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 #include <glib.h>
  32 #include <string.h>
  33 #include <stdlib.h>
  34 #include <glib/gi18n.h>
  35 #include <regex.h>
  36 #include <modest-tny-platform-factory.h>
  37 #include <modest-text-utils.h>
  38 #include <modest-runtime.h>
  39
  40
  41 #ifdef HAVE_CONFIG_H
  42 #include <config.h>
  43 #endif /*HAVE_CONFIG_H */
  44
  45 /* defines */
  46 #define FORWARD_STRING _("-----Forwarded Message-----")
  47 #define FROM_STRING _("From:")
  48 #define SENT_STRING _("Sent:")
  49 #define TO_STRING _("To:")
  50 #define SUBJECT_STRING _("Subject:")
  51 #define EMPTY_STRING ""
  52
  53 /*
  54  * do the hyperlinkification only for texts < 50 Kb,
  55  * as it's quite slow. Without this, e.g. mail with
  56  * an uuencoded part (which is not recognized as attachment,
  57  * will hang modest
  58  */
  59 #define HYPERLINKIFY_MAX_LENGTH (1024*50)
  60
  61 /*
  62  * we need these regexps to find URLs in plain text e-mails
  63  */
  64 typedef struct _url_match_pattern_t url_match_pattern_t;
  65 struct _url_match_pattern_t {
  66         gchar   *regex;
  67         regex_t *preg;
  68         gchar   *prefix;
  69 };
  70
  71 typedef struct _url_match_t url_match_t;
  72 struct _url_match_t {
  73         guint offset;
  74         guint len;
  75         const gchar* prefix;
  76 };
  77
  78 #define MAIL_VIEWER_URL_MATCH_PATTERNS  {                               \
  79         { "(file|rtsp|http|ftp|https)://[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]+[-A-Za-z0-9_$%&=?/~#]",\
  80           NULL, NULL },\
  81         { "www\\.[-a-z0-9.]+[-a-z0-9](:[0-9]*)?(/[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]*[^]}\\),?!;:\"]?)?",\
  82           NULL, "http://" },\
  83         { "ftp\\.[-a-z0-9.]+[-a-z0-9](:[0-9]*)?(/[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]*[^]}\\),?!;:\"]?)?",\
  84           NULL, "ftp://" },\
  85         { "(voipto|callto|chatto|jabberto|xmpp):[-_a-z@0-9.\\+]+", \
  86            NULL, NULL},                                             \
  87         { "mailto:[-_a-z0-9.\\+]+@[-_a-z0-9.]+",                    \
  88           NULL, NULL},\
  89         { "[-_a-z0-9.\\+]+@[-_a-z0-9.]+",\
  90           NULL, "mailto:"}\
  91         }
  92
  93 /* private */
  94 static gchar*   cite                    (const time_t sent_date, const gchar *from);
  95 static void     hyperlinkify_plain_text (GString *txt);
  96 static gint     cmp_offsets_reverse     (const url_match_t *match1, const url_match_t *match2);
  97 static GSList*  get_url_matches         (GString *txt);
  98
  99 static GString* get_next_line           (const char *b, const gsize blen, const gchar * iter);
 100 static int      get_indent_level        (const char *l);
 101 static void     unquote_line            (GString * l);
 102 static void     append_quoted           (GString * buf, const int indent, const GString * str,
 103                                          const int cutpoint);
 104 static int      get_breakpoint_utf8     (const gchar * s, const gint indent, const gint limit);
 105 static int      get_breakpoint_ascii    (const gchar * s, const gint indent, const gint limit);
 106 static int      get_breakpoint          (const gchar * s, const gint indent, const gint limit);
 107
 108 static gchar*   modest_text_utils_quote_plain_text (const gchar *text,
 109                                                     const gchar *cite,
 110                                                     const gchar *signature,
 111                                                     GList *attachments,
 112                                                     int limit);
 113
 114 static gchar*   modest_text_utils_quote_html       (const gchar *text,
 115                                                     const gchar *cite,
 116                                                     const gchar *signature,
 117                                                     GList *attachments,
 118                                                     int limit);
 119 static gchar*   get_email_from_address (const gchar *address);
 120
 121
 122 /* ******************************************************************* */
 123 /* ************************* PUBLIC FUNCTIONS ************************ */
 124 /* ******************************************************************* */
 125
 126 gchar *
 127 modest_text_utils_quote (const gchar *text,
 128                          const gchar *content_type,
 129                          const gchar *signature,
 130                          const gchar *from,
 131                          const time_t sent_date,
 132                          GList *attachments,
 133                          int limit)
 134 {
 135         gchar *retval, *cited;
 136
 137         g_return_val_if_fail (text, NULL);
 138         g_return_val_if_fail (content_type, NULL);
 139
 140         cited = cite (sent_date, from);
 141
 142         if (content_type && strcmp (content_type, "text/html") == 0)
 143                 /* TODO: extract the <body> of the HTML and pass it to
 144                    the function */
 145                 retval = modest_text_utils_quote_html (text, cited, signature, attachments, limit);
 146         else
 147                 retval = modest_text_utils_quote_plain_text (text, cited, signature, attachments, limit);
 148
 149         g_free (cited);
 150
 151         return retval;
 152 }
 153
 154
 155 gchar *
 156 modest_text_utils_cite (const gchar *text,
 157                         const gchar *content_type,
 158                         const gchar *signature,
 159                         const gchar *from,
 160                         time_t sent_date)
 161 {
 162         gchar *retval;
 163         gchar *tmp_sig;
 164
 165         g_return_val_if_fail (text, NULL);
 166         g_return_val_if_fail (content_type, NULL);
 167
 168         if (!signature)
 169                 retval = g_strdup ("");
 170         else if (!strcmp(content_type, "text/html")) {
 171                 tmp_sig = g_strconcat ("\n", signature, NULL);
 172                 retval = modest_text_utils_convert_to_html_body(tmp_sig);
 173                 g_free (tmp_sig);
