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*/
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
414 if (++break_dist == 256) {
415 g_string_append_c (html, '\n');
423 /* this is a temp place holder for '&'; we can only
424 * set the real '&' after hyperlink translation, otherwise
425 * we might screw that up */
426 break; /* ignore embedded \0s and MARK_AMP */
427 case '<' : g_string_append (html, MARK_AMP_STR "lt;"); break;
428 case '>' : g_string_append (html, MARK_AMP_STR "gt;"); break;
429 case '&' : g_string_append (html, MARK_AMP_URI_STR "amp;"); break; /* special case */
430 case '"' : g_string_append (html, MARK_AMP_STR "quot;"); break;
432 /* don't convert ' --> wpeditor will try to re-convert it... */
433 //case '\'' : g_string_append (html, "'"); break;
434 case '\n' : g_string_append (html, "<br>\n");break_dist= 0; break;
435 case '\t' : g_string_append (html, MARK_AMP_STR "nbsp;" MARK_AMP_STR "nbsp;" MARK_AMP_STR "nbsp; ");
436 break_dist=0; break; /* note the space at the end*/
439 if (space_seen) { /* second space in a row */
440 g_string_append (html, " ");
446 /* Optimization to copy single ascii
447 * characters faster */
448 if (kar > 31 && kar < 127) {
449 g_string_append_c (html, kar);
451 /* Important: copy full UTF-8 characters,
452 * don't copy them byte by byte */
453 gunichar c = g_utf8_get_char_validated (data+i, -1);
454 if (c != (gunichar) -1 && c != (gunichar) -2) {
455 const gchar *copyfrom = data + i;
456 int len = g_utf8_next_char(copyfrom) - copyfrom;
457 g_string_append_len (html, copyfrom, len);
460 g_warning ("%s: non-UTF8 byte found, skipping", __FUNCTION__);
470 modest_text_utils_convert_buffer_to_html_finish (GString *html)
473 /* replace all our MARK_AMPs with real ones */
474 for (i = 0; i != html->len; ++i)
475 if ((html->str)[i] == MARK_AMP || (html->str)[i] == MARK_AMP_URI)
476 (html->str)[i] = '&';
481 modest_text_utils_convert_to_html (const gchar *data)
486 g_return_val_if_fail (data, NULL);
492 html = g_string_sized_new (1.5 * len); /* just a guess... */
494 g_string_append_printf (html,
496 "<meta http-equiv=\"content-type\" content=\"text/html; charset=utf8\">"
500 modest_text_utils_convert_buffer_to_html_start (html, data, -1);
502 g_string_append (html, "</body></html>");
504 if (len <= HYPERLINKIFY_MAX_LENGTH)
505 hyperlinkify_plain_text (html);
507 modest_text_utils_convert_buffer_to_html_finish (html);
509 return g_string_free (html, FALSE);
513 modest_text_utils_convert_to_html_body (const gchar *data, gssize n, gboolean hyperlinkify)
517 g_return_val_if_fail (data, NULL);
524 html = g_string_sized_new (1.5 * n); /* just a guess... */
526 modest_text_utils_convert_buffer_to_html_start (html, data, n);
528 if (hyperlinkify && (n < HYPERLINKIFY_MAX_LENGTH))
529 hyperlinkify_plain_text (html);
531 modest_text_utils_convert_buffer_to_html_finish (html);
533 return g_string_free (html, FALSE);
537 modest_text_utils_get_addresses_indexes (const gchar *addresses, GSList **start_indexes, GSList **end_indexes)
539 gchar *current, *start, *last_blank;
540 gint start_offset = 0, current_offset = 0;
542 g_return_if_fail (start_indexes != NULL);
543 g_return_if_fail (end_indexes != NULL);
