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 $(MARK_AMP_URI_STR)"amp;" */
111 #define MAIL_VIEWER_URL_MATCH_PATTERNS { \
112 { "(file|rtsp|http|ftp|https|mms|mmsh|rtsp|rdp|lastfm)://[-a-z0-9_$.+!*(),;:@%=?/~#" MARK_AMP_URI_STR \
113 "]+[-a-z0-9_$%" MARK_AMP_URI_STR "=?/~#]", \
115 { "www\\.[-a-z0-9_$.+!*(),;:@%=?/~#" MARK_AMP_URI_STR "]+[-a-z0-9_$%" MARK_AMP_URI_STR "=?/~#]",\
117 { "ftp\\.[-a-z0-9_$.+!*(),;:@%=?/~#" MARK_AMP_URI_STR "]+[-a-z0-9_$%" MARK_AMP_URI_STR "=?/~#]",\
119 { "(jabberto|voipto|sipto|sip|chatto|xmpp):[-_a-z@0-9.+]+", \
121 { "mailto:[-_a-z0-9.\\+]+@[-_a-z0-9.]+", \
123 { "[-_a-z0-9.\\+]+@[-_a-z0-9.]+",\
127 const gchar account_title_forbidden_chars[] = {
128 '\\', '/', ':', '*', '?', '\'', '<', '>', '|', '^'
130 const gchar folder_name_forbidden_chars[] = {
131 '<', '>', ':', '\'', '/', '\\', '|', '?', '*', '^', '%', '$', '#', '&'
133 const gchar user_name_forbidden_chars[] = {
136 const guint ACCOUNT_TITLE_FORBIDDEN_CHARS_LENGTH = G_N_ELEMENTS (account_title_forbidden_chars);
137 const guint FOLDER_NAME_FORBIDDEN_CHARS_LENGTH = G_N_ELEMENTS (folder_name_forbidden_chars);
138 const guint USER_NAME_FORBIDDEN_CHARS_LENGTH = G_N_ELEMENTS (user_name_forbidden_chars);
141 static gchar* cite (const time_t sent_date, const gchar *from);
142 static void hyperlinkify_plain_text (GString *txt, 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);
149 static void append_quoted (GString * buf, const int indent, const GString * str,
151 static int get_breakpoint_utf8 (const gchar * s, const gint indent, const gint limit);
152 static int get_breakpoint_ascii (const gchar * s, const gint indent, const gint limit);
153 static int get_breakpoint (const gchar * s, const gint indent, const gint limit);
155 static gchar* modest_text_utils_quote_plain_text (const gchar *text,
157 const gchar *signature,
161 static gchar* modest_text_utils_quote_html (const gchar *text,
163 const gchar *signature,
166 static gchar* get_email_from_address (const gchar *address);
169 /* ******************************************************************* */
170 /* ************************* PUBLIC FUNCTIONS ************************ */
171 /* ******************************************************************* */
174 modest_text_utils_quote (const gchar *text,
175 const gchar *content_type,
176 const gchar *signature,
178 const time_t sent_date,
182 gchar *retval, *cited;
184 g_return_val_if_fail (text, NULL);
185 g_return_val_if_fail (content_type, NULL);
187 cited = cite (sent_date, from);
189 if (content_type && strcmp (content_type, "text/html") == 0)
190 /* TODO: extract the <body> of the HTML and pass it to
192 retval = modest_text_utils_quote_html (text, cited, signature, attachments, limit);
194 retval = modest_text_utils_quote_plain_text (text, cited, signature, attachments, limit);
203 modest_text_utils_cite (const gchar *text,
204 const gchar *content_type,
205 const gchar *signature,
212 g_return_val_if_fail (text, NULL);
213 g_return_val_if_fail (content_type, NULL);
216 retval = g_strdup ("");
217 else if (strcmp(content_type, "text/html") == 0) {
218 tmp_sig = g_strconcat ("\n", SIGNATURE_MARKER,"\n", signature, NULL);
219 retval = modest_text_utils_convert_to_html_body(tmp_sig, -1, TRUE);
222 retval = g_strconcat (text, "\n", SIGNATURE_MARKER, "\n", signature, NULL);
229 forward_cite (const gchar *from,
232 const gchar *subject)
234 g_return_val_if_fail (sent, NULL);
236 return g_strdup_printf ("%s\n%s %s\n%s %s\n%s %s\n%s %s\n",
238 FROM_STRING, (from)?from:"",
240 TO_STRING, (to)?to:"",
241 SUBJECT_STRING, (subject)?subject:"");
245 modest_text_utils_inline (const gchar *text,
246 const gchar *content_type,
247 const gchar *signature,
251 const gchar *subject)
257 g_return_val_if_fail (text, NULL);
258 g_return_val_if_fail (content_type, NULL);
260 modest_text_utils_strftime (sent_str, 100, "%c", sent_date);
262 cited = forward_cite (from, sent_str, to, subject);
264 if (content_type && strcmp (content_type, "text/html") == 0)
265 retval = modest_text_utils_quote_html (text, cited, signature, NULL, 80);
267 retval = modest_text_utils_quote_plain_text (text, cited, signature, NULL, 80);
273 /* just to prevent warnings:
274 * warning: `%x' yields only last 2 digits of year in some locales
277 modest_text_utils_strftime(char *s, gsize max, const char *fmt, time_t timet)
281 /* To prevent possible problems in strftime that could leave
282 garbage in the s variable */
286 /* does not work on old maemo glib:
287 * g_date_set_time_t (&date, timet);
289 localtime_r (&timet, &tm);
290 return strftime(s, max, fmt, &tm);
294 modest_text_utils_derived_subject (const gchar *subject, const gchar *prefix)
296 gchar *tmp, *subject_dup, *retval;
299 g_return_val_if_fail (prefix, NULL);
