6 * Copyright (c) 1999-2005 The SquirrelMail Project Team
7 * Licensed under the GNU GPL. For full terms see the file COPYING.
9 * This contains the functions necessary to detect and decode MIME
13 * @package squirrelmail
16 /** The typical includes... */
17 require_once(SM_PATH
. 'functions/imap.php');
18 require_once(SM_PATH
. 'functions/attachment_common.php');
20 /* -------------------------------------------------------------------------- */
22 /* -------------------------------------------------------------------------- */
25 * Get the MIME structure
27 * This function gets the structure of a message and stores it in the "message" class.
28 * It will return this object for use with all relevant header information and
29 * fully parsed into the standard "message" object format.
31 function mime_structure ($bodystructure, $flags=array()) {
33 /* Isolate the body structure and remove beginning and end parenthesis. */
34 $read = trim(substr ($bodystructure, strpos(strtolower($bodystructure), 'bodystructure') +
13));
35 $read = trim(substr ($read, 0, -1));
37 $msg = Message
::parseStructure($read,$i);
38 if (!is_object($msg)) {
39 include_once(SM_PATH
. 'functions/display_messages.php');
40 global $color, $mailbox;
41 /* removed urldecode because $_GET is auto urldecoded ??? */
42 displayPageHeader( $color, $mailbox );
43 $errormessage = _("SquirrelMail could not decode the bodystructure of the message");
44 $errormessage .= '<br />'._("the provided bodystructure by your imap-server").':<br /><br />';
45 $errormessage .= '<pre>' . htmlspecialchars($read) . '</pre>';
46 plain_error_message( $errormessage, $color );
47 echo '</body></html>';
51 foreach ($flags as $flag) {
52 $char = strtoupper($flag{1});
55 if (strtolower($flag) == '\\seen') {
60 if (strtolower($flag) == '\\answered') {
61 $msg->is_answered
= true;
65 if (strtolower($flag) == '\\deleted') {
66 $msg->is_deleted
= true;
70 if (strtolower($flag) == '\\flagged') {
71 $msg->is_flagged
= true;
75 if (strtolower($flag) == '$mdnsent') {
76 $msg->is_mdnsent
= true;
84 // listEntities($msg);
90 /* This starts the parsing of a particular structure. It is called recursively,
91 * so it can be passed different structures. It returns an object of type
93 * First, it checks to see if it is a multipart message. If it is, then it
94 * handles that as it sees is necessary. If it is just a regular entity,
95 * then it parses it and adds the necessary header information (by calling out
96 * to mime_get_elements()
99 function mime_fetch_body($imap_stream, $id, $ent_id=1, $fetch_size=0) {
100 /* Do a bit of error correction. If we couldn't find the entity id, just guess
101 * that it is the first one. That is usually the case anyway.
105 $cmd = "FETCH $id BODY[]";
107 $cmd = "FETCH $id BODY[$ent_id]";
110 if ($fetch_size!=0) $cmd .= "<0.$fetch_size>";
112 $data = sqimap_run_command ($imap_stream, $cmd, true, $response, $message, TRUE);
114 $topline = trim(array_shift($data));
115 } while($topline && ($topline[0] == '*') && !preg_match('/\* [0-9]+ FETCH.*/i', $topline)) ;
117 $wholemessage = implode('', $data);
118 if (ereg('\\{([^\\}]*)\\}', $topline, $regs)) {
119 $ret = substr($wholemessage, 0, $regs[1]);
120 /* There is some information in the content info header that could be important
121 * in order to parse html messages. Let's get them here.
123 // if ($ret{0} == '<') {
124 // $data = sqimap_run_command ($imap_stream, "FETCH $id BODY[$ent_id.MIME]", true, $response, $message, TRUE);
126 } else if (ereg('"([^"]*)"', $topline, $regs)) {
129 global $where, $what, $mailbox, $passed_id, $startMessage;
130 $par = 'mailbox=' . urlencode($mailbox) . '&passed_id=' . $passed_id;
131 if (isset($where) && isset($what)) {
132 $par .= '&where=' . urlencode($where) . '&what=' . urlencode($what);
134 $par .= '&startMessage=' . $startMessage . '&show_more=0';
136 $par .= '&response=' . urlencode($response) .
137 '&message=' . urlencode($message) .
138 '&topline=' . urlencode($topline);
141 '<table width="80%"><tr>' .
142 '<tr><td colspan="2">' .
143 _("Body retrieval error. The reason for this is most probably that the message is malformed.") .
145 '<tr><td><b>' . _("Command:") . "</td><td>$cmd</td></tr>" .
146 '<tr><td><b>' . _("Response:") . "</td><td>$response</td></tr>" .
147 '<tr><td><b>' . _("Message:") . "</td><td>$message</td></tr>" .
148 '<tr><td><b>' . _("FETCH line:") . "</td><td>$topline</td></tr>" .
149 "</table><br /></tt></font><hr />";
151 $data = sqimap_run_command ($imap_stream, "FETCH $passed_id BODY[]", true, $response, $message, TRUE);
153 $wholemessage = implode('', $data);
155 $ret = $wholemessage;
160 function mime_print_body_lines ($imap_stream, $id, $ent_id=1, $encoding) {
162 /* Don't kill the connection if the browser is over a dialup
163 * and it would take over 30 seconds to download it.
164 * Don't call set_time_limit in safe mode.
167 if (!ini_get('safe_mode')) {
170 /* in case of base64 encoded attachments, do not buffer them.
171 Instead, echo the decoded attachment directly to screen */
172 if (strtolower($encoding) == 'base64') {
174 $query = "FETCH $id BODY[]";
176 $query = "FETCH $id BODY[$ent_id]";
178 sqimap_run_command($imap_stream,$query,true,$response,$message,TRUE,'sqimap_base64_decode','php://stdout',true);
180 $body = mime_fetch_body ($imap_stream, $id, $ent_id);
181 echo decodeBody($body, $encoding);
185 TODO, use the same method for quoted printable.
186 However, I assume that quoted printable attachments aren't that large
187 so the performancegain / memory usage drop will be minimal.
188 If we decide to add that then we need to adapt sqimap_fread because
189 we need to split te result on \n and fread doesn't stop at \n. That
190 means we also should provide $results from sqimap_fread (by ref) to
191 te function and set $no_return to false. The $filter function for
192 quoted printable should handle unsetting of $results.
195 TODO 2: find out how we write to the output stream php://stdout. fwrite
196 doesn't work because 'php://stdout isn't a stream.
201 fputs ($imap_stream, "$sid FETCH $id BODY[$ent_id]\r\n");
204 $read = fgets ($imap_stream,8192);
207 // This could be bad -- if the section has sqimap_session_id() . ' OK'
208 // or similar, it will kill the download.
209 while (!ereg("^".$sid_s." (OK|BAD|NO)(.*)$", $read, $regs)) {
210 if (trim($read) == ')==') {
212 $read = fgets ($imap_stream,4096);
213 if (ereg("^".$sid." (OK|BAD|NO)(.*)$", $read, $regs)) {
216 echo decodeBody($read1, $encoding) .
