diff --git a/extlib/HTMLPurifier/HTMLPurifier.includes.php b/extlib/HTMLPurifier/HTMLPurifier.includes.php index fdb58c2d37..e8bce5c850 100644 --- a/extlib/HTMLPurifier/HTMLPurifier.includes.php +++ b/extlib/HTMLPurifier/HTMLPurifier.includes.php @@ -7,7 +7,7 @@ * primary concern and you are using an opcode cache. PLEASE DO NOT EDIT THIS * FILE, changes will be overwritten the next time the script is run. * - * @version 4.7.0 + * @version 4.9.3 * * @warning * You must *not* include any other HTML Purifier files before this file, @@ -137,6 +137,8 @@ require 'HTMLPurifier/AttrTransform/SafeObject.php'; require 'HTMLPurifier/AttrTransform/SafeParam.php'; require 'HTMLPurifier/AttrTransform/ScriptRequired.php'; require 'HTMLPurifier/AttrTransform/TargetBlank.php'; +require 'HTMLPurifier/AttrTransform/TargetNoopener.php'; +require 'HTMLPurifier/AttrTransform/TargetNoreferrer.php'; require 'HTMLPurifier/AttrTransform/Textarea.php'; require 'HTMLPurifier/ChildDef/Chameleon.php'; require 'HTMLPurifier/ChildDef/Custom.php'; @@ -175,6 +177,8 @@ require 'HTMLPurifier/HTMLModule/StyleAttribute.php'; require 'HTMLPurifier/HTMLModule/Tables.php'; require 'HTMLPurifier/HTMLModule/Target.php'; require 'HTMLPurifier/HTMLModule/TargetBlank.php'; +require 'HTMLPurifier/HTMLModule/TargetNoopener.php'; +require 'HTMLPurifier/HTMLModule/TargetNoreferrer.php'; require 'HTMLPurifier/HTMLModule/Text.php'; require 'HTMLPurifier/HTMLModule/Tidy.php'; require 'HTMLPurifier/HTMLModule/XMLCommonAttributes.php'; @@ -225,5 +229,6 @@ require 'HTMLPurifier/URIScheme/https.php'; require 'HTMLPurifier/URIScheme/mailto.php'; require 'HTMLPurifier/URIScheme/news.php'; require 'HTMLPurifier/URIScheme/nntp.php'; +require 'HTMLPurifier/URIScheme/tel.php'; require 'HTMLPurifier/VarParser/Flexible.php'; require 'HTMLPurifier/VarParser/Native.php'; diff --git a/extlib/HTMLPurifier/HTMLPurifier.php b/extlib/HTMLPurifier/HTMLPurifier.php index c6041bc113..b4605ebc6e 100644 --- a/extlib/HTMLPurifier/HTMLPurifier.php +++ b/extlib/HTMLPurifier/HTMLPurifier.php @@ -19,7 +19,7 @@ */ /* - HTML Purifier 4.7.0 - Standards Compliant HTML Filtering + HTML Purifier 4.9.3 - Standards Compliant HTML Filtering Copyright (C) 2006-2008 Edward Z. Yang This library is free software; you can redistribute it and/or @@ -58,12 +58,12 @@ class HTMLPurifier * Version of HTML Purifier. * @type string */ - public $version = '4.7.0'; + public $version = '4.9.3'; /** * Constant with version of HTML Purifier. */ - const VERSION = '4.7.0'; + const VERSION = '4.9.3'; /** * Global configuration object. @@ -104,7 +104,7 @@ class HTMLPurifier /** * Initializes the purifier. * - * @param HTMLPurifier_Config $config Optional HTMLPurifier_Config object + * @param HTMLPurifier_Config|mixed $config Optional HTMLPurifier_Config object * for all instances of the purifier, if omitted, a default * configuration is supplied (which can be overridden on a * per-use basis). diff --git a/extlib/HTMLPurifier/HTMLPurifier.safe-includes.php b/extlib/HTMLPurifier/HTMLPurifier.safe-includes.php index 9dea6d1ed5..a3261f8a32 100644 --- a/extlib/HTMLPurifier/HTMLPurifier.safe-includes.php +++ b/extlib/HTMLPurifier/HTMLPurifier.safe-includes.php @@ -131,6 +131,8 @@ require_once $__dir . '/HTMLPurifier/AttrTransform/SafeObject.php'; require_once $__dir . '/HTMLPurifier/AttrTransform/SafeParam.php'; require_once $__dir . '/HTMLPurifier/AttrTransform/ScriptRequired.php'; require_once $__dir . '/HTMLPurifier/AttrTransform/TargetBlank.php'; +require_once $__dir . '/HTMLPurifier/AttrTransform/TargetNoopener.php'; +require_once $__dir . '/HTMLPurifier/AttrTransform/TargetNoreferrer.php'; require_once $__dir . '/HTMLPurifier/AttrTransform/Textarea.php'; require_once $__dir . '/HTMLPurifier/ChildDef/Chameleon.php'; require_once $__dir . '/HTMLPurifier/ChildDef/Custom.php'; @@ -169,6 +171,8 @@ require_once $__dir . '/HTMLPurifier/HTMLModule/StyleAttribute.php'; require_once $__dir . '/HTMLPurifier/HTMLModule/Tables.php'; require_once $__dir . '/HTMLPurifier/HTMLModule/Target.php'; require_once $__dir . '/HTMLPurifier/HTMLModule/TargetBlank.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/TargetNoopener.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/TargetNoreferrer.php'; require_once $__dir . '/HTMLPurifier/HTMLModule/Text.php'; require_once $__dir . '/HTMLPurifier/HTMLModule/Tidy.php'; require_once $__dir . '/HTMLPurifier/HTMLModule/XMLCommonAttributes.php'; @@ -219,5 +223,6 @@ require_once $__dir . '/HTMLPurifier/URIScheme/https.php'; require_once $__dir . '/HTMLPurifier/URIScheme/mailto.php'; require_once $__dir . '/HTMLPurifier/URIScheme/news.php'; require_once $__dir . '/HTMLPurifier/URIScheme/nntp.php'; +require_once $__dir . '/HTMLPurifier/URIScheme/tel.php'; require_once $__dir . '/HTMLPurifier/VarParser/Flexible.php'; require_once $__dir . '/HTMLPurifier/VarParser/Native.php'; diff --git a/extlib/HTMLPurifier/HTMLPurifier/Arborize.php b/extlib/HTMLPurifier/HTMLPurifier/Arborize.php index 9e6617be5d..d2e9d22a20 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/Arborize.php +++ b/extlib/HTMLPurifier/HTMLPurifier/Arborize.php @@ -19,8 +19,8 @@ class HTMLPurifier_Arborize if ($token instanceof HTMLPurifier_Token_End) { $token->start = null; // [MUT] $r = array_pop($stack); - assert($r->name === $token->name); - assert(empty($token->attr)); + //assert($r->name === $token->name); + //assert(empty($token->attr)); $r->endCol = $token->col; $r->endLine = $token->line; $r->endArmor = $token->armor; @@ -32,7 +32,7 @@ class HTMLPurifier_Arborize $stack[] = $node; } } - assert(count($stack) == 1); + //assert(count($stack) == 1); return $stack[0]; } diff --git a/extlib/HTMLPurifier/HTMLPurifier/AttrCollections.php b/extlib/HTMLPurifier/HTMLPurifier/AttrCollections.php index 4f6c2e39a2..c7b17cf144 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/AttrCollections.php +++ b/extlib/HTMLPurifier/HTMLPurifier/AttrCollections.php @@ -21,6 +21,11 @@ class HTMLPurifier_AttrCollections * @param HTMLPurifier_HTMLModule[] $modules Hash array of HTMLPurifier_HTMLModule members */ public function __construct($attr_types, $modules) + { + $this->doConstruct($attr_types, $modules); + } + + public function doConstruct($attr_types, $modules) { // load extensions from the modules foreach ($modules as $module) { diff --git a/extlib/HTMLPurifier/HTMLPurifier/AttrDef.php b/extlib/HTMLPurifier/HTMLPurifier/AttrDef.php index 5ac06522b9..739646fa7c 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/AttrDef.php +++ b/extlib/HTMLPurifier/HTMLPurifier/AttrDef.php @@ -86,7 +86,13 @@ abstract class HTMLPurifier_AttrDef */ protected function mungeRgb($string) { - return preg_replace('/rgb\((\d+)\s*,\s*(\d+)\s*,\s*(\d+)\)/', 'rgb(\1,\2,\3)', $string); + $p = '\s*(\d+(\.