git.maemo.org Git - modest/blob - src/modest-text-utils.c

   1 /* Copyright (c) 2006, Nokia Corporation
   2  * All rights reserved.
   3  *
   4  * Redistribution and use in source and binary forms, with or without
   5  * modification, are permitted provided that the following conditions are
   6  * met:
   7  *
   8  * * Redistributions of source code must retain the above copyright
   9  *   notice, this list of conditions and the following disclaimer.
  10  * * Redistributions in binary form must reproduce the above copyright
  11  *   notice, this list of conditions and the following disclaimer in the
  12  *   documentation and/or other materials provided with the distribution.
  13  * * Neither the name of the Nokia Corporation nor the names of its
  14  *   contributors may be used to endorse or promote products derived from
  15  *   this software without specific prior written permission.
  16  *
  17  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS
  18  * IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
  19  * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
  20  * PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER
  21  * OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
  22  * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
  23  * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
  24  * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
  25  * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
  26  * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
  27  * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  28  */
  29
  30
  31 #include <glib.h>
  32 #include <string.h>
  33 #include <stdlib.h>
  34 #include <glib/gi18n.h>
  35 #include <regex.h>
  36 #include <modest-tny-platform-factory.h>
  37 #include <modest-text-utils.h>
  38 #include <modest-runtime.h>
  39
  40
  41 #ifdef HAVE_CONFIG_H
  42 #include <config.h>
  43 #endif /*HAVE_CONFIG_H */
  44
  45 /* defines */
  46 #define FORWARD_STRING _("-----Forwarded Message-----")
  47 #define FROM_STRING _("From:")
  48 #define SENT_STRING _("Sent:")
  49 #define TO_STRING _("To:")
  50 #define SUBJECT_STRING _("Subject:")
  51 #define EMPTY_STRING ""
  52
  53 /*
  54  * we need these regexps to find URLs in plain text e-mails
  55  */
  56 typedef struct _url_match_pattern_t url_match_pattern_t;
  57 struct _url_match_pattern_t {
  58         gchar   *regex;
  59         regex_t *preg;
  60         gchar   *prefix;
  61 };
  62
  63 typedef struct _url_match_t url_match_t;
  64 struct _url_match_t {
  65         guint offset;
  66         guint len;
  67         const gchar* prefix;
  68 };
  69
  70 #define MAIL_VIEWER_URL_MATCH_PATTERNS  {                               \
  71         { "(file|rtsp|http|ftp|https)://[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]+[-A-Za-z0-9_$%&=?/~#]",\
  72           NULL, NULL },\
  73         { "www\\.[-a-z0-9.]+[-a-z0-9](:[0-9]*)?(/[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]*[^]}\\),?!;:\"]?)?",\
  74           NULL, "http://" },\
  75         { "ftp\\.[-a-z0-9.]+[-a-z0-9](:[0-9]*)?(/[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]*[^]}\\),?!;:\"]?)?",\
  76           NULL, "ftp://" },\
  77         { "(voipto|callto|chatto|jabberto|xmpp):[-_a-z@0-9.\\+]+", \
  78            NULL, NULL},                                             \
  79         { "mailto:[-_a-z0-9.\\+]+@[-_a-z0-9.]+",                    \
  80           NULL, NULL},\
  81         { "[-_a-z0-9.\\+]+@[-_a-z0-9.]+",\
  82           NULL, "mailto:"}\
  83         }
  84
  85 /* private */
  86 static gchar*   cite                    (const time_t sent_date, const gchar *from);
  87 static void     hyperlinkify_plain_text (GString *txt);
  88 static gint     cmp_offsets_reverse     (const url_match_t *match1, const url_match_t *match2);
  89 static void     chk_partial_match       (const url_match_t *match, guint* offset);
  90 static GSList*  get_url_matches         (GString *txt);
  91
  92 static GString* get_next_line           (const char *b, const gsize blen, const gchar * iter);
  93 static int      get_indent_level        (const char *l);
  94 static void     unquote_line            (GString * l);
  95 static void     append_quoted           (GString * buf, const int indent, const GString * str,
  96                                          const int cutpoint);
  97 static int      get_breakpoint_utf8     (const gchar * s, const gint indent, const gint limit);
  98 static int      get_breakpoint_ascii    (const gchar * s, const gint indent, const gint limit);
  99 static int      get_breakpoint          (const gchar * s, const gint indent, const gint limit);
 100
 101 static gchar*   modest_text_utils_quote_plain_text (const gchar *text,
 102                                                     const gchar *cite,
 103                                                     const gchar *signature,
 104                                                     GList *attachments,
 105                                                     int limit);
 106
 107 static gchar*   modest_text_utils_quote_html       (const gchar *text,
 108                                                     const gchar *cite,
 109                                                     const gchar *signature,
 110                                                     GList *attachments,
 111                                                     int limit);
 112 static gchar*   get_email_from_address (const gchar *address);
 113
 114
 115 /* ******************************************************************* */
 116 /* ************************* PUBLIC FUNCTIONS ************************ */
