<?php
-
/**
- * (X)HTML sanitizer for MediaWiki
+ * XHTML sanitizer for MediaWiki
*
* Copyright (C) 2002-2005 Brion Vibber <brion@pobox.com> et al
* http://www.mediawiki.org/
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
+ * the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
*
* You should have received a copy of the GNU General Public License along
* with this program; if not, write to the Free Software Foundation, Inc.,
- * 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
* http://www.gnu.org/copyleft/gpl.html
*
- * @package MediaWiki
+ * @file
+ * @ingroup Parser
+ */
+
+/**
+ * Regular expression to match various types of character references in
+ * Sanitizer::normalizeCharReferences and Sanitizer::decodeCharReferences
+ */
+define( 'MW_CHAR_REFS_REGEX',
+ '/&([A-Za-z0-9\x80-\xff]+);
+ |&\#([0-9]+);
+ |&\#x([0-9A-Za-z]+);
+ |&\#X([0-9A-Za-z]+);
+ |(&)/x' );
+
+/**
+ * Regular expression to match HTML/XML attribute pairs within a tag.
+ * Allows some... latitude.
+ * Used in Sanitizer::fixTagAttributes and Sanitizer::decodeTagAttributes
+ */
+$attrib = '[A-Za-z0-9]';
+$space = '[\x09\x0a\x0d\x20]';
+define( 'MW_ATTRIBS_REGEX',
+ "/(?:^|$space)($attrib+)
+ ($space*=$space*
+ (?:
+ # The attribute value: quoted or alone
+ \"([^<\"]*)\"
+ | '([^<']*)'
+ | ([a-zA-Z0-9!#$%&()*,\\-.\\/:;<>?@[\\]^_`{|}~]+)
+ | (\#[0-9a-fA-F]+) # Technically wrong, but lots of
+ # colors are specified like this.
+ # We'll be normalizing it.
+ )
+ )?(?=$space|\$)/sx" );
+
+/**
+ * List of all named character entities defined in HTML 4.01
+ * http://www.w3.org/TR/html4/sgml/entities.html
+ * @private
*/
+global $wgHtmlEntities;
+$wgHtmlEntities = array(
+ 'Aacute' => 193,
+ 'aacute' => 225,
+ 'Acirc' => 194,
+ 'acirc' => 226,
+ 'acute' => 180,
+ 'AElig' => 198,
+ 'aelig' => 230,
+ 'Agrave' => 192,
+ 'agrave' => 224,
+ 'alefsym' => 8501,
+ 'Alpha' => 913,
+ 'alpha' => 945,
+ 'amp' => 38,
+ 'and' => 8743,
+ 'ang' => 8736,
+ 'Aring' => 197,
+ 'aring' => 229,
+ 'asymp' => 8776,
+ 'Atilde' => 195,
+ 'atilde' => 227,
+ 'Auml' => 196,
+ 'auml' => 228,
+ 'bdquo' => 8222,
+ 'Beta' => 914,
+ 'beta' => 946,
+ 'brvbar' => 166,
+ 'bull' => 8226,
+ 'cap' => 8745,
+ 'Ccedil' => 199,
+ 'ccedil' => 231,
+ 'cedil' => 184,
+ 'cent' => 162,
+ 'Chi' => 935,
+ 'chi' => 967,
+ 'circ' => 710,
+ 'clubs' => 9827,
+ 'cong' => 8773,
+ 'copy' => 169,
+ 'crarr' => 8629,
+ 'cup' => 8746,
+ 'curren' => 164,
+ 'dagger' => 8224,
+ 'Dagger' => 8225,
+ 'darr' => 8595,
+ 'dArr' => 8659,
+ 'deg' => 176,
+ 'Delta' => 916,
+ 'delta' => 948,
+ 'diams' => 9830,
+ 'divide' => 247,
+ 'Eacute' => 201,
+ 'eacute' => 233,
+ 'Ecirc' => 202,
+ 'ecirc' => 234,
+ 'Egrave' => 200,
+ 'egrave' => 232,
+ 'empty' => 8709,
+ 'emsp' => 8195,
+ 'ensp' => 8194,
+ 'Epsilon' => 917,
+ 'epsilon' => 949,
+ 'equiv' => 8801,
+ 'Eta' => 919,
+ 'eta' => 951,
+ 'ETH' => 208,
+ 'eth' => 240,
+ 'Euml' => 203,
+ 'euml' => 235,
+ 'euro' => 8364,
+ 'exist' => 8707,
+ 'fnof' => 402,
+ 'forall' => 8704,
+ 'frac12' => 189,
+ 'frac14' => 188,
+ 'frac34' => 190,
+ 'frasl' => 8260,
+ 'Gamma' => 915,
+ 'gamma' => 947,
+ 'ge' => 8805,
+ 'gt' => 62,
+ 'harr' => 8596,
+ 'hArr' => 8660,
+ 'hearts' => 9829,
+ 'hellip' => 8230,
+ 'Iacute' => 205,
+ 'iacute' => 237,
+ 'Icirc' => 206,
+ 'icirc' => 238,
+ 'iexcl' => 161,
+ 'Igrave' => 204,
+ 'igrave' => 236,
+ 'image' => 8465,
+ 'infin' => 8734,
+ 'int' => 8747,
+ 'Iota' => 921,
+ 'iota' => 953,
+ 'iquest' => 191,
+ 'isin' => 8712,
+ 'Iuml' => 207,
+ 'iuml' => 239,
+ 'Kappa' => 922,
+ 'kappa' => 954,
+ 'Lambda' => 923,
+ 'lambda' => 955,
+ 'lang' => 9001,
+ 'laquo' => 171,
+ 'larr' => 8592,
+ 'lArr' => 8656,
+ 'lceil' => 8968,
+ 'ldquo' => 8220,
+ 'le' => 8804,
+ 'lfloor' => 8970,
+ 'lowast' => 8727,
+ 'loz' => 9674,
+ 'lrm' => 8206,
+ 'lsaquo' => 8249,
+ 'lsquo' => 8216,
+ 'lt' => 60,
+ 'macr' => 175,
+ 'mdash' => 8212,
+ 'micro' => 181,
+ 'middot' => 183,
+ 'minus' => 8722,
+ 'Mu' => 924,
+ 'mu' => 956,
+ 'nabla' => 8711,
+ 'nbsp' => 160,
+ 'ndash' => 8211,
+ 'ne' => 8800,
+ 'ni' => 8715,
+ 'not' => 172,
+ 'notin' => 8713,
+ 'nsub' => 8836,
+ 'Ntilde' => 209,
+ 'ntilde' => 241,
+ 'Nu' => 925,
+ 'nu' => 957,
+ 'Oacute' => 211,
+ 'oacute' => 243,
+ 'Ocirc' => 212,
+ 'ocirc' => 244,
+ 'OElig' => 338,
+ 'oelig' => 339,
+ 'Ograve' => 210,
+ 'ograve' => 242,
+ 'oline' => 8254,
+ 'Omega' => 937,
+ 'omega' => 969,
+ 'Omicron' => 927,
+ 'omicron' => 959,
+ 'oplus' => 8853,
+ 'or' => 8744,
+ 'ordf' => 170,
+ 'ordm' => 186,
+ 'Oslash' => 216,
+ 'oslash' => 248,
+ 'Otilde' => 213,
+ 'otilde' => 245,
+ 'otimes' => 8855,
+ 'Ouml' => 214,
+ 'ouml' => 246,
+ 'para' => 182,
+ 'part' => 8706,
+ 'permil' => 8240,
+ 'perp' => 8869,
+ 'Phi' => 934,
+ 'phi' => 966,
+ 'Pi' => 928,
+ 'pi' => 960,
+ 'piv' => 982,
+ 'plusmn' => 177,
+ 'pound' => 163,
+ 'prime' => 8242,
+ 'Prime' => 8243,
+ 'prod' => 8719,
+ 'prop' => 8733,
+ 'Psi' => 936,
+ 'psi' => 968,
+ 'quot' => 34,
