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 {@see 'wp_kses_allowed_html'} filter is more powerful and supplies context.
39 * `CUSTOM_TAGS` is not recommended and should be considered deprecated.
41 * @see wp_kses_allowed_html()
45 if ( ! defined( 'CUSTOM_TAGS' ) )
46 define( 'CUSTOM_TAGS', false );
48 // Ensure that these variables are added to the global namespace
49 // (e.g. if using namespaces / autoload in the current PHP environment).
50 global $allowedposttags, $allowedtags, $allowedentitynames;
52 if ( ! CUSTOM_TAGS ) {
54 * Kses global for default allowable HTML tags.
56 * Can be override by using CUSTOM_TAGS constant.
58 * @global array $allowedposttags
61 $allowedposttags = array(
105 'blockquote' => array(
163 'fieldset' => array(),
170 'figcaption' => array(
190 'accept-charset' => true,
314 'cellpadding' => true,
315 'cellspacing' => true,
419 * Kses allowed HTML elements.
421 * @global array $allowedtags
424 $allowedtags = array(
436 'blockquote' => array(
454 $allowedentitynames = array(
455 'nbsp', 'iexcl', 'cent', 'pound', 'curren', 'yen',
456 'brvbar', 'sect', 'uml', 'copy', 'ordf', 'laquo',
457 'not', 'shy', 'reg', 'macr', 'deg', 'plusmn',
458 'acute', 'micro', 'para', 'middot', 'cedil', 'ordm',
459 'raquo', 'iquest', 'Agrave', 'Aacute', 'Acirc', 'Atilde',
460 'Auml', 'Aring', 'AElig', 'Ccedil', 'Egrave', 'Eacute',
461 'Ecirc', 'Euml', 'Igrave', 'Iacute', 'Icirc', 'Iuml',
462 'ETH', 'Ntilde', 'Ograve', 'Oacute', 'Ocirc', 'Otilde',
463 'Ouml', 'times', 'Oslash', 'Ugrave', 'Uacute', 'Ucirc',
464 'Uuml', 'Yacute', 'THORN', 'szlig', 'agrave', 'aacute',
465 'acirc', 'atilde', 'auml', 'aring', 'aelig', 'ccedil',
466 'egrave', 'eacute', 'ecirc', 'euml', 'igrave', 'iacute',
467 'icirc', 'iuml', 'eth', 'ntilde', 'ograve', 'oacute',
468 'ocirc', 'otilde', 'ouml', 'divide', 'oslash', 'ugrave',
469 'uacute', 'ucirc', 'uuml', 'yacute', 'thorn', 'yuml',
470 'quot', 'amp', 'lt', 'gt', 'apos', 'OElig',
471 'oelig', 'Scaron', 'scaron', 'Yuml', 'circ', 'tilde',
472 'ensp', 'emsp', 'thinsp', 'zwnj', 'zwj', 'lrm',
473 'rlm', 'ndash', 'mdash', 'lsquo', 'rsquo', 'sbquo',
474 'ldquo', 'rdquo', 'bdquo', 'dagger', 'Dagger', 'permil',
475 'lsaquo', 'rsaquo', 'euro', 'fnof', 'Alpha', 'Beta',
476 'Gamma', 'Delta', 'Epsilon', 'Zeta', 'Eta', 'Theta',
477 'Iota', 'Kappa', 'Lambda', 'Mu', 'Nu', 'Xi',
478 'Omicron', 'Pi', 'Rho', 'Sigma', 'Tau', 'Upsilon',
479 'Phi', 'Chi', 'Psi', 'Omega', 'alpha', 'beta',
480 'gamma', 'delta', 'epsilon', 'zeta', 'eta', 'theta',
481 'iota', 'kappa', 'lambda', 'mu', 'nu', 'xi',
482 'omicron', 'pi', 'rho', 'sigmaf', 'sigma', 'tau',
483 'upsilon', 'phi', 'chi', 'psi', 'omega', 'thetasym',
484 'upsih', 'piv', 'bull', 'hellip', 'prime', 'Prime',
485 'oline', 'frasl', 'weierp', 'image', 'real', 'trade',
486 'alefsym', 'larr', 'uarr', 'rarr', 'darr', 'harr',
487 'crarr', 'lArr', 'uArr', 'rArr', 'dArr', 'hArr',
488 'forall', 'part', 'exist', 'empty', 'nabla', 'isin',
489 'notin', 'ni', 'prod', 'sum', 'minus', 'lowast',
490 'radic', 'prop', 'infin', 'ang', 'and', 'or',
491 'cap', 'cup', 'int', 'sim', 'cong', 'asymp',
492 'ne', 'equiv', 'le', 'ge', 'sub', 'sup',
493 'nsub', 'sube', 'supe', 'oplus', 'otimes', 'perp',
494 'sdot', 'lceil', 'rceil', 'lfloor', 'rfloor', 'lang',
495 'rang', 'loz', 'spades', 'clubs', 'hearts', 'diams',
496 'sup1', 'sup2', 'sup3', 'frac14', 'frac12', 'frac34',
500 $allowedposttags = array_map( '_wp_add_global_attributes', $allowedposttags );
502 $allowedtags = wp_kses_array_lc( $allowedtags );
503 $allowedposttags = wp_kses_array_lc( $allowedposttags );
507 * Filters content and keeps only allowable HTML elements.
