1 /* Copyright (c) 2006, Nokia Corporation
4 * Redistribution and use in source and binary forms, with or without
5 * modification, are permitted provided that the following conditions are
8 * * Redistributions of source code must retain the above copyright
9 * notice, this list of conditions and the following disclaimer.
10 * * Redistributions in binary form must reproduce the above copyright
11 * notice, this list of conditions and the following disclaimer in the
12 * documentation and/or other materials provided with the distribution.
13 * * Neither the name of the Nokia Corporation nor the names of its
14 * contributors may be used to endorse or promote products derived from
15 * this software without specific prior written permission.
17 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS
18 * IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
19 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
20 * PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER
21 * OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
22 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
23 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
24 * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
25 * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
26 * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
27 * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
34 #endif /*_GNU_SOURCE*/
35 #include <string.h> /* for strcasestr */
40 #include <glib/gi18n.h>
42 #include <modest-tny-platform-factory.h>
43 #include <modest-text-utils.h>
44 #include <modest-runtime.h>
49 #endif /*HAVE_CONFIG_H */
52 #define FORWARD_STRING _("mcen_ia_editor_original_message")
53 #define FROM_STRING _("mail_va_from")
54 #define SENT_STRING _("mcen_fi_message_properties_sent")
55 #define TO_STRING _("mail_va_to")
56 #define SUBJECT_STRING _("mail_va_subject")
57 #define EMPTY_STRING ""
60 * do the hyperlinkification only for texts < 50 Kb,
61 * as it's quite slow. Without this, e.g. mail with
62 * an uuencoded part (which is not recognized as attachment,
65 #define HYPERLINKIFY_MAX_LENGTH (1024*50)
70 * we need these regexps to find URLs in plain text e-mails
72 typedef struct _url_match_pattern_t url_match_pattern_t;
73 struct _url_match_pattern_t {
79 typedef struct _url_match_t url_match_t;
88 * we mark the ampersand with \007 when converting text->html
89 * because after text->html we do hyperlink detecting, which
90 * could be screwed up by the ampersand.
91 * ie. 1<3 ==> 1\007lt;3
93 #define MARK_AMP '\007'
94 #define MARK_AMP_STR "\007"
96 /* mark & separately, because they are parts of urls.
97 * ie. a&b => a\006amp;b, but a>b => a\007gt;b
99 * we need to handle '&' separately, because it can be part of URIs
100 * (as in href="http://foo.bar?a=1&b=1"), so inside those URIs
101 * we need to re-replace \006amp; with '&' again, while outside uri's
104 * yes, it's messy, but a consequence of doing text->html first, then hyperlinkify
106 #define MARK_AMP_URI '\006'
107 #define MARK_AMP_URI_STR "\006"
110 /* note: match MARK_AMP_URI_STR as well, because after txt->html, a '&' will look like $(MARK_AMP_URI_STR)"amp;" */
111 #define MAIL_VIEWER_URL_MATCH_PATTERNS { \
112 { "(file|rtsp|http|ftp|https|mms|mmsh|rtsp|rdp|lastfm)://[-a-z0-9_$.+!*(),;:@%=?/~#" MARK_AMP_URI_STR \
113 "]+[-a-z0-9_$%" MARK_AMP_URI_STR "=?/~#]", \
115 { "www\\.[-a-z0-9_$.+!*(),;:@%=?/~#" MARK_AMP_URI_STR "]+[-a-z0-9_$%" MARK_AMP_URI_STR "=?/~#]",\
117 { "ftp\\.[-a-z0-9_$.+!*(),;:@%=?/~#" MARK_AMP_URI_STR "]+[-a-z0-9_$%" MARK_AMP_URI_STR "=?/~#]",\
119 { "(jabberto|voipto|sipto|sip|chatto|xmpp):[-_a-z@0-9.+]+", \
121 { "mailto:[-_a-z0-9.\\+]+@[-_a-z0-9.]+", \
123 { "[-_a-z0-9.\\+]+@[-_a-z0-9.]+",\
127 const gchar account_title_forbidden_chars[] = {
128 '\\', '/', ':', '*', '?', '\'', '<', '>', '|', '^'
130 const gchar folder_name_forbidden_chars[] = {
131 '<', '>', ':', '\'', '/', '\\', '|', '?', '*', '^', '%', '$', '#', '&'
133 const gchar user_name_forbidden_chars[] = {
136 const guint ACCOUNT_TITLE_FORBIDDEN_CHARS_LENGTH = G_N_ELEMENTS (account_title_forbidden_chars);
137 const guint FOLDER_NAME_FORBIDDEN_CHARS_LENGTH = G_N_ELEMENTS (folder_name_forbidden_chars);
138 const guint USER_NAME_FORBIDDEN_CHARS_LENGTH = G_N_ELEMENTS (user_name_forbidden_chars);
141 static gchar* cite (const time_t sent_date, const gchar *from);
142 static void hyperlinkify_plain_text (GString *txt);
143 static gint cmp_offsets_reverse (const url_match_t *match1, const url_match_t *match2);
144 static GSList* get_url_matches (GString *txt);
146 static GString* get_next_line (const char *b, const gsize blen, const gchar * iter);
147 static int get_indent_level (const char *l);
148 static void unquote_line (GString * l);
149 static void append_quoted (GString * buf, const int indent, const GString * str,
