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(
93 'blockquote' => array(
143 'fieldset' => array(),
150 'figcaption' => array(
170 'accept-charset' => true,
294 'cellpadding' => true,
295 'cellspacing' => true,
380 * Kses allowed HTML elements.
382 * @global array $allowedtags
385 $allowedtags = array(
397 'blockquote' => array(
414 $allowedentitynames = array(
415 'nbsp', 'iexcl', 'cent', 'pound', 'curren', 'yen',
416 'brvbar', 'sect', 'uml', 'copy', 'ordf', 'laquo',
417 'not', 'shy', 'reg', 'macr', 'deg', 'plusmn',
418 'acute', 'micro', 'para', 'middot', 'cedil', 'ordm',
419 'raquo', 'iquest', 'Agrave', 'Aacute', 'Acirc', 'Atilde',
420 'Auml', 'Aring', 'AElig', 'Ccedil', 'Egrave', 'Eacute',
421 'Ecirc', 'Euml', 'Igrave', 'Iacute', 'Icirc', 'Iuml',
422 'ETH', 'Ntilde', 'Ograve', 'Oacute', 'Ocirc', 'Otilde',
423 'Ouml', 'times', 'Oslash', 'Ugrave', 'Uacute', 'Ucirc',
424 'Uuml', 'Yacute', 'THORN', 'szlig', 'agrave', 'aacute',
425 'acirc', 'atilde', 'auml', 'aring', 'aelig', 'ccedil',
426 'egrave', 'eacute', 'ecirc', 'euml', 'igrave', 'iacute',
427 'icirc', 'iuml', 'eth', 'ntilde', 'ograve', 'oacute',
428 'ocirc', 'otilde', 'ouml', 'divide', 'oslash', 'ugrave',
429 'uacute', 'ucirc', 'uuml', 'yacute', 'thorn', 'yuml',
430 'quot', 'amp', 'lt', 'gt', 'apos', 'OElig',
431 'oelig', 'Scaron', 'scaron', 'Yuml', 'circ', 'tilde',
432 'ensp', 'emsp', 'thinsp', 'zwnj', 'zwj', 'lrm',
433 'rlm', 'ndash', 'mdash', 'lsquo', 'rsquo', 'sbquo',
434 'ldquo', 'rdquo', 'bdquo', 'dagger', 'Dagger', 'permil',
435 'lsaquo', 'rsaquo', 'euro', 'fnof', 'Alpha', 'Beta',
436 'Gamma', 'Delta', 'Epsilon', 'Zeta', 'Eta', 'Theta',
437 'Iota', 'Kappa', 'Lambda', 'Mu', 'Nu', 'Xi',
438 'Omicron', 'Pi', 'Rho', 'Sigma', 'Tau', 'Upsilon',
439 'Phi', 'Chi', 'Psi', 'Omega', 'alpha', 'beta',
440 'gamma', 'delta', 'epsilon', 'zeta', 'eta', 'theta',
441 'iota', 'kappa', 'lambda', 'mu', 'nu', 'xi',
442 'omicron', 'pi', 'rho', 'sigmaf', 'sigma', 'tau',
443 'upsilon', 'phi', 'chi', 'psi', 'omega', 'thetasym',
444 'upsih', 'piv', 'bull', 'hellip', 'prime', 'Prime',
445 'oline', 'frasl', 'weierp', 'image', 'real', 'trade',
446 'alefsym', 'larr', 'uarr', 'rarr', 'darr', 'harr',
447 'crarr', 'lArr', 'uArr', 'rArr', 'dArr', 'hArr',
448 'forall', 'part', 'exist', 'empty', 'nabla', 'isin',
449 'notin', 'ni', 'prod', 'sum', 'minus', 'lowast',
450 'radic', 'prop', 'infin', 'ang', 'and', 'or',
451 'cap', 'cup', 'int', 'sim', 'cong', 'asymp',
452 'ne', 'equiv', 'le', 'ge', 'sub', 'sup',
453 'nsub', 'sube', 'supe', 'oplus', 'otimes', 'perp',
454 'sdot', 'lceil', 'rceil', 'lfloor', 'rfloor', 'lang',
455 'rang', 'loz', 'spades', 'clubs', 'hearts', 'diams',
456 'sup1', 'sup2', 'sup3', 'frac14', 'frac12', 'frac34',
460 $allowedposttags = array_map( '_wp_add_global_attributes', $allowedposttags );
462 $allowedtags = wp_kses_array_lc( $allowedtags );
463 $allowedposttags = wp_kses_array_lc( $allowedposttags );
467 * Filters content and keeps only allowable HTML elements.
