Take into account unichar space on converting to plain text quoted text
[modest] / src / modest-text-utils.c
index 0b8927e..99cb710 100644 (file)
@@ -64,8 +64,6 @@
  */
 #define HYPERLINKIFY_MAX_LENGTH (1024*50)
 
-
-
 /*
  * we need these regexps to find URLs in plain text e-mails
  */
@@ -109,14 +107,14 @@ struct _url_match_t {
 
 /* note: match MARK_AMP_URI_STR as well, because after txt->html, a '&' will look like $(MARK_AMP_URI_STR)"amp;" */
 #define MAIL_VIEWER_URL_MATCH_PATTERNS  {                              \
-       { "(file|rtsp|http|ftp|https|mms|mmsh|rtsp|rdp|lastfm)://[-a-z0-9_$.+!*(),;:@%=?/~#" MARK_AMP_URI_STR \
-                       "]+[-a-z0-9_$%" MARK_AMP_URI_STR "=?/~#]",      \
+       { "(feed:|)(file|rtsp|http|ftp|https|mms|mmsh|webcal|feed|rtsp|rdp|lastfm|sip)://[-a-z0-9_$.+!*(),;:@%=\?/~#&" MARK_AMP_URI_STR \
+                       "]+[-a-z0-9_$%&" MARK_AMP_URI_STR "=?/~#]",     \
          NULL, NULL },\
        { "www\\.[-a-z0-9_$.+!*(),;:@%=?/~#" MARK_AMP_URI_STR "]+[-a-z0-9_$%" MARK_AMP_URI_STR "=?/~#]",\
                        NULL, "http://" },                              \
        { "ftp\\.[-a-z0-9_$.+!*(),;:@%=?/~#" MARK_AMP_URI_STR "]+[-a-z0-9_$%" MARK_AMP_URI_STR "=?/~#]",\
          NULL, "ftp://" },\
-       { "(jabberto|voipto|sipto|sip|chatto|xmpp):[-_a-z@0-9.+]+", \
+       { "(jabberto|voipto|sipto|sip|chatto|skype|xmpp):[-_a-z@0-9.+]+", \
           NULL, NULL},                                             \
        { "mailto:[-_a-z0-9.\\+]+@[-_a-z0-9.]+",                    \
          NULL, NULL},\
@@ -139,14 +137,15 @@ const guint USER_NAME_FORBIDDEN_CHARS_LENGTH = G_N_ELEMENTS (user_name_forbidden
 
 /* private */
 static gchar*   cite                    (const time_t sent_date, const gchar *from);
-static void     hyperlinkify_plain_text (GString *txt);
+static void     hyperlinkify_plain_text (GString *txt, gint offset);
 static gint     cmp_offsets_reverse     (const url_match_t *match1, const url_match_t *match2);
-static GSList*  get_url_matches         (GString *txt);
+static GSList*  get_url_matches         (GString *txt, gint offset);
 
 static GString* get_next_line           (const char *b, const gsize blen, const gchar * iter);
 static int      get_indent_level        (const char *l);
-static void     unquote_line            (GString * l);
-static void     append_quoted           (GString * buf, const int indent, const GString * str, 
+static void     unquote_line            (GString * l, const gchar *quote_symbol);
+static void     append_quoted           (GString * buf, const gchar *quote_symbol,
+                                        const int indent, const GString * str, 
                                         const int cutpoint);
 static int      get_breakpoint_utf8     (const gchar * s, const gint indent, const gint limit);
 static int      get_breakpoint_ascii    (const gchar * s, const gint indent, const gint limit);
@@ -212,14 +211,17 @@ modest_text_utils_cite (const gchar *text,
        g_return_val_if_fail (text, NULL);
        g_return_val_if_fail (content_type, NULL);
        
-       if (!signature)
-               retval = g_strdup ("");
-       else if (strcmp(content_type, "text/html") == 0) {
-               tmp_sig = g_strconcat ("\n", signature, NULL);
-               retval = modest_text_utils_convert_to_html_body(tmp_sig, -1, TRUE);
+       if (!signature) {
+               tmp_sig = g_strdup (text);
+       } else {
+               tmp_sig = g_strconcat (text, "\n", MODEST_TEXT_UTILS_SIGNATURE_MARKER, "\n", signature, NULL);
+       }
+
+       if (strcmp (content_type, "text/html") == 0) {
+               retval = modest_text_utils_convert_to_html_body (tmp_sig, -1, TRUE);
                g_free (tmp_sig);
        } else {
-               retval = g_strconcat (text, "\n", signature, NULL);
+               retval = tmp_sig;
        }
 
        return retval;
@@ -278,6 +280,13 @@ modest_text_utils_strftime(char *s, gsize max, const char *fmt, time_t timet)
 {
         struct tm tm;
 
