projects / squirrelmail.git / blame
? search:
summary | shortlog | log | commit | commitdiff | tree
blob | blame (incremental) | history | HEAD
Serialize option types whose values are not scalar. At first, I strongly considered...
[squirrelmail.git] / functions / mime.php
CommitLineData
59177427 1<?php
2ba13803 2
35586184 3/**
8bd0068d 4 * mime.php
5 *
8bd0068d 6 * This contains the functions necessary to detect and decode MIME
7 * messages.
8 *
4b5049de 9 * @copyright &copy; 1999-2007 The SquirrelMail Project Team
4b4abf93 10 * @license http://opensource.org/licenses/gpl-license.php GNU Public License
8bd0068d 11 * @version $Id$
12 * @package squirrelmail
13 */
b74ba498 14
202bcbcc 15/**
16 * dependency information
17 functions dependency
18 mime_structure
19 class/mime/Message.class.php
20 Message::parseStructure
21 functions/page_header.php
22 displayPageHeader
23 functions/display_messages.php
24 plain_error_message
25 mime_fetch_body
26 functions/imap_general.php
27 sqimap_run_command
28 mime_print_body_lines
29
30
31
32functions/imap.php
33functions/attachment_common.php
34functions/display_messages.php
35
36magicHtml => url_parser
37translateText => url_parser
38
39*/
40
8beafbbc 41
7c7b74b3 42/* -------------------------------------------------------------------------- */
43/* MIME DECODING */
44/* -------------------------------------------------------------------------- */
b74ba498 45
d6c32258 46/**
8bd0068d 47 * Get the MIME structure
48 *
49 * This function gets the structure of a message and stores it in the "message" class.
50 * It will return this object for use with all relevant header information and
51 * fully parsed into the standard "message" object format.
52 */
a4a70693 53function mime_structure ($bodystructure, $flags=array()) {
c9d78ab4 54
3d8371be 55 /* Isolate the body structure and remove beginning and end parenthesis. */
a4a70693 56 $read = trim(substr ($bodystructure, strpos(strtolower($bodystructure), 'bodystructure') + 13));
451f74a2 57 $read = trim(substr ($read, 0, -1));
22efa9fb 58 $i = 0;
59 $msg = Message::parseStructure($read,$i);
2b665f28 60
9de42168 61 if (!is_object($msg)) {
3d8371be 62 global $color, $mailbox;
a48eba8f 63 displayPageHeader( $color, $mailbox );
5e8de8b6 64 $errormessage = _("SquirrelMail could not decode the bodystructure of the message");
8bd0068d 65 $errormessage .= '<br />'._("The bodystructure provided by your IMAP server:").'<br /><br />';
472e7acb 66 $errormessage .= '<pre>' . htmlspecialchars($read) . '</pre>';
ce8c6f42 67 plain_error_message( $errormessage );
3d8371be 68 echo '</body></html>';
9de42168 69 exit;
70 }
a4a70693 71 if (count($flags)) {
7a9e9c89 72 foreach ($flags as $flag) {
e1115979 73//FIXME: please document why it is we have to check the first char of the flag but we then go ahead and do a full string comparison anyway. Is this a speed enhancement? If not, let's keep it simple and just compare the full string and forget the switch block.
7a9e9c89 74 $char = strtoupper($flag{1});
75 switch ($char) {
3d8371be 76 case 'S':
77 if (strtolower($flag) == '\\seen') {
78 $msg->is_seen = true;
79 }
80 break;
81 case 'A':
82 if (strtolower($flag) == '\\answered') {
83 $msg->is_answered = true;
84 }
85 break;
86 case 'D':
87 if (strtolower($flag) == '\\deleted') {
88 $msg->is_deleted = true;
89 }
90 break;
91 case 'F':
92 if (strtolower($flag) == '\\flagged') {
93 $msg->is_flagged = true;
94 }
95 break;
96 case 'M':
97 if (strtolower($flag) == '$mdnsent') {
98 $msg->is_mdnsent = true;
99 }
100 break;
101 default:
102 break;
7a9e9c89 103 }
104 }
451f74a2 105 }
7a9e9c89 106 // listEntities($msg);
3d8371be 107 return $msg;
451f74a2 108}
b74ba498 109
22efa9fb 110
111
3d8371be 112/* This starts the parsing of a particular structure. It is called recursively,
8bd0068d 113 * so it can be passed different structures. It returns an object of type
114 * $message.
115 * First, it checks to see if it is a multipart message. If it is, then it
116 * handles that as it sees is necessary. If it is just a regular entity,
117 * then it parses it and adds the necessary header information (by calling out
118 * to mime_get_elements()
119 */
451f74a2 120
4d592352 121function mime_fetch_body($imap_stream, $id, $ent_id=1, $fetch_size=0) {
3d8371be 122 /* Do a bit of error correction. If we couldn't find the entity id, just guess
8bd0068d 123 * that it is the first one. That is usually the case anyway.
124 */
7c7b74b3 125
09a4bde3 126 if (!$ent_id) {
08b7f7cc 127 $cmd = "FETCH $id BODY[]";
1035e159 128 } else {
08b7f7cc 129 $cmd = "FETCH $id BODY[$ent_id]";
09a4bde3 130 }
3d8371be 131
4d592352 132 if ($fetch_size!=0) $cmd .= "<0.$fetch_size>";
da2415c1 133
6201339c 134 $data = sqimap_run_command ($imap_stream, $cmd, true, $response, $message, TRUE);
77b88425 135 do {
3d8371be 136 $topline = trim(array_shift($data));
137 } while($topline && ($topline[0] == '*') && !preg_match('/\* [0-9]+ FETCH.*/i', $topline)) ;
a4a70693 138
451f74a2 139 $wholemessage = implode('', $data);
140 if (ereg('\\{([^\\}]*)\\}', $topline, $regs)) {
3d8371be 141 $ret = substr($wholemessage, 0, $regs[1]);
142 /* There is some information in the content info header that could be important
8bd0068d 143 * in order to parse html messages. Let's get them here.
144 */
0600bdf1 145// if ($ret{0} == '<') {
6201339c 146// $data = sqimap_run_command ($imap_stream, "FETCH $id BODY[$ent_id.MIME]", true, $response, $message, TRUE);
0600bdf1 147// }
451f74a2 148 } else if (ereg('"([^"]*)"', $topline, $regs)) {
149 $ret = $regs[1];
06bcb9c3 150 } else if ((stristr($topline, 'nil') !== false) && (empty($wholemessage))) {
151 $ret = $wholemessage;
451f74a2 152 } else {
153 global $where, $what, $mailbox, $passed_id, $startMessage;
3d8371be 154 $par = 'mailbox=' . urlencode($mailbox) . '&amp;passed_id=' . $passed_id;
451f74a2 155 if (isset($where) && isset($what)) {
3d8371be 156 $par .= '&amp;where=' . urlencode($where) . '&amp;what=' . urlencode($what);
a3daaaf3 157 } else {
3d8371be 158 $par .= '&amp;startMessage=' . $startMessage . '&amp;show_more=0';
451f74a2 159 }
e5ea9327 160 $par .= '&amp;response=' . urlencode($response) .
8bd0068d 161 '&amp;message=' . urlencode($message) .
162 '&amp;topline=' . urlencode($topline);
a019eeb8 163
8bd0068d 164 echo '<tt><br />' .
02474e43 165 '<table width="80%"><tr>' .
166 '<tr><td colspan="2">' .
167 _("Body retrieval error. The reason for this is most probably that the message is malformed.") .
168 '</td></tr>' .
169 '<tr><td><b>' . _("Command:") . "</td><td>$cmd</td></tr>" .
170 '<tr><td><b>' . _("Response:") . "</td><td>$response</td></tr>" .
171 '<tr><td><b>' . _("Message:") . "</td><td>$message</td></tr>" .
172 '<tr><td><b>' . _("FETCH line:") . "</td><td>$topline</td></tr>" .
173 "</table><br /></tt></font><hr />";
346817d4 174
6201339c 175 $data = sqimap_run_command ($imap_stream, "FETCH $passed_id BODY[]", true, $response, $message, TRUE);
451f74a2 176 array_shift($data);
177 $wholemessage = implode('', $data);
a019eeb8 178
346817d4 179 $ret = $wholemessage;
a3daaaf3 180 }
3d8371be 181 return $ret;
451f74a2 182}
d4467150 183
6cc670de 184function mime_print_body_lines ($imap_stream, $id, $ent_id=1, $encoding, $rStream='php://stdout') {
1035e159 185
3d8371be 186 /* Don't kill the connection if the browser is over a dialup
8bd0068d 187 * and it would take over 30 seconds to download it.
188 * Don't call set_time_limit in safe mode.
189 */
b7206e1d 190
3d8371be 191 if (!ini_get('safe_mode')) {
b7206e1d 192 set_time_limit(0);
193 }
7c7b74b3 194 /* in case of base64 encoded attachments, do not buffer them.
8bd0068d 195 Instead, echo the decoded attachment directly to screen */
7c7b74b3 196 if (strtolower($encoding) == 'base64') {
197 if (!$ent_id) {
7591f143 198 $query = "FETCH $id BODY[]";
7c7b74b3 199 } else {
7591f143 200 $query = "FETCH $id BODY[$ent_id]";
7c7b74b3 201 }
7591f143 202 sqimap_run_command($imap_stream,$query,true,$response,$message,TRUE,'sqimap_base64_decode',$rStream,true);
1d142b8d 203 } else {
7591f143 204 $body = mime_fetch_body ($imap_stream, $id, $ent_id);
205 if (is_resource($rStream)) {
206 fputs($rStream,decodeBody($body,$encoding));
207 } else {
208 echo decodeBody($body, $encoding);
209 }
1d142b8d 210 }
346817d4 211
da2415c1 212 /*
8bd0068d 213 TODO, use the same method for quoted printable.
214 However, I assume that quoted printable attachments aren't that large
215 so the performancegain / memory usage drop will be minimal.
216 If we decide to add that then we need to adapt sqimap_fread because
217 we need to split te result on \n and fread doesn't stop at \n. That
218 means we also should provide $results from sqimap_fread (by ref) to
219 te function and set $no_return to false. The $filter function for
220 quoted printable should handle unsetting of $results.
221 */
da2415c1 222 /*
8bd0068d 223 TODO 2: find out how we write to the output stream php://stdout. fwrite
224 doesn't work because 'php://stdout isn't a stream.
225 */
7c7b74b3 226
5d9c6f73 227 return;
451f74a2 228}
beb9e459 229
451f74a2 230/* -[ END MIME DECODING ]----------------------------------------------------------- */
d4467150 231
3d8371be 232/* This is here for debugging purposes. It will print out a list
8bd0068d 233 * of all the entity IDs that are in the $message object.
234 */
451f74a2 235function listEntities ($message) {
3d8371be 236 if ($message) {
3c621ba1 237 echo "<tt>" . $message->entity_id . ' : ' . $message->type0 . '/' . $message->type1 . ' parent = '. $message->parent->entity_id. '<br />';
3d8371be 238 for ($i = 0; isset($message->entities[$i]); $i++) {
239 echo "$i : ";
240 $msg = listEntities($message->entities[$i]);
241
242 if ($msg) {
243 echo "return: ";
244 return $msg;
245 }
246 }
a4a70693 247 }
451f74a2 248}
f0c4dc12 249
f792c641 250function getPriorityStr($priority) {
3d8371be 251 $priority_level = substr($priority,0,1);
252
253 switch($priority_level) {
254 /* Check for a higher then normal priority. */
255 case '1':
256 case '2':
257 $priority_string = _("High");
258 break;
259
260 /* Check for a lower then normal priority. */
261 case '4':
262 case '5':
263 $priority_string = _("Low");
264 break;
265
266 /* Check for a normal priority. */
267 case '3':
268 default:
269 $priority_level = '3';
270 $priority_string = _("Normal");
271 break;
272
273 }
274 return $priority_string;
f792c641 275}
276
451f74a2 277/* returns a $message object for a particular entity id */
278function getEntity ($message, $ent_id) {
a4a70693 279 return $message->getEntity($ent_id);
451f74a2 280}
8beafbbc 281
3d8371be 282/* translateText
8bd0068d 283 * Extracted from strings.php 23/03/2002
284 */
da4c66e8 285
286function translateText(&$body, $wrap_at, $charset) {
3d8371be 287 global $where, $what; /* from searching */
288 global $color; /* color theme */
da4c66e8 289
202bcbcc 290 // require_once(SM_PATH . 'functions/url_parser.php');
da4c66e8 291
292 $body_ary = explode("\n", $body);
da4c66e8 293 for ($i=0; $i < count($body_ary); $i++) {
294 $line = $body_ary[$i];
295 if (strlen($line) - 2 >= $wrap_at) {
c7aff938 296 sqWordWrap($line, $wrap_at, $charset);
da4c66e8 297 }
298 $line = charset_decode($charset, $line);
299 $line = str_replace("\t", ' ', $line);
300
301 parseUrl ($line);
302
3d8371be 303 $quotes = 0;
da4c66e8 304 $pos = 0;
3d8371be 305 $j = strlen($line);
da4c66e8 306
3d8371be 307 while ($pos < $j) {
da4c66e8 308 if ($line[$pos] == ' ') {
3d8371be 309 $pos++;
da4c66e8 310 } else if (strpos($line, '&gt;', $pos) === $pos) {
311 $pos += 4;
3d8371be 312 $quotes++;
da4c66e8 313 } else {
314 break;
315 }
316 }
3d8371be 317
83c94382 318 if ($quotes % 2) {
d0814c02 319 $line = '<span class="quote1">' . $line . '</span>';
4c25967c 320 } elseif ($quotes) {
d0814c02 321 $line = '<span class="quote2">' . $line . '</span>';
da4c66e8 322 }
3d8371be 323
da4c66e8 324 $body_ary[$i] = $line;
325 }
326 $body = '<pre>' . implode("\n", $body_ary) . '</pre>';
327}
328
a2bfcbce 329/**
da1b55ad 330 * This returns a parsed string called $body. That string can then
331 * be displayed as the actual message in the HTML. It contains
332 * everything needed, including HTML Tags, Attachments at the
333 * bottom, etc.
f8a1ed5a 334 *
da1b55ad 335 * Since 1.2.0 function uses message_body hook.
336 * Till 1.3.0 function included output of formatAttachments().
