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)
66 #define SIGNATURE_MARKER "--"
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 $(MARK_AMP_URI_STR)"amp;" */
111 #define MAIL_VIEWER_URL_MATCH_PATTERNS { \
112 { "(file|rtsp|http|ftp|https|mms|mmsh|webcal|feed|rtsp|rdp|lastfm|sip)://[-a-z0-9_$.+!*(),;:@%=?/~#" 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_$%" MARK_AMP_URI_STR "=?/~#]",\
119 { "(jabberto|voipto|sipto|sip|chatto|skype|xmpp):[-_a-z@0-9.+]+", \
121 { "mailto:[-_a-z0-9.\\+]+@[-_a-z0-9.]+", \
123 { "[-_a-z0-9.\\+]+@[-_a-z0-9.]+",\
127 const gchar account_title_forbidden_chars[] = {
128 '\\', '/', ':', '*', '?', '\'', '<', '>', '|', '^'
130 const gchar folder_name_forbidden_chars[] = {
131 '<', '>', ':', '\'', '/', '\\', '|', '?', '*', '^', '%', '$', '#', '&'
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, gint offset);
143 static gint cmp_offsets_reverse (const url_match_t *match1, const url_match_t *match2);
144 static GSList* get_url_matches (GString *txt, gint offset);
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, const gchar *quote_symbol);
149 static void append_quoted (GString * buf, const gchar *quote_symbol,
150 const int indent, const GString * str,
152 static int get_breakpoint_utf8 (const gchar * s, const gint indent, const gint limit);
153 static int get_breakpoint_ascii (const gchar * s, const gint indent, const gint limit);
154 static int get_breakpoint (const gchar * s, const gint indent, const gint limit);
156 static gchar* modest_text_utils_quote_plain_text (const gchar *text,
158 const gchar *signature,
162 static gchar* modest_text_utils_quote_html (const gchar *text,
164 const gchar *signature,
167 static gchar* get_email_from_address (const gchar *address);
170 /* ******************************************************************* */
171 /* ************************* PUBLIC FUNCTIONS ************************ */
172 /* ******************************************************************* */
175 modest_text_utils_quote (const gchar *text,
176 const gchar *content_type,
177 const gchar *signature,
179 const time_t sent_date,
183 gchar *retval, *cited;
185 g_return_val_if_fail (text, NULL);
186 g_return_val_if_fail (content_type, NULL);
188 cited = cite (sent_date, from);
190 if (content_type && strcmp (content_type, "text/html") == 0)
191 /* TODO: extract the <body> of the HTML and pass it to
193 retval = modest_text_utils_quote_html (text, cited, signature, attachments, limit);
195 retval = modest_text_utils_quote_plain_text (text, cited, signature, attachments, limit);
204 modest_text_utils_cite (const gchar *text,
205 const gchar *content_type,
206 const gchar *signature,
213 g_return_val_if_fail (text, NULL);
214 g_return_val_if_fail (content_type, NULL);
217 tmp_sig = g_strdup (text);
219 tmp_sig = g_strconcat (text, "\n", SIGNATURE_MARKER, "\n", signature, NULL);
222 if (strcmp (content_type, "text/html") == 0) {
223 retval = modest_text_utils_convert_to_html_body (tmp_sig, -1, TRUE);
233 forward_cite (const gchar *from,
236 const gchar *subject)
238 g_return_val_if_fail (sent, NULL);
240 return g_strdup_printf ("%s\n%s %s\n%s %s\n%s %s\n%s %s\n",
242 FROM_STRING, (from)?from:"",
244 TO_STRING, (to)?to:"",
245 SUBJECT_STRING, (subject)?subject:"");
249 modest_text_utils_inline (const gchar *text,
250 const gchar *content_type,
251 const gchar *signature,
255 const gchar *subject)
261 g_return_val_if_fail (text, NULL);
262 g_return_val_if_fail (content_type, NULL);
264 modest_text_utils_strftime (sent_str, 100, "%c", sent_date);
266 cited = forward_cite (from, sent_str, to, subject);
268 if (content_type && strcmp (content_type, "text/html") == 0)
269 retval = modest_text_utils_quote_html (text, cited, signature, NULL, 80);
271 retval = modest_text_utils_quote_plain_text (text, cited, signature, NULL, 80);
277 /* just to prevent warnings:
278 * warning: `%x' yields only last 2 digits of year in some locales
281 modest_text_utils_strftime(char *s, gsize max, const char *fmt, time_t timet)
285 /* To prevent possible problems in strftime that could leave
286 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);
323 retval = g_strdup_printf ("%s %s", prefix, tmp);
324 g_free (subject_dup);
330 modest_text_utils_remove_address (const gchar *address_list, const gchar *address)
332 gchar *dup, *token, *ptr = NULL, *result;
333 GString *filtered_emails;
334 gchar *email_address;
336 g_return_val_if_fail (address_list, NULL);
