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