545 start = (gchar *) addresses;
549 while (*current != '\0') {
550 if ((start == current)&&((*current == ' ')||(*current == ',')||(*current == ';'))) {
551 start = g_utf8_next_char (start);
553 last_blank = current;
554 } else if ((*current == ',')||(*current == ';')) {
555 gint *start_index, *end_index;
556 start_index = g_new0(gint, 1);
557 end_index = g_new0(gint, 1);
558 *start_index = start_offset;
559 *end_index = current_offset;
560 *start_indexes = g_slist_prepend (*start_indexes, start_index);
561 *end_indexes = g_slist_prepend (*end_indexes, end_index);
562 start = g_utf8_next_char (current);
563 start_offset = current_offset + 1;
565 } else if (*current == '"') {
566 current = g_utf8_next_char (current);
568 while ((*current != '"')&&(*current != '\0')) {
569 current = g_utf8_next_char (current);
574 current = g_utf8_next_char (current);
578 if (start != current) {
579 gint *start_index, *end_index;
580 start_index = g_new0(gint, 1);
581 end_index = g_new0(gint, 1);
582 *start_index = start_offset;
583 *end_index = current_offset;
584 *start_indexes = g_slist_prepend (*start_indexes, start_index);
585 *end_indexes = g_slist_prepend (*end_indexes, end_index);
588 *start_indexes = g_slist_reverse (*start_indexes);
589 *end_indexes = g_slist_reverse (*end_indexes);
596 modest_text_utils_split_addresses_list (const gchar *addresses)
599 const gchar *my_addrs = addresses;
603 g_return_val_if_fail (addresses, NULL);
605 /* skip any space, ',', ';' at the start */
606 while (my_addrs && (my_addrs[0] == ' ' || my_addrs[0] == ',' || my_addrs[0] == ';'))
609 /* are we at the end of addresses list? */
613 /* nope, we are at the start of some address
614 * now, let's find the end of the address */
616 while (end[0] && end[0] != ',' && end[0] != ';')
619 /* we got the address; copy it and remove trailing whitespace */
620 addr = g_strndup (my_addrs, end - my_addrs);
623 head = g_slist_append (NULL, addr);
624 head->next = modest_text_utils_split_addresses_list (end); /* recurse */
631 modest_text_utils_address_range_at_position (const gchar *recipients_list,
636 gchar *current = NULL;
637 gint range_start = 0;
640 gboolean is_quoted = FALSE;
642 g_return_if_fail (recipients_list);
643 g_return_if_fail (position < g_utf8_strlen(recipients_list, -1));
646 for (current = (gchar *) recipients_list; *current != '\0';
647 current = g_utf8_find_next_char (current, NULL)) {
648 gunichar c = g_utf8_get_char (current);
650 if ((c == ',') && (!is_quoted)) {
651 if (index < position) {
652 range_start = index + 1;
656 } else if (c == '\"') {
657 is_quoted = !is_quoted;
658 } else if ((c == ' ') &&(range_start == index)) {
666 *start = range_start;
672 /* ******************************************************************* */
673 /* ************************* UTILIY FUNCTIONS ************************ */
674 /* ******************************************************************* */
677 get_next_line (const gchar * b, const gsize blen, const gchar * iter)
683 return g_string_new("");
691 gs = g_string_new_len (i0, iter - i0);
695 get_indent_level (const char *l)
712 /* if we hit the signature marker "-- ", we return -(indent + 1). This
713 * stops reformatting.
