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)
68 * we need these regexps to find URLs in plain text e-mails
70 typedef struct _url_match_pattern_t url_match_pattern_t;
71 struct _url_match_pattern_t {
77 typedef struct _url_match_t url_match_t;
86 * we mark the ampersand with \007 when converting text->html
87 * because after text->html we do hyperlink detecting, which
88 * could be screwed up by the ampersand.
89 * ie. 1<3 ==> 1\007lt;3
91 #define MARK_AMP '\007'
92 #define MARK_AMP_STR "\007"
94 /* mark & separately, because they are parts of urls.
95 * ie. a&b => a\006amp;b, but a>b => a\007gt;b
97 * we need to handle '&' separately, because it can be part of URIs
98 * (as in href="http://foo.bar?a=1&b=1"), so inside those URIs
99 * we need to re-replace \006amp; with '&' again, while outside uri's
102 * yes, it's messy, but a consequence of doing text->html first, then hyperlinkify
104 #define MARK_AMP_URI '\006'
105 #define MARK_AMP_URI_STR "\006"
108 /* note: match MARK_AMP_URI_STR as well, because after txt->html, a '&' will look like $(MARK_AMP_URI_STR)"amp;" */
109 #define MAIL_VIEWER_URL_MATCH_PATTERNS { \
110 { "(feed:|)(file|rtsp|http|ftp|https|mms|mmsh|webcal|feed|rtsp|rdp|lastfm|sip)://[-a-z0-9_$.+!*(),;:@%=\?/~#&" MARK_AMP_URI_STR \
111 "]+[-a-z0-9_$%&" MARK_AMP_URI_STR "=?/~#]", \
113 { "www\\.[-a-z0-9_$.+!*(),;:@%=?/~#" MARK_AMP_URI_STR "]+[-a-z0-9_$%" MARK_AMP_URI_STR "=?/~#]",\
115 { "ftp\\.[-a-z0-9_$.+!*(),;:@%=?/~#" MARK_AMP_URI_STR "]+[-a-z0-9_$%" MARK_AMP_URI_STR "=?/~#]",\
117 { "(jabberto|voipto|sipto|sip|chatto|skype|xmpp):[-_a-z@0-9.+]+", \
119 { "mailto:[-_a-z0-9.\\+]+@[-_a-z0-9.]+", \
121 { "[-_a-z0-9.\\+]+@[-_a-z0-9.]+",\
125 const gchar account_title_forbidden_chars[] = {
126 '\\', '/', ':', '*', '?', '\'', '<', '>', '|', '^'
128 const gchar folder_name_forbidden_chars[] = {
129 '<', '>', ':', '\'', '/', '\\', '|', '?', '*', '^', '%', '$', '#', '&'
131 const gchar user_name_forbidden_chars[] = {
134 const guint ACCOUNT_TITLE_FORBIDDEN_CHARS_LENGTH = G_N_ELEMENTS (account_title_forbidden_chars);
135 const guint FOLDER_NAME_FORBIDDEN_CHARS_LENGTH = G_N_ELEMENTS (folder_name_forbidden_chars);
136 const guint USER_NAME_FORBIDDEN_CHARS_LENGTH = G_N_ELEMENTS (user_name_forbidden_chars);
139 static gchar* cite (const time_t sent_date, const gchar *from);
140 static void hyperlinkify_plain_text (GString *txt, gint offset);
141 static gint cmp_offsets_reverse (const url_match_t *match1, const url_match_t *match2);
142 static GSList* get_url_matches (GString *txt, gint offset);
144 static GString* get_next_line (const char *b, const gsize blen, const gchar * iter);
145 static int get_indent_level (const char *l);
146 static void unquote_line (GString * l, const gchar *quote_symbol);
147 static void append_quoted (GString * buf, const gchar *quote_symbol,
148 const int indent, const GString * str,
150 static int get_breakpoint_utf8 (const gchar * s, const gint indent, const gint limit);
151 static int get_breakpoint_ascii (const gchar * s, const gint indent, const gint limit);
152 static int get_breakpoint (const gchar * s, const gint indent, const gint limit);
154 static gchar* modest_text_utils_quote_plain_text (const gchar *text,
156 const gchar *signature,
160 static gchar* modest_text_utils_quote_html (const gchar *text,
162 const gchar *signature,
165 static gchar* get_email_from_address (const gchar *address);
168 /* ******************************************************************* */
169 /* ************************* PUBLIC FUNCTIONS ************************ */
170 /* ******************************************************************* */
173 modest_text_utils_quote (const gchar *text,
174 const gchar *content_type,
175 const gchar *signature,
177 const time_t sent_date,
181 gchar *retval, *cited;
183 g_return_val_if_fail (text, NULL);
184 g_return_val_if_fail (content_type, NULL);
186 cited = cite (sent_date, from);
188 if (content_type && strcmp (content_type, "text/html") == 0)
189 /* TODO: extract the <body> of the HTML and pass it to
191 retval = modest_text_utils_quote_html (text, cited, signature, attachments, limit);
193 retval = modest_text_utils_quote_plain_text (text, cited, signature, attachments, limit);
202 modest_text_utils_cite (const gchar *text,
203 const gchar *content_type,
204 const gchar *signature,
211 g_return_val_if_fail (text, NULL);
212 g_return_val_if_fail (content_type, NULL);
215 tmp_sig = g_strdup (text);
217 tmp_sig = g_strconcat (text, "\n", MODEST_TEXT_UTILS_SIGNATURE_MARKER, "\n", signature, NULL);
220 if (strcmp (content_type, "text/html") == 0) {
221 retval = modest_text_utils_convert_to_html_body (tmp_sig, -1, TRUE);
231 forward_cite (const gchar *from,
234 const gchar *subject)
236 g_return_val_if_fail (sent, NULL);
238 return g_strdup_printf ("%s\n%s %s\n%s %s\n%s %s\n%s %s\n",
240 FROM_STRING, (from)?from:"",
242 TO_STRING, (to)?to:"",
243 SUBJECT_STRING, (subject)?subject:"");
247 modest_text_utils_inline (const gchar *text,
248 const gchar *content_type,
249 const gchar *signature,
253 const gchar *subject)
259 g_return_val_if_fail (text, NULL);
260 g_return_val_if_fail (content_type, NULL);
262 modest_text_utils_strftime (sent_str, 100, "%c", sent_date);
264 cited = forward_cite (from, sent_str, to, subject);
