1 /*
2 * encodings.c - this file is part of Geany, a fast and lightweight IDE
3 *
4 * Copyright 2005 The Geany contributors
5 *
6 * This program is free software; you can redistribute it and/or modify
7 * it under the terms of the GNU General Public License as published by
8 * the Free Software Foundation; either version 2 of the License, or
9 * (at your option) any later version.
10 *
11 * This program is distributed in the hope that it will be useful,
12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 * GNU General Public License for more details.
15 *
16 * You should have received a copy of the GNU General Public License along
17 * with this program; if not, write to the Free Software Foundation, Inc.,
18 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
19 */
20
21 /*
22 * Encoding conversion and Byte Order Mark (BOM) handling.
23 */
24
25 /*
26 * Modified by the gedit Team, 2002. See the gedit AUTHORS file for a
27 * list of people on the gedit Team.
28 * See the gedit ChangeLog files for a list of changes.
29 */
30 /* Stolen from anjuta */
31
32 #ifdef HAVE_CONFIG_H
33 # include "config.h"
34 #endif
35
36 #include "encodings.h"
37 #include "encodingsprivate.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
init_encodings(void)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. */
encodings_charset_equals(const gchar * a,const gchar * b)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
encodings_get_idx_from_charset(const gchar * charset)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
encodings_get_from_charset(const gchar * charset)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
encodings_normalize_charset(const gchar * charset)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
encodings_get_from_index(gint idx)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 @nullable The charset according to idx, or @c NULL if the index is invalid.
263 *
264 * @since 0.13
265 **/
266 GEANY_API_SYMBOL
encodings_get_charset_from_index(gint idx)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
encodings_to_string(const GeanyEncoding * enc)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
encodings_get_charset(const GeanyEncoding * enc)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
encodings_select_radio_item(const gchar * charset)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 */
regex_compile(const gchar * pattern)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
regex_match(GRegex * preg,const gchar * buffer,gsize size)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
encodings_radio_item_change_cb(GtkCheckMenuItem * menuitem,gpointer user_data)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
encodings_reload_radio_item_change_cb(GtkMenuItem * menuitem,gpointer user_data)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
encodings_finalize(void)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
encodings_init(void)405 void encodings_init(void)
406 {
407 GtkWidget *menu[2];
408 GCallback cb_func[2];
409 gint group_sizes[GEANY_ENCODING_GROUPS_MAX] = { 0 };
410 const gchar *const groups[GEANY_ENCODING_GROUPS_MAX] =
411 {
412 [NONE] = NULL,
413 [WESTEUROPEAN] = N_("_West European"),
414 [EASTEUROPEAN] = N_("_East European"),
415 [EASTASIAN] = N_("East _Asian"),
416 [ASIAN] = N_("_SE & SW Asian"),
417 [MIDDLEEASTERN] = N_("_Middle Eastern"),
418 [UNICODE] = N_("_Unicode"),
419 };
420
421 init_encodings();
422
423 if (! pregs_loaded)
424 {
425 pregs[0] = regex_compile(PATTERN_HTMLMETA);
426 pregs[1] = regex_compile(PATTERN_CODING);