+       /* To prevent possible problems in strftime that could leave
+          garbage in the s variable */
+       if (s)
+               s[0] = '\0';
+       else
+               return 0;
+
        /* does not work on old maemo glib: 
         *   g_date_set_time_t (&date, timet);
         */
@@ -288,23 +297,84 @@ modest_text_utils_strftime(char *s, gsize max, const char *fmt, time_t timet)
 gchar *
 modest_text_utils_derived_subject (const gchar *subject, const gchar *prefix)
 {
-       gchar *tmp;
+       gchar *tmp, *subject_dup, *retval;
+       gint prefix_len;
 
        g_return_val_if_fail (prefix, NULL);
 
        if (!subject || subject[0] == '\0')
                subject = _("mail_va_no_subject");
 
-       tmp = g_strchug (g_strdup (subject));
+       subject_dup = g_strdup (subject);
+       tmp = g_strchug (subject_dup);
 
-       if (!strncmp (tmp, prefix, strlen (prefix))) {
-               return tmp;
-       } else {
-               g_free (tmp);
-               return g_strdup_printf ("%s %s", prefix, subject);
+       /* We do not want things like "Re: Re: Re:" or "Fw: Fw:" so
+          delete the previous ones */
+       prefix_len = strlen (prefix);
+       do {
+               if (g_str_has_prefix (tmp, prefix)) {
+                       tmp += prefix_len;
+                       tmp = g_strchug (tmp);
+               } else {
+                       gchar *prefix_down, *tmp_down;
+
+                       /* We need this to properly check the cases of
+                          some clients adding FW: instead of Fw: for
+                          example */
+                       prefix_down = g_utf8_strdown (prefix, -1);
+                       tmp_down = g_utf8_strdown (tmp, -1);
+                       if (g_str_has_prefix (tmp_down, prefix_down)) {
+                               tmp += prefix_len;
+                               tmp = g_strchug (tmp);
+                               g_free (prefix_down);
+                               g_free (tmp_down);
+                       } else {
+                               g_free (prefix_down);
+                               g_free (tmp_down);
+                               break;
+                       }
+               }
+       } while (tmp);
+
+       retval = g_strdup_printf ("%s %s", prefix, tmp);
+       g_free (subject_dup);
+
+       return retval;
+}
+
+
+/* Performs a case-insensitive strstr for ASCII strings */
+static const gchar *
+ascii_stristr(const gchar *haystack, const gchar *needle)
+{
+       int needle_len;
+       int haystack_len;
+       const gchar *pos;
+       const gchar *max_pos;
+
+       if (haystack == NULL || needle == NULL) {
+               return haystack;  /* as in strstr */
+       }
+
+       needle_len = strlen(needle);
+
+       if (needle_len == 0) {
+               return haystack;  /* as in strstr */
+       }
+
+       haystack_len = strlen(haystack);
+       max_pos = haystack + haystack_len - needle_len;
+
+       for (pos = haystack; pos <= max_pos; pos++) {
+               if (g_ascii_strncasecmp (pos, needle, needle_len) == 0) {
+                       return pos;
+               }
        }
+
+       return NULL;
 }
 
+
 gchar*
 modest_text_utils_remove_address (const gchar *address_list, const gchar *address)
 {
@@ -313,26 +383,26 @@ modest_text_utils_remove_address (const gchar *address_list, const gchar *addres
        gchar *email_address;
 
        g_return_val_if_fail (address_list, NULL);
-       
+
        if (!address)
                return g_strdup (address_list);
 
        email_address = get_email_from_address (address);
-       
+
        /* search for substring */
-       if (!strstr ((const char *) address_list, (const char *) email_address)) {
+       if (!ascii_stristr ((const char *) address_list, (const char *) email_address)) {
                g_free (email_address);
                return g_strdup (address_list);
        }
 
        dup = g_strdup (address_list);
        filtered_emails = g_string_new (NULL);
-       
+
        token = strtok_r (dup, ",", &ptr);
 
        while (token != NULL) {
                /* Add to list if not found */
-               if (!strstr ((const char *) token, (const char *) email_address)) {
+               if (!ascii_stristr ((const char *) token, (const char *) email_address)) {
                        if (filtered_emails->len == 0)
                                g_string_append_printf (filtered_emails, "%s", g_strstrip (token));
                        else
@@ -356,34 +426,48 @@ modest_text_utils_remove_duplicate_addresses (const gchar *address_list)
 {
        GSList *addresses, *cursor;
        GHashTable *table;
-       gchar *new_list;
+       gchar *new_list = NULL;
        
        g_return_val_if_fail (address_list, NULL);
 
-       table = g_hash_table_new (g_str_hash, g_str_equal);
+       table = g_hash_table_new_full (g_str_hash, g_str_equal, g_free, NULL);
        addresses = modest_text_utils_split_addresses_list (address_list);
 
-       new_list = g_strdup("");
        cursor = addresses;
        while (cursor) {
                const gchar* address = (const gchar*)cursor->data;
 