715 if (strcmp (l, "-- ") == 0) {
723 unquote_line (GString * l)
738 g_string_erase (l, 0, p - l->str);
742 append_quoted (GString * buf, int indent, const GString * str,
747 indent = indent < 0 ? abs (indent) - 1 : indent;
748 for (i = 0; i <= indent; i++) {
749 g_string_append (buf, "> ");
752 g_string_append_len (buf, str->str, cutpoint);
754 g_string_append (buf, str->str);
756 g_string_append (buf, "\n");
760 get_breakpoint_utf8 (const gchar * s, gint indent, const gint limit)
763 const gchar *pos, *last;
766 indent = indent < 0 ? abs (indent) - 1 : indent;
770 uni = g_utf8_to_ucs4_fast (s, -1, NULL);
772 if ((index + 2 * indent > limit) && last) {
776 if (g_unichar_isspace (uni[index])) {
779 pos = g_utf8_next_char (pos);
787 get_breakpoint_ascii (const gchar * s, const gint indent, const gint limit)
792 if (last + 2 * indent < limit)
795 for (i = strlen (s); i > 0; i--) {
797 if (i + 2 * indent <= limit) {
808 get_breakpoint (const gchar * s, const gint indent, const gint limit)
811 if (g_utf8_validate (s, -1, NULL)) {
812 return get_breakpoint_utf8 (s, indent, limit);
813 } else { /* assume ASCII */
814 //g_warning("invalid UTF-8 in msg");
815 return get_breakpoint_ascii (s, indent, limit);
820 cite (const time_t sent_date, const gchar *from)
822 return g_strdup (_("mcen_ia_editor_original_message"));
826 quoted_attachments (GList *attachments)
829 GString *result = g_string_new ("");
830 for (node = attachments; node != NULL; node = g_list_next (node)) {
831 gchar *filename = (gchar *) node->data;
832 g_string_append_printf ( result, "%s %s\n", _("mcen_ia_editor_attach_filename"), filename);
835 return g_string_free (result, FALSE);
840 modest_text_utils_quote_plain_text (const gchar *text,
842 const gchar *signature,
847 gint indent, breakpoint, rem_indent = 0;
848 GString *q, *l, *remaining;
850 gchar *attachments_string = NULL;
852 /* remaining will store the rest of the line if we have to break it */
853 q = g_string_new ("\n");
854 q = g_string_append (q, cite);
855 q = g_string_append_c (q, '\n');
856 remaining = g_string_new ("");
861 l = get_next_line (text, len, iter);
862 iter = iter + l->len + 1;
863 indent = get_indent_level (l->str);
866 if (remaining->len) {
867 if (l->len && indent == rem_indent) {
868 g_string_prepend (l, " ");
869 g_string_prepend (l, remaining->str);
873 get_breakpoint (remaining->str,
876 append_quoted (q, rem_indent,
877 remaining, breakpoint);
878 g_string_erase (remaining, 0,
880 if (remaining->str[0] == ' ') {
881 g_string_erase (remaining, 0,
884 } while (remaining->len);
887 g_string_free (remaining, TRUE);
888 breakpoint = get_breakpoint (l->str, indent, limit);
889 remaining = g_string_new (l->str + breakpoint);
890 if (remaining->str[0] == ' ') {
891 g_string_erase (remaining, 0, 1);
894 append_quoted (q, indent, l, breakpoint);
895 g_string_free (l, TRUE);
896 } while ((iter < text + len) || (remaining->str[0]));
898 attachments_string = quoted_attachments (attachments);
899 q = g_string_append (q, attachments_string);
900 g_free (attachments_string);
902 if (signature != NULL) {
903 q = g_string_append_c (q, '\n');
904 q = g_string_append (q, signature);
907 return g_string_free (q, FALSE);
911 modest_text_utils_quote_html (const gchar *text,
913 const gchar *signature,
917 gchar *result = NULL;
918 gchar *signature_result = NULL;
919 const gchar *format = \
920 "<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\">\n" \
924 "<pre>%s<br/>%s<br/>%s</pre>\n" \
927 gchar *attachments_string = NULL;
928 gchar *q_attachments_string = NULL;
929 gchar *q_cite = NULL;
930 gchar *html_text = NULL;
932 if (signature == NULL)
933 signature_result = g_strdup ("");
935 signature_result = modest_text_utils_convert_to_html_body (signature, -1, TRUE);
937 attachments_string = quoted_attachments (attachments);
938 q_attachments_string = modest_text_utils_convert_to_html_body (attachments_string, -1, TRUE);
