?  search:
summary | log | graphiclog1 | graphiclog2 | commit | commitdiff | tree | refs | edit | fork
blame | history | raw | HEAD
Merge pull request #613 from SiegeLord/even_more_rust_updates
[geany-mirror.git] / src / encodings.c
blob84b757834bcfae89eae3e9296341b26737e1ec15
1 /*
2 * encodings.c - this file is part of Geany, a fast and lightweight IDE
3 *
4 * Copyright 2005-2012 Enrico Tröger <enrico(dot)troeger(at)uvena(dot)de>
5 * Copyright 2006-2012 Nick Treleaven <nick(dot)treleaven(at)btinternet(dot)com>
6 *
7 * This program is free software; you can redistribute it and/or modify
8 * it under the terms of the GNU General Public License as published by
9 * the Free Software Foundation; either version 2 of the License, or
10 * (at your option) any later version.
11 *
12 * This program is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 * GNU General Public License for more details.
16 *
17 * You should have received a copy of the GNU General Public License along
18 * with this program; if not, write to the Free Software Foundation, Inc.,
19 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
20 */
21
22 /*
23 * Encoding conversion and Byte Order Mark (BOM) handling.
24 */
25
26 /*
27 * Modified by the gedit Team, 2002. See the gedit AUTHORS file for a
28 * list of people on the gedit Team.
29 * See the gedit ChangeLog files for a list of changes.
30 */
31 /* Stolen from anjuta */
32
33 #ifdef HAVE_CONFIG_H
34 # include "config.h"
35 #endif
36
37 #include "encodings.h"
38
39 #include "app.h"
40 #include "callbacks.h"
41 #include "documentprivate.h"
42 #include "support.h"
43 #include "ui_utils.h"
44 #include "utils.h"
45
46 #include <string.h>
47
48
49 /* <meta http-equiv="content-type" content="text/html; charset=UTF-8" /> */
50 #define PATTERN_HTMLMETA "<meta\\s+http-equiv\\s*=\\s*\"?content-type\"?\\s+content\\s*=\\s*\"text/x?html;\\s*charset=([a-z0-9_-]+)\"\\s*/?>"
51 /* " geany_encoding=utf-8 " or " coding: utf-8 " */
52 #define PATTERN_CODING "coding[\t ]*[:=][\t ]*\"?([a-z0-9-]+)\"?[\t ]*"
53
54 /* precompiled regexps */
55 static GRegex *pregs[2];
56 static gboolean pregs_loaded = FALSE;
57
58
59 GeanyEncoding encodings[GEANY_ENCODINGS_MAX];
60
61
62 #define fill(Order, Group, Idx, Charset, Name) \
63 encodings[Idx].idx = Idx; \
64 encodings[Idx].order = Order; \
65 encodings[Idx].group = Group; \
66 encodings[Idx].charset = Charset; \
67 encodings[Idx].name = Name;
68
69 static void init_encodings(void)
70 {
71 fill(0, WESTEUROPEAN, GEANY_ENCODING_ISO_8859_14, "ISO-8859-14", _("Celtic"));
72 fill(1, WESTEUROPEAN, GEANY_ENCODING_ISO_8859_7, "ISO-8859-7", _("Greek"));
73 fill(2, WESTEUROPEAN, GEANY_ENCODING_WINDOWS_1253, "WINDOWS-1253", _("Greek"));
74 fill(3, WESTEUROPEAN, GEANY_ENCODING_ISO_8859_10, "ISO-8859-10", _("Nordic"));
75 fill(4, WESTEUROPEAN, GEANY_ENCODING_ISO_8859_3, "ISO-8859-3", _("South European"));
76 fill(5, WESTEUROPEAN, GEANY_ENCODING_IBM_850, "IBM850", _("Western"));
77 fill(6, WESTEUROPEAN, GEANY_ENCODING_ISO_8859_1, "ISO-8859-1", _("Western"));
78 fill(7, WESTEUROPEAN, GEANY_ENCODING_ISO_8859_15, "ISO-8859-15", _("Western"));
79 fill(8, WESTEUROPEAN, GEANY_ENCODING_WINDOWS_1252, "WINDOWS-1252", _("Western"));
80
81 fill(0, EASTEUROPEAN, GEANY_ENCODING_ISO_8859_4, "ISO-8859-4", _("Baltic"));
82 fill(1, EASTEUROPEAN, GEANY_ENCODING_ISO_8859_13, "ISO-8859-13", _("Baltic"));
83 fill(2, EASTEUROPEAN, GEANY_ENCODING_WINDOWS_1257, "WINDOWS-1257", _("Baltic"));
84 fill(3, EASTEUROPEAN, GEANY_ENCODING_IBM_852, "IBM852", _("Central European"));
85 fill(4, EASTEUROPEAN, GEANY_ENCODING_ISO_8859_2, "ISO-8859-2", _("Central European"));
86 fill(5, EASTEUROPEAN, GEANY_ENCODING_WINDOWS_1250, "WINDOWS-1250", _("Central European"));
87 fill(6, EASTEUROPEAN, GEANY_ENCODING_IBM_855, "IBM855", _("Cyrillic"));
