6 * Copyright (c) 1999-2003 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
15 require_once(SM_PATH
. 'functions/imap.php');
16 require_once(SM_PATH
. 'functions/attachment_common.php');
18 /* --------------------------------------------------------------------------------- */
20 /* --------------------------------------------------------------------------------- */
22 /* This function gets the structure of a message and stores it in the "message" class.
23 * It will return this object for use with all relevant header information and
24 * fully parsed into the standard "message" object format.
27 function mime_structure ($bodystructure, $flags=array()) {
29 /* Isolate the body structure and remove beginning and end parenthesis. */
30 $read = trim(substr ($bodystructure, strpos(strtolower($bodystructure), 'bodystructure') +
13));
31 $read = trim(substr ($read, 0, -1));
33 $msg = Message
::parseStructure($read,$i);
34 if (!is_object($msg)) {
35 include_once(SM_PATH
. 'functions/display_messages.php');
36 global $color, $mailbox;
37 /* removed urldecode because $_GET is auto urldecoded ??? */
38 displayPageHeader( $color, $mailbox );
39 echo "<BODY TEXT=\"$color[8]\" BGCOLOR=\"$color[4]\" LINK=\"$color[7]\" VLINK=\"$color[7]\" ALINK=\"$color[7]\">\n\n" .
41 $errormessage = _("SquirrelMail could not decode the bodystructure of the message");
42 $errormessage .= '<BR>'._("the provided bodystructure by your imap-server").':<BR><BR>';
43 $errormessage .= '<table><tr><td>' . htmlspecialchars($read) . '</td></tr></table>';
44 plain_error_message( $errormessage, $color );
45 echo '</body></html>';
49 foreach ($flags as $flag) {
50 $char = strtoupper($flag{1});
53 if (strtolower($flag) == '\\seen') {
58 if (strtolower($flag) == '\\answered') {
59 $msg->is_answered
= true;
63 if (strtolower($flag) == '\\deleted') {
64 $msg->is_deleted
= true;
68 if (strtolower($flag) == '\\flagged') {
69 $msg->is_flagged
= true;
73 if (strtolower($flag) == '$mdnsent') {
74 $msg->is_mdnsent
= true;
82 // listEntities($msg);
88 /* This starts the parsing of a particular structure. It is called recursively,
89 * so it can be passed different structures. It returns an object of type
91 * First, it checks to see if it is a multipart message. If it is, then it
92 * handles that as it sees is necessary. If it is just a regular entity,
93 * then it parses it and adds the necessary header information (by calling out
94 * to mime_get_elements()
97 function mime_fetch_body($imap_stream, $id, $ent_id=1) {
99 /* Do a bit of error correction. If we couldn't find the entity id, just guess
100 * that it is the first one. That is usually the case anyway.
103 $cmd = "FETCH $id BODY[]";
105 $cmd = "FETCH $id BODY[$ent_id]";
108 $data = sqimap_run_command ($imap_stream, $cmd, true, $response, $message, $uid_support);
110 $topline = trim(array_shift($data));
111 } while($topline && ($topline[0] == '*') && !preg_match('/\* [0-9]+ FETCH.*/i', $topline)) ;
113 $wholemessage = implode('', $data);
114 if (ereg('\\{([^\\}]*)\\}', $topline, $regs)) {
115 $ret = substr($wholemessage, 0, $regs[1]);
116 /* There is some information in the content info header that could be important
117 * in order to parse html messages. Let's get them here.
119 // if ($ret{0} == '<') {
120 // $data = sqimap_run_command ($imap_stream, "FETCH $id BODY[$ent_id.MIME]", true, $response, $message, $uid_support);
122 } else if (ereg('"([^"]*)"', $topline, $regs)) {
125 global $where, $what, $mailbox, $passed_id, $startMessage;
126 $par = 'mailbox=' . urlencode($mailbox) . '&passed_id=' . $passed_id;
127 if (isset($where) && isset($what)) {
128 $par .= '&where=' . urlencode($where) . '&what=' . urlencode($what);
130 $par .= '&startMessage=' . $startMessage . '&show_more=0';
132 $par .= '&response=' . urlencode($response) .
133 '&message=' . urlencode($message) .
134 '&topline=' . urlencode($topline);
137 '<table width="80%"><tr>' .
138 '<tr><td colspan=2>' .
139 _("Body retrieval error. The reason for this is most probably that the message is malformed.") .
141 '<tr><td><b>' . _("Command:") . "</td><td>$cmd</td></tr>" .
142 '<tr><td><b>' . _("Response:") . "</td><td>$response</td></tr>" .
143 '<tr><td><b>' . _("Message:") . "</td><td>$message</td></tr>" .
144 '<tr><td><b>' . _("FETCH line:") . "</td><td>$topline</td></tr>" .
145 "</table><BR></tt></font><hr>";
147 $data = sqimap_run_command ($imap_stream, "FETCH $passed_id BODY[]", true, $response, $message, $uid_support);
149 $wholemessage = implode('', $data);
151 $ret = $wholemessage;
156 function mime_print_body_lines ($imap_stream, $id, $ent_id=1, $encoding) {
159 $sid = sqimap_session_id($uid_support);
160 /* Don't kill the connection if the browser is over a dialup
161 * and it would take over 30 seconds to download it.
162 * DonĀ“t call set_time_limit in safe mode.