939 q_cite = modest_text_utils_convert_to_html_body (cite, -1, TRUE);
940 html_text = modest_text_utils_convert_to_html_body (text, -1, TRUE);
941 result = g_strdup_printf (format, signature_result, q_cite, html_text, q_attachments_string);
944 g_free (attachments_string);
945 g_free (q_attachments_string);
946 g_free (signature_result);
952 cmp_offsets_reverse (const url_match_t *match1, const url_match_t *match2)
954 return match2->offset - match1->offset;
957 static gboolean url_matches_block = 0;
958 static url_match_pattern_t patterns[] = MAIL_VIEWER_URL_MATCH_PATTERNS;
965 const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
966 for (i = 0; i != pattern_num; ++i) {
967 patterns[i].preg = g_slice_new0 (regex_t);
969 /* this should not happen */
970 if (regcomp (patterns[i].preg, patterns[i].regex,
971 REG_ICASE|REG_EXTENDED|REG_NEWLINE) != 0) {
972 g_warning ("%s: error in regexp:\n%s\n", __FUNCTION__, patterns[i].regex);
983 const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
984 for (i = 0; i != pattern_num; ++i) {
985 regfree (patterns[i].preg);
986 g_slice_free (regex_t, patterns[i].preg);
987 } /* don't free patterns itself -- it's static */
991 modest_text_utils_hyperlinkify_begin (void)
993 if (url_matches_block == 0)
995 url_matches_block ++;
999 modest_text_utils_hyperlinkify_end (void)
1001 url_matches_block--;
1002 if (url_matches_block <= 0)
1008 get_url_matches (GString *txt)
1011 guint rv, i, offset = 0;
1012 GSList *match_list = NULL;
1014 const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
1016 /* initalize the regexps */
1017 modest_text_utils_hyperlinkify_begin ();
1019 /* find all the matches */
1020 for (i = 0; i != pattern_num; ++i) {
1024 gboolean is_submatch;
1027 if ((rv = regexec (patterns[i].preg, txt->str + offset, 1, &rm, 0)) != 0) {
1028 g_return_val_if_fail (rv == REG_NOMATCH, NULL); /* this should not happen */
1029 break; /* try next regexp */
1034 is_submatch = FALSE;
1035 /* check old matches to see if this has already been matched */
1036 cursor = match_list;
1037 while (cursor && !is_submatch) {
1038 const url_match_t *old_match =
1039 (const url_match_t *) cursor->data;
1040 guint new_offset = offset + rm.rm_so;
1041 is_submatch = (new_offset > old_match->offset &&
1042 new_offset < old_match->offset + old_match->len);
1043 cursor = g_slist_next (cursor);
1047 /* make a list of our matches (<offset, len, prefix> tupels)*/
1048 match = g_slice_new (url_match_t);
1049 match->offset = offset + rm.rm_so;
1050 match->len = rm.rm_eo - rm.rm_so;
1051 match->prefix = patterns[i].prefix;
1052 match_list = g_slist_prepend (match_list, match);
1058 modest_text_utils_hyperlinkify_end ();
1060 /* now sort the list, so the matches are in reverse order of occurence.
1061 * that way, we can do the replacements starting from the end, so we don't need
1062 * to recalculate the offsets
1064 match_list = g_slist_sort (match_list,
1065 (GCompareFunc)cmp_offsets_reverse);
1071 /* replace all occurences of needle in haystack with repl*/
1073 replace_string (const gchar *haystack, const gchar *needle, gchar repl)
1075 gchar *str, *cursor;
1077 if (!haystack || !needle || strlen(needle) == 0)
1078 return haystack ? g_strdup(haystack) : NULL;
1080 str = g_strdup (haystack);
1082 for (cursor = str; cursor && *cursor; ++cursor) {
1083 if (g_str_has_prefix (cursor, needle)) {
1085 memmove (cursor + 1,
1086 cursor + strlen (needle),
1087 strlen (cursor + strlen (needle)) + 1);
1095 hyperlinkify_plain_text (GString *txt)
1098 GSList *match_list = get_url_matches (txt);
1100 /* we will work backwards, so the offsets stay valid */
1101 for (cursor = match_list; cursor; cursor = cursor->next) {
1103 url_match_t *match = (url_match_t*) cursor->data;
1104 gchar *url = g_strndup (txt->str + match->offset, match->len);
1105 gchar *repl = NULL; /* replacement */
1107 /* the string still contains $(MARK_AMP_URI_STR)"amp;" for each
1108 * '&' in the original, because of the text->html conversion.
