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 // Ensure that these variables are added to the global namespace
48 // (e.g. if using namespaces / autoload in the current PHP environment).
49 global $allowedposttags, $allowedtags, $allowedentitynames;
51 if ( ! CUSTOM_TAGS ) {
53 * Kses global for default allowable HTML tags.
55 * Can be override by using CUSTOM_TAGS constant.
57 * @global array $allowedposttags
60 $allowedposttags = array(
104 'blockquote' => array(
162 'fieldset' => array(),
169 'figcaption' => array(
189 'accept-charset' => true,
313 'cellpadding' => true,
314 'cellspacing' => true,
417 * Kses allowed HTML elements.
419 * @global array $allowedtags
422 $allowedtags = array(
434 'blockquote' => array(
452 $allowedentitynames = array(
453 'nbsp', 'iexcl', 'cent', 'pound', 'curren', 'yen',
454 'brvbar', 'sect', 'uml', 'copy', 'ordf', 'laquo',
455 'not', 'shy', 'reg', 'macr', 'deg', 'plusmn',
456 'acute', 'micro', 'para', 'middot', 'cedil', 'ordm',
457 'raquo', 'iquest', 'Agrave', 'Aacute', 'Acirc', 'Atilde',
458 'Auml', 'Aring', 'AElig', 'Ccedil', 'Egrave', 'Eacute',
459 'Ecirc', 'Euml', 'Igrave', 'Iacute', 'Icirc', 'Iuml',
460 'ETH', 'Ntilde', 'Ograve', 'Oacute', 'Ocirc', 'Otilde',
461 'Ouml', 'times', 'Oslash', 'Ugrave', 'Uacute', 'Ucirc',
462 'Uuml', 'Yacute', 'THORN', 'szlig', 'agrave', 'aacute',
463 'acirc', 'atilde', 'auml', 'aring', 'aelig', 'ccedil',
464 'egrave', 'eacute', 'ecirc', 'euml', 'igrave', 'iacute',
465 'icirc', 'iuml', 'eth', 'ntilde', 'ograve', 'oacute',
466 'ocirc', 'otilde', 'ouml', 'divide', 'oslash', 'ugrave',
467 'uacute', 'ucirc', 'uuml', 'yacute', 'thorn', 'yuml',
468 'quot', 'amp', 'lt', 'gt', 'apos', 'OElig',
469 'oelig', 'Scaron', 'scaron', 'Yuml', 'circ', 'tilde',
470 'ensp', 'emsp', 'thinsp', 'zwnj', 'zwj', 'lrm',
471 'rlm', 'ndash', 'mdash', 'lsquo', 'rsquo', 'sbquo',
472 'ldquo', 'rdquo', 'bdquo', 'dagger', 'Dagger', 'permil',
473 'lsaquo', 'rsaquo', 'euro', 'fnof', 'Alpha', 'Beta',
474 'Gamma', 'Delta', 'Epsilon', 'Zeta', 'Eta', 'Theta',
475 'Iota', 'Kappa', 'Lambda', 'Mu', 'Nu', 'Xi',
476 'Omicron', 'Pi', 'Rho', 'Sigma', 'Tau', 'Upsilon',
477 'Phi', 'Chi', 'Psi', 'Omega', 'alpha', 'beta',
478 'gamma', 'delta', 'epsilon', 'zeta', 'eta', 'theta',
479 'iota', 'kappa', 'lambda', 'mu', 'nu', 'xi',
480 'omicron', 'pi', 'rho', 'sigmaf', 'sigma', 'tau',
481 'upsilon', 'phi', 'chi', 'psi', 'omega', 'thetasym',
482 'upsih', 'piv', 'bull', 'hellip', 'prime', 'Prime',
483 'oline', 'frasl', 'weierp', 'image', 'real', 'trade',
484 'alefsym', 'larr', 'uarr', 'rarr', 'darr', 'harr',
485 'crarr', 'lArr', 'uArr', 'rArr', 'dArr', 'hArr',
486 'forall', 'part', 'exist', 'empty', 'nabla', 'isin',
487 'notin', 'ni', 'prod', 'sum', 'minus', 'lowast',
488 'radic', 'prop', 'infin', 'ang', 'and', 'or',
489 'cap', 'cup', 'int', 'sim', 'cong', 'asymp',
490 'ne', 'equiv', 'le', 'ge', 'sub', 'sup',
491 'nsub', 'sube', 'supe', 'oplus', 'otimes', 'perp',
492 'sdot', 'lceil', 'rceil', 'lfloor', 'rfloor', 'lang',
493 'rang', 'loz', 'spades', 'clubs', 'hearts', 'diams',
494 'sup1', 'sup2', 'sup3', 'frac14', 'frac12', 'frac34',
498 $allowedposttags = array_map( '_wp_add_global_attributes', $allowedposttags );
500 $allowedtags = wp_kses_array_lc( $allowedtags );
501 $allowedposttags = wp_kses_array_lc( $allowedposttags );
505 * Filters content and keeps only allowable HTML elements.
507 * This function makes sure that only the allowed HTML element names, attribute
508 * names and attribute values plus only sane HTML entities will occur in
509 * $string. You have to remove any slashes from PHP's magic quotes before you
510 * call this function.