165 if (!ini_get('safe_mode')) {
169 $sid_s = substr($sid,0,strpos($sid, ' '));
174 $body = mime_fetch_body ($imap_stream, $id, $ent_id);
175 echo decodeBody($body, $encoding);
178 fputs ($imap_stream, "$sid FETCH $id BODY[$ent_id]\r\n");
181 $read = fgets ($imap_stream,8192);
184 // This could be bad -- if the section has sqimap_session_id() . ' OK'
185 // or similar, it will kill the download.
186 while (!ereg("^".$sid_s." (OK|BAD|NO)(.*)$", $read, $regs)) {
187 if (trim($read) == ')==') {
189 $read = fgets ($imap_stream,4096);
190 if (ereg("^".$sid." (OK|BAD|NO)(.*)$", $read, $regs)) {
193 echo decodeBody($read1, $encoding) .
194 decodeBody($read, $encoding);
197 echo decodeBody($read, $encoding);
199 $read = fgets ($imap_stream,4096);
206 /* -[ END MIME DECODING ]----------------------------------------------------------- */
208 /* This is here for debugging purposes. It will print out a list
209 * of all the entity IDs that are in the $message object.
211 function listEntities ($message) {
213 echo "<tt>" . $message->entity_id
. ' : ' . $message->type0
. '/' . $message->type1
. ' parent = '. $message->parent
->entity_id
. '<br>';
214 for ($i = 0; isset($message->entities
[$i]); $i++
) {
216 $msg = listEntities($message->entities
[$i]);
226 function getPriorityStr($priority) {
227 $priority_level = substr($priority,0,1);
229 switch($priority_level) {
230 /* Check for a higher then normal priority. */
233 $priority_string = _("High");
236 /* Check for a lower then normal priority. */
239 $priority_string = _("Low");
242 /* Check for a normal priority. */
245 $priority_level = '3';
246 $priority_string = _("Normal");
250 return $priority_string;
253 /* returns a $message object for a particular entity id */
254 function getEntity ($message, $ent_id) {
255 return $message->getEntity($ent_id);
259 * Extracted from strings.php 23/03/2002
262 function translateText(&$body, $wrap_at, $charset) {
263 global $where, $what; /* from searching */
264 global $color; /* color theme */
266 require_once(SM_PATH
. 'functions/url_parser.php');
268 $body_ary = explode("\n", $body);
269 for ($i=0; $i < count($body_ary); $i++
) {
270 $line = $body_ary[$i];
271 if (strlen($line) - 2 >= $wrap_at) {
272 sqWordWrap($line, $wrap_at);
274 $line = charset_decode($charset, $line);
275 $line = str_replace("\t", ' ', $line);
284 if ($line[$pos] == ' ') {
286 } else if (strpos($line, '>', $pos) === $pos) {
295 if (!isset($color[13])) {
296 $color[13] = '#800000';
298 $line = '<font color="' . $color[13] . '">' . $line . '</font>';
300 if (!isset($color[14])) {
301 $color[14] = '#FF0000';
303 $line = '<font color="' . $color[14] . '">' . $line . '</font>';
306 $body_ary[$i] = $line;
308 $body = '<pre>' . implode("\n", $body_ary) . '</pre>';
311 /* This returns a parsed string called $body. That string can then
312 * be displayed as the actual message in the HTML. It contains
313 * everything needed, including HTML Tags, Attachments at the
316 function formatBody($imap_stream, $message, $color, $wrap_at, $ent_num, $id, $mailbox='INBOX') {
317 /* This if statement checks for the entity to show as the
318 * primary message. To add more of them, just put them in the
319 * order that is their priority.
321 global $startMessage, $username, $key, $imapServerAddress, $imapPort,
322 $show_html_default, $sort, $has_unsafe_images, $passed_ent_id;
323 global $languages, $squirrelmail_language;
325 if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET
) ) {
326 $view_unsafe_images = false;
330 $urlmailbox = urlencode($mailbox);
331 $body_message = getEntity($message, $ent_num);
332 if (($body_message->header
->type0
== 'text') ||
333 ($body_message->header
->type0
== 'rfc822')) {
334 $body = mime_fetch_body ($imap_stream, $id, $ent_num);
335 $body = decodeBody($body, $body_message->header
->encoding
);
337 if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
338 function_exists($languages[$squirrelmail_language]['XTRA_CODE'])) {
339 if (mb_detect_encoding($body) != 'ASCII') {
340 $body = $languages[$squirrelmail_language]['XTRA_CODE']('decode', $body);
343 $hookResults = do_hook("message_body", $body);
344 $body = $hookResults[1];
346 /* If there are other types that shouldn't be formatted, add
350 if ($body_message->header
->type1
== 'html') {
351 if ($show_html_default <> 1) {
352 $entity_conv = array(' ' => ' ',
359 $body = strtr($body, $entity_conv);
360 $body = strip_tags($body);
362 translateText($body, $wrap_at,
363 $body_message->header
->getParameter('charset'));
365 $body = magicHTML($body, $id, $message, $mailbox);
368 translateText($body, $wrap_at,
369 $body_message->header
->getParameter('charset'));
371 $link = 'read_body.php?passed_id=' . $id . '&ent_id='.$ent_num.
372 '&mailbox=' . $urlmailbox .'&sort=' . $sort .
373 '&startMessage=' . $startMessage . '&show_more=0';
374 if (isset($passed_ent_id)) {
375 $link .= '&passed_ent_id='.$passed_ent_id;
377 if ($view_unsafe_images) {
378 $text = _("Hide Unsafe Images");
380 if (isset($has_unsafe_images) && $has_unsafe_images) {
381 $link .= '&view_unsafe_images=1';
382 $text = _("View Unsafe Images");
387 $body .= '<center><small><a href="'.$link.'">'.$text.
