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")) {
 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 ("\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                 case '\'' : g_string_append (html, "&apos;"); break;
 358                 case '\n' : g_string_append (html, "<br>\n");              break_dist= 0; break;
 359                 case '\t' : g_string_append (html, "&nbsp;&nbsp;&nbsp; "); break_dist=0; break; /* note the space at the end*/
 360                 case ' ':
 361                         break_dist = 0;
 362                         if (space_seen) { /* second space in a row */
 363                                 g_string_append (html, "&nbsp; ");
 364                                 space_seen = FALSE;
 365                         } else
 366                                 space_seen = TRUE;
 367                         break;
 368                 default:
 369                         g_string_append_c (html, kar);
 370                 }
 371         }
 372 }
 373
 374 gchar*
 375 modest_text_utils_convert_to_html (const gchar *data)
 376 {
 377         GString         *html;
 378         gsize           len;
 379
 380         if (!data)
 381                 return NULL;
 382
 383         len = strlen (data);
 384         html = g_string_sized_new (1.5 * len);  /* just a  guess... */
 385
 386         g_string_append_printf (html,
 387                                 "<html><head>"
 388                                 "<meta http-equiv=\"content-type\" content=\"text/html; charset=utf8\">"
 389                                 "</head>"
 390                                 "<body>");
 391
 392         modest_text_utils_convert_buffer_to_html (html, data);
 393
 394         g_string_append (html, "</body></html>");
 395
 396         if (len <= HYPERLINKIFY_MAX_LENGTH)
 397                 hyperlinkify_plain_text (html);
 398
 399         return g_string_free (html, FALSE);
 400 }
 401
 402 gchar *
 403 modest_text_utils_convert_to_html_body (const gchar *data)
 404 {
 405         GString         *html;
 406         gsize           len;
 407
 408         if (!data)
 409                 return NULL;
 410
 411         len = strlen (data);
 412         html = g_string_sized_new (1.5 * len);  /* just a  guess... */
 413
 414         modest_text_utils_convert_buffer_to_html (html, data);
 415
 416         if (len < HYPERLINKIFY_MAX_LENGTH)
 417                 hyperlinkify_plain_text (html);
 418
 419         return g_string_free (html, FALSE);
 420 }
 421
 422 void
 423 modest_text_utils_get_addresses_indexes (const gchar *addresses, GSList **start_indexes, GSList **end_indexes)
 424 {
 425         gchar *current, *start, *last_blank;
 426         gint start_offset = 0, current_offset = 0;
 427
 428         g_return_if_fail (start_indexes != NULL);
 429         g_return_if_fail (end_indexes != NULL);
 430
 431         start = (gchar *) addresses;
 432         current = start;
 433         last_blank = start;
 434
 435         while (*current != '\0') {
 436                 if ((start == current)&&((*current == ' ')||(*current == ',')||(*current == ';'))) {
 437                         start = g_utf8_next_char (start);
 438                         start_offset++;
 439                         last_blank = current;
 440                 } else if ((*current == ',')||(*current == ';')) {
 441                         gint *start_index, *end_index;
 442                         start_index = g_new0(gint, 1);
 443                         end_index = g_new0(gint, 1);
 444                         *start_index = start_offset;
 445                         *end_index = current_offset;
 446                         *start_indexes = g_slist_prepend (*start_indexes, start_index);
 447                         *end_indexes = g_slist_prepend (*end_indexes, end_index);
 448                         start = g_utf8_next_char (current);
 449                         start_offset = current_offset + 1;
 450                         last_blank = start;
 451                 } else if (*current == '"') {
 452                         current = g_utf8_next_char (current);
 453                         current_offset ++;
 454                         while ((*current != '"')&&(*current != '\0')) {
 455                                 current = g_utf8_next_char (current);
 456                                 current_offset ++;
 457                         }
 458                 }
 459
 460                 current = g_utf8_next_char (current);
 461                 current_offset ++;
 462         }
 463
 464         if (start != current) {
 465                         gint *start_index, *end_index;
 466                         start_index = g_new0(gint, 1);
 467                         end_index = g_new0(gint, 1);
 468                         *start_index = start_offset;
 469                         *end_index = current_offset;
