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 * 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA or visit
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.
39 if ( ! defined( 'CUSTOM_TAGS' ) )
40 define( 'CUSTOM_TAGS', false );
42 if ( ! CUSTOM_TAGS ) {
44 * Kses global for default allowable HTML tags.
46 * Can be override by using CUSTOM_TAGS constant.
48 * @global array $allowedposttags
51 $allowedposttags = array(
73 'xml:lang' => array(),
81 'xml:lang' => array(),
85 'blockquote' => array(
90 'xml:lang' => array()),
94 'disabled' => array (),
100 'class' => array ()),
105 'title' => array ()),
111 'charoff' => array (),
115 'valign' => array (),
116 'width' => array ()),
118 'datetime' => array ()),
127 'xml:lang' => array(),
135 'xml:lang' => array()),
139 'fieldset' => array(),
146 'xml:lang' => array(),
148 'figcaption' => array(
154 'xml:lang' => array(),
166 'xml:lang' => array(),
169 'action' => array (),
170 'accept' => array (),
171 'accept-charset' => array (),
172 'enctype' => array (),
173 'method' => array (),
175 'target' => array ()),
180 'style' => array ()),
185 'style' => array ()),
190 'style' => array ()),
195 'style' => array ()),
200 'style' => array ()),
205 'style' => array ()),
212 'xml:lang' => array(),
220 'xml:lang' => array(),
225 'noshade' => array (),
227 'width' => array ()),
232 'border' => array (),
234 'height' => array (),
235 'hspace' => array (),
236 'longdesc' => array (),
237 'vspace' => array (),
240 'width' => array ()),
242 'datetime' => array (),
248 'align' => array ()),
251 'class' => array ()),
262 'xml:lang' => array(),
270 'xml:lang' => array()),
273 'width' => array ()),
284 'xml:lang' => array()),
291 'xml:lang' => array(),
302 'xml:lang' => array(),
307 'bgcolor' => array (),
308 'border' => array (),
309 'cellpadding' => array (),
310 'cellspacing' => array (),
316 'summary' => array (),
317 'width' => array ()),
321 'charoff' => array (),
322 'valign' => array ()),
327 'bgcolor' => array (),
329 'charoff' => array (),
331 'colspan' => array (),
333 'headers' => array (),
334 'height' => array (),
335 'nowrap' => array (),
336 'rowspan' => array (),
339 'valign' => array (),
340 'width' => array ()),
344 'disabled' => array (),
346 'readonly' => array ()),
351 'charoff' => array (),
352 'valign' => array ()),
357 'bgcolor' => array (),
359 'charoff' => array (),
361 'colspan' => array (),
362 'headers' => array (),
363 'height' => array (),
364 'nowrap' => array (),
365 'rowspan' => array (),
367 'valign' => array (),
368 'width' => array ()),
372 'charoff' => array (),
374 'valign' => array ()),
378 'bgcolor' => array (),
380 'charoff' => array (),
383 'valign' => array ()),
398 * Kses allowed HTML elements.
