git.maemo.org Git - modest/blob - src/modest-text-utils.c

   1 /* Copyright (c) 2006, Nokia Corporation
   2  * All rights reserved.
   3  *
   4  * Redistribution and use in source and binary forms, with or without
   5  * modification, are permitted provided that the following conditions are
   6  * met:
   7  *
   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.
  16  *
  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.
  28  */
  29
  30
  31
  32 #ifndef _GNU_SOURCE
  33 #define _GNU_SOURCE
  34 #endif /*_GNU_SOURCE*/
  35 #include <string.h> /* for strcasestr */
  36
  37
  38 #include <glib.h>
  39 #include <stdlib.h>
  40 #include <glib/gi18n.h>
  41 #include <regex.h>
  42 #include <modest-tny-platform-factory.h>
  43 #include <modest-text-utils.h>
  44 #include <modest-runtime.h>
  45
  46
  47 #ifdef HAVE_CONFIG_H
  48 #include <config.h>
  49 #endif /*HAVE_CONFIG_H */
  50
  51 /* defines */
  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 ""
  58
  59 /*
  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,
  63  * will hang modest
  64  */
  65 #define HYPERLINKIFY_MAX_LENGTH (1024*50)
  66
  67 /*
  68  * we need these regexps to find URLs in plain text e-mails
  69  */
  70 typedef struct _url_match_pattern_t url_match_pattern_t;
  71 struct _url_match_pattern_t {
  72         gchar   *regex;
  73         regex_t *preg;
  74         gchar   *prefix;
  75 };
  76
  77 typedef struct _url_match_t url_match_t;
  78 struct _url_match_t {
  79         guint offset;
  80         guint len;
  81         const gchar* prefix;
  82 };
  83
  84 #define MAIL_VIEWER_URL_MATCH_PATTERNS  {                               \
  85         { "(file|rtsp|http|ftp|https)://[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]+[-A-Za-z0-9_$%&=?/~#]",\
  86           NULL, NULL },\
  87         { "www\\.[-a-z0-9.]+[-a-z0-9](:[0-9]*)?(/[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]*[^]}\\),?!;:\"]?)?",\
  88                         NULL, "http://" },                              \
  89         { "ftp\\.[-a-z0-9.]+[-a-z0-9](:[0-9]*)?(/[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]*[^]}\\),?!;:\"]?)?",\
  90           NULL, "ftp://" },\
  91         { "(voipto|callto|chatto|jabberto|xmpp):[-_a-z@0-9.\\+]+", \
  92            NULL, NULL},                                             \
  93         { "mailto:[-_a-z0-9.\\+]+@[-_a-z0-9.]+",                    \
  94           NULL, NULL},\
  95         { "[-_a-z0-9.\\+]+@[-_a-z0-9.]+",\
  96           NULL, "mailto:"}\
  97         }
  98
  99 const gchar account_title_forbidden_chars[] = {
 100         '\\', '/', ':', '*', '?', '\'', '<', '>', '|', '^'
 101 };
 102 const gchar folder_name_forbidden_chars[] = {
 103         '<', '>', ':', '\'', '/', '\\', '|', '?', '*', '^', '%', '$'
 104 };
 105 const gchar user_name_forbidden_chars[] = {
 106         '<', '>'
 107 };
 108 const guint ACCOUNT_TITLE_FORBIDDEN_CHARS_LENGTH = G_N_ELEMENTS (account_title_forbidden_chars);
 109 const guint FOLDER_NAME_FORBIDDEN_CHARS_LENGTH = G_N_ELEMENTS (folder_name_forbidden_chars);
 110 const guint USER_NAME_FORBIDDEN_CHARS_LENGTH = G_N_ELEMENTS (user_name_forbidden_chars);
 111
 112 /* private */
 113 static gchar*   cite                    (const time_t sent_date, const gchar *from);
 114 static void     hyperlinkify_plain_text (GString *txt);
 115 static gint     cmp_offsets_reverse     (const url_match_t *match1, const url_match_t *match2);
 116 static GSList*  get_url_matches         (GString *txt);
 117
 118 static GString* get_next_line           (const char *b, const gsize blen, const gchar * iter);
 119 static int      get_indent_level        (const char *l);
 120 static void     unquote_line            (GString * l);
 121 static void     append_quoted           (GString * buf, const int indent, const GString * str,
 122                                          const int cutpoint);
 123 static int      get_breakpoint_utf8     (const gchar * s, const gint indent, const gint limit);
 124 static int      get_breakpoint_ascii    (const gchar * s, const gint indent, const gint limit);
 125 static int      get_breakpoint          (const gchar * s, const gint indent, const gint limit);
 126
 127 static gchar*   modest_text_utils_quote_plain_text (const gchar *text,
 128                                                     const gchar *cite,
 129                                                     const gchar *signature,
 130                                                     GList *attachments,
 131                                                     int limit);
 132
 133 static gchar*   modest_text_utils_quote_html       (const gchar *text,
 134                                                     const gchar *cite,
 135                                                     const gchar *signature,
 136                                                     GList *attachments,
 137                                                     int limit);
 138 static gchar*   get_email_from_address (const gchar *address);
 139
 140
 141 /* ******************************************************************* */
 142 /* ************************* PUBLIC FUNCTIONS ************************ */
 143 /* ******************************************************************* */
 144
 145 gchar *
 146 modest_text_utils_quote (const gchar *text,
 147                          const gchar *content_type,
 148                          const gchar *signature,
 149                          const gchar *from,
 150                          const time_t sent_date,
 151                          GList *attachments,
 152                          int limit)
 153 {
 154         gchar *retval, *cited;
 155
 156         g_return_val_if_fail (text, NULL);
 157         g_return_val_if_fail (content_type, NULL);
 158
 159         cited = cite (sent_date, from);
 160
 161         if (content_type && strcmp (content_type, "text/html") == 0)
 162                 /* TODO: extract the <body> of the HTML and pass it to
 163                    the function */
 164                 retval = modest_text_utils_quote_html (text, cited, signature, attachments, limit);
 165         else
 166                 retval = modest_text_utils_quote_plain_text (text, cited, signature, attachments, limit);
 167
 168         g_free (cited);
 169
 170         return retval;
 171 }
 172
 173
 174 gchar *
 175 modest_text_utils_cite (const gchar *text,
 176                         const gchar *content_type,
 177                         const gchar *signature,
 178                         const gchar *from,
 179                         time_t sent_date)
 180 {
 181         gchar *retval;
 182         gchar *tmp_sig;
 183
 184         g_return_val_if_fail (text, NULL);
 185         g_return_val_if_fail (content_type, NULL);
 186
 187         if (!signature)
 188                 retval = g_strdup ("");
 189         else if (strcmp(content_type, "text/html") == 0) {
 190                 tmp_sig = g_strconcat ("\n", signature, NULL);