 174         } else {
 175                 retval = g_strconcat ("\n", signature, NULL);
 176         }
 177
 178         return retval;
 179 }
 180
 181 static gchar *
 182 forward_cite (const gchar *from,
 183                     const gchar *sent,
 184                     const gchar *to,
 185                     const gchar *subject)
 186 {
 187         return g_strdup_printf ("%s\n%s %s\n%s %s\n%s %s\n%s %s\n",
 188                                 FORWARD_STRING,
 189                                 FROM_STRING, (from)?from:"",
 190                                 SENT_STRING, sent,
 191                                 TO_STRING, (to)?to:"",
 192                                 SUBJECT_STRING, (subject)?subject:"");
 193 }
 194
 195 gchar *
 196 modest_text_utils_inline (const gchar *text,
 197                           const gchar *content_type,
 198                           const gchar *signature,
 199                           const gchar *from,
 200                           time_t sent_date,
 201                           const gchar *to,
 202                           const gchar *subject)
 203 {
 204         gchar sent_str[101];
 205         gchar *cited;
 206         gchar *retval;
 207
 208         g_return_val_if_fail (text, NULL);
 209         g_return_val_if_fail (content_type, NULL);
 210
 211         modest_text_utils_strftime (sent_str, 100, "%c", sent_date);
 212
 213         cited = forward_cite (from, sent_str, to, subject);
 214
 215         if (content_type && strcmp (content_type, "text/html") == 0)
 216                 retval = modest_text_utils_quote_html (text, cited, signature, NULL, 80);
 217         else
 218                 retval = modest_text_utils_quote_plain_text (text, cited, signature, NULL, 80);
 219
 220         g_free (cited);
 221         return retval;
 222 }
 223
 224 /* just to prevent warnings:
 225  * warning: `%x' yields only last 2 digits of year in some locales
 226  */
 227 gsize
 228 modest_text_utils_strftime(char *s, gsize max, const char *fmt, time_t timet)
 229 {
 230         struct tm tm;
 231
 232         /* does not work on old maemo glib:
 233          *   g_date_set_time_t (&date, timet);
 234          */
 235         localtime_r (&timet, &tm);
 236         return strftime(s, max, fmt, &tm);
 237 }
 238
 239 gchar *
 240 modest_text_utils_derived_subject (const gchar *subject, const gchar *prefix)
 241 {
 242         gchar *tmp;
 243
 244         g_return_val_if_fail (prefix, NULL);
 245
 246         if (!subject)
 247                 return g_strdup (prefix);
 248
 249         tmp = g_strchug (g_strdup (subject));
 250
 251         if (!strncmp (tmp, prefix, strlen (prefix))) {
 252                 return tmp;
 253         } else {
 254                 g_free (tmp);
 255                 return g_strdup_printf ("%s %s", prefix, subject);
 256         }
 257 }
 258
 259 gchar*
 260 modest_text_utils_remove_address (const gchar *address_list, const gchar *address)
 261 {
 262         gchar *dup, *token, *ptr, *result;
 263         GString *filtered_emails;
 264         gchar *email_address;
 265
 266         g_return_val_if_fail (address_list, NULL);
 267
 268         if (!address)
 269                 return g_strdup (address_list);
 270
 271         email_address = get_email_from_address (address);
 272
 273         /* search for substring */
 274         if (!strstr ((const char *) address_list, (const char *) email_address)) {
 275                 g_free (email_address);
 276                 return g_strdup (address_list);
 277         }
 278
 279         dup = g_strdup (address_list);
 280         filtered_emails = g_string_new (NULL);
 281
 282         token = strtok_r (dup, ",", &ptr);
 283
 284         while (token != NULL) {
 285                 /* Add to list if not found */
 286                 if (!strstr ((const char *) token, (const char *) email_address)) {
 287                         if (filtered_emails->len == 0)
 288                                 g_string_append_printf (filtered_emails, "%s", g_strstrip (token));
 289                         else
 290                                 g_string_append_printf (filtered_emails, ",%s", g_strstrip (token));
 291                 }
 292                 token = strtok_r (NULL, ",", &ptr);
 293         }
 294         result = filtered_emails->str;
 295
 296         /* Clean */
 297         g_free (email_address);
 298         g_free (dup);
 299         g_string_free (filtered_emails, FALSE);
 300
 301         return result;
 302 }
 303
 304 static void
 305 modest_text_utils_convert_buffer_to_html (GString *html, const gchar *data)
 306 {
 307         guint            i;
 308         gboolean        space_seen = FALSE;
 309         gsize           len;
 310         guint           break_dist = 0; /* distance since last break point */
 311
 312         len = strlen (data);
 313
 314         /* replace with special html chars where needed*/
 315         for (i = 0; i != len; ++i)  {
 316                 char kar = data[i];
 317
 318                 if (space_seen && kar != ' ') {
 319                         g_string_append_c (html, ' ');
 320                         space_seen = FALSE;
 321                 }
 322
 323                 /* we artificially insert a breakpoint (newline)
 324                  * after 256, to make sure our lines are not so long
 325                  * they will DOS the regexping later
 326                  */
 327                 if (++break_dist == 256) {
 328                         g_string_append_c (html, '\n');
 329                         break_dist = 0;
 330                 }
 331
 332                 switch (kar) {
 333                 case 0:  break; /* ignore embedded \0s */
 334                 case '<'  : g_string_append (html, "&lt;");   break;
 335                 case '>'  : g_string_append (html, "&gt;");   break;
 336                 case '&'  : g_string_append (html, "&amp;");  break;
 337                 case '"'  : g_string_append (html, "&quot;");  break;
 338                 case '\'' : g_string_append (html, "&apos;"); break;
 339                 case '\n' : g_string_append (html, "<br>\n");              break_dist= 0; break;
