projects / squirrelmail.git / blame
? search:
summary | shortlog | log | commit | commitdiff | tree
blob | blame (incremental) | history | HEAD
tweak comments
[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 *
47ccfad4 9 * @copyright &copy; 1999-2006 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;
c96c32f4 63 /* removed urldecode because $_GET is auto urldecoded ??? */
a48eba8f 64 displayPageHeader( $color, $mailbox );
5e8de8b6 65 $errormessage = _("SquirrelMail could not decode the bodystructure of the message");
8bd0068d 66 $errormessage .= '<br />'._("The bodystructure provided by your IMAP server:").'<br /><br />';
472e7acb 67 $errormessage .= '<pre>' . htmlspecialchars($read) . '</pre>';
9de42168 68 plain_error_message( $errormessage, $color );
3d8371be 69 echo '</body></html>';
9de42168 70 exit;
71 }
a4a70693 72 if (count($flags)) {
7a9e9c89 73 foreach ($flags as $flag) {
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];
150 } else {
151 global $where, $what, $mailbox, $passed_id, $startMessage;
3d8371be 152 $par = 'mailbox=' . urlencode($mailbox) . '&amp;passed_id=' . $passed_id;
451f74a2 153 if (isset($where) && isset($what)) {
3d8371be 154 $par .= '&amp;where=' . urlencode($where) . '&amp;what=' . urlencode($what);
a3daaaf3 155 } else {
3d8371be 156 $par .= '&amp;startMessage=' . $startMessage . '&amp;show_more=0';
451f74a2 157 }
e5ea9327 158 $par .= '&amp;response=' . urlencode($response) .
8bd0068d 159 '&amp;message=' . urlencode($message) .
160 '&amp;topline=' . urlencode($topline);
a019eeb8 161
8bd0068d 162 echo '<tt><br />' .
02474e43 163 '<table width="80%"><tr>' .
164 '<tr><td colspan="2">' .
165 _("Body retrieval error. The reason for this is most probably that the message is malformed.") .
166 '</td></tr>' .
167 '<tr><td><b>' . _("Command:") . "</td><td>$cmd</td></tr>" .
168 '<tr><td><b>' . _("Response:") . "</td><td>$response</td></tr>" .
169 '<tr><td><b>' . _("Message:") . "</td><td>$message</td></tr>" .
170 '<tr><td><b>' . _("FETCH line:") . "</td><td>$topline</td></tr>" .
171 "</table><br /></tt></font><hr />";
346817d4 172
6201339c 173 $data = sqimap_run_command ($imap_stream, "FETCH $passed_id BODY[]", true, $response, $message, TRUE);
451f74a2 174 array_shift($data);
175 $wholemessage = implode('', $data);
a019eeb8 176
346817d4 177 $ret = $wholemessage;
a3daaaf3 178 }
3d8371be 179 return $ret;
451f74a2 180}
d4467150 181
6cc670de 182function mime_print_body_lines ($imap_stream, $id, $ent_id=1, $encoding, $rStream='php://stdout') {
1035e159 183
3d8371be 184 /* Don't kill the connection if the browser is over a dialup
8bd0068d 185 * and it would take over 30 seconds to download it.
186 * Don't call set_time_limit in safe mode.
187 */
b7206e1d 188
3d8371be 189 if (!ini_get('safe_mode')) {
b7206e1d 190 set_time_limit(0);
191 }
7c7b74b3 192 /* in case of base64 encoded attachments, do not buffer them.
8bd0068d 193 Instead, echo the decoded attachment directly to screen */
7c7b74b3 194 if (strtolower($encoding) == 'base64') {
195 if (!$ent_id) {
7591f143 196 $query = "FETCH $id BODY[]";
7c7b74b3 197 } else {
7591f143 198 $query = "FETCH $id BODY[$ent_id]";
7c7b74b3 199 }
7591f143 200 sqimap_run_command($imap_stream,$query,true,$response,$message,TRUE,'sqimap_base64_decode',$rStream,true);
1d142b8d 201 } else {
7591f143 202 $body = mime_fetch_body ($imap_stream, $id, $ent_id);
203 if (is_resource($rStream)) {
204 fputs($rStream,decodeBody($body,$encoding));
205 } else {
206 echo decodeBody($body, $encoding);
207 }
1d142b8d 208 }
346817d4 209
da2415c1 210 /*
8bd0068d 211 TODO, use the same method for quoted printable.
212 However, I assume that quoted printable attachments aren't that large
213 so the performancegain / memory usage drop will be minimal.
214 If we decide to add that then we need to adapt sqimap_fread because
215 we need to split te result on \n and fread doesn't stop at \n. That
216 means we also should provide $results from sqimap_fread (by ref) to
217 te function and set $no_return to false. The $filter function for
218 quoted printable should handle unsetting of $results.
219 */
da2415c1 220 /*
8bd0068d 221 TODO 2: find out how we write to the output stream php://stdout. fwrite
222 doesn't work because 'php://stdout isn't a stream.
223 */
7c7b74b3 224
5d9c6f73 225 return;
451f74a2 226}
beb9e459 227
451f74a2 228/* -[ END MIME DECODING ]----------------------------------------------------------- */
d4467150 229
3d8371be 230/* This is here for debugging purposes. It will print out a list
8bd0068d 231 * of all the entity IDs that are in the $message object.
232 */
451f74a2 233function listEntities ($message) {
3d8371be 234 if ($message) {
3c621ba1 235 echo "<tt>" . $message->entity_id . ' : ' . $message->type0 . '/' . $message->type1 . ' parent = '. $message->parent->entity_id. '<br />';
3d8371be 236 for ($i = 0; isset($message->entities[$i]); $i++) {
237 echo "$i : ";
238 $msg = listEntities($message->entities[$i]);
239
240 if ($msg) {
241 echo "return: ";
242 return $msg;
243 }
244 }
a4a70693 245 }
451f74a2 246}
f0c4dc12 247
f792c641 248function getPriorityStr($priority) {
3d8371be 249 $priority_level = substr($priority,0,1);
250
251 switch($priority_level) {
252 /* Check for a higher then normal priority. */
253 case '1':
254 case '2':
255 $priority_string = _("High");
256 break;
257
258 /* Check for a lower then normal priority. */
259 case '4':
260 case '5':
261 $priority_string = _("Low");
262 break;
263
264 /* Check for a normal priority. */
265 case '3':
266 default:
267 $priority_level = '3';
268 $priority_string = _("Normal");
269 break;
270
271 }
272 return $priority_string;
f792c641 273}
274
451f74a2 275/* returns a $message object for a particular entity id */
276function getEntity ($message, $ent_id) {
a4a70693 277 return $message->getEntity($ent_id);
451f74a2 278}
8beafbbc 279
3d8371be 280/* translateText
8bd0068d 281 * Extracted from strings.php 23/03/2002
282 */
da4c66e8 283
284function translateText(&$body, $wrap_at, $charset) {
3d8371be 285 global $where, $what; /* from searching */
286 global $color; /* color theme */
da4c66e8 287
202bcbcc 288 // require_once(SM_PATH . 'functions/url_parser.php');
da4c66e8 289
290 $body_ary = explode("\n", $body);
da4c66e8 291 for ($i=0; $i < count($body_ary); $i++) {
292 $line = $body_ary[$i];
293 if (strlen($line) - 2 >= $wrap_at) {
c7aff938 294 sqWordWrap($line, $wrap_at, $charset);
da4c66e8 295 }
296 $line = charset_decode($charset, $line);
297 $line = str_replace("\t", ' ', $line);
298
299 parseUrl ($line);
300
3d8371be 301 $quotes = 0;
da4c66e8 302 $pos = 0;
3d8371be 303 $j = strlen($line);
da4c66e8 304
3d8371be 305 while ($pos < $j) {
da4c66e8 306 if ($line[$pos] == ' ') {
3d8371be 307 $pos++;
da4c66e8 308 } else if (strpos($line, '&gt;', $pos) === $pos) {
309 $pos += 4;
3d8371be 310 $quotes++;
da4c66e8 311 } else {
312 break;
313 }
314 }
3d8371be 315
83c94382 316 if ($quotes % 2) {
d0814c02 317 $line = '<span class="quote1">' . $line . '</span>';
4c25967c 318 } elseif ($quotes) {
d0814c02 319 $line = '<span class="quote2">' . $line . '</span>';
da4c66e8 320 }
3d8371be 321
da4c66e8 322 $body_ary[$i] = $line;
323 }
324 $body = '<pre>' . implode("\n", $body_ary) . '</pre>';
325}
326
a2bfcbce 327/**
da1b55ad 328 * This returns a parsed string called $body. That string can then
329 * be displayed as the actual message in the HTML. It contains
330 * everything needed, including HTML Tags, Attachments at the
331 * bottom, etc.
f8a1ed5a 332 *
da1b55ad 333 * Since 1.2.0 function uses message_body hook.
334 * Till 1.3.0 function included output of formatAttachments().
335 *
336 * @param resource $imap_stream imap connection resource
337 * @param object $message squirrelmail message object
338 * @param array $color squirrelmail color theme array
339 * @param integer $wrap_at number of characters per line
340 * @param string $ent_num (since 1.3.0) message part id
341 * @param integer $id (since 1.3.0) message id
342 * @param string $mailbox (since 1.3.0) imap folder name
41f701c1 343 * @param boolean $clean (since 1.5.1) Do not output stuff that's irrelevant for the printable version.
da1b55ad 344 * @return string html formated message text
345 */
a2bfcbce 346function formatBody($imap_stream, $message, $color, $wrap_at, $ent_num, $id, $mailbox='INBOX', $clean=FALSE) {
3d8371be 347 /* This if statement checks for the entity to show as the
8bd0068d 348 * primary message. To add more of them, just put them in the
349 * order that is their priority.
350 */
ce68b76b 351 global $startMessage, $languages, $squirrelmail_language,
40a34e57 352 $show_html_default, $sort, $has_unsafe_images, $passed_ent_id,
14c85e39 353 $use_iframe, $iframe_height, $download_and_unsafe_link,
354 $download_href, $unsafe_image_toggle_href, $unsafe_image_toggle_text;
2c25d36a 355
356 // workaround for not updated config.php
357 if (! isset($use_iframe)) $use_iframe = false;
77bfbd2e 358
5262d9a6 359 if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET) ) {
360 $view_unsafe_images = false;
77bfbd2e 361 }
d03c24f4 362
cc34b00d 363 $body = '';
23bcec6f 364 $urlmailbox = urlencode($mailbox);
451f74a2 365 $body_message = getEntity($message, $ent_num);
366 if (($body_message->header->type0 == 'text') ||
8bd0068d 367 ($body_message->header->type0 == 'rfc822')) {
3d8371be 368 $body = mime_fetch_body ($imap_stream, $id, $ent_num);
451f74a2 369 $body = decodeBody($body, $body_message->header->encoding);
e842b215 370
371 if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
8bd0068d 372 function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_decode')) {
e842b215 373 if (mb_detect_encoding($body) != 'ASCII') {
33a55f5a 374 $body = call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_decode',$body);
e842b215 375 }
376 }
451f74a2 377 $hookResults = do_hook("message_body", $body);
378 $body = $hookResults[1];
23bcec6f 379
3d8371be 380 /* If there are other types that shouldn't be formatted, add
8bd0068d 381 * them here.
382 */
3d8371be 383
451f74a2 384 if ($body_message->header->type1 == 'html') {
3d8371be 385 if ($show_html_default <> 1) {
85015544 386 $entity_conv = array('&nbsp;' => ' ',
8bd0068d 387 '<p>' => "\n",
388 '<P>' => "\n",
389 '<br>' => "\n",
390 '<BR>' => "\n",
391 '<br />' => "\n",
392 '<BR />' => "\n",
393 '&gt;' => '>',
394 '&lt;' => '<');
85015544 395 $body = strtr($body, $entity_conv);
3d8371be 396 $body = strip_tags($body);
85015544 397 $body = trim($body);
398 translateText($body, $wrap_at,
8bd0068d 399 $body_message->header->getParameter('charset'));
2c25d36a 400 } elseif ($use_iframe && ! $clean) {
401 // $clean is used to remove iframe in printable view.
402
84410f31 403 /**
404 * If we don't add html message between iframe tags,
405 * we must detect unsafe images and modify $has_unsafe_images.
f8a1ed5a 406 */
758a7889 407 $html_body = magicHTML($body, $id, $message, $mailbox);
b6c52e61 408 // Convert character set in order to display html mails in different character set
409 $html_body = charset_decode($body_message->header->getParameter('charset'),$html_body,false,true);
84410f31 410
2c25d36a 411 // creating iframe url
412 $iframeurl=sqm_baseuri().'src/view_html.php?'
f8a1ed5a 413 . 'mailbox=' . $urlmailbox
2c25d36a 414 . '&amp;passed_id=' . $id
415 . '&amp;ent_id=' . $ent_num
416 . '&amp;view_unsafe_images=' . (int) $view_unsafe_images;
417
79d58d4c 418 global $oTemplate;
419 $oTemplate->assign('iframe_url', $iframeurl);
420 $oTemplate->assign('html_body', $html_body);
2b665f28 421
79d58d4c 422 $body = $oTemplate->fetch('read_html_iframe.tpl');
a3daaaf3 423 } else {
2c25d36a 424 // old way of html rendering
f8a1ed5a 425 $body = magicHTML($body, $id, $message, $mailbox);
b6c52e61 426 /**
758a7889 427 * convert character set. charset_decode does not remove html special chars
b6c52e61 428 * applied by magicHTML functions and does not sanitize them second time if
758a7889 429 * fourth argument is true.
