30 define(
'MW_CHAR_REFS_REGEX',
31 '/&([A-Za-z0-9\x80-\xff]+); 44 define(
'MW_ATTRIBS_REGEX',
45 "/(?:^|$space)($attrib+) 48 # The attribute value: quoted or alone 51 | ([a-zA-Z0-9!#$%&()*,\\-.\\/:;<>?@[\\]^_`{|}~]+) 52 | (\#[0-9a-fA-F]+) # Technically wrong, but lots of 53 # colors are specified like this. 54 # We'll be normalizing it. 56 )?(?=$space|\$)/sx" );
342 static function removeHTMLtags( $text, $processCallback = null, $args = array() ) {
345 static $htmlpairs, $htmlsingle, $htmlsingleonly, $htmlnest, $tabletags,
346 $htmllist, $listtags, $htmlsingleallowed, $htmlelements, $staticInitialised;
348 wfProfileIn( __METHOD__ );
350 if ( !$staticInitialised ) {
352 $htmlpairs = array( # Tags that must be closed
353 'b',
'del',
'i',
'ins',
'u',
'font',
'big',
'small',
'sub',
'sup',
'h1',
354 'h2',
'h3',
'h4',
'h5',
'h6',
'cite',
'code',
'em',
's',
355 'strike',
'strong',
'tt',
'var',
'div',
'center',
356 'blockquote',
'ol',
'ul',
'dl',
'table',
'caption',
'pre',
357 'ruby',
'rt' ,
'rb' ,
'rp',
'p',
'span',
'u' 360 'br',
'hr',
'li',
'dt',
'dd' 362 $htmlsingleonly = array( # Elements that cannot have close tags
365 $htmlnest = array( # Tags that can be nested--??
366 'table',
'tr',
'td',
'th',
'div',
'blockquote',
'ol',
'ul',
367 'dl',
'font',
'big',
'small',
'sub',
'sup',
'span' 369 $tabletags = array( # Can only appear inside table, we will close them
372 $htmllist = array( # Tags used by list
375 $listtags = array( # Tags that can appear in a list
379 $htmlsingleallowed = array_merge( $htmlsingle, $tabletags );
380 $htmlelements = array_merge( $htmlsingle, $htmlpairs, $htmlnest );
382 # Convert them all to hashtables for faster lookup 383 $vars = array(
'htmlpairs',
'htmlsingle',
'htmlsingleonly',
'htmlnest',
'tabletags',
384 'htmllist',
'listtags',
'htmlsingleallowed',
'htmlelements' );
385 foreach ( $vars as $var ) {
386 $$var = array_flip( $$var );
388 $staticInitialised =
true;
391 # Remove HTML comments 393 $bits = explode(
'<', $text );
394 $text = str_replace(
'>',
'>', array_shift( $bits ) );
396 $tagstack = $tablestack = array();
397 foreach ( $bits as $x ) {
399 if( preg_match(
'!^(/?)(\\w+)([^>]*?)(/{0,1}>)([^<]*)$!', $x, $regs ) ) {
400 list( , $slash,
$t, $params, $brace,
$rest ) = $regs;
402 $slash =
$t = $params = $brace =
$rest = null;
406 if ( isset( $htmlelements[
$t = strtolower(
$t )] ) ) {
410 if( isset( $htmlsingleonly[
$t] ) ) {
412 } elseif ( ( $ot = @array_pop( $tagstack ) ) != $t ) {
413 if ( isset( $htmlsingleallowed[$ot] ) ) {
414 # Pop all elements with an optional close tag 415 # and see if we find a match below them 417 array_push ($optstack, $ot);
418 while ( ( ( $ot = @array_pop( $tagstack ) ) != $t ) &&
419 isset( $htmlsingleallowed[$ot] ) )
421 array_push ($optstack, $ot);
424 # No match. Push the optinal elements back again 426 while ( $ot = @array_pop( $optstack ) ) {
427 array_push( $tagstack, $ot );
431 @array_push( $tagstack, $ot );
432 # <li> can be nested in <ul> or <ol>, skip those cases: 433 if(!(isset( $htmllist[$ot] ) && isset( $listtags[$t] ) )) {
438 if ( $t ==
'table' ) {
439 $tagstack = array_pop( $tablestack );
444 # Keep track for later 445 if ( isset( $tabletags[$t] ) &&
446 ! in_array(
'table', $tagstack ) ) {
448 }
else if ( in_array( $t, $tagstack ) &&
449 ! isset( $htmlnest [$t ] ) ) {