\d+)?([%]?))\s*'; + + if (preg_match('/(rgba|hsla)\(/', $string)) { + return preg_replace('/(rgba|hsla)\('.$p.','.$p.','.$p.','.$p.'\)/', '\1(\2,\5,\8,\11)', $string); + } + + return preg_replace('/(rgb|hsl)\('.$p.','.$p.','.$p.'\)/', '\1(\2,\5,\8)', $string); } /** diff --git a/extlib/HTMLPurifier/HTMLPurifier/AttrDef/CSS.php b/extlib/HTMLPurifier/HTMLPurifier/AttrDef/CSS.php index 02c1641fb2..ad2cb90ad1 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/AttrDef/CSS.php +++ b/extlib/HTMLPurifier/HTMLPurifier/AttrDef/CSS.php @@ -25,15 +25,42 @@ class HTMLPurifier_AttrDef_CSS extends HTMLPurifier_AttrDef $css = $this->parseCDATA($css); $definition = $config->getCSSDefinition(); + $allow_duplicates = $config->get("CSS.AllowDuplicates"); - // we're going to break the spec and explode by semicolons. - // This is because semicolon rarely appears in escaped form - // Doing this is generally flaky but fast - // IT MIGHT APPEAR IN URIs, see HTMLPurifier_AttrDef_CSSURI - // for details - $declarations = explode(';', $css); + // According to the CSS2.1 spec, the places where a + // non-delimiting semicolon can appear are in strings + // escape sequences. So here is some dumb hack to + // handle quotes. + $len = strlen($css); + $accum = ""; + $declarations = array(); + $quoted = false; + for ($i = 0; $i < $len; $i++) { + $c = strcspn($css, ";'\"", $i); + $accum .= substr($css, $i, $c); + $i += $c; + if ($i == $len) break; + $d = $css[$i]; + if ($quoted) { + $accum .= $d; + if ($d == $quoted) { + $quoted = false; + } + } else { + if ($d == ";") { + $declarations[] = $accum; + $accum = ""; + } else { + $accum .= $d; + $quoted = $d; + } + } + } + if ($accum != "") $declarations[] = $accum; + $propvalues = array(); + $new_declarations = ''; /** * Name of the current CSS property being validated. @@ -83,7 +110,11 @@ class HTMLPurifier_AttrDef_CSS extends HTMLPurifier_AttrDef if ($result === false) { continue; } - $propvalues[$property] = $result; + if ($allow_duplicates) { + $new_declarations .= "$property:$result;"; + } else { + $propvalues[$property] = $result; + } } $context->destroy('CurrentCSSProperty'); @@ -92,7 +123,6 @@ class HTMLPurifier_AttrDef_CSS extends HTMLPurifier_AttrDef // slightly inefficient, but it's the only way of getting rid of // duplicates. Perhaps config to optimize it, but not now. - $new_declarations = ''; foreach ($propvalues as $prop => $value) { $new_declarations .= "$prop:$value;"; } diff --git a/extlib/HTMLPurifier/HTMLPurifier/AttrDef/CSS/Color.php b/extlib/HTMLPurifier/HTMLPurifier/AttrDef/CSS/Color.php index 16d2a6b98c..d7287a00c2 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/AttrDef/CSS/Color.php +++ b/extlib/HTMLPurifier/HTMLPurifier/AttrDef/CSS/Color.php @@ -6,6 +6,16 @@ class HTMLPurifier_AttrDef_CSS_Color extends HTMLPurifier_AttrDef { + /** + * @type HTMLPurifier_AttrDef_CSS_AlphaValue + */ + protected $alpha; + + public function __construct() + { + $this->alpha = new HTMLPurifier_AttrDef_CSS_AlphaValue(); + } + /** * @param string $color * @param HTMLPurifier_Config $config @@ -29,59 +39,104 @@ class HTMLPurifier_AttrDef_CSS_Color extends HTMLPurifier_AttrDef return $colors[$lower]; } - if (strpos($color, 'rgb(') !== false) { - // rgb literal handling + if (preg_match('#(rgb|rgba|hsl|hsla)\(#', $color, $matches) === 1) { $length = strlen($color); if (strpos($color, ')') !== $length - 1) { return false; } - $triad = substr($color, 4, $length - 4 - 1); - $parts = explode(',', $triad); - if (count($parts) !== 3) { + + // get used function : rgb, rgba, hsl or hsla + $function = $matches[1]; + + $parameters_size = 3; + $alpha_channel = false; + if (substr($function, -1) === 'a') { + $parameters_size = 4; + $alpha_channel = true; + } + + /* + * Allowed types for values : + * parameter_position => [type => max_value] + */ + $allowed_types = array( + 1 => array('percentage' => 100, 'integer' => 255), + 2 => array('percentage' => 100, 'integer' => 255), + 3 => array('percentage' => 100, 'integer' => 255), + ); + $allow_different_types = false; + + if (strpos($function, 'hsl') !== false) { + $allowed_types = array( + 1 => array('integer' => 360), + 2 => array('percentage' => 100), + 3 => array('percentage' => 100), + ); + $allow_different_types = true; + } + + $values = trim(str_replace($function, '', $color), ' ()'); + + $parts = explode(',', $values); + if (count($parts) !== $parameters_size) { return false; } - $type = false; // to ensure that they're all the same type + + $type = false; $new_parts = array(); + $i = 0; + foreach ($parts as $part) { + $i++; $part = trim($part); + if ($part === '') { return false; } - $length = strlen($part); - if ($part[$length - 1] === '%') { - // handle percents - if (!$type) { - $type = 'percentage'; - } elseif ($type !== 'percentage') { + + // different check for alpha channel + if ($alpha_channel === true && $i === count($parts)) { + $result = $this->alpha->validate($part, $config, $context); + + if ($result === false) { return false; } - $num = (float)substr($part, 0, $length - 1); - if ($num < 0) { - $num = 0; - } - if ($num > 100) { - $num = 100; - } - $new_parts[] = "$num%"; + + $new_parts[] = (string)$result; + continue; + } + + if (substr($part, -1) === '%') { + $current_type = 'percentage'; } else { - // handle integers - if (!$type) { - $type = 'integer'; - } elseif ($type !== 'integer') { - return false; - } - $num = (int)$part; - if ($num < 0) { - $num = 0; - } - if ($num > 255) { - $num = 255; - } - $new_parts[] = (string)$num; + $current_type = 'integer'; + } + + if (!array_key_exists($current_type, $allowed_types[$i])) { + return false; + } + + if (!