217 decodeBody($read, $encoding);
220 echo decodeBody($read, $encoding);
222 $read = fgets ($imap_stream,4096);
229 /* -[ END MIME DECODING ]----------------------------------------------------------- */
231 /* This is here for debugging purposes. It will print out a list
232 * of all the entity IDs that are in the $message object.
234 function listEntities ($message) {
236 echo "<tt>" . $message->entity_id
. ' : ' . $message->type0
. '/' . $message->type1
. ' parent = '. $message->parent
->entity_id
. '<br />';
237 for ($i = 0; isset($message->entities
[$i]); $i++
) {
239 $msg = listEntities($message->entities
[$i]);
249 function getPriorityStr($priority) {
250 $priority_level = substr($priority,0,1);
252 switch($priority_level) {
253 /* Check for a higher then normal priority. */
256 $priority_string = _("High");
259 /* Check for a lower then normal priority. */
262 $priority_string = _("Low");
265 /* Check for a normal priority. */
268 $priority_level = '3';
269 $priority_string = _("Normal");
273 return $priority_string;
276 /* returns a $message object for a particular entity id */
277 function getEntity ($message, $ent_id) {
278 return $message->getEntity($ent_id);
282 * Extracted from strings.php 23/03/2002
285 function translateText(&$body, $wrap_at, $charset) {
286 global $where, $what; /* from searching */
287 global $color; /* color theme */
289 require_once(SM_PATH
. 'functions/url_parser.php');
291 $body_ary = explode("\n", $body);
292 for ($i=0; $i < count($body_ary); $i++
) {
293 $line = $body_ary[$i];
294 if (strlen($line) - 2 >= $wrap_at) {
295 sqWordWrap($line, $wrap_at, $charset);
297 $line = charset_decode($charset, $line);
298 $line = str_replace("\t", ' ', $line);
307 if ($line[$pos] == ' ') {
309 } else if (strpos($line, '>', $pos) === $pos) {
318 if (!isset($color[13])) {
319 $color[13] = '#800000';
321 $line = '<font color="' . $color[13] . '">' . $line . '</font>';
323 if (!isset($color[14])) {
324 $color[14] = '#FF0000';
326 $line = '<font color="' . $color[14] . '">' . $line . '</font>';
329 $body_ary[$i] = $line;
331 $body = '<pre>' . implode("\n", $body_ary) . '</pre>';
335 * This returns a parsed string called $body. That string can then
336 * be displayed as the actual message in the HTML. It contains
337 * everything needed, including HTML Tags, Attachments at the
339 * @param clean Do not output stuff that's irrelevant for the printable version.
341 function formatBody($imap_stream, $message, $color, $wrap_at, $ent_num, $id, $mailbox='INBOX', $clean=FALSE) {
342 /* This if statement checks for the entity to show as the
343 * primary message. To add more of them, just put them in the
344 * order that is their priority.
346 global $startMessage, $languages, $squirrelmail_language,
347 $show_html_default, $sort, $has_unsafe_images, $passed_ent_id;
349 if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET
) ) {
350 $view_unsafe_images = false;
354 $urlmailbox = urlencode($mailbox);
355 $body_message = getEntity($message, $ent_num);
356 if (($body_message->header
->type0
== 'text') ||
357 ($body_message->header
->type0
== 'rfc822')) {
358 $body = mime_fetch_body ($imap_stream, $id, $ent_num);
359 $body = decodeBody($body, $body_message->header
->encoding
);
361 if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
362 function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_decode')) {
363 if (mb_detect_encoding($body) != 'ASCII') {
364 $body = call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_decode',$body);
367 $hookResults = do_hook("message_body", $body);
368 $body = $hookResults[1];
370 /* If there are other types that shouldn't be formatted, add
374 if ($body_message->header
->type1
== 'html') {
375 if ($show_html_default <> 1) {
376 $entity_conv = array(' ' => ' ',
385 $body = strtr($body, $entity_conv);
386 $body = strip_tags($body);
388 translateText($body, $wrap_at,
389 $body_message->header
->getParameter('charset'));
391 $body = magicHTML($body, $id, $message, $mailbox);
394 translateText($body, $wrap_at,
395 $body_message->header
->getParameter('charset'));
398 // if this is the clean display (i.e. printer friendly), stop here.
403 $link = 'passed_id=' . $id . '&ent_id='.$ent_num.
404 '&mailbox=' . $urlmailbox .'&sort=' . $sort .
405 '&startMessage=' . $startMessage . '&show_more=0';
406 if (isset($passed_ent_id)) {
407 $link .= '&passed_ent_id='.$passed_ent_id;
409 $body .= '<center><small><a href="download.php?absolute_dl=true&' .
410 $link . '">' . _("Download this as a file") . '</a>';
411 if ($view_unsafe_images) {
412 $text = _("Hide Unsafe Images");
414 if (isset($has_unsafe_images) && $has_unsafe_images) {
415 $link .= '&view_unsafe_images=1';
416 $text = _("View Unsafe Images");
422 $body .= ' | <a href="read_body.php?' . $link . '">' . $text . '</a>';
424 $body .= '</small></center><br />' . "\n";
430 function formatAttachments($message, $exclude_id, $mailbox, $id) {
431 global $where, $what, $startMessage, $color, $passed_ent_id;
433 $att_ar = $message->getAttachments($exclude_id);
435 if (!count($att_ar)) return '';
439 $urlMailbox = urlencode($mailbox);
441 foreach ($att_ar as $att) {
442 $ent = $att->entity_id
;
443 $header = $att->header
;
444 $type0 = strtolower($header->type0
);
445 $type1 = strtolower($header->type1
);
447 $links['download link']['text'] = _("Download");
448 $links['download link']['href'] = SM_PATH
.
449 "src/download.php?absolute_dl=true&passed_id=$id&mailbox=$urlMailbox&ent_id=$ent";
450 if ($type0 =='message' && $type1 == 'rfc822') {
451 $default_page = SM_PATH
. 'src/read_body.php';
452 $rfc822_header = $att->rfc822_header
;
453 $filename = $rfc822_header->subject
;
454 if (trim( $filename ) == '') {
455 $filename = 'untitled-[' . $ent . ']' ;
457 $from_o = $rfc822_header->from
;
458 if (is_object($from_o)) {
459 $from_name = decodeHeader($from_o->getAddress(false));
461 $from_name = _("Unknown sender");
463 $description = $from_name;
465 $default_page = SM_PATH
. 'src/download.php';
466 $filename = $att->getFilename();
467 if ($header->description
) {
468 $description = decodeHeader($header->description
);
474 $display_filename = $filename;
475 if (isset($passed_ent_id)) {
476 $passed_ent_id_link = '&passed_ent_id='.$passed_ent_id;
478 $passed_ent_id_link = '';
480 $defaultlink = $default_page . "?startMessage=$startMessage"
481 . "&passed_id=$id&mailbox=$urlMailbox"
482 . '&ent_id='.$ent.$passed_ent_id_link;
483 if ($where && $what) {
484 $defaultlink .= '&where='. urlencode($where).'&what='.urlencode($what);
487 /* This executes the attachment hook with a specific MIME-type.