 470                         *start_indexes = g_slist_prepend (*start_indexes, start_index);
 471                         *end_indexes = g_slist_prepend (*end_indexes, end_index);
 472         }
 473
 474         *start_indexes = g_slist_reverse (*start_indexes);
 475         *end_indexes = g_slist_reverse (*end_indexes);
 476
 477         return;
 478 }
 479
 480 GSList *
 481 modest_text_utils_split_addresses_list (const gchar *addresses)
 482 {
 483         gchar *current, *start, *last_blank;
 484         GSList *result = NULL;
 485
 486         start = (gchar *) addresses;
 487         current = start;
 488         last_blank = start;
 489
 490         while (*current != '\0') {
 491                 if ((start == current)&&((*current == ' ')||(*current == ',')||(*current == ';'))) {
 492                         start = g_utf8_next_char (start);
 493                         last_blank = current;
 494                 } else if ((*current == ',')||(*current == ';')) {
 495                         gchar *new_address = NULL;
 496                         new_address = g_strndup (start, current - last_blank);
 497                         result = g_slist_prepend (result, new_address);
 498                         start = g_utf8_next_char (current);
 499                         last_blank = start;
 500                 } else if (*current == '\"') {
 501                         if (current == start) {
 502                                 current = g_utf8_next_char (current);
 503                                 start = g_utf8_next_char (start);
 504                         }
 505                         while ((*current != '\"')&&(*current != '\0'))
 506                                 current = g_utf8_next_char (current);
 507                 }
 508
 509                 current = g_utf8_next_char (current);
 510         }
 511
 512         if (start != current) {
 513                 gchar *new_address = NULL;
 514                 new_address = g_strndup (start, current - last_blank);
 515                 result = g_slist_prepend (result, new_address);
 516         }
 517
 518         result = g_slist_reverse (result);
 519         return result;
 520
 521 }
 522
 523 void
 524 modest_text_utils_address_range_at_position (const gchar *recipients_list,
 525                                              gint position,
 526                                              gint *start,
 527                                              gint *end)
 528 {
 529         gchar *current = NULL;
 530         gint range_start = 0;
 531         gint range_end = 0;
 532         gint index;
 533         gboolean is_quoted = FALSE;
 534
 535         index = 0;
 536         for (current = (gchar *) recipients_list; *current != '\0'; current = g_utf8_find_next_char (current, NULL)) {
 537                 gunichar c = g_utf8_get_char (current);
 538
 539                 if ((c == ',') && (!is_quoted)) {
 540                         if (index < position) {
 541                                 range_start = index + 1;
 542                         } else {
 543                                 break;
 544                         }
 545                 } else if (c == '\"') {
 546                         is_quoted = !is_quoted;
 547                 } else if ((c == ' ') &&(range_start == index)) {
 548                         range_start ++;
 549                 }
 550                 index ++;
 551                 range_end = index;
 552         }
 553
 554         if (start)
 555                 *start = range_start;
 556         if (end)
 557                 *end = range_end;
 558 }
 559
 560
 561 /* ******************************************************************* */
 562 /* ************************* UTILIY FUNCTIONS ************************ */
 563 /* ******************************************************************* */
 564
 565 static GString *
 566 get_next_line (const gchar * b, const gsize blen, const gchar * iter)
 567 {
 568         GString *gs;
 569         const gchar *i0;
 570
 571         if (iter > b + blen)
 572                 return g_string_new("");
 573
 574         i0 = iter;
 575         while (iter[0]) {
 576                 if (iter[0] == '\n')
 577                         break;
 578                 iter++;
 579         }
 580         gs = g_string_new_len (i0, iter - i0);
 581         return gs;
 582 }
 583 static int
 584 get_indent_level (const char *l)
 585 {
 586         int indent = 0;
 587
 588         while (l[0]) {
 589                 if (l[0] == '>') {
 590                         indent++;
 591                         if (l[1] == ' ') {
 592                                 l++;
 593                         }
 594                 } else {
 595                         break;
 596                 }
 597                 l++;
 598
 599         }
 600
 601         /*      if we hit the signature marker "-- ", we return -(indent + 1). This
 602          *      stops reformatting.
