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(
101 'blockquote' => array(
159 'fieldset' => array(),
166 'figcaption' => array(
186 'accept-charset' => true,
310 'cellpadding' => true,
311 'cellspacing' => true,
414 * Kses allowed HTML elements.
416 * @global array $allowedtags
419 $allowedtags = array(
431 'blockquote' => array(
449 $allowedentitynames = array(
450 'nbsp', 'iexcl', 'cent', 'pound', 'curren', 'yen',
451 'brvbar', 'sect', 'uml', 'copy', 'ordf', 'laquo',
452 'not', 'shy', 'reg', 'macr', 'deg', 'plusmn',
453 'acute', 'micro', 'para', 'middot', 'cedil', 'ordm',
454 'raquo', 'iquest', 'Agrave', 'Aacute', 'Acirc', 'Atilde',
455 'Auml', 'Aring', 'AElig', 'Ccedil', 'Egrave', 'Eacute',
456 'Ecirc', 'Euml', 'Igrave', 'Iacute', 'Icirc', 'Iuml',
457 'ETH', 'Ntilde', 'Ograve', 'Oacute', 'Ocirc', 'Otilde',
458 'Ouml', 'times', 'Oslash', 'Ugrave', 'Uacute', 'Ucirc',
459 'Uuml', 'Yacute', 'THORN', 'szlig', 'agrave', 'aacute',
460 'acirc', 'atilde', 'auml', 'aring', 'aelig', 'ccedil',
461 'egrave', 'eacute', 'ecirc', 'euml', 'igrave', 'iacute',
462 'icirc', 'iuml', 'eth', 'ntilde', 'ograve', 'oacute',
463 'ocirc', 'otilde', 'ouml', 'divide', 'oslash', 'ugrave',
464 'uacute', 'ucirc', 'uuml', 'yacute', 'thorn', 'yuml',
465 'quot', 'amp', 'lt', 'gt', 'apos', 'OElig',
466 'oelig', 'Scaron', 'scaron', 'Yuml', 'circ', 'tilde',
467 'ensp', 'emsp', 'thinsp', 'zwnj', 'zwj', 'lrm',
468 'rlm', 'ndash', 'mdash', 'lsquo', 'rsquo', 'sbquo',
469 'ldquo', 'rdquo', 'bdquo', 'dagger', 'Dagger', 'permil',
470 'lsaquo', 'rsaquo', 'euro', 'fnof', 'Alpha', 'Beta',
471 'Gamma', 'Delta', 'Epsilon', 'Zeta', 'Eta', 'Theta',
472 'Iota', 'Kappa', 'Lambda', 'Mu', 'Nu', 'Xi',
473 'Omicron', 'Pi', 'Rho', 'Sigma', 'Tau', 'Upsilon',
474 'Phi', 'Chi', 'Psi', 'Omega', 'alpha', 'beta',
475 'gamma', 'delta', 'epsilon', 'zeta', 'eta', 'theta',
476 'iota', 'kappa', 'lambda', 'mu', 'nu', 'xi',
477 'omicron', 'pi', 'rho', 'sigmaf', 'sigma', 'tau',
478 'upsilon', 'phi', 'chi', 'psi', 'omega', 'thetasym',
479 'upsih', 'piv', 'bull', 'hellip', 'prime', 'Prime',
480 'oline', 'frasl', 'weierp', 'image', 'real', 'trade',
481 'alefsym', 'larr', 'uarr', 'rarr', 'darr', 'harr',
482 'crarr', 'lArr', 'uArr', 'rArr', 'dArr', 'hArr',
483 'forall', 'part', 'exist', 'empty', 'nabla', 'isin',
484 'notin', 'ni', 'prod', 'sum', 'minus', 'lowast',
485 'radic', 'prop', 'infin', 'ang', 'and', 'or',
486 'cap', 'cup', 'int', 'sim', 'cong', 'asymp',
487 'ne', 'equiv', 'le', 'ge', 'sub', 'sup',
488 'nsub', 'sube', 'supe', 'oplus', 'otimes', 'perp',
489 'sdot', 'lceil', 'rceil', 'lfloor', 'rfloor', 'lang',
490 'rang', 'loz', 'spades', 'clubs', 'hearts', 'diams',
491 'sup1', 'sup2', 'sup3', 'frac14', 'frac12', 'frac34',
495 $allowedposttags = array_map( '_wp_add_global_attributes', $allowedposttags );
497 $allowedtags = wp_kses_array_lc( $allowedtags );
498 $allowedposttags = wp_kses_array_lc( $allowedposttags );
502 * Filters content and keeps only allowable HTML elements.
504 * This function makes sure that only the allowed HTML element names, attribute
505 * names and attribute values plus only sane HTML entities will occur in
506 * $string. You have to remove any slashes from PHP's magic quotes before you
507 * call this function.
509 * The default allowed protocols are 'http', 'https', 'ftp', 'mailto', 'news',
510 * 'irc', 'gopher', 'nntp', 'feed', 'telnet, 'mms', 'rtsp' and 'svn'. This
511 * covers all common link protocols, except for 'javascript' which should not
512 * be allowed for untrusted users.
