3 * kses 0.2.2 - HTML/XHTML filter that only allows some elements and attributes
4 * Copyright (C) 2002, 2003, 2005 Ulf Harnhammar
6 * This program is free software and open source software; you can redistribute
7 * it and/or modify it under the terms of the GNU General Public License as
8 * published by the Free Software Foundation; either version 2 of the License,
9 * or (at your option) any later version.
11 * This program is distributed in the hope that it will be useful, but WITHOUT
12 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
13 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
16 * You should have received a copy of the GNU General Public License along
17 * with this program; if not, write to the Free Software Foundation, Inc.,
18 * 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA
19 * http://www.gnu.org/licenses/gpl.html
21 * [kses strips evil scripts!]
23 * Added wp_ prefix to avoid conflicts with existing kses users
26 * @copyright (C) 2002, 2003, 2005
27 * @author Ulf Harnhammar <http://advogato.org/person/metaur/>
35 * You can override this in a plugin.
37 * The wp_kses_allowed_html filter is more powerful and supplies context.
38 * CUSTOM_TAGS is not recommended and should be considered deprecated.
40 * @see wp_kses_allowed_html()
44 if ( ! defined( 'CUSTOM_TAGS' ) )
45 define( 'CUSTOM_TAGS', false );
47 if ( ! CUSTOM_TAGS ) {
49 * Kses global for default allowable HTML tags.
51 * Can be override by using CUSTOM_TAGS constant.
53 * @global array $allowedposttags
56 $allowedposttags = array(
89 'blockquote' => array(
138 'fieldset' => array(),
145 'figcaption' => array(
165 'accept-charset' => true,
287 'cellpadding' => true,
288 'cellspacing' => true,
373 * Kses allowed HTML elements.
375 * @global array $allowedtags
378 $allowedtags = array(
390 'blockquote' => array(
407 $allowedentitynames = array(
408 'nbsp', 'iexcl', 'cent', 'pound', 'curren', 'yen',
409 'brvbar', 'sect', 'uml', 'copy', 'ordf', 'laquo',
410 'not', 'shy', 'reg', 'macr', 'deg', 'plusmn',
411 'acute', 'micro', 'para', 'middot', 'cedil', 'ordm',
412 'raquo', 'iquest', 'Agrave', 'Aacute', 'Acirc', 'Atilde',
413 'Auml', 'Aring', 'AElig', 'Ccedil', 'Egrave', 'Eacute',
414 'Ecirc', 'Euml', 'Igrave', 'Iacute', 'Icirc', 'Iuml',
415 'ETH', 'Ntilde', 'Ograve', 'Oacute', 'Ocirc', 'Otilde',
416 'Ouml', 'times', 'Oslash', 'Ugrave', 'Uacute', 'Ucirc',
417 'Uuml', 'Yacute', 'THORN', 'szlig', 'agrave', 'aacute',
418 'acirc', 'atilde', 'auml', 'aring', 'aelig', 'ccedil',
419 'egrave', 'eacute', 'ecirc', 'euml', 'igrave', 'iacute',
420 'icirc', 'iuml', 'eth', 'ntilde', 'ograve', 'oacute',
421 'ocirc', 'otilde', 'ouml', 'divide', 'oslash', 'ugrave',
422 'uacute', 'ucirc', 'uuml', 'yacute', 'thorn', 'yuml',
423 'quot', 'amp', 'lt', 'gt', 'apos', 'OElig',
424 'oelig', 'Scaron', 'scaron', 'Yuml', 'circ', 'tilde',
425 'ensp', 'emsp', 'thinsp', 'zwnj', 'zwj', 'lrm',
426 'rlm', 'ndash', 'mdash', 'lsquo', 'rsquo', 'sbquo',
427 'ldquo', 'rdquo', 'bdquo', 'dagger', 'Dagger', 'permil',
428 'lsaquo', 'rsaquo', 'euro', 'fnof', 'Alpha', 'Beta',
429 'Gamma', 'Delta', 'Epsilon', 'Zeta', 'Eta', 'Theta',
430 'Iota', 'Kappa', 'Lambda', 'Mu', 'Nu', 'Xi',
431 'Omicron', 'Pi', 'Rho', 'Sigma', 'Tau', 'Upsilon',
432 'Phi', 'Chi', 'Psi', 'Omega', 'alpha', 'beta',
433 'gamma', 'delta', 'epsilon', 'zeta', 'eta', 'theta',
434 'iota', 'kappa', 'lambda', 'mu', 'nu', 'xi',
435 'omicron', 'pi', 'rho', 'sigmaf', 'sigma', 'tau',
436 'upsilon', 'phi', 'chi', 'psi', 'omega', 'thetasym',
