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 $(MARK_AMP_URI_STR)"amp;" */
111 #define MAIL_VIEWER_URL_MATCH_PATTERNS { \
112 { "(feed:|)(file|rtsp|http|ftp|https|mms|mmsh|webcal|feed|rtsp|rdp|lastfm|sip)://[-a-z0-9_$.+!*(),;:@%=?/~#" 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 */
292 /* does not work on old maemo glib:
293 * g_date_set_time_t (&date, timet);
295 localtime_r (&timet, &tm);
296 return strftime(s, max, fmt, &tm);
300 modest_text_utils_derived_subject (const gchar *subject, const gchar *prefix)
302 gchar *tmp, *subject_dup, *retval;
305 g_return_val_if_fail (prefix, NULL);
307 if (!subject || subject[0] == '\0')
308 subject = _("mail_va_no_subject");
310 subject_dup = g_strdup (subject);
311 tmp = g_strchug (subject_dup);
313 /* We do not want things like "Re: Re: Re:" or "Fw: Fw:" so
314 delete the previous ones */
315 prefix_len = strlen (prefix);
317 if (g_str_has_prefix (tmp, prefix)) {
319 tmp = g_strchug (tmp);
325 retval = g_strdup_printf ("%s %s", prefix, tmp);
326 g_free (subject_dup);
332 modest_text_utils_remove_address (const gchar *address_list, const gchar *address)
334 gchar *dup, *token, *ptr = NULL, *result;
335 GString *filtered_emails;
336 gchar *email_address;
338 g_return_val_if_fail (address_list, NULL);
341 return g_strdup (address_list);
343 email_address = get_email_from_address (address);
345 /* search for substring */
346 if (!strstr ((const char *) address_list, (const char *) email_address)) {
347 g_free (email_address);
348 return g_strdup (address_list);
351 dup = g_strdup (address_list);
352 filtered_emails = g_string_new (NULL);
354 token = strtok_r (dup, ",", &ptr);
356 while (token != NULL) {
357 /* Add to list if not found */
358 if (!strstr ((const char *) token, (const char *) email_address)) {
359 if (filtered_emails->len == 0)
360 g_string_append_printf (filtered_emails, "%s", g_strstrip (token));
362 g_string_append_printf (filtered_emails, ",%s", g_strstrip (token));
364 token = strtok_r (NULL, ",", &ptr);
366 result = filtered_emails->str;
369 g_free (email_address);
371 g_string_free (filtered_emails, FALSE);
378 modest_text_utils_remove_duplicate_addresses (const gchar *address_list)
380 GSList *addresses, *cursor;
382 gchar *new_list = NULL;
384 g_return_val_if_fail (address_list, NULL);
386 table = g_hash_table_new_full (g_str_hash, g_str_equal, g_free, NULL);
387 addresses = modest_text_utils_split_addresses_list (address_list);
391 const gchar* address = (const gchar*)cursor->data;
393 /* We need only the email to just compare it and not
394 the full address which would make "a <a@a.com>"
395 different from "a@a.com" */
396 const gchar *email = get_email_from_address (address);
398 /* ignore the address if already seen */
399 if (g_hash_table_lookup (table, email) == 0) {
402 /* Include the full address and not only the
403 email in the returned list */
405 tmp = g_strdup (address);
407 tmp = g_strjoin (",", new_list, address, NULL);
412 g_hash_table_insert (table, (gchar*)email, GINT_TO_POINTER(1));
414 cursor = g_slist_next (cursor);
417 g_hash_table_unref (table);
418 g_slist_foreach (addresses, (GFunc)g_free, NULL);
419 g_slist_free (addresses);
426 modest_text_utils_convert_buffer_to_html_start (GString *html, const gchar *data, gssize n)
429 gboolean space_seen = FALSE;
430 guint break_dist = 0; /* distance since last break point */
435 /* replace with special html chars where needed*/
436 for (i = 0; i != n; ++i) {
437 guchar kar = data[i];
439 if (space_seen && kar != ' ') {
440 g_string_append_c (html, ' ');
444 /* we artificially insert a breakpoint (newline)
445 * after 256, to make sure our lines are not so long
446 * they will DOS the regexping later
447 * Also, check that kar is ASCII to make sure that we
448 * don't break a UTF8 char in two
450 if (++break_dist >= 256 && kar < 127) {
451 g_string_append_c (html, '\n');
459 /* this is a temp place holder for '&'; we can only
460 * set the real '&' after hyperlink translation, otherwise
461 * we might screw that up */
462 break; /* ignore embedded \0s and MARK_AMP */
463 case '<' : g_string_append (html, MARK_AMP_STR "lt;"); break;
464 case '>' : g_string_append (html, MARK_AMP_STR "gt;"); break;
465 case '&' : g_string_append (html, MARK_AMP_URI_STR "amp;"); break; /* special case */
466 case '"' : g_string_append (html, MARK_AMP_STR "quot;"); break;
468 /* don't convert ' --> wpeditor will try to re-convert it... */