516 * @param string $string Content to filter through kses
517 * @param array $allowed_html List of allowed HTML elements
518 * @param array $allowed_protocols Optional. Allowed protocol in links.
519 * @return string Filtered content with only allowed HTML elements
521 function wp_kses( $string, $allowed_html, $allowed_protocols = array() ) {
522 if ( empty( $allowed_protocols ) )
523 $allowed_protocols = wp_allowed_protocols();
524 $string = wp_kses_no_null( $string, array( 'slash_zero' => 'keep' ) );
525 $string = wp_kses_js_entities($string);
526 $string = wp_kses_normalize_entities($string);
527 $string = wp_kses_hook($string, $allowed_html, $allowed_protocols); // WP changed the order of these funcs and added args to wp_kses_hook
528 return wp_kses_split($string, $allowed_html, $allowed_protocols);
532 * Filters one attribute only and ensures its value is allowed.
534 * This function has the advantage of being more secure than esc_attr() and can
535 * escape data in some situations where wp_kses() must strip the whole attribute.
539 * @param string $string The 'whole' attribute, including name and value.
540 * @param string $element The element name to which the attribute belongs.
541 * @return string Filtered attribute.
543 function wp_kses_one_attr( $string, $element ) {
544 $uris = array('xmlns', 'profile', 'href', 'src', 'cite', 'classid', 'codebase', 'data', 'usemap', 'longdesc', 'action');
545 $allowed_html = wp_kses_allowed_html( 'post' );
546 $allowed_protocols = wp_allowed_protocols();
547 $string = wp_kses_no_null( $string, array( 'slash_zero' => 'keep' ) );
548 $string = wp_kses_js_entities( $string );
550 // Preserve leading and trailing whitespace.
552 preg_match('/^\s*/', $string, $matches);
554 preg_match('/\s*$/', $string, $matches);
555 $trail = $matches[0];
556 if ( empty( $trail ) ) {
557 $string = substr( $string, strlen( $lead ) );
559 $string = substr( $string, strlen( $lead ), -strlen( $trail ) );
562 // Parse attribute name and value from input.
563 $split = preg_split( '/\s*=\s*/', $string, 2 );
565 if ( count( $split ) == 2 ) {
568 // Remove quotes surrounding $value.
569 // Also guarantee correct quoting in $string for this one attribute.
570 if ( '' == $value ) {
575 if ( '"' == $quote || "'" == $quote ) {
576 if ( substr( $value, -1 ) != $quote ) {
579 $value = substr( $value, 1, -1 );
584 // Sanitize quotes, angle braces, and entities.
585 $value = esc_attr( $value );
587 // Sanitize URI values.
588 if ( in_array( strtolower( $name ), $uris ) ) {
589 $value = wp_kses_bad_protocol( $value, $allowed_protocols );
592 $string = "$name=$quote$value$quote";
599 // Sanitize attribute by name.
600 wp_kses_attr_check( $name, $value, $string, $vless, $element, $allowed_html );
602 // Restore whitespace.
603 return $lead . $string . $trail;
607 * Return a list of allowed tags and attributes for a given context.
611 * @global array $allowedposttags
612 * @global array $allowedtags
613 * @global array $allowedentitynames
615 * @param string $context The context for which to retrieve tags.
616 * Allowed values are post, strip, data,entities, or
617 * the name of a field filter such as pre_user_description.
618 * @return array List of allowed tags and their allowed attributes.
620 function wp_kses_allowed_html( $context = '' ) {
621 global $allowedposttags, $allowedtags, $allowedentitynames;
623 if ( is_array( $context ) ) {
625 * Filter HTML elements allowed for a given context.
629 * @param string $tags Allowed tags, attributes, and/or entities.
630 * @param string $context Context to judge allowed tags by. Allowed values are 'post',
631 * 'data', 'strip', 'entities', 'explicit', or the name of a filter.
633 return apply_filters( 'wp_kses_allowed_html', $context, 'explicit' );
636 switch ( $context ) {
638 /** This filter is documented in wp-includes/kses.php */
639 return apply_filters( 'wp_kses_allowed_html', $allowedposttags, $context );
641 case 'user_description':
642 case 'pre_user_description':
643 $tags = $allowedtags;
644 $tags['a']['rel'] = true;
645 /** This filter is documented in wp-includes/kses.php */
646 return apply_filters( 'wp_kses_allowed_html', $tags, $context );
649 /** This filter is documented in wp-includes/kses.php */
650 return apply_filters( 'wp_kses_allowed_html', array(), $context );
653 /** This filter is documented in wp-includes/kses.php */
654 return apply_filters( 'wp_kses_allowed_html', $allowedentitynames, $context);
658 /** This filter is documented in wp-includes/kses.php */
659 return apply_filters( 'wp_kses_allowed_html', $allowedtags, $context );
664 * You add any kses hooks here.
666 * There is currently only one kses WordPress hook and it is called here. All
667 * parameters are passed to the hooks and expected to receive a string.
