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);
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 );
549 $string = wp_kses_normalize_entities( $string );
551 // Preserve leading and trailing whitespace.
553 preg_match('/^\s*/', $string, $matches);
555 preg_match('/\s*$/', $string, $matches);
556 $trail = $matches[0];
557 if ( empty( $trail ) ) {
558 $string = substr( $string, strlen( $lead ) );
560 $string = substr( $string, strlen( $lead ), -strlen( $trail ) );
563 // Parse attribute name and value from input.
564 $split = preg_split( '/\s*=\s*/', $string, 2 );
566 if ( count( $split ) == 2 ) {
569 // Remove quotes surrounding $value.
570 // Also guarantee correct quoting in $string for this one attribute.
571 if ( '' == $value ) {
576 if ( '"' == $quote || "'" == $quote ) {
577 if ( substr( $value, -1 ) != $quote ) {
580 $value = substr( $value, 1, -1 );
585 // Sanitize quotes and angle braces.
586 $value = htmlspecialchars( $value, ENT_QUOTES, null, false );
588 // Sanitize URI values.
589 if ( in_array( strtolower( $name ), $uris ) ) {
590 $value = wp_kses_bad_protocol( $value, $allowed_protocols );
593 $string = "$name=$quote$value$quote";
600 // Sanitize attribute by name.
601 wp_kses_attr_check( $name, $value, $string, $vless, $element, $allowed_html );
603 // Restore whitespace.
604 return $lead . $string . $trail;
608 * Return a list of allowed tags and attributes for a given context.
612 * @param string $context The context for which to retrieve tags. Allowed values are
613 * post | strip | data | entities or the name of a field filter such as pre_user_description.
614 * @return array List of allowed tags and their allowed attributes.
616 function wp_kses_allowed_html( $context = '' ) {
617 global $allowedposttags, $allowedtags, $allowedentitynames;
619 if ( is_array( $context ) ) {
621 * Filter HTML elements allowed for a given context.
625 * @param string $tags Allowed tags, attributes, and/or entities.
626 * @param string $context Context to judge allowed tags by. Allowed values are 'post',
627 * 'data', 'strip', 'entities', 'explicit', or the name of a filter.
629 return apply_filters( 'wp_kses_allowed_html', $context, 'explicit' );
632 switch ( $context ) {
634 /** This filter is documented in wp-includes/kses.php */
635 return apply_filters( 'wp_kses_allowed_html', $allowedposttags, $context );
637 case 'user_description':
638 case 'pre_user_description':
639 $tags = $allowedtags;
640 $tags['a']['rel'] = true;
641 /** This filter is documented in wp-includes/kses.php */
642 return apply_filters( 'wp_kses_allowed_html', $tags, $context );
645 /** This filter is documented in wp-includes/kses.php */
646 return apply_filters( 'wp_kses_allowed_html', array(), $context );
649 /** This filter is documented in wp-includes/kses.php */
650 return apply_filters( 'wp_kses_allowed_html', $allowedentitynames, $context);
654 /** This filter is documented in wp-includes/kses.php */
655 return apply_filters( 'wp_kses_allowed_html', $allowedtags, $context );
660 * You add any kses hooks here.
662 * There is currently only one kses WordPress hook and it is called here. All
663 * parameters are passed to the hooks and expected to receive a string.
667 * @param string $string Content to filter through kses
668 * @param array $allowed_html List of allowed HTML elements
669 * @param array $allowed_protocols Allowed protocol in links
670 * @return string Filtered content through 'pre_kses' hook
672 function wp_kses_hook( $string, $allowed_html, $allowed_protocols ) {
674 * Filter content to be run through kses.
678 * @param string $string Content to run through kses.
679 * @param array $allowed_html Allowed HTML elements.
680 * @param array $allowed_protocols Allowed protocol in links.
682 $string = apply_filters( 'pre_kses', $string, $allowed_html, $allowed_protocols );
687 * This function returns kses' version number.
691 * @return string KSES Version Number
693 function wp_kses_version() {
698 * Searches for HTML tags, no matter how malformed.
700 * It also matches stray ">" characters.