337 *
338 * @param resource $imap_stream imap connection resource
339 * @param object $message squirrelmail message object
340 * @param array $color squirrelmail color theme array
341 * @param integer $wrap_at number of characters per line
342 * @param string $ent_num (since 1.3.0) message part id
343 * @param integer $id (since 1.3.0) message id
344 * @param string $mailbox (since 1.3.0) imap folder name
41f701c1 345 * @param boolean $clean (since 1.5.1) Do not output stuff that's irrelevant for the printable version.
da1b55ad 346 * @return string html formated message text
347 */
a2bfcbce 348function formatBody($imap_stream, $message, $color, $wrap_at, $ent_num, $id, $mailbox='INBOX', $clean=FALSE) {
3d8371be 349 /* This if statement checks for the entity to show as the
8bd0068d 350 * primary message. To add more of them, just put them in the
351 * order that is their priority.
352 */
ce68b76b 353 global $startMessage, $languages, $squirrelmail_language,
40a34e57 354 $show_html_default, $sort, $has_unsafe_images, $passed_ent_id,
14c85e39 355 $use_iframe, $iframe_height, $download_and_unsafe_link,
955bfc8f 356 $download_href, $unsafe_image_toggle_href, $unsafe_image_toggle_text,
357 $oTemplate;
358
359 $nbsp = $oTemplate->fetch('non_breaking_space.tpl');
2c25d36a 360
361 // workaround for not updated config.php
362 if (! isset($use_iframe)) $use_iframe = false;
77bfbd2e 363
5262d9a6 364 if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET) ) {
365 $view_unsafe_images = false;
77bfbd2e 366 }
d03c24f4 367
cc34b00d 368 $body = '';
23bcec6f 369 $urlmailbox = urlencode($mailbox);
451f74a2 370 $body_message = getEntity($message, $ent_num);
371 if (($body_message->header->type0 == 'text') ||
8bd0068d 372 ($body_message->header->type0 == 'rfc822')) {
3d8371be 373 $body = mime_fetch_body ($imap_stream, $id, $ent_num);
451f74a2 374 $body = decodeBody($body, $body_message->header->encoding);
e842b215 375
376 if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
8bd0068d 377 function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_decode')) {
e842b215 378 if (mb_detect_encoding($body) != 'ASCII') {
33a55f5a 379 $body = call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_decode',$body);
e842b215 380 }
381 }
d849b570 382
383 /* As of 1.5.2, $body is passed (and modified) by reference */
384 do_hook('message_body', $body);
23bcec6f 385
3d8371be 386 /* If there are other types that shouldn't be formatted, add
8bd0068d 387 * them here.
388 */
3d8371be 389
451f74a2 390 if ($body_message->header->type1 == 'html') {
3d8371be 391 if ($show_html_default <> 1) {
85015544 392 $entity_conv = array('&nbsp;' => ' ',
8bd0068d 393 '<p>' => "\n",
394 '<P>' => "\n",
395 '<br>' => "\n",
396 '<BR>' => "\n",
397 '<br />' => "\n",
398 '<BR />' => "\n",
399 '&gt;' => '>',
400 '&lt;' => '<');
85015544 401 $body = strtr($body, $entity_conv);
3d8371be 402 $body = strip_tags($body);
85015544 403 $body = trim($body);
404 translateText($body, $wrap_at,
8bd0068d 405 $body_message->header->getParameter('charset'));
2c25d36a 406 } elseif ($use_iframe && ! $clean) {
407 // $clean is used to remove iframe in printable view.
408
84410f31 409 /**
410 * If we don't add html message between iframe tags,
411 * we must detect unsafe images and modify $has_unsafe_images.
f8a1ed5a 412 */
758a7889 413 $html_body = magicHTML($body, $id, $message, $mailbox);
b6c52e61 414 // Convert character set in order to display html mails in different character set
415 $html_body = charset_decode($body_message->header->getParameter('charset'),$html_body,false,true);
84410f31 416
2c25d36a 417 // creating iframe url
418 $iframeurl=sqm_baseuri().'src/view_html.php?'
f8a1ed5a 419 . 'mailbox=' . $urlmailbox
2c25d36a 420 . '&amp;passed_id=' . $id
421 . '&amp;ent_id=' . $ent_num
422 . '&amp;view_unsafe_images=' . (int) $view_unsafe_images;
423
79d58d4c 424 global $oTemplate;
425 $oTemplate->assign('iframe_url', $iframeurl);
426 $oTemplate->assign('html_body', $html_body);
2b665f28 427
79d58d4c 428 $body = $oTemplate->fetch('read_html_iframe.tpl');
a3daaaf3 429 } else {
2c25d36a 430 // old way of html rendering
b6c52e61 431 /**
758a7889 432 * convert character set. charset_decode does not remove html special chars
b6c52e61 433 * applied by magicHTML functions and does not sanitize them second time if
758a7889 434 * fourth argument is true.
435 */
567dc524 436 $charset = $body_message->header->getParameter('charset');
437 if (!empty($charset)) {
438 $body = charset_decode($charset,$body,false,true);
439 }
440 $body = magicHTML($body, $id, $message, $mailbox);
a3daaaf3 441 }
451f74a2 442 } else {
3d8371be 443 translateText($body, $wrap_at,
8bd0068d 444 $body_message->header->getParameter('charset'));
451f74a2 445 }
a2bfcbce 446
447 // if this is the clean display (i.e. printer friendly), stop here.
448 if ( $clean ) {
449 return $body;
450 }
451
40a34e57 452 $download_and_unsafe_link = '';
453
83cf04bd 454 $link = 'passed_id=' . $id . '&amp;ent_id='.$ent_num.
8bd0068d 455 '&amp;mailbox=' . $urlmailbox .'&amp;sort=' . $sort .
456 '&amp;startMessage=' . $startMessage . '&amp;show_more=0';
08b7f7cc 457 if (isset($passed_ent_id)) {
458 $link .= '&amp;passed_ent_id='.$passed_ent_id;
459 }
14c85e39 460 $download_href = SM_PATH . 'src/download.php?absolute_dl=true&amp;' . $link;
955bfc8f 461 $download_and_unsafe_link .= "$nbsp|$nbsp"
462 . create_hyperlink($download_href, _("Download this as a file"));
7aad7b77 463 if ($view_unsafe_images) {
23f617b8 464 $text = _("Hide Unsafe Images");
7aad7b77 465 } else {
08b7f7cc 466 if (isset($has_unsafe_images) && $has_unsafe_images) {
467 $link .= '&amp;view_unsafe_images=1';
468 $text = _("View Unsafe Images");
469 } else {
470 $text = '';
471 }
3d8371be 472 }
83cf04bd 473 if($text != '') {
14c85e39 474 $unsafe_image_toggle_href = SM_PATH . 'src/read_body.php?'.$link;
475 $unsafe_image_toggle_text = $text;
955bfc8f 476 $download_and_unsafe_link .= "$nbsp|$nbsp"
477 . create_hyperlink($unsafe_image_toggle_href, $text);
83cf04bd 478 }
3d8371be 479 }
480 return $body;
451f74a2 481}
b74ba498 482
da1b55ad 483/**
d67f519a 484 * Generate attachments array for passing to templates. Separated from
485 * formatAttachments() below so that the same array can be given to the
486 * print-friendly version.
2b665f28 487 *
d67f519a 488 * @since 1.5.2
da1b55ad 489 * @param object $message SquirrelMail message object
490 * @param array $exclude_id message parts that are not attachments.
491 * @param string $mailbox mailbox name
492 * @param integer $id message id
da1b55ad 493 */
d67f519a 494function buildAttachmentArray($message, $exclude_id, $mailbox, $id) {
495 global $where, $what, $startMessage, $color, $passed_ent_id, $base_uri;
451f74a2 496
23bcec6f 497 $att_ar = $message->getAttachments($exclude_id);
23bcec6f 498 $urlMailbox = urlencode($mailbox);
499
d67f519a 500 $attachments = array();
23bcec6f 501 foreach ($att_ar as $att) {
fdc9d9b5 502 $ent = $att->entity_id;
2e25760a 503 $header = $att->header;
f0c4dc12 504 $type0 = strtolower($header->type0);
505 $type1 = strtolower($header->type1);
2e25760a 506 $name = '';
d0187bd6 507 $links = array();
21dab2dc 508 $links['download link']['text'] = _("Download");
202bcbcc 509 $links['download link']['href'] = $base_uri .
8bd0068d 510 "src/download.php?absolute_dl=true&amp;passed_id=$id&amp;mailbox=$urlMailbox&amp;ent_id=$ent";
2b665f28 511
2e25760a 512 if ($type0 =='message' && $type1 == 'rfc822') {
202bcbcc 513 $default_page = $base_uri . 'src/read_body.php';
2e25760a 514 $rfc822_header = $att->rfc822_header;
098ea084 515 $filename = $rfc822_header->subject;
6cc08d8b 516 if (trim( $filename ) == '') {
517 $filename = 'untitled-[' . $ent . ']' ;
08b7f7cc 518 }
2e25760a 519 $from_o = $rfc822_header->from;
520 if (is_object($from_o)) {
04ea844e 521 $from_name = decodeHeader($from_o->getAddress(false));
7e697748 522 } elseif (is_array($from_o) && count($from_o) && is_object($from_o[0])) {
523 // something weird happens when a digest message is opened and you return to the digest
524 // now the from object is part of an array. Probably the parseHeader call overwrites the info
525 // retrieved from the bodystructure in a different way. We need to fix this later.
526 // possible starting point, do not fetch header we already have and inspect how
527 // the rfc822_header object behaves.
528 $from_name = decodeHeader($from_o[0]->getAddress(false));
2e25760a 529 } else {
530 $from_name = _("Unknown sender");
f0c4dc12 531 }
d0187bd6 532 $description = _("From").': '.$from_name;
23bcec6f 533 } else {
202bcbcc 534 $default_page = $base_uri . 'src/download.php';
02474e43 535 $filename = $att->getFilename();
f810c0b2 536 if ($header->description) {
098ea084 537 $description = decodeHeader($header->description);
f810c0b2 538 } else {
3d8371be 539 $description = '';
540 }
2e25760a 541 }
542
543 $display_filename = $filename;
3d8371be 544 if (isset($passed_ent_id)) {
545 $passed_ent_id_link = '&amp;passed_ent_id='.$passed_ent_id;
546 } else {
547 $passed_ent_id_link = '';
548 }
549 $defaultlink = $default_page . "?startMessage=$startMessage"
8bd0068d 550 . "&amp;passed_id=$id&amp;mailbox=$urlMailbox"
551 . '&amp;ent_id='.$ent.$passed_ent_id_link;
2e25760a 552 if ($where && $what) {
8bd0068d 553 $defaultlink .= '&amp;where='. urlencode($where).'&amp;what='.urlencode($what);
2e25760a 554 }
7e2ff844 555 // IE does make use of mime content sniffing. Forcing a download
556 // prohibit execution of XSS inside an application/octet-stream attachment
557 if ($type0 == 'application' && $type1 == 'octet-stream') {
558 $defaultlink .= '&amp;absolute_dl=true';
559 }
2b665f28 560
3d8371be 561 /* This executes the attachment hook with a specific MIME-type.
8bd0068d 562 * If that doesn't have results, it tries if there's a rule
9b94c54d 563 * for a more generic type. Finally, a hook for ALL attachment
564 * types is run as well.
8bd0068d 565 */
d849b570 566 /* The API for this hook has changed as of 1.5.2 so that all plugin
567 arguments are passed in an array instead of each their own plugin
568 argument, and arguments are passed by reference, so instead of
569 returning any changes, changes should simply be made to the original
570 arguments themselves. */
571 do_hook("attachment $type0/$type1", $temp=array(&$links,
572 &$startMessage, &$id, &$urlMailbox, &$ent, &$defaultlink,
573 &$display_filename, &$where, &$what));
574 if(count($links) <= 1) {
575 /* The API for this hook has changed as of 1.5.2 so that all plugin
576 arguments are passed in an array instead of each their own plugin
577 argument, and arguments are passed by reference, so instead of
578 returning any changes, changes should simply be made to the original
579 arguments themselves. */
580 do_hook("attachment $type0/*", $temp=array(&$links,
581 &$startMessage, &$id, &$urlMailbox, &$ent, &$defaultlink,
582 &$display_filename, &$where, &$what));
2e25760a 583 }
d849b570 584 /* The API for this hook has changed as of 1.5.2 so that all plugin
585 arguments are passed in an array instead of each their own plugin
586 argument, and arguments are passed by reference, so instead of
587 returning any changes, changes should simply be made to the original
588 arguments themselves. */
589 do_hook("attachment */*", $temp=array(&$links,
590 &$startMessage, &$id, &$urlMailbox, &$ent, &$defaultlink,
591 &$display_filename, &$where, &$what));
77b88425 592
d67f519a 593 $this_attachment = array();
594 $this_attachment['Name'] = decodeHeader($display_filename);
595 $this_attachment['Description'] = $description;
596 $this_attachment['DefaultHREF'] = $defaultlink;
597 $this_attachment['DownloadHREF'] = $links['download link']['href'];
598 $this_attachment['ViewHREF'] = isset($links['attachment_common']) ? $links['attachment_common']['href'] : '';
599 $this_attachment['Size'] = $header->size;
600 $this_attachment['ContentType'] = htmlspecialchars($type0 .'/'. $type1);
601 $this_attachment['OtherLinks'] = array();
3d8371be 602 foreach ($links as $val) {
d0187bd6 603 if ($val['text']==_("Download") || $val['text'] == _("View"))
604 continue;
605 if (empty($val['text']) && empty($val['extra']))
606 continue;
2b665f28 607
d67f519a 608 $temp = array();
609 $temp['HREF'] = $val['href'];
610 $temp['Text'] = (empty($val['text']) ? '' : $val['text']) . (empty($val['extra']) ? '' : $val['extra']);
611 $this_attachment['OtherLinks'][] = $temp;
2e25760a 612 }
d67f519a 613 $attachments[] = $this_attachment;
2b665f28 614
3d8371be 615 unset($links);
2e25760a 616 }
2b665f28 617
d67f519a 618 return $attachments;
619}
620
621/**
622 * Displays attachment links and information
623 *
624 * Since 1.3.0 function is not included in formatBody() call.
625 *
626 * Since 1.0.2 uses attachment $type0/$type1 hook.
627 * Since 1.2.5 uses attachment $type0/* hook.
628 * Since 1.5.0 uses attachments_bottom hook.
629 * Since 1.5.2 uses templates and does *not* return a value.
630 *
631 * @param object $message SquirrelMail message object
632 * @param array $exclude_id message parts that are not attachments.