509 * This function makes sure that only the allowed HTML element names, attribute
510 * names and attribute values plus only sane HTML entities will occur in
511 * $string. You have to remove any slashes from PHP's magic quotes before you
512 * call this function.
514 * The default allowed protocols are 'http', 'https', 'ftp', 'mailto', 'news',
515 * 'irc', 'gopher', 'nntp', 'feed', 'telnet, 'mms', 'rtsp' and 'svn'. This
516 * covers all common link protocols, except for 'javascript' which should not
517 * be allowed for untrusted users.
521 * @param string $string Content to filter through kses
522 * @param array $allowed_html List of allowed HTML elements
523 * @param array $allowed_protocols Optional. Allowed protocol in links.
524 * @return string Filtered content with only allowed HTML elements
526 function wp_kses( $string, $allowed_html, $allowed_protocols = array() ) {
527 if ( empty( $allowed_protocols ) )
528 $allowed_protocols = wp_allowed_protocols();
529 $string = wp_kses_no_null( $string, array( 'slash_zero' => 'keep' ) );
530 $string = wp_kses_js_entities($string);
531 $string = wp_kses_normalize_entities($string);
532 $string = wp_kses_hook($string, $allowed_html, $allowed_protocols); // WP changed the order of these funcs and added args to wp_kses_hook
533 return wp_kses_split($string, $allowed_html, $allowed_protocols);
537 * Filters one attribute only and ensures its value is allowed.
539 * This function has the advantage of being more secure than esc_attr() and can
540 * escape data in some situations where wp_kses() must strip the whole attribute.
544 * @param string $string The 'whole' attribute, including name and value.
545 * @param string $element The element name to which the attribute belongs.
546 * @return string Filtered attribute.
548 function wp_kses_one_attr( $string, $element ) {
549 $uris = array('xmlns', 'profile', 'href', 'src', 'cite', 'classid', 'codebase', 'data', 'usemap', 'longdesc', 'action');
550 $allowed_html = wp_kses_allowed_html( 'post' );
551 $allowed_protocols = wp_allowed_protocols();
552 $string = wp_kses_no_null( $string, array( 'slash_zero' => 'keep' ) );
553 $string = wp_kses_js_entities( $string );
555 // Preserve leading and trailing whitespace.
557 preg_match('/^\s*/', $string, $matches);
559 preg_match('/\s*$/', $string, $matches);
560 $trail = $matches[0];
561 if ( empty( $trail ) ) {
562 $string = substr( $string, strlen( $lead ) );
564 $string = substr( $string, strlen( $lead ), -strlen( $trail ) );
567 // Parse attribute name and value from input.
568 $split = preg_split( '/\s*=\s*/', $string, 2 );
570 if ( count( $split ) == 2 ) {
573 // Remove quotes surrounding $value.
574 // Also guarantee correct quoting in $string for this one attribute.
575 if ( '' == $value ) {
580 if ( '"' == $quote || "'" == $quote ) {
581 if ( substr( $value, -1 ) != $quote ) {
584 $value = substr( $value, 1, -1 );
589 // Sanitize quotes, angle braces, and entities.
590 $value = esc_attr( $value );
592 // Sanitize URI values.
593 if ( in_array( strtolower( $name ), $uris ) ) {
594 $value = wp_kses_bad_protocol( $value, $allowed_protocols );
597 $string = "$name=$quote$value$quote";
604 // Sanitize attribute by name.
605 wp_kses_attr_check( $name, $value, $string, $vless, $element, $allowed_html );
607 // Restore whitespace.
608 return $lead . $string . $trail;
612 * Return a list of allowed tags and attributes for a given context.
616 * @global array $allowedposttags
617 * @global array $allowedtags
618 * @global array $allowedentitynames
620 * @param string $context The context for which to retrieve tags.
621 * Allowed values are post, strip, data,entities, or
622 * the name of a field filter such as pre_user_description.
623 * @return array List of allowed tags and their allowed attributes.
625 function wp_kses_allowed_html( $context = '' ) {
626 global $allowedposttags, $allowedtags, $allowedentitynames;
628 if ( is_array( $context ) ) {
630 * Filters HTML elements allowed for a given context.
634 * @param string $tags Allowed tags, attributes, and/or entities.
635 * @param string $context Context to judge allowed tags by. Allowed values are 'post',
636 * 'data', 'strip', 'entities', 'explicit', or the name of a filter.
638 return apply_filters( 'wp_kses_allowed_html', $context, 'explicit' );
641 switch ( $context ) {
643 /** This filter is documented in wp-includes/kses.php */
644 return apply_filters( 'wp_kses_allowed_html', $allowedposttags, $context );
646 case 'user_description':
647 case 'pre_user_description':
648 $tags = $allowedtags;
649 $tags['a']['rel'] = true;
650 /** This filter is documented in wp-includes/kses.php */
651 return apply_filters( 'wp_kses_allowed_html', $tags, $context );
654 /** This filter is documented in wp-includes/kses.php */
655 return apply_filters( 'wp_kses_allowed_html', array(), $context );
658 /** This filter is documented in wp-includes/kses.php */
659 return apply_filters( 'wp_kses_allowed_html', $allowedentitynames, $context);
663 /** This filter is documented in wp-includes/kses.php */
664 return apply_filters( 'wp_kses_allowed_html', $allowedtags, $context );
669 * You add any kses hooks here.