388 '</a></small></center><br>' . "\n";
394 function formatAttachments($message, $exclude_id, $mailbox, $id) {
395 global $where, $what, $startMessage, $color;
396 static $ShownHTML = 0;
398 $att_ar = $message->getAttachments($exclude_id);
400 if (!count($att_ar)) return '';
404 $urlMailbox = urlencode($mailbox);
406 foreach ($att_ar as $att) {
407 $ent = $att->entity_id
;
408 $header = $att->header
;
409 $type0 = strtolower($header->type0
);
410 $type1 = strtolower($header->type1
);
412 $links['download link']['text'] = _("download");
413 $links['download link']['href'] = SM_PATH
.
414 "src/download.php?absolute_dl=true&passed_id=$id&mailbox=$urlMailbox&ent_id=$ent";
416 if ($type0 =='message' && $type1 == 'rfc822') {
417 $default_page = SM_PATH
. 'src/read_body.php';
418 $rfc822_header = $att->rfc822_header
;
419 $filename = $rfc822_header->subject
;
420 if (trim( $filename ) == '') {
421 $filename = 'untitled-[' . $ent . ']' ;
423 $from_o = $rfc822_header->from
;
424 if (is_object($from_o)) {
425 $from_name = $from_o->getAddress(false);
427 $from_name = _("Unknown sender");
429 $from_name = decodeHeader(($from_name));
430 $description = $from_name;
432 $default_page = SM_PATH
. 'src/download.php';
433 if (is_object($header->disposition
)) {
434 $filename = $header->disposition
->getProperty('filename');
435 if (trim($filename) == '') {
436 $name = decodeHeader($header->disposition
->getProperty('name'));
437 if (trim($name) == '') {
438 $name = $header->getParameter('name');
439 if(trim($name) == '') {
440 if (trim( $header->id
) == '') {
441 $filename = 'untitled-[' . $ent . ']' ;
443 $filename = 'cid: ' . $header->id
;
453 $filename = $header->getParameter('name');
454 if (!trim($filename)) {
455 if (trim( $header->id
) == '') {
456 $filename = 'untitled-[' . $ent . ']' ;
458 $filename = 'cid: ' . $header->id
;
462 if ($header->description
) {
463 $description = decodeHeader($header->description
);
469 $display_filename = $filename;
470 if (isset($passed_ent_id)) {
471 $passed_ent_id_link = '&passed_ent_id='.$passed_ent_id;
473 $passed_ent_id_link = '';
475 $defaultlink = $default_page . "?startMessage=$startMessage"
476 . "&passed_id=$id&mailbox=$urlMailbox"
477 . '&ent_id='.$ent.$passed_ent_id_link;
478 if ($where && $what) {
479 $defaultlink .= '&where='. urlencode($where).'&what='.urlencode($what);
481 /* This executes the attachment hook with a specific MIME-type.
482 * If that doesn't have results, it tries if there's a rule
483 * for a more generic type.
485 $hookresults = do_hook("attachment $type0/$type1", $links,
486 $startMessage, $id, $urlMailbox, $ent, $defaultlink,
487 $display_filename, $where, $what);
488 if(count($hookresults[1]) <= 1) {
489 $hookresults = do_hook("attachment $type0/*", $links,
490 $startMessage, $id, $urlMailbox, $ent, $defaultlink,
491 $display_filename, $where, $what);
494 $links = $hookresults[1];
495 $defaultlink = $hookresults[6];
497 $attachments .= '<TR><TD>' .
498 '<A HREF="'.$defaultlink.'">'.decodeHeader($display_filename).'</A> </TD>' .
499 '<TD><SMALL><b>' . show_readable_size($header->size
) .
500 '</b> </small></TD>' .
501 "<TD><SMALL>[ $type0/$type1 ] </SMALL></TD>" .
503 $attachments .= '<b>' . $description . '</b>';
504 $attachments .= '</SMALL></TD><TD><SMALL> ';
507 foreach ($links as $val) {
511 $attachments .= ' | ';
513 $attachments .= '<a href="' . $val['href'] . '">' . $val['text'] . '</a>';
516 $attachments .= "</TD></TR>\n";
521 /* This function decodes the body depending on the encoding type. */
522 function decodeBody($body, $encoding) {
523 global $show_html_default;
525 $body = str_replace("\r\n", "\n", $body);
526 $encoding = strtolower($encoding);
528 if ($encoding == 'quoted-printable' ||
529 $encoding == 'quoted_printable') {
530 $body = quoted_printable_decode($body);
532 while (ereg("=\n", $body)) {
533 $body = ereg_replace ("=\n", '', $body);
536 } else if ($encoding == 'base64') {
537 $body = base64_decode($body);
540 // All other encodings are returned raw.
545 * This functions decode strings that is encoded according to
546 * RFC1522 (MIME Part Two: Message Header Extensions for Non-ASCII Text).
547 * Patched by Christian Schmidt <christian@ostenfeld.dk> 23/03/2002
549 function decodeHeader ($string, $utfencode=true,$htmlsave=true) {
550 global $languages, $squirrelmail_language;
551 if (is_array($string)) {
552 $string = implode("\n", $string);
555 if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
556 function_exists($languages[$squirrelmail_language]['XTRA_CODE'])) {
557 $string = $languages[$squirrelmail_language]['XTRA_CODE']('decodeheader', $string);
558 // Do we need to return at this point?
