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.
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 if ( empty( $allowed_protocols ) )
504 $allowed_protocols = wp_allowed_protocols();
505 $string = wp_kses_no_null($string);
506 $string = wp_kses_js_entities($string);
507 $string = wp_kses_normalize_entities($string);
508 $allowed_html_fixed = wp_kses_array_lc($allowed_html);
509 $string = wp_kses_hook($string, $allowed_html_fixed, $allowed_protocols); // WP changed the order of these funcs and added args to wp_kses_hook
510 return wp_kses_split($string, $allowed_html_fixed, $allowed_protocols);
514 * You add any kses hooks here.
516 * There is currently only one kses WordPress hook and it is called here. All
517 * parameters are passed to the hooks and expected to receive a string.
521 * @param string $string Content to filter through kses
522 * @param array $allowed_html List of allowed HTML elements
523 * @param array $allowed_protocols Allowed protocol in links
524 * @return string Filtered content through 'pre_kses' hook
526 function wp_kses_hook($string, $allowed_html, $allowed_protocols) {
527 $string = apply_filters('pre_kses', $string, $allowed_html, $allowed_protocols);
532 * This function returns kses' version number.
536 * @return string KSES Version Number
538 function wp_kses_version() {
543 * Searches for HTML tags, no matter how malformed.
545 * It also matches stray ">" characters.
549 * @param string $string Content to filter
550 * @param array $allowed_html Allowed HTML elements
551 * @param array $allowed_protocols Allowed protocols to keep
552 * @return string Content with fixed HTML tags
554 function wp_kses_split($string, $allowed_html, $allowed_protocols) {
555 global $pass_allowed_html, $pass_allowed_protocols;
556 $pass_allowed_html = $allowed_html;
557 $pass_allowed_protocols = $allowed_protocols;
558 return preg_replace_callback( '%(<!--.*?(-->|$))|(<[^>]*(>|$)|>)%', '_wp_kses_split_callback', $string );
562 * Callback for wp_kses_split.
567 function _wp_kses_split_callback( $match ) {
568 global $pass_allowed_html, $pass_allowed_protocols;
569 return wp_kses_split2( $match[0], $pass_allowed_html, $pass_allowed_protocols );
573 * Callback for wp_kses_split for fixing malformed HTML tags.
575 * This function does a lot of work. It rejects some very malformed things like
576 * <:::>. It returns an empty string, if the element isn't allowed (look ma, no
577 * strip_tags()!). Otherwise it splits the tag into an element and an attribute
580 * After the tag is split into an element and an attribute list, it is run
581 * through another filter which will remove illegal attributes and once that is
582 * completed, will be returned.
586 * @uses wp_kses_attr()
588 * @param string $string Content to filter
589 * @param array $allowed_html Allowed HTML elements
590 * @param array $allowed_protocols Allowed protocols to keep
591 * @return string Fixed HTML element
593 function wp_kses_split2($string, $allowed_html, $allowed_protocols) {
594 $string = wp_kses_stripslashes($string);
596 if (substr($string, 0, 1) != '<')
598 # It matched a ">" character
600 if ( '<!--' == substr( $string, 0, 4 ) ) {
601 $string = str_replace( array('<!--', '-->'), '', $string );
602 while ( $string != ($newstring = wp_kses($string, $allowed_html, $allowed_protocols)) )
603 $string = $newstring;
606 // prevent multiple dashes in comments
607 $string = preg_replace('/--+/', '-', $string);
608 // prevent three dashes closing a comment
609 $string = preg_replace('/-$/', '', $string);
610 return "<!--{$string}-->";
612 # Allow HTML comments
614 if (!preg_match('%^<\s*(/\s*)?([a-zA-Z0-9]+)([^>]*)>?$%', $string, $matches))
616 # It's seriously malformed
618 $slash = trim($matches[1]);
620 $attrlist = $matches[3];
622 if ( ! isset($allowed_html[strtolower($elem)]) )
624 # They are using a not allowed HTML element
628 # No attributes are allowed for closing elements
630 return wp_kses_attr( $elem, $attrlist, $allowed_html, $allowed_protocols );
634 * Removes all attributes, if none are allowed for this element.
636 * If some are allowed it calls wp_kses_hair() to split them further, and then
637 * it builds up new HTML code from the data that kses_hair() returns. It also
638 * removes "<" and ">" characters, if there are any left. One more thing it does
639 * is to check if the tag has a closing XHTML slash, and if it does, it puts one
640 * in the returned code as well.
