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(
138 'fieldset' => array(),
145 'figcaption' => array(
165 'accept-charset' => true,
286 'cellpadding' => true,
287 'cellspacing' => true,
372 * Kses allowed HTML elements.
374 * @global array $allowedtags
377 $allowedtags = array(
389 'blockquote' => array(
406 $allowedentitynames = array(
407 'nbsp', 'iexcl', 'cent', 'pound', 'curren', 'yen',
408 'brvbar', 'sect', 'uml', 'copy', 'ordf', 'laquo',
409 'not', 'shy', 'reg', 'macr', 'deg', 'plusmn',
410 'acute', 'micro', 'para', 'middot', 'cedil', 'ordm',
411 'raquo', 'iquest', 'Agrave', 'Aacute', 'Acirc', 'Atilde',
412 'Auml', 'Aring', 'AElig', 'Ccedil', 'Egrave', 'Eacute',
413 'Ecirc', 'Euml', 'Igrave', 'Iacute', 'Icirc', 'Iuml',
414 'ETH', 'Ntilde', 'Ograve', 'Oacute', 'Ocirc', 'Otilde',
415 'Ouml', 'times', 'Oslash', 'Ugrave', 'Uacute', 'Ucirc',
416 'Uuml', 'Yacute', 'THORN', 'szlig', 'agrave', 'aacute',
417 'acirc', 'atilde', 'auml', 'aring', 'aelig', 'ccedil',
418 'egrave', 'eacute', 'ecirc', 'euml', 'igrave', 'iacute',
419 'icirc', 'iuml', 'eth', 'ntilde', 'ograve', 'oacute',
420 'ocirc', 'otilde', 'ouml', 'divide', 'oslash', 'ugrave',
421 'uacute', 'ucirc', 'uuml', 'yacute', 'thorn', 'yuml',
422 'quot', 'amp', 'lt', 'gt', 'apos', 'OElig',
423 'oelig', 'Scaron', 'scaron', 'Yuml', 'circ', 'tilde',
424 'ensp', 'emsp', 'thinsp', 'zwnj', 'zwj', 'lrm',
425 'rlm', 'ndash', 'mdash', 'lsquo', 'rsquo', 'sbquo',
426 'ldquo', 'rdquo', 'bdquo', 'dagger', 'Dagger', 'permil',
427 'lsaquo', 'rsaquo', 'euro', 'fnof', 'Alpha', 'Beta',
428 'Gamma', 'Delta', 'Epsilon', 'Zeta', 'Eta', 'Theta',
429 'Iota', 'Kappa', 'Lambda', 'Mu', 'Nu', 'Xi',
430 'Omicron', 'Pi', 'Rho', 'Sigma', 'Tau', 'Upsilon',
431 'Phi', 'Chi', 'Psi', 'Omega', 'alpha', 'beta',
432 'gamma', 'delta', 'epsilon', 'zeta', 'eta', 'theta',
433 'iota', 'kappa', 'lambda', 'mu', 'nu', 'xi',
434 'omicron', 'pi', 'rho', 'sigmaf', 'sigma', 'tau',
435 'upsilon', 'phi', 'chi', 'psi', 'omega', 'thetasym',
436 'upsih', 'piv', 'bull', 'hellip', 'prime', 'Prime',
437 'oline', 'frasl', 'weierp', 'image', 'real', 'trade',
438 'alefsym', 'larr', 'uarr', 'rarr', 'darr', 'harr',
439 'crarr', 'lArr', 'uArr', 'rArr', 'dArr', 'hArr',
440 'forall', 'part', 'exist', 'empty', 'nabla', 'isin',
441 'notin', 'ni', 'prod', 'sum', 'minus', 'lowast',
442 'radic', 'prop', 'infin', 'ang', 'and', 'or',
443 'cap', 'cup', 'int', 'sim', 'cong', 'asymp',
444 'ne', 'equiv', 'le', 'ge', 'sub', 'sup',
445 'nsub', 'sube', 'supe', 'oplus', 'otimes', 'perp',
446 'sdot', 'lceil', 'rceil', 'lfloor', 'rfloor', 'lang',
447 'rang', 'loz', 'spades', 'clubs', 'hearts', 'diams',
450 $allowedposttags = array_map( '_wp_add_global_attributes', $allowedposttags );
452 $allowedtags = wp_kses_array_lc( $allowedtags );
453 $allowedposttags = wp_kses_array_lc( $allowedposttags );
457 * Filters content and keeps only allowable HTML elements.