633 * @param string $mailbox mailbox name
634 * @param integer $id message id
635 */
636function formatAttachments($message, $exclude_id, $mailbox, $id) {
637 global $oTemplate;
2b665f28 638
d67f519a 639 $attach = buildAttachmentArray($message, $exclude_id, $mailbox, $id);
d0187bd6 640
641 $oTemplate->assign('attachments', $attach);
642 $oTemplate->display('read_attachments.tpl');
451f74a2 643}
b74ba498 644
7c7b74b3 645function sqimap_base64_decode(&$string) {
7c0ec1d8 646
b17a8968 647 // Base64 encoded data goes in pairs of 4 bytes. To achieve on the
7c0ec1d8 648 // fly decoding (to reduce memory usage) you have to check if the
649 // data has incomplete pairs
650
b17a8968 651 // Remove the noise in order to check if the 4 bytes pairs are complete
7c0ec1d8 652 $string = str_replace(array("\r\n","\n", "\r", " "),array('','','',''),$string);
653
42ce44f8 654 $sStringRem = '';
7c0ec1d8 655 $iMod = strlen($string) % 4;
656 if ($iMod) {
657 $sStringRem = substr($string,-$iMod);
b17a8968 658 // Check if $sStringRem contains padding characters
7c0ec1d8 659 if (substr($sStringRem,-1) != '=') {
660 $string = substr($string,0,-$iMod);
661 } else {
662 $sStringRem = '';
663 }
664 }
7c7b74b3 665 $string = base64_decode($string);
7c0ec1d8 666 return $sStringRem;
7c7b74b3 667}
668
fdf7cef1 669/**
670 * Decodes encoded message body
671 *
672 * This function decodes the body depending on the encoding type.
673 * Currently quoted-printable and base64 encodings are supported.
674 * decode_body hook was added to this function in 1.4.2/1.5.0
675 * @param string $body encoded message body
676 * @param string $encoding used encoding
677 * @return string decoded string
678 * @since 1.0
679 */
451f74a2 680function decodeBody($body, $encoding) {
83be314a 681
b583c3e8 682 $body = str_replace("\r\n", "\n", $body);
683 $encoding = strtolower($encoding);
3d8371be 684
d849b570 685 $encoding_handler = do_hook('decode_body', $encoding);
5166f86a 686
687
688 // plugins get first shot at decoding the body
689 //
690 if (!empty($encoding_handler) && function_exists($encoding_handler)) {
691 $body = $encoding_handler('decode', $body);
692
fdf7cef1 693 } elseif ($encoding == 'quoted-printable' ||
8bd0068d 694 $encoding == 'quoted_printable') {
fdf7cef1 695 /**
696 * quoted_printable_decode() function is broken in older
697 * php versions. Text with \r\n decoding was fixed only
758a7889 698 * in php 4.3.0. Minimal code requirement 4.0.4 +
fdf7cef1 699 * str_replace("\r\n", "\n", $body); call.
700 */
b583c3e8 701 $body = quoted_printable_decode($body);
fdf7cef1 702 } elseif ($encoding == 'base64') {
b583c3e8 703 $body = base64_decode($body);
704 }
3d8371be 705
b583c3e8 706 // All other encodings are returned raw.
3d8371be 707 return $body;
451f74a2 708}
709
9f7f68c3 710/**
8bd0068d 711 * Decodes headers
712 *
e1115979 713 * This function decodes strings that are encoded according to
8bd0068d 714 * RFC1522 (MIME Part Two: Message Header Extensions for Non-ASCII Text).
715 * Patched by Christian Schmidt <christian@ostenfeld.dk> 23/03/2002
716 *
717 * @param string $string header string that has to be made readable
718 * @param boolean $utfencode change message in order to be readable on user's charset. defaults to true
719 * @param boolean $htmlsave preserve spaces and sanitize html special characters. defaults to true
720 * @param boolean $decide decide if string can be utfencoded. defaults to false
721 * @return string decoded header string
722 */
9f7f68c3 723function decodeHeader ($string, $utfencode=true,$htmlsave=true,$decide=false) {
d6f584fc 724 global $languages, $squirrelmail_language,$default_charset;
79e07c7e 725 if (is_array($string)) {
726 $string = implode("\n", $string);
727 }
da2415c1 728
10dec454 729 if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
8bd0068d 730 function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_decodeheader')) {
33a55f5a 731 $string = call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_decodeheader', $string);
08b7f7cc 732 // Do we need to return at this point?
733 // return $string;
83be314a 734 }
79e07c7e 735 $i = 0;
08b7f7cc 736 $iLastMatch = -2;
db65b6b0 737 $encoded = true;
0a06275a 738
098ea084 739 $aString = explode(' ',$string);
08b7f7cc 740 $ret = '';
098ea084 741 foreach ($aString as $chunk) {
358a78a1 742 if ($encoded && $chunk === '') {
08b7f7cc 743 continue;
358a78a1 744 } elseif ($chunk === '') {
08b7f7cc 745 $ret .= ' ';
746 continue;
747 }
098ea084 748 $encoded = false;
08b7f7cc 749 /* if encoded words are not separated by a linear-space-white we still catch them */
750 $j = $i-1;
7e6ca3e8 751
08b7f7cc 752 while ($match = preg_match('/^(.*)=\?([^?]*)\?(Q|B)\?([^?]*)\?=(.*)$/Ui',$chunk,$res)) {
753 /* if the last chunk isn't an encoded string then put back the space, otherwise don't */
754 if ($iLastMatch !== $j) {
755 if ($htmlsave) {
9f7f68c3 756 $ret .= '&#32;';
08b7f7cc 757 } else {
758 $ret .= ' ';
759 }
760 }
761 $iLastMatch = $i;
762 $j = $i;
cb718de0 763 if ($htmlsave) {
764 $ret .= htmlspecialchars($res[1]);
765 } else {
766 $ret .= $res[1];
767 }
098ea084 768 $encoding = ucfirst($res[3]);
d6f584fc 769
770 /* decide about valid decoding */
771 if ($decide && is_conversion_safe($res[2])) {
8bd0068d 772 $utfencode=true;
773 $can_be_encoded=true;
d6f584fc 774 } else {
8bd0068d 775 $can_be_encoded=false;
d6f584fc 776 }
098ea084 777 switch ($encoding)
778 {
8bd0068d 779 case 'B':
780 $replace = base64_decode($res[4]);
781 if ($utfencode) {
782 if ($can_be_encoded) {
783 /* convert string to different charset,
784 * if functions asks for it (usually in compose)
785 */
bc6c0fba 786 $ret .= charset_convert($res[2],$replace,$default_charset,$htmlsave);
8bd0068d 787 } else {
788 // convert string to html codes in order to display it
789 $ret .= charset_decode($res[2],$replace);
790 }
fab65ca9 791 } else {
8bd0068d 792 if ($htmlsave) {
793 $replace = htmlspecialchars($replace);
794 }
795 $ret.= $replace;
fab65ca9 796 }
8bd0068d 797 break;
798 case 'Q':
799 $replace = str_replace('_', ' ', $res[4]);
800 $replace = preg_replace('/=([0-9a-f]{2})/ie', 'chr(hexdec("\1"))',
801 $replace);
802 if ($utfencode) {
803 if ($can_be_encoded) {
804 /* convert string to different charset,
805 * if functions asks for it (usually in compose)
806 */
bc6c0fba 807 $replace = charset_convert($res[2], $replace,$default_charset,$htmlsave);
8bd0068d 808 } else {
809 // convert string to html codes in order to display it
810 $replace = charset_decode($res[2], $replace);
811 }
812 } else {
813 if ($htmlsave) {
814 $replace = htmlspecialchars($replace);
815 }
098ea084 816 }
8bd0068d 817 $ret .= $replace;
818 break;
819 default:
820 break;
79e07c7e 821 }
098ea084 822 $chunk = $res[5];
823 $encoded = true;
08b7f7cc 824 }
825 if (!$encoded) {
826 if ($htmlsave) {
9f7f68c3 827 $ret .= '&#32;';
08b7f7cc 828 } else {
829 $ret .= ' ';
da2415c1 830 }
08b7f7cc 831 }
dc3d13a7 832
833 if (!$encoded && $htmlsave) {
834 $ret .= htmlspecialchars($chunk);
835 } else {
836 $ret .= $chunk;
837 }
098ea084 838 ++$i;
839 }
fd81e884 840 /* remove the first added space */
841 if ($ret) {
842 if ($htmlsave) {
9f7f68c3 843 $ret = substr($ret,5);
fd81e884 844 } else {
845 $ret = substr($ret,1);
846 }
847 }
da2415c1 848
08b7f7cc 849 return $ret;
451f74a2 850}
851
9f7f68c3 852/**
a24cf710 853 * Encodes header
8bd0068d 854 *
a24cf710 855 * Function uses XTRA_CODE _encodeheader function, if such function exists.
a24cf710 856 *
758a7889 857 * Function uses Q encoding by default and encodes a string according to RFC
858 * 1522 for use in headers if it contains 8-bit characters or anything that
a24cf710 859 * looks like it should be encoded.
8bd0068d 860 *
758a7889 861 * Function switches to B encoding and encodeHeaderBase64() function, if
862 * string is 8bit and multibyte character set supported by mbstring extension
863 * is used. It can cause E_USER_NOTICE errors, if interface is used with
f270a6eb 864 * multibyte character set unsupported by mbstring extension.
865 *
8bd0068d 866 * @param string $string header string, that has to be encoded
867 * @return string quoted-printable encoded string
f270a6eb 868 * @todo make $mb_charsets system wide constant
8bd0068d 869 */
451f74a2 870function encodeHeader ($string) {
6fbd125b 871 global $default_charset, $languages, $squirrelmail_language;
83be314a 872
10dec454 873 if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
8bd0068d 874 function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_encodeheader')) {
33a55f5a 875 return call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_encodeheader', $string);
83be314a 876 }
793cc001 877
a24cf710 878 // Use B encoding for multibyte charsets
f270a6eb 879 $mb_charsets = array('utf-8','big5','gb2313','euc-kr');
880 if (in_array($default_charset,$mb_charsets) &&
881 in_array($default_charset,sq_mb_list_encodings()) &&
882 sq_is8bit($string)) {
883 return encodeHeaderBase64($string,$default_charset);
884 } elseif (in_array($default_charset,$mb_charsets) &&
885 sq_is8bit($string) &&
886 ! in_array($default_charset,sq_mb_list_encodings())) {
887 // Add E_USER_NOTICE error here (can cause 'Cannot add header information' warning in compose.php)
888 // trigger_error('encodeHeader: Multibyte character set unsupported by mbstring extension.',E_USER_NOTICE);
889 }
a24cf710 890
451f74a2 891 // Encode only if the string contains 8-bit characters or =?
3d8371be 892 $j = strlen($string);
098ea084 893 $max_l = 75 - strlen($default_charset) - 7;
894 $aRet = array();
451f74a2 895 $ret = '';
c96c32f4 896 $iEncStart = $enc_init = false;
0d53f0f9 897 $cur_l = $iOffset = 0;
3d8371be 898 for($i = 0; $i < $j; ++$i) {
c96c32f4 899 switch($string{$i})
900 {
8bd0068d 901 case '=':
902 case '<':
903 case '>':
904 case ',':
905 case '?':
906 case '_':
907 if ($iEncStart === false) {
908 $iEncStart = $i;
909 }
910 $cur_l+=3;
911 if ($cur_l > ($max_l-2)) {
912 /* if there is an stringpart that doesn't need encoding, add it */
08b7f7cc 913 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
c96c32f4 914 $aRet[] = "=?$default_charset?Q?$ret?=";
915 $iOffset = $i;
916 $cur_l = 0;
917 $ret = '';
918 $iEncStart = false;
08b7f7cc 919 } else {
8bd0068d 920 $ret .= sprintf("=%02X",ord($string{$i}));
c96c32f4 921 }
8bd0068d 922 break;
923 case '(':
924 case ')':
925 if ($iEncStart !== false) {
08b7f7cc 926 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
8bd0068d 927 $aRet[] = "=?$default_charset?Q?$ret?=";
c96c32f4 928 $iOffset = $i;
8bd0068d 929 $cur_l = 0;
930 $ret = '';
931 $iEncStart = false;
c96c32f4 932 }
8bd0068d 933 break;
934 case ' ':
c96c32f4 935 if ($iEncStart !== false) {
098ea084 936 $cur_l++;
937 if ($cur_l > $max_l) {
08b7f7cc 938 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
c96c32f4 939 $aRet[] = "=?$default_charset?Q?$ret?=";
c96c32f4 940 $iOffset = $i;
941 $cur_l = 0;
098ea084 942 $ret = '';
8bd0068d 943 $iEncStart = false;
08b7f7cc 944 } else {
8bd0068d 945 $ret .= '_';
c96c32f4 946 }
3d8371be 947 }
8bd0068d 948 break;
949 default:
950 $k = ord($string{$i});
951 if ($k > 126) {
952 if ($iEncStart === false) {
953 // do not start encoding in the middle of a string, also take the rest of the word.
954 $sLeadString = substr($string,0,$i);
955 $aLeadString = explode(' ',$sLeadString);
956 $sToBeEncoded = array_pop($aLeadString);
957 $iEncStart = $i - strlen($sToBeEncoded);
958 $ret .= $sToBeEncoded;
959 $cur_l += strlen($sToBeEncoded);
960 }
961 $cur_l += 3;
962 /* first we add the encoded string that reached it's max size */
963 if ($cur_l > ($max_l-2)) {
964 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
965 $aRet[] = "=?$default_charset?Q?$ret?= "; /* the next part is also encoded => separate by space */
966 $cur_l = 3;
967 $ret = '';
968 $iOffset = $i;
969 $iEncStart = $i;
970 }
971 $enc_init = true;
972 $ret .= sprintf("=%02X", $k);
973 } else {
974 if ($iEncStart !== false) {
975 $cur_l++;
976 if ($cur_l > $max_l) {
977 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
978 $aRet[] = "=?$default_charset?Q?$ret?=";
979 $iEncStart = false;
980 $iOffset = $i;
981 $cur_l = 0;
982 $ret = '';
983 } else {
984 $ret .= $string{$i};
985 }
986 }
987 }
988 break;
f7b3ba37 989 }
451f74a2 990 }
793cc001 991
c96c32f4 992 if ($enc_init) {
993 if ($iEncStart !== false) {
994 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
995 $aRet[] = "=?$default_charset?Q?$ret?=";
996 } else {
997 $aRet[] = substr($string,$iOffset);
998 }
999 $string = implode('',$aRet);
451f74a2 1000 }
3d8371be 1001 return $string;
451f74a2 1002}
b74ba498 1003
f270a6eb 1004/**
1005 * Encodes string according to rfc2047 B encoding header formating rules
1006 *
758a7889 1007 * It is recommended way to encode headers with character sets that store
f270a6eb 1008 * symbols in more than one byte.
1009 *
1010 * Function requires mbstring support. If required mbstring functions are missing,
1011 * function returns false and sets E_USER_WARNING level error message.