151 static int get_breakpoint_utf8 (const gchar * s, const gint indent, const gint limit);
152 static int get_breakpoint_ascii (const gchar * s, const gint indent, const gint limit);
153 static int get_breakpoint (const gchar * s, const gint indent, const gint limit);
155 static gchar* modest_text_utils_quote_plain_text (const gchar *text,
157 const gchar *signature,
161 static gchar* modest_text_utils_quote_html (const gchar *text,
163 const gchar *signature,
166 static gchar* get_email_from_address (const gchar *address);
169 /* ******************************************************************* */
170 /* ************************* PUBLIC FUNCTIONS ************************ */
171 /* ******************************************************************* */
174 modest_text_utils_quote (const gchar *text,
175 const gchar *content_type,
176 const gchar *signature,
178 const time_t sent_date,
182 gchar *retval, *cited;
184 g_return_val_if_fail (text, NULL);
185 g_return_val_if_fail (content_type, NULL);
187 cited = cite (sent_date, from);
189 if (content_type && strcmp (content_type, "text/html") == 0)
190 /* TODO: extract the <body> of the HTML and pass it to
192 retval = modest_text_utils_quote_html (text, cited, signature, attachments, limit);
194 retval = modest_text_utils_quote_plain_text (text, cited, signature, attachments, limit);
203 modest_text_utils_cite (const gchar *text,
204 const gchar *content_type,
205 const gchar *signature,
212 g_return_val_if_fail (text, NULL);
213 g_return_val_if_fail (content_type, NULL);
216 retval = g_strdup ("");
217 else if (strcmp(content_type, "text/html") == 0) {
218 tmp_sig = g_strconcat ("\n", signature, NULL);
219 retval = modest_text_utils_convert_to_html_body(tmp_sig, -1, TRUE);
222 retval = g_strconcat (text, "\n", signature, NULL);
229 forward_cite (const gchar *from,
232 const gchar *subject)
234 g_return_val_if_fail (sent, NULL);
236 return g_strdup_printf ("%s\n%s %s\n%s %s\n%s %s\n%s %s\n",
238 FROM_STRING, (from)?from:"",
240 TO_STRING, (to)?to:"",
241 SUBJECT_STRING, (subject)?subject:"");
245 modest_text_utils_inline (const gchar *text,
246 const gchar *content_type,
247 const gchar *signature,
251 const gchar *subject)
257 g_return_val_if_fail (text, NULL);
258 g_return_val_if_fail (content_type, NULL);
260 modest_text_utils_strftime (sent_str, 100, "%c", sent_date);
262 cited = forward_cite (from, sent_str, to, subject);
264 if (content_type && strcmp (content_type, "text/html") == 0)
265 retval = modest_text_utils_quote_html (text, cited, signature, NULL, 80);
267 retval = modest_text_utils_quote_plain_text (text, cited, signature, NULL, 80);
273 /* just to prevent warnings:
274 * warning: `%x' yields only last 2 digits of year in some locales
277 modest_text_utils_strftime(char *s, gsize max, const char *fmt, time_t timet)
281 /* does not work on old maemo glib:
282 * g_date_set_time_t (&date, timet);
284 localtime_r (&timet, &tm);
285 return strftime(s, max, fmt, &tm);
289 modest_text_utils_derived_subject (const gchar *subject, const gchar *prefix)
293 g_return_val_if_fail (prefix, NULL);
295 if (!subject || subject[0] == '\0')
296 subject = _("mail_va_no_subject");
298 tmp = g_strchug (g_strdup (subject));
300 if (!strncmp (tmp, prefix, strlen (prefix))) {
304 return g_strdup_printf ("%s %s", prefix, subject);
309 modest_text_utils_remove_address (const gchar *address_list, const gchar *address)
311 gchar *dup, *token, *ptr = NULL, *result;
312 GString *filtered_emails;
313 gchar *email_address;
315 g_return_val_if_fail (address_list, NULL);
318 return g_strdup (address_list);
320 email_address = get_email_from_address (address);
322 /* search for substring */
323 if (!strstr ((const char *) address_list, (const char *) email_address)) {
324 g_free (email_address);
325 return g_strdup (address_list);
328 dup = g_strdup (address_list);
329 filtered_emails = g_string_new (NULL);
331 token = strtok_r (dup, ",", &ptr);
333 while (token != NULL) {
334 /* Add to list if not found */
335 if (!strstr ((const char *) token, (const char *) email_address)) {
336 if (filtered_emails->len == 0)
337 g_string_append_printf (filtered_emails, "%s", g_strstrip (token));
339 g_string_append_printf (filtered_emails, ",%s", g_strstrip (token));
341 token = strtok_r (NULL, ",", &ptr);
343 result = filtered_emails->str;
346 g_free (email_address);
348 g_string_free (filtered_emails, FALSE);
355 modest_text_utils_remove_duplicate_addresses (const gchar *address_list)
357 GSList *addresses, *cursor;
361 g_return_val_if_fail (address_list, NULL);
363 table = g_hash_table_new (g_str_hash, g_str_equal);
364 addresses = modest_text_utils_split_addresses_list (address_list);
366 new_list = g_strdup("");
369 const gchar* address = (const gchar*)cursor->data;
371 /* ignore the address if already seen */
372 if (g_hash_table_lookup (table, address) == 0) {