+               /* We need only the email to just compare it and not
+                  the full address which would make "a <a@a.com>"
+                  different from "a@a.com" */
+               const gchar *email = get_email_from_address (address);
+
                /* ignore the address if already seen */
-               if (g_hash_table_lookup (table, address) == 0) {
-               
-                       gchar *tmp = g_strjoin (",", new_list, address, NULL);
-                       g_free (new_list);
+               if (g_hash_table_lookup (table, email) == 0) {
+                       gchar *tmp;
+
+                       /* Include the full address and not only the
+                          email in the returned list */
+                       if (!new_list) {
+                               tmp = g_strdup (address);
+                       } else {
+                               tmp = g_strjoin (",", new_list, address, NULL);
+                               g_free (new_list);
+                       }
                        new_list = tmp;
                        
-                       g_hash_table_insert (table, (gchar*)address, GINT_TO_POINTER(1));
+                       g_hash_table_insert (table, (gchar*)email, GINT_TO_POINTER(1));
                }
                cursor = g_slist_next (cursor);
        }
 
-       g_hash_table_destroy (table);
+       g_hash_table_unref (table);
        g_slist_foreach (addresses, (GFunc)g_free, NULL);
        g_slist_free (addresses);
 
+       if (new_list == NULL)
+               new_list = g_strdup ("");
+
        return new_list;
 }
 
@@ -403,7 +487,7 @@ modest_text_utils_convert_buffer_to_html_start (GString *html, const gchar *data
                guchar kar = data[i];
                
                if (space_seen && kar != ' ') {
-                       g_string_append_c (html, ' ');
+                       g_string_append (html, " ");
                        space_seen = FALSE;
                }
                
@@ -440,7 +524,6 @@ modest_text_utils_convert_buffer_to_html_start (GString *html, const gchar *data
                        break_dist = 0;
                        if (space_seen) { /* second space in a row */
                                g_string_append (html, "&nbsp; ");
-                               space_seen = FALSE;
                        } else
                                space_seen = TRUE;
                        break;
@@ -487,7 +570,7 @@ modest_text_utils_convert_to_html (const gchar *data)
        g_string_append (html, "</body></html>");
 
        if (len <= HYPERLINKIFY_MAX_LENGTH)
-               hyperlinkify_plain_text (html);
+               hyperlinkify_plain_text (html, 0);
 
        modest_text_utils_convert_buffer_to_html_finish (html);
        
@@ -511,7 +594,7 @@ modest_text_utils_convert_to_html_body (const gchar *data, gssize n, gboolean hy
        modest_text_utils_convert_buffer_to_html_start (html, data, n);
 
        if (hyperlinkify && (n < HYPERLINKIFY_MAX_LENGTH))
-               hyperlinkify_plain_text (html);
+               hyperlinkify_plain_text (html, 0);
 
        modest_text_utils_convert_buffer_to_html_finish (html);
        
@@ -584,6 +667,7 @@ modest_text_utils_split_addresses_list (const gchar *addresses)
        const gchar *my_addrs = addresses;
        const gchar *end;
        gchar *addr;
+       gboolean after_at = FALSE;
 
        g_return_val_if_fail (addresses, NULL);
        
@@ -598,8 +682,20 @@ modest_text_utils_split_addresses_list (const gchar *addresses)
        /* nope, we are at the start of some address
         * now, let's find the end of the address */
        end = my_addrs + 1;
-       while (end[0] && end[0] != ',' && end[0] != ';')
+       while (end[0] && end[0] != ';' && !(after_at && end[0] == ',')) {
+               if (end[0] == '\"') {
+                       while (end[0] && end[0] != '\"')
+                               ++end;
+               }
+               if (end[0] == '@') {
+                       after_at = TRUE;
+               }
+               if ((end[0] && end[0] == '>')&&(end[1] && end[1] == ',')) {
+                       ++end;
+                       break;
+               }
                ++end;
+       }
 
        /* we got the address; copy it and remove trailing whitespace */
        addr = g_strndup (my_addrs, end - my_addrs);
@@ -721,7 +817,7 @@ get_indent_level (const char *l)
        /*      if we hit the signature marker "-- ", we return -(indent + 1). This
         *      stops reformatting.
         */
-       if (strcmp (l, "-- ") == 0) {
+       if (strcmp (l, MODEST_TEXT_UTILS_SIGNATURE_MARKER) == 0) {
                return -1 - indent;
        } else {
                return indent;
@@ -729,15 +825,17 @@ get_indent_level (const char *l)
 }
 
 static void
-unquote_line (GString * l)
+unquote_line (GString * l, const gchar *quote_symbol)
 {
        gchar *p;
+       gint quote_len;
 
        p = l->str;
+       quote_len = strlen (quote_symbol);
        while (p[0]) {
-               if (p[0] == '>') {
-                       if (p[1] == ' ') {
-                               p++;
+               if (g_str_has_prefix (p, quote_symbol)) {
+                       if (p[quote_len] == ' ') {
+                               p += quote_len;
                        }
                } else {
                        break;
@@ -748,15 +846,19 @@ unquote_line (GString * l)
 }
 
 static void
-append_quoted (GString * buf, int indent, const GString * str,
+append_quoted (GString * buf, const gchar *quote_symbol,
+              int indent, const GString * str,
               const int cutpoint)
 {
        int i;
+       gchar *quote_concat;
 
        indent = indent < 0 ? abs (indent) - 1 : indent;
+       quote_concat = g_strconcat (quote_symbol, " ", NULL);
        for (i = 0; i <= indent; i++) {
-               g_string_append (buf, "> ");
+               g_string_append (buf, quote_concat);
        }
+       g_free (quote_concat);
        if (cutpoint > 0) {
                g_string_append_len (buf, str->str, cutpoint);
        } else {
@@ -845,37 +947,25 @@ quoted_attachments (GList *attachments)
 