704 * @param string $string Content to filter
705 * @param array $allowed_html Allowed HTML elements
706 * @param array $allowed_protocols Allowed protocols to keep
707 * @return string Content with fixed HTML tags
709 function wp_kses_split( $string, $allowed_html, $allowed_protocols ) {
710 global $pass_allowed_html, $pass_allowed_protocols;
711 $pass_allowed_html = $allowed_html;
712 $pass_allowed_protocols = $allowed_protocols;
713 return preg_replace_callback( '%(<!--.*?(-->|$))|(<[^>]*(>|$)|>)%', '_wp_kses_split_callback', $string );
717 * Callback for wp_kses_split.
722 function _wp_kses_split_callback( $match ) {
723 global $pass_allowed_html, $pass_allowed_protocols;
724 return wp_kses_split2( $match[0], $pass_allowed_html, $pass_allowed_protocols );
728 * Callback for wp_kses_split for fixing malformed HTML tags.
730 * This function does a lot of work. It rejects some very malformed things like
731 * <:::>. It returns an empty string, if the element isn't allowed (look ma, no
732 * strip_tags()!). Otherwise it splits the tag into an element and an attribute
735 * After the tag is split into an element and an attribute list, it is run
736 * through another filter which will remove illegal attributes and once that is
737 * completed, will be returned.
742 * @param string $string Content to filter
743 * @param array $allowed_html Allowed HTML elements
744 * @param array $allowed_protocols Allowed protocols to keep
745 * @return string Fixed HTML element
747 function wp_kses_split2($string, $allowed_html, $allowed_protocols) {
748 $string = wp_kses_stripslashes($string);
750 if (substr($string, 0, 1) != '<')
752 // It matched a ">" character
754 if ( '<!--' == substr( $string, 0, 4 ) ) {
755 $string = str_replace( array('<!--', '-->'), '', $string );
756 while ( $string != ($newstring = wp_kses($string, $allowed_html, $allowed_protocols)) )
757 $string = $newstring;
760 // prevent multiple dashes in comments
761 $string = preg_replace('/--+/', '-', $string);
762 // prevent three dashes closing a comment
763 $string = preg_replace('/-$/', '', $string);
764 return "<!--{$string}-->";
766 // Allow HTML comments
768 if (!preg_match('%^<\s*(/\s*)?([a-zA-Z0-9]+)([^>]*)>?$%', $string, $matches))
770 // It's seriously malformed
772 $slash = trim($matches[1]);
774 $attrlist = $matches[3];
776 if ( ! is_array( $allowed_html ) )
777 $allowed_html = wp_kses_allowed_html( $allowed_html );
779 if ( ! isset($allowed_html[strtolower($elem)]) )
781 // They are using a not allowed HTML element
785 // No attributes are allowed for closing elements
787 return wp_kses_attr( $elem, $attrlist, $allowed_html, $allowed_protocols );
791 * Removes all attributes, if none are allowed for this element.
793 * If some are allowed it calls wp_kses_hair() to split them further, and then
794 * it builds up new HTML code from the data that kses_hair() returns. It also
795 * removes "<" and ">" characters, if there are any left. One more thing it does
796 * is to check if the tag has a closing XHTML slash, and if it does, it puts one
797 * in the returned code as well.
801 * @param string $element HTML element/tag
802 * @param string $attr HTML attributes from HTML element to closing HTML element tag
803 * @param array $allowed_html Allowed HTML elements
804 * @param array $allowed_protocols Allowed protocols to keep
805 * @return string Sanitized HTML element
807 function wp_kses_attr($element, $attr, $allowed_html, $allowed_protocols) {
808 // Is there a closing XHTML slash at the end of the attributes?
810 if ( ! is_array( $allowed_html ) )
811 $allowed_html = wp_kses_allowed_html( $allowed_html );
814 if (preg_match('%\s*/\s*$%', $attr))
817 // Are any attributes allowed at all for this element?
