?  search:
summary | log | graphiclog1 | graphiclog2 | commit | commitdiff | tree | refs | edit | fork
blame | history | raw | HEAD
Fix bug #3574: Template addressing
[claws.git] / src / codeconv.c
blob11b23112c8b127fd15a653f9c8c04086ac8eec1c
1 /*
2 * Sylpheed -- a GTK+ based, lightweight, and fast e-mail client
3 * Copyright (C) 1999-2012 Hiroyuki Yamamoto and the Claws Mail team
4 *
5 * This program is free software; you can redistribute it and/or modify
6 * it under the terms of the GNU General Public License as published by
7 * the Free Software Foundation; either version 3 of the License, or
8 * (at your option) any later version.
9 *
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
13 * GNU General Public License for more details.
14 *
15 * You should have received a copy of the GNU General Public License
16 * along with this program. If not, see <http://www.gnu.org/licenses/>.
17 *
18 */
19
20 #ifdef HAVE_CONFIG_H
21 # include "config.h"
22 #include "claws-features.h"
23 #endif
24
25 #include "defs.h"
26
27 #include <glib.h>
28 #include <glib/gi18n.h>
29 #include <string.h>
30 #include <ctype.h>
31 #include <stdlib.h>
32 #include <errno.h>
33
34 #if HAVE_LOCALE_H
35 # include <locale.h>
36 #endif
37
38 #include "codeconv.h"
39 #include "unmime.h"
40 #include "quoted-printable.h"
41 #include "utils.h"
42 #include "prefs_common.h"
43
44 /* For unknown reasons the inconv.m4 macro undefs that macro if no
45 const is needed. This would break the code below so we define it. */
46 #ifndef ICONV_CONST
47 #define ICONV_CONST
48 #endif
49
50 typedef enum
51 {
52 JIS_ASCII,
53 JIS_KANJI,
54 JIS_HWKANA,
55 JIS_AUXKANJI
56 } JISState;
57
58 #define SUBST_CHAR 0x5f;
59 #define ESC '\033'
60
61 #define iseuckanji(c) \
62 (((c) & 0xff) >= 0xa1 && ((c) & 0xff) <= 0xfe)
63 #define iseuchwkana1(c) \
64 (((c) & 0xff) == 0x8e)
65 #define iseuchwkana2(c) \
66 (((c) & 0xff) >= 0xa1 && ((c) & 0xff) <= 0xdf)
67 #define iseucaux(c) \
68 (((c) & 0xff) == 0x8f)
69 #define issjiskanji1(c) \
70 ((((c) & 0xff) >= 0x81 && ((c) & 0xff) <= 0x9f) || \
71 (((c) & 0xff) >= 0xe0 && ((c) & 0xff) <= 0xfc))
72 #define issjiskanji2(c) \
73 ((((c) & 0xff) >= 0x40 && ((c) & 0xff) <= 0x7e) || \
74 (((c) & 0xff) >= 0x80 && ((c) & 0xff) <= 0xfc))
75 #define issjishwkana(c) \
76 (((c) & 0xff) >= 0xa1 && ((c) & 0xff) <= 0xdf)
77
78 #define K_IN() \
79 if (state != JIS_KANJI) { \
80 *out++ = ESC; \
81 *out++ = '$'; \
82 *out++ = 'B'; \
83 state = JIS_KANJI; \
84 }
85
86 #define K_OUT() \
87 if (state != JIS_ASCII) { \
88 *out++ = ESC; \
89 *out++ = '('; \
90 *out++ = 'B'; \
91 state = JIS_ASCII; \
92 }
93
94 #define HW_IN() \
95 if (state != JIS_HWKANA) { \
96 *out++ = ESC; \
97 *out++ = '('; \
98 *out++ = 'I'; \
99 state = JIS_HWKANA; \
100 }
101
102 #define AUX_IN() \
103 if (state != JIS_AUXKANJI) { \
104 *out++ = ESC; \
105 *out++ = '$'; \
106 *out++ = '('; \
107 *out++ = 'D'; \
108 state = JIS_AUXKANJI; \
109 }
110
111 static CodeConvFunc conv_get_code_conv_func (const gchar *src_charset_str,
112 const gchar *dest_charset_str);
113
114 static gchar *conv_iconv_strdup_with_cd (const gchar *inbuf,
115 iconv_t cd);
116
117 static gchar *conv_iconv_strdup (const gchar *inbuf,
118 const gchar *src_code,
119 const gchar *dest_code);
120
121 static CharSet conv_get_locale_charset (void);
122 static CharSet conv_get_outgoing_charset (void);
123 static CharSet conv_guess_ja_encoding(const gchar *str);
124 static gboolean conv_is_ja_locale (void);
125
126 static gint conv_jistoeuc(gchar *outbuf, gint outlen, const gchar *inbuf);
127 static gint conv_euctojis(gchar *outbuf, gint outlen, const gchar *inbuf);
128 static gint conv_sjistoeuc(gchar *outbuf, gint outlen, const gchar *inbuf);
129
130 static gint conv_jistoutf8(gchar *outbuf, gint outlen, const gchar *inbuf);
131 static gint conv_sjistoutf8(gchar *outbuf, gint outlen, const gchar *inbuf);
132 static gint conv_euctoutf8(gchar *outbuf, gint outlen, const gchar *inbuf);
133 static gint conv_anytoutf8(gchar *outbuf, gint outlen, const gchar *inbuf);
134
135 static gint conv_utf8toeuc(gchar *outbuf, gint outlen, const gchar *inbuf);
136 static gint conv_utf8tojis(gchar *outbuf, gint outlen, const gchar *inbuf);
137
138 static void conv_unreadable_8bit(gchar *str);
139
140 static gint conv_jistodisp(gchar *outbuf, gint outlen, const gchar *inbuf);
141 static gint conv_sjistodisp(gchar *outbuf, gint outlen, const gchar *inbuf);
142 static gint conv_euctodisp(gchar *outbuf, gint outlen, const gchar *inbuf);
143
144 static gint conv_anytodisp(gchar *outbuf, gint outlen, const gchar *inbuf);
145 static gint conv_ustodisp(gchar *outbuf, gint outlen, const gchar *inbuf);
146 static gint conv_noconv(gchar *outbuf, gint outlen, const gchar *inbuf);
147
148 static gboolean strict_mode = FALSE;
149
150 void codeconv_set_strict(gboolean mode)
151 {
152 strict_mode = mode;
153 }
154
155 static gint conv_jistoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
156 {
157 const guchar *in = inbuf;
158 gchar *out = outbuf;
159 JISState state = JIS_ASCII;
160
161 /*
162 * Loop outputs up to 3 bytes in each pass (aux kanji) and we
163 * need 1 byte to terminate the output
164 */
165 while (*in != '\0' && (out - outbuf) < outlen - 4) {
166 if (*in == ESC) {
167 in++;
168 if (*in == '$') {
169 if (*(in + 1) == '@' || *(in + 1) == 'B') {
170 state = JIS_KANJI;
171 in += 2;
172 } else if (*(in + 1) == '(' &&
173 *(in + 2) == 'D') {
174 state = JIS_AUXKANJI;
175 in += 3;
176 } else {
177 /* unknown escape sequence */
178 state = JIS_ASCII;
179 }
180 } else if (*in == '(') {
181 if (*(in + 1) == 'B' || *(in + 1) == 'J') {
182 state = JIS_ASCII;
183 in += 2;
184 } else if (*(in + 1) == 'I') {
185 state = JIS_HWKANA;
186 in += 2;
187 } else {
188 /* unknown escape sequence */
189 state = JIS_ASCII;
190 }
191 } else {
192 /* unknown escape sequence */
193 state = JIS_ASCII;
194 }
195 } else if (*in == 0x0e) {
196 state = JIS_HWKANA;
197 in++;
198 } else if (*in == 0x0f) {
199 state = JIS_ASCII;
200 in++;
201 } else {
202 switch (state) {
203 case JIS_ASCII:
204 *out++ = *in++;
205 break;
206 case JIS_KANJI:
207 *out++ = *in++ | 0x80;
208 if (*in == '\0') break;