430 */
b6c52e61 431 $body = charset_decode($body_message->header->getParameter('charset'),$body,false,true);
a3daaaf3 432 }
451f74a2 433 } else {
3d8371be 434 translateText($body, $wrap_at,
8bd0068d 435 $body_message->header->getParameter('charset'));
451f74a2 436 }
a2bfcbce 437
438 // if this is the clean display (i.e. printer friendly), stop here.
439 if ( $clean ) {
440 return $body;
441 }
442
40a34e57 443 $download_and_unsafe_link = '';
444
83cf04bd 445 $link = 'passed_id=' . $id . '&amp;ent_id='.$ent_num.
8bd0068d 446 '&amp;mailbox=' . $urlmailbox .'&amp;sort=' . $sort .
447 '&amp;startMessage=' . $startMessage . '&amp;show_more=0';
08b7f7cc 448 if (isset($passed_ent_id)) {
449 $link .= '&amp;passed_ent_id='.$passed_ent_id;
450 }
14c85e39 451 $download_href = SM_PATH . 'src/download.php?absolute_dl=true&amp;' . $link;
452 $download_and_unsafe_link .= '&nbsp;|&nbsp;<a href="'. $download_href .'">' . _("Download this as a file") . '</a>';
7aad7b77 453 if ($view_unsafe_images) {
23f617b8 454 $text = _("Hide Unsafe Images");
7aad7b77 455 } else {
08b7f7cc 456 if (isset($has_unsafe_images) && $has_unsafe_images) {
457 $link .= '&amp;view_unsafe_images=1';
458 $text = _("View Unsafe Images");
459 } else {
460 $text = '';
461 }
3d8371be 462 }
83cf04bd 463 if($text != '') {
14c85e39 464 $unsafe_image_toggle_href = SM_PATH . 'src/read_body.php?'.$link;
465 $unsafe_image_toggle_text = $text;
466 $download_and_unsafe_link .= '&nbsp;|&nbsp;<a href="'. $unsafe_image_toggle_href .'">' . $text . '</a>';
83cf04bd 467 }
3d8371be 468 }
469 return $body;
451f74a2 470}
b74ba498 471
da1b55ad 472/**
d67f519a 473 * Generate attachments array for passing to templates. Separated from
474 * formatAttachments() below so that the same array can be given to the
475 * print-friendly version.
2b665f28 476 *
d67f519a 477 * @since 1.5.2
da1b55ad 478 * @param object $message SquirrelMail message object
479 * @param array $exclude_id message parts that are not attachments.
480 * @param string $mailbox mailbox name
481 * @param integer $id message id
da1b55ad 482 */
d67f519a 483function buildAttachmentArray($message, $exclude_id, $mailbox, $id) {
484 global $where, $what, $startMessage, $color, $passed_ent_id, $base_uri;
451f74a2 485
23bcec6f 486 $att_ar = $message->getAttachments($exclude_id);
23bcec6f 487 $urlMailbox = urlencode($mailbox);
488
d67f519a 489 $attachments = array();
23bcec6f 490 foreach ($att_ar as $att) {
fdc9d9b5 491 $ent = $att->entity_id;
2e25760a 492 $header = $att->header;
f0c4dc12 493 $type0 = strtolower($header->type0);
494 $type1 = strtolower($header->type1);
2e25760a 495 $name = '';
d0187bd6 496 $links = array();
21dab2dc 497 $links['download link']['text'] = _("Download");
202bcbcc 498 $links['download link']['href'] = $base_uri .
8bd0068d 499 "src/download.php?absolute_dl=true&amp;passed_id=$id&amp;mailbox=$urlMailbox&amp;ent_id=$ent";
2b665f28 500
2e25760a 501 if ($type0 =='message' && $type1 == 'rfc822') {
202bcbcc 502 $default_page = $base_uri . 'src/read_body.php';
2e25760a 503 $rfc822_header = $att->rfc822_header;
098ea084 504 $filename = $rfc822_header->subject;
6cc08d8b 505 if (trim( $filename ) == '') {
506 $filename = 'untitled-[' . $ent . ']' ;
08b7f7cc 507 }
2e25760a 508 $from_o = $rfc822_header->from;
509 if (is_object($from_o)) {
04ea844e 510 $from_name = decodeHeader($from_o->getAddress(false));
7e697748 511 } elseif (is_array($from_o) && count($from_o) && is_object($from_o[0])) {
512 // something weird happens when a digest message is opened and you return to the digest
513 // now the from object is part of an array. Probably the parseHeader call overwrites the info
514 // retrieved from the bodystructure in a different way. We need to fix this later.
515 // possible starting point, do not fetch header we already have and inspect how
516 // the rfc822_header object behaves.
517 $from_name = decodeHeader($from_o[0]->getAddress(false));
2e25760a 518 } else {
519 $from_name = _("Unknown sender");
f0c4dc12 520 }
d0187bd6 521 $description = _("From").': '.$from_name;
23bcec6f 522 } else {
202bcbcc 523 $default_page = $base_uri . 'src/download.php';
02474e43 524 $filename = $att->getFilename();
f810c0b2 525 if ($header->description) {
098ea084 526 $description = decodeHeader($header->description);
f810c0b2 527 } else {
3d8371be 528 $description = '';
529 }
2e25760a 530 }
531
532 $display_filename = $filename;
3d8371be 533 if (isset($passed_ent_id)) {
534 $passed_ent_id_link = '&amp;passed_ent_id='.$passed_ent_id;
535 } else {
536 $passed_ent_id_link = '';
537 }
538 $defaultlink = $default_page . "?startMessage=$startMessage"
8bd0068d 539 . "&amp;passed_id=$id&amp;mailbox=$urlMailbox"
540 . '&amp;ent_id='.$ent.$passed_ent_id_link;
2e25760a 541 if ($where && $what) {
8bd0068d 542 $defaultlink .= '&amp;where='. urlencode($where).'&amp;what='.urlencode($what);
2e25760a 543 }
7e2ff844 544 // IE does make use of mime content sniffing. Forcing a download
545 // prohibit execution of XSS inside an application/octet-stream attachment
546 if ($type0 == 'application' && $type1 == 'octet-stream') {
547 $defaultlink .= '&amp;absolute_dl=true';
548 }
2b665f28 549
3d8371be 550 /* This executes the attachment hook with a specific MIME-type.
8bd0068d 551 * If that doesn't have results, it tries if there's a rule
9b94c54d 552 * for a more generic type. Finally, a hook for ALL attachment
553 * types is run as well.
8bd0068d 554 */
3d8371be 555 $hookresults = do_hook("attachment $type0/$type1", $links,
8bd0068d 556 $startMessage, $id, $urlMailbox, $ent, $defaultlink,
557 $display_filename, $where, $what);
3d8371be 558 if(count($hookresults[1]) <= 1) {
559 $hookresults = do_hook("attachment $type0/*", $links,
8bd0068d 560 $startMessage, $id, $urlMailbox, $ent, $defaultlink,
561 $display_filename, $where, $what);
2e25760a 562 }
9b94c54d 563 $hookresults = do_hook("attachment */*", $hookresults[1],
2b665f28 564 $startMessage, $id, $urlMailbox, $ent, $hookresults[6],
565 $display_filename, $where, $what);
451f74a2 566
3d8371be 567 $links = $hookresults[1];
568 $defaultlink = $hookresults[6];
77b88425 569
d67f519a 570 $this_attachment = array();
571 $this_attachment['Name'] = decodeHeader($display_filename);
572 $this_attachment['Description'] = $description;
573 $this_attachment['DefaultHREF'] = $defaultlink;
574 $this_attachment['DownloadHREF'] = $links['download link']['href'];
575 $this_attachment['ViewHREF'] = isset($links['attachment_common']) ? $links['attachment_common']['href'] : '';
576 $this_attachment['Size'] = $header->size;
577 $this_attachment['ContentType'] = htmlspecialchars($type0 .'/'. $type1);
578 $this_attachment['OtherLinks'] = array();
3d8371be 579 foreach ($links as $val) {
d0187bd6 580 if ($val['text']==_("Download") || $val['text'] == _("View"))
581 continue;
582 if (empty($val['text']) && empty($val['extra']))
583 continue;
2b665f28 584
d67f519a 585 $temp = array();
586 $temp['HREF'] = $val['href'];
587 $temp['Text'] = (empty($val['text']) ? '' : $val['text']) . (empty($val['extra']) ? '' : $val['extra']);
588 $this_attachment['OtherLinks'][] = $temp;
2e25760a 589 }
d67f519a 590 $attachments[] = $this_attachment;
2b665f28 591
3d8371be 592 unset($links);
2e25760a 593 }
2b665f28 594
d67f519a 595 return $attachments;
596}
597
598/**
599 * Displays attachment links and information
600 *
601 * Since 1.3.0 function is not included in formatBody() call.
602 *
603 * Since 1.0.2 uses attachment $type0/$type1 hook.
604 * Since 1.2.5 uses attachment $type0/* hook.
605 * Since 1.5.0 uses attachments_bottom hook.
606 * Since 1.5.2 uses templates and does *not* return a value.
607 *
608 * @param object $message SquirrelMail message object
609 * @param array $exclude_id message parts that are not attachments.
610 * @param string $mailbox mailbox name
611 * @param integer $id message id
612 */
613function formatAttachments($message, $exclude_id, $mailbox, $id) {
614 global $oTemplate;
2b665f28 615
d67f519a 616 $attach = buildAttachmentArray($message, $exclude_id, $mailbox, $id);
d0187bd6 617
618 $oTemplate->assign('attachments', $attach);
619 $oTemplate->display('read_attachments.tpl');
451f74a2 620}
b74ba498 621
7c7b74b3 622function sqimap_base64_decode(&$string) {
7c0ec1d8 623
b17a8968 624 // Base64 encoded data goes in pairs of 4 bytes. To achieve on the
7c0ec1d8 625 // fly decoding (to reduce memory usage) you have to check if the
626 // data has incomplete pairs
627
b17a8968 628 // Remove the noise in order to check if the 4 bytes pairs are complete
7c0ec1d8 629 $string = str_replace(array("\r\n","\n", "\r", " "),array('','','',''),$string);
630
42ce44f8 631 $sStringRem = '';
7c0ec1d8 632 $iMod = strlen($string) % 4;
633 if ($iMod) {
634 $sStringRem = substr($string,-$iMod);
b17a8968 635 // Check if $sStringRem contains padding characters
7c0ec1d8 636 if (substr($sStringRem,-1) != '=') {
637 $string = substr($string,0,-$iMod);
638 } else {
639 $sStringRem = '';
640 }
641 }
7c7b74b3 642 $string = base64_decode($string);
7c0ec1d8 643 return $sStringRem;
7c7b74b3 644}
645
fdf7cef1 646/**
647 * Decodes encoded message body
648 *
649 * This function decodes the body depending on the encoding type.
650 * Currently quoted-printable and base64 encodings are supported.
651 * decode_body hook was added to this function in 1.4.2/1.5.0
652 * @param string $body encoded message body
653 * @param string $encoding used encoding
654 * @return string decoded string
655 * @since 1.0
656 */
451f74a2 657function decodeBody($body, $encoding) {
83be314a 658
b583c3e8 659 $body = str_replace("\r\n", "\n", $body);
660 $encoding = strtolower($encoding);
3d8371be 661
5166f86a 662 $encoding_handler = do_hook_function('decode_body', $encoding);
663
664
665 // plugins get first shot at decoding the body
666 //
667 if (!empty($encoding_handler) && function_exists($encoding_handler)) {
668 $body = $encoding_handler('decode', $body);
669
fdf7cef1 670 } elseif ($encoding == 'quoted-printable' ||
8bd0068d 671 $encoding == 'quoted_printable') {
fdf7cef1 672 /**
673 * quoted_printable_decode() function is broken in older
674 * php versions. Text with \r\n decoding was fixed only
758a7889 675 * in php 4.3.0. Minimal code requirement 4.0.4 +
fdf7cef1 676 * str_replace("\r\n", "\n", $body); call.
677 */
b583c3e8 678 $body = quoted_printable_decode($body);
fdf7cef1 679 } elseif ($encoding == 'base64') {
b583c3e8 680 $body = base64_decode($body);
681 }
3d8371be 682
b583c3e8 683 // All other encodings are returned raw.
3d8371be 684 return $body;
451f74a2 685}
686
9f7f68c3 687/**
8bd0068d 688 * Decodes headers
689 *
690 * This functions decode strings that is encoded according to
691 * RFC1522 (MIME Part Two: Message Header Extensions for Non-ASCII Text).
692 * Patched by Christian Schmidt <christian@ostenfeld.dk> 23/03/2002
693 *
694 * @param string $string header string that has to be made readable
695 * @param boolean $utfencode change message in order to be readable on user's charset. defaults to true
696 * @param boolean $htmlsave preserve spaces and sanitize html special characters. defaults to true
697 * @param boolean $decide decide if string can be utfencoded. defaults to false
698 * @return string decoded header string
699 */
9f7f68c3 700function decodeHeader ($string, $utfencode=true,$htmlsave=true,$decide=false) {
d6f584fc 701 global $languages, $squirrelmail_language,$default_charset;
79e07c7e 702 if (is_array($string)) {
703 $string = implode("\n", $string);
704 }
da2415c1 705
10dec454 706 if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
8bd0068d 707 function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_decodeheader')) {
33a55f5a 708 $string = call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_decodeheader', $string);
08b7f7cc 709 // Do we need to return at this point?