 340                 case '\t' : g_string_append (html, "&nbsp;&nbsp;&nbsp; "); break_dist=0; break; /* note the space at the end*/
 341                 case ' ':
 342                         break_dist = 0;
 343                         if (space_seen) { /* second space in a row */
 344                                 g_string_append (html, "&nbsp; ");
 345                                 space_seen = FALSE;
 346                         } else
 347                                 space_seen = TRUE;
 348                         break;
 349                 default:
 350                         g_string_append_c (html, kar);
 351                 }
 352         }
 353 }
 354
 355 gchar*
 356 modest_text_utils_convert_to_html (const gchar *data)
 357 {
 358         GString         *html;
 359         gsize           len;
 360
 361         if (!data)
 362                 return NULL;
 363
 364         len = strlen (data);
 365         html = g_string_sized_new (1.5 * len);  /* just a  guess... */
 366
 367         g_string_append_printf (html,
 368                                 "<html><head>"
 369                                 "<meta http-equiv=\"content-type\" content=\"text/html; charset=utf8\">"
 370                                 "</head>"
 371                                 "<body>");
 372
 373         modest_text_utils_convert_buffer_to_html (html, data);
 374
 375         g_string_append (html, "</body></html>");
 376
 377         if (len <= HYPERLINKIFY_MAX_LENGTH)
 378                 hyperlinkify_plain_text (html);
 379
 380         return g_string_free (html, FALSE);
 381 }
 382
 383 gchar *
 384 modest_text_utils_convert_to_html_body (const gchar *data)
 385 {
 386         GString         *html;
 387         gsize           len;
 388
 389         if (!data)
 390                 return NULL;
 391
 392         len = strlen (data);
 393         html = g_string_sized_new (1.5 * len);  /* just a  guess... */
 394
 395         modest_text_utils_convert_buffer_to_html (html, data);
 396
 397         if (len < HYPERLINKIFY_MAX_LENGTH)
 398                 hyperlinkify_plain_text (html);
 399
 400         return g_string_free (html, FALSE);
 401 }
 402
 403 void
 404 modest_text_utils_get_addresses_indexes (const gchar *addresses, GSList **start_indexes, GSList **end_indexes)
 405 {
 406         gchar *current, *start, *last_blank;
 407         gint start_offset = 0, current_offset = 0;
 408
 409         g_return_if_fail (start_indexes != NULL);
 410         g_return_if_fail (end_indexes != NULL);
 411
 412         start = (gchar *) addresses;
 413         current = start;
 414         last_blank = start;
 415
 416         while (*current != '\0') {
 417                 if ((start == current)&&((*current == ' ')||(*current == ',')||(*current == ';'))) {
 418                         start = g_utf8_next_char (start);
 419                         start_offset++;
 420                         last_blank = current;
 421                 } else if ((*current == ',')||(*current == ';')) {
 422                         gint *start_index, *end_index;
 423                         start_index = g_new0(gint, 1);
 424                         end_index = g_new0(gint, 1);
 425                         *start_index = start_offset;
 426                         *end_index = current_offset;
 427                         *start_indexes = g_slist_prepend (*start_indexes, start_index);
 428                         *end_indexes = g_slist_prepend (*end_indexes, end_index);
 429                         start = g_utf8_next_char (current);
 430                         start_offset = current_offset + 1;
 431                         last_blank = start;
 432                 } else if (*current == '"') {
 433                         current = g_utf8_next_char (current);
 434                         current_offset ++;
 435                         while ((*current != '"')&&(*current != '\0')) {
 436                                 current = g_utf8_next_char (current);
 437                                 current_offset ++;
 438                         }
 439                 }
 440
 441                 current = g_utf8_next_char (current);
 442                 current_offset ++;
 443         }
 444
 445         if (start != current) {
 446                         gint *start_index, *end_index;
 447                         start_index = g_new0(gint, 1);
 448                         end_index = g_new0(gint, 1);
 449                         *start_index = start_offset;
 450                         *end_index = current_offset;
 451                         *start_indexes = g_slist_prepend (*start_indexes, start_index);
 452                         *end_indexes = g_slist_prepend (*end_indexes, end_index);
 453         }
 454
 455         *start_indexes = g_slist_reverse (*start_indexes);
 456         *end_indexes = g_slist_reverse (*end_indexes);
 457
 458         return;
 459 }
 460
 461 GSList *
 462 modest_text_utils_split_addresses_list (const gchar *addresses)
 463 {
 464         gchar *current, *start, *last_blank;
 465         GSList *result = NULL;
 466
 467         start = (gchar *) addresses;
 468         current = start;
 469         last_blank = start;
 470
 471         while (*current != '\0') {
 472                 if ((start == current)&&((*current == ' ')||(*current == ',')||(*current == ';'))) {
 473                         start = g_utf8_next_char (start);
 474                         last_blank = current;
 475                 } else if ((*current == ',')||(*current == ';')) {
 476                         gchar *new_address = NULL;
 477                         new_address = g_strndup (start, current - last_blank);
 478                         result = g_slist_prepend (result, new_address);
 479                         start = g_utf8_next_char (current);
 480                         last_blank = start;
 481                 } else if (*current == '\"') {
 482                         if (current == start) {
 483                                 current = g_utf8_next_char (current);
 484                                 start = g_utf8_next_char (start);
 485                         }
 486                         while ((*current != '\"')&&(*current != '\0'))
 487                                 current = g_utf8_next_char (current);