818 if ( ! isset($allowed_html[strtolower($element)]) || count($allowed_html[strtolower($element)]) == 0 )
819 return "<$element$xhtml_slash>";
822 $attrarr = wp_kses_hair($attr, $allowed_protocols);
824 // Go through $attrarr, and save the allowed attributes for this element
827 foreach ( $attrarr as $arreach ) {
828 if ( wp_kses_attr_check( $arreach['name'], $arreach['value'], $arreach['whole'], $arreach['vless'], $element, $allowed_html ) ) {
829 $attr2 .= ' '.$arreach['whole'];
833 // Remove any "<" or ">" characters
834 $attr2 = preg_replace('/[<>]/', '', $attr2);
836 return "<$element$attr2$xhtml_slash>";
840 * Determine whether an attribute is allowed.
844 * @param string $name The attribute name. Returns empty string when not allowed.
845 * @param string $value The attribute value. Returns a filtered value.
846 * @param string $whole The name=value input. Returns filtered input.
847 * @param string $vless 'y' when attribute like "enabled", otherwise 'n'.
848 * @param string $element The name of the element to which this attribute belongs.
849 * @param array $allowed_html The full list of allowed elements and attributes.
850 * @return bool Is the attribute allowed?
852 function wp_kses_attr_check( &$name, &$value, &$whole, $vless, $element, $allowed_html ) {
853 $allowed_attr = $allowed_html[strtolower( $element )];
855 $name_low = strtolower( $name );
856 if ( ! isset( $allowed_attr[$name_low] ) || '' == $allowed_attr[$name_low] ) {
857 $name = $value = $whole = '';
861 if ( 'style' == $name_low ) {
862 $new_value = safecss_filter_attr( $value );
864 if ( empty( $new_value ) ) {
865 $name = $value = $whole = '';
869 $whole = str_replace( $value, $new_value, $whole );
873 if ( is_array( $allowed_attr[$name_low] ) ) {
874 // there are some checks
875 foreach ( $allowed_attr[$name_low] as $currkey => $currval ) {
876 if ( ! wp_kses_check_attr_val( $value, $vless, $currkey, $currval ) ) {
877 $name = $value = $whole = '';
887 * Builds an attribute list from string containing attributes.
889 * This function does a lot of work. It parses an attribute list into an array
890 * with attribute data, and tries to do the right thing even if it gets weird
891 * input. It will add quotes around attribute values that don't have any quotes
892 * or apostrophes around them, to make it easier to produce HTML code that will
893 * conform to W3C's HTML specification. It will also remove bad URL protocols
894 * from attribute values. It also reduces duplicate attributes by using the
895 * attribute defined first (foo='bar' foo='baz' will result in foo='bar').
899 * @param string $attr Attribute list from HTML element to closing HTML element tag
900 * @param array $allowed_protocols Allowed protocols to keep
901 * @return array List of attributes after parsing
903 function wp_kses_hair($attr, $allowed_protocols) {
907 $uris = array('xmlns', 'profile', 'href', 'src', 'cite', 'classid', 'codebase', 'data', 'usemap', 'longdesc', 'action');
909 // Loop through the whole attribute list
911 while (strlen($attr) != 0) {
912 $working = 0; // Was the last operation successful?
915 case 0 : // attribute name, href for instance
917 if ( preg_match('/^([-a-zA-Z:]+)/', $attr, $match ) ) {
918 $attrname = $match[1];
919 $working = $mode = 1;
920 $attr = preg_replace( '/^[-a-zA-Z:]+/', '', $attr );
925 case 1 : // equals sign or valueless ("selected")
927 if (preg_match('/^\s*=\s*/', $attr)) // equals sign
931 $attr = preg_replace('/^\s*=\s*/', '', $attr);
935 if (preg_match('/^\s+/', $attr)) // valueless
939 if(false === array_key_exists($attrname, $attrarr)) {
940 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
942 $attr = preg_replace('/^\s+/', '', $attr);
947 case 2 : // attribute value, a URL after href= for instance
949 if (preg_match('%^"([^"]*)"(\s+|/?$)%', $attr, $match))
952 $thisval = $match[1];
953 if ( in_array(strtolower($attrname), $uris) )
954 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
956 if(false === array_key_exists($attrname, $attrarr)) {
957 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n');
961 $attr = preg_replace('/^"[^"]*"(\s+|$)/', '', $attr);
965 if (preg_match("%^'([^']*)'(\s+|/?$)%", $attr, $match))
968 $thisval = $match[1];
969 if ( in_array(strtolower($attrname), $uris) )
970 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
972 if(false === array_key_exists($attrname, $attrarr)) {
973 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname='$thisval'", 'vless' => 'n');
977 $attr = preg_replace("/^'[^']*'(\s+|$)/", '', $attr);
981 if (preg_match("%^([^\s\"']+)(\s+|/?$)%", $attr, $match))
984 $thisval = $match[1];
985 if ( in_array(strtolower($attrname), $uris) )
986 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
988 if(false === array_key_exists($attrname, $attrarr)) {
989 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n');
991 // We add quotes to conform to W3C's HTML spec.