427 pregs_loaded = TRUE;
428 }
429
430 /* create encodings submenu in document menu */
431 menu[0] = ui_lookup_widget(main_widgets.window, "set_encoding1_menu");
432 menu[1] = ui_lookup_widget(main_widgets.window, "menu_reload_as1_menu");
433 cb_func[0] = G_CALLBACK(encodings_radio_item_change_cb);
434 cb_func[1] = G_CALLBACK(encodings_reload_radio_item_change_cb);
435
436 for (guint i = 0; i < G_N_ELEMENTS(encodings); i++)
437 group_sizes[encodings[i].group]++;
438
439 for (guint k = 0; k < 2; k++)
440 {
441 GSList *group = NULL;
442 GtkWidget *submenus[GEANY_ENCODING_GROUPS_MAX];
443 gint orders[GEANY_ENCODING_GROUPS_MAX] = { 0 };
444 guint n_added = 0;
445
446 for (guint i = 0; i < GEANY_ENCODING_GROUPS_MAX; i++)
447 {
448 if (! groups[i]) /* NONE */
449 submenus[i] = menu[k];
450 else
451 {
452 GtkWidget *item = gtk_menu_item_new_with_mnemonic(_(groups[i]));
453 submenus[i] = gtk_menu_new();
454 gtk_menu_item_set_submenu(GTK_MENU_ITEM(item), submenus[i]);
455 gtk_container_add(GTK_CONTAINER(menu[k]), item);
456 gtk_widget_show_all(item);
457 }
458 }
459
460 /** TODO can it be optimized? ATM 882 runs at line "if (encodings[i].order ...)" */
461 do
462 {
463 for (guint i = 0; i < G_N_ELEMENTS(encodings); i++)
464 {
465 if (encodings[i].order == orders[encodings[i].group])
466 {
467 GtkWidget *item;
468 gchar *label = encodings_to_string(&encodings[i]);
469
470 if (k == 0) /* Set Encoding menu */
471 {
472 item = gtk_radio_menu_item_new_with_label(group, label);
473 group = gtk_radio_menu_item_get_group(GTK_RADIO_MENU_ITEM(item));
474 radio_items[i] = item;
475 }
476 else
477 item = gtk_menu_item_new_with_label(label);
478 gtk_widget_show(item);
479 gtk_container_add(GTK_CONTAINER(submenus[encodings[i].group]), item);
480 g_signal_connect(item, "activate", cb_func[k],
481 (gpointer) encodings[i].charset);
482 g_free(label);
483
484 orders[encodings[i].group]++;
485 n_added++;
486 }
487 }
488 }
489 while (n_added < G_N_ELEMENTS(encodings));
490 }
491 }
492
493
encoding_combo_store_sort_func(GtkTreeModel * model,GtkTreeIter * a,GtkTreeIter * b,gpointer data)494 static gint encoding_combo_store_sort_func(GtkTreeModel *model,
495 GtkTreeIter *a,
496 GtkTreeIter *b,
497 gpointer data)
498 {
499 gboolean a_has_child = gtk_tree_model_iter_has_child(model, a);
500 gboolean b_has_child = gtk_tree_model_iter_has_child(model, b);
501 gchar *a_string;
502 gchar *b_string;
503 gint cmp_res;
504
505 if (a_has_child != b_has_child)
506 return a_has_child ? -1 : 1;
507
508 gtk_tree_model_get(model, a, 1, &a_string, -1);
509 gtk_tree_model_get(model, b, 1, &b_string, -1);
510 cmp_res = strcmp(a_string, b_string);
511 g_free(a_string);
512 g_free(b_string);
513 return cmp_res;
514 }
515
516
encodings_encoding_store_new(gboolean has_detect)517 GtkTreeStore *encodings_encoding_store_new(gboolean has_detect)
518 {
519 GtkTreeStore *store;
520 GtkTreeIter iter_current, iter_westeuro, iter_easteuro, iter_eastasian,
521 iter_asian, iter_utf8, iter_middleeast;
522 GtkTreeIter *iter_parent;
523 gint i;
524
525 store = gtk_tree_store_new(2, G_TYPE_INT, G_TYPE_STRING);
526
527 if (has_detect)
528 {
529 gtk_tree_store_append(store, &iter_current, NULL);
530 gtk_tree_store_set(store, &iter_current, 0, GEANY_ENCODINGS_MAX, 1, _("Detect from file"), -1);
531 }
532
533 gtk_tree_store_append(store, &iter_westeuro, NULL);
534 gtk_tree_store_set(store, &iter_westeuro, 0, -1, 1, _("West European"), -1);
535 gtk_tree_store_append(store, &iter_easteuro, NULL);
536 gtk_tree_store_set(store, &iter_easteuro, 0, -1, 1, _("East European"), -1);