 488                 }
 489
 490                 current = g_utf8_next_char (current);
 491         }
 492
 493         if (start != current) {
 494                 gchar *new_address = NULL;
 495                 new_address = g_strndup (start, current - last_blank);
 496                 result = g_slist_prepend (result, new_address);
 497         }
 498
 499         result = g_slist_reverse (result);
 500         return result;
 501
 502 }
 503
 504 void
 505 modest_text_utils_address_range_at_position (const gchar *recipients_list,
 506                                              gint position,
 507                                              gint *start,
 508                                              gint *end)
 509 {
 510         gchar *current = NULL;
 511         gint range_start = 0;
 512         gint range_end = 0;
 513         gint index;
 514         gboolean is_quoted = FALSE;
 515
 516         index = 0;
 517         for (current = (gchar *) recipients_list; *current != '\0'; current = g_utf8_find_next_char (current, NULL)) {
 518                 gunichar c = g_utf8_get_char (current);
 519
 520                 if ((c == ',') && (!is_quoted)) {
 521                         if (index < position) {
 522                                 range_start = index + 1;
 523                         } else {
 524                                 break;
 525                         }
 526                 } else if (c == '\"') {
 527                         is_quoted = !is_quoted;
 528                 } else if ((c == ' ') &&(range_start == index)) {
 529                         range_start ++;
 530                 }
 531                 index ++;
 532                 range_end = index;
 533         }
 534
 535         if (start)
 536                 *start = range_start;
 537         if (end)
 538                 *end = range_end;
 539 }
 540
 541
 542 /* ******************************************************************* */
 543 /* ************************* UTILIY FUNCTIONS ************************ */
 544 /* ******************************************************************* */
 545
 546 static GString *
 547 get_next_line (const gchar * b, const gsize blen, const gchar * iter)
 548 {
 549         GString *gs;
 550         const gchar *i0;
 551
 552         if (iter > b + blen)
 553                 return g_string_new("");
 554
 555         i0 = iter;
 556         while (iter[0]) {
 557                 if (iter[0] == '\n')
 558                         break;
 559                 iter++;
 560         }
 561         gs = g_string_new_len (i0, iter - i0);
 562         return gs;
 563 }
 564 static int
 565 get_indent_level (const char *l)
 566 {
 567         int indent = 0;
 568
 569         while (l[0]) {
 570                 if (l[0] == '>') {
 571                         indent++;
 572                         if (l[1] == ' ') {
 573                                 l++;
 574                         }
 575                 } else {
 576                         break;
 577                 }
 578                 l++;
 579
 580         }
 581
 582         /*      if we hit the signature marker "-- ", we return -(indent + 1). This
 583          *      stops reformatting.
 584          */
 585         if (strcmp (l, "-- ") == 0) {
 586                 return -1 - indent;
 587         } else {
 588                 return indent;
 589         }
 590 }
 591
 592 static void
 593 unquote_line (GString * l)
 594 {
 595         gchar *p;
 596
 597         p = l->str;
 598         while (p[0]) {
 599                 if (p[0] == '>') {
 600                         if (p[1] == ' ') {
 601                                 p++;
 602                         }
 603                 } else {
 604                         break;
 605                 }
 606                 p++;
 607         }
 608         g_string_erase (l, 0, p - l->str);
 609 }
 610
 611 static void
 612 append_quoted (GString * buf, int indent, const GString * str,
 613                const int cutpoint)
 614 {
 615         int i;
 616
 617         indent = indent < 0 ? abs (indent) - 1 : indent;
 618         for (i = 0; i <= indent; i++) {
 619                 g_string_append (buf, "> ");
 620         }
 621         if (cutpoint > 0) {
 622                 g_string_append_len (buf, str->str, cutpoint);
 623         } else {
 624                 g_string_append (buf, str->str);
 625         }
 626         g_string_append (buf, "\n");
 627 }
 628
 629 static int
 630 get_breakpoint_utf8 (const gchar * s, gint indent, const gint limit)
 631 {
 632         gint index = 0;
 633         const gchar *pos, *last;
 634         gunichar *uni;
 635
 636         indent = indent < 0 ? abs (indent) - 1 : indent;
 637
 638         last = NULL;
 639         pos = s;
 640         uni = g_utf8_to_ucs4_fast (s, -1, NULL);
 641         while (pos[0]) {
 642                 if ((index + 2 * indent > limit) && last) {
 643                         g_free (uni);
 644                         return last - s;
 645                 }
 646                 if (g_unichar_isspace (uni[index])) {
 647                         last = pos;
 648                 }
 649                 pos = g_utf8_next_char (pos);
 650                 index++;
 651         }
 652         g_free (uni);
 653         return strlen (s);
 654 }
 655
 656 static int
 657 get_breakpoint_ascii (const gchar * s, const gint indent, const gint limit)
 658 {
 659         gint i, last;
 660
 661         last = strlen (s);
 662         if (last + 2 * indent < limit)
 663                 return last;
 664
 665         for (i = strlen (s); i > 0; i--) {
 666                 if (s[i] == ' ') {
 667                         if (i + 2 * indent <= limit) {
 668                                 return i;
 669                         } else {
 670                                 last = i;
 671                         }
 672                 }
 673         }
 674         return last;
 675 }
 676
 677 static int
 678 get_breakpoint (const gchar * s, const gint indent, const gint limit)
 679 {
 680
 681         if (g_utf8_validate (s, -1, NULL)) {
 682                 return get_breakpoint_utf8 (s, indent, limit);
 683         } else {                /* assume ASCII */