459 * This function makes sure that only the allowed HTML element names, attribute
460 * names and attribute values plus only sane HTML entities will occur in
461 * $string. You have to remove any slashes from PHP's magic quotes before you
462 * call this function.
464 * The default allowed protocols are 'http', 'https', 'ftp', 'mailto', 'news',
465 * 'irc', 'gopher', 'nntp', 'feed', 'telnet, 'mms', 'rtsp' and 'svn'. This
466 * covers all common link protocols, except for 'javascript' which should not
467 * be allowed for untrusted users.
471 * @param string $string Content to filter through kses
472 * @param array $allowed_html List of allowed HTML elements
473 * @param array $allowed_protocols Optional. Allowed protocol in links.
474 * @return string Filtered content with only allowed HTML elements
476 function wp_kses( $string, $allowed_html, $allowed_protocols = array() ) {
477 if ( empty( $allowed_protocols ) )
478 $allowed_protocols = wp_allowed_protocols();
479 $string = wp_kses_no_null($string);
480 $string = wp_kses_js_entities($string);
481 $string = wp_kses_normalize_entities($string);
482 $string = wp_kses_hook($string, $allowed_html, $allowed_protocols); // WP changed the order of these funcs and added args to wp_kses_hook
483 return wp_kses_split($string, $allowed_html, $allowed_protocols);
487 * Return a list of allowed tags and attributes for a given context.
491 * @param string $context The context for which to retrieve tags. Allowed values are
492 * post | strip | data | entities or the name of a field filter such as pre_user_description.
493 * @return array List of allowed tags and their allowed attributes.
495 function wp_kses_allowed_html( $context = '' ) {
496 global $allowedposttags, $allowedtags, $allowedentitynames;
498 if ( is_array( $context ) )
499 return apply_filters( 'wp_kses_allowed_html', $context, 'explicit' );
501 switch ( $context ) {
503 return apply_filters( 'wp_kses_allowed_html', $allowedposttags, $context );
505 case 'user_description':
506 case 'pre_user_description':
507 $tags = $allowedtags;
508 $tags['a']['rel'] = true;
509 return apply_filters( 'wp_kses_allowed_html', $tags, $context );
512 return apply_filters( 'wp_kses_allowed_html', array(), $context );
515 return apply_filters( 'wp_kses_allowed_html', $allowedentitynames, $context);
519 return apply_filters( 'wp_kses_allowed_html', $allowedtags, $context );
524 * You add any kses hooks here.
526 * There is currently only one kses WordPress hook and it is called here. All
527 * parameters are passed to the hooks and expected to receive a string.
531 * @param string $string Content to filter through kses
532 * @param array $allowed_html List of allowed HTML elements
533 * @param array $allowed_protocols Allowed protocol in links
534 * @return string Filtered content through 'pre_kses' hook
536 function wp_kses_hook( $string, $allowed_html, $allowed_protocols ) {
537 $string = apply_filters('pre_kses', $string, $allowed_html, $allowed_protocols);
542 * This function returns kses' version number.
546 * @return string KSES Version Number
548 function wp_kses_version() {
553 * Searches for HTML tags, no matter how malformed.
555 * It also matches stray ">" characters.
559 * @param string $string Content to filter
560 * @param array $allowed_html Allowed HTML elements
561 * @param array $allowed_protocols Allowed protocols to keep
562 * @return string Content with fixed HTML tags
564 function wp_kses_split( $string, $allowed_html, $allowed_protocols ) {
565 global $pass_allowed_html, $pass_allowed_protocols;
566 $pass_allowed_html = $allowed_html;
567 $pass_allowed_protocols = $allowed_protocols;
568 return preg_replace_callback( '%(<!--.*?(-->|$))|(<[^>]*(>|$)|>)%', '_wp_kses_split_callback', $string );
572 * Callback for wp_kses_split.