1109 * in the href-URL (and only there), we must convert that back to
1112 gchar *href_url = replace_string (url, MARK_AMP_URI_STR "amp;", '&');
1114 /* the prefix is NULL: use the one that is already there */
1115 repl = g_strdup_printf ("<a href=\"%s%s\">%s</a>",
1116 match->prefix ? match->prefix : EMPTY_STRING,
1119 /* replace the old thing with our hyperlink
1120 * replacement thing */
1121 g_string_erase (txt, match->offset, match->len);
1122 g_string_insert (txt, match->offset, repl);
1128 g_slice_free (url_match_t, match);
1131 g_slist_free (match_list);
1135 /* for optimization reasons, we change the string in-place */
1137 modest_text_utils_get_display_address (gchar *address)
1141 g_return_if_fail (address);
1146 /* should not be needed, and otherwise, we probably won't screw up the address
1147 * more than it already is :)
1148 * g_return_val_if_fail (g_utf8_validate (address, -1, NULL), NULL);
1151 /* remove leading whitespace */
1152 if (address[0] == ' ')
1153 g_strchug (address);
1155 for (i = 0; address[i]; ++i) {
1156 if (address[i] == '<') {
1157 if (G_UNLIKELY(i == 0))
1158 return; /* there's nothing else, leave it */
1160 address[i] = '\0'; /* terminate the string here */
1172 modest_text_utils_get_email_address (const gchar *full_address)
1174 const gchar *left, *right;
1176 g_return_val_if_fail (full_address, NULL);
1181 g_return_val_if_fail (g_utf8_validate (full_address, -1, NULL), NULL);
1183 left = g_strrstr_len (full_address, strlen(full_address), "<");
1185 return g_strdup (full_address);
1187 right = g_strstr_len (left, strlen(left), ">");
1189 return g_strdup (full_address);
1191 return g_strndup (left + 1, right - left - 1);
1195 modest_text_utils_get_subject_prefix_len (const gchar *sub)
1197 gint prefix_len = 0;
1199 g_return_val_if_fail (sub, 0);
1204 /* optimization: "Re", "RE", "re","Fwd", "FWD", "fwd","FW","Fw", "fw" */
1205 if (sub[0] != 'R' && sub[0] != 'F' && sub[0] != 'r' && sub[0] != 'f')
1207 else if (sub[0] && sub[1] != 'e' && sub[1] != 'E' && sub[1] != 'w' && sub[1] != 'W')
1214 /* skip over a [...] block */
1215 if (sub[prefix_len] == '[') {
1216 int c = prefix_len + 1;
1217 while (sub[c] && sub[c] != ']')
1220 return 0; /* no end to the ']' found */
1225 /* did we find the ':' ? */
1226 if (sub[prefix_len] == ':') {
1228 if (sub[prefix_len] == ' ')
1230 prefix_len += modest_text_utils_get_subject_prefix_len (sub + prefix_len);
1231 /* g_warning ("['%s','%s']", sub, (char*) sub + prefix_len); */
1239 modest_text_utils_utf8_strcmp (const gchar* s1, const gchar *s2, gboolean insensitive)
1242 /* work even when s1 and/or s2 == NULL */
1243 if (G_UNLIKELY(s1 == s2))
1245 if (G_UNLIKELY(!s1))
1247 if (G_UNLIKELY(!s2))
1250 /* if it's not case sensitive */
1253 /* optimization: shortcut if first char is ascii */
1254 if (((s1[0] & 0xf0)== 0) && ((s2[0] & 0xf0) == 0))
1255 return s1[0] - s2[0];
1257 return g_utf8_collate (s1, s2);
1263 /* optimization: short cut iif first char is ascii */
1264 if (((s1[0] & 0xf0) == 0) && ((s2[0] & 0xf0) == 0))
1265 return tolower(s1[0]) - tolower(s2[0]);
1267 n1 = g_utf8_strdown (s1, -1);
1268 n2 = g_utf8_strdown (s2, -1);
1270 result = g_utf8_collate (n1, n2);
