2007-08-04 Johannes Schmid <johannes.schmid@openismus.com>
[modest] / src / modest-text-utils.c
index 96cd573..51e340e 100644 (file)
  */
 
 
  */
 
 
+
+#ifndef _GNU_SOURCE
+#define _GNU_SOURCE
+#endif /*_GNU_SOURCE*/
+#include <string.h> /* for strcasestr */
+
+
 #include <glib.h>
 #include <glib.h>
-#include <string.h>
 #include <stdlib.h>
 #include <glib/gi18n.h>
 #include <regex.h>
 #include <stdlib.h>
 #include <glib/gi18n.h>
 #include <regex.h>
 #define EMPTY_STRING ""
 
 /*
 #define EMPTY_STRING ""
 
 /*
+ * do the hyperlinkification only for texts < 50 Kb,
+ * as it's quite slow. Without this, e.g. mail with
+ * an uuencoded part (which is not recognized as attachment,
+ * will hang modest
+ */
+#define HYPERLINKIFY_MAX_LENGTH (1024*50)
+
+/*
  * we need these regexps to find URLs in plain text e-mails
  */
 typedef struct _url_match_pattern_t url_match_pattern_t;
  * we need these regexps to find URLs in plain text e-mails
  */
 typedef struct _url_match_pattern_t url_match_pattern_t;
@@ -71,7 +85,7 @@ struct _url_match_t {
        { "(file|rtsp|http|ftp|https)://[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]+[-A-Za-z0-9_$%&=?/~#]",\
          NULL, NULL },\
        { "www\\.[-a-z0-9.]+[-a-z0-9](:[0-9]*)?(/[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]*[^]}\\),?!;:\"]?)?",\
        { "(file|rtsp|http|ftp|https)://[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]+[-A-Za-z0-9_$%&=?/~#]",\
          NULL, NULL },\
        { "www\\.[-a-z0-9.]+[-a-z0-9](:[0-9]*)?(/[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]*[^]}\\),?!;:\"]?)?",\
-         NULL, "http://" },\
+                       NULL, "http://" },                              \
        { "ftp\\.[-a-z0-9.]+[-a-z0-9](:[0-9]*)?(/[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]*[^]}\\),?!;:\"]?)?",\
          NULL, "ftp://" },\
        { "(voipto|callto|chatto|jabberto|xmpp):[-_a-z@0-9.\\+]+", \
        { "ftp\\.[-a-z0-9.]+[-a-z0-9](:[0-9]*)?(/[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]*[^]}\\),?!;:\"]?)?",\
          NULL, "ftp://" },\
        { "(voipto|callto|chatto|jabberto|xmpp):[-_a-z@0-9.\\+]+", \
@@ -86,7 +100,6 @@ struct _url_match_t {
 static gchar*   cite                    (const time_t sent_date, const gchar *from);
 static void     hyperlinkify_plain_text (GString *txt);
 static gint     cmp_offsets_reverse     (const url_match_t *match1, const url_match_t *match2);
 static gchar*   cite                    (const time_t sent_date, const gchar *from);
 static void     hyperlinkify_plain_text (GString *txt);
 static gint     cmp_offsets_reverse     (const url_match_t *match1, const url_match_t *match2);
-static void     chk_partial_match       (const url_match_t *match, guint* offset);
 static GSList*  get_url_matches         (GString *txt);
 
 static GString* get_next_line           (const char *b, const gsize blen, const gchar * iter);
 static GSList*  get_url_matches         (GString *txt);
 
 static GString* get_next_line           (const char *b, const gsize blen, const gchar * iter);
@@ -100,11 +113,16 @@ static int      get_breakpoint          (const gchar * s, const gint indent, con
 
 static gchar*   modest_text_utils_quote_plain_text (const gchar *text, 
                                                    const gchar *cite, 
 
 static gchar*   modest_text_utils_quote_plain_text (const gchar *text, 
                                                    const gchar *cite, 
+                                                   const gchar *signature,
+                                                   GList *attachments, 
                                                    int limit);
 
 static gchar*   modest_text_utils_quote_html       (const gchar *text, 
                                                    int limit);
 
 static gchar*   modest_text_utils_quote_html       (const gchar *text, 
-                                                   const gchar *cite, 
+                                                   const gchar *cite,
+                                                   const gchar *signature,
+                                                   GList *attachments,
                                                    int limit);
                                                    int limit);
+static gchar*   get_email_from_address (const gchar *address);
 
 
 /* ******************************************************************* */
 
 
 /* ******************************************************************* */
@@ -114,8 +132,10 @@ static gchar*   modest_text_utils_quote_html       (const gchar *text,
 gchar *
 modest_text_utils_quote (const gchar *text, 
                         const gchar *content_type,
 gchar *
 modest_text_utils_quote (const gchar *text, 
                         const gchar *content_type,
+                        const gchar *signature,
                         const gchar *from,
                         const time_t sent_date, 
                         const gchar *from,
                         const time_t sent_date, 
+                        GList *attachments,
                         int limit)
 {
        gchar *retval, *cited;
                         int limit)
 {
        gchar *retval, *cited;
@@ -128,12 +148,12 @@ modest_text_utils_quote (const gchar *text,
        if (content_type && strcmp (content_type, "text/html") == 0)
                /* TODO: extract the <body> of the HTML and pass it to
                   the function */
        if (content_type && strcmp (content_type, "text/html") == 0)
                /* TODO: extract the <body> of the HTML and pass it to
                   the function */
-               retval = modest_text_utils_quote_html (text, cited, limit);
+               retval = modest_text_utils_quote_html (text, cited, signature, attachments, limit);
        else
        else
-               retval = modest_text_utils_quote_plain_text (text, cited, limit);
+               retval = modest_text_utils_quote_plain_text (text, cited, signature, attachments, limit);
        
