3 * kses 0.2.2 - HTML/XHTML filter that only allows some elements and attributes
4 * Copyright (C) 2002, 2003, 2005 Ulf Harnhammar
6 * This program is free software and open source software; you can redistribute
7 * it and/or modify it under the terms of the GNU General Public License as
8 * published by the Free Software Foundation; either version 2 of the License,
9 * or (at your option) any later version.
11 * This program is distributed in the hope that it will be useful, but WITHOUT
12 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
13 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
16 * You should have received a copy of the GNU General Public License along
17 * with this program; if not, write to the Free Software Foundation, Inc.,
18 * 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA
19 * http://www.gnu.org/licenses/gpl.html
21 * [kses strips evil scripts!]
23 * Added wp_ prefix to avoid conflicts with existing kses users
26 * @copyright (C) 2002, 2003, 2005
27 * @author Ulf Harnhammar <http://advogato.org/person/metaur/>
35 * You can override this in a plugin.
37 * The wp_kses_allowed_html filter is more powerful and supplies context.
38 * CUSTOM_TAGS is not recommended and should be considered deprecated.
40 * @see wp_kses_allowed_html()
44 if ( ! defined( 'CUSTOM_TAGS' ) )
45 define( 'CUSTOM_TAGS', false );
47 if ( ! CUSTOM_TAGS ) {
49 * Kses global for default allowable HTML tags.
51 * Can be override by using CUSTOM_TAGS constant.
53 * @global array $allowedposttags
56 $allowedposttags = array(
89 'blockquote' => array(
139 'fieldset' => array(),
146 'figcaption' => array(
166 'accept-charset' => true,
290 'cellpadding' => true,
291 'cellspacing' => true,
376 * Kses allowed HTML elements.
378 * @global array $allowedtags
381 $allowedtags = array(
393 'blockquote' => array(
410 $allowedentitynames = array(
411 'nbsp', 'iexcl', 'cent', 'pound', 'curren', 'yen',
412 'brvbar', 'sect', 'uml', 'copy', 'ordf', 'laquo',
413 'not', 'shy', 'reg', 'macr', 'deg', 'plusmn',
414 'acute', 'micro', 'para', 'middot', 'cedil', 'ordm',
415 'raquo', 'iquest', 'Agrave', 'Aacute', 'Acirc', 'Atilde',
416 'Auml', 'Aring', 'AElig', 'Ccedil', 'Egrave', 'Eacute',
417 'Ecirc', 'Euml', 'Igrave', 'Iacute', 'Icirc', 'Iuml',
418 'ETH', 'Ntilde', 'Ograve', 'Oacute', 'Ocirc', 'Otilde',
419 'Ouml', 'times', 'Oslash', 'Ugrave', 'Uacute', 'Ucirc',
420 'Uuml', 'Yacute', 'THORN', 'szlig', 'agrave', 'aacute',
421 'acirc', 'atilde', 'auml', 'aring', 'aelig', 'ccedil',
422 'egrave', 'eacute', 'ecirc', 'euml', 'igrave', 'iacute',
423 'icirc', 'iuml', 'eth', 'ntilde', 'ograve', 'oacute',
424 'ocirc', 'otilde', 'ouml', 'divide', 'oslash', 'ugrave',
425 'uacute', 'ucirc', 'uuml', 'yacute', 'thorn', 'yuml',
426 'quot', 'amp', 'lt', 'gt', 'apos', 'OElig',
427 'oelig', 'Scaron', 'scaron', 'Yuml', 'circ', 'tilde',
428 'ensp', 'emsp', 'thinsp', 'zwnj', 'zwj', 'lrm',
429 'rlm', 'ndash', 'mdash', 'lsquo', 'rsquo', 'sbquo',
430 'ldquo', 'rdquo', 'bdquo', 'dagger', 'Dagger', 'permil',
431 'lsaquo', 'rsaquo', 'euro', 'fnof', 'Alpha', 'Beta',
432 'Gamma', 'Delta', 'Epsilon', 'Zeta', 'Eta', 'Theta',
433 'Iota', 'Kappa', 'Lambda', 'Mu', 'Nu', 'Xi',
434 'Omicron', 'Pi', 'Rho', 'Sigma', 'Tau', 'Upsilon',
435 'Phi', 'Chi', 'Psi', 'Omega', 'alpha', 'beta',
436 'gamma', 'delta', 'epsilon', 'zeta', 'eta', 'theta',
437 'iota', 'kappa', 'lambda', 'mu', 'nu', 'xi',
438 'omicron', 'pi', 'rho', 'sigmaf', 'sigma', 'tau',
439 'upsilon', 'phi', 'chi', 'psi', 'omega', 'thetasym',
440 'upsih', 'piv', 'bull', 'hellip', 'prime', 'Prime',
441 'oline', 'frasl', 'weierp', 'image', 'real', 'trade',
442 'alefsym', 'larr', 'uarr', 'rarr', 'darr', 'harr',
443 'crarr', 'lArr', 'uArr', 'rArr', 'dArr', 'hArr',