451 # Is it a self closed htmlpair ? (bug 5487) 452 }
else if( $brace ==
'/>' &&
453 isset( $htmlpairs[$t] ) ) {
455 } elseif( isset( $htmlsingleonly[$t] ) ) {
456 # Hack to force empty tag for uncloseable elements 458 }
else if( isset( $htmlsingle[$t] ) ) {
459 # Hack to not close $htmlsingle tags 461 }
else if( isset( $tabletags[$t] )
462 && in_array($t ,$tagstack) ) {
466 if ( $t ==
'table' ) {
467 array_push( $tablestack, $tagstack );
470 array_push( $tagstack, $t );
473 # Replace any variables or template parameters with 475 if( is_callable( $processCallback ) ) {
476 call_user_func_array( $processCallback, array( &$params, $args ) );
479 # Strip non-approved attributes from the tag 484 $close = ( $brace ==
'/>' && !$slash ) ?
' /' :
'';
485 $text .=
"<$slash$t$newparams$close>$rest";
489 $text .=
'<' . str_replace(
'>',
'>', $x);
491 # Close off any remaining tags 492 while ( is_array( $tagstack ) && ($t = array_pop( $tagstack )) ) {
494 if ( $t ==
'table' ) { $tagstack = array_pop( $tablestack ); }
497 # this might be possible using tidy itself 498 foreach ( $bits as $x ) {
499 preg_match(
'/^(\\/?)(\\w+)([^>]*?)(\\/{0,1}>)([^<]*)$/',
501 @list( , $slash, $t, $params, $brace,
$rest ) = $regs;
502 if ( isset( $htmlelements[$t = strtolower( $t )] ) ) {
503 if( is_callable( $processCallback ) ) {
504 call_user_func_array( $processCallback, array( &$params, $args ) );
508 $text .=
"<$slash$t$newparams$brace$rest";
510 $text .=
'<' . str_replace(
'>',
'>', $x);
514 wfProfileOut( __METHOD__ );
529 wfProfileIn( __METHOD__ );
530 while (($start = strpos($text,
'<!--')) !==
false) {
531 $end = strpos($text,
'-->', $start + 4);
532 if ($end ===
false) {
533 # Unterminated comment; bail out 539 # Trim space and newline if the comment is both 540 # preceded and followed by a newline 541 $spaceStart = max($start - 1, 0);
542 $spaceLen = $end - $spaceStart;
543 while (substr($text, $spaceStart, 1) ===
' ' && $spaceStart > 0) {
547 while (substr($text, $spaceStart + $spaceLen, 1) ===
' ')
549 if (substr($text, $spaceStart, 1) ===
"\n" and substr($text, $spaceStart + $spaceLen, 1) ===
"\n") {
550 # Remove the comment, leading and trailing 551 # spaces, and leave only one newline. 552 $text = substr_replace($text,
"\n", $spaceStart, $spaceLen + 1);
555 # Remove just the comment. 556 $text = substr_replace($text,
'', $start, $end - $start);
559 wfProfileOut( __METHOD__ );
580 foreach( $attribs as $attribute => $value ) {
581 if( !isset( $whitelist[$attribute] ) ) {
584 # Strip javascript "expression" from stylesheets. 585 # http://msdn.microsoft.com/workshop/author/dhtml/overview/recalc.asp 586 if( $attribute ==
'style' ) {
588 if( $value ===
false ) {
594 if ( $attribute ===
'id' )
599 $out[$attribute] = $value;
613 static function checkCss( $value ) {
617 $stripped = StringUtils::delimiterReplace(
'/*',
'*/',
' ', $stripped );
622 $stripped = preg_replace(
'!\\\\([0-9A-Fa-f]{1,6})[ \\n\\r\\t\\f]?!e',
623 'codepointToUtf8(hexdec("$1"))', $stripped );
624 $stripped = str_replace(
'\\',
'', $stripped );
625 if( preg_match(
'/(?:expression|tps*:\/\/|url\\s*\().*/is',
654 if( trim( $text ) ==
'' ) {
662 foreach( $stripped as $attribute => $value ) {
663 $encAttribute = htmlspecialchars( $attribute );
666 $attribs[] =
"$encAttribute=\"$encValue\"";
668 return count( $attribs ) ?