        g_free (cited);
        
        g_free (cited);
-
+       
        return retval;
 }
 
        return retval;
 }
 
@@ -141,60 +161,70 @@ modest_text_utils_quote (const gchar *text,
 gchar *
 modest_text_utils_cite (const gchar *text,
                        const gchar *content_type,
 gchar *
 modest_text_utils_cite (const gchar *text,
                        const gchar *content_type,
+                       const gchar *signature,
                        const gchar *from,
                        time_t sent_date)
 {
                        const gchar *from,
                        time_t sent_date)
 {
-       gchar *tmp, *retval;
+       gchar *retval;
+       gchar *tmp_sig;
 
        g_return_val_if_fail (text, NULL);
        g_return_val_if_fail (content_type, NULL);
 
 
        g_return_val_if_fail (text, NULL);
        g_return_val_if_fail (content_type, NULL);
 
-       tmp = cite (sent_date, from);
-       retval = g_strdup_printf ("%s%s\n", tmp, text);
-       g_free (tmp);
+       if (!signature)
+               retval = g_strdup ("");
+       else if (!strcmp(content_type, "text/html")) {
+               tmp_sig = g_strconcat ("\n", signature, NULL);
+               retval = modest_text_utils_convert_to_html_body(tmp_sig);
+               g_free (tmp_sig);
+       } else {
+               retval = g_strconcat ("\n", signature, NULL);
+       }
 
        return retval;
 }
 
 
        return retval;
 }
 
+static gchar *
+forward_cite (const gchar *from,
+                   const gchar *sent,
+                   const gchar *to,
+                   const gchar *subject)
+{
+       return g_strdup_printf ("%s\n%s %s\n%s %s\n%s %s\n%s %s\n", 
+                               FORWARD_STRING, 
+                               FROM_STRING, (from)?from:"",
+                               SENT_STRING, sent,
+                               TO_STRING, (to)?to:"",
+                               SUBJECT_STRING, (subject)?subject:"");
+}
+
 gchar * 
 modest_text_utils_inline (const gchar *text,
                          const gchar *content_type,
 gchar * 
 modest_text_utils_inline (const gchar *text,
                          const gchar *content_type,
+                         const gchar *signature,
                          const gchar *from,
                          time_t sent_date,
                          const gchar *to,
                          const gchar *subject)
 {
        gchar sent_str[101];
                          const gchar *from,
                          time_t sent_date,
                          const gchar *to,
                          const gchar *subject)
 {
        gchar sent_str[101];
-       const gchar *plain_format = "%s\n%s %s\n%s %s\n%s %s\n%s %s\n\n%s";
-       const gchar *html_format = \
-               "%s<br>\n<table width=\"100%\" border=\"0\" cellspacing=\"2\" cellpadding=\"2\">\n" \
-               "<tr><td>%s</td><td>%s</td></tr>\n" \
-               "<tr><td>%s</td><td>%s</td></tr>\n" \
-               "<tr><td>%s</td><td>%s</td></tr>\n" \
-               "<tr><td>%s</td><td>%s</td></tr>\n" \
-               "<br><br>%s";
-       const gchar *format;
-
+       gchar *cited;
+       gchar *retval;
+       
        g_return_val_if_fail (text, NULL);
        g_return_val_if_fail (content_type, NULL);
        g_return_val_if_fail (text, NULL);
        g_return_val_if_fail (content_type, NULL);
-       g_return_val_if_fail (text, NULL);
        
        modest_text_utils_strftime (sent_str, 100, "%c", sent_date);
 
        
        modest_text_utils_strftime (sent_str, 100, "%c", sent_date);
 
-       if (!strcmp (content_type, "text/html"))
-               /* TODO: extract the <body> of the HTML and pass it to
-                  the function */
-               format = html_format;
+       cited = forward_cite (from, sent_str, to, subject);
+       
+       if (content_type && strcmp (content_type, "text/html") == 0)
+               retval = modest_text_utils_quote_html (text, cited, signature, NULL, 80);
        else
        else
-               format = plain_format;
-
-       return g_strdup_printf (format, 
-                               FORWARD_STRING,
-                               FROM_STRING, (from) ? from : EMPTY_STRING,
-                               SENT_STRING, sent_str,
-                               TO_STRING, (to) ? to : EMPTY_STRING,
-                               SUBJECT_STRING, (subject) ? subject : EMPTY_STRING,
-                               text);
+               retval = modest_text_utils_quote_plain_text (text, cited, signature, NULL, 80);
+       
+       g_free (cited);
+       return retval;
 }
 
 /* just to prevent warnings:
 }
 
 /* just to prevent warnings:
@@ -203,14 +233,13 @@ modest_text_utils_inline (const gchar *text,
 gsize
 modest_text_utils_strftime(char *s, gsize max, const char *fmt, time_t timet)
 {
 gsize
 modest_text_utils_strftime(char *s, gsize max, const char *fmt, time_t timet)
 {
-       static GDate date;
+        struct tm tm;
 
        /* does not work on old maemo glib: 
         *   g_date_set_time_t (&date, timet);
         */
 