 603          */
 604         if (strcmp (l, "-- ") == 0) {
 605                 return -1 - indent;
 606         } else {
 607                 return indent;
 608         }
 609 }
 610
 611 static void
 612 unquote_line (GString * l)
 613 {
 614         gchar *p;
 615
 616         p = l->str;
 617         while (p[0]) {
 618                 if (p[0] == '>') {
 619                         if (p[1] == ' ') {
 620                                 p++;
 621                         }
 622                 } else {
 623                         break;
 624                 }
 625                 p++;
 626         }
 627         g_string_erase (l, 0, p - l->str);
 628 }
 629
 630 static void
 631 append_quoted (GString * buf, int indent, const GString * str,
 632                const int cutpoint)
 633 {
 634         int i;
 635
 636         indent = indent < 0 ? abs (indent) - 1 : indent;
 637         for (i = 0; i <= indent; i++) {
 638                 g_string_append (buf, "> ");
 639         }
 640         if (cutpoint > 0) {
 641                 g_string_append_len (buf, str->str, cutpoint);
 642         } else {
 643                 g_string_append (buf, str->str);
 644         }
 645         g_string_append (buf, "\n");
 646 }
 647
 648 static int
 649 get_breakpoint_utf8 (const gchar * s, gint indent, const gint limit)
 650 {
 651         gint index = 0;
 652         const gchar *pos, *last;
 653         gunichar *uni;
 654
 655         indent = indent < 0 ? abs (indent) - 1 : indent;
 656
 657         last = NULL;
 658         pos = s;
 659         uni = g_utf8_to_ucs4_fast (s, -1, NULL);
 660         while (pos[0]) {
 661                 if ((index + 2 * indent > limit) && last) {
 662                         g_free (uni);
 663                         return last - s;
 664                 }
 665                 if (g_unichar_isspace (uni[index])) {
 666                         last = pos;
 667                 }
 668                 pos = g_utf8_next_char (pos);
 669                 index++;
 670         }
 671         g_free (uni);
 672         return strlen (s);
 673 }
 674
 675 static int
 676 get_breakpoint_ascii (const gchar * s, const gint indent, const gint limit)
 677 {
 678         gint i, last;
 679
 680         last = strlen (s);
 681         if (last + 2 * indent < limit)
 682                 return last;
 683
 684         for (i = strlen (s); i > 0; i--) {
 685                 if (s[i] == ' ') {
 686                         if (i + 2 * indent <= limit) {
 687                                 return i;
 688                         } else {
 689                                 last = i;
 690                         }
 691                 }
 692         }
 693         return last;
 694 }
 695
 696 static int
 697 get_breakpoint (const gchar * s, const gint indent, const gint limit)
 698 {
 699
 700         if (g_utf8_validate (s, -1, NULL)) {
 701                 return get_breakpoint_utf8 (s, indent, limit);
 702         } else {                /* assume ASCII */
 703                 //g_warning("invalid UTF-8 in msg");
 704                 return get_breakpoint_ascii (s, indent, limit);
 705         }
 706 }
 707
 708 static gchar *
 709 cite (const time_t sent_date, const gchar *from)
 710 {
 711         return g_strdup (_("mcen_ia_editor_original_message"));
 712 }
 713
 714 static gchar *
 715 quoted_attachments (GList *attachments)
 716 {
 717         GList *node = NULL;
 718         GString *result = g_string_new ("");
 719         for (node = attachments; node != NULL; node = g_list_next (node)) {
 720                 gchar *filename = (gchar *) node->data;
 721                 g_string_append_printf ( result, "%s %s\n", _("mcen_ia_editor_attach_filename"), filename);
 722         }
 723
 724         return g_string_free (result, FALSE);
 725
 726 }
 727
 728 static gchar *
 729 modest_text_utils_quote_plain_text (const gchar *text,
 730                                     const gchar *cite,
 731                                     const gchar *signature,
 732                                     GList *attachments,
 733                                     int limit)
 734 {
 735         const gchar *iter;
 736         gint indent, breakpoint, rem_indent = 0;
 737         GString *q, *l, *remaining;
 738         gsize len;
 739         gchar *attachments_string = NULL;
 740
 741         /* remaining will store the rest of the line if we have to break it */
 742         q = g_string_new ("\n");
 743         q = g_string_append (q, cite);
 744         q = g_string_append_c (q, '\n');
 745         remaining = g_string_new ("");
 746
 747         iter = text;
 748         len = strlen(text);
 749         do {
 750                 l = get_next_line (text, len, iter);
 751                 iter = iter + l->len + 1;
 752                 indent = get_indent_level (l->str);
 753                 unquote_line (l);
 754
 755                 if (remaining->len) {
 756                         if (l->len && indent == rem_indent) {
 757                                 g_string_prepend (l, " ");
 758                                 g_string_prepend (l, remaining->str);
 759                         } else {
 760                                 do {
 761                                         breakpoint =
 762                                                 get_breakpoint (remaining->str,
 763                                                                 rem_indent,
 764                                                                 limit);
 765                                         append_quoted (q, rem_indent,
 766                                                        remaining, breakpoint);
 767                                         g_string_erase (remaining, 0,
 768                                                         breakpoint);
 769                                         if (remaining->str[0] == ' ') {
 770                                                 g_string_erase (remaining, 0,
 771                                                                 1);
 772                                         }
 773                                 } while (remaining->len);