400 * @global array $allowedtags
403 $allowedtags = array(
406 'title' => array ()),
408 'title' => array ()),
410 'title' => array ()),
412 'blockquote' => array(
418 'datetime' => array ()),
422 'em' => array (), 'i' => array (),
423 // 'ins' => array('datetime' => array(), 'cite' => array()),
437 $allowedentitynames = array(
438 'nbsp', 'iexcl', 'cent', 'pound', 'curren', 'yen',
439 'brvbar', 'sect', 'uml', 'copy', 'ordf', 'laquo',
440 'not', 'shy', 'reg', 'macr', 'deg', 'plusmn',
441 'acute', 'micro', 'para', 'middot', 'cedil', 'ordm',
442 'raquo', 'iquest', 'Agrave', 'Aacute', 'Acirc', 'Atilde',
443 'Auml', 'Aring', 'AElig', 'Ccedil', 'Egrave', 'Eacute',
444 'Ecirc', 'Euml', 'Igrave', 'Iacute', 'Icirc', 'Iuml',
445 'ETH', 'Ntilde', 'Ograve', 'Oacute', 'Ocirc', 'Otilde',
446 'Ouml', 'times', 'Oslash', 'Ugrave', 'Uacute', 'Ucirc',
447 'Uuml', 'Yacute', 'THORN', 'szlig', 'agrave', 'aacute',
448 'acirc', 'atilde', 'auml', 'aring', 'aelig', 'ccedil',
449 'egrave', 'eacute', 'ecirc', 'euml', 'igrave', 'iacute',
450 'icirc', 'iuml', 'eth', 'ntilde', 'ograve', 'oacute',
451 'ocirc', 'otilde', 'ouml', 'divide', 'oslash', 'ugrave',
452 'uacute', 'ucirc', 'uuml', 'yacute', 'thorn', 'yuml',
453 'quot', 'amp', 'lt', 'gt', 'apos', 'OElig',
454 'oelig', 'Scaron', 'scaron', 'Yuml', 'circ', 'tilde',
455 'ensp', 'emsp', 'thinsp', 'zwnj', 'zwj', 'lrm',
456 'rlm', 'ndash', 'mdash', 'lsquo', 'rsquo', 'sbquo',
457 'ldquo', 'rdquo', 'bdquo', 'dagger', 'Dagger', 'permil',
458 'lsaquo', 'rsaquo', 'euro', 'fnof', 'Alpha', 'Beta',
459 'Gamma', 'Delta', 'Epsilon', 'Zeta', 'Eta', 'Theta',
460 'Iota', 'Kappa', 'Lambda', 'Mu', 'Nu', 'Xi',
461 'Omicron', 'Pi', 'Rho', 'Sigma', 'Tau', 'Upsilon',
462 'Phi', 'Chi', 'Psi', 'Omega', 'alpha', 'beta',
463 'gamma', 'delta', 'epsilon', 'zeta', 'eta', 'theta',
464 'iota', 'kappa', 'lambda', 'mu', 'nu', 'xi',
465 'omicron', 'pi', 'rho', 'sigmaf', 'sigma', 'tau',
466 'upsilon', 'phi', 'chi', 'psi', 'omega', 'thetasym',
467 'upsih', 'piv', 'bull', 'hellip', 'prime', 'Prime',
468 'oline', 'frasl', 'weierp', 'image', 'real', 'trade',
469 'alefsym', 'larr', 'uarr', 'rarr', 'darr', 'harr',
470 'crarr', 'lArr', 'uArr', 'rArr', 'dArr', 'hArr',
471 'forall', 'part', 'exist', 'empty', 'nabla', 'isin',
472 'notin', 'ni', 'prod', 'sum', 'minus', 'lowast',
473 'radic', 'prop', 'infin', 'ang', 'and', 'or',
474 'cap', 'cup', 'int', 'sim', 'cong', 'asymp',
475 'ne', 'equiv', 'le', 'ge', 'sub', 'sup',
476 'nsub', 'sube', 'supe', 'oplus', 'otimes', 'perp',
477 'sdot', 'lceil', 'rceil', 'lfloor', 'rfloor', 'lang',
478 'rang', 'loz', 'spades', 'clubs', 'hearts', 'diams',
483 * Filters content and keeps only allowable HTML elements.
485 * This function makes sure that only the allowed HTML element names, attribute
486 * names and attribute values plus only sane HTML entities will occur in
487 * $string. You have to remove any slashes from PHP's magic quotes before you
488 * call this function.
490 * The default allowed protocols are 'http', 'https', 'ftp', 'mailto', 'news',
491 * 'irc', 'gopher', 'nntp', 'feed', 'telnet, 'mms', 'rtsp' and 'svn'. This
492 * covers all common link protocols, except for 'javascript' which should not
493 * be allowed for untrusted users.
497 * @param string $string Content to filter through kses
498 * @param array $allowed_html List of allowed HTML elements
499 * @param array $allowed_protocols Optional. Allowed protocol in links.
500 * @return string Filtered content with only allowed HTML elements
502 function wp_kses($string, $allowed_html, $allowed_protocols = array ()) {
503 $allowed_protocols = wp_parse_args( $allowed_protocols, apply_filters('kses_allowed_protocols', array ('http', 'https', 'ftp', 'ftps', 'mailto', 'news', 'irc', 'gopher', 'nntp', 'feed', 'telnet', 'mms', 'rtsp', 'svn') ));
504 $string = wp_kses_no_null($string);
505 $string = wp_kses_js_entities($string);
506 $string = wp_kses_normalize_entities($string);
507 $allowed_html_fixed = wp_kses_array_lc($allowed_html);
508 $string = wp_kses_hook($string, $allowed_html_fixed, $allowed_protocols); // WP changed the order of these funcs and added args to wp_kses_hook
509 return wp_kses_split($string, $allowed_html_fixed, $allowed_protocols);
513 * You add any kses hooks here.