671 * There is currently only one kses WordPress hook, {@see 'pre_kses'}, and it is called here.
672 * All parameters are passed to the hooks and expected to receive a string.
676 * @param string $string Content to filter through kses
677 * @param array $allowed_html List of allowed HTML elements
678 * @param array $allowed_protocols Allowed protocol in links
679 * @return string Filtered content through {@see 'pre_kses'} hook.
681 function wp_kses_hook( $string, $allowed_html, $allowed_protocols ) {
683 * Filters content to be run through kses.
687 * @param string $string Content to run through kses.
688 * @param array $allowed_html Allowed HTML elements.
689 * @param array $allowed_protocols Allowed protocol in links.
691 return apply_filters( 'pre_kses', $string, $allowed_html, $allowed_protocols );
695 * This function returns kses' version number.
699 * @return string KSES Version Number
701 function wp_kses_version() {
706 * Searches for HTML tags, no matter how malformed.
708 * It also matches stray ">" characters.
712 * @global array $pass_allowed_html
713 * @global array $pass_allowed_protocols
715 * @param string $string Content to filter
716 * @param array $allowed_html Allowed HTML elements
717 * @param array $allowed_protocols Allowed protocols to keep
718 * @return string Content with fixed HTML tags
720 function wp_kses_split( $string, $allowed_html, $allowed_protocols ) {
721 global $pass_allowed_html, $pass_allowed_protocols;
722 $pass_allowed_html = $allowed_html;
723 $pass_allowed_protocols = $allowed_protocols;
724 return preg_replace_callback( '%(<!--.*?(-->|$))|(<[^>]*(>|$)|>)%', '_wp_kses_split_callback', $string );
728 * Callback for wp_kses_split.
733 * @global array $pass_allowed_html
734 * @global array $pass_allowed_protocols
738 function _wp_kses_split_callback( $match ) {
739 global $pass_allowed_html, $pass_allowed_protocols;
740 return wp_kses_split2( $match[0], $pass_allowed_html, $pass_allowed_protocols );
744 * Callback for wp_kses_split for fixing malformed HTML tags.
746 * This function does a lot of work. It rejects some very malformed things like
747 * <:::>. It returns an empty string, if the element isn't allowed (look ma, no
748 * strip_tags()!). Otherwise it splits the tag into an element and an attribute
751 * After the tag is split into an element and an attribute list, it is run
752 * through another filter which will remove illegal attributes and once that is
753 * completed, will be returned.
758 * @param string $string Content to filter
759 * @param array $allowed_html Allowed HTML elements
760 * @param array $allowed_protocols Allowed protocols to keep
761 * @return string Fixed HTML element
763 function wp_kses_split2($string, $allowed_html, $allowed_protocols) {
764 $string = wp_kses_stripslashes($string);
766 if (substr($string, 0, 1) != '<')
768 // It matched a ">" character
770 if ( '<!--' == substr( $string, 0, 4 ) ) {
771 $string = str_replace( array('<!--', '-->'), '', $string );
772 while ( $string != ($newstring = wp_kses($string, $allowed_html, $allowed_protocols)) )
773 $string = $newstring;
776 // prevent multiple dashes in comments
777 $string = preg_replace('/--+/', '-', $string);
778 // prevent three dashes closing a comment
779 $string = preg_replace('/-$/', '', $string);
780 return "<!--{$string}-->";
782 // Allow HTML comments
784 if (!preg_match('%^<\s*(/\s*)?([a-zA-Z0-9]+)([^>]*)>?$%', $string, $matches))
786 // It's seriously malformed
788 $slash = trim($matches[1]);
790 $attrlist = $matches[3];
792 if ( ! is_array( $allowed_html ) )
793 $allowed_html = wp_kses_allowed_html( $allowed_html );
795 if ( ! isset($allowed_html[strtolower($elem)]) )
797 // They are using a not allowed HTML element
801 // No attributes are allowed for closing elements
803 return wp_kses_attr( $elem, $attrlist, $allowed_html, $allowed_protocols );
807 * Removes all attributes, if none are allowed for this element.
809 * If some are allowed it calls wp_kses_hair() to split them further, and then
810 * it builds up new HTML code from the data that kses_hair() returns. It also
811 * removes "<" and ">" characters, if there are any left. One more thing it does
812 * is to check if the tag has a closing XHTML slash, and if it does, it puts one
813 * in the returned code as well.
817 * @param string $element HTML element/tag
818 * @param string $attr HTML attributes from HTML element to closing HTML element tag
819 * @param array $allowed_html Allowed HTML elements
820 * @param array $allowed_protocols Allowed protocols to keep
821 * @return string Sanitized HTML element
823 function wp_kses_attr($element, $attr, $allowed_html, $allowed_protocols) {
824 if ( ! is_array( $allowed_html ) )
825 $allowed_html = wp_kses_allowed_html( $allowed_html );
827 // Is there a closing XHTML slash at the end of the attributes?