537 gtk_tree_store_append(store, &iter_eastasian, NULL);
538 gtk_tree_store_set(store, &iter_eastasian, 0, -1, 1, _("East Asian"), -1);
539 gtk_tree_store_append(store, &iter_asian, NULL);
540 gtk_tree_store_set(store, &iter_asian, 0, -1, 1, _("SE & SW Asian"), -1);
541 gtk_tree_store_append(store, &iter_middleeast, NULL);
542 gtk_tree_store_set(store, &iter_middleeast, 0, -1, 1, _("Middle Eastern"), -1);
543 gtk_tree_store_append(store, &iter_utf8, NULL);
544 gtk_tree_store_set(store, &iter_utf8, 0, -1, 1, _("Unicode"), -1);
545
546 for (i = 0; i < GEANY_ENCODINGS_MAX; i++)
547 {
548 gchar *encoding_string;
549
550 switch (encodings[i].group)
551 {
552 case WESTEUROPEAN: iter_parent = &iter_westeuro; break;
553 case EASTEUROPEAN: iter_parent = &iter_easteuro; break;
554 case EASTASIAN: iter_parent = &iter_eastasian; break;
555 case ASIAN: iter_parent = &iter_asian; break;
556 case MIDDLEEASTERN: iter_parent = &iter_middleeast; break;
557 case UNICODE: iter_parent = &iter_utf8; break;
558 case NONE:
559 default: iter_parent = NULL;
560 }
561 gtk_tree_store_append(store, &iter_current, iter_parent);
562 encoding_string = encodings_to_string(&encodings[i]);
563 gtk_tree_store_set(store, &iter_current, 0, i, 1, encoding_string, -1);
564 g_free(encoding_string);
565 }
566
567 gtk_tree_sortable_set_sort_column_id(GTK_TREE_SORTABLE(store), 1, GTK_SORT_ASCENDING);
568 gtk_tree_sortable_set_sort_func(GTK_TREE_SORTABLE(store), 1, encoding_combo_store_sort_func, NULL, NULL);
569
570 return store;
571 }
572
573
encodings_encoding_store_get_encoding(GtkTreeStore * store,GtkTreeIter * iter)574 gint encodings_encoding_store_get_encoding(GtkTreeStore *store, GtkTreeIter *iter)
575 {
576 gint enc;
577 gtk_tree_model_get(GTK_TREE_MODEL(store), iter, 0, &enc, -1);
578 return enc;
579 }
580
581
encodings_encoding_store_get_iter(GtkTreeStore * store,GtkTreeIter * iter,gint enc)582 gboolean encodings_encoding_store_get_iter(GtkTreeStore *store, GtkTreeIter *iter, gint enc)
583 {
584 if (gtk_tree_model_get_iter_first(GTK_TREE_MODEL(store), iter))
585 {
586 do
587 {
588 if (encodings_encoding_store_get_encoding(store, iter) == enc)
589 return TRUE;
590 }
591 while (ui_tree_model_iter_any_next(GTK_TREE_MODEL(store), iter, TRUE));
592 }
593 return FALSE;
594 }
595
596
encodings_encoding_store_cell_data_func(GtkCellLayout * cell_layout,GtkCellRenderer * cell,GtkTreeModel * tree_model,GtkTreeIter * iter,gpointer data)597 void encodings_encoding_store_cell_data_func(GtkCellLayout *cell_layout,
598 GtkCellRenderer *cell,
599 GtkTreeModel *tree_model,
600 GtkTreeIter *iter,
601 gpointer data)
602 {
603 gboolean sensitive = !gtk_tree_model_iter_has_child(tree_model, iter);
604 gchar *text;
605
606 gtk_tree_model_get(tree_model, iter, 1, &text, -1);
607 g_object_set(cell, "sensitive", sensitive, "text", text, NULL);
608 g_free(text);
609 }
610
611
612 /**
613 * Tries to convert @a buffer into UTF-8 encoding from the encoding specified with @a charset.
614 * If @a fast is not set, additional checks to validate the converted string are performed.
615 *
616 * @param buffer The input string to convert.
617 * @param size The length of the string, or -1 if the string is nul-terminated.
618 * @param charset The charset to be used for conversion.
619 * @param fast @c TRUE to only convert the input and skip extended checks on the converted string.
620 *
621 * @return If the conversion was successful, a newly allocated nul-terminated string,
622 * which must be freed with @c g_free(). Otherwise @c NULL.