        /* does not work on old maemo glib: 
         *   g_date_set_time_t (&date, timet);
         */
-       g_date_set_time (&date, (GTime) timet); 
-
-       return g_date_strftime (s, max, fmt, (const GDate*) &date);
+       localtime_r (&timet, &tm);
+       return strftime(s, max, fmt, &tm);
 }
 
 gchar *
 }
 
 gchar *
@@ -219,9 +248,9 @@ modest_text_utils_derived_subject (const gchar *subject, const gchar *prefix)
        gchar *tmp;
 
        g_return_val_if_fail (prefix, NULL);
        gchar *tmp;
 
        g_return_val_if_fail (prefix, NULL);
-       
-       if (!subject)
-               return g_strdup (prefix);
+
+       if (!subject || subject[0] == '\0')
+               subject = _("mail_va_no_subject");
 
        tmp = g_strchug (g_strdup (subject));
 
 
        tmp = g_strchug (g_strdup (subject));
 
@@ -238,15 +267,20 @@ modest_text_utils_remove_address (const gchar *address_list, const gchar *addres
 {
        gchar *dup, *token, *ptr, *result;
        GString *filtered_emails;
 {
        gchar *dup, *token, *ptr, *result;
        GString *filtered_emails;
+       gchar *email_address;
 
        g_return_val_if_fail (address_list, NULL);
 
        if (!address)
                return g_strdup (address_list);
 
        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 */
        
        /* search for substring */
-       if (!strstr ((const char *) address_list, (const char *) address))
+       if (!strstr ((const char *) address_list, (const char *) email_address)) {
+               g_free (email_address);
                return g_strdup (address_list);
                return g_strdup (address_list);
+       }
 
        dup = g_strdup (address_list);
        filtered_emails = g_string_new (NULL);
 
        dup = g_strdup (address_list);
        filtered_emails = g_string_new (NULL);
@@ -255,7 +289,7 @@ modest_text_utils_remove_address (const gchar *address_list, const gchar *addres
 
        while (token != NULL) {
                /* Add to list if not found */
 
        while (token != NULL) {
                /* Add to list if not found */
-               if (!strstr ((const char *) token, (const char *) address)) {
+               if (!strstr ((const char *) token, (const char *) email_address)) {
                        if (filtered_emails->len == 0)
                                g_string_append_printf (filtered_emails, "%s", g_strstrip (token));
                        else
                        if (filtered_emails->len == 0)
                                g_string_append_printf (filtered_emails, "%s", g_strstrip (token));
                        else
@@ -266,71 +300,168 @@ modest_text_utils_remove_address (const gchar *address_list, const gchar *addres
        result = filtered_emails->str;
 
        /* Clean */
        result = filtered_emails->str;
 
        /* Clean */
+       g_free (email_address);
        g_free (dup);
        g_string_free (filtered_emails, FALSE);
 
        return result;
 }
 
        g_free (dup);
        g_string_free (filtered_emails, FALSE);
 
        return result;
 }
 
+static void
+modest_text_utils_convert_buffer_to_html (GString *html, const gchar *data)
+{
+       guint            i;
+       gboolean        space_seen = FALSE;
+       gsize           len;
+       guint           break_dist = 0; /* distance since last break point */
+
+       len = strlen (data);
+
+       /* replace with special html chars where needed*/
+       for (i = 0; i != len; ++i)  {
+               char kar = data[i];
+               
+               if (space_seen && kar != ' ') {
+                       g_string_append_c (html, ' ');
+                       space_seen = FALSE;
+               }
+               
+               /* we artificially insert a breakpoint (newline)
+                * after 256, to make sure our lines are not so long
+                * they will DOS the regexping later
+                */
+               if (++break_dist == 256) {
+                       g_string_append_c (html, '\n');
+                       break_dist = 0;
+               }
+               
+               switch (kar) {
+               case 0:  break; /* ignore embedded \0s */       
+               case '<'  : g_string_append (html, "&lt;");   break;
+               case '>'  : g_string_append (html, "&gt;");   break;
+               case '&'  : g_string_append (html, "&amp;");  break;
+               case '"'  : g_string_append (html, "&quot;");  break;
+               case '\'' : g_string_append (html, "&apos;"); break;
+               case '\n' : g_string_append (html, "<br>\n");              break_dist= 0; break;
+               case '\t' : g_string_append (html, "&nbsp;&nbsp;&nbsp; "); break_dist=0; break; /* note the space at the end*/
+               case ' ':
+                       break_dist = 0;
+                       if (space_seen) { /* second space in a row */
+                               g_string_append (html, "&nbsp; ");
+                               space_seen = FALSE;
+                       } else
+                               space_seen = TRUE;
+                       break;
+               default:
+                       g_string_append_c (html, kar);
+               }
+       }
+}
+
 gchar*
 modest_text_utils_convert_to_html (const gchar *data)
 {
 gchar*
 modest_text_utils_convert_to_html (const gchar *data)
 {
-       guint            i;
-       gboolean         first_space = TRUE;
        GString         *html;      
        gsize           len;
        GString         *html;      
        gsize           len;
-
+       
        if (!data)
                return NULL;
 
        len = strlen (data);
        if (!data)
                return NULL;
 
        len = strlen (data);
-       html = g_string_sized_new (len + 100);  /* just a  guess... */
-       
+       html = g_string_sized_new (1.5 * len);  /* just a  guess... */
+
        g_string_append_printf (html,
        g_string_append_printf (html,
-                               "<html>"
-                               "<head>"
-                               "<meta http-equiv=\"content-type\""
-                               " content=\"text/html; charset=utf8\">"
+                               "<html><head>"
+                               "<meta http-equiv=\"content-type\" content=\"text/html; charset=utf8\">"
                                "</head>"
                                "</head>"
-                               "<body><tt>");
+                               "<body>");
+
+       modest_text_utils_convert_buffer_to_html (html, data);
        