469 //case '\'' : g_string_append (html, "'"); break;
470 case '\n' : g_string_append (html, "<br>\n");break_dist= 0; break;
471 case '\t' : g_string_append (html, MARK_AMP_STR "nbsp;" MARK_AMP_STR "nbsp;" MARK_AMP_STR "nbsp; ");
472 break_dist=0; break; /* note the space at the end*/
475 if (space_seen) { /* second space in a row */
476 g_string_append (html, " ");
482 g_string_append_c (html, kar);
489 modest_text_utils_convert_buffer_to_html_finish (GString *html)
492 /* replace all our MARK_AMPs with real ones */
493 for (i = 0; i != html->len; ++i)
494 if ((html->str)[i] == MARK_AMP || (html->str)[i] == MARK_AMP_URI)
495 (html->str)[i] = '&';
500 modest_text_utils_convert_to_html (const gchar *data)
505 g_return_val_if_fail (data, NULL);
511 html = g_string_sized_new (1.5 * len); /* just a guess... */
513 g_string_append_printf (html,
515 "<meta http-equiv=\"content-type\" content=\"text/html; charset=utf8\">"
519 modest_text_utils_convert_buffer_to_html_start (html, data, -1);
521 g_string_append (html, "</body></html>");
523 if (len <= HYPERLINKIFY_MAX_LENGTH)
524 hyperlinkify_plain_text (html, 0);
526 modest_text_utils_convert_buffer_to_html_finish (html);
528 return g_string_free (html, FALSE);
532 modest_text_utils_convert_to_html_body (const gchar *data, gssize n, gboolean hyperlinkify)
536 g_return_val_if_fail (data, NULL);
543 html = g_string_sized_new (1.5 * n); /* just a guess... */
545 modest_text_utils_convert_buffer_to_html_start (html, data, n);
547 if (hyperlinkify && (n < HYPERLINKIFY_MAX_LENGTH))
548 hyperlinkify_plain_text (html, 0);
550 modest_text_utils_convert_buffer_to_html_finish (html);
552 return g_string_free (html, FALSE);
556 modest_text_utils_get_addresses_indexes (const gchar *addresses, GSList **start_indexes, GSList **end_indexes)
558 gchar *current, *start, *last_blank;
559 gint start_offset = 0, current_offset = 0;
561 g_return_if_fail (start_indexes != NULL);
562 g_return_if_fail (end_indexes != NULL);
564 start = (gchar *) addresses;
568 while (*current != '\0') {
569 if ((start == current)&&((*current == ' ')||(*current == ',')||(*current == ';'))) {
570 start = g_utf8_next_char (start);
572 last_blank = current;
573 } else if ((*current == ',')||(*current == ';')) {
574 gint *start_index, *end_index;
575 start_index = g_new0(gint, 1);
576 end_index = g_new0(gint, 1);
577 *start_index = start_offset;
578 *end_index = current_offset;
579 *start_indexes = g_slist_prepend (*start_indexes, start_index);
580 *end_indexes = g_slist_prepend (*end_indexes, end_index);
581 start = g_utf8_next_char (current);
582 start_offset = current_offset + 1;
584 } else if (*current == '"') {
585 current = g_utf8_next_char (current);
587 while ((*current != '"')&&(*current != '\0')) {
588 current = g_utf8_next_char (current);
593 current = g_utf8_next_char (current);
597 if (start != current) {
598 gint *start_index, *end_index;
599 start_index = g_new0(gint, 1);
600 end_index = g_new0(gint, 1);
601 *start_index = start_offset;
602 *end_index = current_offset;
603 *start_indexes = g_slist_prepend (*start_indexes, start_index);
604 *end_indexes = g_slist_prepend (*end_indexes, end_index);
607 *start_indexes = g_slist_reverse (*start_indexes);
608 *end_indexes = g_slist_reverse (*end_indexes);
615 modest_text_utils_split_addresses_list (const gchar *addresses)
618 const gchar *my_addrs = addresses;
621 gboolean after_at = FALSE;
623 g_return_val_if_fail (addresses, NULL);
625 /* skip any space, ',', ';' at the start */
626 while (my_addrs && (my_addrs[0] == ' ' || my_addrs[0] == ',' || my_addrs[0] == ';'))
629 /* are we at the end of addresses list? */
633 /* nope, we are at the start of some address
634 * now, let's find the end of the address */
636 while (end[0] && end[0] != ';' && !(after_at && end[0] == ',')) {
637 if (end[0] == '\"') {
638 while (end[0] && end[0] != '\"')
644 if ((end[0] && end[0] == '>')&&(end[1] && end[1] == ',')) {
651 /* we got the address; copy it and remove trailing whitespace */
652 addr = g_strndup (my_addrs, end - my_addrs);
655 head = g_slist_append (NULL, addr);
656 head->next = modest_text_utils_split_addresses_list (end); /* recurse */
663 modest_text_utils_address_range_at_position (const gchar *recipients_list,
668 gchar *current = NULL;
669 gint range_start = 0;
672 gboolean is_quoted = FALSE;
674 g_return_if_fail (recipients_list);
675 g_return_if_fail (position < g_utf8_strlen(recipients_list, -1));