488 * If that doesn't have results, it tries if there's a rule
489 * for a more generic type.
491 $hookresults = do_hook("attachment $type0/$type1", $links,
492 $startMessage, $id, $urlMailbox, $ent, $defaultlink,
493 $display_filename, $where, $what);
494 if(count($hookresults[1]) <= 1) {
495 $hookresults = do_hook("attachment $type0/*", $links,
496 $startMessage, $id, $urlMailbox, $ent, $defaultlink,
497 $display_filename, $where, $what);
500 $links = $hookresults[1];
501 $defaultlink = $hookresults[6];
503 $attachments .= '<tr><td>' .
504 '<a href="'.$defaultlink.'">'.decodeHeader($display_filename).'</a> </td>' .
505 '<td><small><b>' . show_readable_size($header->size
) .
506 '</b> </small></td>' .
507 '<td><small>[ '.htmlspecialchars($type0).'/'.htmlspecialchars($type1).' ] </small></td>' .
509 $attachments .= '<b>' . $description . '</b>';
510 $attachments .= '</small></td><td><small> ';
513 foreach ($links as $val) {
517 $attachments .= ' | ';
519 $attachments .= '<a href="' . $val['href'] . '">' . $val['text'] . '</a>';
522 $attachments .= "</td></tr>\n";
524 $attachmentadd = do_hook_function('attachments_bottom',$attachments);
525 if ($attachmentadd != '')
526 $attachments = $attachmentadd;
530 function sqimap_base64_decode(&$string) {
532 // Base64 encoded data goes in pairs of 4 bytes. To achieve on the
533 // fly decoding (to reduce memory usage) you have to check if the
534 // data has incomplete pairs
536 // Remove the noise in order to check if the 4 bytes pairs are complete
537 $string = str_replace(array("\r\n","\n", "\r", " "),array('','','',''),$string);
540 $iMod = strlen($string) %
4;
542 $sStringRem = substr($string,-$iMod);
543 // Check if $sStringRem contains padding characters
544 if (substr($sStringRem,-1) != '=') {
545 $string = substr($string,0,-$iMod);
550 $string = base64_decode($string);
555 /* This function decodes the body depending on the encoding type. */
556 function decodeBody($body, $encoding) {
557 global $show_html_default;
559 $body = str_replace("\r\n", "\n", $body);
560 $encoding = strtolower($encoding);
562 $encoding_handler = do_hook_function('decode_body', $encoding);
565 // plugins get first shot at decoding the body
567 if (!empty($encoding_handler) && function_exists($encoding_handler)) {
568 $body = $encoding_handler('decode', $body);
570 } else if ($encoding == 'quoted-printable' ||
571 $encoding == 'quoted_printable') {
572 $body = quoted_printable_decode($body);
574 while (ereg("=\n", $body)) {
575 $body = ereg_replace ("=\n", '', $body);
578 } else if ($encoding == 'base64') {
579 $body = base64_decode($body);
582 // All other encodings are returned raw.
589 * This functions decode strings that is encoded according to
590 * RFC1522 (MIME Part Two: Message Header Extensions for Non-ASCII Text).
591 * Patched by Christian Schmidt <christian@ostenfeld.dk> 23/03/2002
593 * @param string $string header string that has to be made readable
594 * @param boolean $utfencode change message in order to be readable on user's charset. defaults to true
595 * @param boolean $htmlsave preserve spaces and sanitize html special characters. defaults to true
596 * @param boolean $decide decide if string can be utfencoded. defaults to false
597 * @return string decoded header string
599 function decodeHeader ($string, $utfencode=true,$htmlsave=true,$decide=false) {
600 global $languages, $squirrelmail_language,$default_charset;
601 if (is_array($string)) {
602 $string = implode("\n", $string);
605 if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
606 function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_decodeheader')) {
607 $string = call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_decodeheader', $string);
608 // Do we need to return at this point?
615 $aString = explode(' ',$string);
617 foreach ($aString as $chunk) {
618 if ($encoded && $chunk === '') {
620 } elseif ($chunk === '') {
625 /* if encoded words are not separated by a linear-space-white we still catch them */
628 while ($match = preg_match('/^(.*)=\?([^?]*)\?(Q|B)\?([^?]*)\?=(.*)$/Ui',$chunk,$res)) {
629 /* if the last chunk isn't an encoded string then put back the space, otherwise don't */
630 if ($iLastMatch !== $j) {
640 $ret .= htmlspecialchars($res[1]);
644 $encoding = ucfirst($res[3]);
646 /* decide about valid decoding */
647 if ($decide && is_conversion_safe($res[2])) {
649 $can_be_encoded=true;
651 $can_be_encoded=false;
656 $replace = base64_decode($res[4]);
658 if ($can_be_encoded) {
659 /* convert string to different charset,
660 * if functions asks for it (usually in compose)
662 $ret .= charset_convert($res[2],$replace,$default_charset);
664 // convert string to html codes in order to display it
665 $ret .= charset_decode($res[2],$replace);
669 $replace = htmlspecialchars($replace);
675 $replace = str_replace('_', ' ', $res[4]);
676 $replace = preg_replace('/=([0-9a-f]{2})/ie', 'chr(hexdec("\1"))',
679 if ($can_be_encoded) {
680 /* convert string to different charset,
681 * if functions asks for it (usually in compose)
683 $replace = charset_convert($res[2], $replace,$default_charset);
685 // convert string to html codes in order to display it
686 $replace = charset_decode($res[2], $replace);
690 $replace = htmlspecialchars($replace);
709 if (!$encoded && $htmlsave) {
710 $ret .= htmlspecialchars($chunk);
716 /* remove the first added space */
719 $ret = substr($ret,5);
721 $ret = substr($ret,1);
729 * Encodes header as quoted-printable
731 * Encode a string according to RFC 1522 for use in headers if it
732 * contains 8-bit characters or anything that looks like it should
735 * @param string $string header string, that has to be encoded
736 * @return string quoted-printable encoded string
738 function encodeHeader ($string) {
739 global $default_charset, $languages, $squirrelmail_language;
741 if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
742 function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_encodeheader')) {
743 return call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_encodeheader', $string);
746 // Encode only if the string contains 8-bit characters or =?