469 * This function makes sure that only the allowed HTML element names, attribute
470 * names and attribute values plus only sane HTML entities will occur in
471 * $string. You have to remove any slashes from PHP's magic quotes before you
472 * call this function.
474 * The default allowed protocols are 'http', 'https', 'ftp', 'mailto', 'news',
475 * 'irc', 'gopher', 'nntp', 'feed', 'telnet, 'mms', 'rtsp' and 'svn'. This
476 * covers all common link protocols, except for 'javascript' which should not
477 * be allowed for untrusted users.
481 * @param string $string Content to filter through kses
482 * @param array $allowed_html List of allowed HTML elements
483 * @param array $allowed_protocols Optional. Allowed protocol in links.
484 * @return string Filtered content with only allowed HTML elements
486 function wp_kses( $string, $allowed_html, $allowed_protocols = array() ) {
487 if ( empty( $allowed_protocols ) )
488 $allowed_protocols = wp_allowed_protocols();
489 $string = wp_kses_no_null($string);
490 $string = wp_kses_js_entities($string);
491 $string = wp_kses_normalize_entities($string);
492 $string = wp_kses_hook($string, $allowed_html, $allowed_protocols); // WP changed the order of these funcs and added args to wp_kses_hook
493 return wp_kses_split($string, $allowed_html, $allowed_protocols);
497 * Return a list of allowed tags and attributes for a given context.
501 * @param string $context The context for which to retrieve tags. Allowed values are
502 * post | strip | data | entities or the name of a field filter such as pre_user_description.
503 * @return array List of allowed tags and their allowed attributes.
505 function wp_kses_allowed_html( $context = '' ) {
506 global $allowedposttags, $allowedtags, $allowedentitynames;
508 if ( is_array( $context ) ) {
510 * Filter HTML elements allowed for a given context.
514 * @param string $tags Allowed tags, attributes, and/or entities.
515 * @param string $context Context to judge allowed tags by. Allowed values are 'post',
516 * 'data', 'strip', 'entities', 'explicit', or the name of a filter.
518 return apply_filters( 'wp_kses_allowed_html', $context, 'explicit' );
521 switch ( $context ) {
523 /** This filter is documented in wp-includes/kses.php */
524 return apply_filters( 'wp_kses_allowed_html', $allowedposttags, $context );
526 case 'user_description':
527 case 'pre_user_description':
528 $tags = $allowedtags;
529 $tags['a']['rel'] = true;
530 /** This filter is documented in wp-includes/kses.php */
531 return apply_filters( 'wp_kses_allowed_html', $tags, $context );
534 /** This filter is documented in wp-includes/kses.php */
535 return apply_filters( 'wp_kses_allowed_html', array(), $context );
538 /** This filter is documented in wp-includes/kses.php */
539 return apply_filters( 'wp_kses_allowed_html', $allowedentitynames, $context);
543 /** This filter is documented in wp-includes/kses.php */
544 return apply_filters( 'wp_kses_allowed_html', $allowedtags, $context );
549 * You add any kses hooks here.
551 * There is currently only one kses WordPress hook and it is called here. All
552 * parameters are passed to the hooks and expected to receive a string.
556 * @param string $string Content to filter through kses
557 * @param array $allowed_html List of allowed HTML elements
558 * @param array $allowed_protocols Allowed protocol in links
559 * @return string Filtered content through 'pre_kses' hook
561 function wp_kses_hook( $string, $allowed_html, $allowed_protocols ) {
563 * Filter content to be run through kses.
567 * @param string $string Content to run through kses.
568 * @param array $allowed_html Allowed HTML elements.
569 * @param array $allowed_protocols Allowed protocol in links.
571 $string = apply_filters( 'pre_kses', $string, $allowed_html, $allowed_protocols );
576 * This function returns kses' version number.
580 * @return string KSES Version Number
582 function wp_kses_version() {
587 * Searches for HTML tags, no matter how malformed.
589 * It also matches stray ">" characters.