671 * @param string $string Content to filter through kses
672 * @param array $allowed_html List of allowed HTML elements
673 * @param array $allowed_protocols Allowed protocol in links
674 * @return string Filtered content through 'pre_kses' hook
676 function wp_kses_hook( $string, $allowed_html, $allowed_protocols ) {
678 * Filter content to be run through kses.
682 * @param string $string Content to run through kses.
683 * @param array $allowed_html Allowed HTML elements.
684 * @param array $allowed_protocols Allowed protocol in links.
686 return apply_filters( 'pre_kses', $string, $allowed_html, $allowed_protocols );
690 * This function returns kses' version number.
694 * @return string KSES Version Number
696 function wp_kses_version() {
701 * Searches for HTML tags, no matter how malformed.
703 * It also matches stray ">" characters.
707 * @global array $pass_allowed_html
708 * @global array $pass_allowed_protocols
710 * @param string $string Content to filter
711 * @param array $allowed_html Allowed HTML elements
712 * @param array $allowed_protocols Allowed protocols to keep
713 * @return string Content with fixed HTML tags
715 function wp_kses_split( $string, $allowed_html, $allowed_protocols ) {
716 global $pass_allowed_html, $pass_allowed_protocols;
717 $pass_allowed_html = $allowed_html;
718 $pass_allowed_protocols = $allowed_protocols;
719 return preg_replace_callback( '%(<!--.*?(-->|$))|(<[^>]*(>|$)|>)%', '_wp_kses_split_callback', $string );
723 * Callback for wp_kses_split.
728 * @global array $pass_allowed_html
729 * @global array $pass_allowed_protocols
733 function _wp_kses_split_callback( $match ) {
734 global $pass_allowed_html, $pass_allowed_protocols;
735 return wp_kses_split2( $match[0], $pass_allowed_html, $pass_allowed_protocols );
739 * Callback for wp_kses_split for fixing malformed HTML tags.
741 * This function does a lot of work. It rejects some very malformed things like
742 * <:::>. It returns an empty string, if the element isn't allowed (look ma, no
743 * strip_tags()!). Otherwise it splits the tag into an element and an attribute
746 * After the tag is split into an element and an attribute list, it is run
747 * through another filter which will remove illegal attributes and once that is
748 * completed, will be returned.
753 * @param string $string Content to filter
754 * @param array $allowed_html Allowed HTML elements
755 * @param array $allowed_protocols Allowed protocols to keep
756 * @return string Fixed HTML element
758 function wp_kses_split2($string, $allowed_html, $allowed_protocols) {
759 $string = wp_kses_stripslashes($string);
761 if (substr($string, 0, 1) != '<')
763 // It matched a ">" character
765 if ( '<!--' == substr( $string, 0, 4 ) ) {
766 $string = str_replace( array('<!--', '-->'), '', $string );
767 while ( $string != ($newstring = wp_kses($string, $allowed_html, $allowed_protocols)) )
768 $string = $newstring;
771 // prevent multiple dashes in comments
772 $string = preg_replace('/--+/', '-', $string);
773 // prevent three dashes closing a comment
774 $string = preg_replace('/-$/', '', $string);
775 return "<!--{$string}-->";
777 // Allow HTML comments
779 if (!preg_match('%^<\s*(/\s*)?([a-zA-Z0-9]+)([^>]*)>?$%', $string, $matches))
781 // It's seriously malformed
783 $slash = trim($matches[1]);
785 $attrlist = $matches[3];
787 if ( ! is_array( $allowed_html ) )
788 $allowed_html = wp_kses_allowed_html( $allowed_html );
790 if ( ! isset($allowed_html[strtolower($elem)]) )
792 // They are using a not allowed HTML element
796 // No attributes are allowed for closing elements
798 return wp_kses_attr( $elem, $attrlist, $allowed_html, $allowed_protocols );
802 * Removes all attributes, if none are allowed for this element.
804 * If some are allowed it calls wp_kses_hair() to split them further, and then
805 * it builds up new HTML code from the data that kses_hair() returns. It also
806 * removes "<" and ">" characters, if there are any left. One more thing it does
807 * is to check if the tag has a closing XHTML slash, and if it does, it puts one
808 * in the returned code as well.
812 * @param string $element HTML element/tag
813 * @param string $attr HTML attributes from HTML element to closing HTML element tag
814 * @param array $allowed_html Allowed HTML elements
815 * @param array $allowed_protocols Allowed protocols to keep
816 * @return string Sanitized HTML element
818 function wp_kses_attr($element, $attr, $allowed_html, $allowed_protocols) {
819 if ( ! is_array( $allowed_html ) )
820 $allowed_html = wp_kses_allowed_html( $allowed_html );
822 // Is there a closing XHTML slash at the end of the attributes?
824 if (preg_match('%\s*/\s*$%', $attr))
827 // Are any attributes allowed at all for this element?