512 * The default allowed protocols are 'http', 'https', 'ftp', 'mailto', 'news',
513 * 'irc', 'gopher', 'nntp', 'feed', 'telnet, 'mms', 'rtsp' and 'svn'. This
514 * covers all common link protocols, except for 'javascript' which should not
515 * be allowed for untrusted users.
519 * @param string $string Content to filter through kses
520 * @param array $allowed_html List of allowed HTML elements
521 * @param array $allowed_protocols Optional. Allowed protocol in links.
522 * @return string Filtered content with only allowed HTML elements
524 function wp_kses( $string, $allowed_html, $allowed_protocols = array() ) {
525 if ( empty( $allowed_protocols ) )
526 $allowed_protocols = wp_allowed_protocols();
527 $string = wp_kses_no_null( $string, array( 'slash_zero' => 'keep' ) );
528 $string = wp_kses_js_entities($string);
529 $string = wp_kses_normalize_entities($string);
530 $string = wp_kses_hook($string, $allowed_html, $allowed_protocols); // WP changed the order of these funcs and added args to wp_kses_hook
531 return wp_kses_split($string, $allowed_html, $allowed_protocols);
535 * Filters one attribute only and ensures its value is allowed.
537 * This function has the advantage of being more secure than esc_attr() and can
538 * escape data in some situations where wp_kses() must strip the whole attribute.
542 * @param string $string The 'whole' attribute, including name and value.
543 * @param string $element The element name to which the attribute belongs.
544 * @return string Filtered attribute.
546 function wp_kses_one_attr( $string, $element ) {
547 $uris = array('xmlns', 'profile', 'href', 'src', 'cite', 'classid', 'codebase', 'data', 'usemap', 'longdesc', 'action');
548 $allowed_html = wp_kses_allowed_html( 'post' );
549 $allowed_protocols = wp_allowed_protocols();
550 $string = wp_kses_no_null( $string, array( 'slash_zero' => 'keep' ) );
551 $string = wp_kses_js_entities( $string );
553 // Preserve leading and trailing whitespace.
555 preg_match('/^\s*/', $string, $matches);
557 preg_match('/\s*$/', $string, $matches);
558 $trail = $matches[0];
559 if ( empty( $trail ) ) {
560 $string = substr( $string, strlen( $lead ) );
562 $string = substr( $string, strlen( $lead ), -strlen( $trail ) );
565 // Parse attribute name and value from input.
566 $split = preg_split( '/\s*=\s*/', $string, 2 );
568 if ( count( $split ) == 2 ) {
571 // Remove quotes surrounding $value.
572 // Also guarantee correct quoting in $string for this one attribute.
573 if ( '' == $value ) {
578 if ( '"' == $quote || "'" == $quote ) {
579 if ( substr( $value, -1 ) != $quote ) {
582 $value = substr( $value, 1, -1 );
587 // Sanitize quotes, angle braces, and entities.
588 $value = esc_attr( $value );
590 // Sanitize URI values.
591 if ( in_array( strtolower( $name ), $uris ) ) {
592 $value = wp_kses_bad_protocol( $value, $allowed_protocols );
595 $string = "$name=$quote$value$quote";
602 // Sanitize attribute by name.
603 wp_kses_attr_check( $name, $value, $string, $vless, $element, $allowed_html );
605 // Restore whitespace.
606 return $lead . $string . $trail;
610 * Return a list of allowed tags and attributes for a given context.
614 * @global array $allowedposttags
615 * @global array $allowedtags
616 * @global array $allowedentitynames
618 * @param string $context The context for which to retrieve tags.
619 * Allowed values are post, strip, data,entities, or
620 * the name of a field filter such as pre_user_description.
621 * @return array List of allowed tags and their allowed attributes.
623 function wp_kses_allowed_html( $context = '' ) {
624 global $allowedposttags, $allowedtags, $allowedentitynames;
626 if ( is_array( $context ) ) {
628 * Filter HTML elements allowed for a given context.
632 * @param string $tags Allowed tags, attributes, and/or entities.
633 * @param string $context Context to judge allowed tags by. Allowed values are 'post',
634 * 'data', 'strip', 'entities', 'explicit', or the name of a filter.
636 return apply_filters( 'wp_kses_allowed_html', $context, 'explicit' );
639 switch ( $context ) {
641 /** This filter is documented in wp-includes/kses.php */
642 return apply_filters( 'wp_kses_allowed_html', $allowedposttags, $context );
644 case 'user_description':
645 case 'pre_user_description':
646 $tags = $allowedtags;
647 $tags['a']['rel'] = true;
648 /** This filter is documented in wp-includes/kses.php */
649 return apply_filters( 'wp_kses_allowed_html', $tags, $context );
652 /** This filter is documented in wp-includes/kses.php */
653 return apply_filters( 'wp_kses_allowed_html', array(), $context );
656 /** This filter is documented in wp-includes/kses.php */
657 return apply_filters( 'wp_kses_allowed_html', $allowedentitynames, $context);
661 /** This filter is documented in wp-includes/kses.php */
662 return apply_filters( 'wp_kses_allowed_html', $allowedtags, $context );
667 * You add any kses hooks here.