593 * @param string $string Content to filter
594 * @param array $allowed_html Allowed HTML elements
595 * @param array $allowed_protocols Allowed protocols to keep
596 * @return string Content with fixed HTML tags
598 function wp_kses_split( $string, $allowed_html, $allowed_protocols ) {
599 global $pass_allowed_html, $pass_allowed_protocols;
600 $pass_allowed_html = $allowed_html;
601 $pass_allowed_protocols = $allowed_protocols;
602 return preg_replace_callback( '%(<!--.*?(-->|$))|(<[^>]*(>|$)|>)%', '_wp_kses_split_callback', $string );
606 * Callback for wp_kses_split.
611 function _wp_kses_split_callback( $match ) {
612 global $pass_allowed_html, $pass_allowed_protocols;
613 return wp_kses_split2( $match[0], $pass_allowed_html, $pass_allowed_protocols );
617 * Callback for wp_kses_split for fixing malformed HTML tags.
619 * This function does a lot of work. It rejects some very malformed things like
620 * <:::>. It returns an empty string, if the element isn't allowed (look ma, no
621 * strip_tags()!). Otherwise it splits the tag into an element and an attribute
624 * After the tag is split into an element and an attribute list, it is run
625 * through another filter which will remove illegal attributes and once that is
626 * completed, will be returned.
630 * @uses wp_kses_attr()
632 * @param string $string Content to filter
633 * @param array $allowed_html Allowed HTML elements
634 * @param array $allowed_protocols Allowed protocols to keep
635 * @return string Fixed HTML element
637 function wp_kses_split2($string, $allowed_html, $allowed_protocols) {
638 $string = wp_kses_stripslashes($string);
640 if (substr($string, 0, 1) != '<')
642 # It matched a ">" character
644 if ( '<!--' == substr( $string, 0, 4 ) ) {
645 $string = str_replace( array('<!--', '-->'), '', $string );
646 while ( $string != ($newstring = wp_kses($string, $allowed_html, $allowed_protocols)) )
647 $string = $newstring;
650 // prevent multiple dashes in comments
651 $string = preg_replace('/--+/', '-', $string);
652 // prevent three dashes closing a comment
653 $string = preg_replace('/-$/', '', $string);
654 return "<!--{$string}-->";
656 # Allow HTML comments
658 if (!preg_match('%^<\s*(/\s*)?([a-zA-Z0-9]+)([^>]*)>?$%', $string, $matches))
660 # It's seriously malformed
662 $slash = trim($matches[1]);
664 $attrlist = $matches[3];
666 if ( ! is_array( $allowed_html ) )
667 $allowed_html = wp_kses_allowed_html( $allowed_html );
669 if ( ! isset($allowed_html[strtolower($elem)]) )
671 # They are using a not allowed HTML element
675 # No attributes are allowed for closing elements
677 return wp_kses_attr( $elem, $attrlist, $allowed_html, $allowed_protocols );
681 * Removes all attributes, if none are allowed for this element.
683 * If some are allowed it calls wp_kses_hair() to split them further, and then
684 * it builds up new HTML code from the data that kses_hair() returns. It also
685 * removes "<" and ">" characters, if there are any left. One more thing it does
686 * is to check if the tag has a closing XHTML slash, and if it does, it puts one
687 * in the returned code as well.
691 * @param string $element HTML element/tag
692 * @param string $attr HTML attributes from HTML element to closing HTML element tag
693 * @param array $allowed_html Allowed HTML elements
694 * @param array $allowed_protocols Allowed protocols to keep
695 * @return string Sanitized HTML element
697 function wp_kses_attr($element, $attr, $allowed_html, $allowed_protocols) {
698 # Is there a closing XHTML slash at the end of the attributes?
700 if ( ! is_array( $allowed_html ) )
701 $allowed_html = wp_kses_allowed_html( $allowed_html );
704 if (preg_match('%\s*/\s*$%', $attr))
707 # Are any attributes allowed at all for this element?