1012 *
758a7889 1013 * Minimal requirements - php 4.0.6 with mbstring extension. Please note,
1014 * that mbstring functions will generate E_WARNING errors, if unsupported
f270a6eb 1015 * character set is used. mb_encode_mimeheader function provided by php
1016 * mbstring extension is not used in order to get better control of header
1017 * encoding.
1018 *
758a7889 1019 * Used php code functions - function_exists(), trigger_error(), strlen()
1020 * (is used with charset names and base64 strings). Used php mbstring
f270a6eb 1021 * functions - mb_strlen and mb_substr.
1022 *
758a7889 1023 * Related documents: rfc 2045 (BASE64 encoding), rfc 2047 (mime header
f270a6eb 1024 * encoding), rfc 2822 (header folding)
1025 *
1026 * @param string $string header string that must be encoded
758a7889 1027 * @param string $charset character set. Must be supported by mbstring extension.
f270a6eb 1028 * Use sq_mb_list_encodings() to detect supported charsets.
1029 * @return string string encoded according to rfc2047 B encoding formating rules
1030 * @since 1.5.1
1031 * @todo First header line can be wrapped to $iMaxLength - $HeaderFieldLength - 1
1032 * @todo Do we want to control max length of header?
1033 * @todo Do we want to control EOL (end-of-line) marker?
1034 * @todo Do we want to translate error message?
1035 */
1036function encodeHeaderBase64($string,$charset) {
1037 /**
1038 * Check mbstring function requirements.
1039 */
1040 if (! function_exists('mb_strlen') ||
1041 ! function_exists('mb_substr')) {
1042 // set E_USER_WARNING
1043 trigger_error('encodeHeaderBase64: Required mbstring functions are missing.',E_USER_WARNING);
1044 // return false
1045 return false;
1046 }
1047
1048 // initial return array
1049 $aRet = array();
1050
1051 /**
1052 * header length = 75 symbols max (same as in encodeHeader)
1053 * remove $charset length
1054 * remove =? ? ?= (5 chars)
1055 * remove 2 more chars (\r\n ?)
1056 */
1057 $iMaxLength = 75 - strlen($charset) - 7;
1058
1059 // set first character position
1060 $iStartCharNum = 0;
1061
1062 // loop through all characters. count characters and not bytes.
1063 for ($iCharNum=1; $iCharNum<=mb_strlen($string,$charset); $iCharNum++) {
1064 // encode string from starting character to current character.
1065 $encoded_string = base64_encode(mb_substr($string,$iStartCharNum,$iCharNum-$iStartCharNum,$charset));
1066
1067 // Check encoded string length
1068 if(strlen($encoded_string)>$iMaxLength) {
1069 // if string exceeds max length, reduce number of encoded characters and add encoded string part to array
1070 $aRet[] = base64_encode(mb_substr($string,$iStartCharNum,$iCharNum-$iStartCharNum-1,$charset));
1071
1072 // set new starting character
1073 $iStartCharNum = $iCharNum-1;
1074
1075 // encode last char (in case it is last character in string)
1076 $encoded_string = base64_encode(mb_substr($string,$iStartCharNum,$iCharNum-$iStartCharNum,$charset));
1077 } // if string is shorter than max length - add next character
1078 }
1079
1080 // add last encoded string to array
1081 $aRet[] = $encoded_string;
1082
1083 // set initial return string
1084 $sRet = '';
1085
1086 // loop through encoded strings
1087 foreach($aRet as $string) {
1088 // TODO: Do we want to control EOL (end-of-line) marker
1089 if ($sRet!='') $sRet.= " ";
1090
1091 // add header tags and encoded string to return string
1092 $sRet.= '=?'.$charset.'?B?'.$string.'?=';
1093 }
1094
1095 return $sRet;
1096}
1097
691a2d25 1098/* This function trys to locate the entity_id of a specific mime element */
3d8371be 1099function find_ent_id($id, $message) {
a171b359 1100 for ($i = 0, $ret = ''; $ret == '' && $i < count($message->entities); $i++) {
1101 if ($message->entities[$i]->header->type0 == 'multipart') {
3d8371be 1102 $ret = find_ent_id($id, $message->entities[$i]);
451f74a2 1103 } else {
3d8371be 1104 if (strcasecmp($message->entities[$i]->header->id, $id) == 0) {
d8cffbab 1105// if (sq_check_save_extension($message->entities[$i])) {
8bd0068d 1106 return $message->entities[$i]->entity_id;
da2415c1 1107// }
c8f5f606 1108 } elseif (!empty($message->entities[$i]->header->parameters['name'])) {
1109 /**
1110 * This is part of a fix for Outlook Express 6.x generating
1111 * cid URLs without creating content-id headers
1112 * @@JA - 20050207
1113 */
1114 if (strcasecmp($message->entities[$i]->header->parameters['name'], $id) == 0) {
1115 return $message->entities[$i]->entity_id;
1116 }
3d8371be 1117 }
a3daaaf3 1118 }
451f74a2 1119 }
3d8371be 1120 return $ret;
451f74a2 1121}
a3daaaf3 1122
e5e9381a 1123function sq_check_save_extension($message) {
1124 $filename = $message->getFilename();
1125 $ext = substr($filename, strrpos($filename,'.')+1);
1126 $save_extensions = array('jpg','jpeg','gif','png','bmp');
3d8371be 1127 return in_array($ext, $save_extensions);
e5e9381a 1128}
1129
1130
691a2d25 1131/**
8bd0068d 1132 ** HTMLFILTER ROUTINES
1133 */
451f74a2 1134
2dd879b8 1135/**
0493ed11 1136 * This function checks attribute values for entity-encoded values
1137 * and returns them translated into 8-bit strings so we can run
1138 * checks on them.
8bd0068d 1139 *
0493ed11 1140 * @param $attvalue A string to run entity check against.
1141 * @return Nothing, modifies a reference value.
8bd0068d 1142 */
0493ed11 1143function sq_defang(&$attvalue){
1144 $me = 'sq_defang';
2dd879b8 1145 /**
0493ed11 1146 * Skip this if there aren't ampersands or backslashes.
8bd0068d 1147 */
0493ed11 1148 if (strpos($attvalue, '&') === false
1149 && strpos($attvalue, '\\') === false){
1150 return;
2dd879b8 1151 }
0493ed11 1152 $m = false;
2b665f28 1153 // before deent, translate the dangerous unicode characters and ... to safe values
1154 // otherwise the regular expressions do not match.
1155
1156
1157
0493ed11 1158 do {
1159 $m = false;
1160 $m = $m || sq_deent($attvalue, '/\&#0*(\d+);*/s');
1161 $m = $m || sq_deent($attvalue, '/\&#x0*((\d|[a-f])+);*/si', true);
1162 $m = $m || sq_deent($attvalue, '/\\\\(\d+)/s', true);
1163 } while ($m == true);
1164 $attvalue = stripslashes($attvalue);
2dd879b8 1165}
1166
1167/**
8bd0068d 1168 * Kill any tabs, newlines, or carriage returns. Our friends the
1169 * makers of the browser with 95% market value decided that it'd
1170 * be funny to make "java[tab]script" be just as good as "javascript".
1171 *
1172 * @param attvalue The attribute value before extraneous spaces removed.
0493ed11 1173 * @return attvalue Nothing, modifies a reference value.
8bd0068d 1174 */
0493ed11 1175function sq_unspace(&$attvalue){
1176 $me = 'sq_unspace';
1177 if (strcspn($attvalue, "\t\r\n\0 ") != strlen($attvalue)){
1178 $attvalue = str_replace(Array("\t", "\r", "\n", "\0", " "),
1179 Array('', '', '', '', ''), $attvalue);
2dd879b8 1180 }
2dd879b8 1181}
1182
2b665f28 1183/**
88de4926 1184 * Translate all dangerous Unicode or Shift_JIS characters which are accepted by
2b665f28 1185 * IE as regular characters.
1186 *
1187 * @param attvalue The attribute value before dangerous characters are translated.
1188 * @return attvalue Nothing, modifies a reference value.
1189 * @author Marc Groot Koerkamp.
1190 */
1191function sq_fixIE_idiocy(&$attvalue) {
1192 // remove NUL
1193 $attvalue = str_replace("\0", "", $attvalue);
1194 // remove comments
1195 $attvalue = preg_replace("/(\/\*.*?\*\/)/","",$attvalue);
1196
88de4926 1197 // IE has the evil habit of accepting every possible value for the attribute expression.
1198 // The table below contains characters which are parsed by IE if they are used in the "expression"
2b665f28 1199 // attribute value.
1200 $aDangerousCharsReplacementTable = array(
1201 array('&#x029F;', '&#0671;' ,/* L UNICODE IPA Extension */
1202 '&#x0280;', '&#0640;' ,/* R UNICODE IPA Extension */
1203 '&#x0274;', '&#0628;' ,/* N UNICODE IPA Extension */
567dc524 1204 '&#xFF25;', '&#65317;' ,/* Unicode FULLWIDTH LATIN CAPITAL LETTER E */
1205 '&#xFF45;', '&#65349;' ,/* Unicode FULLWIDTH LATIN SMALL LETTER E */
2b665f28 1206 '&#xFF38;', '&#65336;',/* Unicode FULLWIDTH LATIN CAPITAL LETTER X */
1207 '&#xFF58;', '&#65368;',/* Unicode FULLWIDTH LATIN SMALL LETTER X */
1208 '&#xFF30;', '&#65328;',/* Unicode FULLWIDTH LATIN CAPITAL LETTER P */
1209 '&#xFF50;', '&#65360;',/* Unicode FULLWIDTH LATIN SMALL LETTER P */
1210 '&#xFF32;', '&#65330;',/* Unicode FULLWIDTH LATIN CAPITAL LETTER R */
1211 '&#xFF52;', '&#65362;',/* Unicode FULLWIDTH LATIN SMALL LETTER R */
1212 '&#xFF33;', '&#65331;',/* Unicode FULLWIDTH LATIN CAPITAL LETTER S */
1213 '&#xFF53;', '&#65363;',/* Unicode FULLWIDTH LATIN SMALL LETTER S */
1214 '&#xFF29;', '&#65321;',/* Unicode FULLWIDTH LATIN CAPITAL LETTER I */
1215 '&#xFF49;', '&#65353;',/* Unicode FULLWIDTH LATIN SMALL LETTER I */
1216 '&#xFF2F;', '&#65327;',/* Unicode FULLWIDTH LATIN CAPITAL LETTER O */
1217 '&#xFF4F;', '&#65359;',/* Unicode FULLWIDTH LATIN SMALL LETTER O */
1218 '&#xFF2E;', '&#65326;',/* Unicode FULLWIDTH LATIN CAPITAL LETTER N */
1219 '&#xFF4E;', '&#65358;',/* Unicode FULLWIDTH LATIN SMALL LETTER N */
1220 '&#xFF2C;', '&#65324;',/* Unicode FULLWIDTH LATIN CAPITAL LETTER L */
1221 '&#xFF4C;', '&#65356;',/* Unicode FULLWIDTH LATIN SMALL LETTER L */
1222 '&#xFF35;', '&#65333;',/* Unicode FULLWIDTH LATIN CAPITAL LETTER U */
1223 '&#xFF55;', '&#65365;',/* Unicode FULLWIDTH LATIN SMALL LETTER U */
1224 '&#x207F;', '&#8319;' ,/* Unicode SUPERSCRIPT LATIN SMALL LETTER N */
567dc524 1225 "\xEF\xBC\xA5", /* Shift JIS FULLWIDTH LATIN CAPITAL LETTER E */ // in unicode this is some Chinese char range
1226 "\xEF\xBD\x85", /* Shift JIS FULLWIDTH LATIN SMALL LETTER E */
1227 "\xEF\xBC\xB8", /* Shift JIS FULLWIDTH LATIN CAPITAL LETTER X */
1228 "\xEF\xBD\x98", /* Shift JIS FULLWIDTH LATIN SMALL LETTER X */
1229 "\xEF\xBC\xB0", /* Shift JIS FULLWIDTH LATIN CAPITAL LETTER P */
1230 "\xEF\xBD\x90", /* Shift JIS FULLWIDTH LATIN SMALL LETTER P */
1231 "\xEF\xBC\xB2", /* Shift JIS FULLWIDTH LATIN CAPITAL LETTER R */
1232 "\xEF\xBD\x92", /* Shift JIS FULLWIDTH LATIN SMALL LETTER R */
1233 "\xEF\xBC\xB3", /* Shift JIS FULLWIDTH LATIN CAPITAL LETTER S */
1234 "\xEF\xBD\x93", /* Shift JIS FULLWIDTH LATIN SMALL LETTER S */
1235 "\xEF\xBC\xA9", /* Shift JIS FULLWIDTH LATIN CAPITAL LETTER I */
1236 "\xEF\xBD\x89", /* Shift JIS FULLWIDTH LATIN SMALL LETTER I */
1237 "\xEF\xBC\xAF", /* Shift JIS FULLWIDTH LATIN CAPITAL LETTER O */
1238 "\xEF\xBD\x8F", /* Shift JIS FULLWIDTH LATIN SMALL LETTER O */
1239 "\xEF\xBC\xAE", /* Shift JIS FULLWIDTH LATIN CAPITAL LETTER N */
1240 "\xEF\xBD\x8E", /* Shift JIS FULLWIDTH LATIN SMALL LETTER N */
1241 "\xEF\xBC\xAC", /* Shift JIS FULLWIDTH LATIN CAPITAL LETTER L */
1242 "\xEF\xBD\x8C", /* Shift JIS FULLWIDTH LATIN SMALL LETTER L */
1243 "\xEF\xBC\xB5", /* Shift JIS FULLWIDTH LATIN CAPITAL LETTER U */
1244 "\xEF\xBD\x95", /* Shift JIS FULLWIDTH LATIN SMALL LETTER U */
1245 "\xE2\x81\xBF", /* Shift JIS FULLWIDTH SUPERSCRIPT N */
1246 "\xCA\x9F", /* L UNICODE IPA Extension */
1247 "\xCA\x80", /* R UNICODE IPA Extension */
1248 "\xC9\xB4"), /* N UNICODE IPA Extension */
2b665f28 1249 array('l', 'l', 'r','r','n','n',
567dc524 1250 'E','E','e','e','X','X','x','x','P','P','p','p','R','R','r','r','S','S','s','s','I','I',
1251 'i','i','O','O','o','o','N','N','n','n','L','L','l','l','U','U','u','u','n','n',
1252 'E','e','X','x','P','p','R','r','S','s','I','i','O','o','N','n','L','l','U','u','n','l','r','n'));
2b665f28 1253 $attvalue = str_replace($aDangerousCharsReplacementTable[0],$aDangerousCharsReplacementTable[1],$attvalue);
1254
88de4926 1255 // Escapes are useful for special characters like "{}[]()'&. In other cases they are
1256 // used for XSS.