747 $j = strlen($string);
748 $max_l = 75 - strlen($default_charset) - 7;
751 $iEncStart = $enc_init = false;
752 $cur_l = $iOffset = 0;
753 for($i = 0; $i < $j; ++
$i) {
762 if ($iEncStart === false) {
766 if ($cur_l > ($max_l-2)) {
767 /* if there is an stringpart that doesn't need encoding, add it */
768 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
769 $aRet[] = "=?$default_charset?Q?$ret?=";
775 $ret .= sprintf("=%02X",ord($string{$i}));
780 if ($iEncStart !== false) {
781 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
782 $aRet[] = "=?$default_charset?Q?$ret?=";
790 if ($iEncStart !== false) {
792 if ($cur_l > $max_l) {
793 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
794 $aRet[] = "=?$default_charset?Q?$ret?=";
805 $k = ord($string{$i});
807 if ($iEncStart === false) {
808 // do not start encoding in the middle of a string, also take the rest of the word.
809 $sLeadString = substr($string,0,$i);
810 $aLeadString = explode(' ',$sLeadString);
811 $sToBeEncoded = array_pop($aLeadString);
812 $iEncStart = $i - strlen($sToBeEncoded);
813 $ret .= $sToBeEncoded;
814 $cur_l +
= strlen($sToBeEncoded);
817 /* first we add the encoded string that reached it's max size */
818 if ($cur_l > ($max_l-2)) {
819 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
820 $aRet[] = "=?$default_charset?Q?$ret?= "; /* the next part is also encoded => separate by space */
827 $ret .= sprintf("=%02X", $k);
829 if ($iEncStart !== false) {
831 if ($cur_l > $max_l) {
832 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
833 $aRet[] = "=?$default_charset?Q?$ret?=";
848 if ($iEncStart !== false) {
849 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
850 $aRet[] = "=?$default_charset?Q?$ret?=";
852 $aRet[] = substr($string,$iOffset);
854 $string = implode('',$aRet);
859 /* This function trys to locate the entity_id of a specific mime element */
860 function find_ent_id($id, $message) {
861 for ($i = 0, $ret = ''; $ret == '' && $i < count($message->entities
); $i++
) {
862 if ($message->entities
[$i]->header
->type0
== 'multipart') {
863 $ret = find_ent_id($id, $message->entities
[$i]);
865 if (strcasecmp($message->entities
[$i]->header
->id
, $id) == 0) {
866 // if (sq_check_save_extension($message->entities[$i])) {
867 return $message->entities
[$i]->entity_id
;
869 } elseif (!empty($message->entities
[$i]->header
->parameters
['name'])) {
871 * This is part of a fix for Outlook Express 6.x generating
872 * cid URLs without creating content-id headers
875 if (strcasecmp($message->entities
[$i]->header
->parameters
['name'], $id) == 0) {
876 return $message->entities
[$i]->entity_id
;
884 function sq_check_save_extension($message) {
885 $filename = $message->getFilename();
886 $ext = substr($filename, strrpos($filename,'.')+
1);
887 $save_extensions = array('jpg','jpeg','gif','png','bmp');
888 return in_array($ext, $save_extensions);
893 ** HTMLFILTER ROUTINES
897 * This function is more or less a wrapper around stripslashes. Apparently
898 * Explorer is stupid enough to just remove the backslashes and then
899 * execute the content of the attribute as if nothing happened.
902 * @param attvalue The value of the attribute
903 * @return attvalue The value of the attribute stripslashed.
905 function sq_unbackslash($attvalue){
907 * Remove any backslashes. See if there are any first.
910 if (strstr($attvalue, '\\') !== false){
911 $attvalue = stripslashes($attvalue);
917 * Kill any tabs, newlines, or carriage returns. Our friends the
918 * makers of the browser with 95% market value decided that it'd
919 * be funny to make "java[tab]script" be just as good as "javascript".
921 * @param attvalue The attribute value before extraneous spaces removed.
922 * @return attvalue The attribute value after extraneous spaces removed.
924 function sq_unspace($attvalue){
925 if (strcspn($attvalue, "\t\r\n") != strlen($attvalue)){
926 $attvalue = str_replace(Array("\t", "\r", "\n"), Array('', '', ''),
933 * This function returns the final tag out of the tag name, an array
934 * of attributes, and the type of the tag. This function is called by
935 * sq_sanitize internally.
937 * @param $tagname the name of the tag.
938 * @param $attary the array of attributes and their values
939 * @param $tagtype The type of the tag (see in comments).
940 * @return a string with the final tag representation.
942 function sq_tagprint($tagname, $attary, $tagtype){
946 $fulltag = '</' . $tagname . '>';
948 $fulltag = '<' . $tagname;
949 if (is_array($attary) && sizeof($attary)){
951 while (list($attname, $attvalue) = each($attary)){
952 array_push($atts, "$attname=$attvalue");
954 $fulltag .= ' ' . join(" ", $atts);
965 * A small helper function to use with array_walk. Modifies a by-ref
966 * value and makes it lowercase.
968 * @param $val a value passed by-ref.
969 * @return void since it modifies a by-ref value.
971 function sq_casenormalize(&$val){
972 $val = strtolower($val);
976 * This function skips any whitespace from the current position within
977 * a string and to the next non-whitespace value.
979 * @param $body the string
980 * @param $offset the offset within the string where we should start
981 * looking for the next non-whitespace character.
982 * @return the location within the $body where the next
983 * non-whitespace char is located.
985 function sq_skipspace($body, $offset){
986 $me = 'sq_skipspace';
987 preg_match('/^(\s*)/s', substr($body, $offset), $matches);
988 if (sizeof($matches{1})){
989 $count = strlen($matches{1});
996 * This function looks for the next character within a string. It's
997 * really just a glorified "strpos", except it catches if failures
1000 * @param $body The string to look for needle in.
1001 * @param $offset Start looking from this position.
1002 * @param $needle The character/string to look for.
1003 * @return location of the next occurance of the needle, or
1004 * strlen($body) if needle wasn't found.
1006 function sq_findnxstr($body, $offset, $needle){
1007 $me = 'sq_findnxstr';
1008 $pos = strpos($body, $needle, $offset);
1009 if ($pos === FALSE){
1010 $pos = strlen($body);
1016 * This function takes a PCRE-style regexp and tries to match it
1017 * within the string.
1019 * @param $body The string to look for needle in.
1020 * @param $offset Start looking from here.
1021 * @param $reg A PCRE-style regex to match.
1022 * @return Returns a false if no matches found, or an array
1023 * with the following members:
1024 * - integer with the location of the match within $body
1025 * - string with whatever content between offset and the match
1026 * - string with whatever it is we matched
1028 function sq_findnxreg($body, $offset, $reg){
1029 $me = 'sq_findnxreg';
1032 preg_match("%^(.*?)($reg)%si", substr($body, $offset), $matches);
1033 if (!isset($matches{0}) ||
!$matches{0}){
1036 $retarr{0} = $offset +
strlen($matches{1});
1037 $retarr{1} = $matches{1};
1038 $retarr{2} = $matches{2};
1044 * This function looks for the next tag.
1046 * @param $body String where to look for the next tag.
1047 * @param $offset Start looking from here.