515 * There is currently only one kses WordPress hook and it is called here. All
516 * parameters are passed to the hooks and expected to recieve a string.
520 * @param string $string Content to filter through kses
521 * @param array $allowed_html List of allowed HTML elements
522 * @param array $allowed_protocols Allowed protocol in links
523 * @return string Filtered content through 'pre_kses' hook
525 function wp_kses_hook($string, $allowed_html, $allowed_protocols) {
526 $string = apply_filters('pre_kses', $string, $allowed_html, $allowed_protocols);
531 * This function returns kses' version number.
535 * @return string KSES Version Number
537 function wp_kses_version() {
542 * Searches for HTML tags, no matter how malformed.
544 * It also matches stray ">" characters.
548 * @param string $string Content to filter
549 * @param array $allowed_html Allowed HTML elements
550 * @param array $allowed_protocols Allowed protocols to keep
551 * @return string Content with fixed HTML tags
553 function wp_kses_split($string, $allowed_html, $allowed_protocols) {
554 global $pass_allowed_html, $pass_allowed_protocols;
555 $pass_allowed_html = $allowed_html;
556 $pass_allowed_protocols = $allowed_protocols;
557 return preg_replace_callback( '%((<!--.*?(-->|$))|(<[^>]*(>|$)|>))%', '_wp_kses_split_callback', $string );
561 * Callback for wp_kses_split.
566 function _wp_kses_split_callback( $match ) {
567 global $pass_allowed_html, $pass_allowed_protocols;
568 return wp_kses_split2( $match[1], $pass_allowed_html, $pass_allowed_protocols );
572 * Callback for wp_kses_split for fixing malformed HTML tags.
574 * This function does a lot of work. It rejects some very malformed things like
575 * <:::>. It returns an empty string, if the element isn't allowed (look ma, no
576 * strip_tags()!). Otherwise it splits the tag into an element and an attribute
579 * After the tag is split into an element and an attribute list, it is run
580 * through another filter which will remove illegal attributes and once that is
581 * completed, will be returned.
585 * @uses wp_kses_attr()
587 * @param string $string Content to filter
588 * @param array $allowed_html Allowed HTML elements
589 * @param array $allowed_protocols Allowed protocols to keep
590 * @return string Fixed HTML element
592 function wp_kses_split2($string, $allowed_html, $allowed_protocols) {
593 $string = wp_kses_stripslashes($string);
595 if (substr($string, 0, 1) != '<')
597 # It matched a ">" character
599 if (preg_match('%^<!--(.*?)(-->)?$%', $string, $matches)) {
600 $string = str_replace(array('<!--', '-->'), '', $matches[1]);
601 while ( $string != $newstring = wp_kses($string, $allowed_html, $allowed_protocols) )
602 $string = $newstring;
605 // prevent multiple dashes in comments
606 $string = preg_replace('/--+/', '-', $string);
607 // prevent three dashes closing a comment
608 $string = preg_replace('/-$/', '', $string);
609 return "<!--{$string}-->";
611 # Allow HTML comments
613 if (!preg_match('%^<\s*(/\s*)?([a-zA-Z0-9]+)([^>]*)>?$%', $string, $matches))
615 # It's seriously malformed
617 $slash = trim($matches[1]);
619 $attrlist = $matches[3];
621 if (!@isset($allowed_html[strtolower($elem)]))
623 # They are using a not allowed HTML element
626 return "<$slash$elem>";
627 # No attributes are allowed for closing elements
629 return wp_kses_attr("$slash$elem", $attrlist, $allowed_html, $allowed_protocols);
633 * Removes all attributes, if none are allowed for this element.
635 * If some are allowed it calls wp_kses_hair() to split them further, and then
636 * it builds up new HTML code from the data that kses_hair() returns. It also
637 * removes "<" and ">" characters, if there are any left. One more thing it does
638 * is to check if the tag has a closing XHTML slash, and if it does, it puts one
639 * in the returned code as well.