1281 modest_text_utils_get_display_date (time_t date)
1283 #define DATE_BUF_SIZE 64
1284 static gchar date_buf[DATE_BUF_SIZE];
1286 /* calculate the # of days since epoch for
1287 * for today and for the date provided
1288 * based on idea from pvanhoof */
1289 int day = time(NULL) / (24 * 60 * 60);
1290 int date_day = date / (24 * 60 * 60);
1292 /* if it's today, show the time, if it's not today, show the date instead */
1294 if (day == date_day) /* is the date today? */
1295 modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, "%X", date);
1297 modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, "%x", date);
1299 return date_buf; /* this is a static buffer, don't free! */
1305 modest_text_utils_validate_folder_name (const gchar *folder_name)
1307 /* based on http://msdn2.microsoft.com/en-us/library/aa365247.aspx,
1308 * with some extras */
1312 const gchar **cursor = NULL;
1313 const gchar *forbidden_names[] = { /* windows does not like these */
1314 "CON", "PRN", "AUX", "NUL", "COM1", "COM2", "COM3", "COM4", "COM5", "COM6",
1315 "COM7", "COM8", "COM9", "LPT1", "LPT2", "LPT3", "LPT4", "LPT5", "LPT6", "LPT7", "LPT8", "LPT9",
1319 /* cannot be NULL */
1323 /* cannot be empty */
1324 len = strlen(folder_name);
1328 /* cannot start or end with a space */
1329 if (g_ascii_isspace(folder_name[0]) || g_ascii_isspace(folder_name[len - 1]))
1332 /* cannot contain a forbidden char */
1333 for (i = 0; i < len; i++)
1334 if (modest_text_utils_is_forbidden_char (folder_name[i], FOLDER_NAME_FORBIDDEN_CHARS))
1337 /* cannot contain a forbidden word */
1339 for (cursor = forbidden_names; cursor && *cursor; ++cursor) {
1340 if (g_ascii_strcasecmp (folder_name, *cursor) == 0)
1345 return TRUE; /* it's valid! */
1351 modest_text_utils_validate_domain_name (const gchar *domain)
1353 gboolean valid = FALSE;
1355 const gchar* domain_regex = "^([a-z0-9-]*[a-z0-9]\\.)+[a-z0-9-]*[a-z0-9]$";
1357 g_return_val_if_fail (domain, FALSE);
1362 memset (&rx, 0, sizeof(regex_t)); /* coverity wants this... */
1364 /* domain name: all alphanum or '-' or '.',
1365 * but beginning/ending in alphanum */
1366 if (regcomp (&rx, domain_regex, REG_ICASE|REG_EXTENDED|REG_NOSUB)) {
1367 g_warning ("BUG: error in regexp");
1371 valid = (regexec (&rx, domain, 1, NULL, 0) == 0);
1380 modest_text_utils_validate_email_address (const gchar *email_address,
1381 const gchar **invalid_char_position)
1384 const gchar *c = NULL, *domain = NULL;
1385 static gchar *rfc822_specials = "()<>@,;:\\\"[]&";
1387 if (invalid_char_position)
1388 *invalid_char_position = NULL;
1390 g_return_val_if_fail (email_address, FALSE);
1392 /* check that the email adress contains exactly one @ */
1393 if (!strstr(email_address, "@") ||
1394 (strstr(email_address, "@") != g_strrstr(email_address, "@")))
1397 /* first we validate the name portion (name@domain) */
1398 for (c = email_address; *c; c++) {
1400 (c == email_address ||
1402 *(c - 1) == '\"')) {
1406 if (*c == '\\' && (*++c == ' '))
1408 if (*c <= ' ' || *c >= 127)
1421 if (*c <= ' ' || *c >= 127)
1423 if (strchr(rfc822_specials, *c)) {
1424 if (invalid_char_position)
1425 *invalid_char_position = c;
1429 if (c == email_address || *(c - 1) == '.')