437 'upsih', 'piv', 'bull', 'hellip', 'prime', 'Prime',
438 'oline', 'frasl', 'weierp', 'image', 'real', 'trade',
439 'alefsym', 'larr', 'uarr', 'rarr', 'darr', 'harr',
440 'crarr', 'lArr', 'uArr', 'rArr', 'dArr', 'hArr',
441 'forall', 'part', 'exist', 'empty', 'nabla', 'isin',
442 'notin', 'ni', 'prod', 'sum', 'minus', 'lowast',
443 'radic', 'prop', 'infin', 'ang', 'and', 'or',
444 'cap', 'cup', 'int', 'sim', 'cong', 'asymp',
445 'ne', 'equiv', 'le', 'ge', 'sub', 'sup',
446 'nsub', 'sube', 'supe', 'oplus', 'otimes', 'perp',
447 'sdot', 'lceil', 'rceil', 'lfloor', 'rfloor', 'lang',
448 'rang', 'loz', 'spades', 'clubs', 'hearts', 'diams',
451 $allowedposttags = array_map( '_wp_add_global_attributes', $allowedposttags );
453 $allowedtags = wp_kses_array_lc( $allowedtags );
454 $allowedposttags = wp_kses_array_lc( $allowedposttags );
458 * Filters content and keeps only allowable HTML elements.
460 * This function makes sure that only the allowed HTML element names, attribute
461 * names and attribute values plus only sane HTML entities will occur in
462 * $string. You have to remove any slashes from PHP's magic quotes before you
463 * call this function.
465 * The default allowed protocols are 'http', 'https', 'ftp', 'mailto', 'news',
466 * 'irc', 'gopher', 'nntp', 'feed', 'telnet, 'mms', 'rtsp' and 'svn'. This
467 * covers all common link protocols, except for 'javascript' which should not
468 * be allowed for untrusted users.
472 * @param string $string Content to filter through kses
473 * @param array $allowed_html List of allowed HTML elements
474 * @param array $allowed_protocols Optional. Allowed protocol in links.
475 * @return string Filtered content with only allowed HTML elements
477 function wp_kses( $string, $allowed_html, $allowed_protocols = array() ) {
478 if ( empty( $allowed_protocols ) )
479 $allowed_protocols = wp_allowed_protocols();
480 $string = wp_kses_no_null($string);
481 $string = wp_kses_js_entities($string);
482 $string = wp_kses_normalize_entities($string);
483 $string = wp_kses_hook($string, $allowed_html, $allowed_protocols); // WP changed the order of these funcs and added args to wp_kses_hook
484 return wp_kses_split($string, $allowed_html, $allowed_protocols);
488 * Return a list of allowed tags and attributes for a given context.
492 * @param string $context The context for which to retrieve tags. Allowed values are
493 * post | strip | data | entities or the name of a field filter such as pre_user_description.
494 * @return array List of allowed tags and their allowed attributes.
496 function wp_kses_allowed_html( $context = '' ) {
497 global $allowedposttags, $allowedtags, $allowedentitynames;
499 if ( is_array( $context ) )
500 return apply_filters( 'wp_kses_allowed_html', $context, 'explicit' );
502 switch ( $context ) {
504 return apply_filters( 'wp_kses_allowed_html', $allowedposttags, $context );
506 case 'user_description':
507 case 'pre_user_description':
508 $tags = $allowedtags;
509 $tags['a']['rel'] = true;
510 return apply_filters( 'wp_kses_allowed_html', $tags, $context );
513 return apply_filters( 'wp_kses_allowed_html', array(), $context );
516 return apply_filters( 'wp_kses_allowed_html', $allowedentitynames, $context);
520 return apply_filters( 'wp_kses_allowed_html', $allowedtags, $context );
525 * You add any kses hooks here.
527 * There is currently only one kses WordPress hook and it is called here. All
528 * parameters are passed to the hooks and expected to receive a string.