339 return g_strdup (address_list);
341 email_address = get_email_from_address (address);
343 /* search for substring */
344 if (!strstr ((const char *) address_list, (const char *) email_address)) {
345 g_free (email_address);
346 return g_strdup (address_list);
349 dup = g_strdup (address_list);
350 filtered_emails = g_string_new (NULL);
352 token = strtok_r (dup, ",", &ptr);
354 while (token != NULL) {
355 /* Add to list if not found */
356 if (!strstr ((const char *) token, (const char *) email_address)) {
357 if (filtered_emails->len == 0)
358 g_string_append_printf (filtered_emails, "%s", g_strstrip (token));
360 g_string_append_printf (filtered_emails, ",%s", g_strstrip (token));
362 token = strtok_r (NULL, ",", &ptr);
364 result = filtered_emails->str;
367 g_free (email_address);
369 g_string_free (filtered_emails, FALSE);
376 modest_text_utils_remove_duplicate_addresses (const gchar *address_list)
378 GSList *addresses, *cursor;
380 gchar *new_list = NULL;
382 g_return_val_if_fail (address_list, NULL);
384 table = g_hash_table_new_full (g_str_hash, g_str_equal, g_free, NULL);
385 addresses = modest_text_utils_split_addresses_list (address_list);
389 const gchar* address = (const gchar*)cursor->data;
391 /* We need only the email to just compare it and not
392 the full address which would make "a <a@a.com>"
393 different from "a@a.com" */
394 const gchar *email = get_email_from_address (address);
396 /* ignore the address if already seen */
397 if (g_hash_table_lookup (table, email) == 0) {
400 /* Include the full address and not only the
401 email in the returned list */
403 tmp = g_strdup (address);
405 tmp = g_strjoin (",", new_list, address, NULL);
410 g_hash_table_insert (table, (gchar*)email, GINT_TO_POINTER(1));
412 cursor = g_slist_next (cursor);
415 g_hash_table_unref (table);
416 g_slist_foreach (addresses, (GFunc)g_free, NULL);
417 g_slist_free (addresses);
424 modest_text_utils_convert_buffer_to_html_start (GString *html, const gchar *data, gssize n)
427 gboolean space_seen = FALSE;
428 guint break_dist = 0; /* distance since last break point */
433 /* replace with special html chars where needed*/
434 for (i = 0; i != n; ++i) {
435 guchar kar = data[i];
437 if (space_seen && kar != ' ') {
438 g_string_append_c (html, ' ');
442 /* we artificially insert a breakpoint (newline)
443 * after 256, to make sure our lines are not so long
444 * they will DOS the regexping later
445 * Also, check that kar is ASCII to make sure that we
446 * don't break a UTF8 char in two
448 if (++break_dist >= 256 && kar < 127) {
449 g_string_append_c (html, '\n');
457 /* this is a temp place holder for '&'; we can only
458 * set the real '&' after hyperlink translation, otherwise
459 * we might screw that up */
460 break; /* ignore embedded \0s and MARK_AMP */
461 case '<' : g_string_append (html, MARK_AMP_STR "lt;"); break;
462 case '>' : g_string_append (html, MARK_AMP_STR "gt;"); break;
463 case '&' : g_string_append (html, MARK_AMP_URI_STR "amp;"); break; /* special case */
464 case '"' : g_string_append (html, MARK_AMP_STR "quot;"); break;
466 /* don't convert ' --> wpeditor will try to re-convert it... */
467 //case '\'' : g_string_append (html, "'"); break;
468 case '\n' : g_string_append (html, "<br>\n");break_dist= 0; break;
469 case '\t' : g_string_append (html, MARK_AMP_STR "nbsp;" MARK_AMP_STR "nbsp;" MARK_AMP_STR "nbsp; ");
470 break_dist=0; break; /* note the space at the end*/
473 if (space_seen) { /* second space in a row */
474 g_string_append (html, " ");
480 g_string_append_c (html, kar);
487 modest_text_utils_convert_buffer_to_html_finish (GString *html)
490 /* replace all our MARK_AMPs with real ones */
491 for (i = 0; i != html->len; ++i)
492 if ((html->str)[i] == MARK_AMP || (html->str)[i] == MARK_AMP_URI)
493 (html->str)[i] = '&';
498 modest_text_utils_convert_to_html (const gchar *data)
503 g_return_val_if_fail (data, NULL);
509 html = g_string_sized_new (1.5 * len); /* just a guess... */
511 g_string_append_printf (html,
513 "<meta http-equiv=\"content-type\" content=\"text/html; charset=utf8\">"
517 modest_text_utils_convert_buffer_to_html_start (html, data, -1);
519 g_string_append (html, "</body></html>");
521 if (len <= HYPERLINKIFY_MAX_LENGTH)
522 hyperlinkify_plain_text (html, 0);
524 modest_text_utils_convert_buffer_to_html_finish (html);
526 return g_string_free (html, FALSE);
530 modest_text_utils_convert_to_html_body (const gchar *data, gssize n, gboolean hyperlinkify)