2b665f28 1257 $attvalue = preg_replace("/(\\\\)([a-zA-Z]{1})/",'$2',$attvalue);
1258}
1259
691a2d25 1260/**
8bd0068d 1261 * This function returns the final tag out of the tag name, an array
1262 * of attributes, and the type of the tag. This function is called by
1263 * sq_sanitize internally.
1264 *
1265 * @param $tagname the name of the tag.
1266 * @param $attary the array of attributes and their values
1267 * @param $tagtype The type of the tag (see in comments).
1268 * @return a string with the final tag representation.
1269 */
691a2d25 1270function sq_tagprint($tagname, $attary, $tagtype){
b583c3e8 1271 $me = 'sq_tagprint';
3d8371be 1272
691a2d25 1273 if ($tagtype == 2){
1274 $fulltag = '</' . $tagname . '>';
1275 } else {
1276 $fulltag = '<' . $tagname;
1277 if (is_array($attary) && sizeof($attary)){
1278 $atts = Array();
1279 while (list($attname, $attvalue) = each($attary)){
1280 array_push($atts, "$attname=$attvalue");
1281 }
1282 $fulltag .= ' ' . join(" ", $atts);
1283 }
1284 if ($tagtype == 3){
b583c3e8 1285 $fulltag .= ' /';
691a2d25 1286 }
b583c3e8 1287 $fulltag .= '>';
451f74a2 1288 }
691a2d25 1289 return $fulltag;
451f74a2 1290}
a3daaaf3 1291
691a2d25 1292/**
8bd0068d 1293 * A small helper function to use with array_walk. Modifies a by-ref
1294 * value and makes it lowercase.
1295 *
1296 * @param $val a value passed by-ref.
1297 * @return void since it modifies a by-ref value.
1298 */
691a2d25 1299function sq_casenormalize(&$val){
1300 $val = strtolower($val);
1301}
451f74a2 1302
691a2d25 1303/**
8bd0068d 1304 * This function skips any whitespace from the current position within
1305 * a string and to the next non-whitespace value.
1306 *
1307 * @param $body the string
1308 * @param $offset the offset within the string where we should start
1309 * looking for the next non-whitespace character.
1310 * @return the location within the $body where the next
1311 * non-whitespace char is located.
1312 */
691a2d25 1313function sq_skipspace($body, $offset){
b583c3e8 1314 $me = 'sq_skipspace';
3d8371be 1315 preg_match('/^(\s*)/s', substr($body, $offset), $matches);
691a2d25 1316 if (sizeof($matches{1})){
1317 $count = strlen($matches{1});
1318 $offset += $count;
451f74a2 1319 }
691a2d25 1320 return $offset;
451f74a2 1321}
a3daaaf3 1322
691a2d25 1323/**
8bd0068d 1324 * This function looks for the next character within a string. It's
1325 * really just a glorified "strpos", except it catches if failures
1326 * nicely.
1327 *
1328 * @param $body The string to look for needle in.
1329 * @param $offset Start looking from this position.
1330 * @param $needle The character/string to look for.
1331 * @return location of the next occurance of the needle, or
1332 * strlen($body) if needle wasn't found.
1333 */
691a2d25 1334function sq_findnxstr($body, $offset, $needle){
3d8371be 1335 $me = 'sq_findnxstr';
691a2d25 1336 $pos = strpos($body, $needle, $offset);
1337 if ($pos === FALSE){
1338 $pos = strlen($body);
451f74a2 1339 }
691a2d25 1340 return $pos;
451f74a2 1341}
a3daaaf3 1342
691a2d25 1343/**
8bd0068d 1344 * This function takes a PCRE-style regexp and tries to match it
1345 * within the string.
1346 *
1347 * @param $body The string to look for needle in.
1348 * @param $offset Start looking from here.
1349 * @param $reg A PCRE-style regex to match.
1350 * @return Returns a false if no matches found, or an array
1351 * with the following members:
1352 * - integer with the location of the match within $body
1353 * - string with whatever content between offset and the match
1354 * - string with whatever it is we matched
1355 */
691a2d25 1356function sq_findnxreg($body, $offset, $reg){
b583c3e8 1357 $me = 'sq_findnxreg';
691a2d25 1358 $matches = Array();
1359 $retarr = Array();
7d06541f 1360 preg_match("%^(.*?)($reg)%si", substr($body, $offset), $matches);
1361 if (!isset($matches{0}) || !$matches{0}){
691a2d25 1362 $retarr = false;
1363 } else {
1364 $retarr{0} = $offset + strlen($matches{1});
1365 $retarr{1} = $matches{1};
1366 $retarr{2} = $matches{2};
1367 }
1368 return $retarr;
1369}
a3daaaf3 1370
691a2d25 1371/**
8bd0068d 1372 * This function looks for the next tag.
1373 *
1374 * @param $body String where to look for the next tag.
1375 * @param $offset Start looking from here.
1376 * @return false if no more tags exist in the body, or
1377 * an array with the following members:
1378 * - string with the name of the tag
1379 * - array with attributes and their values
1380 * - integer with tag type (1, 2, or 3)
1381 * - integer where the tag starts (starting "<")
1382 * - integer where the tag ends (ending ">")
1383 * first three members will be false, if the tag is invalid.
1384 */
691a2d25 1385function sq_getnxtag($body, $offset){
b583c3e8 1386 $me = 'sq_getnxtag';
691a2d25 1387 if ($offset > strlen($body)){
1388 return false;
1389 }
1390 $lt = sq_findnxstr($body, $offset, "<");
1391 if ($lt == strlen($body)){
1392 return false;
1393 }
1394 /**
8bd0068d 1395 * We are here:
1396 * blah blah <tag attribute="value">
1397 * \---------^
1398 */
691a2d25 1399 $pos = sq_skipspace($body, $lt+1);
1400 if ($pos >= strlen($body)){
1401 return Array(false, false, false, $lt, strlen($body));
1402 }
1403 /**
8bd0068d 1404 * There are 3 kinds of tags:
1405 * 1. Opening tag, e.g.:
1406 * <a href="blah">
1407 * 2. Closing tag, e.g.:
1408 * </a>
1409 * 3. XHTML-style content-less tag, e.g.:
1410 * <img src="blah" />
1411 */
691a2d25 1412 $tagtype = false;
1413 switch (substr($body, $pos, 1)){
3d8371be 1414 case '/':
1415 $tagtype = 2;
1416 $pos++;
1417 break;
1418 case '!':
1419 /**
8bd0068d 1420 * A comment or an SGML declaration.
1421 */
3d8371be 1422 if (substr($body, $pos+1, 2) == "--"){
1423 $gt = strpos($body, "-->", $pos);
1424 if ($gt === false){
1425 $gt = strlen($body);
1426 } else {
1427 $gt += 2;
1428 }
1429 return Array(false, false, false, $lt, $gt);
bb8d0799 1430 } else {
3d8371be 1431 $gt = sq_findnxstr($body, $pos, ">");
1432 return Array(false, false, false, $lt, $gt);
1433 }
1434 break;
1435 default:
1436 /**
8bd0068d 1437 * Assume tagtype 1 for now. If it's type 3, we'll switch values
1438 * later.
1439 */
3d8371be 1440 $tagtype = 1;
1441 break;
691a2d25 1442 }
a3daaaf3 1443
0493ed11 1444 $tag_start = $pos;
691a2d25 1445 $tagname = '';
1446 /**
8bd0068d 1447 * Look for next [\W-_], which will indicate the end of the tag name.
1448 */
691a2d25 1449 $regary = sq_findnxreg($body, $pos, "[^\w\-_]");
1450 if ($regary == false){
1451 return Array(false, false, false, $lt, strlen($body));
1452 }
1453 list($pos, $tagname, $match) = $regary;
1454 $tagname = strtolower($tagname);
1455
1456 /**
8bd0068d 1457 * $match can be either of these:
1458 * '>' indicating the end of the tag entirely.
1459 * '\s' indicating the end of the tag name.
1460 * '/' indicating that this is type-3 xhtml tag.
1461 *
1462 * Whatever else we find there indicates an invalid tag.
1463 */
691a2d25 1464 switch ($match){
3d8371be 1465 case '/':
691a2d25 1466 /**
8bd0068d 1467 * This is an xhtml-style tag with a closing / at the
1468 * end, like so: <img src="blah" />. Check if it's followed
1469 * by the closing bracket. If not, then this tag is invalid
1470 */
3d8371be 1471 if (substr($body, $pos, 2) == "/>"){
1472 $pos++;
1473 $tagtype = 3;
1474 } else {
1475 $gt = sq_findnxstr($body, $pos, ">");
1476 $retary = Array(false, false, false, $lt, $gt);
1477 return $retary;
1478 }
1479 case '>':
1480 return Array($tagname, false, $tagtype, $lt, $pos);
1481 break;
1482 default:
1483 /**
8bd0068d 1484 * Check if it's whitespace
1485 */
3d8371be 1486 if (!preg_match('/\s/', $match)){
1487 /**
8bd0068d 1488 * This is an invalid tag! Look for the next closing ">".
1489 */
7d06541f 1490 $gt = sq_findnxstr($body, $lt, ">");
3d8371be 1491 return Array(false, false, false, $lt, $gt);
1492 }
1493 break;
691a2d25 1494 }
3d8371be 1495
691a2d25 1496 /**
8bd0068d 1497 * At this point we're here:
1498 * <tagname attribute='blah'>
1499 * \-------^
1500 *
1501 * At this point we loop in order to find all attributes.
1502 */
691a2d25 1503 $attname = '';
0493ed11 1504 $atttype = false;
691a2d25 1505 $attary = Array();
1506
1507 while ($pos <= strlen($body)){
1508 $pos = sq_skipspace($body, $pos);
1509 if ($pos == strlen($body)){
1510 /**
8bd0068d 1511 * Non-closed tag.
1512 */
691a2d25 1513 return Array(false, false, false, $lt, $pos);
1514 }
1515 /**
8bd0068d 1516 * See if we arrived at a ">" or "/>", which means that we reached
1517 * the end of the tag.
1518 */
691a2d25 1519 $matches = Array();
164800ad 1520 if (preg_match("%^(\s*)(>|/>)%s", substr($body, $pos), $matches)) {
c828931c 1521 /**
8bd0068d 1522 * Yep. So we did.
1523 */
c828931c 1524 $pos += strlen($matches{1});
1525 if ($matches{2} == "/>"){
1526 $tagtype = 3;
1527 $pos++;
1528 }
1529 return Array($tagname, $attary, $tagtype, $lt, $pos);
1530 }
a3daaaf3 1531
cca46357 1532 /**
8bd0068d 1533 * There are several types of attributes, with optional
1534 * [:space:] between members.
1535 * Type 1:
1536 * attrname[:space:]=[:space:]'CDATA'
1537 * Type 2:
1538 * attrname[:space:]=[:space:]"CDATA"
1539 * Type 3:
1540 * attr[:space:]=[:space:]CDATA
1541 * Type 4:
1542 * attrname
1543 *
1544 * We leave types 1 and 2 the same, type 3 we check for
1545 * '"' and convert to "&quot" if needed, then wrap in
1546 * double quotes. Type 4 we convert into:
1547 * attrname="yes".
1548 */
3f7c623f 1549 $regary = sq_findnxreg($body, $pos, "[^:\w\-_]");
691a2d25 1550 if ($regary == false){
1551 /**
8bd0068d 1552 * Looks like body ended before the end of tag.
1553 */
691a2d25 1554 return Array(false, false, false, $lt, strlen($body));
cca46357 1555 }
691a2d25 1556 list($pos, $attname, $match) = $regary;
1557 $attname = strtolower($attname);
1558 /**
8bd0068d 1559 * We arrived at the end of attribute name. Several things possible
1560 * here:
1561 * '>' means the end of the tag and this is attribute type 4
1562 * '/' if followed by '>' means the same thing as above
1563 * '\s' means a lot of things -- look what it's followed by.
1564 * anything else means the attribute is invalid.
1565 */
691a2d25 1566 switch($match){
3d8371be 1567 case '/':
691a2d25 1568 /**
8bd0068d 1569 * This is an xhtml-style tag with a closing / at the
1570 * end, like so: <img src="blah" />. Check if it's followed
1571 * by the closing bracket. If not, then this tag is invalid
1572 */
3d8371be 1573 if (substr($body, $pos, 2) == "/>"){
691a2d25 1574 $pos++;
3d8371be 1575 $tagtype = 3;
691a2d25 1576 } else {
3d8371be 1577 $gt = sq_findnxstr($body, $pos, ">");
1578 $retary = Array(false, false, false, $lt, $gt);
1579 return $retary;
1580 }
1581 case '>':
1582 $attary{$attname} = '"yes"';
1583 return Array($tagname, $attary, $tagtype, $lt, $pos);
1584 break;
1585 default:
1586 /**
8bd0068d 1587 * Skip whitespace and see what we arrive at.
1588 */
3d8371be 1589 $pos = sq_skipspace($body, $pos);
1590 $char = substr($body, $pos, 1);
1591 /**
8bd0068d 1592 * Two things are valid here:
1593 * '=' means this is attribute type 1 2 or 3.
1594 * \w means this was attribute type 4.
1595 * anything else we ignore and re-loop. End of tag and
1596 * invalid stuff will be caught by our checks at the beginning
1597 * of the loop.
1598 */
3d8371be 1599 if ($char == "="){
1600 $pos++;
1601 $pos = sq_skipspace($body, $pos);
691a2d25 1602 /**
8bd0068d 1603 * Here are 3 possibilities:
1604 * "'" attribute type 1
1605 * '"' attribute type 2
1606 * everything else is the content of tag type 3
1607 */
3d8371be 1608 $quot = substr($body, $pos, 1);
1609 if ($quot == "'"){
1610 $regary = sq_findnxreg($body, $pos+1, "\'");
1611 if ($regary == false){
1612 return Array(false, false, false, $lt, strlen($body));
1613 }
1614 list($pos, $attval, $match) = $regary;
1615 $pos++;
1616 $attary{$attname} = "'" . $attval . "'";
1617 } else if ($quot == '"'){
1618 $regary = sq_findnxreg($body, $pos+1, '\"');
1619 if ($regary == false){
1620 return Array(false, false, false, $lt, strlen($body));
1621 }
1622 list($pos, $attval, $match) = $regary;
1623 $pos++;
1624 $attary{$attname} = '"' . $attval . '"';
1625 } else {
1626 /**
8bd0068d 1627 * These are hateful. Look for \s, or >.