209 *out++ = *in++ | 0x80;
210 break;
211 case JIS_HWKANA:
212 *out++ = 0x8e;
213 *out++ = *in++ | 0x80;
214 break;
215 case JIS_AUXKANJI:
216 *out++ = 0x8f;
217 *out++ = *in++ | 0x80;
218 if (*in == '\0') break;
219 *out++ = *in++ | 0x80;
220 break;
221 }
222 }
223 }
224
225 *out = '\0';
226 return 0;
227 }
228
229 #define JIS_HWDAKUTEN 0x5e
230 #define JIS_HWHANDAKUTEN 0x5f
231
232 static gint conv_jis_hantozen(guchar *outbuf, guchar jis_code, guchar sound_sym)
233 {
234 static guint16 h2z_tbl[] = {
235 /* 0x20 - 0x2f */
236 0x0000, 0x2123, 0x2156, 0x2157, 0x2122, 0x2126, 0x2572, 0x2521,
237 0x2523, 0x2525, 0x2527, 0x2529, 0x2563, 0x2565, 0x2567, 0x2543,
238 /* 0x30 - 0x3f */
239 0x213c, 0x2522, 0x2524, 0x2526, 0x2528, 0x252a, 0x252b, 0x252d,
240 0x252f, 0x2531, 0x2533, 0x2535, 0x2537, 0x2539, 0x253b, 0x253d,
241 /* 0x40 - 0x4f */
242 0x253f, 0x2541, 0x2544, 0x2546, 0x2548, 0x254a, 0x254b, 0x254c,
243 0x254d, 0x254e, 0x254f, 0x2552, 0x2555, 0x2558, 0x255b, 0x255e,
244 /* 0x50 - 0x5f */
245 0x255f, 0x2560, 0x2561, 0x2562, 0x2564, 0x2566, 0x2568, 0x2569,
246 0x256a, 0x256b, 0x256c, 0x256d, 0x256f, 0x2573, 0x212b, 0x212c
247 };
248
249 static guint16 dakuten_tbl[] = {
250 /* 0x30 - 0x3f */
251 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x252c, 0x252e,
252 0x2530, 0x2532, 0x2534, 0x2536, 0x2538, 0x253a, 0x253c, 0x253e,
253 /* 0x40 - 0x4f */
254 0x2540, 0x2542, 0x2545, 0x2547, 0x2549, 0x0000, 0x0000, 0x0000,
255 0x0000, 0x0000, 0x2550, 0x2553, 0x2556, 0x2559, 0x255c, 0x0000
256 };
257
258 static guint16 handakuten_tbl[] = {
259 /* 0x4a - 0x4e */
260 0x2551, 0x2554, 0x2557, 0x255a, 0x255d
261 };
262
263 guint16 out_code;
264
265 jis_code &= 0x7f;
266 sound_sym &= 0x7f;
267
268 if (jis_code < 0x21 || jis_code > 0x5f)
269 return 0;
270
271 if (sound_sym == JIS_HWDAKUTEN &&
272 jis_code >= 0x36 && jis_code <= 0x4e) {
273 out_code = dakuten_tbl[jis_code - 0x30];
274 if (out_code != 0) {
275 *outbuf = out_code >> 8;
276 *(outbuf + 1) = out_code & 0xff;
277 return 2;
278 }
279 }
280
281 if (sound_sym == JIS_HWHANDAKUTEN &&
282 jis_code >= 0x4a && jis_code <= 0x4e) {
283 out_code = handakuten_tbl[jis_code - 0x4a];
284 *outbuf = out_code >> 8;
285 *(outbuf + 1) = out_code & 0xff;
286 return 2;
287 }
288
289 out_code = h2z_tbl[jis_code - 0x20];
290 *outbuf = out_code >> 8;
291 *(outbuf + 1) = out_code & 0xff;
292 return 1;
293 }
294
295 static gint conv_euctojis(gchar *outbuf, gint outlen, const gchar *inbuf)
296 {
297 const guchar *in = inbuf;
298 gchar *out = outbuf;
299 JISState state = JIS_ASCII;
300
301 /*
302 * Loop outputs up to 6 bytes in each pass (aux shift + aux
303 * kanji) and we need up to 4 bytes to terminate the output
304 * (ASCII shift + null)
305 */
306 while (*in != '\0' && (out - outbuf) < outlen - 10) {
307 if (IS_ASCII(*in)) {
308 K_OUT();
309 *out++ = *in++;
310 } else if (iseuckanji(*in)) {
311 if (iseuckanji(*(in + 1))) {
312 K_IN();
313 *out++ = *in++ & 0x7f;
314 *out++ = *in++ & 0x7f;
315 } else {
316 K_OUT();
317 *out++ = SUBST_CHAR;
318 in++;
319 if (*in != '\0' && !IS_ASCII(*in)) {
320 *out++ = SUBST_CHAR;
321 in++;
322 }
323 }
324 } else if (iseuchwkana1(*in)) {
325 if (iseuchwkana2(*(in + 1))) {
326 if (prefs_common.allow_jisx0201_kana) {
327 HW_IN();
328 in++;
329 *out++ = *in++ & 0x7f;
330 } else {
331 guchar jis_ch[2];
332 gint len;
333
334 if (iseuchwkana1(*(in + 2)) &&
335 iseuchwkana2(*(in + 3)))
336 len = conv_jis_hantozen
337 (jis_ch,
338 *(in + 1), *(in + 3));
339 else
340 len = conv_jis_hantozen
341 (jis_ch,
342 *(in + 1), '\0');
343 if (len == 0)
344 in += 2;
345 else {
346 K_IN();
347 in += len * 2;
348 *out++ = jis_ch[0];
349 *out++ = jis_ch[1];
350 }
351 }
352 } else {
353 K_OUT();
354 in++;
355 if (*in != '\0' && !IS_ASCII(*in)) {
356 *out++ = SUBST_CHAR;
357 in++;
358 }
359 }
360 } else if (iseucaux(*in)) {
361 in++;
362 if (iseuckanji(*in) && iseuckanji(*(in + 1))) {
363 AUX_IN();
364 *out++ = *in++ & 0x7f;
365 *out++ = *in++ & 0x7f;
366 } else {
367 K_OUT();
368 if (*in != '\0' && !IS_ASCII(*in)) {
369 *out++ = SUBST_CHAR;
370 in++;
371 if (*in != '\0' && !IS_ASCII(*in)) {
372 *out++ = SUBST_CHAR;
373 in++;
374 }
375 }
376 }
377 } else {
378 K_OUT();
379 *out++ = SUBST_CHAR;
380 in++;
381 }
382 }
383
384 K_OUT();
385 *out = '\0';
386 return 0;
387 }
388
389 static gint conv_sjistoeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
390 {
391 const guchar *in = inbuf;
392 gchar *out = outbuf;
393
394 /*
395 * Loop outputs up to 2 bytes in each pass and we need 1 byte
396 * to terminate the output
397 */
398 while (*in != '\0' && (out - outbuf) < outlen - 3) {
399 if (IS_ASCII(*in)) {
400 *out++ = *in++;
401 } else if (issjiskanji1(*in)) {
402 if (issjiskanji2(*(in + 1))) {
403 guchar out1 = *in;
404 guchar out2 = *(in + 1);
405 guchar row;
406
407 row = out1 < 0xa0 ? 0x70 : 0xb0;
408 if (out2 < 0x9f) {
409 out1 = (out1 - row) * 2 - 1;
410 out2 -= out2 > 0x7f ? 0x20 : 0x1f;
411 } else {
412 out1 = (out1 - row) * 2;
413 out2 -= 0x7e;
414 }
415
416 *out++ = out1 | 0x80;
417 *out++ = out2 | 0x80;
418 in += 2;
419 } else {
420 *out++ = SUBST_CHAR;
421 in++;
422 if (*in != '\0' && !IS_ASCII(*in)) {
423 *out++ = SUBST_CHAR;
424 in++;
425 }
426 }
427 } else if (issjishwkana(*in)) {
428 *out++ = 0x8e;
429 *out++ = *in++;
430 } else {
431 *out++ = SUBST_CHAR;
432 in++;
433 }
434 }
435
436 *out = '\0';
437 return 0;
438 }
439
440 static gint conv_jistoutf8(gchar *outbuf, gint outlen, const gchar *inbuf)
441 {
442 gchar *eucstr;
443
444 Xalloca(eucstr, outlen, return -1);
445
446 if (conv_jistoeuc(eucstr, outlen, inbuf) <0)
447 return -1;
448 if (conv_euctoutf8(outbuf, outlen, eucstr) < 0)
449 return -1;
450 return 0;
451 }
452
453 static gint conv_sjistoutf8(gchar *outbuf, gint outlen, const gchar *inbuf)
454 {
455 gchar *tmpstr;
456
457 tmpstr = conv_iconv_strdup(inbuf, CS_SHIFT_JIS, CS_UTF_8);
458 if (tmpstr) {
459 strncpy2(outbuf, tmpstr, outlen);
460 g_free(tmpstr);
461 return 0;
462 } else {
463 strncpy2(outbuf, inbuf, outlen);
464 return -1;
465 }
466 }
467
468 static gint conv_euctoutf8(gchar *outbuf, gint outlen, const gchar *inbuf)
469 {
470 static iconv_t cd = (iconv_t)-1;
471 static gboolean iconv_ok = TRUE;