374 gchar *tmp = g_strjoin (",", new_list, address, NULL);
378 g_hash_table_insert (table, (gchar*)address, GINT_TO_POINTER(1));
380 cursor = g_slist_next (cursor);
383 g_hash_table_destroy (table);
384 g_slist_foreach (addresses, (GFunc)g_free, NULL);
385 g_slist_free (addresses);
392 modest_text_utils_convert_buffer_to_html_start (GString *html, const gchar *data, gssize n)
395 gboolean space_seen = FALSE;
396 guint break_dist = 0; /* distance since last break point */
401 /* replace with special html chars where needed*/
402 for (i = 0; i != n; ++i) {
403 guchar kar = data[i];
405 if (space_seen && kar != ' ') {
406 g_string_append_c (html, ' ');
410 /* we artificially insert a breakpoint (newline)
411 * after 256, to make sure our lines are not so long
412 * they will DOS the regexping later
413 * Also, check that kar is ASCII to make sure that we
414 * don't break a UTF8 char in two
416 if (++break_dist >= 256 && kar < 127) {
417 g_string_append_c (html, '\n');
425 /* this is a temp place holder for '&'; we can only
426 * set the real '&' after hyperlink translation, otherwise
427 * we might screw that up */
428 break; /* ignore embedded \0s and MARK_AMP */
429 case '<' : g_string_append (html, MARK_AMP_STR "lt;"); break;
430 case '>' : g_string_append (html, MARK_AMP_STR "gt;"); break;
431 case '&' : g_string_append (html, MARK_AMP_URI_STR "amp;"); break; /* special case */
432 case '"' : g_string_append (html, MARK_AMP_STR "quot;"); break;
434 /* don't convert ' --> wpeditor will try to re-convert it... */
435 //case '\'' : g_string_append (html, "'"); break;
436 case '\n' : g_string_append (html, "<br>\n");break_dist= 0; break;
437 case '\t' : g_string_append (html, MARK_AMP_STR "nbsp;" MARK_AMP_STR "nbsp;" MARK_AMP_STR "nbsp; ");
438 break_dist=0; break; /* note the space at the end*/
441 if (space_seen) { /* second space in a row */
442 g_string_append (html, " ");
448 g_string_append_c (html, kar);
455 modest_text_utils_convert_buffer_to_html_finish (GString *html)
458 /* replace all our MARK_AMPs with real ones */
459 for (i = 0; i != html->len; ++i)
460 if ((html->str)[i] == MARK_AMP || (html->str)[i] == MARK_AMP_URI)
461 (html->str)[i] = '&';
466 modest_text_utils_convert_to_html (const gchar *data)
471 g_return_val_if_fail (data, NULL);
477 html = g_string_sized_new (1.5 * len); /* just a guess... */
479 g_string_append_printf (html,
481 "<meta http-equiv=\"content-type\" content=\"text/html; charset=utf8\">"
485 modest_text_utils_convert_buffer_to_html_start (html, data, -1);
487 g_string_append (html, "</body></html>");
489 if (len <= HYPERLINKIFY_MAX_LENGTH)
490 hyperlinkify_plain_text (html);
492 modest_text_utils_convert_buffer_to_html_finish (html);
494 return g_string_free (html, FALSE);
498 modest_text_utils_convert_to_html_body (const gchar *data, gssize n, gboolean hyperlinkify)
502 g_return_val_if_fail (data, NULL);
509 html = g_string_sized_new (1.5 * n); /* just a guess... */
511 modest_text_utils_convert_buffer_to_html_start (html, data, n);
513 if (hyperlinkify && (n < HYPERLINKIFY_MAX_LENGTH))
514 hyperlinkify_plain_text (html);
516 modest_text_utils_convert_buffer_to_html_finish (html);
518 return g_string_free (html, FALSE);
522 modest_text_utils_get_addresses_indexes (const gchar *addresses, GSList **start_indexes, GSList **end_indexes)
524 gchar *current, *start, *last_blank;
525 gint start_offset = 0, current_offset = 0;
527 g_return_if_fail (start_indexes != NULL);
528 g_return_if_fail (end_indexes != NULL);
530 start = (gchar *) addresses;
534 while (*current != '\0') {
535 if ((start == current)&&((*current == ' ')||(*current == ',')||(*current == ';'))) {
536 start = g_utf8_next_char (start);
538 last_blank = current;
539 } else if ((*current == ',')||(*current == ';')) {
540 gint *start_index, *end_index;
541 start_index = g_new0(gint, 1);
542 end_index = g_new0(gint, 1);
543 *start_index = start_offset;
544 *end_index = current_offset;
545 *start_indexes = g_slist_prepend (*start_indexes, start_index);
546 *end_indexes = g_slist_prepend (*end_indexes, end_index);
547 start = g_utf8_next_char (current);
548 start_offset = current_offset + 1;
550 } else if (*current == '"') {
551 current = g_utf8_next_char (current);
553 while ((*current != '"')&&(*current != '\0')) {
554 current = g_utf8_next_char (current);
559 current = g_utf8_next_char (current);
563 if (start != current) {
564 gint *start_index, *end_index;
565 start_index = g_new0(gint, 1);
566 end_index = g_new0(gint, 1);
567 *start_index = start_offset;
568 *end_index = current_offset;
569 *start_indexes = g_slist_prepend (*start_indexes, start_index);
570 *end_indexes = g_slist_prepend (*end_indexes, end_index);
573 *start_indexes = g_slist_reverse (*start_indexes);
574 *end_indexes = g_slist_reverse (*end_indexes);
581 modest_text_utils_split_addresses_list (const gchar *addresses)