678 for (current = (gchar *) recipients_list; *current != '\0';
679 current = g_utf8_find_next_char (current, NULL)) {
680 gunichar c = g_utf8_get_char (current);
682 if ((c == ',') && (!is_quoted)) {
683 if (index < position) {
684 range_start = index + 1;
688 } else if (c == '\"') {
689 is_quoted = !is_quoted;
690 } else if ((c == ' ') &&(range_start == index)) {
698 *start = range_start;
704 modest_text_utils_address_with_standard_length (const gchar *recipients_list)
708 GString *buffer = g_string_new ("");
710 splitted = g_strsplit (recipients_list, "\n", 0);
714 if (current != splitted)
715 buffer = g_string_append_c (buffer, '\n');
716 line = g_strndup (*splitted, 1000);
717 buffer = g_string_append (buffer, line);
722 g_strfreev (splitted);
724 return g_string_free (buffer, FALSE);
728 /* ******************************************************************* */
729 /* ************************* UTILIY FUNCTIONS ************************ */
730 /* ******************************************************************* */
733 get_next_line (const gchar * b, const gsize blen, const gchar * iter)
739 return g_string_new("");
747 gs = g_string_new_len (i0, iter - i0);
751 get_indent_level (const char *l)
768 /* if we hit the signature marker "-- ", we return -(indent + 1). This
769 * stops reformatting.
771 if (strcmp (l, "-- ") == 0) {
779 unquote_line (GString * l, const gchar *quote_symbol)
785 quote_len = strlen (quote_symbol);
787 if (g_str_has_prefix (p, quote_symbol)) {
788 if (p[quote_len] == ' ') {
796 g_string_erase (l, 0, p - l->str);
800 append_quoted (GString * buf, const gchar *quote_symbol,
801 int indent, const GString * str,
807 indent = indent < 0 ? abs (indent) - 1 : indent;
808 quote_concat = g_strconcat (quote_symbol, " ", NULL);
809 for (i = 0; i <= indent; i++) {
810 g_string_append (buf, quote_concat);
812 g_free (quote_concat);
814 g_string_append_len (buf, str->str, cutpoint);
816 g_string_append (buf, str->str);
818 g_string_append (buf, "\n");
822 get_breakpoint_utf8 (const gchar * s, gint indent, const gint limit)
825 const gchar *pos, *last;
828 indent = indent < 0 ? abs (indent) - 1 : indent;
832 uni = g_utf8_to_ucs4_fast (s, -1, NULL);
834 if ((index + 2 * indent > limit) && last) {
838 if (g_unichar_isspace (uni[index])) {
841 pos = g_utf8_next_char (pos);
849 get_breakpoint_ascii (const gchar * s, const gint indent, const gint limit)
854 if (last + 2 * indent < limit)
857 for (i = strlen (s); i > 0; i--) {
859 if (i + 2 * indent <= limit) {
870 get_breakpoint (const gchar * s, const gint indent, const gint limit)
873 if (g_utf8_validate (s, -1, NULL)) {
874 return get_breakpoint_utf8 (s, indent, limit);
875 } else { /* assume ASCII */
876 //g_warning("invalid UTF-8 in msg");
877 return get_breakpoint_ascii (s, indent, limit);
882 cite (const time_t sent_date, const gchar *from)
884 return g_strdup (_("mcen_ia_editor_original_message"));
888 quoted_attachments (GList *attachments)
891 GString *result = g_string_new ("");
892 for (node = attachments; node != NULL; node = g_list_next (node)) {
893 gchar *filename = (gchar *) node->data;
894 g_string_append_printf ( result, "%s %s\n", _("mcen_ia_editor_attach_filename"), filename);
897 return g_string_free (result, FALSE);
902 modest_text_utils_quote_body (GString *output, const gchar *text,
903 const gchar *quote_symbol,
909 gint indent, breakpoint, rem_indent = 0;
910 GString *l, *remaining;
914 remaining = g_string_new ("");
916 l = get_next_line (text, len, iter);
917 iter = iter + l->len + 1;
918 indent = get_indent_level (l->str);
919 unquote_line (l, quote_symbol);
921 if (remaining->len) {
922 if (l->len && indent == rem_indent) {
923 g_string_prepend (l, " ");
924 g_string_prepend (l, remaining->str);
928 get_breakpoint (remaining->str,
931 append_quoted (output, quote_symbol, rem_indent,
932 remaining, breakpoint);
933 g_string_erase (remaining, 0,
935 if (remaining->str[0] == ' ') {
936 g_string_erase (remaining, 0,
939 } while (remaining->len);
942 g_string_free (remaining, TRUE);
943 breakpoint = get_breakpoint (l->str, indent, limit);
944 remaining = g_string_new (l->str + breakpoint);
945 if (remaining->str[0] == ' ') {
946 g_string_erase (remaining, 0, 1);
949 append_quoted (output, quote_symbol, indent, l, breakpoint);
950 g_string_free (l, TRUE);
951 } while ((iter < text + len) || (remaining->str[0]));