828 if ( ! isset($allowed_html[strtolower($element)]) || count($allowed_html[strtolower($element)]) == 0 )
829 return "<$element$xhtml_slash>";
832 $attrarr = wp_kses_hair($attr, $allowed_protocols);
834 // Go through $attrarr, and save the allowed attributes for this element
837 foreach ( $attrarr as $arreach ) {
838 if ( wp_kses_attr_check( $arreach['name'], $arreach['value'], $arreach['whole'], $arreach['vless'], $element, $allowed_html ) ) {
839 $attr2 .= ' '.$arreach['whole'];
843 // Remove any "<" or ">" characters
844 $attr2 = preg_replace('/[<>]/', '', $attr2);
846 return "<$element$attr2$xhtml_slash>";
850 * Determine whether an attribute is allowed.
854 * @param string $name The attribute name. Returns empty string when not allowed.
855 * @param string $value The attribute value. Returns a filtered value.
856 * @param string $whole The name=value input. Returns filtered input.
857 * @param string $vless 'y' when attribute like "enabled", otherwise 'n'.
858 * @param string $element The name of the element to which this attribute belongs.
859 * @param array $allowed_html The full list of allowed elements and attributes.
860 * @return bool Is the attribute allowed?
862 function wp_kses_attr_check( &$name, &$value, &$whole, $vless, $element, $allowed_html ) {
863 $allowed_attr = $allowed_html[strtolower( $element )];
865 $name_low = strtolower( $name );
866 if ( ! isset( $allowed_attr[$name_low] ) || '' == $allowed_attr[$name_low] ) {
867 $name = $value = $whole = '';
871 if ( 'style' == $name_low ) {
872 $new_value = safecss_filter_attr( $value );
874 if ( empty( $new_value ) ) {
875 $name = $value = $whole = '';
879 $whole = str_replace( $value, $new_value, $whole );
883 if ( is_array( $allowed_attr[$name_low] ) ) {
884 // there are some checks
885 foreach ( $allowed_attr[$name_low] as $currkey => $currval ) {
886 if ( ! wp_kses_check_attr_val( $value, $vless, $currkey, $currval ) ) {
887 $name = $value = $whole = '';
897 * Builds an attribute list from string containing attributes.
899 * This function does a lot of work. It parses an attribute list into an array
900 * with attribute data, and tries to do the right thing even if it gets weird
901 * input. It will add quotes around attribute values that don't have any quotes
902 * or apostrophes around them, to make it easier to produce HTML code that will
903 * conform to W3C's HTML specification. It will also remove bad URL protocols
904 * from attribute values. It also reduces duplicate attributes by using the
905 * attribute defined first (foo='bar' foo='baz' will result in foo='bar').
909 * @param string $attr Attribute list from HTML element to closing HTML element tag
910 * @param array $allowed_protocols Allowed protocols to keep
911 * @return array List of attributes after parsing
913 function wp_kses_hair($attr, $allowed_protocols) {
917 $uris = array('xmlns', 'profile', 'href', 'src', 'cite', 'classid', 'codebase', 'data', 'usemap', 'longdesc', 'action');
919 // Loop through the whole attribute list
921 while (strlen($attr) != 0) {
922 $working = 0; // Was the last operation successful?
925 case 0 : // attribute name, href for instance
927 if ( preg_match('/^([-a-zA-Z:]+)/', $attr, $match ) ) {
928 $attrname = $match[1];
929 $working = $mode = 1;
930 $attr = preg_replace( '/^[-a-zA-Z:]+/', '', $attr );
935 case 1 : // equals sign or valueless ("selected")
937 if (preg_match('/^\s*=\s*/', $attr)) // equals sign
941 $attr = preg_replace('/^\s*=\s*/', '', $attr);
945 if (preg_match('/^\s+/', $attr)) // valueless
949 if(false === array_key_exists($attrname, $attrarr)) {
950 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
952 $attr = preg_replace('/^\s+/', '', $attr);
957 case 2 : // attribute value, a URL after href= for instance
959 if (preg_match('%^"([^"]*)"(\s+|/?$)%', $attr, $match))
962 $thisval = $match[1];
963 if ( in_array(strtolower($attrname), $uris) )
964 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
966 if(false === array_key_exists($attrname, $attrarr)) {
967 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n');
971 $attr = preg_replace('/^"[^"]*"(\s+|$)/', '', $attr);
975 if (preg_match("%^'([^']*)'(\s+|/?$)%", $attr, $match))
978 $thisval = $match[1];
979 if ( in_array(strtolower($attrname), $uris) )
980 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
982 if(false === array_key_exists($attrname, $attrarr)) {
983 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname='$thisval'", 'vless' => 'n');
987 $attr = preg_replace("/^'[^']*'(\s+|$)/", '', $attr);
991 if (preg_match("%^([^\s\"']+)(\s+|/?$)%", $attr, $match))
994 $thisval = $match[1];
995 if ( in_array(strtolower($attrname), $uris) )
996 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
998 if(false === array_key_exists($attrname, $attrarr)) {
999 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n');
1001 // We add quotes to conform to W3C's HTML spec.