 }
 
-static gchar *
-modest_text_utils_quote_plain_text (const gchar *text, 
-                                   const gchar *cite, 
-                                   const gchar *signature,
-                                   GList *attachments,
-                                   int limit)
+static GString *
+modest_text_utils_quote_body (GString *output, const gchar *text,
+                             const gchar *quote_symbol,
+                             int limit)
 {
+
        const gchar *iter;
-       gint indent, breakpoint, rem_indent = 0;
-       GString *q, *l, *remaining;
        gsize len;
-       gchar *attachments_string = NULL;
-
-       q = g_string_new ("\n");
-       if (signature != NULL) {
-               q = g_string_append (q, signature);
-               q = g_string_append_c (q, '\n');
-       }
-       q = g_string_append (q, cite);
-       q = g_string_append_c (q, '\n');
-
-       /* remaining will store the rest of the line if we have to break it */
-       remaining = g_string_new ("");
+       gint indent, breakpoint, rem_indent = 0;
+       GString *l, *remaining;
 
        iter = text;
        len = strlen(text);
+       remaining = g_string_new ("");
        do {
                l = get_next_line (text, len, iter);
                iter = iter + l->len + 1;
                indent = get_indent_level (l->str);
-               unquote_line (l);
+               unquote_line (l, quote_symbol);
 
                if (remaining->len) {
                        if (l->len && indent == rem_indent) {
@@ -883,17 +973,20 @@ modest_text_utils_quote_plain_text (const gchar *text,
                                g_string_prepend (l, remaining->str);
                        } else {
                                do {
+                                       gunichar remaining_first;
                                        breakpoint =
                                                get_breakpoint (remaining->str,
                                                                rem_indent,
                                                                limit);
-                                       append_quoted (q, rem_indent,
+                                       append_quoted (output, quote_symbol, rem_indent,
                                                       remaining, breakpoint);
                                        g_string_erase (remaining, 0,
                                                        breakpoint);
-                                       if (remaining->str[0] == ' ') {
-                                               g_string_erase (remaining, 0,
-                                                               1);
+                                       remaining_first = g_utf8_get_char_validated (remaining->str, remaining->len);
+                                       if (remaining_first != ((gunichar) -1)) {
+                                               if (g_unichar_isspace (remaining_first)) {
+                                                       g_string_erase (remaining, 0, g_utf8_next_char (remaining->str) - remaining->str);
+                                               }
                                        }
                                } while (remaining->len);
                        }
@@ -905,10 +998,36 @@ modest_text_utils_quote_plain_text (const gchar *text,
                        g_string_erase (remaining, 0, 1);
                }
                rem_indent = indent;
-               append_quoted (q, indent, l, breakpoint);
+               append_quoted (output, quote_symbol, indent, l, breakpoint);
                g_string_free (l, TRUE);
        } while ((iter < text + len) || (remaining->str[0]));
 
+       return output;
+}
+
+static gchar *
+modest_text_utils_quote_plain_text (const gchar *text, 
+                                   const gchar *cite, 
+                                   const gchar *signature,
+                                   GList *attachments,
+                                   int limit)
+{
+       GString *q;
+       gchar *attachments_string = NULL;
+
+       q = g_string_new ("");
+
+       if (signature != NULL) {
+               g_string_append_printf (q, "\n%s\n", MODEST_TEXT_UTILS_SIGNATURE_MARKER);
+               q = g_string_append (q, signature);
+       }
+
+       q = g_string_append (q, "\n");
+       q = g_string_append (q, cite);
+       q = g_string_append_c (q, '\n');
+
+       q = modest_text_utils_quote_body (q, text, ">", limit);
+
        attachments_string = quoted_attachments (attachments);
        q = g_string_append (q, attachments_string);
        g_free (attachments_string);
@@ -916,6 +1035,17 @@ modest_text_utils_quote_plain_text (const gchar *text,
        return g_string_free (q, FALSE);
 }
 