444 'forall', 'part', 'exist', 'empty', 'nabla', 'isin',
445 'notin', 'ni', 'prod', 'sum', 'minus', 'lowast',
446 'radic', 'prop', 'infin', 'ang', 'and', 'or',
447 'cap', 'cup', 'int', 'sim', 'cong', 'asymp',
448 'ne', 'equiv', 'le', 'ge', 'sub', 'sup',
449 'nsub', 'sube', 'supe', 'oplus', 'otimes', 'perp',
450 'sdot', 'lceil', 'rceil', 'lfloor', 'rfloor', 'lang',
451 'rang', 'loz', 'spades', 'clubs', 'hearts', 'diams',
452 'sup1', 'sup2', 'sup3', 'frac14', 'frac12', 'frac34',
456 $allowedposttags = array_map( '_wp_add_global_attributes', $allowedposttags );
458 $allowedtags = wp_kses_array_lc( $allowedtags );
459 $allowedposttags = wp_kses_array_lc( $allowedposttags );
463 * Filters content and keeps only allowable HTML elements.
465 * This function makes sure that only the allowed HTML element names, attribute
466 * names and attribute values plus only sane HTML entities will occur in
467 * $string. You have to remove any slashes from PHP's magic quotes before you
468 * call this function.
470 * The default allowed protocols are 'http', 'https', 'ftp', 'mailto', 'news',
471 * 'irc', 'gopher', 'nntp', 'feed', 'telnet, 'mms', 'rtsp' and 'svn'. This
472 * covers all common link protocols, except for 'javascript' which should not
473 * be allowed for untrusted users.
477 * @param string $string Content to filter through kses
478 * @param array $allowed_html List of allowed HTML elements
479 * @param array $allowed_protocols Optional. Allowed protocol in links.
480 * @return string Filtered content with only allowed HTML elements
482 function wp_kses( $string, $allowed_html, $allowed_protocols = array() ) {
483 if ( empty( $allowed_protocols ) )
484 $allowed_protocols = wp_allowed_protocols();
485 $string = wp_kses_no_null($string);
486 $string = wp_kses_js_entities($string);
487 $string = wp_kses_normalize_entities($string);
488 $string = wp_kses_hook($string, $allowed_html, $allowed_protocols); // WP changed the order of these funcs and added args to wp_kses_hook
489 return wp_kses_split($string, $allowed_html, $allowed_protocols);
493 * Return a list of allowed tags and attributes for a given context.
497 * @param string $context The context for which to retrieve tags. Allowed values are
498 * post | strip | data | entities or the name of a field filter such as pre_user_description.
499 * @return array List of allowed tags and their allowed attributes.
501 function wp_kses_allowed_html( $context = '' ) {
502 global $allowedposttags, $allowedtags, $allowedentitynames;
504 if ( is_array( $context ) ) {
506 * Filter HTML elements allowed for a given context.
510 * @param string $tags Allowed tags, attributes, and/or entities.
511 * @param string $context Context to judge allowed tags by. Allowed values are 'post',
512 * 'data', 'strip', 'entities', 'explicit', or the name of a filter.
514 return apply_filters( 'wp_kses_allowed_html', $context, 'explicit' );
517 switch ( $context ) {
519 /** This filter is documented in wp-includes/kses.php */
520 return apply_filters( 'wp_kses_allowed_html', $allowedposttags, $context );
522 case 'user_description':
523 case 'pre_user_description':
524 $tags = $allowedtags;
525 $tags['a']['rel'] = true;
526 /** This filter is documented in wp-includes/kses.php */
527 return apply_filters( 'wp_kses_allowed_html', $tags, $context );
530 /** This filter is documented in wp-includes/kses.php */
531 return apply_filters( 'wp_kses_allowed_html', array(), $context );
534 /** This filter is documented in wp-includes/kses.php */
535 return apply_filters( 'wp_kses_allowed_html', $allowedentitynames, $context);
539 /** This filter is documented in wp-includes/kses.php */
540 return apply_filters( 'wp_kses_allowed_html', $allowedtags, $context );
545 * You add any kses hooks here.