$type) { + $type = $current_type; + } + + if ($allow_different_types === false && $type != $current_type) { + return false; + } + + $max_value = $allowed_types[$i][$current_type]; + + if ($current_type == 'integer') { + // Return value between range 0 -> $max_value + $new_parts[] = (int)max(min($part, $max_value), 0); + } elseif ($current_type == 'percentage') { + $new_parts[] = (float)max(min(rtrim($part, '%'), $max_value), 0) . '%'; } } - $new_triad = implode(',', $new_parts); - $color = "rgb($new_triad)"; + + $new_values = implode(',', $new_parts); + + $color = $function . '(' . $new_values . ')'; } else { // hexadecimal handling if ($color[0] === '#') { @@ -100,6 +155,7 @@ class HTMLPurifier_AttrDef_CSS_Color extends HTMLPurifier_AttrDef } return $color; } + } // vim: et sw=4 sts=4 diff --git a/extlib/HTMLPurifier/HTMLPurifier/AttrDef/CSS/FontFamily.php b/extlib/HTMLPurifier/HTMLPurifier/AttrDef/CSS/FontFamily.php index 86101020dc..74e24c8816 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/AttrDef/CSS/FontFamily.php +++ b/extlib/HTMLPurifier/HTMLPurifier/AttrDef/CSS/FontFamily.php @@ -130,6 +130,8 @@ class HTMLPurifier_AttrDef_CSS_FontFamily extends HTMLPurifier_AttrDef // . See // the CSS3 spec for more examples: // + // You can see live samples of these on the Internet: + // // However, most of these fonts have ASCII equivalents: // for example, 'MS Mincho', and it's considered // professional to use ASCII font names instead of diff --git a/extlib/HTMLPurifier/HTMLPurifier/AttrDef/CSS/URI.php b/extlib/HTMLPurifier/HTMLPurifier/AttrDef/CSS/URI.php index f9434230e2..6617acace5 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/AttrDef/CSS/URI.php +++ b/extlib/HTMLPurifier/HTMLPurifier/AttrDef/CSS/URI.php @@ -33,6 +33,9 @@ class HTMLPurifier_AttrDef_CSS_URI extends HTMLPurifier_AttrDef_URI return false; } $uri_string = substr($uri_string, 4); + if (strlen($uri_string) == 0) { + return false; + } $new_length = strlen($uri_string) - 1; if ($uri_string[$new_length] != ')') { return false; diff --git a/extlib/HTMLPurifier/HTMLPurifier/AttrDef/HTML/ID.php b/extlib/HTMLPurifier/HTMLPurifier/AttrDef/HTML/ID.php index 3d86efb44c..4ba45610fe 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/AttrDef/HTML/ID.php +++ b/extlib/HTMLPurifier/HTMLPurifier/AttrDef/HTML/ID.php @@ -72,18 +72,26 @@ class HTMLPurifier_AttrDef_HTML_ID extends HTMLPurifier_AttrDef // we purposely avoid using regex, hopefully this is faster - if (ctype_alpha($id)) { - $result = true; - } else { - if (!ctype_alpha(@$id[0])) { + if ($config->get('Attr.ID.HTML5') === true) { + if (preg_match('/[\t\n\x0b\x0c ]/', $id)) { return false; } - // primitive style of regexps, I suppose - $trim = trim( - $id, - 'A..Za..z0..9:-._' - ); - $result = ($trim === ''); + } else { + if (ctype_alpha($id)) { + // OK + } else { + if (!ctype_alpha(@$id[0])) { + return false; + } + // primitive style of regexps, I suppose + $trim = trim( + $id, + 'A..Za..z0..9:-._' + ); + if ($trim !== '') { + return false; + } + } } $regexp = $config->get('Attr.IDBlacklistRegexp'); @@ -91,14 +99,14 @@ class HTMLPurifier_AttrDef_HTML_ID extends HTMLPurifier_AttrDef return false; } - if (!$this->selector && $result) { + if (!$this->selector) { $id_accumulator->add($id); } // if no change was made to the ID, return the result // else, return the new id if stripping whitespace made it // valid, or return false. - return $result ? $id : false; + return $id; } } diff --git a/extlib/HTMLPurifier/HTMLPurifier/AttrDef/URI/Host.php b/extlib/HTMLPurifier/HTMLPurifier/AttrDef/URI/Host.php index e7df800b1e..3b4d186743 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/AttrDef/URI/Host.php +++ b/extlib/HTMLPurifier/HTMLPurifier/AttrDef/URI/Host.php @@ -76,24 +76,33 @@ class HTMLPurifier_AttrDef_URI_Host extends HTMLPurifier_AttrDef // fairly well supported. $underscore = $config->get('Core.AllowHostnameUnderscore') ? '_' : ''; + // Based off of RFC 1738, but amended so that + // as per RFC 3696, the top label need only not be all numeric. // The productions describing this are: $a = '[a-z]'; // alpha $an = '[a-z0-9]'; // alphanum $and = "[a-z0-9-$underscore]"; // alphanum | "-" // domainlabel = alphanum | alphanum *( alphanum | "-" ) alphanum - $domainlabel = "$an($and*$an)?"; - // toplabel = alpha | alpha *( alphanum | "-" ) alphanum - $toplabel = "$a($and*$an)?"; + $domainlabel = "$an(?:$and*$an)?"; + // AMENDED as per RFC 3696 + // toplabel = alphanum | alphanum *( alphanum | "-" ) alphanum + // side condition: not all numeric + $toplabel = "$an(?:$and*$an)?"; // hostname = *( domainlabel "." ) toplabel [ "." ] - if (preg_match("/^($domainlabel\.)*$toplabel\.?$/i", $string)) { - return $string; + if (preg_match("/^(?:$domainlabel\.)*($toplabel)\.?$/i", $string, $matches)) { + if (!ctype_digit($matches[1])) { + return $string; + } } + // PHP 5.3 and later support this functionality natively + if (function_exists('idn_to_ascii')) { + $string = idn_to_ascii($string); + // If we have Net_IDNA2 support, we can support IRIs by // punycoding them. (This is the most portable thing to do, // since otherwise we have to assume browsers support - - if ($config->get('Core.EnableIDNA')) { + } elseif ($config->get('Core.EnableIDNA')) { $idna = new Net_IDNA2(array('encoding' => 'utf8', 'overlong' => false, 'strict' => true)); // we need to encode each period separately $parts = explode('.', $string); @@ -114,13 +123,14 @@ class HTMLPurifier_AttrDef_URI_Host extends HTMLPurifier_AttrDef } } $string = implode('.', $new_parts); - if (preg_match("/^($domainlabel\.)*$toplabel\.?$/i", $string)) { - return $string; - } } catch (Exception $e) { // XXX error reporting } } + // Try again + if (preg_match("/^($domainlabel\.)*$toplabel\.?$/i", $string)) { + return $string; + } return false; } } diff --git a/extlib/HTMLPurifier/HTMLPurifier/AttrTransform/ImgRequired.php b/extlib/HTMLPurifier/HTMLPurifier/AttrTransform/ImgRequired.php index 7df6cb3e1b..235ebb34b6 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/AttrTransform/ImgRequired.php +++ b/extlib/HTMLPurifier/HTMLPurifier/AttrTransform/ImgRequired.php @@ -32,8 +32,7 @@ class HTMLPurifier_AttrTransform_ImgRequired extends HTMLPurifier_AttrTransform if ($src) { $alt = $config->get('Attr.DefaultImageAlt'); if ($alt === null) { - // truncate if the alt is too long - $attr['alt'] = substr(basename($attr['src']), 0, 40); + $attr['alt'] = basename($attr['src']); } else { $attr['alt'] = $alt; } diff --git a/extlib/HTMLPurifier/HTMLPurifier/AttrTransform/TargetNoopener.php b/extlib/HTMLPurifier/HTMLPurifier/AttrTransform/TargetNoopener.php new file mode 100644 index 0000000000..1db3c6c09e --- /dev/null +++ b/extlib/HTMLPurifier/HTMLPurifier/AttrTransform/TargetNoopener.php @@ -0,0 +1,37 @@ +get('CSS.MaxImgLength'); + $this->info['min-width'] = + $this->info['max-width'] = + $this->info['min-height'] = + $this->info['max-height'] = $this->info['width'] = $this->info['height'] = $max === null ? @@ -370,6 +374,19 @@ class HTMLPurifier_CSSDefinition extends HTMLPurifier_Definition ); $this->info['page-break-inside'] = new