+ 'radic' => 8730,
+ 'rang' => 9002,
+ 'raquo' => 187,
+ 'rarr' => 8594,
+ 'rArr' => 8658,
+ 'rceil' => 8969,
+ 'rdquo' => 8221,
+ 'real' => 8476,
+ 'reg' => 174,
+ 'rfloor' => 8971,
+ 'Rho' => 929,
+ 'rho' => 961,
+ 'rlm' => 8207,
+ 'rsaquo' => 8250,
+ 'rsquo' => 8217,
+ 'sbquo' => 8218,
+ 'Scaron' => 352,
+ 'scaron' => 353,
+ 'sdot' => 8901,
+ 'sect' => 167,
+ 'shy' => 173,
+ 'Sigma' => 931,
+ 'sigma' => 963,
+ 'sigmaf' => 962,
+ 'sim' => 8764,
+ 'spades' => 9824,
+ 'sub' => 8834,
+ 'sube' => 8838,
+ 'sum' => 8721,
+ 'sup' => 8835,
+ 'sup1' => 185,
+ 'sup2' => 178,
+ 'sup3' => 179,
+ 'supe' => 8839,
+ 'szlig' => 223,
+ 'Tau' => 932,
+ 'tau' => 964,
+ 'there4' => 8756,
+ 'Theta' => 920,
+ 'theta' => 952,
+ 'thetasym' => 977,
+ 'thinsp' => 8201,
+ 'THORN' => 222,
+ 'thorn' => 254,
+ 'tilde' => 732,
+ 'times' => 215,
+ 'trade' => 8482,
+ 'Uacute' => 218,
+ 'uacute' => 250,
+ 'uarr' => 8593,
+ 'uArr' => 8657,
+ 'Ucirc' => 219,
+ 'ucirc' => 251,
+ 'Ugrave' => 217,
+ 'ugrave' => 249,
+ 'uml' => 168,
+ 'upsih' => 978,
+ 'Upsilon' => 933,
+ 'upsilon' => 965,
+ 'Uuml' => 220,
+ 'uuml' => 252,
+ 'weierp' => 8472,
+ 'Xi' => 926,
+ 'xi' => 958,
+ 'Yacute' => 221,
+ 'yacute' => 253,
+ 'yen' => 165,
+ 'Yuml' => 376,
+ 'yuml' => 255,
+ 'Zeta' => 918,
+ 'zeta' => 950,
+ 'zwj' => 8205,
+ 'zwnj' => 8204 );
+/**
+ * Character entity aliases accepted by MediaWiki
+ */
+global $wgHtmlEntityAliases;
+$wgHtmlEntityAliases = array(
+ 'רלמ' => 'rlm',
+ 'رلم' => 'rlm',
+);
+
+
+/**
+ * XHTML sanitizer for MediaWiki
+ * @ingroup Parser
+ */
class Sanitizer {
/**
* Cleans up HTML, removes dangerous tags and attributes, and
* removes HTML comments
- * @access private
+ * @private
+ * @param string $text
+ * @param callback $processCallback to do any variable or parameter replacements in HTML attribute values
+ * @param array $args for the processing callback
+ * @return string
*/
- function removeHTMLtags( $text ) {
- global $wgUseTidy, $wgUserHtml;
- $fname = 'Parser::removeHTMLtags';
- wfProfileIn( $fname );
+ static function removeHTMLtags( $text, $processCallback = null, $args = array(), $extratags = array() ) {
+ global $wgUseTidy;
+
+ static $htmlpairs, $htmlsingle, $htmlsingleonly, $htmlnest, $tabletags,
+ $htmllist, $listtags, $htmlsingleallowed, $htmlelements, $staticInitialised;
- if( $wgUserHtml ) {
- $htmlpairs = array( # Tags that must be closed
+ wfProfileIn( __METHOD__ );
+
+ if ( !$staticInitialised ) {
+
+ $htmlpairs = array_merge( $extratags, array( # Tags that must be closed
'b', 'del', 'i', 'ins', 'u', 'font', 'big', 'small', 'sub', 'sup', 'h1',
'h2', 'h3', 'h4', 'h5', 'h6', 'cite', 'code', 'em', 's',
'strike', 'strong', 'tt', 'var', 'div', 'center',
'blockquote', 'ol', 'ul', 'dl', 'table', 'caption', 'pre',
- 'ruby', 'rt' , 'rb' , 'rp', 'p', 'span'
- );
+ 'ruby', 'rt' , 'rb' , 'rp', 'p', 'span', 'u'
+ ) );
$htmlsingle = array(
'br', 'hr', 'li', 'dt', 'dd'
);
+ $htmlsingleonly = array( # Elements that cannot have close tags
+ 'br', 'hr'
+ );
$htmlnest = array( # Tags that can be nested--??
'table', 'tr', 'td', 'th', 'div', 'blockquote', 'ol', 'ul',
'dl', 'font', 'big', 'small', 'sub', 'sup', 'span'
);
- $tabletags = array( # Can only appear inside table
- 'td', 'th', 'tr'
+ $tabletags = array( # Can only appear inside table, we will close them
+ 'td', 'th', 'tr',
);
- } else {
- $htmlpairs = array();
- $htmlsingle = array();
- $htmlnest = array();
- $tabletags = array();
- }
+ $htmllist = array( # Tags used by list
+ 'ul','ol',
+ );
+ $listtags = array( # Tags that can appear in a list
+ 'li',
+ );
+
+ $htmlsingleallowed = array_merge( $htmlsingle, $tabletags );
+ $htmlelements = array_merge( $htmlsingle, $htmlpairs, $htmlnest );
- $htmlsingle = array_merge( $tabletags, $htmlsingle );
- $htmlelements = array_merge( $htmlsingle, $htmlpairs );
+ # Convert them all to hashtables for faster lookup
+ $vars = array( 'htmlpairs', 'htmlsingle', 'htmlsingleonly', 'htmlnest', 'tabletags',
+ 'htmllist', 'listtags', 'htmlsingleallowed', 'htmlelements' );
+ foreach ( $vars as $var ) {
+ $$var = array_flip( $$var );
+ }
+ $staticInitialised = true;
+ }
# Remove HTML comments
$text = Sanitizer::removeHTMLcomments( $text );
-
$bits = explode( '<', $text );
- $text = array_shift( $bits );
+ $text = str_replace( '>', '>', array_shift( $bits ) );
if(!$wgUseTidy) {
- $tagstack = array(); $tablestack = array();
+ $tagstack = $tablestack = array();
foreach ( $bits as $x ) {
- $prev = error_reporting( E_ALL & ~( E_NOTICE | E_WARNING ) );
- preg_match( '/^(\\/?)(\\w+)([^>]*)(\\/{0,1}>)([^<]*)$/',
- $x, $regs );
- list( $qbar, $slash, $t, $params, $brace, $rest ) = $regs;
- error_reporting( $prev );
+ $regs = array();
+ if( preg_match( '!^(/?)(\\w+)([^>]*?)(/{0,1}>)([^<]*)$!', $x, $regs ) ) {
+ list( /* $qbar */, $slash, $t, $params, $brace, $rest ) = $regs;
+ } else {
+ $slash = $t = $params = $brace = $rest = null;
+ }
$badtag = 0 ;
