X-Git-Url: http://git.heureux-cyclage.org/?a=blobdiff_plain;f=includes%2FSanitizer.php;h=113a85e1125f0d1a8fb9f341dbc7629a13806fbc;hb=c00d63dda3df84fd728dc21ffe8bab8060d76ebc;hp=b422c4e4bb0c708e1737f2bf6790eff482db52f3;hpb=74a21f3bd1692dac958ddf3e09226a72b7bc65b7;p=lhc%2Fweb%2Fwiklou.git diff --git a/includes/Sanitizer.php b/includes/Sanitizer.php index b422c4e4bb..113a85e112 100644 --- a/includes/Sanitizer.php +++ b/includes/Sanitizer.php @@ -2,7 +2,7 @@ /** * XHTML sanitizer for MediaWiki * - * Copyright (C) 2002-2005 Brion Vibber et al + * Copyright © 2002-2005 Brion Vibber et al * http://www.mediawiki.org/ * * This program is free software; you can redistribute it and/or modify @@ -25,322 +25,325 @@ */ /** - * 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 + * XHTML sanitizer for MediaWiki + * @ingroup Parser */ -$attrib = '[A-Za-z0-9]'; -$space = '[\x09\x0a\x0d\x20]'; -define( 'MW_ATTRIBS_REGEX', - "/(?:^|$space)((?:xml:|xmlns:)?$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" ); +class Sanitizer { + /** + * Regular expression to match various types of character references in + * Sanitizer::normalizeCharReferences and Sanitizer::decodeCharReferences + */ + const CHAR_REFS_REGEX = + '/&([A-Za-z0-9\x80-\xff]+); + |&\#([0-9]+); + |&\#[xX]([0-9A-Fa-f]+); + |(&)/x'; -/** - * Regular expression to match URIs that could trigger script execution - */ -define( 'MW_EVIL_URI_PATTERN', '!(^|\s|\*/\s*)(javascript|vbscript)([^\w]|$)!i' ); + const EVIL_URI_PATTERN = '!(^|\s|\*/\s*)(javascript|vbscript)([^\w]|$)!i'; + const XMLNS_ATTRIBUTE_PATTERN = "/^xmlns:[:A-Z_a-z-.0-9]+$/"; -/** - * Regular expression to match namespace attributes - */ -define( 'MW_XMLNS_ATTRIBUTE_PATTRN', "/^xmlns:$attrib+$/" ); + /** + * List of all named character entities defined in HTML 4.01 + * http://www.w3.org/TR/html4/sgml/entities.html + * As well as ' which is only defined starting in XHTML1. + * @private + */ + static $htmlEntities = 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, + 'apos' => 39, + '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 + ); -/** - * 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 + */ + static $htmlEntityAliases = array( + 'רלמ' => 'rlm', + 'رلم' => 'rlm', + ); -/** - * Character entity aliases accepted by MediaWiki - */ -global $wgHtmlEntityAliases; -$wgHtmlEntityAliases = array( - 'רלמ' => 'rlm', - 'رلم' => 'rlm', -); + /** + * Lazy-initialised attributes regex, see getAttribsRegex() + */ + static $attribsRegex; + /** + * Regular expression to match HTML/XML attribute pairs within a tag. + * Allows some... latitude. + * Used in Sanitizer::fixTagAttributes and Sanitizer::decodeTagAttributes + */ + static function getAttribsRegex() { + if ( self::$attribsRegex === null ) { + $attribFirst = '[:A-Z_a-z0-9]'; + $attrib = '[:A-Z_a-z-.0-9]'; + $space = '[\x09\x0a\x0d\x20]'; + self::$attribsRegex = + "/(?:^|$space)({$attribFirst}{$attrib}*) + ($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. + ) + )?(?