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
340 * Since 1.2.0 function uses message_body hook.
341 * Till 1.3.0 function included output of formatAttachments().
343 * @param resource $imap_stream imap connection resource
344 * @param object $message squirrelmail message object
345 * @param array $color squirrelmail color theme array
346 * @param integer $wrap_at number of characters per line
347 * @param string $ent_num (since 1.3.0) message part id
348 * @param integer $id (since 1.3.0) message id
349 * @param string $mailbox (since 1.3.0) imap folder name
350 * @param boolean $clean (since 1.5.1) Do not output stuff that's irrelevant for the printable version.
351 * @return string html formated message text
353 function formatBody($imap_stream, $message, $color, $wrap_at, $ent_num, $id, $mailbox='INBOX', $clean=FALSE) {
354 /* This if statement checks for the entity to show as the
355 * primary message. To add more of them, just put them in the
356 * order that is their priority.
358 global $startMessage, $languages, $squirrelmail_language,
359 $show_html_default, $sort, $has_unsafe_images, $passed_ent_id;
361 if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET
) ) {
362 $view_unsafe_images = false;
366 $urlmailbox = urlencode($mailbox);
367 $body_message = getEntity($message, $ent_num);
368 if (($body_message->header
->type0
== 'text') ||
369 ($body_message->header
->type0
== 'rfc822')) {
370 $body = mime_fetch_body ($imap_stream, $id, $ent_num);
371 $body = decodeBody($body, $body_message->header
->encoding
);
373 if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
374 function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_decode')) {
375 if (mb_detect_encoding($body) != 'ASCII') {
376 $body = call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_decode',$body);
379 $hookResults = do_hook("message_body", $body);
380 $body = $hookResults[1];
382 /* If there are other types that shouldn't be formatted, add
386 if ($body_message->header
->type1
== 'html') {
387 if ($show_html_default <> 1) {
388 $entity_conv = array(' ' => ' ',
397 $body = strtr($body, $entity_conv);
398 $body = strip_tags($body);
400 translateText($body, $wrap_at,
401 $body_message->header
->getParameter('charset'));
403 $body = magicHTML($body, $id, $message, $mailbox);
406 translateText($body, $wrap_at,
407 $body_message->header
->getParameter('charset'));
410 // if this is the clean display (i.e. printer friendly), stop here.
415 $link = 'passed_id=' . $id . '&ent_id='.$ent_num.
416 '&mailbox=' . $urlmailbox .'&sort=' . $sort .
417 '&startMessage=' . $startMessage . '&show_more=0';
418 if (isset($passed_ent_id)) {
419 $link .= '&passed_ent_id='.$passed_ent_id;
421 $body .= '<center><small><a href="download.php?absolute_dl=true&' .
422 $link . '">' . _("Download this as a file") . '</a>';
423 if ($view_unsafe_images) {
424 $text = _("Hide Unsafe Images");
426 if (isset($has_unsafe_images) && $has_unsafe_images) {
427 $link .= '&view_unsafe_images=1';
428 $text = _("View Unsafe Images");
434 $body .= ' | <a href="read_body.php?' . $link . '">' . $text . '</a>';
436 $body .= '</small></center><br />' . "\n";
442 * Displays attachment links and information
443 * FIXME: SM_PATH is used in URLs
445 * Since 1.3.0 function is not included in formatBody() call.
447 * Since 1.0.2 uses attachment $type0/$type1 hook.
448 * Since 1.2.5 uses attachment $type0/* hook.
449 * Since 1.5.0 uses attachments_bottom hook.
451 * @param object $message SquirrelMail message object
452 * @param array $exclude_id message parts that are not attachments.
453 * @param string $mailbox mailbox name
454 * @param integer $id message id
455 * @return string html formated attachment information.
457 function formatAttachments($message, $exclude_id, $mailbox, $id) {
458 global $where, $what, $startMessage, $color, $passed_ent_id;
460 $att_ar = $message->getAttachments($exclude_id);
462 if (!count($att_ar)) return '';
466 $urlMailbox = urlencode($mailbox);
468 foreach ($att_ar as $att) {
469 $ent = $att->entity_id
;
470 $header = $att->header
;
471 $type0 = strtolower($header->type0
);
472 $type1 = strtolower($header->type1
);
474 $links['download link']['text'] = _("Download");
475 $links['download link']['href'] = SM_PATH
.
476 "src/download.php?absolute_dl=true&passed_id=$id&mailbox=$urlMailbox&ent_id=$ent";
477 if ($type0 =='message' && $type1 == 'rfc822') {
478 $default_page = SM_PATH
. 'src/read_body.php';
479 $rfc822_header = $att->rfc822_header
;
480 $filename = $rfc822_header->subject
;
481 if (trim( $filename ) == '') {
482 $filename = 'untitled-[' . $ent . ']' ;
484 $from_o = $rfc822_header->from
;
485 if (is_object($from_o)) {
486 $from_name = decodeHeader($from_o->getAddress(false));
488 $from_name = _("Unknown sender");
490 $description = $from_name;
492 $default_page = SM_PATH
. 'src/download.php';
493 $filename = $att->getFilename();
494 if ($header->description
) {
495 $description = decodeHeader($header->description
);
501 $display_filename = $filename;
502 if (isset($passed_ent_id)) {
503 $passed_ent_id_link = '&passed_ent_id='.$passed_ent_id;
505 $passed_ent_id_link = '';
507 $defaultlink = $default_page . "?startMessage=$startMessage"
508 . "&passed_id=$id&mailbox=$urlMailbox"
509 . '&ent_id='.$ent.$passed_ent_id_link;
510 if ($where && $what) {
511 $defaultlink .= '&where='. urlencode($where).'&what='.urlencode($what);
514 /* This executes the attachment hook with a specific MIME-type.
515 * If that doesn't have results, it tries if there's a rule
516 * for a more generic type.
518 $hookresults = do_hook("attachment $type0/$type1", $links,
519 $startMessage, $id, $urlMailbox, $ent, $defaultlink,
520 $display_filename, $where, $what);
521 if(count($hookresults[1]) <= 1) {
522 $hookresults = do_hook("attachment $type0/*", $links,
523 $startMessage, $id, $urlMailbox, $ent, $defaultlink,
524 $display_filename, $where, $what);
527 $links = $hookresults[1];
528 $defaultlink = $hookresults[6];
530 $attachments .= '<tr><td>' .
531 '<a href="'.$defaultlink.'">'.decodeHeader($display_filename).'</a> </td>' .