565 $aString = explode(' ',$string);
567 foreach ($aString as $chunk) {
568 if ($encoded && !$chunk) {
575 /* if encoded words are not separated by a linear-space-white we still catch them */
577 if ($chunk{0} === '=') { /* performance, saves an unnessecarry preg call */
578 while ($match = preg_match('/^(.*)=\?([^?]*)\?(Q|B)\?([^?]*)\?=(.*)$/Ui',$chunk,$res)) {
579 /* if the last chunk isn't an encoded string then put back the space, otherwise don't */
580 if ($iLastMatch !== $j) {
590 $encoding = ucfirst($res[3]);
594 $replace = base64_decode($res[4]);
595 $ret .= charset_decode($res[2],$replace);
598 $replace = str_replace('_', ' ', $res[4]);
599 $replace = preg_replace('/=([0-9a-f]{2})/ie', 'chr(hexdec("\1"))',
601 /* Only encode into entities by default. Some places
602 * don't need the encoding, like the compose form.
605 $replace = charset_decode($res[2], $replace);
608 $replace = htmlspecialchars($replace);
628 if (!$encoded && $htmlsave) {
629 $ret .= htmlspecialchars($chunk);
635 /* remove the first added space */
638 $ret = substr($ret,6);
640 $ret = substr($ret,1);
648 * Encode a string according to RFC 1522 for use in headers if it
649 * contains 8-bit characters or anything that looks like it should
652 function encodeHeader ($string) {
653 global $default_charset, $languages, $squirrelmail_language;
655 if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
656 function_exists($languages[$squirrelmail_language]['XTRA_CODE'])) {
657 return $languages[$squirrelmail_language]['XTRA_CODE']('encodeheader', $string);
659 if (strtolower($default_charset) == 'iso-8859-1') {
660 $string = str_replace("\240",' ',$string);
663 // Encode only if the string contains 8-bit characters or =?
664 $j = strlen($string);
665 $max_l = 75 - strlen($default_charset) - 7;
668 $iEncStart = $enc_init = false;
669 $cur_l = $iOffset = 0;
670 for($i = 0; $i < $j; ++
$i) {
679 if ($iEncStart === false) {
683 if ($cur_l > ($max_l-2)) {
684 /* if there is an stringpart that doesn't need encoding, add it */
685 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
686 $aRet[] = "=?$default_charset?Q?$ret?=";
692 $ret .= sprintf("=%02X",ord($string{$i}));
697 if ($iEncStart !== false) {
698 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
699 $aRet[] = "=?$default_charset?Q?$ret?=";
707 if ($iEncStart !== false) {
709 if ($cur_l > $max_l) {
710 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
711 $aRet[] = "=?$default_charset?Q?$ret?=";
722 $k = ord($string{$i});
724 if ($iEncStart === false) {
728 /* first we add the encoded string that reached it's max size */
729 if ($cur_l > ($max_l-2)) {
730 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
731 $aRet[] = "=?$default_charset?Q?$ret?= "; /* the next part is also encoded => separate by space */
738 $ret .= sprintf("=%02X", $k);
740 if ($iEncStart !== false) {
742 if ($cur_l > $max_l) {
743 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
744 $aRet[] = "=?$default_charset?Q?$ret?=";
759 if ($iEncStart !== false) {
760 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
761 $aRet[] = "=?$default_charset?Q?$ret?=";
763 $aRet[] = substr($string,$iOffset);
765 $string = implode('',$aRet);
770 /* This function trys to locate the entity_id of a specific mime element */
771 function find_ent_id($id, $message) {
772 for ($i = 0, $ret = ''; $ret == '' && $i < count($message->entities
); $i++
) {
773 if ($message->entities
[$i]->header
->type0
== 'multipart') {
774 $ret = find_ent_id($id, $message->entities
[$i]);
776 if (strcasecmp($message->entities
[$i]->header
->id
, $id) == 0) {
777 // if (sq_check_save_extension($message->entities[$i])) {
778 return $message->entities
[$i]->entity_id
;
786 function sq_check_save_extension($message) {
787 $filename = $message->getFilename();
788 $ext = substr($filename, strrpos($filename,'.')+
1);
789 $save_extensions = array('jpg','jpeg','gif','png','bmp');
790 return in_array($ext, $save_extensions);
795 ** HTMLFILTER ROUTINES
799 * This function returns the final tag out of the tag name, an array
800 * of attributes, and the type of the tag. This function is called by
801 * sq_sanitize internally.
803 * @param $tagname the name of the tag.
804 * @param $attary the array of attributes and their values
805 * @param $tagtype The type of the tag (see in comments).
806 * @return a string with the final tag representation.
808 function sq_tagprint($tagname, $attary, $tagtype){
812 $fulltag = '</' . $tagname . '>';
814 $fulltag = '<' . $tagname;
815 if (is_array($attary) && sizeof($attary)){
817 while (list($attname, $attvalue) = each($attary)){
818 array_push($atts, "$attname=$attvalue");
820 $fulltag .= ' ' . join(" ", $atts);
831 * A small helper function to use with array_walk. Modifies a by-ref
832 * value and makes it lowercase.
834 * @param $val a value passed by-ref.
835 * @return void since it modifies a by-ref value.
837 function sq_casenormalize(&$val){
838 $val = strtolower($val);
842 * This function skips any whitespace from the current position within
843 * a string and to the next non-whitespace value.
845 * @param $body the string
846 * @param $offset the offset within the string where we should start
847 * looking for the next non-whitespace character.
848 * @return the location within the $body where the next
849 * non-whitespace char is located.