88 fill(7, EASTEUROPEAN, GEANY_ENCODING_ISO_8859_5, "ISO-8859-5", _("Cyrillic"));
89 /* ISO-IR-111 not available on Windows */
90 fill(8, EASTEUROPEAN, GEANY_ENCODING_ISO_IR_111, "ISO-IR-111", _("Cyrillic"));
91 fill(9, EASTEUROPEAN, GEANY_ENCODING_KOI8_R, "KOI8-R", _("Cyrillic"));
92 fill(10, EASTEUROPEAN, GEANY_ENCODING_WINDOWS_1251, "WINDOWS-1251", _("Cyrillic"));
93 fill(11, EASTEUROPEAN, GEANY_ENCODING_CP_866, "CP866", _("Cyrillic/Russian"));
94 fill(12, EASTEUROPEAN, GEANY_ENCODING_KOI8_U, "KOI8-U", _("Cyrillic/Ukrainian"));
95 fill(13, EASTEUROPEAN, GEANY_ENCODING_ISO_8859_16, "ISO-8859-16", _("Romanian"));
96
97 fill(0, MIDDLEEASTERN, GEANY_ENCODING_IBM_864, "IBM864", _("Arabic"));
98 fill(1, MIDDLEEASTERN, GEANY_ENCODING_ISO_8859_6, "ISO-8859-6", _("Arabic"));
99 fill(2, MIDDLEEASTERN, GEANY_ENCODING_WINDOWS_1256, "WINDOWS-1256", _("Arabic"));
100 fill(3, MIDDLEEASTERN, GEANY_ENCODING_IBM_862, "IBM862", _("Hebrew"));
101 /* not available at all, ? */
102 fill(4, MIDDLEEASTERN, GEANY_ENCODING_ISO_8859_8_I, "ISO-8859-8-I", _("Hebrew"));
103 fill(5, MIDDLEEASTERN, GEANY_ENCODING_WINDOWS_1255, "WINDOWS-1255", _("Hebrew"));
104 fill(6, MIDDLEEASTERN, GEANY_ENCODING_ISO_8859_8, "ISO-8859-8", _("Hebrew Visual"));
105
106 fill(0, ASIAN, GEANY_ENCODING_ARMSCII_8, "ARMSCII-8", _("Armenian"));
107 fill(1, ASIAN, GEANY_ENCODING_GEOSTD8, "GEORGIAN-ACADEMY", _("Georgian"));
108 fill(2, ASIAN, GEANY_ENCODING_TIS_620, "TIS-620", _("Thai"));
109 fill(3, ASIAN, GEANY_ENCODING_IBM_857, "IBM857", _("Turkish"));
110 fill(4, ASIAN, GEANY_ENCODING_WINDOWS_1254, "WINDOWS-1254", _("Turkish"));
111 fill(5, ASIAN, GEANY_ENCODING_ISO_8859_9, "ISO-8859-9", _("Turkish"));
112 fill(6, ASIAN, GEANY_ENCODING_TCVN, "TCVN", _("Vietnamese"));
113 fill(7, ASIAN, GEANY_ENCODING_VISCII, "VISCII", _("Vietnamese"));
114 fill(8, ASIAN, GEANY_ENCODING_WINDOWS_1258, "WINDOWS-1258", _("Vietnamese"));
115
116 fill(0, UNICODE, GEANY_ENCODING_UTF_7, "UTF-7", _("Unicode"));
117 fill(1, UNICODE, GEANY_ENCODING_UTF_8, "UTF-8", _("Unicode"));
118 fill(2, UNICODE, GEANY_ENCODING_UTF_16LE, "UTF-16LE", _("Unicode"));
119 fill(3, UNICODE, GEANY_ENCODING_UTF_16BE, "UTF-16BE", _("Unicode"));
120 fill(4, UNICODE, GEANY_ENCODING_UCS_2LE, "UCS-2LE", _("Unicode"));
121 fill(5, UNICODE, GEANY_ENCODING_UCS_2BE, "UCS-2BE", _("Unicode"));
122 fill(6, UNICODE, GEANY_ENCODING_UTF_32LE, "UTF-32LE", _("Unicode"));
123 fill(7, UNICODE, GEANY_ENCODING_UTF_32BE, "UTF-32BE", _("Unicode"));
124
125 fill(0, EASTASIAN, GEANY_ENCODING_GB18030, "GB18030", _("Chinese Simplified"));
126 fill(1, EASTASIAN, GEANY_ENCODING_GB2312, "GB2312", _("Chinese Simplified"));
127 fill(2, EASTASIAN, GEANY_ENCODING_GBK, "GBK", _("Chinese Simplified"));
128 /* maybe not available on Linux */
129 fill(3, EASTASIAN, GEANY_ENCODING_HZ, "HZ", _("Chinese Simplified"));
130 fill(4, EASTASIAN, GEANY_ENCODING_BIG5, "BIG5", _("Chinese Traditional"));
131 fill(5, EASTASIAN, GEANY_ENCODING_BIG5_HKSCS, "BIG5-HKSCS", _("Chinese Traditional"));
132 fill(6, EASTASIAN, GEANY_ENCODING_EUC_TW, "EUC-TW", _("Chinese Traditional"));
133 fill(7, EASTASIAN, GEANY_ENCODING_EUC_JP, "EUC-JP", _("Japanese"));
134 fill(8, EASTASIAN, GEANY_ENCODING_ISO_2022_JP, "ISO-2022-JP", _("Japanese"));
135 fill(9, EASTASIAN, GEANY_ENCODING_SHIFT_JIS, "SHIFT_JIS", _("Japanese"));
136 fill(10, EASTASIAN, GEANY_ENCODING_CP_932, "CP932", _("Japanese"));
137 fill(11, EASTASIAN, GEANY_ENCODING_EUC_KR, "EUC-KR", _("Korean"));
138 fill(12, EASTASIAN, GEANY_ENCODING_ISO_2022_KR, "ISO-2022-KR", _("Korean"));
139 fill(13, EASTASIAN, GEANY_ENCODING_JOHAB, "JOHAB", _("Korean"));
140 fill(14, EASTASIAN, GEANY_ENCODING_UHC, "UHC", _("Korean"));
141
142 fill(0, NONE, GEANY_ENCODING_NONE, "None", _("Without encoding"));
143 }
144
145
146 /* compares two encoding names in a permissive fashion.