301 if (!subject || subject[0] == '\0')
302 subject = _("mail_va_no_subject");
304 subject_dup = g_strdup (subject);
305 tmp = g_strchug (subject_dup);
307 /* We do not want things like "Re: Re: Re:" or "Fw: Fw:" so
308 delete the previous ones */
309 prefix_len = strlen (prefix);
311 if (g_str_has_prefix (tmp, prefix)) {
313 tmp = g_strchug (tmp);
319 retval = g_strdup_printf ("%s %s", prefix, tmp);
320 g_free (subject_dup);
326 modest_text_utils_remove_address (const gchar *address_list, const gchar *address)
328 gchar *dup, *token, *ptr = NULL, *result;
329 GString *filtered_emails;
330 gchar *email_address;
332 g_return_val_if_fail (address_list, NULL);
335 return g_strdup (address_list);
337 email_address = get_email_from_address (address);
339 /* search for substring */
340 if (!strstr ((const char *) address_list, (const char *) email_address)) {
341 g_free (email_address);
342 return g_strdup (address_list);
345 dup = g_strdup (address_list);
346 filtered_emails = g_string_new (NULL);
348 token = strtok_r (dup, ",", &ptr);
350 while (token != NULL) {
351 /* Add to list if not found */
352 if (!strstr ((const char *) token, (const char *) email_address)) {
353 if (filtered_emails->len == 0)
354 g_string_append_printf (filtered_emails, "%s", g_strstrip (token));
356 g_string_append_printf (filtered_emails, ",%s", g_strstrip (token));
358 token = strtok_r (NULL, ",", &ptr);
360 result = filtered_emails->str;
363 g_free (email_address);
365 g_string_free (filtered_emails, FALSE);
372 modest_text_utils_remove_duplicate_addresses (const gchar *address_list)
374 GSList *addresses, *cursor;
376 gchar *new_list = NULL;
378 g_return_val_if_fail (address_list, NULL);
380 table = g_hash_table_new_full (g_str_hash, g_str_equal, g_free, NULL);
381 addresses = modest_text_utils_split_addresses_list (address_list);
385 const gchar* address = (const gchar*)cursor->data;
387 /* We need only the email to just compare it and not
388 the full address which would make "a <a@a.com>"
389 different from "a@a.com" */
390 const gchar *email = get_email_from_address (address);
392 /* ignore the address if already seen */
393 if (g_hash_table_lookup (table, email) == 0) {
396 /* Include the full address and not only the
397 email in the returned list */
399 tmp = g_strdup (address);
401 tmp = g_strjoin (",", new_list, address, NULL);
406 g_hash_table_insert (table, (gchar*)email, GINT_TO_POINTER(1));
408 cursor = g_slist_next (cursor);
411 g_hash_table_unref (table);
412 g_slist_foreach (addresses, (GFunc)g_free, NULL);
413 g_slist_free (addresses);
420 modest_text_utils_convert_buffer_to_html_start (GString *html, const gchar *data, gssize n)
423 gboolean space_seen = FALSE;
424 guint break_dist = 0; /* distance since last break point */
429 /* replace with special html chars where needed*/
430 for (i = 0; i != n; ++i) {
431 guchar kar = data[i];
433 if (space_seen && kar != ' ') {
434 g_string_append_c (html, ' ');
438 /* we artificially insert a breakpoint (newline)
439 * after 256, to make sure our lines are not so long
440 * they will DOS the regexping later
441 * Also, check that kar is ASCII to make sure that we
442 * don't break a UTF8 char in two
444 if (++break_dist >= 256 && kar < 127) {
445 g_string_append_c (html, '\n');
453 /* this is a temp place holder for '&'; we can only
454 * set the real '&' after hyperlink translation, otherwise
455 * we might screw that up */
456 break; /* ignore embedded \0s and MARK_AMP */
457 case '<' : g_string_append (html, MARK_AMP_STR "lt;"); break;
458 case '>' : g_string_append (html, MARK_AMP_STR "gt;"); break;
459 case '&' : g_string_append (html, MARK_AMP_URI_STR "amp;"); break; /* special case */
460 case '"' : g_string_append (html, MARK_AMP_STR "quot;"); break;
462 /* don't convert ' --> wpeditor will try to re-convert it... */
463 //case '\'' : g_string_append (html, "'"); break;
464 case '\n' : g_string_append (html, "<br>\n");break_dist= 0; break;
465 case '\t' : g_string_append (html, MARK_AMP_STR "nbsp;" MARK_AMP_STR "nbsp;" MARK_AMP_STR "nbsp; ");
466 break_dist=0; break; /* note the space at the end*/
469 if (space_seen) { /* second space in a row */
470 g_string_append (html, " ");
476 g_string_append_c (html, kar);
483 modest_text_utils_convert_buffer_to_html_finish (GString *html)
486 /* replace all our MARK_AMPs with real ones */
487 for (i = 0; i != html->len; ++i)
488 if ((html->str)[i] == MARK_AMP || (html->str)[i] == MARK_AMP_URI)
489 (html->str)[i] = '&';
494 modest_text_utils_convert_to_html (const gchar *data)
499 g_return_val_if_fail (data, NULL);
505 html = g_string_sized_new (1.5 * len); /* just a guess... */
507 g_string_append_printf (html,
509 "<meta http-equiv=\"content-type\" content=\"text/html; charset=utf8\">"