994 $attr = preg_replace("%^[^\s\"']+(\s+|$)%", '', $attr);
1000 if ($working == 0) // not well formed, remove and try again
1002 $attr = wp_kses_html_error($attr);
1007 if ($mode == 1 && false === array_key_exists($attrname, $attrarr))
1008 // special case, for when the attribute list ends with a valueless
1009 // attribute like "selected"
1010 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
1016 * Finds all attributes of an HTML element.
1018 * Does not modify input. May return "evil" output.
1020 * Based on wp_kses_split2() and wp_kses_attr()
1024 * @param string $element HTML element/tag
1025 * @return array|bool List of attributes found in $element. Returns false on failure.
1027 function wp_kses_attr_parse( $element ) {
1028 $valid = preg_match('%^(<\s*)(/\s*)?([a-zA-Z0-9]+\s*)([^>]*)(>?)$%', $element, $matches);
1029 if ( 1 !== $valid ) {
1033 $begin = $matches[1];
1034 $slash = $matches[2];
1035 $elname = $matches[3];
1036 $attr = $matches[4];
1039 if ( '' !== $slash ) {
1040 // Closing elements do not get parsed.
1044 // Is there a closing XHTML slash at the end of the attributes?
1045 if ( 1 === preg_match( '%\s*/\s*$%', $attr, $matches ) ) {
1046 $xhtml_slash = $matches[0];
1047 $attr = substr( $attr, 0, -strlen( $xhtml_slash ) );
1053 $attrarr = wp_kses_hair_parse( $attr );
1054 if ( false === $attrarr ) {
1058 // Make sure all input is returned by adding front and back matter.
1059 array_unshift( $attrarr, $begin . $slash . $elname );
1060 array_push( $attrarr, $xhtml_slash . $end );
1066 * Builds an attribute list from string containing attributes.
1068 * Does not modify input. May return "evil" output.
1069 * In case of unexpected input, returns false instead of stripping things.
1071 * Based on wp_kses_hair() but does not return a multi-dimensional array.
1075 * @param string $attr Attribute list from HTML element to closing HTML element tag
1076 * @return array|bool List of attributes found in $attr. Returns false on failure.
1078 function wp_kses_hair_parse( $attr ) {
1079 if ( '' === $attr ) {
1085 . '[-a-zA-Z:]+' // Attribute name.
1087 . '\[\[?[^\[\]]+\]\]?' // Shortcode in the name position implies unfiltered_html.
1089 . '(?:' // Attribute value.
1090 . '\s*=\s*' // All values begin with '='
1092 . '"[^"]*"' // Double-quoted
1094 . "'[^']*'" // Single-quoted
1096 . '[^\s"\']+' // Non-quoted
1097 . '(?:\s|$)' // Must have a space
1100 . '(?:\s|$)' // If attribute has no value, space is required.
1102 . '\s*'; // Trailing space is optional except as mentioned above.
1104 // Although it is possible to reduce this procedure to a single regexp,
1105 // we must run that regexp twice to get exactly the expected result.
1107 $validation = "%^($regex)+$%";
1108 $extraction = "%$regex%";
1110 if ( 1 === preg_match( $validation, $attr ) ) {
1111 preg_match_all( $extraction, $attr, $attrarr );
1119 * Performs different checks for attribute values.
1121 * The currently implemented checks are "maxlen", "minlen", "maxval", "minval"
1126 * @param string $value Attribute value
1127 * @param string $vless Whether the value is valueless. Use 'y' or 'n'
1128 * @param string $checkname What $checkvalue is checking for.