534 g_return_val_if_fail (data, NULL);
541 html = g_string_sized_new (1.5 * n); /* just a guess... */
543 modest_text_utils_convert_buffer_to_html_start (html, data, n);
545 if (hyperlinkify && (n < HYPERLINKIFY_MAX_LENGTH))
546 hyperlinkify_plain_text (html, 0);
548 modest_text_utils_convert_buffer_to_html_finish (html);
550 return g_string_free (html, FALSE);
554 modest_text_utils_get_addresses_indexes (const gchar *addresses, GSList **start_indexes, GSList **end_indexes)
556 gchar *current, *start, *last_blank;
557 gint start_offset = 0, current_offset = 0;
559 g_return_if_fail (start_indexes != NULL);
560 g_return_if_fail (end_indexes != NULL);
562 start = (gchar *) addresses;
566 while (*current != '\0') {
567 if ((start == current)&&((*current == ' ')||(*current == ',')||(*current == ';'))) {
568 start = g_utf8_next_char (start);
570 last_blank = current;
571 } else if ((*current == ',')||(*current == ';')) {
572 gint *start_index, *end_index;
573 start_index = g_new0(gint, 1);
574 end_index = g_new0(gint, 1);
575 *start_index = start_offset;
576 *end_index = current_offset;
577 *start_indexes = g_slist_prepend (*start_indexes, start_index);
578 *end_indexes = g_slist_prepend (*end_indexes, end_index);
579 start = g_utf8_next_char (current);
580 start_offset = current_offset + 1;
582 } else if (*current == '"') {
583 current = g_utf8_next_char (current);
585 while ((*current != '"')&&(*current != '\0')) {
586 current = g_utf8_next_char (current);
591 current = g_utf8_next_char (current);
595 if (start != current) {
596 gint *start_index, *end_index;
597 start_index = g_new0(gint, 1);
598 end_index = g_new0(gint, 1);
599 *start_index = start_offset;
600 *end_index = current_offset;
601 *start_indexes = g_slist_prepend (*start_indexes, start_index);
602 *end_indexes = g_slist_prepend (*end_indexes, end_index);
605 *start_indexes = g_slist_reverse (*start_indexes);
606 *end_indexes = g_slist_reverse (*end_indexes);
613 modest_text_utils_split_addresses_list (const gchar *addresses)
616 const gchar *my_addrs = addresses;
619 gboolean after_at = FALSE;
621 g_return_val_if_fail (addresses, NULL);
623 /* skip any space, ',', ';' at the start */
624 while (my_addrs && (my_addrs[0] == ' ' || my_addrs[0] == ',' || my_addrs[0] == ';'))
627 /* are we at the end of addresses list? */
631 /* nope, we are at the start of some address
632 * now, let's find the end of the address */
634 while (end[0] && end[0] != ';' && !(after_at && end[0] == ',')) {
635 if (end[0] == '\"') {
636 while (end[0] && end[0] != '\"')
642 if ((end[0] && end[0] == '>')&&(end[1] && end[1] == ',')) {
649 /* we got the address; copy it and remove trailing whitespace */
650 addr = g_strndup (my_addrs, end - my_addrs);
653 head = g_slist_append (NULL, addr);
654 head->next = modest_text_utils_split_addresses_list (end); /* recurse */
661 modest_text_utils_address_range_at_position (const gchar *recipients_list,
666 gchar *current = NULL;
667 gint range_start = 0;
670 gboolean is_quoted = FALSE;
672 g_return_if_fail (recipients_list);
673 g_return_if_fail (position < g_utf8_strlen(recipients_list, -1));
676 for (current = (gchar *) recipients_list; *current != '\0';
677 current = g_utf8_find_next_char (current, NULL)) {
678 gunichar c = g_utf8_get_char (current);
680 if ((c == ',') && (!is_quoted)) {
681 if (index < position) {
682 range_start = index + 1;
686 } else if (c == '\"') {
687 is_quoted = !is_quoted;
688 } else if ((c == ' ') &&(range_start == index)) {
696 *start = range_start;
702 modest_text_utils_address_with_standard_length (const gchar *recipients_list)
706 GString *buffer = g_string_new ("");
708 splitted = g_strsplit (recipients_list, "\n", 0);
712 if (current != splitted)
713 buffer = g_string_append_c (buffer, '\n');
714 line = g_strndup (*splitted, 1000);
715 buffer = g_string_append (buffer, line);
720 g_strfreev (splitted);
722 return g_string_free (buffer, FALSE);
726 /* ******************************************************************* */
727 /* ************************* UTILIY FUNCTIONS ************************ */
728 /* ******************************************************************* */
731 get_next_line (const gchar * b, const gsize blen, const gchar * iter)
737 return g_string_new("");
745 gs = g_string_new_len (i0, iter - i0);
749 get_indent_level (const char *l)
766 /* if we hit the signature marker "-- ", we return -(indent + 1). This
767 * stops reformatting.