710 // return $string;
83be314a 711 }
79e07c7e 712 $i = 0;
08b7f7cc 713 $iLastMatch = -2;
db65b6b0 714 $encoded = true;
0a06275a 715
098ea084 716 $aString = explode(' ',$string);
08b7f7cc 717 $ret = '';
098ea084 718 foreach ($aString as $chunk) {
358a78a1 719 if ($encoded && $chunk === '') {
08b7f7cc 720 continue;
358a78a1 721 } elseif ($chunk === '') {
08b7f7cc 722 $ret .= ' ';
723 continue;
724 }
098ea084 725 $encoded = false;
08b7f7cc 726 /* if encoded words are not separated by a linear-space-white we still catch them */
727 $j = $i-1;
7e6ca3e8 728
08b7f7cc 729 while ($match = preg_match('/^(.*)=\?([^?]*)\?(Q|B)\?([^?]*)\?=(.*)$/Ui',$chunk,$res)) {
730 /* if the last chunk isn't an encoded string then put back the space, otherwise don't */
731 if ($iLastMatch !== $j) {
732 if ($htmlsave) {
9f7f68c3 733 $ret .= '&#32;';
08b7f7cc 734 } else {
735 $ret .= ' ';
736 }
737 }
738 $iLastMatch = $i;
739 $j = $i;
cb718de0 740 if ($htmlsave) {
741 $ret .= htmlspecialchars($res[1]);
742 } else {
743 $ret .= $res[1];
744 }
098ea084 745 $encoding = ucfirst($res[3]);
d6f584fc 746
747 /* decide about valid decoding */
748 if ($decide && is_conversion_safe($res[2])) {
8bd0068d 749 $utfencode=true;
750 $can_be_encoded=true;
d6f584fc 751 } else {
8bd0068d 752 $can_be_encoded=false;
d6f584fc 753 }
098ea084 754 switch ($encoding)
755 {
8bd0068d 756 case 'B':
757 $replace = base64_decode($res[4]);
758 if ($utfencode) {
759 if ($can_be_encoded) {
760 /* convert string to different charset,
761 * if functions asks for it (usually in compose)
762 */
bc6c0fba 763 $ret .= charset_convert($res[2],$replace,$default_charset,$htmlsave);
8bd0068d 764 } else {
765 // convert string to html codes in order to display it
766 $ret .= charset_decode($res[2],$replace);
767 }
fab65ca9 768 } else {
8bd0068d 769 if ($htmlsave) {
770 $replace = htmlspecialchars($replace);
771 }
772 $ret.= $replace;
fab65ca9 773 }
8bd0068d 774 break;
775 case 'Q':
776 $replace = str_replace('_', ' ', $res[4]);
777 $replace = preg_replace('/=([0-9a-f]{2})/ie', 'chr(hexdec("\1"))',
778 $replace);
779 if ($utfencode) {
780 if ($can_be_encoded) {
781 /* convert string to different charset,
782 * if functions asks for it (usually in compose)
783 */
bc6c0fba 784 $replace = charset_convert($res[2], $replace,$default_charset,$htmlsave);
8bd0068d 785 } else {
786 // convert string to html codes in order to display it
787 $replace = charset_decode($res[2], $replace);
788 }
789 } else {
790 if ($htmlsave) {
791 $replace = htmlspecialchars($replace);
792 }
098ea084 793 }
8bd0068d 794 $ret .= $replace;
795 break;
796 default:
797 break;
79e07c7e 798 }
098ea084 799 $chunk = $res[5];
800 $encoded = true;
08b7f7cc 801 }
802 if (!$encoded) {
803 if ($htmlsave) {
9f7f68c3 804 $ret .= '&#32;';
08b7f7cc 805 } else {
806 $ret .= ' ';
da2415c1 807 }
08b7f7cc 808 }
dc3d13a7 809
810 if (!$encoded && $htmlsave) {
811 $ret .= htmlspecialchars($chunk);
812 } else {
813 $ret .= $chunk;
814 }
098ea084 815 ++$i;
816 }
fd81e884 817 /* remove the first added space */
818 if ($ret) {
819 if ($htmlsave) {
9f7f68c3 820 $ret = substr($ret,5);
fd81e884 821 } else {
822 $ret = substr($ret,1);
823 }
824 }
da2415c1 825
08b7f7cc 826 return $ret;
451f74a2 827}
828
9f7f68c3 829/**
a24cf710 830 * Encodes header
8bd0068d 831 *
a24cf710 832 * Function uses XTRA_CODE _encodeheader function, if such function exists.
a24cf710 833 *
758a7889 834 * Function uses Q encoding by default and encodes a string according to RFC
835 * 1522 for use in headers if it contains 8-bit characters or anything that
a24cf710 836 * looks like it should be encoded.
8bd0068d 837 *
758a7889 838 * Function switches to B encoding and encodeHeaderBase64() function, if
839 * string is 8bit and multibyte character set supported by mbstring extension
840 * is used. It can cause E_USER_NOTICE errors, if interface is used with
f270a6eb 841 * multibyte character set unsupported by mbstring extension.
842 *
8bd0068d 843 * @param string $string header string, that has to be encoded
844 * @return string quoted-printable encoded string
f270a6eb 845 * @todo make $mb_charsets system wide constant
8bd0068d 846 */
451f74a2 847function encodeHeader ($string) {
6fbd125b 848 global $default_charset, $languages, $squirrelmail_language;
83be314a 849
10dec454 850 if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
8bd0068d 851 function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_encodeheader')) {
33a55f5a 852 return call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_encodeheader', $string);
83be314a 853 }
793cc001 854
a24cf710 855 // Use B encoding for multibyte charsets
f270a6eb 856 $mb_charsets = array('utf-8','big5','gb2313','euc-kr');
857 if (in_array($default_charset,$mb_charsets) &&
858 in_array($default_charset,sq_mb_list_encodings()) &&
859 sq_is8bit($string)) {
860 return encodeHeaderBase64($string,$default_charset);
861 } elseif (in_array($default_charset,$mb_charsets) &&
862 sq_is8bit($string) &&
863 ! in_array($default_charset,sq_mb_list_encodings())) {
864 // Add E_USER_NOTICE error here (can cause 'Cannot add header information' warning in compose.php)
865 // trigger_error('encodeHeader: Multibyte character set unsupported by mbstring extension.',E_USER_NOTICE);
866 }
a24cf710 867
451f74a2 868 // Encode only if the string contains 8-bit characters or =?
3d8371be 869 $j = strlen($string);
098ea084 870 $max_l = 75 - strlen($default_charset) - 7;
871 $aRet = array();
451f74a2 872 $ret = '';
c96c32f4 873 $iEncStart = $enc_init = false;
0d53f0f9 874 $cur_l = $iOffset = 0;
3d8371be 875 for($i = 0; $i < $j; ++$i) {
c96c32f4 876 switch($string{$i})
877 {
8bd0068d 878 case '=':
879 case '<':
880 case '>':
881 case ',':
882 case '?':
883 case '_':
884 if ($iEncStart === false) {
885 $iEncStart = $i;
886 }
887 $cur_l+=3;
888 if ($cur_l > ($max_l-2)) {
889 /* if there is an stringpart that doesn't need encoding, add it */
08b7f7cc 890 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
c96c32f4 891 $aRet[] = "=?$default_charset?Q?$ret?=";
892 $iOffset = $i;
893 $cur_l = 0;
894 $ret = '';
895 $iEncStart = false;
08b7f7cc 896 } else {
8bd0068d 897 $ret .= sprintf("=%02X",ord($string{$i}));
c96c32f4 898 }
8bd0068d 899 break;
900 case '(':
901 case ')':
902 if ($iEncStart !== false) {
08b7f7cc 903 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
8bd0068d 904 $aRet[] = "=?$default_charset?Q?$ret?=";
c96c32f4 905 $iOffset = $i;
8bd0068d 906 $cur_l = 0;
907 $ret = '';
908 $iEncStart = false;
c96c32f4 909 }
8bd0068d 910 break;
911 case ' ':
c96c32f4 912 if ($iEncStart !== false) {
098ea084 913 $cur_l++;
914 if ($cur_l > $max_l) {
08b7f7cc 915 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
c96c32f4 916 $aRet[] = "=?$default_charset?Q?$ret?=";
c96c32f4 917 $iOffset = $i;
918 $cur_l = 0;
098ea084 919 $ret = '';
8bd0068d 920 $iEncStart = false;
08b7f7cc 921 } else {
8bd0068d 922 $ret .= '_';
c96c32f4 923 }
3d8371be 924 }
8bd0068d 925 break;
926 default:
927 $k = ord($string{$i});
928 if ($k > 126) {
929 if ($iEncStart === false) {
930 // do not start encoding in the middle of a string, also take the rest of the word.
931 $sLeadString = substr($string,0,$i);
932 $aLeadString = explode(' ',$sLeadString);
933 $sToBeEncoded = array_pop($aLeadString);
934 $iEncStart = $i - strlen($sToBeEncoded);
935 $ret .= $sToBeEncoded;
936 $cur_l += strlen($sToBeEncoded);
937 }
938 $cur_l += 3;
939 /* first we add the encoded string that reached it's max size */
940 if ($cur_l > ($max_l-2)) {
941 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
942 $aRet[] = "=?$default_charset?Q?$ret?= "; /* the next part is also encoded => separate by space */
943 $cur_l = 3;
944 $ret = '';
945 $iOffset = $i;
946 $iEncStart = $i;
947 }
948 $enc_init = true;
949 $ret .= sprintf("=%02X", $k);
950 } else {
951 if ($iEncStart !== false) {
952 $cur_l++;
953 if ($cur_l > $max_l) {
954 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
955 $aRet[] = "=?$default_charset?Q?$ret?=";
956 $iEncStart = false;
957 $iOffset = $i;
958 $cur_l = 0;
959 $ret = '';
960 } else {
961 $ret .= $string{$i};
962 }
963 }
964 }
965 break;
f7b3ba37 966 }
451f74a2 967 }
793cc001 968
c96c32f4 969 if ($enc_init) {
970 if ($iEncStart !== false) {
971 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
972 $aRet[] = "=?$default_charset?Q?$ret?=";
973 } else {
974 $aRet[] = substr($string,$iOffset);
975 }
976 $string = implode('',$aRet);
451f74a2 977 }
3d8371be 978 return $string;
451f74a2 979}
b74ba498 980
f270a6eb 981/**
982 * Encodes string according to rfc2047 B encoding header formating rules
983 *
758a7889 984 * It is recommended way to encode headers with character sets that store
f270a6eb 985 * symbols in more than one byte.
986 *
987 * Function requires mbstring support. If required mbstring functions are missing,
988 * function returns false and sets E_USER_WARNING level error message.
989 *
758a7889 990 * Minimal requirements - php 4.0.6 with mbstring extension. Please note,
991 * that mbstring functions will generate E_WARNING errors, if unsupported
f270a6eb 992 * character set is used. mb_encode_mimeheader function provided by php
993 * mbstring extension is not used in order to get better control of header
994 * encoding.
995 *
758a7889 996 * Used php code functions - function_exists(), trigger_error(), strlen()
997 * (is used with charset names and base64 strings). Used php mbstring
f270a6eb 998 * functions - mb_strlen and mb_substr.
999 *
758a7889 1000 * Related documents: rfc 2045 (BASE64 encoding), rfc 2047 (mime header
f270a6eb 1001 * encoding), rfc 2822 (header folding)
1002 *
1003 * @param string $string header string that must be encoded
758a7889 1004 * @param string $charset character set. Must be supported by mbstring extension.
f270a6eb 1005 * Use sq_mb_list_encodings() to detect supported charsets.
1006 * @return string string encoded according to rfc2047 B encoding formating rules
1007 * @since 1.5.1
1008 * @todo First header line can be wrapped to $iMaxLength - $HeaderFieldLength - 1
1009 * @todo Do we want to control max length of header?
1010 * @todo Do we want to control EOL (end-of-line) marker?
1011 * @todo Do we want to translate error message?
1012 */
1013function encodeHeaderBase64($string,$charset) {
1014 /**
1015 * Check mbstring function requirements.
1016 */
1017 if (! function_exists('mb_strlen') ||
1018 ! function_exists('mb_substr')) {
1019 // set E_USER_WARNING
1020 trigger_error('encodeHeaderBase64: Required mbstring functions are missing.',E_USER_WARNING);
1021 // return false
1022 return false;
1023 }
1024
1025 // initial return array
1026 $aRet = array();
1027
1028 /**
1029 * header length = 75 symbols max (same as in encodeHeader)
1030 * remove $charset length
1031 * remove =? ? ?= (5 chars)
1032 * remove 2 more chars (\r\n ?)
1033 */
1034 $iMaxLength = 75 - strlen($charset) - 7;
1035
1036 // set first character position
1037 $iStartCharNum = 0;
1038
1039 // loop through all characters. count characters and not bytes.
1040 for ($iCharNum=1; $iCharNum<=mb_strlen($string,$charset); $iCharNum++) {
1041 // encode string from starting character to current character.