957 modest_text_utils_quote_plain_text (const gchar *text,
959 const gchar *signature,
964 gchar *attachments_string = NULL;
966 q = g_string_new ("");
968 if (signature != NULL) {
969 q = g_string_append (q, "\n--\n");
970 q = g_string_append (q, signature);
973 q = g_string_append (q, "\n");
974 q = g_string_append (q, cite);
975 q = g_string_append_c (q, '\n');
977 q = modest_text_utils_quote_body (q, text, ">", limit);
979 attachments_string = quoted_attachments (attachments);
980 q = g_string_append (q, attachments_string);
981 g_free (attachments_string);
983 return g_string_free (q, FALSE);
987 quote_html_add_to_gstring (GString *string,
990 if (text && strcmp (text, "")) {
991 gchar *html_text = modest_text_utils_convert_to_html_body (text, -1, TRUE);
992 g_string_append_printf (string, "%s<br/>", html_text);
998 modest_text_utils_quote_html (const gchar *text,
1000 const gchar *signature,
1004 GString *result_string;
1008 "<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\">\n" \
1012 if (text || cite || signature) {
1013 GString *quoted_text;
1014 g_string_append (result_string, "<pre>\n");
1016 quote_html_add_to_gstring (result_string, SIGNATURE_MARKER);
1017 quote_html_add_to_gstring (result_string, signature);
1019 quote_html_add_to_gstring (result_string, cite);
1020 quoted_text = g_string_new ("");
1021 quoted_text = modest_text_utils_quote_body (quoted_text, (text) ? text : "", ">", limit);
1022 quote_html_add_to_gstring (result_string, quoted_text->str);
1023 g_string_free (quoted_text, TRUE);
1025 gchar *attachments_string = quoted_attachments (attachments);
1026 quote_html_add_to_gstring (result_string, attachments_string);
1027 g_free (attachments_string);
1029 g_string_append (result_string, "</pre>");
1031 g_string_append (result_string, "</body>");
1032 g_string_append (result_string, "</html>");
1034 return g_string_free (result_string, FALSE);
1038 cmp_offsets_reverse (const url_match_t *match1, const url_match_t *match2)
1040 return match2->offset - match1->offset;
1043 static gint url_matches_block = 0;
1044 static url_match_pattern_t patterns[] = MAIL_VIEWER_URL_MATCH_PATTERNS;
1051 const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
1052 for (i = 0; i != pattern_num; ++i) {
1053 patterns[i].preg = g_slice_new0 (regex_t);
1055 /* this should not happen */
1056 if (regcomp (patterns[i].preg, patterns[i].regex,
1057 REG_ICASE|REG_EXTENDED|REG_NEWLINE) != 0) {
1058 g_warning ("%s: error in regexp:\n%s\n", __FUNCTION__, patterns[i].regex);
1069 const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
1070 for (i = 0; i != pattern_num; ++i) {
1071 regfree (patterns[i].preg);
1072 g_slice_free (regex_t, patterns[i].preg);
1073 } /* don't free patterns itself -- it's static */
1077 modest_text_utils_hyperlinkify_begin (void)
1079 if (url_matches_block == 0)
1080 compile_patterns ();
1081 url_matches_block ++;
1085 modest_text_utils_hyperlinkify_end (void)
1087 url_matches_block--;
1088 if (url_matches_block <= 0)
1094 get_url_matches (GString *txt, gint offset)
1097 guint rv, i, tmp_offset = 0;
1098 GSList *match_list = NULL;
1100 const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
1102 /* initalize the regexps */
1103 modest_text_utils_hyperlinkify_begin ();
1105 /* find all the matches */
1106 for (i = 0; i != pattern_num; ++i) {
1107 tmp_offset = offset;
1110 gboolean is_submatch;
1113 if ((rv = regexec (patterns[i].preg, txt->str + tmp_offset, 1, &rm, 0)) != 0) {
1114 g_return_val_if_fail (rv == REG_NOMATCH, NULL); /* this should not happen */
1115 break; /* try next regexp */
1120 is_submatch = FALSE;
1121 /* check old matches to see if this has already been matched */
1122 cursor = match_list;
1123 while (cursor && !is_submatch) {
1124 const url_match_t *old_match =
1125 (const url_match_t *) cursor->data;
1126 guint new_offset = tmp_offset + rm.rm_so;
1127 is_submatch = (new_offset > old_match->offset &&
1128 new_offset < old_match->offset + old_match->len);
1129 cursor = g_slist_next (cursor);
1133 /* make a list of our matches (<offset, len, prefix> tupels)*/
1134 match = g_slice_new (url_match_t);
1135 match->offset = tmp_offset + rm.rm_so;
1136 match->len = rm.rm_eo - rm.rm_so;
1137 match->prefix = patterns[i].prefix;
1138 match_list = g_slist_prepend (match_list, match);
1140 tmp_offset += rm.rm_eo;
1144 modest_text_utils_hyperlinkify_end ();
1146 /* now sort the list, so the matches are in reverse order of occurence.