        
-       /* replace with special html chars where needed*/
-       for (i = 0; i != len; ++i)  {
-               char    kar = data[i]; 
-               switch (kar) {
-                       
-               case 0:  break; /* ignore embedded \0s */       
-               case '<' : g_string_append   (html, "&lt;"); break;
-               case '>' : g_string_append   (html, "&gt;"); break;
-               case '&' : g_string_append   (html, "&quot;"); break;
-               case '\n': g_string_append   (html, "<br>\n"); break;
-               default:
-                       if (kar == ' ') {
-                               g_string_append (html, first_space ? " " : "&nbsp;");
-                               first_space = FALSE;
-                       } else  if (kar == '\t')
-                               g_string_append (html, "&nbsp; &nbsp;&nbsp;");
-                       else {
-                               int charnum = 0;
-                               first_space = TRUE;
-                               /* optimization trick: accumulate 'normal' chars, then copy */
-                               do {
-                                       kar = data [++charnum + i];
-                                       
-                               } while ((i + charnum < len) &&
-                                        (kar > '>' || (kar != '<' && kar != '>'
-                                                       && kar != '&' && kar !=  ' '
-                                                       && kar != '\n' && kar != '\t')));
-                               g_string_append_len (html, &data[i], charnum);
-                               i += (charnum  - 1);
+       g_string_append (html, "</body></html>");
+
+       if (len <= HYPERLINKIFY_MAX_LENGTH)
+               hyperlinkify_plain_text (html);
+
+       return g_string_free (html, FALSE);
+}
+
+gchar *
+modest_text_utils_convert_to_html_body (const gchar *data)
+{
+       GString         *html;      
+       gsize           len;
+       
+       if (!data)
+               return NULL;
+
+       len = strlen (data);
+       html = g_string_sized_new (1.5 * len);  /* just a  guess... */
+
+       modest_text_utils_convert_buffer_to_html (html, data);
+
+       if (len < HYPERLINKIFY_MAX_LENGTH)
+               hyperlinkify_plain_text (html);
+
+       return g_string_free (html, FALSE);
+}
+
+void
+modest_text_utils_get_addresses_indexes (const gchar *addresses, GSList **start_indexes, GSList **end_indexes)
+{
+       gchar *current, *start, *last_blank;
+       gint start_offset = 0, current_offset = 0;
+
+       g_return_if_fail (start_indexes != NULL);
+       g_return_if_fail (end_indexes != NULL);
+
+       start = (gchar *) addresses;
+       current = start;
+       last_blank = start;
+
+       while (*current != '\0') {
+               if ((start == current)&&((*current == ' ')||(*current == ',')||(*current == ';'))) {
+                       start = g_utf8_next_char (start);
+                       start_offset++;
+                       last_blank = current;
+               } else if ((*current == ',')||(*current == ';')) {
+                       gint *start_index, *end_index;
+                       start_index = g_new0(gint, 1);
+                       end_index = g_new0(gint, 1);
+                       *start_index = start_offset;
+                       *end_index = current_offset;
+                       *start_indexes = g_slist_prepend (*start_indexes, start_index);
+                       *end_indexes = g_slist_prepend (*end_indexes, end_index);
+                       start = g_utf8_next_char (current);
+                       start_offset = current_offset + 1;
+                       last_blank = start;
+               } else if (*current == '"') {
+                       current = g_utf8_next_char (current);
+                       current_offset ++;
+                       while ((*current != '"')&&(*current != '\0')) {
+                               current = g_utf8_next_char (current);
+                               current_offset ++;
                        }
                }
                        }
                }
+                               
+               current = g_utf8_next_char (current);
+               current_offset ++;
+       }
+
+       if (start != current) {
+                       gint *start_index, *end_index;
+                       start_index = g_new0(gint, 1);
+                       end_index = g_new0(gint, 1);
+                       *start_index = start_offset;
+                       *end_index = current_offset;
+                       *start_indexes = g_slist_prepend (*start_indexes, start_index);
+                       *end_indexes = g_slist_prepend (*end_indexes, end_index);
        }
        
        }
        
-       g_string_append (html, "</tt></body></html>");
-       hyperlinkify_plain_text (html);
+       *start_indexes = g_slist_reverse (*start_indexes);
+       *end_indexes = g_slist_reverse (*end_indexes);
 
 
-       return g_string_free (html, FALSE);
+       return;
 }
 
 GSList *
 }
 
 GSList *
@@ -344,25 +475,25 @@ modest_text_utils_split_addresses_list (const gchar *addresses)
        last_blank = start;
 
        while (*current != '\0') {
        last_blank = start;
 
        while (*current != '\0') {
-               if ((start == current)&&((*current == ' ')||(*current == ','))) {
-                       start++;
+               if ((start == current)&&((*current == ' ')||(*current == ',')||(*current == ';'))) {
+                       start = g_utf8_next_char (start);
                        last_blank = current;
                        last_blank = current;
-               } else if (*current == ',') {
+               } else if ((*current == ',')||(*current == ';')) {
                        gchar *new_address = NULL;
                        new_address = g_strndup (start, current - last_blank);
                        result = g_slist_prepend (result, new_address);
                        gchar *new_address = NULL;
                        new_address = g_strndup (start, current - last_blank);
                        result = g_slist_prepend (result, new_address);
-                       start = current + 1;
+                       start = g_utf8_next_char (current);
                        last_blank = start;
                } else if (*current == '\"') {
                        if (current == start) {
                        last_blank = start;
                } else if (*current == '\"') {
                        if (current == start) {
-                               current++;
-                               start++;
+                               current = g_utf8_next_char (current);
+                               start = g_utf8_next_char (start);
                        }
                        while ((*current != '\"')&&(*current != '\0'))
                        }
                        while ((*current != '\"')&&(*current != '\0'))
-                               current++;
+                               current = g_utf8_next_char (current);
                }
                                