1129 * @param mixed $checkvalue What constraint the value should pass
1130 * @return bool Whether check passes
1132 function wp_kses_check_attr_val($value, $vless, $checkname, $checkvalue) {
1135 switch (strtolower($checkname)) {
1137 // The maxlen check makes sure that the attribute value has a length not
1138 // greater than the given value. This can be used to avoid Buffer Overflows
1139 // in WWW clients and various Internet servers.
1141 if (strlen($value) > $checkvalue)
1146 // The minlen check makes sure that the attribute value has a length not
1147 // smaller than the given value.
1149 if (strlen($value) < $checkvalue)
1154 // The maxval check does two things: it checks that the attribute value is
1155 // an integer from 0 and up, without an excessive amount of zeroes or
1156 // whitespace (to avoid Buffer Overflows). It also checks that the attribute
1157 // value is not greater than the given value.
1158 // This check can be used to avoid Denial of Service attacks.
1160 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
1162 if ($value > $checkvalue)
1167 // The minval check makes sure that the attribute value is a positive integer,
1168 // and that it is not smaller than the given value.
1170 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
1172 if ($value < $checkvalue)
1177 // The valueless check makes sure if the attribute has a value
1178 // (like <a href="blah">) or not (<option selected>). If the given value
1179 // is a "y" or a "Y", the attribute must not have a value.
1180 // If the given value is an "n" or an "N", the attribute must have one.
1182 if (strtolower($checkvalue) != $vless)
1191 * Sanitize string from bad protocols.
1193 * This function removes all non-allowed protocols from the beginning of
1194 * $string. It ignores whitespace and the case of the letters, and it does
1195 * understand HTML entities. It does its work in a while loop, so it won't be
1196 * fooled by a string like "javascript:javascript:alert(57)".
1200 * @param string $string Content to filter bad protocols from
1201 * @param array $allowed_protocols Allowed protocols to keep
1202 * @return string Filtered content
1204 function wp_kses_bad_protocol($string, $allowed_protocols) {
1205 $string = wp_kses_no_null($string);
1209 $original_string = $string;
1210 $string = wp_kses_bad_protocol_once($string, $allowed_protocols);
1211 } while ( $original_string != $string && ++$iterations < 6 );
1213 if ( $original_string != $string )
1220 * Removes any invalid control characters in $string.
1222 * Also removes any instance of the '\0' string.
1226 * @param string $string
1229 function wp_kses_no_null($string) {
1230 $string = preg_replace('/[\x00-\x08\x0B\x0C\x0E-\x1F]/', '', $string);
1231 $string = preg_replace('/(\\\\0)+/', '', $string);
1237 * Strips slashes from in front of quotes.
1239 * This function changes the character sequence \" to just ". It leaves all
1240 * other slashes alone. It's really weird, but the quoting from
1241 * preg_replace(//e) seems to require this.
1245 * @param string $string String to strip slashes
1246 * @return string Fixed string with quoted slashes
1248 function wp_kses_stripslashes($string) {
1249 return preg_replace('%\\\\"%', '"', $string);
1253 * Goes through an array and changes the keys to all lower case.
1257 * @param array $inarray Unfiltered array
1258 * @return array Fixed array with all lowercase keys
1260 function wp_kses_array_lc($inarray) {
1261 $outarray = array ();
1263 foreach ( (array) $inarray as $inkey => $inval) {
1264 $outkey = strtolower($inkey);
1265 $outarray[$outkey] = array ();
1267 foreach ( (array) $inval as $inkey2 => $inval2) {
1268 $outkey2 = strtolower($inkey2);
1269 $outarray[$outkey][$outkey2] = $inval2;
1271 } // foreach $inarray
1277 * Removes the HTML JavaScript entities found in early versions of Netscape 4.
1281 * @param string $string
1284 function wp_kses_js_entities($string) {
1285 return preg_replace('%&\s*\{[^}]*(\}\s*;?|$)%', '', $string);
1289 * Handles parsing errors in wp_kses_hair().
1291 * The general plan is to remove everything to and including some whitespace,
1292 * but it deals with quotes and apostrophes as well.