1147 * that way, we can do the replacements starting from the end, so we don't need
1148 * to recalculate the offsets
1150 match_list = g_slist_sort (match_list,
1151 (GCompareFunc)cmp_offsets_reverse);
1157 /* replace all occurences of needle in haystack with repl*/
1159 replace_string (const gchar *haystack, const gchar *needle, gchar repl)
1161 gchar *str, *cursor;
1163 if (!haystack || !needle || strlen(needle) == 0)
1164 return haystack ? g_strdup(haystack) : NULL;
1166 str = g_strdup (haystack);
1168 for (cursor = str; cursor && *cursor; ++cursor) {
1169 if (g_str_has_prefix (cursor, needle)) {
1171 memmove (cursor + 1,
1172 cursor + strlen (needle),
1173 strlen (cursor + strlen (needle)) + 1);
1181 hyperlinkify_plain_text (GString *txt, gint offset)
1184 GSList *match_list = get_url_matches (txt, offset);
1186 /* we will work backwards, so the offsets stay valid */
1187 for (cursor = match_list; cursor; cursor = cursor->next) {
1189 url_match_t *match = (url_match_t*) cursor->data;
1190 gchar *url = g_strndup (txt->str + match->offset, match->len);
1191 gchar *repl = NULL; /* replacement */
1193 /* the string still contains $(MARK_AMP_URI_STR)"amp;" for each
1194 * '&' in the original, because of the text->html conversion.
1195 * in the href-URL (and only there), we must convert that back to
1198 gchar *href_url = replace_string (url, MARK_AMP_URI_STR "amp;", '&');
1200 /* the prefix is NULL: use the one that is already there */
1201 repl = g_strdup_printf ("<a href=\"%s%s\">%s</a>",
1202 match->prefix ? match->prefix : EMPTY_STRING,
1205 /* replace the old thing with our hyperlink
1206 * replacement thing */
1207 g_string_erase (txt, match->offset, match->len);
1208 g_string_insert (txt, match->offset, repl);
1214 g_slice_free (url_match_t, match);
1217 g_slist_free (match_list);
1221 modest_text_utils_hyperlinkify (GString *string_buffer)
1226 after_body = strstr (string_buffer->str, "<body>");
1227 if (after_body != NULL)
1228 offset = after_body - string_buffer->str;
1229 hyperlinkify_plain_text (string_buffer, offset);
1233 /* for optimization reasons, we change the string in-place */
1235 modest_text_utils_get_display_address (gchar *address)
1239 g_return_if_fail (address);
1244 /* should not be needed, and otherwise, we probably won't screw up the address
1245 * more than it already is :)
1246 * g_return_val_if_fail (g_utf8_validate (address, -1, NULL), NULL);
1249 /* remove leading whitespace */
1250 if (address[0] == ' ')
1251 g_strchug (address);
1253 for (i = 0; address[i]; ++i) {
1254 if (address[i] == '<') {
1255 if (G_UNLIKELY(i == 0)) {
1256 break; /* there's nothing else, leave it */
1258 address[i] = '\0'; /* terminate the string here */
1264 g_strchomp (address);
1272 modest_text_utils_get_email_address (const gchar *full_address)
1274 const gchar *left, *right;
1276 g_return_val_if_fail (full_address, NULL);
1281 g_return_val_if_fail (g_utf8_validate (full_address, -1, NULL), NULL);
1283 left = g_strrstr_len (full_address, strlen(full_address), "<");
1285 return g_strdup (full_address);