643 * @param string $element HTML element/tag
644 * @param string $attr HTML attributes from HTML element to closing HTML element tag
645 * @param array $allowed_html Allowed HTML elements
646 * @param array $allowed_protocols Allowed protocols to keep
647 * @return string Sanitized HTML element
649 function wp_kses_attr($element, $attr, $allowed_html, $allowed_protocols) {
650 # Is there a closing XHTML slash at the end of the attributes?
653 if (preg_match('%\s*/\s*$%', $attr))
656 # Are any attributes allowed at all for this element?
658 if (@ count($allowed_html[strtolower($element)]) == 0)
659 return "<$element$xhtml_slash>";
663 $attrarr = wp_kses_hair($attr, $allowed_protocols);
665 # Go through $attrarr, and save the allowed attributes for this element
670 foreach ($attrarr as $arreach) {
671 if (!@ isset ($allowed_html[strtolower($element)][strtolower($arreach['name'])]))
672 continue; # the attribute is not allowed
674 $current = $allowed_html[strtolower($element)][strtolower($arreach['name'])];
676 continue; # the attribute is not allowed
678 if (!is_array($current))
679 $attr2 .= ' '.$arreach['whole'];
680 # there are no checks
683 # there are some checks
685 foreach ($current as $currkey => $currval)
686 if (!wp_kses_check_attr_val($arreach['value'], $arreach['vless'], $currkey, $currval)) {
691 if ( strtolower($arreach['name']) == 'style' ) {
692 $orig_value = $arreach['value'];
694 $value = safecss_filter_attr($orig_value);
699 $arreach['value'] = $value;
701 $arreach['whole'] = str_replace($orig_value, $value, $arreach['whole']);
705 $attr2 .= ' '.$arreach['whole']; # it passed them
706 } # if !is_array($current)
709 # Remove any "<" or ">" characters
711 $attr2 = preg_replace('/[<>]/', '', $attr2);
713 return "<$element$attr2$xhtml_slash>";
717 * Builds an attribute list from string containing attributes.
719 * This function does a lot of work. It parses an attribute list into an array
720 * with attribute data, and tries to do the right thing even if it gets weird
721 * input. It will add quotes around attribute values that don't have any quotes
722 * or apostrophes around them, to make it easier to produce HTML code that will
723 * conform to W3C's HTML specification. It will also remove bad URL protocols
724 * from attribute values. It also reduces duplicate attributes by using the
725 * attribute defined first (foo='bar' foo='baz' will result in foo='bar').
729 * @param string $attr Attribute list from HTML element to closing HTML element tag
730 * @param array $allowed_protocols Allowed protocols to keep
731 * @return array List of attributes after parsing
733 function wp_kses_hair($attr, $allowed_protocols) {
737 $uris = array('xmlns', 'profile', 'href', 'src', 'cite', 'classid', 'codebase', 'data', 'usemap', 'longdesc', 'action');
739 # Loop through the whole attribute list
741 while (strlen($attr) != 0) {
742 $working = 0; # Was the last operation successful?
745 case 0 : # attribute name, href for instance
747 if (preg_match('/^([-a-zA-Z]+)/', $attr, $match)) {
748 $attrname = $match[1];
749 $working = $mode = 1;
750 $attr = preg_replace('/^[-a-zA-Z]+/', '', $attr);
755 case 1 : # equals sign or valueless ("selected")
757 if (preg_match('/^\s*=\s*/', $attr)) # equals sign
761 $attr = preg_replace('/^\s*=\s*/', '', $attr);
765 if (preg_match('/^\s+/', $attr)) # valueless
769 if(FALSE === array_key_exists($attrname, $attrarr)) {
770 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
772 $attr = preg_replace('/^\s+/', '', $attr);
777 case 2 : # attribute value, a URL after href= for instance
779 if (preg_match('%^"([^"]*)"(\s+|/?$)%', $attr, $match))
782 $thisval = $match[1];
783 if ( in_array(strtolower($attrname), $uris) )
784 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
786 if(FALSE === array_key_exists($attrname, $attrarr)) {
787 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n');
791 $attr = preg_replace('/^"[^"]*"(\s+|$)/', '', $attr);
795 if (preg_match("%^'([^']*)'(\s+|/?$)%", $attr, $match))
798 $thisval = $match[1];
799 if ( in_array(strtolower($attrname), $uris) )
800 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
802 if(FALSE === array_key_exists($attrname, $attrarr)) {
803 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname='$thisval'", 'vless' => 'n');
807 $attr = preg_replace("/^'[^']*'(\s+|$)/", '', $attr);
811 if (preg_match("%^([^\s\"']+)(\s+|/?$)%", $attr, $match))
814 $thisval = $match[1];
815 if ( in_array(strtolower($attrname), $uris) )
816 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
818 if(FALSE === array_key_exists($attrname, $attrarr)) {
819 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n');
821 # We add quotes to conform to W3C's HTML spec.