1042 $encoded_string = base64_encode(mb_substr($string,$iStartCharNum,$iCharNum-$iStartCharNum,$charset));
1043
1044 // Check encoded string length
1045 if(strlen($encoded_string)>$iMaxLength) {
1046 // if string exceeds max length, reduce number of encoded characters and add encoded string part to array
1047 $aRet[] = base64_encode(mb_substr($string,$iStartCharNum,$iCharNum-$iStartCharNum-1,$charset));
1048
1049 // set new starting character
1050 $iStartCharNum = $iCharNum-1;
1051
1052 // encode last char (in case it is last character in string)
1053 $encoded_string = base64_encode(mb_substr($string,$iStartCharNum,$iCharNum-$iStartCharNum,$charset));
1054 } // if string is shorter than max length - add next character
1055 }
1056
1057 // add last encoded string to array
1058 $aRet[] = $encoded_string;
1059
1060 // set initial return string
1061 $sRet = '';
1062
1063 // loop through encoded strings
1064 foreach($aRet as $string) {
1065 // TODO: Do we want to control EOL (end-of-line) marker
1066 if ($sRet!='') $sRet.= " ";
1067
1068 // add header tags and encoded string to return string
1069 $sRet.= '=?'.$charset.'?B?'.$string.'?=';
1070 }
1071
1072 return $sRet;
1073}
1074
691a2d25 1075/* This function trys to locate the entity_id of a specific mime element */
3d8371be 1076function find_ent_id($id, $message) {
a171b359 1077 for ($i = 0, $ret = ''; $ret == '' && $i < count($message->entities); $i++) {
1078 if ($message->entities[$i]->header->type0 == 'multipart') {
3d8371be 1079 $ret = find_ent_id($id, $message->entities[$i]);
451f74a2 1080 } else {
3d8371be 1081 if (strcasecmp($message->entities[$i]->header->id, $id) == 0) {
d8cffbab 1082// if (sq_check_save_extension($message->entities[$i])) {
8bd0068d 1083 return $message->entities[$i]->entity_id;
da2415c1 1084// }
c8f5f606 1085 } elseif (!empty($message->entities[$i]->header->parameters['name'])) {
1086 /**
1087 * This is part of a fix for Outlook Express 6.x generating
1088 * cid URLs without creating content-id headers
1089 * @@JA - 20050207
1090 */
1091 if (strcasecmp($message->entities[$i]->header->parameters['name'], $id) == 0) {
1092 return $message->entities[$i]->entity_id;
1093 }
3d8371be 1094 }
a3daaaf3 1095 }
451f74a2 1096 }
3d8371be 1097 return $ret;
451f74a2 1098}
a3daaaf3 1099
e5e9381a 1100function sq_check_save_extension($message) {
1101 $filename = $message->getFilename();
1102 $ext = substr($filename, strrpos($filename,'.')+1);
1103 $save_extensions = array('jpg','jpeg','gif','png','bmp');
3d8371be 1104 return in_array($ext, $save_extensions);
e5e9381a 1105}
1106
1107
691a2d25 1108/**
8bd0068d 1109 ** HTMLFILTER ROUTINES
1110 */
451f74a2 1111
2dd879b8 1112/**
0493ed11 1113 * This function checks attribute values for entity-encoded values
1114 * and returns them translated into 8-bit strings so we can run
1115 * checks on them.
8bd0068d 1116 *
0493ed11 1117 * @param $attvalue A string to run entity check against.
1118 * @return Nothing, modifies a reference value.
8bd0068d 1119 */
0493ed11 1120function sq_defang(&$attvalue){
1121 $me = 'sq_defang';
2dd879b8 1122 /**
0493ed11 1123 * Skip this if there aren't ampersands or backslashes.
8bd0068d 1124 */
0493ed11 1125 if (strpos($attvalue, '&') === false
1126 && strpos($attvalue, '\\') === false){
1127 return;
2dd879b8 1128 }
0493ed11 1129 $m = false;
2b665f28 1130 // before deent, translate the dangerous unicode characters and ... to safe values
1131 // otherwise the regular expressions do not match.
1132
1133
1134
0493ed11 1135 do {
1136 $m = false;
1137 $m = $m || sq_deent($attvalue, '/\&#0*(\d+);*/s');
1138 $m = $m || sq_deent($attvalue, '/\&#x0*((\d|[a-f])+);*/si', true);
1139 $m = $m || sq_deent($attvalue, '/\\\\(\d+)/s', true);
1140 } while ($m == true);
1141 $attvalue = stripslashes($attvalue);
2dd879b8 1142}
1143
1144/**
8bd0068d 1145 * Kill any tabs, newlines, or carriage returns. Our friends the
1146 * makers of the browser with 95% market value decided that it'd
1147 * be funny to make "java[tab]script" be just as good as "javascript".
1148 *
1149 * @param attvalue The attribute value before extraneous spaces removed.
0493ed11 1150 * @return attvalue Nothing, modifies a reference value.
8bd0068d 1151 */
0493ed11 1152function sq_unspace(&$attvalue){
1153 $me = 'sq_unspace';
1154 if (strcspn($attvalue, "\t\r\n\0 ") != strlen($attvalue)){
1155 $attvalue = str_replace(Array("\t", "\r", "\n", "\0", " "),
1156 Array('', '', '', '', ''), $attvalue);
2dd879b8 1157 }
2dd879b8 1158}
1159
2b665f28 1160/**
88de4926 1161 * Translate all dangerous Unicode or Shift_JIS characters which are accepted by
2b665f28 1162 * IE as regular characters.
1163 *
1164 * @param attvalue The attribute value before dangerous characters are translated.
1165 * @return attvalue Nothing, modifies a reference value.
1166 * @author Marc Groot Koerkamp.
1167 */
1168function sq_fixIE_idiocy(&$attvalue) {
1169 // remove NUL
1170 $attvalue = str_replace("\0", "", $attvalue);
1171 // remove comments
1172 $attvalue = preg_replace("/(\/\*.*?\*\/)/","",$attvalue);
1173
88de4926 1174 // IE has the evil habit of accepting every possible value for the attribute expression.
1175 // The table below contains characters which are parsed by IE if they are used in the "expression"
2b665f28 1176 // attribute value.
1177 $aDangerousCharsReplacementTable = array(
1178 array('&#x029F;', '&#0671;' ,/* L UNICODE IPA Extension */
1179 '&#x0280;', '&#0640;' ,/* R UNICODE IPA Extension */
1180 '&#x0274;', '&#0628;' ,/* N UNICODE IPA Extension */
1181 '&#xFF25;', '&#65317' ,/* Unicode FULLWIDTH LATIN CAPITAL LETTER E */
1182 '&#xFF45;', '&#65349' ,/* Unicode FULLWIDTH LATIN SMALL LETTER E */
1183 '&#xFF38;', '&#65336;',/* Unicode FULLWIDTH LATIN CAPITAL LETTER X */
1184 '&#xFF58;', '&#65368;',/* Unicode FULLWIDTH LATIN SMALL LETTER X */
1185 '&#xFF30;', '&#65328;',/* Unicode FULLWIDTH LATIN CAPITAL LETTER P */
1186 '&#xFF50;', '&#65360;',/* Unicode FULLWIDTH LATIN SMALL LETTER P */
1187 '&#xFF32;', '&#65330;',/* Unicode FULLWIDTH LATIN CAPITAL LETTER R */
1188 '&#xFF52;', '&#65362;',/* Unicode FULLWIDTH LATIN SMALL LETTER R */
1189 '&#xFF33;', '&#65331;',/* Unicode FULLWIDTH LATIN CAPITAL LETTER S */
1190 '&#xFF53;', '&#65363;',/* Unicode FULLWIDTH LATIN SMALL LETTER S */
1191 '&#xFF29;', '&#65321;',/* Unicode FULLWIDTH LATIN CAPITAL LETTER I */
1192 '&#xFF49;', '&#65353;',/* Unicode FULLWIDTH LATIN SMALL LETTER I */
1193 '&#xFF2F;', '&#65327;',/* Unicode FULLWIDTH LATIN CAPITAL LETTER O */
1194 '&#xFF4F;', '&#65359;',/* Unicode FULLWIDTH LATIN SMALL LETTER O */
1195 '&#xFF2E;', '&#65326;',/* Unicode FULLWIDTH LATIN CAPITAL LETTER N */
1196 '&#xFF4E;', '&#65358;',/* Unicode FULLWIDTH LATIN SMALL LETTER N */
1197 '&#xFF2C;', '&#65324;',/* Unicode FULLWIDTH LATIN CAPITAL LETTER L */
1198 '&#xFF4C;', '&#65356;',/* Unicode FULLWIDTH LATIN SMALL LETTER L */
1199 '&#xFF35;', '&#65333;',/* Unicode FULLWIDTH LATIN CAPITAL LETTER U */
1200 '&#xFF55;', '&#65365;',/* Unicode FULLWIDTH LATIN SMALL LETTER U */
1201 '&#x207F;', '&#8319;' ,/* Unicode SUPERSCRIPT LATIN SMALL LETTER N */
88de4926 1202 '&#x8264;', /* Shift JIS FULLWIDTH LATIN CAPITAL LETTER E */ // in unicode this is some Chinese char range
2b665f28 1203 '&#x8285;', /* Shift JIS FULLWIDTH LATIN SMALL LETTER E */
1204 '&#x8277;', /* Shift JIS FULLWIDTH LATIN CAPITAL LETTER X */
1205 '&#x8298;', /* Shift JIS FULLWIDTH LATIN SMALL LETTER X */
1206 '&#x826F;', /* Shift JIS FULLWIDTH LATIN CAPITAL LETTER P */
1207 '&#x8290;', /* Shift JIS FULLWIDTH LATIN SMALL LETTER P */
1208 '&#x8271;', /* Shift JIS FULLWIDTH LATIN CAPITAL LETTER R */
1209 '&#x8292;', /* Shift JIS FULLWIDTH LATIN SMALL LETTER R */
1210 '&#x8272;', /* Shift JIS FULLWIDTH LATIN CAPITAL LETTER S */
1211 '&#x8293;', /* Shift JIS FULLWIDTH LATIN SMALL LETTER S */
1212 '&#x8268;', /* Shift JIS FULLWIDTH LATIN CAPITAL LETTER I */
1213 '&#x8289;', /* Shift JIS FULLWIDTH LATIN SMALL LETTER I */
1214 '&#x826E;', /* Shift JIS FULLWIDTH LATIN CAPITAL LETTER O */
1215 '&#x828F;', /* Shift JIS FULLWIDTH LATIN SMALL LETTER O */
1216 '&#x826D;', /* Shift JIS FULLWIDTH LATIN CAPITAL LETTER N */
1217 '&#x828E;'), /* Shift JIS FULLWIDTH LATIN SMALL LETTER N */
1218 array('l', 'l', 'r','r','n','n',
1219 'E','E','e','e','X','X','x','x','P','P','p','p','S','S','s','s','I','I',
1220 'i','i','O','O','o','o','N','N','n','n','L','L','l','l','U','U','u','u','n',
1221 'E','e','X','x','P','p','S','s','I','i','O','o','N','n'));
1222 $attvalue = str_replace($aDangerousCharsReplacementTable[0],$aDangerousCharsReplacementTable[1],$attvalue);
1223
88de4926 1224 // Escapes are useful for special characters like "{}[]()'&. In other cases they are
1225 // used for XSS.
2b665f28 1226 $attvalue = preg_replace("/(\\\\)([a-zA-Z]{1})/",'$2',$attvalue);
1227}
1228
691a2d25 1229/**
8bd0068d 1230 * This function returns the final tag out of the tag name, an array
1231 * of attributes, and the type of the tag. This function is called by
1232 * sq_sanitize internally.
1233 *
1234 * @param $tagname the name of the tag.
1235 * @param $attary the array of attributes and their values
1236 * @param $tagtype The type of the tag (see in comments).
1237 * @return a string with the final tag representation.
1238 */
691a2d25 1239function sq_tagprint($tagname, $attary, $tagtype){
b583c3e8 1240 $me = 'sq_tagprint';
3d8371be 1241
691a2d25 1242 if ($tagtype == 2){
1243 $fulltag = '</' . $tagname . '>';
1244 } else {
1245 $fulltag = '<' . $tagname;
1246 if (is_array($attary) && sizeof($attary)){
1247 $atts = Array();
1248 while (list($attname, $attvalue) = each($attary)){
1249 array_push($atts, "$attname=$attvalue");
1250 }
1251 $fulltag .= ' ' . join(" ", $atts);
1252 }
1253 if ($tagtype == 3){
b583c3e8 1254 $fulltag .= ' /';
691a2d25 1255 }
b583c3e8 1256 $fulltag .= '>';
451f74a2 1257 }
691a2d25 1258 return $fulltag;
451f74a2 1259}
a3daaaf3 1260
691a2d25 1261/**
8bd0068d 1262 * A small helper function to use with array_walk. Modifies a by-ref
1263 * value and makes it lowercase.
1264 *
1265 * @param $val a value passed by-ref.
1266 * @return void since it modifies a by-ref value.
1267 */
691a2d25 1268function sq_casenormalize(&$val){
1269 $val = strtolower($val);
1270}
451f74a2 1271
691a2d25 1272/**
8bd0068d 1273 * This function skips any whitespace from the current position within
1274 * a string and to the next non-whitespace value.
1275 *
1276 * @param $body the string
1277 * @param $offset the offset within the string where we should start
1278 * looking for the next non-whitespace character.
1279 * @return the location within the $body where the next
1280 * non-whitespace char is located.