' ' . implode(
' ', $attribs ) :
'';
677 $encValue = htmlspecialchars( $text );
682 $encValue = strtr( $encValue, array(
700 # Templates and links may be expanded in later parsing, 701 # creating invalid or dangerous output. Suppress this. 702 $encValue = strtr( $encValue, array(
708 "''" =>
'''',
709 'ISBN' =>
'ISBN',
711 'PMID' =>
'PMID',
717 $encValue = preg_replace_callback(
719 array(
'Sanitizer',
'armorLinksCallback' ),
739 static $replace = array(
746 return str_replace( array_keys( $replace ), array_values( $replace ), $id );
762 return rtrim(preg_replace(
763 array(
'/(^[0-9\\-])|[\\x00-\\x20!"#$%&\'()*+,.\\/:;<=>?@[\\]^`{|}~]|\\xC2\\xA0/',
'/_+/'),
775 return str_replace(
':',
':', $matches[1] );
789 if( trim( $text ) ==
'' ) {
802 foreach( $pairs as $set ) {
803 $attribute = strtolower( $set[1] );
807 $value = preg_replace(
'/[\t\r\n ]+/',
' ', $value );
808 $value = trim( $value );
825 if( isset( $set[6] ) ) {
826 # Illegal #XXXXXX color with no quotes. 828 } elseif( isset( $set[5] ) ) {
831 } elseif( isset( $set[4] ) ) {
834 } elseif( isset( $set[3] ) ) {
837 } elseif( !isset( $set[2] ) ) {
838 # In XHTML, attributes must have a value. 839 # For 'reduced' form, return explicitly the attribute name here. 842 throw new MWException(
"Tag conditions not met. This should never happen and is a bug." );
859 return str_replace(
'"',
'"',
860 self::normalizeWhitespace(
866 '/\r\n|[\x20\x0d\x0a\x09]/',
886 return preg_replace_callback(
888 array(
'Sanitizer',
'normalizeCharReferencesCallback' ),
897 if( $matches[1] !=
'' ) {
899 } elseif( $matches[2] !=
'' ) {
901 } elseif( $matches[3] !=
'' ) {
903 } elseif( $matches[4] !=
'' ) {
906 if( is_null(
$ret ) ) {
907 return htmlspecialchars( $matches[0] );
925 if ( isset( $wgHtmlEntityAliases[$name] ) ) {
926 return "&{$wgHtmlEntityAliases[$name]};";
927 } elseif( isset( $wgHtmlEntities[$name] ) ) {
930 return "&$name;";
935 $point = intval( $codepoint );
937 return sprintf(
'&#%d;', $point );
944 $point = hexdec( $codepoint );
946 return sprintf(
'&#x%x;', $point );
958 return ($codepoint == 0x09)
959 || ($codepoint == 0x0a)
960 || ($codepoint == 0x0d)
961 || ($codepoint >= 0x20 && $codepoint <= 0xd7ff)
962 || ($codepoint >= 0xe000 && $codepoint <= 0xfffd)
963 || ($codepoint >= 0x10000 && $codepoint <= 0x10ffff);
976 return preg_replace_callback(
978 array(
'Sanitizer',
'decodeCharReferencesCallback' ),
987 if( $matches[1] !=
'' ) {
989 } elseif( $matches[2] !=
'' ) {
991 } elseif( $matches[3] !=
'' ) {
993 } elseif( $matches[4] !=
'' ) {
996 # Last case should be an ampersand by itself 1026 if ( isset( $wgHtmlEntityAliases[$name] ) ) {
1027 $name = $wgHtmlEntityAliases[$name];
1029 if( isset( $wgHtmlEntities[$name] ) ) {
1045 if( !isset( $list ) ) {
1048 return isset( $list[$element] )
1058 $common = array(
'id',
'class',
'lang',
'dir',
'title',
'style' );
1059 $block = array_merge( $common, array(
'align' ) );
1060 $tablealign = array(
'align',
'char',
'charoff',
'valign' );
1061 $tablecell = array(
'abbr',
1067 'nowrap', # deprecated
1068 'width', # deprecated
1069 'height', # deprecated
1070 'bgcolor' # deprecated
1073 # Numbers refer to sections in HTML 4.01 standard describing the element. 1074 # See: http://www.w3.org/TR/html4/ 1075 $whitelist = array (
1078 'center' => $common, # deprecated
1079 'span' => $block, # ??