 191                 retval = modest_text_utils_convert_to_html_body(tmp_sig);
 192                 g_free (tmp_sig);
 193         } else {
 194                 retval = g_strconcat (text, "\n", signature, NULL);
 195         }
 196
 197         return retval;
 198 }
 199
 200 static gchar *
 201 forward_cite (const gchar *from,
 202               const gchar *sent,
 203               const gchar *to,
 204               const gchar *subject)
 205 {
 206         return g_strdup_printf ("%s\n%s %s\n%s %s\n%s %s\n%s %s\n",
 207                                 FORWARD_STRING,
 208                                 FROM_STRING, (from)?from:"",
 209                                 SENT_STRING, sent,
 210                                 TO_STRING, (to)?to:"",
 211                                 SUBJECT_STRING, (subject)?subject:"");
 212 }
 213
 214 gchar *
 215 modest_text_utils_inline (const gchar *text,
 216                           const gchar *content_type,
 217                           const gchar *signature,
 218                           const gchar *from,
 219                           time_t sent_date,
 220                           const gchar *to,
 221                           const gchar *subject)
 222 {
 223         gchar sent_str[101];
 224         gchar *cited;
 225         gchar *retval;
 226
 227         g_return_val_if_fail (text, NULL);
 228         g_return_val_if_fail (content_type, NULL);
 229
 230         modest_text_utils_strftime (sent_str, 100, "%c", sent_date);
 231
 232         cited = forward_cite (from, sent_str, to, subject);
 233
 234         if (content_type && strcmp (content_type, "text/html") == 0)
 235                 retval = modest_text_utils_quote_html (text, cited, signature, NULL, 80);
 236         else
 237                 retval = modest_text_utils_quote_plain_text (text, cited, signature, NULL, 80);
 238
 239         g_free (cited);
 240         return retval;
 241 }
 242
 243 /* just to prevent warnings:
 244  * warning: `%x' yields only last 2 digits of year in some locales
 245  */
 246 gsize
 247 modest_text_utils_strftime(char *s, gsize max, const char *fmt, time_t timet)
 248 {
 249         struct tm tm;
 250
 251         /* does not work on old maemo glib:
 252          *   g_date_set_time_t (&date, timet);
 253          */
 254         localtime_r (&timet, &tm);
 255         return strftime(s, max, fmt, &tm);
 256 }
 257
 258 gchar *
 259 modest_text_utils_derived_subject (const gchar *subject, const gchar *prefix)
 260 {
 261         gchar *tmp;
 262
 263         g_return_val_if_fail (prefix, NULL);
 264
 265         if (!subject || subject[0] == '\0')
 266                 subject = _("mail_va_no_subject");
 267
 268         tmp = g_strchug (g_strdup (subject));
 269
 270         if (!strncmp (tmp, prefix, strlen (prefix))) {
 271                 return tmp;
 272         } else {
 273                 g_free (tmp);
 274                 return g_strdup_printf ("%s %s", prefix, subject);
 275         }
 276 }
 277
 278 gchar*
 279 modest_text_utils_remove_address (const gchar *address_list, const gchar *address)
 280 {
 281         gchar *dup, *token, *ptr, *result;
 282         GString *filtered_emails;
 283         gchar *email_address;
 284
 285         g_return_val_if_fail (address_list, NULL);
 286
 287         if (!address)
 288                 return g_strdup (address_list);
 289
 290         email_address = get_email_from_address (address);
 291
 292         /* search for substring */
 293         if (!strstr ((const char *) address_list, (const char *) email_address)) {
 294                 g_free (email_address);
 295                 return g_strdup (address_list);
 296         }
 297
 298         dup = g_strdup (address_list);
 299         filtered_emails = g_string_new (NULL);
 300
 301         token = strtok_r (dup, ",", &ptr);
 302
 303         while (token != NULL) {
 304                 /* Add to list if not found */
 305                 if (!strstr ((const char *) token, (const char *) email_address)) {
 306                         if (filtered_emails->len == 0)
 307                                 g_string_append_printf (filtered_emails, "%s", g_strstrip (token));
 308                         else
 309                                 g_string_append_printf (filtered_emails, ",%s", g_strstrip (token));
 310                 }
 311                 token = strtok_r (NULL, ",", &ptr);
 312         }
 313         result = filtered_emails->str;
 314
 315         /* Clean */
 316         g_free (email_address);
 317         g_free (dup);
 318         g_string_free (filtered_emails, FALSE);
 319
 320         return result;
 321 }
 322
 323 static void
 324 modest_text_utils_convert_buffer_to_html (GString *html, const gchar *data)
 325 {
 326         guint            i;
 327         gboolean        space_seen = FALSE;
 328         gsize           len;
 329         guint           break_dist = 0; /* distance since last break point */
 330
 331         len = strlen (data);
 332
 333         /* replace with special html chars where needed*/
 334         for (i = 0; i != len; ++i)  {
 335                 char kar = data[i];
 336
 337                 if (space_seen && kar != ' ') {
 338                         g_string_append_c (html, ' ');
 339                         space_seen = FALSE;
 340                 }
 341
 342                 /* we artificially insert a breakpoint (newline)
 343                  * after 256, to make sure our lines are not so long
 344                  * they will DOS the regexping later
 345                  */
 346                 if (++break_dist == 256) {
 347                         g_string_append_c (html, '\n');
 348                         break_dist = 0;
 349                 }
 350
 351                 switch (kar) {
 352                 case 0:  break; /* ignore embedded \0s */
 353                 case '<'  : g_string_append (html, "&lt;");   break;
 354                 case '>'  : g_string_append (html, "&gt;");   break;
 355                 case '&'  : g_string_append (html, "&amp;");  break;
 356                 case '"'  : g_string_append (html, "&quot;");  break;
 357
 358                 /* don't convert &apos; --> wpeditor will try to re-convert it... */
 359                 //case '\'' : g_string_append (html, "&apos;"); break;
 360                 case '\n' : g_string_append (html, "<br>\n");              break_dist= 0; break;
 361                 case '\t' : g_string_append (html, "&nbsp;&nbsp;&nbsp; "); break_dist=0; break; /* note the space at the end*/
 362                 case ' ':
 363                         break_dist = 0;
 364                         if (space_seen) { /* second space in a row */
 365                                 g_string_append (html, "&nbsp; ");
 366                                 space_seen = FALSE;
 367                         } else
 368                                 space_seen = TRUE;
 369                         break;
 370                 default:
 371                         g_string_append_c (html, kar);
 372                 }
 373         }