1628 */
3d8371be 1629 $regary = sq_findnxreg($body, $pos, "[\s>]");
1630 if ($regary == false){
1631 return Array(false, false, false, $lt, strlen($body));
1632 }
1633 list($pos, $attval, $match) = $regary;
1634 /**
8bd0068d 1635 * If it's ">" it will be caught at the top.
1636 */
3d8371be 1637 $attval = preg_replace("/\"/s", "&quot;", $attval);
1638 $attary{$attname} = '"' . $attval . '"';
7e235a1a 1639 }
3d8371be 1640 } else if (preg_match("|[\w/>]|", $char)) {
691a2d25 1641 /**
8bd0068d 1642 * That was attribute type 4.
1643 */
3d8371be 1644 $attary{$attname} = '"yes"';
1645 } else {
1646 /**
8bd0068d 1647 * An illegal character. Find next '>' and return.
1648 */
3d8371be 1649 $gt = sq_findnxstr($body, $pos, ">");
1650 return Array(false, false, false, $lt, $gt);
451f74a2 1651 }
3d8371be 1652 break;
691a2d25 1653 }
1654 }
1655 /**
8bd0068d 1656 * The fact that we got here indicates that the tag end was never
1657 * found. Return invalid tag indication so it gets stripped.
1658 */
691a2d25 1659 return Array(false, false, false, $lt, strlen($body));
1660}
1661
1662/**
0493ed11 1663 * Translates entities into literal values so they can be checked.
8bd0068d 1664 *
0493ed11 1665 * @param $attvalue the by-ref value to check.
1666 * @param $regex the regular expression to check against.
1667 * @param $hex whether the entites are hexadecimal.
1668 * @return True or False depending on whether there were matches.
8bd0068d 1669 */
0493ed11 1670function sq_deent(&$attvalue, $regex, $hex=false){
b583c3e8 1671 $me = 'sq_deent';
0493ed11 1672 $ret_match = false;
2b665f28 1673 // remove comments
1674 //$attvalue = preg_replace("/(\/\*.*\*\/)/","",$attvalue);
0493ed11 1675 preg_match_all($regex, $attvalue, $matches);
1676 if (is_array($matches) && sizeof($matches[0]) > 0){
1677 $repl = Array();
1678 for ($i = 0; $i < sizeof($matches[0]); $i++){
1679 $numval = $matches[1][$i];
1680 if ($hex){
1681 $numval = hexdec($numval);
a3daaaf3 1682 }
0493ed11 1683 $repl{$matches[0][$i]} = chr($numval);
691a2d25 1684 }
0493ed11 1685 $attvalue = strtr($attvalue, $repl);
1686 return true;
1687 } else {
1688 return false;
691a2d25 1689 }
691a2d25 1690}
1691
1692/**
8bd0068d 1693 * This function runs various checks against the attributes.
1694 *
1695 * @param $tagname String with the name of the tag.
1696 * @param $attary Array with all tag attributes.
1697 * @param $rm_attnames See description for sq_sanitize
1698 * @param $bad_attvals See description for sq_sanitize
1699 * @param $add_attr_to_tag See description for sq_sanitize
1700 * @param $message message object
1701 * @param $id message id
1702 * @return Array with modified attributes.
1703 */
da2415c1 1704function sq_fixatts($tagname,
1705 $attary,
691a2d25 1706 $rm_attnames,
1707 $bad_attvals,
1708 $add_attr_to_tag,
1709 $message,
b3af12ef 1710 $id,
3d8371be 1711 $mailbox
691a2d25 1712 ){
b583c3e8 1713 $me = 'sq_fixatts';
691a2d25 1714 while (list($attname, $attvalue) = each($attary)){
1715 /**
8bd0068d 1716 * See if this attribute should be removed.
1717 */
691a2d25 1718 foreach ($rm_attnames as $matchtag=>$matchattrs){
1719 if (preg_match($matchtag, $tagname)){
1720 foreach ($matchattrs as $matchattr){
1721 if (preg_match($matchattr, $attname)){
1722 unset($attary{$attname});
1723 continue;
1724 }
451f74a2 1725 }
451f74a2 1726 }
691a2d25 1727 }
2b665f28 1728 /**
1729 * Workaround for IE quirks
1730 */
1731 sq_fixIE_idiocy($attvalue);
1732
691a2d25 1733 /**
8bd0068d 1734 * Remove any backslashes, entities, and extraneous whitespace.
1735 */
2b665f28 1736
1737 $oldattvalue = $attvalue;
0493ed11 1738 sq_defang($attvalue);
2b665f28 1739 if ($attname == 'style' && $attvalue !== $oldattvalue) {
1740 // entities are used in the attribute value. In 99% of the cases it's there as XSS
1741 // i.e.<div style="{ left:exp&#x0280;essio&#x0274;( alert('XSS') ) }">
1742 $attvalue = "idiocy";
1743 $attary{$attname} = $attvalue;
1744 }
0493ed11 1745 sq_unspace($attvalue);
af861a34 1746
691a2d25 1747 /**
8bd0068d 1748 * Now let's run checks on the attvalues.
1749 * I don't expect anyone to comprehend this. If you do,
1750 * get in touch with me so I can drive to where you live and
1751 * shake your hand personally. :)
1752 */
691a2d25 1753 foreach ($bad_attvals as $matchtag=>$matchattrs){
1754 if (preg_match($matchtag, $tagname)){
1755 foreach ($matchattrs as $matchattr=>$valary){
1756 if (preg_match($matchattr, $attname)){
1757 /**
8bd0068d 1758 * There are two arrays in valary.
1759 * First is matches.
1760 * Second one is replacements
1761 */
691a2d25 1762 list($valmatch, $valrepl) = $valary;
da2415c1 1763 $newvalue =
691a2d25 1764 preg_replace($valmatch, $valrepl, $attvalue);
1765 if ($newvalue != $attvalue){
1766 $attary{$attname} = $newvalue;
567dc524 1767 $attvalue = $newvalue;
691a2d25 1768 }
1769 }
1770 }
451f74a2 1771 }
a3daaaf3 1772 }
567dc524 1773 if ($attname == 'style') {
1774 if (preg_match('/[\0-\37\200-\377]+/',$attvalue)) {
1775 // 8bit and control characters in style attribute values can be used for XSS, remove them
1776 $attary{$attname} = '"disallowed character"';
1777 }
1778 preg_match_all("/url\s*\((.+)\)/si",$attvalue,$aMatch);
1779 if (count($aMatch)) {
1780 foreach($aMatch[1] as $sMatch) {
1781 // url value
1782 $urlvalue = $sMatch;
1783 sq_fix_url($attname, $urlvalue, $message, $id, $mailbox,"'");
1784 $attary{$attname} = str_replace($sMatch,$urlvalue,$attvalue);
1785 }
1786 }
691a2d25 1787 }
ff940ebc 1788 /**
567dc524 1789 * Use white list based filtering on attributes which can contain url's
ff940ebc 1790 */
567dc524 1791 else if ($attname == 'href' || $attname == 'src' || $attname == 'background') {
1792 sq_fix_url($attname, $attvalue, $message, $id, $mailbox);
1793 $attary{$attname} = $attvalue;
ff940ebc 1794 }
a3daaaf3 1795 }
691a2d25 1796 /**
8bd0068d 1797 * See if we need to append any attributes to this tag.
1798 */
691a2d25 1799 foreach ($add_attr_to_tag as $matchtag=>$addattary){
1800 if (preg_match($matchtag, $tagname)){
1801 $attary = array_merge($attary, $addattary);
1802 }
1803 }
1804 return $attary;
451f74a2 1805}
a3daaaf3 1806
567dc524 1807/**
1808 * This function filters url's
1809 *
1810 * @param $attvalue String with attribute value to filter
1811 * @param $message message object
1812 * @param $id message id
1813 * @param $mailbox mailbox
1814 * @param $sQuote quoting characters around url's
1815 */
1816function sq_fix_url($attname, &$attvalue, $message, $id, $mailbox,$sQuote = '"') {
1817 $attvalue = trim($attvalue);
1818 if ($attvalue && ($attvalue[0] =='"'|| $attvalue[0] == "'")) {
1819 // remove the double quotes
1820 $sQuote = $attvalue[0];
1821 $attvalue = trim(substr($attvalue,1,-1));
1822 }
1823
1824 if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET) ) {
1825 $view_unsafe_images = false;
1826 }
1827 $secremoveimg = '../images/' . _("sec_remove_eng.png");
1828
1829 /**
1830 * Replace empty src tags with the blank image. src is only used
1831 * for frames, images, and image inputs. Doing a replace should
1832 * not affect them working as should be, however it will stop
1833 * IE from being kicked off when src for img tags are not set
1834 */
1835 if ($attvalue == '') {
1836 $attvalue = '"' . SM_PATH . 'images/blank.png"';
1837 } else {
1838 // first, disallow 8 bit characters and control characters
1839 if (preg_match('/[\0-\37\200-\377]+/',$attvalue)) {
1840 switch ($attname) {
1841 case 'href':
1842 $attvalue = $sQuote . 'http://invalid-stuff-detected.example.com' . $sQuote;
1843 break;
1844 default:
1845 $attvalue = $sQuote . SM_PATH . 'images/blank.png'. $sQuote;
1846 break;
1847 }
1848 } else {
1849 $aUrl = parse_url($attvalue);
1850 if (isset($aUrl['scheme'])) {
1851 switch(strtolower($aUrl['scheme'])) {
d75e755b 1852 case 'mailto':
567dc524 1853 case 'http':
1854 case 'https':
1855 case 'ftp':
1856 if ($attname != 'href') {
1857 if ($view_unsafe_images == false) {
1858 $attvalue = $sQuote . $secremoveimg . $sQuote;
1859 } else {
1860 if (isset($aUrl['path'])) {
1861 // validate image extension.
1862 $ext = strtolower(substr($aUrl['path'],strrpos($aUrl['path'],'.')));
1863 if (!in_array($ext,array('.jpeg','.jpg','xjpeg','.gif','.bmp','.jpe','.png','.xbm'))) {
1864 $attvalue = $sQuote . SM_PATH . 'images/blank.png'. $sQuote;
1865 }
1866 } else {
1867 $attvalue = $sQuote . SM_PATH . 'images/blank.png'. $sQuote;
1868 }
1869 }
1870 }
1871 break;
1872 case 'outbind':
1873 /**
1874 * "Hack" fix for Outlook using propriatary outbind:// protocol in img tags.
1875 * One day MS might actually make it match something useful, for now, falling
1876 * back to using cid2http, so we can grab the blank.png.
1877 */
1878 $attvalue = sq_cid2http($message, $id, $attvalue, $mailbox);
1879 break;
1880 case 'cid':
1881 /**
1882 * Turn cid: urls into http-friendly ones.
1883 */
1884 $attvalue = sq_cid2http($message, $id, $attvalue, $mailbox);
1885 break;
1886 default:
1887 $attvalue = $sQuote . SM_PATH . 'images/blank.png' . $sQuote;
1888 break;
1889 }
1890 } else {
1891 if (!(isset($aUrl['path']) && $aUrl['path'] == $secremoveimg)) {
1892 // parse_url did not lead to satisfying result
1893 $attvalue = $sQuote . SM_PATH . 'images/blank.png' . $sQuote;
1894 }
1895 }
1896 }
1897 }
1898}
1899
691a2d25 1900/**
8bd0068d 1901 * This function edits the style definition to make them friendly and
1902 * usable in SquirrelMail.
1903 *
1904 * @param $message the message object
1905 * @param $id the message id
1906 * @param $content a string with whatever is between <style> and </style>
1907 * @param $mailbox the message mailbox
1908 * @return a string with edited content.
1909 */
e60a299a 1910function sq_fixstyle($body, $pos, $message, $id, $mailbox){
b583c3e8 1911 $me = 'sq_fixstyle';
7e2ff844 1912 // workaround for </style> in between comments
1913 $iCurrentPos = $pos;
1914 $content = '';
1915 $sToken = '';
1916 $bSucces = false;
1917 $bEndTag = false;
1918 for ($i=$pos,$iCount=strlen($body);$i<$iCount;++$i) {
1919 $char = $body{$i};
1920 switch ($char) {
1921 case '<':
1922 $sToken .= $char;
1923 break;
1924 case '/':
1925 if ($sToken == '<') {
1926 $sToken .= $char;
1927 $bEndTag = true;
1928 } else {
1929 $content .= $char;
1930 }
1931 break;
1932 case '>':
1933 if ($bEndTag) {
1934 $sToken .= $char;
1935 if (preg_match('/\<\/\s*style\s*\>/i',$sToken,$aMatch)) {
1936 $newpos = $i + 1;
1937 $bSucces = true;
1938 break 2;
1939 } else {
1940 $content .= $sToken;
1941 }
1942 $bEndTag = false;
1943 } else {
1944 $content .= $char;
1945 }
1946 break;
1947 case '!':
1948 if ($sToken == '<') {
1949 // possible comment
1950 if (isset($body{$i+2}) && substr($body,$i,3) == '!--') {
1951 $i = strpos($body,'-->',$i+3);
2b665f28 1952 if ($i === false) { // no end comment
1953 $i = strlen($body);
1954 }
7e2ff844 1955 $sToken = '';
1956 }
1957 } else {
1958 $content .= $char;
1959 }
1960 break;
1961 default:
1962 if ($bEndTag) {
1963 $sToken .= $char;
1964 } else {
1965 $content .= $char;
1966 }
1967 break;
1968 }
1969 }
1970 if ($bSucces == FALSE){
7d06541f 1971 return array(FALSE, strlen($body));
1972 }
7e2ff844 1973
2b665f28 1974
1975
691a2d25 1976 /**
0493ed11 1977 * First look for general BODY style declaration, which would be
1978 * like so:
1979 * body {background: blah-blah}
1980 * and change it to .bodyclass so we can just assign it to a <div>
1981 */
691a2d25 1982 $content = preg_replace("|body(\s*\{.*?\})|si", ".bodyclass\\1", $content);
3d8371be 1983 $secremoveimg = '../images/' . _("sec_remove_eng.png");
691a2d25 1984 /**
0493ed11 1985 * Fix url('blah') declarations.
1986 */
1987 // $content = preg_replace("|url\s*\(\s*([\'\"])\s*\S+script\s*:.*?([\'\"])\s*\)|si",
1988 // "url(\\1$secremoveimg\\2)", $content);
2b665f28 1989
567dc524 1990 // first check for 8bit sequences and disallowed control characters
1991 if (preg_match('/[\16-\37\200-\377]+/',$content)) {
1992 $content = '<!-- style block removed by html filter due to presence of 8bit characters -->';
1993 return array($content, $newpos);
1994 }
1995
2b665f28 1996 // IE Sucks hard. We have a special function for it.