+static void
+quote_html_add_to_gstring (GString *string,
+                          const gchar *text)
+{
+       if (text && strcmp (text, "")) {
+               gchar *html_text = modest_text_utils_convert_to_html_body (text, -1, TRUE);
+               g_string_append_printf (string, "%s<br/>", html_text);
+               g_free (html_text);
+       }
+}
+
 static gchar*
 modest_text_utils_quote_html (const gchar *text, 
                              const gchar *cite, 
@@ -923,38 +1053,37 @@ modest_text_utils_quote_html (const gchar *text,
                              GList *attachments,
                              int limit)
 {
-       gchar *result = NULL;
-       gchar *signature_result = NULL;
-       const gchar *format = \
-               "<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\">\n" \
-               "<html>\n" \
-               "<body>\n" \
-               "<br/>%s<br/>" \
-               "<pre>%s<br/>%s<br/>%s</pre>\n" \
-               "</body>\n" \
-               "</html>\n";
-       gchar *attachments_string = NULL;
-       gchar *q_attachments_string = NULL;
-       gchar *q_cite = NULL;
-       gchar *html_text = NULL;
-
-       if (signature == NULL)
-               signature_result = g_strdup ("");
-       else
-               signature_result = modest_text_utils_convert_to_html_body (signature, -1, TRUE);
+       GString *result_string;
+
+       result_string = 
+               g_string_new ( \
+                             "<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\">\n" \
+                             "<html>\n"                                \
+                             "<body>\n<br/>\n");
+
+       if (text || cite || signature) {
+               GString *quoted_text;
+               g_string_append (result_string, "<pre>\n");
+               if (signature) {
+                       quote_html_add_to_gstring (result_string, MODEST_TEXT_UTILS_SIGNATURE_MARKER);
+                       quote_html_add_to_gstring (result_string, signature);
+               }
+               quote_html_add_to_gstring (result_string, cite);
+               quoted_text = g_string_new ("");
+               quoted_text = modest_text_utils_quote_body (quoted_text, (text) ? text : "", ">", limit);
+               quote_html_add_to_gstring (result_string, quoted_text->str);
+               g_string_free (quoted_text, TRUE);
+               if (attachments) {
+                       gchar *attachments_string = quoted_attachments (attachments);
+                       quote_html_add_to_gstring (result_string, attachments_string);
+                       g_free (attachments_string);
+               }
+               g_string_append (result_string, "</pre>");
+       }
+       g_string_append (result_string, "</body>");
+       g_string_append (result_string, "</html>");
 
-       attachments_string = quoted_attachments (attachments);
-       q_attachments_string = modest_text_utils_convert_to_html_body (attachments_string, -1, TRUE);
-       q_cite = modest_text_utils_convert_to_html_body (cite, -1, TRUE);
-       html_text = modest_text_utils_convert_to_html_body (text, -1, TRUE);
-       result = g_strdup_printf (format, signature_result, q_cite, html_text, q_attachments_string);
-       g_free (q_cite);
-       g_free (html_text);
-       g_free (attachments_string);
-       g_free (q_attachments_string);
-       g_free (signature_result);
-       
-       return result;
+       return g_string_free (result_string, FALSE);
 }
 
 static gint 
@@ -963,8 +1092,9 @@ cmp_offsets_reverse (const url_match_t *match1, const url_match_t *match2)
        return match2->offset - match1->offset;
 }
 
-static gboolean url_matches_block = 0;
+static gint url_matches_block = 0;
 static url_match_pattern_t patterns[] = MAIL_VIEWER_URL_MATCH_PATTERNS;
+static GMutex *url_patterns_mutex = NULL;
 
 
 static gboolean
@@ -999,25 +1129,33 @@ free_patterns ()
 void
 modest_text_utils_hyperlinkify_begin (void)
 {
+
+       if (url_patterns_mutex == NULL) {
+               url_patterns_mutex = g_mutex_new ();
+       }
+       g_mutex_lock (url_patterns_mutex);
        if (url_matches_block == 0)
                compile_patterns ();
        url_matches_block ++;
+       g_mutex_unlock (url_patterns_mutex);
 }
 
 void
 modest_text_utils_hyperlinkify_end (void)
 {
+       g_mutex_lock (url_patterns_mutex);
        url_matches_block--;
        if (url_matches_block <= 0)
                free_patterns ();
+       g_mutex_unlock (url_patterns_mutex);
 }
 
 
 static GSList*
-get_url_matches (GString *txt)
+get_url_matches (GString *txt, gint offset)
 {
        regmatch_t rm;
-        guint rv, i, offset = 0;
+        guint rv, i, tmp_offset = 0;
         GSList *match_list = NULL;
 
        const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
@@ -1027,13 +1165,13 @@ get_url_matches (GString *txt)
 
         /* find all the matches */
        for (i = 0; i != pattern_num; ++i) {
-               offset     = 0; 
+               tmp_offset     = offset;        
                while (1) {
                        url_match_t *match;
                        gboolean is_submatch;
                        GSList *cursor;
                        