532 * @param string $string Content to filter through kses
533 * @param array $allowed_html List of allowed HTML elements
534 * @param array $allowed_protocols Allowed protocol in links
535 * @return string Filtered content through 'pre_kses' hook
537 function wp_kses_hook( $string, $allowed_html, $allowed_protocols ) {
538 $string = apply_filters('pre_kses', $string, $allowed_html, $allowed_protocols);
543 * This function returns kses' version number.
547 * @return string KSES Version Number
549 function wp_kses_version() {
554 * Searches for HTML tags, no matter how malformed.
556 * It also matches stray ">" characters.
560 * @param string $string Content to filter
561 * @param array $allowed_html Allowed HTML elements
562 * @param array $allowed_protocols Allowed protocols to keep
563 * @return string Content with fixed HTML tags
565 function wp_kses_split( $string, $allowed_html, $allowed_protocols ) {
566 global $pass_allowed_html, $pass_allowed_protocols;
567 $pass_allowed_html = $allowed_html;
568 $pass_allowed_protocols = $allowed_protocols;
569 return preg_replace_callback( '%(<!--.*?(-->|$))|(<[^>]*(>|$)|>)%', '_wp_kses_split_callback', $string );
573 * Callback for wp_kses_split.
578 function _wp_kses_split_callback( $match ) {
579 global $pass_allowed_html, $pass_allowed_protocols;
580 return wp_kses_split2( $match[0], $pass_allowed_html, $pass_allowed_protocols );
584 * Callback for wp_kses_split for fixing malformed HTML tags.
586 * This function does a lot of work. It rejects some very malformed things like
587 * <:::>. It returns an empty string, if the element isn't allowed (look ma, no
588 * strip_tags()!). Otherwise it splits the tag into an element and an attribute
591 * After the tag is split into an element and an attribute list, it is run
592 * through another filter which will remove illegal attributes and once that is
593 * completed, will be returned.
597 * @uses wp_kses_attr()
599 * @param string $string Content to filter
600 * @param array $allowed_html Allowed HTML elements
601 * @param array $allowed_protocols Allowed protocols to keep
602 * @return string Fixed HTML element
604 function wp_kses_split2($string, $allowed_html, $allowed_protocols) {
605 $string = wp_kses_stripslashes($string);
607 if (substr($string, 0, 1) != '<')
609 # It matched a ">" character
611 if ( '<!--' == substr( $string, 0, 4 ) ) {
612 $string = str_replace( array('<!--', '-->'), '', $string );
613 while ( $string != ($newstring = wp_kses($string, $allowed_html, $allowed_protocols)) )
614 $string = $newstring;
617 // prevent multiple dashes in comments
618 $string = preg_replace('/--+/', '-', $string);
619 // prevent three dashes closing a comment
620 $string = preg_replace('/-$/', '', $string);
621 return "<!--{$string}-->";
623 # Allow HTML comments
625 if (!preg_match('%^<\s*(/\s*)?([a-zA-Z0-9]+)([^>]*)>?$%', $string, $matches))
627 # It's seriously malformed
629 $slash = trim($matches[1]);
631 $attrlist = $matches[3];
633 if ( ! is_array( $allowed_html ) )
634 $allowed_html = wp_kses_allowed_html( $allowed_html );
636 if ( ! isset($allowed_html[strtolower($elem)]) )
638 # They are using a not allowed HTML element
642 # No attributes are allowed for closing elements
644 return wp_kses_attr( $elem, $attrlist, $allowed_html, $allowed_protocols );
648 * Removes all attributes, if none are allowed for this element.
650 * If some are allowed it calls wp_kses_hair() to split them further, and then
651 * it builds up new HTML code from the data that kses_hair() returns. It also
652 * removes "<" and ">" characters, if there are any left. One more thing it does
653 * is to check if the tag has a closing XHTML slash, and if it does, it puts one
654 * in the returned code as well.
658 * @param string $element HTML element/tag
659 * @param string $attr HTML attributes from HTML element to closing HTML element tag
660 * @param array $allowed_html Allowed HTML elements
661 * @param array $allowed_protocols Allowed protocols to keep
662 * @return string Sanitized HTML element
664 function wp_kses_attr($element, $attr, $allowed_html, $allowed_protocols) {
665 # Is there a closing XHTML slash at the end of the attributes?
667 if ( ! is_array( $allowed_html ) )
668 $allowed_html = wp_kses_allowed_html( $allowed_html );
671 if (preg_match('%\s*/\s*$%', $attr))
674 # Are any attributes allowed at all for this element?