769 if (strcmp (l, "-- ") == 0) {
777 unquote_line (GString * l, const gchar *quote_symbol)
783 quote_len = strlen (quote_symbol);
785 if (g_str_has_prefix (p, quote_symbol)) {
786 if (p[quote_len] == ' ') {
794 g_string_erase (l, 0, p - l->str);
798 append_quoted (GString * buf, const gchar *quote_symbol,
799 int indent, const GString * str,
805 indent = indent < 0 ? abs (indent) - 1 : indent;
806 quote_concat = g_strconcat (quote_symbol, " ", NULL);
807 for (i = 0; i <= indent; i++) {
808 g_string_append (buf, quote_concat);
810 g_free (quote_concat);
812 g_string_append_len (buf, str->str, cutpoint);
814 g_string_append (buf, str->str);
816 g_string_append (buf, "\n");
820 get_breakpoint_utf8 (const gchar * s, gint indent, const gint limit)
823 const gchar *pos, *last;
826 indent = indent < 0 ? abs (indent) - 1 : indent;
830 uni = g_utf8_to_ucs4_fast (s, -1, NULL);
832 if ((index + 2 * indent > limit) && last) {
836 if (g_unichar_isspace (uni[index])) {
839 pos = g_utf8_next_char (pos);
847 get_breakpoint_ascii (const gchar * s, const gint indent, const gint limit)
852 if (last + 2 * indent < limit)
855 for (i = strlen (s); i > 0; i--) {
857 if (i + 2 * indent <= limit) {
868 get_breakpoint (const gchar * s, const gint indent, const gint limit)
871 if (g_utf8_validate (s, -1, NULL)) {
872 return get_breakpoint_utf8 (s, indent, limit);
873 } else { /* assume ASCII */
874 //g_warning("invalid UTF-8 in msg");
875 return get_breakpoint_ascii (s, indent, limit);
880 cite (const time_t sent_date, const gchar *from)
882 return g_strdup (_("mcen_ia_editor_original_message"));
886 quoted_attachments (GList *attachments)
889 GString *result = g_string_new ("");
890 for (node = attachments; node != NULL; node = g_list_next (node)) {
891 gchar *filename = (gchar *) node->data;
892 g_string_append_printf ( result, "%s %s\n", _("mcen_ia_editor_attach_filename"), filename);
895 return g_string_free (result, FALSE);
900 modest_text_utils_quote_body (GString *output, const gchar *text,
901 const gchar *quote_symbol,
907 gint indent, breakpoint, rem_indent = 0;
908 GString *l, *remaining;
912 remaining = g_string_new ("");
914 l = get_next_line (text, len, iter);
915 iter = iter + l->len + 1;
916 indent = get_indent_level (l->str);
917 unquote_line (l, quote_symbol);
919 if (remaining->len) {
920 if (l->len && indent == rem_indent) {
921 g_string_prepend (l, " ");
922 g_string_prepend (l, remaining->str);
926 get_breakpoint (remaining->str,
929 append_quoted (output, quote_symbol, rem_indent,
930 remaining, breakpoint);
931 g_string_erase (remaining, 0,
933 if (remaining->str[0] == ' ') {
934 g_string_erase (remaining, 0,
937 } while (remaining->len);
940 g_string_free (remaining, TRUE);
941 breakpoint = get_breakpoint (l->str, indent, limit);
942 remaining = g_string_new (l->str + breakpoint);
943 if (remaining->str[0] == ' ') {
944 g_string_erase (remaining, 0, 1);
947 append_quoted (output, quote_symbol, indent, l, breakpoint);
948 g_string_free (l, TRUE);
949 } while ((iter < text + len) || (remaining->str[0]));
955 modest_text_utils_quote_plain_text (const gchar *text,
957 const gchar *signature,
962 gchar *attachments_string = NULL;
964 q = g_string_new ("");
966 if (signature != NULL) {
967 q = g_string_append (q, "\n--\n");
968 q = g_string_append (q, signature);
971 q = g_string_append (q, "\n");
972 q = g_string_append (q, cite);
973 q = g_string_append_c (q, '\n');
975 q = modest_text_utils_quote_body (q, text, ">", limit);
977 attachments_string = quoted_attachments (attachments);
978 q = g_string_append (q, attachments_string);
979 g_free (attachments_string);
981 return g_string_free (q, FALSE);
985 quote_html_add_to_gstring (GString *string,
988 if (text && strcmp (text, "")) {
989 gchar *html_text = modest_text_utils_convert_to_html_body (text, -1, TRUE);
990 g_string_append_printf (string, "%s<br/>", html_text);
996 modest_text_utils_quote_html (const gchar *text,
998 const gchar *signature,
1002 GString *result_string;
1006 "<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\">\n" \
1010 if (text || cite || signature) {
1011 GString *quoted_text;
1012 g_string_append (result_string, "<pre>\n");
1014 quote_html_add_to_gstring (result_string, SIGNATURE_MARKER);
1015 quote_html_add_to_gstring (result_string, signature);
1017 quote_html_add_to_gstring (result_string, cite);