                }
                                
-               current++;
+               current = g_utf8_next_char (current);
        }
 
        if (start != current) {
        }
 
        if (start != current) {
@@ -564,28 +695,40 @@ get_breakpoint (const gchar * s, const gint indent, const gint limit)
 static gchar *
 cite (const time_t sent_date, const gchar *from)
 {
 static gchar *
 cite (const time_t sent_date, const gchar *from)
 {
-       gchar sent_str[101];
-
-       /* format sent_date */
-       modest_text_utils_strftime (sent_str, 100, "%c", sent_date);
-       return g_strdup_printf (N_("On %s, %s wrote:\n"), 
-                               sent_str, 
-                               (from) ? from : EMPTY_STRING);
+       return g_strdup (_("mcen_ia_editor_original_message"));
 }
 
 }
 
+static gchar *
+quoted_attachments (GList *attachments)
+{
+       GList *node = NULL;
+       GString *result = g_string_new ("");
+       for (node = attachments; node != NULL; node = g_list_next (node)) {
+               gchar *filename = (gchar *) node->data;
+               g_string_append_printf ( result, "%s %s\n", _("mcen_ia_editor_attach_filename"), filename);
+       }
+
+       return g_string_free (result, FALSE);
+
+}
 
 static gchar *
 modest_text_utils_quote_plain_text (const gchar *text, 
                                    const gchar *cite, 
 
 static gchar *
 modest_text_utils_quote_plain_text (const gchar *text, 
                                    const gchar *cite, 
+                                   const gchar *signature,
+                                   GList *attachments,
                                    int limit)
 {
        const gchar *iter;
        gint indent, breakpoint, rem_indent = 0;
        GString *q, *l, *remaining;
        gsize len;
                                    int limit)
 {
        const gchar *iter;
        gint indent, breakpoint, rem_indent = 0;
        GString *q, *l, *remaining;
        gsize len;
+       gchar *attachments_string = NULL;
 
        /* remaining will store the rest of the line if we have to break it */
 
        /* remaining will store the rest of the line if we have to break it */
-       q = g_string_new (cite);
+       q = g_string_new ("\n");
+       q = g_string_append (q, cite);
+       q = g_string_append_c (q, '\n');
        remaining = g_string_new ("");
 
        iter = text;
        remaining = g_string_new ("");
 
        iter = text;
@@ -628,24 +771,56 @@ modest_text_utils_quote_plain_text (const gchar *text,
                g_string_free (l, TRUE);
        } while ((iter < text + len) || (remaining->str[0]));
 
                g_string_free (l, TRUE);
        } while ((iter < text + len) || (remaining->str[0]));
 
+       attachments_string = quoted_attachments (attachments);
+       q = g_string_append (q, attachments_string);
+       g_free (attachments_string);
+
+       if (signature != NULL) {
+               q = g_string_append_c (q, '\n');
+               q = g_string_append (q, signature);
+       }
+
        return g_string_free (q, FALSE);
 }
 
 static gchar*
 modest_text_utils_quote_html (const gchar *text, 
                              const gchar *cite, 
        return g_string_free (q, FALSE);
 }
 
 static gchar*
 modest_text_utils_quote_html (const gchar *text, 
                              const gchar *cite, 
+                             const gchar *signature,
+                             GList *attachments,
                              int limit)
 {
                              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" \
        const gchar *format = \
                "<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\">\n" \
                "<html>\n" \
                "<body>\n" \
-               "%s" \
-               "<blockquote type=\"cite\">\n%s\n</blockquote>\n" \
+               "<br/>%s<br/>" \
+               "<pre>%s<br/>%s<br/>%s</pre>\n" \
                "</body>\n" \
                "</html>\n";
                "</body>\n" \
                "</html>\n";
+       gchar *attachments_string = NULL;
+       gchar *q_attachments_string = NULL;
+       gchar *q_cite = NULL;
+       gchar *html_text = NULL;
 
 
-       return g_strdup_printf (format, cite, text);
+       if (signature == NULL)
+               signature_result = g_strdup ("");
+       else
+               signature_result = modest_text_utils_convert_to_html_body (signature);
+
+       attachments_string = quoted_attachments (attachments);
+       q_attachments_string = modest_text_utils_convert_to_html_body (attachments_string);
+       q_cite = modest_text_utils_convert_to_html_body (cite);
+       html_text = modest_text_utils_convert_to_html_body (text);
+       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;
 }
 
 static gint 
 }
 
 static gint 
@@ -655,16 +830,6 @@ cmp_offsets_reverse (const url_match_t *match1, const url_match_t *match2)
 }
 
 
 }
 
 
-
-/*
- * check if the match is inside an existing match... */
-static void
-chk_partial_match (const url_match_t *match, guint* offset)
-{
-       if (*offset >= match->offset && *offset < match->offset + match->len)
-               *offset = -1;
-}
-
 static GSList*
 get_url_matches (GString *txt)
 {
 static GSList*
 get_url_matches (GString *txt)
 {
@@ -687,28 +852,39 @@ get_url_matches (GString *txt)
        for (i = 0; i != pattern_num; ++i) {
                offset     = 0; 
                while (1) {
        for (i = 0; i != pattern_num; ++i) {
                offset     = 0; 
                while (1) {
-                       int test_offset;
+                       url_match_t *match;
+                       gboolean is_submatch;
+                       GSList *cursor;
+                       
                        if ((rv = regexec (patterns[i].preg, txt->str + offset, 1, &rm, 0)) != 0) {
                                g_return_val_if_fail (rv == REG_NOMATCH, NULL); /* this should not happen */
                                break; /* try next regexp */ 
                        }
                        if (rm.rm_so == -1)
                                break;
                        if ((rv = regexec (patterns[i].preg, txt->str + offset, 1, &rm, 0)) != 0) {
                                g_return_val_if_fail (rv == REG_NOMATCH, NULL); /* this should not happen */
                                break; /* try next regexp */ 
                        }
                        if (rm.rm_so == -1)
                                break;
-
-                       /* FIXME: optimize this */
-                       /* to avoid partial matches on something that was already found... */
-                       /* check_partial_match will put -1 in the data ptr if that is the case */
-                       test_offset = offset + rm.rm_so;
-                       g_slist_foreach (match_list, (GFunc)chk_partial_match, &test_offset);
                        