547 * There is currently only one kses WordPress hook and it is called here. All
548 * parameters are passed to the hooks and expected to receive a string.
552 * @param string $string Content to filter through kses
553 * @param array $allowed_html List of allowed HTML elements
554 * @param array $allowed_protocols Allowed protocol in links
555 * @return string Filtered content through 'pre_kses' hook
557 function wp_kses_hook( $string, $allowed_html, $allowed_protocols ) {
559 * Filter content to be run through kses.
563 * @param string $string Content to run through kses.
564 * @param array $allowed_html Allowed HTML elements.
565 * @param array $allowed_protocols Allowed protocol in links.
567 $string = apply_filters( 'pre_kses', $string, $allowed_html, $allowed_protocols );
572 * This function returns kses' version number.
576 * @return string KSES Version Number
578 function wp_kses_version() {
583 * Searches for HTML tags, no matter how malformed.
585 * It also matches stray ">" characters.
589 * @param string $string Content to filter
590 * @param array $allowed_html Allowed HTML elements
591 * @param array $allowed_protocols Allowed protocols to keep
592 * @return string Content with fixed HTML tags
594 function wp_kses_split( $string, $allowed_html, $allowed_protocols ) {
595 global $pass_allowed_html, $pass_allowed_protocols;
596 $pass_allowed_html = $allowed_html;
597 $pass_allowed_protocols = $allowed_protocols;
598 return preg_replace_callback( '%(<!--.*?(-->|$))|(<[^>]*(>|$)|>)%', '_wp_kses_split_callback', $string );
602 * Callback for wp_kses_split.
607 function _wp_kses_split_callback( $match ) {
608 global $pass_allowed_html, $pass_allowed_protocols;
609 return wp_kses_split2( $match[0], $pass_allowed_html, $pass_allowed_protocols );
613 * Callback for wp_kses_split for fixing malformed HTML tags.
615 * This function does a lot of work. It rejects some very malformed things like
616 * <:::>. It returns an empty string, if the element isn't allowed (look ma, no
617 * strip_tags()!). Otherwise it splits the tag into an element and an attribute
620 * After the tag is split into an element and an attribute list, it is run
621 * through another filter which will remove illegal attributes and once that is
622 * completed, will be returned.
626 * @uses wp_kses_attr()
628 * @param string $string Content to filter
629 * @param array $allowed_html Allowed HTML elements
630 * @param array $allowed_protocols Allowed protocols to keep
631 * @return string Fixed HTML element
633 function wp_kses_split2($string, $allowed_html, $allowed_protocols) {
634 $string = wp_kses_stripslashes($string);
636 if (substr($string, 0, 1) != '<')
638 # It matched a ">" character
640 if ( '<!--' == substr( $string, 0, 4 ) ) {
641 $string = str_replace( array('<!--', '-->'), '', $string );
642 while ( $string != ($newstring = wp_kses($string, $allowed_html, $allowed_protocols)) )
643 $string = $newstring;
646 // prevent multiple dashes in comments
647 $string = preg_replace('/--+/', '-', $string);
648 // prevent three dashes closing a comment
649 $string = preg_replace('/-$/', '', $string);
650 return "<!--{$string}-->";
652 # Allow HTML comments
654 if (!preg_match('%^<\s*(/\s*)?([a-zA-Z0-9]+)([^>]*)>?$%', $string, $matches))
656 # It's seriously malformed
658 $slash = trim($matches[1]);
660 $attrlist = $matches[3];
662 if ( ! is_array( $allowed_html ) )
663 $allowed_html = wp_kses_allowed_html( $allowed_html );
665 if ( ! isset($allowed_html[strtolower($elem)]) )
667 # They are using a not allowed HTML element
671 # No attributes are allowed for closing elements
673 return wp_kses_attr( $elem, $attrlist, $allowed_html, $allowed_protocols );
677 * Removes all attributes, if none are allowed for this element.
679 * If some are allowed it calls wp_kses_hair() to split them further, and then
680 * it builds up new HTML code from the data that kses_hair() returns. It also
681 * removes "<" and ">" characters, if there are any left. One more thing it does
682 * is to check if the tag has a closing XHTML slash, and if it does, it puts one
683 * in the returned code as well.