623 **/
624 GEANY_API_SYMBOL
encodings_convert_to_utf8_from_charset(const gchar * buffer,gssize size,const gchar * charset,gboolean fast)625 gchar *encodings_convert_to_utf8_from_charset(const gchar *buffer, gssize size,
626 const gchar *charset, gboolean fast)
627 {
628 gchar *utf8_content = NULL;
629 GError *conv_error = NULL;
630 gchar* converted_contents = NULL;
631 gsize bytes_written;
632
633 g_return_val_if_fail(buffer != NULL, NULL);
634 g_return_val_if_fail(charset != NULL, NULL);
635
636 converted_contents = g_convert(buffer, size, "UTF-8", charset, NULL,
637 &bytes_written, &conv_error);
638
639 if (fast)
640 {
641 utf8_content = converted_contents;
642 if (conv_error != NULL) g_error_free(conv_error);
643 }
644 else if (conv_error != NULL || ! g_utf8_validate(converted_contents, bytes_written, NULL))
645 {
646 if (conv_error != NULL)
647 {
648 geany_debug("Couldn't convert from %s to UTF-8 (%s).", charset, conv_error->message);
649 g_error_free(conv_error);
650 conv_error = NULL;
651 }
652 else
653 geany_debug("Couldn't convert from %s to UTF-8.", charset);
654
655 utf8_content = NULL;
656 g_free(converted_contents);
657 }
658 else
659 {
660 geany_debug("Converted from %s to UTF-8.", charset);
661 utf8_content = converted_contents;
662 }
663
664 return utf8_content;
665 }
666
667
encodings_check_regexes(const gchar * buffer,gsize size)668 static gchar *encodings_check_regexes(const gchar *buffer, gsize size)
669 {
670 guint i;
671
672 for (i = 0; i < G_N_ELEMENTS(pregs); i++)
673 {
674 gchar *charset;
675
676 if ((charset = regex_match(pregs[i], buffer, size)) != NULL)
677 return charset;
678 }
679 return NULL;
680 }
681
682
encodings_convert_to_utf8_with_suggestion(const gchar * buffer,gssize size,const gchar * suggested_charset,gchar ** used_encoding)683 static gchar *encodings_convert_to_utf8_with_suggestion(const gchar *buffer, gssize size,
684 const gchar *suggested_charset, gchar **used_encoding)
685 {
686 const gchar *locale_charset = NULL;
687 const gchar *charset;
688 gchar *utf8_content;
689 gboolean check_suggestion = suggested_charset != NULL;
690 gboolean check_locale = FALSE;
691 gint i, preferred_charset;
692
693 if (size == -1)
694 {
695 size = strlen(buffer);
696 }
697
698 /* current locale is not UTF-8, we have to check this charset */
699 check_locale = ! g_get_charset(&locale_charset);
700
701 /* First check for preferred charset, if specified */
702 preferred_charset = file_prefs.default_open_encoding;
703
704 if (preferred_charset == (gint) encodings[GEANY_ENCODING_NONE].idx ||
705 preferred_charset < 0 ||
706 preferred_charset >= GEANY_ENCODINGS_MAX)
707 {
708 preferred_charset = -1;
709 }
710
711 /* -1 means "Preferred charset" */
712 for (i = -1; i < GEANY_ENCODINGS_MAX; i++)
713 {
714 if (G_UNLIKELY(i == (gint) encodings[GEANY_ENCODING_NONE].idx))
715 continue;
716
717 if (check_suggestion)
718 {
719 check_suggestion = FALSE;
720 charset = encodings_normalize_charset(suggested_charset);
721 if (charset == NULL) /* we failed at normalizing suggested encoding, try it as is */
722 charset = suggested_charset;
723 i = -2; /* keep i below the start value to have it again at -1 on the next loop run */
724 }
725 else if (check_locale)
726 {
727 check_locale = FALSE;
728 charset = locale_charset;
729 i = -2; /* keep i below the start value to have it again at -1 on the next loop run */
730 }
731 else if (i == -1)
732 {
733 if (preferred_charset >= 0)
734 {
735 charset = encodings[preferred_charset].charset;
736 geany_debug("Using preferred charset: %s", charset);
737 }
738 else
739 continue;
740 }
741 else if (i >= 0)
742 charset = encodings[i].charset;
743 else /* in this case we have i == -2, continue to increase i and go ahead */
744 continue;
745
746 if (G_UNLIKELY(charset == NULL))
747 continue;
748
749 geany_debug("Trying to convert %" G_GSIZE_FORMAT " bytes of data from %s into UTF-8.",
750 size, charset);
751 utf8_content = encodings_convert_to_utf8_from_charset(buffer, size, charset, FALSE);
752
753 if (G_LIKELY(utf8_content != NULL))
754 {
755 if (used_encoding != NULL)
756 {
757 if (G_UNLIKELY(*used_encoding != NULL))
758 {
759 geany_debug("%s:%d", __FILE__, __LINE__);
760 g_free(*used_encoding);
761 }
762 *used_encoding = g_strdup(charset);
763 }
764 return utf8_content;
765 }
766 }
767
768 return NULL;
769 }
770
771
772 /**
773 * Tries to convert @a buffer into UTF-8 encoding and store the detected original encoding in
774 * @a used_encoding.