708 if ( ! isset($allowed_html[strtolower($element)]) || count($allowed_html[strtolower($element)]) == 0 )
709 return "<$element$xhtml_slash>";
712 $attrarr = wp_kses_hair($attr, $allowed_protocols);
714 # Go through $attrarr, and save the allowed attributes for this element
718 $allowed_attr = $allowed_html[strtolower($element)];
719 foreach ($attrarr as $arreach) {
720 if ( ! isset( $allowed_attr[strtolower($arreach['name'])] ) )
721 continue; # the attribute is not allowed
723 $current = $allowed_attr[strtolower($arreach['name'])];
724 if ( $current == '' )
725 continue; # the attribute is not allowed
727 if ( strtolower( $arreach['name'] ) == 'style' ) {
728 $orig_value = $arreach['value'];
729 $value = safecss_filter_attr( $orig_value );
731 if ( empty( $value ) )
734 $arreach['value'] = $value;
735 $arreach['whole'] = str_replace( $orig_value, $value, $arreach['whole'] );
738 if ( ! is_array($current) ) {
739 $attr2 .= ' '.$arreach['whole'];
740 # there are no checks
743 # there are some checks
745 foreach ($current as $currkey => $currval) {
746 if ( ! wp_kses_check_attr_val($arreach['value'], $arreach['vless'], $currkey, $currval) ) {
753 $attr2 .= ' '.$arreach['whole']; # it passed them
754 } # if !is_array($current)
757 # Remove any "<" or ">" characters
758 $attr2 = preg_replace('/[<>]/', '', $attr2);
760 return "<$element$attr2$xhtml_slash>";
764 * Builds an attribute list from string containing attributes.
766 * This function does a lot of work. It parses an attribute list into an array
767 * with attribute data, and tries to do the right thing even if it gets weird
768 * input. It will add quotes around attribute values that don't have any quotes
769 * or apostrophes around them, to make it easier to produce HTML code that will
770 * conform to W3C's HTML specification. It will also remove bad URL protocols
771 * from attribute values. It also reduces duplicate attributes by using the
772 * attribute defined first (foo='bar' foo='baz' will result in foo='bar').
776 * @param string $attr Attribute list from HTML element to closing HTML element tag
777 * @param array $allowed_protocols Allowed protocols to keep
778 * @return array List of attributes after parsing
780 function wp_kses_hair($attr, $allowed_protocols) {
784 $uris = array('xmlns', 'profile', 'href', 'src', 'cite', 'classid', 'codebase', 'data', 'usemap', 'longdesc', 'action');
786 # Loop through the whole attribute list
788 while (strlen($attr) != 0) {
789 $working = 0; # Was the last operation successful?
792 case 0 : # attribute name, href for instance
794 if ( preg_match('/^([-a-zA-Z:]+)/', $attr, $match ) ) {
795 $attrname = $match[1];
796 $working = $mode = 1;
797 $attr = preg_replace( '/^[-a-zA-Z:]+/', '', $attr );
802 case 1 : # equals sign or valueless ("selected")
804 if (preg_match('/^\s*=\s*/', $attr)) # equals sign
808 $attr = preg_replace('/^\s*=\s*/', '', $attr);
812 if (preg_match('/^\s+/', $attr)) # valueless
816 if(false === array_key_exists($attrname, $attrarr)) {
817 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
819 $attr = preg_replace('/^\s+/', '', $attr);
824 case 2 : # attribute value, a URL after href= for instance
826 if (preg_match('%^"([^"]*)"(\s+|/?$)%', $attr, $match))
829 $thisval = $match[1];
830 if ( in_array(strtolower($attrname), $uris) )
831 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
833 if(false === array_key_exists($attrname, $attrarr)) {
834 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n');
838 $attr = preg_replace('/^"[^"]*"(\s+|$)/', '', $attr);
842 if (preg_match("%^'([^']*)'(\s+|/?$)%", $attr, $match))
845 $thisval = $match[1];
846 if ( in_array(strtolower($attrname), $uris) )
847 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
849 if(false === array_key_exists($attrname, $attrarr)) {
850 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname='$thisval'", 'vless' => 'n');
854 $attr = preg_replace("/^'[^']*'(\s+|$)/", '', $attr);
858 if (preg_match("%^([^\s\"']+)(\s+|/?$)%", $attr, $match))
861 $thisval = $match[1];
862 if ( in_array(strtolower($attrname), $uris) )
863 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
865 if(false === array_key_exists($attrname, $attrarr)) {
866 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n');
868 # We add quotes to conform to W3C's HTML spec.