                        
-                       /* make a list of our matches (<offset, len, prefix> tupels)*/
-                       if (test_offset != -1) {
-                               url_match_t *match = g_slice_new (url_match_t);
+                       is_submatch = FALSE;
+                       /* check  old matches to see if this has already been matched */
+                       cursor = match_list;
+                       while (cursor && !is_submatch) {
+                               const url_match_t *old_match =
+                                       (const url_match_t *) cursor->data;
+                               guint new_offset = offset + rm.rm_so;
+                               is_submatch = (new_offset >  old_match->offset &&
+                                              new_offset <  old_match->offset + old_match->len);
+                               cursor = g_slist_next (cursor);
+                       }
+
+                       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->len    = rm.rm_eo - rm.rm_so;
                                match->prefix = patterns[i].prefix;
                                match->offset = offset + rm.rm_so;
                                match->len    = rm.rm_eo - rm.rm_so;
                                match->prefix = patterns[i].prefix;
+                               g_warning ("<%d, %d, %s>",  match->offset, match->len, match->prefix);
                                match_list = g_slist_prepend (match_list, match);
                        }
                                match_list = g_slist_prepend (match_list, match);
                        }
+                               
                        offset += rm.rm_eo;
                }
        }
                        offset += rm.rm_eo;
                }
        }
@@ -770,9 +946,9 @@ modest_text_utils_get_display_address (gchar *address)
        
        if (!address)
                return NULL;
        
        if (!address)
                return NULL;
-
+       
        g_return_val_if_fail (g_utf8_validate (address, -1, NULL), NULL);
        g_return_val_if_fail (g_utf8_validate (address, -1, NULL), NULL);
-
+       
        g_strchug (address); /* remove leading whitespace */
 
        /*  <email@address> from display name */
        g_strchug (address); /* remove leading whitespace */
 
        /*  <email@address> from display name */
@@ -794,17 +970,37 @@ modest_text_utils_get_display_address (gchar *address)
        return address;
 }
 
        return address;
 }
 
+gchar *
+modest_text_utils_get_email_address (const gchar *full_address)
+{
+       const gchar *left, *right;
+       
+       if (!full_address)
+               return NULL;
+       
+       g_return_val_if_fail (g_utf8_validate (full_address, -1, NULL), NULL);
+       
+       left = g_strrstr_len (full_address, strlen(full_address), "<");
+       if (left == NULL)
+               return g_strdup (full_address);
+
+       right = g_strstr_len (left, strlen(left), ">");
+       if (right == NULL)
+               return g_strdup (full_address);
 
 
+       return g_strndup (left + 1, right - left - 1);
+}
 
 gint 
 modest_text_utils_get_subject_prefix_len (const gchar *sub)
 {
        gint i;
        static const gchar* prefix[] = {
 
 gint 
 modest_text_utils_get_subject_prefix_len (const gchar *sub)
 {
        gint i;
        static const gchar* prefix[] = {
-               "Re:", "RE:", "Fwd:", "FWD:", "FW:", NULL
+               "Re:", "RE:", "RV:", "re:"
+               "Fwd:", "FWD:", "FW:", "fwd:", "Fw:", "fw:", NULL
        };
                
        };
                
-       if (!sub || (sub[0] != 'R' && sub[0] != 'F')) /* optimization */
+       if (!sub || (sub[0] != 'R' && sub[0] != 'F' && sub[0] != 'r' && sub[0] != 'f')) /* optimization */
                return 0;
 
        i = 0;
                return 0;
 
        i = 0;
@@ -852,29 +1048,123 @@ gchar*
 modest_text_utils_get_display_date (time_t date)
 {
        time_t now;
 modest_text_utils_get_display_date (time_t date)
 {
        time_t now;
-       const guint BUF_SIZE = 64; 
+       static const guint BUF_SIZE = 64; 
+       static const guint ONE_DAY = 24 * 60 * 60; /* seconds in one day */
        gchar date_buf[BUF_SIZE];  
        gchar date_buf[BUF_SIZE];  
-       gchar now_buf [BUF_SIZE];  
-       
+       gchar today_buf [BUF_SIZE];  
+
+       modest_text_utils_strftime (date_buf, BUF_SIZE, "%x", date); 
+
        now = time (NULL);
 
        now = time (NULL);
 