-                       if ((rv = regexec (patterns[i].preg, txt->str + offset, 1, &rm, 0)) != 0) {
+                       if ((rv = regexec (patterns[i].preg, txt->str + tmp_offset, 1, &rm, 0)) != 0) {
                                g_return_val_if_fail (rv == REG_NOMATCH, NULL); /* this should not happen */
                                break; /* try next regexp */ 
                        }
@@ -1046,7 +1184,7 @@ get_url_matches (GString *txt)
                        while (cursor && !is_submatch) {
                                const url_match_t *old_match =
                                        (const url_match_t *) cursor->data;
-                               guint new_offset = offset + rm.rm_so;
+                               guint new_offset = tmp_offset + rm.rm_so;
                                is_submatch = (new_offset >  old_match->offset &&
                                               new_offset <  old_match->offset + old_match->len);
                                cursor = g_slist_next (cursor);
@@ -1055,12 +1193,12 @@ get_url_matches (GString *txt)
                        if (!is_submatch) {
                                /* make a list of our matches (<offset, len, prefix> tupels)*/
                                match = g_slice_new (url_match_t);
-                               match->offset = offset + rm.rm_so;
+                               match->offset = tmp_offset + rm.rm_so;
                                match->len    = rm.rm_eo - rm.rm_so;
                                match->prefix = patterns[i].prefix;
                                match_list = g_slist_prepend (match_list, match);
                        }               
-                       offset += rm.rm_eo;
+                       tmp_offset += rm.rm_eo;
                }
        }
 
@@ -1101,10 +1239,10 @@ replace_string (const gchar *haystack, const gchar *needle, gchar repl)
 }
 
 static void
-hyperlinkify_plain_text (GString *txt)
+hyperlinkify_plain_text (GString *txt, gint offset)
 {
        GSList *cursor;
-       GSList *match_list = get_url_matches (txt);
+       GSList *match_list = get_url_matches (txt, offset);
 
        /* we will work backwards, so the offsets stay valid */
        for (cursor = match_list; cursor; cursor = cursor->next) {
@@ -1140,6 +1278,18 @@ hyperlinkify_plain_text (GString *txt)
        g_slist_free (match_list);
 }
 
+void
+modest_text_utils_hyperlinkify (GString *string_buffer)
+{
+       gchar *after_body;
+       gint offset = 0;
+
+       after_body = strstr (string_buffer->str, "<body>");
+       if (after_body != NULL)
+               offset = after_body - string_buffer->str;
+       hyperlinkify_plain_text (string_buffer, offset);
+}
+
 
 /* for optimization reasons, we change the string in-place */
 void
@@ -1163,18 +1313,50 @@ modest_text_utils_get_display_address (gchar *address)
                
        for (i = 0; address[i]; ++i) {
                if (address[i] == '<') {
-                       if (G_UNLIKELY(i == 0))
-                               return; /* there's nothing else, leave it */
-                       else {
+                       if (G_UNLIKELY(i == 0)) {
+                               break; /* there's nothing else, leave it */
+                       }else {
                                address[i] = '\0'; /* terminate the string here */
-                               return;
+                               break;
                        }
                }
        }
+
+       g_strchomp (address);
 }
 
 
+gchar *
+modest_text_utils_get_display_addresses (const gchar *recipients)
+{
+       gchar *addresses;
+       GSList *recipient_list;
+
+       addresses = NULL;
+       recipient_list = modest_text_utils_split_addresses_list (recipients);
+       if (recipient_list) {
+               GString *add_string = g_string_sized_new (strlen (recipients));
+               GSList *iter = recipient_list;
+               gboolean first = TRUE;
+
+               while (iter) {
+                       /* Strings are changed in place */
+                       modest_text_utils_get_display_address ((gchar *) iter->data);
+                       if (G_UNLIKELY (first)) {
+                               g_string_append_printf (add_string, "%s", (gchar *) iter->data);
+                               first = FALSE;
+                       } else {
+                               g_string_append_printf (add_string, ", %s", (gchar *) iter->data);
+                       }
+                       iter = g_slist_next (iter);
+               }
+               g_slist_foreach (recipient_list, (GFunc) g_free, NULL);
+               g_slist_free (recipient_list);
+               addresses = g_string_free (add_string, FALSE);
+       }
 
+       return addresses;
+}
 
 
 gchar *
@@ -1225,7 +1407,7 @@ modest_text_utils_get_subject_prefix_len (const gchar *sub)
                int c = prefix_len + 1;
                while (sub[c] && sub[c] != ']')
                        ++c;
-               if (sub[c])
+               if (!sub[c])
                        return 0; /* no end to the ']' found */
                else
                        prefix_len = c + 1;
@@ -1260,7 +1442,8 @@ modest_text_utils_utf8_strcmp (const gchar* s1, const gchar *s2, gboolean insens
        if (!insensitive) {
 
                /* optimization: shortcut if first char is ascii */ 
-               if (((s1[0] & 0xf0)== 0) && ((s2[0] & 0xf0) == 0)) 
+               if (((s1[0] & 0x80)== 0) && ((s2[0] & 0x80) == 0) &&
+                   (s1[0] != s2[0])) 
                        return s1[0] - s2[0];
                
                return g_utf8_collate (s1, s2);
@@ -1269,8 +1452,9 @@ modest_text_utils_utf8_strcmp (const gchar* s1, const gchar *s2, gboolean insens
                gint result;
                gchar *n1, *n2;
 