147 * e.g. "utf8" matches "UTF-8", "iso8859_1" matches "ISO-8859-1", etc. */
148 static gboolean encodings_charset_equals(const gchar *a, const gchar *b)
149 {
150 gboolean was_alpha = FALSE; /* whether last character of previous word was a letter */
151 gboolean need_sep = FALSE; /* whether we're expecting an implicit separator */
152
153 while (*a && *b)
154 {
155 gboolean is_alpha;
156
157 if (g_ascii_toupper(*a) == g_ascii_toupper(*b) &&
158 ((is_alpha = g_ascii_isalpha(*a)) || g_ascii_isdigit(*a)))
159 {
160 /* either there was a real separator, or we need a implicit one (a chage from alpha to
161 * numeric or so) */
162 if (! need_sep || (was_alpha != is_alpha))
163 {
164 a++;
165 b++;
166 was_alpha = is_alpha;
167 need_sep = FALSE;
168 }
169 else
170 return FALSE;
171 }
172 else
173 {
174 guint n_sep = 0;
175
176 if (! g_ascii_isalnum(*a))
177 {
178 a++;
179 n_sep++;
180 }
181 if (! g_ascii_isalnum(*b))
182 {
183 b++;
184 n_sep++;
185 }
186 if (n_sep < 1)
187 return FALSE;
188 else if (n_sep < 2)
189 need_sep = TRUE;
190 }
191 }
192 return *a == *b;
193 }
194
195
196 GeanyEncodingIndex encodings_get_idx_from_charset(const gchar *charset)
197 {
198 gint i;
199
200 if (charset == NULL)
201 return GEANY_ENCODING_UTF_8;
202
203 i = 0;
204 while (i < GEANY_ENCODINGS_MAX)
205 {
206 if (encodings_charset_equals(charset, encodings[i].charset))
207 return i;
208
209 ++i;
210 }
211 return GEANY_ENCODING_UTF_8;
212 }
213
214
215 const GeanyEncoding *encodings_get_from_charset(const gchar *charset)
216 {
217 gint i;
218
219 if (charset == NULL)
220 return &encodings[GEANY_ENCODING_UTF_8];
221
222 i = 0;
223 while (i < GEANY_ENCODINGS_MAX)
224 {
225 if (encodings_charset_equals(charset, encodings[i].charset))
226 return &encodings[i];
227
228 ++i;
229 }
230
231 return NULL;
232 }
233
234
235 static const gchar *encodings_normalize_charset(const gchar *charset)
236 {
237 const GeanyEncoding *encoding;
238
239 encoding = encodings_get_from_charset(charset);
240 if (encoding != NULL)
241 return encoding->charset;
242
243 return NULL;
244 }
245
246
247 const GeanyEncoding *encodings_get_from_index(gint idx)
248 {
249 g_return_val_if_fail(idx >= 0 && idx < GEANY_ENCODINGS_MAX, NULL);
250
251 return &encodings[idx];
252 }
253
254
255 /**
256 * Gets the character set name of the specified index e.g. for use with
257 * @ref document_set_encoding().
258 *
259 * @param idx @ref GeanyEncodingIndex to retrieve the corresponding character set.
260 *
261 *
262 * @return The charset according to idx, or @c NULL if the index is invalid.
263 *
264 * @since 0.13
265 **/
266 GEANY_API_SYMBOL
267 const gchar* encodings_get_charset_from_index(gint idx)
268 {
269 g_return_val_if_fail(idx >= 0 && idx < GEANY_ENCODINGS_MAX, NULL);
270
271 return encodings[idx].charset;
272 }
273
274
275 gchar *encodings_to_string(const GeanyEncoding* enc)
276 {
277 g_return_val_if_fail(enc != NULL, NULL);
278 g_return_val_if_fail(enc->name != NULL, NULL);
279 g_return_val_if_fail(enc->charset != NULL, NULL);
280
281 return g_strdup_printf("%s (%s)", enc->name, enc->charset);
282 }
283
284
285 const gchar *encodings_get_charset(const GeanyEncoding* enc)
286 {
287 g_return_val_if_fail(enc != NULL, NULL);
288 g_return_val_if_fail(enc->charset != NULL, NULL);
289
290 return enc->charset;
291 }
292
293
294 static GtkWidget *radio_items[GEANY_ENCODINGS_MAX];
295
296
297 void encodings_select_radio_item(const gchar *charset)
298 {
299 gint i;
300
301 g_return_if_fail(charset != NULL);
302
303 i = 0;
304 while (i < GEANY_ENCODINGS_MAX)
305 {
306 if (utils_str_equal(charset, encodings[i].charset))
307 break;
308 i++;
309 }
310 if (i == GEANY_ENCODINGS_MAX)
311 i = GEANY_ENCODING_UTF_8; /* fallback to UTF-8 */
312
313 /* ignore_callback has to be set by the caller */
314 gtk_check_menu_item_set_active(GTK_CHECK_MENU_ITEM(radio_items[i]), TRUE);
315 }
316
317
318 /* Regexp detection of file encoding declared in the file itself.
319 * Idea and parts of code taken from Bluefish, thanks.
320 * regex_compile() is used to compile regular expressions on program init and keep it in memory
321 * for faster access when opening a file. Pre-compiled regexps will be freed on program exit.