1296 * @param string $string
1299 function wp_kses_html_error($string) {
1300 return preg_replace('/^("[^"]*("|$)|\'[^\']*(\'|$)|\S)*\s*/', '', $string);
1304 * Sanitizes content from bad protocols and other characters.
1306 * This function searches for URL protocols at the beginning of $string, while
1307 * handling whitespace and HTML entities.
1311 * @param string $string Content to check for bad protocols
1312 * @param string $allowed_protocols Allowed protocols
1313 * @return string Sanitized content
1315 function wp_kses_bad_protocol_once($string, $allowed_protocols, $count = 1 ) {
1316 $string2 = preg_split( '/:|�*58;|�*3a;/i', $string, 2 );
1317 if ( isset($string2[1]) && ! preg_match('%/\?%', $string2[0]) ) {
1318 $string = trim( $string2[1] );
1319 $protocol = wp_kses_bad_protocol_once2( $string2[0], $allowed_protocols );
1320 if ( 'feed:' == $protocol ) {
1323 $string = wp_kses_bad_protocol_once( $string, $allowed_protocols, ++$count );
1324 if ( empty( $string ) )
1327 $string = $protocol . $string;
1334 * Callback for wp_kses_bad_protocol_once() regular expression.
1336 * This function processes URL protocols, checks to see if they're in the
1337 * whitelist or not, and returns different data depending on the answer.
1342 * @param string $string URI scheme to check against the whitelist
1343 * @param string $allowed_protocols Allowed protocols
1344 * @return string Sanitized content
1346 function wp_kses_bad_protocol_once2( $string, $allowed_protocols ) {
1347 $string2 = wp_kses_decode_entities($string);
1348 $string2 = preg_replace('/\s/', '', $string2);
1349 $string2 = wp_kses_no_null($string2);
1350 $string2 = strtolower($string2);
1353 foreach ( (array) $allowed_protocols as $one_protocol )
1354 if ( strtolower($one_protocol) == $string2 ) {
1366 * Converts and fixes HTML entities.
1368 * This function normalizes HTML entities. It will convert `AT&T` to the correct
1369 * `AT&T`, `:` to `:`, `&#XYZZY;` to `&#XYZZY;` and so on.
1373 * @param string $string Content to normalize entities
1374 * @return string Content with normalized entities
1376 function wp_kses_normalize_entities($string) {
1377 // Disarm all entities by converting & to &
1379 $string = str_replace('&', '&', $string);
1381 // Change back the allowed entities in our entity whitelist
1383 $string = preg_replace_callback('/&([A-Za-z]{2,8}[0-9]{0,2});/', 'wp_kses_named_entities', $string);
1384 $string = preg_replace_callback('/&#(0*[0-9]{1,7});/', 'wp_kses_normalize_entities2', $string);
1385 $string = preg_replace_callback('/&#[Xx](0*[0-9A-Fa-f]{1,6});/', 'wp_kses_normalize_entities3', $string);
1391 * Callback for wp_kses_normalize_entities() regular expression.
1393 * This function only accepts valid named entity references, which are finite,
1394 * case-sensitive, and highly scrutinized by HTML and XML validators.
1398 * @param array $matches preg_replace_callback() matches array
1399 * @return string Correctly encoded entity
1401 function wp_kses_named_entities($matches) {
1402 global $allowedentitynames;
1404 if ( empty($matches[1]) )
1408 return ( ( ! in_array($i, $allowedentitynames) ) ? "&$i;" : "&$i;" );
1412 * Callback for wp_kses_normalize_entities() regular expression.
1414 * This function helps {@see wp_kses_normalize_entities()} to only accept 16-bit
1415 * values and nothing more for `&#number;` entities.
1420 * @param array $matches preg_replace_callback() matches array
1421 * @return string Correctly encoded entity
1423 function wp_kses_normalize_entities2($matches) {
1424 if ( empty($matches[1]) )
1428 if (valid_unicode($i)) {
1429 $i = str_pad(ltrim($i,'0'), 3, '0', STR_PAD_LEFT);
1439 * Callback for wp_kses_normalize_entities() for regular expression.