 117 /* ******************************************************************* */
 118
 119 gchar *
 120 modest_text_utils_quote (const gchar *text,
 121                          const gchar *content_type,
 122                          const gchar *signature,
 123                          const gchar *from,
 124                          const time_t sent_date,
 125                          GList *attachments,
 126                          int limit)
 127 {
 128         gchar *retval, *cited;
 129
 130         g_return_val_if_fail (text, NULL);
 131         g_return_val_if_fail (content_type, NULL);
 132
 133         cited = cite (sent_date, from);
 134
 135         if (content_type && strcmp (content_type, "text/html") == 0)
 136                 /* TODO: extract the <body> of the HTML and pass it to
 137                    the function */
 138                 retval = modest_text_utils_quote_html (text, cited, signature, attachments, limit);
 139         else
 140                 retval = modest_text_utils_quote_plain_text (text, cited, signature, attachments, limit);
 141
 142         g_free (cited);
 143
 144         return retval;
 145 }
 146
 147
 148 gchar *
 149 modest_text_utils_cite (const gchar *text,
 150                         const gchar *content_type,
 151                         const gchar *signature,
 152                         const gchar *from,
 153                         time_t sent_date)
 154 {
 155         gchar *retval;
 156         gchar *tmp_sig;
 157
 158         g_return_val_if_fail (text, NULL);
 159         g_return_val_if_fail (content_type, NULL);
 160
 161         if (!signature)
 162                 retval = g_strdup ("");
 163         else if (!strcmp(content_type, "text/html")) {
 164                 tmp_sig = g_strconcat ("\n", signature, NULL);
 165                 retval = modest_text_utils_convert_to_html_body(tmp_sig);
 166                 g_free (tmp_sig);
 167         } else {
 168                 retval = g_strconcat ("\n", signature, NULL);
 169         }
 170
 171         return retval;
 172 }
 173
 174 static gchar *
 175 forward_cite (const gchar *from,
 176                     const gchar *sent,
 177                     const gchar *to,
 178                     const gchar *subject)
 179 {
 180         return g_strdup_printf ("%s\n%s %s\n%s %s\n%s %s\n%s %s\n",
 181                                 FORWARD_STRING,
 182                                 FROM_STRING, (from)?from:"",
 183                                 SENT_STRING, sent,
 184                                 TO_STRING, (to)?to:"",
 185                                 SUBJECT_STRING, (subject)?subject:"");
 186 }
 187
 188 gchar *
 189 modest_text_utils_inline (const gchar *text,
 190                           const gchar *content_type,
 191                           const gchar *signature,
 192                           const gchar *from,
 193                           time_t sent_date,
 194                           const gchar *to,
 195                           const gchar *subject)
 196 {
 197         gchar sent_str[101];
 198         gchar *cited;
 199         gchar *retval;
 200
 201         g_return_val_if_fail (text, NULL);
 202         g_return_val_if_fail (content_type, NULL);
 203
 204         modest_text_utils_strftime (sent_str, 100, "%c", sent_date);
 205
 206         cited = forward_cite (from, sent_str, to, subject);
 207
 208         if (content_type && strcmp (content_type, "text/html") == 0)
 209                 retval = modest_text_utils_quote_html (text, cited, signature, NULL, 80);
 210         else
 211                 retval = modest_text_utils_quote_plain_text (text, cited, signature, NULL, 80);
 212
 213         g_free (cited);
 214         return retval;
 215 }
 216
 217 /* just to prevent warnings:
 218  * warning: `%x' yields only last 2 digits of year in some locales
 219  */
 220 gsize
 221 modest_text_utils_strftime(char *s, gsize max, const char *fmt, time_t timet)
 222 {
 223         struct tm tm;
 224
 225         /* does not work on old maemo glib:
 226          *   g_date_set_time_t (&date, timet);
 227          */
 228         localtime_r (&timet, &tm);
 229         return strftime(s, max, fmt, &tm);
 230 }
 231
 232 gchar *
 233 modest_text_utils_derived_subject (const gchar *subject, const gchar *prefix)
 234 {
 235         gchar *tmp;
 236
 237         g_return_val_if_fail (prefix, NULL);
 238
 239         if (!subject)
 240                 return g_strdup (prefix);
 241
 242         tmp = g_strchug (g_strdup (subject));
 243
 244         if (!strncmp (tmp, prefix, strlen (prefix))) {
 245                 return tmp;
 246         } else {
 247                 g_free (tmp);
 248                 return g_strdup_printf ("%s %s", prefix, subject);
 249         }
 250 }
 251
 252 gchar*
 253 modest_text_utils_remove_address (const gchar *address_list, const gchar *address)
 254 {
 255         gchar *dup, *token, *ptr, *result;
 256         GString *filtered_emails;
 257         gchar *email_address;
 258
 259         g_return_val_if_fail (address_list, NULL);
 260
 261         if (!address)
 262                 return g_strdup (address_list);
 263
 264         email_address = get_email_from_address (address);
 265
 266         /* search for substring */
 267         if (!strstr ((const char *) address_list, (const char *) email_address)) {
 268                 g_free (email_address);
 269                 return g_strdup (address_list);
 270         }
 271
 272         dup = g_strdup (address_list);
 273         filtered_emails = g_string_new (NULL);
 274
 275         token = strtok_r (dup, ",", &ptr);
 276