HTMLPurifier_AttrDef_Enum(array('auto', 'avoid')); + $border_radius = new HTMLPurifier_AttrDef_CSS_Composite( + array( + new HTMLPurifier_AttrDef_CSS_Percentage(true), // disallow negative + new HTMLPurifier_AttrDef_CSS_Length('0') // disallow negative + )); + + $this->info['border-top-left-radius'] = + $this->info['border-top-right-radius'] = + $this->info['border-bottom-right-radius'] = + $this->info['border-bottom-left-radius'] = new HTMLPurifier_AttrDef_CSS_Multiple($border_radius, 2); + // TODO: support SLASH syntax + $this->info['border-radius'] = new HTMLPurifier_AttrDef_CSS_Multiple($border_radius, 4); + } /** diff --git a/extlib/HTMLPurifier/HTMLPurifier/ChildDef/List.php b/extlib/HTMLPurifier/HTMLPurifier/ChildDef/List.php index 891b9f6f5b..4fc70e0efa 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/ChildDef/List.php +++ b/extlib/HTMLPurifier/HTMLPurifier/ChildDef/List.php @@ -38,13 +38,19 @@ class HTMLPurifier_ChildDef_List extends HTMLPurifier_ChildDef return false; } + // if li is not allowed, delete parent node + if (!isset($config->getHTMLDefinition()->info['li'])) { + trigger_error("Cannot allow ul/ol without allowing li", E_USER_WARNING); + return false; + } + // the new set of children $result = array(); // a little sanity check to make sure it's not ALL whitespace $all_whitespace = true; - $current_li = false; + $current_li = null; foreach ($children as $node) { if (!empty($node->is_whitespace)) { @@ -65,7 +71,7 @@ class HTMLPurifier_ChildDef_List extends HTMLPurifier_ChildDef // to handle non-list elements; non-list elements should // not be appended to an existing li; only li created // for non-list. This distinction is not currently made. - if ($current_li === false) { + if ($current_li === null) { $current_li = new HTMLPurifier_Node_Element('li'); $result[] = $current_li; } diff --git a/extlib/HTMLPurifier/HTMLPurifier/ChildDef/Table.php b/extlib/HTMLPurifier/HTMLPurifier/ChildDef/Table.php index 3e4a0f2182..cb6b3e6cdc 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/ChildDef/Table.php +++ b/extlib/HTMLPurifier/HTMLPurifier/ChildDef/Table.php @@ -203,7 +203,7 @@ class HTMLPurifier_ChildDef_Table extends HTMLPurifier_ChildDef $current_tr_tbody->children[] = $node; break; case '#PCDATA': - assert($node->is_whitespace); + //assert($node->is_whitespace); if ($current_tr_tbody === null) { $ret[] = $node; } else { diff --git a/extlib/HTMLPurifier/HTMLPurifier/Config.php b/extlib/HTMLPurifier/HTMLPurifier/Config.php index 2b2db0c264..3648364b30 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/Config.php +++ b/extlib/HTMLPurifier/HTMLPurifier/Config.php @@ -21,7 +21,7 @@ class HTMLPurifier_Config * HTML Purifier's version * @type string */ - public $version = '4.7.0'; + public $version = '4.9.3'; /** * Whether or not to automatically finalize @@ -333,7 +333,7 @@ class HTMLPurifier_Config } // Raw type might be negative when using the fully optimized form - // of stdclass, which indicates allow_null == true + // of stdClass, which indicates allow_null == true $rtype = is_int($def) ? $def : $def->type; if ($rtype < 0) { $type = -$rtype; diff --git a/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema.php b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema.php index bfbb0f92f5..655c0e97ae 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema.php +++ b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema.php @@ -24,11 +24,11 @@ class HTMLPurifier_ConfigSchema * * array( * 'Namespace' => array( - * 'Directive' => new stdclass(), + * 'Directive' => new stdClass(), * ) * ) * - * The stdclass may have the following properties: + * The stdClass may have the following properties: * * - If isAlias isn't set: * - type: Integer type of directive, see HTMLPurifier_VarParser for definitions @@ -39,8 +39,8 @@ class HTMLPurifier_ConfigSchema * - namespace: Namespace this directive aliases to * - name: Directive name this directive aliases to * - * In certain degenerate cases, stdclass will actually be an integer. In - * that case, the value is equivalent to an stdclass with the type + * In certain degenerate cases, stdClass will actually be an integer. In + * that case, the value is equivalent to an stdClass with the type * property set to the integer. If the integer is negative, type is * equal to the absolute value of integer, and allow_null is true. * @@ -105,7 +105,7 @@ class HTMLPurifier_ConfigSchema */ public function add($key, $default, $type, $allow_null) { - $obj = new stdclass(); + $obj = new stdClass(); $obj->type = is_int($type) ? $type : HTMLPurifier_VarParser::$types[$type]; if ($allow_null) { $obj->allow_null = true; @@ -152,14 +152,14 @@ class HTMLPurifier_ConfigSchema */ public function addAlias($key, $new_key) { - $obj = new stdclass; + $obj = new stdClass; $obj->key = $new_key; $obj->isAlias = true; $this->info[$key] = $obj; } /** - * Replaces any stdclass that only has the type property with type integer. + * Replaces any stdClass that only has the type property with type integer. */ public function postProcess() { diff --git a/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema.ser b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema.ser index 1e6ccd2275..371e948f1c 100644 Binary files a/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema.ser and b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema.ser differ diff --git a/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/Attr.ID.HTML5.txt b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/Attr.ID.HTML5.txt new file mode 100644 index 0000000000..735d4b7a10 --- /dev/null +++ b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/Attr.ID.HTML5.txt @@ -0,0 +1,10 @@ +Attr.ID.HTML5 +TYPE: bool/null +DEFAULT: null +VERSION: 4.8.0 +--DESCRIPTION-- +In HTML5, restrictions on the format of the id attribute have been significantly +relaxed, such that any string is valid so long as it contains no spaces and +is at least one character. In lieu of a general HTML5 compatibility flag, +set this configuration directive to true to use the relaxed rules. +--# vim: et sw=4 sts=4 diff --git a/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/CSS.AllowDuplicates.txt b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/CSS.AllowDuplicates.txt new file mode 100644 index 0000000000..4d054b1f07 --- /dev/null +++ b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/CSS.AllowDuplicates.txt @@ -0,0 +1,11 @@ +CSS.AllowDuplicates +TYPE: bool +DEFAULT: false +VERSION: 4.8.0 +--DESCRIPTION-- +