577 function _wp_kses_split_callback( $match ) {
578 global $pass_allowed_html, $pass_allowed_protocols;
579 return wp_kses_split2( $match[0], $pass_allowed_html, $pass_allowed_protocols );
583 * Callback for wp_kses_split for fixing malformed HTML tags.
585 * This function does a lot of work. It rejects some very malformed things like
586 * <:::>. It returns an empty string, if the element isn't allowed (look ma, no
587 * strip_tags()!). Otherwise it splits the tag into an element and an attribute
590 * After the tag is split into an element and an attribute list, it is run
591 * through another filter which will remove illegal attributes and once that is
592 * completed, will be returned.
596 * @uses wp_kses_attr()
598 * @param string $string Content to filter
599 * @param array $allowed_html Allowed HTML elements
600 * @param array $allowed_protocols Allowed protocols to keep
601 * @return string Fixed HTML element
603 function wp_kses_split2($string, $allowed_html, $allowed_protocols) {
604 $string = wp_kses_stripslashes($string);
606 if (substr($string, 0, 1) != '<')
608 # It matched a ">" character
610 if ( '<!--' == substr( $string, 0, 4 ) ) {
611 $string = str_replace( array('<!--', '-->'), '', $string );
612 while ( $string != ($newstring = wp_kses($string, $allowed_html, $allowed_protocols)) )
613 $string = $newstring;
616 // prevent multiple dashes in comments
617 $string = preg_replace('/--+/', '-', $string);
618 // prevent three dashes closing a comment
619 $string = preg_replace('/-$/', '', $string);
620 return "<!--{$string}-->";
622 # Allow HTML comments
624 if (!preg_match('%^<\s*(/\s*)?([a-zA-Z0-9]+)([^>]*)>?$%', $string, $matches))
626 # It's seriously malformed
628 $slash = trim($matches[1]);
630 $attrlist = $matches[3];
632 if ( ! is_array( $allowed_html ) )
633 $allowed_html = wp_kses_allowed_html( $allowed_html );
635 if ( ! isset($allowed_html[strtolower($elem)]) )
637 # They are using a not allowed HTML element
641 # No attributes are allowed for closing elements
643 return wp_kses_attr( $elem, $attrlist, $allowed_html, $allowed_protocols );
647 * Removes all attributes, if none are allowed for this element.
649 * If some are allowed it calls wp_kses_hair() to split them further, and then
650 * it builds up new HTML code from the data that kses_hair() returns. It also
651 * removes "<" and ">" characters, if there are any left. One more thing it does
652 * is to check if the tag has a closing XHTML slash, and if it does, it puts one
653 * in the returned code as well.
657 * @param string $element HTML element/tag
658 * @param string $attr HTML attributes from HTML element to closing HTML element tag
659 * @param array $allowed_html Allowed HTML elements
660 * @param array $allowed_protocols Allowed protocols to keep
661 * @return string Sanitized HTML element
663 function wp_kses_attr($element, $attr, $allowed_html, $allowed_protocols) {
664 # Is there a closing XHTML slash at the end of the attributes?
666 if ( ! is_array( $allowed_html ) )
667 $allowed_html = wp_kses_allowed_html( $allowed_html );
670 if (preg_match('%\s*/\s*$%', $attr))
673 # Are any attributes allowed at all for this element?
674 if ( ! isset($allowed_html[strtolower($element)]) || count($allowed_html[strtolower($element)]) == 0 )
675 return "<$element$xhtml_slash>";
678 $attrarr = wp_kses_hair($attr, $allowed_protocols);
680 # Go through $attrarr, and save the allowed attributes for this element
684 $allowed_attr = $allowed_html[strtolower($element)];
685 foreach ($attrarr as $arreach) {
686 if ( ! isset( $allowed_attr[strtolower($arreach['name'])] ) )
687 continue; # the attribute is not allowed
689 $current = $allowed_attr[strtolower($arreach['name'])];
690 if ( $current == '' )
691 continue; # the attribute is not allowed
693 if ( strtolower( $arreach['name'] ) == 'style' ) {
694 $orig_value = $arreach['value'];
695 $value = safecss_filter_attr( $orig_value );
697 if ( empty( $value ) )
700 $arreach['value'] = $value;
701 $arreach['whole'] = str_replace( $orig_value, $value, $arreach['whole'] );
704 if ( ! is_array($current) ) {
705 $attr2 .= ' '.$arreach['whole'];
706 # there are no checks
709 # there are some checks
711 foreach ($current as $currkey => $currval) {
712 if ( ! wp_kses_check_attr_val($arreach['value'], $arreach['vless'], $currkey, $currval) ) {
719 $attr2 .= ' '.$arreach['whole']; # it passed them
720 } # if !is_array($current)
723 # Remove any "<" or ">" characters
724 $attr2 = preg_replace('/[<>]/', '', $attr2);
726 return "<$element$attr2$xhtml_slash>";
730 * Builds an attribute list from string containing attributes.