 277         while (token != NULL) {
 278                 /* Add to list if not found */
 279                 if (!strstr ((const char *) token, (const char *) email_address)) {
 280                         if (filtered_emails->len == 0)
 281                                 g_string_append_printf (filtered_emails, "%s", g_strstrip (token));
 282                         else
 283                                 g_string_append_printf (filtered_emails, ",%s", g_strstrip (token));
 284                 }
 285                 token = strtok_r (NULL, ",", &ptr);
 286         }
 287         result = filtered_emails->str;
 288
 289         /* Clean */
 290         g_free (email_address);
 291         g_free (dup);
 292         g_string_free (filtered_emails, FALSE);
 293
 294         return result;
 295 }
 296
 297 static void
 298 modest_text_utils_convert_buffer_to_html (GString *html, const gchar *data)
 299 {
 300         guint            i;
 301         gboolean        space_seen = FALSE;
 302         gsize           len;
 303         guint           break_dist = 0; /* distance since last break point */
 304
 305         len = strlen (data);
 306
 307         /* replace with special html chars where needed*/
 308         for (i = 0; i != len; ++i)  {
 309                 char kar = data[i];
 310
 311                 if (space_seen && kar != ' ') {
 312                         g_string_append_c (html, ' ');
 313                         space_seen = FALSE;
 314                 }
 315
 316                 /* we artificially insert a breakpoint (newline)
 317                  * after 256, to make sure our lines are not so long
 318                  * they will DOS the regexping later
 319                  */
 320                 if (++break_dist == 256) {
 321                         g_string_append_c (html, '\n');
 322                         break_dist = 0;
 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_dist= 0; break;
 333                 case '\t' : g_string_append (html, "&nbsp;&nbsp;&nbsp; "); break_dist=0; break; /* note the space at the end*/
 334                 case ' ':
 335                         break_dist = 0;
 336                         if (space_seen) { /* second space in a row */
 337                                 g_string_append (html, "&nbsp; ");
 338                                 space_seen = FALSE;
 339                         } else
 340                                 space_seen = TRUE;
 341                         break;
 342                 default:
 343                         g_string_append_c (html, kar);
 344                 }
 345         }
 346 }
 347
 348 gchar*
 349 modest_text_utils_convert_to_html (const gchar *data)
 350 {
 351         GString         *html;
 352         gsize           len;
 353
 354         if (!data)
 355                 return NULL;
 356
 357         len = strlen (data);
 358         html = g_string_sized_new (1.5 * len);  /* just a  guess... */
 359
 360         g_string_append_printf (html,
 361                                 "<html><head>"
 362                                 "<meta http-equiv=\"content-type\" content=\"text/html; charset=utf8\">"
 363                                 "</head>"
 364                                 "<body>");
 365
 366         modest_text_utils_convert_buffer_to_html (html, data);
 367
 368         g_string_append (html, "</body></html>");
 369         hyperlinkify_plain_text (html);
 370
 371         return g_string_free (html, FALSE);
 372 }
 373
 374 gchar *
 375 modest_text_utils_convert_to_html_body (const gchar *data)
 376 {
 377         GString         *html;
 378         gsize           len;
 379
 380         if (!data)
 381                 return NULL;
 382
 383         len = strlen (data);
 384         html = g_string_sized_new (1.5 * len);  /* just a  guess... */
 385
 386         modest_text_utils_convert_buffer_to_html (html, data);
 387
 388         hyperlinkify_plain_text (html);
 389
 390         return g_string_free (html, FALSE);
 391 }
 392
 393 void
 394 modest_text_utils_get_addresses_indexes (const gchar *addresses, GSList **start_indexes, GSList **end_indexes)
 395 {
 396         gchar *current, *start, *last_blank;
 397         gint start_offset = 0, current_offset = 0;
 398
 399         g_return_if_fail (start_indexes != NULL);
 400         g_return_if_fail (end_indexes != NULL);
 401
 402         start = (gchar *) addresses;
 403         current = start;
 404         last_blank = start;
 405
 406         while (*current != '\0') {
 407                 if ((start == current)&&((*current == ' ')||(*current == ',')||(*current == ';'))) {
 408                         start = g_utf8_next_char (start);
 409                         start_offset++;
 410                         last_blank = current;
 411                 } else if ((*current == ',')||(*current == ';')) {
 412                         gint *start_index, *end_index;
 413                         start_index = g_new0(gint, 1);
 414                         end_index = g_new0(gint, 1);
 415                         *start_index = start_offset;
 416                         *end_index = current_offset;
 417                         *start_indexes = g_slist_prepend (*start_indexes, start_index);
 418                         *end_indexes = g_slist_prepend (*end_indexes, end_index);
 419                         start = g_utf8_next_char (current);
 420                         start_offset = current_offset + 1;
 421                         last_blank = start;
 422                 } else if (*current == '"') {
 423                         current = g_utf8_next_char (current);
 424                         current_offset ++;