1997 sq_fixIE_idiocy($content);
1998
1999 // remove @import line
2000 $content = preg_replace("/^\s*(@import.*)$/mi","\n<!-- @import rules forbidden -->\n",$content);
2001
5b4884be 2002 // translate ur\l and variations (IE parses that)
2b665f28 2003 // TODO check if the sq_fixIE_idiocy function already handles this.
5b4884be 2004 $content = preg_replace("/(\\\\)?u(\\\\)?r(\\\\)?l(\\\\)?/i", 'url', $content);
567dc524 2005 preg_match_all("/url\s*\((.+)\)/si",$content,$aMatch);
2006 if (count($aMatch)) {
2007 $aValue = $aReplace = array();
2008 foreach($aMatch[1] as $sMatch) {
2009 // url value
2010 $urlvalue = $sMatch;
2011 sq_fix_url('style',$urlvalue, $message, $id, $mailbox,"'");
2012 $aValue[] = $sMatch;
2013 $aReplace[] = $urlvalue;
0493ed11 2014 }
567dc524 2015 $content = str_replace($aValue,$aReplace,$content);
691a2d25 2016 }
567dc524 2017
0493ed11 2018 /**
2019 * Remove any backslashes, entities, and extraneous whitespace.
2020 */
2021 $contentTemp = $content;
2022 sq_defang($contentTemp);
2023 sq_unspace($contentTemp);
a3daaaf3 2024
691a2d25 2025 /**
8bd0068d 2026 * Fix stupid css declarations which lead to vulnerabilities
2027 * in IE.
2028 */
5db90261 2029 $match = Array('/\/\*.*\*\//',
2030 '/expression/i',
0493ed11 2031 '/behaviou*r/i',
2032 '/binding/i',
2b665f28 2033 '/include-source/i',
2034 '/javascript/i',
2035 '/script/i');
2036 $replace = Array('','idiocy', 'idiocy', 'idiocy', 'idiocy', 'idiocy', 'idiocy');
0493ed11 2037 $contentNew = preg_replace($match, $replace, $contentTemp);
2038 if ($contentNew !== $contentTemp) {
2039 // insecure css declarations are used. From now on we don't care
2040 // anymore if the css is destroyed by sq_deent, sq_unspace or sq_unbackslash
2041 $content = $contentNew;
2042 }
7d06541f 2043 return array($content, $newpos);
691a2d25 2044}
a3daaaf3 2045
0493ed11 2046
691a2d25 2047/**
8bd0068d 2048 * This function converts cid: url's into the ones that can be viewed in
2049 * the browser.
2050 *
2051 * @param $message the message object
2052 * @param $id the message id
2053 * @param $cidurl the cid: url.
2054 * @param $mailbox the message mailbox
2055 * @return a string with a http-friendly url
2056 */
b3af12ef 2057function sq_cid2http($message, $id, $cidurl, $mailbox){
691a2d25 2058 /**
8bd0068d 2059 * Get rid of quotes.
2060 */
691a2d25 2061 $quotchar = substr($cidurl, 0, 1);
2dd879b8 2062 if ($quotchar == '"' || $quotchar == "'"){
2063 $cidurl = str_replace($quotchar, "", $cidurl);
2064 } else {
2065 $quotchar = '';
2066 }
691a2d25 2067 $cidurl = substr(trim($cidurl), 4);
0493ed11 2068
2069 $match_str = '/\{.*?\}\//';
2070 $str_rep = '';
2071 $cidurl = preg_replace($match_str, $str_rep, $cidurl);
2072
e5e9381a 2073 $linkurl = find_ent_id($cidurl, $message);
2074 /* in case of non-save cid links $httpurl should be replaced by a sort of
8bd0068d 2075 unsave link image */
e5e9381a 2076 $httpurl = '';
c8f5f606 2077
8bd0068d 2078 /**
2079 * This is part of a fix for Outlook Express 6.x generating
2080 * cid URLs without creating content-id headers. These images are
2081 * not part of the multipart/related html mail. The html contains
2082 * <img src="cid:{some_id}/image_filename.ext"> references to
2083 * attached images with as goal to render them inline although
2084 * the attachment disposition property is not inline.
2085 */
c8f5f606 2086
2087 if (empty($linkurl)) {
2088 if (preg_match('/{.*}\//', $cidurl)) {
2089 $cidurl = preg_replace('/{.*}\//','', $cidurl);
2090 if (!empty($cidurl)) {
2091 $linkurl = find_ent_id($cidurl, $message);
2092 }
2093 }
2094 }
f8a1ed5a 2095
c8f5f606 2096 if (!empty($linkurl)) {
6b04287c 2097 $httpurl = $quotchar . SM_PATH . 'src/download.php?absolute_dl=true&amp;' .
8bd0068d 2098 "passed_id=$id&amp;mailbox=" . urlencode($mailbox) .
2099 '&amp;ent_id=' . $linkurl . $quotchar;
c8f5f606 2100 } else {
2101 /**
2102 * If we couldn't generate a proper img url, drop in a blank image
2103 * instead of sending back empty, otherwise it causes unusual behaviour
2104 */
bc017c1d 2105 $httpurl = $quotchar . SM_PATH . 'images/blank.png' . $quotchar;
e5e9381a 2106 }
f8a1ed5a 2107
691a2d25 2108 return $httpurl;
2109}
2110
2111/**
8bd0068d 2112 * This function changes the <body> tag into a <div> tag since we
2113 * can't really have a body-within-body.
2114 *
2115 * @param $attary an array of attributes and values of <body>
2116 * @param $mailbox mailbox we're currently reading (for cid2http)
2117 * @param $message current message (for cid2http)
2118 * @param $id current message id (for cid2http)
2119 * @return a modified array of attributes to be set for <div>
2120 */
2dd879b8 2121function sq_body2div($attary, $mailbox, $message, $id){
b583c3e8 2122 $me = 'sq_body2div';
3d8371be 2123 $divattary = Array('class' => "'bodyclass'");
b583c3e8 2124 $text = '#000000';
c189a963 2125 $has_bgc_stl = $has_txt_stl = false;
b583c3e8 2126 $styledef = '';
691a2d25 2127 if (is_array($attary) && sizeof($attary) > 0){
2128 foreach ($attary as $attname=>$attvalue){
2129 $quotchar = substr($attvalue, 0, 1);
2130 $attvalue = str_replace($quotchar, "", $attvalue);
2131 switch ($attname){
3d8371be 2132 case 'background':
8bd0068d 2133 $attvalue = sq_cid2http($message, $id, $attvalue, $mailbox);
3d8371be 2134 $styledef .= "background-image: url('$attvalue'); ";
2135 break;
2136 case 'bgcolor':
c189a963 2137 $has_bgc_stl = true;
3d8371be 2138 $styledef .= "background-color: $attvalue; ";
2139 break;
2140 case 'text':
c189a963 2141 $has_txt_stl = true;
3d8371be 2142 $styledef .= "color: $attvalue; ";
2143 break;
691a2d25 2144 }
a3daaaf3 2145 }
c189a963 2146 // Outlook defines a white bgcolor and no text color. This can lead to
2147 // white text on a white bg with certain themes.
2148 if ($has_bgc_stl && !$has_txt_stl) {
2149 $styledef .= "color: $text; ";
2150 }
691a2d25 2151 if (strlen($styledef) > 0){
2152 $divattary{"style"} = "\"$styledef\"";
2153 }
2154 }
2155 return $divattary;
2156}
a3daaaf3 2157
691a2d25 2158/**
8bd0068d 2159 * This is the main function and the one you should actually be calling.
2160 * There are several variables you should be aware of an which need
2161 * special description.
2162 *
2163 * Since the description is quite lengthy, see it here:
2164 * http://linux.duke.edu/projects/mini/htmlfilter/
2165 *
2166 * @param $body the string with HTML you wish to filter
2167 * @param $tag_list see description above
2168 * @param $rm_tags_with_content see description above
2169 * @param $self_closing_tags see description above
2170 * @param $force_tag_closing see description above
2171 * @param $rm_attnames see description above
2172 * @param $bad_attvals see description above
2173 * @param $add_attr_to_tag see description above
2174 * @param $message message object
2175 * @param $id message id
2176 * @return sanitized html safe to show on your pages.
2177 */
da2415c1 2178function sq_sanitize($body,
8bd0068d 2179 $tag_list,
2180 $rm_tags_with_content,
2181 $self_closing_tags,
2182 $force_tag_closing,
2183 $rm_attnames,
2184 $bad_attvals,
2185 $add_attr_to_tag,
2186 $message,
2187 $id,
2188 $mailbox
2189 ){
b583c3e8 2190 $me = 'sq_sanitize';
7d06541f 2191 $rm_tags = array_shift($tag_list);
691a2d25 2192 /**
8bd0068d 2193 * Normalize rm_tags and rm_tags_with_content.
2194 */
7d06541f 2195 @array_walk($tag_list, 'sq_casenormalize');
691a2d25 2196 @array_walk($rm_tags_with_content, 'sq_casenormalize');
2197 @array_walk($self_closing_tags, 'sq_casenormalize');
2198 /**
8bd0068d 2199 * See if tag_list is of tags to remove or tags to allow.
2200 * false means remove these tags
2201 * true means allow these tags
2202 */
691a2d25 2203 $curpos = 0;
2204 $open_tags = Array();
2dd879b8 2205 $trusted = "\n<!-- begin sanitized html -->\n";
691a2d25 2206 $skip_content = false;
bb8d0799 2207 /**
8bd0068d 2208 * Take care of netscape's stupid javascript entities like
2209 * &{alert('boo')};
2210 */
bb8d0799 2211 $body = preg_replace("/&(\{.*?\};)/si", "&amp;\\1", $body);
691a2d25 2212
7d06541f 2213 while (($curtag = sq_getnxtag($body, $curpos)) != FALSE){
691a2d25 2214 list($tagname, $attary, $tagtype, $lt, $gt) = $curtag;
2215 $free_content = substr($body, $curpos, $lt-$curpos);
2216 /**
8bd0068d 2217 * Take care of <style>
2218 */
7d06541f 2219 if ($tagname == "style" && $tagtype == 1){
da2415c1 2220 list($free_content, $curpos) =
e60a299a 2221 sq_fixstyle($body, $gt+1, $message, $id, $mailbox);
7d06541f 2222 if ($free_content != FALSE){
2223 $trusted .= sq_tagprint($tagname, $attary, $tagtype);
2224 $trusted .= $free_content;
2225 $trusted .= sq_tagprint($tagname, false, 2);
2226 }
2227 continue;
691a2d25 2228 }
2229 if ($skip_content == false){
2230 $trusted .= $free_content;
691a2d25 2231 }
2232 if ($tagname != FALSE){
2233 if ($tagtype == 2){
2234 if ($skip_content == $tagname){
2235 /**
8bd0068d 2236 * Got to the end of tag we needed to remove.
2237 */
691a2d25 2238 $tagname = false;
2239 $skip_content = false;
2240 } else {
2241 if ($skip_content == false){
c828931c 2242 if ($tagname == "body"){
2243 $tagname = "div";
2dd879b8 2244 }
da2415c1 2245 if (isset($open_tags{$tagname}) &&
8bd0068d 2246 $open_tags{$tagname} > 0){
2dd879b8 2247 $open_tags{$tagname}--;
691a2d25 2248 } else {
2dd879b8 2249 $tagname = false;
691a2d25 2250 }
691a2d25 2251 }
2252 }
2253 } else {
2254 /**
8bd0068d 2255 * $rm_tags_with_content
2256 */
691a2d25 2257 if ($skip_content == false){
2258 /**
8bd0068d 2259 * See if this is a self-closing type and change
2260 * tagtype appropriately.
2261 */
691a2d25 2262 if ($tagtype == 1
8bd0068d 2263 && in_array($tagname, $self_closing_tags)){
2dd879b8 2264 $tagtype = 3;
691a2d25 2265 }
2266 /**
8bd0068d 2267 * See if we should skip this tag and any content
2268 * inside it.
2269 */
691a2d25 2270 if ($tagtype == 1 &&
8bd0068d 2271 in_array($tagname, $rm_tags_with_content)){
691a2d25 2272 $skip_content = $tagname;
2273 } else {
da2415c1 2274 if (($rm_tags == false
8bd0068d 2275 && in_array($tagname, $tag_list)) ||
2276 ($rm_tags == true &&
2277 !in_array($tagname, $tag_list))){
691a2d25 2278 $tagname = false;
2279 } else {
2dd879b8 2280 /**
8bd0068d 2281 * Convert body into div.
2282 */
2dd879b8 2283 if ($tagname == "body"){
2284 $tagname = "div";
da2415c1 2285 $attary = sq_body2div($attary, $mailbox,
8bd0068d 2286 $message, $id);
2dd879b8 2287 }
691a2d25 2288 if ($tagtype == 1){
2289 if (isset($open_tags{$tagname})){
2290 $open_tags{$tagname}++;
2291 } else {
2292 $open_tags{$tagname}=1;
2293 }
2294 }
2295 /**
8bd0068d 2296 * This is where we run other checks.
2297 */
691a2d25 2298 if (is_array($attary) && sizeof($attary) > 0){
2299 $attary = sq_fixatts($tagname,
8bd0068d 2300 $attary,
2301 $rm_attnames,
2302 $bad_attvals,
2303 $add_attr_to_tag,
2304 $message,
2305 $id,
2306 $mailbox
2307 );
691a2d25 2308 }
2309 }
2310 }
691a2d25 2311 }
2312 }
2313 if ($tagname != false && $skip_content == false){
2314 $trusted .= sq_tagprint($tagname, $attary, $tagtype);
2315 }
691a2d25 2316 }
2317 $curpos = $gt+1;
a3daaaf3 2318 }
691a2d25 2319 $trusted .= substr($body, $curpos, strlen($body)-$curpos);
2320 if ($force_tag_closing == true){
2321 foreach ($open_tags as $tagname=>$opentimes){
2322 while ($opentimes > 0){
2323 $trusted .= '</' . $tagname . '>';
2324 $opentimes--;
2325 }
2326 }
2327 $trusted .= "\n";
2328 }
2329 $trusted .= "<!-- end sanitized html -->\n";
2330 return $trusted;
2331}
451f74a2 2332
691a2d25 2333/**
8bd0068d 2334 * This is a wrapper function to call html sanitizing routines.
2335 *
2336 * @param $body the body of the message
2337 * @param $id the id of the message
c189a963 2338
2339 * @param $message
2340 * @param $mailbox
2341 * @param boolean $take_mailto_links When TRUE, converts mailto: links
2342 * into internal SM compose links
2343 * (optional; default = TRUE)
8bd0068d 2344 * @return a string with html safe to display in the browser.