829 if (preg_match('%\s*/\s*$%', $attr))
832 // Are any attributes allowed at all for this element?
833 if ( ! isset($allowed_html[strtolower($element)]) || count($allowed_html[strtolower($element)]) == 0 )
834 return "<$element$xhtml_slash>";
837 $attrarr = wp_kses_hair($attr, $allowed_protocols);
839 // Go through $attrarr, and save the allowed attributes for this element
842 foreach ( $attrarr as $arreach ) {
843 if ( wp_kses_attr_check( $arreach['name'], $arreach['value'], $arreach['whole'], $arreach['vless'], $element, $allowed_html ) ) {
844 $attr2 .= ' '.$arreach['whole'];
848 // Remove any "<" or ">" characters
849 $attr2 = preg_replace('/[<>]/', '', $attr2);
851 return "<$element$attr2$xhtml_slash>";
855 * Determine whether an attribute is allowed.
859 * @param string $name The attribute name. Returns empty string when not allowed.
860 * @param string $value The attribute value. Returns a filtered value.
861 * @param string $whole The name=value input. Returns filtered input.
862 * @param string $vless 'y' when attribute like "enabled", otherwise 'n'.
863 * @param string $element The name of the element to which this attribute belongs.
864 * @param array $allowed_html The full list of allowed elements and attributes.
865 * @return bool Is the attribute allowed?
867 function wp_kses_attr_check( &$name, &$value, &$whole, $vless, $element, $allowed_html ) {
868 $allowed_attr = $allowed_html[strtolower( $element )];
870 $name_low = strtolower( $name );
871 if ( ! isset( $allowed_attr[$name_low] ) || '' == $allowed_attr[$name_low] ) {
872 $name = $value = $whole = '';
876 if ( 'style' == $name_low ) {
877 $new_value = safecss_filter_attr( $value );
879 if ( empty( $new_value ) ) {
880 $name = $value = $whole = '';
884 $whole = str_replace( $value, $new_value, $whole );
888 if ( is_array( $allowed_attr[$name_low] ) ) {
889 // there are some checks
890 foreach ( $allowed_attr[$name_low] as $currkey => $currval ) {
891 if ( ! wp_kses_check_attr_val( $value, $vless, $currkey, $currval ) ) {
892 $name = $value = $whole = '';
902 * Builds an attribute list from string containing attributes.
904 * This function does a lot of work. It parses an attribute list into an array
905 * with attribute data, and tries to do the right thing even if it gets weird
906 * input. It will add quotes around attribute values that don't have any quotes
907 * or apostrophes around them, to make it easier to produce HTML code that will
908 * conform to W3C's HTML specification. It will also remove bad URL protocols
909 * from attribute values. It also reduces duplicate attributes by using the
910 * attribute defined first (foo='bar' foo='baz' will result in foo='bar').
914 * @param string $attr Attribute list from HTML element to closing HTML element tag
915 * @param array $allowed_protocols Allowed protocols to keep
916 * @return array List of attributes after parsing
918 function wp_kses_hair($attr, $allowed_protocols) {
922 $uris = array('xmlns', 'profile', 'href', 'src', 'cite', 'classid', 'codebase', 'data', 'usemap', 'longdesc', 'action');
924 // Loop through the whole attribute list
926 while (strlen($attr) != 0) {
927 $working = 0; // Was the last operation successful?
930 case 0 : // attribute name, href for instance
932 if ( preg_match('/^([-a-zA-Z:]+)/', $attr, $match ) ) {
933 $attrname = $match[1];
934 $working = $mode = 1;
935 $attr = preg_replace( '/^[-a-zA-Z:]+/', '', $attr );
940 case 1 : // equals sign or valueless ("selected")
942 if (preg_match('/^\s*=\s*/', $attr)) // equals sign
946 $attr = preg_replace('/^\s*=\s*/', '', $attr);
950 if (preg_match('/^\s+/', $attr)) // valueless
954 if(false === array_key_exists($attrname, $attrarr)) {
955 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
957 $attr = preg_replace('/^\s+/', '', $attr);
962 case 2 : // attribute value, a URL after href= for instance
964 if (preg_match('%^"([^"]*)"(\s+|/?$)%', $attr, $match))
967 $thisval = $match[1];
968 if ( in_array(strtolower($attrname), $uris) )
969 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
971 if(false === array_key_exists($attrname, $attrarr)) {
972 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n');
976 $attr = preg_replace('/^"[^"]*"(\s+|$)/', '', $attr);
980 if (preg_match("%^'([^']*)'(\s+|/?$)%", $attr, $match))
983 $thisval = $match[1];
984 if ( in_array(strtolower($attrname), $uris) )
985 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
987 if(false === array_key_exists($attrname, $attrarr)) {
988 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname='$thisval'", 'vless' => 'n');
992 $attr = preg_replace("/^'[^']*'(\s+|$)/", '', $attr);
996 if (preg_match("%^([^\s\"']+)(\s+|/?$)%", $attr, $match))
999 $thisval = $match[1];
1000 if ( in_array(strtolower($attrname), $uris) )
1001 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
1003 if(false === array_key_exists($attrname, $attrarr)) {
1004 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n');
1006 // We add quotes to conform to W3C's HTML spec.