1287 right = g_strstr_len (left, strlen(left), ">");
1289 return g_strdup (full_address);
1291 return g_strndup (left + 1, right - left - 1);
1295 modest_text_utils_get_subject_prefix_len (const gchar *sub)
1297 gint prefix_len = 0;
1299 g_return_val_if_fail (sub, 0);
1304 /* optimization: "Re", "RE", "re","Fwd", "FWD", "fwd","FW","Fw", "fw" */
1305 if (sub[0] != 'R' && sub[0] != 'F' && sub[0] != 'r' && sub[0] != 'f')
1307 else if (sub[0] && sub[1] != 'e' && sub[1] != 'E' && sub[1] != 'w' && sub[1] != 'W')
1314 /* skip over a [...] block */
1315 if (sub[prefix_len] == '[') {
1316 int c = prefix_len + 1;
1317 while (sub[c] && sub[c] != ']')
1320 return 0; /* no end to the ']' found */
1325 /* did we find the ':' ? */
1326 if (sub[prefix_len] == ':') {
1328 if (sub[prefix_len] == ' ')
1330 prefix_len += modest_text_utils_get_subject_prefix_len (sub + prefix_len);
1331 /* g_warning ("['%s','%s']", sub, (char*) sub + prefix_len); */
1339 modest_text_utils_utf8_strcmp (const gchar* s1, const gchar *s2, gboolean insensitive)
1342 /* work even when s1 and/or s2 == NULL */
1343 if (G_UNLIKELY(s1 == s2))
1345 if (G_UNLIKELY(!s1))
1347 if (G_UNLIKELY(!s2))
1350 /* if it's not case sensitive */
1353 /* optimization: shortcut if first char is ascii */
1354 if (((s1[0] & 0x80)== 0) && ((s2[0] & 0x80) == 0) &&
1356 return s1[0] - s2[0];
1358 return g_utf8_collate (s1, s2);
1364 /* optimization: shortcut if first char is ascii */
1365 if (((s1[0] & 0x80) == 0) && ((s2[0] & 0x80) == 0) &&
1366 (tolower(s1[0]) != tolower (s2[0])))
1367 return tolower(s1[0]) - tolower(s2[0]);
1369 n1 = g_utf8_strdown (s1, -1);
1370 n2 = g_utf8_strdown (s2, -1);
1372 result = g_utf8_collate (n1, n2);
1383 modest_text_utils_get_display_date (time_t date)
1385 #define DATE_BUF_SIZE 64
1386 static gchar date_buf[DATE_BUF_SIZE];
1388 /* calculate the # of days since epoch for
1389 * for today and for the date provided
1390 * based on idea from pvanhoof */
1391 int day = time(NULL) / (24 * 60 * 60);
1392 int date_day = date / (24 * 60 * 60);
1394 /* if it's today, show the time, if it's not today, show the date instead */
1396 /* TODO: take into account the system config for 24/12h */
1397 if (day == date_day) /* is the date today? */
1398 modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, _HL("wdgt_va_24h_time"), date);
1400 modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, _HL("wdgt_va_date"), date);
1402 return date_buf; /* this is a static buffer, don't free! */
1408 modest_text_utils_validate_folder_name (const gchar *folder_name)
1410 /* based on http://msdn2.microsoft.com/en-us/library/aa365247.aspx,
1411 * with some extras */
1415 const gchar **cursor = NULL;
1416 const gchar *forbidden_names[] = { /* windows does not like these */
1417 "CON", "PRN", "AUX", "NUL", ".", "..", "cur", "tmp", "new",
1418 NULL /* cur, tmp, new are reserved for Maildir */
1421 /* cannot be NULL */
1425 /* cannot be empty */
1426 len = strlen(folder_name);
1430 /* cannot start with a dot, vfat does not seem to like that */
1431 if (folder_name[0] == '.')