472 gchar *tmpstr;
473
474 if (cd == (iconv_t)-1) {
475 if (!iconv_ok) {
476 strncpy2(outbuf, inbuf, outlen);
477 return -1;
478 }
479 cd = iconv_open(CS_UTF_8, CS_EUC_JP_MS);
480 if (cd == (iconv_t)-1) {
481 cd = iconv_open(CS_UTF_8, CS_EUC_JP);
482 if (cd == (iconv_t)-1) {
483 g_warning("conv_euctoutf8(): %s",
484 g_strerror(errno));
485 iconv_ok = FALSE;
486 strncpy2(outbuf, inbuf, outlen);
487 return -1;
488 }
489 }
490 }
491
492 tmpstr = conv_iconv_strdup_with_cd(inbuf, cd);
493 if (tmpstr) {
494 strncpy2(outbuf, tmpstr, outlen);
495 g_free(tmpstr);
496 return 0;
497 } else {
498 strncpy2(outbuf, inbuf, outlen);
499 return -1;
500 }
501 }
502
503 static gint conv_anytoutf8(gchar *outbuf, gint outlen, const gchar *inbuf)
504 {
505 gint r = -1;
506 switch (conv_guess_ja_encoding(inbuf)) {
507 case C_ISO_2022_JP:
508 r = conv_jistoutf8(outbuf, outlen, inbuf);
509 break;
510 case C_SHIFT_JIS:
511 r = conv_sjistoutf8(outbuf, outlen, inbuf);
512 break;
513 case C_EUC_JP:
514 r = conv_euctoutf8(outbuf, outlen, inbuf);
515 break;
516 default:
517 r = 0;
518 strncpy2(outbuf, inbuf, outlen);
519 break;
520 }
521
522 return r;
523 }
524
525 static gint conv_utf8toeuc(gchar *outbuf, gint outlen, const gchar *inbuf)
526 {
527 static iconv_t cd = (iconv_t)-1;
528 static gboolean iconv_ok = TRUE;
529 gchar *tmpstr;
530
531 if (cd == (iconv_t)-1) {
532 if (!iconv_ok) {
533 strncpy2(outbuf, inbuf, outlen);
534 return -1;
535 }
536 cd = iconv_open(CS_EUC_JP_MS, CS_UTF_8);
537 if (cd == (iconv_t)-1) {
538 cd = iconv_open(CS_EUC_JP, CS_UTF_8);
539 if (cd == (iconv_t)-1) {
540 g_warning("conv_utf8toeuc(): %s",
541 g_strerror(errno));
542 iconv_ok = FALSE;
543 strncpy2(outbuf, inbuf, outlen);
544 return -1;
545 }
546 }
547 }
548
549 tmpstr = conv_iconv_strdup_with_cd(inbuf, cd);
550 if (tmpstr) {
551 strncpy2(outbuf, tmpstr, outlen);
552 g_free(tmpstr);
553 return 0;
554 } else {
555 strncpy2(outbuf, inbuf, outlen);
556 return -1;
557 }
558 }
559
560 static gint conv_utf8tojis(gchar *outbuf, gint outlen, const gchar *inbuf)
561 {
562 gchar *eucstr;
563
564 Xalloca(eucstr, outlen, return -1);
565
566 if (conv_utf8toeuc(eucstr, outlen, inbuf) < 0)
567 return -1;
568 if (conv_euctojis(outbuf, outlen, eucstr) < 0)
569 return -1;
570
571 return 0;
572 }
573
574 static void conv_unreadable_8bit(gchar *str)
575 {
576 register guchar *p = str;
577
578 while (*p != '\0') {
579 /* convert CR+LF -> LF */
580 if (*p == '\r' && *(p + 1) == '\n')
581 memmove(p, p + 1, strlen(p));
582 else if (!IS_ASCII(*p)) *p = SUBST_CHAR;
583 p++;
584 }
585 }
586
587 static CharSet conv_guess_ja_encoding(const gchar *str)
588 {
589 const guchar *p = str;
590 CharSet guessed = C_US_ASCII;
591
592 while (*p != '\0') {
593 if (*p == ESC && (*(p + 1) == '$' || *(p + 1) == '(')) {
594 if (guessed == C_US_ASCII)
595 return C_ISO_2022_JP;
596 p += 2;
597 } else if (IS_ASCII(*p)) {
598 p++;
599 } else if (iseuckanji(*p) && iseuckanji(*(p + 1))) {
600 if (*p >= 0xfd && *p <= 0xfe)
601 return C_EUC_JP;
602 else if (guessed == C_SHIFT_JIS) {
603 if ((issjiskanji1(*p) &&
604 issjiskanji2(*(p + 1))) ||
605 issjishwkana(*p))
606 guessed = C_SHIFT_JIS;
607 else
608 guessed = C_EUC_JP;
609 } else
610 guessed = C_EUC_JP;
611 p += 2;
612 } else if (issjiskanji1(*p) && issjiskanji2(*(p + 1))) {
613 if (iseuchwkana1(*p) && iseuchwkana2(*(p + 1)))
614 guessed = C_SHIFT_JIS;
615 else
616 return C_SHIFT_JIS;
617 p += 2;
618 } else if (issjishwkana(*p)) {
619 guessed = C_SHIFT_JIS;
620 p++;
621 } else {
622 p++;
623 }
624 }
625
626 return guessed;
627 }
628
629 static gint conv_jistodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
630 {
631 return conv_jistoutf8(outbuf, outlen, inbuf);
632 }
633
634 static gint conv_sjistodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
635 {
636 return conv_sjistoutf8(outbuf, outlen, inbuf);
637 }
638
639 static gint conv_euctodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
640 {
641 return conv_euctoutf8(outbuf, outlen, inbuf);
642 }
643
644 void conv_utf8todisp(gchar *outbuf, gint outlen, const gchar *inbuf)
645 {
646 if (g_utf8_validate(inbuf, -1, NULL) == TRUE)
647 strncpy2(outbuf, inbuf, outlen);
648 else
649 conv_ustodisp(outbuf, outlen, inbuf);
650 }
651
652 static gint conv_anytodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
653 {
654 gint r = 0;
655 if (conv_anytoutf8(outbuf, outlen, inbuf) < 0)
656 r = -1;
657 if (g_utf8_validate(outbuf, -1, NULL) != TRUE)
658 conv_unreadable_8bit(outbuf);
659 return r;
660 }
661
662 static gint conv_ustodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
663 {
664 strncpy2(outbuf, inbuf, outlen);
665 conv_unreadable_8bit(outbuf);
666
667 return 0;
668 }
669
670 void conv_localetodisp(gchar *outbuf, gint outlen, const gchar *inbuf)
671 {
672 gchar *tmpstr;
673
674 codeconv_set_strict(TRUE);
675 tmpstr = conv_iconv_strdup(inbuf, conv_get_locale_charset_str(),
676 CS_INTERNAL);
677 codeconv_set_strict(FALSE);
678 if (tmpstr && g_utf8_validate(tmpstr, -1, NULL)) {
679 strncpy2(outbuf, tmpstr, outlen);
680 g_free(tmpstr);
681 return;
682 } else if (tmpstr && !g_utf8_validate(tmpstr, -1, NULL)) {
683 g_free(tmpstr);
684 codeconv_set_strict(TRUE);
685 tmpstr = conv_iconv_strdup(inbuf,
686 conv_get_locale_charset_str_no_utf8(),
687 CS_INTERNAL);
688 codeconv_set_strict(FALSE);
689 }
690 if (tmpstr && g_utf8_validate(tmpstr, -1, NULL)) {
691 strncpy2(outbuf, tmpstr, outlen);
692 g_free(tmpstr);
693 return;
694 } else {
695 g_free(tmpstr);
696 conv_utf8todisp(outbuf, outlen, inbuf);
697 }
698 }
699
700 static gint conv_noconv(gchar *outbuf, gint outlen, const gchar *inbuf)
701 {
702 strncpy2(outbuf, inbuf, outlen);
703 return 0;
704 }
705
706 static const gchar *
707 conv_get_fallback_for_private_encoding(const gchar *encoding)
708 {
709 if (encoding && (encoding[0] == 'X' || encoding[0] == 'x') &&
710 encoding[1] == '-') {
711 if (!g_ascii_strcasecmp(encoding, CS_X_MACCYR))
712 return CS_MACCYR;
713 if (!g_ascii_strcasecmp(encoding, CS_X_GBK))
714 return CS_GBK;
715 }
716
717 return encoding;
718 }
719
720 CodeConverter *conv_code_converter_new(const gchar *src_charset)
721 {
722 CodeConverter *conv;
723