1004 $attr = preg_replace("%^[^\s\"']+(\s+|$)%", '', $attr);
1010 if ($working == 0) // not well formed, remove and try again
1012 $attr = wp_kses_html_error($attr);
1017 if ($mode == 1 && false === array_key_exists($attrname, $attrarr))
1018 // special case, for when the attribute list ends with a valueless
1019 // attribute like "selected"
1020 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
1026 * Finds all attributes of an HTML element.
1028 * Does not modify input. May return "evil" output.
1030 * Based on wp_kses_split2() and wp_kses_attr()
1034 * @param string $element HTML element/tag
1035 * @return array|bool List of attributes found in $element. Returns false on failure.
1037 function wp_kses_attr_parse( $element ) {
1038 $valid = preg_match('%^(<\s*)(/\s*)?([a-zA-Z0-9]+\s*)([^>]*)(>?)$%', $element, $matches);
1039 if ( 1 !== $valid ) {
1043 $begin = $matches[1];
1044 $slash = $matches[2];
1045 $elname = $matches[3];
1046 $attr = $matches[4];
1049 if ( '' !== $slash ) {
1050 // Closing elements do not get parsed.
1054 // Is there a closing XHTML slash at the end of the attributes?
1055 if ( 1 === preg_match( '%\s*/\s*$%', $attr, $matches ) ) {
1056 $xhtml_slash = $matches[0];
1057 $attr = substr( $attr, 0, -strlen( $xhtml_slash ) );
1063 $attrarr = wp_kses_hair_parse( $attr );
1064 if ( false === $attrarr ) {
1068 // Make sure all input is returned by adding front and back matter.
1069 array_unshift( $attrarr, $begin . $slash . $elname );
1070 array_push( $attrarr, $xhtml_slash . $end );
1076 * Builds an attribute list from string containing attributes.
1078 * Does not modify input. May return "evil" output.
1079 * In case of unexpected input, returns false instead of stripping things.
1081 * Based on wp_kses_hair() but does not return a multi-dimensional array.
1085 * @param string $attr Attribute list from HTML element to closing HTML element tag
1086 * @return array|bool List of attributes found in $attr. Returns false on failure.
1088 function wp_kses_hair_parse( $attr ) {
1089 if ( '' === $attr ) {
1095 . '[-a-zA-Z:]+' // Attribute name.
1097 . '\[\[?[^\[\]]+\]\]?' // Shortcode in the name position implies unfiltered_html.
1099 . '(?:' // Attribute value.
1100 . '\s*=\s*' // All values begin with '='
1102 . '"[^"]*"' // Double-quoted
1104 . "'[^']*'" // Single-quoted
1106 . '[^\s"\']+' // Non-quoted
1107 . '(?:\s|$)' // Must have a space
1110 . '(?:\s|$)' // If attribute has no value, space is required.
1112 . '\s*'; // Trailing space is optional except as mentioned above.
1114 // Although it is possible to reduce this procedure to a single regexp,
1115 // we must run that regexp twice to get exactly the expected result.
1117 $validation = "%^($regex)+$%";
1118 $extraction = "%$regex%";
1120 if ( 1 === preg_match( $validation, $attr ) ) {
1121 preg_match_all( $extraction, $attr, $attrarr );
1129 * Performs different checks for attribute values.
1131 * The currently implemented checks are "maxlen", "minlen", "maxval", "minval"
1136 * @param string $value Attribute value
1137 * @param string $vless Whether the value is valueless. Use 'y' or 'n'
1138 * @param string $checkname What $checkvalue is checking for.
1139 * @param mixed $checkvalue What constraint the value should pass
1140 * @return bool Whether check passes
1142 function wp_kses_check_attr_val($value, $vless, $checkname, $checkvalue) {
1145 switch (strtolower($checkname)) {
1147 // The maxlen check makes sure that the attribute value has a length not
1148 // greater than the given value. This can be used to avoid Buffer Overflows
1149 // in WWW clients and various Internet servers.
1151 if (strlen($value) > $checkvalue)
1156 // The minlen check makes sure that the attribute value has a length not
1157 // smaller than the given value.
1159 if (strlen($value) < $checkvalue)
1164 // The maxval check does two things: it checks that the attribute value is
1165 // an integer from 0 and up, without an excessive amount of zeroes or
1166 // whitespace (to avoid Buffer Overflows). It also checks that the attribute
1167 // value is not greater than the given value.
1168 // This check can be used to avoid Denial of Service attacks.
1170 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
1172 if ($value > $checkvalue)
1177 // The minval check makes sure that the attribute value is a positive integer,
1178 // and that it is not smaller than the given value.
1180 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
1182 if ($value < $checkvalue)
1187 // The valueless check makes sure if the attribute has a value
1188 // (like <a href="blah">) or not (<option selected>). If the given value
1189 // is a "y" or a "Y", the attribute must not have a value.
1190 // If the given value is an "n" or an "N", the attribute must have one.
1192 if (strtolower($checkvalue) != $vless)
1201 * Sanitize string from bad protocols.
1203 * This function removes all non-allowed protocols from the beginning of
1204 * $string. It ignores whitespace and the case of the letters, and it does
1205 * understand HTML entities. It does its work in a while loop, so it won't be
1206 * fooled by a string like "javascript:javascript:alert(57)".
