2 * Copyright (C) 2000 Edmund Grimley Evans <edmundo@rano.org>
4 * This program is free software; you can redistribute it and/or modify
5 * it under the terms of the GNU General Public License as published by
6 * the Free Software Foundation; either version 2 of the License, or
7 * (at your option) any later version.
9 * This program is distributed in the hope that it will be useful,
10 * but WITHOUT ANY WARRANTY; without even the implied warranty of
11 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12 * GNU General Public License for more details.
14 * You should have received a copy of the GNU General Public License
15 * along with this program; if not, write to the Free Software
16 * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111, USA.
20 * Japanese support by TAKIZAWA Takashi <taki@luna.email.ne.jp>.
39 int Charset_is_utf8 = 0;
42 static int charset_is_ja = 0;
43 static iconv_t charset_to_utf8 = (iconv_t) (-1);
44 static iconv_t charset_from_utf8 = (iconv_t) (-1);
47 void mutt_set_charset (char *charset)
51 mutt_canonical_charset (buffer, sizeof (buffer), charset);
56 if (charset_to_utf8 != (iconv_t) (-1)) {
57 iconv_close (charset_to_utf8);
58 charset_to_utf8 = (iconv_t) (-1);
60 if (charset_from_utf8 != (iconv_t) (-1)) {
61 iconv_close (charset_from_utf8);
62 charset_from_utf8 = (iconv_t) (-1);
66 if (!strcmp (buffer, "utf-8"))
69 else if (!ascii_strcasecmp (buffer, "euc-jp")
70 || !ascii_strcasecmp (buffer, "shift_jis")
71 || !ascii_strcasecmp (buffer, "cp932")
72 || !ascii_strcasecmp (buffer, "eucJP-ms")) {
74 charset_to_utf8 = iconv_open ("UTF-8", charset);
75 charset_from_utf8 = iconv_open (charset, "UTF-8");
79 #ifdef HAVE_BIND_TEXTDOMAIN_CODESET
80 bind_textdomain_codeset (PACKAGE, buffer);
87 * For systems that don't have them, we provide here our own
88 * implementations of wcrtomb(), mbrtowc(), iswprint() and wcwidth().
89 * Instead of using the locale, as these functions normally would,
90 * we use Mutt's Charset variable. We support 3 types of charset:
91 * (1) For 8-bit charsets, wchar_t uses the same encoding as char.
92 * (2) For UTF-8, wchar_t uses UCS.
93 * (3) For stateless Japanese encodings, we use UCS and convert
94 * via UTF-8 using iconv.
95 * Unfortunately, we can't handle non-stateless encodings.
98 static size_t wcrtomb_iconv (char *s, wchar_t wc, iconv_t cd)
100 char buf[MB_LEN_MAX];
101 ICONV_CONST char *ib;
106 ibl = mutt_wctoutf8 (buf, wc);
107 if (ibl == (size_t) (-1))
108 return (size_t) (-1);
112 r = iconv (cd, &ib, &ibl, &ob, &obl);
119 r = iconv (cd, &ib, &ibl, &ob, &obl);
124 size_t wcrtomb (char *s, wchar_t wc, mbstate_t * ps)
126 /* We only handle stateless encodings, so we can ignore ps. */
129 return mutt_wctoutf8 (s, wc);
130 else if (charset_from_utf8 != (iconv_t) (-1))
131 return wcrtomb_iconv (s, wc, charset_from_utf8);
140 return (size_t) (-1);
144 size_t mbrtowc_iconv (wchar_t * pwc, const char *s, size_t n,
145 mbstate_t * ps, iconv_t cd)
147 static mbstate_t mbstate;
148 ICONV_CONST char *ib, *ibmax;
150 size_t ibl, obl, k, r;
151 char bufi[8], bufo[6];
154 return (size_t) (-2);
156 t = memchr (ps, 0, sizeof (*ps));
157 k = t ? (t - (char *) ps) : sizeof (*ps);
158 if (k > sizeof (bufi))
161 /* use the buffer for input */
162 memcpy (bufi, ps, k);
164 ibmax = bufi + (k + n < sizeof (bufi) ? k + n : sizeof (bufi));
165 memcpy (bufi + k, s, ibmax - bufi - k);
168 /* use the real input */
178 r = iconv (cd, &ib, &ibl, &ob, &obl);
179 if (ob > bufo && (!k || ib > bufi + k)) {
180 /* we have a character */
181 memset (ps, 0, sizeof (*ps));
182 utf8rtowc (pwc, bufo, ob - bufo, &mbstate);
183 return (pwc && *pwc) ? (ib - (k ? bufi + k : s)) : 0;