 425                         while ((*current != '"')&&(*current != '\0')) {
 426                                 current = g_utf8_next_char (current);
 427                                 current_offset ++;
 428                         }
 429                 }
 430
 431                 current = g_utf8_next_char (current);
 432                 current_offset ++;
 433         }
 434
 435         if (start != current) {
 436                         gint *start_index, *end_index;
 437                         start_index = g_new0(gint, 1);
 438                         end_index = g_new0(gint, 1);
 439                         *start_index = start_offset;
 440                         *end_index = current_offset;
 441                         *start_indexes = g_slist_prepend (*start_indexes, start_index);
 442                         *end_indexes = g_slist_prepend (*end_indexes, end_index);
 443         }
 444
 445         *start_indexes = g_slist_reverse (*start_indexes);
 446         *end_indexes = g_slist_reverse (*end_indexes);
 447
 448         return;
 449 }
 450
 451 GSList *
 452 modest_text_utils_split_addresses_list (const gchar *addresses)
 453 {
 454         gchar *current, *start, *last_blank;
 455         GSList *result = NULL;
 456
 457         start = (gchar *) addresses;
 458         current = start;
 459         last_blank = start;
 460
 461         while (*current != '\0') {
 462                 if ((start == current)&&((*current == ' ')||(*current == ',')||(*current == ';'))) {
 463                         start = g_utf8_next_char (start);
 464                         last_blank = current;
 465                 } else if ((*current == ',')||(*current == ';')) {
 466                         gchar *new_address = NULL;
 467                         new_address = g_strndup (start, current - last_blank);
 468                         result = g_slist_prepend (result, new_address);
 469                         start = g_utf8_next_char (current);
 470                         last_blank = start;
 471                 } else if (*current == '\"') {
 472                         if (current == start) {
 473                                 current = g_utf8_next_char (current);
 474                                 start = g_utf8_next_char (start);
 475                         }
 476                         while ((*current != '\"')&&(*current != '\0'))
 477                                 current = g_utf8_next_char (current);
 478                 }
 479
 480                 current = g_utf8_next_char (current);
 481         }
 482
 483         if (start != current) {
 484                 gchar *new_address = NULL;
 485                 new_address = g_strndup (start, current - last_blank);
 486                 result = g_slist_prepend (result, new_address);
 487         }
 488
 489         result = g_slist_reverse (result);
 490         return result;
 491
 492 }
 493
 494 void
 495 modest_text_utils_address_range_at_position (const gchar *recipients_list,
 496                                              gint position,
 497                                              gint *start,
 498                                              gint *end)
 499 {
 500         gchar *current = NULL;
 501         gint range_start = 0;
 502         gint range_end = 0;
 503         gint index;
 504         gboolean is_quoted = FALSE;
 505
 506         index = 0;
 507         for (current = (gchar *) recipients_list; *current != '\0'; current = g_utf8_find_next_char (current, NULL)) {
 508                 gunichar c = g_utf8_get_char (current);
 509
 510                 if ((c == ',') && (!is_quoted)) {
 511                         if (index < position) {
 512                                 range_start = index + 1;
 513                         } else {
 514                                 break;
 515                         }
 516                 } else if (c == '\"') {
 517                         is_quoted = !is_quoted;
 518                 } else if ((c == ' ') &&(range_start == index)) {
 519                         range_start ++;
 520                 }
 521                 index ++;
 522                 range_end = index;
 523         }
 524
 525         if (start)
 526                 *start = range_start;
 527         if (end)
 528                 *end = range_end;
 529 }
 530
 531
 532 /* ******************************************************************* */
 533 /* ************************* UTILIY FUNCTIONS ************************ */
 534 /* ******************************************************************* */
 535
 536 static GString *
 537 get_next_line (const gchar * b, const gsize blen, const gchar * iter)
 538 {
 539         GString *gs;
 540         const gchar *i0;
 541
 542         if (iter > b + blen)
 543                 return g_string_new("");
 544
 545         i0 = iter;
 546         while (iter[0]) {
 547                 if (iter[0] == '\n')
 548                         break;
 549                 iter++;
 550         }
 551         gs = g_string_new_len (i0, iter - i0);
 552         return gs;
 553 }
 554 static int
 555 get_indent_level (const char *l)
 556 {
 557         int indent = 0;
 558
 559         while (l[0]) {
 560                 if (l[0] == '>') {
 561                         indent++;
 562                         if (l[1] == ' ') {
 563                                 l++;
 564                         }
 565                 } else {
 566                         break;
 567                 }
 568                 l++;
 569
 570         }
 571
 572         /*      if we hit the signature marker "-- ", we return -(indent + 1). This
 573          *      stops reformatting.