 774                         }
 775                 }
 776                 g_string_free (remaining, TRUE);
 777                 breakpoint = get_breakpoint (l->str, indent, limit);
 778                 remaining = g_string_new (l->str + breakpoint);
 779                 if (remaining->str[0] == ' ') {
 780                         g_string_erase (remaining, 0, 1);
 781                 }
 782                 rem_indent = indent;
 783                 append_quoted (q, indent, l, breakpoint);
 784                 g_string_free (l, TRUE);
 785         } while ((iter < text + len) || (remaining->str[0]));
 786
 787         attachments_string = quoted_attachments (attachments);
 788         q = g_string_append (q, attachments_string);
 789         g_free (attachments_string);
 790
 791         if (signature != NULL) {
 792                 q = g_string_append_c (q, '\n');
 793                 q = g_string_append (q, signature);
 794         }
 795
 796         return g_string_free (q, FALSE);
 797 }
 798
 799 static gchar*
 800 modest_text_utils_quote_html (const gchar *text,
 801                               const gchar *cite,
 802                               const gchar *signature,
 803                               GList *attachments,
 804                               int limit)
 805 {
 806         gchar *result = NULL;
 807         gchar *signature_result = NULL;
 808         const gchar *format = \
 809                 "<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\">\n" \
 810                 "<html>\n" \
 811                 "<body>\n" \
 812                 "<br/>%s<br/>" \
 813                 "<pre>%s<br/>%s<br/>%s</pre>\n" \
 814                 "</body>\n" \
 815                 "</html>\n";
 816         gchar *attachments_string = NULL;
 817         gchar *q_attachments_string = NULL;
 818         gchar *q_cite = NULL;
 819         gchar *html_text = NULL;
 820
 821         if (signature == NULL)
 822                 signature_result = g_strdup ("");
 823         else
 824                 signature_result = modest_text_utils_convert_to_html_body (signature);
 825
 826         attachments_string = quoted_attachments (attachments);
 827         q_attachments_string = modest_text_utils_convert_to_html_body (attachments_string);
 828         q_cite = modest_text_utils_convert_to_html_body (cite);
 829         html_text = modest_text_utils_convert_to_html_body (text);
 830         result = g_strdup_printf (format, signature_result, q_cite, html_text, q_attachments_string);
 831         g_free (q_cite);
 832         g_free (html_text);
 833         g_free (attachments_string);
 834         g_free (q_attachments_string);
 835         g_free (signature_result);
 836         return result;
 837 }
 838
 839 static gint
 840 cmp_offsets_reverse (const url_match_t *match1, const url_match_t *match2)
 841 {
 842         return match2->offset - match1->offset;
 843 }
 844
 845
 846 static GSList*
 847 get_url_matches (GString *txt)
 848 {
 849         regmatch_t rm;
 850         guint rv, i, offset = 0;
 851         GSList *match_list = NULL;
 852
 853         static url_match_pattern_t patterns[] = MAIL_VIEWER_URL_MATCH_PATTERNS;
 854         const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
 855
 856         /* initalize the regexps */
 857         for (i = 0; i != pattern_num; ++i) {
 858                 patterns[i].preg = g_slice_new0 (regex_t);
 859
 860                 /* this should not happen */
 861                 g_return_val_if_fail (regcomp (patterns[i].preg, patterns[i].regex,
 862                                                REG_ICASE|REG_EXTENDED|REG_NEWLINE) == 0, NULL);
 863         }
 864         /* find all the matches */
 865         for (i = 0; i != pattern_num; ++i) {
 866                 offset     = 0;
 867                 while (1) {
 868                         url_match_t *match;
 869                         gboolean is_submatch;
 870                         GSList *cursor;
 871
 872                         if ((rv = regexec (patterns[i].preg, txt->str + offset, 1, &rm, 0)) != 0) {
 873                                 g_return_val_if_fail (rv == REG_NOMATCH, NULL); /* this should not happen */
 874                                 break; /* try next regexp */
 875                         }
 876                         if (rm.rm_so == -1)
 877                                 break;
 878
 879                         is_submatch = FALSE;
 880                         /* check  old matches to see if this has already been matched */
 881                         cursor = match_list;
 882                         while (cursor && !is_submatch) {
 883                                 const url_match_t *old_match =
 884                                         (const url_match_t *) cursor->data;
 885                                 guint new_offset = offset + rm.rm_so;
 886                                 is_submatch = (new_offset >  old_match->offset &&
 887                                                new_offset <  old_match->offset + old_match->len);
 888                                 cursor = g_slist_next (cursor);
 889                         }
 890
 891                         if (!is_submatch) {
 892                                 /* make a list of our matches (<offset, len, prefix> tupels)*/
 893                                 match = g_slice_new (url_match_t);
 894                                 match->offset = offset + rm.rm_so;
 895                                 match->len    = rm.rm_eo - rm.rm_so;
 896                                 match->prefix = patterns[i].prefix;
 897                                 match_list = g_slist_prepend (match_list, match);
 898                         }
 899
 900                         offset += rm.rm_eo;
 901                 }
 902         }
 903
 904         for (i = 0; i != pattern_num; ++i) {
 905                 regfree (patterns[i].preg);
 906                 g_slice_free  (regex_t, patterns[i].preg);
 907         } /* don't free patterns itself -- it's static */
 908
 909         /* now sort the list, so the matches are in reverse order of occurence.
