any. */
319
modifier = strchr (dot, '@');
320
if (modifier == NULL)
322
if (modifier - dot < sizeof (buf))
324
memcpy (buf, dot, modifier - dot);
325
buf [modifier - dot] = '\0';
330
/* Resolve through the charset.alias file. */
335
/* OS/2 has a function returning the locale's codepage as a number. */
336
if (DosQueryCp (sizeof (cp), cp, &cplen))
340
sprintf (buf, "CP%u", cp[0]);
348
/* The canonical name cannot be determined. */
352
for (aliases = get_charset_aliases ();
354
aliases += strlen (aliases) + 1, aliases += strlen (aliases) + 1)
355
if (strcmp (codeset, aliases) == 0
356
|| (aliases[0] == '*' && aliases[1] == '\0'))
358
codeset = aliases + strlen (aliases) + 1;
362
/* Don't return an empty string. GNU libc and GNU libiconv interpret
363
the empty string as denoting "the locale's character encoding",
364
thus GNU libiconv would call this function a second time. */
365
if (codeset[0] == '\0')