1097 'strong' => $common,
1108 'blockquote' => array_merge( $common, array(
'cite' ) ),
1119 'br' => array(
'id',
'class',
'title',
'style',
'clear' ),
1122 'pre' => array_merge( $common, array(
'width' ) ),
1125 'ins' => array_merge( $common, array(
'cite',
'datetime' ) ),
1126 'del' => array_merge( $common, array(
'cite',
'datetime' ) ),
1129 'ul' => array_merge( $common, array(
'type' ) ),
1130 'ol' => array_merge( $common, array(
'type',
'start' ) ),
1131 'li' => array_merge( $common, array(
'type',
'value' ) ),
1139 'table' => array_merge( $common,
1140 array(
'summary',
'width',
'border',
'frame',
1141 'rules',
'cellspacing',
'cellpadding',
1146 'caption' => array_merge( $common, array(
'align' ) ),
1149 'thead' => array_merge( $common, $tablealign ),
1150 'tfoot' => array_merge( $common, $tablealign ),
1151 'tbody' => array_merge( $common, $tablealign ),
1154 'colgroup' => array_merge( $common, array(
'span',
'width' ), $tablealign ),
1155 'col' => array_merge( $common, array(
'span',
'width' ), $tablealign ),
1158 'tr' => array_merge( $common, array(
'bgcolor' ), $tablealign ),
1161 'td' => array_merge( $common, $tablecell, $tablealign ),
1162 'th' => array_merge( $common, $tablecell, $tablealign ),
1170 'strike' => $common,
1175 'font' => array_merge( $common, array(
'size',
'color',
'face' ) ),
1179 'hr' => array_merge( $common, array(
'noshade',
'size',
'width' ) ),
1181 # XHTML Ruby annotation text module, simple ruby only.
1187 'rt' => $common, #array_merge( $common, array(
'rbspan' ) ),
1205 $text = StringUtils::delimiterReplace(
'<',
'>',
'', $text );
1207 # Normalize &entities and whitespace 1208 $text = self::decodeCharReferences( $text );
1209 $text = self::normalizeWhitespace( $text );
1226 $out =
"<!DOCTYPE html [\n";
1227 foreach( $wgHtmlEntities as $entity => $codepoint ) {
1228 $out .=
"<!ENTITY $entity \"&#$codepoint;\">";
1234 static function cleanUrl( $url, $hostname=
true ) {
1235 # Normalize any HTML entities in input. They will be 1236 # re-escaped by makeExternalLink(). 1240 # Escape any control characters introduced by the above step 1241 $url = preg_replace(
'/[\][<>"\\x00-\\x20\\x7F]/e',
"urlencode('\\0')", $url );
1243 # Validate hostname portion 1245 if( preg_match(
'!^([^:]+:)(//[^/]+)?(.*)$!iD', $url, $matches ) ) {
1246 list( , $protocol, $host,
$rest ) = $matches;
1252 \\s| # general whitespace 1253 \xc2\xad| # 00ad SOFT HYPHEN 1254 \xe1\xa0\x86| # 1806 MONGOLIAN TODO SOFT HYPHEN 1255 \xe2\x80\x8b| # 200b ZERO WIDTH SPACE 1256 \xe2\x81\xa0| # 2060 WORD JOINER 1257 \xef\xbb\xbf| # feff ZERO WIDTH NO-BREAK SPACE 1258 \xcd\x8f| # 034f COMBINING GRAPHEME JOINER 1259 \xe1\xa0\x8b| # 180b MONGOLIAN FREE VARIATION SELECTOR ONE 1260 \xe1\xa0\x8c| # 180c MONGOLIAN FREE VARIATION SELECTOR TWO 1261 \xe1\xa0\x8d| # 180d MONGOLIAN FREE VARIATION SELECTOR THREE 1262 \xe2\x80\x8c| # 200c ZERO WIDTH NON-JOINER 1263 \xe2\x80\x8d| # 200d ZERO WIDTH JOINER 1264 [\xef\xb8\x80-\xef\xb8\x8f] # fe00-fe00f VARIATION SELECTOR-1-16 1267 $host = preg_replace( $strip,
'', $host );
1271 return $protocol . $host .