871 $attr = preg_replace("%^[^\s\"']+(\s+|$)%", '', $attr);
877 if ($working == 0) # not well formed, remove and try again
879 $attr = wp_kses_html_error($attr);
884 if ($mode == 1 && false === array_key_exists($attrname, $attrarr))
885 # special case, for when the attribute list ends with a valueless
886 # attribute like "selected"
887 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
893 * Performs different checks for attribute values.
895 * The currently implemented checks are "maxlen", "minlen", "maxval", "minval"
900 * @param string $value Attribute value
901 * @param string $vless Whether the value is valueless. Use 'y' or 'n'
902 * @param string $checkname What $checkvalue is checking for.
903 * @param mixed $checkvalue What constraint the value should pass
904 * @return bool Whether check passes
906 function wp_kses_check_attr_val($value, $vless, $checkname, $checkvalue) {
909 switch (strtolower($checkname)) {
911 # The maxlen check makes sure that the attribute value has a length not
912 # greater than the given value. This can be used to avoid Buffer Overflows
913 # in WWW clients and various Internet servers.
915 if (strlen($value) > $checkvalue)
920 # The minlen check makes sure that the attribute value has a length not
921 # smaller than the given value.
923 if (strlen($value) < $checkvalue)
928 # The maxval check does two things: it checks that the attribute value is
929 # an integer from 0 and up, without an excessive amount of zeroes or
930 # whitespace (to avoid Buffer Overflows). It also checks that the attribute
931 # value is not greater than the given value.
932 # This check can be used to avoid Denial of Service attacks.
934 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
936 if ($value > $checkvalue)
941 # The minval check makes sure that the attribute value is a positive integer,
942 # and that it is not smaller than the given value.
944 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
946 if ($value < $checkvalue)
951 # The valueless check makes sure if the attribute has a value
952 # (like <a href="blah">) or not (<option selected>). If the given value
953 # is a "y" or a "Y", the attribute must not have a value.
954 # If the given value is an "n" or an "N", the attribute must have one.
956 if (strtolower($checkvalue) != $vless)
965 * Sanitize string from bad protocols.
967 * This function removes all non-allowed protocols from the beginning of
968 * $string. It ignores whitespace and the case of the letters, and it does
969 * understand HTML entities. It does its work in a while loop, so it won't be
970 * fooled by a string like "javascript:javascript:alert(57)".
974 * @param string $string Content to filter bad protocols from
975 * @param array $allowed_protocols Allowed protocols to keep
976 * @return string Filtered content
978 function wp_kses_bad_protocol($string, $allowed_protocols) {
979 $string = wp_kses_no_null($string);
983 $original_string = $string;
984 $string = wp_kses_bad_protocol_once($string, $allowed_protocols);
985 } while ( $original_string != $string && ++$iterations < 6 );
987 if ( $original_string != $string )
994 * Removes any invalid control characters in $string.
996 * Also removes any instance of the '\0' string.
1000 * @param string $string
1003 function wp_kses_no_null($string) {
1004 $string = preg_replace('/[\x00-\x08\x0B\x0C\x0E-\x1F]/', '', $string);
1005 $string = preg_replace('/(\\\\0)+/', '', $string);
1011 * Strips slashes from in front of quotes.
1013 * This function changes the character sequence \" to just ". It leaves all
1014 * other slashes alone. It's really weird, but the quoting from
1015 * preg_replace(//e) seems to require this.
1019 * @param string $string String to strip slashes
1020 * @return string Fixed string with quoted slashes
1022 function wp_kses_stripslashes($string) {
1023 return preg_replace('%\\\\"%', '"', $string);
1027 * Goes through an array and changes the keys to all lower case.
1031 * @param array $inarray Unfiltered array
1032 * @return array Fixed array with all lowercase keys
1034 function wp_kses_array_lc($inarray) {
1035 $outarray = array ();
1037 foreach ( (array) $inarray as $inkey => $inval) {
1038 $outkey = strtolower($inkey);
1039 $outarray[$outkey] = array ();
1041 foreach ( (array) $inval as $inkey2 => $inval2) {
1042 $outkey2 = strtolower($inkey2);
1043 $outarray[$outkey][$outkey2] = $inval2;
1045 } # foreach $inarray
1051 * Removes the HTML JavaScript entities found in early versions of Netscape 4.
1055 * @param string $string
1058 function wp_kses_js_entities($string) {
1059 return preg_replace('%&\s*\{[^}]*(\}\s*;?|$)%', '', $string);
1063 * Handles parsing errors in wp_kses_hair().