 684                 //g_warning("invalid UTF-8 in msg");
 685                 return get_breakpoint_ascii (s, indent, limit);
 686         }
 687 }
 688
 689 static gchar *
 690 cite (const time_t sent_date, const gchar *from)
 691 {
 692         return g_strdup (_("mcen_ia_editor_original_message"));
 693 }
 694
 695 static gchar *
 696 quoted_attachments (GList *attachments)
 697 {
 698         GList *node = NULL;
 699         GString *result = g_string_new ("");
 700         for (node = attachments; node != NULL; node = g_list_next (node)) {
 701                 gchar *filename = (gchar *) node->data;
 702                 g_string_append_printf ( result, "%s %s\n", _("mcen_ia_editor_attach_filename"), filename);
 703         }
 704
 705         return g_string_free (result, FALSE);
 706
 707 }
 708
 709 static gchar *
 710 modest_text_utils_quote_plain_text (const gchar *text,
 711                                     const gchar *cite,
 712                                     const gchar *signature,
 713                                     GList *attachments,
 714                                     int limit)
 715 {
 716         const gchar *iter;
 717         gint indent, breakpoint, rem_indent = 0;
 718         GString *q, *l, *remaining;
 719         gsize len;
 720         gchar *attachments_string = NULL;
 721
 722         /* remaining will store the rest of the line if we have to break it */
 723         q = g_string_new ("\n");
 724         q = g_string_append (q, cite);
 725         q = g_string_append_c (q, '\n');
 726         remaining = g_string_new ("");
 727
 728         iter = text;
 729         len = strlen(text);
 730         do {
 731                 l = get_next_line (text, len, iter);
 732                 iter = iter + l->len + 1;
 733                 indent = get_indent_level (l->str);
 734                 unquote_line (l);
 735
 736                 if (remaining->len) {
 737                         if (l->len && indent == rem_indent) {
 738                                 g_string_prepend (l, " ");
 739                                 g_string_prepend (l, remaining->str);
 740                         } else {
 741                                 do {
 742                                         breakpoint =
 743                                                 get_breakpoint (remaining->str,
 744                                                                 rem_indent,
 745                                                                 limit);
 746                                         append_quoted (q, rem_indent,
 747                                                        remaining, breakpoint);
 748                                         g_string_erase (remaining, 0,
 749                                                         breakpoint);
 750                                         if (remaining->str[0] == ' ') {
 751                                                 g_string_erase (remaining, 0,
 752                                                                 1);
 753                                         }
 754                                 } while (remaining->len);
 755                         }
 756                 }
 757                 g_string_free (remaining, TRUE);
 758                 breakpoint = get_breakpoint (l->str, indent, limit);
 759                 remaining = g_string_new (l->str + breakpoint);
 760                 if (remaining->str[0] == ' ') {
 761                         g_string_erase (remaining, 0, 1);
 762                 }
 763                 rem_indent = indent;
 764                 append_quoted (q, indent, l, breakpoint);
 765                 g_string_free (l, TRUE);
 766         } while ((iter < text + len) || (remaining->str[0]));
 767
 768         attachments_string = quoted_attachments (attachments);
 769         q = g_string_append (q, attachments_string);
 770         g_free (attachments_string);
 771
 772         if (signature != NULL) {
 773                 q = g_string_append_c (q, '\n');
 774                 q = g_string_append (q, signature);
 775         }
 776
 777         return g_string_free (q, FALSE);
 778 }
 779
 780 static gchar*
 781 modest_text_utils_quote_html (const gchar *text,
 782                               const gchar *cite,
 783                               const gchar *signature,
 784                               GList *attachments,
 785                               int limit)
 786 {
 787         gchar *result = NULL;
 788         gchar *signature_result = NULL;
 789         const gchar *format = \
 790                 "<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\">\n" \
 791                 "<html>\n" \
 792                 "<body>\n" \
 793                 "<br/>%s<br/>" \
 794                 "<pre>%s<br/>%s<br/>%s</pre>\n" \
 795                 "</body>\n" \
 796                 "</html>\n";
 797         gchar *attachments_string = NULL;
 798         gchar *q_attachments_string = NULL;
 799         gchar *q_cite = NULL;
 800         gchar *html_text = NULL;
 801
 802         if (signature == NULL)
 803                 signature_result = g_strdup ("");
 804         else
 805                 signature_result = modest_text_utils_convert_to_html_body (signature);
 806
 807         attachments_string = quoted_attachments (attachments);
 808         q_attachments_string = modest_text_utils_convert_to_html_body (attachments_string);
 809         q_cite = modest_text_utils_convert_to_html_body (cite);
 810         html_text = modest_text_utils_convert_to_html_body (text);
 811         result = g_strdup_printf (format, signature_result, q_cite, html_text, q_attachments_string);
 812         g_free (q_cite);
 813         g_free (html_text);
 814         g_free (attachments_string);
 815         g_free (q_attachments_string);
 816         g_free (signature_result);
 817         return result;
 818 }
 819
 820 static gint
 821 cmp_offsets_reverse (const url_match_t *match1, const url_match_t *match2)
 822 {
 823         return match2->offset - match1->offset;
 824 }
 825
 826
 827 static GSList*
 828 get_url_matches (GString *txt)
 829 {
 830         regmatch_t rm;
 831         guint rv, i, offset = 0;
 832         GSList *match_list = NULL;
 833