+ By default, HTML Purifier removes duplicate CSS properties, + like color:red; color:blue. If this is set to + true, duplicate properties are allowed. +

+--# vim: et sw=4 sts=4 diff --git a/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/Cache.SerializerPermissions.txt b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/Cache.SerializerPermissions.txt index b2b83d9ab6..2e0cc81044 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/Cache.SerializerPermissions.txt +++ b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/Cache.SerializerPermissions.txt @@ -1,5 +1,5 @@ Cache.SerializerPermissions -TYPE: int +TYPE: int/null VERSION: 4.3.0 DEFAULT: 0755 --DESCRIPTION-- @@ -8,4 +8,9 @@ DEFAULT: 0755 Directory permissions of the files and directories created inside the DefinitionCache/Serializer or other custom serializer path.

+

+ In HTML Purifier 4.8.0, this also supports NULL, + which means that no chmod'ing or directory creation shall + occur. +

--# vim: et sw=4 sts=4 diff --git a/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/Core.AggressivelyRemoveScript.txt b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/Core.AggressivelyRemoveScript.txt new file mode 100644 index 0000000000..b2b6ab1496 --- /dev/null +++ b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/Core.AggressivelyRemoveScript.txt @@ -0,0 +1,16 @@ +Core.AggressivelyRemoveScript +TYPE: bool +VERSION: 4.9.0 +DEFAULT: true +--DESCRIPTION-- +

+ This directive enables aggressive pre-filter removal of + script tags. This is not necessary for security, + but it can help work around a bug in libxml where embedded + HTML elements inside script sections cause the parser to + choke. To revert to pre-4.9.0 behavior, set this to false. + This directive has no effect if %Core.Trusted is true, + %Core.RemoveScriptContents is false, or %Core.HiddenElements + does not contain script. +

+--# vim: et sw=4 sts=4 diff --git a/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/Core.LegacyEntityDecoder.txt b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/Core.LegacyEntityDecoder.txt new file mode 100644 index 0000000000..392b436493 --- /dev/null +++ b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/Core.LegacyEntityDecoder.txt @@ -0,0 +1,36 @@ +Core.LegacyEntityDecoder +TYPE: bool +VERSION: 4.9.0 +DEFAULT: false +--DESCRIPTION-- +