532 '<td><small><b>' . show_readable_size($header->size
) .
533 '</b> </small></td>' .
534 '<td><small>[ '.htmlspecialchars($type0).'/'.htmlspecialchars($type1).' ] </small></td>' .
536 $attachments .= '<b>' . $description . '</b>';
537 $attachments .= '</small></td><td><small> ';
540 foreach ($links as $val) {
544 $attachments .= ' | ';
546 $attachments .= '<a href="' . $val['href'] . '">' . (isset($val['text']) && !empty($val['text']) ?
$val['text'] : '') . (isset($val['extra']) && !empty($val['extra']) ?
$val['extra'] : '') . '</a>';
549 $attachments .= "</td></tr>\n";
551 $attachmentadd = do_hook_function('attachments_bottom',$attachments);
552 if ($attachmentadd != '')
553 $attachments = $attachmentadd;
557 function sqimap_base64_decode(&$string) {
559 // Base64 encoded data goes in pairs of 4 bytes. To achieve on the
560 // fly decoding (to reduce memory usage) you have to check if the
561 // data has incomplete pairs
563 // Remove the noise in order to check if the 4 bytes pairs are complete
564 $string = str_replace(array("\r\n","\n", "\r", " "),array('','','',''),$string);
567 $iMod = strlen($string) %
4;
569 $sStringRem = substr($string,-$iMod);
570 // Check if $sStringRem contains padding characters
571 if (substr($sStringRem,-1) != '=') {
572 $string = substr($string,0,-$iMod);
577 $string = base64_decode($string);
582 /* This function decodes the body depending on the encoding type. */
583 function decodeBody($body, $encoding) {
584 global $show_html_default;
586 $body = str_replace("\r\n", "\n", $body);
587 $encoding = strtolower($encoding);
589 $encoding_handler = do_hook_function('decode_body', $encoding);
592 // plugins get first shot at decoding the body
594 if (!empty($encoding_handler) && function_exists($encoding_handler)) {
595 $body = $encoding_handler('decode', $body);
597 } else if ($encoding == 'quoted-printable' ||
598 $encoding == 'quoted_printable') {
599 $body = quoted_printable_decode($body);
601 while (ereg("=\n", $body)) {
602 $body = ereg_replace ("=\n", '', $body);
605 } else if ($encoding == 'base64') {
606 $body = base64_decode($body);
609 // All other encodings are returned raw.
616 * This functions decode strings that is encoded according to
617 * RFC1522 (MIME Part Two: Message Header Extensions for Non-ASCII Text).
618 * Patched by Christian Schmidt <christian@ostenfeld.dk> 23/03/2002
620 * @param string $string header string that has to be made readable
621 * @param boolean $utfencode change message in order to be readable on user's charset. defaults to true
622 * @param boolean $htmlsave preserve spaces and sanitize html special characters. defaults to true
623 * @param boolean $decide decide if string can be utfencoded. defaults to false
624 * @return string decoded header string
626 function decodeHeader ($string, $utfencode=true,$htmlsave=true,$decide=false) {
627 global $languages, $squirrelmail_language,$default_charset;
628 if (is_array($string)) {
629 $string = implode("\n", $string);
632 if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
633 function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_decodeheader')) {
634 $string = call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_decodeheader', $string);
635 // Do we need to return at this point?
642 $aString = explode(' ',$string);
644 foreach ($aString as $chunk) {
645 if ($encoded && $chunk === '') {
647 } elseif ($chunk === '') {
652 /* if encoded words are not separated by a linear-space-white we still catch them */
655 while ($match = preg_match('/^(.*)=\?([^?]*)\?(Q|B)\?([^?]*)\?=(.*)$/Ui',$chunk,$res)) {
656 /* if the last chunk isn't an encoded string then put back the space, otherwise don't */
657 if ($iLastMatch !== $j) {
667 $ret .= htmlspecialchars($res[1]);
671 $encoding = ucfirst($res[3]);
673 /* decide about valid decoding */
674 if ($decide && is_conversion_safe($res[2])) {
676 $can_be_encoded=true;
678 $can_be_encoded=false;
683 $replace = base64_decode($res[4]);
685 if ($can_be_encoded) {
686 /* convert string to different charset,
687 * if functions asks for it (usually in compose)
689 $ret .= charset_convert($res[2],$replace,$default_charset);
691 // convert string to html codes in order to display it
692 $ret .= charset_decode($res[2],$replace);
696 $replace = htmlspecialchars($replace);
702 $replace = str_replace('_', ' ', $res[4]);
703 $replace = preg_replace('/=([0-9a-f]{2})/ie', 'chr(hexdec("\1"))',
706 if ($can_be_encoded) {
707 /* convert string to different charset,
708 * if functions asks for it (usually in compose)
710 $replace = charset_convert($res[2], $replace,$default_charset);
712 // convert string to html codes in order to display it
713 $replace = charset_decode($res[2], $replace);
717 $replace = htmlspecialchars($replace);
736 if (!$encoded && $htmlsave) {
737 $ret .= htmlspecialchars($chunk);
743 /* remove the first added space */
746 $ret = substr($ret,5);
748 $ret = substr($ret,1);
756 * Encodes header as quoted-printable
758 * Encode a string according to RFC 1522 for use in headers if it
759 * contains 8-bit characters or anything that looks like it should
762 * @param string $string header string, that has to be encoded
763 * @return string quoted-printable encoded string
765 function encodeHeader ($string) {
766 global $default_charset, $languages, $squirrelmail_language;
768 if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
769 function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_encodeheader')) {
770 return call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_encodeheader', $string);
773 // Encode only if the string contains 8-bit characters or =?
774 $j = strlen($string);
775 $max_l = 75 - strlen($default_charset) - 7;
778 $iEncStart = $enc_init = false;
779 $cur_l = $iOffset = 0;
780 for($i = 0; $i < $j; ++
$i) {
789 if ($iEncStart === false) {
793 if ($cur_l > ($max_l-2)) {
794 /* if there is an stringpart that doesn't need encoding, add it */
795 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
796 $aRet[] = "=?$default_charset?Q?$ret?=";
802 $ret .= sprintf("=%02X",ord($string{$i}));
807 if ($iEncStart !== false) {
808 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
809 $aRet[] = "=?$default_charset?Q?$ret?=";
817 if ($iEncStart !== false) {
819 if ($cur_l > $max_l) {
820 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
821 $aRet[] = "=?$default_charset?Q?$ret?=";
832 $k = ord($string{$i});
834 if ($iEncStart === false) {
835 // do not start encoding in the middle of a string, also take the rest of the word.