266 if (content_type && strcmp (content_type, "text/html") == 0)
267 retval = modest_text_utils_quote_html (text, cited, signature, NULL, 80);
269 retval = modest_text_utils_quote_plain_text (text, cited, signature, NULL, 80);
275 /* just to prevent warnings:
276 * warning: `%x' yields only last 2 digits of year in some locales
279 modest_text_utils_strftime(char *s, gsize max, const char *fmt, time_t timet)
283 /* To prevent possible problems in strftime that could leave
284 garbage in the s variable */
290 /* does not work on old maemo glib:
291 * g_date_set_time_t (&date, timet);
293 localtime_r (&timet, &tm);
294 return strftime(s, max, fmt, &tm);
298 modest_text_utils_derived_subject (const gchar *subject, const gchar *prefix)
300 gchar *tmp, *subject_dup, *retval;
303 g_return_val_if_fail (prefix, NULL);
305 if (!subject || subject[0] == '\0')
306 subject = _("mail_va_no_subject");
308 subject_dup = g_strdup (subject);
309 tmp = g_strchug (subject_dup);
311 /* We do not want things like "Re: Re: Re:" or "Fw: Fw:" so
312 delete the previous ones */
313 prefix_len = strlen (prefix);
315 if (g_str_has_prefix (tmp, prefix)) {
317 tmp = g_strchug (tmp);
319 gchar *prefix_down, *tmp_down;
321 /* We need this to properly check the cases of
322 some clients adding FW: instead of Fw: for
324 prefix_down = g_utf8_strdown (prefix, -1);
325 tmp_down = g_utf8_strdown (tmp, -1);
326 if (g_str_has_prefix (tmp_down, prefix_down)) {
328 tmp = g_strchug (tmp);
329 g_free (prefix_down);
332 g_free (prefix_down);
339 retval = g_strdup_printf ("%s %s", prefix, tmp);
340 g_free (subject_dup);
346 /* Performs a case-insensitive strstr for ASCII strings */
348 ascii_stristr(const gchar *haystack, const gchar *needle)
353 const gchar *max_pos;
355 if (haystack == NULL || needle == NULL) {
356 return haystack; /* as in strstr */
359 needle_len = strlen(needle);
361 if (needle_len == 0) {
362 return haystack; /* as in strstr */
365 haystack_len = strlen(haystack);
366 max_pos = haystack + haystack_len - needle_len;
368 for (pos = haystack; pos <= max_pos; pos++) {
369 if (g_ascii_strncasecmp (pos, needle, needle_len) == 0) {
379 modest_text_utils_remove_address (const gchar *address_list, const gchar *address)
381 gchar *dup, *token, *ptr = NULL, *result;
382 GString *filtered_emails;
383 gchar *email_address;
385 g_return_val_if_fail (address_list, NULL);
388 return g_strdup (address_list);
390 email_address = get_email_from_address (address);
392 /* search for substring */
393 if (!ascii_stristr ((const char *) address_list, (const char *) email_address)) {
394 g_free (email_address);
395 return g_strdup (address_list);
398 dup = g_strdup (address_list);
399 filtered_emails = g_string_new (NULL);
401 token = strtok_r (dup, ",", &ptr);
403 while (token != NULL) {
404 /* Add to list if not found */
405 if (!ascii_stristr ((const char *) token, (const char *) email_address)) {
406 if (filtered_emails->len == 0)
407 g_string_append_printf (filtered_emails, "%s", g_strstrip (token));
409 g_string_append_printf (filtered_emails, ",%s", g_strstrip (token));
411 token = strtok_r (NULL, ",", &ptr);
413 result = filtered_emails->str;
416 g_free (email_address);
418 g_string_free (filtered_emails, FALSE);
425 modest_text_utils_remove_duplicate_addresses (const gchar *address_list)
427 GSList *addresses, *cursor;
429 gchar *new_list = NULL;
431 g_return_val_if_fail (address_list, NULL);
433 table = g_hash_table_new_full (g_str_hash, g_str_equal, g_free, NULL);
434 addresses = modest_text_utils_split_addresses_list (address_list);
438 const gchar* address = (const gchar*)cursor->data;
440 /* We need only the email to just compare it and not
441 the full address which would make "a <a@a.com>"
442 different from "a@a.com" */
443 const gchar *email = get_email_from_address (address);
445 /* ignore the address if already seen */
446 if (g_hash_table_lookup (table, email) == 0) {
449 /* Include the full address and not only the
450 email in the returned list */
452 tmp = g_strdup (address);
454 tmp = g_strjoin (",", new_list, address, NULL);
459 g_hash_table_insert (table, (gchar*)email, GINT_TO_POINTER(1));
461 cursor = g_slist_next (cursor);
464 g_hash_table_unref (table);
465 g_slist_foreach (addresses, (GFunc)g_free, NULL);
466 g_slist_free (addresses);
468 if (new_list == NULL)
469 new_list = g_strdup ("");
476 modest_text_utils_convert_buffer_to_html_start (GString *html, const gchar *data, gssize n)
479 gboolean space_seen = FALSE;
480 guint break_dist = 0; /* distance since last break point */
485 /* replace with special html chars where needed*/
486 for (i = 0; i != n; ++i) {
487 guchar kar = data[i];
489 if (space_seen && kar != ' ') {
490 g_string_append (html, " ");
494 /* we artificially insert a breakpoint (newline)
495 * after 256, to make sure our lines are not so long
496 * they will DOS the regexping later
497 * Also, check that kar is ASCII to make sure that we
498 * don't break a UTF8 char in two
500 if (++break_dist >= 256 && kar < 127) {
501 g_string_append_c (html, '\n');
509 /* this is a temp place holder for '&'; we can only
510 * set the real '&' after hyperlink translation, otherwise
511 * we might screw that up */
512 break; /* ignore embedded \0s and MARK_AMP */
513 case '<' : g_string_append (html, MARK_AMP_STR "lt;"); break;