584 const gchar *my_addrs = addresses;
588 g_return_val_if_fail (addresses, NULL);
590 /* skip any space, ',', ';' at the start */
591 while (my_addrs && (my_addrs[0] == ' ' || my_addrs[0] == ',' || my_addrs[0] == ';'))
594 /* are we at the end of addresses list? */
598 /* nope, we are at the start of some address
599 * now, let's find the end of the address */
601 while (end[0] && end[0] != ',' && end[0] != ';')
604 /* we got the address; copy it and remove trailing whitespace */
605 addr = g_strndup (my_addrs, end - my_addrs);
608 head = g_slist_append (NULL, addr);
609 head->next = modest_text_utils_split_addresses_list (end); /* recurse */
616 modest_text_utils_address_range_at_position (const gchar *recipients_list,
621 gchar *current = NULL;
622 gint range_start = 0;
625 gboolean is_quoted = FALSE;
627 g_return_if_fail (recipients_list);
628 g_return_if_fail (position < g_utf8_strlen(recipients_list, -1));
631 for (current = (gchar *) recipients_list; *current != '\0';
632 current = g_utf8_find_next_char (current, NULL)) {
633 gunichar c = g_utf8_get_char (current);
635 if ((c == ',') && (!is_quoted)) {
636 if (index < position) {
637 range_start = index + 1;
641 } else if (c == '\"') {
642 is_quoted = !is_quoted;
643 } else if ((c == ' ') &&(range_start == index)) {
651 *start = range_start;
657 /* ******************************************************************* */
658 /* ************************* UTILIY FUNCTIONS ************************ */
659 /* ******************************************************************* */
662 get_next_line (const gchar * b, const gsize blen, const gchar * iter)
668 return g_string_new("");
676 gs = g_string_new_len (i0, iter - i0);
680 get_indent_level (const char *l)
697 /* if we hit the signature marker "-- ", we return -(indent + 1). This
698 * stops reformatting.
700 if (strcmp (l, "-- ") == 0) {
708 unquote_line (GString * l)
723 g_string_erase (l, 0, p - l->str);
727 append_quoted (GString * buf, int indent, const GString * str,
732 indent = indent < 0 ? abs (indent) - 1 : indent;
733 for (i = 0; i <= indent; i++) {
734 g_string_append (buf, "> ");
737 g_string_append_len (buf, str->str, cutpoint);
739 g_string_append (buf, str->str);
741 g_string_append (buf, "\n");
745 get_breakpoint_utf8 (const gchar * s, gint indent, const gint limit)
748 const gchar *pos, *last;
751 indent = indent < 0 ? abs (indent) - 1 : indent;
755 uni = g_utf8_to_ucs4_fast (s, -1, NULL);
757 if ((index + 2 * indent > limit) && last) {
761 if (g_unichar_isspace (uni[index])) {
764 pos = g_utf8_next_char (pos);
772 get_breakpoint_ascii (const gchar * s, const gint indent, const gint limit)
777 if (last + 2 * indent < limit)
780 for (i = strlen (s); i > 0; i--) {
782 if (i + 2 * indent <= limit) {
793 get_breakpoint (const gchar * s, const gint indent, const gint limit)
796 if (g_utf8_validate (s, -1, NULL)) {
797 return get_breakpoint_utf8 (s, indent, limit);
798 } else { /* assume ASCII */
799 //g_warning("invalid UTF-8 in msg");
800 return get_breakpoint_ascii (s, indent, limit);
805 cite (const time_t sent_date, const gchar *from)
807 return g_strdup (_("mcen_ia_editor_original_message"));
811 quoted_attachments (GList *attachments)
814 GString *result = g_string_new ("");
815 for (node = attachments; node != NULL; node = g_list_next (node)) {
816 gchar *filename = (gchar *) node->data;
817 g_string_append_printf ( result, "%s %s\n", _("mcen_ia_editor_attach_filename"), filename);
820 return g_string_free (result, FALSE);
825 modest_text_utils_quote_plain_text (const gchar *text,
827 const gchar *signature,
832 gint indent, breakpoint, rem_indent = 0;
833 GString *q, *l, *remaining;
835 gchar *attachments_string = NULL;
837 /* remaining will store the rest of the line if we have to break it */
838 q = g_string_new ("\n");
839 q = g_string_append (q, cite);
840 q = g_string_append_c (q, '\n');
841 remaining = g_string_new ("");
846 l = get_next_line (text, len, iter);
847 iter = iter + l->len + 1;
848 indent = get_indent_level (l->str);
851 if (remaining->len) {
852 if (l->len && indent == rem_indent) {
853 g_string_prepend (l, " ");
854 g_string_prepend (l, remaining->str);
858 get_breakpoint (remaining->str,
861 append_quoted (q, rem_indent,
862 remaining, breakpoint);
863 g_string_erase (remaining, 0,
865 if (remaining->str[0] == ' ') {
866 g_string_erase (remaining, 0,
869 } while (remaining->len);
872 g_string_free (remaining, TRUE);
873 breakpoint = get_breakpoint (l->str, indent, limit);
874 remaining = g_string_new (l->str + breakpoint);
875 if (remaining->str[0] == ' ') {
876 g_string_erase (remaining, 0, 1);
879 append_quoted (q, indent, l, breakpoint);
880 g_string_free (l, TRUE);
881 } while ((iter < text + len) || (remaining->str[0]));
883 attachments_string = quoted_attachments (attachments);
884 q = g_string_append (q, attachments_string);