=$space|\$)/sx"; + } + return self::$attribsRegex; + } -/** - * XHTML sanitizer for MediaWiki - * @ingroup Parser - */ -class Sanitizer { /** * Cleans up HTML, removes dangerous tags and attributes, and * removes HTML comments @@ -367,7 +370,8 @@ class Sanitizer { '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', 'u', 'abbr' + 'ruby', 'rt' , 'rb' , 'rp', 'p', 'span', 'abbr', 'dfn', + 'kbd', 'samp', 'thead', 'tbody', 'tfoot' ); $htmlsingle = array( 'br', 'hr', 'li', 'dt', 'dd' @@ -626,7 +630,7 @@ class Sanitizer { * @todo Check for unique id attribute :P */ static function validateAttributes( $attribs, $whitelist ) { - global $wgAllowRdfaAttributes, $wgAllowMicrodataAttributes; + global $wgAllowRdfaAttributes, $wgAllowMicrodataAttributes, $wgHtml5; $whitelist = array_flip( $whitelist ); $hrefExp = '/^(' . wfUrlProtocols() . ')[^\s]+$/'; @@ -634,15 +638,16 @@ class Sanitizer { $out = array(); foreach( $attribs as $attribute => $value ) { #allow XML namespace declaration if RDFa is enabled - if ( $wgAllowRdfaAttributes && preg_match( MW_XMLNS_ATTRIBUTE_PATTRN, $attribute ) ) { - if ( !preg_match( MW_EVIL_URI_PATTERN, $value ) ) { + if ( $wgAllowRdfaAttributes && preg_match( self::XMLNS_ATTRIBUTE_PATTERN, $attribute ) ) { + if ( !preg_match( self::EVIL_URI_PATTERN, $value ) ) { $out[$attribute] = $value; } continue; } - if( !isset( $whitelist[$attribute] ) ) { + # Allow any attribute beginning with "data-", if in HTML5 mode + if ( !($wgHtml5 && preg_match( '/^data-/i', $attribute )) && !isset( $whitelist[$attribute] ) ) { continue; } @@ -664,7 +669,7 @@ class Sanitizer { $attribute === 'itemscope' || $attribute === 'itemtype' ) { #HTML5 microdata //Paranoia. Allow "simple" values but suppress javascript - if ( preg_match( MW_EVIL_URI_PATTERN, $value ) ) { + if ( preg_match( self::EVIL_URI_PATTERN, $value ) ) { continue; } } @@ -684,19 +689,6 @@ class Sanitizer { } if ( $wgAllowMicrodataAttributes ) { - # There are some complicated validity constraints we need to - # enforce here. First of all, we don't want to allow non-standard - # itemtypes. - $allowedTypes = array( - 'http://microformats.org/profile/hcard', - 'http://microformats.org/profile/hcalendar#vevent', - 'http://n.whatwg.org/work', - ); - if ( isset( $out['itemtype'] ) && !in_array( $out['itemtype'], - $allowedTypes ) ) { - # Kill everything - unset( $out['itemscope'] ); - } # itemtype, itemid, itemref don't make sense without itemscope if ( !array_key_exists( 'itemscope', $out ) ) { unset( $out['itemtype'] ); @@ -732,22 +724,35 @@ class Sanitizer { /** * Pick apart some CSS and check it for forbidden or unsafe structures. - * Returns a sanitized string, or false if it was just too evil. + * Returns a sanitized string. This sanitized string will have + * character references and escape sequences decoded, and comments + * stripped. If the input is just too evil, only a comment complaining + * about evilness will be returned. * * Currently URL references, 'expression', 'tps' are forbidden. * + * NOTE: Despite the fact that character references are decoded, the + * returned string may contain character references given certain + * clever input strings. These character references must + * be escaped before the return value is embedded in HTML. + * * @param $value String - * @return Mixed + * @return String */ static function checkCss( $value ) { + // Decode character references like { $value = Sanitizer::decodeCharReferences( $value ); - // Remove any comments; IE gets token splitting wrong - $value = StringUtils::delimiterReplace( '/*', '*/', ' ', $value ); - // Decode escape sequences and line continuation // See the grammar in the CSS 2 spec, appendix D. - static $decodeRegex, $reencodeTable; + // This has to be done AFTER decoding character references. + // This means it isn't possible for this function to return + // unsanitized escape sequences. It is possible to manufacture + // input that contains character references that decode to + // escape sequences that decode to character references, but + // it's OK for the return value to contain character references + // because the caller is supposed to escape those anyway. + static $decodeRegex; if ( !