514 case '>' : g_string_append (html, MARK_AMP_STR "gt;"); break;
515 case '&' : g_string_append (html, MARK_AMP_URI_STR "amp;"); break; /* special case */
516 case '"' : g_string_append (html, MARK_AMP_STR "quot;"); break;
518 /* don't convert ' --> wpeditor will try to re-convert it... */
519 //case '\'' : g_string_append (html, "'"); break;
520 case '\n' : g_string_append (html, "<br>\n");break_dist= 0; break;
521 case '\t' : g_string_append (html, MARK_AMP_STR "nbsp;" MARK_AMP_STR "nbsp;" MARK_AMP_STR "nbsp; ");
522 break_dist=0; break; /* note the space at the end*/
525 if (space_seen) { /* second space in a row */
526 g_string_append (html, " ");
531 g_string_append_c (html, kar);
538 modest_text_utils_convert_buffer_to_html_finish (GString *html)
541 /* replace all our MARK_AMPs with real ones */
542 for (i = 0; i != html->len; ++i)
543 if ((html->str)[i] == MARK_AMP || (html->str)[i] == MARK_AMP_URI)
544 (html->str)[i] = '&';
549 modest_text_utils_convert_to_html (const gchar *data)
554 g_return_val_if_fail (data, NULL);
560 html = g_string_sized_new (1.5 * len); /* just a guess... */
562 g_string_append_printf (html,
564 "<meta http-equiv=\"content-type\" content=\"text/html; charset=utf8\">"
568 modest_text_utils_convert_buffer_to_html_start (html, data, -1);
570 g_string_append (html, "</body></html>");
572 if (len <= HYPERLINKIFY_MAX_LENGTH)
573 hyperlinkify_plain_text (html, 0);
575 modest_text_utils_convert_buffer_to_html_finish (html);
577 return g_string_free (html, FALSE);
581 modest_text_utils_convert_to_html_body (const gchar *data, gssize n, gboolean hyperlinkify)
585 g_return_val_if_fail (data, NULL);
592 html = g_string_sized_new (1.5 * n); /* just a guess... */
594 modest_text_utils_convert_buffer_to_html_start (html, data, n);
596 if (hyperlinkify && (n < HYPERLINKIFY_MAX_LENGTH))
597 hyperlinkify_plain_text (html, 0);
599 modest_text_utils_convert_buffer_to_html_finish (html);
601 return g_string_free (html, FALSE);
605 modest_text_utils_get_addresses_indexes (const gchar *addresses, GSList **start_indexes, GSList **end_indexes)
610 str = g_string_new ("");
611 start = (gchar*) addresses;
612 cur = (gchar*) addresses;
614 for (cur = start; *cur != '\0'; cur = g_utf8_next_char (cur)) {
615 if (*cur == ',' || *cur == ';') {
616 if (!g_utf8_strchr (start, (cur - start + 1), g_utf8_get_char ("@")))
619 gint *start_index, *end_index;
620 start_index = g_new0 (gint, 1);
621 end_index = g_new0 (gint, 1);
622 *start_index = g_utf8_pointer_to_offset (addresses, start);
623 *end_index = g_utf8_pointer_to_offset (addresses, cur);;
624 *start_indexes = g_slist_prepend (*start_indexes, start_index);
625 *end_indexes = g_slist_prepend (*end_indexes, end_index);
626 start = g_utf8_next_char (cur);
631 *start_indexes = g_slist_reverse (*start_indexes);
632 *end_indexes = g_slist_reverse (*end_indexes);
637 modest_text_utils_split_addresses_list (const gchar *addresses)
640 const gchar *my_addrs = addresses;
643 gboolean after_at = FALSE;
645 g_return_val_if_fail (addresses, NULL);
647 /* skip any space, ',', ';' at the start */
648 while (my_addrs && (my_addrs[0] == ' ' || my_addrs[0] == ',' || my_addrs[0] == ';'))
651 /* are we at the end of addresses list? */
655 /* nope, we are at the start of some address
656 * now, let's find the end of the address */
658 while (end[0] && end[0] != ';' && !(after_at && end[0] == ',')) {
659 if (end[0] == '\"') {
660 while (end[0] && end[0] != '\"')
666 if ((end[0] && end[0] == '>')&&(end[1] && end[1] == ',')) {
673 /* we got the address; copy it and remove trailing whitespace */
674 addr = g_strndup (my_addrs, end - my_addrs);
677 head = g_slist_append (NULL, addr);
678 head->next = modest_text_utils_split_addresses_list (end); /* recurse */
685 modest_text_utils_address_range_at_position (const gchar *recipients_list,
690 gchar *current = NULL;
691 gint range_start = 0;
694 gboolean is_quoted = FALSE;
696 g_return_if_fail (recipients_list);
697 g_return_if_fail (position < g_utf8_strlen(recipients_list, -1));
700 for (current = (gchar *) recipients_list; *current != '\0';
701 current = g_utf8_find_next_char (current, NULL)) {
702 gunichar c = g_utf8_get_char (current);
704 if ((c == ',') && (!is_quoted)) {
705 if (index < position) {
706 range_start = index + 1;
710 } else if (c == '\"') {
711 is_quoted = !is_quoted;
712 } else if ((c == ' ') &&(range_start == index)) {
720 *start = range_start;
726 modest_text_utils_address_with_standard_length (const gchar *recipients_list)
730 GString *buffer = g_string_new ("");
732 splitted = g_strsplit (recipients_list, "\n", 0);
736 if (current != splitted)
737 buffer = g_string_append_c (buffer, '\n');
738 line = g_strndup (*splitted, 1000);
739 buffer = g_string_append (buffer, line);
744 g_strfreev (splitted);
746 return g_string_free (buffer, FALSE);
750 /* ******************************************************************* */
751 /* ************************* UTILIY FUNCTIONS ************************ */
752 /* ******************************************************************* */
755 get_next_line (const gchar * b, const gsize blen, const gchar * iter)
761 return g_string_new("");
769 gs = g_string_new_len (i0, iter - i0);
773 get_indent_level (const char *l)
790 /* if we hit the signature marker "-- ", we return -(indent + 1). This
791 * stops reformatting.