775 *
776 * @param buffer the input string to convert.
777 * @param size the length of the string, or -1 if the string is nul-terminated.
778 * @param used_encoding @out @optional return location of the detected encoding of the input string, or @c NULL.
779 *
780 * @return @nullable If the conversion was successful, a newly allocated nul-terminated string,
781 * which must be freed with @c g_free(). Otherwise @c NULL.
782 **/
783 GEANY_API_SYMBOL
encodings_convert_to_utf8(const gchar * buffer,gssize size,gchar ** used_encoding)784 gchar *encodings_convert_to_utf8(const gchar *buffer, gssize size, gchar **used_encoding)
785 {
786 gchar *regex_charset;
787 gchar *utf8;
788
789 /* first try to read the encoding from the file content */
790 regex_charset = encodings_check_regexes(buffer, size);
791 utf8 = encodings_convert_to_utf8_with_suggestion(buffer, size, regex_charset, used_encoding);
792 g_free(regex_charset);
793
794 return utf8;
795 }
796
797
798 /* If there's a BOM, return a corresponding GEANY_ENCODING_UTF_* index,
799 * otherwise GEANY_ENCODING_NONE.
800 * */
encodings_scan_unicode_bom(const gchar * string,gsize len,guint * bom_len)801 GeanyEncodingIndex encodings_scan_unicode_bom(const gchar *string, gsize len, guint *bom_len)
802 {
803 if (len >= 3)
804 {
805 if (bom_len)
806 *bom_len = 3;
807
808 if ((guchar)string[0] == 0xef && (guchar)string[1] == 0xbb &&
809 (guchar)string[2] == 0xbf)
810 {
811 return GEANY_ENCODING_UTF_8;
812 }
813 }
814 if (len >= 4)
815 {
816 if (bom_len)
817 *bom_len = 4;
818
819 if ((guchar)string[0] == 0x00 && (guchar)string[1] == 0x00 &&
820 (guchar)string[2] == 0xfe && (guchar)string[3] == 0xff)
821 {
822 return GEANY_ENCODING_UTF_32BE; /* Big endian */
823 }
824 if ((guchar)string[0] == 0xff && (guchar)string[1] == 0xfe &&
825 (guchar)string[2] == 0x00 && (guchar)string[3] == 0x00)
826 {
827 return GEANY_ENCODING_UTF_32LE; /* Little endian */
828 }
829 if ((string[0] == 0x2b && string[1] == 0x2f && string[2] == 0x76) &&
830 (string[3] == 0x38 || string[3] == 0x39 || string[3] == 0x2b || string[3] == 0x2f))
831 {
832 return GEANY_ENCODING_UTF_7;
833 }
834 }
835 if (len >= 2)
836 {
837 if (bom_len)
838 *bom_len = 2;
839
840 if ((guchar)string[0] == 0xfe && (guchar)string[1] == 0xff)
841 {
842 return GEANY_ENCODING_UTF_16BE; /* Big endian */
843 }
844 if ((guchar)string[0] == 0xff && (guchar)string[1] == 0xfe)
845 {
846 return GEANY_ENCODING_UTF_16LE; /* Little endian */
847 }
848 }
849 if (bom_len)
850 *bom_len = 0;
851 return GEANY_ENCODING_NONE;
852 }
853
854
encodings_is_unicode_charset(const gchar * string)855 gboolean encodings_is_unicode_charset(const gchar *string)
856 {
857 if (string != NULL &&
858 (strncmp(string, "UTF", 3) == 0 || strncmp(string, "UCS", 3) == 0))
859 {
860 return TRUE;
861 }
862 return FALSE;
863 }
864
865
866 typedef struct
867 {
868 gchar *data; /* null-terminated data */
869 gsize size; /* actual data size */
870 gsize len; /* string length of data */
871 gchar *enc;
872 gboolean bom;
873 gboolean partial;
874 } BufferData;
875
876
877 /* convert data with the specified encoding */
878 static gboolean
handle_forced_encoding(BufferData * buffer,const gchar * forced_enc)879 handle_forced_encoding(BufferData *buffer, const gchar *forced_enc)
880 {
881 GeanyEncodingIndex enc_idx;
882
883 if (utils_str_equal(forced_enc, "UTF-8"))
884 {