1009 $attr = preg_replace("%^[^\s\"']+(\s+|$)%", '', $attr);
1015 if ($working == 0) // not well formed, remove and try again
1017 $attr = wp_kses_html_error($attr);
1022 if ($mode == 1 && false === array_key_exists($attrname, $attrarr))
1023 // special case, for when the attribute list ends with a valueless
1024 // attribute like "selected"
1025 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
1031 * Finds all attributes of an HTML element.
1033 * Does not modify input. May return "evil" output.
1035 * Based on wp_kses_split2() and wp_kses_attr()
1039 * @param string $element HTML element/tag
1040 * @return array|bool List of attributes found in $element. Returns false on failure.
1042 function wp_kses_attr_parse( $element ) {
1043 $valid = preg_match('%^(<\s*)(/\s*)?([a-zA-Z0-9]+\s*)([^>]*)(>?)$%', $element, $matches);
1044 if ( 1 !== $valid ) {
1048 $begin = $matches[1];
1049 $slash = $matches[2];
1050 $elname = $matches[3];
1051 $attr = $matches[4];
1054 if ( '' !== $slash ) {
1055 // Closing elements do not get parsed.
1059 // Is there a closing XHTML slash at the end of the attributes?
1060 if ( 1 === preg_match( '%\s*/\s*$%', $attr, $matches ) ) {
1061 $xhtml_slash = $matches[0];
1062 $attr = substr( $attr, 0, -strlen( $xhtml_slash ) );
1068 $attrarr = wp_kses_hair_parse( $attr );
1069 if ( false === $attrarr ) {
1073 // Make sure all input is returned by adding front and back matter.
1074 array_unshift( $attrarr, $begin . $slash . $elname );
1075 array_push( $attrarr, $xhtml_slash . $end );
1081 * Builds an attribute list from string containing attributes.
1083 * Does not modify input. May return "evil" output.
1084 * In case of unexpected input, returns false instead of stripping things.
1086 * Based on wp_kses_hair() but does not return a multi-dimensional array.
1090 * @param string $attr Attribute list from HTML element to closing HTML element tag
1091 * @return array|bool List of attributes found in $attr. Returns false on failure.
1093 function wp_kses_hair_parse( $attr ) {
1094 if ( '' === $attr ) {
1100 . '[-a-zA-Z:]+' // Attribute name.
1102 . '\[\[?[^\[\]]+\]\]?' // Shortcode in the name position implies unfiltered_html.
1104 . '(?:' // Attribute value.
1105 . '\s*=\s*' // All values begin with '='
1107 . '"[^"]*"' // Double-quoted
1109 . "'[^']*'" // Single-quoted
1111 . '[^\s"\']+' // Non-quoted
1112 . '(?:\s|$)' // Must have a space
1115 . '(?:\s|$)' // If attribute has no value, space is required.
1117 . '\s*'; // Trailing space is optional except as mentioned above.
1119 // Although it is possible to reduce this procedure to a single regexp,
1120 // we must run that regexp twice to get exactly the expected result.
1122 $validation = "%^($regex)+$%";
1123 $extraction = "%$regex%";
1125 if ( 1 === preg_match( $validation, $attr ) ) {
1126 preg_match_all( $extraction, $attr, $attrarr );
1134 * Performs different checks for attribute values.
1136 * The currently implemented checks are "maxlen", "minlen", "maxval", "minval"
1141 * @param string $value Attribute value
1142 * @param string $vless Whether the value is valueless. Use 'y' or 'n'
1143 * @param string $checkname What $checkvalue is checking for.
1144 * @param mixed $checkvalue What constraint the value should pass
1145 * @return bool Whether check passes
1147 function wp_kses_check_attr_val($value, $vless, $checkname, $checkvalue) {
1150 switch (strtolower($checkname)) {
1152 // The maxlen check makes sure that the attribute value has a length not
1153 // greater than the given value. This can be used to avoid Buffer Overflows
1154 // in WWW clients and various Internet servers.
1156 if (strlen($value) > $checkvalue)
1161 // The minlen check makes sure that the attribute value has a length not
1162 // smaller than the given value.
1164 if (strlen($value) < $checkvalue)
1169 // The maxval check does two things: it checks that the attribute value is
1170 // an integer from 0 and up, without an excessive amount of zeroes or
1171 // whitespace (to avoid Buffer Overflows). It also checks that the attribute
1172 // value is not greater than the given value.
1173 // This check can be used to avoid Denial of Service attacks.
1175 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
1177 if ($value > $checkvalue)
1182 // The minval check makes sure that the attribute value is a positive integer,
1183 // and that it is not smaller than the given value.
1185 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
1187 if ($value < $checkvalue)
1192 // The valueless check makes sure if the attribute has a value
1193 // (like <a href="blah">) or not (<option selected>). If the given value
1194 // is a "y" or a "Y", the attribute must not have a value.