732 * This function does a lot of work. It parses an attribute list into an array
733 * with attribute data, and tries to do the right thing even if it gets weird
734 * input. It will add quotes around attribute values that don't have any quotes
735 * or apostrophes around them, to make it easier to produce HTML code that will
736 * conform to W3C's HTML specification. It will also remove bad URL protocols
737 * from attribute values. It also reduces duplicate attributes by using the
738 * attribute defined first (foo='bar' foo='baz' will result in foo='bar').
742 * @param string $attr Attribute list from HTML element to closing HTML element tag
743 * @param array $allowed_protocols Allowed protocols to keep
744 * @return array List of attributes after parsing
746 function wp_kses_hair($attr, $allowed_protocols) {
750 $uris = array('xmlns', 'profile', 'href', 'src', 'cite', 'classid', 'codebase', 'data', 'usemap', 'longdesc', 'action');
752 # Loop through the whole attribute list
754 while (strlen($attr) != 0) {
755 $working = 0; # Was the last operation successful?
758 case 0 : # attribute name, href for instance
760 if (preg_match('/^([-a-zA-Z]+)/', $attr, $match)) {
761 $attrname = $match[1];
762 $working = $mode = 1;
763 $attr = preg_replace('/^[-a-zA-Z]+/', '', $attr);
768 case 1 : # equals sign or valueless ("selected")
770 if (preg_match('/^\s*=\s*/', $attr)) # equals sign
774 $attr = preg_replace('/^\s*=\s*/', '', $attr);
778 if (preg_match('/^\s+/', $attr)) # valueless
782 if(false === array_key_exists($attrname, $attrarr)) {
783 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
785 $attr = preg_replace('/^\s+/', '', $attr);
790 case 2 : # attribute value, a URL after href= for instance
792 if (preg_match('%^"([^"]*)"(\s+|/?$)%', $attr, $match))
795 $thisval = $match[1];
796 if ( in_array(strtolower($attrname), $uris) )
797 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
799 if(false === array_key_exists($attrname, $attrarr)) {
800 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n');
804 $attr = preg_replace('/^"[^"]*"(\s+|$)/', '', $attr);
808 if (preg_match("%^'([^']*)'(\s+|/?$)%", $attr, $match))
811 $thisval = $match[1];
812 if ( in_array(strtolower($attrname), $uris) )
813 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
815 if(false === array_key_exists($attrname, $attrarr)) {
816 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname='$thisval'", 'vless' => 'n');
820 $attr = preg_replace("/^'[^']*'(\s+|$)/", '', $attr);
824 if (preg_match("%^([^\s\"']+)(\s+|/?$)%", $attr, $match))
827 $thisval = $match[1];
828 if ( in_array(strtolower($attrname), $uris) )
829 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
831 if(false === array_key_exists($attrname, $attrarr)) {
832 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n');
834 # We add quotes to conform to W3C's HTML spec.
837 $attr = preg_replace("%^[^\s\"']+(\s+|$)%", '', $attr);
843 if ($working == 0) # not well formed, remove and try again
845 $attr = wp_kses_html_error($attr);
850 if ($mode == 1 && false === array_key_exists($attrname, $attrarr))
851 # special case, for when the attribute list ends with a valueless
852 # attribute like "selected"
853 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
859 * Performs different checks for attribute values.