 574          */
 575         if (strcmp (l, "-- ") == 0) {
 576                 return -1 - indent;
 577         } else {
 578                 return indent;
 579         }
 580 }
 581
 582 static void
 583 unquote_line (GString * l)
 584 {
 585         gchar *p;
 586
 587         p = l->str;
 588         while (p[0]) {
 589                 if (p[0] == '>') {
 590                         if (p[1] == ' ') {
 591                                 p++;
 592                         }
 593                 } else {
 594                         break;
 595                 }
 596                 p++;
 597         }
 598         g_string_erase (l, 0, p - l->str);
 599 }
 600
 601 static void
 602 append_quoted (GString * buf, int indent, const GString * str,
 603                const int cutpoint)
 604 {
 605         int i;
 606
 607         indent = indent < 0 ? abs (indent) - 1 : indent;
 608         for (i = 0; i <= indent; i++) {
 609                 g_string_append (buf, "> ");
 610         }
 611         if (cutpoint > 0) {
 612                 g_string_append_len (buf, str->str, cutpoint);
 613         } else {
 614                 g_string_append (buf, str->str);
 615         }
 616         g_string_append (buf, "\n");
 617 }
 618
 619 static int
 620 get_breakpoint_utf8 (const gchar * s, gint indent, const gint limit)
 621 {
 622         gint index = 0;
 623         const gchar *pos, *last;
 624         gunichar *uni;
 625
 626         indent = indent < 0 ? abs (indent) - 1 : indent;
 627
 628         last = NULL;
 629         pos = s;
 630         uni = g_utf8_to_ucs4_fast (s, -1, NULL);
 631         while (pos[0]) {
 632                 if ((index + 2 * indent > limit) && last) {
 633                         g_free (uni);
 634                         return last - s;
 635                 }
 636                 if (g_unichar_isspace (uni[index])) {
 637                         last = pos;
 638                 }
 639                 pos = g_utf8_next_char (pos);
 640                 index++;
 641         }
 642         g_free (uni);
 643         return strlen (s);
 644 }
 645
 646 static int
 647 get_breakpoint_ascii (const gchar * s, const gint indent, const gint limit)
 648 {
 649         gint i, last;
 650
 651         last = strlen (s);
 652         if (last + 2 * indent < limit)
 653                 return last;
 654
 655         for (i = strlen (s); i > 0; i--) {
 656                 if (s[i] == ' ') {
 657                         if (i + 2 * indent <= limit) {
 658                                 return i;
 659                         } else {
 660                                 last = i;
 661                         }
 662                 }
 663         }
 664         return last;
 665 }
 666
 667 static int
 668 get_breakpoint (const gchar * s, const gint indent, const gint limit)
 669 {
 670
 671         if (g_utf8_validate (s, -1, NULL)) {
 672                 return get_breakpoint_utf8 (s, indent, limit);
 673         } else {                /* assume ASCII */
 674                 //g_warning("invalid UTF-8 in msg");
 675                 return get_breakpoint_ascii (s, indent, limit);
 676         }
 677 }
 678
 679 static gchar *
 680 cite (const time_t sent_date, const gchar *from)
 681 {
 682         return g_strdup (_("mcen_ia_editor_original_message"));
 683 }
 684
 685 static gchar *
 686 quoted_attachments (GList *attachments)
 687 {
 688         GList *node = NULL;
 689         GString *result = g_string_new ("");
 690         for (node = attachments; node != NULL; node = g_list_next (node)) {
 691                 gchar *filename = (gchar *) node->data;
 692                 g_string_append_printf ( result, "%s %s\n", _("mcen_ia_editor_attach_filename"), filename);
 693         }
 694
 695         return g_string_free (result, FALSE);
 696
 697 }
 698
 699 static gchar *
 700 modest_text_utils_quote_plain_text (const gchar *text,
 701                                     const gchar *cite,
 702                                     const gchar *signature,
 703                                     GList *attachments,
 704                                     int limit)
 705 {
 706         const gchar *iter;
 707         gint indent, breakpoint, rem_indent = 0;
 708         GString *q, *l, *remaining;
 709         gsize len;
 710         gchar *attachments_string = NULL;
 711
 712         /* remaining will store the rest of the line if we have to break it */
 713         q = g_string_new ("\n");
 714         q = g_string_append (q, cite);
 715         q = g_string_append_c (q, '\n');
 716         remaining = g_string_new ("");
 717
 718         iter = text;
 719         len = strlen(text);
 720         do {
 721                 l = get_next_line (text, len, iter);
 722                 iter = iter + l->len + 1;
 723                 indent = get_indent_level (l->str);
 724                 unquote_line (l);
 725
 726                 if (remaining->len) {
 727                         if (l->len && indent == rem_indent) {
 728                                 g_string_prepend (l, " ");
 729                                 g_string_prepend (l, remaining->str);
 730                         } else {
 731                                 do {
 732                                         breakpoint =
 733                                                 get_breakpoint (remaining->str,
 734                                                                 rem_indent,
 735                                                                 limit);
 736                                         append_quoted (q, rem_indent,
 737                                                        remaining, breakpoint);
 738                                         g_string_erase (remaining, 0,
 739                                                         breakpoint);
 740                                         if (remaining->str[0] == ' ') {
 741                                                 g_string_erase (remaining, 0,
 742                                                                 1);