513 modest_text_utils_convert_buffer_to_html_start (html, data, -1);
515 g_string_append (html, "</body></html>");
517 if (len <= HYPERLINKIFY_MAX_LENGTH)
518 hyperlinkify_plain_text (html, 0);
520 modest_text_utils_convert_buffer_to_html_finish (html);
522 return g_string_free (html, FALSE);
526 modest_text_utils_convert_to_html_body (const gchar *data, gssize n, gboolean hyperlinkify)
530 g_return_val_if_fail (data, NULL);
537 html = g_string_sized_new (1.5 * n); /* just a guess... */
539 modest_text_utils_convert_buffer_to_html_start (html, data, n);
541 if (hyperlinkify && (n < HYPERLINKIFY_MAX_LENGTH))
542 hyperlinkify_plain_text (html, 0);
544 modest_text_utils_convert_buffer_to_html_finish (html);
546 return g_string_free (html, FALSE);
550 modest_text_utils_get_addresses_indexes (const gchar *addresses, GSList **start_indexes, GSList **end_indexes)
552 gchar *current, *start, *last_blank;
553 gint start_offset = 0, current_offset = 0;
555 g_return_if_fail (start_indexes != NULL);
556 g_return_if_fail (end_indexes != NULL);
558 start = (gchar *) addresses;
562 while (*current != '\0') {
563 if ((start == current)&&((*current == ' ')||(*current == ',')||(*current == ';'))) {
564 start = g_utf8_next_char (start);
566 last_blank = current;
567 } else if ((*current == ',')||(*current == ';')) {
568 gint *start_index, *end_index;
569 start_index = g_new0(gint, 1);
570 end_index = g_new0(gint, 1);
571 *start_index = start_offset;
572 *end_index = current_offset;
573 *start_indexes = g_slist_prepend (*start_indexes, start_index);
574 *end_indexes = g_slist_prepend (*end_indexes, end_index);
575 start = g_utf8_next_char (current);
576 start_offset = current_offset + 1;
578 } else if (*current == '"') {
579 current = g_utf8_next_char (current);
581 while ((*current != '"')&&(*current != '\0')) {
582 current = g_utf8_next_char (current);
587 current = g_utf8_next_char (current);
591 if (start != current) {
592 gint *start_index, *end_index;
593 start_index = g_new0(gint, 1);
594 end_index = g_new0(gint, 1);
595 *start_index = start_offset;
596 *end_index = current_offset;
597 *start_indexes = g_slist_prepend (*start_indexes, start_index);
598 *end_indexes = g_slist_prepend (*end_indexes, end_index);
601 *start_indexes = g_slist_reverse (*start_indexes);
602 *end_indexes = g_slist_reverse (*end_indexes);
609 modest_text_utils_split_addresses_list (const gchar *addresses)
612 const gchar *my_addrs = addresses;
616 g_return_val_if_fail (addresses, NULL);
618 /* skip any space, ',', ';' at the start */
619 while (my_addrs && (my_addrs[0] == ' ' || my_addrs[0] == ',' || my_addrs[0] == ';'))
622 /* are we at the end of addresses list? */
626 /* nope, we are at the start of some address
627 * now, let's find the end of the address */
629 while (end[0] && end[0] != ';') {
630 if (end[0] == '\"') {
631 while (end[0] && end[0] != '\"')
634 if ((end[0] && end[0] == '>')&&(end[1] && end[1] == ',')) {
641 /* we got the address; copy it and remove trailing whitespace */
642 addr = g_strndup (my_addrs, end - my_addrs);
645 head = g_slist_append (NULL, addr);
646 head->next = modest_text_utils_split_addresses_list (end); /* recurse */
653 modest_text_utils_address_range_at_position (const gchar *recipients_list,
658 gchar *current = NULL;
659 gint range_start = 0;
662 gboolean is_quoted = FALSE;
664 g_return_if_fail (recipients_list);
665 g_return_if_fail (position < g_utf8_strlen(recipients_list, -1));
668 for (current = (gchar *) recipients_list; *current != '\0';
669 current = g_utf8_find_next_char (current, NULL)) {
670 gunichar c = g_utf8_get_char (current);
672 if ((c == ',') && (!is_quoted)) {
673 if (index < position) {
674 range_start = index + 1;
678 } else if (c == '\"') {
679 is_quoted = !is_quoted;
680 } else if ((c == ' ') &&(range_start == index)) {
688 *start = range_start;
694 modest_text_utils_address_with_standard_length (const gchar *recipients_list)
698 GString *buffer = g_string_new ("");
700 splitted = g_strsplit (recipients_list, "\n", 0);
704 if (current != splitted)
705 buffer = g_string_append_c (buffer, '\n');
706 line = g_strndup (*splitted, 1000);
707 buffer = g_string_append (buffer, line);
712 g_strfreev (splitted);
714 return g_string_free (buffer, FALSE);
718 /* ******************************************************************* */
719 /* ************************* UTILIY FUNCTIONS ************************ */
720 /* ******************************************************************* */
723 get_next_line (const gchar * b, const gsize blen, const gchar * iter)
729 return g_string_new("");
737 gs = g_string_new_len (i0, iter - i0);
741 get_indent_level (const char *l)
758 /* if we hit the signature marker "-- ", we return -(indent + 1). This
759 * stops reformatting.