 374 }
 375
 376 gchar*
 377 modest_text_utils_convert_to_html (const gchar *data)
 378 {
 379         GString         *html;
 380         gsize           len;
 381
 382         if (!data)
 383                 return NULL;
 384
 385         len = strlen (data);
 386         html = g_string_sized_new (1.5 * len);  /* just a  guess... */
 387
 388         g_string_append_printf (html,
 389                                 "<html><head>"
 390                                 "<meta http-equiv=\"content-type\" content=\"text/html; charset=utf8\">"
 391                                 "</head>"
 392                                 "<body>");
 393
 394         modest_text_utils_convert_buffer_to_html (html, data);
 395
 396         g_string_append (html, "</body></html>");
 397
 398         if (len <= HYPERLINKIFY_MAX_LENGTH)
 399                 hyperlinkify_plain_text (html);
 400
 401         return g_string_free (html, FALSE);
 402 }
 403
 404 gchar *
 405 modest_text_utils_convert_to_html_body (const gchar *data)
 406 {
 407         GString         *html;
 408         gsize           len;
 409
 410         if (!data)
 411                 return NULL;
 412
 413         len = strlen (data);
 414         html = g_string_sized_new (1.5 * len);  /* just a  guess... */
 415
 416         modest_text_utils_convert_buffer_to_html (html, data);
 417
 418         if (len < HYPERLINKIFY_MAX_LENGTH)
 419                 hyperlinkify_plain_text (html);
 420
 421         return g_string_free (html, FALSE);
 422 }
 423
 424 void
 425 modest_text_utils_get_addresses_indexes (const gchar *addresses, GSList **start_indexes, GSList **end_indexes)
 426 {
 427         gchar *current, *start, *last_blank;
 428         gint start_offset = 0, current_offset = 0;
 429
 430         g_return_if_fail (start_indexes != NULL);
 431         g_return_if_fail (end_indexes != NULL);
 432
 433         start = (gchar *) addresses;
 434         current = start;
 435         last_blank = start;
 436
 437         while (*current != '\0') {
 438                 if ((start == current)&&((*current == ' ')||(*current == ',')||(*current == ';'))) {
 439                         start = g_utf8_next_char (start);
 440                         start_offset++;
 441                         last_blank = current;
 442                 } else if ((*current == ',')||(*current == ';')) {
 443                         gint *start_index, *end_index;
 444                         start_index = g_new0(gint, 1);
 445                         end_index = g_new0(gint, 1);
 446                         *start_index = start_offset;
 447                         *end_index = current_offset;
 448                         *start_indexes = g_slist_prepend (*start_indexes, start_index);
 449                         *end_indexes = g_slist_prepend (*end_indexes, end_index);
 450                         start = g_utf8_next_char (current);
 451                         start_offset = current_offset + 1;
 452                         last_blank = start;
 453                 } else if (*current == '"') {
 454                         current = g_utf8_next_char (current);
 455                         current_offset ++;
 456                         while ((*current != '"')&&(*current != '\0')) {
 457                                 current = g_utf8_next_char (current);
 458                                 current_offset ++;
 459                         }
 460                 }
 461
 462                 current = g_utf8_next_char (current);
 463                 current_offset ++;
 464         }
 465
 466         if (start != current) {
 467                         gint *start_index, *end_index;
 468                         start_index = g_new0(gint, 1);
 469                         end_index = g_new0(gint, 1);
 470                         *start_index = start_offset;
 471                         *end_index = current_offset;
 472                         *start_indexes = g_slist_prepend (*start_indexes, start_index);
 473                         *end_indexes = g_slist_prepend (*end_indexes, end_index);
 474         }
 475
 476         *start_indexes = g_slist_reverse (*start_indexes);
 477         *end_indexes = g_slist_reverse (*end_indexes);
 478
 479         return;
 480 }
 481
 482 GSList *
 483 modest_text_utils_split_addresses_list (const gchar *addresses)
 484 {
 485         gchar *current, *start, *last_blank;
 486         GSList *result = NULL;
 487
 488         start = (gchar *) addresses;
 489         current = start;
 490         last_blank = start;
 491
 492         while (*current != '\0') {
 493                 if ((start == current)&&((*current == ' ')||(*current == ',')||(*current == ';'))) {
 494                         start = g_utf8_next_char (start);
 495                         last_blank = current;
 496                 } else if ((*current == ',')||(*current == ';')) {
 497                         gchar *new_address = NULL;
 498                         new_address = g_strndup (start, current - last_blank);
 499                         result = g_slist_prepend (result, new_address);
 500                         start = g_utf8_next_char (current);
 501                         last_blank = start;
 502                 } else if (*current == '\"') {
 503                         if (current == start) {
 504                                 current = g_utf8_next_char (current);
 505                                 start = g_utf8_next_char (start);
 506                         }
 507                         while ((*current != '\"')&&(*current != '\0'))
 508                                 current = g_utf8_next_char (current);
 509                 }
 510
 511                 current = g_utf8_next_char (current);
 512         }
 513
 514         if (start != current) {
 515                 gchar *new_address = NULL;
 516                 new_address = g_strndup (start, current - last_blank);
 517                 result = g_slist_prepend (result, new_address);
 518         }
 519
 520         result = g_slist_reverse (result);
 521         return result;
 522
 523 }
 524
 525 void
 526 modest_text_utils_address_range_at_position (const gchar *recipients_list,
 527                                              gint position,
 528                                              gint *start,
 529                                              gint *end)
 530 {
 531         gchar *current = NULL;
 532         gint range_start = 0;
 533         gint range_end = 0;
 534         gint index;
 535         gboolean is_quoted = FALSE;
 536
 537         index = 0;
 538         for (current = (gchar *) recipients_list; *current != '\0'; current = g_utf8_find_next_char (current, NULL)) {
 539                 gunichar c = g_utf8_get_char (current);
 540
 541                 if ((c == ',') && (!is_quoted)) {
 542                         if (index < position) {
 543                                 range_start = index + 1;
 544                         } else {
 545                                 break;
 546                         }