 743                                         }
 744                                 } while (remaining->len);
 745                         }
 746                 }
 747                 g_string_free (remaining, TRUE);
 748                 breakpoint = get_breakpoint (l->str, indent, limit);
 749                 remaining = g_string_new (l->str + breakpoint);
 750                 if (remaining->str[0] == ' ') {
 751                         g_string_erase (remaining, 0, 1);
 752                 }
 753                 rem_indent = indent;
 754                 append_quoted (q, indent, l, breakpoint);
 755                 g_string_free (l, TRUE);
 756         } while ((iter < text + len) || (remaining->str[0]));
 757
 758         attachments_string = quoted_attachments (attachments);
 759         q = g_string_append (q, attachments_string);
 760         g_free (attachments_string);
 761
 762         if (signature != NULL) {
 763                 q = g_string_append_c (q, '\n');
 764                 q = g_string_append (q, signature);
 765         }
 766
 767         return g_string_free (q, FALSE);
 768 }
 769
 770 static gchar*
 771 modest_text_utils_quote_html (const gchar *text,
 772                               const gchar *cite,
 773                               const gchar *signature,
 774                               GList *attachments,
 775                               int limit)
 776 {
 777         gchar *result = NULL;
 778         gchar *signature_result = NULL;
 779         const gchar *format = \
 780                 "<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\">\n" \
 781                 "<html>\n" \
 782                 "<body>\n" \
 783                 "<br/>%s<br/>" \
 784                 "<pre>%s<br/>%s<br/>%s</pre>\n" \
 785                 "</body>\n" \
 786                 "</html>\n";
 787         gchar *attachments_string = NULL;
 788         gchar *q_attachments_string = NULL;
 789         gchar *q_cite = NULL;
 790         gchar *html_text = NULL;
 791
 792         if (signature == NULL)
 793                 signature_result = g_strdup ("");
 794         else
 795                 signature_result = modest_text_utils_convert_to_html_body (signature);
 796
 797         attachments_string = quoted_attachments (attachments);
 798         q_attachments_string = modest_text_utils_convert_to_html_body (attachments_string);
 799         q_cite = modest_text_utils_convert_to_html_body (cite);
 800         html_text = modest_text_utils_convert_to_html_body (text);
 801         result = g_strdup_printf (format, signature_result, q_cite, html_text, q_attachments_string);
 802         g_free (q_cite);
 803         g_free (html_text);
 804         g_free (attachments_string);
 805         g_free (q_attachments_string);
 806         g_free (signature_result);
 807         return result;
 808 }
 809
 810 static gint
 811 cmp_offsets_reverse (const url_match_t *match1, const url_match_t *match2)
 812 {
 813         return match2->offset - match1->offset;
 814 }
 815
 816
 817
 818 /*
 819  * check if the match is inside an existing match... */
 820 static void
 821 chk_partial_match (const url_match_t *match, guint* offset)
 822 {
 823         if (*offset >= match->offset && *offset < match->offset + match->len)
 824                 *offset = -1;
 825 }
 826
 827 static GSList*
 828 get_url_matches (GString *txt)
 829 {
 830         regmatch_t rm;
 831         guint rv, i, offset = 0;
 832         GSList *match_list = NULL;
 833
 834         static url_match_pattern_t patterns[] = MAIL_VIEWER_URL_MATCH_PATTERNS;
 835         const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t);
 836
 837         /* initalize the regexps */
 838         for (i = 0; i != pattern_num; ++i) {
 839                 patterns[i].preg = g_slice_new0 (regex_t);
 840
 841                 /* this should not happen */
 842                 g_return_val_if_fail (regcomp (patterns[i].preg, patterns[i].regex,
 843                                                REG_ICASE|REG_EXTENDED|REG_NEWLINE) == 0, NULL);
 844         }
 845         /* find all the matches */
 846         for (i = 0; i != pattern_num; ++i) {
 847                 offset     = 0;
 848                 while (1) {
 849                         int test_offset;
 850                         if ((rv = regexec (patterns[i].preg, txt->str + offset, 1, &rm, 0)) != 0) {
 851                                 g_return_val_if_fail (rv == REG_NOMATCH, NULL); /* this should not happen */
 852                                 break; /* try next regexp */
 853                         }
 854                         if (rm.rm_so == -1)
 855                                 break;
 856
 857                         /* FIXME: optimize this */
 858                         /* to avoid partial matches on something that was already found... */
 859                         /* check_partial_match will put -1 in the data ptr if that is the case */
 860                         test_offset = offset + rm.rm_so;
 861                         g_slist_foreach (match_list, (GFunc)chk_partial_match, &test_offset);
 862
 863                         /* make a list of our matches (<offset, len, prefix> tupels)*/
 864                         if (test_offset != -1) {
 865                                 url_match_t *match = g_slice_new (url_match_t);
 866                                 match->offset = offset + rm.rm_so;
 867                                 match->len    = rm.rm_eo - rm.rm_so;
 868                                 match->prefix = patterns[i].prefix;
 869                                 match_list = g_slist_prepend (match_list, match);
 870                         }
 871                         offset += rm.rm_eo;
 872                 }
 873         }
 874
 875         for (i = 0; i != pattern_num; ++i) {
 876                 regfree (patterns[i].preg);
 877                 g_slice_free  (regex_t, patterns[i].preg);
 878         } /* don't free patterns itself -- it's static */
 879
 880         /* now sort the list, so the matches are in reverse order of occurence.