669 * There is currently only one kses WordPress hook and it is called here. All
670 * parameters are passed to the hooks and expected to receive a string.
674 * @param string $string Content to filter through kses
675 * @param array $allowed_html List of allowed HTML elements
676 * @param array $allowed_protocols Allowed protocol in links
677 * @return string Filtered content through 'pre_kses' hook
679 function wp_kses_hook( $string, $allowed_html, $allowed_protocols ) {
681 * Filter content to be run through kses.
685 * @param string $string Content to run through kses.
686 * @param array $allowed_html Allowed HTML elements.
687 * @param array $allowed_protocols Allowed protocol in links.
689 return apply_filters( 'pre_kses', $string, $allowed_html, $allowed_protocols );
693 * This function returns kses' version number.
697 * @return string KSES Version Number
699 function wp_kses_version() {
704 * Searches for HTML tags, no matter how malformed.
706 * It also matches stray ">" characters.
710 * @global array $pass_allowed_html
711 * @global array $pass_allowed_protocols
713 * @param string $string Content to filter
714 * @param array $allowed_html Allowed HTML elements
715 * @param array $allowed_protocols Allowed protocols to keep
716 * @return string Content with fixed HTML tags
718 function wp_kses_split( $string, $allowed_html, $allowed_protocols ) {
719 global $pass_allowed_html, $pass_allowed_protocols;
720 $pass_allowed_html = $allowed_html;
721 $pass_allowed_protocols = $allowed_protocols;
722 return preg_replace_callback( '%(<!--.*?(-->|$))|(<[^>]*(>|$)|>)%', '_wp_kses_split_callback', $string );
726 * Callback for wp_kses_split.
731 * @global array $pass_allowed_html
732 * @global array $pass_allowed_protocols
736 function _wp_kses_split_callback( $match ) {
737 global $pass_allowed_html, $pass_allowed_protocols;
738 return wp_kses_split2( $match[0], $pass_allowed_html, $pass_allowed_protocols );
742 * Callback for wp_kses_split for fixing malformed HTML tags.
744 * This function does a lot of work. It rejects some very malformed things like
745 * <:::>. It returns an empty string, if the element isn't allowed (look ma, no
746 * strip_tags()!). Otherwise it splits the tag into an element and an attribute
749 * After the tag is split into an element and an attribute list, it is run
750 * through another filter which will remove illegal attributes and once that is
751 * completed, will be returned.
756 * @param string $string Content to filter
757 * @param array $allowed_html Allowed HTML elements
758 * @param array $allowed_protocols Allowed protocols to keep
759 * @return string Fixed HTML element
761 function wp_kses_split2($string, $allowed_html, $allowed_protocols) {
762 $string = wp_kses_stripslashes($string);
764 if (substr($string, 0, 1) != '<')
766 // It matched a ">" character
768 if ( '<!--' == substr( $string, 0, 4 ) ) {
769 $string = str_replace( array('<!--', '-->'), '', $string );
770 while ( $string != ($newstring = wp_kses($string, $allowed_html, $allowed_protocols)) )
771 $string = $newstring;
774 // prevent multiple dashes in comments
775 $string = preg_replace('/--+/', '-', $string);
776 // prevent three dashes closing a comment
777 $string = preg_replace('/-$/', '', $string);
778 return "<!--{$string}-->";
780 // Allow HTML comments
782 if (!preg_match('%^<\s*(/\s*)?([a-zA-Z0-9]+)([^>]*)>?$%', $string, $matches))
784 // It's seriously malformed
786 $slash = trim($matches[1]);
788 $attrlist = $matches[3];
790 if ( ! is_array( $allowed_html ) )
791 $allowed_html = wp_kses_allowed_html( $allowed_html );
793 if ( ! isset($allowed_html[strtolower($elem)]) )
795 // They are using a not allowed HTML element
799 // No attributes are allowed for closing elements
801 return wp_kses_attr( $elem, $attrlist, $allowed_html, $allowed_protocols );
805 * Removes all attributes, if none are allowed for this element.
807 * If some are allowed it calls wp_kses_hair() to split them further, and then
808 * it builds up new HTML code from the data that kses_hair() returns. It also
809 * removes "<" and ">" characters, if there are any left. One more thing it does
810 * is to check if the tag has a closing XHTML slash, and if it does, it puts one
811 * in the returned code as well.
815 * @param string $element HTML element/tag
816 * @param string $attr HTML attributes from HTML element to closing HTML element tag
817 * @param array $allowed_html Allowed HTML elements
818 * @param array $allowed_protocols Allowed protocols to keep
819 * @return string Sanitized HTML element
821 function wp_kses_attr($element, $attr, $allowed_html, $allowed_protocols) {
822 if ( ! is_array( $allowed_html ) )
823 $allowed_html = wp_kses_allowed_html( $allowed_html );
825 // Is there a closing XHTML slash at the end of the attributes?
827 if (preg_match('%\s*/\s*$%', $attr))
830 // Are any attributes allowed at all for this element?