687 * @param string $element HTML element/tag
688 * @param string $attr HTML attributes from HTML element to closing HTML element tag
689 * @param array $allowed_html Allowed HTML elements
690 * @param array $allowed_protocols Allowed protocols to keep
691 * @return string Sanitized HTML element
693 function wp_kses_attr($element, $attr, $allowed_html, $allowed_protocols) {
694 # Is there a closing XHTML slash at the end of the attributes?
696 if ( ! is_array( $allowed_html ) )
697 $allowed_html = wp_kses_allowed_html( $allowed_html );
700 if (preg_match('%\s*/\s*$%', $attr))
703 # Are any attributes allowed at all for this element?
704 if ( ! isset($allowed_html[strtolower($element)]) || count($allowed_html[strtolower($element)]) == 0 )
705 return "<$element$xhtml_slash>";
708 $attrarr = wp_kses_hair($attr, $allowed_protocols);
710 # Go through $attrarr, and save the allowed attributes for this element
714 $allowed_attr = $allowed_html[strtolower($element)];
715 foreach ($attrarr as $arreach) {
716 if ( ! isset( $allowed_attr[strtolower($arreach['name'])] ) )
717 continue; # the attribute is not allowed
719 $current = $allowed_attr[strtolower($arreach['name'])];
720 if ( $current == '' )
721 continue; # the attribute is not allowed
723 if ( strtolower( $arreach['name'] ) == 'style' ) {
724 $orig_value = $arreach['value'];
725 $value = safecss_filter_attr( $orig_value );
727 if ( empty( $value ) )
730 $arreach['value'] = $value;
731 $arreach['whole'] = str_replace( $orig_value, $value, $arreach['whole'] );
734 if ( ! is_array($current) ) {
735 $attr2 .= ' '.$arreach['whole'];
736 # there are no checks
739 # there are some checks
741 foreach ($current as $currkey => $currval) {
742 if ( ! wp_kses_check_attr_val($arreach['value'], $arreach['vless'], $currkey, $currval) ) {
749 $attr2 .= ' '.$arreach['whole']; # it passed them
750 } # if !is_array($current)
753 # Remove any "<" or ">" characters
754 $attr2 = preg_replace('/[<>]/', '', $attr2);
756 return "<$element$attr2$xhtml_slash>";
760 * Builds an attribute list from string containing attributes.
762 * This function does a lot of work. It parses an attribute list into an array
763 * with attribute data, and tries to do the right thing even if it gets weird
764 * input. It will add quotes around attribute values that don't have any quotes
765 * or apostrophes around them, to make it easier to produce HTML code that will
766 * conform to W3C's HTML specification. It will also remove bad URL protocols
767 * from attribute values. It also reduces duplicate attributes by using the
768 * attribute defined first (foo='bar' foo='baz' will result in foo='bar').
772 * @param string $attr Attribute list from HTML element to closing HTML element tag
773 * @param array $allowed_protocols Allowed protocols to keep
774 * @return array List of attributes after parsing
776 function wp_kses_hair($attr, $allowed_protocols) {
780 $uris = array('xmlns', 'profile', 'href', 'src', 'cite', 'classid', 'codebase', 'data', 'usemap', 'longdesc', 'action');
782 # Loop through the whole attribute list
784 while (strlen($attr) != 0) {
785 $working = 0; # Was the last operation successful?
788 case 0 : # attribute name, href for instance
790 if ( preg_match('/^([-a-zA-Z:]+)/', $attr, $match ) ) {
791 $attrname = $match[1];
792 $working = $mode = 1;
793 $attr = preg_replace( '/^[-a-zA-Z:]+/', '', $attr );
798 case 1 : # equals sign or valueless ("selected")
800 if (preg_match('/^\s*=\s*/', $attr)) # equals sign
804 $attr = preg_replace('/^\s*=\s*/', '', $attr);
808 if (preg_match('/^\s+/', $attr)) # valueless
812 if(false === array_key_exists($attrname, $attrarr)) {
813 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
815 $attr = preg_replace('/^\s+/', '', $attr);
820 case 2 : # attribute value, a URL after href= for instance
822 if (preg_match('%^"([^"]*)"(\s+|/?$)%', $attr, $match))
825 $thisval = $match[1];
826 if ( in_array(strtolower($attrname), $uris) )
827 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
829 if(false === array_key_exists($attrname, $attrarr)) {
830 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n');
834 $attr = preg_replace('/^"[^"]*"(\s+|$)/', '', $attr);
838 if (preg_match("%^'([^']*)'(\s+|/?$)%", $attr, $match))
841 $thisval = $match[1];
842 if ( in_array(strtolower($attrname), $uris) )
843 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
845 if(false === array_key_exists($attrname, $attrarr)) {
846 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname='$thisval'", 'vless' => 'n');
850 $attr = preg_replace("/^'[^']*'(\s+|$)/", '', $attr);
854 if (preg_match("%^([^\s\"']+)(\s+|/?$)%", $attr, $match))
857 $thisval = $match[1];
858 if ( in_array(strtolower($attrname), $uris) )
859 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
861 if(false === array_key_exists($attrname, $attrarr)) {
862 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n');
864 # We add quotes to conform to W3C's HTML spec.