 834         static url_match_pattern_t patterns[] = MAIL_VIEWER_URL_MATCH_PATTERNS;
 835         const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
 836
 837         /* initalize the regexps */
 838         for (i = 0; i != pattern_num; ++i) {
 839                 patterns[i].preg = g_slice_new0 (regex_t);
 840
 841                 /* this should not happen */
 842                 g_return_val_if_fail (regcomp (patterns[i].preg, patterns[i].regex,
 843                                                REG_ICASE|REG_EXTENDED|REG_NEWLINE) == 0, NULL);
 844         }
 845         /* find all the matches */
 846         for (i = 0; i != pattern_num; ++i) {
 847                 offset     = 0;
 848                 while (1) {
 849                         url_match_t *match;
 850                         gboolean is_submatch;
 851                         GSList *cursor;
 852
 853                         if ((rv = regexec (patterns[i].preg, txt->str + offset, 1, &rm, 0)) != 0) {
 854                                 g_return_val_if_fail (rv == REG_NOMATCH, NULL); /* this should not happen */
 855                                 break; /* try next regexp */
 856                         }
 857                         if (rm.rm_so == -1)
 858                                 break;
 859
 860                         is_submatch = FALSE;
 861                         /* check  old matches to see if this has already been matched */
 862                         cursor = match_list;
 863                         while (cursor && !is_submatch) {
 864                                 const url_match_t *old_match =
 865                                         (const url_match_t *) cursor->data;
 866                                 guint new_offset = offset + rm.rm_so;
 867                                 is_submatch = (new_offset >  old_match->offset &&
 868                                                new_offset <  old_match->offset + old_match->len);
 869                                 cursor = g_slist_next (cursor);
 870                         }
 871
 872                         if (!is_submatch) {
 873                                 /* make a list of our matches (<offset, len, prefix> tupels)*/
 874                                 match = g_slice_new (url_match_t);
 875                                 match->offset = offset + rm.rm_so;
 876                                 match->len    = rm.rm_eo - rm.rm_so;
 877                                 match->prefix = patterns[i].prefix;
 878                                 g_warning ("<%d, %d, %s>",  match->offset, match->len, match->prefix);
 879                                 match_list = g_slist_prepend (match_list, match);
 880                         }
 881
 882                         offset += rm.rm_eo;
 883                 }
 884         }
 885
 886         for (i = 0; i != pattern_num; ++i) {
 887                 regfree (patterns[i].preg);
 888                 g_slice_free  (regex_t, patterns[i].preg);
 889         } /* don't free patterns itself -- it's static */
 890
 891         /* now sort the list, so the matches are in reverse order of occurence.
 892          * that way, we can do the replacements starting from the end, so we don't need
 893          * to recalculate the offsets
 894          */
 895         match_list = g_slist_sort (match_list,
 896                                    (GCompareFunc)cmp_offsets_reverse);
 897         return match_list;
 898 }
 899
 900
 901
 902 static void
 903 hyperlinkify_plain_text (GString *txt)
 904 {
 905         GSList *cursor;
 906         GSList *match_list = get_url_matches (txt);
 907
 908         /* we will work backwards, so the offsets stay valid */
 909         for (cursor = match_list; cursor; cursor = cursor->next) {
 910
 911                 url_match_t *match = (url_match_t*) cursor->data;
 912                 gchar *url  = g_strndup (txt->str + match->offset, match->len);
 913                 gchar *repl = NULL; /* replacement  */
 914
 915                 /* the prefix is NULL: use the one that is already there */
 916                 repl = g_strdup_printf ("<a href=\"%s%s\">%s</a>",
 917                                         match->prefix ? match->prefix : EMPTY_STRING,
 918                                         url, url);
 919
 920                 /* replace the old thing with our hyperlink
 921                  * replacement thing */
 922                 g_string_erase  (txt, match->offset, match->len);
 923                 g_string_insert (txt, match->offset, repl);
 924
 925                 g_free (url);
 926                 g_free (repl);
 927
 928                 g_slice_free (url_match_t, match);
 929         }
 930
 931         g_slist_free (match_list);
 932 }
 933
 934
 935
 936 gchar*
 937 modest_text_utils_get_display_address (gchar *address)
 938 {
 939         gchar *cursor;
 940
 941         if (!address)
 942                 return NULL;
 943
 944         g_return_val_if_fail (g_utf8_validate (address, -1, NULL), NULL);
 945
 946         g_strchug (address); /* remove leading whitespace */
 947
 948         /*  <email@address> from display name */
 949         cursor = g_strstr_len (address, strlen(address), "<");
 950         if (cursor == address) /* there's nothing else? leave it */
 951                 return address;
 952         if (cursor)
 953                 cursor[0]='\0';
 954
 955         /* remove (bla bla) from display name */
 956         cursor = g_strstr_len (address, strlen(address), "(");
 957         if (cursor == address) /* there's nothing else? leave it */
 958                 return address;
 959         if (cursor)
 960                 cursor[0]='\0';
 961
 962         g_strchomp (address); /* remove trailing whitespace */
 963
 964         return address;
 965 }
 966
 967
 968
 969 gint
 970 modest_text_utils_get_subject_prefix_len (const gchar *sub)
 971 {
 972         gint i;
 973         static const gchar* prefix[] = {
 974                 "Re:", "RE:", "RV:", "re:"
 975                 "Fwd:", "FWD:", "FW:", "fwd:", "Fw:", "fw:", NULL
 976         };
 977
 978         if (!sub || (sub[0] != 'R' && sub[0] != 'F' && sub[0] != 'r' && sub[0] != 'f')) /* optimization */
 979                 return 0;