885 if (! g_utf8_validate(buffer->data, buffer->len, NULL))
886 {
887 return FALSE;
888 }
889 }
890 else
891 {
892 gchar *converted_text = encodings_convert_to_utf8_from_charset(
893 buffer->data, buffer->size, forced_enc, FALSE);
894 if (converted_text == NULL)
895 {
896 return FALSE;
897 }
898 else
899 {
900 SETPTR(buffer->data, converted_text);
901 buffer->len = strlen(converted_text);
902 }
903 }
904 enc_idx = encodings_scan_unicode_bom(buffer->data, buffer->size, NULL);
905 buffer->bom = (enc_idx == GEANY_ENCODING_UTF_8);
906 buffer->enc = g_strdup(forced_enc);
907 return TRUE;
908 }
909
910
911 /* detect encoding and convert to UTF-8 if necessary */
912 static gboolean
handle_encoding(BufferData * buffer,GeanyEncodingIndex enc_idx)913 handle_encoding(BufferData *buffer, GeanyEncodingIndex enc_idx)
914 {
915 g_return_val_if_fail(buffer->enc == NULL, FALSE);
916 g_return_val_if_fail(buffer->bom == FALSE, FALSE);
917
918 if (buffer->size == 0)
919 {
920 /* we have no data so assume UTF-8, buffer->len can be 0 even we have an empty
921 * e.g. UTF32 file with a BOM(so size is 4, len is 0) */
922 buffer->enc = g_strdup("UTF-8");
923 }
924 else
925 {
926 /* first check for a BOM */
927 if (enc_idx != GEANY_ENCODING_NONE)
928 {
929 buffer->enc = g_strdup(encodings[enc_idx].charset);
930 buffer->bom = TRUE;
931
932 if (enc_idx != GEANY_ENCODING_UTF_8) /* the BOM indicated something else than UTF-8 */
933 {
934 gchar *converted_text = encodings_convert_to_utf8_from_charset(
935 buffer->data, buffer->size, buffer->enc, FALSE);
936 if (converted_text != NULL)
937 {
938 SETPTR(buffer->data, converted_text);
939 buffer->len = strlen(converted_text);
940 }
941 else
942 {
943 /* there was a problem converting data from BOM encoding type */
944 SETPTR(buffer->enc, NULL);
945 buffer->bom = FALSE;
946 }
947 }
948 }
949
950 if (buffer->enc == NULL) /* either there was no BOM or the BOM encoding failed */
951 {
952 /* first try to read the encoding from the file content */
953 gchar *regex_charset = encodings_check_regexes(buffer->data, buffer->size);
954
955 /* try UTF-8 first */
956 if (encodings_get_idx_from_charset(regex_charset) == GEANY_ENCODING_UTF_8 &&
957 (buffer->size == buffer->len) && g_utf8_validate(buffer->data, buffer->len, NULL))
958 {
959 buffer->enc = g_strdup("UTF-8");
960 }
961 else
962 {
963 /* detect the encoding */
964 gchar *converted_text = encodings_convert_to_utf8_with_suggestion(buffer->data,
965 buffer->size, regex_charset, &buffer->enc);
966
967 if (converted_text == NULL)
968 {
969 g_free(regex_charset);
970 return FALSE;
971 }
972 SETPTR(buffer->data, converted_text);
973 buffer->len = strlen(converted_text);
974 }
975 g_free(regex_charset);
976 }
977 }
978 return TRUE;
979 }
980
981
982 static void
handle_bom(BufferData * buffer)983 handle_bom(BufferData *buffer)
984 {
985 guint bom_len;
986
987 encodings_scan_unicode_bom(buffer->data, buffer->size, &bom_len);
988 g_return_if_fail(bom_len != 0);
989
990 /* use filedata->len here because the contents are already converted into UTF-8 */
991 buffer->len -= bom_len;
992 /* overwrite the BOM with the remainder of the file contents, plus the NULL terminator. */
993 memmove(buffer->data, buffer->data + bom_len, buffer->len + 1);
994 buffer->data = g_realloc(buffer->data, buffer->len + 1);
995 }
996
997
998 /* loads textfile data, verifies and converts to forced_enc or UTF-8. Also handles BOM. */