867 $attr = preg_replace("%^[^\s\"']+(\s+|$)%", '', $attr);
873 if ($working == 0) # not well formed, remove and try again
875 $attr = wp_kses_html_error($attr);
880 if ($mode == 1 && false === array_key_exists($attrname, $attrarr))
881 # special case, for when the attribute list ends with a valueless
882 # attribute like "selected"
883 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
889 * Performs different checks for attribute values.
891 * The currently implemented checks are "maxlen", "minlen", "maxval", "minval"
896 * @param string $value Attribute value
897 * @param string $vless Whether the value is valueless. Use 'y' or 'n'
898 * @param string $checkname What $checkvalue is checking for.
899 * @param mixed $checkvalue What constraint the value should pass
900 * @return bool Whether check passes
902 function wp_kses_check_attr_val($value, $vless, $checkname, $checkvalue) {
905 switch (strtolower($checkname)) {
907 # The maxlen check makes sure that the attribute value has a length not
908 # greater than the given value. This can be used to avoid Buffer Overflows
909 # in WWW clients and various Internet servers.
911 if (strlen($value) > $checkvalue)
916 # The minlen check makes sure that the attribute value has a length not
917 # smaller than the given value.
919 if (strlen($value) < $checkvalue)
924 # The maxval check does two things: it checks that the attribute value is
925 # an integer from 0 and up, without an excessive amount of zeroes or
926 # whitespace (to avoid Buffer Overflows). It also checks that the attribute
927 # value is not greater than the given value.
928 # This check can be used to avoid Denial of Service attacks.
930 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
932 if ($value > $checkvalue)
937 # The minval check makes sure that the attribute value is a positive integer,
938 # and that it is not smaller than the given value.
940 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
942 if ($value < $checkvalue)
947 # The valueless check makes sure if the attribute has a value
948 # (like <a href="blah">) or not (<option selected>). If the given value
949 # is a "y" or a "Y", the attribute must not have a value.
950 # If the given value is an "n" or an "N", the attribute must have one.
952 if (strtolower($checkvalue) != $vless)
961 * Sanitize string from bad protocols.
963 * This function removes all non-allowed protocols from the beginning of
964 * $string. It ignores whitespace and the case of the letters, and it does
965 * understand HTML entities. It does its work in a while loop, so it won't be
966 * fooled by a string like "javascript:javascript:alert(57)".
970 * @param string $string Content to filter bad protocols from
971 * @param array $allowed_protocols Allowed protocols to keep
972 * @return string Filtered content
974 function wp_kses_bad_protocol($string, $allowed_protocols) {
975 $string = wp_kses_no_null($string);
979 $original_string = $string;
980 $string = wp_kses_bad_protocol_once($string, $allowed_protocols);
981 } while ( $original_string != $string && ++$iterations < 6 );
983 if ( $original_string != $string )
990 * Removes any null characters in $string.
994 * @param string $string
997 function wp_kses_no_null($string) {
998 $string = preg_replace('/\0+/', '', $string);
999 $string = preg_replace('/(\\\\0)+/', '', $string);
1005 * Strips slashes from in front of quotes.
1007 * This function changes the character sequence \" to just ". It leaves all
1008 * other slashes alone. It's really weird, but the quoting from
1009 * preg_replace(//e) seems to require this.
1013 * @param string $string String to strip slashes
1014 * @return string Fixed string with quoted slashes
1016 function wp_kses_stripslashes($string) {
1017 return preg_replace('%\\\\"%', '"', $string);
1021 * Goes through an array and changes the keys to all lower case.
1025 * @param array $inarray Unfiltered array
1026 * @return array Fixed array with all lowercase keys
1028 function wp_kses_array_lc($inarray) {
1029 $outarray = array ();
1031 foreach ( (array) $inarray as $inkey => $inval) {
1032 $outkey = strtolower($inkey);
1033 $outarray[$outkey] = array ();
1035 foreach ( (array) $inval as $inkey2 => $inval2) {
1036 $outkey2 = strtolower($inkey2);
1037 $outarray[$outkey][$outkey2] = $inval2;
1039 } # foreach $inarray
1045 * Removes the HTML JavaScript entities found in early versions of Netscape 4.