 980
 981         i = 0;
 982
 983         while (prefix[i]) {
 984                 if (g_str_has_prefix(sub, prefix[i])) {
 985                         int prefix_len = strlen(prefix[i]);
 986                         if (sub[prefix_len] == ' ')
 987                                 ++prefix_len; /* ignore space after prefix as well */
 988                         return prefix_len;
 989                 }
 990                 ++i;
 991         }
 992         return 0;
 993 }
 994
 995
 996 gint
 997 modest_text_utils_utf8_strcmp (const gchar* s1, const gchar *s2, gboolean insensitive)
 998 {
 999         gint result = 0;
1000         gchar *n1, *n2;
1001
1002         /* work even when s1 and/or s2 == NULL */
1003         if (G_UNLIKELY(s1 == s2))
1004                 return 0;
1005
1006         /* if it's not case sensitive */
1007         if (!insensitive)
1008                 return strcmp (s1 ? s1 : "", s2 ? s2 : "");
1009
1010         n1 = g_utf8_collate_key (s1 ? s1 : "", -1);
1011         n2 = g_utf8_collate_key (s2 ? s2 : "", -1);
1012
1013         result = strcmp (n1, n2);
1014
1015         g_free (n1);
1016         g_free (n2);
1017
1018         return result;
1019 }
1020
1021
1022 gchar*
1023 modest_text_utils_get_display_date (time_t date)
1024 {
1025         time_t now;
1026         static const guint BUF_SIZE = 64;
1027         static const guint ONE_DAY = 24 * 60 * 60; /* seconds in one day */
1028         gchar date_buf[BUF_SIZE];
1029         gchar today_buf [BUF_SIZE];
1030
1031         modest_text_utils_strftime (date_buf, BUF_SIZE, "%x", date);
1032
1033         now = time (NULL);
1034
1035         /* we check if the date is within the last 24h, if not, we don't
1036          * have to do the extra, expensive strftime, which was very visible
1037          * in the profiles.
1038          */
1039         if (abs(now - date) < ONE_DAY) {
1040
1041                 /* it's within the last 24 hours, but double check */
1042                 /* use the localized dates */
1043                 modest_text_utils_strftime (today_buf,  BUF_SIZE, "%x", now);
1044
1045                 /* if it's today, use the time instead */
1046                 if (strcmp (date_buf, today_buf) == 0)
1047                         modest_text_utils_strftime (date_buf, BUF_SIZE, "%X", date);
1048         }
1049
1050         return g_strdup(date_buf);
1051 }
1052
1053 gboolean
1054 modest_text_utils_validate_email_address (const gchar *email_address, const gchar **invalid_char_position)
1055 {
1056         int count = 0;
1057         const gchar *c = NULL, *domain = NULL;
1058         static gchar *rfc822_specials = "()<>@,;:\\\"[]&";
1059
1060         if (invalid_char_position != NULL)
1061                 *invalid_char_position = NULL;
1062
1063         /* first we validate the name portion (name@domain) */
1064         for (c = email_address;  *c;  c++) {
1065                 if (*c == '\"' &&
1066                     (c == email_address ||
1067                      *(c - 1) == '.' ||
1068                      *(c - 1) == '\"')) {
1069                         while (*++c) {
1070                                 if (*c == '\"')
1071                                         break;
1072                                 if (*c == '\\' && (*++c == ' '))
1073                                         continue;
1074                                 if (*c <= ' ' || *c >= 127)
1075                                         return FALSE;
1076                         }
1077                         if (!*c++)
1078                                 return FALSE;
1079                         if (*c == '@')
1080                                 break;
1081                         if (*c != '.')
1082                                 return FALSE;
1083                         continue;
1084                 }
1085                 if (*c == '@')
1086                         break;
1087                 if (*c <= ' ' || *c >= 127)
1088                         return FALSE;
1089                 if (strchr(rfc822_specials, *c)) {
1090                         if (invalid_char_position)
1091                                 *invalid_char_position = c;
1092                         return FALSE;
1093                 }
1094         }
1095         if (c == email_address || *(c - 1) == '.')
1096                 return FALSE;
1097
1098         /* next we validate the domain portion (name@domain) */
1099         if (!*(domain = ++c))
1100                 return FALSE;
1101         do {
1102                 if (*c == '.') {
1103                         if (c == domain || *(c - 1) == '.')
1104                                 return FALSE;
1105                         count++;
1106                 }
1107                 if (*c <= ' ' || *c >= 127)
1108                         return FALSE;
1109                 if (strchr(rfc822_specials, *c)) {
1110                         if (invalid_char_position)
1111                                 *invalid_char_position = c;
1112                         return FALSE;
1113                 }
1114         } while (*++c);
1115
1116         return (count >= 1) ? TRUE : FALSE;
1117 }
1118
1119 gboolean
1120 modest_text_utils_validate_recipient (const gchar *recipient, const gchar **invalid_char_position)
1121 {
1122         gchar *stripped, *current;
1123         gchar *right_part;
1124         gboolean has_error = FALSE;
1125
1126         if (modest_text_utils_validate_email_address (recipient, invalid_char_position))
1127                 return TRUE;
1128         stripped = g_strdup (recipient);
1129         stripped = g_strstrip (stripped);
1130         current = stripped;
1131
1132         if (*current == '\0') {
1133                 g_free (stripped);
1134                 return FALSE;
1135         }
1136
1137         /* quoted string */
1138         if (*current == '\"') {
1139                 current = g_utf8_next_char (current);
1140                 has_error = TRUE;
1141                 for (; *current != '\0'; current = g_utf8_next_char (current)) {
1142                         if (*current == '\\') {
1143                                 /* TODO: This causes a warning, which breaks the build,
1144                                  * because a gchar cannot be < 0.