1018 quoted_text = g_string_new ("");
1019 quoted_text = modest_text_utils_quote_body (quoted_text, text, ">", limit);
1020 quote_html_add_to_gstring (result_string, quoted_text->str);
1021 g_string_free (quoted_text, TRUE);
1023 gchar *attachments_string = quoted_attachments (attachments);
1024 quote_html_add_to_gstring (result_string, attachments_string);
1025 g_free (attachments_string);
1027 g_string_append (result_string, "</pre>");
1029 g_string_append (result_string, "</body>");
1030 g_string_append (result_string, "</html>");
1032 return g_string_free (result_string, FALSE);
1036 cmp_offsets_reverse (const url_match_t *match1, const url_match_t *match2)
1038 return match2->offset - match1->offset;
1041 static gint url_matches_block = 0;
1042 static url_match_pattern_t patterns[] = MAIL_VIEWER_URL_MATCH_PATTERNS;
1049 const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
1050 for (i = 0; i != pattern_num; ++i) {
1051 patterns[i].preg = g_slice_new0 (regex_t);
1053 /* this should not happen */
1054 if (regcomp (patterns[i].preg, patterns[i].regex,
1055 REG_ICASE|REG_EXTENDED|REG_NEWLINE) != 0) {
1056 g_warning ("%s: error in regexp:\n%s\n", __FUNCTION__, patterns[i].regex);
1067 const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
1068 for (i = 0; i != pattern_num; ++i) {
1069 regfree (patterns[i].preg);
1070 g_slice_free (regex_t, patterns[i].preg);
1071 } /* don't free patterns itself -- it's static */
1075 modest_text_utils_hyperlinkify_begin (void)
1077 if (url_matches_block == 0)
1078 compile_patterns ();
1079 url_matches_block ++;
1083 modest_text_utils_hyperlinkify_end (void)
1085 url_matches_block--;
1086 if (url_matches_block <= 0)
1092 get_url_matches (GString *txt, gint offset)
1095 guint rv, i, tmp_offset = 0;
1096 GSList *match_list = NULL;
1098 const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
1100 /* initalize the regexps */
1101 modest_text_utils_hyperlinkify_begin ();
1103 /* find all the matches */
1104 for (i = 0; i != pattern_num; ++i) {
1105 tmp_offset = offset;
1108 gboolean is_submatch;
1111 if ((rv = regexec (patterns[i].preg, txt->str + tmp_offset, 1, &rm, 0)) != 0) {
1112 g_return_val_if_fail (rv == REG_NOMATCH, NULL); /* this should not happen */
1113 break; /* try next regexp */
1118 is_submatch = FALSE;
1119 /* check old matches to see if this has already been matched */
1120 cursor = match_list;
1121 while (cursor && !is_submatch) {
1122 const url_match_t *old_match =
1123 (const url_match_t *) cursor->data;
1124 guint new_offset = tmp_offset + rm.rm_so;
1125 is_submatch = (new_offset > old_match->offset &&
1126 new_offset < old_match->offset + old_match->len);
1127 cursor = g_slist_next (cursor);
1131 /* make a list of our matches (<offset, len, prefix> tupels)*/
1132 match = g_slice_new (url_match_t);
1133 match->offset = tmp_offset + rm.rm_so;
1134 match->len = rm.rm_eo - rm.rm_so;
1135 match->prefix = patterns[i].prefix;
1136 match_list = g_slist_prepend (match_list, match);
1138 tmp_offset += rm.rm_eo;
1142 modest_text_utils_hyperlinkify_end ();
1144 /* now sort the list, so the matches are in reverse order of occurence.
1145 * that way, we can do the replacements starting from the end, so we don't need
1146 * to recalculate the offsets
1148 match_list = g_slist_sort (match_list,
1149 (GCompareFunc)cmp_offsets_reverse);
1155 /* replace all occurences of needle in haystack with repl*/
1157 replace_string (const gchar *haystack, const gchar *needle, gchar repl)
1159 gchar *str, *cursor;
1161 if (!haystack || !needle || strlen(needle) == 0)
1162 return haystack ? g_strdup(haystack) : NULL;
1164 str = g_strdup (haystack);
1166 for (cursor = str; cursor && *cursor; ++cursor) {
1167 if (g_str_has_prefix (cursor, needle)) {
1169 memmove (cursor + 1,
1170 cursor + strlen (needle),
1171 strlen (cursor + strlen (needle)) + 1);
1179 hyperlinkify_plain_text (GString *txt, gint offset)
1182 GSList *match_list = get_url_matches (txt, offset);
1184 /* we will work backwards, so the offsets stay valid */
1185 for (cursor = match_list; cursor; cursor = cursor->next) {
1187 url_match_t *match = (url_match_t*) cursor->data;
1188 gchar *url = g_strndup (txt->str + match->offset, match->len);
1189 gchar *repl = NULL; /* replacement */
1191 /* the string still contains $(MARK_AMP_URI_STR)"amp;" for each
1192 * '&' in the original, because of the text->html conversion.