-               /* optimization: short cut iif first char is ascii */ 
-               if (((s1[0] & 0xf0) == 0) && ((s2[0] & 0xf0) == 0)) 
+               /* optimization: shortcut if first char is ascii */ 
+               if (((s1[0] & 0x80) == 0) && ((s2[0] & 0x80) == 0) &&
+                   (tolower(s1[0]) != tolower (s2[0]))) 
                        return tolower(s1[0]) - tolower(s2[0]);
                
                n1 = g_utf8_strdown (s1, -1);
@@ -1300,10 +1484,18 @@ modest_text_utils_get_display_date (time_t date)
 
        /* if it's today, show the time, if it's not today, show the date instead */
 
+       /* TODO: take into account the system config for 24/12h */
+#ifdef MODEST_TOOLKIT_HILDON2
+       if (day == date_day) /* is the date today? */
+               modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, _HL("wdgt_va_24h_time"), date);
+       else 
+               modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, _HL("wdgt_va_date"), date); 
+#else
        if (day == date_day) /* is the date today? */
                modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, "%X", date);
        else 
                modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, "%x", date); 
+#endif
 
        return date_buf; /* this is a static buffer, don't free! */
 }
@@ -1320,9 +1512,8 @@ modest_text_utils_validate_folder_name (const gchar *folder_name)
        gint i;
        const gchar **cursor = NULL;
        const gchar *forbidden_names[] = { /* windows does not like these */
-               "CON", "PRN", "AUX", "NUL", "COM1", "COM2", "COM3", "COM4", "COM5", "COM6",
-               "COM7", "COM8", "COM9", "LPT1", "LPT2", "LPT3", "LPT4", "LPT5", "LPT6", "LPT7", "LPT8", "LPT9",
-               ".", "..", "cur", "tmp", "new", NULL /* cur, tmp new  are reserved for Maildir */
+               "CON", "PRN", "AUX", "NUL", ".", "..", "cur", "tmp", "new", 
+               NULL /* cur, tmp, new are reserved for Maildir */
        };
        
        /* cannot be NULL */
@@ -1346,6 +1537,25 @@ modest_text_utils_validate_folder_name (const gchar *folder_name)
        for (i = 0; i < len; i++)
                if (modest_text_utils_is_forbidden_char (folder_name[i], FOLDER_NAME_FORBIDDEN_CHARS))
                        return FALSE;
+
+       /* Cannot contain Windows port numbers. I'd like to use GRegex
+          but it's still not available in Maemo. sergio */
+       if (!g_ascii_strncasecmp (folder_name, "LPT", 3) ||
+           !g_ascii_strncasecmp (folder_name, "COM", 3)) {
+               glong val;
+               gchar *endptr;
+
+               /* We skip the first 3 characters for the
+                  comparison */
+               val = strtol(folder_name+3, &endptr, 10);
+
+               /* If the conversion to long succeeded then the string
+                  is not valid for us */
+               if (*endptr == '\0')
+                       return FALSE;
+               else
+                       return TRUE;
+       }
        
        /* cannot contain a forbidden word */
        if (len <= 4) {
@@ -1551,19 +1761,19 @@ modest_text_utils_get_display_size (guint64 size)
        const guint GB=1024 * MB;
 
        if (size == 0)
-               return g_strdup_printf(_FM("sfil_li_size_kb"), 0);
-       if (0 < size && size < KB)
-               return g_strdup_printf (_FM("sfil_li_size_kb"), 1);
+               return g_strdup_printf (_FM("sfil_li_size_kb"), (int) 0);
+       if (0 <= size && size < KB)
+               return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), (int) 1);
        else if (KB <= size && size < 100 * KB)
-               return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), size / KB);
+               return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), (int) size / KB);
        else if (100*KB <= size && size < MB)
-               return g_strdup_printf (_FM("sfil_li_size_100kb_1mb"), (float) size / MB);
+               return g_strdup_printf (_FM("sfil_li_size_100kb_1mb"), (int) size / KB);
        else if (MB <= size && size < 10*MB)
                return g_strdup_printf (_FM("sfil_li_size_1mb_10mb"), (float) size / MB);
        else if (10*MB <= size && size < GB)
-               return g_strdup_printf (_FM("sfil_li_size_10mb_1gb"), size / MB);
+               return g_strdup_printf (_FM("sfil_li_size_10mb_1gb"), (float) size / MB);
        else
-               return g_strdup_printf (_FM("sfil_li_size_1gb_or_greater"), (float) size / GB); 
+               return g_strdup_printf (_FM("sfil_li_size_1gb_or_greater"), (float) size / GB);
 }
 
 static gchar *
@@ -1580,11 +1790,11 @@ get_email_from_address (const gchar * address)
                return g_strndup (left_limit + 1, (right_limit - left_limit) - 1);
 }
 