- if ( in_array( $t = strtolower( $t ), $htmlelements ) ) {
+ if ( isset( $htmlelements[$t = strtolower( $t )] ) ) {
# Check our stack
if ( $slash ) {
# Closing a tag...
- if ( ! in_array( $t, $htmlsingle ) &&
- ( $ot = @array_pop( $tagstack ) ) != $t ) {
- @array_push( $tagstack, $ot );
+ if( isset( $htmlsingleonly[$t] ) ) {
$badtag = 1;
+ } elseif ( ( $ot = @array_pop( $tagstack ) ) != $t ) {
+ if ( isset( $htmlsingleallowed[$ot] ) ) {
+ # Pop all elements with an optional close tag
+ # and see if we find a match below them
+ $optstack = array();
+ array_push ($optstack, $ot);
+ while ( ( ( $ot = @array_pop( $tagstack ) ) != $t ) &&
+ isset( $htmlsingleallowed[$ot] ) )
+ {
+ array_push ($optstack, $ot);
+ }
+ if ( $t != $ot ) {
+ # No match. Push the optinal elements back again
+ $badtag = 1;
+ while ( $ot = @array_pop( $optstack ) ) {
+ array_push( $tagstack, $ot );
+ }
+ }
+ } else {
+ @array_push( $tagstack, $ot );
+ # <li> can be nested in <ul> or <ol>, skip those cases:
+ if(!(isset( $htmllist[$ot] ) && isset( $listtags[$t] ) )) {
+ $badtag = 1;
+ }
+ }
} else {
if ( $t == 'table' ) {
$tagstack = array_pop( $tablestack );
}
- $newparams = '';
}
+ $newparams = '';
} else {
# Keep track for later
- if ( in_array( $t, $tabletags ) &&
+ if ( isset( $tabletags[$t] ) &&
! in_array( 'table', $tagstack ) ) {
$badtag = 1;
} else if ( in_array( $t, $tagstack ) &&
- ! in_array ( $t , $htmlnest ) ) {
+ ! isset( $htmlnest [$t ] ) ) {
$badtag = 1 ;
- } else if ( ! in_array( $t, $htmlsingle ) ) {
+ # Is it a self closed htmlpair ? (bug 5487)
+ } else if( $brace == '/>' &&
+ isset( $htmlpairs[$t] ) ) {
+ $badtag = 1;
+ } elseif( isset( $htmlsingleonly[$t] ) ) {
+ # Hack to force empty tag for uncloseable elements
+ $brace = '/>';
+ } else if( isset( $htmlsingle[$t] ) ) {
+ # Hack to not close $htmlsingle tags
+ $brace = NULL;
+ } else if( isset( $tabletags[$t] )
+ && in_array($t ,$tagstack) ) {
+ // New table tag but forgot to close the previous one
+ $text .= "</$t>";
+ } else {
if ( $t == 'table' ) {
array_push( $tablestack, $tagstack );
$tagstack = array();
}
array_push( $tagstack, $t );
}
+
+ # Replace any variables or template parameters with
+ # plaintext results.
+ if( is_callable( $processCallback ) ) {
+ call_user_func_array( $processCallback, array( &$params, $args ) );
+ }
+
# Strip non-approved attributes from the tag
$newparams = Sanitizer::fixTagAttributes( $params, $t );
}
if ( ! $badtag ) {
$rest = str_replace( '>', '>', $rest );
- $text .= "<$slash$t$newparams$brace$rest";
+ $close = ( $brace == '/>' && !$slash ) ? ' /' : '';
+ $text .= "<$slash$t$newparams$close>$rest";
continue;
}
}
} else {
# this might be possible using tidy itself
foreach ( $bits as $x ) {
- preg_match( '/^(\\/?)(\\w+)([^>]*)(\\/{0,1}>)([^<]*)$/',
+ preg_match( '/^(\\/?)(\\w+)([^>]*?)(\\/{0,1}>)([^<]*)$/',
$x, $regs );
- @list( $qbar, $slash, $t, $params, $brace, $rest ) = $regs;
- if ( in_array( $t = strtolower( $t ), $htmlelements ) ) {
+ @list( /* $qbar */, $slash, $t, $params, $brace, $rest ) = $regs;
+ if ( isset( $htmlelements[$t = strtolower( $t )] ) ) {
+ if( is_callable( $processCallback ) ) {
+ call_user_func_array( $processCallback, array( &$params, $args ) );
+ }
$newparams = Sanitizer::fixTagAttributes( $params, $t );
$rest = str_replace( '>', '>', $rest );
$text .= "<$slash$t$newparams$brace$rest";
}
}
}
- wfProfileOut( $fname );
+ wfProfileOut( __METHOD__ );
return $text;
}
* To avoid leaving blank lines, when a comment is both preceded
* and followed by a newline (ignoring spaces), trim leading and
* trailing spaces and one of the newlines.
- *
- * @access private
+ *
+ * @private
+ * @param string $text
+ * @return string
*/
- function removeHTMLcomments( $text ) {
- $fname='Parser::removeHTMLcomments';
- wfProfileIn( $fname );
+ static function removeHTMLcomments( $text ) {
+ wfProfileIn( __METHOD__ );
while (($start = strpos($text, '<!--')) !== false) {
$end = strpos($text, '-->', $start + 4);
if ($end === false) {
$text = substr_replace($text, '', $start, $end - $start);
}
}
- wfProfileOut( $fname );
+ wfProfileOut( __METHOD__ );
return $text;
}
+ /**
+ * Take an array of attribute names and values and normalize or discard
+ * illegal values for the given element type.
+ *
+ * - Discards attributes not on a whitelist for the given element
+ * - Unsafe style attributes are discarded
+ * - Invalid id attributes are reencoded
+ *
+ * @param array $attribs
+ * @param string $element
+ * @return array
+ *
+ * @todo Check for legal values where the DTD limits things.
+ * @todo Check for unique id attribute :P
+ */
+ static function validateTagAttributes( $attribs, $element ) {
+ return Sanitizer::validateAttributes( $attribs,
+ Sanitizer::attributeWhitelist( $element ) );
+ }
+
+ /**
+ * Take an array of attribute names and values and normalize or discard
+ * illegal values for the given whitelist.
+ *
+ * - Discards attributes not the given whitelist
+ * - Unsafe style attributes are discarded
+ * - Invalid id attributes are reencoded
+ *
+ * @param array $attribs
+ * @param array $whitelist list of allowed attribute names
+ * @return array
+ *
+ * @todo Check for legal values where the DTD limits things.