761 if (strcmp (l, "-- ") == 0) {
769 unquote_line (GString * l)
784 g_string_erase (l, 0, p - l->str);
788 append_quoted (GString * buf, int indent, const GString * str,
793 indent = indent < 0 ? abs (indent) - 1 : indent;
794 for (i = 0; i <= indent; i++) {
795 g_string_append (buf, "> ");
798 g_string_append_len (buf, str->str, cutpoint);
800 g_string_append (buf, str->str);
802 g_string_append (buf, "\n");
806 get_breakpoint_utf8 (const gchar * s, gint indent, const gint limit)
809 const gchar *pos, *last;
812 indent = indent < 0 ? abs (indent) - 1 : indent;
816 uni = g_utf8_to_ucs4_fast (s, -1, NULL);
818 if ((index + 2 * indent > limit) && last) {
822 if (g_unichar_isspace (uni[index])) {
825 pos = g_utf8_next_char (pos);
833 get_breakpoint_ascii (const gchar * s, const gint indent, const gint limit)
838 if (last + 2 * indent < limit)
841 for (i = strlen (s); i > 0; i--) {
843 if (i + 2 * indent <= limit) {
854 get_breakpoint (const gchar * s, const gint indent, const gint limit)
857 if (g_utf8_validate (s, -1, NULL)) {
858 return get_breakpoint_utf8 (s, indent, limit);
859 } else { /* assume ASCII */
860 //g_warning("invalid UTF-8 in msg");
861 return get_breakpoint_ascii (s, indent, limit);
866 cite (const time_t sent_date, const gchar *from)
868 return g_strdup (_("mcen_ia_editor_original_message"));
872 quoted_attachments (GList *attachments)
875 GString *result = g_string_new ("");
876 for (node = attachments; node != NULL; node = g_list_next (node)) {
877 gchar *filename = (gchar *) node->data;
878 g_string_append_printf ( result, "%s %s\n", _("mcen_ia_editor_attach_filename"), filename);
881 return g_string_free (result, FALSE);
886 modest_text_utils_quote_plain_text (const gchar *text,
888 const gchar *signature,
893 gint indent, breakpoint, rem_indent = 0;
894 GString *q, *l, *remaining;
896 gchar *attachments_string = NULL;
898 q = g_string_new ("");
900 if (signature != NULL) {
901 q = g_string_append (q, "\n--\n");
902 q = g_string_append (q, signature);
905 q = g_string_append (q, "\n");
906 q = g_string_append (q, cite);
907 q = g_string_append_c (q, '\n');
909 /* remaining will store the rest of the line if we have to break it */
910 remaining = g_string_new ("");
915 l = get_next_line (text, len, iter);
916 iter = iter + l->len + 1;
917 indent = get_indent_level (l->str);
920 if (remaining->len) {
921 if (l->len && indent == rem_indent) {
922 g_string_prepend (l, " ");
923 g_string_prepend (l, remaining->str);
927 get_breakpoint (remaining->str,
930 append_quoted (q, rem_indent,
931 remaining, breakpoint);
932 g_string_erase (remaining, 0,
934 if (remaining->str[0] == ' ') {
935 g_string_erase (remaining, 0,
938 } while (remaining->len);
941 g_string_free (remaining, TRUE);
942 breakpoint = get_breakpoint (l->str, indent, limit);
943 remaining = g_string_new (l->str + breakpoint);
944 if (remaining->str[0] == ' ') {
945 g_string_erase (remaining, 0, 1);
948 append_quoted (q, indent, l, breakpoint);
949 g_string_free (l, TRUE);
950 } while ((iter < text + len) || (remaining->str[0]));
952 attachments_string = quoted_attachments (attachments);
953 q = g_string_append (q, attachments_string);
954 g_free (attachments_string);
956 return g_string_free (q, FALSE);
960 quote_html_add_to_gstring (GString *string,
963 if (text && strcmp (text, "")) {
964 gchar *html_text = modest_text_utils_convert_to_html_body (text, -1, TRUE);
965 g_string_append_printf (string, "%s<br/>", html_text);
971 modest_text_utils_quote_html (const gchar *text,
973 const gchar *signature,
977 GString *result_string;
981 "<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\">\n" \
985 if (text || cite || signature) {
986 g_string_append (result_string, "<pre>\n");
988 quote_html_add_to_gstring (result_string, SIGNATURE_MARKER);
989 quote_html_add_to_gstring (result_string, signature);
991 quote_html_add_to_gstring (result_string, cite);
992 quote_html_add_to_gstring (result_string, text);
994 gchar *attachments_string = quoted_attachments (attachments);
995 quote_html_add_to_gstring (result_string, attachments_string);