824 $attr = preg_replace("%^[^\s\"']+(\s+|$)%", '', $attr);
830 if ($working == 0) # not well formed, remove and try again
832 $attr = wp_kses_html_error($attr);
837 if ($mode == 1 && FALSE === array_key_exists($attrname, $attrarr))
838 # special case, for when the attribute list ends with a valueless
839 # attribute like "selected"
840 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
846 * Performs different checks for attribute values.
848 * The currently implemented checks are "maxlen", "minlen", "maxval", "minval"
849 * and "valueless" with even more checks to come soon.
853 * @param string $value Attribute value
854 * @param string $vless Whether the value is valueless. Use 'y' or 'n'
855 * @param string $checkname What $checkvalue is checking for.
856 * @param mixed $checkvalue What constraint the value should pass
857 * @return bool Whether check passes
859 function wp_kses_check_attr_val($value, $vless, $checkname, $checkvalue) {
862 switch (strtolower($checkname)) {
864 # The maxlen check makes sure that the attribute value has a length not
865 # greater than the given value. This can be used to avoid Buffer Overflows
866 # in WWW clients and various Internet servers.
868 if (strlen($value) > $checkvalue)
873 # The minlen check makes sure that the attribute value has a length not
874 # smaller than the given value.
876 if (strlen($value) < $checkvalue)
881 # The maxval check does two things: it checks that the attribute value is
882 # an integer from 0 and up, without an excessive amount of zeroes or
883 # whitespace (to avoid Buffer Overflows). It also checks that the attribute
884 # value is not greater than the given value.
885 # This check can be used to avoid Denial of Service attacks.
887 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
889 if ($value > $checkvalue)
894 # The minval check checks that the attribute value is a positive integer,
895 # and that it is not smaller than the given value.
897 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
899 if ($value < $checkvalue)
904 # The valueless check checks if the attribute has a value
905 # (like <a href="blah">) or not (<option selected>). If the given value
906 # is a "y" or a "Y", the attribute must not have a value.
907 # If the given value is an "n" or an "N", the attribute must have one.
909 if (strtolower($checkvalue) != $vless)
918 * Sanitize string from bad protocols.
920 * This function removes all non-allowed protocols from the beginning of
921 * $string. It ignores whitespace and the case of the letters, and it does
922 * understand HTML entities. It does its work in a while loop, so it won't be
923 * fooled by a string like "javascript:javascript:alert(57)".
927 * @param string $string Content to filter bad protocols from
928 * @param array $allowed_protocols Allowed protocols to keep
929 * @return string Filtered content
931 function wp_kses_bad_protocol($string, $allowed_protocols) {
932 $string = wp_kses_no_null($string);
933 $string2 = $string.'a';
935 while ($string != $string2) {
937 $string = wp_kses_bad_protocol_once($string, $allowed_protocols);
944 * Removes any NULL characters in $string.
948 * @param string $string
951 function wp_kses_no_null($string) {
952 $string = preg_replace('/\0+/', '', $string);
953 $string = preg_replace('/(\\\\0)+/', '', $string);
959 * Strips slashes from in front of quotes.
961 * This function changes the character sequence \" to just ". It leaves all
962 * other slashes alone. It's really weird, but the quoting from
963 * preg_replace(//e) seems to require this.
967 * @param string $string String to strip slashes
968 * @return string Fixed strings with quoted slashes
970 function wp_kses_stripslashes($string) {
971 return preg_replace('%\\\\"%', '"', $string);
975 * Goes through an array and changes the keys to all lower case.