+ Prior to HTML Purifier 4.9.0, entities were decoded by performing + a global search replace for all entities whose decoded versions + did not have special meanings under HTML, and replaced them with + their decoded versions. We would match all entities, even if they did + not have a trailing semicolon, but only if there weren't any trailing + alphanumeric characters. +

+ + + + + + +
OriginalTextAttribute
&yen;¥¥
&yen¥¥
&yena&yena&yena
&yen=¥=¥=
+

+ In HTML Purifier 4.9.0, we changed the behavior of entity parsing + to match entities that had missing trailing semicolons in less + cases, to more closely match HTML5 parsing behavior: +

+ + + + + + +
OriginalTextAttribute
&yen;¥¥
&yen¥¥
&yena¥a&yena
&yen=¥=&yen=
+

+ This flag reverts back to pre-HTML Purifier 4.9.0 behavior. +

+--# vim: et sw=4 sts=4 diff --git a/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/HTML.TargetNoopener.txt b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/HTML.TargetNoopener.txt new file mode 100644 index 0000000000..dd514c0def --- /dev/null +++ b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/HTML.TargetNoopener.txt @@ -0,0 +1,10 @@ +--# vim: et sw=4 sts=4 +HTML.TargetNoopener +TYPE: bool +VERSION: 4.8.0 +DEFAULT: TRUE +--DESCRIPTION-- +If enabled, noopener rel attributes are added to links which have +a target attribute associated with them. This prevents malicious +destinations from overwriting the original window. +--# vim: et sw=4 sts=4 diff --git a/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/HTML.TargetNoreferrer.txt b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/HTML.TargetNoreferrer.txt new file mode 100644 index 0000000000..cb5a0b0e5e --- /dev/null +++ b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/HTML.TargetNoreferrer.txt @@ -0,0 +1,9 @@ +HTML.TargetNoreferrer +TYPE: bool +VERSION: 4.8.0 +DEFAULT: TRUE +--DESCRIPTION-- +If enabled, noreferrer rel attributes are added to links which have +a target attribute associated with them. This prevents malicious +destinations from overwriting the original window. +--# vim: et sw=4 sts=4 diff --git a/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/URI.AllowedSchemes.txt b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/URI.AllowedSchemes.txt index 666635a5ff..eb97307e20 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/URI.AllowedSchemes.txt +++ b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/URI.AllowedSchemes.txt @@ -8,6 +8,7 @@ array ( 'ftp' => true, 'nntp' => true, 'news' => true, + 'tel' => true, ) --DESCRIPTION-- Whitelist that defines the schemes that a URI is allowed to have. This diff --git a/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/URI.DefaultScheme.txt b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/URI.DefaultScheme.txt index 728e378cbe..834bc08c0b 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/URI.DefaultScheme.txt +++ b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/URI.DefaultScheme.txt @@ -1,5 +1,5 @@ URI.DefaultScheme -TYPE: string +TYPE: string/null DEFAULT: 'http' --DESCRIPTION-- @@ -7,4 +7,9 @@ DEFAULT: 'http' Defines through what scheme the output will be served, in order to select the proper object validator when no scheme information is present.

+ +

+ Starting with HTML Purifier 4.9.0, the default scheme can be null, in + which case we reject all URIs which do not have explicit schemes. +

--# vim: et sw=4 sts=4 diff --git a/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/URI.Munge.txt b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/URI.Munge.txt index 179f94eb03..58c81dcc44 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/URI.Munge.txt +++ b/extlib/HTMLPurifier/HTMLPurifier/ConfigSchema/schema/URI.Munge.txt @@ -9,75 +9,75 @@ DEFAULT: NULL absolute URIs into another URI, usually a URI redirection service. This directive accepts a URI, formatted with a %s where the url-encoded original URI should be inserted (sample: - https://searx.laquadrature.net/?q=%s). -

-

+ http://www.google.com/url?q=%s). +

+

Uses for this directive: -

-
    +

    +
    • - Prevent PageRank leaks, while being fairly transparent - to users (you may also want to add some client side JavaScript to - override the text in the statusbar). Notice: - Many security experts believe that this form of protection does not deter spam-bots. + Prevent PageRank leaks, while being fairly transparent + to users (you may also want to add some client side JavaScript to + override the text in the statusbar). Notice: + Many security experts believe that this form of protection does not deter spam-bots.
    • - Redirect users to a splash page telling them they are leaving your - website. While this is poor usability practice, it is often mandated - in corporate environments. + Redirect users to a splash page telling them they are leaving your + website. While this is poor usability practice, it is often mandated + in corporate environments.
      • -
    -

    +

+

Prior to HTML Purifier 3.1.1, this directive also enabled the munging of browsable external resources, which could break things if your redirection script was a splash page or used meta tags. To revert to previous behavior, please use %URI.MungeResources. -

-

+

+

You may want to also use %URI.MungeSecretKey along with this directive in order to enforce what URIs your redirector script allows. Open redirector scripts can be a security risk and negatively affect the reputation of your domain name. -

-

+

+

Starting with HTML Purifier 3.1.1, there is also these substitutions: -

- +

+
- - - - - + + + + + - - - - - - - - - - - - - - - - - - - - + + + + + + + + + + + + + + + + + + + + -
KeyDescriptionExample <a href="">
KeyDescriptionExample <a href="">
%r1 - The URI embeds a resource
(blank) - The URI is merely a link
%nThe name of the tag this URI came froma
%mThe name of the attribute this URI came fromhref
%pThe name of the CSS property this URI came from, or blank if irrelevant
%r1 - The URI embeds a resource
(blank) - The URI is merely a link
%nThe name of the tag this URI came froma
%mThe name of the attribute this URI came fromhref
%pThe name of the CSS property this URI came from, or blank if irrelevant
-

+ +

Admittedly, these letters are somewhat arbitrary; the only stipulation was that they couldn't be a through f. r is for resource (I would have preferred e, but you take what you can get), n is for name, m was picked because it came after n (and I couldn't use a), p is for property. -