836 $sLeadString = substr($string,0,$i);
837 $aLeadString = explode(' ',$sLeadString);
838 $sToBeEncoded = array_pop($aLeadString);
839 $iEncStart = $i - strlen($sToBeEncoded);
840 $ret .= $sToBeEncoded;
841 $cur_l +
= strlen($sToBeEncoded);
844 /* first we add the encoded string that reached it's max size */
845 if ($cur_l > ($max_l-2)) {
846 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
847 $aRet[] = "=?$default_charset?Q?$ret?= "; /* the next part is also encoded => separate by space */
854 $ret .= sprintf("=%02X", $k);
856 if ($iEncStart !== false) {
858 if ($cur_l > $max_l) {
859 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
860 $aRet[] = "=?$default_charset?Q?$ret?=";
875 if ($iEncStart !== false) {
876 $aRet[] = substr($string,$iOffset,$iEncStart-$iOffset);
877 $aRet[] = "=?$default_charset?Q?$ret?=";
879 $aRet[] = substr($string,$iOffset);
881 $string = implode('',$aRet);
886 /* This function trys to locate the entity_id of a specific mime element */
887 function find_ent_id($id, $message) {
888 for ($i = 0, $ret = ''; $ret == '' && $i < count($message->entities
); $i++
) {
889 if ($message->entities
[$i]->header
->type0
== 'multipart') {
890 $ret = find_ent_id($id, $message->entities
[$i]);
892 if (strcasecmp($message->entities
[$i]->header
->id
, $id) == 0) {
893 // if (sq_check_save_extension($message->entities[$i])) {
894 return $message->entities
[$i]->entity_id
;
896 } elseif (!empty($message->entities
[$i]->header
->parameters
['name'])) {
898 * This is part of a fix for Outlook Express 6.x generating
899 * cid URLs without creating content-id headers
902 if (strcasecmp($message->entities
[$i]->header
->parameters
['name'], $id) == 0) {
903 return $message->entities
[$i]->entity_id
;
911 function sq_check_save_extension($message) {
912 $filename = $message->getFilename();
913 $ext = substr($filename, strrpos($filename,'.')+
1);
914 $save_extensions = array('jpg','jpeg','gif','png','bmp');
915 return in_array($ext, $save_extensions);
920 ** HTMLFILTER ROUTINES
924 * This function is more or less a wrapper around stripslashes. Apparently
925 * Explorer is stupid enough to just remove the backslashes and then
926 * execute the content of the attribute as if nothing happened.
929 * @param attvalue The value of the attribute
930 * @return attvalue The value of the attribute stripslashed.
932 function sq_unbackslash($attvalue){
934 * Remove any backslashes. See if there are any first.
937 if (strstr($attvalue, '\\') !== false){
938 $attvalue = stripslashes($attvalue);
944 * Kill any tabs, newlines, or carriage returns. Our friends the
945 * makers of the browser with 95% market value decided that it'd
946 * be funny to make "java[tab]script" be just as good as "javascript".
948 * @param attvalue The attribute value before extraneous spaces removed.
949 * @return attvalue The attribute value after extraneous spaces removed.
951 function sq_unspace($attvalue){
952 if (strcspn($attvalue, "\t\r\n") != strlen($attvalue)){
953 $attvalue = str_replace(Array("\t", "\r", "\n"), Array('', '', ''),
960 * This function returns the final tag out of the tag name, an array
961 * of attributes, and the type of the tag. This function is called by
962 * sq_sanitize internally.
964 * @param $tagname the name of the tag.
965 * @param $attary the array of attributes and their values
966 * @param $tagtype The type of the tag (see in comments).
967 * @return a string with the final tag representation.
969 function sq_tagprint($tagname, $attary, $tagtype){
973 $fulltag = '</' . $tagname . '>';
975 $fulltag = '<' . $tagname;
976 if (is_array($attary) && sizeof($attary)){
978 while (list($attname, $attvalue) = each($attary)){
979 array_push($atts, "$attname=$attvalue");
981 $fulltag .= ' ' . join(" ", $atts);
992 * A small helper function to use with array_walk. Modifies a by-ref
993 * value and makes it lowercase.
995 * @param $val a value passed by-ref.
996 * @return void since it modifies a by-ref value.
998 function sq_casenormalize(&$val){
999 $val = strtolower($val);
1003 * This function skips any whitespace from the current position within
1004 * a string and to the next non-whitespace value.
1006 * @param $body the string
1007 * @param $offset the offset within the string where we should start
1008 * looking for the next non-whitespace character.
1009 * @return the location within the $body where the next
1010 * non-whitespace char is located.
1012 function sq_skipspace($body, $offset){
1013 $me = 'sq_skipspace';
1014 preg_match('/^(\s*)/s', substr($body, $offset), $matches);
1015 if (sizeof($matches{1})){
1016 $count = strlen($matches{1});
1023 * This function looks for the next character within a string. It's
1024 * really just a glorified "strpos", except it catches if failures
1027 * @param $body The string to look for needle in.
1028 * @param $offset Start looking from this position.
1029 * @param $needle The character/string to look for.
1030 * @return location of the next occurance of the needle, or
1031 * strlen($body) if needle wasn't found.
1033 function sq_findnxstr($body, $offset, $needle){
1034 $me = 'sq_findnxstr';
1035 $pos = strpos($body, $needle, $offset);
1036 if ($pos === FALSE){
1037 $pos = strlen($body);
1043 * This function takes a PCRE-style regexp and tries to match it
1044 * within the string.
1046 * @param $body The string to look for needle in.
1047 * @param $offset Start looking from here.
1048 * @param $reg A PCRE-style regex to match.
1049 * @return Returns a false if no matches found, or an array
1050 * with the following members:
1051 * - integer with the location of the match within $body
1052 * - string with whatever content between offset and the match
1053 * - string with whatever it is we matched
1055 function sq_findnxreg($body, $offset, $reg){
1056 $me = 'sq_findnxreg';
1059 preg_match("%^(.*?)($reg)%si", substr($body, $offset), $matches);
1060 if (!isset($matches{0}) ||
!$matches{0}){
1063 $retarr{0} = $offset +
strlen($matches{1});
1064 $retarr{1} = $matches{1};
1065 $retarr{2} = $matches{2};
1071 * This function looks for the next tag.