$decodeRegex ) { $space = '[\\x20\\t\\r\\n\\f]'; $nl = '(?:\\n|\\r\\n|\\r|\\f)'; @@ -762,6 +767,21 @@ class Sanitizer { } $value = preg_replace_callback( $decodeRegex, array( __CLASS__, 'cssDecodeCallback' ), $value ); + + // Remove any comments; IE gets token splitting wrong + // This must be done AFTER decoding character references and + // escape sequences, because those steps can introduce comments + // This step cannot introduce character references or escape + // sequences, because it replaces comments with spaces rather + // than removing them completely. + $value = StringUtils::delimiterReplace( '/*', '*/', ' ', $value ); + + // Remove anything after a comment-start token, to guard against + // incorrect client implementations. + $commentPos = strpos( $value, '/*' ); + if ( $commentPos !== false ) { + $value = substr( $value, 0, $commentPos ); + } // Reject problematic keywords and control characters if ( preg_match( '/[\000-\010\016-\037\177]/', $value ) ) { @@ -772,6 +792,10 @@ class Sanitizer { return $value; } + /** + * @param $matches array + * @return String + */ static function cssDecodeCallback( $matches ) { if ( $matches[1] !== '' ) { // Line continuation @@ -892,7 +916,9 @@ class Sanitizer { * * To ensure we don't have to bother escaping anything, we also strip ', ", * & even if $wgExperimentalIds is true. TODO: Is this the best tactic? - * We also strip # because it upsets IE6. + * We also strip # because it upsets IE, and % because it could be + * ambiguous if it's part of something that looks like a percent escape + * (which don't work reliably in fragments cross-browser). * * @see http://www.w3.org/TR/html401/types.html#type-name Valid characters * in the id and @@ -918,7 +944,7 @@ class Sanitizer { if ( $wgHtml5 && $wgExperimentalHtmlIds && !in_array( 'legacy', $options ) ) { $id = Sanitizer::decodeCharReferences( $id ); - $id = preg_replace( '/[ \t\n\r\f_\'"&#]+/', '_', $id ); + $id = preg_replace( '/[ \t\n\r\f_\'"&#%]+/', '_', $id ); $id = trim( $id, '_' ); if ( $id === '' ) { # Must have been all whitespace to start with. @@ -972,11 +998,10 @@ class Sanitizer { * @return String: escaped input */ static function escapeHtmlAllowEntities( $html ) { + $html = Sanitizer::decodeCharReferences( $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; } @@ -1005,7 +1030,7 @@ class Sanitizer { $attribs = array(); $pairs = array(); if( !preg_match_all( - MW_ATTRIBS_REGEX, + self::getAttribsRegex(), $text, $pairs, PREG_SET_ORDER ) ) { @@ -1028,7 +1053,7 @@ class Sanitizer { /** * Pick the appropriate attribute value from a match set from the - * MW_ATTRIBS_REGEX matches. + * attribs regex matches. * * @param $set Array * @return String @@ -1072,6 +1097,10 @@ class Sanitizer { Sanitizer::normalizeCharReferences( $text ) ) ); } + /** + * @param $text string + * @return mixed + */ private static function normalizeWhitespace( $text ) { return preg_replace( '/\r\n|[\x20\x0d\x0a\x09]/', @@ -1079,12 +1108,25 @@ class Sanitizer { $text ); } + /** + * Normalizes whitespace in a section name, such as might be returned + * by Parser::stripSectionName(), for use in the id's that are used for + * section links. + * + * @param $section String + * @return String + */ + static