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