644 * @param string $element HTML element/tag
645 * @param string $attr HTML attributes from HTML element to closing HTML element tag
646 * @param array $allowed_html Allowed HTML elements
647 * @param array $allowed_protocols Allowed protocols to keep
648 * @return string Sanitized HTML element
650 function wp_kses_attr($element, $attr, $allowed_html, $allowed_protocols) {
651 # Is there a closing XHTML slash at the end of the attributes?
654 if (preg_match('%\s*/\s*$%', $attr))
657 # Are any attributes allowed at all for this element?
658 if ( ! isset($allowed_html[strtolower($element)]) || count($allowed_html[strtolower($element)]) == 0 )
659 return "<$element$xhtml_slash>";
662 $attrarr = wp_kses_hair($attr, $allowed_protocols);
664 # Go through $attrarr, and save the allowed attributes for this element
668 $allowed_attr = $allowed_html[strtolower($element)];
669 foreach ($attrarr as $arreach) {
670 if ( ! isset( $allowed_attr[strtolower($arreach['name'])] ) )
671 continue; # the attribute is not allowed
673 $current = $allowed_attr[strtolower($arreach['name'])];
674 if ( $current == '' )
675 continue; # the attribute is not allowed
677 if ( ! is_array($current) ) {
678 $attr2 .= ' '.$arreach['whole'];
679 # there are no checks
682 # there are some checks
684 foreach ($current as $currkey => $currval) {
685 if ( ! wp_kses_check_attr_val($arreach['value'], $arreach['vless'], $currkey, $currval) ) {
691 if ( strtolower($arreach['name']) == 'style' ) {
692 $orig_value = $arreach['value'];
693 $value = safecss_filter_attr($orig_value);
698 $arreach['value'] = $value;
699 $arreach['whole'] = str_replace($orig_value, $value, $arreach['whole']);
703 $attr2 .= ' '.$arreach['whole']; # it passed them
704 } # if !is_array($current)
707 # Remove any "<" or ">" characters
708 $attr2 = preg_replace('/[<>]/', '', $attr2);
710 return "<$element$attr2$xhtml_slash>";
714 * Builds an attribute list from string containing attributes.
716 * This function does a lot of work. It parses an attribute list into an array
717 * with attribute data, and tries to do the right thing even if it gets weird
718 * input. It will add quotes around attribute values that don't have any quotes
719 * or apostrophes around them, to make it easier to produce HTML code that will
720 * conform to W3C's HTML specification. It will also remove bad URL protocols
721 * from attribute values. It also reduces duplicate attributes by using the
722 * attribute defined first (foo='bar' foo='baz' will result in foo='bar').
726 * @param string $attr Attribute list from HTML element to closing HTML element tag
727 * @param array $allowed_protocols Allowed protocols to keep
728 * @return array List of attributes after parsing
730 function wp_kses_hair($attr, $allowed_protocols) {
734 $uris = array('xmlns', 'profile', 'href', 'src', 'cite', 'classid', 'codebase', 'data', 'usemap', 'longdesc', 'action');
736 # Loop through the whole attribute list
738 while (strlen($attr) != 0) {
739 $working = 0; # Was the last operation successful?
742 case 0 : # attribute name, href for instance
744 if (preg_match('/^([-a-zA-Z]+)/', $attr, $match)) {
745 $attrname = $match[1];
746 $working = $mode = 1;
747 $attr = preg_replace('/^[-a-zA-Z]+/', '', $attr);
752 case 1 : # equals sign or valueless ("selected")
754 if (preg_match('/^\s*=\s*/', $attr)) # equals sign
758 $attr = preg_replace('/^\s*=\s*/', '', $attr);
762 if (preg_match('/^\s+/', $attr)) # valueless
766 if(FALSE === array_key_exists($attrname, $attrarr)) {
767 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
769 $attr = preg_replace('/^\s+/', '', $attr);
774 case 2 : # attribute value, a URL after href= for instance
776 if (preg_match('%^"([^"]*)"(\s+|/?$)%', $attr, $match))
779 $thisval = $match[1];
780 if ( in_array(strtolower($attrname), $uris) )
781 $thisval = wp_kses_bad_protocol($thisval, $allowed_protocols);
783 if(FALSE === array_key_exists($attrname, $attrarr)) {
784 $attrarr[$attrname] = array ('name' => $attrname, 'value' => $thisval, 'whole' => "$attrname=\"$thisval\"", 'vless' => 'n');
788 $attr = preg_replace('/^"[^"]*"(\s+|$)/', '', $attr);
792 if (preg_match("%^'([^']*)'(\s+|/?$)%", $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 $attr = preg_replace("/^'[^']*'(\s+|$)/", '', $attr);
808 if (preg_match("%^([^\s\"']+)(\s+|/?$)%", $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');
818 # We add quotes to conform to W3C's HTML spec.