885 g_free (attachments_string);
887 if (signature != NULL) {
888 q = g_string_append_c (q, '\n');
889 q = g_string_append (q, signature);
892 return g_string_free (q, FALSE);
896 modest_text_utils_quote_html (const gchar *text,
898 const gchar *signature,
902 gchar *result = NULL;
903 gchar *signature_result = NULL;
904 const gchar *format = \
905 "<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\">\n" \
909 "<pre>%s<br/>%s<br/>%s</pre>\n" \
912 gchar *attachments_string = NULL;
913 gchar *q_attachments_string = NULL;
914 gchar *q_cite = NULL;
915 gchar *html_text = NULL;
917 if (signature == NULL)
918 signature_result = g_strdup ("");
920 signature_result = modest_text_utils_convert_to_html_body (signature, -1, TRUE);
922 attachments_string = quoted_attachments (attachments);
923 q_attachments_string = modest_text_utils_convert_to_html_body (attachments_string, -1, TRUE);
924 q_cite = modest_text_utils_convert_to_html_body (cite, -1, TRUE);
925 html_text = modest_text_utils_convert_to_html_body (text, -1, TRUE);
926 result = g_strdup_printf (format, signature_result, q_cite, html_text, q_attachments_string);
929 g_free (attachments_string);
930 g_free (q_attachments_string);
931 g_free (signature_result);
937 cmp_offsets_reverse (const url_match_t *match1, const url_match_t *match2)
939 return match2->offset - match1->offset;
942 static gboolean url_matches_block = 0;
943 static url_match_pattern_t patterns[] = MAIL_VIEWER_URL_MATCH_PATTERNS;
950 const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
951 for (i = 0; i != pattern_num; ++i) {
952 patterns[i].preg = g_slice_new0 (regex_t);
954 /* this should not happen */
955 if (regcomp (patterns[i].preg, patterns[i].regex,
956 REG_ICASE|REG_EXTENDED|REG_NEWLINE) != 0) {
957 g_warning ("%s: error in regexp:\n%s\n", __FUNCTION__, patterns[i].regex);
968 const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
969 for (i = 0; i != pattern_num; ++i) {
970 regfree (patterns[i].preg);
971 g_slice_free (regex_t, patterns[i].preg);
972 } /* don't free patterns itself -- it's static */
976 modest_text_utils_hyperlinkify_begin (void)
978 if (url_matches_block == 0)
980 url_matches_block ++;
984 modest_text_utils_hyperlinkify_end (void)
987 if (url_matches_block <= 0)
993 get_url_matches (GString *txt)
996 guint rv, i, offset = 0;
997 GSList *match_list = NULL;
999 const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
1001 /* initalize the regexps */
1002 modest_text_utils_hyperlinkify_begin ();
1004 /* find all the matches */
1005 for (i = 0; i != pattern_num; ++i) {
1009 gboolean is_submatch;
1012 if ((rv = regexec (patterns[i].preg, txt->str + offset, 1, &rm, 0)) != 0) {
1013 g_return_val_if_fail (rv == REG_NOMATCH, NULL); /* this should not happen */
1014 break; /* try next regexp */
1019 is_submatch = FALSE;
1020 /* check old matches to see if this has already been matched */
1021 cursor = match_list;
1022 while (cursor && !is_submatch) {
1023 const url_match_t *old_match =
1024 (const url_match_t *) cursor->data;
1025 guint new_offset = offset + rm.rm_so;
1026 is_submatch = (new_offset > old_match->offset &&
1027 new_offset < old_match->offset + old_match->len);
1028 cursor = g_slist_next (cursor);
1032 /* make a list of our matches (<offset, len, prefix> tupels)*/
1033 match = g_slice_new (url_match_t);
1034 match->offset = offset + rm.rm_so;
1035 match->len = rm.rm_eo - rm.rm_so;
1036 match->prefix = patterns[i].prefix;
1037 match_list = g_slist_prepend (match_list, match);
1043 modest_text_utils_hyperlinkify_end ();
1045 /* now sort the list, so the matches are in reverse order of occurence.
1046 * that way, we can do the replacements starting from the end, so we don't need
1047 * to recalculate the offsets
1049 match_list = g_slist_sort (match_list,
1050 (GCompareFunc)cmp_offsets_reverse);
1056 /* replace all occurences of needle in haystack with repl*/
1058 replace_string (const gchar *haystack, const gchar *needle, gchar repl)
1060 gchar *str, *cursor;
1062 if (!haystack || !needle || strlen(needle) == 0)
1063 return haystack ? g_strdup(haystack) : NULL;
1065 str = g_strdup (haystack);
1067 for (cursor = str; cursor && *cursor; ++cursor) {
1068 if (g_str_has_prefix (cursor, needle)) {
1070 memmove (cursor + 1,
1071 cursor + strlen (needle),
1072 strlen (cursor + strlen (needle)) + 1);
1080 hyperlinkify_plain_text (GString *txt)
1083 GSList *match_list = get_url_matches (txt);
1085 /* we will work backwards, so the offsets stay valid */
1086 for (cursor = match_list; cursor; cursor = cursor->next) {
1088 url_match_t *match = (url_match_t*) cursor->data;
1089 gchar *url = g_strndup (txt->str + match->offset, match->len);
1090 gchar *repl = NULL; /* replacement */
1092 /* the string still contains $(MARK_AMP_URI_STR)"amp;" for each
1093 * '&' in the original, because of the text->html conversion.