1195 // If the given value is an "n" or an "N", the attribute must have one.
1197 if (strtolower($checkvalue) != $vless)
1206 * Sanitize string from bad protocols.
1208 * This function removes all non-allowed protocols from the beginning of
1209 * $string. It ignores whitespace and the case of the letters, and it does
1210 * understand HTML entities. It does its work in a while loop, so it won't be
1211 * fooled by a string like "javascript:javascript:alert(57)".
1215 * @param string $string Content to filter bad protocols from
1216 * @param array $allowed_protocols Allowed protocols to keep
1217 * @return string Filtered content
1219 function wp_kses_bad_protocol($string, $allowed_protocols) {
1220 $string = wp_kses_no_null($string);
1224 $original_string = $string;
1225 $string = wp_kses_bad_protocol_once($string, $allowed_protocols);
1226 } while ( $original_string != $string && ++$iterations < 6 );
1228 if ( $original_string != $string )
1235 * Removes any invalid control characters in $string.
1237 * Also removes any instance of the '\0' string.
1241 * @param string $string
1242 * @param array $options Set 'slash_zero' => 'keep' when '\0' is allowed. Default is 'remove'.
1245 function wp_kses_no_null( $string, $options = null ) {
1246 if ( ! isset( $options['slash_zero'] ) ) {
1247 $options = array( 'slash_zero' => 'remove' );
1250 $string = preg_replace( '/[\x00-\x08\x0B\x0C\x0E-\x1F]/', '', $string );
1251 if ( 'remove' == $options['slash_zero'] ) {
1252 $string = preg_replace( '/\\\\+0+/', '', $string );
1259 * Strips slashes from in front of quotes.
1261 * This function changes the character sequence \" to just ". It leaves all
1262 * other slashes alone. It's really weird, but the quoting from
1263 * preg_replace(//e) seems to require this.
1267 * @param string $string String to strip slashes
1268 * @return string Fixed string with quoted slashes
1270 function wp_kses_stripslashes($string) {
1271 return preg_replace('%\\\\"%', '"', $string);
1275 * Goes through an array and changes the keys to all lower case.
1279 * @param array $inarray Unfiltered array
1280 * @return array Fixed array with all lowercase keys
1282 function wp_kses_array_lc($inarray) {
1283 $outarray = array ();
1285 foreach ( (array) $inarray as $inkey => $inval) {
1286 $outkey = strtolower($inkey);
1287 $outarray[$outkey] = array ();
1289 foreach ( (array) $inval as $inkey2 => $inval2) {
1290 $outkey2 = strtolower($inkey2);
1291 $outarray[$outkey][$outkey2] = $inval2;
1293 } // foreach $inarray
1299 * Removes the HTML JavaScript entities found in early versions of Netscape 4.
1303 * @param string $string
1306 function wp_kses_js_entities($string) {
1307 return preg_replace('%&\s*\{[^}]*(\}\s*;?|$)%', '', $string);
1311 * Handles parsing errors in wp_kses_hair().
1313 * The general plan is to remove everything to and including some whitespace,
1314 * but it deals with quotes and apostrophes as well.
1318 * @param string $string
1321 function wp_kses_html_error($string) {
1322 return preg_replace('/^("[^"]*("|$)|\'[^\']*(\'|$)|\S)*\s*/', '', $string);
1326 * Sanitizes content from bad protocols and other characters.
1328 * This function searches for URL protocols at the beginning of $string, while
1329 * handling whitespace and HTML entities.
1333 * @param string $string Content to check for bad protocols
1334 * @param string $allowed_protocols Allowed protocols
1335 * @return string Sanitized content
1337 function wp_kses_bad_protocol_once($string, $allowed_protocols, $count = 1 ) {
1338 $string2 = preg_split( '/:|�*58;|�*3a;/i', $string, 2 );
1339 if ( isset($string2[1]) && ! preg_match('%/\?%', $string2[0]) ) {
1340 $string = trim( $string2[1] );
1341 $protocol = wp_kses_bad_protocol_once2( $string2[0], $allowed_protocols );
1342 if ( 'feed:' == $protocol ) {
1345 $string = wp_kses_bad_protocol_once( $string, $allowed_protocols, ++$count );
1346 if ( empty( $string ) )
1349 $string = $protocol . $string;
1356 * Callback for wp_kses_bad_protocol_once() regular expression.
1358 * This function processes URL protocols, checks to see if they're in the
1359 * whitelist or not, and returns different data depending on the answer.
1364 * @param string $string URI scheme to check against the whitelist
1365 * @param string $allowed_protocols Allowed protocols
1366 * @return string Sanitized content
1368 function wp_kses_bad_protocol_once2( $string, $allowed_protocols ) {
1369 $string2 = wp_kses_decode_entities($string);
1370 $string2 = preg_replace('/\s/', '', $string2);
1371 $string2 = wp_kses_no_null($string2);
1372 $string2 = strtolower($string2);
1375 foreach ( (array) $allowed_protocols as $one_protocol )
1376 if ( strtolower($one_protocol) == $string2 ) {
1388 * Converts and fixes HTML entities.