+ * @todo Check for unique id attribute :P
+ */
+ static function validateAttributes( $attribs, $whitelist ) {
+ $whitelist = array_flip( $whitelist );
+ $out = array();
+ foreach( $attribs as $attribute => $value ) {
+ if( !isset( $whitelist[$attribute] ) ) {
+ continue;
+ }
+ # Strip javascript "expression" from stylesheets.
+ # http://msdn.microsoft.com/workshop/author/dhtml/overview/recalc.asp
+ if( $attribute == 'style' ) {
+ $value = Sanitizer::checkCss( $value );
+ if( $value === false ) {
+ # haxx0r
+ continue;
+ }
+ }
+
+ if ( $attribute === 'id' ) {
+ global $wgEnforceHtmlIds;
+ $value = Sanitizer::escapeId( $value,
+ $wgEnforceHtmlIds ? array() : 'xml' );
+ }
+
+ // If this attribute was previously set, override it.
+ // Output should only have one attribute of each name.
+ $out[$attribute] = $value;
+ }
+ return $out;
+ }
+
+ /**
+ * Merge two sets of HTML attributes. Conflicting items in the second set
+ * will override those in the first, except for 'class' attributes which
+ * will be combined (if they're both strings).
+ *
+ * @todo implement merging for other attributes such as style
+ * @param array $a
+ * @param array $b
+ * @return array
+ */
+ static function mergeAttributes( $a, $b ) {
+ $out = array_merge( $a, $b );
+ if( isset( $a['class'] ) && isset( $b['class'] )
+ && is_string( $a['class'] ) && is_string( $b['class'] )
+ && $a['class'] !== $b['class'] ) {
+ $classes = preg_split( '/\s+/', "{$a['class']} {$b['class']}",
+ -1, PREG_SPLIT_NO_EMPTY );
+ $out['class'] = implode( ' ', array_unique( $classes ) );
+ }
+ return $out;
+ }
+
+ /**
+ * Pick apart some CSS and check it for forbidden or unsafe structures.
+ * Returns a sanitized string, or false if it was just too evil.
+ *
+ * Currently URL references, 'expression', 'tps' are forbidden.
+ *
+ * @param string $value
+ * @return mixed
+ */
+ static function checkCss( $value ) {
+ $stripped = Sanitizer::decodeCharReferences( $value );
+
+ // Remove any comments; IE gets token splitting wrong
+ $stripped = StringUtils::delimiterReplace( '/*', '*/', ' ', $stripped );
+
+ $value = $stripped;
+
+ // ... and continue checks
+ $stripped = preg_replace( '!\\\\([0-9A-Fa-f]{1,6})[ \\n\\r\\t\\f]?!e',
+ 'codepointToUtf8(hexdec("$1"))', $stripped );
+ $stripped = str_replace( '\\', '', $stripped );
+ if( preg_match( '/(?:expression|tps*:\/\/|url\\s*\().*/is',
+ $stripped ) ) {
+ # haxx0r
+ return false;
+ }
+
+ return $value;
+ }
+
/**
* Take a tag soup fragment listing an HTML element's attributes
* and normalize it to well-formed XML, discarding unwanted attributes.
+ * Output is safe for further wikitext processing, with escaping of
+ * values that could trigger problems.
*
* - Normalizes attribute names to lowercase
* - Discards attributes not on a whitelist for the given element
* @param string $text
* @param string $element
* @return string
- *
- * @todo Check for legal values where the DTD limits things.
- * @todo Check for unique id attribute :P
*/
- function fixTagAttributes( $text, $element ) {
+ static function fixTagAttributes( $text, $element ) {
if( trim( $text ) == '' ) {
return '';
}
-
- $attrib = '[A-Za-z0-9]'; #FIXME
- $space = '[\x09\x0a\x0d\x20]';
+
+ $stripped = Sanitizer::validateTagAttributes(
+ Sanitizer::decodeTagAttributes( $text ), $element );
+
+ $attribs = array();
+ foreach( $stripped as $attribute => $value ) {
+ $encAttribute = htmlspecialchars( $attribute );
+ $encValue = Sanitizer::safeEncodeAttribute( $value );
+
+ $attribs[] = "$encAttribute=\"$encValue\"";
+ }
+ return count( $attribs ) ? ' ' . implode( ' ', $attribs ) : '';
+ }
+
+ /**
+ * Encode an attribute value for HTML output.
+ * @param $text
+ * @return HTML-encoded text fragment
+ */
+ static function encodeAttribute( $text ) {
+ $encValue = htmlspecialchars( $text, ENT_QUOTES );
+
+ // Whitespace is normalized during attribute decoding,
+ // so if we've been passed non-spaces we must encode them
+ // ahead of time or they won't be preserved.
+ $encValue = strtr( $encValue, array(
+ "\n" => ' ',
+ "\r" => ' ',
+ "\t" => '	',
+ ) );
+
+ return $encValue;
+ }
+
+ /**
+ * Encode an attribute value for HTML tags, with extra armoring
+ * against further wiki processing.
+ * @param $text
+ * @return HTML-encoded text fragment
+ */
+ static function safeEncodeAttribute( $text ) {
+ $encValue = Sanitizer::encodeAttribute( $text );
+
+ # Templates and links may be expanded in later parsing,
+ # creating invalid or dangerous output. Suppress this.
+ $encValue = strtr( $encValue, array(
+ '<' => '<', // This should never happen,
+ '>' => '>', // we've received invalid input
+ '"' => '"', // which should have been escaped.
+ '{' => '{',
+ '[' => '[',
+ "''" => '''',
+ 'ISBN' => 'ISBN',
+ 'RFC' => 'RFC',
+ 'PMID' => 'PMID',
+ '|' => '|',
+ '__' => '__',
+ ) );
+
+ # Stupid hack
+ $encValue = preg_replace_callback(
+ '/(' . wfUrlProtocols() . ')/',
+ array( 'Sanitizer', 'armorLinksCallback' ),
+ $encValue );
+ return $encValue;
+ }
+
+ /**
+ * Given a value escape it so that it can be used in an id attribute and
+ * return it, this does not validate the value however (see first link)
+ *
+ * @see http://www.w3.org/TR/html401/types.html#type-name Valid characters
+ * in the id and
+ * name attributes
+ * @see http://www.w3.org/TR/html401/struct/links.html#h-12.2.3 Anchors with the id attribute
+ *
+ * @param string $id Id to validate
+ * @param mixed $options String or array of strings (default is array()):
+ * 'noninitial': This is a non-initial fragment of an id, not a full id,
+ * so don't pay attention if the first character isn't valid at the
+ * beginning of an id.
+ * 'xml': Don't restrict the id to be HTML4-compatible. This option
+ * allows any alphabetic character to be used, per the XML standard.
+ * Therefore, it also completely changes the type of escaping: instead
+ * of weird dot-encoding, runs of invalid characters (mostly
+ * whitespace) are just compressed into a single underscore.
+ * @return string
+ */
+ static function escapeId( $id, $options = array() ) {
+ $options = (array)$options;
+
+ if ( !in_array( 'xml', $options ) ) {
+ # HTML4-style escaping
+ static $replace = array(
+ '%3A' => ':',
+ '%' => '.'