724 src_charset = conv_get_fallback_for_private_encoding(src_charset);
725
726 conv = g_new0(CodeConverter, 1);
727 conv->code_conv_func = conv_get_code_conv_func(src_charset, NULL);
728 conv->charset_str = g_strdup(src_charset);
729 conv->charset = conv_get_charset_from_str(src_charset);
730
731 return conv;
732 }
733
734 void conv_code_converter_destroy(CodeConverter *conv)
735 {
736 g_free(conv->charset_str);
737 g_free(conv);
738 }
739
740 gint conv_convert(CodeConverter *conv, gchar *outbuf, gint outlen,
741 const gchar *inbuf)
742 {
743 if (conv->code_conv_func != conv_noconv)
744 return conv->code_conv_func(outbuf, outlen, inbuf);
745 else {
746 gchar *str;
747
748 str = conv_iconv_strdup(inbuf, conv->charset_str, NULL);
749 if (!str)
750 return -1;
751 else {
752 strncpy2(outbuf, str, outlen);
753 g_free(str);
754 }
755 }
756
757 return 0;
758 }
759
760 gchar *conv_codeset_strdup(const gchar *inbuf,
761 const gchar *src_code, const gchar *dest_code)
762 {
763 gchar *buf;
764 size_t len;
765 CodeConvFunc conv_func;
766
767 if (!strcmp2(src_code, dest_code)) {
768 CharSet dest_charset = conv_get_charset_from_str(dest_code);
769 if (strict_mode && dest_charset == C_UTF_8) {
770 /* ensure valid UTF-8 if target is UTF-8 */
771 if (!g_utf8_validate(inbuf, -1, NULL)) {
772 return NULL;
773 }
774 }
775 /* otherwise, try for a lucky day */
776 return g_strdup(inbuf);
777 }
778
779 src_code = conv_get_fallback_for_private_encoding(src_code);
780 conv_func = conv_get_code_conv_func(src_code, dest_code);
781 if (conv_func == conv_ustodisp && strict_mode && !is_ascii_str(inbuf))
782 return NULL;
783
784 if (conv_func != conv_noconv) {
785 len = (strlen(inbuf) + 1) * 3;
786 buf = g_malloc(len);
787 if (!buf) return NULL;
788
789 if (conv_func(buf, len, inbuf) == 0 || !strict_mode)
790 return g_realloc(buf, strlen(buf) + 1);
791 else {
792 g_free(buf);
793 return NULL;
794 }
795 }
796
797 return conv_iconv_strdup(inbuf, src_code, dest_code);
798 }
799
800 static CodeConvFunc conv_get_code_conv_func(const gchar *src_charset_str,
801 const gchar *dest_charset_str)
802 {
803 CodeConvFunc code_conv = conv_noconv;
804 CharSet src_charset;
805 CharSet dest_charset;
806
807 if (!src_charset_str)
808 src_charset = conv_get_locale_charset();
809 else
810 src_charset = conv_get_charset_from_str(src_charset_str);
811
812 /* auto detection mode */
813 if (!src_charset_str && !dest_charset_str) {
814 if (conv_is_ja_locale())
815 return conv_anytodisp;
816 else
817 return conv_noconv;
818 }
819
820 dest_charset = conv_get_charset_from_str(dest_charset_str);
821
822 if (dest_charset == C_US_ASCII)
823 return conv_ustodisp;
824
825 switch (src_charset) {
826 case C_US_ASCII:
827 case C_ISO_8859_1:
828 case C_ISO_8859_2:
829 case C_ISO_8859_3:
830 case C_ISO_8859_4:
831 case C_ISO_8859_5:
832 case C_ISO_8859_6:
833 case C_ISO_8859_7:
834 case C_ISO_8859_8:
835 case C_ISO_8859_9:
836 case C_ISO_8859_10:
837 case C_ISO_8859_11:
838 case C_ISO_8859_13:
839 case C_ISO_8859_14:
840 case C_ISO_8859_15:
841 break;
842 case C_ISO_2022_JP:
843 case C_ISO_2022_JP_2:
844 case C_ISO_2022_JP_3:
845 if (dest_charset == C_AUTO)
846 code_conv = conv_jistodisp;
847 else if (dest_charset == C_EUC_JP)
848 code_conv = conv_jistoeuc;
849 else if (dest_charset == C_UTF_8)
850 code_conv = conv_jistoutf8;
851 break;
852 case C_SHIFT_JIS:
853 if (dest_charset == C_AUTO)
854 code_conv = conv_sjistodisp;
855 else if (dest_charset == C_EUC_JP)
856 code_conv = conv_sjistoeuc;
857 else if (dest_charset == C_UTF_8)
858 code_conv = conv_sjistoutf8;
859 break;
860 case C_EUC_JP:
861 if (dest_charset == C_AUTO)
862 code_conv = conv_euctodisp;
863 else if (dest_charset == C_ISO_2022_JP ||
864 dest_charset == C_ISO_2022_JP_2 ||
865 dest_charset == C_ISO_2022_JP_3)
866 code_conv = conv_euctojis;
867 else if (dest_charset == C_UTF_8)
868 code_conv = conv_euctoutf8;
869 break;
870 case C_UTF_8:
871 if (dest_charset == C_EUC_JP)
872 code_conv = conv_utf8toeuc;
873 else if (dest_charset == C_ISO_2022_JP ||
874 dest_charset == C_ISO_2022_JP_2 ||
875 dest_charset == C_ISO_2022_JP_3)
876 code_conv = conv_utf8tojis;
877 break;
878 default:
879 break;
880 }
881
882 return code_conv;
883 }
884
885 static gchar *conv_iconv_strdup(const gchar *inbuf,
886 const gchar *src_code, const gchar *dest_code)
887 {
888 iconv_t cd;
889 gchar *outbuf;
890
891 if (!src_code && !dest_code &&
892 g_utf8_validate(inbuf, -1, NULL))
893 return g_strdup(inbuf);
894
895 if (!src_code)
896 src_code = conv_get_outgoing_charset_str();
897 if (!dest_code)
898 dest_code = CS_INTERNAL;
899
900 /* don't convert if src and dest codeset are identical */
901 if (!strcasecmp(src_code, dest_code))
902 return g_strdup(inbuf);
903
904 /* don't convert if dest codeset is US-ASCII */
905 if (!strcasecmp(src_code, CS_US_ASCII))
906 return g_strdup(inbuf);
907
908 /* don't convert if dest codeset is US-ASCII */
909 if (!strcasecmp(dest_code, CS_US_ASCII))
910 return g_strdup(inbuf);
911
912 cd = iconv_open(dest_code, src_code);
913 if (cd == (iconv_t)-1)
914 return NULL;
915
916 outbuf = conv_iconv_strdup_with_cd(inbuf, cd);
917
918 iconv_close(cd);
919
920 return outbuf;
921 }
922
923 gchar *conv_iconv_strdup_with_cd(const gchar *inbuf, iconv_t cd)
924 {
925 const gchar *inbuf_p;
926 gchar *outbuf;
927 gchar *outbuf_p;
928 size_t in_size;
929 size_t in_left;
930 size_t out_size;
931 size_t out_left;
932 size_t n_conv;
933 size_t len;
934
935 inbuf_p = inbuf;
936 in_size = strlen(inbuf);
937 in_left = in_size;
938 out_size = (in_size + 1) * 2;
939 outbuf = g_malloc(out_size);
940 outbuf_p = outbuf;
941 out_left = out_size;
942
943 #define EXPAND_BUF() \
944 { \
945 len = outbuf_p - outbuf; \
946 out_size *= 2; \
947 outbuf = g_realloc(outbuf, out_size); \
948 outbuf_p = outbuf + len; \
949 out_left = out_size - len; \
950 }
951
952 while ((n_conv = iconv(cd, (ICONV_CONST gchar **)&inbuf_p, &in_left,
953 &outbuf_p, &out_left)) == (size_t)-1) {
954 if (EILSEQ == errno) {
955 if (strict_mode) {
956 g_free(outbuf);
957 return NULL;
958 }
959 //g_print("iconv(): at %d: %s\n", in_size - in_left, g_strerror(errno));
960 inbuf_p++;
961 in_left--;
962 if (out_left == 0) {
963 EXPAND_BUF();
964 }