 881          * that way, we can do the replacements starting from the end, so we don't need
 882          * to recalculate the offsets
 883          */
 884         match_list = g_slist_sort (match_list,
 885                                    (GCompareFunc)cmp_offsets_reverse);
 886         return match_list;
 887 }
 888
 889
 890
 891 static void
 892 hyperlinkify_plain_text (GString *txt)
 893 {
 894         GSList *cursor;
 895         GSList *match_list = get_url_matches (txt);
 896
 897         /* we will work backwards, so the offsets stay valid */
 898         for (cursor = match_list; cursor; cursor = cursor->next) {
 899
 900                 url_match_t *match = (url_match_t*) cursor->data;
 901                 gchar *url  = g_strndup (txt->str + match->offset, match->len);
 902                 gchar *repl = NULL; /* replacement  */
 903
 904                 /* the prefix is NULL: use the one that is already there */
 905                 repl = g_strdup_printf ("<a href=\"%s%s\">%s</a>",
 906                                         match->prefix ? match->prefix : EMPTY_STRING,
 907                                         url, url);
 908
 909                 /* replace the old thing with our hyperlink
 910                  * replacement thing */
 911                 g_string_erase  (txt, match->offset, match->len);
 912                 g_string_insert (txt, match->offset, repl);
 913
 914                 g_free (url);
 915                 g_free (repl);
 916
 917                 g_slice_free (url_match_t, match);
 918         }
 919
 920         g_slist_free (match_list);
 921 }
 922
 923
 924
 925 gchar*
 926 modest_text_utils_get_display_address (gchar *address)
 927 {
 928         gchar *cursor;
 929
 930         if (!address)
 931                 return NULL;
 932
 933         g_return_val_if_fail (g_utf8_validate (address, -1, NULL), NULL);
 934
 935         g_strchug (address); /* remove leading whitespace */
 936
 937         /*  <email@address> from display name */
 938         cursor = g_strstr_len (address, strlen(address), "<");
 939         if (cursor == address) /* there's nothing else? leave it */
 940                 return address;
 941         if (cursor)
 942                 cursor[0]='\0';
 943
 944         /* remove (bla bla) from display name */
 945         cursor = g_strstr_len (address, strlen(address), "(");
 946         if (cursor == address) /* there's nothing else? leave it */
 947                 return address;
 948         if (cursor)
 949                 cursor[0]='\0';
 950
 951         g_strchomp (address); /* remove trailing whitespace */
 952
 953         return address;
 954 }
 955
 956
 957
 958 gint
 959 modest_text_utils_get_subject_prefix_len (const gchar *sub)
 960 {
 961         gint i;
 962         static const gchar* prefix[] = {
 963                 "Re:", "RE:", "RV:", "re:"
 964                 "Fwd:", "FWD:", "FW:", "fwd:", "Fw:", "fw:", NULL
 965         };
 966
 967         if (!sub || (sub[0] != 'R' && sub[0] != 'F' && sub[0] != 'r' && sub[0] != 'f')) /* optimization */
 968                 return 0;
 969
 970         i = 0;
 971
 972         while (prefix[i]) {
 973                 if (g_str_has_prefix(sub, prefix[i])) {
 974                         int prefix_len = strlen(prefix[i]);
 975                         if (sub[prefix_len] == ' ')
 976                                 ++prefix_len; /* ignore space after prefix as well */
 977                         return prefix_len;
 978                 }
 979                 ++i;
 980         }
 981         return 0;
 982 }
 983
 984
 985 gint
 986 modest_text_utils_utf8_strcmp (const gchar* s1, const gchar *s2, gboolean insensitive)
 987 {
 988         gint result = 0;
 989         gchar *n1, *n2;
 990
 991         /* work even when s1 and/or s2 == NULL */
 992         if (G_UNLIKELY(s1 == s2))
 993                 return 0;
 994
 995         /* if it's not case sensitive */
 996         if (!insensitive)
 997                 return strcmp (s1 ? s1 : "", s2 ? s2 : "");
 998
 999         n1 = g_utf8_collate_key (s1 ? s1 : "", -1);
1000         n2 = g_utf8_collate_key (s2 ? s2 : "", -1);
1001
1002         result = strcmp (n1, n2);
1003
1004         g_free (n1);
1005         g_free (n2);
1006
1007         return result;
1008 }
1009
1010
1011 gchar*
1012 modest_text_utils_get_display_date (time_t date)
1013 {
1014         time_t now;
1015         const guint BUF_SIZE = 64;
1016         gchar date_buf[BUF_SIZE];
1017         gchar now_buf [BUF_SIZE];
1018
1019         now = time (NULL);
1020
1021         /* use the localized dates */
1022         modest_text_utils_strftime (date_buf, BUF_SIZE, "%x", date);
1023         modest_text_utils_strftime (now_buf,  BUF_SIZE, "%x", now);
1024
1025         /* if this is today, get the time instead of the date */
1026         if (strcmp (date_buf, now_buf) == 0)
1027                 modest_text_utils_strftime (date_buf, BUF_SIZE, "%X", date);
1028
1029         return g_strdup(date_buf);
1030 }
1031
1032 gboolean
1033 modest_text_utils_validate_email_address (const gchar *email_address, const gchar **invalid_char_position)
1034 {
1035         int count = 0;
1036         const gchar *c = NULL, *domain = NULL;
1037         static gchar *rfc822_specials = "()<>@,;:\\\"[]&";
1038
1039         if (invalid_char_position != NULL)
1040                 *invalid_char_position = NULL;
1041
1042         /* first we validate the name portion (name@domain) */
1043         for (c = email_address;  *c;  c++) {
1044                 if (*c == '\"' &&
1045                     (c == email_address ||
1046                      *(c - 1) == '.' ||
1047                      *(c - 1) == '\"')) {
1048                         while (*++c) {
1049                                 if (*c == '\"')
1050                                         break;
1051                                 if (*c == '\\' && (*++c == ' '))
1052                                         continue;
1053                                 if (*c <= ' ' || *c >= 127)
1054                                         return FALSE;
1055                         }
1056                         if (!*c++)
1057                                 return FALSE;
1058                         if (*c == '@')
1059                                 break;
1060                         if (*c != '.')
1061                                 return FALSE;
1062                         continue;
1063                 }
1064                 if (*c == '@')
1065                         break;
1066                 if (*c <= ' ' || *c >= 127)
1067                         return FALSE;
1068                 if (strchr(rfc822_specials, *c)) {
1069                         if (invalid_char_position)
1070                                 *invalid_char_position = c;
1071                         return FALSE;
1072                 }
1073         }
1074         if (c == email_address || *(c - 1) == '.')
1075                 return FALSE;
1076
1077         /* next we validate the domain portion (name@domain) */
1078         if (!*(domain = ++c))
1079                 return FALSE;
1080         do {
1081                 if (*c == '.') {
1082                         if (c == domain || *(c - 1) == '.')