2345 */
c189a963 2346function magicHTML($body, $id, $message, $mailbox = 'INBOX', $take_mailto_links =true) {
2347
202bcbcc 2348 // require_once(SM_PATH . 'functions/url_parser.php'); // for $MailTo_PReg_Match
c189a963 2349
691a2d25 2350 global $attachment_common_show_images, $view_unsafe_images,
8bd0068d 2351 $has_unsafe_images;
691a2d25 2352 /**
8bd0068d 2353 * Don't display attached images in HTML mode.
2b665f28 2354 *
d0187bd6 2355 * SB: why?
8bd0068d 2356 */
691a2d25 2357 $attachment_common_show_images = false;
2358 $tag_list = Array(
8bd0068d 2359 false,
2360 "object",
2361 "meta",
2362 "html",
2363 "head",
2364 "base",
2365 "link",
2366 "frame",
2367 "iframe",
2368 "plaintext",
2369 "marquee"
2370 );
691a2d25 2371
2372 $rm_tags_with_content = Array(
8bd0068d 2373 "script",
2374 "applet",
2375 "embed",
2376 "title",
2377 "frameset",
0493ed11 2378 "xmp",
8bd0068d 2379 "xml"
2380 );
691a2d25 2381
2382 $self_closing_tags = Array(
8bd0068d 2383 "img",
2384 "br",
2385 "hr",
2386 "input",
2387 "outbind"
2388 );
691a2d25 2389
2dd879b8 2390 $force_tag_closing = true;
691a2d25 2391
2392 $rm_attnames = Array(
8bd0068d 2393 "/.*/" =>
2394 Array(
2395 "/target/i",
2396 "/^on.*/i",
2397 "/^dynsrc/i",
2398 "/^data.*/i",
2399 "/^lowsrc.*/i"
2400 )
2401 );
691a2d25 2402
2403 $secremoveimg = "../images/" . _("sec_remove_eng.png");
2404 $bad_attvals = Array(
8bd0068d 2405 "/.*/" =>
691a2d25 2406 Array(
0a6ec9b5 2407 "/^src|background/i" =>
8bd0068d 2408 Array(
691a2d25 2409 Array(
8bd0068d 2410 "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si",
2411 "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si",
2412 "/^([\'\"])\s*about\s*:.*([\'\"])/si"
691a2d25 2413 ),
8bd0068d 2414 Array(
2415 "\\1$secremoveimg\\2",
2416 "\\1$secremoveimg\\2",
2417 "\\1$secremoveimg\\2",
8bd0068d 2418 )
2419 ),
0a6ec9b5 2420 "/^href|action/i" =>
8bd0068d 2421 Array(
0a6ec9b5 2422 Array(
8bd0068d 2423 "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si",
2424 "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si",
2425 "/^([\'\"])\s*about\s*:.*([\'\"])/si"
0a6ec9b5 2426 ),
691a2d25 2427 Array(
8bd0068d 2428 "\\1#\\1",
2429 "\\1#\\1",
2430 "\\1#\\1"
02474e43 2431 )
8bd0068d 2432 ),
2433 "/^style/i" =>
2434 Array(
2435 Array(
5db90261 2436 "/\/\*.*\*\//",
8bd0068d 2437 "/expression/i",
2438 "/binding/i",
2439 "/behaviou*r/i",
2440 "/include-source/i",
2441 "/position\s*:\s*absolute/i",
1d935bc2 2442 "/(\\\\)?u(\\\\)?r(\\\\)?l(\\\\)?/i",
8bd0068d 2443 "/url\s*\(\s*([\'\"])\s*\S+script\s*:.*([\'\"])\s*\)/si",
2444 "/url\s*\(\s*([\'\"])\s*mocha\s*:.*([\'\"])\s*\)/si",
2445 "/url\s*\(\s*([\'\"])\s*about\s*:.*([\'\"])\s*\)/si",
2446 "/(.*)\s*:\s*url\s*\(\s*([\'\"]*)\s*\S+script\s*:.*([\'\"]*)\s*\)/si"
2447 ),
2448 Array(
5db90261 2449 "",
8bd0068d 2450 "idiocy",
2451 "idiocy",
2452 "idiocy",
2453 "idiocy",
567dc524 2454 "idiocy",
5b4884be 2455 "url",
8bd0068d 2456 "url(\\1#\\1)",
2457 "url(\\1#\\1)",
2458 "url(\\1#\\1)",
8bd0068d 2459 "\\1:url(\\2#\\3)"
2460 )
691a2d25 2461 )
8bd0068d 2462 )
691a2d25 2463 );
5262d9a6 2464 if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET) ) {
2dd879b8 2465 $view_unsafe_images = false;
45071bd6 2466 }
691a2d25 2467 if (!$view_unsafe_images){
2468 /**
8bd0068d 2469 * Remove any references to http/https if view_unsafe_images set
2470 * to false.
2471 */
02474e43 2472 array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[0],
8bd0068d 2473 '/^([\'\"])\s*https*:.*([\'\"])/si');
02474e43 2474 array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[1],
8bd0068d 2475 "\\1$secremoveimg\\1");
02474e43 2476 array_push($bad_attvals{'/.*/'}{'/^style/i'}[0],
7ef0b415 2477 '/url\([\'\"]?https?:[^\)]*[\'\"]?\)/si');
02474e43 2478 array_push($bad_attvals{'/.*/'}{'/^style/i'}[1],
8bd0068d 2479 "url(\\1$secremoveimg\\1)");
691a2d25 2480 }
451f74a2 2481
691a2d25 2482 $add_attr_to_tag = Array(
8bd0068d 2483 "/^a$/i" =>
c25f2fbb 2484 Array('target'=>'"_blank"',
02474e43 2485 'title'=>'"'._("This external link will open in a new window").'"'
8bd0068d 2486 )
2487 );
da2415c1 2488 $trusted = sq_sanitize($body,
8bd0068d 2489 $tag_list,
2490 $rm_tags_with_content,
2491 $self_closing_tags,
2492 $force_tag_closing,
2493 $rm_attnames,
2494 $bad_attvals,
2495 $add_attr_to_tag,
2496 $message,
2497 $id,
2498 $mailbox
2499 );
567dc524 2500 if (strpos($trusted,$secremoveimg)){
691a2d25 2501 $has_unsafe_images = true;
da2415c1 2502 }
c189a963 2503
2504 // we want to parse mailto's in HTML output, change to SM compose links
2505 // this is a modified version of code from url_parser.php... but Marc is
2506 // right: we need a better filtering implementation; adding this randomly
2507 // here is not a great solution
2508 //
2509 if ($take_mailto_links) {
2510 // parseUrl($trusted); // this even parses URLs inside of tags... too aggressive
2511 global $MailTo_PReg_Match;
202bcbcc 2512 $MailTo_PReg_Match = '/mailto:' . substr($MailTo_PReg_Match, 1) ;
c189a963 2513 if ((preg_match_all($MailTo_PReg_Match, $trusted, $regs)) && ($regs[0][0] != '')) {
2514 foreach ($regs[0] as $i => $mailto_before) {
2515 $mailto_params = $regs[10][$i];
2516 // get rid of any tailing quote since we have to add send_to to the end
2517 //
2518 if (substr($mailto_before, strlen($mailto_before) - 1) == '"')
2519 $mailto_before = substr($mailto_before, 0, strlen($mailto_before) - 1);
2520 if (substr($mailto_params, strlen($mailto_params) - 1) == '"')
2521 $mailto_params = substr($mailto_params, 0, strlen($mailto_params) - 1);
2522
2523 if ($regs[1][$i]) { //if there is an email addr before '?', we need to merge it with the params
2524 $to = 'to=' . $regs[1][$i];
2525 if (strpos($mailto_params, 'to=') > -1) //already a 'to='
2526 $mailto_params = str_replace('to=', $to . '%2C%20', $mailto_params);
2527 else {
2528 if ($mailto_params) //already some params, append to them
2529 $mailto_params .= '&amp;' . $to;
2530 else
2531 $mailto_params .= '?' . $to;
2532 }
2533 }
2534
2535 $url_str = preg_replace(array('/to=/i', '/(?<!b)cc=/i', '/bcc=/i'), array('send_to=', 'send_to_cc=', 'send_to_bcc='), $mailto_params);
2536
2537 // we'll already have target=_blank, no need to allow comp_in_new
2538 // here (which would be a lot more work anyway)
2539 //
2540 global $compose_new_win;
2541 $temp_comp_in_new = $compose_new_win;
2542 $compose_new_win = 0;
2543 $comp_uri = makeComposeLink('src/compose.php' . $url_str, $mailto_before);
2544 $compose_new_win = $temp_comp_in_new;
2545
2546 // remove <a href=" and anything after the next quote (we only
2547 // need the uri, not the link HTML) in compose uri
2548 //
2549 $comp_uri = substr($comp_uri, 9);
2550 $comp_uri = substr($comp_uri, 0, strpos($comp_uri, '"', 1));
2551 $trusted = str_replace($mailto_before, $comp_uri, $trusted);
2552 }
2553 }
2554 }
2555
691a2d25 2556 return $trusted;
451f74a2 2557}
a4a70693 2558
da2415c1 2559/**
8bd0068d 2560 * function SendDownloadHeaders - send file to the browser
2561 *
2562 * Original Source: SM core src/download.php
2563 * moved here to make it available to other code, and separate
2564 * front end from back end functionality.
2565 *
2566 * @param string $type0 first half of mime type
2567 * @param string $type1 second half of mime type
2568 * @param string $filename filename to tell the browser for downloaded file
2569 * @param boolean $force whether to force the download dialog to pop
2570 * @param optional integer $filesize send the Content-Header and length to the browser
2571 * @return void
2572 */
02474e43 2573function SendDownloadHeaders($type0, $type1, $filename, $force, $filesize=0) {
2574 global $languages, $squirrelmail_language;
cfffd60b 2575 $isIE = $isIE6plus = false;
02474e43 2576
2577 sqgetGlobalVar('HTTP_USER_AGENT', $HTTP_USER_AGENT, SQ_SERVER);
2578
2579 if (strstr($HTTP_USER_AGENT, 'compatible; MSIE ') !== false &&
8bd0068d 2580 strstr($HTTP_USER_AGENT, 'Opera') === false) {
cfffd60b 2581 $isIE = true;
02474e43 2582 }
2583
cfffd60b 2584 if (preg_match('/compatible; MSIE ([0-9]+)/', $HTTP_USER_AGENT, $match) &&
2585 ((int)$match[1]) >= 6 && strstr($HTTP_USER_AGENT, 'Opera') === false) {
2586 $isIE6plus = true;
02474e43 2587 }
2588
2589 if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
8bd0068d 2590 function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_downloadfilename')) {
02474e43 2591 $filename =
8bd0068d 2592 call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_downloadfilename', $filename, $HTTP_USER_AGENT);
02474e43 2593 } else {
2594 $filename = ereg_replace('[\\/:\*\?"<>\|;]', '_', str_replace('&nbsp;', ' ', $filename));
2595 }
2596
2597 // A Pox on Microsoft and it's Internet Explorer!
2598 //
2599 // IE has lots of bugs with file downloads.
2600 // It also has problems with SSL. Both of these cause problems
2601 // for us in this function.
2602 //
2603 // See this article on Cache Control headers and SSL
2604 // http://support.microsoft.com/default.aspx?scid=kb;en-us;323308
2605 //
2606 // The best thing you can do for IE is to upgrade to the latest
2607 // version
2608 //set all the Cache Control Headers for IE
2609 if ($isIE) {
2610 $filename=rawurlencode($filename);
2611 header ("Pragma: public");
8bd0068d 2612 header ("Cache-Control: no-store, max-age=0, no-cache, must-revalidate"); // HTTP/1.1
02474e43 2613 header ("Cache-Control: post-check=0, pre-check=0", false);
8bd0068d 2614 header ("Cache-Control: private");
02474e43 2615
2616 //set the inline header for IE, we'll add the attachment header later if we need it
2617 header ("Content-Disposition: inline; filename=$filename");
2618 }
2619
2620 if (!$force) {
2621 // Try to show in browser window
2622 header ("Content-Disposition: inline; filename=\"$filename\"");
2623 header ("Content-Type: $type0/$type1; name=\"$filename\"");
2624 } else {
2625 // Try to pop up the "save as" box
2626
2627 // IE makes this hard. It pops up 2 save boxes, or none.
2628 // http://support.microsoft.com/support/kb/articles/Q238/5/88.ASP
2629 // http://support.microsoft.com/default.aspx?scid=kb;EN-US;260519
2630 // But, according to Microsoft, it is "RFC compliant but doesn't
2631 // take into account some deviations that allowed within the
2632 // specification." Doesn't that mean RFC non-compliant?
2633 // http://support.microsoft.com/support/kb/articles/Q258/4/52.ASP
2634
2635 // all browsers need the application/octet-stream header for this
2636 header ("Content-Type: application/octet-stream; name=\"$filename\"");
2637
2638 // http://support.microsoft.com/support/kb/articles/Q182/3/15.asp
2639 // Do not have quotes around filename, but that applied to
2640 // "attachment"... does it apply to inline too?
2641 header ("Content-Disposition: attachment; filename=\"$filename\"");
2642
cfffd60b 2643 if ($isIE && !$isIE6plus) {
02474e43 2644 // This combination seems to work mostly. IE 5.5 SP 1 has
2645 // known issues (see the Microsoft Knowledge Base)
2646
2647 // This works for most types, but doesn't work with Word files
2648 header ("Content-Type: application/download; name=\"$filename\"");
7e2ff844 2649 header ("Content-Type: application/force-download; name=\"$filename\"");
02474e43 2650 // These are spares, just in case. :-)
2651 //header("Content-Type: $type0/$type1; name=\"$filename\"");
2652 //header("Content-Type: application/x-msdownload; name=\"$filename\"");
2653 //header("Content-Type: application/octet-stream; name=\"$filename\"");
7e2ff844 2654 } else if ($isIE) {
2655 // This is to prevent IE for MIME sniffing and auto open a file in IE
2656 header ("Content-Type: application/force-download; name=\"$filename\"");
02474e43 2657 } else {
2658 // another application/octet-stream forces download for Netscape
2659 header ("Content-Type: application/octet-stream; name=\"$filename\"");
2660 }
2661 }
2662
2663 //send the content-length header if the calling function provides it
2664 if ($filesize > 0) {
2665 header("Content-Length: $filesize");
2666 }
07c49f57 2667
8d863f64 2668} // end fn SendDownloadHeaders
Unnamed repository; edit this file 'description' to name the repository.