1049 * @param string $string
1052 function wp_kses_js_entities($string) {
1053 return preg_replace('%&\s*\{[^}]*(\}\s*;?|$)%', '', $string);
1057 * Handles parsing errors in wp_kses_hair().
1059 * The general plan is to remove everything to and including some whitespace,
1060 * but it deals with quotes and apostrophes as well.
1064 * @param string $string
1067 function wp_kses_html_error($string) {
1068 return preg_replace('/^("[^"]*("|$)|\'[^\']*(\'|$)|\S)*\s*/', '', $string);
1072 * Sanitizes content from bad protocols and other characters.
1074 * This function searches for URL protocols at the beginning of $string, while
1075 * handling whitespace and HTML entities.
1079 * @param string $string Content to check for bad protocols
1080 * @param string $allowed_protocols Allowed protocols
1081 * @return string Sanitized content
1083 function wp_kses_bad_protocol_once($string, $allowed_protocols, $count = 1 ) {
1084 $string2 = preg_split( '/:|�*58;|�*3a;/i', $string, 2 );
1085 if ( isset($string2[1]) && ! preg_match('%/\?%', $string2[0]) ) {
1086 $string = trim( $string2[1] );
1087 $protocol = wp_kses_bad_protocol_once2( $string2[0], $allowed_protocols );
1088 if ( 'feed:' == $protocol ) {
1091 $string = wp_kses_bad_protocol_once( $string, $allowed_protocols, ++$count );
1092 if ( empty( $string ) )
1095 $string = $protocol . $string;
1102 * Callback for wp_kses_bad_protocol_once() regular expression.
1104 * This function processes URL protocols, checks to see if they're in the
1105 * whitelist or not, and returns different data depending on the answer.
1110 * @param string $string URI scheme to check against the whitelist
1111 * @param string $allowed_protocols Allowed protocols
1112 * @return string Sanitized content
1114 function wp_kses_bad_protocol_once2( $string, $allowed_protocols ) {
1115 $string2 = wp_kses_decode_entities($string);
1116 $string2 = preg_replace('/\s/', '', $string2);
1117 $string2 = wp_kses_no_null($string2);
1118 $string2 = strtolower($string2);
1121 foreach ( (array) $allowed_protocols as $one_protocol )
1122 if ( strtolower($one_protocol) == $string2 ) {
1134 * Converts and fixes HTML entities.
1136 * This function normalizes HTML entities. It will convert "AT&T" to the correct
1137 * "AT&T", ":" to ":", "&#XYZZY;" to "&#XYZZY;" and so on.
1141 * @param string $string Content to normalize entities
1142 * @return string Content with normalized entities
1144 function wp_kses_normalize_entities($string) {
1145 # Disarm all entities by converting & to &
1147 $string = str_replace('&', '&', $string);
1149 # Change back the allowed entities in our entity whitelist
1151 $string = preg_replace_callback('/&([A-Za-z]{2,8}[0-9]{0,2});/', 'wp_kses_named_entities', $string);
1152 $string = preg_replace_callback('/&#(0*[0-9]{1,7});/', 'wp_kses_normalize_entities2', $string);
1153 $string = preg_replace_callback('/&#[Xx](0*[0-9A-Fa-f]{1,6});/', 'wp_kses_normalize_entities3', $string);
1159 * Callback for wp_kses_normalize_entities() regular expression.
1161 * This function only accepts valid named entity references, which are finite,
1162 * case-sensitive, and highly scrutinized by HTML and XML validators.
1166 * @param array $matches preg_replace_callback() matches array
1167 * @return string Correctly encoded entity
1169 function wp_kses_named_entities($matches) {
1170 global $allowedentitynames;
1172 if ( empty($matches[1]) )
1176 return ( ( ! in_array($i, $allowedentitynames) ) ? "&$i;" : "&$i;" );
1180 * Callback for wp_kses_normalize_entities() regular expression.
1182 * This function helps wp_kses_normalize_entities() to only accept 16-bit values
1183 * and nothing more for &#number; entities.