861 * The currently implemented checks are "maxlen", "minlen", "maxval", "minval"
866 * @param string $value Attribute value
867 * @param string $vless Whether the value is valueless. Use 'y' or 'n'
868 * @param string $checkname What $checkvalue is checking for.
869 * @param mixed $checkvalue What constraint the value should pass
870 * @return bool Whether check passes
872 function wp_kses_check_attr_val($value, $vless, $checkname, $checkvalue) {
875 switch (strtolower($checkname)) {
877 # The maxlen check makes sure that the attribute value has a length not
878 # greater than the given value. This can be used to avoid Buffer Overflows
879 # in WWW clients and various Internet servers.
881 if (strlen($value) > $checkvalue)
886 # The minlen check makes sure that the attribute value has a length not
887 # smaller than the given value.
889 if (strlen($value) < $checkvalue)
894 # The maxval check does two things: it checks that the attribute value is
895 # an integer from 0 and up, without an excessive amount of zeroes or
896 # whitespace (to avoid Buffer Overflows). It also checks that the attribute
897 # value is not greater than the given value.
898 # This check can be used to avoid Denial of Service attacks.
900 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
902 if ($value > $checkvalue)
907 # The minval check makes sure that the attribute value is a positive integer,
908 # and that it is not smaller than the given value.
910 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
912 if ($value < $checkvalue)
917 # The valueless check makes sure if the attribute has a value
918 # (like <a href="blah">) or not (<option selected>). If the given value
919 # is a "y" or a "Y", the attribute must not have a value.
920 # If the given value is an "n" or an "N", the attribute must have one.
922 if (strtolower($checkvalue) != $vless)
931 * Sanitize string from bad protocols.
933 * This function removes all non-allowed protocols from the beginning of
934 * $string. It ignores whitespace and the case of the letters, and it does
935 * understand HTML entities. It does its work in a while loop, so it won't be
936 * fooled by a string like "javascript:javascript:alert(57)".
940 * @param string $string Content to filter bad protocols from
941 * @param array $allowed_protocols Allowed protocols to keep
942 * @return string Filtered content
944 function wp_kses_bad_protocol($string, $allowed_protocols) {
945 $string = wp_kses_no_null($string);
949 $original_string = $string;
950 $string = wp_kses_bad_protocol_once($string, $allowed_protocols);
951 } while ( $original_string != $string && ++$iterations < 6 );
953 if ( $original_string != $string )
960 * Removes any null characters in $string.
964 * @param string $string
967 function wp_kses_no_null($string) {
968 $string = preg_replace('/\0+/', '', $string);
969 $string = preg_replace('/(\\\\0)+/', '', $string);
975 * Strips slashes from in front of quotes.
977 * This function changes the character sequence \" to just ". It leaves all
978 * other slashes alone. It's really weird, but the quoting from
979 * preg_replace(//e) seems to require this.
983 * @param string $string String to strip slashes
984 * @return string Fixed string with quoted slashes
986 function wp_kses_stripslashes($string) {
987 return preg_replace('%\\\\"%', '"', $string);
991 * Goes through an array and changes the keys to all lower case.
995 * @param array $inarray Unfiltered array
996 * @return array Fixed array with all lowercase keys
998 function wp_kses_array_lc($inarray) {
999 $outarray = array ();
1001 foreach ( (array) $inarray as $inkey => $inval) {
1002 $outkey = strtolower($inkey);
1003 $outarray[$outkey] = array ();
1005 foreach ( (array) $inval as $inkey2 => $inval2) {
1006 $outkey2 = strtolower($inkey2);
1007 $outarray[$outkey][$outkey2] = $inval2;
1009 } # foreach $inarray
1015 * Removes the HTML JavaScript entities found in early versions of Netscape 4.
1019 * @param string $string
1022 function wp_kses_js_entities($string) {
1023 return preg_replace('%&\s*\{[^}]*(\}\s*;?|$)%', '', $string);
1027 * Handles parsing errors in wp_kses_hair().
1029 * The general plan is to remove everything to and including some whitespace,
1030 * but it deals with quotes and apostrophes as well.
1034 * @param string $string
1037 function wp_kses_html_error($string) {
1038 return preg_replace('/^("[^"]*("|$)|\'[^\']*(\'|$)|\S)*\s*/', '', $string);
1042 * Sanitizes content from bad protocols and other characters.