1210 * @param string $string Content to filter bad protocols from
1211 * @param array $allowed_protocols Allowed protocols to keep
1212 * @return string Filtered content
1214 function wp_kses_bad_protocol($string, $allowed_protocols) {
1215 $string = wp_kses_no_null($string);
1219 $original_string = $string;
1220 $string = wp_kses_bad_protocol_once($string, $allowed_protocols);
1221 } while ( $original_string != $string && ++$iterations < 6 );
1223 if ( $original_string != $string )
1230 * Removes any invalid control characters in $string.
1232 * Also removes any instance of the '\0' string.
1236 * @param string $string
1237 * @param array $options Set 'slash_zero' => 'keep' when '\0' is allowed. Default is 'remove'.
1240 function wp_kses_no_null( $string, $options = null ) {
1241 if ( ! isset( $options['slash_zero'] ) ) {
1242 $options = array( 'slash_zero' => 'remove' );
1245 $string = preg_replace( '/[\x00-\x08\x0B\x0C\x0E-\x1F]/', '', $string );
1246 if ( 'remove' == $options['slash_zero'] ) {
1247 $string = preg_replace( '/\\\\+0+/', '', $string );
1254 * Strips slashes from in front of quotes.
1256 * This function changes the character sequence \" to just ". It leaves all
1257 * other slashes alone. It's really weird, but the quoting from
1258 * preg_replace(//e) seems to require this.
1262 * @param string $string String to strip slashes
1263 * @return string Fixed string with quoted slashes
1265 function wp_kses_stripslashes($string) {
1266 return preg_replace('%\\\\"%', '"', $string);
1270 * Goes through an array and changes the keys to all lower case.
1274 * @param array $inarray Unfiltered array
1275 * @return array Fixed array with all lowercase keys
1277 function wp_kses_array_lc($inarray) {
1278 $outarray = array ();
1280 foreach ( (array) $inarray as $inkey => $inval) {
1281 $outkey = strtolower($inkey);
1282 $outarray[$outkey] = array ();
1284 foreach ( (array) $inval as $inkey2 => $inval2) {
1285 $outkey2 = strtolower($inkey2);
1286 $outarray[$outkey][$outkey2] = $inval2;
1288 } // foreach $inarray
1294 * Removes the HTML JavaScript entities found in early versions of Netscape 4.
1298 * @param string $string
1301 function wp_kses_js_entities($string) {
1302 return preg_replace('%&\s*\{[^}]*(\}\s*;?|$)%', '', $string);
1306 * Handles parsing errors in wp_kses_hair().
1308 * The general plan is to remove everything to and including some whitespace,
1309 * but it deals with quotes and apostrophes as well.
1313 * @param string $string
1316 function wp_kses_html_error($string) {
1317 return preg_replace('/^("[^"]*("|$)|\'[^\']*(\'|$)|\S)*\s*/', '', $string);
1321 * Sanitizes content from bad protocols and other characters.
1323 * This function searches for URL protocols at the beginning of $string, while
1324 * handling whitespace and HTML entities.
1328 * @param string $string Content to check for bad protocols
1329 * @param string $allowed_protocols Allowed protocols
1330 * @return string Sanitized content
1332 function wp_kses_bad_protocol_once($string, $allowed_protocols, $count = 1 ) {
1333 $string2 = preg_split( '/:|�*58;|�*3a;/i', $string, 2 );
1334 if ( isset($string2[1]) && ! preg_match('%/\?%', $string2[0]) ) {
1335 $string = trim( $string2[1] );
1336 $protocol = wp_kses_bad_protocol_once2( $string2[0], $allowed_protocols );
1337 if ( 'feed:' == $protocol ) {
1340 $string = wp_kses_bad_protocol_once( $string, $allowed_protocols, ++$count );
1341 if ( empty( $string ) )
1344 $string = $protocol . $string;
1351 * Callback for wp_kses_bad_protocol_once() regular expression.
1353 * This function processes URL protocols, checks to see if they're in the
1354 * whitelist or not, and returns different data depending on the answer.
1359 * @param string $string URI scheme to check against the whitelist
1360 * @param string $allowed_protocols Allowed protocols
1361 * @return string Sanitized content
1363 function wp_kses_bad_protocol_once2( $string, $allowed_protocols ) {
1364 $string2 = wp_kses_decode_entities($string);
1365 $string2 = preg_replace('/\s/', '', $string2);
1366 $string2 = wp_kses_no_null($string2);
1367 $string2 = strtolower($string2);
1370 foreach ( (array) $allowed_protocols as $one_protocol )
1371 if ( strtolower($one_protocol) == $string2 ) {
1383 * Converts and fixes HTML entities.
1385 * This function normalizes HTML entities. It will convert `AT&T` to the correct
1386 * `AT&T`, `:` to `:`, `&#XYZZY;` to `&#XYZZY;` and so on.