979 * @param array $inarray Unfiltered array
980 * @return array Fixed array with all lowercase keys
982 function wp_kses_array_lc($inarray) {
983 $outarray = array ();
985 foreach ( (array) $inarray as $inkey => $inval) {
986 $outkey = strtolower($inkey);
987 $outarray[$outkey] = array ();
989 foreach ( (array) $inval as $inkey2 => $inval2) {
990 $outkey2 = strtolower($inkey2);
991 $outarray[$outkey][$outkey2] = $inval2;
999 * Removes the HTML JavaScript entities found in early versions of Netscape 4.
1003 * @param string $string
1006 function wp_kses_js_entities($string) {
1007 return preg_replace('%&\s*\{[^}]*(\}\s*;?|$)%', '', $string);
1011 * Handles parsing errors in wp_kses_hair().
1013 * The general plan is to remove everything to and including some whitespace,
1014 * but it deals with quotes and apostrophes as well.
1018 * @param string $string
1021 function wp_kses_html_error($string) {
1022 return preg_replace('/^("[^"]*("|$)|\'[^\']*(\'|$)|\S)*\s*/', '', $string);
1026 * Sanitizes content from bad protocols and other characters.
1028 * This function searches for URL protocols at the beginning of $string, while
1029 * handling whitespace and HTML entities.
1033 * @param string $string Content to check for bad protocols
1034 * @param string $allowed_protocols Allowed protocols
1035 * @return string Sanitized content
1037 function wp_kses_bad_protocol_once($string, $allowed_protocols) {
1038 $string2 = preg_split( '/:|�*58;|�*3a;/i', $string, 2 );
1039 if ( isset($string2[1]) && ! preg_match('%/\?%', $string2[0]) )
1040 $string = wp_kses_bad_protocol_once2( $string2[0], $allowed_protocols ) . trim( $string2[1] );
1046 * Callback for wp_kses_bad_protocol_once() regular expression.
1048 * This function processes URL protocols, checks to see if they're in the
1049 * white-list or not, and returns different data depending on the answer.
1054 * @param string $string URI scheme to check against the whitelist
1055 * @param string $allowed_protocols Allowed protocols
1056 * @return string Sanitized content
1058 function wp_kses_bad_protocol_once2( $string, $allowed_protocols ) {
1059 $string2 = wp_kses_decode_entities($string);
1060 $string2 = preg_replace('/\s/', '', $string2);
1061 $string2 = wp_kses_no_null($string2);
1062 $string2 = strtolower($string2);
1065 foreach ( (array) $allowed_protocols as $one_protocol )
1066 if ( strtolower($one_protocol) == $string2 ) {
1078 * Converts and fixes HTML entities.
1080 * This function normalizes HTML entities. It will convert "AT&T" to the correct
1081 * "AT&T", ":" to ":", "&#XYZZY;" to "&#XYZZY;" and so on.
1085 * @param string $string Content to normalize entities
1086 * @return string Content with normalized entities
1088 function wp_kses_normalize_entities($string) {
1089 # Disarm all entities by converting & to &
1091 $string = str_replace('&', '&', $string);
1093 # Change back the allowed entities in our entity whitelist
1095 $string = preg_replace_callback('/&([A-Za-z]{2,8});/', 'wp_kses_named_entities', $string);
1096 $string = preg_replace_callback('/&#(0*[0-9]{1,7});/', 'wp_kses_normalize_entities2', $string);
1097 $string = preg_replace_callback('/&#[Xx](0*[0-9A-Fa-f]{1,6});/', 'wp_kses_normalize_entities3', $string);
1103 * Callback for wp_kses_normalize_entities() regular expression.
1105 * This function only accepts valid named entity references, which are finite,
1106 * case-sensitive, and highly scrutinized by HTML and XML validators.
1110 * @param array $matches preg_replace_callback() matches array
1111 * @return string Correctly encoded entity
1113 function wp_kses_named_entities($matches) {
1114 global $allowedentitynames;
1116 if ( empty($matches[1]) )
1120 return ( ( ! in_array($i, $allowedentitynames) ) ? "&$i;" : "&$i;" );
1124 * Callback for wp_kses_normalize_entities() regular expression.
1126 * This function helps wp_kses_normalize_entities() to only accept 16 bit values
1127 * and nothing more for &#number; entities.