1073 * @param $body String where to look for the next tag.
1074 * @param $offset Start looking from here.
1075 * @return false if no more tags exist in the body, or
1076 * an array with the following members:
1077 * - string with the name of the tag
1078 * - array with attributes and their values
1079 * - integer with tag type (1, 2, or 3)
1080 * - integer where the tag starts (starting "<")
1081 * - integer where the tag ends (ending ">")
1082 * first three members will be false, if the tag is invalid.
1084 function sq_getnxtag($body, $offset){
1085 $me = 'sq_getnxtag';
1086 if ($offset > strlen($body)){
1089 $lt = sq_findnxstr($body, $offset, "<");
1090 if ($lt == strlen($body)){
1095 * blah blah <tag attribute="value">
1098 $pos = sq_skipspace($body, $lt+
1);
1099 if ($pos >= strlen($body)){
1100 return Array(false, false, false, $lt, strlen($body));
1103 * There are 3 kinds of tags:
1104 * 1. Opening tag, e.g.:
1106 * 2. Closing tag, e.g.:
1108 * 3. XHTML-style content-less tag, e.g.:
1109 * <img src="blah" />
1112 switch (substr($body, $pos, 1)){
1119 * A comment or an SGML declaration.
1121 if (substr($body, $pos+
1, 2) == "--"){
1122 $gt = strpos($body, "-->", $pos);
1124 $gt = strlen($body);
1128 return Array(false, false, false, $lt, $gt);
1130 $gt = sq_findnxstr($body, $pos, ">");
1131 return Array(false, false, false, $lt, $gt);
1136 * Assume tagtype 1 for now. If it's type 3, we'll switch values
1145 * Look for next [\W-_], which will indicate the end of the tag name.
1147 $regary = sq_findnxreg($body, $pos, "[^\w\-_]");
1148 if ($regary == false){
1149 return Array(false, false, false, $lt, strlen($body));
1151 list($pos, $tagname, $match) = $regary;
1152 $tagname = strtolower($tagname);
1155 * $match can be either of these:
1156 * '>' indicating the end of the tag entirely.
1157 * '\s' indicating the end of the tag name.
1158 * '/' indicating that this is type-3 xhtml tag.
1160 * Whatever else we find there indicates an invalid tag.
1165 * This is an xhtml-style tag with a closing / at the
1166 * end, like so: <img src="blah" />. Check if it's followed
1167 * by the closing bracket. If not, then this tag is invalid
1169 if (substr($body, $pos, 2) == "/>"){
1173 $gt = sq_findnxstr($body, $pos, ">");
1174 $retary = Array(false, false, false, $lt, $gt);
1178 return Array($tagname, false, $tagtype, $lt, $pos);
1182 * Check if it's whitespace
1184 if (!preg_match('/\s/', $match)){
1186 * This is an invalid tag! Look for the next closing ">".
1188 $gt = sq_findnxstr($body, $lt, ">");
1189 return Array(false, false, false, $lt, $gt);
1195 * At this point we're here:
1196 * <tagname attribute='blah'>
1199 * At this point we loop in order to find all attributes.
1204 while ($pos <= strlen($body)){
1205 $pos = sq_skipspace($body, $pos);
1206 if ($pos == strlen($body)){
1210 return Array(false, false, false, $lt, $pos);
1213 * See if we arrived at a ">" or "/>", which means that we reached
1214 * the end of the tag.
1217 if (preg_match("%^(\s*)(>|/>)%s", substr($body, $pos), $matches)) {
1221 $pos +
= strlen($matches{1});
1222 if ($matches{2} == "/>"){
1226 return Array($tagname, $attary, $tagtype, $lt, $pos);
1230 * There are several types of attributes, with optional
1231 * [:space:] between members.
1233 * attrname[:space:]=[:space:]'CDATA'
1235 * attrname[:space:]=[:space:]"CDATA"
1237 * attr[:space:]=[:space:]CDATA
1241 * We leave types 1 and 2 the same, type 3 we check for
1242 * '"' and convert to """ if needed, then wrap in
1243 * double quotes. Type 4 we convert into:
1246 $regary = sq_findnxreg($body, $pos, "[^:\w\-_]");
1247 if ($regary == false){
1249 * Looks like body ended before the end of tag.
1251 return Array(false, false, false, $lt, strlen($body));
1253 list($pos, $attname, $match) = $regary;
1254 $attname = strtolower($attname);
1256 * We arrived at the end of attribute name. Several things possible
1258 * '>' means the end of the tag and this is attribute type 4
1259 * '/' if followed by '>' means the same thing as above
1260 * '\s' means a lot of things -- look what it's followed by.
1261 * anything else means the attribute is invalid.
1266 * This is an xhtml-style tag with a closing / at the
1267 * end, like so: <img src="blah" />. Check if it's followed
1268 * by the closing bracket. If not, then this tag is invalid
1270 if (substr($body, $pos, 2) == "/>"){
1274 $gt = sq_findnxstr($body, $pos, ">");
1275 $retary = Array(false, false, false, $lt, $gt);
1279 $attary{$attname} = '"yes"';
1280 return Array($tagname, $attary, $tagtype, $lt, $pos);
1284 * Skip whitespace and see what we arrive at.
1286 $pos = sq_skipspace($body, $pos);
1287 $char = substr($body, $pos, 1);
1289 * Two things are valid here:
1290 * '=' means this is attribute type 1 2 or 3.
1291 * \w means this was attribute type 4.
1292 * anything else we ignore and re-loop. End of tag and
1293 * invalid stuff will be caught by our checks at the beginning
1298 $pos = sq_skipspace($body, $pos);
1300 * Here are 3 possibilities:
1301 * "'" attribute type 1
1302 * '"' attribute type 2
1303 * everything else is the content of tag type 3
1305 $quot = substr($body, $pos, 1);
1307 $regary = sq_findnxreg($body, $pos+
1, "\'");
1308 if ($regary == false){
1309 return Array(false, false, false, $lt, strlen($body));
1311 list($pos, $attval, $match) = $regary;
1313 $attary{$attname} = "'" . $attval . "'";
1314 } else if ($quot == '"'){
1315 $regary = sq_findnxreg($body, $pos+
1, '\"');
1316 if ($regary == false){
1317 return Array(false, false, false, $lt, strlen($body));
1319 list($pos, $attval, $match) = $regary;
1321 $attary{$attname} = '"' . $attval . '"';
1324 * These are hateful. Look for \s, or >.