675 if ( ! isset($allowed_html[strtolower($element)]) || count($allowed_html[strtolower($element)]) == 0 )
676 return "<$element$xhtml_slash>";
679 $attrarr = wp_kses_hair($attr, $allowed_protocols);
681 # Go through $attrarr, and save the allowed attributes for this element
685 $allowed_attr = $allowed_html[strtolower($element)];
686 foreach ($attrarr as $arreach) {
687 if ( ! isset( $allowed_attr[strtolower($arreach['name'])] ) )
688 continue; # the attribute is not allowed
690 $current = $allowed_attr[strtolower($arreach['name'])];
691 if ( $current == '' )
692 continue; # the attribute is not allowed
694 if ( strtolower( $arreach['name'] ) == 'style' ) {
695 $orig_value = $arreach['value'];
696 $value = safecss_filter_attr( $orig_value );
698 if ( empty( $value ) )
701 $arreach['value'] = $value;
702 $arreach['whole'] = str_replace( $orig_value, $value, $arreach['whole'] );
705 if ( ! is_array($current) ) {
706 $attr2 .= ' '.$arreach['whole'];
707 # there are no checks
710 # there are some checks
712 foreach ($current as $currkey => $currval) {
713 if ( ! wp_kses_check_attr_val($arreach['value'], $arreach['vless'], $currkey, $currval) ) {
720 $attr2 .= ' '.$arreach['whole']; # it passed them
721 } # if !is_array($current)
724 # Remove any "<" or ">" characters
725 $attr2 = preg_replace('/[<>]/', '', $attr2);
727 return "<$element$attr2$xhtml_slash>";
731 * Builds an attribute list from string containing attributes.
733 * This function does a lot of work. It parses an attribute list into an array
734 * with attribute data, and tries to do the right thing even if it gets weird
735 * input. It will add quotes around attribute values that don't have any quotes
736 * or apostrophes around them, to make it easier to produce HTML code that will
737 * conform to W3C's HTML specification. It will also remove bad URL protocols
738 * from attribute values. It also reduces duplicate attributes by using the
739 * attribute defined first (foo='bar' foo='baz' will result in foo='bar').
743 * @param string $attr Attribute list from HTML element to closing HTML element tag
744 * @param array $allowed_protocols Allowed protocols to keep
745 * @return array List of attributes after parsing
747 function wp_kses_hair($attr, $allowed_protocols) {
751 $uris = array('xmlns', 'profile', 'href', 'src', 'cite', 'classid', 'codebase', 'data', 'usemap', 'longdesc', 'action');
753 # Loop through the whole attribute list
755 while (strlen($attr) != 0) {
756 $working = 0; # Was the last operation successful?
759 case 0 : # attribute name, href for instance
761 if (preg_match('/^([-a-zA-Z]+)/', $attr, $match)) {
762 $attrname = $match[1];
763 $working = $mode = 1;
764 $attr = preg_replace('/^[-a-zA-Z]+/', '', $attr);
769 case 1 : # equals sign or valueless ("selected")
771 if (preg_match('/^\s*=\s*/', $attr)) # equals sign
775 $attr = preg_replace('/^\s*=\s*/', '', $attr);
779 if (preg_match('/^\s+/', $attr)) # valueless
783 if(false === array_key_exists($attrname, $attrarr)) {
784 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
786 $attr = preg_replace('/^\s+/', '', $attr);
791 case 2 : # attribute value, a URL after href= for instance
793 if (preg_match('%^"([^"]*)"(\s+|/?$)%', $attr, $match))
796 $thisval = $match[1];
797 if ( in_array(strtolower($attrname), $uris) )
798 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
800 if(false === array_key_exists($attrname, $attrarr)) {
801 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n');
805 $attr = preg_replace('/^"[^"]*"(\s+|$)/', '', $attr);
809 if (preg_match("%^'([^']*)'(\s+|/?$)%", $attr, $match))
812 $thisval = $match[1];
813 if ( in_array(strtolower($attrname), $uris) )
814 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
816 if(false === array_key_exists($attrname, $attrarr)) {
817 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname='$thisval'", 'vless' => 'n');
821 $attr = preg_replace("/^'[^']*'(\s+|$)/", '', $attr);
825 if (preg_match("%^([^\s\"']+)(\s+|/?$)%", $attr, $match))
828 $thisval = $match[1];
829 if ( in_array(strtolower($attrname), $uris) )
830 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
832 if(false === array_key_exists($attrname, $attrarr)) {
833 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n');
835 # We add quotes to conform to W3C's HTML spec.