1390 * @param string $string Content to normalize entities
1391 * @return string Content with normalized entities
1393 function wp_kses_normalize_entities($string) {
1394 // Disarm all entities by converting & to &
1395 $string = str_replace('&', '&', $string);
1397 // Change back the allowed entities in our entity whitelist
1398 $string = preg_replace_callback('/&([A-Za-z]{2,8}[0-9]{0,2});/', 'wp_kses_named_entities', $string);
1399 $string = preg_replace_callback('/&#(0*[0-9]{1,7});/', 'wp_kses_normalize_entities2', $string);
1400 $string = preg_replace_callback('/&#[Xx](0*[0-9A-Fa-f]{1,6});/', 'wp_kses_normalize_entities3', $string);
1406 * Callback for wp_kses_normalize_entities() regular expression.
1408 * This function only accepts valid named entity references, which are finite,
1409 * case-sensitive, and highly scrutinized by HTML and XML validators.
1413 * @global array $allowedentitynames
1415 * @param array $matches preg_replace_callback() matches array
1416 * @return string Correctly encoded entity
1418 function wp_kses_named_entities($matches) {
1419 global $allowedentitynames;
1421 if ( empty($matches[1]) )
1425 return ( ! in_array( $i, $allowedentitynames ) ) ? "&$i;" : "&$i;";
1429 * Callback for wp_kses_normalize_entities() regular expression.
1431 * This function helps {@see wp_kses_normalize_entities()} to only accept 16-bit
1432 * values and nothing more for `&#number;` entities.
1437 * @param array $matches preg_replace_callback() matches array
1438 * @return string Correctly encoded entity
1440 function wp_kses_normalize_entities2($matches) {
1441 if ( empty($matches[1]) )
1445 if (valid_unicode($i)) {
1446 $i = str_pad(ltrim($i,'0'), 3, '0', STR_PAD_LEFT);
1456 * Callback for wp_kses_normalize_entities() for regular expression.
1458 * This function helps wp_kses_normalize_entities() to only accept valid Unicode
1459 * numeric entities in hex form.
1463 * @param array $matches preg_replace_callback() matches array
1464 * @return string Correctly encoded entity
1466 function wp_kses_normalize_entities3($matches) {
1467 if ( empty($matches[1]) )
1470 $hexchars = $matches[1];
1471 return ( ! valid_unicode( hexdec( $hexchars ) ) ) ? "&#x$hexchars;" : '&#x'.ltrim($hexchars,'0').';';
1475 * Helper function to determine if a Unicode value is valid.
1477 * @param int $i Unicode value
1478 * @return bool True if the value was a valid Unicode number
1480 function valid_unicode($i) {
1481 return ( $i == 0x9 || $i == 0xa || $i == 0xd ||
1482 ($i >= 0x20 && $i <= 0xd7ff) ||
1483 ($i >= 0xe000 && $i <= 0xfffd) ||
1484 ($i >= 0x10000 && $i <= 0x10ffff) );
1488 * Convert all entities to their character counterparts.
1490 * This function decodes numeric HTML entities (`A` and `A`).
1491 * It doesn't do anything with other entities like ä, but we don't
1492 * need them in the URL protocol whitelisting system anyway.
1496 * @param string $string Content to change entities
1497 * @return string Content after decoded entities
1499 function wp_kses_decode_entities($string) {
1500 $string = preg_replace_callback('/&#([0-9]+);/', '_wp_kses_decode_entities_chr', $string);
1501 $string = preg_replace_callback('/&#[Xx]([0-9A-Fa-f]+);/', '_wp_kses_decode_entities_chr_hexdec', $string);
1507 * Regex callback for wp_kses_decode_entities()
1509 * @param array $match preg match
1512 function _wp_kses_decode_entities_chr( $match ) {
1513 return chr( $match[1] );
1517 * Regex callback for wp_kses_decode_entities()
1519 * @param array $match preg match
1522 function _wp_kses_decode_entities_chr_hexdec( $match ) {
1523 return chr( hexdec( $match[1] ) );
1527 * Sanitize content with allowed HTML Kses rules.
1531 * @param string $data Content to filter, expected to be escaped with slashes
1532 * @return string Filtered content
1534 function wp_filter_kses( $data ) {
1535 return addslashes( wp_kses( stripslashes( $data ), current_filter() ) );
1539 * Sanitize content with allowed HTML Kses rules.
1543 * @param string $data Content to filter, expected to not be escaped
1544 * @return string Filtered content
1546 function wp_kses_data( $data ) {
1547 return wp_kses( $data, current_filter() );
1551 * Sanitize content for allowed HTML tags for post content.
1553 * Post content refers to the page contents of the 'post' type and not $_POST
1558 * @param string $data Post content to filter, expected to be escaped with slashes
1559 * @return string Filtered post content with allowed HTML tags and attributes intact.
1561 function wp_filter_post_kses( $data ) {
1562 return addslashes( wp_kses( stripslashes( $data ), 'post' ) );
1566 * Sanitize content for allowed HTML tags for post content.
1568 * Post content refers to the page contents of the 'post' type and not $_POST
1573 * @param string $data Post content to filter
1574 * @return string Filtered post content with allowed HTML tags and attributes intact.