- --# vim: et sw=4 sts=4 +

+--# vim: et sw=4 sts=4 diff --git a/extlib/HTMLPurifier/HTMLPurifier/DefinitionCache.php b/extlib/HTMLPurifier/HTMLPurifier/DefinitionCache.php index 67bb5b1e69..9aa8ff354f 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/DefinitionCache.php +++ b/extlib/HTMLPurifier/HTMLPurifier/DefinitionCache.php @@ -118,7 +118,7 @@ abstract class HTMLPurifier_DefinitionCache /** * Clears all expired (older version or revision) objects from cache - * @note Be carefuly implementing this method as flush. Flush must + * @note Be careful implementing this method as flush. Flush must * not interfere with other Definition types, and cleanup() * should not be repeatedly called by userland code. * @param HTMLPurifier_Config $config diff --git a/extlib/HTMLPurifier/HTMLPurifier/DefinitionCache/Serializer.php b/extlib/HTMLPurifier/HTMLPurifier/DefinitionCache/Serializer.php index ce268d91b4..952e48d470 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/DefinitionCache/Serializer.php +++ b/extlib/HTMLPurifier/HTMLPurifier/DefinitionCache/Serializer.php @@ -97,6 +97,12 @@ class HTMLPurifier_DefinitionCache_Serializer extends HTMLPurifier_DefinitionCac } $dir = $this->generateDirectoryPath($config); $dh = opendir($dir); + // Apparently, on some versions of PHP, readdir will return + // an empty string if you pass an invalid argument to readdir. + // So you need this test. See #49. + if (false === $dh) { + return false; + } while (false !== ($filename = readdir($dh))) { if (empty($filename)) { continue; @@ -106,6 +112,8 @@ class HTMLPurifier_DefinitionCache_Serializer extends HTMLPurifier_DefinitionCac } unlink($dir . '/' . $filename); } + closedir($dh); + return true; } /** @@ -119,6 +127,10 @@ class HTMLPurifier_DefinitionCache_Serializer extends HTMLPurifier_DefinitionCac } $dir = $this->generateDirectoryPath($config); $dh = opendir($dir); + // See #49 (and above). + if (false === $dh) { + return false; + } while (false !== ($filename = readdir($dh))) { if (empty($filename)) { continue; @@ -131,6 +143,8 @@ class HTMLPurifier_DefinitionCache_Serializer extends HTMLPurifier_DefinitionCac unlink($dir . '/' . $filename); } } + closedir($dh); + return true; } /** @@ -186,11 +200,9 @@ class HTMLPurifier_DefinitionCache_Serializer extends HTMLPurifier_DefinitionCac if ($result !== false) { // set permissions of the new file (no execute) $chmod = $config->get('Cache.SerializerPermissions'); - if (!$chmod) { - $chmod = 0644; // invalid config or simpletest + if ($chmod !== null) { + chmod($file, $chmod & 0666); } - $chmod = $chmod & 0666; - chmod($file, $chmod); } return $result; } @@ -204,8 +216,10 @@ class HTMLPurifier_DefinitionCache_Serializer extends HTMLPurifier_DefinitionCac { $directory = $this->generateDirectoryPath($config); $chmod = $config->get('Cache.SerializerPermissions'); - if (!$chmod) { - $chmod = 0755; // invalid config or simpletest + if ($chmod === null) { + // TODO: This races + if (is_dir($directory)) return true; + return mkdir($directory); } if (!is_dir($directory)) { $base = $this->generateBaseDirectoryPath($config); @@ -219,15 +233,16 @@ class HTMLPurifier_DefinitionCache_Serializer extends HTMLPurifier_DefinitionCac } elseif (!$this->_testPermissions($base, $chmod)) { return false; } - mkdir($directory, $chmod); - if (!$this->_testPermissions($directory, $chmod)) { + if (!mkdir($directory, $chmod)) { trigger_error( - 'Base directory ' . $base . ' does not exist, - please create or change using %Cache.SerializerPath', + 'Could not create directory ' . $directory . '', E_USER_WARNING ); return false; } + if (!$this->_testPermissions($directory, $chmod)) { + return false; + } } elseif (!$this->_testPermissions($directory, $chmod)) { return false; } @@ -256,7 +271,7 @@ class HTMLPurifier_DefinitionCache_Serializer extends HTMLPurifier_DefinitionCac ); return false; } - if (function_exists('posix_getuid')) { + if (function_exists('posix_getuid') && $chmod !== null) { // POSIX system, we can give more specific advice if (fileowner($dir) === posix_getuid()) { // we can chmod it ourselves diff --git a/extlib/HTMLPurifier/HTMLPurifier/Encoder.php b/extlib/HTMLPurifier/HTMLPurifier/Encoder.php index fef9b58906..b94f175423 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/Encoder.php +++ b/extlib/HTMLPurifier/HTMLPurifier/Encoder.php @@ -101,6 +101,14 @@ class HTMLPurifier_Encoder * It will parse according to UTF-8 and return a valid UTF8 string, with * non-SGML codepoints excluded. * + * Specifically, it will permit: + * \x{9}\x{A}\x{D}\x{20}-\x{7E}\x{A0}-\x{D7FF}\x{E000}-\x{FFFD}\x{10000}-\x{10FFFF} + * Source: https://www.w3.org/TR/REC-xml/#NT-Char + * Arguably this function should be modernized to the HTML5 set + * of allowed characters: + * https://www.w3.org/TR/html5/syntax.html#preprocessing-the-input-stream + * which simultaneously expand and restrict the set of allowed characters. + * * @param string $str The string to clean * @param bool $force_php * @return string @@ -122,15 +130,12 @@ class HTMLPurifier_Encoder * function that needs to be able to understand UTF-8 characters. * As of right now, only smart lossless character encoding converters * would need that, and I'm probably not going to implement them. - * Once again, PHP 6 should solve all our problems. */ public static function cleanUTF8($str, $force_php = false) { // UTF-8 validity is checked since PHP 4.3.5 // This is an optimization: if the string is already valid UTF-8, no // need to do PHP stuff. 