+ );
+
+ $id = urlencode( Sanitizer::decodeCharReferences( strtr( $id, ' ', '_' ) ) );
+ $id = str_replace( array_keys( $replace ), array_values( $replace ), $id );
+
+ /*
+ // Let's not go changing our links just yet, since we plan to
+ // rip this out anyway and switch to the prettier Unicode-friendly
+ // links as XHTML and HTML 5 allow and browsers seem to think is
+ // just a-ok to do.
+ //
+ if ( !preg_match( '/^[a-zA-Z]/', $id )
+ && !in_array( 'noninitial', $options ) ) {
+ // Initial character must be a letter!
+ $id = "x$id";
+ }
+ */
+ return $id;
+ }
+
+ # XML-style escaping. For the patterns used, see the XML 1.0 standard,
+ # 5th edition, NameStartChar and NameChar: <http://www.w3.org/TR/REC-xml/>
+ $nameStartChar = ':a-zA-Z_\xC0-\xD6\xD8-\xF6\xF8-\x{2FF}\x{370}-\x{37D}'
+ . '\x{37F}-\x{1FFF}\x{200C}-\x{200D}\x{2070}-\x{218F}\x{2C00}-\x{2FEF}'
+ . '\x{3001}-\x{D7FF}\x{F900}-\x{FDCF}\x{FDF0}-\x{FFFD}\x{10000}-\x{EFFFF}';
+ $nameChar = $nameStartChar . '.\-0-9\xB7\x{0300}-\x{036F}'
+ . '\x{203F}-\x{2040}';
+ # Replace _ as well so we don't get multiple consecutive underscores
+ $id = preg_replace( "/([^$nameChar]|_)+/u", '_', $id );
+ $id = trim( $id, '_' );
+
+ if ( !preg_match( "/^[$nameStartChar]/u", $id )
+ && !in_array( 'noninitial', $options ) ) {
+ $id = "_$id";
+ }
+
+ return $id;
+ }
+
+ /**
+ * Given a value, escape it so that it can be used as a CSS class and
+ * return it.
+ *
+ * @todo For extra validity, input should be validated UTF-8.
+ *
+ * @see http://www.w3.org/TR/CSS21/syndata.html Valid characters/format
+ *
+ * @param string $class
+ * @return string
+ */
+ static function escapeClass( $class ) {
+ // Convert ugly stuff to underscores and kill underscores in ugly places
+ return rtrim(preg_replace(
+ array('/(^[0-9\\-])|[\\x00-\\x20!"#$%&\'()*+,.\\/:;<=>?@[\\]^`{|}~]|\\xC2\\xA0/','/_+/'),
+ '_',
+ $class ), '_');
+ }
+
+ /**
+ * Given HTML input, escape with htmlspecialchars but un-escape entites.
+ * This allows (generally harmless) entities like to survive.
+ *
+ * @param string $html String to escape
+ * @return string Escaped input
+ */
+ static function escapeHtmlAllowEntities( $html ) {
+ # It seems wise to escape ' as well as ", as a matter of course. Can't
+ # hurt.
+ $html = htmlspecialchars( $html, ENT_QUOTES );
+ $html = str_replace( '&', '&', $html );
+ $html = Sanitizer::normalizeCharReferences( $html );
+ return $html;
+ }
+
+ /**
+ * Regex replace callback for armoring links against further processing.
+ * @param array $matches
+ * @return string
+ * @private
+ */
+ private static function armorLinksCallback( $matches ) {
+ return str_replace( ':', ':', $matches[1] );
+ }
+
+ /**
+ * Return an associative array of attribute names and values from
+ * a partial tag string. Attribute names are forces to lowercase,
+ * character references are decoded to UTF-8 text.
+ *
+ * @param string
+ * @return array
+ */
+ public static function decodeTagAttributes( $text ) {
+ $attribs = array();
+
+ if( trim( $text ) == '' ) {
+ return $attribs;
+ }
+
+ $pairs = array();
if( !preg_match_all(
- "/(?:^|$space)($attrib+)
- ($space*=$space*
- (?:
- # The attribute value: quoted or alone
- \"([^<\"]*)\"
- | '([^<']*)'
- | ([a-zA-Z0-9._:-]+)
- | (\#[0-9a-fA-F]+) # Technically wrong, but lots of
- # colors are specified like this.
- # We'll be normalizing it.
- )
- )?(?=$space|\$)/sx",
+ MW_ATTRIBS_REGEX,
$text,
$pairs,
PREG_SET_ORDER ) ) {
- return '';
+ return $attribs;
}
- $whitelist = array_flip( Sanitizer::attributeWhitelist( $element ) );
- $attribs = array();
foreach( $pairs as $set ) {
$attribute = strtolower( $set[1] );
- if( !isset( $whitelist[$attribute] ) ) {
- continue;
- }
- if( $set[2] == '' ) {
- # In XHTML, attributes must have a value.
- $value = $set[1];
- } elseif( $set[3] != '' ) {
- # Double-quoted
- $value = Sanitizer::normalizeAttributeValue( $set[3] );
- } elseif( $set[4] != '' ) {
- # Single-quoted
- $value = str_replace( '"', '"',
- Sanitizer::normalizeAttributeValue( $set[4] ) );
- } elseif( $set[5] != '' ) {
- # No quotes.
- $value = Sanitizer::normalizeAttributeValue( $set[5] );
- } elseif( $set[6] != '' ) {
- # Illegal #XXXXXX color with no quotes.
- $value = Sanitizer::normalizeAttributeValue( $set[6] );
- } else {
- wfDebugDieBacktrace( "Tag conditions not met. Something's very odd." );
- }
-
- # Strip javascript "expression" from stylesheets.
- # http://msdn.microsoft.com/workshop/author/dhtml/overview/recalc.asp
- if( $attribute == 'style' && preg_match(
- '/(expression|tps*:\/\/|url\\s*\().*/is',
- wfMungeToUtf8( $value ) ) ) {
- # haxx0r
- continue;
- }
-
- if( !isset( $attribs[$attribute] ) ) {
- $attribs[$attribute] = "$attribute=\"$value\"";
- }
+ $value = Sanitizer::getTagAttributeCallback( $set );
+
+ // Normalize whitespace
+ $value = preg_replace( '/[\t\r\n ]+/', ' ', $value );
+ $value = trim( $value );
+
+ // Decode character references
+ $attribs[$attribute] = Sanitizer::decodeCharReferences( $value );
}
- if( empty( $attribs ) ) {
- return '';
+ return $attribs;
+ }
+
+ /**
+ * Pick the appropriate attribute value from a match set from the
+ * MW_ATTRIBS_REGEX matches.
+ *
+ * @param array $set
+ * @return string
+ * @private
+ */
+ private static function getTagAttributeCallback( $set ) {
+ if( isset( $set[6] ) ) {
+ # Illegal #XXXXXX color with no quotes.
+ return $set[6];
+ } elseif( isset( $set[5] ) ) {
+ # No quotes.
+ return $set[5];
+ } elseif( isset( $set[4] ) ) {
+ # Single-quoted
+ return $set[4];
+ } elseif( isset( $set[3] ) ) {
+ # Double-quoted
+ return $set[3];
+ } elseif( !isset( $set[2] ) ) {
+ # In XHTML, attributes must have a value.
