($space*=$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";
}
public static function removeHTMLtags( $text, $processCallback = null,
$args = array(), $extratags = array(), $removetags = array()
) {
- global $wgUseTidy;
-
extract( self::getRecognizedTagData( $extratags, $removetags ) );
# Remove HTML comments
$text = Sanitizer::removeHTMLcomments( $text );
$bits = explode( '<', $text );
$text = str_replace( '>', '>', array_shift( $bits ) );
- if ( !$wgUseTidy ) {
+ if ( !MWTidy::isEnabled() ) {
$tagstack = $tablestack = array();
foreach ( $bits as $x ) {
$regs = array();
$value = self::normalizeCss( $value );
// Reject problematic keywords and control characters
- if ( preg_match( '/[\000-\010\013\016-\037\177]/', $value ) ) {
+ if ( preg_match( '/[\000-\010\013\016-\037\177]/', $value ) ||
+ strpos( $value, UtfNormal\Constants::UTF8_REPLACEMENT ) !== false ) {
return '/* invalid control char */';
} elseif ( preg_match(
'! expression
* @return string
*/
private static function getTagAttributeCallback( $set ) {
- if ( isset( $set[6] ) ) {
- # Illegal #XXXXXX color with no quotes.
- return $set[6];
- } elseif ( isset( $set[5] ) ) {
+ if ( isset( $set[5] ) ) {
# No quotes.
return $set[5];
} elseif ( isset( $set[4] ) ) {
# 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];
+ # In XHTML, attributes must have a value so return an empty string.
+ # See "Empty attribute syntax",
+ # http://www.w3.org/TR/html5/syntax.html#syntax-attribute-name
+ return "";
} else {
throw new MWException( "Tag conditions not met. This should never happen and is a bug." );
}
}
/**
- * Returns true if a given Unicode codepoint is a valid character in XML.
+ * Returns true if a given Unicode codepoint is a valid character in
+ * both HTML5 and XML.
* @param int $codepoint
* @return bool
*/
private static function validateCodepoint( $codepoint ) {
+ # U+000C is valid in HTML5 but not allowed in XML.
+ # U+000D is valid in XML but not allowed in HTML5.
+ # U+007F - U+009F are disallowed in HTML5 (control characters).
return $codepoint == 0x09
|| $codepoint == 0x0a
- || $codepoint == 0x0d
- || ( $codepoint >= 0x20 && $codepoint <= 0xd7ff )
+ || ( $codepoint >= 0x20 && $codepoint <= 0x7e )
+ || ( $codepoint >= 0xa0 && $codepoint <= 0xd7ff )
|| ( $codepoint >= 0xe000 && $codepoint <= 0xfffd )
|| ( $codepoint >= 0x10000 && $codepoint <= 0x10ffff );
}
$host = preg_replace( $strip, '', $host );
+ // IPv6 host names are bracketed with []. Url-decode these.
+ if ( substr_compare( "//%5B", $host, 0, 5 ) === 0 && preg_match( '!^//%5B([0-9A-Fa-f:.]+)%5D((:\d+)?)$!', $host, $matches ) ) {
+ $host = '//[' . $matches[1] . ']' . $matches[2];
+ }
+
// @todo FIXME: Validate hostnames here
return $protocol . $host . $rest;