 910          * that way, we can do the replacements starting from the end, so we don't need
 911          * to recalculate the offsets
 912          */
 913         match_list = g_slist_sort (match_list,
 914                                    (GCompareFunc)cmp_offsets_reverse);
 915         return match_list;
 916 }
 917
 918
 919
 920 static void
 921 hyperlinkify_plain_text (GString *txt)
 922 {
 923         GSList *cursor;
 924         GSList *match_list = get_url_matches (txt);
 925
 926         /* we will work backwards, so the offsets stay valid */
 927         for (cursor = match_list; cursor; cursor = cursor->next) {
 928
 929                 url_match_t *match = (url_match_t*) cursor->data;
 930                 gchar *url  = g_strndup (txt->str + match->offset, match->len);
 931                 gchar *repl = NULL; /* replacement  */
 932
 933                 /* the prefix is NULL: use the one that is already there */
 934                 repl = g_strdup_printf ("<a href=\"%s%s\">%s</a>",
 935                                         match->prefix ? match->prefix : EMPTY_STRING,
 936                                         url, url);
 937
 938                 /* replace the old thing with our hyperlink
 939                  * replacement thing */
 940                 g_string_erase  (txt, match->offset, match->len);
 941                 g_string_insert (txt, match->offset, repl);
 942
 943                 g_free (url);
 944                 g_free (repl);
 945
 946                 g_slice_free (url_match_t, match);
 947         }
 948
 949         g_slist_free (match_list);
 950 }
 951
 952
 953
 954 gchar*
 955 modest_text_utils_get_display_address (gchar *address)
 956 {
 957         gchar *cursor;
 958
 959         if (!address)
 960                 return NULL;
 961
 962         g_return_val_if_fail (g_utf8_validate (address, -1, NULL), NULL);
 963
 964         g_strchug (address); /* remove leading whitespace */
 965
 966         /*  <email@address> from display name */
 967         cursor = g_strstr_len (address, strlen(address), "<");
 968         if (cursor == address) /* there's nothing else? leave it */
 969                 return address;
 970         if (cursor)
 971                 cursor[0]='\0';
 972
 973         /* remove (bla bla) from display name */
 974         cursor = g_strstr_len (address, strlen(address), "(");
 975         if (cursor == address) /* there's nothing else? leave it */
 976                 return address;
 977         if (cursor)
 978                 cursor[0]='\0';
 979
 980         g_strchomp (address); /* remove trailing whitespace */
 981
 982         return address;
 983 }
 984
 985 gchar *
 986 modest_text_utils_get_email_address (const gchar *full_address)
 987 {
 988         const gchar *left, *right;
 989
 990         if (!full_address)
 991                 return NULL;
 992
 993         g_return_val_if_fail (g_utf8_validate (full_address, -1, NULL), NULL);
 994
 995         left = g_strrstr_len (full_address, strlen(full_address), "<");
 996         if (left == NULL)
 997                 return g_strdup (full_address);
 998
 999         right = g_strstr_len (left, strlen(left), ">");
1000         if (right == NULL)
1001                 return g_strdup (full_address);
1002
1003         return g_strndup (left + 1, right - left - 1);
1004 }
1005
1006 gint
1007 modest_text_utils_get_subject_prefix_len (const gchar *sub)
1008 {
1009         gint i;
1010         static const gchar* prefix[] = {
1011                 "Re:", "RE:", "RV:", "re:"
1012                 "Fwd:", "FWD:", "FW:", "fwd:", "Fw:", "fw:", NULL
1013         };
1014
1015         if (!sub || (sub[0] != 'R' && sub[0] != 'F' && sub[0] != 'r' && sub[0] != 'f')) /* optimization */
1016                 return 0;
1017
1018         i = 0;
1019
1020         while (prefix[i]) {
1021                 if (g_str_has_prefix(sub, prefix[i])) {
1022                         int prefix_len = strlen(prefix[i]);
1023                         if (sub[prefix_len] == ' ')
1024                                 ++prefix_len; /* ignore space after prefix as well */
1025                         return prefix_len;
1026                 }
1027                 ++i;
1028         }
1029         return 0;
1030 }
1031
1032
1033 gint
1034 modest_text_utils_utf8_strcmp (const gchar* s1, const gchar *s2, gboolean insensitive)
1035 {
1036         gint result = 0;
1037         gchar *n1, *n2;
1038
1039         /* work even when s1 and/or s2 == NULL */
1040         if (G_UNLIKELY(s1 == s2))
1041                 return 0;
1042
1043         /* if it's not case sensitive */
1044         if (!insensitive)
1045                 return strcmp (s1 ? s1 : "", s2 ? s2 : "");
1046
1047         n1 = g_utf8_collate_key (s1 ? s1 : "", -1);
1048         n2 = g_utf8_collate_key (s2 ? s2 : "", -1);
1049
1050         result = strcmp (n1, n2);
1051
1052         g_free (n1);
1053         g_free (n2);
1054
1055         return result;
1056 }
1057
1058
1059 gchar*
1060 modest_text_utils_get_display_date (time_t date)
1061 {
1062         time_t now;
1063         static const guint BUF_SIZE = 64;
1064         static const guint ONE_DAY = 24 * 60 * 60; /* seconds in one day */
1065         gchar date_buf[BUF_SIZE];
1066         gchar today_buf [BUF_SIZE];
1067
1068         modest_text_utils_strftime (date_buf, BUF_SIZE, "%x", date);
1069
1070         now = time (NULL);
1071
1072         /* we check if the date is within the last 24h, if not, we don't
1073          * have to do the extra, expensive strftime, which was very visible
1074          * in the profiles.
