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>.
35 int Charset_is_utf8 = 0;
37 static int charset_is_ja = 0;
38 static iconv_t charset_to_utf8 = (iconv_t)(-1);
39 static iconv_t charset_from_utf8 = (iconv_t)(-1);
42 void mutt_set_charset (char *charset)
46 mutt_canonical_charset (buffer, sizeof (buffer), charset);
51 if (charset_to_utf8 != (iconv_t)(-1))
53 iconv_close (charset_to_utf8);
54 charset_to_utf8 = (iconv_t)(-1);
56 if (charset_from_utf8 != (iconv_t)(-1))
58 iconv_close (charset_from_utf8);
59 charset_from_utf8 = (iconv_t)(-1);
63 if (!strcmp(buffer, "utf-8"))
66 else if (!strcmp(buffer, "euc-jp") || !strcmp(buffer, "shift_jis")
67 || !strcmp(buffer, "cp932"))
70 charset_to_utf8 = iconv_open ("UTF-8", charset);
71 charset_from_utf8 = iconv_open (charset, "UTF-8");
75 #ifdef HAVE_BIND_TEXTDOMAIN_CODESET
76 bind_textdomain_codeset(PACKAGE, buffer);
83 * For systems that don't have them, we provide here our own
84 * implementations of wcrtomb(), mbrtowc(), iswprint() and wcwidth().
85 * Instead of using the locale, as these functions normally would,
86 * we use Mutt's Charset variable. We support 3 types of charset:
87 * (1) For 8-bit charsets, wchar_t uses the same encoding as char.
88 * (2) For UTF-8, wchar_t uses UCS.
89 * (3) For stateless Japanese encodings, we use UCS and convert
90 * via UTF-8 using iconv.
91 * Unfortunately, we can't handle non-stateless encodings.
94 static size_t wcrtomb_iconv (char *s, wchar_t wc, iconv_t cd)
103 ibl = mutt_wctoutf8 (buf, wc);
104 if (ibl == (size_t)(-1))
109 r = iconv (cd, &ib, &ibl, &ob, &obl);
117 r = iconv (cd, &ib, &ibl, &ob, &obl);
122 size_t wcrtomb (char *s, wchar_t wc, mbstate_t *ps)
124 /* We only handle stateless encodings, so we can ignore ps. */
127 return mutt_wctoutf8 (s, wc);
128 else if (charset_from_utf8 != (iconv_t)(-1))
129 return wcrtomb_iconv (s, wc, charset_from_utf8);
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];
156 t = memchr (ps, 0, sizeof (*ps));
157 k = t ? (t - (char *)ps) : sizeof (*ps);
158 if (k > sizeof (bufi))
162 /* use the buffer for input */
163 memcpy (bufi, ps, k);
165 ibmax = bufi + (k + n < sizeof (bufi) ? k + n : sizeof (bufi));
166 memcpy (bufi + k, s, ibmax - bufi - k);
170 /* use the real input */
181 r = iconv (cd, &ib, &ibl, &ob, &obl);
182 if (ob > bufo && (!k || ib > bufi + k))
184 /* we have a character */
185 memset (ps, 0, sizeof (*ps));
186 utf8rtowc (pwc, bufo, ob - bufo, &mbstate);
187 return (pwc && *pwc) ? (ib - (k ? bufi + k : s)) : 0;
189 else if (!r || (r == (size_t)(-1) && errno == EINVAL))
191 if (ib + ibl < ibmax)
192 /* try using more input */
194 else if (k && ib > bufi + k && bufi + k + n > ibmax)
196 /* switch to using real input */
197 ib = s + (ib - bufi - k);
204 /* save the state and give up */
205 memset (ps, 0, sizeof (*ps));
206 if (ibl <= sizeof (mbstate_t)) /* need extra condition here! */
207 memcpy (ps, ib, ibl);
220 size_t mbrtowc (wchar_t *pwc, const char *s, size_t n, mbstate_t *ps)
222 static mbstate_t mbstate;
228 return utf8rtowc (pwc, s, n, ps);