1145                                  * murrayc.
1146                                 if (current[1] <0) {
1147                                         has_error = TRUE;
1148                                         break;
1149                                 }
1150                                 */
1151                         } else if (*current == '\"') {
1152                                 has_error = FALSE;
1153                                 current = g_utf8_next_char (current);
1154                                 break;
1155                         }
1156                 }
1157         } else {
1158                 has_error = TRUE;
1159                 for (current = stripped ; *current != '\0'; current = g_utf8_next_char (current)) {
1160                         if (*current == '<') {
1161                                 has_error = FALSE;
1162                                 break;
1163                         }
1164                 }
1165         }
1166
1167         if (has_error) {
1168                 g_free (stripped);
1169                 return FALSE;
1170         }
1171
1172         right_part = g_strdup (current);
1173         g_free (stripped);
1174         right_part = g_strstrip (right_part);
1175
1176         if (g_str_has_prefix (right_part, "<") &&
1177             g_str_has_suffix (right_part, ">")) {
1178                 gchar *address;
1179                 gboolean valid;
1180
1181                 address = g_strndup (right_part+1, strlen (right_part) - 2);
1182                 g_free (right_part);
1183                 valid = modest_text_utils_validate_email_address (address, invalid_char_position);
1184                 g_free (address);
1185                 return valid;
1186         } else {
1187                 g_free (right_part);
1188                 return FALSE;
1189         }
1190 }
1191
1192
1193 gchar *
1194 modest_text_utils_get_display_size (guint64 size)
1195 {
1196         const guint KB=1024;
1197         const guint MB=1024 * KB;
1198         const guint GB=1024 * MB;
1199
1200         if (size == 0)
1201                 return g_strdup_printf(_FM("sfil_li_size_kb"), 0);
1202         if (0 < size && size < KB)
1203                 return g_strdup_printf (_FM("sfil_li_size_kb"), 1);
1204         else if (KB <= size && size < 100 * KB)
1205                 return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), size / KB);
1206         else if (100*KB <= size && size < MB)
1207                 return g_strdup_printf (_FM("sfil_li_size_100kb_1mb"), (float) size / MB);
1208         else if (MB <= size && size < 10*MB)
1209                 return g_strdup_printf (_FM("sfil_li_size_1mb_10mb"), (float) size / MB);
1210         else if (10*MB <= size && size < GB)
1211                 return g_strdup_printf (_FM("sfil_li_size_10mb_1gb"), size / MB);
1212         else
1213                 return g_strdup_printf (_FM("sfil_li_size_1gb_or_greater"), (float) size / GB);
1214 }
1215
1216 static gchar *
1217 get_email_from_address (const gchar * address)
1218 {
1219         gchar *left_limit, *right_limit;
1220
1221         left_limit = strstr (address, "<");
1222         right_limit = g_strrstr (address, ">");
1223
1224         if ((left_limit == NULL)||(right_limit == NULL)|| (left_limit > right_limit))
1225                 return g_strdup (address);
1226         else
1227                 return g_strndup (left_limit + 1, (right_limit - left_limit) - 1);
1228 }
1229
1230 gchar *
1231 modest_text_utils_get_color_string (GdkColor *color)
1232 {
1233
1234         return g_strdup_printf ("#%x%x%x%x%x%x%x%x%x%x%x%x",
1235                                 (color->red >> 12)   & 0xf, (color->red >> 8)   & 0xf,
1236                                 (color->red >>  4)   & 0xf, (color->red)        & 0xf,
1237                                 (color->green >> 12) & 0xf, (color->green >> 8) & 0xf,
1238                                 (color->green >>  4) & 0xf, (color->green)      & 0xf,
1239                                 (color->blue >> 12)  & 0xf, (color->blue >> 8)  & 0xf,
1240                                 (color->blue >>  4)  & 0xf, (color->blue)       & 0xf);
1241 }
1242
1243 gchar *
1244 modest_text_utils_text_buffer_get_text (GtkTextBuffer *buffer)
1245 {
1246         GtkTextIter start, end;
1247         gchar *slice, *current;
1248         GString *result = g_string_new ("");
1249
1250         g_return_val_if_fail (GTK_IS_TEXT_BUFFER (buffer), NULL);
1251
1252         gtk_text_buffer_get_start_iter (buffer, &start);
1253         gtk_text_buffer_get_end_iter (buffer, &end);
1254
1255         slice = gtk_text_buffer_get_slice (buffer, &start, &end, FALSE);
1256         current = slice;
1257
1258         while (current && current != '\0') {
1259                 if (g_utf8_get_char (current) == 0xFFFC) {
1260                         result = g_string_append_c (result, ' ');
1261                         current = g_utf8_next_char (current);
1262                 } else {
1263                         gchar *next = g_utf8_strchr (current, -1, 0xFFFC);
1264                         if (next == NULL) {
1265                                 result = g_string_append (result, current);
1266                         } else {
1267                                 result = g_string_append_len (result, current, next - current);
1268                         }
1269                         current = next;
1270                 }
1271         }
1272         g_free (slice);
1273
1274         return g_string_free (result, FALSE);
1275
1276 }