-gchar *      
+gchar *
 modest_text_utils_get_color_string (GdkColor *color)
 {
        g_return_val_if_fail (color, NULL);
-       
+
        return g_strdup_printf ("#%x%x%x%x%x%x%x%x%x%x%x%x",
                                (color->red >> 12)   & 0xf, (color->red >> 8)   & 0xf,
                                (color->red >>  4)   & 0xf, (color->red)        & 0xf,
@@ -1716,6 +1926,17 @@ modest_text_utils_buffer_selection_is_valid (GtkTextBuffer *buffer)
        return result;
 }
 
+static void
+remove_quotes (gchar **quotes)
+{
+       if (g_str_has_prefix (*quotes, "\"") && g_str_has_suffix (*quotes, "\"")) {
+               gchar *result;
+               result = g_strndup ((*quotes)+1, strlen (*quotes) - 2);
+               g_free (*quotes);
+               *quotes = result;
+       }
+}
+
 gchar *
 modest_text_utils_escape_mnemonics (const gchar *text)
 {
@@ -1735,3 +1956,125 @@ modest_text_utils_escape_mnemonics (const gchar *text)
        
        return g_string_free (result, FALSE);
 }
+
+gchar *
+modest_text_utils_simplify_recipients (const gchar *recipients)
+{
+       GSList *addresses, *node;
+       GString *result;
+       gboolean is_first = TRUE;
+
+       if (recipients == NULL)
+               return g_strdup ("");
+
+       addresses = modest_text_utils_split_addresses_list (recipients);
+       result = g_string_new ("");
+
+       for (node = addresses; node != NULL; node = g_slist_next (node)) {
+               const gchar *address = (const gchar *) node->data;
+               gchar *left_limit, *right_limit;
+
+               left_limit = strstr (address, "<");
+               right_limit = g_strrstr (address, ">");
+
+               if (is_first)
+                       is_first = FALSE;
+               else
+                       result = g_string_append (result, ", ");
+
+               if ((left_limit == NULL)||(right_limit == NULL)|| (left_limit > right_limit)) {
+                       result = g_string_append (result, address);
+               } else {
+                       gchar *name_side;
+                       gchar *email_side;
+                       name_side = g_strndup (address, left_limit - address);
+                       name_side = g_strstrip (name_side);
+                       remove_quotes (&name_side);
+                       email_side = get_email_from_address (address);
+                       if (name_side && email_side && !strcmp (name_side, email_side)) {
+                               result = g_string_append (result, email_side);
+                       } else {
+                               result = g_string_append (result, address);
+                       }
+                       g_free (name_side);
+                       g_free (email_side);
+               }
+
+       }
+       g_slist_foreach (addresses, (GFunc)g_free, NULL);
+       g_slist_free (addresses);
+
+       return g_string_free (result, FALSE);
+
+}
+
+GSList *
+modest_text_utils_remove_duplicate_addresses_list (GSList *address_list)
+{
+       GSList *new_list, *iter;
+       GHashTable *table;
+
+       g_return_val_if_fail (address_list, NULL);
+
+       table = g_hash_table_new_full (g_str_hash, g_str_equal, g_free, NULL);
+
+       new_list = address_list;
+       iter = address_list;
+       while (iter) {
+               const gchar* address = (const gchar*)iter->data;
+
+               /* We need only the email to just compare it and not
+                  the full address which would make "a <a@a.com>"
+                  different from "a@a.com" */
+               const gchar *email = get_email_from_address (address);
+
+               /* ignore the address if already seen */
+               if (g_hash_table_lookup (table, email) == 0) {
+                       g_hash_table_insert (table, (gchar*)email, GINT_TO_POINTER(1));
+                       iter = g_slist_next (iter);
+               } else {
+                       GSList *tmp = g_slist_next (iter);
+                       new_list = g_slist_delete_link (new_list, iter);
+                       iter = tmp;
+               }
+       }
+
+       g_hash_table_unref (table);
+
+       return new_list;
+}
+
+gchar *
+modest_text_utils_get_secure_header (const gchar *value,
+                                    const gchar *header)
+{
+       const gint max_len = 128;
+       gchar *new_value = NULL;
+       gchar *needle = g_strrstr (value, header);
+
+       if (needle && value != needle)
+               new_value = g_strdup (needle + strlen (header));
+
+       if (!new_value)
+               new_value = g_strdup (value);
+
+       /* Do a max length check to prevent DoS attacks caused by huge
+          malformed headers */
+       if (g_utf8_validate (new_value, -1, NULL)) {
+               if (g_utf8_strlen (new_value, -1) > max_len) {
+                       gchar *tmp = g_malloc0 (max_len * 4);
+                       g_utf8_strncpy (tmp, (const gchar *) new_value, max_len);
+                       g_free (new_value);
+                       new_value = tmp;
+               }
+       } else {
+               if (strlen (new_value) > max_len) {
+                       gchar *tmp = g_malloc0 (max_len);
+                       strncpy (new_value, tmp, max_len);
+                       g_free (new_value);
+                       new_value = tmp;
+               }
+       }
+
+       return new_value;
+}