1326 $regary = sq_findnxreg($body, $pos, "[\s>]");
1327 if ($regary == false){
1328 return Array(false, false, false, $lt, strlen($body));
1330 list($pos, $attval, $match) = $regary;
1332 * If it's ">" it will be caught at the top.
1334 $attval = preg_replace("/\"/s", """, $attval);
1335 $attary{$attname} = '"' . $attval . '"';
1337 } else if (preg_match("|[\w/>]|", $char)) {
1339 * That was attribute type 4.
1341 $attary{$attname} = '"yes"';
1344 * An illegal character. Find next '>' and return.
1346 $gt = sq_findnxstr($body, $pos, ">");
1347 return Array(false, false, false, $lt, $gt);
1353 * The fact that we got here indicates that the tag end was never
1354 * found. Return invalid tag indication so it gets stripped.
1356 return Array(false, false, false, $lt, strlen($body));
1360 * This function checks attribute values for entity-encoded values
1361 * and returns them translated into 8-bit strings so we can run
1364 * @param $attvalue A string to run entity check against.
1365 * @return Translated value.
1368 function sq_deent($attvalue){
1371 * See if we have to run the checks first. All entities must start
1374 if (strpos($attvalue, '&') === false){
1378 * Check named entities first.
1380 $trans = get_html_translation_table(HTML_ENTITIES
);
1382 * Leave " in, as it can mess us up.
1384 $trans = array_flip($trans);
1385 unset($trans{'"'});
1386 while (list($ent, $val) = each($trans)){
1387 $attvalue = preg_replace('/' . $ent . '*/si', $val, $attvalue);
1390 * Now translate numbered entities from 1 to 255 if needed.
1392 if (strpos($attvalue, '#') !== false){
1393 $omit = Array(34, 39);
1394 for ($asc = 256; $asc >= 0; $asc--){
1395 if (!in_array($asc, $omit)){
1397 $octrule = '/\�*' . $asc . ';*/si';
1398 $hexrule = '/\�*' . dechex($asc) . ';*/si';
1399 $attvalue = preg_replace($octrule, $chr, $attvalue);
1400 $attvalue = preg_replace($hexrule, $chr, $attvalue);
1408 * This function runs various checks against the attributes.
1410 * @param $tagname String with the name of the tag.
1411 * @param $attary Array with all tag attributes.
1412 * @param $rm_attnames See description for sq_sanitize
1413 * @param $bad_attvals See description for sq_sanitize
1414 * @param $add_attr_to_tag See description for sq_sanitize
1415 * @param $message message object
1416 * @param $id message id
1417 * @return Array with modified attributes.
1419 function sq_fixatts($tagname,
1429 while (list($attname, $attvalue) = each($attary)){
1431 * See if this attribute should be removed.
1433 foreach ($rm_attnames as $matchtag=>$matchattrs){
1434 if (preg_match($matchtag, $tagname)){
1435 foreach ($matchattrs as $matchattr){
1436 if (preg_match($matchattr, $attname)){
1437 unset($attary{$attname});
1444 * Remove any backslashes, entities, and extraneous whitespace.
1446 $attvalue = sq_unbackslash($attvalue);
1447 $attvalue = sq_deent($attvalue);
1448 $attvalue = sq_unspace($attvalue);
1451 * Remove \r \n \t \0 " " "\\"
1453 $attvalue = str_replace(Array("\r", "\n", "\t", "\0", " ", "\\"),
1454 Array('', '','','','',''), $attvalue);
1457 * Now let's run checks on the attvalues.
1458 * I don't expect anyone to comprehend this. If you do,
1459 * get in touch with me so I can drive to where you live and
1460 * shake your hand personally. :)
1462 foreach ($bad_attvals as $matchtag=>$matchattrs){
1463 if (preg_match($matchtag, $tagname)){
1464 foreach ($matchattrs as $matchattr=>$valary){
1465 if (preg_match($matchattr, $attname)){
1467 * There are two arrays in valary.
1469 * Second one is replacements
1471 list($valmatch, $valrepl) = $valary;
1473 preg_replace($valmatch, $valrepl, $attvalue);
1474 if ($newvalue != $attvalue){
1475 $attary{$attname} = $newvalue;
1484 * Replace empty src tags with the blank image. src is only used
1485 * for frames, images, and image inputs. Doing a replace should
1486 * not affect them working as should be, however it will stop
1487 * IE from being kicked off when src for img tags are not set
1489 if (($attname == 'src') && ($attvalue == '""')) {
1490 $attary{$attname} = '"' . SM_PATH
. 'images/blank.png"';
1494 * Turn cid: urls into http-friendly ones.
1496 if (preg_match("/^[\'\"]\s*cid:/si", $attvalue)){
1497 $attary{$attname} = sq_cid2http($message, $id, $attvalue, $mailbox);
1501 * "Hack" fix for Outlook using propriatary outbind:// protocol in img tags.
1502 * One day MS might actually make it match something useful, for now, falling
1503 * back to using cid2http, so we can grab the blank.png.
1505 if (preg_match("/^[\'\"]\s*outbind:\/\//si", $attvalue)) {
1506 $attary{$attname} = sq_cid2http($message, $id, $attvalue, $mailbox);
1511 * See if we need to append any attributes to this tag.
1513 foreach ($add_attr_to_tag as $matchtag=>$addattary){
1514 if (preg_match($matchtag, $tagname)){
1515 $attary = array_merge($attary, $addattary);
1522 * This function edits the style definition to make them friendly and
1523 * usable in SquirrelMail.
1525 * @param $message the message object
1526 * @param $id the message id
1527 * @param $content a string with whatever is between <style> and </style>
1528 * @param $mailbox the message mailbox
1529 * @return a string with edited content.
1531 function sq_fixstyle($body, $pos, $message, $id, $mailbox){
1532 global $view_unsafe_images;
1533 $me = 'sq_fixstyle';
1534 $ret = sq_findnxreg($body, $pos, '</\s*style\s*>');
1536 return array(FALSE, strlen($body));
1538 $newpos = $ret[0] +
strlen($ret[2]);
1541 * First look for general BODY style declaration, which would be
1543 * body {background: blah-blah}
1544 * and change it to .bodyclass so we can just assign it to a <div>
1546 $content = preg_replace("|body(\s*\{.*?\})|si", ".bodyclass\\1", $content);
1547 $secremoveimg = '../images/' . _("sec_remove_eng.png");
1549 * Fix url('blah') declarations.