+ # For 'reduced' form, return explicitly the attribute name here.
+ return $set[1];
} else {
- return ' ' . implode( ' ', $attribs );
+ throw new MWException( "Tag conditions not met. This should never happen and is a bug." );
}
}
-
+
/**
* Normalize whitespace and character references in an XML source-
* encoded text for an attribute value.
*
* @param string $text
* @return string
- * @access private
+ * @private
*/
- function normalizeAttributeValue( $text ) {
+ private static function normalizeAttributeValue( $text ) {
+ return str_replace( '"', '"',
+ self::normalizeWhitespace(
+ Sanitizer::normalizeCharReferences( $text ) ) );
+ }
+
+ private static function normalizeWhitespace( $text ) {
return preg_replace(
'/\r\n|[\x20\x0d\x0a\x09]/',
' ',
- Sanitizer::normalizeCharReferences( $text ) );
- return $spaced;
+ $text );
}
-
+
/**
* Ensure that any entities and character references are legal
* for XML and XHTML specifically. Any stray bits will be
*
* @param string $text
* @return string
- * @access private
+ * @private
*/
- function normalizeCharReferences( $text ) {
+ static function normalizeCharReferences( $text ) {
return preg_replace_callback(
- '/&([A-Za-z0-9]+);
- |&\#([0-9]+);
- |&\#x([0-9A-Za-z]+);
- |&\#X([0-9A-Za-z]+);
- |(&)/x',
+ MW_CHAR_REFS_REGEX,
array( 'Sanitizer', 'normalizeCharReferencesCallback' ),
$text );
}
-
- function normalizeCharReferencesCallback( $matches ) {
+ /**
+ * @param string $matches
+ * @return string
+ */
+ static function normalizeCharReferencesCallback( $matches ) {
$ret = null;
if( $matches[1] != '' ) {
$ret = Sanitizer::normalizeEntity( $matches[1] );
return $ret;
}
}
-
+
/**
* If the named entity is defined in the HTML 4.0/XHTML 1.0 DTD,
- * return the named entity reference as is. Otherwise, returns
- * HTML-escaped text of pseudo-entity source (eg &foo;)
+ * return the named entity reference as is. If the entity is a
+ * MediaWiki-specific alias, returns the HTML equivalent. Otherwise,
+ * returns HTML-escaped text of pseudo-entity source (eg &foo;)
*
+ * @param string $name
* @return string
+ * @static
*/
- function normalizeEntity( $name ) {
- # List of all named character entities defined in HTML 4.01
- # http://www.w3.org/TR/html4/sgml/entities.html
- static $htmlEntities = array(
- 'aacute' => true,
- 'Aacute' => true,
- 'acirc' => true,
- 'Acirc' => true,
- 'acute' => true,
- 'aelig' => true,
- 'AElig' => true,
- 'agrave' => true,
- 'Agrave' => true,
- 'alefsym' => true,
- 'alpha' => true,
- 'Alpha' => true,
- 'amp' => true,
- 'and' => true,
- 'ang' => true,
- 'apos' => true,
- 'aring' => true,
- 'Aring' => true,
- 'asymp' => true,
- 'atilde' => true,
- 'Atilde' => true,
- 'auml' => true,
- 'Auml' => true,
- 'bdquo' => true,
- 'beta' => true,
- 'Beta' => true,
- 'brvbar' => true,
- 'bull' => true,
- 'cap' => true,
- 'ccedil' => true,
- 'Ccedil' => true,
- 'cedil' => true,
- 'cent' => true,
- 'chi' => true,
- 'Chi' => true,
- 'circ' => true,
- 'clubs' => true,
- 'cong' => true,
- 'copy' => true,
- 'crarr' => true,
- 'cup' => true,
- 'curren' => true,
- 'dagger' => true,
- 'Dagger' => true,
- 'darr' => true,
- 'dArr' => true,
- 'deg' => true,
- 'delta' => true,
- 'Delta' => true,
- 'diams' => true,
- 'divide' => true,
- 'eacute' => true,
- 'Eacute' => true,
- 'ecirc' => true,
- 'Ecirc' => true,
- 'egrave' => true,
- 'Egrave' => true,
- 'empty' => true,
- 'emsp' => true,
- 'ensp' => true,
- 'epsilon' => true,
- 'Epsilon' => true,
- 'equiv' => true,
- 'eta' => true,
- 'Eta' => true,
- 'eth' => true,
- 'ETH' => true,
- 'euml' => true,
- 'Euml' => true,
- 'euro' => true,
- 'exist' => true,
- 'fnof' => true,
- 'forall' => true,
- 'frac12' => true,
- 'frac14' => true,
- 'frac34' => true,
- 'frasl' => true,
- 'gamma' => true,
- 'Gamma' => true,
- 'ge' => true,
- 'gt' => true,
- 'harr' => true,
- 'hArr' => true,
- 'hearts' => true,
- 'hellip' => true,
- 'iacute' => true,
- 'Iacute' => true,
- 'icirc' => true,
- 'Icirc' => true,
- 'iexcl' => true,
- 'igrave' => true,
- 'Igrave' => true,
- 'image' => true,
- 'infin' => true,
- 'int' => true,
- 'iota' => true,
- 'Iota' => true,
- 'iquest' => true,
- 'isin' => true,
- 'iuml' => true,
- 'Iuml' => true,
- 'kappa' => true,
- 'Kappa' => true,
- 'lambda' => true,
- 'Lambda' => true,
- 'lang' => true,
- 'laquo' => true,
- 'larr' => true,
- 'lArr' => true,
- 'lceil' => true,
- 'ldquo' => true,
- 'le' => true,
- 'lfloor' => true,
- 'lowast' => true,
- 'loz' => true,
- 'lrm' => true,
- 'lsaquo' => true,
- 'lsquo' => true,
- 'lt' => true,
- 'macr' => true,
- 'mdash' => true,
- 'micro' => true,
- 'middot' => true,
- 'minus' => true,
- 'mu' => true,
- 'Mu' => true,
- 'nabla' => true,
- 'nbsp' => true,
- 'ndash' => true,
- 'ne' => true,
- 'ni' => true,
- 'not' => true,
- 'notin' => true,
- 'nsub' => true,
- 'ntilde' => true,
- 'Ntilde' => true,
- 'nu' => true,
- 'Nu' => true,
- 'oacute' => true,
- 'Oacute' => true,
- 'ocirc' => true,
- 'Ocirc' => true,
- 'oelig' => true,
- 'OElig' => true,
- 'ograve' => true,
- 'Ograve' => true,
- 'oline' => true,
- 'omega' => true,
- 'Omega' => true,
- 'omicron' => true,
- 'Omicron' => true,
- 'oplus' => true,
- 'or' => true,
- 'ordf' => true,
- 'ordm' => true,
- 'oslash' => true,
- 'Oslash' => true,
- 'otilde' => true,
- 'Otilde' => true,
- 'otimes' => true,
- 'ouml' => true,
- 'Ouml' => true,
- 'para' => true,
- 'part' => true,
- 'permil' => true,
- 'perp' => true,
- 'phi' => true,
- 'Phi' => true,
- 'pi' => true,
- 'Pi' => true,
- 'piv' => true,
- 'plusmn' => true,
- 'pound' => true,
- 'prime' => true,
- 'Prime' => true,
- 'prod' => true,
- 'prop' => true,
- 'psi' => true,