1441 * This function helps wp_kses_normalize_entities() to only accept valid Unicode
1442 * numeric entities in hex form.
1446 * @param array $matches preg_replace_callback() matches array
1447 * @return string Correctly encoded entity
1449 function wp_kses_normalize_entities3($matches) {
1450 if ( empty($matches[1]) )
1453 $hexchars = $matches[1];
1454 return ( ( ! valid_unicode(hexdec($hexchars)) ) ? "&#x$hexchars;" : '&#x'.ltrim($hexchars,'0').';' );
1458 * Helper function to determine if a Unicode value is valid.
1460 * @param int $i Unicode value
1461 * @return bool True if the value was a valid Unicode number
1463 function valid_unicode($i) {
1464 return ( $i == 0x9 || $i == 0xa || $i == 0xd ||
1465 ($i >= 0x20 && $i <= 0xd7ff) ||
1466 ($i >= 0xe000 && $i <= 0xfffd) ||
1467 ($i >= 0x10000 && $i <= 0x10ffff) );
1471 * Convert all entities to their character counterparts.
1473 * This function decodes numeric HTML entities (`A` and `A`).
1474 * It doesn't do anything with other entities like ä, but we don't
1475 * need them in the URL protocol whitelisting system anyway.
1479 * @param string $string Content to change entities
1480 * @return string Content after decoded entities
1482 function wp_kses_decode_entities($string) {
1483 $string = preg_replace_callback('/&#([0-9]+);/', '_wp_kses_decode_entities_chr', $string);
1484 $string = preg_replace_callback('/&#[Xx]([0-9A-Fa-f]+);/', '_wp_kses_decode_entities_chr_hexdec', $string);
1490 * Regex callback for wp_kses_decode_entities()
1492 * @param array $match preg match
1495 function _wp_kses_decode_entities_chr( $match ) {
1496 return chr( $match[1] );
1500 * Regex callback for wp_kses_decode_entities()
1502 * @param array $match preg match
1505 function _wp_kses_decode_entities_chr_hexdec( $match ) {
1506 return chr( hexdec( $match[1] ) );
1510 * Sanitize content with allowed HTML Kses rules.
1514 * @param string $data Content to filter, expected to be escaped with slashes
1515 * @return string Filtered content
1517 function wp_filter_kses( $data ) {
1518 return addslashes( wp_kses( stripslashes( $data ), current_filter() ) );
1522 * Sanitize content with allowed HTML Kses rules.
1526 * @param string $data Content to filter, expected to not be escaped
1527 * @return string Filtered content
1529 function wp_kses_data( $data ) {
1530 return wp_kses( $data , current_filter() );
1534 * Sanitize content for allowed HTML tags for post content.
1536 * Post content refers to the page contents of the 'post' type and not $_POST
1541 * @param string $data Post content to filter, expected to be escaped with slashes
1542 * @return string Filtered post content with allowed HTML tags and attributes intact.
1544 function wp_filter_post_kses($data) {
1545 return addslashes ( wp_kses( stripslashes( $data ), 'post' ) );
1549 * Sanitize content for allowed HTML tags for post content.
1551 * Post content refers to the page contents of the 'post' type and not $_POST
1556 * @param string $data Post content to filter
1557 * @return string Filtered post content with allowed HTML tags and attributes intact.
1559 function wp_kses_post($data) {
1560 return wp_kses( $data , 'post' );
1564 * Strips all of the HTML in the content.
1568 * @param string $data Content to strip all HTML from
1569 * @return string Filtered content without any HTML
1571 function wp_filter_nohtml_kses( $data ) {
1572 return addslashes ( wp_kses( stripslashes( $data ), 'strip' ) );
1576 * Adds all Kses input form content filters.
1578 * All hooks have default priority. The wp_filter_kses() function is added to
1579 * the 'pre_comment_content' and 'title_save_pre' hooks.
1581 * The wp_filter_post_kses() function is added to the 'content_save_pre',
1582 * 'excerpt_save_pre', and 'content_filtered_save_pre' hooks.