793 if (strcmp (l, MODEST_TEXT_UTILS_SIGNATURE_MARKER) == 0) {
801 unquote_line (GString * l, const gchar *quote_symbol)
807 quote_len = strlen (quote_symbol);
809 if (g_str_has_prefix (p, quote_symbol)) {
810 if (p[quote_len] == ' ') {
818 g_string_erase (l, 0, p - l->str);
822 append_quoted (GString * buf, const gchar *quote_symbol,
823 int indent, const GString * str,
829 indent = indent < 0 ? abs (indent) - 1 : indent;
830 quote_concat = g_strconcat (quote_symbol, " ", NULL);
831 for (i = 0; i <= indent; i++) {
832 g_string_append (buf, quote_concat);
834 g_free (quote_concat);
836 g_string_append_len (buf, str->str, cutpoint);
838 g_string_append (buf, str->str);
840 g_string_append (buf, "\n");
844 get_breakpoint_utf8 (const gchar * s, gint indent, const gint limit)
847 const gchar *pos, *last;
850 indent = indent < 0 ? abs (indent) - 1 : indent;
854 uni = g_utf8_to_ucs4_fast (s, -1, NULL);
856 if ((index + 2 * indent > limit) && last) {
860 if (g_unichar_isspace (uni[index])) {
863 pos = g_utf8_next_char (pos);
871 get_breakpoint_ascii (const gchar * s, const gint indent, const gint limit)
876 if (last + 2 * indent < limit)
879 for (i = strlen (s); i > 0; i--) {
881 if (i + 2 * indent <= limit) {
892 get_breakpoint (const gchar * s, const gint indent, const gint limit)
895 if (g_utf8_validate (s, -1, NULL)) {
896 return get_breakpoint_utf8 (s, indent, limit);
897 } else { /* assume ASCII */
898 //g_warning("invalid UTF-8 in msg");
899 return get_breakpoint_ascii (s, indent, limit);
904 cite (const time_t sent_date, const gchar *from)
906 return g_strdup (_("mcen_ia_editor_original_message"));
910 quoted_attachments (GList *attachments)
913 GString *result = g_string_new ("");
914 for (node = attachments; node != NULL; node = g_list_next (node)) {
915 gchar *filename = (gchar *) node->data;
916 g_string_append_printf ( result, "%s %s\n", _("mcen_ia_editor_attach_filename"), filename);
919 return g_string_free (result, FALSE);
924 modest_text_utils_quote_body (GString *output, const gchar *text,
925 const gchar *quote_symbol,
931 gint indent, breakpoint, rem_indent = 0;
932 GString *l, *remaining;
936 remaining = g_string_new ("");
938 l = get_next_line (text, len, iter);
939 iter = iter + l->len + 1;
940 indent = get_indent_level (l->str);
941 unquote_line (l, quote_symbol);
943 if (remaining->len) {
944 if (l->len && indent == rem_indent) {
945 g_string_prepend (l, " ");
946 g_string_prepend (l, remaining->str);
949 gunichar remaining_first;
951 get_breakpoint (remaining->str,
954 append_quoted (output, quote_symbol, rem_indent,
955 remaining, breakpoint);
956 g_string_erase (remaining, 0,
958 remaining_first = g_utf8_get_char_validated (remaining->str, remaining->len);
959 if (remaining_first != ((gunichar) -1)) {
960 if (g_unichar_isspace (remaining_first)) {
961 g_string_erase (remaining, 0, g_utf8_next_char (remaining->str) - remaining->str);
964 } while (remaining->len);
967 g_string_free (remaining, TRUE);
968 breakpoint = get_breakpoint (l->str, indent, limit);
969 remaining = g_string_new (l->str + breakpoint);
970 if (remaining->str[0] == ' ') {
971 g_string_erase (remaining, 0, 1);
974 append_quoted (output, quote_symbol, indent, l, breakpoint);
975 g_string_free (l, TRUE);
976 } while ((iter < text + len) || (remaining->str[0]));
982 modest_text_utils_quote_plain_text (const gchar *text,
984 const gchar *signature,
989 gchar *attachments_string = NULL;
991 q = g_string_new ("");
993 if (signature != NULL) {
994 g_string_append_printf (q, "\n%s\n", MODEST_TEXT_UTILS_SIGNATURE_MARKER);
995 q = g_string_append (q, signature);
998 q = g_string_append (q, "\n");
999 q = g_string_append (q, cite);
1000 q = g_string_append_c (q, '\n');
1002 q = modest_text_utils_quote_body (q, text, ">", limit);
1004 attachments_string = quoted_attachments (attachments);
1005 q = g_string_append (q, attachments_string);
1006 g_free (attachments_string);
1008 return g_string_free (q, FALSE);
1012 quote_html_add_to_gstring (GString *string,
1015 if (text && strcmp (text, "")) {
1016 gchar *html_text = modest_text_utils_convert_to_html_body (text, -1, TRUE);
1017 g_string_append_printf (string, "%s<br/>", html_text);
1023 modest_text_utils_quote_html (const gchar *text,
1025 const gchar *signature,
1029 GString *result_string;
1033 "<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\">\n" \
1037 if (text || cite || signature) {
1038 GString *quoted_text;
1039 g_string_append (result_string, "<pre>\n");
1041 quote_html_add_to_gstring (result_string, MODEST_TEXT_UTILS_SIGNATURE_MARKER);
1042 quote_html_add_to_gstring (result_string, signature);
1044 quote_html_add_to_gstring (result_string, cite);
1045 quoted_text = g_string_new ("");
1046 quoted_text = modest_text_utils_quote_body (quoted_text, (text) ? text : "", ">", limit);
1047 quote_html_add_to_gstring (result_string, quoted_text->str);
1048 g_string_free (quoted_text, TRUE);
1050 gchar *attachments_string = quoted_attachments (attachments);
1051 quote_html_add_to_gstring (result_string, attachments_string);
1052 g_free (attachments_string);
1054 g_string_append (result_string, "</pre>");
1056 g_string_append (result_string, "</body>");
1057 g_string_append (result_string, "</html>");
1059 return g_string_free (result_string, FALSE);
1063 cmp_offsets_reverse (const url_match_t *match1, const url_match_t *match2)
1065 return match2->offset - match1->offset;
1068 static gint url_matches_block = 0;
1069 static url_match_pattern_t patterns[] = MAIL_VIEWER_URL_MATCH_PATTERNS;
1070 static GMutex *url_patterns_mutex = NULL;
1077 const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
1078 for (i = 0; i != pattern_num; ++i) {
1079 patterns[i].preg = g_slice_new0 (regex_t);
1081 /* this should not happen */
1082 if (regcomp (patterns[i].preg, patterns[i].regex,
1083 REG_ICASE|REG_EXTENDED|REG_NEWLINE) != 0) {
1084 g_warning ("%s: error in regexp:\n%s\n", __FUNCTION__, patterns[i].regex);
1095 const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
1096 for (i = 0; i != pattern_num; ++i) {
1097 regfree (patterns[i].preg);
1098 g_slice_free (regex_t, patterns[i].preg);
1099 } /* don't free patterns itself -- it's static */
1103 modest_text_utils_hyperlinkify_begin (void)
1106 if (url_patterns_mutex == NULL) {
1107 url_patterns_mutex = g_mutex_new ();
1109 g_mutex_lock (url_patterns_mutex);
1110 if (url_matches_block == 0)
1111 compile_patterns ();
1112 url_matches_block ++;
1113 g_mutex_unlock (url_patterns_mutex);
1117 modest_text_utils_hyperlinkify_end (void)
1119 g_mutex_lock (url_patterns_mutex);
1120 url_matches_block--;
1121 if (url_matches_block <= 0)
1123 g_mutex_unlock (url_patterns_mutex);
1128 get_url_matches (GString *txt, gint offset)
1131 guint rv, i, tmp_offset = 0;
1132 GSList *match_list = NULL;
1134 const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
1136 /* initalize the regexps */
1137 modest_text_utils_hyperlinkify_begin ();
1139 /* find all the matches */
1140 for (i = 0; i != pattern_num; ++i) {
1141 tmp_offset = offset;
1144 gboolean is_submatch;
1147 if ((rv = regexec (patterns[i].preg, txt->str + tmp_offset, 1, &rm, 0)) != 0) {
1148 g_return_val_if_fail (rv == REG_NOMATCH, NULL); /* this should not happen */
1149 break; /* try next regexp */
1154 is_submatch = FALSE;
1155 /* check old matches to see if this has already been matched */
1156 cursor = match_list;
1157 while (cursor && !is_submatch) {
1158 const url_match_t *old_match =
1159 (const url_match_t *) cursor->data;
1160 guint new_offset = tmp_offset + rm.rm_so;
1161 is_submatch = (new_offset > old_match->offset &&
1162 new_offset < old_match->offset + old_match->len);
1163 cursor = g_slist_next (cursor);
1167 /* make a list of our matches (<offset, len, prefix> tupels)*/
1168 match = g_slice_new (url_match_t);
1169 match->offset = tmp_offset + rm.rm_so;
1170 match->len = rm.rm_eo - rm.rm_so;
1171 match->prefix = patterns[i].prefix;
1172 match_list = g_slist_prepend (match_list, match);
1174 tmp_offset += rm.rm_eo;
1178 modest_text_utils_hyperlinkify_end ();
1180 /* now sort the list, so the matches are in reverse order of occurence.