- 'Psi' => true,
- 'quot' => true,
- 'radic' => true,
- 'rang' => true,
- 'raquo' => true,
- 'rarr' => true,
- 'rArr' => true,
- 'rceil' => true,
- 'rdquo' => true,
- 'real' => true,
- 'reg' => true,
- 'rfloor' => true,
- 'rho' => true,
- 'Rho' => true,
- 'rlm' => true,
- 'rsaquo' => true,
- 'rsquo' => true,
- 'sbquo' => true,
- 'scaron' => true,
- 'Scaron' => true,
- 'sdot' => true,
- 'sect' => true,
- 'shy' => true,
- 'sigma' => true,
- 'Sigma' => true,
- 'sigmaf' => true,
- 'sim' => true,
- 'spades' => true,
- 'sub' => true,
- 'sube' => true,
- 'sum' => true,
- 'sup' => true,
- 'sup1' => true,
- 'sup2' => true,
- 'sup3' => true,
- 'supe' => true,
- 'szlig' => true,
- 'tau' => true,
- 'Tau' => true,
- 'there4' => true,
- 'theta' => true,
- 'Theta' => true,
- 'thetasym' => true,
- 'thinsp' => true,
- 'thorn' => true,
- 'THORN' => true,
- 'tilde' => true,
- 'times' => true,
- 'trade' => true,
- 'uacute' => true,
- 'Uacute' => true,
- 'uarr' => true,
- 'uArr' => true,
- 'ucirc' => true,
- 'Ucirc' => true,
- 'ugrave' => true,
- 'Ugrave' => true,
- 'uml' => true,
- 'upsih' => true,
- 'upsilon' => true,
- 'Upsilon' => true,
- 'uuml' => true,
- 'Uuml' => true,
- 'weierp' => true,
- 'xi' => true,
- 'Xi' => true,
- 'yacute' => true,
- 'Yacute' => true,
- 'yen' => true,
- 'yuml' => true,
- 'Yuml' => true,
- 'zeta' => true,
- 'Zeta' => true,
- 'zwj' => true,
- 'zwnj' => true );
- if( isset( $htmlEntities[$name] ) ) {
+ static function normalizeEntity( $name ) {
+ global $wgHtmlEntities, $wgHtmlEntityAliases;
+ if ( isset( $wgHtmlEntityAliases[$name] ) ) {
+ return "&{$wgHtmlEntityAliases[$name]};";
+ } elseif( isset( $wgHtmlEntities[$name] ) ) {
return "&$name;";
} else {
return "&$name;";
}
}
-
- function decCharReference( $codepoint ) {
- $point = IntVal( $codepoint );
+
+ static function decCharReference( $codepoint ) {
+ $point = intval( $codepoint );
if( Sanitizer::validateCodepoint( $point ) ) {
return sprintf( '&#%d;', $point );
} else {
return null;
}
}
-
- function hexCharReference( $codepoint ) {
+
+ static function hexCharReference( $codepoint ) {
$point = hexdec( $codepoint );
if( Sanitizer::validateCodepoint( $point ) ) {
return sprintf( '&#x%x;', $point );
return null;
}
}
-
+
/**
* Returns true if a given Unicode codepoint is a valid character in XML.
* @param int $codepoint
* @return bool
*/
- function validateCodepoint( $codepoint ) {
+ private static function validateCodepoint( $codepoint ) {
return ($codepoint == 0x09)
|| ($codepoint == 0x0a)
|| ($codepoint == 0x0d)
|| ($codepoint >= 0x10000 && $codepoint <= 0x10ffff);
}
+ /**
+ * Decode any character references, numeric or named entities,
+ * in the text and return a UTF-8 string.
+ *
+ * @param string $text
+ * @return string
+ * @public
+ * @static
+ */
+ public static function decodeCharReferences( $text ) {
+ return preg_replace_callback(
+ MW_CHAR_REFS_REGEX,
+ array( 'Sanitizer', 'decodeCharReferencesCallback' ),
+ $text );
+ }
+
+ /**
+ * @param string $matches
+ * @return string
+ */
+ static function decodeCharReferencesCallback( $matches ) {
+ if( $matches[1] != '' ) {
+ return Sanitizer::decodeEntity( $matches[1] );
+ } elseif( $matches[2] != '' ) {
+ return Sanitizer::decodeChar( intval( $matches[2] ) );
+ } elseif( $matches[3] != '' ) {
+ return Sanitizer::decodeChar( hexdec( $matches[3] ) );
+ } elseif( $matches[4] != '' ) {
+ return Sanitizer::decodeChar( hexdec( $matches[4] ) );
+ }
+ # Last case should be an ampersand by itself
+ return $matches[0];
+ }
+
+ /**
+ * Return UTF-8 string for a codepoint if that is a valid
+ * character reference, otherwise U+FFFD REPLACEMENT CHARACTER.
+ * @param int $codepoint
+ * @return string
+ * @private
+ */
+ static function decodeChar( $codepoint ) {
+ if( Sanitizer::validateCodepoint( $codepoint ) ) {
+ return codepointToUtf8( $codepoint );
+ } else {
+ return UTF8_REPLACEMENT;
+ }
+ }
+
+ /**
+ * If the named entity is defined in the HTML 4.0/XHTML 1.0 DTD,
+ * return the UTF-8 encoding of that character. Otherwise, returns
+ * pseudo-entity source (eg &foo;)
+ *
+ * @param string $name
+ * @return string
+ */
+ static function decodeEntity( $name ) {
+ global $wgHtmlEntities, $wgHtmlEntityAliases;
+ if ( isset( $wgHtmlEntityAliases[$name] ) ) {
+ $name = $wgHtmlEntityAliases[$name];
+ }
+ if( isset( $wgHtmlEntities[$name] ) ) {
+ return codepointToUtf8( $wgHtmlEntities[$name] );
+ } else {
+ return "&$name;";
+ }
+ }
+
/**
* Fetch the whitelist of acceptable attributes for a given
* element name.
* @param string $element
* @return array
*/
- function attributeWhitelist( $element ) {
- $list = Sanitizer::setupAttributeWhitelist();
+ static function attributeWhitelist( $element ) {
+ static $list;
+ if( !isset( $list ) ) {
+ $list = Sanitizer::setupAttributeWhitelist();
+ }
return isset( $list[$element] )
? $list[$element]
: array();
}
-
+
/**
+ * Foreach array key (an allowed HTML element), return an array
+ * of allowed attributes
* @return array
*/
- function setupAttributeWhitelist() {
+ static function setupAttributeWhitelist() {
$common = array( 'id', 'class', 'lang', 'dir', 'title', 'style' );
$block = array_merge( $common, array( 'align' ) );
$tablealign = array( 'align', 'char', 'charoff', 'valign' );
'rowspan',
'colspan',
'nowrap', # deprecated
- 'width', # deprecated
- 'height' # deprecated
+ 'width', # deprecated
+ 'height', # deprecated
+ 'bgcolor' # deprecated
);
-
+
# Numbers refer to sections in HTML 4.01 standard describing the element.
# See: http://www.w3.org/TR/html4/
$whitelist = array (
'div' => $block,
'center' => $common, # deprecated
'span' => $block, # ??