1065 * The general plan is to remove everything to and including some whitespace,
1066 * but it deals with quotes and apostrophes as well.
1070 * @param string $string
1073 function wp_kses_html_error($string) {
1074 return preg_replace('/^("[^"]*("|$)|\'[^\']*(\'|$)|\S)*\s*/', '', $string);
1078 * Sanitizes content from bad protocols and other characters.
1080 * This function searches for URL protocols at the beginning of $string, while
1081 * handling whitespace and HTML entities.
1085 * @param string $string Content to check for bad protocols
1086 * @param string $allowed_protocols Allowed protocols
1087 * @return string Sanitized content
1089 function wp_kses_bad_protocol_once($string, $allowed_protocols, $count = 1 ) {
1090 $string2 = preg_split( '/:|�*58;|�*3a;/i', $string, 2 );
1091 if ( isset($string2[1]) && ! preg_match('%/\?%', $string2[0]) ) {
1092 $string = trim( $string2[1] );
1093 $protocol = wp_kses_bad_protocol_once2( $string2[0], $allowed_protocols );
1094 if ( 'feed:' == $protocol ) {
1097 $string = wp_kses_bad_protocol_once( $string, $allowed_protocols, ++$count );
1098 if ( empty( $string ) )
1101 $string = $protocol . $string;
1108 * Callback for wp_kses_bad_protocol_once() regular expression.
1110 * This function processes URL protocols, checks to see if they're in the
1111 * whitelist or not, and returns different data depending on the answer.
1116 * @param string $string URI scheme to check against the whitelist
1117 * @param string $allowed_protocols Allowed protocols
1118 * @return string Sanitized content
1120 function wp_kses_bad_protocol_once2( $string, $allowed_protocols ) {
1121 $string2 = wp_kses_decode_entities($string);
1122 $string2 = preg_replace('/\s/', '', $string2);
1123 $string2 = wp_kses_no_null($string2);
1124 $string2 = strtolower($string2);
1127 foreach ( (array) $allowed_protocols as $one_protocol )
1128 if ( strtolower($one_protocol) == $string2 ) {
1140 * Converts and fixes HTML entities.
1142 * This function normalizes HTML entities. It will convert "AT&T" to the correct
1143 * "AT&T", ":" to ":", "&#XYZZY;" to "&#XYZZY;" and so on.
1147 * @param string $string Content to normalize entities
1148 * @return string Content with normalized entities
1150 function wp_kses_normalize_entities($string) {
1151 # Disarm all entities by converting & to &
1153 $string = str_replace('&', '&', $string);
1155 # Change back the allowed entities in our entity whitelist
1157 $string = preg_replace_callback('/&([A-Za-z]{2,8}[0-9]{0,2});/', 'wp_kses_named_entities', $string);
1158 $string = preg_replace_callback('/&#(0*[0-9]{1,7});/', 'wp_kses_normalize_entities2', $string);
1159 $string = preg_replace_callback('/&#[Xx](0*[0-9A-Fa-f]{1,6});/', 'wp_kses_normalize_entities3', $string);
1165 * Callback for wp_kses_normalize_entities() regular expression.
1167 * This function only accepts valid named entity references, which are finite,
1168 * case-sensitive, and highly scrutinized by HTML and XML validators.
1172 * @param array $matches preg_replace_callback() matches array
1173 * @return string Correctly encoded entity
1175 function wp_kses_named_entities($matches) {
1176 global $allowedentitynames;
1178 if ( empty($matches[1]) )
1182 return ( ( ! in_array($i, $allowedentitynames) ) ? "&$i;" : "&$i;" );
1186 * Callback for wp_kses_normalize_entities() regular expression.
1188 * This function helps wp_kses_normalize_entities() to only accept 16-bit values
1189 * and nothing more for &#number; entities.
1194 * @param array $matches preg_replace_callback() matches array
1195 * @return string Correctly encoded entity
1197 function wp_kses_normalize_entities2($matches) {
1198 if ( empty($matches[1]) )
1202 if (valid_unicode($i)) {
1203 $i = str_pad(ltrim($i,'0'), 3, '0', STR_PAD_LEFT);
1213 * Callback for wp_kses_normalize_entities() for regular expression.
1215 * This function helps wp_kses_normalize_entities() to only accept valid Unicode
1216 * numeric entities in hex form.