851 function sq_skipspace($body, $offset){
852 $me = 'sq_skipspace';
853 preg_match('/^(\s*)/s', substr($body, $offset), $matches);
854 if (sizeof($matches{1})){
855 $count = strlen($matches{1});
862 * This function looks for the next character within a string. It's
863 * really just a glorified "strpos", except it catches if failures
866 * @param $body The string to look for needle in.
867 * @param $offset Start looking from this position.
868 * @param $needle The character/string to look for.
869 * @return location of the next occurance of the needle, or
870 * strlen($body) if needle wasn't found.
872 function sq_findnxstr($body, $offset, $needle){
873 $me = 'sq_findnxstr';
874 $pos = strpos($body, $needle, $offset);
876 $pos = strlen($body);
882 * This function takes a PCRE-style regexp and tries to match it
885 * @param $body The string to look for needle in.
886 * @param $offset Start looking from here.
887 * @param $reg A PCRE-style regex to match.
888 * @return Returns a false if no matches found, or an array
889 * with the following members:
890 * - integer with the location of the match within $body
891 * - string with whatever content between offset and the match
892 * - string with whatever it is we matched
894 function sq_findnxreg($body, $offset, $reg){
895 $me = 'sq_findnxreg';
898 preg_match("%^(.*?)($reg)%s", substr($body, $offset), $matches);
902 $retarr{0} = $offset +
strlen($matches{1});
903 $retarr{1} = $matches{1};
904 $retarr{2} = $matches{2};
910 * This function looks for the next tag.
912 * @param $body String where to look for the next tag.
913 * @param $offset Start looking from here.
914 * @return false if no more tags exist in the body, or
915 * an array with the following members:
916 * - string with the name of the tag
917 * - array with attributes and their values
918 * - integer with tag type (1, 2, or 3)
919 * - integer where the tag starts (starting "<")
920 * - integer where the tag ends (ending ">")
921 * first three members will be false, if the tag is invalid.
923 function sq_getnxtag($body, $offset){
925 if ($offset > strlen($body)){
928 $lt = sq_findnxstr($body, $offset, "<");
929 if ($lt == strlen($body)){
934 * blah blah <tag attribute="value">
937 $pos = sq_skipspace($body, $lt+
1);
938 if ($pos >= strlen($body)){
939 return Array(false, false, false, $lt, strlen($body));
942 * There are 3 kinds of tags:
943 * 1. Opening tag, e.g.:
945 * 2. Closing tag, e.g.:
947 * 3. XHTML-style content-less tag, e.g.:
951 switch (substr($body, $pos, 1)){
958 * A comment or an SGML declaration.
960 if (substr($body, $pos+
1, 2) == "--"){
961 $gt = strpos($body, "-->", $pos);
967 return Array(false, false, false, $lt, $gt);
969 $gt = sq_findnxstr($body, $pos, ">");
970 return Array(false, false, false, $lt, $gt);
975 * Assume tagtype 1 for now. If it's type 3, we'll switch values
985 * Look for next [\W-_], which will indicate the end of the tag name.
987 $regary = sq_findnxreg($body, $pos, "[^\w\-_]");
988 if ($regary == false){
989 return Array(false, false, false, $lt, strlen($body));
991 list($pos, $tagname, $match) = $regary;
992 $tagname = strtolower($tagname);
995 * $match can be either of these:
996 * '>' indicating the end of the tag entirely.
997 * '\s' indicating the end of the tag name.
998 * '/' indicating that this is type-3 xhtml tag.
1000 * Whatever else we find there indicates an invalid tag.
1005 * This is an xhtml-style tag with a closing / at the
1006 * end, like so: <img src="blah"/>. Check if it's followed
1007 * by the closing bracket. If not, then this tag is invalid
1009 if (substr($body, $pos, 2) == "/>"){
1013 $gt = sq_findnxstr($body, $pos, ">");
1014 $retary = Array(false, false, false, $lt, $gt);
1018 return Array($tagname, false, $tagtype, $lt, $pos);
1022 * Check if it's whitespace
1024 if (!preg_match('/\s/', $match)){
1026 * This is an invalid tag! Look for the next closing ">".
1028 $gt = sq_findnxstr($body, $offset, ">");
1029 return Array(false, false, false, $lt, $gt);
1035 * At this point we're here:
1036 * <tagname attribute='blah'>
1039 * At this point we loop in order to find all attributes.
1045 while ($pos <= strlen($body)){
1046 $pos = sq_skipspace($body, $pos);
1047 if ($pos == strlen($body)){
1051 return Array(false, false, false, $lt, $pos);
1054 * See if we arrived at a ">" or "/>", which means that we reached
1055 * the end of the tag.
1058 if (preg_match("%^(\s*)(>|/>)%s", substr($body, $pos), $matches)) {
1062 $pos +
= strlen($matches{1});
1063 if ($matches{2} == "/>"){
1067 return Array($tagname, $attary, $tagtype, $lt, $pos);
1071 * There are several types of attributes, with optional
1072 * [:space:] between members.
1074 * attrname[:space:]=[:space:]'CDATA'
1076 * attrname[:space:]=[:space:]"CDATA"
1078 * attr[:space:]=[:space:]CDATA
1082 * We leave types 1 and 2 the same, type 3 we check for
1083 * '"' and convert to """ if needed, then wrap in
1084 * double quotes. Type 4 we convert into:
1087 $regary = sq_findnxreg($body, $pos, "[^\w\-_]");
1088 if ($regary == false){
1090 * Looks like body ended before the end of tag.