831 if ( ! isset($allowed_html[strtolower($element)]) || count($allowed_html[strtolower($element)]) == 0 )
832 return "<$element$xhtml_slash>";
835 $attrarr = wp_kses_hair($attr, $allowed_protocols);
837 // Go through $attrarr, and save the allowed attributes for this element
840 foreach ( $attrarr as $arreach ) {
841 if ( wp_kses_attr_check( $arreach['name'], $arreach['value'], $arreach['whole'], $arreach['vless'], $element, $allowed_html ) ) {
842 $attr2 .= ' '.$arreach['whole'];
846 // Remove any "<" or ">" characters
847 $attr2 = preg_replace('/[<>]/', '', $attr2);
849 return "<$element$attr2$xhtml_slash>";
853 * Determine whether an attribute is allowed.
857 * @param string $name The attribute name. Returns empty string when not allowed.
858 * @param string $value The attribute value. Returns a filtered value.
859 * @param string $whole The name=value input. Returns filtered input.
860 * @param string $vless 'y' when attribute like "enabled", otherwise 'n'.
861 * @param string $element The name of the element to which this attribute belongs.
862 * @param array $allowed_html The full list of allowed elements and attributes.
863 * @return bool Is the attribute allowed?
865 function wp_kses_attr_check( &$name, &$value, &$whole, $vless, $element, $allowed_html ) {
866 $allowed_attr = $allowed_html[strtolower( $element )];
868 $name_low = strtolower( $name );
869 if ( ! isset( $allowed_attr[$name_low] ) || '' == $allowed_attr[$name_low] ) {
870 $name = $value = $whole = '';
874 if ( 'style' == $name_low ) {
875 $new_value = safecss_filter_attr( $value );
877 if ( empty( $new_value ) ) {
878 $name = $value = $whole = '';
882 $whole = str_replace( $value, $new_value, $whole );
886 if ( is_array( $allowed_attr[$name_low] ) ) {
887 // there are some checks
888 foreach ( $allowed_attr[$name_low] as $currkey => $currval ) {
889 if ( ! wp_kses_check_attr_val( $value, $vless, $currkey, $currval ) ) {
890 $name = $value = $whole = '';
900 * Builds an attribute list from string containing attributes.
902 * This function does a lot of work. It parses an attribute list into an array
903 * with attribute data, and tries to do the right thing even if it gets weird
904 * input. It will add quotes around attribute values that don't have any quotes
905 * or apostrophes around them, to make it easier to produce HTML code that will
906 * conform to W3C's HTML specification. It will also remove bad URL protocols
907 * from attribute values. It also reduces duplicate attributes by using the
908 * attribute defined first (foo='bar' foo='baz' will result in foo='bar').
912 * @param string $attr Attribute list from HTML element to closing HTML element tag
913 * @param array $allowed_protocols Allowed protocols to keep
914 * @return array List of attributes after parsing
916 function wp_kses_hair($attr, $allowed_protocols) {
920 $uris = array('xmlns', 'profile', 'href', 'src', 'cite', 'classid', 'codebase', 'data', 'usemap', 'longdesc', 'action');
922 // Loop through the whole attribute list
924 while (strlen($attr) != 0) {
925 $working = 0; // Was the last operation successful?
928 case 0 : // attribute name, href for instance
930 if ( preg_match('/^([-a-zA-Z:]+)/', $attr, $match ) ) {
931 $attrname = $match[1];
932 $working = $mode = 1;
933 $attr = preg_replace( '/^[-a-zA-Z:]+/', '', $attr );
938 case 1 : // equals sign or valueless ("selected")
940 if (preg_match('/^\s*=\s*/', $attr)) // equals sign
944 $attr = preg_replace('/^\s*=\s*/', '', $attr);
948 if (preg_match('/^\s+/', $attr)) // valueless
952 if(false === array_key_exists($attrname, $attrarr)) {
953 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
955 $attr = preg_replace('/^\s+/', '', $attr);
960 case 2 : // attribute value, a URL after href= for instance
962 if (preg_match('%^"([^"]*)"(\s+|/?$)%', $attr, $match))
965 $thisval = $match[1];
966 if ( in_array(strtolower($attrname), $uris) )
967 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
969 if(false === array_key_exists($attrname, $attrarr)) {
970 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n');
974 $attr = preg_replace('/^"[^"]*"(\s+|$)/', '', $attr);
978 if (preg_match("%^'([^']*)'(\s+|/?$)%", $attr, $match))
981 $thisval = $match[1];
982 if ( in_array(strtolower($attrname), $uris) )
983 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
985 if(false === array_key_exists($attrname, $attrarr)) {
986 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname='$thisval'", 'vless' => 'n');
990 $attr = preg_replace("/^'[^']*'(\s+|$)/", '', $attr);
994 if (preg_match("%^([^\s\"']+)(\s+|/?$)%", $attr, $match))
997 $thisval = $match[1];
998 if ( in_array(strtolower($attrname), $uris) )
999 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
1001 if(false === array_key_exists($attrname, $attrarr)) {
1002 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n');
1004 // We add quotes to conform to W3C's HTML spec.