996 g_free (attachments_string);
998 g_string_append (result_string, "</pre>");
1000 g_string_append (result_string, "</body>");
1001 g_string_append (result_string, "</html>");
1003 return g_string_free (result_string, FALSE);
1007 cmp_offsets_reverse (const url_match_t *match1, const url_match_t *match2)
1009 return match2->offset - match1->offset;
1012 static gint url_matches_block = 0;
1013 static url_match_pattern_t patterns[] = MAIL_VIEWER_URL_MATCH_PATTERNS;
1020 const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
1021 for (i = 0; i != pattern_num; ++i) {
1022 patterns[i].preg = g_slice_new0 (regex_t);
1024 /* this should not happen */
1025 if (regcomp (patterns[i].preg, patterns[i].regex,
1026 REG_ICASE|REG_EXTENDED|REG_NEWLINE) != 0) {
1027 g_warning ("%s: error in regexp:\n%s\n", __FUNCTION__, patterns[i].regex);
1038 const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
1039 for (i = 0; i != pattern_num; ++i) {
1040 regfree (patterns[i].preg);
1041 g_slice_free (regex_t, patterns[i].preg);
1042 } /* don't free patterns itself -- it's static */
1046 modest_text_utils_hyperlinkify_begin (void)
1048 if (url_matches_block == 0)
1049 compile_patterns ();
1050 url_matches_block ++;
1054 modest_text_utils_hyperlinkify_end (void)
1056 url_matches_block--;
1057 if (url_matches_block <= 0)
1063 get_url_matches (GString *txt, gint offset)
1066 guint rv, i, tmp_offset = 0;
1067 GSList *match_list = NULL;
1069 const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
1071 /* initalize the regexps */
1072 modest_text_utils_hyperlinkify_begin ();
1074 /* find all the matches */
1075 for (i = 0; i != pattern_num; ++i) {
1076 tmp_offset = offset;
1079 gboolean is_submatch;
1082 if ((rv = regexec (patterns[i].preg, txt->str + tmp_offset, 1, &rm, 0)) != 0) {
1083 g_return_val_if_fail (rv == REG_NOMATCH, NULL); /* this should not happen */
1084 break; /* try next regexp */
1089 is_submatch = FALSE;
1090 /* check old matches to see if this has already been matched */
1091 cursor = match_list;
1092 while (cursor && !is_submatch) {
1093 const url_match_t *old_match =
1094 (const url_match_t *) cursor->data;
1095 guint new_offset = tmp_offset + rm.rm_so;
1096 is_submatch = (new_offset > old_match->offset &&
1097 new_offset < old_match->offset + old_match->len);
1098 cursor = g_slist_next (cursor);
1102 /* make a list of our matches (<offset, len, prefix> tupels)*/
1103 match = g_slice_new (url_match_t);
1104 match->offset = tmp_offset + rm.rm_so;
1105 match->len = rm.rm_eo - rm.rm_so;
1106 match->prefix = patterns[i].prefix;
1107 match_list = g_slist_prepend (match_list, match);
1109 tmp_offset += rm.rm_eo;
1113 modest_text_utils_hyperlinkify_end ();
1115 /* now sort the list, so the matches are in reverse order of occurence.
1116 * that way, we can do the replacements starting from the end, so we don't need
1117 * to recalculate the offsets
1119 match_list = g_slist_sort (match_list,
1120 (GCompareFunc)cmp_offsets_reverse);
1126 /* replace all occurences of needle in haystack with repl*/
1128 replace_string (const gchar *haystack, const gchar *needle, gchar repl)
1130 gchar *str, *cursor;
1132 if (!haystack || !needle || strlen(needle) == 0)
1133 return haystack ? g_strdup(haystack) : NULL;
1135 str = g_strdup (haystack);
1137 for (cursor = str; cursor && *cursor; ++cursor) {
1138 if (g_str_has_prefix (cursor, needle)) {
1140 memmove (cursor + 1,
1141 cursor + strlen (needle),
1142 strlen (cursor + strlen (needle)) + 1);
1150 hyperlinkify_plain_text (GString *txt, gint offset)
1153 GSList *match_list = get_url_matches (txt, offset);
1155 /* we will work backwards, so the offsets stay valid */
1156 for (cursor = match_list; cursor; cursor = cursor->next) {
1158 url_match_t *match = (url_match_t*) cursor->data;
1159 gchar *url = g_strndup (txt->str + match->offset, match->len);
1160 gchar *repl = NULL; /* replacement */
1162 /* the string still contains $(MARK_AMP_URI_STR)"amp;" for each
1163 * '&' in the original, because of the text->html conversion.