1181 * that way, we can do the replacements starting from the end, so we don't need
1182 * to recalculate the offsets
1184 match_list = g_slist_sort (match_list,
1185 (GCompareFunc)cmp_offsets_reverse);
1191 /* replace all occurences of needle in haystack with repl*/
1193 replace_string (const gchar *haystack, const gchar *needle, gchar repl)
1195 gchar *str, *cursor;
1197 if (!haystack || !needle || strlen(needle) == 0)
1198 return haystack ? g_strdup(haystack) : NULL;
1200 str = g_strdup (haystack);
1202 for (cursor = str; cursor && *cursor; ++cursor) {
1203 if (g_str_has_prefix (cursor, needle)) {
1205 memmove (cursor + 1,
1206 cursor + strlen (needle),
1207 strlen (cursor + strlen (needle)) + 1);
1215 hyperlinkify_plain_text (GString *txt, gint offset)
1218 GSList *match_list = get_url_matches (txt, offset);
1220 /* we will work backwards, so the offsets stay valid */
1221 for (cursor = match_list; cursor; cursor = cursor->next) {
1223 url_match_t *match = (url_match_t*) cursor->data;
1224 gchar *url = g_strndup (txt->str + match->offset, match->len);
1225 gchar *repl = NULL; /* replacement */
1227 /* the string still contains $(MARK_AMP_URI_STR)"amp;" for each
1228 * '&' in the original, because of the text->html conversion.
1229 * in the href-URL (and only there), we must convert that back to
1232 gchar *href_url = replace_string (url, MARK_AMP_URI_STR "amp;", '&');
1234 /* the prefix is NULL: use the one that is already there */
1235 repl = g_strdup_printf ("<a href=\"%s%s\">%s</a>",
1236 match->prefix ? match->prefix : EMPTY_STRING,
1239 /* replace the old thing with our hyperlink
1240 * replacement thing */
1241 g_string_erase (txt, match->offset, match->len);
1242 g_string_insert (txt, match->offset, repl);
1248 g_slice_free (url_match_t, match);
1251 g_slist_free (match_list);
1255 modest_text_utils_hyperlinkify (GString *string_buffer)
1260 after_body = strstr (string_buffer->str, "<body>");
1261 if (after_body != NULL)
1262 offset = after_body - string_buffer->str;
1263 hyperlinkify_plain_text (string_buffer, offset);
1267 /* for optimization reasons, we change the string in-place */
1269 modest_text_utils_get_display_address (gchar *address)
1273 g_return_if_fail (address);
1278 /* should not be needed, and otherwise, we probably won't screw up the address
1279 * more than it already is :)
1280 * g_return_val_if_fail (g_utf8_validate (address, -1, NULL), NULL);
1283 /* remove leading whitespace */
1284 if (address[0] == ' ')
1285 g_strchug (address);
1287 for (i = 0; address[i]; ++i) {
1288 if (address[i] == '<') {
1289 if (G_UNLIKELY(i == 0)) {
1290 break; /* there's nothing else, leave it */
1292 address[i] = '\0'; /* terminate the string here */
1298 g_strchomp (address);
1303 modest_text_utils_get_display_addresses (const gchar *recipients)
1306 GSList *recipient_list;
1309 recipient_list = modest_text_utils_split_addresses_list (recipients);
1310 if (recipient_list) {
1311 GString *add_string = g_string_sized_new (strlen (recipients));
1312 GSList *iter = recipient_list;
1313 gboolean first = TRUE;
1316 /* Strings are changed in place */
1317 modest_text_utils_get_display_address ((gchar *) iter->data);
1318 if (G_UNLIKELY (first)) {
1319 g_string_append_printf (add_string, "%s", (gchar *) iter->data);
1322 g_string_append_printf (add_string, ", %s", (gchar *) iter->data);
1324 iter = g_slist_next (iter);
1326 g_slist_foreach (recipient_list, (GFunc) g_free, NULL);
1327 g_slist_free (recipient_list);
1328 addresses = g_string_free (add_string, FALSE);
1336 modest_text_utils_get_email_address (const gchar *full_address)
1338 const gchar *left, *right;
1340 g_return_val_if_fail (full_address, NULL);
1345 g_return_val_if_fail (g_utf8_validate (full_address, -1, NULL), NULL);
1347 left = g_strrstr_len (full_address, strlen(full_address), "<");
1349 return g_strdup (full_address);
1351 right = g_strstr_len (left, strlen(left), ">");
1353 return g_strdup (full_address);
1355 return g_strndup (left + 1, right - left - 1);
1359 modest_text_utils_get_subject_prefix_len (const gchar *sub)
1361 gint prefix_len = 0;
1363 g_return_val_if_fail (sub, 0);
1368 /* optimization: "Re", "RE", "re","Fwd", "FWD", "fwd","FW","Fw", "fw" */
1369 if (sub[0] != 'R' && sub[0] != 'F' && sub[0] != 'r' && sub[0] != 'f')
1371 else if (sub[0] && sub[1] != 'e' && sub[1] != 'E' && sub[1] != 'w' && sub[1] != 'W')
1378 /* skip over a [...] block */
1379 if (sub[prefix_len] == '[') {
1380 int c = prefix_len + 1;
1381 while (sub[c] && sub[c] != ']')
1384 return 0; /* no end to the ']' found */
1389 /* did we find the ':' ? */
1390 if (sub[prefix_len] == ':') {
1392 if (sub[prefix_len] == ' ')
1394 prefix_len += modest_text_utils_get_subject_prefix_len (sub + prefix_len);
1395 /* g_warning ("['%s','%s']", sub, (char*) sub + prefix_len); */