1007 $attr = preg_replace("%^[^\s\"']+(\s+|$)%", '', $attr);
1013 if ($working == 0) // not well formed, remove and try again
1015 $attr = wp_kses_html_error($attr);
1020 if ($mode == 1 && false === array_key_exists($attrname, $attrarr))
1021 // special case, for when the attribute list ends with a valueless
1022 // attribute like "selected"
1023 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
1029 * Finds all attributes of an HTML element.
1031 * Does not modify input. May return "evil" output.
1033 * Based on wp_kses_split2() and wp_kses_attr()
1037 * @param string $element HTML element/tag
1038 * @return array|bool List of attributes found in $element. Returns false on failure.
1040 function wp_kses_attr_parse( $element ) {
1041 $valid = preg_match('%^(<\s*)(/\s*)?([a-zA-Z0-9]+\s*)([^>]*)(>?)$%', $element, $matches);
1042 if ( 1 !== $valid ) {
1046 $begin = $matches[1];
1047 $slash = $matches[2];
1048 $elname = $matches[3];
1049 $attr = $matches[4];
1052 if ( '' !== $slash ) {
1053 // Closing elements do not get parsed.
1057 // Is there a closing XHTML slash at the end of the attributes?
1058 if ( 1 === preg_match( '%\s*/\s*$%', $attr, $matches ) ) {
1059 $xhtml_slash = $matches[0];
1060 $attr = substr( $attr, 0, -strlen( $xhtml_slash ) );
1066 $attrarr = wp_kses_hair_parse( $attr );
1067 if ( false === $attrarr ) {
1071 // Make sure all input is returned by adding front and back matter.
1072 array_unshift( $attrarr, $begin . $slash . $elname );
1073 array_push( $attrarr, $xhtml_slash . $end );
1079 * Builds an attribute list from string containing attributes.
1081 * Does not modify input. May return "evil" output.
1082 * In case of unexpected input, returns false instead of stripping things.
1084 * Based on wp_kses_hair() but does not return a multi-dimensional array.
1088 * @param string $attr Attribute list from HTML element to closing HTML element tag
1089 * @return array|bool List of attributes found in $attr. Returns false on failure.
1091 function wp_kses_hair_parse( $attr ) {
1092 if ( '' === $attr ) {
1098 . '[-a-zA-Z:]+' // Attribute name.
1100 . '\[\[?[^\[\]]+\]\]?' // Shortcode in the name position implies unfiltered_html.
1102 . '(?:' // Attribute value.
1103 . '\s*=\s*' // All values begin with '='
1105 . '"[^"]*"' // Double-quoted
1107 . "'[^']*'" // Single-quoted
1109 . '[^\s"\']+' // Non-quoted
1110 . '(?:\s|$)' // Must have a space
1113 . '(?:\s|$)' // If attribute has no value, space is required.
1115 . '\s*'; // Trailing space is optional except as mentioned above.
1117 // Although it is possible to reduce this procedure to a single regexp,
1118 // we must run that regexp twice to get exactly the expected result.
1120 $validation = "%^($regex)+$%";
1121 $extraction = "%$regex%";
1123 if ( 1 === preg_match( $validation, $attr ) ) {
1124 preg_match_all( $extraction, $attr, $attrarr );
1132 * Performs different checks for attribute values.
1134 * The currently implemented checks are "maxlen", "minlen", "maxval", "minval"
1139 * @param string $value Attribute value
1140 * @param string $vless Whether the value is valueless. Use 'y' or 'n'
1141 * @param string $checkname What $checkvalue is checking for.
1142 * @param mixed $checkvalue What constraint the value should pass
1143 * @return bool Whether check passes
1145 function wp_kses_check_attr_val($value, $vless, $checkname, $checkvalue) {
1148 switch (strtolower($checkname)) {
1150 // The maxlen check makes sure that the attribute value has a length not
1151 // greater than the given value. This can be used to avoid Buffer Overflows
1152 // in WWW clients and various Internet servers.
1154 if (strlen($value) > $checkvalue)
1159 // The minlen check makes sure that the attribute value has a length not
1160 // smaller than the given value.
1162 if (strlen($value) < $checkvalue)
1167 // The maxval check does two things: it checks that the attribute value is
1168 // an integer from 0 and up, without an excessive amount of zeroes or
1169 // whitespace (to avoid Buffer Overflows). It also checks that the attribute
1170 // value is not greater than the given value.
1171 // This check can be used to avoid Denial of Service attacks.
1173 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
1175 if ($value > $checkvalue)
1180 // The minval check makes sure that the attribute value is a positive integer,
1181 // and that it is not smaller than the given value.
1183 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
1185 if ($value < $checkvalue)
1190 // The valueless check makes sure if the attribute has a value
1191 // (like <a href="blah">) or not (<option selected>). If the given value
1192 // is a "y" or a "Y", the attribute must not have a value.
1193 // If the given value is an "n" or an "N", the attribute must have one.
1195 if (strtolower($checkvalue) != $vless)
1204 * Sanitize string from bad protocols.
1206 * This function removes all non-allowed protocols from the beginning of
1207 * $string. It ignores whitespace and the case of the letters, and it does
1208 * understand HTML entities. It does its work in a while loop, so it won't be
1209 * fooled by a string like "javascript:javascript:alert(57)".