 547                 } else if (c == '\"') {
 548                         is_quoted = !is_quoted;
 549                 } else if ((c == ' ') &&(range_start == index)) {
 550                         range_start ++;
 551                 }
 552                 index ++;
 553                 range_end = index;
 554         }
 555
 556         if (start)
 557                 *start = range_start;
 558         if (end)
 559                 *end = range_end;
 560 }
 561
 562
 563 /* ******************************************************************* */
 564 /* ************************* UTILIY FUNCTIONS ************************ */
 565 /* ******************************************************************* */
 566
 567 static GString *
 568 get_next_line (const gchar * b, const gsize blen, const gchar * iter)
 569 {
 570         GString *gs;
 571         const gchar *i0;
 572
 573         if (iter > b + blen)
 574                 return g_string_new("");
 575
 576         i0 = iter;
 577         while (iter[0]) {
 578                 if (iter[0] == '\n')
 579                         break;
 580                 iter++;
 581         }
 582         gs = g_string_new_len (i0, iter - i0);
 583         return gs;
 584 }
 585 static int
 586 get_indent_level (const char *l)
 587 {
 588         int indent = 0;
 589
 590         while (l[0]) {
 591                 if (l[0] == '>') {
 592                         indent++;
 593                         if (l[1] == ' ') {
 594                                 l++;
 595                         }
 596                 } else {
 597                         break;
 598                 }
 599                 l++;
 600
 601         }
 602
 603         /*      if we hit the signature marker "-- ", we return -(indent + 1). This
 604          *      stops reformatting.
 605          */
 606         if (strcmp (l, "-- ") == 0) {
 607                 return -1 - indent;
 608         } else {
 609                 return indent;
 610         }
 611 }
 612
 613 static void
 614 unquote_line (GString * l)
 615 {
 616         gchar *p;
 617
 618         p = l->str;
 619         while (p[0]) {
 620                 if (p[0] == '>') {
 621                         if (p[1] == ' ') {
 622                                 p++;
 623                         }
 624                 } else {
 625                         break;
 626                 }
 627                 p++;
 628         }
 629         g_string_erase (l, 0, p - l->str);
 630 }
 631
 632 static void
 633 append_quoted (GString * buf, int indent, const GString * str,
 634                const int cutpoint)
 635 {
 636         int i;
 637
 638         indent = indent < 0 ? abs (indent) - 1 : indent;
 639         for (i = 0; i <= indent; i++) {
 640                 g_string_append (buf, "> ");
 641         }
 642         if (cutpoint > 0) {
 643                 g_string_append_len (buf, str->str, cutpoint);
 644         } else {
 645                 g_string_append (buf, str->str);
 646         }
 647         g_string_append (buf, "\n");
 648 }
 649
 650 static int
 651 get_breakpoint_utf8 (const gchar * s, gint indent, const gint limit)
 652 {
 653         gint index = 0;
 654         const gchar *pos, *last;
 655         gunichar *uni;
 656
 657         indent = indent < 0 ? abs (indent) - 1 : indent;
 658
 659         last = NULL;
 660         pos = s;
 661         uni = g_utf8_to_ucs4_fast (s, -1, NULL);
 662         while (pos[0]) {
 663                 if ((index + 2 * indent > limit) && last) {
 664                         g_free (uni);
 665                         return last - s;
 666                 }
 667                 if (g_unichar_isspace (uni[index])) {
 668                         last = pos;
 669                 }
 670                 pos = g_utf8_next_char (pos);
 671                 index++;
 672         }
 673         g_free (uni);
 674         return strlen (s);
 675 }
 676
 677 static int
 678 get_breakpoint_ascii (const gchar * s, const gint indent, const gint limit)
 679 {
 680         gint i, last;
 681
 682         last = strlen (s);
 683         if (last + 2 * indent < limit)
 684                 return last;
 685
 686         for (i = strlen (s); i > 0; i--) {
 687                 if (s[i] == ' ') {
 688                         if (i + 2 * indent <= limit) {
 689                                 return i;
 690                         } else {
 691                                 last = i;
 692                         }
 693                 }
 694         }
 695         return last;
 696 }
 697
 698 static int
 699 get_breakpoint (const gchar * s, const gint indent, const gint limit)
 700 {
 701
 702         if (g_utf8_validate (s, -1, NULL)) {
 703                 return get_breakpoint_utf8 (s, indent, limit);
 704         } else {                /* assume ASCII */
 705                 //g_warning("invalid UTF-8 in msg");
 706                 return get_breakpoint_ascii (s, indent, limit);
 707         }
 708 }
 709
 710 static gchar *
 711 cite (const time_t sent_date, const gchar *from)
 712 {
 713         return g_strdup (_("mcen_ia_editor_original_message"));
 714 }
 715
 716 static gchar *
 717 quoted_attachments (GList *attachments)
 718 {
 719         GList *node = NULL;
 720         GString *result = g_string_new ("");
 721         for (node = attachments; node != NULL; node = g_list_next (node)) {
 722                 gchar *filename = (gchar *) node->data;
 723                 g_string_append_printf ( result, "%s %s\n", _("mcen_ia_editor_attach_filename"), filename);
 724         }
 725
 726         return g_string_free (result, FALSE);
 727
 728 }
 729
 730 static gchar *
 731 modest_text_utils_quote_plain_text (const gchar *text,
 732                                     const gchar *cite,
 733                                     const gchar *signature,
 734                                     GList *attachments,
 735                                     int limit)
 736 {
 737         const gchar *iter;
 738         gint indent, breakpoint, rem_indent = 0;
 739         GString *q, *l, *remaining;
 740         gsize len;
 741         gchar *attachments_string = NULL;
 742
 743         /* remaining will store the rest of the line if we have to break it */
 744         q = g_string_new ("\n");
 745         q = g_string_append (q, cite);
 746         q = g_string_append_c (q, '\n');
 747         remaining = g_string_new ("");
 748
 749         iter = text;
 750         len = strlen(text);
 751         do {
 752                 l = get_next_line (text, len, iter);
 753                 iter = iter + l->len + 1;
 754                 indent = get_indent_level (l->str);
 755                 unquote_line (l);
 756
 757                 if (remaining->len) {
 758                         if (l->len && indent == rem_indent) {
 759                                 g_string_prepend (l, " ");
 760                                 g_string_prepend (l, remaining->str);
 761                         } else {
 762                                 do {
 763                                         breakpoint =