1075          */
1076         if (abs(now - date) < ONE_DAY) {
1077
1078                 /* it's within the last 24 hours, but double check */
1079                 /* use the localized dates */
1080                 modest_text_utils_strftime (today_buf,  BUF_SIZE, "%x", now);
1081
1082                 /* if it's today, use the time instead */
1083                 if (strcmp (date_buf, today_buf) == 0)
1084                         modest_text_utils_strftime (date_buf, BUF_SIZE, "%X", date);
1085         }
1086
1087         return g_strdup(date_buf);
1088 }
1089
1090
1091 gboolean
1092 modest_text_utils_validate_folder_name (const gchar *folder_name)
1093 {
1094         /* based on http://msdn2.microsoft.com/en-us/library/aa365247.aspx,
1095          * with some extras */
1096
1097         guint len;
1098         gint i;
1099         const gchar **cursor = NULL;
1100         const gchar *forbidden_names[] = { /* windows does not like these */
1101                 "CON", "PRN", "AUX", "NUL", "COM1", "COM2", "COM3", "COM4", "COM5", "COM6",
1102                 "COM7", "COM8", "COM9", "LPT1", "LPT2", "LPT3", "LPT4", "LPT5", "LPT6", "LPT7", "LPT8", "LPT9",
1103                 ".", "..", NULL
1104         };
1105
1106         /* cannot be NULL */
1107         if (!folder_name)
1108                 return FALSE;
1109
1110         /* cannot be empty */
1111         len = strlen(folder_name);
1112         if (len == 0)
1113                 return FALSE;
1114
1115         /* cannot start or end with a space */
1116         if (g_ascii_isspace(folder_name[0]) || g_ascii_isspace(folder_name[len - 1]))
1117                 return FALSE;
1118
1119         /* cannot contain a forbidden char */
1120         for (i = 0; i < len; i++)
1121                 if (modest_text_utils_is_forbidden_char (folder_name[i], FOLDER_NAME_FORBIDDEN_CHARS))
1122                         return FALSE;
1123
1124         /* cannot contain a forbidden word */
1125         if (len <= 4) {
1126                 for (cursor = forbidden_names; cursor && *cursor; ++cursor) {
1127                         if (g_ascii_strcasecmp (folder_name, *cursor) == 0)
1128                                 return FALSE;
1129                 }
1130         }
1131         return TRUE; /* it's valid! */
1132 }
1133
1134
1135
1136 gboolean
1137 modest_text_utils_validate_domain_name (const gchar *domain)
1138 {
1139         gboolean valid = FALSE;
1140         regex_t rx;
1141         const gchar* domain_regex = "^[a-z0-9]([.]?[a-z0-9-])*[a-z0-9]$";
1142
1143         if (!domain)
1144                 return FALSE;
1145
1146         /* domain name: all alphanum or '-' or '.',
1147          * but beginning/ending in alphanum */
1148         if (regcomp (&rx, domain_regex, REG_ICASE|REG_EXTENDED|REG_NOSUB)) {
1149                 g_warning ("BUG: error in regexp");
1150                 return FALSE;
1151         }
1152
1153         valid = (regexec (&rx, domain, 1, NULL, 0) == 0);
1154         regfree (&rx);
1155
1156         return valid;
1157 }
1158
1159
1160
1161 gboolean
1162 modest_text_utils_validate_email_address (const gchar *email_address, const gchar **invalid_char_position)
1163 {
1164         int count = 0;
1165         const gchar *c = NULL, *domain = NULL;
1166         static gchar *rfc822_specials = "()<>@,;:\\\"[]&";
1167
1168         if (invalid_char_position != NULL)
1169                 *invalid_char_position = NULL;
1170
1171         /* check that the email adress contains exactly one @ */
1172         if (!strstr(email_address, "@") ||
1173                         (strstr(email_address, "@") != g_strrstr(email_address, "@")))
1174         {
1175                 return FALSE;
1176         }
1177
1178         /* first we validate the name portion (name@domain) */
1179         for (c = email_address;  *c;  c++) {
1180                 if (*c == '\"' &&
1181                     (c == email_address ||
1182                      *(c - 1) == '.' ||
1183                      *(c - 1) == '\"')) {
1184                         while (*++c) {
1185                                 if (*c == '\"')
1186                                         break;
1187                                 if (*c == '\\' && (*++c == ' '))
1188                                         continue;
1189                                 if (*c <= ' ' || *c >= 127)
1190                                         return FALSE;
1191                         }
1192                         if (!*c++)
1193                                 return FALSE;
1194                         if (*c == '@')
1195                                 break;
1196                         if (*c != '.')