322 */
323 static GRegex *regex_compile(const gchar *pattern)
324 {
325 GError *error = NULL;
326 GRegex *regex = g_regex_new(pattern, G_REGEX_CASELESS, 0, &error);
327
328 if (!regex)
329 {
330 geany_debug("Failed to compile encoding regex (%s)", error->message);
331 g_error_free(error);
332 }
333 return regex;
334 }
335
336
337 static gchar *regex_match(GRegex *preg, const gchar *buffer, gsize size)
338 {
339 gchar *encoding = NULL;
340 GMatchInfo *minfo;
341
342 if (G_UNLIKELY(! pregs_loaded || buffer == NULL))
343 return NULL;
344
345 /* scan only the first 512 characters in the buffer */
346 size = MIN(size, 512);
347
348 if (g_regex_match_full(preg, buffer, size, 0, 0, &minfo, NULL) &&
349 g_match_info_get_match_count(minfo) >= 2)
350 {
351 encoding = g_match_info_fetch(minfo, 1);
352 geany_debug("Detected encoding by regex search: %s", encoding);
353
354 SETPTR(encoding, g_utf8_strup(encoding, -1));
355 }
356 g_match_info_free(minfo);
357 return encoding;
358 }
359
360
361 static void encodings_radio_item_change_cb(GtkCheckMenuItem *menuitem, gpointer user_data)
362 {
363 GeanyDocument *doc = document_get_current();
364 const gchar *charset = user_data;
365
366 if (ignore_callback || doc == NULL || charset == NULL ||
367 ! gtk_check_menu_item_get_active(menuitem) ||
368 utils_str_equal(charset, doc->encoding))
369 return;
370
371 if (doc->readonly)
372 {
373 utils_beep();
374 return;
375 }
376 document_undo_add(doc, UNDO_ENCODING, g_strdup(doc->encoding));
377
378 document_set_encoding(doc, charset);
379 }
380
381 static void encodings_reload_radio_item_change_cb(GtkMenuItem *menuitem, gpointer user_data)
382 {
383 GeanyDocument *doc = document_get_current();
384
385 g_return_if_fail(doc != NULL);
386
387 document_reload_prompt(doc, user_data);
388 }
389
390
391 void encodings_finalize(void)
392 {
393 if (pregs_loaded)
394 {
395 guint i, len;
396 len = G_N_ELEMENTS(pregs);
397 for (i = 0; i < len; i++)
398 {
399 g_regex_unref(pregs[i]);
400 }
401 }
402 }
403
404
405 void encodings_init(void)
406 {
407 GtkWidget *item, *menu[2], *submenu, *menu_westeuro, *menu_easteuro, *menu_eastasian, *menu_asian,
408 *menu_utf8, *menu_middleeast, *item_westeuro, *item_easteuro, *item_eastasian,
409 *item_asian, *item_utf8, *item_middleeast;
410 GCallback cb_func[2];
411 GSList *group = NULL;
412 gchar *label;
413 gint order, group_size;
414 guint i, j, k;
415
416 init_encodings();
417
418 if (! pregs_loaded)
419 {
420 pregs[0] = regex_compile(PATTERN_HTMLMETA);
421 pregs[1] = regex_compile(PATTERN_CODING);
422 pregs_loaded = TRUE;
423 }
424
425 /* create encodings submenu in document menu */
426 menu[0] = ui_lookup_widget(main_widgets.window, "set_encoding1_menu");
427 menu[1] = ui_lookup_widget(main_widgets.window, "menu_reload_as1_menu");
428 cb_func[0] = G_CALLBACK(encodings_radio_item_change_cb);
429 cb_func[1] = G_CALLBACK(encodings_reload_radio_item_change_cb);
430
431 for (k = 0; k < 2; k++)
432 {
433 menu_westeuro = gtk_menu_new();
434 item_westeuro = gtk_menu_item_new_with_mnemonic(_("_West European"));
435 gtk_menu_item_set_submenu(GTK_MENU_ITEM(item_westeuro), menu_westeuro);
436 gtk_container_add(GTK_CONTAINER(menu[k]), item_westeuro);
437 gtk_widget_show_all(item_westeuro);
438
439 menu_easteuro = gtk_menu_new();
440 item_easteuro = gtk_menu_item_new_with_mnemonic(_("_East European"));
441 gtk_menu_item_set_submenu(GTK_MENU_ITEM(item_easteuro), menu_easteuro);
442 gtk_container_add(GTK_CONTAINER(menu[k]), item_easteuro);
443 gtk_widget_show_all(item_easteuro);
444
445 menu_eastasian = gtk_menu_new();
446 item_eastasian = gtk_menu_item_new_with_mnemonic(_("East _Asian"));
447 gtk_menu_item_set_submenu(GTK_MENU_ITEM(item_eastasian), menu_eastasian);
448 gtk_container_add(GTK_CONTAINER(menu[k]), item_eastasian);
449 gtk_widget_show_all(item_eastasian);
450
451 menu_asian = gtk_menu_new();
452 item_asian = gtk_menu_item_new_with_mnemonic(_("_SE & SW Asian"));
453 gtk_menu_item_set_submenu(GTK_MENU_ITEM(item_asian), menu_asian);
454 gtk_container_add(GTK_CONTAINER(menu[k]), item_asian);
455 gtk_widget_show_all(item_asian);
456
457 menu_middleeast = gtk_menu_new();
458 item_middleeast = gtk_menu_item_new_with_mnemonic(_("_Middle Eastern"));
459 gtk_menu_item_set_submenu(GTK_MENU_ITEM(item_middleeast), menu_middleeast);
460 gtk_container_add(GTK_CONTAINER(menu[k]), item_middleeast);
461 gtk_widget_show_all(item_middleeast);
462
463 menu_utf8 = gtk_menu_new();
464 item_utf8 = gtk_menu_item_new_with_mnemonic(_("_Unicode"));
465 gtk_menu_item_set_submenu(GTK_MENU_ITEM(item_utf8), menu_utf8);
466 gtk_container_add(GTK_CONTAINER(menu[k]), item_utf8);
467 gtk_widget_show_all(item_utf8);
468
469 /** TODO can it be optimized? ATM 3782 runs at line "if (encodings[j].group ...)" */
470 for (i = 0; i < GEANY_ENCODING_GROUPS_MAX; i++)
471 {
472 order = 0;
473 switch (i)
474 {
475 case WESTEUROPEAN: submenu = menu_westeuro; group_size = 9; break;
476 case EASTEUROPEAN: submenu = menu_easteuro; group_size = 14; break;
477 case EASTASIAN: submenu = menu_eastasian; group_size = 14; break;