 764                                                 get_breakpoint (remaining->str,
 765                                                                 rem_indent,
 766                                                                 limit);
 767                                         append_quoted (q, rem_indent,
 768                                                        remaining, breakpoint);
 769                                         g_string_erase (remaining, 0,
 770                                                         breakpoint);
 771                                         if (remaining->str[0] == ' ') {
 772                                                 g_string_erase (remaining, 0,
 773                                                                 1);
 774                                         }
 775                                 } while (remaining->len);
 776                         }
 777                 }
 778                 g_string_free (remaining, TRUE);
 779                 breakpoint = get_breakpoint (l->str, indent, limit);
 780                 remaining = g_string_new (l->str + breakpoint);
 781                 if (remaining->str[0] == ' ') {
 782                         g_string_erase (remaining, 0, 1);
 783                 }
 784                 rem_indent = indent;
 785                 append_quoted (q, indent, l, breakpoint);
 786                 g_string_free (l, TRUE);
 787         } while ((iter < text + len) || (remaining->str[0]));
 788
 789         attachments_string = quoted_attachments (attachments);
 790         q = g_string_append (q, attachments_string);
 791         g_free (attachments_string);
 792
 793         if (signature != NULL) {
 794                 q = g_string_append_c (q, '\n');
 795                 q = g_string_append (q, signature);
 796         }
 797
 798         return g_string_free (q, FALSE);
 799 }
 800
 801 static gchar*
 802 modest_text_utils_quote_html (const gchar *text,
 803                               const gchar *cite,
 804                               const gchar *signature,
 805                               GList *attachments,
 806                               int limit)
 807 {
 808         gchar *result = NULL;
 809         gchar *signature_result = NULL;
 810         const gchar *format = \
 811                 "<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\">\n" \
 812                 "<html>\n" \
 813                 "<body>\n" \
 814                 "<br/>%s<br/>" \
 815                 "<pre>%s<br/>%s<br/>%s</pre>\n" \
 816                 "</body>\n" \
 817                 "</html>\n";
 818         gchar *attachments_string = NULL;
 819         gchar *q_attachments_string = NULL;
 820         gchar *q_cite = NULL;
 821         gchar *html_text = NULL;
 822
 823         if (signature == NULL)
 824                 signature_result = g_strdup ("");
 825         else
 826                 signature_result = modest_text_utils_convert_to_html_body (signature);
 827
 828         attachments_string = quoted_attachments (attachments);
 829         q_attachments_string = modest_text_utils_convert_to_html_body (attachments_string);
 830         q_cite = modest_text_utils_convert_to_html_body (cite);
 831         html_text = modest_text_utils_convert_to_html_body (text);
 832         result = g_strdup_printf (format, signature_result, q_cite, html_text, q_attachments_string);
 833         g_free (q_cite);
 834         g_free (html_text);
 835         g_free (attachments_string);
 836         g_free (q_attachments_string);
 837         g_free (signature_result);
 838
 839         return result;
 840 }
 841
 842 static gint
 843 cmp_offsets_reverse (const url_match_t *match1, const url_match_t *match2)
 844 {
 845         return match2->offset - match1->offset;
 846 }
 847
 848
 849 static GSList*
 850 get_url_matches (GString *txt)
 851 {
 852         regmatch_t rm;
 853         guint rv, i, offset = 0;
 854         GSList *match_list = NULL;
 855
 856         static url_match_pattern_t patterns[] = MAIL_VIEWER_URL_MATCH_PATTERNS;
 857         const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
 858
 859         /* initalize the regexps */
 860         for (i = 0; i != pattern_num; ++i) {
 861                 patterns[i].preg = g_slice_new0 (regex_t);
 862
 863                 /* this should not happen */
 864                 g_return_val_if_fail (regcomp (patterns[i].preg, patterns[i].regex,
 865                                                REG_ICASE|REG_EXTENDED|REG_NEWLINE) == 0, NULL);
 866         }
 867         /* find all the matches */
 868         for (i = 0; i != pattern_num; ++i) {
 869                 offset     = 0;
 870                 while (1) {
 871                         url_match_t *match;
 872                         gboolean is_submatch;
 873                         GSList *cursor;
 874
 875                         if ((rv = regexec (patterns[i].preg, txt->str + offset, 1, &rm, 0)) != 0) {
 876                                 g_return_val_if_fail (rv == REG_NOMATCH, NULL); /* this should not happen */
 877                                 break; /* try next regexp */
 878                         }
 879                         if (rm.rm_so == -1)
 880                                 break;
 881
 882                         is_submatch = FALSE;
 883                         /* check  old matches to see if this has already been matched */
 884                         cursor = match_list;
 885                         while (cursor && !is_submatch) {
 886                                 const url_match_t *old_match =
 887                                         (const url_match_t *) cursor->data;
 888                                 guint new_offset = offset + rm.rm_so;
 889                                 is_submatch = (new_offset >  old_match->offset &&
 890                                                new_offset <  old_match->offset + old_match->len);
 891                                 cursor = g_slist_next (cursor);
 892                         }
 893
 894                         if (!is_submatch) {
 895                                 /* make a list of our matches (<offset, len, prefix> tupels)*/
 896                                 match = g_slice_new (url_match_t);
 897                                 match->offset = offset + rm.rm_so;
 898                                 match->len    = rm.rm_eo - rm.rm_so;
 899                                 match->prefix = patterns[i].prefix;
 900                                 match_list = g_slist_prepend (match_list, match);
 901                         }
 902
 903                         offset += rm.rm_eo;
 904                 }
 905         }
 906
 907         for (i = 0; i != pattern_num; ++i) {
 908                 regfree (patterns[i].preg);
 909                 g_slice_free  (regex_t, patterns[i].preg);
 910         } /* don't free patterns itself -- it's static */
 911
 912         /* now sort the list, so the matches are in reverse order of occurence.