1044 * This function searches for URL protocols at the beginning of $string, while
1045 * handling whitespace and HTML entities.
1049 * @param string $string Content to check for bad protocols
1050 * @param string $allowed_protocols Allowed protocols
1051 * @return string Sanitized content
1053 function wp_kses_bad_protocol_once($string, $allowed_protocols, $count = 1 ) {
1054 $string2 = preg_split( '/:|�*58;|�*3a;/i', $string, 2 );
1055 if ( isset($string2[1]) && ! preg_match('%/\?%', $string2[0]) ) {
1056 $string = trim( $string2[1] );
1057 $protocol = wp_kses_bad_protocol_once2( $string2[0], $allowed_protocols );
1058 if ( 'feed:' == $protocol ) {
1061 $string = wp_kses_bad_protocol_once( $string, $allowed_protocols, ++$count );
1062 if ( empty( $string ) )
1065 $string = $protocol . $string;
1072 * Callback for wp_kses_bad_protocol_once() regular expression.
1074 * This function processes URL protocols, checks to see if they're in the
1075 * whitelist or not, and returns different data depending on the answer.
1080 * @param string $string URI scheme to check against the whitelist
1081 * @param string $allowed_protocols Allowed protocols
1082 * @return string Sanitized content
1084 function wp_kses_bad_protocol_once2( $string, $allowed_protocols ) {
1085 $string2 = wp_kses_decode_entities($string);
1086 $string2 = preg_replace('/\s/', '', $string2);
1087 $string2 = wp_kses_no_null($string2);
1088 $string2 = strtolower($string2);
1091 foreach ( (array) $allowed_protocols as $one_protocol )
1092 if ( strtolower($one_protocol) == $string2 ) {
1104 * Converts and fixes HTML entities.
1106 * This function normalizes HTML entities. It will convert "AT&T" to the correct
1107 * "AT&T", ":" to ":", "&#XYZZY;" to "&#XYZZY;" and so on.
1111 * @param string $string Content to normalize entities
1112 * @return string Content with normalized entities
1114 function wp_kses_normalize_entities($string) {
1115 # Disarm all entities by converting & to &
1117 $string = str_replace('&', '&', $string);
1119 # Change back the allowed entities in our entity whitelist
1121 $string = preg_replace_callback('/&([A-Za-z]{2,8});/', 'wp_kses_named_entities', $string);
1122 $string = preg_replace_callback('/&#(0*[0-9]{1,7});/', 'wp_kses_normalize_entities2', $string);
1123 $string = preg_replace_callback('/&#[Xx](0*[0-9A-Fa-f]{1,6});/', 'wp_kses_normalize_entities3', $string);
1129 * Callback for wp_kses_normalize_entities() regular expression.
1131 * This function only accepts valid named entity references, which are finite,
1132 * case-sensitive, and highly scrutinized by HTML and XML validators.
1136 * @param array $matches preg_replace_callback() matches array
1137 * @return string Correctly encoded entity
1139 function wp_kses_named_entities($matches) {
1140 global $allowedentitynames;
1142 if ( empty($matches[1]) )
1146 return ( ( ! in_array($i, $allowedentitynames) ) ? "&$i;" : "&$i;" );
1150 * Callback for wp_kses_normalize_entities() regular expression.
1152 * This function helps wp_kses_normalize_entities() to only accept 16-bit values
1153 * and nothing more for &#number; entities.
1158 * @param array $matches preg_replace_callback() matches array
1159 * @return string Correctly encoded entity
1161 function wp_kses_normalize_entities2($matches) {
1162 if ( empty($matches[1]) )
1166 if (valid_unicode($i)) {
1167 $i = str_pad(ltrim($i,'0'), 3, '0', STR_PAD_LEFT);
1177 * Callback for wp_kses_normalize_entities() for regular expression.
1179 * This function helps wp_kses_normalize_entities() to only accept valid Unicode
1180 * numeric entities in hex form.