478 case ASIAN: submenu = menu_asian; group_size = 9; break;
479 case MIDDLEEASTERN: submenu = menu_middleeast; group_size = 7; break;
480 case UNICODE: submenu = menu_utf8; group_size = 8; break;
481 default: submenu = menu[k]; group_size = 1;
482 }
483
484 while (order < group_size) /* the biggest group has 13 elements */
485 {
486 for (j = 0; j < GEANY_ENCODINGS_MAX; j++)
487 {
488 if (encodings[j].group == i && encodings[j].order == order)
489 {
490 label = encodings_to_string(&encodings[j]);
491 if (k == 0)
492 {
493 item = gtk_radio_menu_item_new_with_label(group, label);
494 group = gtk_radio_menu_item_get_group(GTK_RADIO_MENU_ITEM(item));
495 radio_items[j] = item;
496 }
497 else
498 item = gtk_menu_item_new_with_label(label);
499 gtk_widget_show(item);
500 gtk_container_add(GTK_CONTAINER(submenu), item);
501 g_signal_connect(item, "activate", cb_func[k],
502 (gpointer) encodings[j].charset);
503 g_free(label);
504 break;
505 }
506 }
507 order++;
508 }
509 }
510 }
511 }
512
513
514 static gint encoding_combo_store_sort_func(GtkTreeModel *model,
515 GtkTreeIter *a,
516 GtkTreeIter *b,
517 gpointer data)
518 {
519 gboolean a_has_child = gtk_tree_model_iter_has_child(model, a);
520 gboolean b_has_child = gtk_tree_model_iter_has_child(model, b);
521 gchar *a_string;
522 gchar *b_string;
523 gint cmp_res;
524
525 if (a_has_child != b_has_child)
526 return a_has_child ? -1 : 1;
527
528 gtk_tree_model_get(model, a, 1, &a_string, -1);
529 gtk_tree_model_get(model, b, 1, &b_string, -1);
530 cmp_res = strcmp(a_string, b_string);
531 g_free(a_string);
532 g_free(b_string);
533 return cmp_res;
534 }
535
536
537 GtkTreeStore *encodings_encoding_store_new(gboolean has_detect)
538 {
539 GtkTreeStore *store;
540 GtkTreeIter iter_current, iter_westeuro, iter_easteuro, iter_eastasian,
541 iter_asian, iter_utf8, iter_middleeast;
542 GtkTreeIter *iter_parent;
543 gchar *encoding_string;
544 gint i;
545
546 store = gtk_tree_store_new(2, G_TYPE_INT, G_TYPE_STRING);
547
548 if (has_detect)
549 {
550 gtk_tree_store_append(store, &iter_current, NULL);
551 gtk_tree_store_set(store, &iter_current, 0, GEANY_ENCODINGS_MAX, 1, _("Detect from file"), -1);
552 }
553
554 gtk_tree_store_append(store, &iter_westeuro, NULL);
555 gtk_tree_store_set(store, &iter_westeuro, 0, -1, 1, _("West European"), -1);
556 gtk_tree_store_append(store, &iter_easteuro, NULL);
557 gtk_tree_store_set(store, &iter_easteuro, 0, -1, 1, _("East European"), -1);
558 gtk_tree_store_append(store, &iter_eastasian, NULL);
559 gtk_tree_store_set(store, &iter_eastasian, 0, -1, 1, _("East Asian"), -1);
560 gtk_tree_store_append(store, &iter_asian, NULL);
561 gtk_tree_store_set(store, &iter_asian, 0, -1, 1, _("SE & SW Asian"), -1);
562 gtk_tree_store_append(store, &iter_middleeast, NULL);
563 gtk_tree_store_set(store, &iter_middleeast, 0, -1, 1, _("Middle Eastern"), -1);
564 gtk_tree_store_append(store, &iter_utf8, NULL);
565 gtk_tree_store_set(store, &iter_utf8, 0, -1, 1, _("Unicode"), -1);
566
567 for (i = 0; i < GEANY_ENCODINGS_MAX; i++)
568 {
569 switch (encodings[i].group)
570 {
571 case WESTEUROPEAN: iter_parent = &iter_westeuro; break;
572 case EASTEUROPEAN: iter_parent = &iter_easteuro; break;
573 case EASTASIAN: iter_parent = &iter_eastasian; break;
574 case ASIAN: iter_parent = &iter_asian; break;
575 case MIDDLEEASTERN: iter_parent = &iter_middleeast; break;
576 case UNICODE: iter_parent = &iter_utf8; break;
577 case NONE:
578 default: iter_parent = NULL;
579 }
580 gtk_tree_store_append(store, &iter_current, iter_parent);
581 encoding_string = encodings_to_string(&encodings[i]);
582 gtk_tree_store_set(store, &iter_current, 0, i, 1, encoding_string, -1);
583 g_free(encoding_string);
584 }
585
586 gtk_tree_sortable_set_sort_column_id(GTK_TREE_SORTABLE(store), 1, GTK_SORT_ASCENDING);
587 gtk_tree_sortable_set_sort_func(GTK_TREE_SORTABLE(store), 1, encoding_combo_store_sort_func, NULL, NULL);
588
589 return store;
590 }
591
592
593 gint encodings_encoding_store_get_encoding(GtkTreeStore *store, GtkTreeIter *iter)
594 {
595 gint enc;
596 gtk_tree_model_get(GTK_TREE_MODEL(store), iter, 0, &enc, -1);
597 return enc;
598 }
599
600
601 gboolean encodings_encoding_store_get_iter(GtkTreeStore *store, GtkTreeIter *iter, gint enc)
602 {
603 if (gtk_tree_model_get_iter_first(GTK_TREE_MODEL(store), iter))
604 {
605 do
606 {
607 if (encodings_encoding_store_get_encoding(store, iter) == enc)
608 return TRUE;
609 }
610 while (ui_tree_model_iter_any_next(GTK_TREE_MODEL(store), iter, TRUE));
611 }
612 return FALSE;
613 }
614
615
616 void encodings_encoding_store_cell_data_func(GtkCellLayout *cell_layout,
617 GtkCellRenderer *cell,
618 GtkTreeModel *tree_model,
619 GtkTreeIter *iter,
620 gpointer data)
621 {
622 gboolean sensitive = !gtk_tree_model_iter_has_child(tree_model, iter);
623 gchar *text;
624
625 gtk_tree_model_get(tree_model, iter, 1, &text, -1);
626 g_object_set(cell, "sensitive", sensitive, "text", text, NULL);
627 g_free(text);
628 }
629
630
631 /**
632 * Tries to convert @a buffer into UTF-8 encoding from the encoding specified with @a charset.