-       modest_text_utils_strftime (date_buf, BUF_SIZE, "%d/%m/%Y", date);
-       modest_text_utils_strftime (now_buf,  BUF_SIZE, "%d/%m/%Y",  now); /* today */
-       
-       /* if this is today, get the time instead of the date */
-       if (strcmp (date_buf, now_buf) == 0)
-               modest_text_utils_strftime (date_buf, BUF_SIZE, "%H:%M:%P", date);
+       /* we check if the date is within the last 24h, if not, we don't
+        * have to do the extra, expensive strftime, which was very visible
+        * in the profiles.
+        */
+       if (abs(now - date) < ONE_DAY) {
+               
+               /* it's within the last 24 hours, but double check */
+               /* use the localized dates */
+               modest_text_utils_strftime (today_buf,  BUF_SIZE, "%x", now); 
+
+               /* if it's today, use the time instead */
+               if (strcmp (date_buf, today_buf) == 0)
+                       modest_text_utils_strftime (date_buf, BUF_SIZE, "%X", date);
+       }
        
        return g_strdup(date_buf);
 }
 
        
        return g_strdup(date_buf);
 }
 
-gboolean 
-modest_text_utils_validate_email_address (const gchar *email_address)
+
+gboolean
+modest_text_utils_validate_folder_name (const gchar *folder_name)
+{
+       /* based on http://msdn2.microsoft.com/en-us/library/aa365247.aspx,
+        * with some extras */
+       
+       guint len;
+       const gchar **cursor;
+       const gchar *forbidden_chars[] = {
+               "<", ">", ":", "\"", "/", "\\", "|", "?", "*", "^", "%", "$", NULL
+       };
+       const gchar *forbidden_names[] = { /* windows does not like these */
+               "CON", "PRN", "AUX", "NUL", "COM1", "COM2", "COM3", "COM4", "COM5", "COM6",
+               "COM7", "COM8", "COM9", "LPT1", "LPT2", "LPT3", "LPT4", "LPT5", "LPT6", "LPT7", "LPT8", "LPT9",
+               ".", "..", NULL
+       };
+       
+       /* cannot be NULL */
+       if (!folder_name) 
+               return FALSE;
+
+       /* cannot be empty */
+       len = strlen(folder_name);
+       if (len == 0)
+               return FALSE;
+       
+       /* cannot start or end with a space */
+       if (g_ascii_isspace(folder_name[0]) || g_ascii_isspace(folder_name[len - 1]))
+               return FALSE; 
+
+       /* cannot contain a forbidden char */
+       for (cursor = forbidden_chars; cursor && *cursor; ++cursor)
+               if (strstr(folder_name, *cursor) != NULL)
+                       return FALSE;
+       
+       /* cannot contain a forbidden word */
+       if (len <= 4) {
+               for (cursor = forbidden_names; cursor && *cursor; ++cursor) {
+                       if (g_ascii_strcasecmp (folder_name, *cursor) == 0)
+                               return FALSE;
+               }
+       }
+       return TRUE; /* it's valid! */
+}
+
+
+
+gboolean
+modest_text_utils_validate_domain_name (const gchar *domain)
+{
+       gboolean valid = FALSE;
+       regex_t rx;
+       const gchar* domain_regex = "^[a-z0-9]([.]?[a-z0-9-])*[a-z0-9]$";
+
+       if (!domain)
+               return FALSE;
+       
+       /* domain name: all alphanum or '-' or '.',
+        * but beginning/ending in alphanum */  
+       if (regcomp (&rx, domain_regex, REG_ICASE|REG_EXTENDED|REG_NOSUB)) {
+               g_warning ("BUG: error in regexp");
+               return FALSE;
+       }
+       
+       valid = (regexec (&rx, domain, 1, NULL, 0) == 0);
+       regfree (&rx);
+               
+       return valid;
+}
+
+
+
+gboolean
+modest_text_utils_validate_email_address (const gchar *email_address, const gchar **invalid_char_position)
 {
        int count = 0;
        const gchar *c = NULL, *domain = NULL;
 {
        int count = 0;
        const gchar *c = NULL, *domain = NULL;
-       static gchar *rfc822_specials = "()<>@,;:\\\"[]";
+       static gchar *rfc822_specials = "()<>@,;:\\\"[]&";
 
 
+       if (invalid_char_position != NULL)
+               *invalid_char_position = NULL;
+       
+       /* check that the email adress contains exactly one @ */
+       if (!strstr(email_address, "@") || 
+                       (strstr(email_address, "@") != g_strrstr(email_address, "@")))
+       {
+               return FALSE;
+       }
+       
        /* first we validate the name portion (name@domain) */
        for (c = email_address;  *c;  c++) {
                if (*c == '\"' && 
        /* first we validate the name portion (name@domain) */
        for (c = email_address;  *c;  c++) {
                if (*c == '\"' && 
@@ -901,8 +1191,11 @@ modest_text_utils_validate_email_address (const gchar *email_address)
                        break;
                if (*c <= ' ' || *c >= 127) 
                        return FALSE;
                        break;
                if (*c <= ' ' || *c >= 127) 
                        return FALSE;
-               if (strchr(rfc822_specials, *c)) 
+               if (strchr(rfc822_specials, *c)) {
+                       if (invalid_char_position)
+                               *invalid_char_position = c;
                        return FALSE;
                        return FALSE;
+               }
        }
        if (c == email_address || *(c - 1) == '.') 
                return FALSE;
        }
        if (c == email_address || *(c - 1) == '.') 
                return FALSE;
@@ -912,38 +1205,177 @@ modest_text_utils_validate_email_address (const gchar *email_address)
                return FALSE;
        do {
                if (*c == '.') {
                return FALSE;
        do {
                if (*c == '.') {
-                       if (c == domain || *(c - 1) == '.') 
+                       if (c == domain || *(c - 1) == '.' || *(c + 1) == '\0') 
                                return FALSE;
                        count++;
                }
                if (*c <= ' ' || *c >= 127) 
                        return FALSE;
                                return FALSE;
                        count++;
                }
                if (*c <= ' ' || *c >= 127) 
                        return FALSE;
-               if (strchr(rfc822_specials, *c)) 
+               if (strchr(rfc822_specials, *c)) {
+                       if (invalid_char_position)
+                               *invalid_char_position = c;
                        return FALSE;
                        return FALSE;
+               }
        } while (*++c);
 