1197                                 return FALSE;
1198                         continue;
1199                 }
1200                 if (*c == '@')
1201                         break;
1202                 if (*c <= ' ' || *c >= 127)
1203                         return FALSE;
1204                 if (strchr(rfc822_specials, *c)) {
1205                         if (invalid_char_position)
1206                                 *invalid_char_position = c;
1207                         return FALSE;
1208                 }
1209         }
1210         if (c == email_address || *(c - 1) == '.')
1211                 return FALSE;
1212
1213         /* next we validate the domain portion (name@domain) */
1214         if (!*(domain = ++c))
1215                 return FALSE;
1216         do {
1217                 if (*c == '.') {
1218                         if (c == domain || *(c - 1) == '.' || *(c + 1) == '\0')
1219                                 return FALSE;
1220                         count++;
1221                 }
1222                 if (*c <= ' ' || *c >= 127)
1223                         return FALSE;
1224                 if (strchr(rfc822_specials, *c)) {
1225                         if (invalid_char_position)
1226                                 *invalid_char_position = c;
1227                         return FALSE;
1228                 }
1229         } while (*++c);
1230
1231         return (count >= 1) ? TRUE : FALSE;
1232 }
1233
1234 gboolean
1235 modest_text_utils_validate_recipient (const gchar *recipient, const gchar **invalid_char_position)
1236 {
1237         gchar *stripped, *current;
1238         gchar *right_part;
1239         gboolean has_error = FALSE;
1240
1241         if (modest_text_utils_validate_email_address (recipient, invalid_char_position))
1242                 return TRUE;
1243         stripped = g_strdup (recipient);
1244         stripped = g_strstrip (stripped);
1245         current = stripped;
1246
1247         if (*current == '\0') {
1248                 g_free (stripped);
1249                 return FALSE;
1250         }
1251
1252         /* quoted string */
1253         if (*current == '\"') {
1254                 current = g_utf8_next_char (current);
1255                 has_error = TRUE;
1256                 for (; *current != '\0'; current = g_utf8_next_char (current)) {
1257                         if (*current == '\\') {
1258                                 /* TODO: This causes a warning, which breaks the build,
1259                                  * because a gchar cannot be < 0.
1260                                  * murrayc.
1261                                 if (current[1] <0) {
1262                                         has_error = TRUE;
1263                                         break;
1264                                 }
1265                                 */
1266                         } else if (*current == '\"') {
1267                                 has_error = FALSE;
1268                                 current = g_utf8_next_char (current);
1269                                 break;
1270                         }
1271                 }
1272         } else {
1273                 has_error = TRUE;
1274                 for (current = stripped ; *current != '\0'; current = g_utf8_next_char (current)) {
1275                         if (*current == '<') {
1276                                 has_error = FALSE;
1277                                 break;
1278                         }
1279                 }
1280         }
1281
1282         if (has_error) {
1283                 g_free (stripped);
1284                 return FALSE;
1285         }
1286
1287         right_part = g_strdup (current);
1288         g_free (stripped);
1289         right_part = g_strstrip (right_part);
1290
1291         if (g_str_has_prefix (right_part, "<") &&
1292             g_str_has_suffix (right_part, ">")) {
1293                 gchar *address;
1294                 gboolean valid;
1295
1296                 address = g_strndup (right_part+1, strlen (right_part) - 2);
1297                 g_free (right_part);
1298                 valid = modest_text_utils_validate_email_address (address, invalid_char_position);
1299                 g_free (address);
1300                 return valid;
1301         } else {
1302                 g_free (right_part);
1303                 return FALSE;
1304         }
1305 }
1306
1307
1308 gchar *
1309 modest_text_utils_get_display_size (guint64 size)
1310 {
1311         const guint KB=1024;
1312         const guint MB=1024 * KB;
1313         const guint GB=1024 * MB;
1314
1315         if (size == 0)
1316                 return g_strdup_printf(_FM("sfil_li_size_kb"), 0);