1193 * in the href-URL (and only there), we must convert that back to
1196 gchar *href_url = replace_string (url, MARK_AMP_URI_STR "amp;", '&');
1198 /* the prefix is NULL: use the one that is already there */
1199 repl = g_strdup_printf ("<a href=\"%s%s\">%s</a>",
1200 match->prefix ? match->prefix : EMPTY_STRING,
1203 /* replace the old thing with our hyperlink
1204 * replacement thing */
1205 g_string_erase (txt, match->offset, match->len);
1206 g_string_insert (txt, match->offset, repl);
1212 g_slice_free (url_match_t, match);
1215 g_slist_free (match_list);
1219 modest_text_utils_hyperlinkify (GString *string_buffer)
1224 after_body = strstr (string_buffer->str, "<body>");
1225 if (after_body != NULL)
1226 offset = after_body - string_buffer->str;
1227 hyperlinkify_plain_text (string_buffer, offset);
1231 /* for optimization reasons, we change the string in-place */
1233 modest_text_utils_get_display_address (gchar *address)
1237 g_return_if_fail (address);
1242 /* should not be needed, and otherwise, we probably won't screw up the address
1243 * more than it already is :)
1244 * g_return_val_if_fail (g_utf8_validate (address, -1, NULL), NULL);
1247 /* remove leading whitespace */
1248 if (address[0] == ' ')
1249 g_strchug (address);
1251 for (i = 0; address[i]; ++i) {
1252 if (address[i] == '<') {
1253 if (G_UNLIKELY(i == 0)) {
1254 break; /* there's nothing else, leave it */
1256 address[i] = '\0'; /* terminate the string here */
1262 g_strchomp (address);
1270 modest_text_utils_get_email_address (const gchar *full_address)
1272 const gchar *left, *right;
1274 g_return_val_if_fail (full_address, NULL);
1279 g_return_val_if_fail (g_utf8_validate (full_address, -1, NULL), NULL);
1281 left = g_strrstr_len (full_address, strlen(full_address), "<");
1283 return g_strdup (full_address);
1285 right = g_strstr_len (left, strlen(left), ">");
1287 return g_strdup (full_address);
1289 return g_strndup (left + 1, right - left - 1);
1293 modest_text_utils_get_subject_prefix_len (const gchar *sub)
1295 gint prefix_len = 0;
1297 g_return_val_if_fail (sub, 0);
1302 /* optimization: "Re", "RE", "re","Fwd", "FWD", "fwd","FW","Fw", "fw" */
1303 if (sub[0] != 'R' && sub[0] != 'F' && sub[0] != 'r' && sub[0] != 'f')
1305 else if (sub[0] && sub[1] != 'e' && sub[1] != 'E' && sub[1] != 'w' && sub[1] != 'W')
1312 /* skip over a [...] block */
1313 if (sub[prefix_len] == '[') {
1314 int c = prefix_len + 1;
1315 while (sub[c] && sub[c] != ']')
1318 return 0; /* no end to the ']' found */
1323 /* did we find the ':' ? */
1324 if (sub[prefix_len] == ':') {
1326 if (sub[prefix_len] == ' ')
1328 prefix_len += modest_text_utils_get_subject_prefix_len (sub + prefix_len);
1329 /* g_warning ("['%s','%s']", sub, (char*) sub + prefix_len); */
1337 modest_text_utils_utf8_strcmp (const gchar* s1, const gchar *s2, gboolean insensitive)
1340 /* work even when s1 and/or s2 == NULL */
1341 if (G_UNLIKELY(s1 == s2))
1343 if (G_UNLIKELY(!s1))
1345 if (G_UNLIKELY(!s2))
1348 /* if it's not case sensitive */
1351 /* optimization: shortcut if first char is ascii */
1352 if (((s1[0] & 0x80)== 0) && ((s2[0] & 0x80) == 0) &&
1354 return s1[0] - s2[0];
1356 return g_utf8_collate (s1, s2);
1362 /* optimization: shortcut if first char is ascii */
1363 if (((s1[0] & 0x80) == 0) && ((s2[0] & 0x80) == 0) &&
1364 (tolower(s1[0]) != tolower (s2[0])))
1365 return tolower(s1[0]) - tolower(s2[0]);
1367 n1 = g_utf8_strdown (s1, -1);
1368 n2 = g_utf8_strdown (s2, -1);
1370 result = g_utf8_collate (n1, n2);
1381 modest_text_utils_get_display_date (time_t date)
1383 #define DATE_BUF_SIZE 64
1384 static gchar date_buf[DATE_BUF_SIZE];
1386 /* calculate the # of days since epoch for
1387 * for today and for the date provided
1388 * based on idea from pvanhoof */
1389 int day = time(NULL) / (24 * 60 * 60);
1390 int date_day = date / (24 * 60 * 60);
1392 /* if it's today, show the time, if it's not today, show the date instead */
1394 /* TODO: take into account the system config for 24/12h */
1395 if (day == date_day) /* is the date today? */
1396 modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, _HL("wdgt_va_24h_time"), date);
1398 modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, _HL("wdgt_va_date"), date);
1400 return date_buf; /* this is a static buffer, don't free! */
1406 modest_text_utils_validate_folder_name (const gchar *folder_name)
1408 /* based on http://msdn2.microsoft.com/en-us/library/aa365247.aspx,
1409 * with some extras */
1413 const gchar **cursor = NULL;
1414 const gchar *forbidden_names[] = { /* windows does not like these */
1415 "CON", "PRN", "AUX", "NUL", ".", "..", "cur", "tmp", "new",
1416 NULL /* cur, tmp, new are reserved for Maildir */
1419 /* cannot be NULL */
1423 /* cannot be empty */
1424 len = strlen(folder_name);
1428 /* cannot start with a dot, vfat does not seem to like that */
1429 if (folder_name[0] == '.')