-
+
# 7.5.5
'h1' => $block,
'h2' => $block,
'h4' => $block,
'h5' => $block,
'h6' => $block,
-
+
# 7.5.6
# address
-
+
# 8.2.4
# bdo
-
+
# 9.2.1
'em' => $common,
'strong' => $common,
'var' => $common,
# abbr
# acronym
-
+
# 9.2.2
'blockquote' => array_merge( $common, array( 'cite' ) ),
# q
-
+
# 9.2.3
'sub' => $common,
'sup' => $common,
-
+
# 9.3.1
'p' => $block,
-
+
# 9.3.2
'br' => array( 'id', 'class', 'title', 'style', 'clear' ),
-
+
# 9.3.4
'pre' => array_merge( $common, array( 'width' ) ),
-
+
# 9.4
'ins' => array_merge( $common, array( 'cite', 'datetime' ) ),
'del' => array_merge( $common, array( 'cite', 'datetime' ) ),
-
+
# 10.2
'ul' => array_merge( $common, array( 'type' ) ),
'ol' => array_merge( $common, array( 'type', 'start' ) ),
'li' => array_merge( $common, array( 'type', 'value' ) ),
-
+
# 10.3
'dl' => $common,
'dd' => $common,
'dt' => $common,
-
+
# 11.2.1
'table' => array_merge( $common,
array( 'summary', 'width', 'border', 'frame',
- 'rules', 'cellspacing', 'cellpadding',
- 'align', 'bgcolor', 'frame', 'rules',
- 'border' ) ),
-
+ 'rules', 'cellspacing', 'cellpadding',
+ 'align', 'bgcolor',
+ ) ),
+
# 11.2.2
'caption' => array_merge( $common, array( 'align' ) ),
-
+
# 11.2.3
'thead' => array_merge( $common, $tablealign ),
'tfoot' => array_merge( $common, $tablealign ),
'tbody' => array_merge( $common, $tablealign ),
-
+
# 11.2.4
'colgroup' => array_merge( $common, array( 'span', 'width' ), $tablealign ),
'col' => array_merge( $common, array( 'span', 'width' ), $tablealign ),
-
+
# 11.2.5
'tr' => array_merge( $common, array( 'bgcolor' ), $tablealign ),
-
+
# 11.2.6
'td' => array_merge( $common, $tablecell, $tablealign ),
'th' => array_merge( $common, $tablecell, $tablealign ),
-
+
+ # 13.2
+ # Not usually allowed, but may be used for extension-style hooks
+ # such as <math> when it is rasterized
+ 'img' => array_merge( $common, array( 'alt' ) ),
+
# 15.2.1
'tt' => $common,
'b' => $common,
'strike' => $common,
's' => $common,
'u' => $common,
-
+
# 15.2.2
'font' => array_merge( $common, array( 'size', 'color', 'face' ) ),
# basefont
-
+
# 15.3
'hr' => array_merge( $common, array( 'noshade', 'size', 'width' ) ),
-
+
+ # XHTML Ruby annotation text module, simple ruby only.
+ # http://www.w3c.org/TR/ruby/
+ 'ruby' => $common,
+ # rbc
+ # rtc
'rb' => $common,
+ 'rt' => $common, #array_merge( $common, array( 'rbspan' ) ),
'rp' => $common,
- 'rt' => $common,
- 'ruby' => $common,
+
+ # MathML root element, where used for extensions
+ # 'title' may not be 100% valid here; it's XHTML
+ # http://www.w3.org/TR/REC-MathML/
+ 'math' => array( 'class', 'style', 'id', 'title' ),
);
return $whitelist;
}
-}
+ /**
+ * Take a fragment of (potentially invalid) HTML and return
+ * a version with any tags removed, encoded as plain text.
+ *
+ * Warning: this return value must be further escaped for literal
+ * inclusion in HTML output as of 1.10!
+ *
+ * @param string $text HTML fragment
+ * @return string
+ */
+ static function stripAllTags( $text ) {
+ # Actual <tags>
+ $text = StringUtils::delimiterReplace( '<', '>', '', $text );
+
+ # Normalize &entities and whitespace
+ $text = self::decodeCharReferences( $text );
+ $text = self::normalizeWhitespace( $text );
+
+ return $text;
+ }
+
+ /**
+ * Hack up a private DOCTYPE with HTML's standard entity declarations.
+ * PHP 4 seemed to know these if you gave it an HTML doctype, but
+ * PHP 5.1 doesn't.
+ *
+ * Use for passing XHTML fragments to PHP's XML parsing functions
+ *
+ * @return string
+ * @static
+ */
+ static function hackDocType() {
+ global $wgHtmlEntities;
+ $out = "<!DOCTYPE html [\n";
+ foreach( $wgHtmlEntities as $entity => $codepoint ) {
+ $out .= "<!ENTITY $entity \"&#$codepoint;\">";
+ }
+ $out .= "]>\n";
+ return $out;
+ }
+
+ static function cleanUrl( $url ) {
+ # Normalize any HTML entities in input. They will be
+ # re-escaped by makeExternalLink().
+ $url = Sanitizer::decodeCharReferences( $url );
+
+ # Escape any control characters introduced by the above step
+ $url = preg_replace( '/[\][<>"\\x00-\\x20\\x7F]/e', "urlencode('\\0')", $url );
-?>
\ No newline at end of file
+ # Validate hostname portion
+ $matches = array();
+ if( preg_match( '!^([^:]+:)(//[^/]+)?(.*)$!iD', $url, $matches ) ) {
+ list( /* $whole */, $protocol, $host, $rest ) = $matches;
+
+ // Characters that will be ignored in IDNs.
+ // http://tools.ietf.org/html/3454#section-3.1
+ // Strip them before further processing so blacklists and such work.
+ $strip = "/
+ \\s| # general whitespace
+ \xc2\xad| # 00ad SOFT HYPHEN
+ \xe1\xa0\x86| # 1806 MONGOLIAN TODO SOFT HYPHEN
+ \xe2\x80\x8b| # 200b ZERO WIDTH SPACE
+ \xe2\x81\xa0| # 2060 WORD JOINER
+ \xef\xbb\xbf| # feff ZERO WIDTH NO-BREAK SPACE
+ \xcd\x8f| # 034f COMBINING GRAPHEME JOINER
+ \xe1\xa0\x8b| # 180b MONGOLIAN FREE VARIATION SELECTOR ONE
+ \xe1\xa0\x8c| # 180c MONGOLIAN FREE VARIATION SELECTOR TWO
+ \xe1\xa0\x8d| # 180d MONGOLIAN FREE VARIATION SELECTOR THREE
+ \xe2\x80\x8c| # 200c ZERO WIDTH NON-JOINER
+ \xe2\x80\x8d| # 200d ZERO WIDTH JOINER
+ [\xef\xb8\x80-\xef\xb8\x8f] # fe00-fe00f VARIATION SELECTOR-1-16
+ /xuD";
+
+ $host = preg_replace( $strip, '', $host );
+
+ // @fixme: validate hostnames here
+
+ return $protocol . $host . $rest;
+ } else {
+ return $url;
+ }
+ }
+
+}