633 * If @a fast is not set, additional checks to validate the converted string are performed.
634 *
635 * @param buffer The input string to convert.
636 * @param size The length of the string, or -1 if the string is nul-terminated.
637 * @param charset The charset to be used for conversion.
638 * @param fast @c TRUE to only convert the input and skip extended checks on the converted string.
639 *
640 * @return If the conversion was successful, a newly allocated nul-terminated string,
641 * which must be freed with @c g_free(). Otherwise @c NULL.
642 **/
643 GEANY_API_SYMBOL
644 gchar *encodings_convert_to_utf8_from_charset(const gchar *buffer, gssize size,
645 const gchar *charset, gboolean fast)
646 {
647 gchar *utf8_content = NULL;
648 GError *conv_error = NULL;
649 gchar* converted_contents = NULL;
650 gsize bytes_written;
651
652 g_return_val_if_fail(buffer != NULL, NULL);
653 g_return_val_if_fail(charset != NULL, NULL);
654
655 converted_contents = g_convert(buffer, size, "UTF-8", charset, NULL,
656 &bytes_written, &conv_error);
657
658 if (fast)
659 {
660 utf8_content = converted_contents;
661 if (conv_error != NULL) g_error_free(conv_error);
662 }
663 else if (conv_error != NULL || ! g_utf8_validate(converted_contents, bytes_written, NULL))
664 {
665 if (conv_error != NULL)
666 {
667 geany_debug("Couldn't convert from %s to UTF-8 (%s).", charset, conv_error->message);
668 g_error_free(conv_error);
669 conv_error = NULL;
670 }
671 else
672 geany_debug("Couldn't convert from %s to UTF-8.", charset);
673
674 utf8_content = NULL;
675 g_free(converted_contents);
676 }
677 else
678 {
679 geany_debug("Converted from %s to UTF-8.", charset);
680 utf8_content = converted_contents;
681 }
682
683 return utf8_content;
684 }
685
686
687 static gchar *encodings_check_regexes(const gchar *buffer, gsize size)
688 {
689 guint i;
690
691 for (i = 0; i < G_N_ELEMENTS(pregs); i++)
692 {
693 gchar *charset;
694
695 if ((charset = regex_match(pregs[i], buffer, size)) != NULL)
696 return charset;
697 }
698 return NULL;
699 }
700
701
702 static gchar *encodings_convert_to_utf8_with_suggestion(const gchar *buffer, gssize size,
703 const gchar *suggested_charset, gchar **used_encoding)
704 {
705 const gchar *locale_charset = NULL;
706 const gchar *charset;
707 gchar *utf8_content;
708 gboolean check_suggestion = suggested_charset != NULL;
709 gboolean check_locale = FALSE;
710 gint i, preferred_charset;
711
712 if (size == -1)
713 {
714 size = strlen(buffer);
715 }
716
717 /* current locale is not UTF-8, we have to check this charset */
718 check_locale = ! g_get_charset(&locale_charset);
719
720 /* First check for preferred charset, if specified */
721 preferred_charset = file_prefs.default_open_encoding;
722
723 if (preferred_charset == encodings[GEANY_ENCODING_NONE].idx ||
724 preferred_charset < 0 ||
725 preferred_charset >= GEANY_ENCODINGS_MAX)
726 {
727 preferred_charset = -1;
728 }
729
730 /* -1 means "Preferred charset" */
731 for (i = -1; i < GEANY_ENCODINGS_MAX; i++)
732 {
733 if (G_UNLIKELY(i == encodings[GEANY_ENCODING_NONE].idx))
734 continue;
735
736 if (check_suggestion)
737 {
738 check_suggestion = FALSE;
739 charset = encodings_normalize_charset(suggested_charset);
740 if (charset == NULL) /* we failed at normalizing suggested encoding, try it as is */
741 charset = suggested_charset;
742 i = -2; /* keep i below the start value to have it again at -1 on the next loop run */
743 }
744 else if (check_locale)
745 {
746 check_locale = FALSE;
747 charset = locale_charset;
748 i = -2; /* keep i below the start value to have it again at -1 on the next loop run */
749 }
750 else if (i == -1)
751 {
752 if (preferred_charset >= 0)
753 {
754 charset = encodings[preferred_charset].charset;
755 geany_debug("Using preferred charset: %s", charset);
756 }
757 else
758 continue;
759 }
760 else if (i >= 0)
761 charset = encodings[i].charset;
762 else /* in this case we have i == -2, continue to increase i and go ahead */
763 continue;
764
765 if (G_UNLIKELY(charset == NULL))
766 continue;
767
768 geany_debug("Trying to convert %" G_GSIZE_FORMAT " bytes of data from %s into UTF-8.",
769 size, charset);
770 utf8_content = encodings_convert_to_utf8_from_charset(buffer, size, charset, FALSE);
771
772 if (G_LIKELY(utf8_content != NULL))
773 {
774 if (used_encoding != NULL)
775 {
776 if (G_UNLIKELY(*used_encoding != NULL))
777 {
778 geany_debug("%s:%d", __FILE__, __LINE__);
779 g_free(*used_encoding);
780 }
781 *used_encoding = g_strdup(charset);
782 }
783 return utf8_content;
784 }
785 }
786
787 return NULL;
788 }
789
790
791 /**
792 * Tries to convert @a buffer into UTF-8 encoding and store the detected original encoding in
793 * @a used_encoding.