838 $attr = preg_replace("%^[^\s\"']+(\s+|$)%", '', $attr);
844 if ($working == 0) # not well formed, remove and try again
846 $attr = wp_kses_html_error($attr);
851 if ($mode == 1 && false === array_key_exists($attrname, $attrarr))
852 # special case, for when the attribute list ends with a valueless
853 # attribute like "selected"
854 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
860 * Performs different checks for attribute values.
862 * The currently implemented checks are "maxlen", "minlen", "maxval", "minval"
867 * @param string $value Attribute value
868 * @param string $vless Whether the value is valueless. Use 'y' or 'n'
869 * @param string $checkname What $checkvalue is checking for.
870 * @param mixed $checkvalue What constraint the value should pass
871 * @return bool Whether check passes
873 function wp_kses_check_attr_val($value, $vless, $checkname, $checkvalue) {
876 switch (strtolower($checkname)) {
878 # The maxlen check makes sure that the attribute value has a length not
879 # greater than the given value. This can be used to avoid Buffer Overflows
880 # in WWW clients and various Internet servers.
882 if (strlen($value) > $checkvalue)
887 # The minlen check makes sure that the attribute value has a length not
888 # smaller than the given value.
890 if (strlen($value) < $checkvalue)
895 # The maxval check does two things: it checks that the attribute value is
896 # an integer from 0 and up, without an excessive amount of zeroes or
897 # whitespace (to avoid Buffer Overflows). It also checks that the attribute
898 # value is not greater than the given value.
899 # This check can be used to avoid Denial of Service attacks.
901 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
903 if ($value > $checkvalue)
908 # The minval check makes sure that the attribute value is a positive integer,
909 # and that it is not smaller than the given value.
911 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
913 if ($value < $checkvalue)
918 # The valueless check makes sure if the attribute has a value
919 # (like <a href="blah">) or not (<option selected>). If the given value
920 # is a "y" or a "Y", the attribute must not have a value.
921 # If the given value is an "n" or an "N", the attribute must have one.
923 if (strtolower($checkvalue) != $vless)
932 * Sanitize string from bad protocols.
934 * This function removes all non-allowed protocols from the beginning of
935 * $string. It ignores whitespace and the case of the letters, and it does
936 * understand HTML entities. It does its work in a while loop, so it won't be
937 * fooled by a string like "javascript:javascript:alert(57)".
941 * @param string $string Content to filter bad protocols from
942 * @param array $allowed_protocols Allowed protocols to keep
943 * @return string Filtered content
945 function wp_kses_bad_protocol($string, $allowed_protocols) {
946 $string = wp_kses_no_null($string);
950 $original_string = $string;
951 $string = wp_kses_bad_protocol_once($string, $allowed_protocols);
952 } while ( $original_string != $string && ++$iterations < 6 );
954 if ( $original_string != $string )
961 * Removes any null characters in $string.
965 * @param string $string
968 function wp_kses_no_null($string) {
969 $string = preg_replace('/\0+/', '', $string);
970 $string = preg_replace('/(\\\\0)+/', '', $string);
976 * Strips slashes from in front of quotes.
978 * This function changes the character sequence \" to just ". It leaves all
979 * other slashes alone. It's really weird, but the quoting from
980 * preg_replace(//e) seems to require this.
984 * @param string $string String to strip slashes
985 * @return string Fixed string with quoted slashes
987 function wp_kses_stripslashes($string) {
988 return preg_replace('%\\\\"%', '"', $string);
992 * Goes through an array and changes the keys to all lower case.
996 * @param array $inarray Unfiltered array
997 * @return array Fixed array with all lowercase keys
999 function wp_kses_array_lc($inarray) {
1000 $outarray = array ();
1002 foreach ( (array) $inarray as $inkey => $inval) {
1003 $outkey = strtolower($inkey);
1004 $outarray[$outkey] = array ();
1006 foreach ( (array) $inval as $inkey2 => $inval2) {
1007 $outkey2 = strtolower($inkey2);
1008 $outarray[$outkey][$outkey2] = $inval2;
1010 } # foreach $inarray
1016 * Removes the HTML JavaScript entities found in early versions of Netscape 4.