1317         if (0 < size && size < KB)
1318                 return g_strdup_printf (_FM("sfil_li_size_kb"), 1);
1319         else if (KB <= size && size < 100 * KB)
1320                 return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), size / KB);
1321         else if (100*KB <= size && size < MB)
1322                 return g_strdup_printf (_FM("sfil_li_size_100kb_1mb"), (float) size / MB);
1323         else if (MB <= size && size < 10*MB)
1324                 return g_strdup_printf (_FM("sfil_li_size_1mb_10mb"), (float) size / MB);
1325         else if (10*MB <= size && size < GB)
1326                 return g_strdup_printf (_FM("sfil_li_size_10mb_1gb"), size / MB);
1327         else
1328                 return g_strdup_printf (_FM("sfil_li_size_1gb_or_greater"), (float) size / GB);
1329 }
1330
1331 static gchar *
1332 get_email_from_address (const gchar * address)
1333 {
1334         gchar *left_limit, *right_limit;
1335
1336         left_limit = strstr (address, "<");
1337         right_limit = g_strrstr (address, ">");
1338
1339         if ((left_limit == NULL)||(right_limit == NULL)|| (left_limit > right_limit))
1340                 return g_strdup (address);
1341         else
1342                 return g_strndup (left_limit + 1, (right_limit - left_limit) - 1);
1343 }
1344
1345 gchar *
1346 modest_text_utils_get_color_string (GdkColor *color)
1347 {
1348
1349         return g_strdup_printf ("#%x%x%x%x%x%x%x%x%x%x%x%x",
1350                                 (color->red >> 12)   & 0xf, (color->red >> 8)   & 0xf,
1351                                 (color->red >>  4)   & 0xf, (color->red)        & 0xf,
1352                                 (color->green >> 12) & 0xf, (color->green >> 8) & 0xf,
1353                                 (color->green >>  4) & 0xf, (color->green)      & 0xf,
1354                                 (color->blue >> 12)  & 0xf, (color->blue >> 8)  & 0xf,
1355                                 (color->blue >>  4)  & 0xf, (color->blue)       & 0xf);
1356 }
1357
1358 gchar *
1359 modest_text_utils_text_buffer_get_text (GtkTextBuffer *buffer)
1360 {
1361         GtkTextIter start, end;
1362         gchar *slice, *current;
1363         GString *result = g_string_new ("");
1364
1365         g_return_val_if_fail (GTK_IS_TEXT_BUFFER (buffer), NULL);
1366
1367         gtk_text_buffer_get_start_iter (buffer, &start);
1368         gtk_text_buffer_get_end_iter (buffer, &end);
1369
1370         slice = gtk_text_buffer_get_slice (buffer, &start, &end, FALSE);
1371         current = slice;
1372
1373         while (current && current != '\0') {
1374                 if (g_utf8_get_char (current) == 0xFFFC) {
1375                         result = g_string_append_c (result, ' ');
1376                         current = g_utf8_next_char (current);
1377                 } else {
1378                         gchar *next = g_utf8_strchr (current, -1, 0xFFFC);
1379                         if (next == NULL) {
1380                                 result = g_string_append (result, current);
1381                         } else {
1382                                 result = g_string_append_len (result, current, next - current);
1383                         }
1384                         current = next;
1385                 }
1386         }
1387         g_free (slice);
1388
1389         return g_string_free (result, FALSE);
1390
1391 }
1392
1393 gboolean
1394 modest_text_utils_is_forbidden_char (const gchar character,
1395                                      ModestTextUtilsForbiddenCharType type)
1396 {
1397         gint i, len;
1398         const gchar *forbidden_chars = NULL;
1399
1400         /* We need to get the length in the switch because the
1401            compiler needs to know the size at compile time */
1402         switch (type) {
1403         case ACCOUNT_TITLE_FORBIDDEN_CHARS:
1404                 forbidden_chars = account_title_forbidden_chars;
1405                 len = G_N_ELEMENTS (account_title_forbidden_chars);
1406                 break;
1407         case FOLDER_NAME_FORBIDDEN_CHARS:
1408                 forbidden_chars = folder_name_forbidden_chars;
1409                 len = G_N_ELEMENTS (folder_name_forbidden_chars);
1410                 break;
1411         case USER_NAME_FORBIDDEN_NAMES:
1412                 forbidden_chars = user_name_forbidden_chars;
1413                 len = G_N_ELEMENTS (user_name_forbidden_chars);
1414                 break;
1415         default:
1416                 g_return_val_if_reached (TRUE);
1417         }
1418
1419         for (i = 0; i < len ; i++)
1420                 if (forbidden_chars[i] == character)
1421                         return TRUE;
1422
1423         return FALSE; /* it's valid! */
1424 }