1434 /* cannot start or end with a space */
1435 if (g_ascii_isspace(folder_name[0]) || g_ascii_isspace(folder_name[len - 1]))
1438 /* cannot contain a forbidden char */
1439 for (i = 0; i < len; i++)
1440 if (modest_text_utils_is_forbidden_char (folder_name[i], FOLDER_NAME_FORBIDDEN_CHARS))
1443 /* Cannot contain Windows port numbers. I'd like to use GRegex
1444 but it's still not available in Maemo. sergio */
1445 if (!g_ascii_strncasecmp (folder_name, "LPT", 3) ||
1446 !g_ascii_strncasecmp (folder_name, "COM", 3)) {
1450 /* We skip the first 3 characters for the
1452 val = strtol(folder_name+3, &endptr, 10);
1454 /* If the conversion to long succeeded then the string
1455 is not valid for us */
1456 if (*endptr == '\0')
1462 /* cannot contain a forbidden word */
1464 for (cursor = forbidden_names; cursor && *cursor; ++cursor) {
1465 if (g_ascii_strcasecmp (folder_name, *cursor) == 0)
1470 return TRUE; /* it's valid! */
1476 modest_text_utils_validate_domain_name (const gchar *domain)
1478 gboolean valid = FALSE;
1480 const gchar* domain_regex = "^([a-z0-9-]*[a-z0-9]\\.)+[a-z0-9-]*[a-z0-9]$";
1482 g_return_val_if_fail (domain, FALSE);
1487 memset (&rx, 0, sizeof(regex_t)); /* coverity wants this... */
1489 /* domain name: all alphanum or '-' or '.',
1490 * but beginning/ending in alphanum */
1491 if (regcomp (&rx, domain_regex, REG_ICASE|REG_EXTENDED|REG_NOSUB)) {
1492 g_warning ("BUG: error in regexp");
1496 valid = (regexec (&rx, domain, 1, NULL, 0) == 0);
1505 modest_text_utils_validate_email_address (const gchar *email_address,
1506 const gchar **invalid_char_position)
1509 const gchar *c = NULL, *domain = NULL;
1510 static gchar *rfc822_specials = "()<>@,;:\\\"[]&";
1512 if (invalid_char_position)
1513 *invalid_char_position = NULL;
1515 g_return_val_if_fail (email_address, FALSE);
1517 /* check that the email adress contains exactly one @ */
1518 if (!strstr(email_address, "@") ||
1519 (strstr(email_address, "@") != g_strrstr(email_address, "@")))
1522 /* first we validate the name portion (name@domain) */
1523 for (c = email_address; *c; c++) {
1525 (c == email_address ||
1527 *(c - 1) == '\"')) {
1531 if (*c == '\\' && (*++c == ' '))
1533 if (*c <= ' ' || *c >= 127)
1546 if (*c <= ' ' || *c >= 127)
1548 if (strchr(rfc822_specials, *c)) {
1549 if (invalid_char_position)
1550 *invalid_char_position = c;
1554 if (c == email_address || *(c - 1) == '.')
1557 /* next we validate the domain portion (name@domain) */
1558 if (!*(domain = ++c))
1562 if (c == domain || *(c - 1) == '.' || *(c + 1) == '\0')
1566 if (*c <= ' ' || *c >= 127)
1568 if (strchr(rfc822_specials, *c)) {
1569 if (invalid_char_position)
1570 *invalid_char_position = c;
1575 return (count >= 1) ? TRUE : FALSE;
1579 modest_text_utils_validate_recipient (const gchar *recipient, const gchar **invalid_char_position)
1581 gchar *stripped, *current;
1583 gboolean has_error = FALSE;
1585 if (invalid_char_position)
1586 *invalid_char_position = NULL;
1588 g_return_val_if_fail (recipient, FALSE);
1590 if (modest_text_utils_validate_email_address (recipient, invalid_char_position))
1593 stripped = g_strdup (recipient);
1594 stripped = g_strstrip (stripped);
1597 if (*current == '\0') {
1603 if (*current == '\"') {
1604 current = g_utf8_next_char (current);
1606 for (; *current != '\0'; current = g_utf8_next_char (current)) {
1607 if (*current == '\\') {
1608 /* TODO: This causes a warning, which breaks the build,
1609 * because a gchar cannot be < 0.
1611 if (current[1] <0) {
1616 } else if (*current == '\"') {
1618 current = g_utf8_next_char (current);
1624 for (current = stripped ; *current != '\0'; current = g_utf8_next_char (current)) {
1625 if (*current == '<') {
1637 right_part = g_strdup (current);
1639 right_part = g_strstrip (right_part);
1641 if (g_str_has_prefix (right_part, "<") &&
1642 g_str_has_suffix (right_part, ">")) {
1646 address = g_strndup (right_part+1, strlen (right_part) - 2);
1647 g_free (right_part);
1648 valid = modest_text_utils_validate_email_address (address, invalid_char_position);
1652 g_free (right_part);
1659 modest_text_utils_get_display_size (guint64 size)
1661 const guint KB=1024;
1662 const guint MB=1024 * KB;
1663 const guint GB=1024 * MB;
1666 return g_strdup_printf (_FM("sfil_li_size_kb"), (int) 0);
1667 if (0 <= size && size < KB)
1668 return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), (int) 1);
1669 else if (KB <= size && size < 100 * KB)