1213 * @param string $string Content to filter bad protocols from
1214 * @param array $allowed_protocols Allowed protocols to keep
1215 * @return string Filtered content
1217 function wp_kses_bad_protocol($string, $allowed_protocols) {
1218 $string = wp_kses_no_null($string);
1222 $original_string = $string;
1223 $string = wp_kses_bad_protocol_once($string, $allowed_protocols);
1224 } while ( $original_string != $string && ++$iterations < 6 );
1226 if ( $original_string != $string )
1233 * Removes any invalid control characters in $string.
1235 * Also removes any instance of the '\0' string.
1239 * @param string $string
1240 * @param array $options Set 'slash_zero' => 'keep' when '\0' is allowed. Default is 'remove'.
1243 function wp_kses_no_null( $string, $options = null ) {
1244 if ( ! isset( $options['slash_zero'] ) ) {
1245 $options = array( 'slash_zero' => 'remove' );
1248 $string = preg_replace( '/[\x00-\x08\x0B\x0C\x0E-\x1F]/', '', $string );
1249 if ( 'remove' == $options['slash_zero'] ) {
1250 $string = preg_replace( '/\\\\+0+/', '', $string );
1257 * Strips slashes from in front of quotes.
1259 * This function changes the character sequence \" to just ". It leaves all
1260 * other slashes alone. It's really weird, but the quoting from
1261 * preg_replace(//e) seems to require this.
1265 * @param string $string String to strip slashes
1266 * @return string Fixed string with quoted slashes
1268 function wp_kses_stripslashes($string) {
1269 return preg_replace('%\\\\"%', '"', $string);
1273 * Goes through an array and changes the keys to all lower case.
1277 * @param array $inarray Unfiltered array
1278 * @return array Fixed array with all lowercase keys
1280 function wp_kses_array_lc($inarray) {
1281 $outarray = array ();
1283 foreach ( (array) $inarray as $inkey => $inval) {
1284 $outkey = strtolower($inkey);
1285 $outarray[$outkey] = array ();
1287 foreach ( (array) $inval as $inkey2 => $inval2) {
1288 $outkey2 = strtolower($inkey2);
1289 $outarray[$outkey][$outkey2] = $inval2;
1291 } // foreach $inarray
1297 * Removes the HTML JavaScript entities found in early versions of Netscape 4.
1301 * @param string $string
1304 function wp_kses_js_entities($string) {
1305 return preg_replace('%&\s*\{[^}]*(\}\s*;?|$)%', '', $string);
1309 * Handles parsing errors in wp_kses_hair().
1311 * The general plan is to remove everything to and including some whitespace,
1312 * but it deals with quotes and apostrophes as well.
1316 * @param string $string
1319 function wp_kses_html_error($string) {
1320 return preg_replace('/^("[^"]*("|$)|\'[^\']*(\'|$)|\S)*\s*/', '', $string);
1324 * Sanitizes content from bad protocols and other characters.
1326 * This function searches for URL protocols at the beginning of $string, while
1327 * handling whitespace and HTML entities.
1331 * @param string $string Content to check for bad protocols
1332 * @param string $allowed_protocols Allowed protocols
1333 * @return string Sanitized content
1335 function wp_kses_bad_protocol_once($string, $allowed_protocols, $count = 1 ) {
1336 $string2 = preg_split( '/:|�*58;|�*3a;/i', $string, 2 );
1337 if ( isset($string2[1]) && ! preg_match('%/\?%', $string2[0]) ) {
1338 $string = trim( $string2[1] );
1339 $protocol = wp_kses_bad_protocol_once2( $string2[0], $allowed_protocols );
1340 if ( 'feed:' == $protocol ) {
1343 $string = wp_kses_bad_protocol_once( $string, $allowed_protocols, ++$count );
1344 if ( empty( $string ) )
1347 $string = $protocol . $string;
1354 * Callback for wp_kses_bad_protocol_once() regular expression.
1356 * This function processes URL protocols, checks to see if they're in the
1357 * whitelist or not, and returns different data depending on the answer.
1362 * @param string $string URI scheme to check against the whitelist
1363 * @param string $allowed_protocols Allowed protocols
1364 * @return string Sanitized content
1366 function wp_kses_bad_protocol_once2( $string, $allowed_protocols ) {
1367 $string2 = wp_kses_decode_entities($string);
1368 $string2 = preg_replace('/\s/', '', $string2);
1369 $string2 = wp_kses_no_null($string2);
1370 $string2 = strtolower($string2);
1373 foreach ( (array) $allowed_protocols as $one_protocol )
1374 if ( strtolower($one_protocol) == $string2 ) {
1386 * Converts and fixes HTML entities.
1388 * This function normalizes HTML entities. It will convert `AT&T` to the correct
1389 * `AT&T`, `:` to `:`, `&#XYZZY;` to `&#XYZZY;` and so on.