1586 function kses_init_filters() {
1588 add_filter('title_save_pre', 'wp_filter_kses');
1590 // Comment filtering
1591 if ( current_user_can( 'unfiltered_html' ) )
1592 add_filter( 'pre_comment_content', 'wp_filter_post_kses' );
1594 add_filter( 'pre_comment_content', 'wp_filter_kses' );
1597 add_filter('content_save_pre', 'wp_filter_post_kses');
1598 add_filter('excerpt_save_pre', 'wp_filter_post_kses');
1599 add_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1603 * Removes all Kses input form content filters.
1605 * A quick procedural method to removing all of the filters that kses uses for
1606 * content in WordPress Loop.
1608 * Does not remove the kses_init() function from 'init' hook (priority is
1609 * default). Also does not remove kses_init() function from 'set_current_user'
1610 * hook (priority is also default).
1614 function kses_remove_filters() {
1616 remove_filter('title_save_pre', 'wp_filter_kses');
1618 // Comment filtering
1619 remove_filter( 'pre_comment_content', 'wp_filter_post_kses' );
1620 remove_filter( 'pre_comment_content', 'wp_filter_kses' );
1623 remove_filter('content_save_pre', 'wp_filter_post_kses');
1624 remove_filter('excerpt_save_pre', 'wp_filter_post_kses');
1625 remove_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1629 * Sets up most of the Kses filters for input form content.
1631 * If you remove the kses_init() function from 'init' hook and
1632 * 'set_current_user' (priority is default), then none of the Kses filter hooks
1635 * First removes all of the Kses filters in case the current user does not need
1636 * to have Kses filter the content. If the user does not have unfiltered_html
1637 * capability, then Kses filters are added.
1641 function kses_init() {
1642 kses_remove_filters();
1644 if (current_user_can('unfiltered_html') == false)
1645 kses_init_filters();
1653 function safecss_filter_attr( $css, $deprecated = '' ) {
1654 if ( !empty( $deprecated ) )
1655 _deprecated_argument( __FUNCTION__, '2.8.1' ); // Never implemented
1657 $css = wp_kses_no_null($css);
1658 $css = str_replace(array("\n","\r","\t"), '', $css);
1660 if ( preg_match( '%[\\\\(&=}]|/\*%', $css ) ) // remove any inline css containing \ ( & } = or comments
1663 $css_array = explode( ';', trim( $css ) );
1666 * Filter list of allowed CSS attributes.
1670 * @param array $attr List of allowed CSS attributes.
1672 $allowed_attr = apply_filters( 'safe_style_css', array( 'text-align', 'margin', 'color', 'float',
1673 'border', 'background', 'background-color', 'border-bottom', 'border-bottom-color',
1674 'border-bottom-style', 'border-bottom-width', 'border-collapse', 'border-color', 'border-left',
1675 'border-left-color', 'border-left-style', 'border-left-width', 'border-right', 'border-right-color',
1676 'border-right-style', 'border-right-width', 'border-spacing', 'border-style', 'border-top',
1677 'border-top-color', 'border-top-style', 'border-top-width', 'border-width', 'caption-side',
1678 'clear', 'cursor', 'direction', 'font', 'font-family', 'font-size', 'font-style',
1679 'font-variant', 'font-weight', 'height', 'letter-spacing', 'line-height', 'margin-bottom',
1680 'margin-left', 'margin-right', 'margin-top', 'overflow', 'padding', 'padding-bottom',
1681 'padding-left', 'padding-right', 'padding-top', 'text-decoration', 'text-indent', 'vertical-align',
1684 if ( empty($allowed_attr) )
1688 foreach ( $css_array as $css_item ) {
1689 if ( $css_item == '' )
1691 $css_item = trim( $css_item );
1693 if ( strpos( $css_item, ':' ) === false ) {
1696 $parts = explode( ':', $css_item );
1697 if ( in_array( trim( $parts[0] ), $allowed_attr ) )
1711 * Helper function to add global attributes to a tag in the allowed html list.
1716 * @param array $value An array of attributes.
1717 * @return array The array of attributes with global attributes added.
1719 function _wp_add_global_attributes( $value ) {
1720 $global_attributes = array(
1728 if ( true === $value )
1731 if ( is_array( $value ) )
1732 return array_merge( $value, $global_attributes );