794 *
795 * @param buffer the input string to convert.
796 * @param size the length of the string, or -1 if the string is nul-terminated.
797 * @param used_encoding return location of the detected encoding of the input string, or @c NULL.
798 *
799 * @return If the conversion was successful, a newly allocated nul-terminated string,
800 * which must be freed with @c g_free(). Otherwise @c NULL.
801 **/
802 GEANY_API_SYMBOL
803 gchar *encodings_convert_to_utf8(const gchar *buffer, gssize size, gchar **used_encoding)
804 {
805 gchar *regex_charset;
806 gchar *utf8;
807
808 /* first try to read the encoding from the file content */
809 regex_charset = encodings_check_regexes(buffer, size);
810 utf8 = encodings_convert_to_utf8_with_suggestion(buffer, size, regex_charset, used_encoding);
811 g_free(regex_charset);
812
813 return utf8;
814 }
815
816
817 /* If there's a BOM, return a corresponding GEANY_ENCODING_UTF_* index,
818 * otherwise GEANY_ENCODING_NONE.
819 * */
820 GeanyEncodingIndex encodings_scan_unicode_bom(const gchar *string, gsize len, guint *bom_len)
821 {
822 if (len >= 3)
823 {
824 if (bom_len)
825 *bom_len = 3;
826
827 if ((guchar)string[0] == 0xef && (guchar)string[1] == 0xbb &&
828 (guchar)string[2] == 0xbf)
829 {
830 return GEANY_ENCODING_UTF_8;
831 }
832 }
833 if (len >= 4)
834 {
835 if (bom_len)
836 *bom_len = 4;
837
838 if ((guchar)string[0] == 0x00 && (guchar)string[1] == 0x00 &&
839 (guchar)string[2] == 0xfe && (guchar)string[3] == 0xff)
840 {
841 return GEANY_ENCODING_UTF_32BE; /* Big endian */
842 }
843 if ((guchar)string[0] == 0xff && (guchar)string[1] == 0xfe &&
844 (guchar)string[2] == 0x00 && (guchar)string[3] == 0x00)
845 {
846 return GEANY_ENCODING_UTF_32LE; /* Little endian */
847 }
848 if ((string[0] == 0x2b && string[1] == 0x2f && string[2] == 0x76) &&
849 (string[3] == 0x38 || string[3] == 0x39 || string[3] == 0x2b || string[3] == 0x2f))
850 {
851 return GEANY_ENCODING_UTF_7;
852 }
853 }
854 if (len >= 2)
855 {
856 if (bom_len)
857 *bom_len = 2;
858
859 if ((guchar)string[0] == 0xfe && (guchar)string[1] == 0xff)
860 {
861 return GEANY_ENCODING_UTF_16BE; /* Big endian */
862 }
863 if ((guchar)string[0] == 0xff && (guchar)string[1] == 0xfe)
864 {
865 return GEANY_ENCODING_UTF_16LE; /* Little endian */
866 }
867 }
868 if (bom_len)
869 *bom_len = 0;
870 return GEANY_ENCODING_NONE;
871 }
872
873
874 gboolean encodings_is_unicode_charset(const gchar *string)
875 {
876 if (string != NULL &&
877 (strncmp(string, "UTF", 3) == 0 || strncmp(string, "UCS", 3) == 0))
878 {
879 return TRUE;
880 }
881 return FALSE;
882 }
883
884
885 typedef struct
886 {
887 gchar *data; /* null-terminated data */
888 gsize size; /* actual data size */
889 gsize len; /* string length of data */
890 gchar *enc;
891 gboolean bom;
892 gboolean partial;
893 } BufferData;
894
895
896 /* convert data with the specified encoding */
897 static gboolean
898 handle_forced_encoding(BufferData *buffer, const gchar *forced_enc)
899 {
900 GeanyEncodingIndex enc_idx;
901
902 if (utils_str_equal(forced_enc, "UTF-8"))
903 {
904 if (! g_utf8_validate(buffer->data, buffer->len, NULL))
905 {
906 return FALSE;
907 }
908 }
909 else
910 {
911 gchar *converted_text = encodings_convert_to_utf8_from_charset(
912 buffer->data, buffer->size, forced_enc, FALSE);
913 if (converted_text == NULL)
914 {
915 return FALSE;
916 }
917 else
918 {
919 SETPTR(buffer->data, converted_text);
920 buffer->len = strlen(converted_text);
921 }
922 }
923 enc_idx = encodings_scan_unicode_bom(buffer->data, buffer->size, NULL);
924 buffer->bom = (enc_idx == GEANY_ENCODING_UTF_8);
925 buffer->enc = g_strdup(forced_enc);
926 return TRUE;
927 }
928
929
930 /* detect encoding and convert to UTF-8 if necessary */
931 static gboolean
932 handle_encoding(BufferData *buffer, GeanyEncodingIndex enc_idx)
933 {
934 g_return_val_if_fail(buffer->enc == NULL, FALSE);
935 g_return_val_if_fail(buffer->bom == FALSE, FALSE);
936
937 if (buffer->size == 0)
938 {
939 /* we have no data so assume UTF-8, buffer->len can be 0 even we have an empty
940 * e.g. UTF32 file with a BOM(so size is 4, len is 0) */
941 buffer->enc = g_strdup("UTF-8");
942 }
943 else
944 {
945 /* first check for a BOM */
946 if (enc_idx != GEANY_ENCODING_NONE)
947 {
948 buffer->enc = g_strdup(encodings[enc_idx].charset);
949 buffer->bom = TRUE;
950
951 if (enc_idx != GEANY_ENCODING_UTF_8) /* the BOM indicated something else than UTF-8 */
952 {
953 gchar *converted_text = encodings_convert_to_utf8_from_charset(
954 buffer->data, buffer->size, buffer->enc, FALSE);
955 if (converted_text != NULL)
956 {
957 SETPTR(buffer->data, converted_text);
958 buffer->len = strlen(converted_text);
959 }
960 else
961 {
962 /* there was a problem converting data from BOM encoding type */
963 SETPTR(buffer->enc, NULL);
964 buffer->bom = FALSE;