1432 /* next we validate the domain portion (name@domain) */
1433 if (!*(domain = ++c))
1437 if (c == domain || *(c - 1) == '.' || *(c + 1) == '\0')
1441 if (*c <= ' ' || *c >= 127)
1443 if (strchr(rfc822_specials, *c)) {
1444 if (invalid_char_position)
1445 *invalid_char_position = c;
1450 return (count >= 1) ? TRUE : FALSE;
1454 modest_text_utils_validate_recipient (const gchar *recipient, const gchar **invalid_char_position)
1456 gchar *stripped, *current;
1458 gboolean has_error = FALSE;
1460 if (invalid_char_position)
1461 *invalid_char_position = NULL;
1463 g_return_val_if_fail (recipient, FALSE);
1465 if (modest_text_utils_validate_email_address (recipient, invalid_char_position))
1468 stripped = g_strdup (recipient);
1469 stripped = g_strstrip (stripped);
1472 if (*current == '\0') {
1478 if (*current == '\"') {
1479 current = g_utf8_next_char (current);
1481 for (; *current != '\0'; current = g_utf8_next_char (current)) {
1482 if (*current == '\\') {
1483 /* TODO: This causes a warning, which breaks the build,
1484 * because a gchar cannot be < 0.
1486 if (current[1] <0) {
1491 } else if (*current == '\"') {
1493 current = g_utf8_next_char (current);
1499 for (current = stripped ; *current != '\0'; current = g_utf8_next_char (current)) {
1500 if (*current == '<') {
1512 right_part = g_strdup (current);
1514 right_part = g_strstrip (right_part);
1516 if (g_str_has_prefix (right_part, "<") &&
1517 g_str_has_suffix (right_part, ">")) {
1521 address = g_strndup (right_part+1, strlen (right_part) - 2);
1522 g_free (right_part);
1523 valid = modest_text_utils_validate_email_address (address, invalid_char_position);
1527 g_free (right_part);
1534 modest_text_utils_get_display_size (guint64 size)
1536 const guint KB=1024;
1537 const guint MB=1024 * KB;
1538 const guint GB=1024 * MB;
1541 return g_strdup_printf(_FM("sfil_li_size_kb"), 0);
1542 if (0 < size && size < KB)
1543 return g_strdup_printf (_FM("sfil_li_size_kb"), 1);
1544 else if (KB <= size && size < 100 * KB)
1545 return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), size / KB);
1546 else if (100*KB <= size && size < MB)
1547 return g_strdup_printf (_FM("sfil_li_size_100kb_1mb"), (float) size / MB);
1548 else if (MB <= size && size < 10*MB)
1549 return g_strdup_printf (_FM("sfil_li_size_1mb_10mb"), (float) size / MB);
1550 else if (10*MB <= size && size < GB)
1551 return g_strdup_printf (_FM("sfil_li_size_10mb_1gb"), size / MB);
1553 return g_strdup_printf (_FM("sfil_li_size_1gb_or_greater"), (float) size / GB);
1557 get_email_from_address (const gchar * address)
1559 gchar *left_limit, *right_limit;
1561 left_limit = strstr (address, "<");
1562 right_limit = g_strrstr (address, ">");
1564 if ((left_limit == NULL)||(right_limit == NULL)|| (left_limit > right_limit))
1565 return g_strdup (address);
1567 return g_strndup (left_limit + 1, (right_limit - left_limit) - 1);
1571 modest_text_utils_get_color_string (GdkColor *color)
1573 g_return_val_if_fail (color, NULL);
1575 return g_strdup_printf ("#%x%x%x%x%x%x%x%x%x%x%x%x",
1576 (color->red >> 12) & 0xf, (color->red >> 8) & 0xf,