1164 * in the href-URL (and only there), we must convert that back to
1167 gchar *href_url = replace_string (url, MARK_AMP_URI_STR "amp;", '&');
1169 /* the prefix is NULL: use the one that is already there */
1170 repl = g_strdup_printf ("<a href=\"%s%s\">%s</a>",
1171 match->prefix ? match->prefix : EMPTY_STRING,
1174 /* replace the old thing with our hyperlink
1175 * replacement thing */
1176 g_string_erase (txt, match->offset, match->len);
1177 g_string_insert (txt, match->offset, repl);
1183 g_slice_free (url_match_t, match);
1186 g_slist_free (match_list);
1190 modest_text_utils_hyperlinkify (GString *string_buffer)
1195 after_body = strstr (string_buffer->str, "<body>");
1196 if (after_body != NULL)
1197 offset = after_body - string_buffer->str;
1198 hyperlinkify_plain_text (string_buffer, offset);
1202 /* for optimization reasons, we change the string in-place */
1204 modest_text_utils_get_display_address (gchar *address)
1208 g_return_if_fail (address);
1213 /* should not be needed, and otherwise, we probably won't screw up the address
1214 * more than it already is :)
1215 * g_return_val_if_fail (g_utf8_validate (address, -1, NULL), NULL);
1218 /* remove leading whitespace */
1219 if (address[0] == ' ')
1220 g_strchug (address);
1222 for (i = 0; address[i]; ++i) {
1223 if (address[i] == '<') {
1224 if (G_UNLIKELY(i == 0))
1225 return; /* there's nothing else, leave it */
1227 address[i] = '\0'; /* terminate the string here */
1239 modest_text_utils_get_email_address (const gchar *full_address)
1241 const gchar *left, *right;
1243 g_return_val_if_fail (full_address, NULL);
1248 g_return_val_if_fail (g_utf8_validate (full_address, -1, NULL), NULL);
1250 left = g_strrstr_len (full_address, strlen(full_address), "<");
1252 return g_strdup (full_address);
1254 right = g_strstr_len (left, strlen(left), ">");
1256 return g_strdup (full_address);
1258 return g_strndup (left + 1, right - left - 1);
1262 modest_text_utils_get_subject_prefix_len (const gchar *sub)
1264 gint prefix_len = 0;
1266 g_return_val_if_fail (sub, 0);
1271 /* optimization: "Re", "RE", "re","Fwd", "FWD", "fwd","FW","Fw", "fw" */
1272 if (sub[0] != 'R' && sub[0] != 'F' && sub[0] != 'r' && sub[0] != 'f')
1274 else if (sub[0] && sub[1] != 'e' && sub[1] != 'E' && sub[1] != 'w' && sub[1] != 'W')
1281 /* skip over a [...] block */
1282 if (sub[prefix_len] == '[') {
1283 int c = prefix_len + 1;
1284 while (sub[c] && sub[c] != ']')
1287 return 0; /* no end to the ']' found */
1292 /* did we find the ':' ? */
1293 if (sub[prefix_len] == ':') {
1295 if (sub[prefix_len] == ' ')
1297 prefix_len += modest_text_utils_get_subject_prefix_len (sub + prefix_len);
1298 /* g_warning ("['%s','%s']", sub, (char*) sub + prefix_len); */
1306 modest_text_utils_utf8_strcmp (const gchar* s1, const gchar *s2, gboolean insensitive)
1309 /* work even when s1 and/or s2 == NULL */
1310 if (G_UNLIKELY(s1 == s2))
1312 if (G_UNLIKELY(!s1))
1314 if (G_UNLIKELY(!s2))
1317 /* if it's not case sensitive */
1320 /* optimization: shortcut if first char is ascii */
1321 if (((s1[0] & 0x80)== 0) && ((s2[0] & 0x80) == 0) &&
1323 return s1[0] - s2[0];
1325 return g_utf8_collate (s1, s2);
1331 /* optimization: shortcut if first char is ascii */
1332 if (((s1[0] & 0x80) == 0) && ((s2[0] & 0x80) == 0) &&
1333 (tolower(s1[0]) != tolower (s2[0])))
1334 return tolower(s1[0]) - tolower(s2[0]);
1336 n1 = g_utf8_strdown (s1, -1);
1337 n2 = g_utf8_strdown (s2, -1);
1339 result = g_utf8_collate (n1, n2);
1350 modest_text_utils_get_display_date (time_t date)
1352 #define DATE_BUF_SIZE 64
1353 static gchar date_buf[DATE_BUF_SIZE];
1355 /* calculate the # of days since epoch for
1356 * for today and for the date provided
1357 * based on idea from pvanhoof */
1358 int day = time(NULL) / (24 * 60 * 60);
1359 int date_day = date / (24 * 60 * 60);
1361 /* if it's today, show the time, if it's not today, show the date instead */
1363 /* TODO: take into account the system config for 24/12h */
1364 if (day == date_day) /* is the date today? */
1365 modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, _HL("wdgt_va_24h_time"), date);
1367 modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, _HL("wdgt_va_date"), date);
1369 return date_buf; /* this is a static buffer, don't free! */
1375 modest_text_utils_validate_folder_name (const gchar *folder_name)
1377 /* based on http://msdn2.microsoft.com/en-us/library/aa365247.aspx,
1378 * with some extras */
1382 const gchar **cursor = NULL;
1383 const gchar *forbidden_names[] = { /* windows does not like these */
1384 "CON", "PRN", "AUX", "NUL", ".", "..", "cur", "tmp", "new",
1385 NULL /* cur, tmp, new are reserved for Maildir */
1388 /* cannot be NULL */
1392 /* cannot be empty */
1393 len = strlen(folder_name);
1397 /* cannot start with a dot, vfat does not seem to like that */
1398 if (folder_name[0] == '.')