1094 * in the href-URL (and only there), we must convert that back to
1097 gchar *href_url = replace_string (url, MARK_AMP_URI_STR "amp;", '&');
1099 /* the prefix is NULL: use the one that is already there */
1100 repl = g_strdup_printf ("<a href=\"%s%s\">%s</a>",
1101 match->prefix ? match->prefix : EMPTY_STRING,
1104 /* replace the old thing with our hyperlink
1105 * replacement thing */
1106 g_string_erase (txt, match->offset, match->len);
1107 g_string_insert (txt, match->offset, repl);
1113 g_slice_free (url_match_t, match);
1116 g_slist_free (match_list);
1120 /* for optimization reasons, we change the string in-place */
1122 modest_text_utils_get_display_address (gchar *address)
1126 g_return_if_fail (address);
1131 /* should not be needed, and otherwise, we probably won't screw up the address
1132 * more than it already is :)
1133 * g_return_val_if_fail (g_utf8_validate (address, -1, NULL), NULL);
1136 /* remove leading whitespace */
1137 if (address[0] == ' ')
1138 g_strchug (address);
1140 for (i = 0; address[i]; ++i) {
1141 if (address[i] == '<') {
1142 if (G_UNLIKELY(i == 0))
1143 return; /* there's nothing else, leave it */
1145 address[i] = '\0'; /* terminate the string here */
1157 modest_text_utils_get_email_address (const gchar *full_address)
1159 const gchar *left, *right;
1161 g_return_val_if_fail (full_address, NULL);
1166 g_return_val_if_fail (g_utf8_validate (full_address, -1, NULL), NULL);
1168 left = g_strrstr_len (full_address, strlen(full_address), "<");
1170 return g_strdup (full_address);
1172 right = g_strstr_len (left, strlen(left), ">");
1174 return g_strdup (full_address);
1176 return g_strndup (left + 1, right - left - 1);
1180 modest_text_utils_get_subject_prefix_len (const gchar *sub)
1182 gint prefix_len = 0;
1184 g_return_val_if_fail (sub, 0);
1189 /* optimization: "Re", "RE", "re","Fwd", "FWD", "fwd","FW","Fw", "fw" */
1190 if (sub[0] != 'R' && sub[0] != 'F' && sub[0] != 'r' && sub[0] != 'f')
1192 else if (sub[0] && sub[1] != 'e' && sub[1] != 'E' && sub[1] != 'w' && sub[1] != 'W')
1199 /* skip over a [...] block */
1200 if (sub[prefix_len] == '[') {
1201 int c = prefix_len + 1;
1202 while (sub[c] && sub[c] != ']')
1205 return 0; /* no end to the ']' found */
1210 /* did we find the ':' ? */
1211 if (sub[prefix_len] == ':') {
1213 if (sub[prefix_len] == ' ')
1215 prefix_len += modest_text_utils_get_subject_prefix_len (sub + prefix_len);
1216 /* g_warning ("['%s','%s']", sub, (char*) sub + prefix_len); */
1224 modest_text_utils_utf8_strcmp (const gchar* s1, const gchar *s2, gboolean insensitive)
1227 /* work even when s1 and/or s2 == NULL */
1228 if (G_UNLIKELY(s1 == s2))
1230 if (G_UNLIKELY(!s1))
1232 if (G_UNLIKELY(!s2))
1235 /* if it's not case sensitive */
1238 /* optimization: shortcut if first char is ascii */
1239 if (((s1[0] & 0xf0)== 0) && ((s2[0] & 0xf0) == 0))
1240 return s1[0] - s2[0];
1242 return g_utf8_collate (s1, s2);
1248 /* optimization: short cut iif first char is ascii */
1249 if (((s1[0] & 0xf0) == 0) && ((s2[0] & 0xf0) == 0))
1250 return tolower(s1[0]) - tolower(s2[0]);
1252 n1 = g_utf8_strdown (s1, -1);
1253 n2 = g_utf8_strdown (s2, -1);
1255 result = g_utf8_collate (n1, n2);
1266 modest_text_utils_get_display_date (time_t date)
1268 #define DATE_BUF_SIZE 64
1269 static gchar date_buf[DATE_BUF_SIZE];
1271 /* calculate the # of days since epoch for
1272 * for today and for the date provided
1273 * based on idea from pvanhoof */
1274 int day = time(NULL) / (24 * 60 * 60);
1275 int date_day = date / (24 * 60 * 60);
1277 /* if it's today, show the time, if it's not today, show the date instead */
1279 if (day == date_day) /* is the date today? */
1280 modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, "%X", date);