1403 modest_text_utils_utf8_strcmp (const gchar* s1, const gchar *s2, gboolean insensitive)
1406 /* work even when s1 and/or s2 == NULL */
1407 if (G_UNLIKELY(s1 == s2))
1409 if (G_UNLIKELY(!s1))
1411 if (G_UNLIKELY(!s2))
1414 /* if it's not case sensitive */
1417 /* optimization: shortcut if first char is ascii */
1418 if (((s1[0] & 0x80)== 0) && ((s2[0] & 0x80) == 0) &&
1420 return s1[0] - s2[0];
1422 return g_utf8_collate (s1, s2);
1428 /* optimization: shortcut if first char is ascii */
1429 if (((s1[0] & 0x80) == 0) && ((s2[0] & 0x80) == 0) &&
1430 (tolower(s1[0]) != tolower (s2[0])))
1431 return tolower(s1[0]) - tolower(s2[0]);
1433 n1 = g_utf8_strdown (s1, -1);
1434 n2 = g_utf8_strdown (s2, -1);
1436 result = g_utf8_collate (n1, n2);
1447 modest_text_utils_get_display_date (time_t date)
1449 #define DATE_BUF_SIZE 64
1450 static gchar date_buf[DATE_BUF_SIZE];
1452 /* calculate the # of days since epoch for
1453 * for today and for the date provided
1454 * based on idea from pvanhoof */
1455 int day = time(NULL) / (24 * 60 * 60);
1456 int date_day = date / (24 * 60 * 60);
1458 /* if it's today, show the time, if it's not today, show the date instead */
1460 /* TODO: take into account the system config for 24/12h */
1461 #ifdef MODEST_TOOLKIT_HILDON2
1462 if (day == date_day) /* is the date today? */
1463 modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, _HL("wdgt_va_24h_time"), date);
1465 modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, _HL("wdgt_va_date"), date);
1467 if (day == date_day) /* is the date today? */
1468 modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, "%X", date);
1470 modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, "%x", date);
1473 return date_buf; /* this is a static buffer, don't free! */
1479 modest_text_utils_validate_folder_name (const gchar *folder_name)
1481 /* based on http://msdn2.microsoft.com/en-us/library/aa365247.aspx,
1482 * with some extras */
1486 const gchar **cursor = NULL;
1487 const gchar *forbidden_names[] = { /* windows does not like these */
1488 "CON", "PRN", "AUX", "NUL", ".", "..", "cur", "tmp", "new",
1489 NULL /* cur, tmp, new are reserved for Maildir */
1492 /* cannot be NULL */
1496 /* cannot be empty */
1497 len = strlen(folder_name);
1501 /* cannot start with a dot, vfat does not seem to like that */
1502 if (folder_name[0] == '.')
1505 /* cannot start or end with a space */
1506 if (g_ascii_isspace(folder_name[0]) || g_ascii_isspace(folder_name[len - 1]))
1509 /* cannot contain a forbidden char */
1510 for (i = 0; i < len; i++)
1511 if (modest_text_utils_is_forbidden_char (folder_name[i], FOLDER_NAME_FORBIDDEN_CHARS))
1514 /* Cannot contain Windows port numbers. I'd like to use GRegex
1515 but it's still not available in Maemo. sergio */
1516 if (!g_ascii_strncasecmp (folder_name, "LPT", 3) ||
1517 !g_ascii_strncasecmp (folder_name, "COM", 3)) {
1521 /* We skip the first 3 characters for the
1523 val = strtol(folder_name+3, &endptr, 10);
1525 /* If the conversion to long succeeded then the string
1526 is not valid for us */
1527 if (*endptr == '\0')
1533 /* cannot contain a forbidden word */
1535 for (cursor = forbidden_names; cursor && *cursor; ++cursor) {
1536 if (g_ascii_strcasecmp (folder_name, *cursor) == 0)
1541 return TRUE; /* it's valid! */
1547 modest_text_utils_validate_domain_name (const gchar *domain)
1549 gboolean valid = FALSE;
1551 const gchar* domain_regex = "^([a-z0-9-]*[a-z0-9]\\.)+[a-z0-9-]*[a-z0-9]$";
1553 g_return_val_if_fail (domain, FALSE);
1558 memset (&rx, 0, sizeof(regex_t)); /* coverity wants this... */
1560 /* domain name: all alphanum or '-' or '.',
1561 * but beginning/ending in alphanum */
1562 if (regcomp (&rx, domain_regex, REG_ICASE|REG_EXTENDED|REG_NOSUB)) {
1563 g_warning ("BUG: error in regexp");
1567 valid = (regexec (&rx, domain, 1, NULL, 0) == 0);
1576 modest_text_utils_validate_email_address (const gchar *email_address,
1577 const gchar **invalid_char_position)
1580 const gchar *c = NULL, *domain = NULL;
1581 static gchar *rfc822_specials = "()<>@,;:\\\"[]&";
1583 if (invalid_char_position)
1584 *invalid_char_position = NULL;
1586 g_return_val_if_fail (email_address, FALSE);
1588 /* check that the email adress contains exactly one @ */
1589 if (!strstr(email_address, "@") ||
1590 (strstr(email_address, "@") != g_strrstr(email_address, "@")))
1593 /* first we validate the name portion (name@domain) */
1594 for (c = email_address; *c; c++) {
1596 (c == email_address ||
1598 *(c - 1) == '\"')) {
1602 if (*c == '\\' && (*++c == ' '))
1604 if (*c <= ' ' || *c >= 127)
1617 if (*c <= ' ' || *c >= 127)
1619 if (strchr(rfc822_specials, *c)) {
1620 if (invalid_char_position)
1621 *invalid_char_position = c;
1625 if (c == email_address || *(c - 1) == '.')