1020 * @param string $string
1023 function wp_kses_js_entities($string) {
1024 return preg_replace('%&\s*\{[^}]*(\}\s*;?|$)%', '', $string);
1028 * Handles parsing errors in wp_kses_hair().
1030 * The general plan is to remove everything to and including some whitespace,
1031 * but it deals with quotes and apostrophes as well.
1035 * @param string $string
1038 function wp_kses_html_error($string) {
1039 return preg_replace('/^("[^"]*("|$)|\'[^\']*(\'|$)|\S)*\s*/', '', $string);
1043 * Sanitizes content from bad protocols and other characters.
1045 * This function searches for URL protocols at the beginning of $string, while
1046 * handling whitespace and HTML entities.
1050 * @param string $string Content to check for bad protocols
1051 * @param string $allowed_protocols Allowed protocols
1052 * @return string Sanitized content
1054 function wp_kses_bad_protocol_once($string, $allowed_protocols, $count = 1 ) {
1055 $string2 = preg_split( '/:|�*58;|�*3a;/i', $string, 2 );
1056 if ( isset($string2[1]) && ! preg_match('%/\?%', $string2[0]) ) {
1057 $string = trim( $string2[1] );
1058 $protocol = wp_kses_bad_protocol_once2( $string2[0], $allowed_protocols );
1059 if ( 'feed:' == $protocol ) {
1062 $string = wp_kses_bad_protocol_once( $string, $allowed_protocols, ++$count );
1063 if ( empty( $string ) )
1066 $string = $protocol . $string;
1073 * Callback for wp_kses_bad_protocol_once() regular expression.
1075 * This function processes URL protocols, checks to see if they're in the
1076 * whitelist or not, and returns different data depending on the answer.
1081 * @param string $string URI scheme to check against the whitelist
1082 * @param string $allowed_protocols Allowed protocols
1083 * @return string Sanitized content
1085 function wp_kses_bad_protocol_once2( $string, $allowed_protocols ) {
1086 $string2 = wp_kses_decode_entities($string);
1087 $string2 = preg_replace('/\s/', '', $string2);
1088 $string2 = wp_kses_no_null($string2);
1089 $string2 = strtolower($string2);
1092 foreach ( (array) $allowed_protocols as $one_protocol )
1093 if ( strtolower($one_protocol) == $string2 ) {
1105 * Converts and fixes HTML entities.
1107 * This function normalizes HTML entities. It will convert "AT&T" to the correct
1108 * "AT&T", ":" to ":", "&#XYZZY;" to "&#XYZZY;" and so on.
1112 * @param string $string Content to normalize entities
1113 * @return string Content with normalized entities
1115 function wp_kses_normalize_entities($string) {
1116 # Disarm all entities by converting & to &
1118 $string = str_replace('&', '&', $string);
1120 # Change back the allowed entities in our entity whitelist
1122 $string = preg_replace_callback('/&([A-Za-z]{2,8});/', 'wp_kses_named_entities', $string);
1123 $string = preg_replace_callback('/&#(0*[0-9]{1,7});/', 'wp_kses_normalize_entities2', $string);
1124 $string = preg_replace_callback('/&#[Xx](0*[0-9A-Fa-f]{1,6});/', 'wp_kses_normalize_entities3', $string);
1130 * Callback for wp_kses_normalize_entities() regular expression.
1132 * This function only accepts valid named entity references, which are finite,
1133 * case-sensitive, and highly scrutinized by HTML and XML validators.
1137 * @param array $matches preg_replace_callback() matches array
1138 * @return string Correctly encoded entity
1140 function wp_kses_named_entities($matches) {
1141 global $allowedentitynames;
1143 if ( empty($matches[1]) )
1147 return ( ( ! in_array($i, $allowedentitynames) ) ? "&$i;" : "&$i;" );
1151 * Callback for wp_kses_normalize_entities() regular expression.
1153 * This function helps wp_kses_normalize_entities() to only accept 16-bit values
1154 * and nothing more for &#number; entities.
1159 * @param array $matches preg_replace_callback() matches array
1160 * @return string Correctly encoded entity
1162 function wp_kses_normalize_entities2($matches) {
1163 if ( empty($matches[1]) )
1167 if (valid_unicode($i)) {
1168 $i = str_pad(ltrim($i,'0'), 3, '0', STR_PAD_LEFT);
1178 * Callback for wp_kses_normalize_entities() for regular expression.