1670 return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), (int) size / KB);
1671 else if (100*KB <= size && size < MB)
1672 return g_strdup_printf (_FM("sfil_li_size_100kb_1mb"), (float) size / MB);
1673 else if (MB <= size && size < 10*MB)
1674 return g_strdup_printf (_FM("sfil_li_size_1mb_10mb"), (float) size / MB);
1675 else if (10*MB <= size && size < GB)
1676 return g_strdup_printf (_FM("sfil_li_size_10mb_1gb"), (int) size / MB);
1678 return g_strdup_printf (_FM("sfil_li_size_1gb_or_greater"), (float) size / GB);
1682 get_email_from_address (const gchar * address)
1684 gchar *left_limit, *right_limit;
1686 left_limit = strstr (address, "<");
1687 right_limit = g_strrstr (address, ">");
1689 if ((left_limit == NULL)||(right_limit == NULL)|| (left_limit > right_limit))
1690 return g_strdup (address);
1692 return g_strndup (left_limit + 1, (right_limit - left_limit) - 1);
1696 modest_text_utils_get_color_string (GdkColor *color)
1698 g_return_val_if_fail (color, NULL);
1700 return g_strdup_printf ("#%x%x%x%x%x%x%x%x%x%x%x%x",
1701 (color->red >> 12) & 0xf, (color->red >> 8) & 0xf,
1702 (color->red >> 4) & 0xf, (color->red) & 0xf,
1703 (color->green >> 12) & 0xf, (color->green >> 8) & 0xf,
1704 (color->green >> 4) & 0xf, (color->green) & 0xf,
1705 (color->blue >> 12) & 0xf, (color->blue >> 8) & 0xf,
1706 (color->blue >> 4) & 0xf, (color->blue) & 0xf);
1710 modest_text_utils_text_buffer_get_text (GtkTextBuffer *buffer)
1712 GtkTextIter start, end;
1713 gchar *slice, *current;
1714 GString *result = g_string_new ("");
1716 g_return_val_if_fail (buffer && GTK_IS_TEXT_BUFFER (buffer), NULL);
1718 gtk_text_buffer_get_start_iter (buffer, &start);
1719 gtk_text_buffer_get_end_iter (buffer, &end);
1721 slice = gtk_text_buffer_get_slice (buffer, &start, &end, FALSE);
1724 while (current && current != '\0') {
1725 if (g_utf8_get_char (current) == 0xFFFC) {
1726 result = g_string_append_c (result, ' ');
1727 current = g_utf8_next_char (current);
1729 gchar *next = g_utf8_strchr (current, -1, 0xFFFC);
1731 result = g_string_append (result, current);
1733 result = g_string_append_len (result, current, next - current);
1740 return g_string_free (result, FALSE);
1745 modest_text_utils_is_forbidden_char (const gchar character,
1746 ModestTextUtilsForbiddenCharType type)
1749 const gchar *forbidden_chars = NULL;
1751 /* We need to get the length in the switch because the
1752 compiler needs to know the size at compile time */
1754 case ACCOUNT_TITLE_FORBIDDEN_CHARS:
1755 forbidden_chars = account_title_forbidden_chars;
1756 len = G_N_ELEMENTS (account_title_forbidden_chars);
1758 case FOLDER_NAME_FORBIDDEN_CHARS:
1759 forbidden_chars = folder_name_forbidden_chars;
1760 len = G_N_ELEMENTS (folder_name_forbidden_chars);
1762 case USER_NAME_FORBIDDEN_NAMES:
1763 forbidden_chars = user_name_forbidden_chars;
1764 len = G_N_ELEMENTS (user_name_forbidden_chars);
1767 g_return_val_if_reached (TRUE);
1770 for (i = 0; i < len ; i++)
1771 if (forbidden_chars[i] == character)
1774 return FALSE; /* it's valid! */
1778 modest_text_utils_label_get_selection (GtkLabel *label)
1783 if (gtk_label_get_selection_bounds (GTK_LABEL (label), &start, &end)) {
1784 const gchar *start_offset;
1785 const gchar *end_offset;
1786 start_offset = gtk_label_get_text (GTK_LABEL (label));
1787 start_offset = g_utf8_offset_to_pointer (start_offset, start);
1788 end_offset = gtk_label_get_text (GTK_LABEL (label));
1789 end_offset = g_utf8_offset_to_pointer (end_offset, end);
1790 selection = g_strndup (start_offset, end_offset - start_offset);
1793 return g_strdup ("");
1798 _forward_search_image_char (gunichar ch,
1801 return (ch == 0xFFFC);
1805 modest_text_utils_buffer_selection_is_valid (GtkTextBuffer *buffer)
1808 GtkTextIter start, end;
1810 g_return_val_if_fail (GTK_IS_TEXT_BUFFER (buffer), FALSE);
1812 result = gtk_text_buffer_get_has_selection (GTK_TEXT_BUFFER (buffer));
1814 /* check there are no images in selection */
1816 gtk_text_buffer_get_selection_bounds (buffer, &start, &end);
1817 if (gtk_text_iter_get_char (&start)== 0xFFFC)
1820 gtk_text_iter_backward_char (&end);
1821 if (gtk_text_iter_forward_find_char (&start, _forward_search_image_char,
1832 modest_text_utils_escape_mnemonics (const gchar *text)
1835 GString *result = NULL;
1840 result = g_string_new ("");
1841 for (p = text; *p != '\0'; p++) {
1843 result = g_string_append (result, "__");
1845 result = g_string_append_c (result, *p);
1848 return g_string_free (result, FALSE);