965 }
966 }
967 }
968
969 if (buffer->enc == NULL) /* either there was no BOM or the BOM encoding failed */
970 {
971 /* first try to read the encoding from the file content */
972 gchar *regex_charset = encodings_check_regexes(buffer->data, buffer->size);
973
974 /* try UTF-8 first */
975 if (encodings_get_idx_from_charset(regex_charset) == GEANY_ENCODING_UTF_8 &&
976 (buffer->size == buffer->len) && g_utf8_validate(buffer->data, buffer->len, NULL))
977 {
978 buffer->enc = g_strdup("UTF-8");
979 }
980 else
981 {
982 /* detect the encoding */
983 gchar *converted_text = encodings_convert_to_utf8_with_suggestion(buffer->data,
984 buffer->size, regex_charset, &buffer->enc);
985
986 if (converted_text == NULL)
987 {
988 g_free(regex_charset);
989 return FALSE;
990 }
991 SETPTR(buffer->data, converted_text);
992 buffer->len = strlen(converted_text);
993 }
994 g_free(regex_charset);
995 }
996 }
997 return TRUE;
998 }
999
1000
1001 static void
1002 handle_bom(BufferData *buffer)
1003 {
1004 guint bom_len;
1005
1006 encodings_scan_unicode_bom(buffer->data, buffer->size, &bom_len);
1007 g_return_if_fail(bom_len != 0);
1008
1009 /* use filedata->len here because the contents are already converted into UTF-8 */
1010 buffer->len -= bom_len;
1011 /* overwrite the BOM with the remainder of the file contents, plus the NULL terminator. */
1012 g_memmove(buffer->data, buffer->data + bom_len, buffer->len + 1);
1013 buffer->data = g_realloc(buffer->data, buffer->len + 1);
1014 }
1015
1016
1017 /* loads textfile data, verifies and converts to forced_enc or UTF-8. Also handles BOM. */
1018 static gboolean handle_buffer(BufferData *buffer, const gchar *forced_enc)
1019 {
1020 GeanyEncodingIndex tmp_enc_idx;
1021
1022 /* temporarily retrieve the encoding idx based on the BOM to suppress the following warning
1023 * if we have a BOM */
1024 tmp_enc_idx = encodings_scan_unicode_bom(buffer->data, buffer->size, NULL);
1025
1026 /* check whether the size of the loaded data is equal to the size of the file in the
1027 * filesystem file size may be 0 to allow opening files in /proc/ which have typically a
1028 * file size of 0 bytes */
1029 if (buffer->len != buffer->size && buffer->size != 0 && (
1030 tmp_enc_idx == GEANY_ENCODING_UTF_8 || /* tmp_enc_idx can be UTF-7/8/16/32, UCS and None */
1031 tmp_enc_idx == GEANY_ENCODING_UTF_7)) /* filter UTF-7/8 where no NULL bytes are allowed */
1032 {
1033 buffer->partial = TRUE;
1034 }
1035
1036 /* Determine character encoding and convert to UTF-8 */
1037 if (forced_enc != NULL)
1038 {
1039 /* the encoding should be ignored(requested by user), so open the file "as it is" */
1040 if (utils_str_equal(forced_enc, encodings[GEANY_ENCODING_NONE].charset))
1041 {
1042 buffer->bom = FALSE;
1043 buffer->enc = g_strdup(encodings[GEANY_ENCODING_NONE].charset);
1044 }
1045 else if (! handle_forced_encoding(buffer, forced_enc))
1046 {
1047 return FALSE;
1048 }
1049 }
1050 else if (! handle_encoding(buffer, tmp_enc_idx))
1051 {
1052 return FALSE;
1053 }
1054
1055 if (buffer->bom)
1056 handle_bom(buffer);
1057 return TRUE;
1058 }
1059
1060
1061 /*
1062 * Tries to convert @a buffer into UTF-8 encoding. Unlike encodings_convert_to_utf8()
1063 * and encodings_convert_to_utf8_from_charset() it handles the possible BOM in the data.
1064 *
1065 * @param buf a pointer to modifiable null-terminated buffer to convert.
1066 * It may or may not be modified, and should be freed whatever happens.
1067 * @param size a pointer to the size of the buffer (expected to be e.g. the on-disk
1068 * file size). It will be updated to the new size.
1069 * @param forced_enc forced encoding to use, or @c NULL
1070 * @param used_encoding return location for the actually used encoding, or @c NULL
1071 * @param has_bom return location to store whether the data had a BOM, or @c NULL
1072 * @param partial return location to store whether the conversion may be partial, or @c NULL
1073 *
1074 * @return @C TRUE if the conversion succeeded, @c FALSE otherwise.
1075 */
1076 gboolean encodings_convert_to_utf8_auto(gchar **buf, gsize *size, const gchar *forced_enc,
1077 gchar **used_encoding, gboolean *has_bom, gboolean *partial)
1078 {
1079 BufferData buffer;
1080
1081 buffer.data = *buf;
1082 buffer.size = *size;
1083 /* use strlen to check for null chars */
1084 buffer.len = strlen(buffer.data);
1085 buffer.enc = NULL;
1086 buffer.bom = FALSE;
1087 buffer.partial = FALSE;
1088
1089 if (! handle_buffer(&buffer, forced_enc))
1090 return FALSE;
1091
1092 *size = buffer.len;
1093 if (used_encoding)
1094 *used_encoding = buffer.enc;
1095 else
1096 g_free(buffer.enc);
1097 if (has_bom)
1098 *has_bom = buffer.bom;
1099 if (partial)
1100 *partial = buffer.partial;
1101
1102 *buf = buffer.data;
1103 return TRUE;
1104 }
A fast and lightweight IDE