1048 * @return false if no more tags exist in the body, or
1049 * an array with the following members:
1050 * - string with the name of the tag
1051 * - array with attributes and their values
1052 * - integer with tag type (1, 2, or 3)
1053 * - integer where the tag starts (starting "<")
1054 * - integer where the tag ends (ending ">")
1055 * first three members will be false, if the tag is invalid.
1057 function sq_getnxtag($body, $offset){
1058 $me = 'sq_getnxtag';
1059 if ($offset > strlen($body)){
1062 $lt = sq_findnxstr($body, $offset, "<");
1063 if ($lt == strlen($body)){
1068 * blah blah <tag attribute="value">
1071 $pos = sq_skipspace($body, $lt+
1);
1072 if ($pos >= strlen($body)){
1073 return Array(false, false, false, $lt, strlen($body));
1076 * There are 3 kinds of tags:
1077 * 1. Opening tag, e.g.:
1079 * 2. Closing tag, e.g.:
1081 * 3. XHTML-style content-less tag, e.g.:
1082 * <img src="blah" />
1085 switch (substr($body, $pos, 1)){
1092 * A comment or an SGML declaration.
1094 if (substr($body, $pos+
1, 2) == "--"){
1095 $gt = strpos($body, "-->", $pos);
1097 $gt = strlen($body);
1101 return Array(false, false, false, $lt, $gt);
1103 $gt = sq_findnxstr($body, $pos, ">");
1104 return Array(false, false, false, $lt, $gt);
1109 * Assume tagtype 1 for now. If it's type 3, we'll switch values
1118 * Look for next [\W-_], which will indicate the end of the tag name.
1120 $regary = sq_findnxreg($body, $pos, "[^\w\-_]");
1121 if ($regary == false){
1122 return Array(false, false, false, $lt, strlen($body));
1124 list($pos, $tagname, $match) = $regary;
1125 $tagname = strtolower($tagname);
1128 * $match can be either of these:
1129 * '>' indicating the end of the tag entirely.
1130 * '\s' indicating the end of the tag name.
1131 * '/' indicating that this is type-3 xhtml tag.
1133 * Whatever else we find there indicates an invalid tag.
1138 * This is an xhtml-style tag with a closing / at the
1139 * end, like so: <img src="blah" />. Check if it's followed
1140 * by the closing bracket. If not, then this tag is invalid
1142 if (substr($body, $pos, 2) == "/>"){
1146 $gt = sq_findnxstr($body, $pos, ">");
1147 $retary = Array(false, false, false, $lt, $gt);
1151 return Array($tagname, false, $tagtype, $lt, $pos);
1155 * Check if it's whitespace
1157 if (!preg_match('/\s/', $match)){
1159 * This is an invalid tag! Look for the next closing ">".
1161 $gt = sq_findnxstr($body, $lt, ">");
1162 return Array(false, false, false, $lt, $gt);
1168 * At this point we're here:
1169 * <tagname attribute='blah'>
1172 * At this point we loop in order to find all attributes.
1177 while ($pos <= strlen($body)){
1178 $pos = sq_skipspace($body, $pos);
1179 if ($pos == strlen($body)){
1183 return Array(false, false, false, $lt, $pos);
1186 * See if we arrived at a ">" or "/>", which means that we reached
1187 * the end of the tag.
1190 if (preg_match("%^(\s*)(>|/>)%s", substr($body, $pos), $matches)) {
1194 $pos +
= strlen($matches{1});
1195 if ($matches{2} == "/>"){
1199 return Array($tagname, $attary, $tagtype, $lt, $pos);
1203 * There are several types of attributes, with optional
1204 * [:space:] between members.
1206 * attrname[:space:]=[:space:]'CDATA'
1208 * attrname[:space:]=[:space:]"CDATA"
1210 * attr[:space:]=[:space:]CDATA
1214 * We leave types 1 and 2 the same, type 3 we check for
1215 * '"' and convert to """ if needed, then wrap in
1216 * double quotes. Type 4 we convert into:
1219 $regary = sq_findnxreg($body, $pos, "[^:\w\-_]");
1220 if ($regary == false){
1222 * Looks like body ended before the end of tag.
1224 return Array(false, false, false, $lt, strlen($body));
1226 list($pos, $attname, $match) = $regary;
1227 $attname = strtolower($attname);
1229 * We arrived at the end of attribute name. Several things possible
1231 * '>' means the end of the tag and this is attribute type 4
1232 * '/' if followed by '>' means the same thing as above
1233 * '\s' means a lot of things -- look what it's followed by.
1234 * anything else means the attribute is invalid.
1239 * This is an xhtml-style tag with a closing / at the
1240 * end, like so: <img src="blah" />. Check if it's followed
1241 * by the closing bracket. If not, then this tag is invalid
1243 if (substr($body, $pos, 2) == "/>"){
1247 $gt = sq_findnxstr($body, $pos, ">");
1248 $retary = Array(false, false, false, $lt, $gt);
1252 $attary{$attname} = '"yes"';
1253 return Array($tagname, $attary, $tagtype, $lt, $pos);
1257 * Skip whitespace and see what we arrive at.
1259 $pos = sq_skipspace($body, $pos);
1260 $char = substr($body, $pos, 1);
1262 * Two things are valid here:
1263 * '=' means this is attribute type 1 2 or 3.
1264 * \w means this was attribute type 4.
1265 * anything else we ignore and re-loop. End of tag and
1266 * invalid stuff will be caught by our checks at the beginning
1271 $pos = sq_skipspace($body, $pos);
1273 * Here are 3 possibilities:
1274 * "'" attribute type 1
1275 * '"' attribute type 2
1276 * everything else is the content of tag type 3
1278 $quot = substr($body, $pos, 1);
1280 $regary = sq_findnxreg($body, $pos+
1, "\'");
1281 if ($regary == false){
1282 return Array(false, false, false, $lt, strlen($body));
1284 list($pos, $attval, $match) = $regary;
1286 $attary{$attname} = "'" . $attval . "'";
1287 } else if ($quot == '"'){
1288 $regary = sq_findnxreg($body, $pos+
1, '\"');
1289 if ($regary == false){
1290 return Array(false, false, false, $lt, strlen($body));
1292 list($pos, $attval, $match) = $regary;
1294 $attary{$attname} = '"' . $attval . '"';
1297 * These are hateful. Look for \s, or >.
1299 $regary = sq_findnxreg($body, $pos, "[\s>]");
1300 if ($regary == false){
1301 return Array(false, false, false, $lt, strlen($body));
1303 list($pos, $attval, $match) = $regary;
1305 * If it's ">" it will be caught at the top.
1307 $attval = preg_replace("/\"/s", """, $attval);
1308 $attary{$attname} = '"' . $attval . '"';
1310 } else if (preg_match("|[\w/>]|", $char)) {
1312 * That was attribute type 4.
1314 $attary{$attname} = '"yes"';
1317 * An illegal character. Find next '>' and return.