1390 * This function normalizes HTML entities. It will convert `AT&T` to the correct
1391 * `AT&T`, `:` to `:`, `&#XYZZY;` to `&#XYZZY;` and so on.
1395 * @param string $string Content to normalize entities
1396 * @return string Content with normalized entities
1398 function wp_kses_normalize_entities($string) {
1399 // Disarm all entities by converting & to &
1400 $string = str_replace('&', '&', $string);
1402 // Change back the allowed entities in our entity whitelist
1403 $string = preg_replace_callback('/&([A-Za-z]{2,8}[0-9]{0,2});/', 'wp_kses_named_entities', $string);
1404 $string = preg_replace_callback('/&#(0*[0-9]{1,7});/', 'wp_kses_normalize_entities2', $string);
1405 $string = preg_replace_callback('/&#[Xx](0*[0-9A-Fa-f]{1,6});/', 'wp_kses_normalize_entities3', $string);
1411 * Callback for wp_kses_normalize_entities() regular expression.
1413 * This function only accepts valid named entity references, which are finite,
1414 * case-sensitive, and highly scrutinized by HTML and XML validators.
1418 * @global array $allowedentitynames
1420 * @param array $matches preg_replace_callback() matches array
1421 * @return string Correctly encoded entity
1423 function wp_kses_named_entities($matches) {
1424 global $allowedentitynames;
1426 if ( empty($matches[1]) )
1430 return ( ! in_array( $i, $allowedentitynames ) ) ? "&$i;" : "&$i;";
1434 * Callback for wp_kses_normalize_entities() regular expression.
1436 * This function helps wp_kses_normalize_entities() to only accept 16-bit
1437 * values and nothing more for `&#number;` entities.
1442 * @param array $matches preg_replace_callback() matches array
1443 * @return string Correctly encoded entity
1445 function wp_kses_normalize_entities2($matches) {
1446 if ( empty($matches[1]) )
1450 if (valid_unicode($i)) {
1451 $i = str_pad(ltrim($i,'0'), 3, '0', STR_PAD_LEFT);
1461 * Callback for wp_kses_normalize_entities() for regular expression.
1463 * This function helps wp_kses_normalize_entities() to only accept valid Unicode
1464 * numeric entities in hex form.
1468 * @param array $matches preg_replace_callback() matches array
1469 * @return string Correctly encoded entity
1471 function wp_kses_normalize_entities3($matches) {
1472 if ( empty($matches[1]) )
1475 $hexchars = $matches[1];
1476 return ( ! valid_unicode( hexdec( $hexchars ) ) ) ? "&#x$hexchars;" : '&#x'.ltrim($hexchars,'0').';';
1480 * Helper function to determine if a Unicode value is valid.
1482 * @param int $i Unicode value
1483 * @return bool True if the value was a valid Unicode number
1485 function valid_unicode($i) {
1486 return ( $i == 0x9 || $i == 0xa || $i == 0xd ||
1487 ($i >= 0x20 && $i <= 0xd7ff) ||
1488 ($i >= 0xe000 && $i <= 0xfffd) ||
1489 ($i >= 0x10000 && $i <= 0x10ffff) );
1493 * Convert all entities to their character counterparts.
1495 * This function decodes numeric HTML entities (`A` and `A`).
1496 * It doesn't do anything with other entities like ä, but we don't
1497 * need them in the URL protocol whitelisting system anyway.
1501 * @param string $string Content to change entities
1502 * @return string Content after decoded entities
1504 function wp_kses_decode_entities($string) {
1505 $string = preg_replace_callback('/&#([0-9]+);/', '_wp_kses_decode_entities_chr', $string);
1506 $string = preg_replace_callback('/&#[Xx]([0-9A-Fa-f]+);/', '_wp_kses_decode_entities_chr_hexdec', $string);
1512 * Regex callback for wp_kses_decode_entities()
1514 * @param array $match preg match
1517 function _wp_kses_decode_entities_chr( $match ) {
1518 return chr( $match[1] );
1522 * Regex callback for wp_kses_decode_entities()
1524 * @param array $match preg match
1527 function _wp_kses_decode_entities_chr_hexdec( $match ) {
1528 return chr( hexdec( $match[1] ) );
1532 * Sanitize content with allowed HTML Kses rules.
1536 * @param string $data Content to filter, expected to be escaped with slashes
1537 * @return string Filtered content
1539 function wp_filter_kses( $data ) {
1540 return addslashes( wp_kses( stripslashes( $data ), current_filter() ) );
1544 * Sanitize content with allowed HTML Kses rules.
1548 * @param string $data Content to filter, expected to not be escaped
1549 * @return string Filtered content
1551 function wp_kses_data( $data ) {
1552 return wp_kses( $data, current_filter() );
1556 * Sanitize content for allowed HTML tags for post content.
1558 * Post content refers to the page contents of the 'post' type and not $_POST
1563 * @param string $data Post content to filter, expected to be escaped with slashes
1564 * @return string Filtered post content with allowed HTML tags and attributes intact.
1566 function wp_filter_post_kses( $data ) {
1567 return addslashes( wp_kses( stripslashes( $data ), 'post' ) );
1571 * Sanitize content for allowed HTML tags for post content.
