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