1282 modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, "%x", date);
1284 return date_buf; /* this is a static buffer, don't free! */
1290 modest_text_utils_validate_folder_name (const gchar *folder_name)
1292 /* based on http://msdn2.microsoft.com/en-us/library/aa365247.aspx,
1293 * with some extras */
1297 const gchar **cursor = NULL;
1298 const gchar *forbidden_names[] = { /* windows does not like these */
1299 "CON", "PRN", "AUX", "NUL", "COM1", "COM2", "COM3", "COM4", "COM5", "COM6",
1300 "COM7", "COM8", "COM9", "LPT1", "LPT2", "LPT3", "LPT4", "LPT5", "LPT6", "LPT7", "LPT8", "LPT9",
1304 /* cannot be NULL */
1308 /* cannot be empty */
1309 len = strlen(folder_name);
1313 /* cannot start or end with a space */
1314 if (g_ascii_isspace(folder_name[0]) || g_ascii_isspace(folder_name[len - 1]))
1317 /* cannot contain a forbidden char */
1318 for (i = 0; i < len; i++)
1319 if (modest_text_utils_is_forbidden_char (folder_name[i], FOLDER_NAME_FORBIDDEN_CHARS))
1322 /* cannot contain a forbidden word */
1324 for (cursor = forbidden_names; cursor && *cursor; ++cursor) {
1325 if (g_ascii_strcasecmp (folder_name, *cursor) == 0)
1330 return TRUE; /* it's valid! */
1336 modest_text_utils_validate_domain_name (const gchar *domain)
1338 gboolean valid = FALSE;
1340 const gchar* domain_regex = "^([a-z0-9-]*[a-z0-9]\\.)+[a-z0-9-]*[a-z0-9]$";
1342 g_return_val_if_fail (domain, FALSE);
1347 memset (&rx, 0, sizeof(regex_t)); /* coverity wants this... */
1349 /* domain name: all alphanum or '-' or '.',
1350 * but beginning/ending in alphanum */
1351 if (regcomp (&rx, domain_regex, REG_ICASE|REG_EXTENDED|REG_NOSUB)) {
1352 g_warning ("BUG: error in regexp");
1356 valid = (regexec (&rx, domain, 1, NULL, 0) == 0);
1365 modest_text_utils_validate_email_address (const gchar *email_address,
1366 const gchar **invalid_char_position)
1369 const gchar *c = NULL, *domain = NULL;
1370 static gchar *rfc822_specials = "()<>@,;:\\\"[]&";
1372 if (invalid_char_position)
1373 *invalid_char_position = NULL;
1375 g_return_val_if_fail (email_address, FALSE);
1377 /* check that the email adress contains exactly one @ */
1378 if (!strstr(email_address, "@") ||
1379 (strstr(email_address, "@") != g_strrstr(email_address, "@")))
1382 /* first we validate the name portion (name@domain) */
1383 for (c = email_address; *c; c++) {
1385 (c == email_address ||
1387 *(c - 1) == '\"')) {
1391 if (*c == '\\' && (*++c == ' '))
1393 if (*c <= ' ' || *c >= 127)
1406 if (*c <= ' ' || *c >= 127)
1408 if (strchr(rfc822_specials, *c)) {
1409 if (invalid_char_position)
1410 *invalid_char_position = c;
1414 if (c == email_address || *(c - 1) == '.')
1417 /* next we validate the domain portion (name@domain) */
1418 if (!*(domain = ++c))
1422 if (c == domain || *(c - 1) == '.' || *(c + 1) == '\0')
1426 if (*c <= ' ' || *c >= 127)
1428 if (strchr(rfc822_specials, *c)) {
1429 if (invalid_char_position)
1430 *invalid_char_position = c;
1435 return (count >= 1) ? TRUE : FALSE;
1439 modest_text_utils_validate_recipient (const gchar *recipient, const gchar **invalid_char_position)
1441 gchar *stripped, *current;
1443 gboolean has_error = FALSE;
1445 if (invalid_char_position)
1446 *invalid_char_position = NULL;
1448 g_return_val_if_fail (recipient, FALSE);
1450 if (modest_text_utils_validate_email_address (recipient, invalid_char_position))
1453 stripped = g_strdup (recipient);
1454 stripped = g_strstrip (stripped);
1457 if (*current == '\0') {
1463 if (*current == '\"') {
1464 current = g_utf8_next_char (current);
1466 for (; *current != '\0'; current = g_utf8_next_char (current)) {
1467 if (*current == '\\') {
1468 /* TODO: This causes a warning, which breaks the build,
1469 * because a gchar cannot be < 0.