 913          * that way, we can do the replacements starting from the end, so we don't need
 914          * to recalculate the offsets
 915          */
 916         match_list = g_slist_sort (match_list,
 917                                    (GCompareFunc)cmp_offsets_reverse);
 918         return match_list;
 919 }
 920
 921
 922
 923 static void
 924 hyperlinkify_plain_text (GString *txt)
 925 {
 926         GSList *cursor;
 927         GSList *match_list = get_url_matches (txt);
 928
 929         /* we will work backwards, so the offsets stay valid */
 930         for (cursor = match_list; cursor; cursor = cursor->next) {
 931
 932                 url_match_t *match = (url_match_t*) cursor->data;
 933                 gchar *url  = g_strndup (txt->str + match->offset, match->len);
 934                 gchar *repl = NULL; /* replacement  */
 935
 936                 /* the prefix is NULL: use the one that is already there */
 937                 repl = g_strdup_printf ("<a href=\"%s%s\">%s</a>",
 938                                         match->prefix ? match->prefix : EMPTY_STRING,
 939                                         url, url);
 940
 941                 /* replace the old thing with our hyperlink
 942                  * replacement thing */
 943                 g_string_erase  (txt, match->offset, match->len);
 944                 g_string_insert (txt, match->offset, repl);
 945
 946                 g_free (url);
 947                 g_free (repl);
 948
 949                 g_slice_free (url_match_t, match);
 950         }
 951
 952         g_slist_free (match_list);
 953 }
 954
 955
 956 gchar*
 957 modest_text_utils_get_display_address (const gchar *address)
 958 {
 959         gchar *display;
 960         gchar **tokens;
 961         gint i = 0;
 962
 963         if (!address)
 964                 return NULL;
 965
 966         g_return_val_if_fail (g_utf8_validate (address, -1, NULL), NULL);
 967
 968         tokens = g_strsplit_set ((const gchar*) address, "<>()", 3);
 969
 970         /* Note that if any of the delimiters is the first character
 971            then g_strsplit_set will return "" as the first string */
 972         while (tokens[i] != NULL) {
 973                 if (strlen ((char *) (tokens[i])) != 0)
 974                         break;
 975                 i++;
 976         }
 977
 978         display = g_strdup (tokens [i]);
 979         g_strchug (display);
 980
 981         /* Free the other tokens */
 982         g_strfreev (tokens);
 983
 984         return display;
 985 }
 986
 987 gchar *
 988 modest_text_utils_get_email_address (const gchar *full_address)
 989 {
 990         const gchar *left, *right;
 991
 992         if (!full_address)
 993                 return NULL;
 994
 995         g_return_val_if_fail (g_utf8_validate (full_address, -1, NULL), NULL);
 996
 997         left = g_strrstr_len (full_address, strlen(full_address), "<");
 998         if (left == NULL)
 999                 return g_strdup (full_address);
1000
1001         right = g_strstr_len (left, strlen(left), ">");
1002         if (right == NULL)
1003                 return g_strdup (full_address);
1004
1005         return g_strndup (left + 1, right - left - 1);
1006 }
1007
1008 gint
1009 modest_text_utils_get_subject_prefix_len (const gchar *sub)
1010 {
1011         gint i;
1012         static const gchar* prefix[] = {
1013                 "Re:", "RE:", "RV:", "re:"
1014                 "Fwd:", "FWD:", "FW:", "fwd:", "Fw:", "fw:", NULL
1015         };
1016
1017         if (!sub || (sub[0] != 'R' && sub[0] != 'F' && sub[0] != 'r' && sub[0] != 'f')) /* optimization */
1018                 return 0;
1019
1020         i = 0;
1021
1022         while (prefix[i]) {
1023                 if (g_str_has_prefix(sub, prefix[i])) {
1024                         int prefix_len = strlen(prefix[i]);
1025                         if (sub[prefix_len] == ' ')
1026                                 ++prefix_len; /* ignore space after prefix as well */
1027                         return prefix_len;
1028                 }
1029                 ++i;
1030         }
1031         return 0;
1032 }
1033
1034
1035 gint
1036 modest_text_utils_utf8_strcmp (const gchar* s1, const gchar *s2, gboolean insensitive)
1037 {
1038         gint result = 0;
1039         gchar *n1, *n2;
1040
1041         /* work even when s1 and/or s2 == NULL */
1042         if (G_UNLIKELY(s1 == s2))
1043                 return 0;
1044
1045         /* if it's not case sensitive */
1046         if (!insensitive)
1047                 return strcmp (s1 ? s1 : "", s2 ? s2 : "");
1048
1049         n1 = g_utf8_collate_key (s1 ? s1 : "", -1);
1050         n2 = g_utf8_collate_key (s2 ? s2 : "", -1);
1051
1052         result = strcmp (n1, n2);
1053
1054         g_free (n1);
1055         g_free (n2);
1056
1057         return result;
1058 }
1059
1060
1061 gchar*
1062 modest_text_utils_get_display_date (time_t date)
1063 {
1064         time_t now;
1065         static const guint BUF_SIZE = 64;
1066         static const guint ONE_DAY = 24 * 60 * 60; /* seconds in one day */
1067         gchar date_buf[BUF_SIZE];
1068         gchar today_buf [BUF_SIZE];
1069
1070         modest_text_utils_strftime (date_buf, BUF_SIZE, "%x", date);
1071
1072         now = time (NULL);
1073
1074         /* we check if the date is within the last 24h, if not, we don't
1075          * have to do the extra, expensive strftime, which was very visible
1076          * in the profiles.