1184 * @param array $matches preg_replace_callback() matches array
1185 * @return string Correctly encoded entity
1187 function wp_kses_normalize_entities3($matches) {
1188 if ( empty($matches[1]) )
1191 $hexchars = $matches[1];
1192 return ( ( ! valid_unicode(hexdec($hexchars)) ) ? "&#x$hexchars;" : '&#x'.ltrim($hexchars,'0').';' );
1196 * Helper function to determine if a Unicode value is valid.
1198 * @param int $i Unicode value
1199 * @return bool True if the value was a valid Unicode number
1201 function valid_unicode($i) {
1202 return ( $i == 0x9 || $i == 0xa || $i == 0xd ||
1203 ($i >= 0x20 && $i <= 0xd7ff) ||
1204 ($i >= 0xe000 && $i <= 0xfffd) ||
1205 ($i >= 0x10000 && $i <= 0x10ffff) );
1209 * Convert all entities to their character counterparts.
1211 * This function decodes numeric HTML entities (A and A). It doesn't do
1212 * anything with other entities like ä, but we don't need them in the URL
1213 * protocol whitelisting system anyway.
1217 * @param string $string Content to change entities
1218 * @return string Content after decoded entities
1220 function wp_kses_decode_entities($string) {
1221 $string = preg_replace_callback('/&#([0-9]+);/', '_wp_kses_decode_entities_chr', $string);
1222 $string = preg_replace_callback('/&#[Xx]([0-9A-Fa-f]+);/', '_wp_kses_decode_entities_chr_hexdec', $string);
1228 * Regex callback for wp_kses_decode_entities()
1230 * @param array $match preg match
1233 function _wp_kses_decode_entities_chr( $match ) {
1234 return chr( $match[1] );
1238 * Regex callback for wp_kses_decode_entities()
1240 * @param array $match preg match
1243 function _wp_kses_decode_entities_chr_hexdec( $match ) {
1244 return chr( hexdec( $match[1] ) );
1248 * Sanitize content with allowed HTML Kses rules.
1251 * @uses $allowedtags
1253 * @param string $data Content to filter, expected to be escaped with slashes
1254 * @return string Filtered content
1256 function wp_filter_kses( $data ) {
1257 return addslashes( wp_kses( stripslashes( $data ), current_filter() ) );
1261 * Sanitize content with allowed HTML Kses rules.
1264 * @uses $allowedtags
1266 * @param string $data Content to filter, expected to not be escaped
1267 * @return string Filtered content
1269 function wp_kses_data( $data ) {
1270 return wp_kses( $data , current_filter() );
1274 * Sanitize content for allowed HTML tags for post content.
1276 * Post content refers to the page contents of the 'post' type and not $_POST
1281 * @param string $data Post content to filter, expected to be escaped with slashes
1282 * @return string Filtered post content with allowed HTML tags and attributes intact.
1284 function wp_filter_post_kses($data) {
1285 return addslashes ( wp_kses( stripslashes( $data ), 'post' ) );
1289 * Sanitize content for allowed HTML tags for post content.
1291 * Post content refers to the page contents of the 'post' type and not $_POST
1296 * @param string $data Post content to filter
1297 * @return string Filtered post content with allowed HTML tags and attributes intact.
1299 function wp_kses_post($data) {
1300 return wp_kses( $data , 'post' );
1304 * Strips all of the HTML in the content.
1308 * @param string $data Content to strip all HTML from
1309 * @return string Filtered content without any HTML
1311 function wp_filter_nohtml_kses( $data ) {
1312 return addslashes ( wp_kses( stripslashes( $data ), 'strip' ) );
1316 * Adds all Kses input form content filters.
1318 * All hooks have default priority. The wp_filter_kses() function is added to
1319 * the 'pre_comment_content' and 'title_save_pre' hooks.
1321 * The wp_filter_post_kses() function is added to the 'content_save_pre',
1322 * 'excerpt_save_pre', and 'content_filtered_save_pre' hooks.
1325 * @uses add_filter() See description for what functions are added to what hooks.
1327 function kses_init_filters() {
1329 add_filter('title_save_pre', 'wp_filter_kses');
1331 // Comment filtering
1332 if ( current_user_can( 'unfiltered_html' ) )
1333 add_filter( 'pre_comment_content', 'wp_filter_post_kses' );
1335 add_filter( 'pre_comment_content', 'wp_filter_kses' );
1338 add_filter('content_save_pre', 'wp_filter_post_kses');
1339 add_filter('excerpt_save_pre', 'wp_filter_post_kses');
1340 add_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1344 * Removes all Kses input form content filters.