1393 * @param string $string Content to normalize entities
1394 * @return string Content with normalized entities
1396 function wp_kses_normalize_entities($string) {
1397 // Disarm all entities by converting & to &
1398 $string = str_replace('&', '&', $string);
1400 // Change back the allowed entities in our entity whitelist
1401 $string = preg_replace_callback('/&([A-Za-z]{2,8}[0-9]{0,2});/', 'wp_kses_named_entities', $string);
1402 $string = preg_replace_callback('/&#(0*[0-9]{1,7});/', 'wp_kses_normalize_entities2', $string);
1403 $string = preg_replace_callback('/&#[Xx](0*[0-9A-Fa-f]{1,6});/', 'wp_kses_normalize_entities3', $string);
1409 * Callback for wp_kses_normalize_entities() regular expression.
1411 * This function only accepts valid named entity references, which are finite,
1412 * case-sensitive, and highly scrutinized by HTML and XML validators.
1416 * @global array $allowedentitynames
1418 * @param array $matches preg_replace_callback() matches array
1419 * @return string Correctly encoded entity
1421 function wp_kses_named_entities($matches) {
1422 global $allowedentitynames;
1424 if ( empty($matches[1]) )
1428 return ( ! in_array( $i, $allowedentitynames ) ) ? "&$i;" : "&$i;";
1432 * Callback for wp_kses_normalize_entities() regular expression.
1434 * This function helps {@see wp_kses_normalize_entities()} to only accept 16-bit
1435 * values and nothing more for `&#number;` entities.
1440 * @param array $matches preg_replace_callback() matches array
1441 * @return string Correctly encoded entity
1443 function wp_kses_normalize_entities2($matches) {
1444 if ( empty($matches[1]) )
1448 if (valid_unicode($i)) {
1449 $i = str_pad(ltrim($i,'0'), 3, '0', STR_PAD_LEFT);
1459 * Callback for wp_kses_normalize_entities() for regular expression.
1461 * This function helps wp_kses_normalize_entities() to only accept valid Unicode
1462 * numeric entities in hex form.
1466 * @param array $matches preg_replace_callback() matches array
1467 * @return string Correctly encoded entity
1469 function wp_kses_normalize_entities3($matches) {
1470 if ( empty($matches[1]) )
1473 $hexchars = $matches[1];
1474 return ( ! valid_unicode( hexdec( $hexchars ) ) ) ? "&#x$hexchars;" : '&#x'.ltrim($hexchars,'0').';';
1478 * Helper function to determine if a Unicode value is valid.
1480 * @param int $i Unicode value
1481 * @return bool True if the value was a valid Unicode number
1483 function valid_unicode($i) {
1484 return ( $i == 0x9 || $i == 0xa || $i == 0xd ||
1485 ($i >= 0x20 && $i <= 0xd7ff) ||
1486 ($i >= 0xe000 && $i <= 0xfffd) ||
1487 ($i >= 0x10000 && $i <= 0x10ffff) );
1491 * Convert all entities to their character counterparts.
1493 * This function decodes numeric HTML entities (`A` and `A`).
1494 * It doesn't do anything with other entities like ä, but we don't
1495 * need them in the URL protocol whitelisting system anyway.
1499 * @param string $string Content to change entities
1500 * @return string Content after decoded entities
1502 function wp_kses_decode_entities($string) {
1503 $string = preg_replace_callback('/&#([0-9]+);/', '_wp_kses_decode_entities_chr', $string);
1504 $string = preg_replace_callback('/&#[Xx]([0-9A-Fa-f]+);/', '_wp_kses_decode_entities_chr_hexdec', $string);
1510 * Regex callback for wp_kses_decode_entities()
1512 * @param array $match preg match
1515 function _wp_kses_decode_entities_chr( $match ) {
1516 return chr( $match[1] );
1520 * Regex callback for wp_kses_decode_entities()
1522 * @param array $match preg match
1525 function _wp_kses_decode_entities_chr_hexdec( $match ) {
1526 return chr( hexdec( $match[1] ) );
1530 * Sanitize content with allowed HTML Kses rules.
1534 * @param string $data Content to filter, expected to be escaped with slashes
1535 * @return string Filtered content
1537 function wp_filter_kses( $data ) {
1538 return addslashes( wp_kses( stripslashes( $data ), current_filter() ) );
1542 * Sanitize content with allowed HTML Kses rules.
1546 * @param string $data Content to filter, expected to not be escaped
1547 * @return string Filtered content
1549 function wp_kses_data( $data ) {
1550 return wp_kses( $data, current_filter() );
1554 * Sanitize content for allowed HTML tags for post content.
1556 * Post content refers to the page contents of the 'post' type and not $_POST
1561 * @param string $data Post content to filter, expected to be escaped with slashes
1562 * @return string Filtered post content with allowed HTML tags and attributes intact.
1564 function wp_filter_post_kses( $data ) {
1565 return addslashes( wp_kses( stripslashes( $data ), 'post' ) );
1569 * Sanitize content for allowed HTML tags for post content.
1571 * Post content refers to the page contents of the 'post' type and not $_POST
1576 * @param string $data Post content to filter
1577 * @return string Filtered post content with allowed HTML tags and attributes intact.
1579 function wp_kses_post( $data ) {
1580 return wp_kses( $data, 'post' );
1584 * Navigates through an array, object, or scalar, and sanitizes content for
1585 * allowed HTML tags for post content.
1589 * @param mixed $value The array or string to filter.