1220 * @param array $matches preg_replace_callback() matches array
1221 * @return string Correctly encoded entity
1223 function wp_kses_normalize_entities3($matches) {
1224 if ( empty($matches[1]) )
1227 $hexchars = $matches[1];
1228 return ( ( ! valid_unicode(hexdec($hexchars)) ) ? "&#x$hexchars;" : '&#x'.ltrim($hexchars,'0').';' );
1232 * Helper function to determine if a Unicode value is valid.
1234 * @param int $i Unicode value
1235 * @return bool True if the value was a valid Unicode number
1237 function valid_unicode($i) {
1238 return ( $i == 0x9 || $i == 0xa || $i == 0xd ||
1239 ($i >= 0x20 && $i <= 0xd7ff) ||
1240 ($i >= 0xe000 && $i <= 0xfffd) ||
1241 ($i >= 0x10000 && $i <= 0x10ffff) );
1245 * Convert all entities to their character counterparts.
1247 * This function decodes numeric HTML entities (A and A). It doesn't do
1248 * anything with other entities like ä, but we don't need them in the URL
1249 * protocol whitelisting system anyway.
1253 * @param string $string Content to change entities
1254 * @return string Content after decoded entities
1256 function wp_kses_decode_entities($string) {
1257 $string = preg_replace_callback('/&#([0-9]+);/', '_wp_kses_decode_entities_chr', $string);
1258 $string = preg_replace_callback('/&#[Xx]([0-9A-Fa-f]+);/', '_wp_kses_decode_entities_chr_hexdec', $string);
1264 * Regex callback for wp_kses_decode_entities()
1266 * @param array $match preg match
1269 function _wp_kses_decode_entities_chr( $match ) {
1270 return chr( $match[1] );
1274 * Regex callback for wp_kses_decode_entities()
1276 * @param array $match preg match
1279 function _wp_kses_decode_entities_chr_hexdec( $match ) {
1280 return chr( hexdec( $match[1] ) );
1284 * Sanitize content with allowed HTML Kses rules.
1287 * @uses $allowedtags
1289 * @param string $data Content to filter, expected to be escaped with slashes
1290 * @return string Filtered content
1292 function wp_filter_kses( $data ) {
1293 return addslashes( wp_kses( stripslashes( $data ), current_filter() ) );
1297 * Sanitize content with allowed HTML Kses rules.
1300 * @uses $allowedtags
1302 * @param string $data Content to filter, expected to not be escaped
1303 * @return string Filtered content
1305 function wp_kses_data( $data ) {
1306 return wp_kses( $data , current_filter() );
1310 * Sanitize content for allowed HTML tags for post content.
1312 * Post content refers to the page contents of the 'post' type and not $_POST
1317 * @param string $data Post content to filter, expected to be escaped with slashes
1318 * @return string Filtered post content with allowed HTML tags and attributes intact.
1320 function wp_filter_post_kses($data) {
1321 return addslashes ( wp_kses( stripslashes( $data ), 'post' ) );
1325 * Sanitize content for allowed HTML tags for post content.
1327 * Post content refers to the page contents of the 'post' type and not $_POST
1332 * @param string $data Post content to filter
1333 * @return string Filtered post content with allowed HTML tags and attributes intact.
1335 function wp_kses_post($data) {
1336 return wp_kses( $data , 'post' );
1340 * Strips all of the HTML in the content.
1344 * @param string $data Content to strip all HTML from
1345 * @return string Filtered content without any HTML
1347 function wp_filter_nohtml_kses( $data ) {
1348 return addslashes ( wp_kses( stripslashes( $data ), 'strip' ) );
1352 * Adds all Kses input form content filters.
1354 * All hooks have default priority. The wp_filter_kses() function is added to
1355 * the 'pre_comment_content' and 'title_save_pre' hooks.
1357 * The wp_filter_post_kses() function is added to the 'content_save_pre',
1358 * 'excerpt_save_pre', and 'content_filtered_save_pre' hooks.
1361 * @uses add_filter() See description for what functions are added to what hooks.