185 else if (!r || (r == (size_t) (-1) && errno == EINVAL)) {
186 if (ib + ibl < ibmax)
187 /* try using more input */
189 else if (k && ib > bufi + k && bufi + k + n > ibmax) {
190 /* switch to using real input */
191 ib = s + (ib - bufi - k);
197 /* save the state and give up */
198 memset (ps, 0, sizeof (*ps));
199 if (ibl <= sizeof (mbstate_t)) /* need extra condition here! */
200 memcpy (ps, ib, ibl);
201 return (size_t) (-2);
207 return (size_t) (-1);
212 size_t mbrtowc (wchar_t * pwc, const char *s, size_t n, mbstate_t * ps)
214 static mbstate_t mbstate;
220 return utf8rtowc (pwc, s, n, ps);
221 else if (charset_to_utf8 != (iconv_t) (-1))
222 return mbrtowc_iconv (pwc, s, n, ps, charset_to_utf8);
225 memset (ps, 0, sizeof (*ps));
231 *pwc = (wchar_t) (unsigned char) *s;
236 int iswprint (wint_t wc)
238 if (Charset_is_utf8 || charset_is_ja)
239 return ((0x20 <= wc && wc < 0x7f) || 0xa0 <= wc);
241 return (0 <= wc && wc < 256) ? IsPrint (wc) : 0;
244 int iswspace (wint_t wc)
246 if (Charset_is_utf8 || charset_is_ja)
247 return (9 <= wc && wc <= 13) || wc == 32;
249 return (0 <= wc && wc < 256) ? isspace (wc) : 0;
252 static wint_t towupper_ucs (wint_t x)
254 /* Only works for x < 0x130 */
255 if ((0x60 < x && x < 0x7b) || (0xe0 <= x && x < 0xff && x != 0xf7))
257 else if (0x100 <= x && x < 0x130)
267 static wint_t towlower_ucs (wint_t x)
269 /* Only works for x < 0x130 */
270 if ((0x40 < x && x < 0x5b) || (0xc0 <= x && x < 0xdf && x != 0xd7))
272 else if (0x100 <= x && x < 0x130)
278 static int iswalnum_ucs (wint_t wc)
280 /* Only works for x < 0x220 */
288 return (0x40 < (wc & ~0x20) && (wc & ~0x20) < 0x5b);
290 return (wc == 0xaa || wc == 0xb5 || wc == 0xba);
292 return !(wc == 0xd7 || wc == 0xf7);
295 wint_t towupper (wint_t wc)
297 if (Charset_is_utf8 || charset_is_ja)
298 return towupper_ucs (wc);
300 return (0 <= wc && wc < 256) ? toupper (wc) : wc;
303 wint_t towlower (wint_t wc)
305 if (Charset_is_utf8 || charset_is_ja)
306 return towlower_ucs (wc);
308 return (0 <= wc && wc < 256) ? tolower (wc) : wc;
311 int iswalnum (wint_t wc)
313 if (Charset_is_utf8 || charset_is_ja)
314 return iswalnum_ucs (wc);
316 return (0 <= wc && wc < 256) ? isalnum (wc) : 0;
321 * Symbols, Greek and Cyrillic in JIS X 0208, Japanese Kanji
322 * Character Set, have a column width of 2.
324 int wcwidth_ja (wchar_t ucs)
327 return -1; /* continue with the normal check */
328 /* a rough range for quick check */
329 if ((ucs >= 0x00a1 && ucs <= 0x00fe) || /* Latin-1 Supplement */
330 (ucs >= 0x0391 && ucs <= 0x0451) || /* Greek and Cyrillic */
331 (ucs >= 0x2010 && ucs <= 0x266f) || /* Symbols */
332 (ucs >= 0x3000 && ucs <= 0x3020)) /* CJK Symbols and Punctuation */
338 int wcwidth_ucs (wchar_t ucs);
340 int wcwidth (wchar_t wc)
342 if (!Charset_is_utf8) {
343 if (!charset_is_ja) {
347 else if ((0 <= wc && wc < 256) && IsPrint (wc))
354 int k = wcwidth_ja (wc);
360 return wcwidth_ucs (wc);
363 size_t utf8rtowc (wchar_t * pwc, const char *s, size_t n, mbstate_t * _ps)
365 static wchar_t mbstate;
366 wchar_t *ps = (wchar_t *) _ps;
383 c = (unsigned char) *s;
394 wc = ((c & 0x1f) << 6) + (count = 0);
396 wc = ((c & 0x0f) << 12) + (count = 1);
398 wc = ((c & 0x07) << 18) + (count = 2);
400 wc = ((c & 0x03) << 24) + (count = 3);
402 wc = ((c & 0x01) << 30) + (count = 4);
410 wc = *ps & 0x7fffffff;
411 count = wc & 7; /* if count > 4 it will be caught below */
414 for (; n; ++s, --n, ++k) {
415 c = (unsigned char) *s;
416 if (0x80 <= c && c < 0xc0) {
417 wc |= (c & 0x3f) << (6 * count);
425 if (!(wc >> (11 + count * 5))) {
426 errno = count < 4 ? EILSEQ : EINVAL;
439 #endif /* !HAVE_WC_FUNCS */
441 wchar_t replacement_char ()
443 return Charset_is_utf8 ? 0xfffd : '?';