1551 $content = preg_replace("|url\s*\(\s*([\'\"])\s*\S+script\s*:.*?([\'\"])\s*\)|si",
1552 "url(\\1$secremoveimg\\2)", $content);
1554 * Fix url('https*://.*) declarations but only if $view_unsafe_images
1557 if (!$view_unsafe_images){
1558 $content = preg_replace("|url\s*\(\s*([\'\"])\s*https*:.*?([\'\"])\s*\)|si",
1559 "url(\\1$secremoveimg\\2)", $content);
1563 * Fix urls that refer to cid:
1565 while (preg_match("|url\s*\(\s*([\'\"]\s*cid:.*?[\'\"])\s*\)|si",
1566 $content, $matches)){
1567 $cidurl = $matches{1};
1568 $httpurl = sq_cid2http($message, $id, $cidurl, $mailbox);
1569 $content = preg_replace("|url\s*\(\s*$cidurl\s*\)|si",
1570 "url($httpurl)", $content);
1574 * Fix stupid css declarations which lead to vulnerabilities
1577 $match = Array('/expression/i',
1580 '/include-source/i');
1581 $replace = Array('idiocy', 'idiocy', 'idiocy', 'idiocy');
1582 $content = preg_replace($match, $replace, $content);
1583 return array($content, $newpos);
1587 * This function converts cid: url's into the ones that can be viewed in
1590 * @param $message the message object
1591 * @param $id the message id
1592 * @param $cidurl the cid: url.
1593 * @param $mailbox the message mailbox
1594 * @return a string with a http-friendly url
1596 function sq_cid2http($message, $id, $cidurl, $mailbox){
1598 * Get rid of quotes.
1600 $quotchar = substr($cidurl, 0, 1);
1601 if ($quotchar == '"' ||
$quotchar == "'"){
1602 $cidurl = str_replace($quotchar, "", $cidurl);
1606 $cidurl = substr(trim($cidurl), 4);
1607 $linkurl = find_ent_id($cidurl, $message);
1608 /* in case of non-save cid links $httpurl should be replaced by a sort of
1609 unsave link image */
1613 * This is part of a fix for Outlook Express 6.x generating
1614 * cid URLs without creating content-id headers. These images are
1615 * not part of the multipart/related html mail. The html contains
1616 * <img src="cid:{some_id}/image_filename.ext"> references to
1617 * attached images with as goal to render them inline although
1618 * the attachment disposition property is not inline.
1621 if (empty($linkurl)) {
1622 if (preg_match('/{.*}\//', $cidurl)) {
1623 $cidurl = preg_replace('/{.*}\//','', $cidurl);
1624 if (!empty($cidurl)) {
1625 $linkurl = find_ent_id($cidurl, $message);
1630 if (!empty($linkurl)) {
1631 $httpurl = $quotchar . SM_PATH
. 'src/download.php?absolute_dl=true&' .
1632 "passed_id=$id&mailbox=" . urlencode($mailbox) .
1633 '&ent_id=' . $linkurl . $quotchar;
1636 * If we couldn't generate a proper img url, drop in a blank image
1637 * instead of sending back empty, otherwise it causes unusual behaviour
1639 $httpurl = $quotchar . SM_PATH
. 'images/blank.png';
1646 * This function changes the <body> tag into a <div> tag since we
1647 * can't really have a body-within-body.
1649 * @param $attary an array of attributes and values of <body>
1650 * @param $mailbox mailbox we're currently reading (for cid2http)
1651 * @param $message current message (for cid2http)
1652 * @param $id current message id (for cid2http)
1653 * @return a modified array of attributes to be set for <div>
1655 function sq_body2div($attary, $mailbox, $message, $id){
1656 $me = 'sq_body2div';
1657 $divattary = Array('class' => "'bodyclass'");
1659 $has_bgc_stl = $has_txt_stl = false;
1661 if (is_array($attary) && sizeof($attary) > 0){
1662 foreach ($attary as $attname=>$attvalue){
1663 $quotchar = substr($attvalue, 0, 1);
1664 $attvalue = str_replace($quotchar, "", $attvalue);
1667 $attvalue = sq_cid2http($message, $id,
1668 $attvalue, $mailbox);
1669 $styledef .= "background-image: url('$attvalue'); ";
1672 $has_bgc_stl = true;
1673 $styledef .= "background-color: $attvalue; ";
1676 $has_txt_stl = true;
1677 $styledef .= "color: $attvalue; ";
1681 // Outlook defines a white bgcolor and no text color. This can lead to
1682 // white text on a white bg with certain themes.
1683 if ($has_bgc_stl && !$has_txt_stl) {
1684 $styledef .= "color: $text; ";
1686 if (strlen($styledef) > 0){
1687 $divattary{"style"} = "\"$styledef\"";
1694 * This is the main function and the one you should actually be calling.
1695 * There are several variables you should be aware of an which need
1696 * special description.
1698 * Since the description is quite lengthy, see it here:
1699 * http://linux.duke.edu/projects/mini/htmlfilter/
1701 * @param $body the string with HTML you wish to filter
1702 * @param $tag_list see description above
1703 * @param $rm_tags_with_content see description above
1704 * @param $self_closing_tags see description above
1705 * @param $force_tag_closing see description above
1706 * @param $rm_attnames see description above
1707 * @param $bad_attvals see description above
1708 * @param $add_attr_to_tag see description above
1709 * @param $message message object
1710 * @param $id message id
1711 * @return sanitized html safe to show on your pages.
1713 function sq_sanitize($body,
1715 $rm_tags_with_content,
1725 $me = 'sq_sanitize';
1726 $rm_tags = array_shift($tag_list);
1728 * Normalize rm_tags and rm_tags_with_content.
1730 @array_walk
($tag_list, 'sq_casenormalize');
1731 @array_walk
($rm_tags_with_content, 'sq_casenormalize');
1732 @array_walk
($self_closing_tags, 'sq_casenormalize');
1734 * See if tag_list is of tags to remove or tags to allow.