99% of the time, this will be the case. - // The regexp matches the XML char production, as well as well as excluding - // non-SGML codepoints U+007F to U+009F if (preg_match( '/^[\x{9}\x{A}\x{D}\x{20}-\x{7E}\x{A0}-\x{D7FF}\x{E000}-\x{FFFD}\x{10000}-\x{10FFFF}]*$/Du', $str @@ -255,6 +260,7 @@ class HTMLPurifier_Encoder // 7F-9F is not strictly prohibited by XML, // but it is non-SGML, and thus we don't allow it (0xA0 <= $mUcs4 && 0xD7FF >= $mUcs4) || + (0xE000 <= $mUcs4 && 0xFFFD >= $mUcs4) || (0x10000 <= $mUcs4 && 0x10FFFF >= $mUcs4) ) ) { diff --git a/extlib/HTMLPurifier/HTMLPurifier/EntityParser.php b/extlib/HTMLPurifier/HTMLPurifier/EntityParser.php index 61529dcd9d..c372b5a6a6 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/EntityParser.php +++ b/extlib/HTMLPurifier/HTMLPurifier/EntityParser.php @@ -16,6 +16,138 @@ class HTMLPurifier_EntityParser */ protected $_entity_lookup; + /** + * Callback regex string for entities in text. + * @type string + */ + protected $_textEntitiesRegex; + + /** + * Callback regex string for entities in attributes. + * @type string + */ + protected $_attrEntitiesRegex; + + /** + * Tests if the beginning of a string is a semi-optional regex + */ + protected $_semiOptionalPrefixRegex; + + public function __construct() { + // From + // http://stackoverflow.com/questions/15532252/why-is-reg-being-rendered-as-without-the-bounding-semicolon + $semi_optional = "quot|QUOT|lt|LT|gt|GT|amp|AMP|AElig|Aacute|Acirc|Agrave|Aring|Atilde|Auml|COPY|Ccedil|ETH|Eacute|Ecirc|Egrave|Euml|Iacute|Icirc|Igrave|Iuml|Ntilde|Oacute|Ocirc|Ograve|Oslash|Otilde|Ouml|REG|THORN|Uacute|Ucirc|Ugrave|Uuml|Yacute|aacute|acirc|acute|aelig|agrave|aring|atilde|auml|brvbar|ccedil|cedil|cent|copy|curren|deg|divide|eacute|ecirc|egrave|eth|euml|frac12|frac14|frac34|iacute|icirc|iexcl|igrave|iquest|iuml|laquo|macr|micro|middot|nbsp|not|ntilde|oacute|ocirc|ograve|ordf|ordm|oslash|otilde|ouml|para|plusmn|pound|raquo|reg|sect|shy|sup1|sup2|sup3|szlig|thorn|times|uacute|ucirc|ugrave|uml|uuml|yacute|yen|yuml"; + + // NB: three empty captures to put the fourth match in the right + // place + $this->_semiOptionalPrefixRegex = "/&()()()($semi_optional)/"; + + $this->_textEntitiesRegex = + '/&(?:'. + // hex + '[#]x([a-fA-F0-9]+);?|'. + // dec + '[#]0*(\d+);?|'. + // string (mandatory semicolon) + // NB: order matters: match semicolon preferentially + '([A-Za-z_:][A-Za-z0-9.\-_:]*);|'. + // string (optional semicolon) + "($semi_optional)". + ')/'; + + $this->_attrEntitiesRegex = + '/&(?:'. + // hex + '[#]x([a-fA-F0-9]+);?|'. + // dec + '[#]0*(\d+);?|'. + // string (mandatory semicolon) + // NB: order matters: match semicolon preferentially + '([A-Za-z_:][A-Za-z0-9.\-_:]*);|'. + // string (optional semicolon) + // don't match if trailing is equals or alphanumeric (URL + // like) + "($semi_optional)(?![=;A-Za-z0-9])". + ')/'; + + } + + /** + * Substitute entities with the parsed equivalents. Use this on + * textual data in an HTML document (as opposed to attributes.) + * + * @param string $string String to have entities parsed. + * @return string Parsed string. + */ + public function substituteTextEntities($string) + { + return preg_replace_callback( + $this->_textEntitiesRegex, + array($this, 'entityCallback'), + $string + ); + } + + /** + * Substitute entities with the parsed equivalents. Use this on + * attribute contents in documents. + * + * @param string $string String to have entities parsed. + * @return string Parsed string. + */ + public function substituteAttrEntities($string) + { + return preg_replace_callback( + $this->_attrEntitiesRegex, + array($this, 'entityCallback'), + $string + ); + } + + /** + * Callback function for substituteNonSpecialEntities() that does the work. + * + * @param array $matches PCRE matches array, with 0 the entire match, and + * either index 1, 2 or 3 set with a hex value, dec value, + * or string (respectively). + * @return string Replacement string. + */ + + protected function entityCallback($matches) + { + $entity = $matches[0]; + $hex_part = @$matches[1]; + $dec_part = @$matches[2]; + $named_part = empty($matches[3]) ? @$matches[4] : $matches[3]; + if ($hex_part !== NULL && $hex_part !== "") { + return HTMLPurifier_Encoder::unichr(hexdec($hex_part)); + } elseif ($dec_part !== NULL && $dec_part !== "") { + return HTMLPurifier_Encoder::unichr((int) $dec_part); + } else { + if (!$this->_entity_lookup) { + $this->_entity_lookup = HTMLPurifier_EntityLookup::instance(); + } + if (isset($this->_entity_lookup->table[$named_part])) { + return $this->_entity_lookup->table[$named_part]; + } else { + // exact match didn't match anything, so test if + // any of the semicolon optional match the prefix. + // Test that this is an EXACT match is important to + // prevent infinite loop + if (!empty($matches[3])) { + return preg_replace_callback( + $this->_semiOptionalPrefixRegex, + array($this, 'entityCallback'), + $entity + ); + } + return $entity; + } + } + } + + // LEGACY CODE BELOW + /** * Callback regex string for parsing entities. * @type string @@ -144,7 +276,7 @@ class HTMLPurifier_EntityParser $entity; } else { return isset($this->_special_ent2dec[$matches[3]]) ? - $this->_special_ent2dec[$matches[3]] : + $this->_special_dec2str[$this->_special_ent2dec[$matches[3]]] : $entity; } } diff --git a/extlib/HTMLPurifier/HTMLPurifier/Filter/ExtractStyleBlocks.php b/extlib/HTMLPurifier/HTMLPurifier/Filter/ExtractStyleBlocks.php index 08e62c16bf..66f70b0fc0 100644 --- a/extlib/HTMLPurifier/HTMLPurifier/Filter/ExtractStyleBlocks.php +++ b/extlib/HTMLPurifier/HTMLPurifier/Filter/ExtractStyleBlocks.php @@ -95,7 +95,10 @@ class HTMLPurifier_Filter_ExtractStyleBlocks extends HTMLPurifier_Filter if ($tidy !== null) { $this->_tidy = $tidy; } - $html = preg_replace_callback('#(.+)#isU', array($this, 'styleCallback'), $html); + // NB: this must be NON-greedy because if we have + // + // we must not grab foo