1319 $gt = sq_findnxstr($body, $pos, ">");
1320 return Array(false, false, false, $lt, $gt);
1326 * The fact that we got here indicates that the tag end was never
1327 * found. Return invalid tag indication so it gets stripped.
1329 return Array(false, false, false, $lt, strlen($body));
1333 * This function checks attribute values for entity-encoded values
1334 * and returns them translated into 8-bit strings so we can run
1337 * @param $attvalue A string to run entity check against.
1338 * @return Translated value.
1341 function sq_deent($attvalue){
1344 * See if we have to run the checks first. All entities must start
1347 if (strpos($attvalue, '&') === false){
1351 * Check named entities first.
1353 $trans = get_html_translation_table(HTML_ENTITIES
);
1355 * Leave " in, as it can mess us up.
1357 $trans = array_flip($trans);
1358 unset($trans{'"'});
1359 while (list($ent, $val) = each($trans)){
1360 $attvalue = preg_replace('/' . $ent . '*/si', $val, $attvalue);
1363 * Now translate numbered entities from 1 to 255 if needed.
1365 if (strpos($attvalue, '#') !== false){
1366 $omit = Array(34, 39);
1367 for ($asc = 256; $asc >= 0; $asc--){
1368 if (!in_array($asc, $omit)){
1370 $octrule = '/\�*' . $asc . ';*/si';
1371 $hexrule = '/\�*' . dechex($asc) . ';*/si';
1372 $attvalue = preg_replace($octrule, $chr, $attvalue);
1373 $attvalue = preg_replace($hexrule, $chr, $attvalue);
1381 * This function runs various checks against the attributes.
1383 * @param $tagname String with the name of the tag.
1384 * @param $attary Array with all tag attributes.
1385 * @param $rm_attnames See description for sq_sanitize
1386 * @param $bad_attvals See description for sq_sanitize
1387 * @param $add_attr_to_tag See description for sq_sanitize
1388 * @param $message message object
1389 * @param $id message id
1390 * @return Array with modified attributes.
1392 function sq_fixatts($tagname,
1402 while (list($attname, $attvalue) = each($attary)){
1404 * See if this attribute should be removed.
1406 foreach ($rm_attnames as $matchtag=>$matchattrs){
1407 if (preg_match($matchtag, $tagname)){
1408 foreach ($matchattrs as $matchattr){
1409 if (preg_match($matchattr, $attname)){
1410 unset($attary{$attname});
1417 * Remove any backslashes, entities, and extraneous whitespace.
1419 $attvalue = sq_unbackslash($attvalue);
1420 $attvalue = sq_deent($attvalue);
1421 $attvalue = sq_unspace($attvalue);
1424 * Remove \r \n \t \0 " " "\\"
1426 $attvalue = str_replace(Array("\r", "\n", "\t", "\0", " ", "\\"),
1427 Array('', '','','','',''), $attvalue);
1430 * Now let's run checks on the attvalues.
1431 * I don't expect anyone to comprehend this. If you do,
1432 * get in touch with me so I can drive to where you live and
1433 * shake your hand personally. :)
1435 foreach ($bad_attvals as $matchtag=>$matchattrs){
1436 if (preg_match($matchtag, $tagname)){
1437 foreach ($matchattrs as $matchattr=>$valary){
1438 if (preg_match($matchattr, $attname)){
1440 * There are two arrays in valary.
1442 * Second one is replacements
1444 list($valmatch, $valrepl) = $valary;
1446 preg_replace($valmatch, $valrepl, $attvalue);
1447 if ($newvalue != $attvalue){
1448 $attary{$attname} = $newvalue;
1455 * Turn cid: urls into http-friendly ones.
1457 if (preg_match("/^[\'\"]\s*cid:/si", $attvalue)){
1458 $attary{$attname} = sq_cid2http($message, $id, $attvalue, $mailbox);
1462 * See if we need to append any attributes to this tag.
1464 foreach ($add_attr_to_tag as $matchtag=>$addattary){
1465 if (preg_match($matchtag, $tagname)){
1466 $attary = array_merge($attary, $addattary);
1473 * This function edits the style definition to make them friendly and
1474 * usable in SquirrelMail.
1476 * @param $message the message object
1477 * @param $id the message id
1478 * @param $content a string with whatever is between <style> and </style>
1479 * @param $mailbox the message mailbox
1480 * @return a string with edited content.
1482 function sq_fixstyle($body, $pos, $message, $id, $mailbox){
1483 global $view_unsafe_images;
1484 $me = 'sq_fixstyle';
1485 $ret = sq_findnxreg($body, $pos, '</\s*style\s*>');
1487 return array(FALSE, strlen($body));
1489 $newpos = $ret[0] +
strlen($ret[2]);
1492 * First look for general BODY style declaration, which would be
1494 * body {background: blah-blah}
1495 * and change it to .bodyclass so we can just assign it to a <div>
1497 $content = preg_replace("|body(\s*\{.*?\})|si", ".bodyclass\\1", $content);
1498 $secremoveimg = '../images/' . _("sec_remove_eng.png");
1500 * Fix url('blah') declarations.
1502 $content = preg_replace("|url\s*\(\s*([\'\"])\s*\S+script\s*:.*?([\'\"])\s*\)|si",
1503 "url(\\1$secremoveimg\\2)", $content);
1505 * Fix url('https*://.*) declarations but only if $view_unsafe_images
1508 if (!$view_unsafe_images){
1509 $content = preg_replace("|url\s*\(\s*([\'\"])\s*https*:.*?([\'\"])\s*\)|si",
1510 "url(\\1$secremoveimg\\2)", $content);
1514 * Fix urls that refer to cid:
1516 while (preg_match("|url\s*\(\s*([\'\"]\s*cid:.*?[\'\"])\s*\)|si",
1517 $content, $matches)){
1518 $cidurl = $matches{1};
1519 $httpurl = sq_cid2http($message, $id, $cidurl, $mailbox);
1520 $content = preg_replace("|url\s*\(\s*$cidurl\s*\)|si",
1521 "url($httpurl)", $content);
1525 * Fix stupid css declarations which lead to vulnerabilities
1528 $match = Array('/expression/i',
1531 '/include-source/i');
1532 $replace = Array('idiocy', 'idiocy', 'idiocy', 'idiocy');
1533 $content = preg_replace($match, $replace, $content);
1534 return array($content, $newpos);
1538 * This function converts cid: url's into the ones that can be viewed in
1541 * @param $message the message object
1542 * @param $id the message id
1543 * @param $cidurl the cid: url.
1544 * @param $mailbox the message mailbox
1545 * @return a string with a http-friendly url
1547 function sq_cid2http($message, $id, $cidurl, $mailbox){
1549 * Get rid of quotes.