1735 * false means remove these tags
1736 * true means allow these tags
1739 $open_tags = Array();
1740 $trusted = "\n<!-- begin sanitized html -->\n";
1741 $skip_content = false;
1743 * Take care of netscape's stupid javascript entities like
1746 $body = preg_replace("/&(\{.*?\};)/si", "&\\1", $body);
1748 while (($curtag = sq_getnxtag($body, $curpos)) != FALSE){
1749 list($tagname, $attary, $tagtype, $lt, $gt) = $curtag;
1750 $free_content = substr($body, $curpos, $lt-$curpos);
1752 * Take care of <style>
1754 if ($tagname == "style" && $tagtype == 1){
1755 list($free_content, $curpos) =
1756 sq_fixstyle($body, $gt+
1, $message, $id, $mailbox);
1757 if ($free_content != FALSE){
1758 $trusted .= sq_tagprint($tagname, $attary, $tagtype);
1759 $trusted .= $free_content;
1760 $trusted .= sq_tagprint($tagname, false, 2);
1764 if ($skip_content == false){
1765 $trusted .= $free_content;
1767 if ($tagname != FALSE){
1769 if ($skip_content == $tagname){
1771 * Got to the end of tag we needed to remove.
1774 $skip_content = false;
1776 if ($skip_content == false){
1777 if ($tagname == "body"){
1780 if (isset($open_tags{$tagname}) &&
1781 $open_tags{$tagname} > 0){
1782 $open_tags{$tagname}--;
1790 * $rm_tags_with_content
1792 if ($skip_content == false){
1794 * See if this is a self-closing type and change
1795 * tagtype appropriately.
1798 && in_array($tagname, $self_closing_tags)){
1802 * See if we should skip this tag and any content
1805 if ($tagtype == 1 &&
1806 in_array($tagname, $rm_tags_with_content)){
1807 $skip_content = $tagname;
1809 if (($rm_tags == false
1810 && in_array($tagname, $tag_list)) ||
1811 ($rm_tags == true &&
1812 !in_array($tagname, $tag_list))){
1816 * Convert body into div.
1818 if ($tagname == "body"){
1820 $attary = sq_body2div($attary, $mailbox,
1824 if (isset($open_tags{$tagname})){
1825 $open_tags{$tagname}++
;
1827 $open_tags{$tagname}=1;
1831 * This is where we run other checks.
1833 if (is_array($attary) && sizeof($attary) > 0){
1834 $attary = sq_fixatts($tagname,
1848 if ($tagname != false && $skip_content == false){
1849 $trusted .= sq_tagprint($tagname, $attary, $tagtype);
1854 $trusted .= substr($body, $curpos, strlen($body)-$curpos);
1855 if ($force_tag_closing == true){
1856 foreach ($open_tags as $tagname=>$opentimes){
1857 while ($opentimes > 0){
1858 $trusted .= '</' . $tagname . '>';
1864 $trusted .= "<!-- end sanitized html -->\n";
1869 * This is a wrapper function to call html sanitizing routines.
1871 * @param $body the body of the message
1872 * @param $id the id of the message
1875 * @param boolean $take_mailto_links When TRUE, converts mailto: links
1876 * into internal SM compose links
1877 * (optional; default = TRUE)
1878 * @return a string with html safe to display in the browser.
1880 function magicHTML($body, $id, $message, $mailbox = 'INBOX', $take_mailto_links = true) {
1882 require_once(SM_PATH
. 'functions/url_parser.php'); // for $MailTo_PReg_Match
1884 global $attachment_common_show_images, $view_unsafe_images,
1888 * Don't display attached images in HTML mode.
1890 $attachment_common_show_images = false;
1905 $rm_tags_with_content = Array(
1914 $self_closing_tags = Array(
1922 $force_tag_closing = true;
1924 $rm_attnames = Array(
1935 $secremoveimg = "../images/" . _("sec_remove_eng.png");
1936 $bad_attvals = Array(
1939 "/^src|background/i" =>
1942 "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si",
1943 "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si",
1944 "/^([\'\"])\s*about\s*:.*([\'\"])/si"
1947 "\\1$secremoveimg\\2",
1948 "\\1$secremoveimg\\2",
1949 "\\1$secremoveimg\\2",
1950 "\\1$secremoveimg\\2"
1953 "/^href|action/i" =>
1956 "/^([\'\"])\s*\S+script\s*:.*([\'\"])/si",
1957 "/^([\'\"])\s*mocha\s*:*.*([\'\"])/si",
1958 "/^([\'\"])\s*about\s*:.*([\'\"])/si"
1973 "/include-source/i",
1974 "/position\s*:\s*absolute/i",
1975 "/url\s*\(\s*([\'\"])\s*\S+script\s*:.*([\'\"])\s*\)/si",
1976 "/url\s*\(\s*([\'\"])\s*mocha\s*:.*([\'\"])\s*\)/si",
1977 "/url\s*\(\s*([\'\"])\s*about\s*:.*([\'\"])\s*\)/si",
1978 "/(.*)\s*:\s*url\s*\(\s*([\'\"]*)\s*\S+script\s*:.*([\'\"]*)\s*\)/si"
1995 if( !sqgetGlobalVar('view_unsafe_images', $view_unsafe_images, SQ_GET
) ) {
1996 $view_unsafe_images = false;
1998 if (!$view_unsafe_images){
2000 * Remove any references to http/https if view_unsafe_images set
2003 array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[0],
2004 '/^([\'\"])\s*https*:.*([\'\"])/si');
2005 array_push($bad_attvals{'/.*/'}{'/^src|background/i'}[1],
2006 "\\1$secremoveimg\\1");
2007 array_push($bad_attvals{'/.*/'}{'/^style/i'}[0],
2008 '/url\(([\'\"])\s*https*:.*([\'\"])\)/si');
2009 array_push($bad_attvals{'/.*/'}{'/^style/i'}[1],
2010 "url(\\1$secremoveimg\\1)");
2013 $add_attr_to_tag = Array(
2015 Array('target'=>'"_blank"',
2016 'title'=>'"'._("This external link will open in a new window").'"'