1281 */
691a2d25 1282function sq_skipspace($body, $offset){
b583c3e8 1283 $me = 'sq_skipspace';
3d8371be 1284 preg_match('/^(\s*)/s', substr($body, $offset), $matches);
691a2d25 1285 if (sizeof($matches{1})){
1286 $count = strlen($matches{1});
1287 $offset += $count;
451f74a2 1288 }
691a2d25 1289 return $offset;
451f74a2 1290}
a3daaaf3 1291
691a2d25 1292/**
8bd0068d 1293 * This function looks for the next character within a string. It's
1294 * really just a glorified "strpos", except it catches if failures
1295 * nicely.
1296 *
1297 * @param $body The string to look for needle in.
1298 * @param $offset Start looking from this position.
1299 * @param $needle The character/string to look for.
1300 * @return location of the next occurance of the needle, or
1301 * strlen($body) if needle wasn't found.
1302 */
691a2d25 1303function sq_findnxstr($body, $offset, $needle){
3d8371be 1304 $me = 'sq_findnxstr';
691a2d25 1305 $pos = strpos($body, $needle, $offset);
1306 if ($pos === FALSE){
1307 $pos = strlen($body);
451f74a2 1308 }
691a2d25 1309 return $pos;
451f74a2 1310}
a3daaaf3 1311
691a2d25 1312/**
8bd0068d 1313 * This function takes a PCRE-style regexp and tries to match it
1314 * within the string.
1315 *
1316 * @param $body The string to look for needle in.
1317 * @param $offset Start looking from here.
1318 * @param $reg A PCRE-style regex to match.
1319 * @return Returns a false if no matches found, or an array
1320 * with the following members:
1321 * - integer with the location of the match within $body
1322 * - string with whatever content between offset and the match
1323 * - string with whatever it is we matched
1324 */
691a2d25 1325function sq_findnxreg($body, $offset, $reg){
b583c3e8 1326 $me = 'sq_findnxreg';
691a2d25 1327 $matches = Array();
1328 $retarr = Array();
7d06541f 1329 preg_match("%^(.*?)($reg)%si", substr($body, $offset), $matches);
1330 if (!isset($matches{0}) || !$matches{0}){
691a2d25 1331 $retarr = false;
1332 } else {
1333 $retarr{0} = $offset + strlen($matches{1});
1334 $retarr{1} = $matches{1};
1335 $retarr{2} = $matches{2};
1336 }
1337 return $retarr;
1338}
a3daaaf3 1339
691a2d25 1340/**
8bd0068d 1341 * This function looks for the next tag.
1342 *
1343 * @param $body String where to look for the next tag.
1344 * @param $offset Start looking from here.
1345 * @return false if no more tags exist in the body, or
1346 * an array with the following members:
1347 * - string with the name of the tag
1348 * - array with attributes and their values
1349 * - integer with tag type (1, 2, or 3)
1350 * - integer where the tag starts (starting "<")
1351 * - integer where the tag ends (ending ">")
1352 * first three members will be false, if the tag is invalid.
1353 */
691a2d25 1354function sq_getnxtag($body, $offset){
b583c3e8 1355 $me = 'sq_getnxtag';
691a2d25 1356 if ($offset > strlen($body)){
1357 return false;
1358 }
1359 $lt = sq_findnxstr($body, $offset, "<");
1360 if ($lt == strlen($body)){
1361 return false;
1362 }
1363 /**
8bd0068d 1364 * We are here:
1365 * blah blah <tag attribute="value">
1366 * \---------^
1367 */
691a2d25 1368 $pos = sq_skipspace($body, $lt+1);
1369 if ($pos >= strlen($body)){
1370 return Array(false, false, false, $lt, strlen($body));
1371 }
1372 /**
8bd0068d 1373 * There are 3 kinds of tags:
1374 * 1. Opening tag, e.g.:
1375 * <a href="blah">
1376 * 2. Closing tag, e.g.:
1377 * </a>
1378 * 3. XHTML-style content-less tag, e.g.:
1379 * <img src="blah" />
1380 */
691a2d25 1381 $tagtype = false;
1382 switch (substr($body, $pos, 1)){
3d8371be 1383 case '/':
1384 $tagtype = 2;
1385 $pos++;
1386 break;
1387 case '!':
1388 /**
8bd0068d 1389 * A comment or an SGML declaration.
1390 */
3d8371be 1391 if (substr($body, $pos+1, 2) == "--"){
1392 $gt = strpos($body, "-->", $pos);
1393 if ($gt === false){
1394 $gt = strlen($body);
1395 } else {
1396 $gt += 2;
1397 }
1398 return Array(false, false, false, $lt, $gt);
bb8d0799 1399 } else {
3d8371be 1400 $gt = sq_findnxstr($body, $pos, ">");
1401 return Array(false, false, false, $lt, $gt);
1402 }
1403 break;
1404 default:
1405 /**
8bd0068d 1406 * Assume tagtype 1 for now. If it's type 3, we'll switch values
1407 * later.
1408 */
3d8371be 1409 $tagtype = 1;
1410 break;
691a2d25 1411 }
a3daaaf3 1412
0493ed11 1413 $tag_start = $pos;
691a2d25 1414 $tagname = '';
1415 /**
8bd0068d 1416 * Look for next [\W-_], which will indicate the end of the tag name.
1417 */
691a2d25 1418 $regary = sq_findnxreg($body, $pos, "[^\w\-_]");
1419 if ($regary == false){
1420 return Array(false, false, false, $lt, strlen($body));
1421 }
1422 list($pos, $tagname, $match) = $regary;
1423 $tagname = strtolower($tagname);
1424
1425 /**
8bd0068d 1426 * $match can be either of these:
1427 * '>' indicating the end of the tag entirely.
1428 * '\s' indicating the end of the tag name.
1429 * '/' indicating that this is type-3 xhtml tag.
1430 *
1431 * Whatever else we find there indicates an invalid tag.
1432 */
691a2d25 1433 switch ($match){
3d8371be 1434 case '/':
691a2d25 1435 /**
8bd0068d 1436 * This is an xhtml-style tag with a closing / at the
1437 * end, like so: <img src="blah" />. Check if it's followed
1438 * by the closing bracket. If not, then this tag is invalid
1439 */
3d8371be 1440 if (substr($body, $pos, 2) == "/>"){
1441 $pos++;
1442 $tagtype = 3;
1443 } else {
1444 $gt = sq_findnxstr($body, $pos, ">");
1445 $retary = Array(false, false, false, $lt, $gt);
1446 return $retary;
1447 }
1448 case '>':
1449 return Array($tagname, false, $tagtype, $lt, $pos);
1450 break;
1451 default:
1452 /**
8bd0068d 1453 * Check if it's whitespace
1454 */
3d8371be 1455 if (!preg_match('/\s/', $match)){
1456 /**
8bd0068d 1457 * This is an invalid tag! Look for the next closing ">".
1458 */
7d06541f 1459 $gt = sq_findnxstr($body, $lt, ">");
3d8371be 1460 return Array(false, false, false, $lt, $gt);
1461 }
1462 break;
691a2d25 1463 }
3d8371be 1464
691a2d25 1465 /**
8bd0068d 1466 * At this point we're here:
1467 * <tagname attribute='blah'>
1468 * \-------^
1469 *
1470 * At this point we loop in order to find all attributes.
1471 */
691a2d25 1472 $attname = '';
0493ed11 1473 $atttype = false;
691a2d25 1474 $attary = Array();
1475
1476 while ($pos <= strlen($body)){
1477 $pos = sq_skipspace($body, $pos);
1478 if ($pos == strlen($body)){
1479 /**
8bd0068d 1480 * Non-closed tag.
1481 */
691a2d25 1482 return Array(false, false, false, $lt, $pos);
1483 }
1484 /**
8bd0068d 1485 * See if we arrived at a ">" or "/>", which means that we reached
1486 * the end of the tag.
1487 */
691a2d25 1488 $matches = Array();
164800ad 1489 if (preg_match("%^(\s*)(>|/>)%s", substr($body, $pos), $matches)) {
c828931c 1490 /**
8bd0068d 1491 * Yep. So we did.
1492 */
c828931c 1493 $pos += strlen($matches{1});
1494 if ($matches{2} == "/>"){
1495 $tagtype = 3;
1496 $pos++;
1497 }
1498 return Array($tagname, $attary, $tagtype, $lt, $pos);
1499 }
a3daaaf3 1500
cca46357 1501 /**
8bd0068d 1502 * There are several types of attributes, with optional
1503 * [:space:] between members.
1504 * Type 1:
1505 * attrname[:space:]=[:space:]'CDATA'
1506 * Type 2:
1507 * attrname[:space:]=[:space:]"CDATA"
1508 * Type 3:
1509 * attr[:space:]=[:space:]CDATA
1510 * Type 4:
1511 * attrname
1512 *
1513 * We leave types 1 and 2 the same, type 3 we check for
1514 * '"' and convert to "&quot" if needed, then wrap in
1515 * double quotes. Type 4 we convert into:
1516 * attrname="yes".
1517 */
3f7c623f 1518 $regary = sq_findnxreg($body, $pos, "[^:\w\-_]");
691a2d25 1519 if ($regary == false){
1520 /**
8bd0068d 1521 * Looks like body ended before the end of tag.
1522 */
691a2d25 1523 return Array(false, false, false, $lt, strlen($body));
cca46357 1524 }
691a2d25 1525 list($pos, $attname, $match) = $regary;
1526 $attname = strtolower($attname);
1527 /**
8bd0068d 1528 * We arrived at the end of attribute name. Several things possible
1529 * here:
1530 * '>' means the end of the tag and this is attribute type 4
1531 * '/' if followed by '>' means the same thing as above
1532 * '\s' means a lot of things -- look what it's followed by.
1533 * anything else means the attribute is invalid.
1534 */
691a2d25 1535 switch($match){
3d8371be 1536 case '/':
691a2d25 1537 /**
8bd0068d 1538 * This is an xhtml-style tag with a closing / at the
1539 * end, like so: <img src="blah" />. Check if it's followed
1540 * by the closing bracket. If not, then this tag is invalid
1541 */
3d8371be 1542 if (substr($body, $pos, 2) == "/>"){
691a2d25 1543 $pos++;
3d8371be 1544 $tagtype = 3;
691a2d25 1545 } else {
3d8371be 1546 $gt = sq_findnxstr($body, $pos, ">");
1547 $retary = Array(false, false, false, $lt, $gt);
1548 return $retary;
1549 }
1550 case '>':
1551 $attary{$attname} = '"yes"';
1552 return Array($tagname, $attary, $tagtype, $lt, $pos);
1553 break;
1554 default:
1555 /**
8bd0068d 1556 * Skip whitespace and see what we arrive at.
1557 */
3d8371be 1558 $pos = sq_skipspace($body, $pos);
1559 $char = substr($body, $pos, 1);
1560 /**
8bd0068d 1561 * Two things are valid here:
1562 * '=' means this is attribute type 1 2 or 3.
1563 * \w means this was attribute type 4.
1564 * anything else we ignore and re-loop. End of tag and
1565 * invalid stuff will be caught by our checks at the beginning
1566 * of the loop.
1567 */
3d8371be 1568 if ($char == "="){
1569 $pos++;
1570 $pos = sq_skipspace($body, $pos);
691a2d25 1571 /**
8bd0068d 1572 * Here are 3 possibilities:
1573 * "'" attribute type 1
1574 * '"' attribute type 2
1575 * everything else is the content of tag type 3
1576 */
3d8371be 1577 $quot = substr($body, $pos, 1);
1578 if ($quot == "'"){
1579 $regary = sq_findnxreg($body, $pos+1, "\'");
1580 if ($regary == false){
1581 return Array(false, false, false, $lt, strlen($body));
1582 }
1583 list($pos, $attval, $match) = $regary;
1584 $pos++;
1585 $attary{$attname} = "'" . $attval . "'";
1586 } else if ($quot == '"'){
1587 $regary = sq_findnxreg($body, $pos+1, '\"');
1588 if ($regary == false){
1589 return Array(false, false, false, $lt, strlen($body));
1590 }
1591 list($pos, $attval, $match) = $regary;
1592 $pos++;
1593 $attary{$attname} = '"' . $attval . '"';
1594 } else {
1595 /**
8bd0068d 1596 * These are hateful. Look for \s, or >.
1597 */
3d8371be 1598 $regary = sq_findnxreg($body, $pos, "[\s>]");
1599 if ($regary == false){
1600 return Array(false, false, false, $lt, strlen($body));
1601 }
1602 list($pos, $attval, $match) = $regary;
1603 /**
8bd0068d 1604 * If it's ">" it will be caught at the top.
1605 */
3d8371be 1606 $attval = preg_replace("/\"/s", "&quot;", $attval);
1607 $attary{$attname} = '"' . $attval . '"';
7e235a1a 1608 }
3d8371be 1609 } else if (preg_match("|[\w/>]|", $char)) {
691a2d25 1610 /**
8bd0068d 1611 * That was attribute type 4.
1612 */
3d8371be 1613 $attary{$attname} = '"yes"';
1614 } else {
1615 /**
8bd0068d 1616 * An illegal character. Find next '>' and return.
1617 */
3d8371be 1618 $gt = sq_findnxstr($body, $pos, ">");
1619 return Array(false, false, false, $lt, $gt);
451f74a2 1620 }
3d8371be 1621 break;
691a2d25 1622 }
1623 }
1624 /**
8bd0068d 1625 * The fact that we got here indicates that the tag end was never
1626 * found. Return invalid tag indication so it gets stripped.