965 *outbuf_p++ = SUBST_CHAR;
966 out_left--;
967 } else if (EINVAL == errno) {
968 break;
969 } else if (E2BIG == errno) {
970 EXPAND_BUF();
971 } else {
972 g_warning("conv_iconv_strdup(): %s",
973 g_strerror(errno));
974 break;
975 }
976 }
977
978 while ((n_conv = iconv(cd, NULL, NULL, &outbuf_p, &out_left)) ==
979 (size_t)-1) {
980 if (E2BIG == errno) {
981 EXPAND_BUF();
982 } else {
983 g_warning("conv_iconv_strdup(): %s",
984 g_strerror(errno));
985 break;
986 }
987 }
988
989 #undef EXPAND_BUF
990
991 len = outbuf_p - outbuf;
992 outbuf = g_realloc(outbuf, len + 1);
993 outbuf[len] = '\0';
994
995 return outbuf;
996 }
997
998 static const struct {
999 CharSet charset;
1000 gchar *const name;
1001 } charsets[] = {
1002 {C_US_ASCII, CS_US_ASCII},
1003 {C_US_ASCII, CS_ANSI_X3_4_1968},
1004 {C_UTF_8, CS_UTF_8},
1005 {C_UTF_7, CS_UTF_7},
1006 {C_ISO_8859_1, CS_ISO_8859_1},
1007 {C_ISO_8859_2, CS_ISO_8859_2},
1008 {C_ISO_8859_3, CS_ISO_8859_3},
1009 {C_ISO_8859_4, CS_ISO_8859_4},
1010 {C_ISO_8859_5, CS_ISO_8859_5},
1011 {C_ISO_8859_6, CS_ISO_8859_6},
1012 {C_ISO_8859_7, CS_ISO_8859_7},
1013 {C_ISO_8859_8, CS_ISO_8859_8},
1014 {C_ISO_8859_9, CS_ISO_8859_9},
1015 {C_ISO_8859_10, CS_ISO_8859_10},
1016 {C_ISO_8859_11, CS_ISO_8859_11},
1017 {C_ISO_8859_13, CS_ISO_8859_13},
1018 {C_ISO_8859_14, CS_ISO_8859_14},
1019 {C_ISO_8859_15, CS_ISO_8859_15},
1020 {C_BALTIC, CS_BALTIC},
1021 {C_CP1250, CS_CP1250},
1022 {C_CP1251, CS_CP1251},
1023 {C_CP1252, CS_CP1252},
1024 {C_CP1253, CS_CP1253},
1025 {C_CP1254, CS_CP1254},
1026 {C_CP1255, CS_CP1255},
1027 {C_CP1256, CS_CP1256},
1028 {C_CP1257, CS_CP1257},
1029 {C_CP1258, CS_CP1258},
1030 {C_WINDOWS_1250, CS_WINDOWS_1250},
1031 {C_WINDOWS_1251, CS_WINDOWS_1251},
1032 {C_WINDOWS_1252, CS_WINDOWS_1252},
1033 {C_WINDOWS_1253, CS_WINDOWS_1253},
1034 {C_WINDOWS_1254, CS_WINDOWS_1254},
1035 {C_WINDOWS_1255, CS_WINDOWS_1255},
1036 {C_WINDOWS_1256, CS_WINDOWS_1256},
1037 {C_WINDOWS_1257, CS_WINDOWS_1257},
1038 {C_WINDOWS_1258, CS_WINDOWS_1258},
1039 {C_KOI8_R, CS_KOI8_R},
1040 {C_MACCYR, CS_MACCYR},
1041 {C_KOI8_T, CS_KOI8_T},
1042 {C_KOI8_U, CS_KOI8_U},
1043 {C_ISO_2022_JP, CS_ISO_2022_JP},
1044 {C_ISO_2022_JP_2, CS_ISO_2022_JP_2},
1045 {C_ISO_2022_JP_3, CS_ISO_2022_JP_3},
1046 {C_EUC_JP, CS_EUC_JP},
1047 {C_EUC_JP, CS_EUCJP},
1048 {C_EUC_JP_MS, CS_EUC_JP_MS},
1049 {C_SHIFT_JIS, CS_SHIFT_JIS},
1050 {C_SHIFT_JIS, CS_SHIFT__JIS},
1051 {C_SHIFT_JIS, CS_SJIS},
1052 {C_ISO_2022_KR, CS_ISO_2022_KR},
1053 {C_EUC_KR, CS_EUC_KR},
1054 {C_ISO_2022_CN, CS_ISO_2022_CN},
1055 {C_EUC_CN, CS_EUC_CN},
1056 {C_GB18030, CS_GB18030},
1057 {C_GB2312, CS_GB2312},
1058 {C_GBK, CS_GBK},
1059 {C_EUC_TW, CS_EUC_TW},
1060 {C_BIG5, CS_BIG5},
1061 {C_BIG5_HKSCS, CS_BIG5_HKSCS},
1062 {C_TIS_620, CS_TIS_620},
1063 {C_WINDOWS_874, CS_WINDOWS_874},
1064 {C_GEORGIAN_PS, CS_GEORGIAN_PS},
1065 {C_TCVN5712_1, CS_TCVN5712_1},
1066 };
1067
1068 static const struct {
1069 gchar *const locale;
1070 CharSet charset;
1071 CharSet out_charset;
1072 } locale_table[] = {
1073 {"ja_JP.eucJP" , C_EUC_JP , C_ISO_2022_JP},
1074 {"ja_JP.EUC-JP" , C_EUC_JP , C_ISO_2022_JP},
1075 {"ja_JP.EUC" , C_EUC_JP , C_ISO_2022_JP},
1076 {"ja_JP.ujis" , C_EUC_JP , C_ISO_2022_JP},
1077 {"ja_JP.SJIS" , C_SHIFT_JIS , C_ISO_2022_JP},
1078 {"ja_JP.JIS" , C_ISO_2022_JP , C_ISO_2022_JP},
1079 #ifdef G_OS_WIN32
1080 {"ja_JP" , C_SHIFT_JIS , C_ISO_2022_JP},
1081 #else
1082 {"ja_JP" , C_EUC_JP , C_ISO_2022_JP},
1083 #endif
1084 {"ko_KR.EUC-KR" , C_EUC_KR , C_EUC_KR},
1085 {"ko_KR" , C_EUC_KR , C_EUC_KR},
1086 {"zh_CN.GB18030" , C_GB18030 , C_GB18030},
1087 {"zh_CN.GB2312" , C_GB2312 , C_GB2312},
1088 {"zh_CN.GBK" , C_GBK , C_GBK},
1089 {"zh_CN" , C_GB18030 , C_GB18030},
1090 {"zh_HK" , C_BIG5_HKSCS , C_BIG5_HKSCS},
1091 {"zh_TW.eucTW" , C_EUC_TW , C_BIG5},
1092 {"zh_TW.EUC-TW" , C_EUC_TW , C_BIG5},
1093 {"zh_TW.Big5" , C_BIG5 , C_BIG5},
1094 {"zh_TW" , C_BIG5 , C_BIG5},
1095
1096 {"ru_RU.KOI8-R" , C_KOI8_R , C_KOI8_R},
1097 {"ru_RU.KOI8R" , C_KOI8_R , C_KOI8_R},
1098 {"ru_RU.CP1251" , C_WINDOWS_1251, C_KOI8_R},
1099 #ifdef G_OS_WIN32
1100 {"ru_RU" , C_WINDOWS_1251, C_KOI8_R},
1101 #else
1102 {"ru_RU" , C_ISO_8859_5 , C_KOI8_R},
1103 #endif
1104 {"tg_TJ" , C_KOI8_T , C_KOI8_T},
1105 {"ru_UA" , C_KOI8_U , C_KOI8_U},
1106 {"uk_UA.CP1251" , C_WINDOWS_1251, C_KOI8_U},
1107 {"uk_UA" , C_KOI8_U , C_KOI8_U},
1108
1109 {"be_BY" , C_WINDOWS_1251, C_WINDOWS_1251},
1110 {"bg_BG" , C_WINDOWS_1251, C_WINDOWS_1251},
1111
1112 {"yi_US" , C_WINDOWS_1255, C_WINDOWS_1255},
1113
1114 {"af_ZA" , C_ISO_8859_1 , C_ISO_8859_1},
1115 {"br_FR" , C_ISO_8859_1 , C_ISO_8859_1},
1116 {"ca_ES" , C_ISO_8859_1 , C_ISO_8859_1},
1117 {"da_DK" , C_ISO_8859_1 , C_ISO_8859_1},
1118 {"de_AT" , C_ISO_8859_1 , C_ISO_8859_1},
1119 {"de_BE" , C_ISO_8859_1 , C_ISO_8859_1},
1120 {"de_CH" , C_ISO_8859_1 , C_ISO_8859_1},
1121 {"de_DE" , C_ISO_8859_1 , C_ISO_8859_1},
1122 {"de_LU" , C_ISO_8859_1 , C_ISO_8859_1},
1123 {"en_AU" , C_ISO_8859_1 , C_ISO_8859_1},
1124 {"en_BW" , C_ISO_8859_1 , C_ISO_8859_1},
1125 {"en_CA" , C_ISO_8859_1 , C_ISO_8859_1},
1126 {"en_DK" , C_ISO_8859_1 , C_ISO_8859_1},
1127 {"en_GB" , C_ISO_8859_1 , C_ISO_8859_1},
1128 {"en_HK" , C_ISO_8859_1 , C_ISO_8859_1},
1129 {"en_IE" , C_ISO_8859_1 , C_ISO_8859_1},
1130 {"en_NZ" , C_ISO_8859_1 , C_ISO_8859_1},
1131 {"en_PH" , C_ISO_8859_1 , C_ISO_8859_1},
1132 {"en_SG" , C_ISO_8859_1 , C_ISO_8859_1},
1133 {"en_US" , C_ISO_8859_1 , C_ISO_8859_1},
1134 {"en_ZA" , C_ISO_8859_1 , C_ISO_8859_1},
1135 {"en_ZW" , C_ISO_8859_1 , C_ISO_8859_1},
1136 {"es_AR" , C_ISO_8859_1 , C_ISO_8859_1},
1137 {"es_BO" , C_ISO_8859_1 , C_ISO_8859_1},
1138 {"es_CL" , C_ISO_8859_1 , C_ISO_8859_1},
1139 {"es_CO" , C_ISO_8859_1 , C_ISO_8859_1},
1140 {"es_CR" , C_ISO_8859_1 , C_ISO_8859_1},
1141 {"es_DO" , C_ISO_8859_1 , C_ISO_8859_1},
1142 {"es_EC" , C_ISO_8859_1 , C_ISO_8859_1},
1143 {"es_ES" , C_ISO_8859_1 , C_ISO_8859_1},
1144 {"es_GT" , C_ISO_8859_1 , C_ISO_8859_1},
1145 {"es_HN" , C_ISO_8859_1 , C_ISO_8859_1},
1146 {"es_MX" , C_ISO_8859_1 , C_ISO_8859_1},
1147 {"es_NI" , C_ISO_8859_1 , C_ISO_8859_1},
1148 {"es_PA" , C_ISO_8859_1 , C_ISO_8859_1},
1149 {"es_PE" , C_ISO_8859_1 , C_ISO_8859_1},
1150 {"es_PR" , C_ISO_8859_1 , C_ISO_8859_1},