$rest;
static normalizeWhitespace( $text)
static setupAttributeWhitelist()
static stripAllTags( $text)
Take a fragment of (potentially invalid) HTML and return a version with any tags removed, encoded as plain text.
$attrib
Regular expression to match HTML/XML attribute pairs within a tag.
static hackDocType()
Hack up a private DOCTYPE with HTML's standard entity declarations.
static hexCharReference( $codepoint)
const MW_CHAR_REFS_REGEX
Regular expression to match various types of character references in Sanitizer::normalizeCharReferenc...
global $wgHtmlEntities
List of all named character entities defined in HTML 4.01 http://www.w3.org/TR/html4/sgml/entities.html.
static attributeWhitelist( $element)
Fetch the whitelist of acceptable attributes for a given element name.
static decCharReference( $codepoint)
static armorLinksCallback( $matches)
Regex replace callback for armoring links against further processing.
static validateTagAttributes( $attribs, $element)
Take an array of attribute names and values and normalize or discard illegal values for the given ele...
codepointToUtf8( $codepoint)
Return UTF-8 sequence for a given Unicode code point.
static getTagAttributeCallback( $set)
Pick the appropriate attribute value from a match set from the MW_ATTRIBS_REGEX matches.
static decodeChar( $codepoint)
Return UTF-8 string for a codepoint if that is a valid character reference, otherwise U+FFFD REPLACEM...
static cleanUrl( $url, $hostname=true)
static fixTagAttributes( $text, $element)
Take a tag soup fragment listing an HTML element's attributes and normalize it to well-formed XML...
static decodeTagAttributes( $text)
Return an associative array of attribute names and values from a partial tag string.
static escapeClass( $class)
Given a value, escape it so that it can be used as a CSS class and return it.
static normalizeCharReferences( $text)
Ensure that any entities and character references are legal for XML and XHTML specifically.
static escapeId( $id)
Given a value escape it so that it can be used in an id attribute and return it, this does not valida...
static validateCodepoint( $codepoint)
Returns true if a given Unicode codepoint is a valid character in XML.
static normalizeCharReferencesCallback( $matches)
static decodeCharReferencesCallback( $matches)
static normalizeAttributeValue( $text)
Normalize whitespace and character references in an XML source- encoded text for an attribute value...
global $wgHtmlEntityAliases
Character entity aliases accepted by MediaWiki.
static encodeAttribute( $text)
Encode an attribute value for HTML output.
static removeHTMLcomments( $text)
Remove '', and everything between.
static decodeCharReferences( $text)
Decode any character references, numeric or named entities, in the text and return a UTF-8 string...
static normalizeEntity( $name)
If the named entity is defined in the HTML 4.0/XHTML 1.0 DTD, return the named entity reference as is...
static checkCss( $value)
Pick apart some CSS and check it for forbidden or unsafe structures.
static decodeEntity( $name)
If the named entity is defined in the HTML 4.0/XHTML 1.0 DTD, return the UTF-8 encoding of that chara...
static removeHTMLtags( $text, $processCallback=null, $args=array())
Cleans up HTML, removes dangerous tags and attributes, and removes HTML comments. ...
static safeEncodeAttribute( $text)
Encode an attribute value for HTML tags, with extra armoring against further wiki processing...
wfUrlProtocols()
Returns a regular expression of url protocols.