1627 */
691a2d25 1628 return Array(false, false, false, $lt, strlen($body));
1629}
1630
1631/**
0493ed11 1632 * Translates entities into literal values so they can be checked.
8bd0068d 1633 *
0493ed11 1634 * @param $attvalue the by-ref value to check.
1635 * @param $regex the regular expression to check against.
1636 * @param $hex whether the entites are hexadecimal.
1637 * @return True or False depending on whether there were matches.
8bd0068d 1638 */
0493ed11 1639function sq_deent(&$attvalue, $regex, $hex=false){
b583c3e8 1640 $me = 'sq_deent';
0493ed11 1641 $ret_match = false;
2b665f28 1642 // remove comments
1643 //$attvalue = preg_replace("/(\/\*.*\*\/)/","",$attvalue);
0493ed11 1644 preg_match_all($regex, $attvalue, $matches);
1645 if (is_array($matches) && sizeof($matches[0]) > 0){
1646 $repl = Array();
1647 for ($i = 0; $i < sizeof($matches[0]); $i++){
1648 $numval = $matches[1][$i];
1649 if ($hex){
1650 $numval = hexdec($numval);
a3daaaf3 1651 }
0493ed11 1652 $repl{$matches[0][$i]} = chr($numval);
691a2d25 1653 }
0493ed11 1654 $attvalue = strtr($attvalue, $repl);
1655 return true;
1656 } else {
1657 return false;
691a2d25 1658 }
691a2d25 1659}
1660
1661/**
8bd0068d 1662 * This function runs various checks against the attributes.
1663 *
1664 * @param $tagname String with the name of the tag.
1665 * @param $attary Array with all tag attributes.
1666 * @param $rm_attnames See description for sq_sanitize
1667 * @param $bad_attvals See description for sq_sanitize
1668 * @param $add_attr_to_tag See description for sq_sanitize
1669 * @param $message message object
1670 * @param $id message id
1671 * @return Array with modified attributes.
1672 */
da2415c1 1673function sq_fixatts($tagname,
1674 $attary,
691a2d25 1675 $rm_attnames,
1676 $bad_attvals,
1677 $add_attr_to_tag,
1678 $message,
b3af12ef 1679 $id,
3d8371be 1680 $mailbox
691a2d25 1681 ){
b583c3e8 1682 $me = 'sq_fixatts';
691a2d25 1683 while (list($attname, $attvalue) = each($attary)){
1684 /**
8bd0068d 1685 * See if this attribute should be removed.
1686 */
691a2d25 1687 foreach ($rm_attnames as $matchtag=>$matchattrs){
1688 if (preg_match($matchtag, $tagname)){
1689 foreach ($matchattrs as $matchattr){
1690 if (preg_match($matchattr, $attname)){
1691 unset($attary{$attname});
1692 continue;
1693 }
451f74a2 1694 }
451f74a2 1695 }
691a2d25 1696 }
2b665f28 1697 /**
1698 * Workaround for IE quirks
1699 */
1700 sq_fixIE_idiocy($attvalue);
1701
691a2d25 1702 /**
8bd0068d 1703 * Remove any backslashes, entities, and extraneous whitespace.
1704 */
2b665f28 1705
1706 $oldattvalue = $attvalue;
0493ed11 1707 sq_defang($attvalue);
2b665f28 1708 if ($attname == 'style' && $attvalue !== $oldattvalue) {
1709 // entities are used in the attribute value. In 99% of the cases it's there as XSS
1710 // i.e.<div style="{ left:exp&#x0280;essio&#x0274;( alert('XSS') ) }">
1711 $attvalue = "idiocy";
1712 $attary{$attname} = $attvalue;
1713 }
0493ed11 1714 sq_unspace($attvalue);
af861a34 1715
691a2d25 1716 /**
8bd0068d 1717 * Now let's run checks on the attvalues.
1718 * I don't expect anyone to comprehend this. If you do,
1719 * get in touch with me so I can drive to where you live and
1720 * shake your hand personally. :)
1721 */
691a2d25 1722 foreach ($bad_attvals as $matchtag=>$matchattrs){
1723 if (preg_match($matchtag, $tagname)){
1724 foreach ($matchattrs as $matchattr=>$valary){
1725 if (preg_match($matchattr, $attname)){
1726 /**
8bd0068d 1727 * There are two arrays in valary.
1728 * First is matches.
1729 * Second one is replacements
1730 */
691a2d25 1731 list($valmatch, $valrepl) = $valary;
da2415c1 1732 $newvalue =
691a2d25 1733 preg_replace($valmatch, $valrepl, $attvalue);
1734 if ($newvalue != $attvalue){
1735 $attary{$attname} = $newvalue;
1736 }
1737 }
1738 }
451f74a2 1739 }
a3daaaf3 1740 }
834a1027 1741
834a1027 1742 /**
1743 * Replace empty src tags with the blank image. src is only used
1744 * for frames, images, and image inputs. Doing a replace should
1745 * not affect them working as should be, however it will stop
1746 * IE from being kicked off when src for img tags are not set
1747 */
5a6fde9e 1748 if (($attname == 'src') && ($attvalue == '""')) {
834a1027 1749 $attary{$attname} = '"' . SM_PATH . 'images/blank.png"';
1750 }
1751
691a2d25 1752 /**
8bd0068d 1753 * Turn cid: urls into http-friendly ones.
1754 */
691a2d25 1755 if (preg_match("/^[\'\"]\s*cid:/si", $attvalue)){
b3af12ef 1756 $attary{$attname} = sq_cid2http($message, $id, $attvalue, $mailbox);
691a2d25 1757 }
ff940ebc 1758
1759 /**
1760 * "Hack" fix for Outlook using propriatary outbind:// protocol in img tags.
1761 * One day MS might actually make it match something useful, for now, falling
1762 * back to using cid2http, so we can grab the blank.png.
1763 */
1764 if (preg_match("/^[\'\"]\s*outbind:\/\//si", $attvalue)) {
1765 $attary{$attname} = sq_cid2http($message, $id, $attvalue, $mailbox);
1766 }
1767
a3daaaf3 1768 }
691a2d25 1769 /**
8bd0068d 1770 * See if we need to append any attributes to this tag.
1771 */
691a2d25 1772 foreach ($add_attr_to_tag as $matchtag=>$addattary){
1773 if (preg_match($matchtag, $tagname)){
1774 $attary = array_merge($attary, $addattary);
1775 }
1776 }
1777 return $attary;
451f74a2 1778}
a3daaaf3 1779
691a2d25 1780/**
8bd0068d 1781 * This function edits the style definition to make them friendly and
1782 * usable in SquirrelMail.
1783 *
1784 * @param $message the message object
1785 * @param $id the message id
1786 * @param $content a string with whatever is between <style> and </style>
1787 * @param $mailbox the message mailbox
1788 * @return a string with edited content.
1789 */
e60a299a 1790function sq_fixstyle($body, $pos, $message, $id, $mailbox){
691a2d25 1791 global $view_unsafe_images;
b583c3e8 1792 $me = 'sq_fixstyle';
7e2ff844 1793 // workaround for </style> in between comments
1794 $iCurrentPos = $pos;
1795 $content = '';
1796 $sToken = '';
1797 $bSucces = false;
1798 $bEndTag = false;
1799 for ($i=$pos,$iCount=strlen($body);$i<$iCount;++$i) {
1800 $char = $body{$i};
1801 switch ($char) {
1802 case '<':
1803 $sToken .= $char;
1804 break;
1805 case '/':
1806 if ($sToken == '<') {
1807 $sToken .= $char;
1808 $bEndTag = true;
1809 } else {
1810 $content .= $char;
1811 }
1812 break;
1813 case '>':
1814 if ($bEndTag) {
1815 $sToken .= $char;
1816 if (preg_match('/\<\/\s*style\s*\>/i',$sToken,$aMatch)) {
1817 $newpos = $i + 1;
1818 $bSucces = true;
1819 break 2;
1820 } else {
1821 $content .= $sToken;
1822 }
1823 $bEndTag = false;
1824 } else {
1825 $content .= $char;
1826 }
1827 break;
1828 case '!':
1829 if ($sToken == '<') {
1830 // possible comment
1831 if (isset($body{$i+2}) && substr($body,$i,3) == '!--') {
1832 $i = strpos($body,'-->',$i+3);
2b665f28 1833 if ($i === false) { // no end comment
1834 $i = strlen($body);
1835 }
7e2ff844 1836 $sToken = '';
1837 }
1838 } else {
1839 $content .= $char;
1840 }
1841 break;
1842 default:
1843 if ($bEndTag) {
1844 $sToken .= $char;
1845 } else {
1846 $content .= $char;
1847 }
1848 break;
1849 }
1850 }
1851 if ($bSucces == FALSE){
7d06541f 1852 return array(FALSE, strlen($body));
1853 }
7e2ff844 1854
2b665f28 1855
1856
691a2d25 1857 /**
0493ed11 1858 * First look for general BODY style declaration, which would be
1859 * like so:
1860 * body {background: blah-blah}
1861 * and change it to .bodyclass so we can just assign it to a <div>
1862 */
691a2d25 1863 $content = preg_replace("|body(\s*\{.*?\})|si", ".bodyclass\\1", $content);
3d8371be 1864 $secremoveimg = '../images/' . _("sec_remove_eng.png");
691a2d25 1865 /**
0493ed11 1866 * Fix url('blah') declarations.
1867 */
1868 // $content = preg_replace("|url\s*\(\s*([\'\"])\s*\S+script\s*:.*?([\'\"])\s*\)|si",
1869 // "url(\\1$secremoveimg\\2)", $content);
2b665f28 1870
1871 // IE Sucks hard. We have a special function for it.
1872 sq_fixIE_idiocy($content);
1873
1874 // remove @import line
1875 $content = preg_replace("/^\s*(@import.*)$/mi","\n<!-- @import rules forbidden -->\n",$content);
1876
5b4884be 1877 // translate ur\l and variations (IE parses that)
2b665f28 1878 // TODO check if the sq_fixIE_idiocy function already handles this.
5b4884be 1879 $content = preg_replace("/(\\\\)?u(\\\\)?r(\\\\)?l(\\\\)?/i", 'url', $content);
0493ed11 1880 // NB I insert NUL characters to keep to avoid an infinite loop. They are removed after the loop.
1881 while (preg_match("/url\s*\(\s*[\'\"]?([^:]+):(.*)?[\'\"]?\s*\)/si", $content, $matches)) {
1882 $sProto = strtolower($matches[1]);
1883 switch ($sProto) {
0493ed11 1884 /**
758a7889 1885 * Fix url('https*://.*) declarations but only if $view_unsafe_images
1886 * is false.
0493ed11 1887 */
758a7889 1888 case 'https':
1889 case 'http':
1890 if (!$view_unsafe_images){
1891
1892 $sExpr = "/url\s*\(\s*[\'\"]?\s*$sProto*:.*[\'\"]?\s*\)/si";
1893 $content = preg_replace($sExpr, "u\0r\0l(\\1$secremoveimg\\2)", $content);
1894
1895 } else {
1896 $content = preg_replace('/url/i',"u\0r\0l",$content);
1897 }
1898 break;
1899 /**
1900 * Fix urls that refer to cid:
1901 */
1902 case 'cid':
1903 $cidurl = 'cid:'. $matches[2];
1904 $httpurl = sq_cid2http($message, $id, $cidurl, $mailbox);
1905 // escape parentheses that can modify the regular expression
1906 $cidurl = str_replace(array('(',')'),array('\\(','\\)'),$cidurl);
1907 $content = preg_replace("|url\s*\(\s*$cidurl\s*\)|si",
1908 "u\0r\0l($httpurl)", $content);
1909 break;
1910 default:
1911 /**
1912 * replace url with protocol other then the white list
1913 * http,https and cid by an empty string.
1914 */
1915 $content = preg_replace("/url\s*\(\s*[\'\"]?([^:]+):(.*)?[\'\"]?\s*\)/si",
1916 "", $content);
1917 break;
0493ed11 1918 }
691a2d25 1919 }
0493ed11 1920 // remove NUL
1921 $content = str_replace("\0", "", $content);
0493ed11 1922 /**
1923 * Remove any backslashes, entities, and extraneous whitespace.
1924 */
1925 $contentTemp = $content;
1926 sq_defang($contentTemp);
1927 sq_unspace($contentTemp);
a3daaaf3 1928
691a2d25 1929 /**
8bd0068d 1930 * Fix stupid css declarations which lead to vulnerabilities
1931 * in IE.
1932 */
5db90261 1933 $match = Array('/\/\*.*\*\//',
1934 '/expression/i',
0493ed11 1935 '/behaviou*r/i',
1936 '/binding/i',
2b665f28 1937 '/include-source/i',
1938 '/javascript/i',
1939 '/script/i');
1940 $replace = Array('','idiocy', 'idiocy', 'idiocy', 'idiocy', 'idiocy', 'idiocy');
0493ed11 1941 $contentNew = preg_replace($match, $replace, $contentTemp);
1942 if ($contentNew !== $contentTemp) {
1943 // insecure css declarations are used. From now on we don't care
1944 // anymore if the css is destroyed by sq_deent, sq_unspace or sq_unbackslash
1945 $content = $contentNew;
1946 }
7d06541f 1947 return array($content, $newpos);
691a2d25 1948}
a3daaaf3 1949
0493ed11 1950
691a2d25 1951/**
8bd0068d 1952 * This function converts cid: url's into the ones that can be viewed in
1953 * the browser.