1151 {"es_PY" , C_ISO_8859_1 , C_ISO_8859_1},
1152 {"es_SV" , C_ISO_8859_1 , C_ISO_8859_1},
1153 {"es_US" , C_ISO_8859_1 , C_ISO_8859_1},
1154 {"es_UY" , C_ISO_8859_1 , C_ISO_8859_1},
1155 {"es_VE" , C_ISO_8859_1 , C_ISO_8859_1},
1156 {"et_EE" , C_ISO_8859_1 , C_ISO_8859_1},
1157 {"eu_ES" , C_ISO_8859_1 , C_ISO_8859_1},
1158 {"fi_FI" , C_ISO_8859_1 , C_ISO_8859_1},
1159 {"fo_FO" , C_ISO_8859_1 , C_ISO_8859_1},
1160 {"fr_BE" , C_ISO_8859_1 , C_ISO_8859_1},
1161 {"fr_CA" , C_ISO_8859_1 , C_ISO_8859_1},
1162 {"fr_CH" , C_ISO_8859_1 , C_ISO_8859_1},
1163 {"fr_FR" , C_ISO_8859_1 , C_ISO_8859_1},
1164 {"fr_LU" , C_ISO_8859_1 , C_ISO_8859_1},
1165 {"ga_IE" , C_ISO_8859_1 , C_ISO_8859_1},
1166 {"gl_ES" , C_ISO_8859_1 , C_ISO_8859_1},
1167 {"gv_GB" , C_ISO_8859_1 , C_ISO_8859_1},
1168 {"id_ID" , C_ISO_8859_1 , C_ISO_8859_1},
1169 {"is_IS" , C_ISO_8859_1 , C_ISO_8859_1},
1170 {"it_CH" , C_ISO_8859_1 , C_ISO_8859_1},
1171 {"it_IT" , C_ISO_8859_1 , C_ISO_8859_1},
1172 {"kl_GL" , C_ISO_8859_1 , C_ISO_8859_1},
1173 {"kw_GB" , C_ISO_8859_1 , C_ISO_8859_1},
1174 {"ms_MY" , C_ISO_8859_1 , C_ISO_8859_1},
1175 {"nl_BE" , C_ISO_8859_1 , C_ISO_8859_1},
1176 {"nl_NL" , C_ISO_8859_1 , C_ISO_8859_1},
1177 {"nb_NO" , C_ISO_8859_1 , C_ISO_8859_1},
1178 {"nn_NO" , C_ISO_8859_1 , C_ISO_8859_1},
1179 {"no_NO" , C_ISO_8859_1 , C_ISO_8859_1},
1180 {"oc_FR" , C_ISO_8859_1 , C_ISO_8859_1},
1181 {"pt_BR" , C_ISO_8859_1 , C_ISO_8859_1},
1182 {"pt_PT" , C_ISO_8859_1 , C_ISO_8859_1},
1183 {"sq_AL" , C_ISO_8859_1 , C_ISO_8859_1},
1184 {"sv_FI" , C_ISO_8859_1 , C_ISO_8859_1},
1185 {"sv_SE" , C_ISO_8859_1 , C_ISO_8859_1},
1186 {"tl_PH" , C_ISO_8859_1 , C_ISO_8859_1},
1187 {"uz_UZ" , C_ISO_8859_1 , C_ISO_8859_1},
1188 {"wa_BE" , C_ISO_8859_1 , C_ISO_8859_1},
1189
1190 {"bs_BA" , C_ISO_8859_2 , C_ISO_8859_2},
1191 {"cs_CZ" , C_ISO_8859_2 , C_ISO_8859_2},
1192 {"hr_HR" , C_ISO_8859_2 , C_ISO_8859_2},
1193 {"hu_HU" , C_ISO_8859_2 , C_ISO_8859_2},
1194 {"pl_PL" , C_ISO_8859_2 , C_ISO_8859_2},
1195 {"ro_RO" , C_ISO_8859_2 , C_ISO_8859_2},
1196 {"sk_SK" , C_ISO_8859_2 , C_ISO_8859_2},
1197 {"sl_SI" , C_ISO_8859_2 , C_ISO_8859_2},
1198
1199 {"sr_YU@cyrillic" , C_ISO_8859_5 , C_ISO_8859_5},
1200 {"sr_YU" , C_ISO_8859_2 , C_ISO_8859_2},
1201
1202 {"mt_MT" , C_ISO_8859_3 , C_ISO_8859_3},
1203
1204 {"lt_LT.iso88594" , C_ISO_8859_4 , C_ISO_8859_4},
1205 {"lt_LT.ISO8859-4" , C_ISO_8859_4 , C_ISO_8859_4},
1206 {"lt_LT.ISO_8859-4" , C_ISO_8859_4 , C_ISO_8859_4},
1207 {"lt_LT" , C_ISO_8859_13 , C_ISO_8859_13},
1208
1209 {"mk_MK" , C_ISO_8859_5 , C_ISO_8859_5},
1210
1211 {"ar_AE" , C_ISO_8859_6 , C_ISO_8859_6},
1212 {"ar_BH" , C_ISO_8859_6 , C_ISO_8859_6},
1213 {"ar_DZ" , C_ISO_8859_6 , C_ISO_8859_6},
1214 {"ar_EG" , C_ISO_8859_6 , C_ISO_8859_6},
1215 {"ar_IQ" , C_ISO_8859_6 , C_ISO_8859_6},
1216 {"ar_JO" , C_ISO_8859_6 , C_ISO_8859_6},
1217 {"ar_KW" , C_ISO_8859_6 , C_ISO_8859_6},
1218 {"ar_LB" , C_ISO_8859_6 , C_ISO_8859_6},
1219 {"ar_LY" , C_ISO_8859_6 , C_ISO_8859_6},
1220 {"ar_MA" , C_ISO_8859_6 , C_ISO_8859_6},
1221 {"ar_OM" , C_ISO_8859_6 , C_ISO_8859_6},
1222 {"ar_QA" , C_ISO_8859_6 , C_ISO_8859_6},
1223 {"ar_SA" , C_ISO_8859_6 , C_ISO_8859_6},
1224 {"ar_SD" , C_ISO_8859_6 , C_ISO_8859_6},
1225 {"ar_SY" , C_ISO_8859_6 , C_ISO_8859_6},
1226 {"ar_TN" , C_ISO_8859_6 , C_ISO_8859_6},
1227 {"ar_YE" , C_ISO_8859_6 , C_ISO_8859_6},
1228
1229 {"el_GR" , C_ISO_8859_7 , C_ISO_8859_7},
1230 {"he_IL" , C_ISO_8859_8 , C_ISO_8859_8},
1231 {"iw_IL" , C_ISO_8859_8 , C_ISO_8859_8},
1232 {"tr_TR" , C_ISO_8859_9 , C_ISO_8859_9},
1233
1234 {"lv_LV" , C_ISO_8859_13 , C_ISO_8859_13},
1235 {"mi_NZ" , C_ISO_8859_13 , C_ISO_8859_13},
1236
1237 {"cy_GB" , C_ISO_8859_14 , C_ISO_8859_14},
1238
1239 {"ar_IN" , C_UTF_8 , C_UTF_8},
1240 {"en_IN" , C_UTF_8 , C_UTF_8},
1241 {"se_NO" , C_UTF_8 , C_UTF_8},
1242 {"ta_IN" , C_UTF_8 , C_UTF_8},
1243 {"te_IN" , C_UTF_8 , C_UTF_8},
1244 {"ur_PK" , C_UTF_8 , C_UTF_8},
1245
1246 {"th_TH" , C_TIS_620 , C_TIS_620},
1247 /* {"th_TH" , C_WINDOWS_874}, */
1248 /* {"th_TH" , C_ISO_8859_11}, */
1249
1250 {"ka_GE" , C_GEORGIAN_PS , C_GEORGIAN_PS},
1251 {"vi_VN.TCVN" , C_TCVN5712_1 , C_TCVN5712_1},
1252
1253 {"C" , C_US_ASCII , C_US_ASCII},
1254 {"POSIX" , C_US_ASCII , C_US_ASCII},
1255 {"ANSI_X3.4-1968" , C_US_ASCII , C_US_ASCII},
1256 };
1257
1258 static GHashTable *conv_get_charset_to_str_table(void)
1259 {
1260 static GHashTable *table;
1261 gint i;
1262
1263 if (table)
1264 return table;
1265
1266 table = g_hash_table_new(NULL, g_direct_equal);
1267
1268 for (i = 0; i < sizeof(charsets) / sizeof(charsets[0]); i++) {
1269 if (g_hash_table_lookup(table, GUINT_TO_POINTER(charsets[i].charset))
1270 == NULL) {
1271 g_hash_table_insert
1272 (table, GUINT_TO_POINTER(charsets[i].charset),
1273 charsets[i].name);
1274 }
1275 }
1276
1277 return table;
1278 }
1279
1280 static GHashTable *conv_get_charset_from_str_table(void)
1281 {
1282 static GHashTable *table;
1283 gint i;
1284
1285 if (table)
1286 return table;
1287
1288 table = g_hash_table_new(str_case_hash, str_case_equal);
1289
1290 for (i = 0; i < sizeof(charsets) / sizeof(charsets[0]); i++) {
1291 g_hash_table_insert(table, charsets[i].name,
1292 GUINT_TO_POINTER(charsets[i].charset));
1293 }
1294
1295 return table;
1296 }
1297
1298 const gchar *conv_get_charset_str(CharSet charset)
1299 {
1300 GHashTable *table;
1301
1302 table = conv_get_charset_to_str_table();
1303 return g_hash_table_lookup(table, GUINT_TO_POINTER(charset));
1304 }
1305
1306 CharSet conv_get_charset_from_str(const gchar *charset)
1307 {
1308 GHashTable *table;
1309
1310 if (!charset) return C_AUTO;
1311
1312 table = conv_get_charset_from_str_table();
1313 return GPOINTER_TO_UINT(g_hash_table_lookup(table, charset));
1314 }
1315
1316 static CharSet conv_get_locale_charset(void)
1317 {
1318 static CharSet cur_charset = -1;
1319 const gchar *cur_locale;
1320 const gchar *p;
1321 gint i;
1322
1323 if (cur_charset != -1)
1324 return cur_charset;
1325
1326 cur_locale = conv_get_current_locale();
1327 if (!cur_locale) {
1328 cur_charset = C_US_ASCII;
1329 return cur_charset;
1330 }
1331
1332 if (strcasestr(cur_locale, "UTF-8") ||
1333 strcasestr(cur_locale, "utf8")) {