1432 /* cannot start or end with a space */
1433 if (g_ascii_isspace(folder_name[0]) || g_ascii_isspace(folder_name[len - 1]))
1436 /* cannot contain a forbidden char */
1437 for (i = 0; i < len; i++)
1438 if (modest_text_utils_is_forbidden_char (folder_name[i], FOLDER_NAME_FORBIDDEN_CHARS))
1441 /* Cannot contain Windows port numbers. I'd like to use GRegex
1442 but it's still not available in Maemo. sergio */
1443 if (!g_ascii_strncasecmp (folder_name, "LPT", 3) ||
1444 !g_ascii_strncasecmp (folder_name, "COM", 3)) {
1448 /* We skip the first 3 characters for the
1450 val = strtol(folder_name+3, &endptr, 10);
1452 /* If the conversion to long succeeded then the string
1453 is not valid for us */
1454 if (*endptr == '\0')
1460 /* cannot contain a forbidden word */
1462 for (cursor = forbidden_names; cursor && *cursor; ++cursor) {
1463 if (g_ascii_strcasecmp (folder_name, *cursor) == 0)
1468 return TRUE; /* it's valid! */
1474 modest_text_utils_validate_domain_name (const gchar *domain)
1476 gboolean valid = FALSE;
1478 const gchar* domain_regex = "^([a-z0-9-]*[a-z0-9]\\.)+[a-z0-9-]*[a-z0-9]$";
1480 g_return_val_if_fail (domain, FALSE);
1485 memset (&rx, 0, sizeof(regex_t)); /* coverity wants this... */
1487 /* domain name: all alphanum or '-' or '.',
1488 * but beginning/ending in alphanum */
1489 if (regcomp (&rx, domain_regex, REG_ICASE|REG_EXTENDED|REG_NOSUB)) {
1490 g_warning ("BUG: error in regexp");
1494 valid = (regexec (&rx, domain, 1, NULL, 0) == 0);
1503 modest_text_utils_validate_email_address (const gchar *email_address,
1504 const gchar **invalid_char_position)
1507 const gchar *c = NULL, *domain = NULL;
1508 static gchar *rfc822_specials = "()<>@,;:\\\"[]&";
1510 if (invalid_char_position)
1511 *invalid_char_position = NULL;
1513 g_return_val_if_fail (email_address, FALSE);
1515 /* check that the email adress contains exactly one @ */
1516 if (!strstr(email_address, "@") ||
1517 (strstr(email_address, "@") != g_strrstr(email_address, "@")))
1520 /* first we validate the name portion (name@domain) */
1521 for (c = email_address; *c; c++) {
1523 (c == email_address ||
1525 *(c - 1) == '\"')) {
1529 if (*c == '\\' && (*++c == ' '))
1531 if (*c <= ' ' || *c >= 127)
1544 if (*c <= ' ' || *c >= 127)
1546 if (strchr(rfc822_specials, *c)) {
1547 if (invalid_char_position)
1548 *invalid_char_position = c;
1552 if (c == email_address || *(c - 1) == '.')
1555 /* next we validate the domain portion (name@domain) */
1556 if (!*(domain = ++c))
1560 if (c == domain || *(c - 1) == '.' || *(c + 1) == '\0')
1564 if (*c <= ' ' || *c >= 127)
1566 if (strchr(rfc822_specials, *c)) {
1567 if (invalid_char_position)
1568 *invalid_char_position = c;
1573 return (count >= 1) ? TRUE : FALSE;
1577 modest_text_utils_validate_recipient (const gchar *recipient, const gchar **invalid_char_position)
1579 gchar *stripped, *current;
1581 gboolean has_error = FALSE;
1583 if (invalid_char_position)
1584 *invalid_char_position = NULL;
1586 g_return_val_if_fail (recipient, FALSE);
1588 if (modest_text_utils_validate_email_address (recipient, invalid_char_position))
1591 stripped = g_strdup (recipient);
1592 stripped = g_strstrip (stripped);
1595 if (*current == '\0') {
1601 if (*current == '\"') {
1602 current = g_utf8_next_char (current);
1604 for (; *current != '\0'; current = g_utf8_next_char (current)) {
1605 if (*current == '\\') {
1606 /* TODO: This causes a warning, which breaks the build,
1607 * because a gchar cannot be < 0.