1628 /* next we validate the domain portion (name@domain) */
1629 if (!*(domain = ++c))
1633 if (c == domain || *(c - 1) == '.' || *(c + 1) == '\0')
1637 if (*c <= ' ' || *c >= 127)
1639 if (strchr(rfc822_specials, *c)) {
1640 if (invalid_char_position)
1641 *invalid_char_position = c;
1646 return (count >= 1) ? TRUE : FALSE;
1650 modest_text_utils_validate_recipient (const gchar *recipient, const gchar **invalid_char_position)
1652 gchar *stripped, *current;
1654 gboolean has_error = FALSE;
1656 if (invalid_char_position)
1657 *invalid_char_position = NULL;
1659 g_return_val_if_fail (recipient, FALSE);
1661 if (modest_text_utils_validate_email_address (recipient, invalid_char_position))
1664 stripped = g_strdup (recipient);
1665 stripped = g_strstrip (stripped);
1668 if (*current == '\0') {
1674 if (*current == '\"') {
1675 gchar *last_quote = NULL;
1676 current = g_utf8_next_char (current);
1678 for (; *current != '\0'; current = g_utf8_next_char (current)) {
1679 if (*current == '\\') {
1680 /* TODO: This causes a warning, which breaks the build,
1681 * because a gchar cannot be < 0.
1683 if (current[1] <0) {
1688 } else if (*current == '\"') {
1690 current = g_utf8_next_char (current);
1691 last_quote = current;
1695 current = last_quote;
1698 for (current = stripped ; *current != '\0'; current = g_utf8_next_char (current)) {
1699 if (*current == '<') {
1711 right_part = g_strdup (current);
1713 right_part = g_strstrip (right_part);
1715 if (g_str_has_suffix (right_part, ",") || g_str_has_suffix (right_part, ";"))
1716 right_part [(strlen (right_part) - 1)] = '\0';
1718 if (g_str_has_prefix (right_part, "<") &&
1719 g_str_has_suffix (right_part, ">")) {
1723 address = g_strndup (right_part+1, strlen (right_part) - 2);
1724 g_free (right_part);
1725 valid = modest_text_utils_validate_email_address (address, invalid_char_position);
1729 g_free (right_part);
1736 modest_text_utils_get_display_size (guint64 size)
1738 const guint KB=1024;
1739 const guint MB=1024 * KB;
1740 const guint GB=1024 * MB;
1743 return g_strdup_printf (_FM("sfil_li_size_kb"), (int) 0);
1744 if (0 <= size && size < KB)
1745 return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), (int) 1);
1746 else if (KB <= size && size < 100 * KB)
1747 return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), (int) size / KB);
1748 else if (100*KB <= size && size < MB)
1749 return g_strdup_printf (_FM("sfil_li_size_100kb_1mb"), (int) size / KB);
1750 else if (MB <= size && size < 10*MB)
1751 return g_strdup_printf (_FM("sfil_li_size_1mb_10mb"), (float) size / MB);
1752 else if (10*MB <= size && size < GB)
1753 return g_strdup_printf (_FM("sfil_li_size_10mb_1gb"), (float) size / MB);
1755 return g_strdup_printf (_FM("sfil_li_size_1gb_or_greater"), (float) size / GB);
1759 get_email_from_address (const gchar * address)
1761 gchar *left_limit, *right_limit;
1763 left_limit = strstr (address, "<");
1764 right_limit = g_strrstr (address, ">");
1766 if ((left_limit == NULL)||(right_limit == NULL)|| (left_limit > right_limit))
1767 return g_strdup (address);
1769 return g_strndup (left_limit + 1, (right_limit - left_limit) - 1);
1773 modest_text_utils_get_color_string (GdkColor *color)
1775 g_return_val_if_fail (color, NULL);
1777 return g_strdup_printf ("#%x%x%x%x%x%x%x%x%x%x%x%x",
1778 (color->red >> 12) & 0xf, (color->red >> 8) & 0xf,
1779 (color->red >> 4) & 0xf, (color->red) & 0xf,
1780 (color->green >> 12) & 0xf, (color->green >> 8) & 0xf,
1781 (color->green >> 4) & 0xf, (color->green) & 0xf,
1782 (color->blue >> 12) & 0xf, (color->blue >> 8) & 0xf,
1783 (color->blue >> 4) & 0xf, (color->blue) & 0xf);
1787 modest_text_utils_text_buffer_get_text (GtkTextBuffer *buffer)
1789 GtkTextIter start, end;
1790 gchar *slice, *current;
1791 GString *result = g_string_new ("");
1793 g_return_val_if_fail (buffer && GTK_IS_TEXT_BUFFER (buffer), NULL);
1795 gtk_text_buffer_get_start_iter (buffer, &start);
1796 gtk_text_buffer_get_end_iter (buffer, &end);
1798 slice = gtk_text_buffer_get_slice (buffer, &start, &end, FALSE);
1801 while (current && current != '\0') {
1802 if (g_utf8_get_char (current) == 0xFFFC) {
1803 result = g_string_append_c (result, ' ');
1804 current = g_utf8_next_char (current);
1806 gchar *next = g_utf8_strchr (current, -1, 0xFFFC);
1808 result = g_string_append (result, current);
1810 result = g_string_append_len (result, current, next - current);
1817 return g_string_free (result, FALSE);
1822 modest_text_utils_is_forbidden_char (const gchar character,
1823 ModestTextUtilsForbiddenCharType type)
1826 const gchar *forbidden_chars = NULL;
1828 /* We need to get the length in the switch because the
1829 compiler needs to know the size at compile time */
1831 case ACCOUNT_TITLE_FORBIDDEN_CHARS:
1832 forbidden_chars = account_title_forbidden_chars;
1833 len = G_N_ELEMENTS (account_title_forbidden_chars);
1835 case FOLDER_NAME_FORBIDDEN_CHARS:
1836 forbidden_chars = folder_name_forbidden_chars;
1837 len = G_N_ELEMENTS (folder_name_forbidden_chars);
1839 case USER_NAME_FORBIDDEN_NAMES:
1840 forbidden_chars = user_name_forbidden_chars;
1841 len = G_N_ELEMENTS (user_name_forbidden_chars);
1844 g_return_val_if_reached (TRUE);
1847 for (i = 0; i < len ; i++)
1848 if (forbidden_chars[i] == character)
1851 return FALSE; /* it's valid! */
1855 modest_text_utils_label_get_selection (GtkLabel *label)
1860 if (gtk_label_get_selection_bounds (GTK_LABEL (label), &start, &end)) {
1861 const gchar *start_offset;
1862 const gchar *end_offset;
1863 start_offset = gtk_label_get_text (GTK_LABEL (label));
1864 start_offset = g_utf8_offset_to_pointer (start_offset, start);
1865 end_offset = gtk_label_get_text (GTK_LABEL (label));
1866 end_offset = g_utf8_offset_to_pointer (end_offset, end);
1867 selection = g_strndup (start_offset, end_offset - start_offset);