1334 cur_charset = C_UTF_8;
1335 return cur_charset;
1336 }
1337
1338 if ((p = strcasestr(cur_locale, "@euro")) && p[5] == '\0') {
1339 cur_charset = C_ISO_8859_15;
1340 return cur_charset;
1341 }
1342
1343 for (i = 0; i < sizeof(locale_table) / sizeof(locale_table[0]); i++) {
1344 const gchar *p;
1345
1346 /* "ja_JP.EUC" matches with "ja_JP.eucJP", "ja_JP.EUC" and
1347 "ja_JP". "ja_JP" matches with "ja_JP.xxxx" and "ja" */
1348 if (!g_ascii_strncasecmp(cur_locale, locale_table[i].locale,
1349 strlen(locale_table[i].locale))) {
1350 cur_charset = locale_table[i].charset;
1351 return cur_charset;
1352 } else if ((p = strchr(locale_table[i].locale, '_')) &&
1353 !strchr(p + 1, '.')) {
1354 if (strlen(cur_locale) == 2 &&
1355 !g_ascii_strncasecmp(cur_locale, locale_table[i].locale, 2)) {
1356 cur_charset = locale_table[i].charset;
1357 return cur_charset;
1358 }
1359 }
1360 }
1361
1362 cur_charset = C_AUTO;
1363 return cur_charset;
1364 }
1365
1366 static CharSet conv_get_locale_charset_no_utf8(void)
1367 {
1368 static CharSet cur_charset = -1;
1369 const gchar *cur_locale;
1370 const gchar *p;
1371 gint i;
1372
1373 if (prefs_common.broken_are_utf8) {
1374 cur_charset = C_UTF_8;
1375 return cur_charset;
1376 }
1377
1378 cur_locale = conv_get_current_locale();
1379 if (!cur_locale) {
1380 cur_charset = C_US_ASCII;
1381 return cur_charset;
1382 }
1383
1384 if (strcasestr(cur_locale, "UTF-8") ||
1385 strcasestr(cur_locale, "utf8")) {
1386 cur_charset = C_UTF_8;
1387 return cur_charset;
1388 }
1389
1390 if ((p = strcasestr(cur_locale, "@euro")) && p[5] == '\0') {
1391 cur_charset = C_ISO_8859_15;
1392 return cur_charset;
1393 }
1394
1395 for (i = 0; i < sizeof(locale_table) / sizeof(locale_table[0]); i++) {
1396 const gchar *p;
1397
1398 /* "ja_JP.EUC" matches with "ja_JP.eucJP", "ja_JP.EUC" and
1399 "ja_JP". "ja_JP" matches with "ja_JP.xxxx" and "ja" */
1400 if (!g_ascii_strncasecmp(cur_locale, locale_table[i].locale,
1401 strlen(locale_table[i].locale))) {
1402 cur_charset = locale_table[i].charset;
1403 return cur_charset;
1404 } else if ((p = strchr(locale_table[i].locale, '_')) &&
1405 !strchr(p + 1, '.')) {
1406 if (strlen(cur_locale) == 2 &&
1407 !g_ascii_strncasecmp(cur_locale, locale_table[i].locale, 2)) {
1408 cur_charset = locale_table[i].charset;
1409 return cur_charset;
1410 }
1411 }
1412 }
1413
1414 cur_charset = C_AUTO;
1415 return cur_charset;
1416 }
1417
1418 const gchar *conv_get_locale_charset_str(void)
1419 {
1420 static const gchar *codeset = NULL;
1421
1422 if (!codeset)
1423 codeset = conv_get_charset_str(conv_get_locale_charset());
1424
1425 return codeset ? codeset : CS_INTERNAL;
1426 }
1427
1428 const gchar *conv_get_locale_charset_str_no_utf8(void)
1429 {
1430 static const gchar *codeset = NULL;
1431
1432 if (!codeset)
1433 codeset = conv_get_charset_str(conv_get_locale_charset_no_utf8());
1434
1435 return codeset ? codeset : CS_INTERNAL;
1436 }
1437
1438 static CharSet conv_get_outgoing_charset(void)
1439 {
1440 static CharSet out_charset = -1;
1441 const gchar *cur_locale;
1442 const gchar *p;
1443 gint i;
1444
1445 if (out_charset != -1)
1446 return out_charset;
1447
1448 cur_locale = conv_get_current_locale();
1449 if (!cur_locale) {
1450 out_charset = C_AUTO;
1451 return out_charset;
1452 }
1453
1454 if (strcasestr(cur_locale, "UTF-8") ||
1455 strcasestr(cur_locale, "utf8")) {
1456 out_charset = C_UTF_8;
1457 return out_charset;
1458 }
1459
1460 if ((p = strcasestr(cur_locale, "@euro")) && p[5] == '\0') {
1461 out_charset = C_ISO_8859_15;
1462 return out_charset;
1463 }
1464
1465 for (i = 0; i < sizeof(locale_table) / sizeof(locale_table[0]); i++) {
1466 const gchar *p;
1467
1468 if (!g_ascii_strncasecmp(cur_locale, locale_table[i].locale,
1469 strlen(locale_table[i].locale))) {
1470 out_charset = locale_table[i].out_charset;
1471 break;
1472 } else if ((p = strchr(locale_table[i].locale, '_')) &&
1473 !strchr(p + 1, '.')) {
1474 if (strlen(cur_locale) == 2 &&
1475 !g_ascii_strncasecmp(cur_locale, locale_table[i].locale, 2)) {
1476 out_charset = locale_table[i].out_charset;
1477 break;
1478 }
1479 }
1480 }
1481
1482 return out_charset;
1483 }
1484
1485 const gchar *conv_get_outgoing_charset_str(void)
1486 {
1487 CharSet out_charset;
1488 const gchar *str;
1489
1490 out_charset = conv_get_outgoing_charset();
1491 str = conv_get_charset_str(out_charset);
1492
1493 return str ? str : CS_UTF_8;
1494 }
1495
1496 const gchar *conv_get_current_locale(void)
1497 {
1498 const gchar *cur_locale;
1499
1500 #ifdef G_OS_WIN32
1501 cur_locale = g_win32_getlocale();
1502 #else
1503 cur_locale = g_getenv("LC_ALL");
1504 if (!cur_locale) cur_locale = g_getenv("LC_CTYPE");
1505 if (!cur_locale) cur_locale = g_getenv("LANG");
1506 if (!cur_locale) cur_locale = setlocale(LC_CTYPE, NULL);
1507 #endif /* G_OS_WIN32 */
1508
1509 debug_print("current locale: %s\n",
1510 cur_locale ? cur_locale : "(none)");
1511
1512 return cur_locale;
1513 }
1514
1515 static gboolean conv_is_ja_locale(void)
1516 {
1517 static gint is_ja_locale = -1;
1518 const gchar *cur_locale;
1519
1520 if (is_ja_locale != -1)
1521 return is_ja_locale != 0;
1522
1523 is_ja_locale = 0;
1524 cur_locale = conv_get_current_locale();
1525 if (cur_locale) {
1526 if (g_ascii_strncasecmp(cur_locale, "ja", 2) == 0)
1527 is_ja_locale = 1;
1528 }
1529
1530 return is_ja_locale != 0;
1531 }
1532
1533 gchar *conv_unmime_header(const gchar *str, const gchar *default_encoding,
1534 gboolean addr_field)
1535 {
1536 gchar buf[BUFFSIZE];
1537
1538 if (is_ascii_str(str))
1539 return unmime_header(str, addr_field);
1540
1541 if (default_encoding) {
1542 gchar *utf8_buf;
1543
1544 utf8_buf = conv_codeset_strdup
1545 (str, default_encoding, CS_INTERNAL);
1546 if (utf8_buf) {
1547 gchar *decoded_str;
1548
1549 decoded_str = unmime_header(utf8_buf, addr_field);
1550 g_free(utf8_buf);
1551 return decoded_str;
1552 }
1553 }
1554
1555 if (conv_is_ja_locale())
1556 conv_anytodisp(buf, sizeof(buf), str);
1557 else
1558 conv_localetodisp(buf, sizeof(buf), str);
1559
1560 return unmime_header(buf, addr_field);
1561 }
1562
1563 #define MAX_LINELEN 76
1564 #define MAX_HARD_LINELEN 996
1565 #define MIMESEP_BEGIN "=?"