1551 $quotchar = substr($cidurl, 0, 1);
1552 if ($quotchar == '"' ||
$quotchar == "'"){
1553 $cidurl = str_replace($quotchar, "", $cidurl);
1557 $cidurl = substr(trim($cidurl), 4);
1558 $linkurl = find_ent_id($cidurl, $message);
1559 /* in case of non-save cid links $httpurl should be replaced by a sort of
1560 unsave link image */
1564 * This is part of a fix for Outlook Express 6.x generating
1565 * cid URLs without creating content-id headers. These images are
1566 * not part of the multipart/related html mail. The html contains
1567 * <img src="cid:{some_id}/image_filename.ext"> references to
1568 * attached images with as goal to render them inline although
1569 * the attachment disposition property is not inline.
1572 if (empty($linkurl)) {
1573 if (preg_match('/{.*}\//', $cidurl)) {
1574 $cidurl = preg_replace('/{.*}\//','', $cidurl);
1575 if (!empty($cidurl)) {
1576 $linkurl = find_ent_id($cidurl, $message);
1581 if (!empty($linkurl)) {
1582 $httpurl = $quotchar . SM_PATH
. 'src/download.php?absolute_dl=true&' .
1583 "passed_id=$id&mailbox=" . urlencode($mailbox) .
1584 '&ent_id=' . $linkurl . $quotchar;
1587 * If we couldn't generate a proper img url, drop in a blank image
1588 * instead of sending back empty, otherwise it causes unusual behaviour
1590 $httpurl = $quotechar . SM_PATH
. 'images/blank.png';
1597 * This function changes the <body> tag into a <div> tag since we
1598 * can't really have a body-within-body.
1600 * @param $attary an array of attributes and values of <body>
1601 * @param $mailbox mailbox we're currently reading (for cid2http)
1602 * @param $message current message (for cid2http)
1603 * @param $id current message id (for cid2http)
1604 * @return a modified array of attributes to be set for <div>
1606 function sq_body2div($attary, $mailbox, $message, $id){
1607 $me = 'sq_body2div';
1608 $divattary = Array('class' => "'bodyclass'");
1610 $has_bgc_stl = $has_txt_stl = false;
1612 if (is_array($attary) && sizeof($attary) > 0){
1613 foreach ($attary as $attname=>$attvalue){
1614 $quotchar = substr($attvalue, 0, 1);
1615 $attvalue = str_replace($quotchar, "", $attvalue);
1618 $attvalue = sq_cid2http($message, $id,
1619 $attvalue, $mailbox);
1620 $styledef .= "background-image: url('$attvalue'); ";
1623 $has_bgc_stl = true;
1624 $styledef .= "background-color: $attvalue; ";
1627 $has_txt_stl = true;
1628 $styledef .= "color: $attvalue; ";
1632 // Outlook defines a white bgcolor and no text color. This can lead to
1633 // white text on a white bg with certain themes.
1634 if ($has_bgc_stl && !$has_txt_stl) {
1635 $styledef .= "color: $text; ";
1637 if (strlen($styledef) > 0){
1638 $divattary{"style"} = "\"$styledef\"";
1645 * This is the main function and the one you should actually be calling.
1646 * There are several variables you should be aware of an which need
1647 * special description.
1649 * Since the description is quite lengthy, see it here:
1650 * http://linux.duke.edu/projects/mini/htmlfilter/
1652 * @param $body the string with HTML you wish to filter
1653 * @param $tag_list see description above
1654 * @param $rm_tags_with_content see description above
1655 * @param $self_closing_tags see description above
1656 * @param $force_tag_closing see description above
1657 * @param $rm_attnames see description above
1658 * @param $bad_attvals see description above
1659 * @param $add_attr_to_tag see description above
1660 * @param $message message object
1661 * @param $id message id
1662 * @return sanitized html safe to show on your pages.
1664 function sq_sanitize($body,
1666 $rm_tags_with_content,
1676 $me = 'sq_sanitize';
1677 $rm_tags = array_shift($tag_list);
1679 * Normalize rm_tags and rm_tags_with_content.
1681 @array_walk
($tag_list, 'sq_casenormalize');
1682 @array_walk
($rm_tags_with_content, 'sq_casenormalize');
1683 @array_walk
($self_closing_tags, 'sq_casenormalize');
1685 * See if tag_list is of tags to remove or tags to allow.
1686 * false means remove these tags
1687 * true means allow these tags
1690 $open_tags = Array();
1691 $trusted = "\n<!-- begin sanitized html -->\n";
1692 $skip_content = false;
1694 * Take care of netscape's stupid javascript entities like
1697 $body = preg_replace("/&(\{.*?\};)/si", "&\\1", $body);
1699 while (($curtag = sq_getnxtag($body, $curpos)) != FALSE){
1700 list($tagname, $attary, $tagtype, $lt, $gt) = $curtag;
1701 $free_content = substr($body, $curpos, $lt-$curpos);
1703 * Take care of <style>
1705 if ($tagname == "style" && $tagtype == 1){
1706 list($free_content, $curpos) =
1707 sq_fixstyle($body, $gt+
1, $message, $id, $mailbox);
1708 if ($free_content != FALSE){
1709 $trusted .= sq_tagprint($tagname, $attary, $tagtype);
1710 $trusted .= $free_content;
1711 $trusted .= sq_tagprint($tagname, false, 2);
1715 if ($skip_content == false){
1716 $trusted .= $free_content;
1718 if ($tagname != FALSE){
1720 if ($skip_content == $tagname){
1722 * Got to the end of tag we needed to remove.
1725 $skip_content = false;
1727 if ($skip_content == false){
1728 if ($tagname == "body"){
1731 if (isset($open_tags{$tagname}) &&
1732 $open_tags{$tagname} > 0){
1733 $open_tags{$tagname}--;
1741 * $rm_tags_with_content
1743 if ($skip_content == false){
1745 * See if this is a self-closing type and change
1746 * tagtype appropriately.
1749 && in_array($tagname, $self_closing_tags)){
1753 * See if we should skip this tag and any content
1756 if ($tagtype == 1 &&
1757 in_array($tagname, $rm_tags_with_content)){
1758 $skip_content = $tagname;
1760 if (($rm_tags == false
1761 && in_array($tagname, $tag_list)) ||
1762 ($rm_tags == true &&
1763 !in_array($tagname, $tag_list))){
1767 * Convert body into div.
1769 if ($tagname == "body"){
1771 $attary = sq_body2div($attary, $mailbox,
1775 if (isset($open_tags{$tagname})){
1776 $open_tags{$tagname}++
;
1778 $open_tags{$tagname}=1;
1782 * This is where we run other checks.
1784 if (is_array($attary) && sizeof($attary) > 0){
1785 $attary = sq_fixatts($tagname,
1799 if ($tagname != false && $skip_content == false){
1800 $trusted .= sq_tagprint($tagname, $attary, $tagtype);
1805 $trusted .= substr($body, $curpos, strlen($body)-$curpos);
1806 if ($force_tag_closing == true){
1807 foreach ($open_tags as $tagname=>$opentimes){
1808 while ($opentimes > 0){
1809 $trusted .= '</' . $tagname . '>';
1815 $trusted .= "<!-- end sanitized html -->\n";
1820 * This is a wrapper function to call html sanitizing routines.