function normalizeSectionNameWhitespace( $section ) { + return trim( preg_replace( '/[ _]+/', ' ', $section ) ); + } + /** * Ensure that any entities and character references are legal * for XML and XHTML specifically. Any stray bits will be * &-escaped to result in a valid text fragment. * - * a. any named char refs must be known in XHTML + * a. named char refs can only be < > & ", others are + * numericized (this way we're well-formed even without a DTD) * b. any numeric char refs must be legal chars, not invalid or forbidden * c. use &#x, not &#X * d. fix or reject non-valid attributes @@ -1095,7 +1137,7 @@ class Sanitizer { */ static function normalizeCharReferences( $text ) { return preg_replace_callback( - MW_CHAR_REFS_REGEX, + self::CHAR_REFS_REGEX, array( 'Sanitizer', 'normalizeCharReferencesCallback' ), $text ); } @@ -1111,8 +1153,6 @@ class Sanitizer { $ret = Sanitizer::decCharReference( $matches[2] ); } elseif( $matches[3] != '' ) { $ret = Sanitizer::hexCharReference( $matches[3] ); - } elseif( $matches[4] != '' ) { - $ret = Sanitizer::hexCharReference( $matches[4] ); } if( is_null( $ret ) ) { return htmlspecialchars( $matches[0] ); @@ -1123,24 +1163,31 @@ class Sanitizer { /** * If the named entity is defined in the HTML 4.0/XHTML 1.0 DTD, - * 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;) + * return the equivalent numeric entity reference (except for the core < + * > & "). If the entity is a MediaWiki-specific alias, returns + * the HTML equivalent. Otherwise, returns HTML-escaped text of + * pseudo-entity source (eg &foo;) * * @param $name String * @return String */ static function normalizeEntity( $name ) { - global $wgHtmlEntities, $wgHtmlEntityAliases; - if ( isset( $wgHtmlEntityAliases[$name] ) ) { - return "&{$wgHtmlEntityAliases[$name]};"; - } elseif( isset( $wgHtmlEntities[$name] ) ) { + if ( isset( self::$htmlEntityAliases[$name] ) ) { + return '&' . self::$htmlEntityAliases[$name] . ';'; + } elseif ( in_array( $name, + array( 'lt', 'gt', 'amp', 'quot' ) ) ) { return "&$name;"; + } elseif ( isset( self::$htmlEntities[$name] ) ) { + return '&#' . self::$htmlEntities[$name] . ';'; } else { return "&$name;"; } } + /** + * @param $codepoint + * @return null|string + */ static function decCharReference( $codepoint ) { $point = intval( $codepoint ); if( Sanitizer::validateCodepoint( $point ) ) { @@ -1150,6 +1197,10 @@ class Sanitizer { } } + /** + * @param $codepoint + * @return null|string + */ static function hexCharReference( $codepoint ) { $point = hexdec( $codepoint ); if( Sanitizer::validateCodepoint( $point ) ) { @@ -1182,7 +1233,7 @@ class Sanitizer { */ public static function decodeCharReferences( $text ) { return preg_replace_callback( - MW_CHAR_REFS_REGEX, + self::CHAR_REFS_REGEX, array( 'Sanitizer', 'decodeCharReferencesCallback' ), $text ); } @@ -1200,7 +1251,7 @@ class Sanitizer { public static function decodeCharReferencesAndNormalize( $text ) { global $wgContLang; $text = preg_replace_callback( - MW_CHAR_REFS_REGEX, + self::CHAR_REFS_REGEX, array( 'Sanitizer', 'decodeCharReferencesCallback' ), $text, /* limit */ -1, $count ); @@ -1222,8 +1273,6 @@ class Sanitizer { 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]; @@ -1249,16 +1298,15 @@ class Sanitizer { * return the UTF-8 encoding of that character. Otherwise, returns * pseudo-entity source (eg &foo;) * - * @param $name Strings + * @param $name String * @return String */ static function decodeEntity( $name ) { - global $wgHtmlEntities, $wgHtmlEntityAliases; - if ( isset( $wgHtmlEntityAliases[$name] ) ) { - $name = $wgHtmlEntityAliases[$name]; + if ( isset( self::$htmlEntityAliases[$name] ) ) { + $name = self::$htmlEntityAliases[$name]; } - if( isset( $wgHtmlEntities[$name] ) ) { - return codepointToUtf8( $wgHtmlEntities[$name] ); + if( isset( self::$htmlEntities[$name] ) ) { + return codepointToUtf8( self::$htmlEntities[$name] ); } else { return "&$name;"; } @@ -1344,10 +1392,10 @@ class Sanitizer { 'em' => $common, 'strong' => $common, 'cite' => $common, - # dfn + 'dfn' => $common, 'code' => $common, - # samp - # kbd + 'samp' => $common, + 'kbd' => $common, 'var' => $common, 'abbr' => $common, # acronym @@ -1483,22 +1531,26 @@ class Sanitizer { * @return String */ static function hackDocType() { - global $wgHtmlEntities; $out = " $codepoint ) { + foreach( self::$htmlEntities as $entity => $codepoint ) { $out .= ""; } $out .= "]>\n"; return $out; } + /** + * @param $url string + * @return mixed|string + */ 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 ); + $url = preg_replace_callback( '/[\][<>"\\x00-\\x20\\x7F\|]/', + array( __CLASS__, 'cleanUrlCallback' ), $url ); # Validate hostname portion $matches = array(); @@ -1526,7 +1578,7 @@ class Sanitizer { $host = preg_replace( $strip, '', $host ); - // @todo Fixme: validate hostnames here + // @todo FIXME: Validate hostnames here return $protocol . $host . $rest; } else { @@ -1534,4 +1586,61 @@ class Sanitizer { } } + /** + * @param $matches array + * @return string + */ + static function cleanUrlCallback( $matches ) { + return urlencode( $matches[0] ); + } + + /** + * Does a string look like an e-mail address? + * + * This validates an email address using an HTML5 specification found at: + * http://www.whatwg.org/specs/web-apps/current-work/multipage/states-of-the-type-attribute.html#valid-e-mail-address + * Which as of 2011-01-24 says: + * + * A valid e-mail address is a string that matches the ABNF production + * 1*( atext / "." ) "@" ldh-str *( "." ldh-str ) where atext is defined + * in RFC 5322 section 3.2.3, and ldh-str is defined in RFC 1034 section + * 3.5. + * + * This function is an implementation of the specification as requested in + * bug 22449. + * + * Client-side forms will use the same standard validation rules via JS or + * HTML 5 validation; additional restrictions can be enforced server-side + * by extensions via the 'isValidEmailAddr' hook. + * + * Note that this validation doesn't 100% match RFC 2822, but is believed + * to be liberal enough for wide use. Some invalid addresses will still + * pass validation here. + * + * @param $addr String E-mail address + * @return Bool + */ + public static function validateEmail( $addr ) { + $result = null; + if( !wfRunHooks( 'isValidEmailAddr', array( $addr, &$result ) ) ) { + return $result; + } + + // Please note strings below are enclosed in brackets [], this make the + // hyphen "-" a range indicator. Hence it is double backslashed below. + // See bug 26948 + $rfc5322_atext = "a-z0-9!#$%&'*+\\-\/=?^_`{|}~" ; + $rfc1034_ldh_str = "a-z0-9\\-" ; + + $HTML5_email_regexp = "/ + ^ # start of string + [$rfc5322_atext\\.]+ # user part which is liberal :p + @ # 'apostrophe' + [$rfc1034_ldh_str]+ # First domain part + (\\.[$rfc1034_ldh_str]+)* # Following part prefixed with a dot + $ # End of string + /ix" ; // case Insensitive, eXtended + + return (bool) preg_match( $HTML5_email_regexp, $addr ); + } }