821 $attr = preg_replace("%^[^\s\"']+(\s+|$)%", '', $attr);
827 if ($working == 0) # not well formed, remove and try again
829 $attr = wp_kses_html_error($attr);
834 if ($mode == 1 && FALSE === array_key_exists($attrname, $attrarr))
835 # special case, for when the attribute list ends with a valueless
836 # attribute like "selected"
837 $attrarr[$attrname] = array ('name' => $attrname, 'value' => '', 'whole' => $attrname, 'vless' => 'y');
843 * Performs different checks for attribute values.
845 * The currently implemented checks are "maxlen", "minlen", "maxval", "minval"
850 * @param string $value Attribute value
851 * @param string $vless Whether the value is valueless. Use 'y' or 'n'
852 * @param string $checkname What $checkvalue is checking for.
853 * @param mixed $checkvalue What constraint the value should pass
854 * @return bool Whether check passes
856 function wp_kses_check_attr_val($value, $vless, $checkname, $checkvalue) {
859 switch (strtolower($checkname)) {
861 # The maxlen check makes sure that the attribute value has a length not
862 # greater than the given value. This can be used to avoid Buffer Overflows
863 # in WWW clients and various Internet servers.
865 if (strlen($value) > $checkvalue)
870 # The minlen check makes sure that the attribute value has a length not
871 # smaller than the given value.
873 if (strlen($value) < $checkvalue)
878 # The maxval check does two things: it checks that the attribute value is
879 # an integer from 0 and up, without an excessive amount of zeroes or
880 # whitespace (to avoid Buffer Overflows). It also checks that the attribute
881 # value is not greater than the given value.
882 # This check can be used to avoid Denial of Service attacks.
884 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
886 if ($value > $checkvalue)
891 # The minval check makes sure that the attribute value is a positive integer,
892 # and that it is not smaller than the given value.
894 if (!preg_match('/^\s{0,6}[0-9]{1,6}\s{0,6}$/', $value))
896 if ($value < $checkvalue)
901 # The valueless check makes sure if the attribute has a value
902 # (like <a href="blah">) or not (<option selected>). If the given value
903 # is a "y" or a "Y", the attribute must not have a value.
904 # If the given value is an "n" or an "N", the attribute must have one.
906 if (strtolower($checkvalue) != $vless)
915 * Sanitize string from bad protocols.
917 * This function removes all non-allowed protocols from the beginning of
918 * $string. It ignores whitespace and the case of the letters, and it does
919 * understand HTML entities. It does its work in a while loop, so it won't be
920 * fooled by a string like "javascript:javascript:alert(57)".
924 * @param string $string Content to filter bad protocols from
925 * @param array $allowed_protocols Allowed protocols to keep
926 * @return string Filtered content
928 function wp_kses_bad_protocol($string, $allowed_protocols) {
929 $string = wp_kses_no_null($string);
930 $string2 = $string.'a';
932 while ($string != $string2) {
934 $string = wp_kses_bad_protocol_once($string, $allowed_protocols);
941 * Removes any NULL characters in $string.
945 * @param string $string
948 function wp_kses_no_null($string) {
949 $string = preg_replace('/\0+/', '', $string);
950 $string = preg_replace('/(\\\\0)+/', '', $string);
956 * Strips slashes from in front of quotes.
958 * This function changes the character sequence \" to just ". It leaves all
959 * other slashes alone. It's really weird, but the quoting from
960 * preg_replace(//e) seems to require this.
964 * @param string $string String to strip slashes
965 * @return string Fixed string with quoted slashes
967 function wp_kses_stripslashes($string) {
968 return preg_replace('%\\\\"%', '"', $string);
972 * Goes through an array and changes the keys to all lower case.