handle_buffer(BufferData * buffer,const gchar * forced_enc)999 static gboolean handle_buffer(BufferData *buffer, const gchar *forced_enc)
1000 {
1001 GeanyEncodingIndex tmp_enc_idx;
1002
1003 /* temporarily retrieve the encoding idx based on the BOM to suppress the following warning
1004 * if we have a BOM */
1005 tmp_enc_idx = encodings_scan_unicode_bom(buffer->data, buffer->size, NULL);
1006
1007 /* check whether the size of the loaded data is equal to the size of the file in the
1008 * filesystem file size may be 0 to allow opening files in /proc/ which have typically a
1009 * file size of 0 bytes */
1010 if (buffer->len != buffer->size && buffer->size != 0 && (
1011 tmp_enc_idx == GEANY_ENCODING_UTF_8 || /* tmp_enc_idx can be UTF-7/8/16/32, UCS and None */
1012 tmp_enc_idx == GEANY_ENCODING_UTF_7)) /* filter UTF-7/8 where no NULL bytes are allowed */
1013 {
1014 buffer->partial = TRUE;
1015 }
1016
1017 /* Determine character encoding and convert to UTF-8 */
1018 if (forced_enc != NULL)
1019 {
1020 /* the encoding should be ignored(requested by user), so open the file "as it is" */
1021 if (utils_str_equal(forced_enc, encodings[GEANY_ENCODING_NONE].charset))
1022 {
1023 buffer->bom = FALSE;
1024 buffer->enc = g_strdup(encodings[GEANY_ENCODING_NONE].charset);
1025 }
1026 else if (! handle_forced_encoding(buffer, forced_enc))
1027 {
1028 return FALSE;
1029 }
1030 }
1031 else if (! handle_encoding(buffer, tmp_enc_idx))
1032 {
1033 return FALSE;
1034 }
1035
1036 if (buffer->bom)
1037 handle_bom(buffer);
1038 return TRUE;
1039 }
1040
1041
1042 /*
1043 * Tries to convert @a buffer into UTF-8 encoding. Unlike encodings_convert_to_utf8()
1044 * and encodings_convert_to_utf8_from_charset() it handles the possible BOM in the data.
1045 *
1046 * @param buf a pointer to modifiable null-terminated buffer to convert.
1047 * It may or may not be modified, and should be freed whatever happens.
1048 * @param size a pointer to the size of the buffer (expected to be e.g. the on-disk
1049 * file size). It will be updated to the new size.
1050 * @param forced_enc forced encoding to use, or @c NULL
1051 * @param used_encoding return location for the actually used encoding, or @c NULL
1052 * @param has_bom return location to store whether the data had a BOM, or @c NULL
1053 * @param partial return location to store whether the conversion may be partial, or @c NULL
1054 *
1055 * @return @C TRUE if the conversion succeeded, @c FALSE otherwise.
1056 */
encodings_convert_to_utf8_auto(gchar ** buf,gsize * size,const gchar * forced_enc,gchar ** used_encoding,gboolean * has_bom,gboolean * partial)1057 gboolean encodings_convert_to_utf8_auto(gchar **buf, gsize *size, const gchar *forced_enc,
1058 gchar **used_encoding, gboolean *has_bom, gboolean *partial)
1059 {
1060 BufferData buffer;
1061
1062 buffer.data = *buf;
1063 buffer.size = *size;
1064 /* use strlen to check for null chars */
1065 buffer.len = strlen(buffer.data);
1066 buffer.enc = NULL;
1067 buffer.bom = FALSE;
1068 buffer.partial = FALSE;
1069
1070 if (! handle_buffer(&buffer, forced_enc))
1071 return FALSE;
1072
1073 *size = buffer.len;
1074 if (used_encoding)
1075 *used_encoding = buffer.enc;
1076 else
1077 g_free(buffer.enc);
1078 if (has_bom)
1079 *has_bom = buffer.bom;
1080 if (partial)
1081 *partial = buffer.partial;
1082
1083 *buf = buffer.data;
1084 return TRUE;
1085 }
1086