1132 * @param array $matches preg_replace_callback() matches array
1133 * @return string Correctly encoded entity
1135 function wp_kses_normalize_entities2($matches) {
1136 if ( empty($matches[1]) )
1140 if (valid_unicode($i)) {
1141 $i = str_pad(ltrim($i,'0'), 3, '0', STR_PAD_LEFT);
1151 * Callback for wp_kses_normalize_entities() for regular expression.
1153 * This function helps wp_kses_normalize_entities() to only accept valid Unicode
1154 * numeric entities in hex form.
1158 * @param array $matches preg_replace_callback() matches array
1159 * @return string Correctly encoded entity
1161 function wp_kses_normalize_entities3($matches) {
1162 if ( empty($matches[1]) )
1165 $hexchars = $matches[1];
1166 return ( ( ! valid_unicode(hexdec($hexchars)) ) ? "&#x$hexchars;" : '&#x'.ltrim($hexchars,'0').';' );
1170 * Helper function to determine if a Unicode value is valid.
1172 * @param int $i Unicode value
1173 * @return bool true if the value was a valid Unicode number
1175 function valid_unicode($i) {
1176 return ( $i == 0x9 || $i == 0xa || $i == 0xd ||
1177 ($i >= 0x20 && $i <= 0xd7ff) ||
1178 ($i >= 0xe000 && $i <= 0xfffd) ||
1179 ($i >= 0x10000 && $i <= 0x10ffff) );
1183 * Convert all entities to their character counterparts.
1185 * This function decodes numeric HTML entities (A and A). It doesn't do
1186 * anything with other entities like ä, but we don't need them in the URL
1187 * protocol whitelisting system anyway.
1191 * @param string $string Content to change entities
1192 * @return string Content after decoded entities
1194 function wp_kses_decode_entities($string) {
1195 $string = preg_replace_callback('/&#([0-9]+);/', '_wp_kses_decode_entities_chr', $string);
1196 $string = preg_replace_callback('/&#[Xx]([0-9A-Fa-f]+);/', '_wp_kses_decode_entities_chr_hexdec', $string);
1202 * Regex callback for wp_kses_decode_entities()
1204 * @param array $match preg match
1207 function _wp_kses_decode_entities_chr( $match ) {
1208 return chr( $match[1] );
1212 * Regex callback for wp_kses_decode_entities()
1214 * @param array $match preg match
1217 function _wp_kses_decode_entities_chr_hexdec( $match ) {
1218 return chr( hexdec( $match[1] ) );
1222 * Sanitize content with allowed HTML Kses rules.
1225 * @uses $allowedtags
1227 * @param string $data Content to filter, expected to be escaped with slashes
1228 * @return string Filtered content
1230 function wp_filter_kses($data) {
1231 global $allowedtags;
1232 return addslashes( wp_kses(stripslashes( $data ), $allowedtags) );
1236 * Sanitize content with allowed HTML Kses rules.
1239 * @uses $allowedtags
1241 * @param string $data Content to filter, expected to not be escaped
1242 * @return string Filtered content
1244 function wp_kses_data($data) {
1245 global $allowedtags;
1246 return wp_kses( $data , $allowedtags );
1250 * Sanitize content for allowed HTML tags for post content.
1252 * Post content refers to the page contents of the 'post' type and not $_POST
1256 * @uses $allowedposttags
1258 * @param string $data Post content to filter, expected to be escaped with slashes
1259 * @return string Filtered post content with allowed HTML tags and attributes intact.
1261 function wp_filter_post_kses($data) {
1262 global $allowedposttags;
1263 return addslashes ( wp_kses(stripslashes( $data ), $allowedposttags) );
1267 * Sanitize content for allowed HTML tags for post content.
1269 * Post content refers to the page contents of the 'post' type and not $_POST
1273 * @uses $allowedposttags
1275 * @param string $data Post content to filter
1276 * @return string Filtered post content with allowed HTML tags and attributes intact.
1278 function wp_kses_post($data) {
1279 global $allowedposttags;
1280 return wp_kses( $data , $allowedposttags );
1284 * Strips all of the HTML in the content.
1288 * @param string $data Content to strip all HTML from
1289 * @return string Filtered content without any HTML
1291 function wp_filter_nohtml_kses($data) {
1292 return addslashes ( wp_kses(stripslashes( $data ), array()) );
1296 * Adds all Kses input form content filters.