1870 return g_strdup ("");
1875 _forward_search_image_char (gunichar ch,
1878 return (ch == 0xFFFC);
1882 modest_text_utils_buffer_selection_is_valid (GtkTextBuffer *buffer)
1885 GtkTextIter start, end;
1887 g_return_val_if_fail (GTK_IS_TEXT_BUFFER (buffer), FALSE);
1889 result = gtk_text_buffer_get_has_selection (GTK_TEXT_BUFFER (buffer));
1891 /* check there are no images in selection */
1893 gtk_text_buffer_get_selection_bounds (buffer, &start, &end);
1894 if (gtk_text_iter_get_char (&start)== 0xFFFC)
1897 gtk_text_iter_backward_char (&end);
1898 if (gtk_text_iter_forward_find_char (&start, _forward_search_image_char,
1909 remove_quotes (gchar **quotes)
1911 if (g_str_has_prefix (*quotes, "\"") && g_str_has_suffix (*quotes, "\"")) {
1913 result = g_strndup ((*quotes)+1, strlen (*quotes) - 2);
1920 modest_text_utils_escape_mnemonics (const gchar *text)
1923 GString *result = NULL;
1928 result = g_string_new ("");
1929 for (p = text; *p != '\0'; p++) {
1931 result = g_string_append (result, "__");
1933 result = g_string_append_c (result, *p);
1936 return g_string_free (result, FALSE);
1940 modest_text_utils_simplify_recipients (const gchar *recipients)
1942 GSList *addresses, *node;
1944 gboolean is_first = TRUE;
1946 if (recipients == NULL)
1947 return g_strdup ("");
1949 addresses = modest_text_utils_split_addresses_list (recipients);
1950 result = g_string_new ("");
1952 for (node = addresses; node != NULL; node = g_slist_next (node)) {
1953 const gchar *address = (const gchar *) node->data;
1954 gchar *left_limit, *right_limit;
1956 left_limit = strstr (address, "<");
1957 right_limit = g_strrstr (address, ">");
1962 result = g_string_append (result, ", ");
1964 if ((left_limit == NULL)||(right_limit == NULL)|| (left_limit > right_limit)) {
1965 result = g_string_append (result, address);
1969 name_side = g_strndup (address, left_limit - address);
1970 name_side = g_strstrip (name_side);
1971 remove_quotes (&name_side);
1972 email_side = get_email_from_address (address);
1973 if (name_side && email_side && !strcmp (name_side, email_side)) {
1974 result = g_string_append (result, email_side);
1976 result = g_string_append (result, address);
1979 g_free (email_side);
1983 g_slist_foreach (addresses, (GFunc)g_free, NULL);
1984 g_slist_free (addresses);
1986 return g_string_free (result, FALSE);
1991 modest_text_utils_remove_duplicate_addresses_list (GSList *address_list)
1993 GSList *new_list, *iter;
1996 g_return_val_if_fail (address_list, NULL);
1998 table = g_hash_table_new_full (g_str_hash, g_str_equal, g_free, NULL);
2000 new_list = address_list;
2001 iter = address_list;
2003 const gchar* address = (const gchar*)iter->data;
2005 /* We need only the email to just compare it and not
2006 the full address which would make "a <a@a.com>"
2007 different from "a@a.com" */
2008 const gchar *email = get_email_from_address (address);
2010 /* ignore the address if already seen */
2011 if (g_hash_table_lookup (table, email) == 0) {
2012 g_hash_table_insert (table, (gchar*)email, GINT_TO_POINTER(1));
2013 iter = g_slist_next (iter);
2015 GSList *tmp = g_slist_next (iter);
2016 new_list = g_slist_delete_link (new_list, iter);
2021 g_hash_table_unref (table);
2027 modest_text_utils_get_secure_header (const gchar *value,
2028 const gchar *header)
2030 const gint max_len = 128;
2031 gchar *new_value = NULL;
2032 gchar *needle = g_strrstr (value, header);
2034 if (needle && value != needle)
2035 new_value = g_strdup (needle + strlen (header));
2038 new_value = g_strdup (value);
2040 /* Do a max length check to prevent DoS attacks caused by huge
2041 malformed headers */
2042 if (g_utf8_validate (new_value, -1, NULL)) {
2043 if (g_utf8_strlen (new_value, -1) > max_len) {
2044 gchar *tmp = g_malloc0 (max_len * 4);
2045 g_utf8_strncpy (tmp, (const gchar *) new_value, max_len);
2050 if (strlen (new_value) > max_len) {
2051 gchar *tmp = g_malloc0 (max_len);
2052 strncpy (new_value, tmp, max_len);
2062 is_quoted (const char *start, const gchar *end)
2066 c = (gchar *) start;
2068 c = g_utf8_next_char (c);
2070 if (*c == '\0' || *c != '\"')
2074 while (*c == ' ' && c != start)
2075 c = g_utf8_prev_char (c);
2077 if (c == start || *c != '\"')
2085 quote_name_part (GString **str, gchar **cur, gchar **start)
2088 gint str_len = g_utf8_pointer_to_offset (*start, *cur) -
2089 g_utf8_pointer_to_offset (*start, *start);
2091 while (**start == ' ') {
2092 *start = g_utf8_next_char (*start);
2096 blank = g_utf8_strrchr (*start, str_len, g_utf8_get_char (" "));
2097 if (blank && (blank != *start)) {
2098 if (is_quoted (*start, blank - 1)) {
2099 *str = g_string_append_len (*str, *start, str_len);
2100 *str = g_string_append (*str, "; ");
2101 *start = g_utf8_next_char (*cur);
2103 *str = g_string_append_c (*str, '"');
2104 *str = g_string_append_len (*str, *start,
2105 (g_utf8_pointer_to_offset (*start, blank) -
2106 g_utf8_pointer_to_offset (*start, *start)));
2107 *str = g_string_append_c (*str, '"');
2108 *str = g_string_append_len (*str, blank,
2109 (g_utf8_pointer_to_offset (*start, *cur) -
2110 g_utf8_pointer_to_offset (*start, blank)));
2111 *str = g_string_append (*str, "; ");
2112 *start = g_utf8_next_char (*cur);
2115 *str = g_string_append_len (*str, *start, str_len);
2116 *str = g_string_append (*str, "; ");
2117 *start = g_utf8_next_char (*cur);
2122 modest_text_utils_quote_names (const gchar *recipients)
2127 str = g_string_new ("");
2128 start = (gchar*) recipients;
2129 cur = (gchar*) recipients;
2131 for (cur = start; *cur != '\0'; cur = g_utf8_next_char (cur)) {
2132 if (*cur == ',' || *cur == ';') {
2133 if (!g_utf8_strchr (start, (cur - start + 1), g_utf8_get_char ("@")))
2135 quote_name_part (&str, &cur, &start);
2139 quote_name_part (&str, &cur, &start);
2141 return g_string_free (str, FALSE);