1180 * This function helps wp_kses_normalize_entities() to only accept valid Unicode
1181 * numeric entities in hex form.
1185 * @param array $matches preg_replace_callback() matches array
1186 * @return string Correctly encoded entity
1188 function wp_kses_normalize_entities3($matches) {
1189 if ( empty($matches[1]) )
1192 $hexchars = $matches[1];
1193 return ( ( ! valid_unicode(hexdec($hexchars)) ) ? "&#x$hexchars;" : '&#x'.ltrim($hexchars,'0').';' );
1197 * Helper function to determine if a Unicode value is valid.
1199 * @param int $i Unicode value
1200 * @return bool True if the value was a valid Unicode number
1202 function valid_unicode($i) {
1203 return ( $i == 0x9 || $i == 0xa || $i == 0xd ||
1204 ($i >= 0x20 && $i <= 0xd7ff) ||
1205 ($i >= 0xe000 && $i <= 0xfffd) ||
1206 ($i >= 0x10000 && $i <= 0x10ffff) );
1210 * Convert all entities to their character counterparts.
1212 * This function decodes numeric HTML entities (A and A). It doesn't do
1213 * anything with other entities like ä, but we don't need them in the URL
1214 * protocol whitelisting system anyway.
1218 * @param string $string Content to change entities
1219 * @return string Content after decoded entities
1221 function wp_kses_decode_entities($string) {
1222 $string = preg_replace_callback('/&#([0-9]+);/', '_wp_kses_decode_entities_chr', $string);
1223 $string = preg_replace_callback('/&#[Xx]([0-9A-Fa-f]+);/', '_wp_kses_decode_entities_chr_hexdec', $string);
1229 * Regex callback for wp_kses_decode_entities()
1231 * @param array $match preg match
1234 function _wp_kses_decode_entities_chr( $match ) {
1235 return chr( $match[1] );
1239 * Regex callback for wp_kses_decode_entities()
1241 * @param array $match preg match
1244 function _wp_kses_decode_entities_chr_hexdec( $match ) {
1245 return chr( hexdec( $match[1] ) );
1249 * Sanitize content with allowed HTML Kses rules.
1252 * @uses $allowedtags
1254 * @param string $data Content to filter, expected to be escaped with slashes
1255 * @return string Filtered content
1257 function wp_filter_kses( $data ) {
1258 return addslashes( wp_kses( stripslashes( $data ), current_filter() ) );
1262 * Sanitize content with allowed HTML Kses rules.
1265 * @uses $allowedtags
1267 * @param string $data Content to filter, expected to not be escaped
1268 * @return string Filtered content
1270 function wp_kses_data( $data ) {
1271 return wp_kses( $data , current_filter() );
1275 * Sanitize content for allowed HTML tags for post content.
1277 * Post content refers to the page contents of the 'post' type and not $_POST
1282 * @param string $data Post content to filter, expected to be escaped with slashes
1283 * @return string Filtered post content with allowed HTML tags and attributes intact.
1285 function wp_filter_post_kses($data) {
1286 return addslashes ( wp_kses( stripslashes( $data ), 'post' ) );
1290 * Sanitize content for allowed HTML tags for post content.
1292 * Post content refers to the page contents of the 'post' type and not $_POST
1297 * @param string $data Post content to filter
1298 * @return string Filtered post content with allowed HTML tags and attributes intact.
1300 function wp_kses_post($data) {
1301 return wp_kses( $data , 'post' );
1305 * Strips all of the HTML in the content.
1309 * @param string $data Content to strip all HTML from
1310 * @return string Filtered content without any HTML
1312 function wp_filter_nohtml_kses( $data ) {
1313 return addslashes ( wp_kses( stripslashes( $data ), 'strip' ) );
1317 * Adds all Kses input form content filters.
1319 * All hooks have default priority. The wp_filter_kses() function is added to
1320 * the 'pre_comment_content' and 'title_save_pre' hooks.
1322 * The wp_filter_post_kses() function is added to the 'content_save_pre',
1323 * 'excerpt_save_pre', and 'content_filtered_save_pre' hooks.
1326 * @uses add_filter() See description for what functions are added to what hooks.