1590 * @return mixed $value The filtered content.
1592 function wp_kses_post_deep( $data ) {
1593 return map_deep( $data, 'wp_kses_post' );
1597 * Strips all of the HTML in the content.
1601 * @param string $data Content to strip all HTML from
1602 * @return string Filtered content without any HTML
1604 function wp_filter_nohtml_kses( $data ) {
1605 return addslashes( wp_kses( stripslashes( $data ), 'strip' ) );
1609 * Adds all Kses input form content filters.
1611 * All hooks have default priority. The wp_filter_kses() function is added to
1612 * the 'pre_comment_content' and 'title_save_pre' hooks.
1614 * The wp_filter_post_kses() function is added to the 'content_save_pre',
1615 * 'excerpt_save_pre', and 'content_filtered_save_pre' hooks.
1619 function kses_init_filters() {
1621 add_filter('title_save_pre', 'wp_filter_kses');
1623 // Comment filtering
1624 if ( current_user_can( 'unfiltered_html' ) )
1625 add_filter( 'pre_comment_content', 'wp_filter_post_kses' );
1627 add_filter( 'pre_comment_content', 'wp_filter_kses' );
1630 add_filter('content_save_pre', 'wp_filter_post_kses');
1631 add_filter('excerpt_save_pre', 'wp_filter_post_kses');
1632 add_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1636 * Removes all Kses input form content filters.
1638 * A quick procedural method to removing all of the filters that kses uses for
1639 * content in WordPress Loop.
1641 * Does not remove the kses_init() function from 'init' hook (priority is
1642 * default). Also does not remove kses_init() function from 'set_current_user'
1643 * hook (priority is also default).
1647 function kses_remove_filters() {
1649 remove_filter('title_save_pre', 'wp_filter_kses');
1651 // Comment filtering
1652 remove_filter( 'pre_comment_content', 'wp_filter_post_kses' );
1653 remove_filter( 'pre_comment_content', 'wp_filter_kses' );
1656 remove_filter('content_save_pre', 'wp_filter_post_kses');
1657 remove_filter('excerpt_save_pre', 'wp_filter_post_kses');
1658 remove_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1662 * Sets up most of the Kses filters for input form content.
1664 * If you remove the kses_init() function from 'init' hook and
1665 * 'set_current_user' (priority is default), then none of the Kses filter hooks
1668 * First removes all of the Kses filters in case the current user does not need
1669 * to have Kses filter the content. If the user does not have unfiltered_html
1670 * capability, then Kses filters are added.
1674 function kses_init() {
1675 kses_remove_filters();
1677 if ( ! current_user_can( 'unfiltered_html' ) ) {
1678 kses_init_filters();
1687 function safecss_filter_attr( $css, $deprecated = '' ) {
1688 if ( !empty( $deprecated ) )
1689 _deprecated_argument( __FUNCTION__, '2.8.1' ); // Never implemented
1691 $css = wp_kses_no_null($css);
1692 $css = str_replace(array("\n","\r","\t"), '', $css);
1694 if ( preg_match( '%[\\\\(&=}]|/\*%', $css ) ) // remove any inline css containing \ ( & } = or comments
1697 $css_array = explode( ';', trim( $css ) );
1700 * Filter list of allowed CSS attributes.
1704 * @param array $attr List of allowed CSS attributes.
1706 $allowed_attr = apply_filters( 'safe_style_css', array( 'text-align', 'margin', 'color', 'float',
1707 'border', 'background', 'background-color', 'border-bottom', 'border-bottom-color',
1708 'border-bottom-style', 'border-bottom-width', 'border-collapse', 'border-color', 'border-left',
1709 'border-left-color', 'border-left-style', 'border-left-width', 'border-right', 'border-right-color',
1710 'border-right-style', 'border-right-width', 'border-spacing', 'border-style', 'border-top',
1711 'border-top-color', 'border-top-style', 'border-top-width', 'border-width', 'caption-side',
1712 'clear', 'cursor', 'direction', 'font', 'font-family', 'font-size', 'font-style',
1713 'font-variant', 'font-weight', 'height', 'min-height','max-height' , 'letter-spacing', 'line-height', 'margin-bottom',
1714 'margin-left', 'margin-right', 'margin-top', 'overflow', 'padding', 'padding-bottom',
1715 'padding-left', 'padding-right', 'padding-top', 'text-decoration', 'text-indent', 'vertical-align',
1716 'width', 'min-width', 'max-width' ) );
1718 if ( empty($allowed_attr) )
1722 foreach ( $css_array as $css_item ) {
1723 if ( $css_item == '' )
1725 $css_item = trim( $css_item );
1727 if ( strpos( $css_item, ':' ) === false ) {
1730 $parts = explode( ':', $css_item );
1731 if ( in_array( trim( $parts[0] ), $allowed_attr ) )
1745 * Helper function to add global attributes to a tag in the allowed html list.
1750 * @param array $value An array of attributes.
1751 * @return array The array of attributes with global attributes added.
1753 function _wp_add_global_attributes( $value ) {
1754 $global_attributes = array(
1762 if ( true === $value )
1765 if ( is_array( $value ) )
1766 return array_merge( $value, $global_attributes );