1346 * A quick procedural method to removing all of the filters that kses uses for
1347 * content in WordPress Loop.
1349 * Does not remove the kses_init() function from 'init' hook (priority is
1350 * default). Also does not remove kses_init() function from 'set_current_user'
1351 * hook (priority is also default).
1355 function kses_remove_filters() {
1357 remove_filter('title_save_pre', 'wp_filter_kses');
1359 // Comment filtering
1360 remove_filter( 'pre_comment_content', 'wp_filter_post_kses' );
1361 remove_filter( 'pre_comment_content', 'wp_filter_kses' );
1364 remove_filter('content_save_pre', 'wp_filter_post_kses');
1365 remove_filter('excerpt_save_pre', 'wp_filter_post_kses');
1366 remove_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1370 * Sets up most of the Kses filters for input form content.
1372 * If you remove the kses_init() function from 'init' hook and
1373 * 'set_current_user' (priority is default), then none of the Kses filter hooks
1376 * First removes all of the Kses filters in case the current user does not need
1377 * to have Kses filter the content. If the user does not have unfiltered_html
1378 * capability, then Kses filters are added.
1380 * @uses kses_remove_filters() Removes the Kses filters
1381 * @uses kses_init_filters() Adds the Kses filters back if the user
1382 * does not have unfiltered HTML capability.
1385 function kses_init() {
1386 kses_remove_filters();
1388 if (current_user_can('unfiltered_html') == false)
1389 kses_init_filters();
1392 add_action('init', 'kses_init');
1393 add_action('set_current_user', 'kses_init');
1400 function safecss_filter_attr( $css, $deprecated = '' ) {
1401 if ( !empty( $deprecated ) )
1402 _deprecated_argument( __FUNCTION__, '2.8.1' ); // Never implemented
1404 $css = wp_kses_no_null($css);
1405 $css = str_replace(array("\n","\r","\t"), '', $css);
1407 if ( preg_match( '%[\\(&=}]|/\*%', $css ) ) // remove any inline css containing \ ( & } = or comments
1410 $css_array = explode( ';', trim( $css ) );
1411 $allowed_attr = apply_filters( 'safe_style_css', array( 'text-align', 'margin', 'color', 'float',
1412 'border', 'background', 'background-color', 'border-bottom', 'border-bottom-color',
1413 'border-bottom-style', 'border-bottom-width', 'border-collapse', 'border-color', 'border-left',
1414 'border-left-color', 'border-left-style', 'border-left-width', 'border-right', 'border-right-color',
1415 'border-right-style', 'border-right-width', 'border-spacing', 'border-style', 'border-top',
1416 'border-top-color', 'border-top-style', 'border-top-width', 'border-width', 'caption-side',
1417 'clear', 'cursor', 'direction', 'font', 'font-family', 'font-size', 'font-style',
1418 'font-variant', 'font-weight', 'height', 'letter-spacing', 'line-height', 'margin-bottom',
1419 'margin-left', 'margin-right', 'margin-top', 'overflow', 'padding', 'padding-bottom',
1420 'padding-left', 'padding-right', 'padding-top', 'text-decoration', 'text-indent', 'vertical-align',
1423 if ( empty($allowed_attr) )
1427 foreach ( $css_array as $css_item ) {
1428 if ( $css_item == '' )
1430 $css_item = trim( $css_item );
1432 if ( strpos( $css_item, ':' ) === false ) {
1435 $parts = explode( ':', $css_item );
1436 if ( in_array( trim( $parts[0] ), $allowed_attr ) )
1450 * Helper function to add global attributes to a tag in the allowed html list.
1455 * @param array $value An array of attributes.
1456 * @return array The array of attributes with global attributes added.
1458 function _wp_add_global_attributes( $value ) {
1459 $global_attributes = array(
1466 if ( true === $value )
1469 if ( is_array( $value ) )
1470 return array_merge( $value, $global_attributes );