1954 *
1955 * @param $message the message object
1956 * @param $id the message id
1957 * @param $cidurl the cid: url.
1958 * @param $mailbox the message mailbox
1959 * @return a string with a http-friendly url
1960 */
b3af12ef 1961function sq_cid2http($message, $id, $cidurl, $mailbox){
691a2d25 1962 /**
8bd0068d 1963 * Get rid of quotes.
1964 */
691a2d25 1965 $quotchar = substr($cidurl, 0, 1);
2dd879b8 1966 if ($quotchar == '"' || $quotchar == "'"){
1967 $cidurl = str_replace($quotchar, "", $cidurl);
1968 } else {
1969 $quotchar = '';
1970 }
691a2d25 1971 $cidurl = substr(trim($cidurl), 4);
0493ed11 1972
1973 $match_str = '/\{.*?\}\//';
1974 $str_rep = '';
1975 $cidurl = preg_replace($match_str, $str_rep, $cidurl);
1976
e5e9381a 1977 $linkurl = find_ent_id($cidurl, $message);
1978 /* in case of non-save cid links $httpurl should be replaced by a sort of
8bd0068d 1979 unsave link image */
e5e9381a 1980 $httpurl = '';
c8f5f606 1981
8bd0068d 1982 /**
1983 * This is part of a fix for Outlook Express 6.x generating
1984 * cid URLs without creating content-id headers. These images are
1985 * not part of the multipart/related html mail. The html contains
1986 * <img src="cid:{some_id}/image_filename.ext"> references to
1987 * attached images with as goal to render them inline although
1988 * the attachment disposition property is not inline.
1989 */
c8f5f606 1990
1991 if (empty($linkurl)) {
1992 if (preg_match('/{.*}\//', $cidurl)) {
1993 $cidurl = preg_replace('/{.*}\//','', $cidurl);
1994 if (!empty($cidurl)) {
1995 $linkurl = find_ent_id($cidurl, $message);
1996 }
1997 }
1998 }
f8a1ed5a 1999
c8f5f606 2000 if (!empty($linkurl)) {
6b04287c 2001 $httpurl = $quotchar . SM_PATH . 'src/download.php?absolute_dl=true&amp;' .
8bd0068d 2002 "passed_id=$id&amp;mailbox=" . urlencode($mailbox) .
2003 '&amp;ent_id=' . $linkurl . $quotchar;
c8f5f606 2004 } else {
2005 /**
2006 * If we couldn't generate a proper img url, drop in a blank image
2007 * instead of sending back empty, otherwise it causes unusual behaviour
2008 */
bc017c1d 2009 $httpurl = $quotchar . SM_PATH . 'images/blank.png' . $quotchar;
e5e9381a 2010 }
f8a1ed5a 2011
691a2d25 2012 return $httpurl;
2013}
2014
2015/**
8bd0068d 2016 * This function changes the <body> tag into a <div> tag since we
2017 * can't really have a body-within-body.
2018 *
2019 * @param $attary an array of attributes and values of <body>
2020 * @param $mailbox mailbox we're currently reading (for cid2http)
2021 * @param $message current message (for cid2http)
2022 * @param $id current message id (for cid2http)
2023 * @return a modified array of attributes to be set for <div>
2024 */
2dd879b8 2025function sq_body2div($attary, $mailbox, $message, $id){
b583c3e8 2026 $me = 'sq_body2div';
3d8371be 2027 $divattary = Array('class' => "'bodyclass'");
b583c3e8 2028 $text = '#000000';
c189a963 2029 $has_bgc_stl = $has_txt_stl = false;
b583c3e8 2030 $styledef = '';
691a2d25 2031 if (is_array($attary) && sizeof($attary) > 0){
2032 foreach ($attary as $attname=>$attvalue){
2033 $quotchar = substr($attvalue, 0, 1);
2034 $attvalue = str_replace($quotchar, "", $attvalue);
2035 switch ($attname){
3d8371be 2036 case 'background':
8bd0068d 2037 $attvalue = sq_cid2http($message, $id, $attvalue, $mailbox);
3d8371be 2038 $styledef .= "background-image: url('$attvalue'); ";
2039 break;
2040 case 'bgcolor':
c189a963 2041 $has_bgc_stl = true;
3d8371be 2042 $styledef .= "background-color: $attvalue; ";
2043 break;
2044 case 'text':
c189a963 2045 $has_txt_stl = true;
3d8371be 2046 $styledef .= "color: $attvalue; ";
2047 break;
691a2d25 2048 }
a3daaaf3 2049 }
c189a963 2050 // Outlook defines a white bgcolor and no text color. This can lead to
2051 // white text on a white bg with certain themes.
2052 if ($has_bgc_stl && !$has_txt_stl) {
2053 $styledef .= "color: $text; ";
2054 }
691a2d25 2055 if (strlen($styledef) > 0){
2056 $divattary{"style"} = "\"$styledef\"";
2057 }
2058 }
2059 return $divattary;
2060}
a3daaaf3 2061
691a2d25 2062/**
8bd0068d 2063 * This is the main function and the one you should actually be calling.
2064 * There are several variables you should be aware of an which need
2065 * special description.
2066 *
2067 * Since the description is quite lengthy, see it here:
2068 * http://linux.duke.edu/projects/mini/htmlfilter/
2069 *
2070 * @param $body the string with HTML you wish to filter
2071 * @param $tag_list see description above
2072 * @param $rm_tags_with_content see description above
2073 * @param $self_closing_tags see description above
2074 * @param $force_tag_closing see description above
2075 * @param $rm_attnames see description above
2076 * @param $bad_attvals see description above
2077 * @param $add_attr_to_tag see description above
2078 * @param $message message object
2079 * @param $id message id
2080 * @return sanitized html safe to show on your pages.
2081 */
da2415c1 2082function sq_sanitize($body,
8bd0068d 2083 $tag_list,
2084 $rm_tags_with_content,
2085 $self_closing_tags,
2086 $force_tag_closing,
2087 $rm_attnames,
2088 $bad_attvals,
2089 $add_attr_to_tag,
2090 $message,
2091 $id,
2092 $mailbox
2093 ){
b583c3e8 2094 $me = 'sq_sanitize';
7d06541f 2095 $rm_tags = array_shift($tag_list);
691a2d25 2096 /**
8bd0068d 2097 * Normalize rm_tags and rm_tags_with_content.
2098 */
7d06541f 2099 @array_walk($tag_list, 'sq_casenormalize');
691a2d25 2100 @array_walk($rm_tags_with_content, 'sq_casenormalize');
2101 @array_walk($self_closing_tags, 'sq_casenormalize');
2102 /**
8bd0068d 2103 * See if tag_list is of tags to remove or tags to allow.
2104 * false means remove these tags
2105 * true means allow these tags
2106 */
691a2d25 2107 $curpos = 0;
2108 $open_tags = Array();
2dd879b8 2109 $trusted = "\n<!-- begin sanitized html -->\n";
691a2d25 2110 $skip_content = false;
bb8d0799 2111 /**
8bd0068d 2112 * Take care of netscape's stupid javascript entities like
2113 * &{alert('boo')};
2114 */
bb8d0799 2115 $body = preg_replace("/&(\{.*?\};)/si", "&amp;\\1", $body);
691a2d25 2116
7d06541f 2117 while (($curtag = sq_getnxtag($body, $curpos)) != FALSE){
691a2d25 2118 list($tagname, $attary, $tagtype, $lt, $gt) = $curtag;
2119 $free_content = substr($body, $curpos, $lt-$curpos);
2120 /**
8bd0068d 2121 * Take care of <style>
2122 */
7d06541f 2123 if ($tagname == "style" && $tagtype == 1){
da2415c1 2124 list($free_content, $curpos) =
e60a299a 2125 sq_fixstyle($body, $gt+1, $message, $id, $mailbox);
7d06541f 2126 if ($free_content != FALSE){
2127 $trusted .= sq_tagprint($tagname, $attary, $tagtype);
2128 $trusted .= $free_content;
2129 $trusted .= sq_tagprint($tagname, false, 2);
2130 }
2131 continue;
691a2d25 2132 }
2133 if ($skip_content == false){
2134 $trusted .= $free_content;
691a2d25 2135 }
2136 if ($tagname != FALSE){
2137 if ($tagtype == 2){
2138 if ($skip_content == $tagname){
2139 /**
8bd0068d 2140 * Got to the end of tag we needed to remove.
2141 */
691a2d25 2142 $tagname = false;
2143 $skip_content = false;
2144 } else {
2145 if ($skip_content == false){
c828931c 2146 if ($tagname == "body"){
2147 $tagname = "div";
2dd879b8 2148 }
da2415c1 2149 if (isset($open_tags{$tagname}) &&
8bd0068d 2150 $open_tags{$tagname} > 0){
2dd879b8 2151 $open_tags{$tagname}--;
691a2d25 2152 } else {
2dd879b8 2153 $tagname = false;
691a2d25 2154 }
691a2d25 2155 }
2156 }
2157 } else {
2158 /**
8bd0068d 2159 * $rm_tags_with_content
2160 */
691a2d25 2161 if ($skip_content == false){
2162 /**
8bd0068d 2163 * See if this is a self-closing type and change
2164 * tagtype appropriately.
2165 */
691a2d25 2166 if ($tagtype == 1
8bd0068d 2167 && in_array($tagname, $self_closing_tags)){
2dd879b8 2168 $tagtype = 3;
691a2d25 2169 }
2170 /**
8bd0068d 2171 * See if we should skip this tag and any content
2172 * inside it.
2173 */
691a2d25 2174 if ($tagtype == 1 &&
8bd0068d 2175 in_array($tagname, $rm_tags_with_content)){
691a2d25 2176 $skip_content = $tagname;
2177 } else {
da2415c1 2178 if (($rm_tags == false
8bd0068d 2179 && in_array($tagname, $tag_list)) ||
2180 ($rm_tags == true &&
2181 !in_array($tagname, $tag_list))){
691a2d25 2182 $tagname = false;
2183 } else {
2dd879b8 2184 /**
8bd0068d 2185 * Convert body into div.
2186 */
2dd879b8 2187 if ($tagname == "body"){
2188 $tagname = "div";
da2415c1 2189 $attary = sq_body2div($attary, $mailbox,
8bd0068d 2190 $message, $id);
2dd879b8 2191 }
691a2d25 2192 if ($tagtype == 1){
2193 if (isset($open_tags{$tagname})){
2194 $open_tags{$tagname}++;
2195 } else {
2196 $open_tags{$tagname}=1;
2197 }
2198 }
2199 /**
8bd0068d 2200 * This is where we run other checks.
2201 */
691a2d25 2202 if (is_array($attary) && sizeof($attary) > 0){
2203 $attary = sq_fixatts($tagname,
8bd0068d 2204 $attary,
2205 $rm_attnames,
2206 $bad_attvals,
2207 $add_attr_to_tag,
2208 $message,
2209 $id,
2210 $mailbox
2211 );
691a2d25 2212 }
2213 }
2214 }
691a2d25 2215 }
2216 }
2217 if ($tagname != false && $skip_content == false){
2218 $trusted .= sq_tagprint($tagname, $attary, $tagtype);
2219 }
691a2d25 2220 }
2221 $curpos = $gt+1;
a3daaaf3 2222 }
691a2d25 2223 $trusted .= substr($body, $curpos, strlen($body)-$curpos);
2224 if ($force_tag_closing == true){
2225 foreach ($open_tags as $tagname=>$opentimes){
2226 while ($opentimes > 0){
2227 $trusted .= '</' . $tagname . '>';
2228 $opentimes--;
2229 }
2230 }
2231 $trusted .= "\n";
2232 }
2233 $trusted .= "<!-- end sanitized html -->\n";
2234 return $trusted;
2235}
451f74a2 2236
691a2d25 2237/**
8bd0068d 2238 * This is a wrapper function to call html sanitizing routines.
2239 *
2240 * @param $body the body of the message
2241 * @param $id the id of the message
c189a963 2242
2243 * @param $message
2244 * @param $mailbox
2245 * @param boolean $take_mailto_links When TRUE, converts mailto: links
2246 * into internal SM compose links
2247 * (optional; default = TRUE)
8bd0068d 2248 * @return a string with html safe to display in the browser.
2249 */
c189a963 2250function magicHTML($body, $id, $message, $mailbox = 'INBOX', $take_mailto_links =true) {
2251
202bcbcc 2252 // require_once(SM_PATH . 'functions/url_parser.php'); // for $MailTo_PReg_Match
c189a963 2253
691a2d25 2254 global $attachment_common_show_images, $view_unsafe_images,
8bd0068d 2255 $has_unsafe_images;
691a2d25 2256 /**
8bd0068d 2257 * Don't display attached images in HTML mode.
2b665f28 2258 *
d0187bd6 2259 * SB: why?