1092 return Array(false, false, false, $lt, strlen($body));
1094 list($pos, $attname, $match) = $regary;
1095 $attname = strtolower($attname);
1097 * We arrived at the end of attribute name. Several things possible
1099 * '>' means the end of the tag and this is attribute type 4
1100 * '/' if followed by '>' means the same thing as above
1101 * '\s' means a lot of things -- look what it's followed by.
1102 * anything else means the attribute is invalid.
1107 * This is an xhtml-style tag with a closing / at the
1108 * end, like so: <img src="blah"/>. Check if it's followed
1109 * by the closing bracket. If not, then this tag is invalid
1111 if (substr($body, $pos, 2) == "/>"){
1115 $gt = sq_findnxstr($body, $pos, ">");
1116 $retary = Array(false, false, false, $lt, $gt);
1120 $attary{$attname} = '"yes"';
1121 return Array($tagname, $attary, $tagtype, $lt, $pos);
1125 * Skip whitespace and see what we arrive at.
1127 $pos = sq_skipspace($body, $pos);
1128 $char = substr($body, $pos, 1);
1130 * Two things are valid here:
1131 * '=' means this is attribute type 1 2 or 3.
1132 * \w means this was attribute type 4.
1133 * anything else we ignore and re-loop. End of tag and
1134 * invalid stuff will be caught by our checks at the beginning
1139 $pos = sq_skipspace($body, $pos);
1141 * Here are 3 possibilities:
1142 * "'" attribute type 1
1143 * '"' attribute type 2
1144 * everything else is the content of tag type 3
1146 $quot = substr($body, $pos, 1);
1148 $regary = sq_findnxreg($body, $pos+
1, "\'");
1149 if ($regary == false){
1150 return Array(false, false, false, $lt, strlen($body));
1152 list($pos, $attval, $match) = $regary;
1154 $attary{$attname} = "'" . $attval . "'";
1155 } else if ($quot == '"'){
1156 $regary = sq_findnxreg($body, $pos+
1, '\"');
1157 if ($regary == false){
1158 return Array(false, false, false, $lt, strlen($body));
1160 list($pos, $attval, $match) = $regary;
1162 $attary{$attname} = '"' . $attval . '"';
1165 * These are hateful. Look for \s, or >.
1167 $regary = sq_findnxreg($body, $pos, "[\s>]");
1168 if ($regary == false){
1169 return Array(false, false, false, $lt, strlen($body));
1171 list($pos, $attval, $match) = $regary;
1173 * If it's ">" it will be caught at the top.
1175 $attval = preg_replace("/\"/s", """, $attval);
1176 $attary{$attname} = '"' . $attval . '"';
1178 } else if (preg_match("|[\w/>]|", $char)) {
1180 * That was attribute type 4.
1182 $attary{$attname} = '"yes"';
1185 * An illegal character. Find next '>' and return.
1187 $gt = sq_findnxstr($body, $pos, ">");
1188 return Array(false, false, false, $lt, $gt);
1194 * The fact that we got here indicates that the tag end was never
1195 * found. Return invalid tag indication so it gets stripped.
1197 return Array(false, false, false, $lt, strlen($body));
1201 * This function checks attribute values for entity-encoded values
1202 * and returns them translated into 8-bit strings so we can run
1205 * @param $attvalue A string to run entity check against.
1206 * @return Translated value.
1208 function sq_deent($attvalue){
1211 * See if we have to run the checks first. All entities must start
1214 if (strpos($attvalue, "&") === false){
1218 * Check named entities first.
1220 $trans = get_html_translation_table(HTML_ENTITIES
);
1222 * Leave " in, as it can mess us up.
1224 $trans = array_flip($trans);
1225 unset($trans{"""});
1226 while (list($ent, $val) = each($trans)){
1227 $attvalue = preg_replace("/$ent*(\W)/si", "$val\\1", $attvalue);
1230 * Now translate numbered entities from 1 to 255 if needed.
1232 if (strpos($attvalue, "#") !== false){
1233 $omit = Array(34, 39);
1234 for ($asc=1; $asc<256; $asc++
){
1235 if (!in_array($asc, $omit)){
1237 $attvalue = preg_replace("/\�*$asc;*(\D)/si", "$chr\\1",
1239 $attvalue = preg_replace("/\�*".dechex($asc).";*(\W)/si",
1240 "$chr\\1", $attvalue);
1248 * This function runs various checks against the attributes.
1250 * @param $tagname String with the name of the tag.
1251 * @param $attary Array with all tag attributes.
1252 * @param $rm_attnames See description for sq_sanitize
1253 * @param $bad_attvals See description for sq_sanitize
1254 * @param $add_attr_to_tag See description for sq_sanitize
1255 * @param $message message object
1256 * @param $id message id
1257 * @return Array with modified attributes.
1259 function sq_fixatts($tagname,
1269 while (list($attname, $attvalue) = each($attary)){
1271 * See if this attribute should be removed.
1273 foreach ($rm_attnames as $matchtag=>$matchattrs){
1274 if (preg_match($matchtag, $tagname)){
1275 foreach ($matchattrs as $matchattr){
1276 if (preg_match($matchattr, $attname)){
1277 unset($attary{$attname});
1284 * Remove any entities.
1286 $attvalue = sq_deent($attvalue);
1289 * Now let's run checks on the attvalues.
1290 * I don't expect anyone to comprehend this. If you do,
1291 * get in touch with me so I can drive to where you live and
1292 * shake your hand personally. :)
1294 foreach ($bad_attvals as $matchtag=>$matchattrs){
1295 if (preg_match($matchtag, $tagname)){
1296 foreach ($matchattrs as $matchattr=>$valary){
1297 if (preg_match($matchattr, $attname)){
1299 * There are two arrays in valary.