1471 if (current[1] <0) {
1476 } else if (*current == '\"') {
1478 current = g_utf8_next_char (current);
1484 for (current = stripped ; *current != '\0'; current = g_utf8_next_char (current)) {
1485 if (*current == '<') {
1497 right_part = g_strdup (current);
1499 right_part = g_strstrip (right_part);
1501 if (g_str_has_prefix (right_part, "<") &&
1502 g_str_has_suffix (right_part, ">")) {
1506 address = g_strndup (right_part+1, strlen (right_part) - 2);
1507 g_free (right_part);
1508 valid = modest_text_utils_validate_email_address (address, invalid_char_position);
1512 g_free (right_part);
1519 modest_text_utils_get_display_size (guint64 size)
1521 const guint KB=1024;
1522 const guint MB=1024 * KB;
1523 const guint GB=1024 * MB;
1526 return g_strdup_printf(_FM("sfil_li_size_kb"), 0);
1527 if (0 < size && size < KB)
1528 return g_strdup_printf (_FM("sfil_li_size_kb"), 1);
1529 else if (KB <= size && size < 100 * KB)
1530 return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), size / KB);
1531 else if (100*KB <= size && size < MB)
1532 return g_strdup_printf (_FM("sfil_li_size_100kb_1mb"), (float) size / MB);
1533 else if (MB <= size && size < 10*MB)
1534 return g_strdup_printf (_FM("sfil_li_size_1mb_10mb"), (float) size / MB);
1535 else if (10*MB <= size && size < GB)
1536 return g_strdup_printf (_FM("sfil_li_size_10mb_1gb"), size / MB);
1538 return g_strdup_printf (_FM("sfil_li_size_1gb_or_greater"), (float) size / GB);
1542 get_email_from_address (const gchar * address)
1544 gchar *left_limit, *right_limit;
1546 left_limit = strstr (address, "<");
1547 right_limit = g_strrstr (address, ">");
1549 if ((left_limit == NULL)||(right_limit == NULL)|| (left_limit > right_limit))
1550 return g_strdup (address);
1552 return g_strndup (left_limit + 1, (right_limit - left_limit) - 1);
1556 modest_text_utils_get_color_string (GdkColor *color)
1558 g_return_val_if_fail (color, NULL);
1560 return g_strdup_printf ("#%x%x%x%x%x%x%x%x%x%x%x%x",
1561 (color->red >> 12) & 0xf, (color->red >> 8) & 0xf,
1562 (color->red >> 4) & 0xf, (color->red) & 0xf,
1563 (color->green >> 12) & 0xf, (color->green >> 8) & 0xf,
1564 (color->green >> 4) & 0xf, (color->green) & 0xf,
1565 (color->blue >> 12) & 0xf, (color->blue >> 8) & 0xf,
1566 (color->blue >> 4) & 0xf, (color->blue) & 0xf);
1570 modest_text_utils_text_buffer_get_text (GtkTextBuffer *buffer)
1572 GtkTextIter start, end;
1573 gchar *slice, *current;
1574 GString *result = g_string_new ("");
1576 g_return_val_if_fail (buffer && GTK_IS_TEXT_BUFFER (buffer), NULL);
1578 gtk_text_buffer_get_start_iter (buffer, &start);
1579 gtk_text_buffer_get_end_iter (buffer, &end);
1581 slice = gtk_text_buffer_get_slice (buffer, &start, &end, FALSE);
1584 while (current && current != '\0') {
1585 if (g_utf8_get_char (current) == 0xFFFC) {
1586 result = g_string_append_c (result, ' ');
1587 current = g_utf8_next_char (current);
1589 gchar *next = g_utf8_strchr (current, -1, 0xFFFC);
1591 result = g_string_append (result, current);
1593 result = g_string_append_len (result, current, next - current);
1600 return g_string_free (result, FALSE);
1605 modest_text_utils_is_forbidden_char (const gchar character,
1606 ModestTextUtilsForbiddenCharType type)
1609 const gchar *forbidden_chars = NULL;
1611 /* We need to get the length in the switch because the
1612 compiler needs to know the size at compile time */
1614 case ACCOUNT_TITLE_FORBIDDEN_CHARS:
1615 forbidden_chars = account_title_forbidden_chars;
1616 len = G_N_ELEMENTS (account_title_forbidden_chars);
1618 case FOLDER_NAME_FORBIDDEN_CHARS:
1619 forbidden_chars = folder_name_forbidden_chars;
1620 len = G_N_ELEMENTS (folder_name_forbidden_chars);
1622 case USER_NAME_FORBIDDEN_NAMES:
1623 forbidden_chars = user_name_forbidden_chars;
1624 len = G_N_ELEMENTS (user_name_forbidden_chars);
1627 g_return_val_if_reached (TRUE);
1630 for (i = 0; i < len ; i++)
1631 if (forbidden_chars[i] == character)
1634 return FALSE; /* it's valid! */
1638 modest_text_utils_label_get_selection (GtkLabel *label)
1643 if (gtk_label_get_selection_bounds (GTK_LABEL (label), &start, &end)) {
1644 const gchar *start_offset;
1645 const gchar *end_offset;
1646 start_offset = gtk_label_get_text (GTK_LABEL (label));
1647 start_offset = g_utf8_offset_to_pointer (start_offset, start);
1648 end_offset = gtk_label_get_text (GTK_LABEL (label));
1649 end_offset = g_utf8_offset_to_pointer (end_offset, end);
1650 selection = g_strndup (start_offset, end_offset - start_offset);
1653 return g_strdup ("");
1658 _forward_search_image_char (gunichar ch,
1661 return (ch == 0xFFFC);
1665 modest_text_utils_buffer_selection_is_valid (GtkTextBuffer *buffer)
1668 GtkTextIter start, end;
1670 g_return_val_if_fail (GTK_IS_TEXT_BUFFER (buffer), FALSE);
1672 result = gtk_text_buffer_get_has_selection (GTK_TEXT_BUFFER (buffer));
1674 /* check there are no images in selection */
1676 gtk_text_buffer_get_selection_bounds (buffer, &start, &end);
1677 if (gtk_text_iter_get_char (&start)== 0xFFFC)
1680 gtk_text_iter_backward_char (&end);
1681 if (gtk_text_iter_forward_find_char (&start, _forward_search_image_char,