1822 * @param $body the body of the message
1823 * @param $id the id of the message
1824 * @return a string with html safe to display in the browser.
1826 function magicHTML($body, $id, $message, $mailbox = 'INBOX') {
1827 global $attachment_common_show_images, $view_unsafe_images,
1830 * Don't display attached images in HTML mode.
1832 $attachment_common_show_images = false;
1847 $rm_tags_with_content = Array(
1856 $self_closing_tags = Array(
1863 $force_tag_closing = true;
1865 $rm_attnames = Array(
1876 $secremoveimg = "../images/" . _("sec_remove_eng.png");
1877 $bad_attvals = Array(
1880 "/^src|background/i" =>
1883 "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si",
1884 "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si",
1885 "/^([\'\"])\s*about\s*:.*([\'\"])/si"
1888 "\\1$secremoveimg\\2",
1889 "\\1$secremoveimg\\2",
1890 "\\1$secremoveimg\\2",
1891 "\\1$secremoveimg\\2"
1894 "/^href|action/i" =>
1897 "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si",
1898 "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si",
1899 "/^([\'\"])\s*about\s*:.*([\'\"])/si"
1914 "/include-source/i",
1915 "/url\s*\(\s*([\'\"])\s*\S+script\s*:.*([\'\"])\s*\)/si",
1916 "/url\s*\(\s*([\'\"])\s*mocha\s*:.*([\'\"])\s*\)/si",
1917 "/url\s*\(\s*([\'\"])\s*about\s*:.*([\'\"])\s*\)/si",
1918 "/(.*)\s*:\s*url\s*\(\s*([\'\"]*)\s*\S+script\s*:.*([\'\"]*)\s*\)/si"
1934 if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET
) ) {
1935 $view_unsafe_images = false;
1937 if (!$view_unsafe_images){
1939 * Remove any references to http/https if view_unsafe_images set
1942 array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[0],
1943 '/^([\'\"])\s*https*:.*([\'\"])/si');
1944 array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[1],
1945 "\\1$secremoveimg\\1");
1946 array_push($bad_attvals{'/.*/'}{'/^style/i'}[0],
1947 '/url\(([\'\"])\s*https*:.*([\'\"])\)/si');
1948 array_push($bad_attvals{'/.*/'}{'/^style/i'}[1],
1949 "url(\\1$secremoveimg\\1)");
1952 $add_attr_to_tag = Array(
1954 Array('target'=>'"_new"',
1955 'title'=>'"'._("This external link will open in a new window").'"'
1958 $trusted = sq_sanitize($body,
1960 $rm_tags_with_content,
1970 if (preg_match("|$secremoveimg|i", $trusted)){
1971 $has_unsafe_images = true;
1977 * function SendDownloadHeaders - send file to the browser
1979 * Original Source: SM core src/download.php
1980 * moved here to make it available to other code, and separate
1981 * front end from back end functionality.
1983 * @param string $type0 first half of mime type
1984 * @param string $type1 second half of mime type
1985 * @param string $filename filename to tell the browser for downloaded file
1986 * @param boolean $force whether to force the download dialog to pop
1987 * @param optional integer $filesize send the Content-Header and length to the browser
1990 function SendDownloadHeaders($type0, $type1, $filename, $force, $filesize=0) {
1991 global $languages, $squirrelmail_language;
1994 sqgetGlobalVar('HTTP_USER_AGENT', $HTTP_USER_AGENT, SQ_SERVER
);
1996 if (strstr($HTTP_USER_AGENT, 'compatible; MSIE ') !== false &&
1997 strstr($HTTP_USER_AGENT, 'Opera') === false) {
2001 if (strstr($HTTP_USER_AGENT, 'compatible; MSIE 6') !== false &&
2002 strstr($HTTP_USER_AGENT, 'Opera') === false) {
2006 if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
2007 function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_downloadfilename')) {
2009 call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_downloadfilename', $filename, $HTTP_USER_AGENT);
2011 $filename = ereg_replace('[\\/:\*\?"<>\|;]', '_', str_replace(' ', ' ', $filename));
2014 // A Pox on Microsoft and it's Internet Explorer!
2016 // IE has lots of bugs with file downloads.
2017 // It also has problems with SSL. Both of these cause problems
2018 // for us in this function.
2020 // See this article on Cache Control headers and SSL
2021 // http://support.microsoft.com/default.aspx?scid=kb;en-us;323308
2023 // The best thing you can do for IE is to upgrade to the latest
2025 //set all the Cache Control Headers for IE
2027 $filename=rawurlencode($filename);
2028 header ("Pragma: public");
2029 header ("Cache-Control: no-store, max-age=0, no-cache, must-revalidate"); # HTTP/1.1
2030 header ("Cache-Control: post-check=0, pre-check=0", false);
2031 header ("Cache-control: private");
2033 //set the inline header for IE, we'll add the attachment header later if we need it
2034 header ("Content-Disposition: inline; filename=$filename");
2038 // Try to show in browser window
2039 header ("Content-Disposition: inline; filename=\"$filename\"");
2040 header ("Content-Type: $type0/$type1; name=\"$filename\"");
2042 // Try to pop up the "save as" box
2044 // IE makes this hard. It pops up 2 save boxes, or none.
2045 // http://support.microsoft.com/support/kb/articles/Q238/5/88.ASP
2046 // http://support.microsoft.com/default.aspx?scid=kb;EN-US;260519
2047 // But, according to Microsoft, it is "RFC compliant but doesn't
2048 // take into account some deviations that allowed within the
2049 // specification." Doesn't that mean RFC non-compliant?
2050 // http://support.microsoft.com/support/kb/articles/Q258/4/52.ASP
2052 // all browsers need the application/octet-stream header for this
2053 header ("Content-Type: application/octet-stream; name=\"$filename\"");
2055 // http://support.microsoft.com/support/kb/articles/Q182/3/15.asp
2056 // Do not have quotes around filename, but that applied to
2057 // "attachment"... does it apply to inline too?
2058 header ("Content-Disposition: attachment; filename=\"$filename\"");
2060 if ($isIE && !$isIE6) {
2061 // This combination seems to work mostly. IE 5.5 SP 1 has
2062 // known issues (see the Microsoft Knowledge Base)
2064 // This works for most types, but doesn't work with Word files
2065 header ("Content-Type: application/download; name=\"$filename\"");
2067 // These are spares, just in case. :-)
2068 //header("Content-Type: $type0/$type1; name=\"$filename\"");
2069 //header("Content-Type: application/x-msdownload; name=\"$filename\"");
2070 //header("Content-Type: application/octet-stream; name=\"$filename\"");
2072 // another application/octet-stream forces download for Netscape
2073 header ("Content-Type: application/octet-stream; name=\"$filename\"");
2077 //send the content-length header if the calling function provides it
2078 if ($filesize > 0) {
2079 header("Content-Length: $filesize");
2082 } // end fn SendDownloadHeaders