1401 /* cannot start or end with a space */
1402 if (g_ascii_isspace(folder_name[0]) || g_ascii_isspace(folder_name[len - 1]))
1405 /* cannot contain a forbidden char */
1406 for (i = 0; i < len; i++)
1407 if (modest_text_utils_is_forbidden_char (folder_name[i], FOLDER_NAME_FORBIDDEN_CHARS))
1410 /* Cannot contain Windows port numbers. I'd like to use GRegex
1411 but it's still not available in Maemo. sergio */
1412 if (!g_ascii_strncasecmp (folder_name, "LPT", 3) ||
1413 !g_ascii_strncasecmp (folder_name, "COM", 3)) {
1417 /* We skip the first 3 characters for the
1419 val = strtol(folder_name+3, &endptr, 10);
1421 /* If the conversion to long succeeded then the string
1422 is not valid for us */
1423 if (*endptr == '\0')
1429 /* cannot contain a forbidden word */
1431 for (cursor = forbidden_names; cursor && *cursor; ++cursor) {
1432 if (g_ascii_strcasecmp (folder_name, *cursor) == 0)
1437 return TRUE; /* it's valid! */
1443 modest_text_utils_validate_domain_name (const gchar *domain)
1445 gboolean valid = FALSE;
1447 const gchar* domain_regex = "^([a-z0-9-]*[a-z0-9]\\.)+[a-z0-9-]*[a-z0-9]$";
1449 g_return_val_if_fail (domain, FALSE);
1454 memset (&rx, 0, sizeof(regex_t)); /* coverity wants this... */
1456 /* domain name: all alphanum or '-' or '.',
1457 * but beginning/ending in alphanum */
1458 if (regcomp (&rx, domain_regex, REG_ICASE|REG_EXTENDED|REG_NOSUB)) {
1459 g_warning ("BUG: error in regexp");
1463 valid = (regexec (&rx, domain, 1, NULL, 0) == 0);
1472 modest_text_utils_validate_email_address (const gchar *email_address,
1473 const gchar **invalid_char_position)
1476 const gchar *c = NULL, *domain = NULL;
1477 static gchar *rfc822_specials = "()<>@,;:\\\"[]&";
1479 if (invalid_char_position)
1480 *invalid_char_position = NULL;
1482 g_return_val_if_fail (email_address, FALSE);
1484 /* check that the email adress contains exactly one @ */
1485 if (!strstr(email_address, "@") ||
1486 (strstr(email_address, "@") != g_strrstr(email_address, "@")))
1489 /* first we validate the name portion (name@domain) */
1490 for (c = email_address; *c; c++) {
1492 (c == email_address ||
1494 *(c - 1) == '\"')) {
1498 if (*c == '\\' && (*++c == ' '))
1500 if (*c <= ' ' || *c >= 127)
1513 if (*c <= ' ' || *c >= 127)
1515 if (strchr(rfc822_specials, *c)) {
1516 if (invalid_char_position)
1517 *invalid_char_position = c;
1521 if (c == email_address || *(c - 1) == '.')
1524 /* next we validate the domain portion (name@domain) */
1525 if (!*(domain = ++c))
1529 if (c == domain || *(c - 1) == '.' || *(c + 1) == '\0')
1533 if (*c <= ' ' || *c >= 127)
1535 if (strchr(rfc822_specials, *c)) {
1536 if (invalid_char_position)
1537 *invalid_char_position = c;
1542 return (count >= 1) ? TRUE : FALSE;
1546 modest_text_utils_validate_recipient (const gchar *recipient, const gchar **invalid_char_position)
1548 gchar *stripped, *current;
1550 gboolean has_error = FALSE;
1552 if (invalid_char_position)
1553 *invalid_char_position = NULL;
1555 g_return_val_if_fail (recipient, FALSE);
1557 if (modest_text_utils_validate_email_address (recipient, invalid_char_position))
1560 stripped = g_strdup (recipient);
1561 stripped = g_strstrip (stripped);
1564 if (*current == '\0') {
1570 if (*current == '\"') {
1571 current = g_utf8_next_char (current);
1573 for (; *current != '\0'; current = g_utf8_next_char (current)) {
1574 if (*current == '\\') {
1575 /* TODO: This causes a warning, which breaks the build,
1576 * because a gchar cannot be < 0.