1328 function kses_init_filters() {
1330 add_filter('title_save_pre', 'wp_filter_kses');
1332 // Comment filtering
1333 if ( current_user_can( 'unfiltered_html' ) )
1334 add_filter( 'pre_comment_content', 'wp_filter_post_kses' );
1336 add_filter( 'pre_comment_content', 'wp_filter_kses' );
1339 add_filter('content_save_pre', 'wp_filter_post_kses');
1340 add_filter('excerpt_save_pre', 'wp_filter_post_kses');
1341 add_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1345 * Removes all Kses input form content filters.
1347 * A quick procedural method to removing all of the filters that kses uses for
1348 * content in WordPress Loop.
1350 * Does not remove the kses_init() function from 'init' hook (priority is
1351 * default). Also does not remove kses_init() function from 'set_current_user'
1352 * hook (priority is also default).
1356 function kses_remove_filters() {
1358 remove_filter('title_save_pre', 'wp_filter_kses');
1360 // Comment filtering
1361 remove_filter( 'pre_comment_content', 'wp_filter_post_kses' );
1362 remove_filter( 'pre_comment_content', 'wp_filter_kses' );
1365 remove_filter('content_save_pre', 'wp_filter_post_kses');
1366 remove_filter('excerpt_save_pre', 'wp_filter_post_kses');
1367 remove_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1371 * Sets up most of the Kses filters for input form content.
1373 * If you remove the kses_init() function from 'init' hook and
1374 * 'set_current_user' (priority is default), then none of the Kses filter hooks
1377 * First removes all of the Kses filters in case the current user does not need
1378 * to have Kses filter the content. If the user does not have unfiltered_html
1379 * capability, then Kses filters are added.
1381 * @uses kses_remove_filters() Removes the Kses filters
1382 * @uses kses_init_filters() Adds the Kses filters back if the user
1383 * does not have unfiltered HTML capability.
1386 function kses_init() {
1387 kses_remove_filters();
1389 if (current_user_can('unfiltered_html') == false)
1390 kses_init_filters();
1393 add_action('init', 'kses_init');
1394 add_action('set_current_user', 'kses_init');
1401 function safecss_filter_attr( $css, $deprecated = '' ) {
1402 if ( !empty( $deprecated ) )
1403 _deprecated_argument( __FUNCTION__, '2.8.1' ); // Never implemented
1405 $css = wp_kses_no_null($css);
1406 $css = str_replace(array("\n","\r","\t"), '', $css);
1408 if ( preg_match( '%[\\(&=}]|/\*%', $css ) ) // remove any inline css containing \ ( & } = or comments
1411 $css_array = explode( ';', trim( $css ) );
1412 $allowed_attr = apply_filters( 'safe_style_css', array( 'text-align', 'margin', 'color', 'float',
1413 'border', 'background', 'background-color', 'border-bottom', 'border-bottom-color',
1414 'border-bottom-style', 'border-bottom-width', 'border-collapse', 'border-color', 'border-left',
1415 'border-left-color', 'border-left-style', 'border-left-width', 'border-right', 'border-right-color',
1416 'border-right-style', 'border-right-width', 'border-spacing', 'border-style', 'border-top',
1417 'border-top-color', 'border-top-style', 'border-top-width', 'border-width', 'caption-side',
1418 'clear', 'cursor', 'direction', 'font', 'font-family', 'font-size', 'font-style',
1419 'font-variant', 'font-weight', 'height', 'letter-spacing', 'line-height', 'margin-bottom',
1420 'margin-left', 'margin-right', 'margin-top', 'overflow', 'padding', 'padding-bottom',
1421 'padding-left', 'padding-right', 'padding-top', 'text-decoration', 'text-indent', 'vertical-align',
1424 if ( empty($allowed_attr) )
1428 foreach ( $css_array as $css_item ) {
1429 if ( $css_item == '' )
1431 $css_item = trim( $css_item );
1433 if ( strpos( $css_item, ':' ) === false ) {
1436 $parts = explode( ':', $css_item );
1437 if ( in_array( trim( $parts[0] ), $allowed_attr ) )
1451 * Helper function to add global attributes to a tag in the allowed html list.
1456 * @param array $value An array of attributes.
1457 * @return array The array of attributes with global attributes added.
1459 function _wp_add_global_attributes( $value ) {
1460 $global_attributes = array(
1467 if ( true === $value )
1470 if ( is_array( $value ) )
1471 return array_merge( $value, $global_attributes );