8bd0068d 2260 */
691a2d25 2261 $attachment_common_show_images = false;
2262 $tag_list = Array(
8bd0068d 2263 false,
2264 "object",
2265 "meta",
2266 "html",
2267 "head",
2268 "base",
2269 "link",
2270 "frame",
2271 "iframe",
2272 "plaintext",
2273 "marquee"
2274 );
691a2d25 2275
2276 $rm_tags_with_content = Array(
8bd0068d 2277 "script",
2278 "applet",
2279 "embed",
2280 "title",
2281 "frameset",
0493ed11 2282 "xmp",
8bd0068d 2283 "xml"
2284 );
691a2d25 2285
2286 $self_closing_tags = Array(
8bd0068d 2287 "img",
2288 "br",
2289 "hr",
2290 "input",
2291 "outbind"
2292 );
691a2d25 2293
2dd879b8 2294 $force_tag_closing = true;
691a2d25 2295
2296 $rm_attnames = Array(
8bd0068d 2297 "/.*/" =>
2298 Array(
2299 "/target/i",
2300 "/^on.*/i",
2301 "/^dynsrc/i",
2302 "/^data.*/i",
2303 "/^lowsrc.*/i"
2304 )
2305 );
691a2d25 2306
2307 $secremoveimg = "../images/" . _("sec_remove_eng.png");
2308 $bad_attvals = Array(
8bd0068d 2309 "/.*/" =>
691a2d25 2310 Array(
0a6ec9b5 2311 "/^src|background/i" =>
8bd0068d 2312 Array(
691a2d25 2313 Array(
8bd0068d 2314 "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si",
2315 "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si",
2316 "/^([\'\"])\s*about\s*:.*([\'\"])/si"
691a2d25 2317 ),
8bd0068d 2318 Array(
2319 "\\1$secremoveimg\\2",
2320 "\\1$secremoveimg\\2",
2321 "\\1$secremoveimg\\2",
8bd0068d 2322 )
2323 ),
0a6ec9b5 2324 "/^href|action/i" =>
8bd0068d 2325 Array(
0a6ec9b5 2326 Array(
8bd0068d 2327 "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si",
2328 "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si",
2329 "/^([\'\"])\s*about\s*:.*([\'\"])/si"
0a6ec9b5 2330 ),
691a2d25 2331 Array(
8bd0068d 2332 "\\1#\\1",
2333 "\\1#\\1",
2334 "\\1#\\1"
02474e43 2335 )
8bd0068d 2336 ),
2337 "/^style/i" =>
2338 Array(
2339 Array(
5db90261 2340 "/\/\*.*\*\//",
8bd0068d 2341 "/expression/i",
2342 "/binding/i",
2343 "/behaviou*r/i",
2344 "/include-source/i",
2345 "/position\s*:\s*absolute/i",
1d935bc2 2346 "/(\\\\)?u(\\\\)?r(\\\\)?l(\\\\)?/i",
8bd0068d 2347 "/url\s*\(\s*([\'\"])\s*\S+script\s*:.*([\'\"])\s*\)/si",
2348 "/url\s*\(\s*([\'\"])\s*mocha\s*:.*([\'\"])\s*\)/si",
2349 "/url\s*\(\s*([\'\"])\s*about\s*:.*([\'\"])\s*\)/si",
2350 "/(.*)\s*:\s*url\s*\(\s*([\'\"]*)\s*\S+script\s*:.*([\'\"]*)\s*\)/si"
2351 ),
2352 Array(
5db90261 2353 "",
8bd0068d 2354 "idiocy",
2355 "idiocy",
2356 "idiocy",
2357 "idiocy",
2358 "",
5b4884be 2359 "url",
8bd0068d 2360 "url(\\1#\\1)",
2361 "url(\\1#\\1)",
2362 "url(\\1#\\1)",
8bd0068d 2363 "\\1:url(\\2#\\3)"
2364 )
691a2d25 2365 )
8bd0068d 2366 )
691a2d25 2367 );
5262d9a6 2368 if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET) ) {
2dd879b8 2369 $view_unsafe_images = false;
45071bd6 2370 }
691a2d25 2371 if (!$view_unsafe_images){
2372 /**
8bd0068d 2373 * Remove any references to http/https if view_unsafe_images set
2374 * to false.
2375 */
02474e43 2376 array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[0],
8bd0068d 2377 '/^([\'\"])\s*https*:.*([\'\"])/si');
02474e43 2378 array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[1],
8bd0068d 2379 "\\1$secremoveimg\\1");
02474e43 2380 array_push($bad_attvals{'/.*/'}{'/^style/i'}[0],
7ef0b415 2381 '/url\([\'\"]?https?:[^\)]*[\'\"]?\)/si');
02474e43 2382 array_push($bad_attvals{'/.*/'}{'/^style/i'}[1],
8bd0068d 2383 "url(\\1$secremoveimg\\1)");
691a2d25 2384 }
451f74a2 2385
691a2d25 2386 $add_attr_to_tag = Array(
8bd0068d 2387 "/^a$/i" =>
c25f2fbb 2388 Array('target'=>'"_blank"',
02474e43 2389 'title'=>'"'._("This external link will open in a new window").'"'
8bd0068d 2390 )
2391 );
da2415c1 2392 $trusted = sq_sanitize($body,
8bd0068d 2393 $tag_list,
2394 $rm_tags_with_content,
2395 $self_closing_tags,
2396 $force_tag_closing,
2397 $rm_attnames,
2398 $bad_attvals,
2399 $add_attr_to_tag,
2400 $message,
2401 $id,
2402 $mailbox
2403 );
f83c60a2 2404 if (preg_match("|$secremoveimg|i", $trusted)){
691a2d25 2405 $has_unsafe_images = true;
da2415c1 2406 }
c189a963 2407
2408 // we want to parse mailto's in HTML output, change to SM compose links
2409 // this is a modified version of code from url_parser.php... but Marc is
2410 // right: we need a better filtering implementation; adding this randomly
2411 // here is not a great solution
2412 //
2413 if ($take_mailto_links) {
2414 // parseUrl($trusted); // this even parses URLs inside of tags... too aggressive
2415 global $MailTo_PReg_Match;
202bcbcc 2416 $MailTo_PReg_Match = '/mailto:' . substr($MailTo_PReg_Match, 1) ;
c189a963 2417 if ((preg_match_all($MailTo_PReg_Match, $trusted, $regs)) && ($regs[0][0] != '')) {
2418 foreach ($regs[0] as $i => $mailto_before) {
2419 $mailto_params = $regs[10][$i];
2420 // get rid of any tailing quote since we have to add send_to to the end
2421 //
2422 if (substr($mailto_before, strlen($mailto_before) - 1) == '"')
2423 $mailto_before = substr($mailto_before, 0, strlen($mailto_before) - 1);
2424 if (substr($mailto_params, strlen($mailto_params) - 1) == '"')
2425 $mailto_params = substr($mailto_params, 0, strlen($mailto_params) - 1);
2426
2427 if ($regs[1][$i]) { //if there is an email addr before '?', we need to merge it with the params
2428 $to = 'to=' . $regs[1][$i];
2429 if (strpos($mailto_params, 'to=') > -1) //already a 'to='
2430 $mailto_params = str_replace('to=', $to . '%2C%20', $mailto_params);
2431 else {
2432 if ($mailto_params) //already some params, append to them
2433 $mailto_params .= '&amp;' . $to;
2434 else
2435 $mailto_params .= '?' . $to;
2436 }
2437 }
2438
2439 $url_str = preg_replace(array('/to=/i', '/(?<!b)cc=/i', '/bcc=/i'), array('send_to=', 'send_to_cc=', 'send_to_bcc='), $mailto_params);
2440
2441 // we'll already have target=_blank, no need to allow comp_in_new
2442 // here (which would be a lot more work anyway)
2443 //
2444 global $compose_new_win;
2445 $temp_comp_in_new = $compose_new_win;
2446 $compose_new_win = 0;
2447 $comp_uri = makeComposeLink('src/compose.php' . $url_str, $mailto_before);
2448 $compose_new_win = $temp_comp_in_new;
2449
2450 // remove <a href=" and anything after the next quote (we only
2451 // need the uri, not the link HTML) in compose uri
2452 //
2453 $comp_uri = substr($comp_uri, 9);
2454 $comp_uri = substr($comp_uri, 0, strpos($comp_uri, '"', 1));
2455 $trusted = str_replace($mailto_before, $comp_uri, $trusted);
2456 }
2457 }
2458 }
2459
691a2d25 2460 return $trusted;
451f74a2 2461}
a4a70693 2462
da2415c1 2463/**
8bd0068d 2464 * function SendDownloadHeaders - send file to the browser
2465 *
2466 * Original Source: SM core src/download.php
2467 * moved here to make it available to other code, and separate
2468 * front end from back end functionality.
2469 *
2470 * @param string $type0 first half of mime type
2471 * @param string $type1 second half of mime type
2472 * @param string $filename filename to tell the browser for downloaded file
2473 * @param boolean $force whether to force the download dialog to pop
2474 * @param optional integer $filesize send the Content-Header and length to the browser
2475 * @return void
2476 */
02474e43 2477function SendDownloadHeaders($type0, $type1, $filename, $force, $filesize=0) {
2478 global $languages, $squirrelmail_language;
cfffd60b 2479 $isIE = $isIE6plus = false;
02474e43 2480
2481 sqgetGlobalVar('HTTP_USER_AGENT', $HTTP_USER_AGENT, SQ_SERVER);
2482
2483 if (strstr($HTTP_USER_AGENT, 'compatible; MSIE ') !== false &&
8bd0068d 2484 strstr($HTTP_USER_AGENT, 'Opera') === false) {
cfffd60b 2485 $isIE = true;
02474e43 2486 }
2487
cfffd60b 2488 if (preg_match('/compatible; MSIE ([0-9]+)/', $HTTP_USER_AGENT, $match) &&
2489 ((int)$match[1]) >= 6 && strstr($HTTP_USER_AGENT, 'Opera') === false) {
2490 $isIE6plus = true;
02474e43 2491 }
2492
2493 if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
8bd0068d 2494 function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_downloadfilename')) {
02474e43 2495 $filename =
8bd0068d 2496 call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_downloadfilename', $filename, $HTTP_USER_AGENT);
02474e43 2497 } else {
2498 $filename = ereg_replace('[\\/:\*\?"<>\|;]', '_', str_replace('&nbsp;', ' ', $filename));
2499 }
2500
2501 // A Pox on Microsoft and it's Internet Explorer!
2502 //
2503 // IE has lots of bugs with file downloads.
2504 // It also has problems with SSL. Both of these cause problems
2505 // for us in this function.
2506 //
2507 // See this article on Cache Control headers and SSL
2508 // http://support.microsoft.com/default.aspx?scid=kb;en-us;323308
2509 //
2510 // The best thing you can do for IE is to upgrade to the latest
2511 // version
2512 //set all the Cache Control Headers for IE
2513 if ($isIE) {
2514 $filename=rawurlencode($filename);
2515 header ("Pragma: public");
8bd0068d 2516 header ("Cache-Control: no-store, max-age=0, no-cache, must-revalidate"); // HTTP/1.1
02474e43 2517 header ("Cache-Control: post-check=0, pre-check=0", false);
8bd0068d 2518 header ("Cache-Control: private");
02474e43 2519
2520 //set the inline header for IE, we'll add the attachment header later if we need it
2521 header ("Content-Disposition: inline; filename=$filename");
2522 }
2523
2524 if (!$force) {
2525 // Try to show in browser window
2526 header ("Content-Disposition: inline; filename=\"$filename\"");
2527 header ("Content-Type: $type0/$type1; name=\"$filename\"");
2528 } else {
2529 // Try to pop up the "save as" box
2530
2531 // IE makes this hard. It pops up 2 save boxes, or none.
2532 // http://support.microsoft.com/support/kb/articles/Q238/5/88.ASP
2533 // http://support.microsoft.com/default.aspx?scid=kb;EN-US;260519
2534 // But, according to Microsoft, it is "RFC compliant but doesn't
2535 // take into account some deviations that allowed within the
2536 // specification." Doesn't that mean RFC non-compliant?
2537 // http://support.microsoft.com/support/kb/articles/Q258/4/52.ASP
2538
2539 // all browsers need the application/octet-stream header for this
2540 header ("Content-Type: application/octet-stream; name=\"$filename\"");
2541
2542 // http://support.microsoft.com/support/kb/articles/Q182/3/15.asp
2543 // Do not have quotes around filename, but that applied to
2544 // "attachment"... does it apply to inline too?
2545 header ("Content-Disposition: attachment; filename=\"$filename\"");
2546
cfffd60b 2547 if ($isIE && !$isIE6plus) {
02474e43 2548 // This combination seems to work mostly. IE 5.5 SP 1 has
2549 // known issues (see the Microsoft Knowledge Base)
2550
2551 // This works for most types, but doesn't work with Word files
2552 header ("Content-Type: application/download; name=\"$filename\"");
7e2ff844 2553 header ("Content-Type: application/force-download; name=\"$filename\"");
02474e43 2554 // These are spares, just in case. :-)
2555 //header("Content-Type: $type0/$type1; name=\"$filename\"");
2556 //header("Content-Type: application/x-msdownload; name=\"$filename\"");
2557 //header("Content-Type: application/octet-stream; name=\"$filename\"");
7e2ff844 2558 } else if ($isIE) {
2559 // This is to prevent IE for MIME sniffing and auto open a file in IE
2560 header ("Content-Type: application/force-download; name=\"$filename\"");
02474e43 2561 } else {
2562 // another application/octet-stream forces download for Netscape
2563 header ("Content-Type: application/octet-stream; name=\"$filename\"");
2564 }
2565 }
2566
2567 //send the content-length header if the calling function provides it
2568 if ($filesize > 0) {
2569 header("Content-Length: $filesize");
2570 }
07c49f57 2571
8d863f64 2572} // end fn SendDownloadHeaders
Unnamed repository; edit this file 'description' to name the repository.