2019 $trusted = sq_sanitize($body,
2021 $rm_tags_with_content,
2031 if (preg_match("|$secremoveimg|i", $trusted)){
2032 $has_unsafe_images = true;
2036 // we want to parse mailto's in HTML output, change to SM compose links
2037 // this is a modified version of code from url_parser.php... but Marc is
2038 // right: we need a better filtering implementation; adding this randomly
2039 // here is not a great solution
2041 if ($take_mailto_links) {
2042 // parseUrl($trusted); // this even parses URLs inside of tags... too aggressive
2043 global $MailTo_PReg_Match;
2044 $MailTo_PReg_Match = '/mailto:' . substr($MailTo_PReg_Match, 1);
2045 if ((preg_match_all($MailTo_PReg_Match, $trusted, $regs)) && ($regs[0][0] != '')) {
2046 foreach ($regs[0] as $i => $mailto_before) {
2047 $mailto_params = $regs[10][$i];
2049 // get rid of any tailing quote since we have to add send_to to the end
2051 if (substr($mailto_before, strlen($mailto_before) - 1) == '"')
2052 $mailto_before = substr($mailto_before, 0, strlen($mailto_before) - 1);
2053 if (substr($mailto_params, strlen($mailto_params) - 1) == '"')
2054 $mailto_params = substr($mailto_params, 0, strlen($mailto_params) - 1);
2056 if ($regs[1][$i]) { //if there is an email addr before '?', we need to merge it with the params
2057 $to = 'to=' . $regs[1][$i];
2058 if (strpos($mailto_params, 'to=') > -1) //already a 'to='
2059 $mailto_params = str_replace('to=', $to . '%2C%20', $mailto_params);
2061 if ($mailto_params) //already some params, append to them
2062 $mailto_params .= '&' . $to;
2064 $mailto_params .= '?' . $to;
2068 $url_str = preg_replace(array('/to=/i', '/(?<!b)cc=/i', '/bcc=/i'), array('send_to=', 'send_to_cc=', 'send_to_bcc='), $mailto_params);
2070 // we'll already have target=_blank, no need to allow comp_in_new
2071 // here (which would be a lot more work anyway)
2073 global $compose_new_win;
2074 $temp_comp_in_new = $compose_new_win;
2075 $compose_new_win = 0;
2076 $comp_uri = makeComposeLink('src/compose.php' . $url_str, $mailto_before);
2077 $compose_new_win = $temp_comp_in_new;
2079 // remove <a href=" and anything after the next quote (we only
2080 // need the uri, not the link HTML) in compose uri
2082 $comp_uri = substr($comp_uri, 9);
2083 $comp_uri = substr($comp_uri, 0, strpos($comp_uri, '"', 1));
2084 $trusted = str_replace($mailto_before, $comp_uri, $trusted);
2093 * function SendDownloadHeaders - send file to the browser
2095 * Original Source: SM core src/download.php
2096 * moved here to make it available to other code, and separate
2097 * front end from back end functionality.
2099 * @param string $type0 first half of mime type
2100 * @param string $type1 second half of mime type
2101 * @param string $filename filename to tell the browser for downloaded file
2102 * @param boolean $force whether to force the download dialog to pop
2103 * @param optional integer $filesize send the Content-Header and length to the browser
2106 function SendDownloadHeaders($type0, $type1, $filename, $force, $filesize=0) {
2107 global $languages, $squirrelmail_language;
2110 sqgetGlobalVar('HTTP_USER_AGENT', $HTTP_USER_AGENT, SQ_SERVER
);
2112 if (strstr($HTTP_USER_AGENT, 'compatible; MSIE ') !== false &&
2113 strstr($HTTP_USER_AGENT, 'Opera') === false) {
2117 if (strstr($HTTP_USER_AGENT, 'compatible; MSIE 6') !== false &&
2118 strstr($HTTP_USER_AGENT, 'Opera') === false) {
2122 if (isset($languages[$squirrelmail_language]['XTRA_CODE']) &&
2123 function_exists($languages[$squirrelmail_language]['XTRA_CODE'] . '_downloadfilename')) {
2125 call_user_func($languages[$squirrelmail_language]['XTRA_CODE'] . '_downloadfilename', $filename, $HTTP_USER_AGENT);
2127 $filename = ereg_replace('[\\/:\*\?"<>\|;]', '_', str_replace(' ', ' ', $filename));
2130 // A Pox on Microsoft and it's Internet Explorer!
2132 // IE has lots of bugs with file downloads.
2133 // It also has problems with SSL. Both of these cause problems
2134 // for us in this function.
2136 // See this article on Cache Control headers and SSL
2137 // http://support.microsoft.com/default.aspx?scid=kb;en-us;323308
2139 // The best thing you can do for IE is to upgrade to the latest
2141 //set all the Cache Control Headers for IE
2143 $filename=rawurlencode($filename);
2144 header ("Pragma: public");
2145 header ("Cache-Control: no-store, max-age=0, no-cache, must-revalidate"); # HTTP/1.1
2146 header ("Cache-Control: post-check=0, pre-check=0", false);
2147 header ("Cache-control: private");
2149 //set the inline header for IE, we'll add the attachment header later if we need it
2150 header ("Content-Disposition: inline; filename=$filename");
2154 // Try to show in browser window
2155 header ("Content-Disposition: inline; filename=\"$filename\"");
2156 header ("Content-Type: $type0/$type1; name=\"$filename\"");
2158 // Try to pop up the "save as" box
2160 // IE makes this hard. It pops up 2 save boxes, or none.
2161 // http://support.microsoft.com/support/kb/articles/Q238/5/88.ASP
2162 // http://support.microsoft.com/default.aspx?scid=kb;EN-US;260519
2163 // But, according to Microsoft, it is "RFC compliant but doesn't
2164 // take into account some deviations that allowed within the
2165 // specification." Doesn't that mean RFC non-compliant?
2166 // http://support.microsoft.com/support/kb/articles/Q258/4/52.ASP
2168 // all browsers need the application/octet-stream header for this
2169 header ("Content-Type: application/octet-stream; name=\"$filename\"");
2171 // http://support.microsoft.com/support/kb/articles/Q182/3/15.asp
2172 // Do not have quotes around filename, but that applied to
2173 // "attachment"... does it apply to inline too?
2174 header ("Content-Disposition: attachment; filename=\"$filename\"");
2176 if ($isIE && !$isIE6) {
2177 // This combination seems to work mostly. IE 5.5 SP 1 has
2178 // known issues (see the Microsoft Knowledge Base)
2180 // This works for most types, but doesn't work with Word files
2181 header ("Content-Type: application/download; name=\"$filename\"");
2183 // These are spares, just in case. :-)
2184 //header("Content-Type: $type0/$type1; name=\"$filename\"");
2185 //header("Content-Type: application/x-msdownload; name=\"$filename\"");
2186 //header("Content-Type: application/octet-stream; name=\"$filename\"");
2188 // another application/octet-stream forces download for Netscape
2189 header ("Content-Type: application/octet-stream; name=\"$filename\"");
2193 //send the content-length header if the calling function provides it
2194 if ($filesize > 0) {
2195 header("Content-Length: $filesize");
2198 } // end fn SendDownloadHeaders