1188 * @param array $matches preg_replace_callback() matches array
1189 * @return string Correctly encoded entity
1191 function wp_kses_normalize_entities2($matches) {
1192 if ( empty($matches[1]) )
1196 if (valid_unicode($i)) {
1197 $i = str_pad(ltrim($i,'0'), 3, '0', STR_PAD_LEFT);
1207 * Callback for wp_kses_normalize_entities() for regular expression.
1209 * This function helps wp_kses_normalize_entities() to only accept valid Unicode
1210 * numeric entities in hex form.
1214 * @param array $matches preg_replace_callback() matches array
1215 * @return string Correctly encoded entity
1217 function wp_kses_normalize_entities3($matches) {
1218 if ( empty($matches[1]) )
1221 $hexchars = $matches[1];
1222 return ( ( ! valid_unicode(hexdec($hexchars)) ) ? "&#x$hexchars;" : '&#x'.ltrim($hexchars,'0').';' );
1226 * Helper function to determine if a Unicode value is valid.
1228 * @param int $i Unicode value
1229 * @return bool True if the value was a valid Unicode number
1231 function valid_unicode($i) {
1232 return ( $i == 0x9 || $i == 0xa || $i == 0xd ||
1233 ($i >= 0x20 && $i <= 0xd7ff) ||
1234 ($i >= 0xe000 && $i <= 0xfffd) ||
1235 ($i >= 0x10000 && $i <= 0x10ffff) );
1239 * Convert all entities to their character counterparts.
1241 * This function decodes numeric HTML entities (A and A). It doesn't do
1242 * anything with other entities like ä, but we don't need them in the URL
1243 * protocol whitelisting system anyway.
1247 * @param string $string Content to change entities
1248 * @return string Content after decoded entities
1250 function wp_kses_decode_entities($string) {
1251 $string = preg_replace_callback('/&#([0-9]+);/', '_wp_kses_decode_entities_chr', $string);
1252 $string = preg_replace_callback('/&#[Xx]([0-9A-Fa-f]+);/', '_wp_kses_decode_entities_chr_hexdec', $string);
1258 * Regex callback for wp_kses_decode_entities()
1260 * @param array $match preg match
1263 function _wp_kses_decode_entities_chr( $match ) {
1264 return chr( $match[1] );
1268 * Regex callback for wp_kses_decode_entities()
1270 * @param array $match preg match
1273 function _wp_kses_decode_entities_chr_hexdec( $match ) {
1274 return chr( hexdec( $match[1] ) );
1278 * Sanitize content with allowed HTML Kses rules.
1281 * @uses $allowedtags
1283 * @param string $data Content to filter, expected to be escaped with slashes
1284 * @return string Filtered content
1286 function wp_filter_kses( $data ) {
1287 return addslashes( wp_kses( stripslashes( $data ), current_filter() ) );
1291 * Sanitize content with allowed HTML Kses rules.
1294 * @uses $allowedtags
1296 * @param string $data Content to filter, expected to not be escaped
1297 * @return string Filtered content
1299 function wp_kses_data( $data ) {
1300 return wp_kses( $data , current_filter() );
1304 * Sanitize content for allowed HTML tags for post content.
1306 * Post content refers to the page contents of the 'post' type and not $_POST
1311 * @param string $data Post content to filter, expected to be escaped with slashes
1312 * @return string Filtered post content with allowed HTML tags and attributes intact.
1314 function wp_filter_post_kses($data) {
1315 return addslashes ( wp_kses( stripslashes( $data ), 'post' ) );
1319 * Sanitize content for allowed HTML tags for post content.
1321 * Post content refers to the page contents of the 'post' type and not $_POST
1326 * @param string $data Post content to filter
1327 * @return string Filtered post content with allowed HTML tags and attributes intact.
1329 function wp_kses_post($data) {
1330 return wp_kses( $data , 'post' );
1334 * Strips all of the HTML in the content.
1338 * @param string $data Content to strip all HTML from
1339 * @return string Filtered content without any HTML
1341 function wp_filter_nohtml_kses( $data ) {
1342 return addslashes ( wp_kses( stripslashes( $data ), 'strip' ) );
1346 * Adds all Kses input form content filters.
1348 * All hooks have default priority. The wp_filter_kses() function is added to
1349 * the 'pre_comment_content' and 'title_save_pre' hooks.
1351 * The wp_filter_post_kses() function is added to the 'content_save_pre',
1352 * 'excerpt_save_pre', and 'content_filtered_save_pre' hooks.
1355 * @uses add_filter() See description for what functions are added to what hooks.