1298 * All hooks have default priority. The wp_filter_kses() function is added to
1299 * the 'pre_comment_content' and 'title_save_pre' hooks.
1301 * The wp_filter_post_kses() function is added to the 'content_save_pre',
1302 * 'excerpt_save_pre', and 'content_filtered_save_pre' hooks.
1305 * @uses add_filter() See description for what functions are added to what hooks.
1307 function kses_init_filters() {
1308 // Normal filtering.
1309 add_filter('pre_comment_content', 'wp_filter_kses');
1310 add_filter('title_save_pre', 'wp_filter_kses');
1313 add_filter('content_save_pre', 'wp_filter_post_kses');
1314 add_filter('excerpt_save_pre', 'wp_filter_post_kses');
1315 add_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1319 * Removes all Kses input form content filters.
1321 * A quick procedural method to removing all of the filters that kses uses for
1322 * content in WordPress Loop.
1324 * Does not remove the kses_init() function from 'init' hook (priority is
1325 * default). Also does not remove kses_init() function from 'set_current_user'
1326 * hook (priority is also default).
1330 function kses_remove_filters() {
1331 // Normal filtering.
1332 remove_filter('pre_comment_content', 'wp_filter_kses');
1333 remove_filter('title_save_pre', 'wp_filter_kses');
1336 remove_filter('content_save_pre', 'wp_filter_post_kses');
1337 remove_filter('excerpt_save_pre', 'wp_filter_post_kses');
1338 remove_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1342 * Sets up most of the Kses filters for input form content.
1344 * If you remove the kses_init() function from 'init' hook and
1345 * 'set_current_user' (priority is default), then none of the Kses filter hooks
1348 * First removes all of the Kses filters in case the current user does not need
1349 * to have Kses filter the content. If the user does not have unfiltered html
1350 * capability, then Kses filters are added.
1352 * @uses kses_remove_filters() Removes the Kses filters
1353 * @uses kses_init_filters() Adds the Kses filters back if the user
1354 * does not have unfiltered HTML capability.
1357 function kses_init() {
1358 kses_remove_filters();
1360 if (current_user_can('unfiltered_html') == false)
1361 kses_init_filters();
1364 add_action('init', 'kses_init');
1365 add_action('set_current_user', 'kses_init');
1372 function safecss_filter_attr( $css, $deprecated = '' ) {
1373 if ( !empty( $deprecated ) )
1374 _deprecated_argument( __FUNCTION__, '2.8.1' ); // Never implemented
1376 $css = wp_kses_no_null($css);
1377 $css = str_replace(array("\n","\r","\t"), '', $css);
1379 if ( preg_match( '%[\\(&=}]|/\*%', $css ) ) // remove any inline css containing \ ( & } = or comments
1382 $css_array = explode( ';', trim( $css ) );
1383 $allowed_attr = apply_filters( 'safe_style_css', array( 'text-align', 'margin', 'color', 'float',
1384 'border', 'background', 'background-color', 'border-bottom', 'border-bottom-color',
1385 'border-bottom-style', 'border-bottom-width', 'border-collapse', 'border-color', 'border-left',
1386 'border-left-color', 'border-left-style', 'border-left-width', 'border-right', 'border-right-color',
1387 'border-right-style', 'border-right-width', 'border-spacing', 'border-style', 'border-top',
1388 'border-top-color', 'border-top-style', 'border-top-width', 'border-width', 'caption-side',
1389 'clear', 'cursor', 'direction', 'font', 'font-family', 'font-size', 'font-style',
1390 'font-variant', 'font-weight', 'height', 'letter-spacing', 'line-height', 'margin-bottom',
1391 'margin-left', 'margin-right', 'margin-top', 'overflow', 'padding', 'padding-bottom',
1392 'padding-left', 'padding-right', 'padding-top', 'text-decoration', 'text-indent', 'vertical-align',
1395 if ( empty($allowed_attr) )
1399 foreach ( $css_array as $css_item ) {
1400 if ( $css_item == '' )
1402 $css_item = trim( $css_item );
1404 if ( strpos( $css_item, ':' ) === false ) {
1407 $parts = split( ':', $css_item );
1408 if ( in_array( trim( $parts[0] ), $allowed_attr ) )