1573 * Post content refers to the page contents of the 'post' type and not $_POST
1578 * @param string $data Post content to filter
1579 * @return string Filtered post content with allowed HTML tags and attributes intact.
1581 function wp_kses_post( $data ) {
1582 return wp_kses( $data, 'post' );
1586 * Navigates through an array, object, or scalar, and sanitizes content for
1587 * allowed HTML tags for post content.
1593 * @param mixed $data The array, object, or scalar value to inspect.
1594 * @return mixed The filtered content.
1596 function wp_kses_post_deep( $data ) {
1597 return map_deep( $data, 'wp_kses_post' );
1601 * Strips all of the HTML in the content.
1605 * @param string $data Content to strip all HTML from
1606 * @return string Filtered content without any HTML
1608 function wp_filter_nohtml_kses( $data ) {
1609 return addslashes( wp_kses( stripslashes( $data ), 'strip' ) );
1613 * Adds all Kses input form content filters.
1615 * All hooks have default priority. The wp_filter_kses() function is added to
1616 * the 'pre_comment_content' and 'title_save_pre' hooks.
1618 * The wp_filter_post_kses() function is added to the 'content_save_pre',
1619 * 'excerpt_save_pre', and 'content_filtered_save_pre' hooks.
1623 function kses_init_filters() {
1625 add_filter('title_save_pre', 'wp_filter_kses');
1627 // Comment filtering
1628 if ( current_user_can( 'unfiltered_html' ) )
1629 add_filter( 'pre_comment_content', 'wp_filter_post_kses' );
1631 add_filter( 'pre_comment_content', 'wp_filter_kses' );
1634 add_filter('content_save_pre', 'wp_filter_post_kses');
1635 add_filter('excerpt_save_pre', 'wp_filter_post_kses');
1636 add_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1640 * Removes all Kses input form content filters.
1642 * A quick procedural method to removing all of the filters that kses uses for
1643 * content in WordPress Loop.
1645 * Does not remove the kses_init() function from {@see 'init'} hook (priority is
1646 * default). Also does not remove kses_init() function from {@see 'set_current_user'}
1647 * hook (priority is also default).
1651 function kses_remove_filters() {
1653 remove_filter('title_save_pre', 'wp_filter_kses');
1655 // Comment filtering
1656 remove_filter( 'pre_comment_content', 'wp_filter_post_kses' );
1657 remove_filter( 'pre_comment_content', 'wp_filter_kses' );
1660 remove_filter('content_save_pre', 'wp_filter_post_kses');
1661 remove_filter('excerpt_save_pre', 'wp_filter_post_kses');
1662 remove_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1666 * Sets up most of the Kses filters for input form content.
1668 * If you remove the kses_init() function from {@see 'init'} hook and
1669 * {@see 'set_current_user'} (priority is default), then none of the Kses filter hooks
1672 * First removes all of the Kses filters in case the current user does not need
1673 * to have Kses filter the content. If the user does not have unfiltered_html
1674 * capability, then Kses filters are added.
1678 function kses_init() {
1679 kses_remove_filters();
1681 if ( ! current_user_can( 'unfiltered_html' ) ) {
1682 kses_init_filters();
1691 * @param string $css A string of CSS rules.
1692 * @param string $deprecated Not used.
1693 * @return string Filtered string of CSS rules.
1695 function safecss_filter_attr( $css, $deprecated = '' ) {
1696 if ( !empty( $deprecated ) )
1697 _deprecated_argument( __FUNCTION__, '2.8.1' ); // Never implemented
1699 $css = wp_kses_no_null($css);
1700 $css = str_replace(array("\n","\r","\t"), '', $css);
1702 if ( preg_match( '%[\\\\(&=}]|/\*%', $css ) ) // remove any inline css containing \ ( & } = or comments
1705 $css_array = explode( ';', trim( $css ) );
1708 * Filters list of allowed CSS attributes.
1711 * @since 4.4.0 Added support for `min-height`, `max-height`, `min-width`, and `max-width`.
1712 * @since 4.6.0 Added support for `list-style-type`.
1714 * @param array $attr List of allowed CSS attributes.
1716 $allowed_attr = apply_filters( 'safe_style_css', array(
1725 'border-right-color',
1726 'border-right-style',
1727 'border-right-width',
1729 'border-bottom-color',
1730 'border-bottom-style',
1731 'border-bottom-width',
1733 'border-left-color',
1734 'border-left-style',
1735 'border-left-width',
1787 if ( empty($allowed_attr) )
1791 foreach ( $css_array as $css_item ) {
1792 if ( $css_item == '' )
1794 $css_item = trim( $css_item );
1796 if ( strpos( $css_item, ':' ) === false ) {
1799 $parts = explode( ':', $css_item );
1800 if ( in_array( trim( $parts[0] ), $allowed_attr ) )
1814 * Helper function to add global attributes to a tag in the allowed html list.
1819 * @param array $value An array of attributes.
1820 * @return array The array of attributes with global attributes added.
1822 function _wp_add_global_attributes( $value ) {
1823 $global_attributes = array(
1831 if ( true === $value )
1834 if ( is_array( $value ) )
1835 return array_merge( $value, $global_attributes );