1301 * Second one is replacements
1303 list($valmatch, $valrepl) = $valary;
1305 preg_replace($valmatch, $valrepl, $attvalue);
1306 if ($newvalue != $attvalue){
1307 $attary{$attname} = $newvalue;
1314 * Turn cid: urls into http-friendly ones.
1316 if (preg_match("/^[\'\"]\s*cid:/si", $attvalue)){
1317 $attary{$attname} = sq_cid2http($message, $id, $attvalue, $mailbox);
1321 * See if we need to append any attributes to this tag.
1323 foreach ($add_attr_to_tag as $matchtag=>$addattary){
1324 if (preg_match($matchtag, $tagname)){
1325 $attary = array_merge($attary, $addattary);
1332 * This function edits the style definition to make them friendly and
1333 * usable in squirrelmail.
1335 * @param $message the message object
1336 * @param $id the message id
1337 * @param $content a string with whatever is between <style> and </style>
1338 * @return a string with edited content.
1340 function sq_fixstyle($message, $id, $content){
1341 global $view_unsafe_images;
1342 $me = 'sq_fixstyle';
1344 * First look for general BODY style declaration, which would be
1346 * body {background: blah-blah}
1347 * and change it to .bodyclass so we can just assign it to a <div>
1349 $content = preg_replace("|body(\s*\{.*?\})|si", ".bodyclass\\1", $content);
1350 $secremoveimg = '../images/' . _("sec_remove_eng.png");
1352 * Fix url('blah') declarations.
1354 $content = preg_replace("|url\(([\'\"])\s*\S+script\s*:.*?([\'\"])\)|si",
1355 "url(\\1$secremoveimg\\2)", $content);
1357 * Fix url('https*://.*) declarations but only if $view_unsafe_images
1360 if (!$view_unsafe_images){
1361 $content = preg_replace("|url\(([\'\"])\s*https*:.*?([\'\"])\)|si",
1362 "url(\\1$secremoveimg\\2)", $content);
1366 * Fix urls that refer to cid:
1368 while (preg_match("|url\(([\'\"]\s*cid:.*?[\'\"])\)|si", $content,
1370 $cidurl = $matches{1};
1371 $httpurl = sq_cid2http($message, $id, $cidurl);
1372 $content = preg_replace("|url\($cidurl\)|si",
1373 "url($httpurl)", $content);
1377 * Fix stupid css declarations which lead to vulnerabilities
1380 $match = Array('/expression/i',
1383 $replace = Array('idiocy', 'idiocy', 'idiocy');
1384 $content = preg_replace($match, $replace, $content);
1389 * This function converts cid: url's into the ones that can be viewed in
1392 * @param $message the message object
1393 * @param $id the message id
1394 * @param $cidurl the cid: url.
1395 * @return a string with a http-friendly url
1397 function sq_cid2http($message, $id, $cidurl, $mailbox){
1399 * Get rid of quotes.
1401 $quotchar = substr($cidurl, 0, 1);
1402 $cidurl = str_replace($quotchar, "", $cidurl);
1403 $cidurl = substr(trim($cidurl), 4);
1404 $linkurl = find_ent_id($cidurl, $message);
1405 /* in case of non-save cid links $httpurl should be replaced by a sort of
1406 unsave link image */
1409 $httpurl = $quotchar . SM_PATH
. 'src/download.php?absolute_dl=true&' .
1410 "passed_id=$id&mailbox=" . urlencode($mailbox) .
1411 '&ent_id=' . $linkurl . $quotchar;
1417 * This function changes the <body> tag into a <div> tag since we
1418 * can't really have a body-within-body.
1420 * @param $attary an array of attributes and values of <body>
1421 * @return a modified array of attributes to be set for <div>
1423 function sq_body2div($attary){
1424 $me = 'sq_body2div';
1425 $divattary = Array('class' => "'bodyclass'");
1426 $bgcolor = '#ffffff';
1429 if (is_array($attary) && sizeof($attary) > 0){
1430 foreach ($attary as $attname=>$attvalue){
1431 $quotchar = substr($attvalue, 0, 1);
1432 $attvalue = str_replace($quotchar, "", $attvalue);
1435 $styledef .= "background-image: url('$attvalue'); ";
1438 $styledef .= "background-color: $attvalue; ";
1441 $styledef .= "color: $attvalue; ";
1445 if (strlen($styledef) > 0){
1446 $divattary{"style"} = "\"$styledef\"";
1453 * This is the main function and the one you should actually be calling.
1454 * There are several variables you should be aware of an which need
1455 * special description.
1457 * Since the description is quite lengthy, see it here:
1458 * http://www.mricon.com/html/phpfilter.html
1460 * @param $body the string with HTML you wish to filter
1461 * @param $tag_list see description above
1462 * @param $rm_tags_with_content see description above
1463 * @param $self_closing_tags see description above
1464 * @param $force_tag_closing see description above
1465 * @param $rm_attnames see description above
1466 * @param $bad_attvals see description above
1467 * @param $add_attr_to_tag see description above
1468 * @param $message message object
1469 * @param $id message id
1470 * @return sanitized html safe to show on your pages.
1472 function sq_sanitize($body,
1474 $rm_tags_with_content,
1484 $me = 'sq_sanitize';
1486 * Normalize rm_tags and rm_tags_with_content.
1488 @array_walk
($rm_tags, 'sq_casenormalize');
1489 @array_walk
($rm_tags_with_content, 'sq_casenormalize');
1490 @array_walk
($self_closing_tags, 'sq_casenormalize');
1492 * See if tag_list is of tags to remove or tags to allow.