1577 (color->red >> 4) & 0xf, (color->red) & 0xf,
1578 (color->green >> 12) & 0xf, (color->green >> 8) & 0xf,
1579 (color->green >> 4) & 0xf, (color->green) & 0xf,
1580 (color->blue >> 12) & 0xf, (color->blue >> 8) & 0xf,
1581 (color->blue >> 4) & 0xf, (color->blue) & 0xf);
1585 modest_text_utils_text_buffer_get_text (GtkTextBuffer *buffer)
1587 GtkTextIter start, end;
1588 gchar *slice, *current;
1589 GString *result = g_string_new ("");
1591 g_return_val_if_fail (buffer && GTK_IS_TEXT_BUFFER (buffer), NULL);
1593 gtk_text_buffer_get_start_iter (buffer, &start);
1594 gtk_text_buffer_get_end_iter (buffer, &end);
1596 slice = gtk_text_buffer_get_slice (buffer, &start, &end, FALSE);
1599 while (current && current != '\0') {
1600 if (g_utf8_get_char (current) == 0xFFFC) {
1601 result = g_string_append_c (result, ' ');
1602 current = g_utf8_next_char (current);
1604 gchar *next = g_utf8_strchr (current, -1, 0xFFFC);
1606 result = g_string_append (result, current);
1608 result = g_string_append_len (result, current, next - current);
1615 return g_string_free (result, FALSE);
1620 modest_text_utils_is_forbidden_char (const gchar character,
1621 ModestTextUtilsForbiddenCharType type)
1624 const gchar *forbidden_chars = NULL;
1626 /* We need to get the length in the switch because the
1627 compiler needs to know the size at compile time */
1629 case ACCOUNT_TITLE_FORBIDDEN_CHARS:
1630 forbidden_chars = account_title_forbidden_chars;
1631 len = G_N_ELEMENTS (account_title_forbidden_chars);
1633 case FOLDER_NAME_FORBIDDEN_CHARS:
1634 forbidden_chars = folder_name_forbidden_chars;
1635 len = G_N_ELEMENTS (folder_name_forbidden_chars);
1637 case USER_NAME_FORBIDDEN_NAMES:
1638 forbidden_chars = user_name_forbidden_chars;
1639 len = G_N_ELEMENTS (user_name_forbidden_chars);
1642 g_return_val_if_reached (TRUE);
1645 for (i = 0; i < len ; i++)
1646 if (forbidden_chars[i] == character)
1649 return FALSE; /* it's valid! */
1653 modest_text_utils_label_get_selection (GtkLabel *label)
1658 if (gtk_label_get_selection_bounds (GTK_LABEL (label), &start, &end)) {
1659 const gchar *start_offset;
1660 const gchar *end_offset;
1661 start_offset = gtk_label_get_text (GTK_LABEL (label));
1662 start_offset = g_utf8_offset_to_pointer (start_offset, start);
1663 end_offset = gtk_label_get_text (GTK_LABEL (label));
1664 end_offset = g_utf8_offset_to_pointer (end_offset, end);
1665 selection = g_strndup (start_offset, end_offset - start_offset);
1668 return g_strdup ("");
1673 _forward_search_image_char (gunichar ch,
1676 return (ch == 0xFFFC);
1680 modest_text_utils_buffer_selection_is_valid (GtkTextBuffer *buffer)
1683 GtkTextIter start, end;
1685 g_return_val_if_fail (GTK_IS_TEXT_BUFFER (buffer), FALSE);
1687 result = gtk_text_buffer_get_has_selection (GTK_TEXT_BUFFER (buffer));
1689 /* check there are no images in selection */
1691 gtk_text_buffer_get_selection_bounds (buffer, &start, &end);
1692 if (gtk_text_iter_get_char (&start)== 0xFFFC)
1695 gtk_text_iter_backward_char (&end);
1696 if (gtk_text_iter_forward_find_char (&start, _forward_search_image_char,