229 else if (charset_to_utf8 != (iconv_t)(-1))
230 return mbrtowc_iconv (pwc, s, n, ps, charset_to_utf8);
235 memset(ps, 0, sizeof(*ps));
241 *pwc = (wchar_t)(unsigned char)*s;
246 int iswprint (wint_t wc)
248 if (Charset_is_utf8 || charset_is_ja)
249 return ((0x20 <= wc && wc < 0x7f) || 0xa0 <= wc);
251 return (0 <= wc && wc < 256) ? IsPrint (wc) : 0;
254 int iswspace (wint_t wc)
256 if (Charset_is_utf8 || charset_is_ja)
257 return (9 <= wc && wc <= 13) || wc == 32;
259 return (0 <= wc && wc < 256) ? isspace (wc) : 0;
262 static wint_t towupper_ucs (wint_t x)
264 /* Only works for x < 0x130 */
265 if ((0x60 < x && x < 0x7b) || (0xe0 <= x && x < 0xff && x != 0xf7))
267 else if (0x100 <= x && x < 0x130)
277 static wint_t towlower_ucs (wint_t x)
279 /* Only works for x < 0x130 */
280 if ((0x40 < x && x < 0x5b) || (0xc0 <= x && x < 0xdf && x != 0xd7))
282 else if (0x100 <= x && x < 0x130)
288 static int iswalnum_ucs (wint_t wc)
290 /* Only works for x < 0x220 */
298 return (0x40 < (wc & ~0x20) && (wc & ~0x20) < 0x5b);
300 return (wc == 0xaa || wc == 0xb5 || wc == 0xba);
302 return !(wc == 0xd7 || wc == 0xf7);
305 wint_t towupper (wint_t wc)
307 if (Charset_is_utf8 || charset_is_ja)
308 return towupper_ucs (wc);
310 return (0 <= wc && wc < 256) ? toupper (wc) : wc;
313 wint_t towlower (wint_t wc)
315 if (Charset_is_utf8 || charset_is_ja)
316 return towlower_ucs (wc);
318 return (0 <= wc && wc < 256) ? tolower (wc) : wc;
321 int iswalnum (wint_t wc)
323 if (Charset_is_utf8 || charset_is_ja)
324 return iswalnum_ucs (wc);
326 return (0 <= wc && wc < 256) ? isalnum (wc) : 0;
331 * Symbols, Greek and Cyrillic in JIS X 0208, Japanese Kanji
332 * Character Set, have a column width of 2.
334 int wcwidth_ja (wchar_t ucs)
337 return -1; /* continue with the normal check */
338 /* a rough range for quick check */
339 if ((ucs >= 0x00a1 && ucs <= 0x00fe) || /* Latin-1 Supplement */
340 (ucs >= 0x0391 && ucs <= 0x0451) || /* Greek and Cyrillic */
341 (ucs >= 0x2010 && ucs <= 0x266f) || /* Symbols */
342 (ucs >= 0x3000 && ucs <= 0x3020)) /* CJK Symbols and Punctuation */
348 int wcwidth_ucs(wchar_t ucs);
350 int wcwidth (wchar_t wc)
352 if (!Charset_is_utf8)
359 else if ((0 <= wc && wc < 256) && IsPrint (wc))
367 int k = wcwidth_ja (wc);
372 return wcwidth_ucs (wc);
375 size_t utf8rtowc (wchar_t *pwc, const char *s, size_t n, mbstate_t *_ps)
377 static wchar_t mbstate;
378 wchar_t *ps = (wchar_t *)_ps;
397 c = (unsigned char)*s;
410 wc = ((c & 0x1f) << 6) + (count = 0);
412 wc = ((c & 0x0f) << 12) + (count = 1);
414 wc = ((c & 0x07) << 18) + (count = 2);
416 wc = ((c & 0x03) << 24) + (count = 3);
418 wc = ((c & 0x01) << 30) + (count = 4);
428 wc = *ps & 0x7fffffff;
429 count = wc & 7; /* if count > 4 it will be caught below */
432 for (; n; ++s, --n, ++k)
434 c = (unsigned char)*s;
435 if (0x80 <= c && c < 0xc0)
437 wc |= (c & 0x3f) << (6 * count);
446 if (!(wc >> (11+count*5)))
448 errno = count < 4 ? EILSEQ : EINVAL;
462 #endif /* !HAVE_WC_FUNCS */
464 wchar_t replacement_char ()
466 return Charset_is_utf8 ? 0xfffd : '?';