1566 #define MIMESEP_END "?="
1567
1568 #define LBREAK_IF_REQUIRED(cond, is_plain_text) \
1569 { \
1570 if (len - (destp - (guchar *)dest) < MAX_LINELEN + 2) { \
1571 *destp = '\0'; \
1572 return; \
1573 } \
1574 \
1575 if ((cond) && *srcp) { \
1576 if (destp > (guchar *)dest && left < MAX_LINELEN - 1) { \
1577 if (isspace(*(destp - 1))) \
1578 destp--; \
1579 else if (is_plain_text && isspace(*srcp)) \
1580 srcp++; \
1581 if (*srcp) { \
1582 *destp++ = '\n'; \
1583 *destp++ = ' '; \
1584 left = MAX_LINELEN - 1; \
1585 } \
1586 } else if (destp == (guchar *)dest && left < 7) { \
1587 if (is_plain_text && isspace(*srcp)) \
1588 srcp++; \
1589 if (*srcp) { \
1590 *destp++ = '\n'; \
1591 *destp++ = ' '; \
1592 left = MAX_LINELEN - 1; \
1593 } \
1594 } \
1595 } \
1596 }
1597
1598 #define B64LEN(len) ((len) / 3 * 4 + ((len) % 3 ? 4 : 0))
1599
1600 void conv_encode_header_full(gchar *dest, gint len, const gchar *src,
1601 gint header_len, gboolean addr_field,
1602 const gchar *out_encoding_)
1603 {
1604 const gchar *cur_encoding;
1605 const gchar *out_encoding;
1606 gint mimestr_len;
1607 gchar *mimesep_enc;
1608 gint left;
1609 const guchar *srcp = src;
1610 guchar *destp = dest;
1611 gboolean use_base64;
1612
1613 cm_return_if_fail(g_utf8_validate(src, -1, NULL) == TRUE);
1614 cm_return_if_fail(destp != NULL);
1615
1616 if (MB_CUR_MAX > 1) {
1617 use_base64 = TRUE;
1618 mimesep_enc = "?B?";
1619 } else {
1620 use_base64 = FALSE;
1621 mimesep_enc = "?Q?";
1622 }
1623
1624 cur_encoding = CS_INTERNAL;
1625
1626 if (out_encoding_)
1627 out_encoding = out_encoding_;
1628 else
1629 out_encoding = conv_get_outgoing_charset_str();
1630
1631 if (!strcmp(out_encoding, CS_US_ASCII))
1632 out_encoding = CS_ISO_8859_1;
1633
1634 mimestr_len = strlen(MIMESEP_BEGIN) + strlen(out_encoding) +
1635 strlen(mimesep_enc) + strlen(MIMESEP_END);
1636
1637 left = MAX_LINELEN - header_len;
1638
1639 while (*srcp) {
1640 LBREAK_IF_REQUIRED(left <= 0, TRUE);
1641
1642 while (isspace(*srcp)) {
1643 *destp++ = *srcp++;
1644 left--;
1645 LBREAK_IF_REQUIRED(left <= 0, TRUE);
1646 }
1647
1648 /* output as it is if the next word is ASCII string */
1649 if (!is_next_nonascii(srcp)) {
1650 gint word_len;
1651
1652 word_len = get_next_word_len(srcp);
1653 LBREAK_IF_REQUIRED(left < word_len, TRUE);
1654 while (word_len > 0) {
1655 LBREAK_IF_REQUIRED(left + (MAX_HARD_LINELEN - MAX_LINELEN) <= 0, TRUE)
1656 *destp++ = *srcp++;
1657 left--;
1658 word_len--;
1659 }
1660
1661 continue;
1662 }
1663
1664 /* don't include parentheses and quotes in encoded strings */
1665 if (addr_field && (*srcp == '(' || *srcp == ')' || *srcp == '"')) {
1666 LBREAK_IF_REQUIRED(left < 2, FALSE);
1667 *destp++ = *srcp++;
1668 left--;
1669 }
1670
1671 while (1) {
1672 gint mb_len = 0;
1673 gint cur_len = 0;
1674 gchar *part_str;
1675 gchar *out_str;
1676 gchar *enc_str;
1677 const guchar *p = srcp;
1678 gint out_str_len;
1679 gint out_enc_str_len;
1680 gint mime_block_len;
1681 gboolean cont = FALSE;
1682
1683 while (*p != '\0') {
1684 if (isspace(*p) && !is_next_nonascii(p + 1))
1685 break;
1686 /* don't include parentheses in encoded
1687 strings */
1688 if (addr_field && (*p == '(' || *p == ')' || *p == '"'))
1689 break;
1690
1691 mb_len = g_utf8_skip[*p];
1692
1693 Xstrndup_a(part_str, srcp, cur_len + mb_len, );
1694 out_str = conv_codeset_strdup
1695 (part_str, cur_encoding, out_encoding);
1696 if (!out_str) {
1697 if (strict_mode) {
1698 *dest = '\0';
1699 return;
1700 } else {
1701 g_warning("conv_encode_header(): code conversion failed");
1702 conv_unreadable_8bit(part_str);
1703 out_str = g_strdup(part_str);
1704 }
1705 }
1706 out_str_len = strlen(out_str);
1707
1708 if (use_base64)
1709 out_enc_str_len = B64LEN(out_str_len);
1710 else
1711 out_enc_str_len =
1712 qp_get_q_encoding_len(out_str);
1713
1714 g_free(out_str);
1715
1716 if (mimestr_len + out_enc_str_len <= left) {
1717 cur_len += mb_len;
1718 p += mb_len;
1719 } else if (cur_len == 0) {
1720 left = 0;
1721 LBREAK_IF_REQUIRED(1, FALSE);
1722 continue;
1723 } else {
1724 cont = TRUE;
1725 break;
1726 }
1727 }
1728
1729 if (cur_len > 0) {
1730 Xstrndup_a(part_str, srcp, cur_len, );
1731 out_str = conv_codeset_strdup
1732 (part_str, cur_encoding, out_encoding);
1733 if (!out_str) {
1734 g_warning("conv_encode_header(): code conversion failed");
1735 conv_unreadable_8bit(part_str);
1736 out_str = g_strdup(part_str);
1737 }
1738 out_str_len = strlen(out_str);
1739
1740 if (use_base64)
1741 out_enc_str_len = B64LEN(out_str_len);
1742 else
1743 out_enc_str_len =
1744 qp_get_q_encoding_len(out_str);
1745
1746 if (use_base64)
1747 enc_str = g_base64_encode(out_str, out_str_len);
1748 else {
1749 Xalloca(enc_str, out_enc_str_len + 1, );
1750 qp_q_encode(enc_str, out_str);
1751 }
1752
1753 g_free(out_str);
1754
1755 /* output MIME-encoded string block */
1756 mime_block_len = mimestr_len + strlen(enc_str);
1757 g_snprintf(destp, mime_block_len + 1,
1758 MIMESEP_BEGIN "%s%s%s" MIMESEP_END,
1759 out_encoding, mimesep_enc, enc_str);
1760
1761 if (use_base64)
1762 g_free(enc_str);
1763
1764 destp += mime_block_len;
1765 srcp += cur_len;
1766
1767 left -= mime_block_len;
1768 }
1769
1770 LBREAK_IF_REQUIRED(cont, FALSE);
1771
1772 if (cur_len == 0)
1773 break;
1774 }
1775 }
1776
1777 *destp = '\0';
1778 }
1779
1780 void conv_encode_header(gchar *dest, gint len, const gchar *src,
1781 gint header_len, gboolean addr_field)
1782 {
1783 conv_encode_header_full(dest,len,src,header_len,addr_field,NULL);
1784 }
1785
1786 #undef LBREAK_IF_REQUIRED
1787 #undef B64LEN
1788
1789 gchar *conv_filename_from_utf8(const gchar *utf8_file)
1790 {
1791 gchar *fs_file;
1792 GError *error = NULL;
1793
1794 fs_file = g_filename_from_utf8(utf8_file, -1, NULL, NULL, &error);
1795 if (error) {
1796 debug_print("failed to convert encoding of file name: %s\n",
1797 error->message);
1798 g_error_free(error);
1799 }
1800 if (!fs_file)
1801 fs_file = g_strdup(utf8_file);
1802
1803 return fs_file;
1804 }
1805
1806 gchar *conv_filename_to_utf8(const gchar *fs_file)
1807 {
1808 gchar *utf8_file = NULL;
1809 GError *error = NULL;
1810
1811 utf8_file = g_filename_to_utf8(fs_file, -1, NULL, NULL, &error);
1812 if (error) {
1813 g_warning("failed to convert encoding of file name: %s",
1814 error->message);
1815 g_error_free(error);
1816 }
1817
1818 if (!utf8_file || !g_utf8_validate(utf8_file, -1, NULL)) {
1819 g_free(utf8_file);
1820 utf8_file = g_strdup(fs_file);
1821 conv_unreadable_8bit(utf8_file);
1822 }
1823
1824 return utf8_file;
1825 }