1077          */
1078         if (abs(now - date) < ONE_DAY) {
1079
1080                 /* it's within the last 24 hours, but double check */
1081                 /* use the localized dates */
1082                 modest_text_utils_strftime (today_buf,  BUF_SIZE, "%x", now);
1083
1084                 /* if it's today, use the time instead */
1085                 if (strcmp (date_buf, today_buf) == 0)
1086                         modest_text_utils_strftime (date_buf, BUF_SIZE, "%X", date);
1087         }
1088
1089         return g_strdup(date_buf);
1090 }
1091
1092
1093 gboolean
1094 modest_text_utils_validate_folder_name (const gchar *folder_name)
1095 {
1096         /* based on http://msdn2.microsoft.com/en-us/library/aa365247.aspx,
1097          * with some extras */
1098
1099         guint len;
1100         gint i;
1101         const gchar **cursor = NULL;
1102         const gchar *forbidden_names[] = { /* windows does not like these */
1103                 "CON", "PRN", "AUX", "NUL", "COM1", "COM2", "COM3", "COM4", "COM5", "COM6",
1104                 "COM7", "COM8", "COM9", "LPT1", "LPT2", "LPT3", "LPT4", "LPT5", "LPT6", "LPT7", "LPT8", "LPT9",
1105                 ".", "..", NULL
1106         };
1107
1108         /* cannot be NULL */
1109         if (!folder_name)
1110                 return FALSE;
1111
1112         /* cannot be empty */
1113         len = strlen(folder_name);
1114         if (len == 0)
1115                 return FALSE;
1116
1117         /* cannot start or end with a space */
1118         if (g_ascii_isspace(folder_name[0]) || g_ascii_isspace(folder_name[len - 1]))
1119                 return FALSE;
1120
1121         /* cannot contain a forbidden char */
1122         for (i = 0; i < len; i++)
1123                 if (modest_text_utils_is_forbidden_char (folder_name[i], FOLDER_NAME_FORBIDDEN_CHARS))
1124                         return FALSE;
1125
1126         /* cannot contain a forbidden word */
1127         if (len <= 4) {
1128                 for (cursor = forbidden_names; cursor && *cursor; ++cursor) {
1129                         if (g_ascii_strcasecmp (folder_name, *cursor) == 0)
1130                                 return FALSE;
1131                 }
1132         }
1133         return TRUE; /* it's valid! */
1134 }
1135
1136
1137
1138 gboolean
1139 modest_text_utils_validate_domain_name (const gchar *domain)
1140 {
1141         gboolean valid = FALSE;
1142         regex_t rx;
1143         const gchar* domain_regex = "^[a-z0-9]([.]?[a-z0-9-])*[a-z0-9]$";
1144
1145         if (!domain)
1146                 return FALSE;
1147
1148         /* domain name: all alphanum or '-' or '.',
1149          * but beginning/ending in alphanum */
1150         if (regcomp (&rx, domain_regex, REG_ICASE|REG_EXTENDED|REG_NOSUB)) {
1151                 g_warning ("BUG: error in regexp");
1152                 return FALSE;
1153         }
1154
1155         valid = (regexec (&rx, domain, 1, NULL, 0) == 0);
1156         regfree (&rx);
1157
1158         return valid;
1159 }
1160
1161
1162
1163 gboolean
1164 modest_text_utils_validate_email_address (const gchar *email_address, const gchar **invalid_char_position)
1165 {
1166         int count = 0;
1167         const gchar *c = NULL, *domain = NULL;
1168         static gchar *rfc822_specials = "()<>@,;:\\\"[]&";
1169
1170         if (invalid_char_position != NULL)
1171                 *invalid_char_position = NULL;
1172
1173         /* check that the email adress contains exactly one @ */
1174         if (!strstr(email_address, "@") ||
1175                         (strstr(email_address, "@") != g_strrstr(email_address, "@")))
1176         {
1177                 return FALSE;
1178         }
1179
1180         /* first we validate the name portion (name@domain) */
1181         for (c = email_address;  *c;  c++) {
1182                 if (*c == '\"' &&
1183                     (c == email_address ||
1184                      *(c - 1) == '.' ||
1185                      *(c - 1) == '\"')) {
1186                         while (*++c) {
1187                                 if (*c == '\"')
1188                                         break;
1189                                 if (*c == '\\' && (*++c == ' '))
1190                                         continue;
1191                                 if (*c <= ' ' || *c >= 127)
1192                                         return FALSE;
1193                         }
1194                         if (!*c++)
1195                                 return FALSE;
1196                         if (*c == '@')
1197                                 break;
1198                         if (*c != '.')
1199                                 return FALSE;
1200                         continue;
1201                 }
1202                 if (*c == '@')
1203                         break;
1204                 if (*c <= ' ' || *c >= 127)
1205                         return FALSE;
1206                 if (strchr(rfc822_specials, *c)) {
1207                         if (invalid_char_position)
1208                                 *invalid_char_position = c;
1209                         return FALSE;
1210                 }
1211         }
1212         if (c == email_address || *(c - 1) == '.')
1213                 return FALSE;
1214
1215         /* next we validate the domain portion (name@domain) */
1216         if (!*(domain = ++c))
1217                 return FALSE;
1218         do {
1219                 if (*c == '.') {
1220                         if (c == domain || *(c - 1) == '.' || *(c + 1) == '\0')
1221                                 return FALSE;
1222                         count++;
1223                 }
1224                 if (*c <= ' ' || *c >= 127)
1225                         return FALSE;
1226                 if (strchr(rfc822_specials, *c)) {
1227                         if (invalid_char_position)
1228                                 *invalid_char_position = c;
1229                         return FALSE;
1230                 }
1231         } while (*++c);
1232
1233         return (count >= 1) ? TRUE : FALSE;
1234 }
1235
1236 gboolean
1237 modest_text_utils_validate_recipient (const gchar *recipient, const gchar **invalid_char_position)
1238 {
1239         gchar *stripped, *current;
1240         gchar *right_part;
1241         gboolean has_error = FALSE;
1242
1243         if (modest_text_utils_validate_email_address (recipient, invalid_char_position))
1244                 return TRUE;
1245         stripped = g_strdup (recipient);
1246         stripped = g_strstrip (stripped);
1247         current = stripped;
1248
1249         if (*current == '\0') {
1250                 g_free (stripped);
1251                 return FALSE;
1252         }
1253
1254         /* quoted string */
1255         if (*current == '\"') {
1256                 current = g_utf8_next_char (current);
1257                 has_error = TRUE;
1258                 for (; *current != '\0'; current = g_utf8_next_char (current)) {
1259                         if (*current == '\\') {
1260                                 /* TODO: This causes a warning, which breaks the build,
1261                                  * because a gchar cannot be < 0.
1262                                  * murrayc.