1357 function kses_init_filters() {
1359 add_filter('title_save_pre', 'wp_filter_kses');
1361 // Comment filtering
1362 if ( current_user_can( 'unfiltered_html' ) )
1363 add_filter( 'pre_comment_content', 'wp_filter_post_kses' );
1365 add_filter( 'pre_comment_content', 'wp_filter_kses' );
1368 add_filter('content_save_pre', 'wp_filter_post_kses');
1369 add_filter('excerpt_save_pre', 'wp_filter_post_kses');
1370 add_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1374 * Removes all Kses input form content filters.
1376 * A quick procedural method to removing all of the filters that kses uses for
1377 * content in WordPress Loop.
1379 * Does not remove the kses_init() function from 'init' hook (priority is
1380 * default). Also does not remove kses_init() function from 'set_current_user'
1381 * hook (priority is also default).
1385 function kses_remove_filters() {
1387 remove_filter('title_save_pre', 'wp_filter_kses');
1389 // Comment filtering
1390 remove_filter( 'pre_comment_content', 'wp_filter_post_kses' );
1391 remove_filter( 'pre_comment_content', 'wp_filter_kses' );
1394 remove_filter('content_save_pre', 'wp_filter_post_kses');
1395 remove_filter('excerpt_save_pre', 'wp_filter_post_kses');
1396 remove_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1400 * Sets up most of the Kses filters for input form content.
1402 * If you remove the kses_init() function from 'init' hook and
1403 * 'set_current_user' (priority is default), then none of the Kses filter hooks
1406 * First removes all of the Kses filters in case the current user does not need
1407 * to have Kses filter the content. If the user does not have unfiltered_html
1408 * capability, then Kses filters are added.
1410 * @uses kses_remove_filters() Removes the Kses filters
1411 * @uses kses_init_filters() Adds the Kses filters back if the user
1412 * does not have unfiltered HTML capability.
1415 function kses_init() {
1416 kses_remove_filters();
1418 if (current_user_can('unfiltered_html') == false)
1419 kses_init_filters();
1422 add_action('init', 'kses_init');
1423 add_action('set_current_user', 'kses_init');
1430 function safecss_filter_attr( $css, $deprecated = '' ) {
1431 if ( !empty( $deprecated ) )
1432 _deprecated_argument( __FUNCTION__, '2.8.1' ); // Never implemented
1434 $css = wp_kses_no_null($css);
1435 $css = str_replace(array("\n","\r","\t"), '', $css);
1437 if ( preg_match( '%[\\(&=}]|/\*%', $css ) ) // remove any inline css containing \ ( & } = or comments
1440 $css_array = explode( ';', trim( $css ) );
1443 * Filter list of allowed CSS attributes.
1447 * @param array $attr List of allowed CSS attributes.
1449 $allowed_attr = apply_filters( 'safe_style_css', array( 'text-align', 'margin', 'color', 'float',
1450 'border', 'background', 'background-color', 'border-bottom', 'border-bottom-color',
1451 'border-bottom-style', 'border-bottom-width', 'border-collapse', 'border-color', 'border-left',
1452 'border-left-color', 'border-left-style', 'border-left-width', 'border-right', 'border-right-color',
1453 'border-right-style', 'border-right-width', 'border-spacing', 'border-style', 'border-top',
1454 'border-top-color', 'border-top-style', 'border-top-width', 'border-width', 'caption-side',
1455 'clear', 'cursor', 'direction', 'font', 'font-family', 'font-size', 'font-style',
1456 'font-variant', 'font-weight', 'height', 'letter-spacing', 'line-height', 'margin-bottom',
1457 'margin-left', 'margin-right', 'margin-top', 'overflow', 'padding', 'padding-bottom',
1458 'padding-left', 'padding-right', 'padding-top', 'text-decoration', 'text-indent', 'vertical-align',
1461 if ( empty($allowed_attr) )
1465 foreach ( $css_array as $css_item ) {
1466 if ( $css_item == '' )
1468 $css_item = trim( $css_item );
1470 if ( strpos( $css_item, ':' ) === false ) {
1473 $parts = explode( ':', $css_item );
1474 if ( in_array( trim( $parts[0] ), $allowed_attr ) )
1488 * Helper function to add global attributes to a tag in the allowed html list.
1493 * @param array $value An array of attributes.
1494 * @return array The array of attributes with global attributes added.
1496 function _wp_add_global_attributes( $value ) {
1497 $global_attributes = array(
1505 if ( true === $value )
1508 if ( is_array( $value ) )
1509 return array_merge( $value, $global_attributes );