1493 * false means remove these tags
1494 * true means allow these tags
1496 $rm_tags = array_shift($tag_list);
1498 $open_tags = Array();
1499 $trusted = "<!-- begin sanitized html -->\n";
1500 $skip_content = false;
1502 * Take care of netscape's stupid javascript entities like
1505 $body = preg_replace("/&(\{.*?\};)/si", "&\\1", $body);
1507 while (($curtag=sq_getnxtag($body, $curpos)) != FALSE){
1508 list($tagname, $attary, $tagtype, $lt, $gt) = $curtag;
1509 $free_content = substr($body, $curpos, $lt-$curpos);
1511 * Take care of <style>
1513 if ($tagname == "style" && $tagtype == 2){
1515 * This is a closing </style>. Edit the
1516 * content before we apply it.
1518 $free_content = sq_fixstyle($message, $id, $free_content);
1520 if ($skip_content == false){
1521 $trusted .= $free_content;
1523 if ($tagname != FALSE){
1525 if ($skip_content == $tagname){
1527 * Got to the end of tag we needed to remove.
1530 $skip_content = false;
1532 if ($skip_content == false){
1533 if ($tagname == "body"){
1536 if (isset($open_tags{$tagname}) &&
1537 $open_tags{$tagname} > 0){
1538 $open_tags{$tagname}--;
1547 * $rm_tags_with_content
1549 if ($skip_content == false){
1551 * See if this is a self-closing type and change
1552 * tagtype appropriately.
1555 && in_array($tagname, $self_closing_tags)){
1559 * See if we should skip this tag and any content
1562 if ($tagtype == 1 &&
1563 in_array($tagname, $rm_tags_with_content)){
1564 $skip_content = $tagname;
1566 if (($rm_tags == false
1567 && in_array($tagname, $tag_list)) ||
1568 ($rm_tags == true &&
1569 !in_array($tagname, $tag_list))){
1573 if (isset($open_tags{$tagname})){
1574 $open_tags{$tagname}++
;
1576 $open_tags{$tagname}=1;
1580 * This is where we run other checks.
1582 if (is_array($attary) && sizeof($attary) > 0){
1583 $attary = sq_fixatts($tagname,
1594 * Convert body into div.
1596 if ($tagname == "body"){
1598 $attary = sq_body2div($attary, $message, $id);
1604 if ($tagname != false && $skip_content == false){
1605 $trusted .= sq_tagprint($tagname, $attary, $tagtype);
1610 $trusted .= substr($body, $curpos, strlen($body)-$curpos);
1611 if ($force_tag_closing == true){
1612 foreach ($open_tags as $tagname=>$opentimes){
1613 while ($opentimes > 0){
1614 $trusted .= '</' . $tagname . '>';
1620 $trusted .= "<!-- end sanitized html -->\n";
1625 * This is a wrapper function to call html sanitizing routines.
1627 * @param $body the body of the message
1628 * @param $id the id of the message
1629 * @return a string with html safe to display in the browser.
1631 function magicHTML($body, $id, $message, $mailbox = 'INBOX') {
1632 global $attachment_common_show_images, $view_unsafe_images,
1635 * Don't display attached images in HTML mode.
1637 $attachment_common_show_images = false;
1650 $rm_tags_with_content = Array(
1657 $self_closing_tags = Array(
1664 $force_tag_closing = false;
1666 $rm_attnames = Array(
1677 $secremoveimg = "../images/" . _("sec_remove_eng.png");
1678 $bad_attvals = Array(
1681 "/^src|background/i" =>
1684 "|^([\'\"])\s*\.\./.*([\'\"])|si",
1685 "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si",
1686 "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si",
1687 "/^([\'\"])\s*about\s*:.*([\'\"])/si"
1690 "\\1$secremoveimg\\2",
1691 "\\1$secremoveimg\\2",
1692 "\\1$secremoveimg\\2",
1693 "\\1$secremoveimg\\2"
1696 "/^href|action/i" =>
1699 "|^([\'\"])\s*\.\./.*([\'\"])|si",
1700 "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si",
1701 "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si",
1702 "/^([\'\"])\s*about\s*:.*([\'\"])/si"
1717 "|url\(([\'\"])\s*\.\./.*([\'\"])\)|si",
1718 "/url\(([\'\"])\s*\S+script\s*:.*([\'\"])\)/si",
1719 "/url\(([\'\"])\s*mocha\s*:.*([\'\"])\)/si",
1720 "/url\(([\'\"])\s*about\s*:.*([\'\"])\)/si"
1734 if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET
) ) {
1735 $view_unsafe_images = false;
1737 if (!$view_unsafe_images){
1739 * Remove any references to http/https if view_unsafe_images set
1742 array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[0],
1743 '/^([\'\"])\s*https*:.*([\'\"])/si');
1744 array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[1],
1745 "\\1$secremoveimg\\2");
1746 array_push($bad_attvals{'/.*/'}{'/^style/i'}[0],
1747 '/url\(([\'\"])\s*https*:.*([\'\"])\)/si');
1748 array_push($bad_attvals{'/.*/'}{'/^style/i'}[1],
1749 "url(\\1$secremoveimg\\2)");
1752 $add_attr_to_tag = Array(
1753 "/^a$/i" => Array('target'=>'"_new"')
1755 $trusted = sq_sanitize($body,
1757 $rm_tags_with_content,
1767 if (preg_match("|$secremoveimg|i", $trusted)){
1768 $has_unsafe_images = true;