1083                                 return FALSE;
1084                         count++;
1085                 }
1086                 if (*c <= ' ' || *c >= 127)
1087                         return FALSE;
1088                 if (strchr(rfc822_specials, *c)) {
1089                         if (invalid_char_position)
1090                                 *invalid_char_position = c;
1091                         return FALSE;
1092                 }
1093         } while (*++c);
1094
1095         return (count >= 1) ? TRUE : FALSE;
1096 }
1097
1098 gboolean
1099 modest_text_utils_validate_recipient (const gchar *recipient, const gchar **invalid_char_position)
1100 {
1101         gchar *stripped, *current;
1102         gchar *right_part;
1103         gboolean has_error = FALSE;
1104
1105         if (modest_text_utils_validate_email_address (recipient, invalid_char_position))
1106                 return TRUE;
1107         stripped = g_strdup (recipient);
1108         stripped = g_strstrip (stripped);
1109         current = stripped;
1110
1111         if (*current == '\0') {
1112                 g_free (stripped);
1113                 return FALSE;
1114         }
1115
1116         /* quoted string */
1117         if (*current == '\"') {
1118                 current = g_utf8_next_char (current);
1119                 has_error = TRUE;
1120                 for (; *current != '\0'; current = g_utf8_next_char (current)) {
1121                         if (*current == '\\') {
1122                                 /* TODO: This causes a warning, which breaks the build,
1123                                  * because a gchar cannot be < 0.
1124                                  * murrayc.
1125                                 if (current[1] <0) {
1126                                         has_error = TRUE;
1127                                         break;
1128                                 }
1129                                 */
1130                         } else if (*current == '\"') {
1131                                 has_error = FALSE;
1132                                 current = g_utf8_next_char (current);
1133                                 break;
1134                         }
1135                 }
1136         } else {
1137                 has_error = TRUE;
1138                 for (current = stripped ; *current != '\0'; current = g_utf8_next_char (current)) {
1139                         if (*current == '<') {
1140                                 has_error = FALSE;
1141                                 break;
1142                         }
1143                 }
1144         }
1145
1146         if (has_error) {
1147                 g_free (stripped);
1148                 return FALSE;
1149         }
1150
1151         right_part = g_strdup (current);
1152         g_free (stripped);
1153         right_part = g_strstrip (right_part);
1154
1155         if (g_str_has_prefix (right_part, "<") &&
1156             g_str_has_suffix (right_part, ">")) {
1157                 gchar *address;
1158                 gboolean valid;
1159
1160                 address = g_strndup (right_part+1, strlen (right_part) - 2);
1161                 g_free (right_part);
1162                 valid = modest_text_utils_validate_email_address (address, invalid_char_position);
1163                 g_free (address);
1164                 return valid;
1165         } else {
1166                 g_free (right_part);
1167                 return FALSE;
1168         }
1169 }
1170
1171
1172 gchar *
1173 modest_text_utils_get_display_size (guint64 size)
1174 {
1175         const guint KB=1024;
1176         const guint MB=1024 * KB;
1177         const guint GB=1024 * MB;
1178
1179         if (size == 0)
1180                 return g_strdup_printf(_FM("sfil_li_size_kb"), 0);
1181         if (0 < size && size < KB)
1182                 return g_strdup_printf (_FM("sfil_li_size_kb"), 1);
1183         else if (KB <= size && size < 100 * KB)
1184                 return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), size / KB);
1185         else if (100*KB <= size && size < MB)
1186                 return g_strdup_printf (_FM("sfil_li_size_100kb_1mb"), (float) size / MB);
1187         else if (MB <= size && size < 10*MB)
1188                 return g_strdup_printf (_FM("sfil_li_size_1mb_10mb"), (float) size / MB);
1189         else if (10*MB <= size && size < GB)
1190                 return g_strdup_printf (_FM("sfil_li_size_10mb_1gb"), size / MB);
1191         else
1192                 return g_strdup_printf (_FM("sfil_li_size_1gb_or_greater"), (float) size / GB);
1193 }
1194
1195 static gchar *
1196 get_email_from_address (const gchar * address)
1197 {
1198         gchar *left_limit, *right_limit;
1199
1200         left_limit = strstr (address, "<");
1201         right_limit = g_strrstr (address, ">");
1202
1203         if ((left_limit == NULL)||(right_limit == NULL)|| (left_limit > right_limit))
1204                 return g_strdup (address);
1205         else
1206                 return g_strndup (left_limit + 1, (right_limit - left_limit) - 1);
1207 }
1208
1209 gchar *
1210 modest_text_utils_get_color_string (GdkColor *color)
1211 {
1212
1213         return g_strdup_printf ("#%x%x%x%x%x%x%x%x%x%x%x%x",
1214                                 (color->red >> 12)   & 0xf, (color->red >> 8)   & 0xf,
1215                                 (color->red >>  4)   & 0xf, (color->red)        & 0xf,
1216                                 (color->green >> 12) & 0xf, (color->green >> 8) & 0xf,
1217                                 (color->green >>  4) & 0xf, (color->green)      & 0xf,
1218                                 (color->blue >> 12)  & 0xf, (color->blue >> 8)  & 0xf,
1219                                 (color->blue >>  4)  & 0xf, (color->blue)       & 0xf);
1220 }