1578 if (current[1] <0) {
1583 } else if (*current == '\"') {
1585 current = g_utf8_next_char (current);
1591 for (current = stripped ; *current != '\0'; current = g_utf8_next_char (current)) {
1592 if (*current == '<') {
1604 right_part = g_strdup (current);
1606 right_part = g_strstrip (right_part);
1608 if (g_str_has_prefix (right_part, "<") &&
1609 g_str_has_suffix (right_part, ">")) {
1613 address = g_strndup (right_part+1, strlen (right_part) - 2);
1614 g_free (right_part);
1615 valid = modest_text_utils_validate_email_address (address, invalid_char_position);
1619 g_free (right_part);
1626 modest_text_utils_get_display_size (guint64 size)
1628 const guint KB=1024;
1629 const guint MB=1024 * KB;
1630 const guint GB=1024 * MB;
1633 return g_strdup_printf(_FM("sfil_li_size_kb"), 0);
1634 if (0 < size && size < KB)
1635 return g_strdup_printf (_FM("sfil_li_size_kb"), 1);
1636 else if (KB <= size && size < 100 * KB)
1637 return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), size / KB);
1638 else if (100*KB <= size && size < MB)
1639 return g_strdup_printf (_FM("sfil_li_size_100kb_1mb"), (float) size / MB);
1640 else if (MB <= size && size < 10*MB)
1641 return g_strdup_printf (_FM("sfil_li_size_1mb_10mb"), (float) size / MB);
1642 else if (10*MB <= size && size < GB)
1643 return g_strdup_printf (_FM("sfil_li_size_10mb_1gb"), size / MB);
1645 return g_strdup_printf (_FM("sfil_li_size_1gb_or_greater"), (float) size / GB);
1649 get_email_from_address (const gchar * address)
1651 gchar *left_limit, *right_limit;
1653 left_limit = strstr (address, "<");
1654 right_limit = g_strrstr (address, ">");
1656 if ((left_limit == NULL)||(right_limit == NULL)|| (left_limit > right_limit))
1657 return g_strdup (address);
1659 return g_strndup (left_limit + 1, (right_limit - left_limit) - 1);
1663 modest_text_utils_get_color_string (GdkColor *color)
1665 g_return_val_if_fail (color, NULL);
1667 return g_strdup_printf ("#%x%x%x%x%x%x%x%x%x%x%x%x",
1668 (color->red >> 12) & 0xf, (color->red >> 8) & 0xf,
1669 (color->red >> 4) & 0xf, (color->red) & 0xf,
1670 (color->green >> 12) & 0xf, (color->green >> 8) & 0xf,
1671 (color->green >> 4) & 0xf, (color->green) & 0xf,
1672 (color->blue >> 12) & 0xf, (color->blue >> 8) & 0xf,
1673 (color->blue >> 4) & 0xf, (color->blue) & 0xf);
1677 modest_text_utils_text_buffer_get_text (GtkTextBuffer *buffer)
1679 GtkTextIter start, end;
1680 gchar *slice, *current;
1681 GString *result = g_string_new ("");
1683 g_return_val_if_fail (buffer && GTK_IS_TEXT_BUFFER (buffer), NULL);
1685 gtk_text_buffer_get_start_iter (buffer, &start);
1686 gtk_text_buffer_get_end_iter (buffer, &end);
1688 slice = gtk_text_buffer_get_slice (buffer, &start, &end, FALSE);
1691 while (current && current != '\0') {
1692 if (g_utf8_get_char (current) == 0xFFFC) {
1693 result = g_string_append_c (result, ' ');
1694 current = g_utf8_next_char (current);
1696 gchar *next = g_utf8_strchr (current, -1, 0xFFFC);
1698 result = g_string_append (result, current);
1700 result = g_string_append_len (result, current, next - current);
1707 return g_string_free (result, FALSE);
1712 modest_text_utils_is_forbidden_char (const gchar character,
1713 ModestTextUtilsForbiddenCharType type)
1716 const gchar *forbidden_chars = NULL;
1718 /* We need to get the length in the switch because the
1719 compiler needs to know the size at compile time */
1721 case ACCOUNT_TITLE_FORBIDDEN_CHARS:
1722 forbidden_chars = account_title_forbidden_chars;
1723 len = G_N_ELEMENTS (account_title_forbidden_chars);
1725 case FOLDER_NAME_FORBIDDEN_CHARS:
1726 forbidden_chars = folder_name_forbidden_chars;
1727 len = G_N_ELEMENTS (folder_name_forbidden_chars);
1729 case USER_NAME_FORBIDDEN_NAMES:
1730 forbidden_chars = user_name_forbidden_chars;
1731 len = G_N_ELEMENTS (user_name_forbidden_chars);
1734 g_return_val_if_reached (TRUE);
1737 for (i = 0; i < len ; i++)
1738 if (forbidden_chars[i] == character)
1741 return FALSE; /* it's valid! */
1745 modest_text_utils_label_get_selection (GtkLabel *label)
1750 if (gtk_label_get_selection_bounds (GTK_LABEL (label), &start, &end)) {
1751 const gchar *start_offset;
1752 const gchar *end_offset;
1753 start_offset = gtk_label_get_text (GTK_LABEL (label));
1754 start_offset = g_utf8_offset_to_pointer (start_offset, start);
1755 end_offset = gtk_label_get_text (GTK_LABEL (label));
1756 end_offset = g_utf8_offset_to_pointer (end_offset, end);
1757 selection = g_strndup (start_offset, end_offset - start_offset);
1760 return g_strdup ("");
1765 _forward_search_image_char (gunichar ch,
1768 return (ch == 0xFFFC);
1772 modest_text_utils_buffer_selection_is_valid (GtkTextBuffer *buffer)
1775 GtkTextIter start, end;
1777 g_return_val_if_fail (GTK_IS_TEXT_BUFFER (buffer), FALSE);
1779 result = gtk_text_buffer_get_has_selection (GTK_TEXT_BUFFER (buffer));
1781 /* check there are no images in selection */
1783 gtk_text_buffer_get_selection_bounds (buffer, &start, &end);
1784 if (gtk_text_iter_get_char (&start)== 0xFFFC)
1787 gtk_text_iter_backward_char (&end);
1788 if (gtk_text_iter_forward_find_char (&start, _forward_search_image_char,
1799 modest_text_utils_escape_mnemonics (const gchar *text)
1802 GString *result = NULL;
1807 result = g_string_new ("");
1808 for (p = text; *p != '\0'; p++) {
1810 result = g_string_append (result, "__");
1812 result = g_string_append_c (result, *p);
1815 return g_string_free (result, FALSE);