976 * @param array $inarray Unfiltered array
977 * @return array Fixed array with all lowercase keys
979 function wp_kses_array_lc($inarray) {
980 $outarray = array ();
982 foreach ( (array) $inarray as $inkey => $inval) {
983 $outkey = strtolower($inkey);
984 $outarray[$outkey] = array ();
986 foreach ( (array) $inval as $inkey2 => $inval2) {
987 $outkey2 = strtolower($inkey2);
988 $outarray[$outkey][$outkey2] = $inval2;
996 * Removes the HTML JavaScript entities found in early versions of Netscape 4.
1000 * @param string $string
1003 function wp_kses_js_entities($string) {
1004 return preg_replace('%&\s*\{[^}]*(\}\s*;?|$)%', '', $string);
1008 * Handles parsing errors in wp_kses_hair().
1010 * The general plan is to remove everything to and including some whitespace,
1011 * but it deals with quotes and apostrophes as well.
1015 * @param string $string
1018 function wp_kses_html_error($string) {
1019 return preg_replace('/^("[^"]*("|$)|\'[^\']*(\'|$)|\S)*\s*/', '', $string);
1023 * Sanitizes content from bad protocols and other characters.
1025 * This function searches for URL protocols at the beginning of $string, while
1026 * handling whitespace and HTML entities.
1030 * @param string $string Content to check for bad protocols
1031 * @param string $allowed_protocols Allowed protocols
1032 * @return string Sanitized content
1034 function wp_kses_bad_protocol_once($string, $allowed_protocols) {
1035 $string2 = preg_split( '/:|�*58;|�*3a;/i', $string, 2 );
1036 if ( isset($string2[1]) && ! preg_match('%/\?%', $string2[0]) )
1037 $string = wp_kses_bad_protocol_once2( $string2[0], $allowed_protocols ) . trim( $string2[1] );
1043 * Callback for wp_kses_bad_protocol_once() regular expression.
1045 * This function processes URL protocols, checks to see if they're in the
1046 * whitelist or not, and returns different data depending on the answer.
1051 * @param string $string URI scheme to check against the whitelist
1052 * @param string $allowed_protocols Allowed protocols
1053 * @return string Sanitized content
1055 function wp_kses_bad_protocol_once2( $string, $allowed_protocols ) {
1056 $string2 = wp_kses_decode_entities($string);
1057 $string2 = preg_replace('/\s/', '', $string2);
1058 $string2 = wp_kses_no_null($string2);
1059 $string2 = strtolower($string2);
1062 foreach ( (array) $allowed_protocols as $one_protocol )
1063 if ( strtolower($one_protocol) == $string2 ) {
1075 * Converts and fixes HTML entities.
1077 * This function normalizes HTML entities. It will convert "AT&T" to the correct
1078 * "AT&T", ":" to ":", "&#XYZZY;" to "&#XYZZY;" and so on.
1082 * @param string $string Content to normalize entities
1083 * @return string Content with normalized entities
1085 function wp_kses_normalize_entities($string) {
1086 # Disarm all entities by converting & to &
1088 $string = str_replace('&', '&', $string);
1090 # Change back the allowed entities in our entity whitelist
1092 $string = preg_replace_callback('/&([A-Za-z]{2,8});/', 'wp_kses_named_entities', $string);
1093 $string = preg_replace_callback('/&#(0*[0-9]{1,7});/', 'wp_kses_normalize_entities2', $string);
1094 $string = preg_replace_callback('/&#[Xx](0*[0-9A-Fa-f]{1,6});/', 'wp_kses_normalize_entities3', $string);
1100 * Callback for wp_kses_normalize_entities() regular expression.
1102 * This function only accepts valid named entity references, which are finite,
1103 * case-sensitive, and highly scrutinized by HTML and XML validators.
1107 * @param array $matches preg_replace_callback() matches array
1108 * @return string Correctly encoded entity
1110 function wp_kses_named_entities($matches) {
1111 global $allowedentitynames;
1113 if ( empty($matches[1]) )
1117 return ( ( ! in_array($i, $allowedentitynames) ) ? "&$i;" : "&$i;" );
1121 * Callback for wp_kses_normalize_entities() regular expression.
1123 * This function helps wp_kses_normalize_entities() to only accept 16-bit values
1124 * and nothing more for &#number; entities.