        return (count >= 1) ? TRUE : FALSE;
 }
 
        } while (*++c);
 
        return (count >= 1) ? TRUE : FALSE;
 }
 
+gboolean 
+modest_text_utils_validate_recipient (const gchar *recipient, const gchar **invalid_char_position)
+{
+       gchar *stripped, *current;
+       gchar *right_part;
+       gboolean has_error = FALSE;
+
+       if (modest_text_utils_validate_email_address (recipient, invalid_char_position))
+               return TRUE;
+       stripped = g_strdup (recipient);
+       stripped = g_strstrip (stripped);
+       current = stripped;
+
+       if (*current == '\0') {
+               g_free (stripped);
+               return FALSE;
+       }
+
+       /* quoted string */
+       if (*current == '\"') {
+               current = g_utf8_next_char (current);
+               has_error = TRUE;
+               for (; *current != '\0'; current = g_utf8_next_char (current)) {
+                       if (*current == '\\') {
+                               /* TODO: This causes a warning, which breaks the build, 
+                                * because a gchar cannot be < 0.
+                                * murrayc. 
+                               if (current[1] <0) {
+                                       has_error = TRUE;
+                                       break;
+                               }
+                               */
+                       } else if (*current == '\"') {
+                               has_error = FALSE;
+                               current = g_utf8_next_char (current);
+                               break;
+                       }
+               }
+       } else {
+               has_error = TRUE;
+               for (current = stripped ; *current != '\0'; current = g_utf8_next_char (current)) {
+                       if (*current == '<') {
+                               has_error = FALSE;
+                               break;
+                       }
+               }
+       }
+               
+       if (has_error) {
+               g_free (stripped);
+               return FALSE;
+       }
 
 
+       right_part = g_strdup (current);
+       g_free (stripped);
+       right_part = g_strstrip (right_part);
+
+       if (g_str_has_prefix (right_part, "<") &&
+           g_str_has_suffix (right_part, ">")) {
+               gchar *address;
+               gboolean valid;
+
+               address = g_strndup (right_part+1, strlen (right_part) - 2);
+               g_free (right_part);
+               valid = modest_text_utils_validate_email_address (address, invalid_char_position);
+               g_free (address);
+               return valid;
+       } else {
+               g_free (right_part);
+               return FALSE;
+       }
+}
 
 
 gchar *
 
 
 gchar *
-modest_text_utils_get_display_size (guint size)
+modest_text_utils_get_display_size (guint64 size)
 {
        const guint KB=1024;
        const guint MB=1024 * KB;
        const guint GB=1024 * MB;
 {
        const guint KB=1024;
        const guint MB=1024 * KB;
        const guint GB=1024 * MB;
-       const guint TB=1024 * GB;
-
-       if (size < KB)
-               return g_strdup_printf (_("%0.1f Kb"), (double)size / KB);
-       else if (size < MB)
-               return g_strdup_printf (_("%d Kb"), size / KB);
-       else if (size < GB)
-               return g_strdup_printf (_("%d Mb"), size / MB);
-       else if (size < TB)
-               return g_strdup_printf (_("%d Gb"), size/ GB);
+
+       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);
+       else if (KB <= size && size < 100 * KB)
+               return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), size / KB);
+       else if (100*KB <= size && size < MB)
+               return g_strdup_printf (_FM("sfil_li_size_100kb_1mb"), (float) size / MB);
+       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);
        else
        else
-               return g_strdup_printf (_("Very big"));
+               return g_strdup_printf (_FM("sfil_li_size_1gb_or_greater"), (float) size / GB); 
+}
+
+static gchar *
+get_email_from_address (const gchar * address)
+{
+       gchar *left_limit, *right_limit;
+
+       left_limit = strstr (address, "<");
+       right_limit = g_strrstr (address, ">");
+
+       if ((left_limit == NULL)||(right_limit == NULL)|| (left_limit > right_limit))
+               return g_strdup (address);
+       else
+               return g_strndup (left_limit + 1, (right_limit - left_limit) - 1);
+}
+
+gchar *      
+modest_text_utils_get_color_string (GdkColor *color)
+{
+
+       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,
+                               (color->green >> 12) & 0xf, (color->green >> 8) & 0xf,
+                               (color->green >>  4) & 0xf, (color->green)      & 0xf,
+                               (color->blue >> 12)  & 0xf, (color->blue >> 8)  & 0xf,
+                               (color->blue >>  4)  & 0xf, (color->blue)       & 0xf);
+}
+
+gchar *
+modest_text_utils_text_buffer_get_text (GtkTextBuffer *buffer)
+{
+       GtkTextIter start, end;
+       gchar *slice, *current;
+       GString *result = g_string_new ("");
+
+       g_return_val_if_fail (GTK_IS_TEXT_BUFFER (buffer), NULL);
+
+       gtk_text_buffer_get_start_iter (buffer, &start);
+       gtk_text_buffer_get_end_iter (buffer, &end);
+
+       slice = gtk_text_buffer_get_slice (buffer, &start, &end, FALSE);
+       current = slice;
+
+       while (current && current != '\0') {
+               if (g_utf8_get_char (current) == 0xFFFC) {
+                       result = g_string_append_c (result, ' ');
+                       current = g_utf8_next_char (current);
+               } else {
+                       gchar *next = g_utf8_strchr (current, -1, 0xFFFC);
+                       if (next == NULL) {
+                               result = g_string_append (result, current);
+                       } else {
+                               result = g_string_append_len (result, current, next - current);
+                       }
+                       current = next;
+               }
+       }
+       g_free (slice);
+
+       return g_string_free (result, FALSE);
+       
 }
 }