1609 if (current[1] <0) {
1614 } else if (*current == '\"') {
1616 current = g_utf8_next_char (current);
1622 for (current = stripped ; *current != '\0'; current = g_utf8_next_char (current)) {
1623 if (*current == '<') {
1635 right_part = g_strdup (current);
1637 right_part = g_strstrip (right_part);
1639 if (g_str_has_prefix (right_part, "<") &&
1640 g_str_has_suffix (right_part, ">")) {
1644 address = g_strndup (right_part+1, strlen (right_part) - 2);
1645 g_free (right_part);
1646 valid = modest_text_utils_validate_email_address (address, invalid_char_position);
1650 g_free (right_part);
1657 modest_text_utils_get_display_size (guint64 size)
1659 const guint KB=1024;
1660 const guint MB=1024 * KB;
1661 const guint GB=1024 * MB;
1664 return g_strdup_printf (_FM("sfil_li_size_kb"), (int) 0);
1665 if (0 <= size && size < KB)
1666 return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), (int) 1);
1667 else if (KB <= size && size < 100 * KB)
1668 return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), (int) size / KB);
1669 else if (100*KB <= size && size < MB)
1670 return g_strdup_printf (_FM("sfil_li_size_100kb_1mb"), (float) size / MB);
1671 else if (MB <= size && size < 10*MB)
1672 return g_strdup_printf (_FM("sfil_li_size_1mb_10mb"), (float) size / MB);
1673 else if (10*MB <= size && size < GB)
1674 return g_strdup_printf (_FM("sfil_li_size_10mb_1gb"), (int) size / MB);
1676 return g_strdup_printf (_FM("sfil_li_size_1gb_or_greater"), (float) size / GB);
1680 get_email_from_address (const gchar * address)
1682 gchar *left_limit, *right_limit;
1684 left_limit = strstr (address, "<");
1685 right_limit = g_strrstr (address, ">");
1687 if ((left_limit == NULL)||(right_limit == NULL)|| (left_limit > right_limit))
1688 return g_strdup (address);
1690 return g_strndup (left_limit + 1, (right_limit - left_limit) - 1);
1694 modest_text_utils_get_color_string (GdkColor *color)
1696 g_return_val_if_fail (color, NULL);
1698 return g_strdup_printf ("#%x%x%x%x%x%x%x%x%x%x%x%x",
1699 (color->red >> 12) & 0xf, (color->red >> 8) & 0xf,
1700 (color->red >> 4) & 0xf, (color->red) & 0xf,
1701 (color->green >> 12) & 0xf, (color->green >> 8) & 0xf,
1702 (color->green >> 4) & 0xf, (color->green) & 0xf,
1703 (color->blue >> 12) & 0xf, (color->blue >> 8) & 0xf,
1704 (color->blue >> 4) & 0xf, (color->blue) & 0xf);
1708 modest_text_utils_text_buffer_get_text (GtkTextBuffer *buffer)
1710 GtkTextIter start, end;
1711 gchar *slice, *current;
1712 GString *result = g_string_new ("");
1714 g_return_val_if_fail (buffer && GTK_IS_TEXT_BUFFER (buffer), NULL);
1716 gtk_text_buffer_get_start_iter (buffer, &start);
1717 gtk_text_buffer_get_end_iter (buffer, &end);
1719 slice = gtk_text_buffer_get_slice (buffer, &start, &end, FALSE);
1722 while (current && current != '\0') {
1723 if (g_utf8_get_char (current) == 0xFFFC) {
1724 result = g_string_append_c (result, ' ');
1725 current = g_utf8_next_char (current);
1727 gchar *next = g_utf8_strchr (current, -1, 0xFFFC);
1729 result = g_string_append (result, current);
1731 result = g_string_append_len (result, current, next - current);
1738 return g_string_free (result, FALSE);
1743 modest_text_utils_is_forbidden_char (const gchar character,
1744 ModestTextUtilsForbiddenCharType type)
1747 const gchar *forbidden_chars = NULL;
1749 /* We need to get the length in the switch because the
1750 compiler needs to know the size at compile time */
1752 case ACCOUNT_TITLE_FORBIDDEN_CHARS:
1753 forbidden_chars = account_title_forbidden_chars;
1754 len = G_N_ELEMENTS (account_title_forbidden_chars);
1756 case FOLDER_NAME_FORBIDDEN_CHARS:
1757 forbidden_chars = folder_name_forbidden_chars;
1758 len = G_N_ELEMENTS (folder_name_forbidden_chars);
1760 case USER_NAME_FORBIDDEN_NAMES:
1761 forbidden_chars = user_name_forbidden_chars;
1762 len = G_N_ELEMENTS (user_name_forbidden_chars);
1765 g_return_val_if_reached (TRUE);
1768 for (i = 0; i < len ; i++)
1769 if (forbidden_chars[i] == character)
1772 return FALSE; /* it's valid! */
1776 modest_text_utils_label_get_selection (GtkLabel *label)
1781 if (gtk_label_get_selection_bounds (GTK_LABEL (label), &start, &end)) {
1782 const gchar *start_offset;
1783 const gchar *end_offset;
1784 start_offset = gtk_label_get_text (GTK_LABEL (label));
1785 start_offset = g_utf8_offset_to_pointer (start_offset, start);
1786 end_offset = gtk_label_get_text (GTK_LABEL (label));
1787 end_offset = g_utf8_offset_to_pointer (end_offset, end);
1788 selection = g_strndup (start_offset, end_offset - start_offset);
1791 return g_strdup ("");
1796 _forward_search_image_char (gunichar ch,
1799 return (ch == 0xFFFC);
1803 modest_text_utils_buffer_selection_is_valid (GtkTextBuffer *buffer)
1806 GtkTextIter start, end;
1808 g_return_val_if_fail (GTK_IS_TEXT_BUFFER (buffer), FALSE);
1810 result = gtk_text_buffer_get_has_selection (GTK_TEXT_BUFFER (buffer));
1812 /* check there are no images in selection */
1814 gtk_text_buffer_get_selection_bounds (buffer, &start, &end);
1815 if (gtk_text_iter_get_char (&start)== 0xFFFC)
1818 gtk_text_iter_backward_char (&end);
1819 if (gtk_text_iter_forward_find_char (&start, _forward_search_image_char,
1830 modest_text_utils_escape_mnemonics (const gchar *text)
1833 GString *result = NULL;
1838 result = g_string_new ("");
1839 for (p = text; *p != '\0'; p++) {
1841 result = g_string_append (result, "__");
1843 result = g_string_append_c (result, *p);
1846 return g_string_free (result, FALSE);