1129 * @param array $matches preg_replace_callback() matches array
1130 * @return string Correctly encoded entity
1132 function wp_kses_normalize_entities2($matches) {
1133 if ( empty($matches[1]) )
1137 if (valid_unicode($i)) {
1138 $i = str_pad(ltrim($i,'0'), 3, '0', STR_PAD_LEFT);
1148 * Callback for wp_kses_normalize_entities() for regular expression.
1150 * This function helps wp_kses_normalize_entities() to only accept valid Unicode
1151 * numeric entities in hex form.
1155 * @param array $matches preg_replace_callback() matches array
1156 * @return string Correctly encoded entity
1158 function wp_kses_normalize_entities3($matches) {
1159 if ( empty($matches[1]) )
1162 $hexchars = $matches[1];
1163 return ( ( ! valid_unicode(hexdec($hexchars)) ) ? "&#x$hexchars;" : '&#x'.ltrim($hexchars,'0').';' );
1167 * Helper function to determine if a Unicode value is valid.
1169 * @param int $i Unicode value
1170 * @return bool True if the value was a valid Unicode number
1172 function valid_unicode($i) {
1173 return ( $i == 0x9 || $i == 0xa || $i == 0xd ||
1174 ($i >= 0x20 && $i <= 0xd7ff) ||
1175 ($i >= 0xe000 && $i <= 0xfffd) ||
1176 ($i >= 0x10000 && $i <= 0x10ffff) );
1180 * Convert all entities to their character counterparts.
1182 * This function decodes numeric HTML entities (A and A). It doesn't do
1183 * anything with other entities like ä, but we don't need them in the URL
1184 * protocol whitelisting system anyway.
1188 * @param string $string Content to change entities
1189 * @return string Content after decoded entities
1191 function wp_kses_decode_entities($string) {
1192 $string = preg_replace_callback('/&#([0-9]+);/', '_wp_kses_decode_entities_chr', $string);
1193 $string = preg_replace_callback('/&#[Xx]([0-9A-Fa-f]+);/', '_wp_kses_decode_entities_chr_hexdec', $string);
1199 * Regex callback for wp_kses_decode_entities()
1201 * @param array $match preg match
1204 function _wp_kses_decode_entities_chr( $match ) {
1205 return chr( $match[1] );
1209 * Regex callback for wp_kses_decode_entities()
1211 * @param array $match preg match
1214 function _wp_kses_decode_entities_chr_hexdec( $match ) {
1215 return chr( hexdec( $match[1] ) );
1219 * Sanitize content with allowed HTML Kses rules.
1222 * @uses $allowedtags
1224 * @param string $data Content to filter, expected to be escaped with slashes
1225 * @return string Filtered content
1227 function wp_filter_kses($data) {
1228 global $allowedtags;
1229 return addslashes( wp_kses(stripslashes( $data ), $allowedtags) );
1233 * Sanitize content with allowed HTML Kses rules.
1236 * @uses $allowedtags
1238 * @param string $data Content to filter, expected to not be escaped
1239 * @return string Filtered content
1241 function wp_kses_data($data) {
1242 global $allowedtags;
1243 return wp_kses( $data , $allowedtags );
1247 * Sanitize content for allowed HTML tags for post content.
1249 * Post content refers to the page contents of the 'post' type and not $_POST
1253 * @uses $allowedposttags
1255 * @param string $data Post content to filter, expected to be escaped with slashes
1256 * @return string Filtered post content with allowed HTML tags and attributes intact.
1258 function wp_filter_post_kses($data) {
1259 global $allowedposttags;
1260 return addslashes ( wp_kses(stripslashes( $data ), $allowedposttags) );
1264 * Sanitize content for allowed HTML tags for post content.
1266 * Post content refers to the page contents of the 'post' type and not $_POST
1270 * @uses $allowedposttags
1272 * @param string $data Post content to filter
1273 * @return string Filtered post content with allowed HTML tags and attributes intact.
1275 function wp_kses_post($data) {
1276 global $allowedposttags;
1277 return wp_kses( $data , $allowedposttags );
1281 * Strips all of the HTML in the content.
1285 * @param string $data Content to strip all HTML from
1286 * @return string Filtered content without any HTML
1288 function wp_filter_nohtml_kses($data) {
1289 return addslashes ( wp_kses(stripslashes( $data ), array()) );
1293 * Adds all Kses input form content filters.