1363 function kses_init_filters() {
1365 add_filter('title_save_pre', 'wp_filter_kses');
1367 // Comment filtering
1368 if ( current_user_can( 'unfiltered_html' ) )
1369 add_filter( 'pre_comment_content', 'wp_filter_post_kses' );
1371 add_filter( 'pre_comment_content', 'wp_filter_kses' );
1374 add_filter('content_save_pre', 'wp_filter_post_kses');
1375 add_filter('excerpt_save_pre', 'wp_filter_post_kses');
1376 add_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1380 * Removes all Kses input form content filters.
1382 * A quick procedural method to removing all of the filters that kses uses for
1383 * content in WordPress Loop.
1385 * Does not remove the kses_init() function from 'init' hook (priority is
1386 * default). Also does not remove kses_init() function from 'set_current_user'
1387 * hook (priority is also default).
1391 function kses_remove_filters() {
1393 remove_filter('title_save_pre', 'wp_filter_kses');
1395 // Comment filtering
1396 remove_filter( 'pre_comment_content', 'wp_filter_post_kses' );
1397 remove_filter( 'pre_comment_content', 'wp_filter_kses' );
1400 remove_filter('content_save_pre', 'wp_filter_post_kses');
1401 remove_filter('excerpt_save_pre', 'wp_filter_post_kses');
1402 remove_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1406 * Sets up most of the Kses filters for input form content.
1408 * If you remove the kses_init() function from 'init' hook and
1409 * 'set_current_user' (priority is default), then none of the Kses filter hooks
1412 * First removes all of the Kses filters in case the current user does not need
1413 * to have Kses filter the content. If the user does not have unfiltered_html
1414 * capability, then Kses filters are added.
1416 * @uses kses_remove_filters() Removes the Kses filters
1417 * @uses kses_init_filters() Adds the Kses filters back if the user
1418 * does not have unfiltered HTML capability.
1421 function kses_init() {
1422 kses_remove_filters();
1424 if (current_user_can('unfiltered_html') == false)
1425 kses_init_filters();
1428 add_action('init', 'kses_init');
1429 add_action('set_current_user', 'kses_init');
1436 function safecss_filter_attr( $css, $deprecated = '' ) {
1437 if ( !empty( $deprecated ) )
1438 _deprecated_argument( __FUNCTION__, '2.8.1' ); // Never implemented
1440 $css = wp_kses_no_null($css);
1441 $css = str_replace(array("\n","\r","\t"), '', $css);
1443 if ( preg_match( '%[\\(&=}]|/\*%', $css ) ) // remove any inline css containing \ ( & } = or comments
1446 $css_array = explode( ';', trim( $css ) );
1449 * Filter list of allowed CSS attributes.
1453 * @param array $attr List of allowed CSS attributes.
1455 $allowed_attr = apply_filters( 'safe_style_css', array( 'text-align', 'margin', 'color', 'float',
1456 'border', 'background', 'background-color', 'border-bottom', 'border-bottom-color',
1457 'border-bottom-style', 'border-bottom-width', 'border-collapse', 'border-color', 'border-left',
1458 'border-left-color', 'border-left-style', 'border-left-width', 'border-right', 'border-right-color',
1459 'border-right-style', 'border-right-width', 'border-spacing', 'border-style', 'border-top',
1460 'border-top-color', 'border-top-style', 'border-top-width', 'border-width', 'caption-side',
1461 'clear', 'cursor', 'direction', 'font', 'font-family', 'font-size', 'font-style',
1462 'font-variant', 'font-weight', 'height', 'letter-spacing', 'line-height', 'margin-bottom',
1463 'margin-left', 'margin-right', 'margin-top', 'overflow', 'padding', 'padding-bottom',
1464 'padding-left', 'padding-right', 'padding-top', 'text-decoration', 'text-indent', 'vertical-align',
1467 if ( empty($allowed_attr) )
1471 foreach ( $css_array as $css_item ) {
1472 if ( $css_item == '' )
1474 $css_item = trim( $css_item );
1476 if ( strpos( $css_item, ':' ) === false ) {
1479 $parts = explode( ':', $css_item );
1480 if ( in_array( trim( $parts[0] ), $allowed_attr ) )
1494 * Helper function to add global attributes to a tag in the allowed html list.
1499 * @param array $value An array of attributes.
1500 * @return array The array of attributes with global attributes added.
1502 function _wp_add_global_attributes( $value ) {
1503 $global_attributes = array(
1511 if ( true === $value )
1514 if ( is_array( $value ) )
1515 return array_merge( $value, $global_attributes );