1263                                 if (current[1] <0) {
1264                                         has_error = TRUE;
1265                                         break;
1266                                 }
1267                                 */
1268                         } else if (*current == '\"') {
1269                                 has_error = FALSE;
1270                                 current = g_utf8_next_char (current);
1271                                 break;
1272                         }
1273                 }
1274         } else {
1275                 has_error = TRUE;
1276                 for (current = stripped ; *current != '\0'; current = g_utf8_next_char (current)) {
1277                         if (*current == '<') {
1278                                 has_error = FALSE;
1279                                 break;
1280                         }
1281                 }
1282         }
1283
1284         if (has_error) {
1285                 g_free (stripped);
1286                 return FALSE;
1287         }
1288
1289         right_part = g_strdup (current);
1290         g_free (stripped);
1291         right_part = g_strstrip (right_part);
1292
1293         if (g_str_has_prefix (right_part, "<") &&
1294             g_str_has_suffix (right_part, ">")) {
1295                 gchar *address;
1296                 gboolean valid;
1297
1298                 address = g_strndup (right_part+1, strlen (right_part) - 2);
1299                 g_free (right_part);
1300                 valid = modest_text_utils_validate_email_address (address, invalid_char_position);
1301                 g_free (address);
1302                 return valid;
1303         } else {
1304                 g_free (right_part);
1305                 return FALSE;
1306         }
1307 }
1308
1309
1310 gchar *
1311 modest_text_utils_get_display_size (guint64 size)
1312 {
1313         const guint KB=1024;
1314         const guint MB=1024 * KB;
1315         const guint GB=1024 * MB;
1316
1317         if (size == 0)
1318                 return g_strdup_printf(_FM("sfil_li_size_kb"), 0);
1319         if (0 < size && size < KB)
1320                 return g_strdup_printf (_FM("sfil_li_size_kb"), 1);
1321         else if (KB <= size && size < 100 * KB)
1322                 return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), size / KB);
1323         else if (100*KB <= size && size < MB)
1324                 return g_strdup_printf (_FM("sfil_li_size_100kb_1mb"), (float) size / MB);
1325         else if (MB <= size && size < 10*MB)
1326                 return g_strdup_printf (_FM("sfil_li_size_1mb_10mb"), (float) size / MB);
1327         else if (10*MB <= size && size < GB)
1328                 return g_strdup_printf (_FM("sfil_li_size_10mb_1gb"), size / MB);
1329         else
1330                 return g_strdup_printf (_FM("sfil_li_size_1gb_or_greater"), (float) size / GB);
1331 }
1332
1333 static gchar *
1334 get_email_from_address (const gchar * address)
1335 {
1336         gchar *left_limit, *right_limit;
1337
1338         left_limit = strstr (address, "<");
1339         right_limit = g_strrstr (address, ">");
1340
1341         if ((left_limit == NULL)||(right_limit == NULL)|| (left_limit > right_limit))
1342                 return g_strdup (address);
1343         else
1344                 return g_strndup (left_limit + 1, (right_limit - left_limit) - 1);
1345 }
1346
1347 gchar *
1348 modest_text_utils_get_color_string (GdkColor *color)
1349 {
1350
1351         return g_strdup_printf ("#%x%x%x%x%x%x%x%x%x%x%x%x",
1352                                 (color->red >> 12)   & 0xf, (color->red >> 8)   & 0xf,
1353                                 (color->red >>  4)   & 0xf, (color->red)        & 0xf,
1354                                 (color->green >> 12) & 0xf, (color->green >> 8) & 0xf,
1355                                 (color->green >>  4) & 0xf, (color->green)      & 0xf,
1356                                 (color->blue >> 12)  & 0xf, (color->blue >> 8)  & 0xf,
1357                                 (color->blue >>  4)  & 0xf, (color->blue)       & 0xf);
1358 }
1359
1360 gchar *
1361 modest_text_utils_text_buffer_get_text (GtkTextBuffer *buffer)
1362 {
1363         GtkTextIter start, end;
1364         gchar *slice, *current;
1365         GString *result = g_string_new ("");
1366
1367         g_return_val_if_fail (GTK_IS_TEXT_BUFFER (buffer), NULL);
1368
1369         gtk_text_buffer_get_start_iter (buffer, &start);
1370         gtk_text_buffer_get_end_iter (buffer, &end);
1371
1372         slice = gtk_text_buffer_get_slice (buffer, &start, &end, FALSE);
1373         current = slice;
1374
1375         while (current && current != '\0') {
1376                 if (g_utf8_get_char (current) == 0xFFFC) {
1377                         result = g_string_append_c (result, ' ');
1378                         current = g_utf8_next_char (current);
1379                 } else {
1380                         gchar *next = g_utf8_strchr (current, -1, 0xFFFC);
1381                         if (next == NULL) {
1382                                 result = g_string_append (result, current);
1383                         } else {
1384                                 result = g_string_append_len (result, current, next - current);
1385                         }
1386                         current = next;
1387                 }
1388         }
1389         g_free (slice);
1390
1391         return g_string_free (result, FALSE);
1392
1393 }
1394
1395 gboolean
1396 modest_text_utils_is_forbidden_char (const gchar character,
1397                                      ModestTextUtilsForbiddenCharType type)
1398 {
1399         gint i, len;
1400         const gchar *forbidden_chars = NULL;
1401
1402         /* We need to get the length in the switch because the
1403            compiler needs to know the size at compile time */
1404         switch (type) {
1405         case ACCOUNT_TITLE_FORBIDDEN_CHARS:
1406                 forbidden_chars = account_title_forbidden_chars;
1407                 len = G_N_ELEMENTS (account_title_forbidden_chars);
1408                 break;
1409         case FOLDER_NAME_FORBIDDEN_CHARS:
1410                 forbidden_chars = folder_name_forbidden_chars;
1411                 len = G_N_ELEMENTS (folder_name_forbidden_chars);
1412                 break;
1413         case USER_NAME_FORBIDDEN_NAMES:
1414                 forbidden_chars = user_name_forbidden_chars;
1415                 len = G_N_ELEMENTS (user_name_forbidden_chars);
1416                 break;
1417         default:
1418                 g_return_val_if_reached (TRUE);
1419         }
1420
1421         for (i = 0; i < len ; i++)
1422                 if (forbidden_chars[i] == character)
1423                         return TRUE;
1424
1425         return FALSE; /* it's valid! */
1426 }