1295 * All hooks have default priority. The wp_filter_kses() function is added to
1296 * the 'pre_comment_content' and 'title_save_pre' hooks.
1298 * The wp_filter_post_kses() function is added to the 'content_save_pre',
1299 * 'excerpt_save_pre', and 'content_filtered_save_pre' hooks.
1302 * @uses add_filter() See description for what functions are added to what hooks.
1304 function kses_init_filters() {
1305 // Normal filtering.
1306 add_filter('pre_comment_content', 'wp_filter_kses');
1307 add_filter('title_save_pre', 'wp_filter_kses');
1310 add_filter('content_save_pre', 'wp_filter_post_kses');
1311 add_filter('excerpt_save_pre', 'wp_filter_post_kses');
1312 add_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1316 * Removes all Kses input form content filters.
1318 * A quick procedural method to removing all of the filters that kses uses for
1319 * content in WordPress Loop.
1321 * Does not remove the kses_init() function from 'init' hook (priority is
1322 * default). Also does not remove kses_init() function from 'set_current_user'
1323 * hook (priority is also default).
1327 function kses_remove_filters() {
1328 // Normal filtering.
1329 remove_filter('pre_comment_content', 'wp_filter_kses');
1330 remove_filter('title_save_pre', 'wp_filter_kses');
1333 remove_filter('content_save_pre', 'wp_filter_post_kses');
1334 remove_filter('excerpt_save_pre', 'wp_filter_post_kses');
1335 remove_filter('content_filtered_save_pre', 'wp_filter_post_kses');
1339 * Sets up most of the Kses filters for input form content.
1341 * If you remove the kses_init() function from 'init' hook and
1342 * 'set_current_user' (priority is default), then none of the Kses filter hooks
1345 * First removes all of the Kses filters in case the current user does not need
1346 * to have Kses filter the content. If the user does not have unfiltered_html
1347 * capability, then Kses filters are added.
1349 * @uses kses_remove_filters() Removes the Kses filters
1350 * @uses kses_init_filters() Adds the Kses filters back if the user
1351 * does not have unfiltered HTML capability.
1354 function kses_init() {
1355 kses_remove_filters();
1357 if (current_user_can('unfiltered_html') == false)
1358 kses_init_filters();
1361 add_action('init', 'kses_init');
1362 add_action('set_current_user', 'kses_init');
1369 function safecss_filter_attr( $css, $deprecated = '' ) {
1370 if ( !empty( $deprecated ) )
1371 _deprecated_argument( __FUNCTION__, '2.8.1' ); // Never implemented
1373 $css = wp_kses_no_null($css);
1374 $css = str_replace(array("\n","\r","\t"), '', $css);
1376 if ( preg_match( '%[\\(&=}]|/\*%', $css ) ) // remove any inline css containing \ ( & } = or comments
1379 $css_array = explode( ';', trim( $css ) );
1380 $allowed_attr = apply_filters( 'safe_style_css', array( 'text-align', 'margin', 'color', 'float',
1381 'border', 'background', 'background-color', 'border-bottom', 'border-bottom-color',
1382 'border-bottom-style', 'border-bottom-width', 'border-collapse', 'border-color', 'border-left',
1383 'border-left-color', 'border-left-style', 'border-left-width', 'border-right', 'border-right-color',
1384 'border-right-style', 'border-right-width', 'border-spacing', 'border-style', 'border-top',
1385 'border-top-color', 'border-top-style', 'border-top-width', 'border-width', 'caption-side',
1386 'clear', 'cursor', 'direction', 'font', 'font-family', 'font-size', 'font-style',
1387 'font-variant', 'font-weight', 'height', 'letter-spacing', 'line-height', 'margin-bottom',
1388 'margin-left', 'margin-right', 'margin-top', 'overflow', 'padding', 'padding-bottom',
1389 'padding-left', 'padding-right', 'padding-top', 'text-decoration', 'text-indent', 'vertical-align',
1392 if ( empty($allowed_attr) )
1396 foreach ( $css_array as $css_item ) {
1397 if ( $css_item == '' )
1399 $css_item = trim( $css_item );
1401 if ( strpos( $css_item, ':' ) === false ) {
1404 $parts = split( ':', $css_item );
1405 if ( in_array( trim( $parts[0] ), $allowed_attr ) )