1576 function wp_kses_post( $data ) {
1577 return wp_kses( $data, 'post' );
1581 * Strips all of the HTML in the content.
1585 * @param string $data Content to strip all HTML from
1586 * @return string Filtered content without any HTML
1588 function wp_filter_nohtml_kses( $data ) {
1589 return addslashes( wp_kses( stripslashes( $data ), 'strip' ) );
1593 * Adds all Kses input form content filters.
1595 * All hooks have default priority. The wp_filter_kses() function is added to
1596 * the 'pre_comment_content' and 'title_save_pre' hooks.
1598 * The wp_filter_post_kses() function is added to the 'content_save_pre',
1599 * 'excerpt_save_pre', and 'content_filtered_save_pre' hooks.
1603 function kses_init_filters() {
1605 add_filter('title_save_pre', 'wp_filter_kses');
1607 // Comment filtering
1608 if ( current_user_can( 'unfiltered_html' ) )
1609 add_filter( 'pre_comment_content', 'wp_filter_post_kses' );
1611 add_filter( 'pre_comment_content', 'wp_filter_kses' );
1614 add_filter('content_save_pre', 'wp_filter_post_kses');
1615 add_filter('excerpt_save_pre', 'wp_filter_post_kses');
1616 add_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1620 * Removes all Kses input form content filters.
1622 * A quick procedural method to removing all of the filters that kses uses for
1623 * content in WordPress Loop.
1625 * Does not remove the kses_init() function from 'init' hook (priority is
1626 * default). Also does not remove kses_init() function from 'set_current_user'
1627 * hook (priority is also default).
1631 function kses_remove_filters() {
1633 remove_filter('title_save_pre', 'wp_filter_kses');
1635 // Comment filtering
1636 remove_filter( 'pre_comment_content', 'wp_filter_post_kses' );
1637 remove_filter( 'pre_comment_content', 'wp_filter_kses' );
1640 remove_filter('content_save_pre', 'wp_filter_post_kses');
1641 remove_filter('excerpt_save_pre', 'wp_filter_post_kses');
1642 remove_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1646 * Sets up most of the Kses filters for input form content.
1648 * If you remove the kses_init() function from 'init' hook and
1649 * 'set_current_user' (priority is default), then none of the Kses filter hooks
1652 * First removes all of the Kses filters in case the current user does not need
1653 * to have Kses filter the content. If the user does not have unfiltered_html
1654 * capability, then Kses filters are added.
1658 function kses_init() {
1659 kses_remove_filters();
1661 if ( ! current_user_can( 'unfiltered_html' ) ) {
1662 kses_init_filters();
1671 function safecss_filter_attr( $css, $deprecated = '' ) {
1672 if ( !empty( $deprecated ) )
1673 _deprecated_argument( __FUNCTION__, '2.8.1' ); // Never implemented
1675 $css = wp_kses_no_null($css);
1676 $css = str_replace(array("\n","\r","\t"), '', $css);
1678 if ( preg_match( '%[\\\\(&=}]|/\*%', $css ) ) // remove any inline css containing \ ( & } = or comments
1681 $css_array = explode( ';', trim( $css ) );
1684 * Filter list of allowed CSS attributes.
1688 * @param array $attr List of allowed CSS attributes.
1690 $allowed_attr = apply_filters( 'safe_style_css', array( 'text-align', 'margin', 'color', 'float',
1691 'border', 'background', 'background-color', 'border-bottom', 'border-bottom-color',
1692 'border-bottom-style', 'border-bottom-width', 'border-collapse', 'border-color', 'border-left',
1693 'border-left-color', 'border-left-style', 'border-left-width', 'border-right', 'border-right-color',
1694 'border-right-style', 'border-right-width', 'border-spacing', 'border-style', 'border-top',
1695 'border-top-color', 'border-top-style', 'border-top-width', 'border-width', 'caption-side',
1696 'clear', 'cursor', 'direction', 'font', 'font-family', 'font-size', 'font-style',
1697 'font-variant', 'font-weight', 'height', 'letter-spacing', 'line-height', 'margin-bottom',
1698 'margin-left', 'margin-right', 'margin-top', 'overflow', 'padding', 'padding-bottom',
1699 'padding-left', 'padding-right', 'padding-top', 'text-decoration', 'text-indent', 'vertical-align',
1702 if ( empty($allowed_attr) )
1706 foreach ( $css_array as $css_item ) {
1707 if ( $css_item == '' )
1709 $css_item = trim( $css_item );
1711 if ( strpos( $css_item, ':' ) === false ) {
1714 $parts = explode( ':', $css_item );
1715 if ( in_array( trim( $parts[0] ), $allowed_attr ) )
1729 * Helper function to add global attributes to a tag in the allowed html list.
1734 * @param array $value An array of attributes.
1735 * @return array The array of attributes with global attributes added.
1737 function _wp_add_global_attributes( $value ) {
1738 $global_attributes = array(
1746 if ( true === $value )
1749 if ( is_array( $value ) )
1750 return array_merge( $value, $global_attributes );